Said Salaheldeen Elnashaie Firoozeh Danafar Hassan...

Said Salaheldeen ElnashaieFiroozeh DanafarHassan Hashemipour Rafsanjani

Nanotechnology for Chemical Engineers

Nanotechnology for Chemical Engineers

Said Salaheldeen ElnashaieFiroozeh DanafarHassan Hashemipour Rafsanjani

Nanotechnology forChemical Engineers

123

Said Salaheldeen ElnashaieChemical and Environmental EngineeringDepartment

University Putra Malaysia (UPM)SerdangMalaysia

and

Chemical and Biological EngineeringDepartment

University of British Columbia (UBC)VancouverCanada

Firoozeh DanafarChemical Engineering DepartmentShahid Bahonar University of KermanKermanIran

Hassan Hashemipour RafsanjaniChemical Engineering DepartmentShahid Bahonar University of KermanKermanIran

ISBN 978-981-287-495-5 ISBN 978-981-287-496-2 (eBook)DOI 10.1007/978-981-287-496-2

Library of Congress Control Number: 2015942473

Springer Singapore Heidelberg New York Dordrecht London© Springer Science+Business Media Singapore 2015This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or partof the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmissionor information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilarmethodology now known or hereafter developed.The use of general descriptive names, registered names, trademarks, service marks, etc. in thispublication does not imply, even in the absence of a specific statement, that such names are exempt fromthe relevant protective laws and regulations and therefore free for general use.The publisher, the authors and the editors are safe to assume that the advice and information in thisbook are believed to be true and accurate at the date of publication. Neither the publisher nor theauthors or the editors give a warranty, express or implied, with respect to the material contained herein orfor any errors or omissions that may have been made.

Printed on acid-free paper

Springer Science+Business Media Singapore Pte Ltd. is part of Springer Science+Business Media(www.springer.com)

Preface

“Nanotechnology for Chemical Engineers” is a revolutionary book describing thebasic principles of transforming nanotechnology into nanoengineering with a par-ticular focus on chemical engineering fundamentals. This book aims to provide vitalinformation about differences between descriptive technology and quantitativeengineering for students as well as working professionals in various fields ofnanotechnology. Besides chemical engineering principles, the fundamentals ofnanotechnology is also covered along with detailed explanation of several specificnanoscale processes from chemical engineering point of view. This information ispresented as practical examples and case studies that help the engineers andresearchers to integrate the processes which can meet the commercial production. Itis worth mentioning here that the main challenge in production of nanostructure andnanodevices is nowadays related to the economic point of view.

The uniqueness of this book is a balance between important insight into thesynthetic methods of nanostructures and their applications with chemical engi-neering rules that educates the readers about process design, simulation, modeling,and optimization. Briefly, the book takes the readers through a journey from fun-damentals to frontiers of engineering of processes involved in production ofnanostructures and those products comprising one or more nanostructures andinforms them about industrial perspective research challenges, opportunities, andsynergism in chemical engineering and nanotechnology. Utilizing this informationthe readers can make informed decisions on their career and business.

v

Acknowledgments

We would like to thank our families for their supports.We would also like to express our great gratitude to Mohammad Rezazadeh

Mehrjou who despites of all his duties helped us to accomplish the book.

vii

Contents

1 Chemical Engineering from Technology to Engineering . . . . . . . . . 11.1 Differences Between Technology and Engineering

in General and Focusing on this Difference as Relatedto Chemical Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.2 Concepts, Paradigms and Historical Perspective . . . . . . . . . . . . . 41.2.1 Traditional Paradigms of Chemical Engineering . . . . . . . . 41.2.2 Chemical Engineering in the Twenty-First Century

and the Integrated System Approach (ISA) Basedon System Theory (ST) . . . . . . . . . . . . . . . . . . . . . . . . . 8

1.2.3 Modern Trends in Chemical Engineering. . . . . . . . . . . . . 111.2.4 Multi-disciplinary Approach for Development

of Novel Processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161.2.5 Innovation and Sequential De-Bottle-Necking . . . . . . . . . 191.2.6 Intensification Through Integration and Combination

of Different Processes in Single Units . . . . . . . . . . . . . . . 211.2.7 Chemical Engineering Expectations on 2020 . . . . . . . . . . 251.2.8 Indispensable Tools for the Success of Chemical

and Nanoengineering . . . . . . . . . . . . . . . . . . . . . . . . . . 281.2.9 New Tools, Outlooks and Opportunities for Chemical

Engineering in Relation to the Other EngineeringDisciplines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

1.3 Principles of Chemical Engineering . . . . . . . . . . . . . . . . . . . . . 341.3.1 Generalized Mass, Momentum and Energy Balances

for Multiple Inputs-Multiple Outputs (MIMO),Systems with Multiple Reactions (MRs) . . . . . . . . . . . . . 37

1.3.2 Stationary Non-equilibrium State Modeling Approachfor Chemical Engineering Systems Basedon MIMO-MRs Generalized Massand Energy Balances . . . . . . . . . . . . . . . . . . . . . . . . . . 56

ix

http://dx.doi.org/10.1007/978-981-287-496-2_1

http://dx.doi.org/10.1007/978-981-287-496-2_1

http://dx.doi.org/10.1007/978-981-287-496-2_1#Sec1










































1.3.3 Dynamic Modeling Based on MIMO-MRs . . . . . . . . . . . 621.3.4 Simulation and Optimization of Chemical

Engineering Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . 651.4 Chemical Engineering and New Materials . . . . . . . . . . . . . . . . . 651.5 Preliminary Introduction to Nano Scale Process Engineering . . . . 67Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

2 From Nanotechnology to Nanoengineering. . . . . . . . . . . . . . . . . . . 792.1 Introduction to Nanotechnology . . . . . . . . . . . . . . . . . . . . . . . . 79

2.1.1 Application of Nanotechnology in Different Fields . . . . . . 802.1.2 Nanostructured Materials Synthesis, Concepts

and Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 932.1.3 Routine Tests for Characterization of Nanostructures . . . . 102

2.2 Transforming Nanotechnology into NanoengineeringThrough Chemical Engineering Principles . . . . . . . . . . . . . . . . . 1222.2.1 Nanotechnology in Support of General Science

and Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1242.2.2 An Industrial Perspective Research Challenges

in Nanotechnology . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

3 Learning Synergism in Nanotechnology and ChemicalEngineering by Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1793.1 Chemical Vapor Deposition (CVD) Techniques

in Nanotechnology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1793.1.1 Basic Principles of CVD . . . . . . . . . . . . . . . . . . . . . . . . 1823.1.2 Constraints on CVD Processes . . . . . . . . . . . . . . . . . . . . 1943.1.3 CVD Synthesis of Carbon Nanotubes (CNTs) . . . . . . . . . 195

3.2 Chemical Vapor Synthesis (CVS) of Nanostructures . . . . . . . . . . 2343.2.1 Basic Principles of CVS . . . . . . . . . . . . . . . . . . . . . . . . 2343.2.2 Similarity Between CVS, CVD and Inert

Gas Condensation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2353.2.3 Nanoparticle Synthesis via CVS . . . . . . . . . . . . . . . . . . . 2383.2.4 The CVS Process Simulation . . . . . . . . . . . . . . . . . . . . . 2413.2.5 Sintering Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244

3.3 Precipitation of Nanostructures . . . . . . . . . . . . . . . . . . . . . . . . . 2473.3.1 Definition of the Process . . . . . . . . . . . . . . . . . . . . . . . . 2473.3.2 Effect of the Process Parameters on the Produced

Nanoparticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

x Contents











http://dx.doi.org/10.1007/978-981-287-496-2_1#Bib1

http://dx.doi.org/10.1007/978-981-287-496-2_2

http://dx.doi.org/10.1007/978-981-287-496-2_2





















http://dx.doi.org/10.1007/978-981-287-496-2_3

http://dx.doi.org/10.1007/978-981-287-496-2_3

http://dx.doi.org/10.1007/978-981-287-496-2_3






























3.3.3 General Description of the Nucleationand Growth Steps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252

Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264

4 Conclusions and Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273

About the Authors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275

Contents xi






http://dx.doi.org/10.1007/978-981-287-496-2_4

http://dx.doi.org/10.1007/978-981-287-496-2_4

Chapter 1Chemical Engineering from Technologyto Engineering

1.1 Differences Between Technology and Engineeringin General and Focusing on this Difference as Relatedto Chemical Engineering

Although the two terms “technology” and “engineering” are interlaced, the prin-ciples of each subject and their capabilities are quite different. Generally, definitionof technology is broader than that of engineering, and from a normative standpoint,it is a part of coevolution process with society (Keulartz et al. 2004). Technologyemerges from ideas and wills for creating and utilizing the artifacts that fulfillhuman needs or desires. The outcome of technology is just producing andemploying objects, and it does not depend upon thinking and using design equa-tions, theories, etc. Accordingly, skills and arts without specific knowledge aresufficient in creating and managing the technology. Understanding and improvingthe existing technology, or resolving the problems pertinent to it requires utilizingprinciples of science and engineering knowledge. Technology is more descriptiveand empirical than engineering.

In general, it is usually not enough to create a technically successful product, andit must also meet further requirements. For example, it is essential to introduce anew product that performs as well as expected, does not cause unintended harm tothe public at large, especially the environment, and also achieves sustainability.Engineers take this responsibility to identify, understand and interpret the con-straints on a design in order to produce a successful result. They typically attempt topredict how well their designs will perform to their specifications prior to full-scaleproduction. In developing technology, engineers carefully consider the constraintsincluding available resources, physical or technical limitations, flexibility for futuremodifications and some other factors, such as cost, safety, marketability and ser-viceability. Engineering is then defined as a creative application of scientific,economic, social and practical principles to design, construct, operate, develop andmaintain structures, machines, devices, apparatus, systems, materials and/or

© Springer Science+Business Media Singapore 2015S. Salaheldeen Elnashaie et al., Nanotechnology for Chemical Engineers,DOI 10.1007/978-981-287-496-2_1

1

manufacturing processes as required for an intended function, economics of oper-ation or safety to life and property. With full cognizance of the object or process,engineers forecast the behavior of their designed equipment and/or plants underspecific operating conditions.

Engineering is the ability not only to create a technology but also to solveproblems, improve the process and construct the product by applying the engi-neering principles. Engineering principles involve a systematic and often iterativeapproach to accomplishing goals in order to meet human needs and/or societyconcerns. Engineers should know how to define a solvable problem, to test thepotential solutions and finally to reach an optimal solution by making trade-offsamong multiple concerns, such as functional, ethical, economic, esthetic and socialfactors. Engineers apply mathematics, other sciences and economics accompaniedwith their logic and tacit knowledge to design novel processes or to find appropriatesolutions to existing problems or to improve the status quo. This job is basicallyperformed by creating proper mathematical models (design equations) that allowanalyzing the system and/or its operation. Engineers respond to the interests andneeds of society and in turn affect society and the environment by bringing abouttechnological changes. Consequently, it is fair to note that engineers have animportant role in the coevolution process of society and technology. Engineering ismore quantitative and less empirical than technology.

The distinction between engineering and technology emanates primarily fromdifferences in their educational programs. Engineering curriculums are orientedtoward development of conceptual skills and include a sequence of fundamentalsand courses built on a foundation of complex mathematics as well as sciencecourses. Technology programs are geared toward applications and provide thestudents with only introductory mathematics and science courses accompanied witha qualitative introduction to engineering fundamentals. Relying on these differ-ences, nowadays, engineering and technology programs are evaluated andaccredited using two separate sets of accreditation criteria (Williams 2000;Blumenthal and Grothus 2008). Throughout history, technology has been created tosatisfy human wishes and requirements. Much of modern technology is a product ofscience and engineering, and existing technological tools are used in both fields.Technology education relies on study of the human-made world, including artifacts,processes and their underlying principles and concepts, and the overall aim oftechnology education is to provide students to participate effectively in techno-logically dependent world. There are no constraints in developing and usingtechnology. However, in engineering design, a great attention is made on con-straints including the laws of nature, or science, time, money, available materials,ergonomics, environmental regulations, manufacturability and reparability (Yore2011). In this regard, engineering utilizes concepts in science and management aswell as technological tools to accomplish their responsibilities.

Graduates from engineering programs are called engineers, while graduates oftechnology programs are called technologists. Clearly, engineering graduates’career differs from those with technology background, technologist. Engineeringgraduates are with a breadth and depth of knowledge that allows them to function as

2 1 Chemical Engineering from Technology to Engineering

conceptual designers and operators in product and process development. Onceengineers enter the workforce, they typically spend their time planning, whiletechnologists spend their time making plans work.

Nowadays, the engineering disciplines are thought as the repositories of tech-nological knowledge and their practitioners as the primary agents of technologicalchange in their respective industries. The growth of useful technological knowledgeis the product of what goes on inside the engineering disciplines (Rosenberg 1998;Broens and de Vries 2003; Huber 2012). To learn how technological learningaccumulates, it is necessary to look carefully at the engineering professions. Forexample, in early days of chemical industries, technological knowledge simply wasused in the creation of a particular and singular product or process. However, withthe development of the concept of unit operations and its codification in textbooksof chemical engineering, a given amount of inventive effort led to a larger spreadout to future inventors (Rosenberg 1998). As a result, chemical engineering wasdeveloped and replaced chemical technology. In other words, chemical engineeringcame into existence in response to the emergence of new industries and technol-ogies. Chemical engineering is a body of knowledge about the design of product,which their production involves chemical transformations, as well as their processplants (Rosenberg 1998; Favre et al. 2002; Seider et al. 2009).

In technology, an object is produced to serve a function, and in general, goal orproduct is not researched itself. According to the preliminary definition of tech-nology as “the effort to avoid effort,” the results or outcomes of the technology arenot concerned (Poel and Goldberg 2007). However, the commercialization potentialof any technological object, its further development and its impact on the societyand environment should be well researched and identified. The importance of theseconsiderations was strongly felt when World War II began (Steffensen et al. 2000).After World War II, the empirical training of engineers proved inadequate to meetthe growing demands for new processes and hence chemical engineering educationstarted to change. Chemical engineering was becoming more focused on scienceand mathematics than on engineering tradition. Aris and Amundson (1958a, b)began emphasizing the importance of mathematical modeling using dimensionlessquantities in reactor design. Bird et al. (1960) presented a unified mathematicaldescription of mass, momentum and energy transfer in their now famous text,“Transport Phenomena.” Their work helps encourage greater mathematical com-petence in chemical engineering education (Aris 1999). Engineers are, nowadays,expected to be experts in mathematics and relevant sciences for design, innovationand trouble shooting of a system or process. The engineering tasks are generallycreative and iterative that requires multiplicity of knowledge, which constitutes theessence of research, modeling, fabricating, testing and communicating. Evidently,these are not concepts only associate with the mathematics and sciences learned byan engineer at the university, and engineers must update their knowledge and skills(Winkelman 2009). Besides that, engineers mostly face a great challenge aftergraduation, which is the expectations of industries, as they usually prefer to employmulti-disciplinary and skilled engineers who able to work at different departments,such as design, innovation, maintenance and marketing to provide both the plant

1.1 Differences Between Technology and Engineering … 3

and customer satisfaction (Joshi 2013). Engineering research in industry andresearch institute consists of a wide variety of activities ranging from study ofmaterial properties for possible future application to the testing necessary toestablish design parameters or to verify the adequacy of new design concepts (Mayand Chubin 2003). Engineers are recommended to focus on not only a specificdisciplinary but also multi-disciplinary research. The research in engineering mustuse quantitative as well as qualitative approaches. The impacts of technologicalinnovation on society must be assessed by considering evolving marketing, envi-ronmental, sustainability, legal and commercial implications. One of the mostcritical elements of the innovation process is the long-term research required totransform new knowledge generated by fundamental scientific discoveries into thenovel new products, processes and services required by society. Consequently, inthe engineering carrier, research plays a major role and is the most criticalstep. Engineers spend a lot of time researching about the product or process they areworking on. They also research to learn about the problem they need to resolve aswell as to gain information when they run into difficulties. Accordingly, they needto keep on learning new materials throughout their career that means Life LongLearning (L3) is vital for engineers. Technology and engineering possess two dif-ferent stages of development, i.e., the route to engineering started by science,followed by applied science, followed by technology and then followed by engi-neering. Therefore, we should be careful in using terms of engineering andtechnology.

1.2 Concepts, Paradigms and Historical Perspective

In attempting to construct a set of visions for the readers, the following will presenthow chemical engineering came to what it is today. The main objective of thissection is to outline the development of the profession of chemical engineeringstarting from its origins in the last eighteenth century up to now.

1.2.1 Traditional Paradigms of Chemical Engineering

Ancient people traditionally made chemical products to provide their daily liferequirements. Chemical operations in which raw materials such as coal, oil and saltconverted into a variety of products can be traced back to the area B.C. MiddleEastern artisans’ refined alkali and limestone for the production of glass as early as7000 B.C. (Brill 1970; Barkoudah 2006). The Phoenicians produced soap in thesixth century B.C. (Routhet et al. 1996). In the last decades of the eighteenthcentury, advent of the industrial revolution in Northern Europe boosted industrialactivities and brought about considerable development in the production and con-sumption of chemicals (Landes 2003). Chemical industry in its modern sense


originated with linking together of the sulfuric acid manufacturing and alkalis. Inparticular, increasing demand of alkalis for manufacture of soap, glass, textile andbleach caused the traditional procedure to be deficient for the production of suchchemicals (Rousseau and Porter 1980). In response to the market, a method forconverting common salt into soda ash was established. This process became thecentral operation of the world for about 100 years (Perkins 2003). Chemicaltechnologists who were chemists or mechanical engineers had experienced workingin chemical industries took the responsibilities of the managing, operation anddeveloping of the plants.

Chemists have a vast knowledge of usually only chemistry and laboratoryprocedures. As a consequence, they generally focus on products by working on thefundamentals of chemicals, such as discovering a new compound or new ways toextract or refine a compound. They also consider individual reactions employed inmanufacturing, and they are not concerned with producing that compound on alarge scale or lowering the cost of producing the compound. Mechanical engineersalso focus on machinery, and basically, they do not have any knowledge ofchemistry, which is required in chemical plants. As described, neither chemists normechanical engineers focus on processes or operation conditions for production oflarge variety of products. Despite their effective role in some parts, their lack ofknowledge in solving the problems facing the industry and revolution of existingprocessing plants clarified the necessity for a new branch of engineering thatequally comprised both applied chemistry and traditional engineering.Consequently, engineers having general knowledge of chemistry accompanied witha vast knowledge of engineering principles like thermodynamics, fluid mechanics,mass and energy balance and the underlying mathematics were strongly needed. Inthis respect, some European countries and USA began to develop a new type ofengineering profession to meet the needs of the chemical industry (Perkins 2003).Engineers were preferred, because they have basic knowledge of science with moreabstract approach to problem solving. Accordingly, establishment of chemicalengineering emerged, a profession that applies the knowledge gained throughchemistry into commercial production of useful materials for the society.

In parallel with the industrial developments, the conceptual basis of the chem-istry was being established by enunciating the principle of conservation of mass(Lavoisier 1965):

Nothing is created…, and one can set a principle that in every process there is an equalamount of matter in the beginning and at the end of the process.

The atomic theory was then developed in the first half of the nineteenth centuryby John Dalton, JonsJakob Berzelius and others (Siegfried 2002), who describe thestructure of matter and laws in chemical phenomena. With the emergence of thenew science of chemistry, education and training of technical personnel able towork in burgeoning chemical industry became an issue (Perkins 2003; Rousseauand Porter 1980).

1.2 Concepts, Paradigms and Historical Perspective 5

Traditionally, engineers were educated through a “learning by doing” approachby working in industrial fields until they had acquired enough skills to work on theirown. In the late ninetieth century, an independent academic discipline of chemicalengineering was established. Individuals interested in engineering started enrollingin university courses. In the early twentieth century, chemical engineers, whopossessed a university qualification, started their careers. Chemical engineeringeducation was initially relied on some general principles which guide the con-struction of the chemical engineering curriculum. The emphasis was still on thevalue of hands-on experience; however, some science courses were included toimprove their practical skills (Seely 1999).

A key figure in development of chemical engineering was George E. Davis. Heis also the man most responsible for the initiation of the chemical engineeringprofession. In the first step, in 1880, George E. Davis proposed the formation of a“Society of Chemical Engineers” (Perkins 2003). In 1888, he molded his knowl-edge into a series of 12 lectures about individual chemical operations, a subject thathe empirically gained information about them. Davis analyzed the chemical pro-cesses as combinations or sequences of basic operations. In the early twentiethcentury, Davis accomplished his great job by publishing Handbook of ChemicalEngineering (Davis 1901). In this handbook, he stressed the value of chemicalengineering approaches including large-scale experimentation as the precursor ofthe modern pilot plant, safety practices and unit operations. Anyway, the curriculumtaught in the universities and institutes at that time consisted of separate courses inchemistry and conventional engineering (Cohen 1996). A systematic survey wasconducted on the existing chemical engineering courses. The researchers concludedthat a radical change in courses and teaching methods is beneficial. In the finalreport, they declared: “Chemical engineering…, is not a composite of chemistryand mechanical and civil engineering, but a science of itself, the basis of which isthose unit operations which in their proper sequence and coordination constitute achemical process as conducted on the industrial scale” (Sharifi 2002).

The survey and Davis’ handbook were corner stones for the creation of chemicalengineering curriculum (Sharifi 2002). In the first step, course of unit operation andthen other subjects like material and energy balances, chemical engineering ther-modynamics, process control, reactor design and transport phenomena were addedto the education instruction of chemical engineering. The importance and necessityof these courses are explained in the following paragraphs.

More typically, chemical engineers deal with the processes that turn rawmaterials into valuable products through physical/chemical/biochemical processes.These processes are complex as they are affected by a variety of conditions such astemperature and pressure. The chemical engineering approaches are distinguishedfrom industrial or applied chemistry and from mechanical engineering. Chemicalengineers study these complex processes by breaking them up into smaller units,and any process, on whatever scale conducted, can be resolved into a series of unitprocesses. It is believed that the concept of unit processes was a key to establish theindependent profession and academic discipline of chemical engineering (Perkins2003). Selection, design and operating of unit processes with acceptable and


desirable outcome are a demanding and hard job. The unit processes are generallyclassified into two main categories, physical and chemical operations. Physicaloperations called “Unit Operations” such as evaporation, distillation, distillation,absorption, adsorption, extraction, drying, filtration, agitation, precipitation, flui-dization, emulsification, crystallization and agglomeration. Chemical operationsinclude all process where catalytic, biochemical, electrochemical, photochemicaland agrochemical reactions take place. The number of these basic operations is notlarge, and relatively few of them are involved in any particular process. Althoughunit operations are the cornerstone of chemical engineering, they do not encompassvery important chemical aspects of the chemical industry, the processes in whichchemical reactions take place. In most chemical industries, chemical reactors aresomehow heart of the process. The chemical operations that include chemicalreactions are called unit processes.

The essential skills of chemical engineers encompass all aspects of design,testing, scale-up, operation, control and optimization of the various unit processes.Key to these developments necessitates a detailed understanding of unit operationsand unit processes and a more substantial knowledge of the fundamentals oftransport phenomena. Any separate unit process involves a combination of the basicprinciples in the transfer of heat, mass and momentum and chemical reaction if it isa unit process not a unit operation. Then, a scientific approach to chemical engi-neering can be only acquired through these principles. The publication of TransportPhenomena, in mid-twentieth century, was an important milestone in the estab-lishment of a new course in transport phenomena for chemical engineers (Bird et al.1960). Principles of transport phenomena enabled chemical engineers to formulateand solve a wider variety of problems in the current as well as new areas of theiractivity. The principles of the transport phenomena (including mass, heat andmomentum transfer) were generalized from the unit processes incorporating a greatdeal of physics and mathematics. Unit operations are basically controlled bymaterial and energy balances based on mass and heat transfer rates. In chemicalprocesses besides mass and heat transfer rates, rate(s) of chemical reaction(s) whichare the basis of transforming material and energy balances into sizing equations arealso the basis for design. The principles of transport phenomena were used incombining the chemistry involved in industrial processes to create the third-stagegeneralization in the emergence of chemical reaction engineering. Transport phe-nomena are coupled to chemical reaction to describe reacting systems having massand transfer resistances. In all unit processes, momentum transfer connecting to thefluid mechanics is the next toward real design. Chemical engineering science uti-lizes transport phenomena along with thermodynamics and chemical kinetics toanalyze and improve the unit processes and hence the whole process (Kwauk 2005).Rightfully the four subjects (unit operation, transport phenomena, chemical reactionengineering and thermodynamics) became the established knowledge base forchemical engineering. These four sharply delineate the domain of chemical engi-neering approaches.


1.2.2 Chemical Engineering in the Twenty-First Centuryand the Integrated System Approach (ISA) Basedon System Theory (ST)

The chemical industry has revolutionized human life to such an extent invaded allfields of today modern life. At the beginning of the twenty-first century, chemicalengineers have involved in a diverse range of industries requiring a wide domain ofknowledge in different areas (Elnashaie and Garhyan 2003). These developmentswere accompanied by a steady growth in productivity, sophistication and a highlevel of competition in the chemical industry. At the same time, the need for theexpertise of the chemical engineers greatly expanded with advancing productionskills from commodity chemicals through pharmaceuticals, food processing,environmental protection, to extraterrestrial resource conservation and utilization,especially the provision of functional materials, for instance, for the informationtechnology (IT) industry including device making, e.g., via chemical vapor depo-sition (CVD)—thus providing additional impetus for extending the chemicalengineering knowledge base (Kwauk 2005; Chow 2002).

Today, chemical engineers are expected in their career to deal with diversifiedproblems that need a fundamental basis as well as practical knowledge. Of course,practical knowledge is acquired after graduation working in industries; however,undergraduate education is also an important factor that determines the degree ofsuccess of chemical engineers after graduation. Current challenges in the fields ofhealth, safety and environmental protection, together with the fast expansion in theuse of sophisticated instruments, make the task even more difficult (Elnashaie andGarhyan 2003; Harmsen 2004; Martin et al. 2005). The knowledge base establishedfor educating the chemical engineers in the previous era does not obviously sufficewhen they deal with novel processes. The variety of high added-value and small-lotfunctional materials is another motivating force for ever developing chemicalengineering knowledge base (Kwauk 2005). A chemical engineer graduate must beable to hold the growing opportunities of our modern society as well as to cope withits difficulties and problems. To fulfill the anticipated needs in the ever-advancingproduction skills in the future process industry/engineering, a new educationalstructure for chemical engineering has emerged. The curriculum contains basicscience courses including chemistry, physics, mathematics and biology andchemical engineering disciplines including transport phenomena, chemical reactionengineering, unit operations, process dynamics and control, plant design and eco-nomics. In addition to these two main subjects, supplementary subjects from otherengineering disciplines such as mechanical, electronic, material and even civilengineering were taken into account. Besides all those, training of chemical engi-neers should emphasize a “global outlook to the relation between Engineering andSociety” (Elnashaie and Garhyan 2003). Considering the socioeconomic implica-tions of chemical, biochemical and related industries, it is not wise to producechemical engineers who are not sufficiently aware of the problems of their societyand the connections between these problems and their profession.


However, since the scope of chemical engineering is set to become wider, itcreates clear difficulties for constructors of curricula. Furthermore, the industrialdevelopment and the innovative work of many pioneering chemical engineeringresearchers, coupled with the advancement in computer technology, make thetraining of chemical engineers for the future quite a challenging task. The situationbecomes even more difficult with the expansion of chemical engineering into newfields, especially biotechnology, the electronic industries, new materials, nano-technology and composite membranes. High levels of qualitative logic derived fromexperience or experiment are required before the act of precise problem formula-tion. This qualitative approach is just a primary step, and a quantitative descriptionof the process should be drawn using all applicable physical and chemical infor-mation, conservation laws and rate expressions. At this point, the real purposes ofthe modeling effort must be classified. The automatic computerized analysis andsynthesis of balancing relations among hundreds or thousands of process streamsrequire a systematic approach respecting all possible situations which can occur inpractice. This task can be achieved by looking at a chemical process as an inte-grated system from a system theory point of view. Chemical processes are aggre-gates of simple chemical and physical processes having mathematical descriptionsbased on fundamental chemical and physical laws. System theory is the basis of theintegrated system approach, which is the most efficient methodology for knowledgeclassification, organization, transfer and exchange (Elnashaie and Grace 2007).

Before moving to detail explanation of system theory and integrated systemapproach, let us describe the word system. The word system derives from the Greekword “systema” and means an assemblage of objects united by some form ofregular interaction or interdependence (Elnashaie and Grace 2007). A simpler, morepragmatic description regarding systems includes the following:

• The system is a whole composed of parts (elements).• The concept of a system, sub-system and element is relative and depends on the

degree of analysis; for example, we can take the entire human body as a systemand the heart, the arms, the liver and so forth as the elements. Alternatively, wecan consider these elements as sub-systems and analyze them with respect tosmaller elements (or sub-systems) and so on.

• The parts of the system can be parts in the physical sense of the word or they canbe processes. In the physical sense, the parts of the body or of a chair form asystem. On the other hand, for chemical equipment performing a certain func-tion, we consider the various processes taking place inside the system as theelements which are almost always interacting with each other to give thefunction of the system. A simple chemical engineering example is a chemicalreactor in which processes like mixing, chemical reaction, heat evolution andheat transfer take place to give the function of this reactor, which is the changingof some reactants to some products.

• The properties of the system are not the sum of the properties of its components(elements), although it is, of course, affected by the properties of its components.The properties of the system are the result of the nonlinear interaction among its


components (elements). For example, humans have consciousness which is nota property of any of its components (elements) alone. Also, mass transfer withchemical reaction has certain properties which are not properties of the chemicalreaction or the mass transfer alone (e.g., multiplicity of steady states, as will beshown later in this book).

The term state of the system, rigorously defined through the state variables of thesystem, is used extensively in discussing and modeling/simulation of systems.These state variables are chosen according to the nature of the system. Inputvariables are not state variables. Instead, they are external to the system but affectthe system (i.e., work on the system). For example, the feed temperature andcomposition of the feed stream to a distillation tower or a chemical reactor or thefeed temperature to a heat exchanger are input variables. Design variables areassociated with the design of the system and are usually fixed. Examples are thediameter and height of a continuous stirred tank reactor (CSTR) or of a tubularreactor. A system has boundaries distinguishing it from the surroundings or envi-ronment. The relation between the system and its environment leads to one of themost important classifications of systems:

• Isolated SystemsDo not exchange matter or energy with the surrounding. Example is adiabaticbatch reactors.

• Closed SystemsDo not exchange matter with the environment or surrounding, but do exchangeenergy. Example is a non-adiabatic batch reactor.

• Open SystemsExchange matter and energy with the environment or surrounding (actuallystating that it exchanges matter is sufficient because this implies an exchange ofthe energy in the matter transferred). Example is a CSTR.

Integrated system approach is one of the most important tools for the develop-ment of new knowledge and novel processes, especially in areas wheremulti-disciplinary research and development is a must for innovative solutions(Charpentier 2007). Nanoengineering is one of those areas that aremulti-disciplinary by their very nature. It is formed of a number of sub-systems,each of which is formed by its own elements (or sub-systems of the sub-systemsdepending upon the level of analysis). Sub-systems of nanoengineering includeboth technical and non-technical categories, for example, technology, socioeco-nomic and safety. Focusing on any one of the nanoengineering sub-systems canonly be successful within a framework that has the other sub-systems as a back-ground. Within the technological sub-system of nanoengineering, a structuralhierarchy of sub-systems, followed by sub-systems of sub-systems, down to ele-ments gives the structure and boundaries of this important sub-system, especiallyfrom an engineering point of view. It is useful in this regard to use terminologiesand classifications of system theory coupled with terminology of nonlineardynamics and stability theorem.


Chemical engineering in current century requires many speculations, includingsustainable, environmental, communities and economic enterprises. In the face ofever increasing process complexity and instability, it is essential to move beyond asimplistic steady-state model of processes. Adaptive procedures and models thatenable industries to cope with unexpected challenges must be developed. Thecurrent lack of success in commercial production of novel products, for examplenanostructures, coupled with the challenges of complexity and resilience of thepertaining process indicates that a systematic approach is vital. In other words,collaborative solutions with a multi-disciplinary nature are required. A new productneeds desired properties both in their structure and function. Product quality isinvestigated at the micro- and the nanolevels that require understanding of therelationship between structure and property at both molecular and microscopiclevels. The ability to control microstructure formation that determines the end-useproperties of a product is the key to success of the process (Charpentier 2002). Suchsuccess will therefore help design and control of the product quality. Moving fromlaboratory scale to process level requires an integrated system approach for mod-eling of complex simultaneous and often coupled momentum, heat and masstransfer phenomena taking place in the processes.

The integrated system approach based on system theory depends on definingevery system through its boundary, main processes within this boundary andexchange with the environment through this boundary. It relies upon thermody-namics and information theory and is, therefore, applicable to all kinds of systems,which makes it most suitable for cross-disciplinary investigations and innovation.The basis and principles of system theory rely on mass and energy balance derivedfrom mass and energy conservations laws, respectively. Since chemical industriesare complex involving different unit operations and unit processes. Several tech-nical advances will likely improve the usefulness of models, including rigorousmethodologies for dealing with missing and uncertain information; improvedmethods for interpretation of multivariate data sets and for multi-objective decisionmaking involving trade-offs among conflicting goals; and novel modeling methodsas alternatives to traditional mathematical models. More generally, there is a greatneed for operational definitions and metrics for sustainability and resilience ineconomic, ecological and societal systems (Elnashaie et al. 2013).

1.2.3 Modern Trends in Chemical Engineering

The chemicals and related industries are in a phase of rapid evolution due to bothunprecedented demands and limitations, stemming from public concern overenvironmental and safety issues as well as special consumers trends. The corre-sponding knowledge, including chemical knowledge, is also growing rapidly, andthe rate of discovery increases every day. Not only that but this development ischaracterized by the coupled use of experimental studies, mathematical modelingtechniques and advanced computers and softwares. It is also highly


multi-disciplinary which widen the scope of discovery and decreases the time spacebetween discovery and commercialization. Today, about hundreds thousandsproducts are deliberately synthesized and manufactured to meet the human needs.A few number of the products exist in the market can be found in nature.Nowadays, chemical engineers are expected to answer to the evolution of marketdemands and offer novel products that can compete in the new global economy,where environmental protection and sustainability are the keywords (Lin 2003). Inaddition to that, chemical engineers are qualified and obliged to play a leading rolein the environmental and sustainable development (SD) challenges that are facingthe human society in a dangerous and complicated manner. Accordingly, chemicalengineers are in charge to research innovative processes that are non-polluting,defect free and perfectly safe for production of commodity and intermediateproducts. Indeed, nowadays, the economic constraint is no longer the only issue forprocess design, and other aspects also must be carefully taken into account such assafety, health and environmental aspects, including the value of non-pollutingtechnologies, reduction of raw materials and energy losses as well as products andby-products recyclability (Charpentier 2002).

Generally speaking, chemical engineering is today concerned with the under-standing and development of systematic procedures for the design and optimaloperation of industrial processes such as chemical, petrochemical, pharmaceutical,food, cosmetics, etc. These products are mostly manufactured through a veryintricate chemical processes and methodology (Lin 2003). Chemical engineers arewell positioned to venture into problems of these systems (Riegel and Kent 2003).Not only the scale-up of complex systems challenges chemical engineers, but alsonovel products involving molecular self-assembly, nanostructure and supramolec-ular chemistry are posing nowadays-fresh problems (Gallagher and Appenzeller1999; Grossmann 2004; Charpentier 2007; García-Serna et al. 2007).

Design and evaluation of complex systems, new integrated operations andproducts require further research into novel strategies, methodologies and tools.These should be oriented toward the acquisition of basic knowledge in chemicalengineering science, including thermodynamics, kinetics, rheology, transport phe-nomena and reaction engineering accompanied with improved knowledge in pro-cess modeling, process dynamics, automation and control. Also rigorous andempirical mathematical models and scientific instrumentations are required todesign and install accurate control instrument. In addition, chemical engineeringinvolves a strongest multi-disciplinary collaboration among physicists, chemists,biologists, mathematicians and instrumentation specialists leading to the develop-ment and design of the products and processes with complex and sophisticatedstructures (Charpentier 2002, 2007; Eyinagho 2007; Educational 2007; Smith andIerapepritou 2010). Current modes of production are also more challenged bydecentralization, modularization and miniaturization (Charpentier 2007).Organizing scales and complexity levels in chemical engineering are necessary tounderstand the events at micro- and nanoscales which help to better convert rawmolecules into useful required products. Besides, systemic analytical models basedon the multi-scale integrated approach should be taken into account. Recall that,


this approach considers the global behavior of complex systems as a whole, insteadof looking more into the small-scale mathematical details. Novel principles of theanalytical models in chemical engineering at the highest level of integration arerequired for a good understanding of the behavior of the interactions in the optimalprocess control and operation. For better conversion of raw materials into usefulproducts on the process scale, it is necessary to understand the relationshipsbetween events at nano- and microscales. Figure 1.1 illustrates the scales andcomplexity levels in process engineering.

Recall that, a key to successful commercial production of a novel material withcontrolled quality requires an integrated system approach for a multiscale andmulti-disciplinary modeling of complex processes taking place on different scales oftime and length. Time scales include from femto- and picoseconds for the motion ofatoms in a molecule during a chemical reaction and nanoseconds for molecularvibrations up to the scale of hours for operating industrial processes and of centuriesfor the destruction of pollutants in the environment. Length scales vary fromnano-scale (10−9 m) to mega scale (106 m). Molecular processes and catalytic activesites are examples for nano-scale, bubbles, droplets, particles and eddies in the systemare on the micro-scale, unit processes including different types of unit operations,reactors and heat exchangers are on the meso-scale, production units includingchemical, petrochemical and biochemical plants are on the macro-scale and envi-ronment including atmosphere, oceans and soils are on mega-scale (Charpentier2002). For example, synthesis of various nanostructures has been extensively studiedwith the purpose of commercial production of the nanostructures (Raoet et al. 2006;Zhang and Wang 2006; Paradise and Goswami 2007; Lines 2008; Li and Lin 2010;Dastjerdi andMontazer 2010; Mohapatra and Anand 2010; Hayat et al. 2011). In this

Fig. 1.1 Scales and complexity levels in process engineering (Charpentier 2002)


regards, the evolution of the morphology of nanostructures formed was both exper-imentally and theoretically investigated because understanding of nanostructureformation is prerequisites for a proper reactor design, optimization and control.

To put the problem into perspective, Fig. 1.2 illustrates different length scalesand transfer phenomena involved in synthesis of carbon nanotubes (CNTs) bychemical vapor deposition (this process will be comprehensively discussed inChap. 2) in a fluidized-bed reactor (FBR). In chemical vapor deposition, CNTs aregrown on active sites of catalysts when temperature is generally above 600 °C . Theactive sites are nanoparticles of transition metal (Fig. 1.3) usually anchored on asuitable support. The process condition in the reactor, which is meter scale, directlyaffects the nucleation, which occurs at nanoscale. Considering these length scales,as presented in the schema of Figs. 1.2 and 1.3, the interconnection of the eventstaking place on various scales is evident. The kinetics of CVD reaction is controlledby active sites, which are generally nanoparticles of transition metals, of the cat-alyst. On the other hand, internal mass and energy transport to form the nano-structure takes place at microscale, taking into account that particle–particle,particle–wall interactions affect the fluidization and thus the parameters that havesignificant effects on mass and energy transfer, kinetics of reaction and hence theformation of CNTs. It is evident that all the scales mentioned have important

Fig. 1.2 General schematicof fluidized bed for CNTsynthesis at large scale. 1Reactor, 2 furnace, 3distributor, 4 thermocouple, 5flow meter, 6 carbon source, 7carrier gas, 8 outlet, 9pressure sensor analyticalinstruments like gaschromatography also can beused for outflow


http://dx.doi.org/10.1007/978-981-287-496-2_2

impacts on global behavior of the reactor, macroscale events and hence the use ofenergy, etc., in mega-scale. Accordingly, a fundamental understanding of events atall levels of complexity is absolutely critical. This perspective necessitates theknowledge of engineering rather than technology, and chemical engineering on topof all other engineering disciplines can efficiently be involved in the field ofnanoengineering.

Transitioning from nanotechnology to nanoengineering means move to thedesign and analysis of the unit operations and unit processes which are combinedinto an integrated process. Chemical engineers must design and develop processescapable of commercial manufacturing of nanostructures and nanoproducts (Ottino2005). It is worth mentioning that, nowadays, sustainable methods and tools tomanufacture products of market-defined properties are expected. This approach,SD, is regarded as the forth paradigm of chemical engineering. These challengingproblems cannot be resolved by a mere application of the existing knowledge, andmulti-disciplinary research both in macroscopic scale and nanoscale is required(Schummer 2004; Nicole et al. 2010). A detailed and local analysis in microscopicscale yields a deeper understanding of the underlying phenomena. On the otherhand, integration is achieved by a systems approach which accounts for couplingeffects and complex behavior, and yields phenomenological laws for the behaviorof organized systems. Both approaches should be developed simultaneously basedon their paradigms and principles, which are the main goal for basic research.

Modern chemical engineering structure can be summarized by four mainobjectives:

1. Increase productivity and selectivity of operations toward manufacturing ofproducts of market-defined properties.

2. Develop intelligent algorithms and instrumentations for process control. Thisapproach should provide nano- and micro-tailoring of materials with controlledstructures.

3. Implement multiscale application of computational modeling and simulation toreal-life situations from the molecular scale to the production scale, e.g., in orderto understand how phenomena at a smaller length scale relate to properties andbehavior at a longer length scale.

4. Design novel equipment and new production methods based on scientificprinciples and SD.

Fig. 1.3 Schematic diagrams of CNT growth mechanisms occurred at nanoscale


1.2.4 Multi-disciplinary Approach for Development of NovelProcesses

Modern chemical engineering deals with developing systematic procedures foroptimal design and operation of various chemical processes to satisfy the marketsrequirements, while accurately considering the social and environmental constraintsof the industrial-scale processes. The principles and methodology of chemicalengineering have been instrumental in the development and application of manynew technologies like bioengineering, genetic engineering, microelectronic pro-cessing, micro-fabrication and nanotechnology (Lin 2003).

For example, chemical engineers are responsible for producingnanotechnology-based products to market, and hence, they have to deliver thetechniques to make that happen. Successful performance of any chemical processand the necessary progress are obtained via a multi-disciplinary approach. To makeit clearer, let us consider a case that a novel product is planned to be produced.Figure 1.4 represents production steps of a product from the idea to production.

Figure 1.4 indicates that all branches of engineering (such as chemical,mechanical, electrical, metallurgical and civil) have to cooperate to give a specificproduct from specific raw material. Although a branch of engineering may havemore important rule in a specific branch of the production process, but it isimpossible to find a discrete line between the engineering branches in this coop-eration. It should be mentioned that chemical engineering has a central rule in allabove activities. Therefore, today we observe new interdisciplinary profession inthe research activities. When we insert nanooverview in these activities, nano-technology will be born. CNT is a nanostructure material which has different andunique applications because of its specific properties. Its specified and uniquestructure introduces CNT as symbol of nanotechnology. Today, there are a lot ofbasic and applied researches to improve properties of materials using CNT forchemical, metallurgical, mechanical, electronic, environmental, medical and so onapplications, and thus, day-to-day increasing in its market is observing.

Fig. 1.4 Different aspects in the product


Accordingly, mass production of CNT with predetermined and controlled proper-ties is target of engineers’ investigation. So the rule of scientists and engineers in abranch of nanotechnology is vital. Scientists work on molecular system at thelaboratory scale, while engineers deal with process system. The commodity is firstsynthesized and characterized at the molecular level in the research laboratorymostly using chemistry, physics and biology knowledge and principles. The tran-sition from applied science to engineering involves the design and analysis ofproduction units, which are integrated into a process, and then becomes part of amultiprocess industrial site. At the final state, this site is part of the commercialenterprise driven by market considerations and demands the inclusion of theproduct quality. In this simple example, it is evident that a multi-disciplinaryapproach is strongly needed for successful production of a new product to themarket.

Chemical engineers involved in such a process have a key role and need to mixtheir capabilities with a similarly talented group of other scientists and engineers.To make it clear, now let us discuss about interaction of different aspects of sciencesand technologies in mass production of CNT. Although various researches arecarried out to find out kinetic of catalytic reactions and rule of transport phenomenaon the final properties of the product, all of these investigations are needed asprerequisite of CNT mass production. The main question is determination of theproduction capacity in a large-scale plant. The engineering economy and profit cananswer this question but, several technical and engineering aspects are involved incalculating the economic features of the plants. To clarify this fact, we shouldanswer following questions to determine production scale of the product:

• Which properties of the product are expecting to be improved:• Which applications of the product is considered:• Which operating conditions should be set in the production process:• How can control these operating conditions in the production process:• What is the product purity and the purification process:• What are the list and size of the production and purification processes

equipment:• What is the market of the product:• Is it possible to model and simulate the production and purification processes to

predict the product specifications.

Chemical engineering and its profession subsets have main rule to answer thesequestions and discuss about that. Some general profession subsets of chemicalengineering are transport phenomena, thermodynamic and thermo-kinetic, processdesign and control, modeling and simulation, process optimization, engineeringeconomic and of course nanotechnology. It should be mentioned that todaychemical engineering (such as other engineering) has no discrete boundary, and it isa multidiscipline field which interacts with mechanical engineering, metallurgicalengineering, chemistry, physics, biology, etc. So we can summarize this discussionas following flowchart (Fig. 1.5).


In addition, in order to cope with future demands, chemical engineering researchmust be multi-disciplinary as stressed earlier. Variety of high added-value andsmall-lot functional materials, and the anticipated needs in the ever-advancingproduction skills in the future need micro- to macroscale studies, and hence,chemical engineering as an active role in this area is pointing to multi-disciplinaryskills. Chemical engineers need additional inputs from disciplines such as chem-istry, colloid science, measurement science, computational science, materials sci-ence, systems engineering, environmental science and consumer science (Edwards2006). In this respect, chemical engineers need to have a greater multi-disciplinaryaptitude than most other academic subjects and a highly skilled cadre. The tradi-tional skill sets of reaction engineering, systems engineering, thermodynamics,transport processes and separations may simply be constructed to provide peda-gogical clarity, but they nonetheless remain intensely relevant to today’s problems.Multi-disciplinary research requires teams of people each with their specializedskills and capable of working through whatever interfaces are unavoidable.

Fig. 1.5 Interrelation of chemical engineering with other science and engineering professions inthe field of nanotechnology


1.2.5 Innovation and Sequential De-Bottle-Necking

Process innovation and improvement are critical elements in maintaining compet-itiveness of the chemical industries. There are different forms of process innovationand improvement such as yield enhancement, quality improvement, material andenergy conservation, waste minimization and safety enhancement (El-Halwagi2006). The limited productivity of any process is caused by certain bottleneck(s),parameters or units that restrict the production. A common method used for processimprovement is process debottlenecking, aimed at increasing production byremoval of bottlenecks. These bottlenecks are generally related to either equipmentor resources, demand for various utilities, labor and raw materials and/or arise fromthe reaction rate, mass transfer rate, heat transfer rate, hydrodynamic limitations,etc., or a nonlinear coupling between two or more of these bottlenecks (Chen andElnashaie 2004). Debottlenecking may apply to a specific unit or overall plant,whether it is due to increased throughput or process modifications (Fair and Seibert1996; Summer et al. 1995; Litzen and Bravo 1999; Saremi et al. 2000; Modashiaet al. 2000; Chen and Elnashaie 2004; Tan et al. 2006, Alshekhli et al. 2011). Theconventional approach for debottlenecking has been sequential in nature wherebottlenecks are identified and removed one at a time. The active bottleneck is firstidentified, then relaxed or removed through capacity expansion of the unit, changesin design and operating variables, unit replacement or addition, and stream rero-uting. As a result of debottlenecking of the unit, a new bottleneck appears in thesame unit or somewhere else in the process. The same approach is repeated forremoving the new bottleneck. This sequential debottlenecking approach involvestwo activities: identification of active bottlenecks and removal of bottlenecks.Bottleneck identification may be achieved through various means. Actual processperformance may be analyzed to detect the active bottleneck. Process experiencemay be used to point to likely bottlenecks. Once a bottleneck is identified, acombination of process analysis and process synthesis techniques can be used toscreen debottlenecking alternatives and select a solution.

The hierarchical techniques that rely on intuition, engineering knowledge andphysical principles have been used for debottlenecking (Fisher et al. 1985;Rapoport et al. 1994). Fisher et al. (1985) proposed a method for screening alter-natives and modifying equipment sizes, replacing units and adding new equipment.Rapoport et al. (1994) developed a procedure for equipment design, for capital costsand for economic evaluation. While heuristic approaches utilize engineeringinsights, they are not guaranteed to identify optimum solutions for general cases.Optimization techniques may be used to solve a production maximization problemand identify the bottleneck as the unit for which at least one of its constraintsbecomes active (Ben-Guang et al. 2000). Zhang et al. (2001) proposed a two-stagedebottlenecking approach for refinery operations. In the first stage, a linear pro-gramming model is used to identify major bottlenecks by locating the equipmentthat required extra capacity. In the second stage, debottlenecking stage, high-levelbottlenecks are removed first, and then, a detailed process model is used to remove


the low-level bottlenecks. Hence, this debottlenecking procedure is sequential andmay lead to sub-optimal solutions.

Another example of using sequential debottlenecking is development of steamreforming of hydrocarbons, the most prominent process for hydrogen production(Rostrup-Nielsen 1977; Elnashaie and Elshishini 1993; Christensen 1996; Chenet al. 2003). The first-generation reformer (FGR) is the fixed-bed steam reformer,which is inefficient, is highly polluting and suffers from catalyst deactivation,especially when using higher hydrocarbons as a feedstock (Elnashaie and Elshishini1993; Rostrup-Nielsen 1977). Significant progresses toward overcoming the limi-tations of conventional fixed-bed reforming systems lead to introducing thesecond-generation reformer (SGR), which are bubbling fluidized-bed reformer(Elnashaie and Adris 1989; Adris and Elnashaie 1991; Sammels et al. 2000). Thethird-generation reformer (TGR), a novel circulating fluidized-bed membranereformer (CFMR), has been then introduced. This technology is more efficient andflexible for the production of pure hydrogen (Chen et al. 2003; Prasad andElnashaie 2002). The TGR is a novel configuration that will be more efficient thanboth the FGR and the SGR. Development from the FGR to the SGR and relativelyrecently to the present TGR is based on sequential debottlenecking, which relies onthe simple fact that the limited productivity of any process is caused by certainbottleneck(s). FGRs suffer from a large number of limitations (bottlenecks), themain one of which is the intra-particle diffusion resistance of the catalyst pellets,giving rise to very low effectiveness factors (Elnashaie and Elshishini 1993).The SGR was developed to break all of these limitations (debottlenecking); how-ever, new bottlenecks appeared with this configuration because there are inherenthydrodynamic limitation associated with the SGR, in addition to thermal and otherlimitations (Chen et al. 2003; Prasad and Elnashaie 2002).

The CFMR process for hydrogen production, however, suffers from an importantbottleneck, thermodynamic limitations (Elnashaie and Garhyan 2003). The netreaction in steam reforming process which is highly endothermic can be expressed as

CH4 þ 2H2O � CO2 þ 4H2

The reaction is very fast but limited by thermodynamic equilibrium, and hence, itrequires very high temperatures to increase the equilibrium conversion. Normally theprocess is carried out at 800–900 °C and a pressure of 30–40 bar, resulting in aconversion of 90 %. Cracking of the hydrocarbons at these high temperatures causescarbon formation, which necessitates the use of high steam-to-hydrocarbon ratio toavoid catalyst deactivation (Elnashaie and Elshishini 1993). In order to obtain thesame conversion (90 %) at a lower temperature, hydrogen must be removed selec-tively from the reaction zone during the process. This can be done in a hydrogenselective membrane reactor that gives the same conversion at lower temperature(Prasad and Elnashaie 2002). The product hydrogen permeates selectively throughhydrogen membranes from the reaction side and is carried away by sweep gas suchas low-value steam in the hydrogen membrane tubes. The use of a membrane reactorin steam reforming has several advantages. Because of the lower-temperature


operation, the energy consumption of the process is reduced which results in loweremission of CO2. The lower temperature also requires less expensive catalyst, tubingand other reactor materials. Since hydrogen of sufficient purity is produced directlyfrom the reformer, the downstream shift conversion can be omitted. Moreover, thedimensions of the CO2 removal and final purification units can be reduced. Hence,significant savings in equipment costs can be expected.

It is important to examine the ability of a sequential approach in attaining thetrue potential of the process and in achieving maximum debottlenecking. In par-ticular, the following questions are important:

• Does the sequence of debottlenecking the units (e.g., the active bottlenecks)affect the ultimate extent of debottlenecking the whole process? If so, what is theoptimal sequence of debottlenecking?

• In tackling an active bottleneck, should it be debottlenecked to the maximumextent? If not, then to what extent?

• If no new units are added to the process, is it possible to identify a target formaximum extent of debottlenecking ahead of detailed debottlenecking andwithout commitment to the debottlenecking strategies?

1.2.6 Intensification Through Integration and Combinationof Different Processes in Single Units

Chemical engineers have usually focused on scaling-up the chemical processes fromlaboratory to industry scales by passing pilot plant scale in the most cases. However,in the present era, scaling-down of industrial processes has become more and moreimportant to meet the required criteria of social and economic concerns (Stankiewiczand Moulijn 2000). Scaling-down of industrial processes through process intensifi-cation contributes to SD of the chemical process industry and is now seen as animportant field by the chemical industry (Harmsen et al. 2004; Harmsen 2004;Charpentier 2007; Reay 2008; Becht et al. 2009; Keshav et al. 2009; Narodoslawsky2013). Process intensification is achieved through integrated design of chemicalprocesses, quantitative design and optimization of chemical systems based ondetailed mechanisms (El-Halwagi 2006). The objective of process intensification is todesign more efficient plants with reduction of equipment size, energy consumption,and/or waste production, while both investment and plant operating costs arereduced. Higher safety as well as improved remote control and automation withadequate level of energy and resource saving will be achieved that substantiallyresults in sustainable production (Dvoretsky et al. 2010). In general, process inten-sification offers new prospect for chemical engineering in the world economydevelopment and world wealth protection. The progress of basic research in chemicalengineering has led to a better understanding of elementary phenomena and hencemakes it possible to adopt new operating modes for the same equipment or novelconfiguration of equipment or both based on scientific principles. The field of processintensification involves design of novel production system via three main schemes:


• Process intensification through equipment development such as novel heattransfer and mass transfer devices, process control instruments or new systemsusing alternative sources of energy (light, ultrasound, etc.)

• Process intensification through development of unit process. This scheme isperformed via integrated systems or using multifunctional equipment

• Process intensification through miniaturization (micro-unit operation,micro-reactor)

It is fair to note that chemical engineers are involved in the two last schemes thatwill be explained in details in the following paragraphs.

Integrated systems

Integrated systems have been successfully designed through hybridization of unitoperations or multifunctional reactors.

Hybrid unit operations

Hybrid unit operations are processing methods that integrate two or more differentunit operations, usually separation techniques, in a single unit, making use of thesynergy between them. The most promising hybrid unit operations include:

• Membrane distillation• Membrane absorption/stripping• Adsorptive membranes (membrane chromatography)• Membrane extraction• Extractive distillation• Adsorptive distillation

Integration of membranes with another unit operation presents the mostimportant category of hybrid unit operations. In membrane distillation, two aqueoussolutions held at different temperatures are mechanically separated by a hydro-phobic membrane. Vapors are transported via the membrane from the hot solutionto the cold one. The most important (potential) applications of membrane distil-lation are in water desalination and water decontamination (Alklaibi and Lior 2005;Charcosset 2009). Other possible fields of application include recovery of alcoholsfrom fermentation broths (Lewandowicz et al. 2011), concentration of oil–wateremulsions and removal of water from azeotropic mixtures (Ravanchi and Kargari2009).

In membrane absorption/ stripping, the membrane is used as a permeable barrierbetween the gas and liquid phases. By using membrane modules, large masstransfer areas can be created. Besides, absorption membranes offer operationindependent of gas- and liquid-flow rates, with no entrainment, flooding, chan-neling or foaming (Mansourizadeh and Ismail 2009; Drioli et al. 2011).

Membrane chromatography is a combination of liquid chromatography andmembrane filtration. The most important potential applications of membranechromatography is in bio-processing for separations of biomolecules (Drioli et al.2011; Orr et al. 2013).


In adsorptive distillation, a selective adsorbent is added to a distillation mixturethat increases separation ability. This technique presents an attractive option in theseparation of azeotropes (Wang et al. 2010).

Multifunctional reactors

Multifunctional reactors can be described as reactors in which chemical conversionintegrate at least one more function that traditionally would have to be performed ina separate piece of equipment (Andrzej 2003). Reverse-flow reactors are awell-known example, in which the reaction and heat transfer are integrated in asingle unit (Stankiewicz and Moulijn 2000).

Reactive separation processes are one of the thriving applications of multi-functional reactors. In these processes, reaction and separation operation arecombined to be performed in one unit. The industrially important reactive sepa-ration processes include (Andrzej 2003):

• Reactive distillation• Membrane-based reactive separations• Reactive adsorption• Reactive absorption• Reactive extraction• Reactive crystallization

Separation–reaction processes offer several advantages over conventional pro-cesses in which thermodynamic limitations, like azeotrope, can be overcome andhence the reaction yield is increased (Charpentier 2007). The improved selectivityof product leads to decreasing in raw material consumption and, thus, operatingcosts. Besides that, reduction in the number of process units generally leads to lessinvestment costs and significant energy savings. Consequently, using integratedsystems causes significant reductions in both investment and plant operating costsby optimizing the process.

Reactive distillation process, sometimes addressed as or catalytic distillation, isan important example of this category applied in petrochemical industry (Harmsen2010). In such a system, reaction and distillation take place in one vessel usingstructured catalysts as the enabling element. Both investment and operating costsare far lower than with conventional reaction followed by distillation. This com-bination also results in several outstanding advantageous including

• Obtaining a constant pressure boiling system ensuring precise temperaturecontrol in the catalyst zone.

• Efficient energy utilization as the heat of reaction directly vaporizes the reactionproducts for.

• Break of the reaction equilibrium barrier by distilling the products from thereactants in the reactor.

• Elimination of additional fractionation and reaction stages.• Increasing conversion and improving product quality.


Reactive crystallization-based processes, in which crystallization is combinedwith extraction, are now successfully used to produce numerous chemical, phar-maceuticals, agricultural products, ceramic powders and pigments (Berry and Ng1997). These integrated processes enable bypassing the thermodynamics barriersimposed on the system by the chemical reactions and the solubility of the com-ponents in the mixture (Charpentier 2007). By combining crystallizers with otherunit operations, the stream compositions can be driven to regions within compo-sition space where selective crystallization can occur. Integrated membrane oper-ations, like catalytic membrane reactors, are another example (Andrzej 2003).Chromatographic reactor, which utilizes differences in absorptivity of the differentcomponents involved rather than differences in their volatility, can be addressed asan alternative reaction–separation unit to reactive distillation when the speciesinvolved exhibit small volatility differences or are either nonvolatile or sensitive totemperature, like fine chemical or pharmaceutical applications (Harmsen 2010).Fuel cells are another type of multifunctional reactors attracted a great attention. Infuel cells, chemical reaction is integrated with the generation of electric power(Stankiewicz and Moulijn 2000).

Process intensification through miniaturization (micro-unit operation,micro-reactor).

This scheme includes design and use of systems or units in scale of micro, forexample, micro-reactors, micro-mixers, micro-separators, micro-heat exchangers andmicro-analyzers. (Ehrfeld et al. 2000; Gavriilidis et al. 2002; Hessel et al. 2006).Using micro-units/micro-systems offers several promising advantageous (Ehrfeldet al. 2000; Hessel et al. 2006; Becht et al. 2009). It facilitates accurate control ofprocess conditions like mixing, quenching and temperature profile. A lot of industrialbatch processes can be run continuously by using this new technology which pro-vides more flexibility compared with traditional plants. The high heat and masstransfer rates in micro-reactor allow reactions to be performed under more aggressiveconditions with higher yields that can be achieved with conventional reactors.Micro-reactors with their inherent safety characteristics allow safe production ofchemicals, especially those chemicals with storage and shipping limitations, such ashighly reactive and toxic intermediates (cyanides, peroxides, azides) (Ehrfeld et al.2000). Micro-reactors are equipped with integrated sensor and control units; in casethe micro-reactor fails, it can be isolated and replaced, while other parallel unitscontinue production. Besides, if the micro-reactor fails, a small amount of chemicalsare released accidentally. Moreover, scale-up to production by replication ofmicro-reactors units used in the laboratory would eliminate costly redesign and pilotplant experiments, thereby shortening the development time from laboratory tocommercial-scale production (Exchangers 2000). This approach would be particu-larly advantageous for pharmaceutical and fine chemicals industries where produc-tion amounts are often less than a few metric tons per year (Charpentier 2007).

All the schemes of process intensification are currently possible thanks to con-siderable progress in the use of scientific instrumentation and powerful computa-tional tools and capabilities, needed for modeling and simulation at different scalesand for systematic data collection and experimental verifications (Charpentier 2010).


Large benefits obtained by using integrated process, however, to achieve optimalperformance, it is important to lead a scientific approach to understanding where theintegration of functionalities helps. Besides, several important barriers must beovercome before process intensification is widely adapted, such as the maturity andeconomic competitiveness of the new technologies compared to the conventionaltechnologies. For instance in integrated membrane processes, more collaboration isneeded between many disciplines especially chemical/biological engineers andmaterial scientists to solve problems such as providing inorganic membranes ofperfect integrity involving mechanical and thermal stability and which will allowlarge fluxes of desired species, and also chemical engineers must figure out the heattransfer problems which now threatens successful scale-up of integratedauto-thermal processes. The use of hybrid technologies encountered in a greatnumber of multifunctional reactors is also limited by the resulting problems con-cerning control and simulation, i.e., the interaction between simultaneous reactionand distillation introduces more complex behavior involving the existence of mul-tiple steady states and output multiplicities corresponding to different conversion andproduct selectivity, compared to conventional reactors and ordinary distillationcolumns. This leads to interesting challenging problems in dynamic modeling,design, operation and strong nonlinear control. In this respect, multi-disciplinary andmulti-scale approach of chemical engineering is applied from the scale of themicro-reaction technology up to the scale of multifunctional macro-reactors orequipment.

1.2.7 Chemical Engineering Expectations on 2020

Chemical engineering played an important part in human development in the lastfew decades, and it will continue doing that as it has done that very well, deliveringgreat benefits to different societies. As described in the previous sections, it is thejob of chemical engineers to develop and optimize chemical and physical processesin the transformation of raw materials to products. The efficiency and effectivenessof chemical processes is the heart of the chemical engineering profession. There is anew twist in process design though. Chemical engineering processes should beoptimized not only for economic but also for environmental performance(Westmoreland 2008). This places new constraints and boundary conditions onprocess modeling and design. Competitive classical technologies are required tobreak the undesired cycle of raw material and energy to wastes to save thesesources. Renewable energy resources like agricultural and/or domestic wastes aregood examples that wastes constitute the raw material of the following cycle.Fortunately, chemical engineering is the major discipline involving in this domainas the objective of this engineering is design, scale-up or scale-down operation,control and optimization of industrial processes. Since the state, micro-structure andchemical composition of material changes through physico (bio-) chemical opera-tions, chemical engineering needs to acquire the scientific and technical knowledge


necessary for physicochemical and biological transformations of raw materials andenergy into the targeted products required by the customer. On the other hand,discovery of complex phenomena like bifurcation, instability and chaos associatedwith chemical and biochemical processes makes their responsibilities even moredifficult and critical. Manifestations of these phenomena are evident in typicalchemical engineering systems, and hence, their recognition and implications on thedesign, optimization and control of catalytic and biocatalytic processes are vital(Elnashaie et al. 1993; Elnashaie and Grace 2007; Zhang et al. 2012). Thesephenomena are important because bifurcation, instability and chaos in these sys-tems are generally due to nonlinearity and, specifically, non-monotonicity, which iswidespread in catalytic reactors either as a result of exothermicity or as a result ofthe non-monotonic dependence of the rate of reaction on the concentration of thereactant species (Elnashaie and Garhyan 2003). Many researchers have turned theirattention to the investigation of these complex phenomena in industrial systems(Paladino and Ratto 2000; Merta 2006; Ottino 2006; Karri 2011; Zhang et al. 2012).The issue is engineering education within old paradigms does not offer a good sightof such new phenomena and challenges, such as globalization and sustainable to thegraduates that necessitate transforming engineering education into new paradigmsrequired to meet these new phenomena and challenges. Chemical engineers willcontinue to use their core skills of material and energy balances, process dynamicsand control, fluid dynamics, reaction engineering, transport phenomena, thermo-dynamics, systems analysis and unit operations, etc. to break down complexmanufacturing issues. What is critical to be added to their curriculum includes thosecourses that develop understanding of the process at a range of length scales (frommolecular to global) and repack that understanding into sophisticated processes ablereliably to create products tightly specified at the molecular level and/or by thefunctionality that they possess. These processes will need to be safe, economic,environmentally clean and sustainable. From this view, new chemical engineeringwill focus to cope with the new challenges.

We live in a time of great change, an increasingly global society, driven by theexponential growth of new knowledge and knitted together by rapidly evolvinginformation and communication technologies. There is increasing recognitionthroughout the world that leadership in technological innovation is a key to anation’s prosperity and security in a hypercompetitive, global, knowledge-driveneconomy. In this respect, there have long been calls for engineering to take a moreformal approach to lifelong learning. Other critical elements of the innovationprocess are the long-term research required to transform new knowledge generatedby fundamental scientific discovery into the innovative new products, processes andservices required by society. The requirements of twenty-first century engineeringare considerable: Engineers must be technically competent, globally sophisticated,culturally aware, innovative and entrepreneurial, and nimble, flexible and mobile(Duderstadt 2010). Clearly, new paradigms for engineering education are deman-ded to: (i) respond to the incredible pace of intellectual change (e.g., from reduc-tionism to complexity, from analysis to synthesis, from disciplinary tomulti-disciplinary); (ii) develop and implement new technologies (e.g., from the


microscopic level of info–bio–nano to the macroscopic level of global systems);(iii) accommodate a far more holistic approach to addressing social needs andpriorities, linking social, economic, environmental, legal and political consider-ations with technological design and innovation; and (iv) to reflect in its diversity,quality and rigor the characteristics necessary to serve a twenty-first century nationand world (Duderstadt 2010).

While the traditional chemical engineering concentrates its tools only on theconversion path from raw materials to products, present chemical engineering needsto start from a product of market-defined properties and explores methods and toolsto effectively resolve the problem (García-Serna et al. 2007; Gwehenberger andNarodoslawsky 2008; Smith and Ierapepritou 2010; Jaworski and Zakrzewska2011). Accordingly, required modifications in the product and design methods arestrongly required.

It can also been visaged that the design of chemical and biochemical equipmentin the coming years will be coupled with the nanoscale approach to modeling (bio-)chemical processes on ever-growing scale. Expected properties of the designedchemical/biochemical product established in macroscale can nowadays be preciselypredicted by means of the molecular modeling. The key factor to succeed in pro-ducing the final product with the desired features is the quality control at the level ofnano- and/or microstructure formation (Roco 2004). It is widely accepted thatmodeling carried out at the scale of a whole production plant, together with theprocess simulation of selected equipment units, can lead to cost optimization of theproduct manufacturing.

To achieve the goal, as presented in the previous paragraphs, chemical engineersneed to have in-depth knowledge not only of process efficiency but also of thechemical pathways to desired products using both common and uncommon rawmaterials. Therefore, a higher level of organization of thinking and economy ofknowledge is needed in undergraduate chemical engineering education. Nowadays,the purpose of teaching and basic research in chemical engineering involves thesynthesis of nano- and microstructures materials, design, scale-up/scale-downoperation, control and optimization of industrial processes throughphysical-bio-chemical separations as well as through chemical, catalytic, biochem-ical, electrochemical, photochemical and agrochemical reactions. But the emphasistoday on end-use properties requires also a wide variety of technologies including thenew role of microtechnology, i.e., the use of microstructured mixers and reactors forprocess intensification. Chemical engineers are expected to build a platform forfuture generations of practitioners and theoreticians to facilitate innovation anddevelopment of the conventional process (Westmoreland 2008). Engineeringresearch and education, including chemical engineering, plays a key role in devel-opment of industries related to novel technologies, like nanotechnology (Educational2007). Their role of engineering in nanotechnology is essential since the degree ofcomplexity of systems increases at the nanoscale, and various disciplines of scienceand engineering converge. The capability of nanotechnology today for systematiccontrol and commercial production of nanotechnology-based products necessitatesapplication of engineering principles, most importantly chemical engineering.


1.2.8 Indispensable Tools for the Success of Chemicaland Nanoengineering

The role of chemical and nanoengineers is to design and develop new products andprocesses while reducing costs, increasing production and improving the qualityand safety of the new products (Lin 2003; Edward 2006; Charpentier 2010). Theyare also asked to solve convoluted problems either in industry or in environmentusing innovative methods through SD of the current process and technologies(Byrne and Fitzpatrick 2009). Therefore, academic programs and curriculums ofchemical and nanoengineering should offer students a broad knowledge, includingfundamentals in science and engineering, applied to a variety of problems related tochemical and nanoindustries. As these problems become more intricate, moderntools and novel techniques are necessary and chemical and nanoengineers areexpected to be able to:

1. Apply and develop the fundamental knowledge and understanding of scienceand engineering disciplines as required.Chemical and nanoengineers will need to have a more cross-functional approachin different areas of science/engineering/nature to get a more inclusive focus andinsight. The scientific and engineering subjects needed to function effectivelyand efficiently in the general fields of chemical and nanoengineering include butnot limited to design and conduct engineering experiments, analyze and inter-pret data; design a chemical/nanoprocess to meet desired requirements withinrealistic constraints such as economic, environmental, social, political, ethical,health and safety, manufacturability and sustainability. Mathematical models areindispensable tools in the design and the operation of chemical and nanoplants(Elnashaie et al. 1993). Practical experience in these fields during the last decadehas proven modern equation-oriented simulation techniques superior to theclassical sequential modular approach. The equation-oriented concept stronglyrelies on efficient numerical methods for large, sparse algebraic and differential–algebraic systems of equations. Interfaces abound in chemical and nanoengi-neering operations and interfacial phenomena control not only severalchemical/nanoprocesses but also characteristics of products. Deeper insightsinto diffusion, adsorption and reaction phenomena through the use of sophisti-cated tools such as Monte Carlo (Rubinstein and Kroese 2008) or moleculardynamic simulations (Obot et al. 2013; Oliveira et al. 2013; Ding et al. 2013).will enhance the understanding of the processes at interfaces. Accordingly,interdisciplinary science and engineering where engineering meet science(mathematics, chemistry, physics and biology) help engineers to solve importantchallenges in the workplace through application of basic analytical skills andrelated theory describing the problem. Besides, social science and managerialskills, including conflict resolution, leadership, delegation and risk assessment,are crucial to make smart decisions.


2. Communicate and function effectively in a diverse workplace andmulti-disciplinary teams that need interpersonal and communicative skills tofunction and work confidently and effectively. Besides professional and per-sonal skills, communication skills, like business and/or technical writing andoral presentation are likely to become increasingly fundamental for professionalsuccess of an engineer (Dannels et al. 2003; Rugarcia et al. 2000). Engineeringis becoming global, and teams of engineers around the world will be working ona single project. Accordingly, multilingual skills as well as the skills for com-municating through electronic and data communication tools are mandatory.

3. Understand and incorporate professional and ethical responsibility into engi-neering solutions and business activities that work in a sustainable mode, i.e.,safe and environmentally conscious manner.

As a result of serious attention to human development in the twenty-first century,SD has become a great attention. This development paradigm has strong implica-tions for the professional practice of chemical engineers, particularly the way theydesign industrial processes (Narodoslawsky 2013). The industrial aspect of SDfocuses on using renewable resources and simultaneously reduces its environmentalimpact. Accordingly, the structure of industrial processes will change dramaticallythat pose chemical engineering education to change accordingly. The new skillstaught to students today to make them fit for their carrier in the twenty-first centuryinclude understanding how to integrate processes into the ecosphere, how to set upraw material logistics and to deal with stake holders outside industry(Gwehenberger and Narodoslawsky 2008). Since chemical and nanoengineers havea broad range of careers, a fresh chemical/nanoengineer may feel lack of knowledgeabout the specific field in which he/she starts working. Besides all tools mentioned,they should be aware of the need of lifelong learning (L3) and continuous research,as well as technical knowledge needed to work with engineers from otherdisciplines.

1.2.9 New Tools, Outlooks and Opportunities for ChemicalEngineering in Relation to the Other EngineeringDisciplines

Rapid advancements in technology combined with increasing global connectivity,necessitate having more professional skills and tools for success of chemicalengineers; chemical technology changed in the last four decades from its descriptivenature to quantitative more mathematical chemical engineering. The same applies tothe transformation of nanotechnology to nanoengineering. For both disciplines, anew paradigm of borderless chemical and nanoengineering is emerging andincreasing demands of chemical and nanoengineers with multi-disciplinary trainingare predicted. In the new era, chemical and nanoengineers need to work more


frequently on multi-disciplinary teams with both engineering and non-engineeringprofessionals. The future cutting-edge fields are generally addressed as: alternativeenergy, bioengineering, nanoengineering, water engineering as well as environ-mental engineering (EE) and SD engineering. In addition, engineers, especiallychemical and nanoengineers, in a near future, will meet the essential needs of anexpanding world population for food, clean water, energy and sanitation. In thisrespect, they will strongly require additional competencies and skills to improvetheir capabilities to share information and data among individuals on amulti-disciplinary basis.

The broadest factors that change the engineering profession are a mix of social,economic and political turbulence and globalization. Business and engineering aremerging, and thus, there engineers need to acquire a more thorough knowledge ofthe business world besides technical side of the societal development in a nationalor international infrastructure. Non-engineering skills required for the future suc-cess of chemical and nanoengineers contain those skills linked to global business,economics, management and communications as well as SD. In summary, sincemarket trends are moving at a rapid pace, the future engineers should focus ondeveloping what they see as critical global skills in order to keep pace. In accor-dance with the requirements mentioned, the following areas have been identified forspecial consideration:

1. SD EngineeringThe demands from the society on cleaner technologies are increasing more thancleanup technologies to treat up the waste after it had already been formed. Inaddition to creating important products, chemical and nanoengineers will also beinvolved in protecting the environment by exploring ways to reduce pollutantsemission and wastes, to recycle waste, to develop new sources of environ-mentally clean energy and to design inherently safe, efficient andenvironment-friendly processes using novel technologies. This is only necessarybut not sufficient for SD. Methods for SD need support process planning, thecreation of environmentally friendly industrial solutions and renewable rawmaterials (RRMs). On the most important issue related to scientific world’sattention is producing clean energy from sustainable sources (RRMs). Althoughfossil fuel may not get exhausted soon, it is time for scientific world includingchemical and nanoengineers to seek alternative sources of clean renewableenergies and products as well as novel process for future use. Solar photovoltaiccells, fuel cells and micro-turbine are novel energy devices designed, andhydrogen, biomass, wind and tidal wave are some of the energy resources. Thistopic is multi-disciplinary and needs fundamental knowledge in areas like, butnot limited to, interfacial and electrochemical engineering and electrokinetictransport phenomena, etc.

2. BioengineeringIn biotechnology and bioengineering, chemical engineers are working to pro-duce new products and medicines from new RRMs as well as develop new


medical diagnostics and treatments. In a more multi-disciplinary manner,chemical and nanoengineers employ nanotechnology and biotechnology torevolutionize biosensors, biomolecule and drug delivery systems. Chemical andnanoengineers are also involved in the design of new methods, techniques andoperations in biochemical processes. In the most modern bioprocesses, superpure1 biomaterials are required that generate new challenges in downstreamprocessing. For example, separation of specific protein from a mixture causesgeneration of new separation technologies and methods. Novel bioreactorscombining reactions and separation in one unit are a new challenge as shown inthe next Section.

3. Novel Reactors EngineeringAs a result of process intensification, as well explained in Sect. 1.2.6, newspatial configurations of reactors, which usually include reaction and separationsimultaneously, are emerging. These chemical reactors offer several benefits tothe chemical process industries. For example, due to their rapid start-up andshutdown, micro-reactors suit for portable applications, e.g., for detection ofhazardous chemicals in air and water (Kothare 2006). In such applications,process safety enhances because of requiring only small quantities of hazardousmaterials as samples for analysis. Design, development and control ofmicro-reactors have received significant attention in recent years (Roberge et al.2011). Analysis systems with microscale also reduce the time and cost associ-ated with conventional laboratory methods. Another important application formicro-reactors is in situ production of hydrogen for small-scale fuel cell powerapplications (Huang et al. 2013).There is relative lack of information and understanding of the flow phenomenain the design and scale-up of many industrially relevant novel reactors. It shouldbe kept in mind that intensification is only one aspect of plant design andoperation, as with most issues in engineering, there are trade-offs that must bemade among the various important considerations to achieve optimal design andoperation. Accordingly, design and concept of novel reactor with enhancedselectivity as well as productivity is one of the major research areas of chemicalengineers.

4. Advanced materials engineeringInnovations in material engineering have been leading to control composition ofproducts at molecular level. Improved knowledge about the relationshipbetween the structure of a substance and its function provides a scientific basisfor engineering to produce advanced structural materials that offer specialfunctionalities. Production of such materials requires intelligent chemical andphysical manipulation and control of surface and interfacial properties of thesubstances at molecular level. Chemical and nanoengineers are actively

1The impurity levels are in the parts per billion ranges.


involved in manufacturing and creating new materials needed for spaceexploration, alternative energy sources and faster, self-powered computer chips,etc.

5. NanoengineeringThe basis for synthesizing new materials is shifting to small systems, namelymicro-emulsions, reverse micelles, vesicles, nanoparticles, etc. (Müller-Goymann2004; De Stefano et al. 2009; Singh et al. 2011). The properties and then behaviorsof such systems, where the particles are too small, differ dramatically from bulk.Accordingly, transport phenomena, chemical reactions and thermodynamics posenew conceptual as well as practical challenges. Figure 1.6 illustrates nanotech-nology platforms and process technologies in nanoengineering laboratories. Themain platforms include synthesis of nanostructures and incorporate them asbuilding blocks into final products. Unit operations significantly contribute tosuccessful commercialization of many discoveries in the chemical industry.Understanding and controlling nanotechnology unit operations will be equallyimportant for the commercialization of nanotechnology. In manufacturing sys-tems, besides development of unit operations, scale-up and scale-down, processintegration and intensification also present new challenges (Stankiewicz andMoulijn 2000; Lin 2003).On the other hand, as nanotechnology progresses towardmanufacturing and commercial stages, the effects of nanomaterials on environ-mental or public health will be an important issue that needs more consideration.

6. Water engineeringWith the increase in population, severe shortage of water would be felt in mostparts of the world. To resolve the water problem, water harvesting, waterrecycling or reusing and momentum gaining are considered initiatives. These

Fig. 1.6 Nanotechnology platforms and process technologies in nanoengineering


techniques are not new to chemical engineers, and they certainly can help indesigning and managing water treatment and distribution in both industrial andurban areas.There are many conventional methods applied to remove the impurities orpollutants in water treatment including adsorption, aeration, biological oxidationand chemical oxidation. Among them, the promising process is adsorption,because the used adsorbent can be regenerated by suitable desorption process,and it is highly effective and economical. The most widely used adsorbents areactivated carbon and zeolite. However, these adsorbents suffer from slowkinetics and low adsorption capacities that encourage researchers to investigatenew adsorbents (Fan et al. 2010). CNTs, a new and exciting nanoproduct, havebeen proven to possess great potential as superior adsorbents for removing manykinds of organic and inorganic pollutants (Long and Yang 2001; Díaz et al.2007; Lu and Su 2007; Wang et al. 2007; Shih and Li 2008; Lu et al. 2005;Hyung and Kim 2008; Liao et al. 2008; Chen et al. 2007; Ye et al. 2007).Compared with other adsorbents such as activated carbon by the abovemen-tioned researchers, it is suggested that the CNTs are a promising adsorbent forthe removal of organic compounds.

7. New Modeling ToolsChemical engineers have always been involved in quantitative analysis andunderstanding of various phenomena utilizing mathematical modeling. Byadvent of novel technologies like bio- and nanotechnology, this task becomesmore important and complicated. Chemical engineers have mainly focused onunderstanding of gas and liquid systems compared to the area of solids handling.

Accordingly, the fundamentals for solids flow remain empirical in nature. Solidshandling is of considerable importance in process involved in synthesis of nano-structures, and hence, research in these areas, such as flow of cohesive powders,flow of gas particle in risers and fluidized beds is vital. Improved physicochemicalmodeling accompanied with microscopic modeling (such as molecular dynamicsimulation and monte-carlo simulation) can help to predict the macroscopicbehavior of nanomaterials; nevertheless, their behavior at microscopic level,especially for reacting systems, is still in its infancy. Besides improved modelingtechniques, new mathematical tools, which suit the complexities and uncertainties’in the chemical engineering systems such as black box modeling, artificial neuralnetworks, fuzzy logic, phase-space reconstruction and cellular automata, have beenbrought into the profession (Jahan et al. 2011). These new mathematical tools haveoffered great possibilities of system prediction under specific circumstances, whennot enough information is available. Novel analysis and simulation techniques, likemotion simulation and animation/virtual prototype and computational fluiddynamics, finite element analysis, life cycle analysis and even project management,are also likely to play a strong role in the future of chemical engineering. Taken intoaccount that the results obtained using novel techniques are in the laboratory scale,they need to be developed to become relevant to industrial operations.


1.3 Principles of Chemical Engineering

Chemical engineering is concerned with the study of systems in which materialundergoes changes in composition, energy and morphological structure. Analysis inchemical engineering science is based on the use of physical conservation princi-ples together with phenomenological equations of material behavior within theframework of continuum and statistical mechanics, chemical kinetics and thermo-dynamics (Albright 2008; Martin et al. 2005).

The methodology used by chemical engineering to design, optimize and controlof processes as well as to solve problems related to chemical industries is based onapplying sets of mathematical equations derived from general balances (mass,momentum, energy and compositions) and particular laws (thermodynamic, phys-ics, kinetics). In recent era, chemical engineers also use novel methods, likemolecular theory, artificial intelligence and practical experience admixed withprinciples of economics and SD to satisfy the society requirements.

Before the presentation of the adopted modeling approach, some importantconcepts are essential to discuss. First of all, what is a system? The word systemderives from the Greek word “systema” and means an assemblage of objects unitedby some form of regular interaction or interdependence (Elnashaie and Grace 2007).A simpler, more pragmatic description regarding systems includes the following:

• A system is a whole composed of parts (elements or sub-systems).• The concepts of a system, sub-system and element are relative and depend upon

the level of analysis.• The parts of the system can be parts in the physical sense or they can be

processes. A system can be formed of both (i.e., different parts of the system; areactor and a regenerator combined to form a fluid catalytic cracking unit), eachpart having a number of processes taking place within its boundaries.

• The properties of the system are not necessarily the sum of the properties of itscomponents (elements or sub-systems), although they are, of course, affected bythose components. Instead, the properties of the system result from nonlinearinteraction (synergy) between elements or sub-systems.

The term state of the system, rigorously defined through the state variables of thesystem, is used extensively in discussing and modeling/simulation of systems.These state variables are chosen according to the nature of the system.

A system has boundaries distinguishing it from the surroundings or environ-ment. The relation between the system and its environment leads to one of the mostimportant classifications of systems:

Isolated systems: They do not exchange matter or energy with the environment(surroundings). They tend to the state of thermodynamic equilibrium (maximumentropy). An example is a batch adiabatic reactor (Figure 1.7).

Closed systems: They do not exchange matter with the environment (surroundings),but they do exchange energy. Such systems, again, tend to thermodynamic


equilibrium (maximum entropy). A batch non-adiabatic reactor is an example(Figure 1.8).

Open systems: They exchange matter and energy with the surroundings. A CSTR isan example of these systems (Figure 1.9). These systems do not tend to thermo-dynamic equilibrium but to stationary non-equilibrium state usually called inchemical engineering: steady state.

Variation of the system property (as dependent variable) can vary with theindependent variables (which are time and position). This view leads to anotherimportant classification of systems:

Fig. 1.7 Schematic diagramof an adiabatic batch system

Fig. 1.8 Schematic diagramof a non-adiabatic batchsystem

Fig. 1.9 Schematic diagramof a continuous stirred tankreactor (CSTR)

1.3 Principles of Chemical Engineering 35

Lumped systems: The independent variable of these systems is just time. The batchand CSTR reactor are examples of these systems.

Distributed systems: The dependent variable of these systems varies with theposition (in any geometrical coordinates such as Cartesian, cylindrical or sphericalcoordinate) as independent variables. The plug-flow reactor (PFR) with one geo-metrical coordinate is an example of these systems (Fig. 1.10). At steady state, thesewill be the only independent variables and the system is described by ordinarydifferential equations, and at unsteady state, time will also be an independentvariable and the system is described by partial differential equations. A system withmore than one geometrical coordinate is described by partial differential equationsin both steady-state and unsteady-state conditions.

Role of system theory and mathematical modeling in chemical engineeringeducation is clear (Elnashaie and Garhyan 2003). The modeling methods are usedin the prediction of phenomena and processes employed in the chemical and alliedindustries (Jaworski and Zakrzewska 2011). Besides that, chemical engineers arenow faced with novel requirements and challenging problems which require ori-ginal and novel solutions. In these cases, a detailed and local analysis that yields adeeper understanding of the underlying phenomena will help a chemical engineer tomeet the challenges facing him/her. For example, process integration is achieved bya system approach which accounts for coupling effects and complex behavior in thesystem, and yields phenomenological laws for the behavior of organized systems.

Mass, momentum and energy conservation are the original principles formu-lating the core of chemical engineering. Each of these three disciplines follows acommon rule as general balance which is introduced as:

Input rate of the property-output rate of the property + generation rate of theproperty-consumption rate of the property = Accumulation of the property.

These events are shown schematically in Fig. 1.11.In this respect, the following subsection presents the following:

Fig. 1.10 Schematic diagram of a plug-flow system

Fig. 1.11 Schematic diagram of a system with multiple tasks


1.3.1 Generalized Mass, Momentum and Energy Balancesfor Multiple Inputs-Multiple Outputs (MIMO),Systems with Multiple Reactions (MRs)

Mass, momentum and energy balances based on general balance can be applied toany system. These balances could be made over the entire system, to give “overall”or “macroscopic” balances, or they could be applied to portions of the system ofdifferential size, giving “differential” or “microscopic” balances. All balances areprerequisite to almost all chemical engineering calculations to design new units orprocesses formed of a number of units, analyze existing ones or resolve processengineering problems. The tightly controlled inventory for unit/process usingmaterial and energy balances is important not only for design, operation and controlbut also for tight control over pollution. By definition, the polluting components canbe escaping reactants/products/side products or impurities with the feed. The tightcontrol over the material balances and the comparison with continuous measure-ments after the plant is built is one of the best and most reliable means to predictpollution emissions from the plant (especially fugitive emissions).

Mass balance

Mass balance, based on the law of conservation of mass, states mass will never becreated or destroyed, as long as no nuclear reaction occurs. The overall massbalance equation is expressed as “what remains within the boundaries of a system isthe difference between what was added to the system (input) and what was takenout from the system (output). In other, shorter terms, it is

Rate at which mass enters the system� Rate at which mass leaves the system¼ Rate at which mass accumulates within the system

This expression is an overall mass balance that applies to the total mass withinthe system, presented in terms of rate, where the units of each term are mass/time. Itshould be noticed that mass balance on a component usually describes as com-ponent mole balance. The component mole balance will describe later. The rate ofaccumulation of a conserved quantity within the boundaries of a system is thedifferentiation of total mass of a system with time. To formulate the aboveexpression, consider a system with the specifications shown in Fig. 1.12.

For this system, according to the recent expression, it can be written as shown inEq. 1.1

_min ¼ _mout þ dmdt

ð1:1Þ

where_min mass flow rate into the system_mout mass flow rate out of the systemm total mass content in the system


Equation 1.1 is written in terms of mass information including mass flow ratesand mass inventory. Often, volumetric flow rate is known, but the mass flow rate isrequired in Eq. 1.1. To make the formula appropriate, terms of volumes and den-sities that describe the total mass in each stream can be used for total mass balance.The following Eq. 1.2 is written using volumetric flow rates and volume rather thanmass flow rates and mass inventories, respectively. Figure 1.6 is a schematic dia-gram for Eq. 1.2

Qinqin ¼ Qoutqout þdq̂Vdt

ð1:2Þ

where Q is the volumetric flow rate and ρ is the mass density. Since density of massin the system may change, q̂, which is the average mass density in the system, isused to express the mass in the system, so that q̂V ¼ m.

Example 1

Consider an input stream containing a defined dose of nanoparticles which separatefrom the gas stream in a filter (membrane). A portion of nanoparticles will passfrom the filter and remains in the output gas stream. The portion of nanoparticleseparated from the stream is accumulated on the membrane and makes the filtercake. The mass of cake changes with the time, and it increases because of nano-particles separation. The mass balances of nanoparticles in this system will give:

Input rate of nanoparticles into the filter� output rate of nanoparticles from the filter¼ accumulation rate of nanoparticles in the filter cake

Figure 1.13 shows this process schematically. Nanomembrane is an example of afilter to separate nanoparticles from a fluid stream.

Assume that an input stream with mass flow rate _min (kg/sec) has the nano-particle dosage xin (mg/kg) and nanoparticle dosage of the output stream is xout(mg/kg). If the output stream has the same flow rate as input, the mass balance onthe nanoparticles based on Eq. 1.3 below is

_minxin � _minxout ¼ dm=dt ð1:3Þ

In this equation, the parameter m is total mass of filter cake which is the mass ofnanoparticles separated from the feed stream and collected on the filter (which isfilter cake). This term is varied with time. There are particular laws which predict

Fig. 1.12 Open system with volume V


the output dosage of particles based on the filter type and operating conditions(Cheryan 1998).

Mole balance on a component

When the system contains more than one chemical compound (species), it isrequired to keep track of one or more of those species individually. In such a case,species balance, which is an entire material balance on just one compound, iswritten. If necessary, species balances on the other compounds are also written. Incontrast to total mass, which mass is neither created nor destroyed, a particularchemical compound can be generated or consumed by chemical reactions that occurin the system. For such systems, in addition to input and output of materials,generation and consumption terms should be considered in the general component(s) material balance equations. A number of important individual processes inchemical engineering do not include generation or consumption of a species. Unitoperations are good example of such a case. For instance, operations used forseparation involve the extraction of specific compounds from mixtures of chemi-cals. In such a system, there is no chemical reaction. In this type of system, a massbalance for the individual species of interest is written while ignoring the formationand consumption terms. The procedure is illustrated in the example below.

Example 2

In a crystallizer, an input stream is a saturated salt solution (with defined concen-tration of the salt) which enters the crystallizer (as the system). In the crystallizer,the salt can separate as the nanoparticles in the solid phase from the solutionbecause of the supersaturation creation (for example by cooling). The remainingsalt dissolved in the solution leaves the system as the output stream. In this system,rate of nanoparticles precipitated equals to the input rate minus sum of the outputrate and accumulation rate of the salt.

Consider Qin and Qout (lit/sec) as the volumetric flow rates of input and outputstreams, respectively. Notations Cin and Cout (mol/lit) correspond to the salt con-centration in the input and output stream, respectively. The Cout equals to thesaturation concentration of salt. We use the perfect mixing assumption in thissystem. The rate of salt precipitation is Rpre. (mol/sec). The mole balance of salt inthis system is:

Fig. 1.13 Schematic diagramof filtration process


QinCin � QoutCout � Rpre: ¼ dVCout=dt ð1:4Þ

where VCout is total moles of salt in the system and it changes with the time. In thesteady state, his term is constant, and therefore, the accumulation term (dVCout/dt) isequal to zero.

In this case, the size of precipitated solid can be controlled by controlling the keyterm Rpre.. This term as a particular law is a function of the supersaturation degreeand its relation to several operation parameters such as temperature, resident timeand mixing. The nucleation and growth rate are the main mechanisms of the pre-cipitation that define its functionality (Pratsinis and Spicers 1998). Definition andderivation of the term Rpre. is one of the duties and abilities of chemical engineers inthe nanotechnology field.

Example 3

When the supersaturation condition is made with a chemical reaction, i.e., a reactoris used to produce nanoparticles (precipitator) by controlling the reaction rate. Inthis case, two streams containing A and B components enter the system and the saltwill be produced because of the reaction. The reaction progress causes the super-saturation and then precipitation of the solid salt. Again the solid particles canprecipitate as the nanosized particles by controlling the reaction and the supersat-uration. The mole balance on salt in this system has the generation term and noinput term.

Consider the reaction A + B → C in the liquid phase that produces componentC (the solid product) which has a low saturation concentration in a solute phase andcan precipitate. The precipitation process of component C is shown schematically inFig. 1.14.

The Rc (mole/sec) is the reaction rate which generates the solid C and Rpre.

(mole/sec) is the solid precipitation rate. The mole balance on C component in thissystem is given by (the feedstock is free of the solid C):

Rc � Rpre: � QoutCc;out ¼ VdCc;out=dt ð1:5Þ

Fig. 1.14 Schematic diagramof a CSTR


Again the Qout (lit/sec) is the volumetric flow rate of output stream and CC,out

(mol/lit) is the salt concentration in the output stream which is the same as saltconcentration within the system. V is the system volume. This equation needs aninitial condition which can define as:

at t ¼ 0 CC;out ¼ 0

This system also can operate in the steady state, and in this case, the left-handside of the above equation is equal to zero. Definition and derivation of the RC as afunction of the components concentration and temperature is subject of chemicalkinetics studies. This equation just describes the variation of solid concentration. Ifconcentration term of the other components is observed in the particular lawdescribing Rc and Rpre., we need mole balance on these components.

The conservation of total mass is a general basis to determine some unknownquantities required to obtain complete model of a system. Keep in mind that, unliketotal mass, total moles are not always conserved. In systems involving a chemicalreaction, the number of total moles may change and, thus, an equation similar toEqs. 1.1 or 1.2 cannot be written for total moles. To clarify this statement, considerthe following reaction equation

Aþ B ! C

One mole of A is reacted with one mole of B and one mole C is generated. Twomoles of reactants produce one mole of product. A statement of the number ofmoles or molecules reacting to produce products is given by a chemical equationknown as stoichiometric equation. In a reactive system, a mole balance is writtenfor each component in the following general manner:

Rate at which component enters

the system

Rate at which component is

generated

Rate at which component leves

the system+ =

Rate at which component is

consumed

Accumulation rate of a component in

the system+ +

Xinputstreams

_nA;in þ Rformation;A ¼X

outputstreams

_nA;out þ Rconsumption;A ð1:6Þ

It is supposed that the volumetric flow rate of a stream (Qin lit/sec) is known,along with the concentration of species A in that stream (CA). Under such cir-cumstances, the molar flowrate of A in that stream will be

FAin ¼ Qin � CA ð1:7Þ


The mass flow rate of component A can be calculated according to the followingrelation using molecular weight of the A component (MwA)

_mAin ¼ FAin �MwA ð1:8Þ

It should also be remembered that the sum of mass balances on the all com-ponents in a stream is the same as total mass balance. This means that mass flowrate of component A can be defined as a function of total mass flow rate:

_mAin ¼ _min � xA ð1:9Þ

where xA is the mass fraction of component A in the stream.If chemical reactions take place and the stoichiometry of the reaction is known,

equations based on mole balances are written for the system. The stoichiometry ofchemical reaction is usually known; however, there are some cases where noinformation about reaction stoichiometry is available. As an example, for complexsystems when pseudo-components rather than actual components are used,empirical stoichiometric numbers are used rather than rigorous stoichiometricnumbers. Accordingly, the following questions will lead the calculation procedurethrough the following decision tree:

• Is species information required, or will a total balance suffice?• If species information is required, are the terms associated with formation and/or

consumption necessary?• If the formation/consumption terms are required, is the reaction stoichiometry

known or unknown?

Once the type of process has been identified, the specific approach to modelingthe system can be used. Using material balance equations, it is frequently necessaryto use more than one balance to mathematical describe the system correctly. Inother words, if there are “n” components, we have “n” unknown variables whichare the concentrations of “n” components, and therefore, we need “n” equations. Inthis case, a mass balance for each component (n equations) plus one overall massbalance equation can be formulated. However, from the (n + 1) equations, only “n”equations are independent. For example, in a binary system (n = 2), a balance ontotal mass and a balance on one of the species both may be needed to arrive at aunique solution, or balances on two separate species might be needed. Of course, itis possible to write equations representing additional information as well (givenflow rates, given conversions, etc.). The strategy is to keep writing equations untilthe total number of independent equations equal to the total number of unknowns.

Reaction variables, Kinetics and rates

The chemical reactions are classified into single and multiple reactions from thestoichiometry view point. In a single reaction, there is just one reaction withreactants consumption and products production such as:


Single reaction: aAþ bB!cC þ dD

To define the progress of these reactions, there is just one equation. In themultiple reactions, there is more than one reaction effective on the componentconcentration variation and therefore more than one equation for the definition ofthe reactions progress. Examples of these multiple reactions are as follows:

Parallel reactions: A ! B and A ! C

Parallel reaction is parallel with respect to reactant A giving both B andC products in parallel reactions

Series reactions: A ! B and B ! C

Series reaction is a reaction producing the intermediate product B from thereactant A and B produces the final product C

Parallel � Series reactions: Aþ B ! C and Aþ C ! D

In parallel–series reaction, the reaction is parallel from the view point of com-ponent A and is series reaction from the view point of component C.

A reactor with a single reaction can be completely defined in terms of onevariable. In this system, there are algebraic relations between the componentsconcentration. The variable can be conversion of any one of the reactants, or theyield of any one of the products, or the concentration of one of the components, orthe rate of reaction of one of the components. Once any of the mentioned variablesis determined, all other quantities can be computed in terms of this single variableas long as the stoichiometric equation is fully defined.

The conversion of a reactant component is defined as the number of moles (ormolar flow rate) reacted of the specific component divided by the original numberof moles (or molar flow rate) of the same specific component in the feed. Let usconsider a reactor, depicted in Fig. 1.7 in which an irreversible reaction shownbelow takes place.

Aþ B ! C þ D

The conversion of component A is defined as

conversion ¼ moles (molar rate) of the reactant consumemoles (molar rate) of the reactant in the feed stream

xA ¼ nAf � nAnAf

ð1:10Þ


The conversion of component B is defined based on different bases as

xB ¼ nBf � nBnBf

or x0B ¼ nBf � nBnAf

ð1:11Þ

where nif is molar rate of component i in the feed stream and i = A, B, C, Dcomponent. The relation between xA and xB depends on the stoichiometric numbersof each component as well as their ratio in the feed stream. Of course, the relationbetween xA and xB will be different because of using different ways to define thebasis for each component.

In the chemical reaction, usually the main reaction is coupled with several sidereactions which will produce undesirable by-products. The task of chemical andnanoengineers is to maximize the production of the desired product (with desirableproperties) in the process. Two quantities, yield and selectivity, are basically usedfor this purpose. The yield of a product C is the number of moles of C formed (perunit time for continuous processes) divided by the original number of moles ofreactant A (or per unit time for continuous processes); that is,

Yield ¼ moles (molar rate) of desired product formedmoles (molar rate) of desired products formed if there were

no side reactions and the limiting component reacts completely

Yc ¼ nC � nCfnAf

or Yc ¼ nC � nCfnBf

ð1:12Þ

Also, the yield can be referred to the number of moles of A (or B) that reacted:

Y0c ¼

nC � nCfnAf � nA

or Y0c ¼

nC � nCfnBf � nB

ð1:13Þ

Selectivity is a main variable to describe the reaction progress in the multiplereactions. This variable has many definitions; one of the most popular ones is

Selectivity ¼ moles (molar rate) of desired product formedmoles (molar rate) of u desired products formed

Selectivity ¼ nC � nCfnD � nDf

ð1:14Þ

Example 4The single reaction σAA + σBB → σCC + σDD takes place in a batch reactor withthe initial concentration of the reactants and products nAf, nBf, nCf and nDf. In thiscase, we can use the following relations


nAf � nArA

¼ nBf � nBrB

¼ nC � nCfrC

¼ nD � nDfrD

ð1:15Þ

By dividing the above equation by nAf and rearranging it, we get:

nAf � nAnAf

¼ rArB

nBf � nBnAf

¼ rArC

nC � nCfnAf

¼ rArD

nD � nDfnAf

ð1:16Þ

And in the conversion and yield terms, it is:

xA ¼ aArB

xB ¼ rArC

YC ¼ rArD

YD ð1:17Þ

The rate of reaction in the context of material and energy balance is defined asthe number of moles of component produced per unit time. Of course, produced orconsumed in the definition is a matter of convention. Here, in this book, the con-vention used R is a rate of production. The choice is arbitrary; however, thechemical and nanoengineers should be very clear about the above meanings andsign conventions because in the complex process of design in the chemical andnano, different engineers (and different books and manuals) may choose the con-vention differently, which may cause confusion if the chemical/nanoengineer is notcompletely aware of the above simple and fundamental facts. The equations appliedshould be always correct and consistent under the chosen sign convention.

Basically, the reaction rate is a local function that varies with some other localvariables such as concentration and temperature. The functionality of the reactionrate with these local variables is the subject of kinetic study. From another view-point, discussion about the reaction rate can be treated as an overall (global) term. Inthis case, the reaction rate is defined based on just input and output concentrations.

According to the sign convention selected in this book, the consumption rate(RA) of component A is obviously negative; because actually A is a reactant (it isbeing consumed)

RA ¼ nA � nAf ð1:18Þ

If we write the production rate of A and call it �RA, then it is positive and definedas

�RA ¼ nAf � nA ð1:19Þ

Based on definition of conversion of component A (Eq. 1.9), we have:

RA ¼ �nAf xA ð1:20Þ

Consider again the reaction σAA + σBB → σCC + σDD taking place in the reactorin Fig. 1.15.


We have four rates of production: RA, RB, RC and RD, but it does not mean thatthe single reaction is defined by four different rates. The four rates are related by thestoichiometric numbers. Clearly, they are related based on Eqs. 1.15 and 1.18 asfollows:

RA ¼ rArB

RB ¼ rArC

RC ¼ rArD

RD ð1:21Þ

Consequently, all the rates of production of a single reaction can be all expressedin terms of one rate of production as long as the stoichiometric numbers for thissingle reaction are all known. A single reaction system is then fully defined andsolvable in terms of any conversion for one of the reactants or any yield of one ofthe products.

The rates of production are directly related to conversions (or yields) as well.Therefore, either one conversion (or yield) or one rate of reaction is needed to fullydefine a single reaction system. It is worth mentioning that if a degree of freedom ina problem is not zero, a conversion (or yield) and rate of production cannot be usedas two given relations. In case of using conversion (or yield), any rate of productioninformation is redundant from the degree of freedom point of view and vice versa.

Accordingly, there is a direct relation between RA and xA. Similar relations canbe developed between any rate of reaction and any reactant conversion (the readershould practice deriving these relations).

In conclusion, for given feed conditions in a single reaction system, the outputconditions are completely defined in terms of one and only one variable (conversionof any reactant, or yield of any product, or rate of reaction of any component). Thischemical specie is usually the limiting component of the reaction (which will bediscussed in the next section). Therefore, only one relation related to these variablesor one relation relating the output variables together can be used in the solution ofthe problem (and in the determination of the degrees of freedom).

RA ¼ �nAf xA ¼ � rArA

nAf xA ð1:22Þ

RB ¼ rBrA

RA ¼ � rBrA

nAf xA ð1:23Þ

RC ¼ rCrA

RA ¼ � rCrA

nAf xA ð1:24Þ

Fig. 1.15 Reactor system with multiple-inputs multiple-outputs with single reaction


RD ¼ rDrA

RA ¼ � rDrA

nAf xA ð1:25Þ

To generalize all explanation above, it can be written

Ri ¼ � rirA

nAf xA ð1:26Þ

whereRi rate of production of component i(σi) stoichiometric number of component i(σA) stoichiometric number of a special component (A)

Therefore, the relation between the rate of reaction of any component and theconversion for a specific equation does not depend on the stoichiometric number signconvention. Of course, we can choose any other component other than A as a base forconversion (if it is another reactant, then it will be conversion, but if it is a product, thenit will be called yield, not conversion). Thus, the general form of Eq. (1.25) is

Ri ¼ � rirk

nkf xk ð1:27Þ

where nkf is the feed number of moles on which basis x (conversion) is defined.Now, the generalized rate of reaction is

ni ¼ nif þ rir ð1:28Þ

where

r ¼ Ri

ri¼ � nkf xk

rk

Taken into account that σi is positive for products and σi is negative for reactants.It should be mentioned that in Eq. (1.26), we can use the yield variable of a productcomponent to replace the conversion variable of a reactant component.

The generalized mass balance equation of any component i in a system withsingle-input, single-output and N reactions (as shown in Fig. 1.16) is given by

Fig. 1.16 Reactor system with single-inputs single-outputs with multiple reactions


ni ¼ nif þ Ri ð1:29Þ

where Ri is the rate of production of component i in reaction j,

Ri ¼XNj¼1

ðrijrjÞ ð1:30Þ

where σij is the stoichiometric number of component i in reaction j.Multiple reactions involve two or more stoichiometric equations, each with its

own rate expression. They are often classified as complex reactions. For a systemwith multiple-inputs, multiple-outputs and multiple reactions (the most generalcase), as shown in Fig. 1.17, the mass balance equation is

XKk¼1

noi;k ¼XLl¼1

n fi;l þ

XNj¼1

ðrijri;jÞ ð1:31Þ

The system has L feed streams of component i as n fi;l l ¼ 1; 2. . .; Lð Þ and K output

streams of component i as noi;k k ¼ 1; 2. . .;Kð Þ and N reactions of component i as ri,j(i, j = 1, 2,…, N) and with M components (i = 1, 2,…, M). The parameter σij isstoichiometric number of component i in reaction j. Note that rj is the overall rate ofreaction for the whole unit (it is not per unit volume or per unit mass of catalyst andso on), because these are mass balance equations not sizing equations. Pay attentionthat Eq. (1.30) is M equations which can be coupled and can be applied to allpossible mass balance cases.

Kinetic consideration of a reaction

Basically, the reaction when it takes place the molecules of the reactant(s) interact.In other words, the interactions of the molecules of the reactant(s) with themselves(or each other for bimolecular reactions and higher) cause the reaction occur. In thisstate, any factor affecting the number of interactions will effect on the reaction rate.Reaction rate depends upon interactions among reactants. These interactions isaffected by temperature and reactants concentration.

Because the concentration and temperature are parameters which can vary pointby point of a reactor, therefore as it is mentioned before, the reaction rate is a localfunction.

Fig. 1.17 Reactor system with multiple-inputs multiple-outputs with multiple reactions


Fundamentally, finding reaction rate functionality needs to determine the reac-tion mechanism. The well-known reactions subject of last researches have totallydefined reaction rate. The famous and familiar reaction rate functionality is thepower law and Langmuir–Hinshelwood types (Levenspiel 1999):

Power law functionality:

RA ¼ KCn1A ð1:32Þ

Langmuir–Hinshelwood functionality:

RA ¼ K1CA

K2 þ K3CAð1:33Þ

The reaction rate constants (K) define the effect of temperature on the rate. Thefunctionality of reaction rate constant with temperature is predetermined andaccepted as the Arrhenius law as follows:

K ¼ K0e� E

RT ð1:34Þ

The functionality of reaction rate with the concentrations and temperature is akey point in the reactor modeling, design, simulation and optimization. For com-mercial well-developed systems, the reaction rate is well-known function. But in thenanoengineering since the reaction is the most important step in the nanostructureformation and development, the definition of the reaction rate is a bottle neck ofresearches.

The Limiting Component

This is a very important concept regarding the rational definition of the conversionfor a certain feed component. Consider A and B, which are two reactants of achemical reaction taking place in a reactor:

rAAþ rBB ! rCC þ rDD

At any time t, the molar contents of these two reactants are related by

nA tð Þ � nA0rA

¼ nB tð Þ � nB0rB

Reactant A is the limiting reactant (or component), if at any time (t) mole ofA vanishes before mole of B; hence, the following relation should be satisfied:

nA0rA

�� nB0

rB

�� ð1:35Þ


Note that absolute values are used because the stoichiometric coefficients ofreactants are negative. Equation (1.35) is the mathematical condition for stoichi-ometric proportion of the reactants in the reactor feed. In case of multiple chemicalreactions, the chemical reaction whose stoichiometric coefficients are used inEq. (1.35) is the stoichiometric relation that ties the reactants fed to the desirableproduct. The procedure for identifying the limiting reactant of a chemical reaction is

quite simple. For each reactant (i), calculate ni0ri

�� , where i = A, B,…. The reactant

with the smallest value is the limiting reactant.When the stoichiometric numbers for all reactants are equal, the problem is

rather trivial; it is even more trivial when the stoichiometric numbers of the reac-tants are equal and the feed is equimolar. However, for complex systems, it is notthat obvious. When the stoichiometric numbers of the reactants are not equal and/orthe feed molar flows of the different components are not equal, the problem is nottrivial, although it is very simple. For reactions in which all of the reactants have thesame stoichiometric numbers, the limiting component is the reactant with the lowestnumber of the moles in feed. For this case of equal stoichiometric numbers for thereactants, if we additionally have equimolar feed, then any reactant can be thelimiting reactant.

Consider the chemical reaction with different stoichiometric numbers for dif-ferent components. Now, if we take for this reaction that the limiting component isthe component with the lowest number of moles in the feed, we will fall into aserious mistake. According to Eq. (1.34), the one with larger stoichiometric numbercan be limiting component. If the number of moles are quite small to satisfyEq. (1.35), the limiting component cannot be found by just directly looking at theratio of the stoichiometric numbers.

Energy balances

In all engineering and nanoengineering systems, temperature of the system is a keyvariable for controlling the processes (such as the reactions and the mass transfers)and then the size of the product. This fact indicates the importance of temperatureprediction on the system which can be reached through the energy balance. Thetotal energy balance on a system is exactly similar to the other balances:

Energy Accumulation ¼ Energy Input Rate� Energy Output Rateþ Energy Production Rate� Energy Consumption Rate

In this case, the mechanisms of energy input (and output) rate are enthalpy (inputand output rates with the streams) and heat transfer (input/output rate via the systemboundary). The enthalpy is a type of energy carried by a mass flow, but the heattransfer term is due to temperature difference between the un-isolated system and itssurrounding. This phenomenon usually transports by one of three mechanisms


conduction, convection and radiation which is free of mass transfer. The accu-mulation of energy usually relates to the internal energy.

Example 5

Consider a system that you should control its temperature. For this purpose,nanofluid acts a perfect media to insert desirable conditions. Consider a double-pipeheat exchanger that a nanofluid stream heat (or cool) an environment or heating(cooling) media without reaction. This system is shown in Fig. (1.18):

In this case, the energy balance on the nanofluid gives

dqVCh�Tfdt

¼ _minChTin þ hA _�Tf � T0� �� _moutChTout ð1:36Þ

where Ch and q are heat capacity and density of nanofluid, respectively, h is heattransfer coefficient between nanofluid and the system, _min and _mout are mass flowrate of nanofluid in input and output streams, respectively, Tin and Tout are tem-peratures of nanofluid in input and output streams, respectively, �Tf is averagenanofluid temperature and T0 is the environment or heating (cooling) media tem-perature. If we can assume that the physical properties of nanofluid (such as densityand heat capacity) are independent of temperature, therefore the mass flow inlet isequal of mass flow outlet. In this case, the energy accumulation is just because ofjust temperature variation and no mass variation. One of the important abilities ofchemical engineers in the nanotechnology field is calculation and prediction of heattransfer coefficient of nanofluid. It should be so better than normal fluid, because thenanoparticles enhance the thermal conductivity and fluidity of the base fluid (Jangand Choi 2004).

In case of chemical reaction in a system, there might be generation of energy orexpenditure of energy. Conversion of one species or more to other species causingrelease of heat (exothermic reactions) or absorption of heat (endothermic reactions).Accordingly, it is vital to understand the concept heat of reaction and the methodsof its calculation. The heat of reaction is usually defined as:

Heat of reaction ¼ Enthalpy of products� Enthalpy of reactants

Fig. 1.18 Schematic diagram of a double-pipe heat exchanger


DHR ¼Xi

ðniHiÞproducts �Xi

ðniHiÞreactants ð1:37Þ

where ΔHR is the heats of reaction and it is positive for endothermic reactions andnegative for exothermic reactions.

Of course, scientifically there is nothing wrong in reversing the sign convention;however, the sign convention introduced above is the one adopted worldwide. Theenthalpies of products and reactants here are meant to be in the same molar pro-portions given in the stoichiometric equation. The heats of reaction under standardconditions can be computed from the heats of formation of the any components(Hf 0

i ) in this reaction (reactants and products with their stoichiometric amount andsign which is positive for products and negative for reactants) under standardconditions through the following formulae

DH0R ¼

Xi

riHf 0i ð1:38Þ

where “i” represents any component (reactant and product) in the reaction. Thestandard conditions are usually defined at 25 °C and 1 atm at the same phasewithout phase change. The heats of formation of large number of compounds aretabulated at standard conditions (heat of formation tables available in Perry’sChemical Engineer’s Handbook 1997).

Basically, the functionality of enthalpy of components with temperature, pres-sure (fugacity) and phase is well defined and presented in the thermodynamic books(Van Wyllan 1987). Therefore, in the any condition of the feed (Tf ;Pf ;pif ) and theproduct (T;P; pi), the enthalpy of the stream is known. The heat balance for anon-adiabatic system is given in Fig. 1.19, in which “i” represents the numbering ofall components involved in the reactions (reactants and products). If any productdoes not exist in the feed, then we put its nif = 0; if any reactant does not exist in theoutput, then we put its ni = 0.X

i

nif Hi Tf ;Pf ; pif� �þ Q ¼

Xi

niHi T ;P; pið Þ ð1:39Þ

Fig. 1.19 Heat balance for a non-adiabatic reactor


In a glance, term of heats of reactions cannot be observed in the correlation;however, it is automatically included. To describe this fact, consider the followingexample.

Consider an adiabatic batch reactor that is charged with a reactant(s) feed flow attemperature Tf (and therefore enthalpy Hif for component i). The reactions takeplace and temperature of the product will be Ti (and so the enthalpy Hi) because ofthe heat effects of the reaction. In this case, the term DHi ¼ Hi � Hif is enthalpydifference between input and output. In another view, assume that the reactantcooled (or heated) from temperature Tf to the reference temperature (T0), thenreaction took place in the reference temperature and then the product heated(cooled) from reference temperature to the temperature Ti. These events are shownin Fig. 1.20.

Enthalpy of the reactant cooling and the product heating can be calculated fromfollowing equations if there is no phase changes during heating and cooling (onlysensible heat):

DHif ¼ ZT�

Tf

CpreactantdT ð1:40Þ

DHi ¼ ZTi

T�CpproductdT ð1:41Þ

Finally, we can derive the following main results:

DH ¼ DHif þ DH�r þ DHi ð1:42Þ

The reaction enthalpy is included in the term DHi.Consider that we have a number of reactions taking place simultaneously. The

heat of reaction in the heat balance equation should account for all the heatsproduced/absorbed by all the reactions.

Fig. 1.20 Definition ofenthalpies in a reactor system


If the number of components is M, the number of reactions is N; we use i as thecomponent counter and j as the reaction counter. Therefore, the heat balanceequation becomes

XMi¼1

nif fHif Tf ;Pf ; pif� �� Hir Tr;Pr; pirð Þg þ Q

¼XMi¼1

nifHi T;P; pið Þ � Hir Tr;Pr; pirð Þg þXNj¼1

rjðDHRÞrjð1:43Þ

It is convenient to write these heat balance equations in a shorter form byrealizing that the Hif is the enthalpy at Tf, Pf, πif without writing them explicitlybetween brackets, and so on for other terms. Therefore, Eq. 1.38 can be rewritten asfollows:

XMi¼1

nifDHif þ Q ¼XMi¼1

niDHi þXNj¼1

rjðDHRÞrj ð1:44Þ

In this case, we just sum up the enthalpies differences (at which the heats ofreactions are taken) of the input streams and the output streams. If we have L inputstreams (and we use l as the counter for the input streams) and K output streams (weuse k as the counter for output streams) with L components and N reactions, then wehave the following most general heat balance equation

XLl¼1

ðXMi¼1

niflDHiflÞ þ Q ¼XKk¼1

ðXMi¼1

nikDHikÞ þXNj¼1

rjðDHRÞrj ð1:45Þ

This is the most general heat balance equation for a multiple-inputs,multiple-outputs (MIMO) and multiple reactions (and, may be, multi-components)system.

Example 6

Return to example of component concentration variation with reaction (example 3).In that system, the nanoparticles of C solid component were synthesized during areaction. This system is a lumped continuous reactor (perfect mixing condition)which operates in the unsteady state. Since the physical properties of the stream andespecially the reaction rate are functions of temperature and therefore contenttemperature affects the nanoparticle production, the temperature profile of thesystem should be determined using energy balance on the system. Followingequation describes energy balance on the system:

DHAf þ DHBf� �� DHA þ DHB þ DHCð Þ þ Q� DHrRC ¼ dq=dt ð1:46Þ


where ΔHAf and ΔHBf (J/mol) are total enthalpy difference of component A and B infeed stream andΔHA, ΔHB and ΔHC (J/mol) are total enthalpy difference of A, B andC components in the product streams, Q is heat transfer rate between the system andsurrounding (which is zero at adiabatic condition), ΔHr is reaction enthalpy and q isthe energy content of the system (which is the same of internal energy). Thisequation needs to an initial condition. It can be as follows:

at t ¼ 0 q ¼ q0

where q0 is the initial energy content of the system (before reaction progressing).The two equations for mole and energy balances of the system (1.5 and 1.46) are

coupled to solve, because the reaction constant (in the mole balance) if a function oftemperature (1.34) and the system temperature (in the energy balance) vary with thereaction rate and so with the concentration. Solution of these equations (withrespect of their particular laws) can help us to predict effect of the system param-eters (such as Cin, Tin, Qin and V) on the reaction progress and finally nanoparticleformation.

Momentum Balance

Another important term having important effect on the control of the systemsespecially on the nanosystems is resident time which is directly related to thevelocity of the materials stream. Determination of the stream velocity is carried outwith momentum balance. This balance is exactly similar to the other balances asfollows:

Momentum Accumulation ¼ momentum Input rate�momentum Output rateþmomentum production rate�momentum consumption rate

If the generation and consumption terms of momentum ignite (always theseterms are zero except of special cases such as source or sink events), the momentumbalance rearranges to the second law of Newton, as follows:X

Fco�direction �X

Fcounter�direction ¼ total acceleration ð1:47Þ

Total acceleration is equal mass multiple linear acceleration. It is important thatthis balance is in the vector form because the velocities are in the vector form. Theinserted forces are tangential (shear tension) or perpendicular (pressure force orgravity). Usually the general form of this balance in any directions and any coor-dinates is derived and presented in the fluid mechanics book and named Navier–Stokes equations (Bird et al. 1960). Nowadays, these equations are solvednumerically with different boundary conditions and presented as licensed software(such as Fluent and CFX) which is the subject of CFD (computational fluiddynamics) research field (Anderson 1995). Many publications on the analysis of theflow patterns of the materials in the nanoengineering systems show the importance


of velocity distribution on the product properties. Another advantage of CFD isconsidering interaction of molar and energy balances on the momentum balances(Hoffmann and Chiang 2000).

1.3.2 Stationary Non-equilibrium State Modeling Approachfor Chemical Engineering Systems Basedon MIMO-MRs Generalized Mass and EnergyBalances

The term steady state commonly used in chemical engineering and other disciplinesis not precise enough. A more accurate term should be stationary non-equilibriumstate, which is a characteristic of open systems, distinguishing it from stationaryequilibrium state, associated with isolated and closed systems (batch processes).

Steady state occurs when the state of the system does not change with time, but thesystem is not at thermodynamic equilibrium. This steady state of lumped systems is apoint in a space having the same dimensions as the problem (number of components,temperature, pressure, etc.), whereas that for distributed systems is a profile in thespace coordinate(s) as additional dimension(s). Unsteady state of an open systemstarts at an initial condition and tends with time toward a steady state when the systemis stable (a point for lumped system and profile for distributed systems). In opensystems, when no parameters in the system changewith time, the system does not tendto the thermodynamic equilibrium, but to the state with minimum entropy generation.In chemical engineering, this state is commonly called “steady state”; however, it isnot distinctive enough and a better and more accurate phrase is “stationarynon-equilibrium state.”This phrase is a characteristic of open systems and importantlyis distinguishable from the “stationary equilibrium state,” associated with an isolatedand closed systems (batch systems). For such a process, “stationary equilibrium state”,there would be no accumulation of mass in the system, because an accumulation ofmass would be a change in mass with time. For such a steady-state process, it can bewritten:

Rate of mass entering system ¼ Rate of mass leaving system

Or in the algebraic form:

_min ¼ _mout ð1:48Þ

A more general statement of the steady-state total mass balance for MIMOsystems is:

Fig. 1.21 Schematic diagram of a general system with input and output streams


Xinputstreams

_min ¼X

outputstreams

_mout ð1:49Þ

To make it clearer, consider a typical system such as the one depicted inFig. 1.21, in which several flowing streams bring material into the system andseveral streams take the materials out of it.

For the system illustrated in Fig. 1.21, the concept expressed in Eq. 1.48 iswritten

_m1 þ _m2 þ _m3 ¼ _m4 þ _m5 þ _m6 ð1:50Þ

Again, in steady state, there would be no accumulation of mass of species in thesystem, and for any component A, it can be said:

Rate that A enters the system

Rate that A is generated in the

system

Rate that A leaves the system+ = Rate that A is

consumed in the system

+

To formulate the above statement as a mass balance and in a format that we canuse, we introduce the following definitions:

Rformation A ¼ rate species A is formed; in units of mass=time

Rconsumption A ¼ rate species A is consumed; in units of mass=timeXinput streams

_mA;in þ Rformation A ¼X

output streams

_mA;out þ Rconsumption A ð1:51Þ

Since stationary non-equilibrium state is stipulated for the system, the accu-mulation of heat is zero too. Moreover, the heat generation (consumption) isconsidered for the exothermic (endothermic) reaction in the system volume (V).Considering heat effect of surrounding on the system (Q), we have:

XLl¼1

XMi¼1

nifl Hifl Tf ;Pf ; pif� �� Hir Tr;Pr; pirð Þ� �þ Q

¼XKk¼1

XMi¼1

nikHik T ;P; pið Þ � Hir Tr;Pr; pirð Þð Þ þ VXNj¼1

r0jðDHjÞð1:52Þ

In case no change in phase is involved, and therefore, the change in enthalpies isonly change in sensible heats (heat capacities of the components “Cpi” can befunction of temperature), then the above heat balance equation for a SISOmulti-component multi-reactions system is:


XMi¼1

nifZTf

Tr

CPidT þ Q ¼XMi¼1

niZT

Tr

CPidT þ VXNj¼1

r0jðDHjÞ ð1:53Þ

The stationary non-equilibrium state associated with continuous processes is themost common processing mode in the petrochemical, petroleum refining andnanoindustries.

A simple example for a system with stationary non-equilibrium state is thewell-known idealized CSTR, where the dynamics of the system is described bysimple ordinary differential equations and the steady state (stationarynon-equilibrium state) is described by algebraic equations. In order to illustrate thissimple case, let us consider the very simple uni-molecular irreversible reaction withlinear kinetics taking place in a single isothermal CSTR with constant input con-ditions and no change in the flow rate or physical properties due to reaction.Figure 1.22 shows a schematic diagram for the said reactor.

For the reaction A → B, the rate of reaction per unit volume is given by,

r ¼ kCAmolm3 � h ð1:54Þ

Thus, the rate of consumption of component A and the rate of production ofcomponent B are

�rA ¼ �rB ¼ kCA ð1:55Þ

The simple unsteady-state equation is obtained by performing a material balanceon components A and B

d�nAdt

¼ nAf � nA þ V�rA ð1:56Þ

d�nBdt

¼ nBf � nB þ V�rB ð1:57Þ

Fig. 1.22 Schematic diagramfor the idealized isothermalCSTR


wherenA and nB are the outlet molar flow rates from the reactor of A and B,

respectively;�nA and �nB are the molar holdup inside the reactor;nAf and nBf are the molar feed flow rates;V is the reactor active volume which is assumed constant

The inlet and outlet volumetric flow rates are assumed constant and the same (q).In addition to the assumption of perfect mixing, which implies that concentrationsof A and B at all points of the reactor are equal and equal to the output concen-trations. Then, we can write:

nif ¼ qCif

ni ¼ qCi

�ni ¼ VCi

i can be either A or B.Ci stands for exit concentrations, subscript “f” stands for input variables, and

then, Cif is input concentration of component i. All concentrations are in Kmol/m3.Substituting the recent definition of ni into Eqs. (1.56) and (1.57), the following

simple equations are obtained

VqdCA

dt¼ CAf � CA

� �� VqkCA ð1:58Þ

VqdCB

dt¼ CBf � CB

� �þ VqkCA ð1:59Þ

These equations describe the change in the two state variables CA and CB withtime.

Obviously, any change with time must has a beginning and possibly an end(depending on how we define “end”). For this first very simple case, it can bepositively asserted that the dynamics of the system has a simple “end,” which is thestationary non-equilibrium state (i.e., the steady state in common chemical andnanoengineering terminology).

The beginning is what we usually call the initial conditions, that is the state ofthe system (the value of CA, CB in our case) at some starting time which will bedesignated as time zero. Therefore, the initial conditions for this system are:

at t ¼ 0 CA ¼ CA0 and CB ¼ CB0

The steady-state equations are obtained by setting the unsteady-state terms in theleft-hand side of the equations equal to zero, since at this state there is no change in


state variables with time. Clearly, the exit concentrations will be determined bysolving an easy algebraic equation for each component.

In case of multiple reactions, for example, a consecutive reaction takes place inan isothermal CSTR, with B being the intermediate desired product.

A!k1 B!k2 C

For simplicity, we assume that the feed is pure A (contains neither B nor C), i.e.,CBf = CCf = 0.

Now, the unsteady-state mass balance equation for component A is

VdCA

dt¼ qCAf � qCA � Vk1CA ð1:60Þ

and that for component B is:

VdCB

dt¼ �qCB þ V k1CA � k2CBð Þ ð1:61Þ

with the initial conditions CA = CA0 and CB = CB0 (at t = 0).The objective is to determine the size of the reactor, V, that gives maximum

concentration of the desired product, B, for given q, CAf, k1 and k2. To do this task,the steady-state equations for the CSTR should be written. These equations can besimply obtained by setting the time derivatives in Eqs. (1.60) and (1.61) equal tozero, thus giving

qðCAf � CAÞ ¼ Vk1CA ð1:62Þ

and

qCB ¼ V k1CA � k2CBð Þ ð1:63Þ

Fig. 1.23 Variation ofconcentration of componentsA and B with reactor volume(V)


Some simple manipulations of Eqs. (1.62) and (1.63) give

CB ¼ qk1CAf Vðqþ Vk1Þðqþ Vk2Þ ð1:64Þ

Figure 1.23 shows the functionality of components concentration A and B withthe parameter V (the reactor volume). This figure shows an optimum point in theCB. At this point (Vopt), the maximum concentration of component B with observedand the concentration CA is CAopt.

To obtain Vopt, we differentiate Eq. (1.63) with respect to V to get

dCB

dV¼ qk1CAf

q2 � k1k2V2

ðqþ Vk1Þ2ðqþ Vk2Þ2" #

ð1:65Þ

Then, putting dCBdV ¼ 0 gives Vopt:

Vopt ¼ qffiffiffiffiffiffiffiffiffik1k2

p ð1:66Þ

At the optimum volume, maximum concentration of component B is determinedas:

CBmax ¼CAf

1þffiffiffik1k2

q� �2 ð1:67Þ

This condition will occur at optimum concentration of component A as:

CAopt ¼CAf

1þffiffiffik1k2

q ð1:68Þ

If the reactor is operating at this output concentration, disturbances will cause itto deviate from it.

As it was demonstrated, steady-state models are invaluable in steady-statedesign, optimization and control, in which the operator is not highly concernedabout the dynamic behavior of the system, but is mostly concerned about operatingthe unit at its optimum steady state in the face of long-term external disturbances.This is a situation in which one of the operating parameters changes, for examplefeedstock composition, and the steady-state control question is: What are the inputvariables that need to be changed in order to keep the unit at the same steady stateand producing the desired yield and production rates?

In conclusion, steady-state models are those sets of equations which are timeinvariant and describe the conditions of the system at rest (i.e., when the states ofthe system are not changing with time). This will automatically presuppose that the


system parameters are also time invariants (i.e., input variables, heat transfercoefficients, catalyst activity and so forth are not changing with time). Of course,this is a theoretical concept, for no real system can fulfill these requirements per-fectly. However, this theoretical concept represents the basis for the design andoptimization of almost all chemical/biochemical and nanoengineering equipment.The philosophy is that we assume that the system can attain such a time-invariantstate and design the system on that basis. Then, we design and implement thecontrol system that always “pushes” the system back to its optimally designedsteady state.

1.3.3 Dynamic Modeling Based on MIMO-MRs

As discussed in the previous section, steady-state models are applied for design,scaling-up and optimization purposes. The great majority of chemical and nano-processes are designed for steady-state operation. However, even steady-stateprocesses must occasionally start up and shut down which are both dynamicoperations experiencing change with time. Also, an understanding of processdynamics is necessary to design the control systems needed to handle upsets and toenable operation at steady states that would otherwise be unstable. For start-up andcontrol of units, dynamic models are utilized. Dynamic modeling obtainsunsteady-state equations of an open system (continuous process). The equations aredeveloped in a later stage for the design of the proper control loops in order to keepthe process dynamically operating near its optimum steady-state design in the faceof external disturbances. Dynamic models are used to compensate for the dynamiceffects associated with external disturbances. They are also used for stabilization ofthe desirable unstable steady states by designing the necessary stabilizing controlloops.

For a CSTR depicted in Fig. 1.22 (previous section), it was demonstrated that itcan be written:

VqdCA

dt¼ CAf � CA

� ��VqkCA ð1:69Þ

VqdCB

dt¼ CBf � CB

� �þ VqkCA ð1:70Þ

To solve these equations:

CA sð Þ ¼ CAf

1þ aþ CA0 � CAf

1þ a

� e� 1það Þs ð1:71Þ


CB sð Þ ¼ CBf þ a1þ a

CAf þ CB0 � CBf þ CA0 � CAf� �

e�s

� CA0 � CAf

1þ a

� e� 1það Þs ð1:72Þ

where s ¼ t�qV and a ¼ V �k

q .

Equations (1.71) and (1.72) describe the change in CA and CB with time from theinitial condition CA = CA0 and CB= CB0 at s ¼ 0 up to the end when s ! 1. Finalamount of the components concentration A and B calculated using Eqs. (1.69) and(1.70) at s ! 1 will be CA1 ¼ CAf

1þa and CB1 ¼ CBf þ a1þaCAf . Figure 1.24 illus-

trates change in the concentrations with dimensionless time.In the above case of multiple reactions in the reactor, such as the consecutive

reaction shown above (A → B → C) takes place in an isothermal CSTR with Bbeing the desired product. Assume that the feed is pure A. If the objective is tofollow change in the state variables with time due to a disturbance, say a change inq, the unsteady-state equations should be solved.

If q changes to q′, then we insert q′ in the dynamic equations instead of q andsolve the differential equations from t = 0 to higher values of time in order to followthe change with time. At large values of t, the system will settle down to a newsteady-state corresponding to the new q′. These new values of CA and CB willbe C′A and C′B:

C0A ¼ q0CAf

q0 þ Voptk01� � ð1:73Þ

C0B ¼ q0k1CAf Vopt

q0 þ Voptk1� �ðq0 þ Voptk2Þ

ð1:74Þ

Fig. 1.24 Change incomponents concentrationA and B with dimensionlesstime


q0k01C0Af Vopt


¼ qk1CAf Vopt

qþ Voptk1� �ðqþ Voptk2Þ

ð1:75Þ

Vopt is used here as a symbol to indicate the value of V chosen to give CBmax for q.Note that, for q′, the volume Vopt is no longer optimum. This steady-state

deviation from CBmax can be compensated for by the change in another variablesuch as CAf and/or the operating temperature that will change k1 and k2. Therequirement for this compensation (say, in feed concentration) will be that

q0C0Af

q0 þ Voptk01� � ¼ qCAf

qþ Voptk1� � ð1:76Þ

and

q0k01C

0Af Vopt


¼ qk1CAf Vopt

qþ Voptk1� �ðqþ Voptk2Þ

ð1:77Þ

These are two equations in three unknowns (C′Af, k′1 and k′2). However, note thatboth k′1 and k′2 are functions of the new temperature T′. Equations (1.76) and (1.77)can be solved for C′Af and T′ (gives k′1 and k′2) to find the new feed concentrationand temperature needed to maintain the system at its original optimum state despitethe change in q to q′. This is what can be called steady-state control.

A question arises with regard to the dynamic variation due to change in input.What is to be done when the variation in input parameters is continuous (i.e., whenthe system does not have enough time to settle to any new steady state, i.e., notenough time to achieve steady-state control as described above)? In this case, thedynamic model equations must be used to design a controller that introducescontinuous compensation which is changing with time. It is also important in thisrespect to make it clear that when the disturbances are very slow (e.g., slowlydeactivating catalyst), the quasi-steady-state approximation can be used, where thechange with time is considered as a series of steady states each corresponding to thevalue of the changing variable at the sequence of time intervals.

In industrial catalytic reactors with their heterogeneous and distributed nature(variation of the state variables with respect to the space coordinates), dynamictemperature runaways may occur, especially for highly exothermic reactions.A reliable dynamic model is one of the best ways to discover and monitor thesetemperature runaways, which may cause explosions or, at the least, emergencyshutdowns, which are quite expensive, especially with today’s large-capacity pro-duction lines.


1.3.4 Simulation and Optimization of ChemicalEngineering Systems

Process simulation may be used to recognize active bottlenecks by increasing pro-duction and detecting the first unit to reach its maximum capacity (Litzen and Bravo1999; Tan et al. 2006; Alshekhli et al. 2011). To perform simulation and optimi-zation of any chemical system, it is first required to formulate or model the system.The first step in formulation and modeling of a unit or a process is essentiallyqualitative that involves drawing as schematic diagram of the system to be studied.The second step is collecting all applicable physical and chemical information,conservation laws, and rate expressions. Then, the real purposes of the modelingeffort must be classified and clarified. Determining whether the model is to be usedonly for explaining trends in the operation of an existing piece of equipment? Or isthe model to be used for predictive and/or design purposes? Finally, is it to be usedfor steady-state or transient studies? The scope and depth of these early decisionsdetermine the ultimate complexity of the final mathematical description. The prac-tical advantages gained from the use of steady-state models in design, optimizationand operation of systems are remarkable (Pantelides and Renfro 2013; Yuan et al.2012). Although the success of the process mostly can be achieved through thesteady-state design, in certain cases, inefficient dynamic control may cause seriousproblems or malfunctions in the process (Yuan et al. 2012).

1.4 Chemical Engineering and New Materials

In a multi-disciplinary world of science and engineering, chemical engineers are notjust involved in designing and controlling chemical plants. Besides design, oper-ation and control of chemical processes on a large scale, they are now dealing withthe formation of new materials. Indeed, one of the main tracks for future researchesin chemical engineering involves tailoring of materials with controlled structures,targeted properties and new applications (Favre et al. 2002; Müller-Goymann 2004;Kothare 2006; Dastjerdi and Montazer 2010; Li and Lin 2010). The new materialsoffer novel optical, magnetic, electronic or mechanical characteristics and/or sig-nificantly enhanced properties, such as high surface area. Novel materials are aimednew applications and contributions to solutions of the increasing problems ofenergy and environment (Ashby 2012). In other words, new products withnoticeable properties and new functions will revolutionize much of how we willlive in the years ahead.

As an example, chemical engineers are keenly involved in developing newprocesses to produce nanotechnology-based products, including nanostructuresmaterials like CNT, metal nanoparticles, quantum dots and thin films or materialsincluding nanostructures like nanocomposites, nanofluids, nanocrystalline metaland nanosensors (Comini 2006; Roy et al. 2011; Singh et al. 2011; Nasiri et al.


2013; Sharma et al. 2013; Jajja et al. 2013; Liu and Kumar 2014; Leong et al. 2014;Oueiny et al. 2014). The unique properties of nanostructures have sparked theattention of scientists, engineers and manufacturer to generate materials and deviceswith new physical characteristics and chemical/biochemical functionalities for awide variety of applications, as will be completely explained in Sect. 2.1.1. Fornanostructures, which have the size of below 100 nm, the theories of classical andquantum mechanics are no longer valid and a rich variety of unexpected propertiesare possible. For instance, nanofluids, well dispersion of nanostructure in a fluid,enhance the thermal conductivity of the fluid. Another example is incorporatingCNTs into plastics can lead to a dramatically increased modulus of elasticity andstrength in structural materials (Endo et al. 2008; Paul and Robeson 2008). A greatnovelty with CNTs is that they can achieve high stiffness along with high strength.Bringing these new types of products to the commercial arena is now the mainobjective of major chemical firms and industries. This will definitely achievethrough technical programs and through investments to find appropriate materials,processes and applications. All products that a chemical company sells to its cus-tomer need to have a clearly defined physical shape in order to meet the designedand the desired quality standards. Indeed, clients buy the product which is the mostefficient. The main key characteristics demanded for nanostructures to capture highvalue markets include: producing uniform size of nanoparticle with low levels ofagglomeration and high dispensability (Yang et al. 2010; Aravind et al. 2011; Zhaoand Astruc 2013; Leong et al. 2014). According to these researches referred here, akey obstacle in the development of new materials lies in avoiding any inability todirectly control the structure formation at multiple hierarchical levels. So, newdevelopments increasingly concern highly targeted and specialized materials, activecompounds and special effect chemicals, which are complex in terms of molecularstructure. Dominant elements in producing the most efficient products are control ofthe end-use properties and expertise in design of the relevant process, permanentadjustments to diversity and varying demands along with rapid reacting to marketconditions. Chemical engineering discipline includes the whole of scientific andtechnical knowledge necessary for physical, chemical or biological transformationsof raw materials and energy into the targeted products. Accordingly, this disciplinecovers broad variety of knowledge and technologies emphasizing on the productionof products with specific end-use properties.

Chemical engineers should exploit their theoretical knowledge to design the bestindustrial equipment for the production of a certain material that possesses desiredstructure and properties of use. The unique specialty of chemical engineers is tosimulate the operating conditions of a unit process. Nevertheless, approaches con-cerning the tailoring of materials with controlled structure needs to understand thematerial behavior at the atomic level and to describe relationships between thestructure of a material and its properties. Accordingly, chemical engineers should godown to the nanoscale to control phenomena like molecules interactionsself-organization, regulation and replication (Charpentier 2002). The approach ofutilizing nanostructure building blocks to create multi-functional materials needs theemergence of novel technological concepts for synthesis of nanostructures and their


http://dx.doi.org/10.1007/978-981-287-496-2_2

tailoring, characterizing as well as incorporating to the other matrix (for exampleproducing nanocomposites). Accordingly, chemical engineering requires scopingwith emerging technologies such as nanotechnologies to form nanoengineering.

Chemical engineers are nowadays playing a fundamental role in recent tech-nological revolution with their broad knowledge in chemistry, physicochemistry aswell as processing and product design. However, chemical engineers are facingvarious challenges in fundamentals (structure–activity relationships on molecularlevel, interfacial phenomena, i.e., adhesive forces, molecular modeling, i.e., equi-librium, kinetics, product characterization techniques, etc.), in product design(nucleation growth, internal structure, stabilization, additives, etc.), in processintegration (simulation and design tools based on population balance) and in pro-cess control (sensors and dynamic models) (Charpentier 2002; Becht et al. 2009;Jaworski and Zakrzewska 2011). To satisfy the previous demands, it involvescreating innovative industrial processes with desired characteristics such as beingenergy saving and totally safe and having minimum pollution and producing nodefected products. For example in nanotechnology, the process designed must bescalable, capable of replicating identical nanostructures on a massive scale and doesnot require harsh chemicals or sophisticated equipment. Optimization of primaryconditions for practical applications is also vital (Charpentier 2002).

Besides all areas explained in the previous paragraphs, the integration of scales,as well as the mixing of physical, biological and chemical concepts into novelengineering designs could complement the current practice of disease diagnosis andtreatment, as well as the design of new materials, and therefore unfold manyopportunities for technological innovations (Papazoglou and Parthasarathy 2007;Choloupla and Malam 2010; Pankhurst et al. 2003). The wide impact of usingmaterials science approaches in biology and biomedical sciences, in the context oftissue engineering and regenerative medicine, have started to play an important rolein the biomedical literature (Papazoglou and Parthasarathy 2007; Scholz et al.2011). Tissue engineering can be considered as a subsection of chemical engi-neering applications that has the potential to deliver exciting future technologies.

1.5 Preliminary Introduction to Nano Scale ProcessEngineering

Chemical engineering research has effectively solved most of the major techno-logical problems associated with simulation, design, control, diagnosis, schedulingand planning of operations for large-scale continuous and batch chemical processes(Stephanopoulos et al. 2005). The unique focus of chemical engineering onmolecular transformations, manufacturing processes, multi-scale treatments andquantitative analysis provides an ideal platform for productive interactions on thenanoscale with a number of other science and engineering disciplines at the

1.4 Chemical Engineering and New Materials 67

boundaries that are included in the most exciting areas of modern science andtechnology (Charpentier 2002).

Nanoscience and technology, a high-tech developed since the end of last century,is devoted to the understanding, control and manipulation of matter at the level ofindividual atoms and molecules. Its goal is to create materials, devices and systemswith essentially new properties and functions because of their small structures. Thedevelopment of this technology serves to fill up mankind’s knowledge of many newphenomena and processes at the scale of below 100 nm. At the nanoscale, thephysical, chemical and biological properties of materials differ in fundamental andvaluable ways from the properties of individual atoms and molecules or bulk matter.Research in nanoscale is directed toward understanding and creating improvedmaterials, devices and systems that exploit these new properties.

Chemical engineers are nowadays widely active in the area of nanoscale pro-cessing, which is based on the interdisciplinary nature of nanoscale science andtechnology. The nanoscale processing researches mainly deal with transformationof materials and energy into nanostructured materials and nanodevices and syner-gize the multi-disciplinary convergence between materials science and technology,biotechnology and information technology. Research in nanoscale science andengineering is basically directed toward three main targets, addressed as:(Stephanopoulos et al. 2005)

• Design and manufacturing of materials with passive nanostructures (e.g.,nanostructured coatings, dispersion of nanoparticles and bulk nanostructuredmetals, polymers and ceramics).

• Design, construction and operation of nanodevices (e.g., transistors, amplifiers,targeted drugs and delivery systems, actuators and adaptive structures).

• Design, fabrication and operation of integrated “nanoscale factories,” that is,processes with unit operations and materials movement among these units at thenanoscale, along with the requisite energy supply system and monitoring andcontrol infrastructure (e.g., nanoscaled reactors, separators, molecular tubes,motors, shuttles or pumps, molecular gates or channels).

Chemical engineering is one of the active disciplines on recent researches relatedto nanoscale processing. However, the absence of a systems theory for the engi-neering of such process merges the inherent physical and chemical difficulties intheir design and integration. Construction of different configurations of nanotubevesicle networks is among the first examples of integrated nanoscale units(Stephanopoulos et al. 2005). Consequently, novel theories and tools are required tomanage the design, simulation, operation and control of nanoscale processing. Eachsignificant advance in understanding the physical and chemical properties andfabrication principles, as well as development of predictive methods to controlthem, possibly leads to major advances in the ability to design, fabricate andassemble the nanostructures and nanodevices into a working system (Roco 2004).All engineered nanodevices regardless of composition, structure, function orimplementation have several fundamental characteristics in common such asenvironmental compatibility, controllability and service reliability. All of these


require the development of basic scientific and engineering elements that areconverged at the establishment of clearly defined core technologies, which in turnsupport the realization of novel nanodevices. The nanoscale systems are somehowdifferent from the conventional systems existed with the following distinguishingfeatures (Stephanopoulos et al. 2005):

(a) The “unit operations” are self-assembled supramolecular structures at the scaleof a few nanometers,

(b) The spatial topology of the “process flow sheets” is guided by molecularscaffolds, and the unit operations are positioned in space through directedself-organization mechanisms of independent units and

(c) The operation of such “supramolecular factories” is driven bypre-programmed information encoded in the design of the system itself and isrobustly controllable through local feedback loops with no evidence of cen-tralized coordination mechanisms.

Exercises

1. How integrated system approach helps nanoengineers to achieve their goal?2. Give three examples of chemical process and state whether they are open,

closed, or isolated.3. Define the state variables and the laws governing the processes in terms of the

state variables, and the system input variables for these three systems

• Dose of nanoparticles is separated from the water stream using ananomembrane.

• A nanocrystalline is synthesized in a Crystallizer.

4. Acetylene stream enters the reactor and a solid product, in form of nanoparti-cles, is obtained. Hydrogen and carbon dioxide are also produced as byproducts that leave the reactor.

5. What are the main objectives of modern chemical engineers?6. What are the challenges of nanotechnology? How a chemical engineer con-

tributes in resolving these challenges?7. What are the main schemes of process intensification?8. What are chemical engineering expectations on 2020?9. How chemical engineer profession contributes in nanoengineering?

10. What are the main targets of research in nanoengineering and how a chemicalengineer contributes?

1.5 Preliminary Introduction to Nano Scale Process Engineering 69

11. Consider a reactor similar of Example 3. The production rate of component C isproportional to concentration of the components A and B (Rc = k CACB mol/lit.sec).

A. Calculate the precipitation rate of nanoparticles, which is constant.B. The precipitation rate is the main parameter that affects the size of nano-

particles. Based on derived equation for the precipitation rate, what is themain parameter that influences the precipitation rate of nanoparticle andthen their size?

12. Water in the cooling system of an automobile warms 25 °C. It is aimed to usenanofluid instead of water in this cooling system. If nanofluid improves the heattransfer coefficient (h in Eq. 1.36) 10 %, how the flow rate of nanofluid willdecrease? The other conditions of two systems (water and nanofluid) such aswarming degree, engine conditions are the same. The systems are steady state.

13. Consider the CSTR reactor Example 6. How much energy should be transferredbetween the reactor and surrounding to make isothermal condition in thereactor? The reaction is endothermic. The thermo-physical properties of thestream are constant and the heat effect of the precipitation is negligible. Thesystem is steady state.

14. In a CSTR reactor, two components of A and B are produced via parallelreactions, where B is desirable and C is undesirable component. If fractionalyield is defined as CB/CC, calculate the fractional yield in this reactor anddiscuss about maximizing this parameter.

ðA ! B and A ! C take placeÞ

References

Anderson, J. D. (1995). Computational fluid dynamics. New York: McGraw-Hill.Andrzej, S. (2003). Re-Engineering the Chemical Processing Plant. Process Intensification: CRC

Press.Adris, A. M., Elnashaie, S. S. E. H., & Hughes, R. (1991). A fluidized bed membrane reactor for

the steam reforming of methane. The Canadian Journal of Chemical Engineering, 69(5),1061–1070.

Albright, L. (Ed.). (2008). Albright’s chemical engineering handbook. Boca Raton: CRC Press.Alklaibi, A. M., & Lior, N. (2005). Membrane-distillation desalination: Status and potential.

Desalination, 171(2), 111–131.Alshekhli, O., Foo, D. C., Hii, C. L., & Law, C. L. (2011). Process simulation and debottlenecking

for an industrial cocoa manufacturing process. Food and Bioproducts Processing, 89(4), 528–536.

Aravind, S. J., Baskar, P., Baby, T. T., Sabareesh, R. K., Das, S., & Ramaprabhu, S. (2011).Investigation of structural stability, dispersion, viscosity, and conductive heat transferproperties of functionalized carbon nanotube based nanofluids. The Journal of PhysicalChemistry C, 115(34), 16737–16744.


Aris, R. (1999). Mathematical modeling: A chemical engineer’s perspective (Vol. 1). Waltham:Academic Press.

Aris, R., & Amundson, N. R. (1958a). An analysis of chemical reactor stability and control—I:The possibility of local control, with perfect or imperfect control mechanisms. ChemicalEngineering Science, 7(3), 121–131.

Aris, R., & Amundson, N. R. (1958b). An analysis of chemical reactor stability and control—II:The evolution of proportional control. Chemical Engineering Science, 7(3), 132–147.

Ashby, M. F. (2012). Materials and the environment: Eco-informed material choice. Amsterdam:Elsevier.

Barkoudah, Y., & Henderson, J. (2006). Plant ashes from Syria and the manufacture of ancientglass: Ethnographic and scientific aspects. Journal of Glass Studies, 48, 297–321.

Becht, S., Franke, R., Geißelmann, A., & Hahn, H. (2009). An industrial view of processintensification. Chemical Engineering and Processing: Process Intensification, 48(1), 329–332.

Ben-Guang, R. O. N. G., Fang-Yu, H., Kraslawski, A., & Nyström, L. (2000). Study on themethodology for retrofitting chemical processes. Chemical Engineering and Technology, 23(6), 479–484.

Berry, D. A., & Ng, K. M. (1997). Synthesis of reactive crystallization processes. AIChE Journal,43(7), 1737–1750.

Bird, R. B., Stewart, W. E., & Lightfoot, E. N. (1960). Transport phenomena. Hoboken: Wiley.Blumenthal, P., & Grothus, U. (2008). Developing global competence in engineering students: US

and German approaches. Online Journal for Global Engineering Education, 3(2), 1.Brill, R. H. (1970). The chemical interpretation of the texts. In L. A. Oppenheim et al. (Eds.), Glass

and glassmaking in ancient mesopotamia (pp. 105–128). Corning: The Corning Museum ofGlass Press and London. (Associated University Presses).

Broens, R. C., & de Vries, M. J. (2003). Classifying technological knowledge for presentation tomechanical engineering designers. Design Studies, 24(5), 457–471.

Byrne, E. P., & Fitzpatrick, J. J. (2009). Chemical engineering in an unsustainable world:Obligations and opportunities. Education for Chemical Engineers, 4(4), 51–67.

Charcosset, C. (2009). A review of membrane processes and renewable energies for desalination.Desalination, 245(1), 214–231.

Charpentier, J. C. (2002). The triplet “molecular processes–product–process” engineering: Thefuture of chemical engineering? Chemical Engineering Science, 57(22), 4667–4690.

Charpentier, J. C. (2007). In the frame of globalization and sustainability, process intensification, apath to the future of chemical and process engineering (molecules into money). ChemicalEngineering Journal, 134(1), 84–92.

Charpentier, J. C. (2010). Among the trends for a modern chemical engineering, the thirdparadigm: The time and length multiscale approach as an efficient tool for processintensification and product design and engineering. Chemical Engineering Research andDesign, 88(3), 248–254.

Chen, W., Duan, L., & Zhu, D. (2007). Adsorption of polar and nonpolar organic chemicals tocarbon nanotubes. Environmental Science and Technology, 41(24), 8295–8300.

Chen, Z., & Elnashaie, S. E. H. S. (2004). Bifurcation behavior and efficient pure hydrogenproduction for fuel cells using a novel autothermic membrane circulating fluidized-bed(CFB) reformer: Sequential debottlenecking and the contribution of John Grace. IndustrialEngineering Chemistry Research, 43, 5449–5459.

Chen, Z., Yan, Y., & Elnashaie, S. S. (2003). Modeling and optimization of a novel membranereformer for higher hydrocarbons. AIChE Journal, 49(5), 1250–1265.

Cheryan, M. (1998). Ultrafiltration and microfiltration handbook. Boca Raton: CRC press.Choloupla, K., & Malam, Y. (2010). Nanosilver as a new generation of nano products in

biomedical application. Trends in biotechnology, 11, 580–588.Chow, A. W. (2002). Lab-on-a-chip: Opportunities for chemical engineering. AIChE Journal, 48

(8), 1590–1595.

References 71

Christensen, T. S. (1996). Adiabatic prereforming of hydrocarbons—An important step in syngasproduction. Applied Catalysis, A: General, 138(2), 285–309.

Cohen, C. (1996). The early history of chemical engineering: A reassessment. The British Journalfor the History of Science, 29(02), 171–194.

Comini, E. (2006). Metal oxide nano-crystals for gas sensing. Analytica chimica acta, 568(1),28–40.

Dannels, D. P., Anson, C. M., Bullard, L., & Peretti, S. (2003). Challenges in learningcommunication skills in chemical engineering. Communication Education, 52(1), 50–56.

Dastjerdi, R., & Montazer, M. (2010). A review on the application of inorganic nano-structuredmaterials in the modification of textiles: Focus on anti-microbial properties. Colloids andSurfaces B: Biointerfaces, 79(1), 5–18.

Davis, G. E. (1901). A handbook of chemical engineering: Illustrated with working examples (Vol.1). Peoria: Davis Bros.

De Stefano, L., Maddalena, P., Moretti, L., Rea, I., Rendina, I., De Tommasi, E., et al. (2009).Nano-biosilica from marine diatoms: A brand new material for photonic applications.Superlattices and Microstructures, 46(1), 84–89.

Díaz, E., Ordóñez, S., & Vega, A. (2007). Adsorption of volatile organic compounds onto carbonnanotubes, carbon nanofibers, and high-surface-area graphites. Journal of Colloid andInterface Science, 305(1), 7–16.

Ding, J., Zhang, L., Zhang, Y., & Han, K. L. (2013). A reactive molecular dynamics study ofn-heptane pyrolysis at high temperature. The Journal of Physical Chemistry A, 117(16), 3266–3278.

Drioli, E., Stankiewicz, A. I., & Macedonio, F. (2011). Membrane engineering in processintensification—An overview. Journal of Membrane Science, 380(1), 1–8.

Duderstadt, J. J. (2010). Engineering for a changing world. In Holistic Engineering Education(pp. 17–35). New York: Springer.

Dvoretsky, D. S., Dvoretsky, S. I., Mishchenko, S. V., & Ostrovsky, G. M. (2010). Newapproaches to the integrated synthesis of flexible automated chemical engineering systems.Theoretical Foundations of Chemical Engineering, 44(1), 67–75.

Educational, N. U. (2007). Educating undergraduate Nanoengineers. Nanotechnology: SocietalImplications—Individual Perspectives, 305.

Edwards, M. F. (2006). Product engineering: Some challenges for chemical engineers. ChemicalEngineering Research and Design, 84(4), 255–260.

Endo, M., Strano, M. S., & Ajayan, P. M. (2008). Potential applications of carbon nanotubes. InCarbon nanotubes (pp. 13–61). Springer, Berlin Heidelberg.

Exchangers, M. H. (2000). Microreactors: New technology for modern chemistry. Weinheim:Wiley/VCH.

El-Halwagi, M. M. (2006). Process integration (Vol. 7). Waltham: Academic Press.Elnashaie, S. S. E. H., & Adris, A. (1989, May). A fluidized bed steam reformer for methane. In

Proceedings of the IV International Fluidization Conference, Banff, Canada.Elnashaie, S. E. H. S., & Grace, J. R. (2007). Complexity, bifurcation and chaos in natural and

man-made lumped and distributed systems. Chemical Engineering Science, 62(13), 3295–3325.

Elnashaie, S. S. E. H., Alhabdan, F. M., & Elshishini, S. S. (1993). The vital role of mathematicalmodelling in chemical engineering education. Mathematical and Computer Modelling, 17(3),3–11.

Elnashaie, S .E .H. S., Danafar, F., & Fakhru’l-Razi, A. (2013). Sustainable world throughsustainable materials and integrated biorefineries. Appllied Petrochemical Research, 3, 107–116.

Elnashaie, S. S., & Garhyan, P. (Eds.). (2003). Conservation equations and modeling of chemicaland biochemical processes. Boca Raton: CRC Press.

Elnashaie, S. S. E. H., & Elshishini, S. S. (1993). Modelling, simulation and optimization ofindustrial fixed bed catalytic reactors. London, U.K: Gordon and Breach Science Publishers.


Eyinagho, M. O. (2007). Design and implementation of a microprocessor-based sequencer for asmall-scale groundnut oil production plant. Advanced Materials Research, 18, 107–110.

Fair, J. R., & Seibert, A. F. (1996). Understand distillation-column debottlenecking options.Chemical Engineering Progress, 96, 42–48.

Fan, M., Huang, C. P., Bland, A. E., Wang, Z., Slimane, R., & Wright, I. G. (Eds.). (2010).Environ and technology. Amsterdam: Elsevier.

Favre, E., Marchal-Heusler, L., & Kind, M. (2002). Chemical product engineering: Research andeducational challenges. Chemical Engineering Research and Design, 80(1), 65–74.

Fisher, W. R., Doherty, M. F., & Douglas, J. M. (1985). Evaluating significant economic trade-offsfor process design and steady-state control optimization problems. AIChE Journal, 31, 1538–1547.

Gallagher, R., & Appenzeller, T. (1999). Complex systems. Science, 284(5411), 79–109.García-Serna, J., Pérez-Barrigón, L., & Cocero, M. J. (2007). New trends for design towards

sustainability in chemical engineering: Green engineering. Chemical Engineering Journal, 133(1), 7–30.

Gavriilidis, A., Angeli, P., Cao, E., Yeong, K. K., & Wan, Y. S. S. (2002). Technology andapplications of microengineered reactors. Chemical Engineering Research and Design, 80(1),3–30.

Grossmann, I. E. (2004). Challenges in the new millennium: Product discovery and design,enterprise and supply chain optimization, global life cycle assessment. Computers andChemical Engineering, 29(1), 29–39.

Gwehenberger, G., & Narodoslawsky, M. (2008). Sustainable processes—The challenge of the21st century for chemical engineering. Process Safety and Environmental Protection, 86(5),321–327.

Harmsen, G. J. (2004). Industrial best practices of conceptual process design. ChemicalEngineering and Processing: Process Intensification, 43(5), 671–675.

Harmsen, G. J., Korevaar, G., & Lemkowitz, S. M. (2004). Process intensification contributions tosustainable development. Chemical Industries (pp. 495–522). New York: Marcel Dekker.

Harmsen, J. (2010). Process intensification in the petrochemicals industry: Drivers and hurdles forcommercial implementation. Chemical Engineering and Processing, 49, 70–73.

Hayat, K., Gondal, M. A., Khaled, M. M., Ahmed, S., & Shemsi, A. M. (2011). Nano ZnOsynthesis by modified sol gel method and its application in heterogeneous photocatalyticremoval of phenol from water. Applied Catalysis, A: General, 393(1), 122–129.

Hessel, V., Hardt, S., & Löwe, H. (2006). Chemical micro process engineering: Fundamentals,modelling and reactions. Hoboken: Wiley.

Hoffmann, K. A., & Chiang, S. T. (2000). Computational fluid dynamics. Engineering educationsystem.

Huang, X., Gao, T., Pan, X., Wei, D., Lv, C., Qin, L., & Huang, Y. (2013). A review: Feasibilityof hydrogen generation from the reaction between aluminum and water for fuel cellapplications. Journal of Power Sources, 229, 133–140.

Huber, F. (2012). Do clusters really matter for innovation practices in information technology?Questioning the significance of technological knowledge spillovers. Journal of EconomicGeography, 12(1), 107–126.

Hyung, H., & Kim, J. H. (2008). Natural organic matter (NOM) adsorption to multi-walled carbonnanotubes: Effect of NOM characteristics and water quality parameters. Environmental Scienceand Technology, 42(12), 4416–4421.

Jajja, S. A., Ali, W., & Ali, H. M. (2013). Multiwalled carbon nanotube nanofluid for thermalmanagement of high heat generating computer processor. Heat Transfer—Asian Research.

Jahan, A., Ismail, M. Y., Shuib, S., Norfazidah, D., & Edwards, K. L. (2011). An aggregationtechnique for optimal decision-making in materials selection. Materials and Design, 32(10),4918–4924.

Jang, S. P., & Choi, S. U. (2004). Role of brownian motion in the enhanced thermal conductivityof nanofluids. Applied Physics Letters, 84(21), 4316–4318.

References 73

Jaworski, Z., & Zakrzewska, B. (2011). Towards multiscale modelling in product engineering.Computers and Chemical Engineering, 35, 434–445.

Joshi, J. G. (2013). Some important aspects to enhance the quality of the technical educationsystem for better industry-institute interaction. Strategic Role of Tertiary Education andTechnologies for Sustainable Competitive Advantage, 222.

Karri, R. R. (2011). Evaluating and estimating the complex dynamic phenomena in nonlinearchemical systems. International Journal of Chemical Reactor Engineering, 9(1).

Keshav, A., Chand, S., & Wasewar, K. L. (2009). Process intensification in hydrocarbonindustries: Methodology and applications. Recent trends in exploration, exploitation andprocessing of petroleum resources, 455.

Keulartz, J., Schermer, M., Korthals, M., & Swierstra, T. (2004). Ethics in a technological culture:A programmatic proposal for a pragmatist approach. Science Technology Human, 29(1), 3–29.

Kothare, M. V. (2006). Dynamics and control of integrated microchemical systems withapplication to micro-scale fuel processing. Computers and Chemical Engineering, 30(10),1725–1734.

Kwauk, M. (2005). Extending the knowledge base of chemical engineering. China Particuology, 3(3), 151–164.

Landes, D. S. (2003). The unbound prometheus: Technological change and industrialdevelopment in Western Europe from 1750 to the present. Cambridge: CambridgeUniversity Press.

Lavoisier, A. L. (1965). Elements of chemistry, in a new systematic order: Containing all themodern discoveries. Mineola: Courier Dover Publications.

Leong, K. Y., Mohd, H. N., Mohd, S. R., & Amer, N. H. (2014). The effect of surfactant onstability and thermal conductivity of carbon nanotube based nanofluids. Thermal Science, 00,78–78.

Levenspiel, O. (1999). Chemical reaction engineering, 3th edn, Hoboken: Wiley.Lewandowicz, G., Białas, W., Marczewski, B., & Szymanowska, D. (2011). Application of

membrane distillation for ethanol recovery during fuel ethanol production. Journal ofMembrane Science, 375(1), 212–219.

Li, C., & Lin, J. (2010). Rare earth fluoride nano-/microcrystals: Synthesis, surface modificationand application. Journal of Materials Chemistry, 20(33), 6831–6847.

Liao, Q., Sun, J., & Gao, L. (2008). The adsorption of resorcinol from water using multi-walledcarbon nanotubes. Colloids and Surfaces A: Physicochemical and Engineering Aspects,312(2), 160–165.

Lin, O. C. C. (2003). Formulation of a vision: Chemical engineering in the 21st Century. In R.C. Darton, R. G. H. Prince, and D. G. Wood. Chemical Engineering: Visions of the World.Amsterdam: Elsevier Science.

Lines, M. G. (2008). Nanomaterials for practical functional uses. Journal of Alloys andCompounds, 449(1), 242–245.

Litzen, D. B., & Bravo, J. L. (1999). Uncover low-cost debottlenecking opportunities. ChemicalEngineering Progress, 95(3), 25–32.

Liu, Y., & Kumar, S. (2014). Polymer/Carbon nanotube nano composite fibers–a review. ACSapplied materials & interfaces.

Long, R. Q., & Yang, R. T. (2001). Carbon nanotubes as superior sorbent for dioxin removal.Journal of the American Chemical Society, 123(9), 2058–2059.

Lu, C., Chung, Y. L., & Chang, K. F. (2005). Adsorption of trihalomethanes from water withcarbon nanotubes. Water Research, 39, 1183–1189.

Lu, C., & Su, F. (2007). Adsorption of natural organic matter by carbon nanotubes. SeparationPurification Technology, 58,113–121.

Mansourizadeh, A., & Ismail, A. F. (2009). Hollow fiber gas–liquid membrane contactors for acidgas capture: A review. Journal of Hazardous Materials, 171(1), 38–53.

Martin, R., Maytham, B., Case, J., & Fraser, D. (2005). Engineering graduates’ perceptions of howwell they were prepared for work in industry. European Journal of Engineering Education,30(2), 167–180.


May, G. S., & Chubin, D. E. (2003). A retrospective on undergraduate engineering success forunderrepresented minority students. Journal of Engineering Education, 92(1), 27–39.

Merta, H. (2006). Characteristic time series and operation region of the system of two tank reactors(CSTR) with variable division of recirculation stream. Chaos, Solitons and Fractals, 27(1),279–285.

Modashia, A. D., Karvekar, N. P., & Shah, K. P. (2000). Debottlenecking of methanol-waterdistillation column in ammonia plant. Fertilizer News, 45, 71–75.

Mohapatra, M., & Anand, S. (2010). Synthesis and applications of nano-structured ironoxides/hydroxides—A review. International Journal of Engineering, Science and Technology,2(8).

Müller-Goymann, C. C. (2004). Physicochemical characterization of colloidal drug deliverysystems such as reverse micelles, vesicles, liquid crystals and nanoparticles for topicaladministration. European Journal of Pharmaceutics and Biopharmaceutics, 58(2), 343–356.

Narodoslawsky, M. (2013). Chemical engineering in a sustainable economy. ChemicalEngineering Research and Design, 91, 2021–2028.

Nasiri, A., Rashidi, A., Shariaty-Niasar, M., & Soltanian, H. (2013). Preparation and application ofcarbon nanotube nanofluid as a reinforcement of cement slurry.

Nicole, L., Rozes, L., & Sanchez, C. (2010). Integrative approaches to hybrid multifunctionalmaterials: From multidisciplinary research to applied technologies. Advanced Materials, 22(29), 3208–3214.

Obot, I. B., Obi-Egbedi, N. O., Ebenso, E. E., Afolabi, A. S., & Oguzie, E. E. (2013).Experimental, quantum chemical calculations, and molecular dynamic simulations insight intothe corrosion inhibition properties of 2-(6-methylpyridin-2-yl) oxazolo [5, 4-f][1, 10]phenanthroline on mild steel. Research on Chemical Intermediates, 39(5), 1927–1948.

Oliveira, G. S., Leite, F. L., Amarante, A. M., Franca, E. F., Cunha, R. A., Briggs, J. M., & Freitas,L. C. (2013). Molecular modeling of enzyme attachment on AFM probes. Journal ofMolecular Graphics and Modelling, 45, 128–136.

Orr, V., Zhong, L., Moo-Young, M., & Chou, C. P. (2013). Recent advances in bioprocessingapplication of membrane chromatography. Biotechnology Advances, 31(4), 450–465.

Ottino, J. M. (2005). New tools, new outlooks, new opportunities. AIChE Journal, 51(7), 1839–1845.

Ottino, J. M. (2006). Granular matter as a wind collective systems far from equilibrium,complexity, and scientific prematurity. Chemical Engineering Science, 61, 4165–4171.

Oueiny, C., Berlioz, S., & Perrin, F. X. (2014). Carbon nanotube–polyaniline composites.Progress in Polymer Science, 39(4), 707–748.

Paladino, O., & Ratto, M. (2000). Robust stability and sensitivity of real controlled CSTRs.Chemical Engineering Science, 55(2), 321–330.

Pantelides, C. C., & Renfro, J. G. (2013). The online use of first-principles models in processoperations: Review, current status and future needs. Computers and Chemical Engineering, 51,136–148.

Pankhurst, Q. A., Connolly, J., Jones, S. K., & Dobson, J. (2003). Applications of magneticnanoparticles in biomedicine. Journal of physics D: Applied physics, 36(13), R167.

Paradise, M., & Goswami, T. (2007). Carbon nanotubes—Production and industrial applications.Materials Design, 28, 1477–1489.

Papazoglou, E. S. & Parthasarathy A. (2007). BioNanotechnology, Synthesis Lectures onBiomedical Engineering, Morgan & Claypool Publishers.

Paul, D. R., & Robeson, L. M. (2008). Polymer nanotechnology: Nanocomposite. Polymer, 49,3187–3204.

Perkins, J. D. (2003). Chemical engineering—The first 100 years. Chemical Engineering: Visionsof the World. Amsterdam: Elsevier Science.

Poel, I. D., & Goldberg, D. E. (2007). Abstracts of the Workshop Philosophy and Engineering,ISBN/EAN: 978-90-5638-183-7.

References 75

Prasad, P., & Elnashaie, S. S. E. H. (2002). Novel circulating fluidized bed membrane reformer forthe efficient production of ultraclean fuels from hydrocarbons. Industrial EngineeringChemistry Research, 41, 6518–6527.

Pratsinis, S. E., & Spicer, P. T. (1998). Competition between gas phase and surface oxidation ofTiCl4 during synthesis of TiO2 particles. Chemical Engineering Science, 53(10), 1861–1868.

Rao, C. N. R., Müller, A., & Cheetham, A. K. (Eds.). (2006). The chemistry of nanomaterials:Synthesis, properties and applications (Vol. 1). Hoboken: Wiley.

Rapoport, H., Lavie, R., & Kehat, E. (1994). Retrofit design of new units into an existing plant:Cast study: Adding new units to an aromatics plant. Computer and Chemical Engineering, 18(8), 743–753.

Ravanchi, M. T., & Kargari, A. (2009). New advances in membrane technology. INTECH OpenAccess Publisher.

Reay, D. (2008). The role of process intensification in cutting greenhouse gas emissions. AppliedThermal Engineering, 28(16), 2011–2019.

Riegel, E. R., & Kent, J. A. (2003). Riegel’s handbook of industrial chemistry. New York: PlenumPublisher.

Roberge, D. M., Ducry, L., Bieler, N., & Cretton, P. (2011). Microreactor technology: Arevolution for the fine chemical and pharmaceutical industries? Pharmaceutical industries?Newsletter.

Roco, M. C. (2004). Nanoscale science and engineering: Unifying and transforming tools. AIChEJournal, 50(5), 890–897.

Rosenberg, N. (1998). Chemical engineering as a general purpose technology. General PurposeTechnologies and Economic Growth, 167–192.

Rostrup-Nielsen, J. (1977). Hydrogen via steam reforming of naphtha. Chemical EngineeringProgress, 9, 87–92.

Rousseau, G. S., & Porter, R. (Eds.). (1980). The ferment of knowledge: Studies in thehistoriography of eighteenth-century science. Cambridge: Cambridge University Press.

Routh, H. B., Bhowmik, K. R., Parish, L. C., & Witkowski, J. A. (1996). Soaps: From thephoenicians to the 20th century-a historical review. Clinics in Dermatology, 14(1), 3–6.

Roy, P., Berger, S., & Schmuki, P. (2011). TiO2 nanotubes: Synthesis and applications.Angewandte Chemie International Edition, 50(13), 2904–2939.

Rubinstein R. Y., & Kroese, D. P. (2008). Simulation and the Monte Carlo Method. Hoboken:Wiley.

Rugarcia, A., Felder, R. M., Woods, D. R., & Stice, J. E. (2000). The future of engineeringeducation I. A vision for a new century. Chemical Engineering Education, 34(1), 16–25.

Sammels, A. F., Schwartz, M., Mackay, R. A., Barton, T. F., & Peterson, D. R. (2000). Catalyticmembrane reactors for spontaneous synthesis gas production. Catalysis Today, 56, 325–328.

Saremi, H. S., Abdi, N., & Kashani, M. A. (2000). Blending amines debottlenecks iranian gasplant. Oil and Gas Journal, 98, 60–61.

Scholz, M. S., Blanchfield, J. P., Bloom, L. D., Coburn, B. H., Elkington, M., Fuller, J. D., &Bond, I. P. (2011). The use of composite materials in modern orthopaedic medicine andprosthetic devices: A review. Composites Science and Technology, 71(16), 1791–1803.

Schummer, J. (2004). Multidisciplinarity, interdisciplinarity, and patterns of research collaborationin nanoscience and nanotechnology. Scientometrics, 59(3), 425–465.

Seely, B. E. (1999). The other re-engineering of engineering education, 1900–1965. Journal ofEngineering Education, 88, 285–294.

Seider, W. D., Seader, J. D., & Lewin, D. R. (2009). Product and process design principles:Synthesis, analysis and evaluation, (With CD). Hoboken: Wiley.

Sharifi, F. (2002). Chemical engineering: Professionally ignored?. In Proceedings of the 2002American Society for Engineering Education Annual Conference and Exposition.

Sharma, A., Tomar, M., & Gupta, V. (2013). Enhanced response characteristics of SnO2 thin filmbased NO2 gas sensor integrated with nanoscaled metal oxide clusters. Sensors and ActuatorsB: Chemical, 181, 735–742.


Shih, Y. H., & Li, M. S. (2008). Adsorption of selected volatile organic vapors on multiwallcarbon nanotubes. Journal Hazard Materials, 154, 21–28.

Siegfried, R. (2002). From elements to atoms: A history of chemical composition. Transactions ofthe American Philosophical Society, i–278.

Singh, M., Manikandan, S., & Kumaraguru, A. K. (2011). Nanoparticles: A new technology withwide applications. Research Journal of Nanoscience and Nanotechnology, 1(1), 1–11.

Smith, B. V., & Ierapepritou, M. G. (2010). Integrative chemical product design strategies:Reflecting industry trends and challenges. Computers and Chemical Engineering, 34(6), 857–865.

Stankiewicz, A. I., & Moulijn, J. A. (2000). Process intensification: Transforming chemicalengineering. Chemical Engineering Progress, 96(1), 22–34.

Stankiewicz, A., & Moulijn, J. A. (2003). Re-engineering the chemical processing plant: Processintensification. Boca Raton: CRC Press.

Steffensen, M., Rogers, E. M., & Speakman, K. (2000). Spin-offs from research centers at aresearch university. Journal of Business Venturing, 15(1), 93–111.

Stephanopoulos, N., Solis, E. O., & Stephanopoulos, G. (2005). Nanoscale process systemsengineering: Toward molecular factories, synthetic cells, and adaptive devices. AIChE Journal,51(7), 1858–1869.

Summer, D. R., McGuire, P. J., Resetarits, M. R., Graves, E. C., Harper, S. E., & Angelino, S. J.(1995). High-capacity trays debottleneck Texas C3 splitter. Oil and Gas Journal, 93, 45–50.

Tan, J., Foo, D. C. Y., Kumaresan, S., & Abdul-Aziz, R. (2006). Debottlenecking of a batchpharmaceutical cream production. Pharmaceutical Engineering, 26(4), 72–84.

Van Wyllan. (1987). Classical thermodynamics.Wang, S. G., Liu, X. W., Gong, W. X., Nie, W., Gao, B. Y., & Yue, Q. Y. (2007). Adsorption of

fulvic acids from aqueous solutions by carbon nanotubes. Journal of Chemical TechnologyBiotechnology, 82, 698–704.

Wang, Y., Gong, C., Sun, J., Gao, H., Zheng, S., & Xu, S. (2010). Separation of ethanol/waterazeotrope using compound starch-based adsorbents. Bioresource technology, 101(15), 6170–6176.

Westmoreland, P. R. (2008). Chemical engineering the next 25 years. Chemical EngineeringProgress, 104(11), 31–41.

Williams, J. M. (2000). Transformations in technical communication pedagogy: Engineering,writing, and the ABET engineering criteria. Technology & Teamwork, 0-7803-6431-7/00/.

Winkelman, P. (2009). Perceptions of mathematics in engineering. European Journal ofEngineering Education, 34(4), 305–316.

Yang, S. Y., Cheng, F. Y., Yeh, C. S., & Lee, G. B. (2010). Size-controlled synthesis of goldnanoparticles using a micro-mixing system. Microfluidics and Nanofluidics, 8(3), 303–311.

Ye, C., Gong, Q. M., Lu, F. P., & Liang, J. (2007). Adsorption of uraemic toxins on carbonnanotubes. Separation and Purification Technology, 58, 2–6.

Yore, L. D. (2011). Foundations of scientific, mathematical, and technological literacies—Common themes and theoretical frameworks. In Pacific CRYSTAL Center for Science,Mathematics, and Technology Literacy: Lessons learned (pp. 23–44). Sense Publishers.

Yuan, Z., Chen, B., Sin, G., & Gani, R. (2012). State-of-the-art and progress in theoptimization-based simultaneous design and control for chemical processes. AIChE Journal,58(6), 1640–1659.

Zhao, P., Li, N., & Astruc, D. (2013). State of the art in gold nanoparticle synthesis. CoordinationChemistry Reviews, 257(3), 638–665.

Zhang, D., & Wang, Y. (2006). Synthesis and applications of one-dimensional nano-structuredpolyaniline: An overview. Materials Science and Engineering B, 134(1), 9–19.

Zhang, J., Zhu, X. X., & Towler, G. P. (2001). A level-by-level debottlenecking approach inrefinery operation. Industrial Engineering Chemical Research, 40, 1528–1540.

Zhang, Q., Liu, C., & Zhang, X. (2012). Complexity, analysis and control of singular biologicalsystems (Vol. 421). London: Springer.

References 77

Chapter 2From Nanotechnology to Nanoengineering

2.1 Introduction to Nanotechnology

The world is presently witnessing the advancement and development of a newmultidisciplinary technology, “Nanotechnology.” The concepts that seeded nano-technology were first discussed in 1959 by renowned physicist Richard Feynman inhis talk (Feynman 1961):

There’s Plenty of Room at the Bottom

In those early days, nanotechnology referred to the possibility of synthesis viadirect manipulation of atoms and molecules for fabrication of macroscale products.In the 1990, K. Eric Drexler and M. Minsky used the word “nanotechnology” intheir book “Engines of Creation: The Coming Era of Nanotechnology” (Drexler andMinsky 1990), in which they proposed assembling machines and devices on thescale of molecules, a few nanometers wide. Later on, as nanotechnology became anaccepted concept, the meaning of the word shifted to encompass technologiesrelated to making any type of materials, structures and devices in nanometer scale.A nanometer (nm) is one-billionth of a meter, hundred-thousandth the width of ahuman hair. There is a multidisciplinary convergence of science dedicated to thestudy of a world in such small scale. The US National Nanotechnology Initiative(Roco 2011) has described four generations of nanotechnology development(Fig. 2.1). The first era is a design of passive nanostructures and materials toperform just one task like nanostructured metals, aerosol. The second phaseintroduced active nanostructures for multitasking, for example, actuators, drugdelivery devices and sensors. The third generation featured nanosystems withthousands of interacting components. In this era, integrated nanosystems, hierar-chical systems within systems, have been developed.

Accordingly, a comprehensive definition for Nanotechnology is:

Nanotechnology is art and science of manipulating atoms and molecules to create systems,materials and devices at nanoscale as well as their application in various fields.


79

Nanotechnology can be referred to as a general-purpose technology, as it hassignificant impacts on almost all industries and all areas of society. Nanotechnologyis expected to offer better built, longer-lasting, cleaner, safer and smarter productsfor the home, for communications, for medicine, for transportation, for agricultureand for industry in general. Chemistry and materials science and in some casesbiology are integrated to create new properties of materials in nanoscale. However,engineering principles must be exploited to gain market opportunities.

2.1.1 Application of Nanotechnology in Different Fields

The expectations from nanotechnology as a key technology of the current centuryfor innovative products and new market potentials are high. Some of these potentialapplications of nanotechnology-based products are presented in this section.

2.1.1.1 Nanotechnology in Biotechnology

The size of nanometer is of central importance in the nature and biological systems.Cells are the main element of the living organisms that can be much smaller thanthe submicron size domain. A wide range of biomolecules, like proteins and

Fig. 2.1 Generations of nanotechnology development (Roco 2011)

80 2 From Nanotechnology to Nanoengineering

viruses, is in the nanoscale range (Fig. 2.2). Accordingly, nanotechnology hasbecome a part of the world of biotechnology for understanding the biologicalsystems and their phenomena (Papazoglou and Parthasarathy 2007). Unexpectedgrowth in the field of nanotechnology and biotechnology has brought novel tech-nologies that make it possible to:

• Design new nanostructures by mimicking the structure and function of livingsystems

• Control and alter the biosystems including cellular and subcellular organelles,protein molecules, receptors and cytokines

Proteins, with a typical size of 5 nm, are comparable to the smallest nanopar-ticles that have been made by researchers. This circumstance has led to theimprovement in the health care and medical research through the development ofnanobased products, as a result of research efforts. These products have extensivelyfound bioapplications in the fields of cosmetics, drug delivery, imaging and medicaldiagnosis, tissue engineering, etc. For example, biomolecules can be attached tonanoparticles by electrostatic forces including hydrogen bonding, hydrophobicforces and van der Waals forces. New approaches of using nanostructures forbiomolecule immobilization assist potential applications in biosensors (glucosesensor, DNA sensors), targeted drug delivery and other biocatalytic processes (Kimand Grate 2003; Kim et al. 2006a, b). The optical properties of nanoparticles couldbe also implemented in imaging and characterization of biomolecules, for example,as a marker in TEM, as well as surface enhancement of fluorescent emission andRaman scattering (Wang 2000a, b).

The revolutionary advancements in bionanotechnology and biomedical researchplace a strong foundation for a customized, personalized and quantitative medicinein the future. The vital role of nanobiotechnology in medicine is improvement indiagnostic technologies. Diagnosis is a key stage in health care; the earlier thediagnosis of a disease the more effective the therapy is, both from outcomes andfrom a total cost perspective. The integration of nanotechnology with medicine hasgiven birth to a new field of science called “nanomedicine.” The ultimate goal ofnanomedicine is to develop well-engineered products (tools and materials) thatcould efficiently be used for the prevention, diagnosis and treatment of differentdiseases. These products include drug delivery platforms, imaging systems,

Fig. 2.2 Various nanosized biological systems

2.1 Introduction to Nanotechnology 81

biochips and probes, needles for painless drug infusion or intracellular injections,etc. (Papazoglou and Parthasarathy 2007).

Nanostructures are an important component of biosensing platforms (Bianco andPrato 2003; Davis et al. 2003). The extraordinary properties of nanostructures(specially electrical and optical properties) in the presence of biomolecules makethem possible to be used in biosensors fabrication (Vo-Dinh 2004; Hong et al.2010; Duncan 2011). Biosensors, which act in the aqueous phase, are analyticaldevices incorporating biomolecule (e.g., DNA, enzymes, antibodies, microorgan-isms, etc.) associated with or integrated within a physico-chemical transducer(Vo-Dinh 2004). Biosensors are now employed in a wide range of applications:

• Detecting of diseases, particularly in cancer diagnostics (Fortina et al. 2007;Perfézou et al. 2012)These sensors capable of monitoring in vivo processes within living cells,leading to achieve new information on the inner workings of the entire cell andunderstanding the cellular function, thus revolutionizing cell biology and todifferentiate normal and abnormal cells (Vo-Dinh 2004).

• Organo-phosphorus pesticides and nerve gases (Gan et al. 2010; Liang et al.2012)

• Pathogens and toxins, or total cholesterol and glucose, etc., in blood (Lahiffet al. 2010; Kuila et al. 2011)

• Low concentration of toxic gases (Endo et al. 2008; Asefa et al. 2009).

Nanostructures have been also considered in the controlled release of activecompounds in the application of pharmaceuticals (Kumar 2000) and food additives(Chaudhry et al. 2008). Drug delivery using nanostructures has demonstrated highand versatile loading capacity for therapeutic agents, some selective cargounloading and better cell penetration than many other delivery materials (Farokhzadand Langer 2009; Verma and Stellacci 2010). To achieve the efficient performance,a suitable scheme to conjugate the drug and the nanostructure is required to makenanostructures into viable delivery vehicles (Endo et al. 2008). Another critical stepis to conjugate the nanostructures in such a way that the functionality of the bio-molecules is maintained. The present-day nanomedicine initiatives include a rangeof successful and evolving technologies encompassing targeted drug delivery aimedat minimizing side effects, creation of implantable materials as scaffolds for tissueengineering, development of implantable devices, surface modification anddesigning optimal topology for biomaterial implants, surgical aids, nanorobotics, aswell as high-throughput drug screening and medical diagnostic imaging (Pankhurstet al. 2003; Papazoglou and Parthasarathy 2007). Medical applications of the mostused nanostructures are presented in Fig. 2.3. As presented in the previous para-graph and indicated in the Fig. 2.3, medical applications of nanostructures com-monly include imaging, sensing and delivery. However, magnetic nanoparticles(Fe3O4 and γ-Fe2O3) due to their unique behavior have a specific application forbiomolecule separation selection. This application is especially well suited to the


separation of rare tumor cells from blood and low numbers of target cells(Pankhurst et al. 2003; McCarthy and Weissleder 2008).

Nanostructures have also attracted great attention in tissue engineering fordesign of medical prostheses or implants (Scholz et al. 2011). Designing medicalimplants or scaffolds similar to natural tissue or organs with respect to their per-formance is a challenging task facing materials scientists and engineers. Novelinnovative technology for the fabrication of nanostructural biomaterials increasesthe possibility to use nanostructure in designing and producing implants or pros-theses with all ideal characteristic features in order to function in a natural way inthe body environment (Dastjerdi and Montazer 2010). Nanocomposites of differenttypes of ceramics are mostly used as orthopedic implants and in dental applicationsto repair and replace diseased and damaged parts of the musculoskeletal system(Chevalier and Gremillard 2009). Metallic-based materials that present sufficientfatigue strength to endure the rigors of daily activity, such as walking and chewingare normally used as pins and plates and femoral stems, orthopedic implants, etc.(Minagar et al. 2012). Coating materials in which nanostructures are used canmodify the surface properties of materials used in medical activities (for example,surgery) by their improvements in performance, reliability and biocompatibility(Ben-Nissan and Choi 2006).

Nanostructures are also considered for the food packaging sector, and nano-composites have shown impressive characteristic for this purpose (Sanchez-Garciaet al. 2010; Lagaron and Lopez-Rubio 2011; Duncan 2011). According to thesereviews, the majority of the researches used clay nanoparticles; nevertheless, other

Fig. 2.3 Medicalapplications


types of reinforcing elements like carbon nanotubes (CNTs), metals, nanoparticlesand biobased nanofillers such as starch nanoparticles and biodegradable cellulosenanowhiskers have been applied. The latter biobased nanofillers offer moreadvantages like generating fully biobased formulations and edibility since they canbe made of food hydrocolloids (Le Corre et al. 2010).

The usefulness of using nanostructures in food industry as presented in (Lagaronand Lopez-Rubio 2011; Gemili et al. 2010; Mastromatteo et al. 2010) includes:

• Controlled and/or burst release of substances in active and functional foodpackaging technologies and intelligent food packaging.

• Formulation of active packaging technologies based on bioplastics such as moreefficient antioxidant, oxygen scavenging or antimicrobial biopackaging, whichhas more direct implications in increasing packaged foods quality and safety.

• Important issues associated with the use of bioplastics, such as the non-intendedmigration of plastic components to foods, can also be potentially reduced by theuse of nanoparticles.

While significant progress has been made in the area of nanobiotechnology,there are still many issues left to resolve before clinical use of these materials can berealized. Before nanostructures can be utilized for human body, several importantproperties need to achieve. The nanostructures must be nontoxic,non-immunogenic, stable in blood, biodegradable and applicable to various bio-molecules. From commercial and engineering points of view, the nanostructuresalso must possess scalable and inexpensive manufacturing process. Technical,regulatory and legal challenges, however, exist along the road to implementnanotechnology in the fields of medical pharmaceutical and food industry. A greatchallenge is realizing the long-term impact of nanostructures on human health andtheir interactions (Papazoglou and Parthasarathy 2007). The other issue is lack ofknowledge about the stability of nanostructures during processing. However, theuse of natural additives like clay nanoparticles and nanocellulose may help toovercome this problem because of their biocompatibility.

2.1.1.2 Nanotechnology in Petroleum Industries

Nanotechnology is offering new and improved methods in different areas of the oiland gas industries from exploration and well drilling to refining and distribution.Properties of nanostructures such as lightness, corrosion resistance and mechanicalstrength make them significant elements to be used in the oil industry machines,specially drilling machines (Singh et al. 2010). Nanotechnology represents break-through elements, thanks to the development of innovative monitoring techniquesand smarter nanosensors. Conventional sensors and other measuring tools areunreliable in hostile high-temperature and high-pressure conditions. Improvedperformance sensors for imaging, measuring and controlling reservoirs and oilfieldswill improve all activities in the area of oil industry, from exploring oil well anddrilling to oil transporting and reserving (Matteo et al. 2012). Besides the potential


applications enumerated for nanotechnology in oil and gas industry, nanostructurescan provide extraordinary opportunities to develop more cost-effective and envi-ronmentally friendly upgrading and recovery enhancement of heavy feedstocks(Krishnamoorti 2006; Nassar et al. 2011; Almao 2012). Nanostructures were usedfor reduction in the viscosity of heavy oil, thus benefiting the oil exploitation andupgrading transportation. Using nanostructures in catalysts provides severaladvantages that ultimately increase the economics of the upgrading process. Thecatalyst-improved characteristics include (Hashemi et al. 2014):

(a) High surface area-to-volume ratio, which results in improved catalytic per-formance for processing purposes

(b) Increased probability of contacts between reactants because of their highmobilization inside the reactor

(c) Long run times for conversion as there is no need of catalyst replacementbecause of nanocatalysts implementation inside the medium

(d) Stable long-term high activity

Another promising advantage is possibility of in situ preparation of catalyst.Figure 2.4 illustrates the in situ preparation of nanocatalyst for upgrading andrecovery of bitumen during the steam-assisted gravity drainage (SAGD) process(Nassar et al. 2011). In this method, nanocatalysts are introduced into the porous

Fig. 2.4 In situ heavy oil upgrading and recovery (Nassar et al. 2011)


media through a line to perform upgrading inside the reservoir to convert bitumento lighter products. The ultra-dispersed nanocatalysts will aid the SAGD process,and light oil is produced at the surface, whereas heavy molecules, solids andminerals stay subsurface. Successful in situ processing will reduce the operatingcosts as well as environmental concerns; however, there is a long way and extensiveworks are needed to commercialize the proposed method (Hashemi et al. 2014).

Other emerging applications of nanotechnology in oil industry include devel-opment of:

• New generation membranes for gas separation.• New types of smart fluids for water shutoff and improved/enhanced oil recovery

(Matteo et al. 2012).• Nanocatalyst (Pour et al. 2010; Gharibi et al. 2012).• Unconventional petroleum exploration and exploitation (Caineng et al. 2012).

2.1.1.3 Nanotechnology in Material Science

Nanocomposites are broadly defined as nanofillers bonded to a matrix (Paul andRobeson 2008). Nanocomposites of ceramic, metallic and polymer matrixes haveshown outstanding properties in comparison with composites of the same materialbut using microstructures (Gao et al. 2007; Thostenson et al. 2005; Esawi and Farag2007; Pezzin et al. 2011). The benefits encompass improved mechanical properties,scratch resistance, barrier properties, fire resistance and dimensional stability.Moreover, a small amount of nanofillers can cut weight and reduce cost comparedwith the usual loading of conventional fillers (Thostenson et al. 2005).Nanocomposites can be classified as structural or functional depending on the roleof the nanofiller in each situation (Pezzin et al. 2011). For structural composites, themechanical properties of the nanoparticle, such as high Young’s modulus, tensilestrength and elongation at break, and the ability to resist compression and distortioncan be used to produce lightweight structural materials. On the other hand, infunctional composites, other interesting properties are exploited, such as highelectrical and thermal conductivity, required in the development of thermal resistantmaterials, sensors, electrical conductors, photoemitters, electromagnetic shields orenergy accumulators. Polymer nanocomposites enable substantial improvements inmaterial properties such as shear and bulk modulus, yield strength, toughness,scratch resistance, optical properties, electrical conductivity, gas and solventtransport, with a small loading of nanostructure dispersed in the polymer matrix(Paul and Robeson 2008). Nanostructures also increase the temperature at which thepolymer will start to acquire a softened state, and hence, they have been extensivelyused in synthesis of thermoplastic polymers. Using natural nanostructures todevelop thermoplastic polymers is a promising route toward producing sustainableproducts (Cyras et al. 2008; Schlemmer et al. 2010; Aouada et al. 2011).Nanocellulose, nanostructured minerals (like clay hallo site nanotubes, modifiedbentonites and montmorillonites), organic–inorganic hybrid nanomaterials and


polymer matrix clay-reinforced nanocomposites are some of the examples ofnanostructured natural materials. These natural nanostructures have been alsoexamined for other applications. For example, nanoceramic coatings have beendeveloped that can be applied in household appliances or automobiles instead ofcoatings made from hazardous materials like chromium and toxic heavy metals(Nentwich and Greßler 2012). A small amount of nanoclay (less than 5 wt%)indicated a great enhancement in polymer properties regarding the mechanical andthermal resistance (Powell and Beall 2007; Choudalakis and Gotsis 2009;Meneghetti and Qutubuddin 2006; Du et al. 2010). Nanocellulose is an anotherexample of natural nanostructures that have wide applications in producingmoldable lightweight and high-strength materials, medical implants, electrodes forfuel cells, barrier film for packaging applications (keeping oxygen from spoilingfood), composites for construction, vehicles and furniture (Klemm et al. 2009; Siróand Plackett 2010). Cellulose is abundant, has high strength and stiffness, lowweight and biodegrade ability.

2.1.1.4 Nanotechnology in Environmental Science

Nanotechnology innovations have also raised great applications in the environ-mental sector. Nanostructure-based materials are aimed to improve the environmentthrough direct applications in detecting and removing pollutants from soil andgroundwater. Nanosensors capable of detecting a low concentration of toxic gasesare imperative for environmental monitoring and chemical safety as well as controlof chemical processes and agriculture. Nanoadsorbents and nanomembranes indi-cated their effective role in water and air purification as well as wastewater treat-ment by removal of various types of pollutants including heavy metals, syntheticdyes and biological contaminants (Lu et al. 2005; Yang et al. 2007; Liao et al. 2008;Savage and Diallo 2005; Kwon et al. 2008; Kuo 2009; Diallo 2009; Li et al. 2013;Hu et al. 2009; Mishra et al. 2010; Yao et al. 2010; Bora and Dutta 2014).Nanostructures act as an adsorbent or photocatalyst and have indicated high sen-sitivity, selectivity and efficiency for removal of these contaminants. Commonnanostructures with potential applications for adsorbing pollutants include dendri-mers, zeolites and CNTs (Diallo 2009). Different structures of these materials aredepicted in Fig. 2.5.

There are also some nanostructures with photocatalytic activity, like Zinc Oxide,applied for the treatment of environmental pollution (Sung et al. 2010). Thenanomembranes, nanoadsorbents and nanostructures with photocatalytic activitycan be used to purify indoor air volumes or to separate out contaminates inautomobile tailpipes and factory smokestacks and prevent these contaminantsentering the atmosphere.

Another excellent application of nanotechnology in the water sector is for waterdesalination. As the fresh water resources become increasingly scarce due tooverconsumption and contamination, scientists have begun to consider seawater as


another source for drinking water. However, the seawater has too much salt forhuman consumption, and desalination is required for removing the salt to createnew sources of drinking water. Desalination is an expensive method, and nano-membranes have the potential to reduce its cost (Mansoori et al. 2008; Kar et al.2012).

Besides their role in environment remediation, nanostructure-based products,based on their special properties, have the potential to make products or productionprocesses more environmentally friendly (Yuan 2004). Although positive envi-ronmental effects are rarely the reason for using nanobased products, such aninfluence is an acknowledged side effect. They contribute to environmental andclimate protection by saving the raw materials and energy sources. For example,using nanocomposites increases the durability of the products against mechanicaland thermal stresses or weathering and thus increases the useful life of a product. Itwas indicated that adding nanostructures, like nanosilica or carbon nanofiber ortubes, to tires reduce rolling resistance, which leads to fuel savings (Sun et al.2012a, b; De et al. 2013; Chandra and Bhandari 2013). Nanocomposites also canreduce weight of the vehicle and hence fuel consumption that save energy duringtransport. Nanocoatings, which are resistant to dirt or easy-to-clean, do not need tobe cleaned so often and hence help energy and water saving in facility cleaning.Novel insulation materials in the constructions, in which nanostructures have beenused, improve the energy efficiency of buildings and reduce the energy needed toheat and cool buildings.

One of the outstanding roles of nanotechnology is improving energy andresource efficiency in the chemical industry. The explosion and developments innanotechnology have exhibited significant impacts on the understanding, practiceand applications of catalysis. Nanocatalysts can be used to increase the yield ofchemical reactions and reduce the amount of environmentally damaging sideproducts. Catalysis provides controls over the rates at which chemical bonds are

Fig. 2.5 Typical structures ofnanomaterials with potentialapplications for adsorption:a dendrimer, b fullerene,c zeolite, d carbon nanotube


broken and formed. Therefore, it is the key to energy conversion and environmentalprotection in chemical manufacturing and transportation. In the chemical industry,nanocatalysts with improved properties boost energy and resource efficiency. Highhopes are placed in nanotechnologically optimized products and processes forenergy production and storage. Novel lighting materials with nanoscale layers ofplastic and organic pigments enhance conversion rate from energy to light (Changet al. 2013).

2.1.1.5 Nanotechnology in the Energy Sector

The world demand for energy is expected to become about 30 terawatts by the year2050. Compounding this challenge is to protect our environment by increasingenergy efficiency and developing clean energy sources. Solutions require scientificbreakthroughs and truly revolutionary developments. Within this context, nano-technology presents exciting and requisite approaches for addressing these chal-lenges. Those areas that nanotechnology helps to improve efficiency of energysources have been described as its positive effect in environment (previous section).Another application of nanostructure in the energy sector is for energy production,distribution and storage. For example, electrodes comprising CNTs have been usedto produce high-power lithium batteries, solar cells, fuel cells and several otherelectrochemical applications (Lee et al. 2010; Lota et al. 2011; Liu et al. 2012;Zhang and Dai 2012). Lithium-ion batteries (LIB) consisting of nanostructures alsoindicated improved storage capacity as well as an increased lifespan (Landi et al.2009). LIB (Fig. 2.6) have several applications, ranging from portable electronics toelectric vehicles, due to their superior energy density over other rechargeable bat-tery technologies. Using nanostructure also offers LIB of smaller size and lowerweight that may attract more attentions. Another example of positive effect of

Fig. 2.6 Solar cell and lithium-ion batteries (LIB)


nanostructure is their application in solar cells (Fig. 2.6). For example, solar cellswith nanoscale semiconductor materials have been developed by mimicking naturalphotosynthesis in green plants (Anandan et al. 2006). Solar energy is nowadaysconsidered as a promising renewable source of energy. The benefits of nanotech-nology in solar cells include reducing costs of materials, processing and installationand achieve reaching higher efficiency levels compared to traditional ones.

A fuel cell is an electrochemical device consisting of an electrolyte, an anodeand a cathode which directly and continuously converts the chemical energy of afuel into electrical energy (Hoogers 2002). Fuel cells may replace fossil fuels topower automobiles and reduce our reliance on petroleum; however, their efficiencystill needs to be improved. The efficiency of fuel cells is determined by the rate ofelectron transfer at electrodes. The main factors that determine efficiency of a fuelcell involve the catalysts and electrolyte used and the operating temperature, sincethey influence the reaction kinetics and the ionic transportation in the cell(Kirubakaran et al. 2009). The performance of fuel cells with nanostructures hasbeen found to be superior to other common electrodes in terms of their efficiencyand reversibility. Nanotechnology researches offer more efficient catalyst in fuel cellthrough specific design of catalyst.

Hydrogen is the most abundant element in the universe, and its molecule canstore a great amount of chemical energy, which can be used to generate electricity.This is the base of developing hydrogen fuel cells, in which chemical oxidation ofhydrogen to water produces electricity (Fig. 2.7). Hydrogen fuel cells can be used topower the vehicle or power plants, in place of fossil fuels. A great priority of

Fig. 2.7 Hydrogen fuel cell(Source EnLIST ChemistryWorkshop, University ofIllinois, 2010)


hydrogen fuel cells is that they do not contribute to air pollution. The first step indeveloping fuel cells is to enhance the hydrogen storage. Hydrogen can be com-pressed and reversibly stored in tanks through chemical or physical bonding ofhydrogen with storage material. Nanotechnology has offered great possibility inimproving storage materials; high porous materials able to adsorb hydrogen in theirporous or complex hybrids reversibly bound with hydrogen in their lattice structure(Sakintuna et al. 2007; Zubizarreta et al. 2009). Carbonaceous structures (activatedcarbon, carbon nanofiber and CNTs) are leading adsorbent candidates for gasesincluding hydrogen (Lee et al. 2000; Panella et al. 2005; Zubizarreta et al. 2009;Ioannatos and Verykios 2010). The properties that make carbon materials attractiveinclude their morphology, in the form of a fine powder, with high porosity and theexistence of specific interactions between the carbon atoms and the gas molecules.

Supercapacitors are energy-storage devices that store energy directly andphysically as charge, whereas batteries store energy in chemical reactants capable ofgenerating charge (Izadi-Najafabadi et al. 2010). They have application insmall-scale energy-storage devices in stationary electronics, such as memorybackup devices and solar batteries with semipermanent charge–discharge cycle life(Jurewicz et al. 2001). Supercapacitors compared to the other energy-storagedevices are able to store and deliver energy rapidly and efficiently for a long lifecycle via a simple charge separation process. In addition, their wide range of powercapability makes it possible to hybridize them with other energy-storage devices,such as batteries and fuel cells. The performance of supercapacitors have beenfurther improved by using nanostructures (Frackowiak et al. 2006; Cheng et al.2011; Kong et al. 2013; Hahm et al. 2012; Kim et al. 2012). It is worth mentioninghere that carbon nanostructures compared to the other types of nanostructures havebeen preferred to be used in the electrode materials. Conventional carbon materialshave been extensively used in energy-storage systems due to their good chemicalstability and high electrical conductivity. Carbon nanostructures, besides theadvantages mentioned for conventional carbon materials, have unique electrical andelectronic properties, a wide electrochemical stability window and a highlyaccessible surface area (Kim et al. 2012).

2.1.1.6 Nanotechnology in Other Specific Fields

Different nanostructures (nanoparticles, nanotubes, nanocrystals) contribute to thefabrication of gas sensors (Comini 2006; Lupan et al. 2010; Sun et al. 2012a, b;Moloney and Barrera 2013; Chow et al. 2013; Benkstein et al. 2014). The mainadvantage of these sensors is the nanoscopic size of the sensing element and thecorresponding nanoscopic size of the material required for a response. In addition,the mechanical robustness of the sensing elements and its low buckling forceincrease the sensor lifetime. As an example, the electrical resistivity of single-wallnanotubes has been found to change sensitively on exposure to gaseous ambientcontaining NO2, NH3 and O2. Nanosensors’ response is at least an order of mag-nitude faster than those currently available, and they could be operated at room


temperature or at higher temperatures for sensing applications (Lupan et al. 2010;Chow et al. 2013; Sharma et al. 2013).

The correspondence between mechanical response and electronic transport hasbeen proven potential applications of nanostructures in such applications asnano-electro-mechanical sensors (NEMs). Besides their wide application in lifescience, NEMs are now being researched for their use in automotive industries(Esashi 2009; Hema 2013). A tiny sensor would be able to monitor and report tirepressure to the driver while being able to withstand extreme temperatures andvibrations. NEMs also provide opportunities for smart airport pavement instru-mentation and health monitoring by long-term, continuous, real-time responsemeasurement of transportation infrastructure systems (Yang et al. 2014).

Nanostructures can be also applied in scanning probes of high-resolutionimaging instruments, such as scanning tunneling microscope (STM), atomic forcemicroscope (AFM) and electrostatic force microscopes. More detailed explanationson these instruments can be found in Sect. 2.1.3. They can be also used for surfacemanipulation. For example, on an AFM tip, they can be controlled like tweezers topick up and release nanoscale structures (Takekawa et al. 2005). Nanotweezers alsohave great application in life science (Hashiguchi et al. 2003; Roxworthy andToussaint 2013). Nanostructures are also considered for improving the catalystsperformance, either as active element or as support (Akia et al. 2014; Shen andYoshikawa 2013).

Applications of nanotechnology-based products are not limited to thoseaddressed in this section, and some unique applications are being developed.A unique role of nanostructure is in the area of controlled release of an element.Besides sustained release of drugs, food additives and fragrances, explained in thefield of biotechnology application, they are applied for controlled release of anti-corrosion agents as well as sustained release of herbicides, insecticides, fungicidesand antimicrobials. Another imperative application of nanostructures is in devel-oping antibacterial products. Silver nanoparticles have been extensively used inboth antibacterial and biocidal fabric and agent. These products have biomedicalapplications (YeonáLee et al. 2007; Choloupla and Malam 2010) and can beapplied as preservatives in wood, paints, etc. (Nowack et al. 2011). Nanoparticulatetitanium dioxide (TiO2) is another nanostructure used in synthesis of antimicrobialproducts (Chung et al. 2008). This nanostructure has found more applications suchas in synthesis of flame retardants (Chen and Wang 2010; Kiliaris and Papaspyrides2010), mineral UV filter in sunscreens (Popov et al. 2005; Sadrieh et al. 2010) andself-cleaning products (tiles, windows and textiles) like antifogging car mirrors anddental mirror surfaces (Funakoshi and Nonami 2007; Veronovski et al. 2009).Another emerging application of nanostructure is in the field of refractory tech-nologies, for example, in the steel and cement industries (Kuznetsov et al. 2010;Antonovič et al. 2010).

Despite vast application of nanostructures, there are some uncertainty about theirside effects on human health and environment. Accordingly, a great attempt towarddeveloping healthier and more efficient and sustainable nanotechnology-basedproducts is conducting. In this respect, natural and modified natural nanostructures


are preferred. Their applications are as wide as refractories, textiles, energy, bio-medicals, functional barriers and environmental fields. The most common naturalnanostructures are nanoclays (hallo site and kaolin clays), montmorillonite, nano-cellulose and nanostarch.

2.1.2 Nanostructured Materials Synthesis, Conceptsand Design

Nanostructures defined as materials with at least one dimension of their structure inthe nanometer scale. Nanostructures possess new and unique chemical and physicalproperties compared to their corresponding bulk or isolated atoms and molecules.Nanostructures have a limited number of atoms (or molecules) in which theirarrangement can be controlled during synthesis. Therefore, their chemical,mechanical, optical, electronic and magnetic properties of nanostructures can besignificantly altered. For example, the color or absorption spectrum changes dra-matically with size when the size is small compared to the de Broglie wavelength orthe Bohr excitation radius of the electron (Cao 2004). When a metal particle such asgold is smaller than 10 nm, it essentially exists in a state that is neither liquid norsolid. When a common liquid such as water is confined to a space that is only a fewnanometers in dimension (for example, when water flows in a nanochannel), itsproperties are significantly different from those of the liquid water and solid ice thatwe are familiar with (Cao 2004).

The technological importance of these nanostructures is well demonstrated invarious applications, including in catalytic process, biotechnology, medical andbiomedical, photonic, energy-storage, etc. It is worth mentioning that the nano-structures properties are dependent not only on size but also on morphology andspatial organization. Factors like microstructures of nanoparticles, their size dis-tribution, order of orientation, presence of defects and contaminants also signifi-cantly change the suitability of a nanostructure for integration of any material ordevices. Accordingly, the feature size, shape of nanostructures and its purity need tobe well controlled to attain the properties and functions that have been alreadyestablished. Consequently, besides fabulous potential applications of nanostructuresin different areas, their fascinating properties still remain the main motivation forfurther discovery and exploration.

2.1.2.1 Synthesis Technologies and Challenges

Referring to their wide range of applications, synthesis and manufacturing ofnanotechnology-based products is one of the most active fields in nanoscience andnanoengineering. Nevertheless, advances in this field mainly depend upon theability to synthesize nanostructures of controlled properties. It is well recognized


that properties of nanostructure materials greatly depend on the size, shape, com-position, morphology and its crystalline structure. Accordingly, various approacheshave been developed to control these parameters and, therefore, meet the require-ments for diverse applications. Despite numerous technologies for fabrication ofnanostructures, typically, there are two drastically different approaches, top-downand the bottom-up. The top-down approach is analogous to making a stone statuewhich is starting from bulk size and getting nanosize. A bulk piece of solidmaterials is taken and modified by milling, carving or cutting to create the desirableshape and size. The top-down process involves material wastage and is limited bythe resolution of the tools employed. The smallest sizes of the structures made bythese techniques are also restricted. Anyway, the top-down approaches have beenpracticed with great success by the electronics industry in fabrication of integratedcircuits (Cao 2004).

Examples of this kind of approach include the various types of lithographictechniques (such as photo-, ion beam, electron or X-ray-lithography) and solidtreatment such as milling, cutting, etching and grinding. The bottom-up approachfor creating nanostructure involves starting from unit base of material (atom ormolecule) and getting the bulk size with controlling the unit base arrangement. Thisapproach can be analogous to building a house. Lots of building blocks are takenand put in a specific place to make final bigger structure. There is less wastage inthis technique; however, it is limited in how big the structures can be made.Producing nanostructures is generally carried out through chemical reaction, andstrong covalent bonds will hold the constituent parts together. This approach is themore preferred and efficient method for fabricating a wide variety of nanostructureswith controllable size and properties. Chemical synthesis, self-assembly andmolecular fabrication are all examples of bottom-up techniques (Texter and Tirrell2001). A good example of bottom-up approach can be found in nature; all cells useenzymes to produce DNA by taking the component molecules and binding themtogether to make the final structure. In addition to these two main approaches, thereare some special methods that cannot be classified in these two categories, asdepicted in Fig. 2.8.

As explained in the section of nanotechnology applications in different fields,nanostructures have the potential to provide greatly enhanced performance andcustomer benefits at very low volume use. The practice of large volume manu-facturing for bulk chemicals is unlikely to be the way for development of manu-facturing processes for nanostructures. Instead of large footprint plants, smaller,portable, modular and integrated manufacturing systems will be desirable fornanostructures production (Zhao et al. 2003; Stephanopoulos et al. 2005). Themajor challenge for development of such systems is their scale-up. Scale-up ofmanufacturing processes for nanostructures to technological scales has not beenpursued to an appreciable degree yet.

Another important issue is the matter of kinetics. Bottom-up approach isgrounded in processes that tend toward desirable equilibrium structures. Chemicalprocesses generally should go as fast as possible, consistent with product quality. Itis difficult to envision the processing research fruitfully without data on the rates of


formation process. For design and development purposes, it is required to fullyunderstand the trajectories, through time and structural intermediates, of the pro-cesses. This needs research in reaction kinetics, and the related issues of mecha-nisms have several facets. As the complexity and number of components in theseprocesses increase, predictive models will become more important tools in processand product design. Thorough understanding of chemical bonding, reactionmechanisms, pathways and kinetics is crucial for reactor design. Practical processesmust be controlled to produce nanostructures that possess intricate internal struc-ture. Structural analyses present instrumental challenges, i.e., techniques capable ofmeasuring with reasonable resolution are needed; informative, online measurementis also prerequisite for process control; characterization of defects is particularlyimportant for the applications envisioned (Wang 2000a, b; Gommes et al. 2004;Puretzky et al. 2005; Gancs et al. 2008; Schlemmer et al. 2010; Chow et al. 2013).Moreover, barriers like environmental protection issues and human society safetymust be applied to the development of nanostructures fabrication in the commercialsector.

Top-down Methods

Synthesis of nanomaterials by means of top-down approaches is generally physicalor mechanical approach. The top-down strategies are mainly based on milling,machining and lithography. A categorized diagram of these methods is depicted inFig. 2.9.

Fig. 2.8 Schematicillustrations of the synthesismethods of nanomaterials(Qiao et al. 2011)


Photolithography, in which the entire surface is simultaneously patterned atonce, is the most common top-down technique. This technique is cost-effective andrelatively fast; however, its resolution is limited to typically 0.2–0.5 mm because ofthe optical diffraction effects. Electron and ion-based lithography, in contrast,provide creation of ordered arrays of nanostructures with high resolution of about50 nm (Qiao et al. 2011). Since they have line-by-line generation pattern, bettercontrol over shape and spacing of nanostructures is achieved but at slower ratecompared to photolithography. Dip pen lithography (DPN) is a process in which thetip of an AFM is “dipped” into a chemical fluid and then used to “write” on asurface, like an old-fashioned ink pen onto paper. DPN is also serial technique, andhence, it is not suitable for high-volume manufacturing technologies. Nanoimprint

Fig. 2.9 Classification of the common top-down methods for nanostructural material synthesis orproduction


lithography (NIL) is a process for creating nanoscale features by “stamping” or“printing” them onto a surface. In soft lithography, patterns of small features arestamped and lines at a width nanometer can be printed. These patterns becomemicro-channels for analysis of nucleic acids, proteins or cells in a lab-on-a-chip orlab-on-a-system devices. Other top-down techniques include scanning tunnelingmicroscopy (STM), micro-contact printing (mCP) and NIL (Liu et al. 2008).

Another branch of top-down method in the nanostructure material synthesis ismilling. Milling is generally a conventional method to size reduction in solid. Ittheoretically can use to reduce size of the powder to nanosized range, but in thepractice, this method has several limitations. Milling is a routine method for pro-ducing nanocrystalline powder, which is named powder metallurgy. Besides con-ventional mills, today new generation of mills, planetary mill, producessubmicronized powder with the nanosized crystalline. This method is also using inthe mechanical alloying.

Generally, mechanical milling has proved to be an effective and simple tech-nique without involving high-temperature treatment for the production of nano-powder and nanocrystalline powders, with the possibility of obtaining largequantities of materials with modified properties. In this technique, starting powderparticles are trapped between highly kinetic colliding balls and the inner surface ofthe vial, which causes repeated deformation, re-welding and fragmentation ofpremixed powders resulting in the formation of fine, dispersed particles in thegrain-refined matrix. During the milling operation, two essential processes affect theparticle characteristics. First, the cold welding process leads to an increase inaverage particle size of the composite. The second, fragmentation, process causesthe breaking up of composite particles. Steady-state equilibrium is attained when abalance is achieved between these processes after a certain period of milling (Salahet al. 2011). Another type of nanostructure synthesis is micromachining. In this typeof operation, usually a mechanical machining such as drilling, cutting, turning andso on is done in nanoscale using special apparatus such as laser beam, ion beam,X-ray beam and chemical or electrochemical methods.

Bottom-up Methods

The bottom-up approach is a self-assembly of molecular species, with controllablechemical reactions. This approach involves the creation and utilization of functionalmaterials, devices and systems with novel properties and functions achieved in theforms of control of matter, atom by atom, molecule by molecule or at the macro-molecular level. This fact causes the synthesis to be carried out in a fluid phase,while the top-down methods were done in the solid phase. Compared to top-downapproach, this technique is more efficient and flexible to synthesis vast variety ofnanomaterials with well-controlled shape, size, morphology, structure, surfaceproperties conveniently.

Generally, the bottom-up approach to synthesis of nanostructural materials canbe divided to two main methods which are synthesis in liquid phase and gas


(or vapor) phase. In most of these methods, chemical reactions take place, and inthe others, physical events (such as vaporization, condensation or precipitation)occur. In any case, the unit base of the nanostructure product (atom or molecule) isgenerated in the mobile phase (fluid phase). Then, nanostructure is created withcontrolling of unit base mobility and also their deposit arrangement in an expectedpattern. This controlling of deposit arrangement is the art of nanotechnology. Inother words, when the synthesis condition (such as temperature, pressure, atmo-sphere, initial concentration, reactant type, resident time, type and concentration ofadditives and so on) is controlled and leads sensitively, different nanostructureshapes such as hollow sphere, filled sphere, core–shell, tube, rod, plate, horn shapeand any other one-, two- and three-dimensional shape with desirable compositionand crystallinity are achievable even in the mass production.

Examples for synthesis in liquid phase are precipitation, sol–gel, hydrothermal,microemulsion and electrochemical deposition (anodizing and cathodizing). Themost common types of bottom-up synthesis procedures in the liquid phase are thosebased on the use of sol–gel method. Sol–gel method is referred to a large group ofsynthesis methods where sol is obtained from solution through hydrolysis followedby gel formation through poly-condensation reactions. Sol–gel methods are mostlybased on controlled hydrolysis of metal alkoxides in aqueous or organic mediumwhere there are two distinct reactions: hydrolysis of the alcohol groups andpoly-condensation of the resulting hydroxyl groups as follow:

Hydrolysis:�M�ORþ H2O ! �M�OHþ ROH

Poly-condensation:�M�OHþ�M�OR ! �M�O�M�þ ROH

�M�OHþ�M�OH ! �M�O�Mþ H2O

Metal alkoxides are denoted as M(OR)x where M is symbol of a metal and R ishydrocarbon chain.

The first step in sol–gel processes leads to the formation of a colloidal solution ofmonomers (in dimension about 0.1–1 μm) where only the Brownian motions arepresent, named as sol. Increasing bulk concentration of the dispersed phase or anychanges in pH and/or solvent substitution result in formation of strong contactsbetween particles, and thus, monolithic gel, a solid network containing liquidcomponents, is formed. In gel state, liquid and solid are dispersed in each other,where molecules of solvent are enclosed in a flexible, but fairly stable,three-dimensional grid formed by solid particles. Concentration of sols that leads togel formation is carried out by evaporation at relatively low temperatures, extrac-tion, dialysis or electrodialysis and/or ultra-filtration. Sol–gel processes are gener-ally applied for synthesis of a wide range of nanotechnology-based products such asnanoparticles, thin films, nanofibers. The sol–gel process for synthesis of thin filmsusually consists of four steps:

• The desired colloidal particles once dispersed in a liquid to form a sol.• The deposition of sol solution produces the coatings on the substrates by

spraying, dipping or spinning.


• The particles in sol are polymerized through the removal of the stabilizingcomponents and produce a gel in a state of a continuous network.

• The final heat treatments pyrolyze the remaining organic or inorganic compo-nents and form an amorphous or crystalline coating.

After gel formation, the solvent needs to be removed from the gel. Differenttypes of dried gel (aerogels, xerogels, ambigels, cryogels) can be produceddepending on the method of drying (super critical drying, ambient drying, freezedrying, etc.). The final product (dried gel), pronounced quasi-one-dimensionalstructure, includes the nanostructures whose bulk density can vary by hundreds oftimes with sufficiently high surface area (hundreds of m2/g). When drying causesshrinking in the gel, xerogels are formed, and in case no shrinking happens in thegel, aerogels are obtained. Shrinking and gel shape deformation depend on theconditions applied for drying; for example, ambient drying will result in xerogels,and super critical drying will get aerogels. Figure 2.10 illustrates the consequencesof sol–gel process to produce different types of dried gel and nanostructures.

Examples for synthesis in gas phase, based on bottom-up approach, are chemicalvapor deposition (CVD), chemical vapor synthesis (CVS), physical vapor deposition(PVD), light scattering, laser ablation. CVD as a symbol of gas-phase synthesisprocedure is a technique which is used to produce solid nanomaterials, typically CNTand thin films. In this process, substrates are heated to high temperatures and exposedto precursor materials in the gaseous state. The precursors react or decompose on thesubstrate surface to yield a coating of the required material. CVD has been foundsubsequently to be an excellent, scalable method for the production of high-qualitymaterial. The CVD growth of carbon nanomaterials consists of several stages:

• Substrate heating/conditioning• Growth• Substrate cooling

Fig. 2.10 Schematic diagram of sol–gel method and its nanomaterials products


To start, substrates are heated to the growth temperature. During heating, it iscommon to flow gases appropriate for substrate conditioning (i.e., for removal ofsurface oxides and contamination). After a certain time period, precursor gases areintroduced into the reaction chamber. Given appropriate system conditions (tem-perature, pressure, etc.), these precursors lead to the formation of films ofsurface-bound material. Growth is carried out for a certain period of time, afterwhich precursor supply is terminated. Heating is then switched off and the substratecooled to a temperature at which it can be safely removed. During these stages,several events take place on the precursor and product atoms (molecules) and theseevents affect the product properties. These events are illustrated in Fig. 2.11, whichare as follows:

• Precursor balk flow• Precursor adsorption• Surface and gas-phase reactions• Surface diffusion of precursor and product• Precursor and product desorption• Nucleation and growth of product

For graphene synthesis, CVD processes are typically carried out using metalsubstrates, methane feedstock, low pressures (1–50 Torr) and temperatures of 900–1000 °C. In addition to methane, other process gases, such as argon and hydrogen,may be present. Common substrates include copper and nickel. Copper substratesare particularly attractive, since growth is then dependent on precursor-substratecontact. In this case, growth stops after graphene formation (as opposed tocontinuing on to form thick, nonuniform graphitic material).

Fig. 2.11 Events take place in the CVD process (http://postechlocal.k2web.co.kr/user)


http://postechlocal.k2web.co.kr/user

Recently, researchers have shown that CVD formation of graphene may bepossible at reduced temperatures. This work has involved the use of room tem-perature liquid precursors, e.g., toluene, introduced into CVD systems in vaporform. Reduced growth temperatures should enable CVD processes to becomecompatible with more substrate types. Other work has also demonstrated the for-mation of graphene material from solid precursors, e.g., PMMA.

CVD growth of CNTs is very different. Rather than metal films or foils, CNTsynthesis requires inert surfaces decorated with metallic nanoparticles. For these,transition metals such as iron, nickel and cobalt are commonly used. The metals areapplied to substrates directly in nanoparticle form (e.g., from suitable liquid dis-persions) or as thin films, prior to growth. For the latter case, theheating/conditioning step involves nanoparticle formation through thermal filmcoalescence. During the subsequent growth stage, nanoparticles are exposed tofeedstocks. Typical feedstocks include methane, ethylene and alcohol vapors. Thesethermally decompose to yield carbon at nanoparticle sites that is catalyticallyassembled into CNT structures. These CNTs “sprout” out of the nanoparticles asgrowth continues. The classification of bottom to up methods in nanostructuralmaterials synthesis is shown in Fig. 2.12. It should be noticed that there, othermethods not mentioned in this chart. Only, the general methods are categorized inthis chart and each mentioned method can be divided into several detailed methodsthat follow the same general route. In addition, those methods that are important inthe chemical engineering view point are mentioned.

PVD

CVS

CVD

Gas (vapor) phase

Electrochemical deposition

Hydrotherman

Microemulsion

Sol-gel

Co-precipitation

Liquid phase

Bottom to up

Fig. 2.12 Schematic diagram of categorized methods in the bottom to up procedure in thenanomaterials synthesis


Other Special Synthetic Methods

The most important achievements in nanotechnology are the novel techniques bycombining the methods of lithography and self-assembly, such as lab-on-a-chip andsystem-on-a-chip. There are other special methods, including microwave irradia-tion, photochemical synthesis and bioinspired synthesis. Some of these techniquesare:

• Biomimetic and Bioinspired Methods• Photochemical and Radiation chemical Methods• Ultrasonic-assisted Synthetic Methods• Microwave Synthetic Methods• Ionic liquid-assisted Synthesis Methods• Electrochemical Synthesis

2.1.3 Routine Tests for Characterization of Nanostructures

To determine the success of the nanotechnology-related processing, the keyobjective is to characterize the structural feature as well as chemical and physicalproperties of nanostructured system. Numbers of characterization techniques forboth individual and bulk nanostructured systems have been adopted and developed.High-resolution techniques offer the possibilities to study individual nanostructures,while bulk characterization methods do not provide information of individualnanoparticles. The properties and behaviors observed and measured by thesetechniques are typically group characteristics. Properties of nanomaterials can besubstantially different from that of their bulk encounters. However, bulk charac-terization techniques are essential complement to other high-resolution methods,which provide rather detailed information on only a few particles. Since mostnanostructures have uniform chemical composition and structures, bulk methodsare extensively used in the study of some properties of nanostructures, likemechanical, electronic and optical properties. Nevertheless, all nanostructuresproperties are size dependent, and thus, their properties can be considerably tunedby adjusting the size, shape or extent of agglomeration (Cao 2004). For example,the optical absorption peak of metal particles can shift by hundreds of nanometersvia particle size and shape. Recall that both techniques are complementary in thestudy of nanostructures. This section presents the basic principles of those char-acterization methods mostly applied in nanotechnology researches. Full explanationof all characterization techniques, their technical details, operation procedures andinstrumentations are beyond the aim of this text, and readers interested to obtainmore detailed information can refer to the special books and handbook like (Settle1997; Wang 2000a, b; Wilkening and Koenders 2006).


2.1.3.1 Microscopes

The first step in characterization of nanostructures is to visualize their morphology.Under ideal conditions, the smallest object that the eye can resolve is about0.07 mm. This limit is related to the size of the receptors in the retina of the eye.Microscopes are employed to improve our capacity for observing the objects inmore details. These instruments allow us to observe a magnified image with greaterdetails as they effectively bring the object closer to the eye and hence magnify theimage falling on the retina. The best distance that one can resolve with opticalmicroscope is of the order of 0.25 µm, and hence, this type of microscope is notapplicable for nanostructure visualization. For characterization and manipulation ofindividual nanostructures, it becomes essential to measure at a nanometer scale andresolution of the order of atomic distances or even smaller. High-resolution tech-niques can provide resolution at nanoscale and thus local information on thenanometer scale. High-resolution microscopes are categorized into scanning probemicroscopes and electron microscopes (EM).

Scanning Probe Microscopy

Scanning probe microscopy (SPM) covers a broad group of quantitative measuringinstruments applied for topographical imaging as well as quantification of chemicaland physical properties of surfaces at a resolution down to nanometers (Bonnell2001; Meyer et al. 2004). In SPM, a probe of nanometer dimension is used to tracethe surface of the sample. A sharp tip (3–50 nm radius of curvature) mounted on aflexible cantilever scan across the object surface. Due to flexibility of cantilever, tipcan follow the surface profile and as a result produce topographic image of a surfacewith atomic resolution in all three dimensions (Chi and Röthig 2000). The reso-lution of an image obtained by SPMs depends on the sample, movement control ofthe tip on sample and the inherent nature of the data. The mechanism presented andthe scanning of a nanoscale probe forms the basis of all scanning probe instruments.Depending on the physical interactions used to probe the surface, the scanningprobe microscopes have different names. Atomic force microscopy (AFM) andSTM are the most common SPM used in study of nanostructures.

Generally speaking, AFMs use interatomic or intermolecular forces, while STMsare based on the quantum mechanical tunneling effect. Almost all solid surfaces,whether hard or soft, electrically conductive or not, can all be studied with STM andAFM. Although SPM is a surface image technique, combining with appropriateinstruments, they have found a much broadened range of applications, such asnanoindentation, nanolithography and patterned self-assembly. SPM techniques arealso used in biotechnology researches, for example, study of biomolecule imagingand proteins unfolding, antibody–antigen binding, binding forces of complimentaryDNA strands, etc. (Gaboriaud and Dufrêne 2007; Wei and Liu 2010; Murty et al.2013). Furthermore, SPM has recently entered the production and quality controlenvironment of semiconductor manufacturers (Magonov et al. 2011).


Atomic Force Microscopy

Atomic force microscopy (AFM) is one type of SPM used to image surfacestructures. In addition, as its mechanism depends on the force of attraction betweenmolecules, it is also possible to measure surface forces, i.e., attractive or repulsiveforces between tip and sample (Magonov and Alexander 2008). In AFMs, the probeis a tip at the end of a cantilever which bends in response to the force between thetip and the sample. Figure 2.13 illustrates a simple schematic of an AFM. A sampleis positioned on a piezoelectric scanner. A microscopic tip (curvature radius of*10–50 nm) attached to a cantilever spring moves across the surface of the sample,and deflections are detected by measuring the cantilever’s vertical displacementusing reflections from a laser beam. When the cantilever flexes, the light from thelaser is reflected onto the split photodiode and position of the reflected beamchanges depending on the cantilever deflection. A photodetector converts thischange in an electrical signal, and a displacement map of the surface is depicted thatallows visualization of surface structure at the nanometer scale or even subna-nometer scale (Wang 2000a, b). Since AFMs provide three-dimensional images of asample surface, it probes the sample and makes measurements in three dimensions,x, y and z. A resolution in the x–y plane ranges from 0.1 to 1.0 nm and in thez direction is 0.01 nm (atomic resolution) (Chatterjee et al. 2010).

AFM does not require any special sample, and no current flows between theAFM tip and sample. The later makes AFM suitable for studies of nonconductors.Accordingly, this technique is widely employed for studies and the detection ofatomic scale features of insulating surfaces including ceramic materials, biologicalsamples and polymers. These microscopes can be used in either an ambient orliquid environment, and it does not need a vacuum environment. With all theseadvantages, AFM is capable of measuring topography, surface energy and elasticityof samples at the nanometer, even molecular scale, and it has significantly impactedthe research fields of materials science, physics, chemistry, biotechnology, poly-mers and the specialized field of semiconductors (Magonov et al. 2011). The AFMmicroscope is capable of imaging all kinds of surfaces under atmospheric

Fig. 2.13 Schematic diagramshowing AFM principles (Chiand Röthig 2000)


conditions without the need of special sample preparation. One of the best imagesobtained from AFM scans is topography of a graphite surface as illustrated inFig. 2.14. A good top view image of graphite indicates the graphene lattice in thewhite lines, and different shading of carbon atoms results from the different situ-ation in the atomic layer underneath (Hölscher et al. 2000).

Scanning Tunneling Microscopy

Scanning tunneling microscopies (STM) are addressed as the first instrument togive real-space atomic resolution images (Binnig et al. 1982). STM relies on theelectrical conductivity of the sample, and a weak electrical current flowing betweentip and conductive sample is measured using tunneling (exponential) current.In STM, electrical charges pass from the surface of an object to the point of themicroscope without there being any contact. The current varies strongly with dis-tance. The movement of the point of the microscope is controlled with a specificcurrent value in order to follow exactly the surface of the sample. Figure 2.15illustrates a schematic of STM.

Electron Microscopes

Electron microscopy (EM) is the most powerful technique for structural researchand characterization of nanoparticles. With this technique, several important fea-tures of nanostructures such as size, presence of defects or contamination and

Fig. 2.14 AFM topographyof a graphite surface(Hölscher et al. 2000)


surface quality can be observed (Li 2002; Gommes et al. 2004). Similar to lightmicroscopy where a light source is used to produce an image of higher magnifi-cation from a specimen, in EM, an electron source is used for the same purpose. Inorder to prevent overheating inside the instrument due to collisions between theelectrons and gas molecules, the inner compartments of the EM are maintained in avacuum. EM, like light microscope, contains a number of diaphragms that restrainthe dispersion of the electron beam, and lenses that deflect electrons. Similar to lightmicroscope, a condenser lens concentrates the beam, and an objective lens focusesthe beam on the object.

An EM, which uses the wave nature of the electron to capture an image, iscapable of imaging at a significantly higher resolution and magnification than lightmicroscopes (Egerton 2005; Goodhew et al. 2000). This enables the instrument’suser to examine specimen at nanoscale, thousands of times smaller than the smallestresolvable object in a light microscope. Electron microscopy has a major role inelucidating those micro- and nanostructures in which the reflections in reciprocalspace are not sharp and spread out along lines or planes. Examples of thesestructures are fullerenes and related molecular structures like CNTs (Tendeloo andAmelinckx 2000; Gommes et al. 2004). Electron diffraction technique is able toproduce useful structural information, mainly because the electron–sample inter-action is very strong (Tendeloo and Amelinckx 2000). In electron microscopy,electrons are source of illuminating the sample. The lenses used in EM are elec-tromagnetic lenses, which are widely different from glass lenses, though similar

Fig. 2.15 Schematic diagram showing STM principles (http://www.virlab.virginia.edu/VL/easyScan_STM.htm)


http://www.virlab.virginia.edu/VL/easyScan_STM.htm

http://www.virlab.virginia.edu/VL/easyScan_STM.htm

principles apply in both cases (Pradeep 2007). There are two important types ofelectron microscopy, namely scanning electron microscopy (SEM) and transmis-sion electron microscopy (TEM). The schematic of SEM and TEM machines isillustrated in Figs. 2.16 and 2.17, respectively, and the principal elements of themare presented in the following sections.

Scanning Electron Microscopy

Main parts of Scanning electron microscopy (SEM) include an electron gun, lenses(a condenser lens, an objective lens, stigmator lenses), coils for the x–y scanmovement, specimen chamber and detection device for image formation(Fig. 2.18). An electron gun in SEM provides a stable beam of electrons, and thesurface of solid is scanned with a focused beam of electrons (Goodhew et al. 2000).The electrons interact with atoms in the sample and produce various signals thatcontain information about the physical nature, sample’s surface topography andchemical composition of it. Specimens can be observed in high vacuum, in low

Fig. 2.16 Scanning electronmicroscopy (SEM) machine


vacuum and (in environmental SEM) in wet conditions. SEM is the most widelyused electron microscopy because of its versatility, its various modes of imaging,ease of sample preparation, possibility of spectroscopy and diffraction, as well aseasy interpretation of the images (Pradeep 2007). Another advantage of SEM is thata very wide range of magnification is available which facilitates the visualization ofvirtually every detail. Some SEMs can obtain image resolutions even smaller than1 nm (around 0.5 nm). No special sample preparation and no limitation for samplesize are other benefits of this type of microscopy. The resolution of the SEMapproaches a few nanometers, and the instruments can operate at magnificationsover 1000 (Egerton 2005). Figure 2.19 shows SEM observation of bundles of CNTswith 5000 magnification.

Fig. 2.17 Transmissionelectron microscopy(TEM) machine

Fig. 2.18 Schematic diagram showing the main components of a SEM


Field Emission Scanning Electron Microscopy (FESEM)

A field emission cathode, a thin and sharp tungsten needle (tip diameter 0.1–0.01 µm), in the electron gun of a SEM can liberate narrower probing electronbeam. The beam is accelerated in the direction of the column by a voltage gradient.The electron beam produced by the field emission source is about 1000 timessmaller than in a common SEM that causes improved spatial resolution as well asminimized sample charging and damage (Li 2002). Therefore, quality of imagescaptured by FESEM is markedly better (clearer and less electrostatically distorted)compared to images obtained from SEM. Field emission necessitates an extremevacuum (*10−11 atm) in the column of the microscope. However, in contrast to aconventional tungsten filament used in SEM, a field emission tip lasts theoreticallyfor a lifetime, provided the vacuum is maintained stable. For ultra-high magnifi-cation imaging, like advanced coating thickness and structure uniformity determi-nation, FESEM is the most suitable device. Small contamination feature geometryand elemental composition measurement are also possible with FESEM. The res-olution of SEM is limited to about 1 nm, whereas FESEM can achieve higherresolutions (Bhushan 2010). Figure 2.20 shows FESEM observation of CNT usingdifferent magnifications.

Transmission Electron Microscopy

Transmission electron microscopy (TEM) is an EM technique used to obtainnanostructural information by diffraction and imaging from an ultra-thin specimen(Wang 2000a, b). It can provide image of higher magnification with better reso-lution compared to SEM. A schematic diagram for a common type of TEM isshown in Fig. 2.21. A beam of electrons is transmitted through specimen, inter-acting with the specimen as it passes through the sample. An image is formed fromthe interaction of the electrons transmitted through the specimen. The image can be

Fig. 2.19 SEM monographCNT 5000 magnification


magnified and focused onto an imaging device, such as a fluorescent screen, on alayer of photographic film, or to be detected by a CCD (charge-coupled device)image sensor. Comparing the TEM diagram with SEM diagram depicted inFig. 2.19, two clear distinctions can be found. First, there is an additional projectionlens (beneath the objective lens) to project the image on a screen. The second one isthe location of the object for observation. As depicted in Fig. 2.19, the specimenchamber in SEM is located below the column, while in TEM the specimen chamberis located about half way the column (Fig. 2.22). Difference in position of specimenrelates with the specific mode of image formation in either TEM or SEM.

Having fine TEM observations, thin samples should be prepared due to theimportant absorption of the electrons in the material. A modern TEM is capable ofproducing a fine electron probe of smaller than 2 nm, allowing direct identificationthe local composition of an individual nanostructure. Figures 2.22 and 2.23 illus-trate TEM observation on a CNTs sample using different magnifications.

Fig. 2.20 FESEM images of CNTs at different resolutions (Kim et al. 2006b)

Fig. 2.21 Schematic diagram showing the main components of a TEM


High-Resolution TEM

Conventional TEM uses only the transmitted beams or some of the forward-scatteredbeams to create a diffraction contrast image. High-resolution TEM (HRTEM) usesthe transmitted and the scattered beams to create an interference image (Tendeloo andAmelinckx 2000). This technique is a powerful tool to study properties of materials

Fig. 2.22 TEM monograph CNT at magnification of 20,000

Fig. 2.23 TEM monograph CNT at magnification of a 70,000 b 150,000


on the atomic scale like the surface and inner crystallinity, surface and planar defects(Gancs et al. 2008; Williams and Carter 2009). The morphology and size distributionof the nanostructures are most easily obtained by TEM (Fig. 2.25a), whereas theirpurely morphological features such as shape, diameter and length and their internalstructure can be obtained by HRTEM (Fig. 2.24b). HRTEM can be also used todetermine the 3D shape of small particles (Wang 2000a, b).

2.1.3.2 X-ray Diffraction Analysis

X-ray diffraction analysis (XRD) is an important crystallographic method that haslong been used to obtain all information related to the crystalline phase present inthe solid sample. This information includes lattice constants and geometry, iden-tification of unknown materials, orientation of single crystals, preferred orientationof polycrystals, defects, stresses, etc. (Cullity and Stock 2001). In this technique, acollimated and rather monochromatic beam of X-rays, with a wavelength typicallyranging from 0.7 to 2 Å, is incident on a specimen. Interaction of X-rays withcrystalline matter leads its diffraction produced by the reticular planes that form theatoms of the crystal. A crystal diffracts an X-ray beam passing through it to producebeams at specific angles depending on the X-ray wavelength, the crystal orientationand the structure of the crystal. In the macroscopic version of X-ray diffraction, acertain wavelength of radiation will constructively interfere when partially reflectedbetween surfaces (i.e., the atomic planes) that produce a path difference equal to anintegral number of wavelengths (Zanchet et al. 2000). X-ray diffraction by crystalstructure is described by the Bragg law:

2d sin h ¼ nk ð2:1Þ

Fig. 2.24 TEM a and HRTEM b image of MWCNT (Yu et al. 2006)


where n is an integer, λ is the wavelength of the radiation, d is the spacing betweenatomic planes in the crystalline phase and θ is the angle between the radiation andthe surfaces. The intensity of the diffracted X-rays is measured as a function of thediffraction angle 2θ and the specimen’s orientation. This diffraction pattern is usedto identify the specimen’s crystalline phases and to measure its structural properties.XRD is nondestructive and does not require elaborate sample preparation, whichpartly explains the wide usage of XRD method in materials characterization (Cullityand Stock 2001). In general, the diffraction pattern of crystalline nanoscale mate-rials exhibits broadened and shifted peaks as compared to bulk, and these changesare associated with both size and strain (Souza Filho and Fagan 2011).

This relation demonstrates that interference effects are observable only whenradiation interacts with physical dimensions that are approximately the same size asthe wavelength of the radiation. Since the distances between atoms or ions are onthe order of 1 Å, diffraction methods require radiation in the X-ray region of theelectromagnetic spectrum or beams of electrons or neutrons with similar wave-length. Accordingly, through X-ray spectra one can identify and analyze anycrystalline matter. Each crystalline powder gives a unique diffraction diagram,which is the basis for a qualitative analysis by X-ray diffraction (Belin and Epron2005). Identification is practically always accompanied by the systematic com-parison of the obtained spectrum with a standard one (a pattern), taken from anyX-ray powder data file catalogues, published by the American Society for Testingand Materials (JCPDS). X-ray diffraction is generally used for catalyst character-ization in CNT synthesis research (Tran et al. 2007; Hsieh et al. 2009; Philippe et al.2007), since the crystalline structure of transition metals is not active for CVDreaction. In other words, crystalline configuration of nanostructures cannot catalyzethe CNTs growth reaction. On the other hand, XRD is not basically used for CNTsanalysis as this method can just reveal the graphitized structure of carbon and thereis no difference between CNT and graphite in the spectrum (Belin and Epron 2005).

Another example of using XRD is to study the suitability of heterogeneouscatalyst for CNT formation through CVD method. The suitable heterogeneouscatalyst for CNT production involves a well dispersion of transition metal(s) in asuitable support. Monometallic and bimetallic catalytic particles with the sameamount of metal are compared. Transition metals are the active elements, and theymust be in the form of nanoparticles. Transition metals do not act as catalyst forCNT formation if aggregation of metal particles in the form of crystalline structureis observed. The XRD diffractogram for bimetallic catalytic particles indicates anintense peak just for alumina (Fig. 2.25). The indicated peaks in diffractogram at 2θidentical to 25.56, 35.12, 37.75, 43.32, 52.52, 57.47, 66.47 and 68.17 degree haverelatively high intensity and symmetry. These peaks correspond to the crystallinestructure of alumina (Tran et al. 2007; Hsieh et al. 2009; Philippe et al. 2009).Hence, the predominant species on these samples are only alumina, which obvi-ously comes from the catalyst support particles. Accordingly, as there are nodetectable peaks for metals, and just intense peaks corresponding to the alumina areobserved, the following conclusions can be inferred. The crystallinity of the sam-ples was mainly due to the alumina used as a support but not due to the presence of


any metallic phases. In other words, there is no accumulation of metals (iron orcobalt) or metal compounds in the bimetallic catalytic samples, and the smallparticles of metals were well dispersed in the support.

2.1.3.3 Particle Characterization

Gas sorption (both adsorption and desorption) at the clean surface of dry solidpowders is the most popular method for determining the surface area of the finepowders and porous materials as well as their pore size distribution (Lowell 2004;Condon 2006). In gas sorption methods, the specimen is first heated and degassedby vacuum force or inert gas (such as nitrogen, krypton, or argon) purging toremove adsorbed foreign molecules. The sample is then placed in a vacuumchamber at a constant and very low temperature (usually at the temperature of liquidnitrogen, −195.6 °C), and subjected to a wide range of pressures, to generateadsorption and desorption isotherms (Xu 2001). The amounts of gas moleculesadsorbed or desorbed are determined by the pressure variations due to theadsorption or desorption of the gas molecules by the adsorbent (the sample).Knowing the area occupied by one adsorbate molecule (for example, nitrogen) andusing an adsorption model, the total surface area of the sample can be determined.The most well-known and most widely used adsorption model is the BET equationfor multilayer adsorption (Brunauer et al. 1938):

PnðPs � PÞ ¼

1cnm

þ c� 1cnm

PPs

ð2:2Þ

In Eq. 2.2,P is adsorption pressurePs is saturation vapor pressureC is a constant (so-called) BET constantn is the amount adsorbed (moles per gram of adsorbent) at the relative pressure

Fig. 2.25 Diffractogramobtained by XRD for catalyticparticles comprising iron andcobalt with mass ratioof 8–8 %


nm is the monolayer capacity (moles of molecules needed to make a monolayercoverage on the surface of one gram of adsorbent).

By plotting the quantity on the left of this equation, P/[n(Ps −P) against (P/Ps),versus P/Ps, the terms C and nm can be determined through the slope and interceptof this plot. The specific area, S, can then be derived:

S ¼ NAnmr ð2:3Þ

In Eq. 2.3, NA is Avogadro’s number and σ is the area occupied by one adsorbatemolecule.

The most common adsorbate used is nitrogen with the value σ = 16.2 Å (Xu2001).

Note that, the plot should be taken over the 0.05–0.35 P/Ps range, since beyondthese values the linearity of the plot breaks down. Other parameters, identifiedthrough gas sorption method, are the porosity in terms of pore size, pore volume andpore size distribution. The range of pore sizes that can be measured using gas sorptionis from a few angstroms up to about half a micron. Pore size of porous materials andits distribution are determined using adsorption/desorption isotherm based on anassessment model, such as the t-plot, the MP method, the Dubinin–Radushkevichmethod and the Barrett–Joyner–Halenda (BJH) model, etc. (Condon 2006).

2.1.3.4 Chemical Analysis

Chemical characterization involves determining the surface and interior atoms andcompounds as well as their spatial distributions (Cao 2004). According to theprinciple, different substances produce distinctive spectral lines, and spectroscopytechniques have been developed. Spectroscopy is a general methodology based onprinciple of interaction of electromagnetic radiation on materials. In addition towavelength, other characteristics of the light, such as its intensity, can also provideuseful information, and thus, spectroscopy has been adapted in many ways toextract the information required. In X-ray spectroscopy, for example, when X-raysbombard a substance, the electrons in the inner shells of the atoms are excited andthen de-excite emitting radiation. This radiation comes out at different frequencies,depending on the atom and chemical bonds present. Accordingly, with spectro-scopic techniques, it is possible to determine what elements in what quantities andwhat chemical bonds are present. Different spectroscopic techniques operate overdifferent, limited frequency ranges within this broad, depending on the processesand magnitudes of the energy changes. A vast number of spectroscopy techniqueshave been developed that makes hard to have exact classification for them.Spectroscopy techniques are basically different in terms of nature of energy sourceutilized, type of interaction between the energy and the material and type ofmeasurement.


1. Nature of energy source utilized, for example

• Optical spectroscopy

– X-ray photoelectron spectroscopy (XPS)– UV-vis spectroscopy– Fourier transform Infrared spectroscopy (FTIR)– γ-ray spectroscopy (Mossbauer spectroscopy)

• Electron spectroscopy

– Energy Dispersive Spectroscopy (EDS)– Auger Electron Spectroscopy (AES)– Mass Spectroscopy (MS)

• Ionic spectroscopy

– Rutherford backscattering spectrometry (RBS)– Ion scattering spectroscopy (ISS)

There are some spectroscopy techniques classified as electron spectroscopy, inwhich the kinetic energy of electrons emitted from a substance is measured. Inelectron spectroscopy, the substance is bombarded with ionizing radiation, andthus, electrons are excited by absorbing photon energy from an initiallow-energy state to a higher-energy state. When an electron is ejected from aninner shell of an atom, the resultant vacancy can be filled by either a radiative(X-ray) or nonradiative (Auger) process. Because the energy of these electronsis approximately equal to the difference between the two shells, X-rays or Augerelectron can be a characteristic of the element from which it was released andthe shell energy of that element. The famous examples of this technique includeenergy dispersive X-ray spectroscopy, electron energy loss spectroscopy, X-rayphotoelectron spectroscopy and Auger electron spectroscopy.

2. Type of interaction between the energy and the material

• Emission• Absorption• Vibration (infrared spectroscopy, Raman spectroscopy)Absorption and emission spectroscopy determines the electronic structures ofatoms, ions, molecules or crystals through exciting electrons from the ground toexcited states (absorption) and relaxing from the excited to ground states(emission). Emission spectroscopy can be fluorescence (emission from excitedelectronic singlet states) or phosphorescence (emission from excited electronictriplet states). Fluorescence spectroscopy is commonly used in biology andmedicine, as its damaging effects are less than other methods and because someorganic molecules are naturally fluorescent.Vibrational spectroscopy is employed to derive information on the vibrationalexcitation of molecules. Atoms are spherically symmetric, but molecules haveshapes which permit them to vibrate and rotate. A substance can be thought ofas systems of balls (atoms) connected by springs (chemical bonds). These balls


(atoms) can vibrate or rotate with frequencies determined by their mass (atomicweight) and by the stiffness of the springs (bond strengths). These oscillationsinduced provide new energy levels that can be quantized to extract informationabout chemical bonds in the detecting samples (Cao 2004). Vibrational energiesare much smaller as compared to the chemical bond energies; therefore, evenminute changes in the local atmosphere of a sample are reflected in the spectra(Pradeep 2007).

3. Type of measurement

Infrared spectroscopy is known as Fourier transform infrared spectroscopy(FTIR), because a Fourier transform (a mathematical process) is required to convertthe raw data, intensity-time output, into the actual spectrum (intensity-frequency).FTIR is a powerful technique for identifying the identities, surrounding environ-ments or atomic arrangement, and concentrations of the chemical bonds in thesample.

Due to great development and improvement in spectroscopy techniques, they arenow employed for obtaining wide range of information including energies ofelectronic, vibrational, rotational states, structure and symmetry of molecules,dynamic information. The most common spectroscopic techniques employed incharacterization of nanostructures are presented:

UV-Vis Spectrophotometer

A UV-Vis spectrophotometer measures the amount of light absorbed at eachwavelength of the UV and visible regions of the electromagnetic spectrum. In astandard UV-Vis spectrophotometer, a beam of light is split; one half of the beam(the sample beam) is directed through a transparent cell containing a solution of thecompound being analyzed; and one half (the reference beam) is directed through anidentical cell that does not contain the compound but contains the solvent. UV-Visspectrophotometery is a powerful technique to obtain useful information aboutoptical properties of nanostructure and their size (Burda et al. 2000). For example,reduction in the silver ion to silver nanoparticles (SNPs) results the visual change ofcolor from yellow to red. The UV-Vis spectrum of the SNPs indicates an absorptionband at 420–430 nm as a result of surface plasmon vibrations SNPs (He et al.2004).

Raman Spectroscopy

Raman spectroscopy is based on the inelastic light scattering by the lattice vibra-tions (phonons). In bulk materials, the scattering obeys the momentum selectionrule which states that only the phonon wave vector equal to zero is allowed (SouzaFilho and Fagan 2011). Raman spectrum is very sensitive to the lengths, strengthsand arrangements of chemical bonds in a material, but less sensitive to the chemicalcomposition (Cao 2004). A promising application of Raman spectroscopy is forstructural analysis of CNTs. Different types of CNTs, single-wall, double-wall andmulti-wall CNTs, can be distinguished by Raman spectroscopy (Dresselhaus et al.2005). The various forms of CNT can be distinguished by the position and the line


width of the frequency bands in the Raman spectrum. Figure 2.26 demonstrates aRaman spectrum of one SWNT taken over a broad frequency range.

Breathing mode (RBM) is usually located between 75 and 300 cm−1 from theexciting line; an illustration of the spectrum resulting from this mode is displayed inthe figure. The frequency of the RBM is directly linked to the reciprocal of thenanotube diameter. Basically, D band position at 1285–1300 cm−1 and a linewidthof 10–30 cm−1 are characteristic of SWNTs, while MWCNTs form has a typicalposition of 1305–1330 cm−1 and a width of about 30–60 cm−1 (Belin and Epron2005). The D band is expected to be observed in multi-walled carbon nanotube(MWCNT). However, when it is observed in single-walled carbon nanotube(SWNT), it is assumed that it is due to defects in the tubes. Raman spectroscopy hasalso provided a powerful tool to study the vibrational properties and electronicstructures of SWCNT (Lefrant 2002). The position, width and relative intensity ofthe bands in the Raman spectrum are modified according to the type of CNT. Forexample, the Raman line shape differs between metallic and semiconductor nano-tubes and thus allows distinguishing the two types.

Energy Dispersive Spectroscopy

Quantitative analysis of chemical composition for a specimen can be obtained byX-ray energy dispersive spectrometry (EDS), sometimes is referred as EDX. In thistechnique, a specimen is excited by the incident electrons. The X-rays emitted fromthe sample atoms represent the characteristics of the elements, and their intensitydistribution represents the thickness-projected atom densities in the specimen(Wang 2000a, b). EDS is generally a complementary tool with EM, SEM or TEM.For example, in order to understand heterogeneous catalysis, information about thenature and structure of the upper atomic layers is required (Hagen 1999). SEMimage combined with EDS spectrum of a catalytic sample can be provide infor-mation about the morphology, structure and composition of it. EDS analysis can

Fig. 2.26 Raman spectrum ofone SWNT taken over a broadfrequency range (Dresselhauset al. 2002)


qualify and quantify the presence of metal elements on the surface of the catalyticparticles, and EM observation indicates surface morphology. Figure 2.27 showsSEM/EDX analysis result for a catalyst sample made from iron and cobalt as activeelements supported on alumina. With SEM/EDX, both topographical informationand knowledge about elemental composition are obtained.

Electron Energy Loss Spectroscopy (EELS)

Electron energy loss spectroscopy (EELS), which is another common spectroscopytechnique available on many EM, is often spoken of as being complementary toEDS.

A proper application of EELS can provide additional information which is notpossible in the case of EDS. For example, EELS is capable of measuring atomiccomposition, chemical bonding, valence and conduction band electronic properties,surface properties and element-specific pair distance distribution functions (Wang2000a, b; Egerton 2005). With EELS, it is possible to differentiate different forms ofcarbon, diamond, graphite, amorphous carbon and “mineral” carbon (such as thecarbon appearing in carbonates). The spectra of 3d transition metals can be alsoanalyzed to identify the oxidation states of the atoms transitions. EELS is based onthe inelastic collisions of a monochromatic beam of electrons and the study of thekinetic energy of the electrons. The energy loss of the sample corresponds toexcitations in the sample (Bhushan 2010).

Fig. 2.27 SEM/EDX illustration of catalyst surface (Fe–Co: 2.5–2.5 %)


Auger Electron Spectroscopy

Auger electron spectroscopy (AES) derives its name from the effect, first observed byPierre Auger, a French Physicist (Hawkins 1977). Under the impact of an incidentelectron, the electrons bounded to the atoms may be excited, either to a free electronstate or to an unoccupied energy level with a higher energy. The quantum transitionsassociated with these excitations will emit photons (or X-rays) and electrons such assecondary electrons, Auger electrons and ionized electrons. By discriminatingbetween Auger electrons of various energies, quantitative chemical and electronicstructural analysis can be obtained (Olsson et al. 1997). AES, which utilizes theemission of low-energy electrons in the Auger process, provides elemental analysisof surface layers of a specimen with high sensitivity (typically ca. 1 % monolayer)for all elements except H and He (Chourasia and Chopra 1997). AES is based onthree basic steps, atomic ionization, electron emission and analysis of the emittedAuger electrons. A specimen atom, excited by the incident electron, emits some ofthe energy by one of the higher level electrons coming down, by emitting a secondelectron with a characteristic energy. The elements with higher Auger electron yieldshave lower X-ray emission and vice versa. Thus, the AES is more sensitive to lightelements, while EDS is to heavier elements (Wang 2000a, b).

X-Ray Photoelectron Spectroscopy

X-ray photoelectron spectroscopy (XPS) is based on detecting photoelectronsejected by X-rays. This technique provides determination of the chemical bondingof species present on the surface of solid materials. XPS, like AES, is applicable todetecting almost all the elements in the Periodic Table, with few exceptions,whereas EDS can only detect elements with Z > 11. XPS, and to a much lesserextent AES, is capable of readily providing information on the nature of chemicalbonding and valence states (Brune et al. 1997; Wang 2000a, b). XPS is animportant tool in materials engineering with applications such as corrosion,embrittlement of metals and powder metallurgy. In polymer technology, XPS iswidely used for analysis of functional groups and determination of thickness anddistribution of thin liquid film on a substrate (Bhushan 2010).

Ionic Spectroscopy

When a beam of ions hits the surface of amaterial, a part of projectiles will be scatteredback into the vacuum after one or more collisions. Measuring the energy of back-scattered particles can be used to identify the mass of these atoms, which is the base ofISS. The technique is classified depending on the energy of the primary ion beam:

• Low-energy ion scattering (LEIS) spectroscopy is referred to primary energiesin the range of 100 eV–10 keV,

• medium-energy ion scattering (MEIS) to a range from 100 to 200 keV and• high-energy scattering (HEIS) to energies between 1 and several MeV.

Often the LEIS technique is called ISS, while HEIS technique is best known asRutherford backscattering spectroscopy (RBS). RBS is quantitatively a precise


technique for analysis of chemical composition, and it is a popular thin filmcharacterization technique. However, RBS is restricted to only selected combina-tions of elements whose spectra do not overlap. LEIS (or ISS) has some advantagesand disadvantages depending on the specific problem being examined. The quan-tification analysis using LEIS is impeded by the uncertainty of the inelastic lossesand the neutralization rate depending on ion trajectories. The practical use of ISS isdetermined by its extreme sensitivity to only the top surface or two monolayers. It isnecessary to note that ISS is strongly affected by surface contamination and ele-ments of similar molecular mass cannot be differentiated. For example, in a systemcomprising iron and cobalt, these two elements cannot be distinguished becausetheir molecular masses are so close to each other (Taglauer 1991).

Mass Spectroscopy

Mass spectrometry is based on slightly different principles to the other spectroscopicmethods. Mass spectroscopy determines the mass/charge ratio (m/z) in the vaporphase of a specimen. With mass spectroscopy, it is possible to determine exactmolecular mass and structure of the molecule (Downard 2004). The physics behindmass spectrometry is that a charged particle passing through a magnetic field isdeflected along a circular path on a radius proportional to the mass/charge ratio.Bombardment of a sample surface with a primary ion beam followed by massspectrometry of the emitted secondary ions constitutes secondary ion mass spec-trometry (SIMS) (Benninghoven et al. 1987). Today, SIMS is widely used foranalysis of trace elements in solid materials, especially semiconductors and thinfilms. The SIMS ion source is one of only a few to produce ions from solid sampleswithout prior vaporization. The SIMS primary ion beam can be focused to less than1 µm in diameter. Controlling where the primary ion beam strikes the samplesurface provides for microanalysis, the measurement of the lateral distribution ofelements on a microscopic scale (Brune et al. 1997).

2.1.3.5 Thermal Analysis

Thermal analysis is a general name for measurement of certain physical andchemical properties, like enthalpy, heat capacity, mass and coefficient of thermalexpansion as a function of temperature. The most common thermal analysis tech-niques include thermogravimetric analysis (TGA), differential thermal analysis(DTA) and differential scanning calorimetry (DSC). In DTA and DSC, temperatureof a sample is compared with that of an inert reference material during a pro-grammed change of temperature. Thermal analysis techniques have found impor-tant roles to play in analysis and development of nanostructures andnanocomposites. TGA is an analytical technique used to determine a material’sthermal stability and its fraction of volatile components by measuring the changesin mass of a substance as a function of temperature or time when the specimen isheated. Figure 2.28 shows common schematic of machine used for TGA. Themeasurement is normally carried out in air or in an inert atmosphere, such as helium


or argon, and the weight is recorded as a function of increasing temperature.Sometimes, the measurement is performed in a lean oxygen atmosphere (1–5 % O2

in N2 or He) to slow down oxidation.As an example for TGA, this analysis method has been adopted to distinguish

the different forms of the carbon deposits (e.g., amorphous carbon and CNTs)according to their different thermal stabilities (McKee and Vecchio 2006). A samplecollected from reactor after reaction accomplished involves carbon formed andcatalytic particles. Nanotubes have a higher thermal stability than amorphous car-bon, and thus, they burn off at higher temperatures. Increasing amorphous carboncontamination in the sample, the temperature corresponding to the peak burn-offrate will shift lower. In addition, after the mass loss has stabilized (after all nanotubeand amorphous carbon product have burned off), the remaining mass gives anindication of the original non-carbon content of the sample (typically the mass ofthe residual catalyst). Figure 2.29 illustrates TGA result of product obtained fromCVD of ethanol. To conduct TGA, a ceramic crucible having certain amount ofsample is placed on the weighing pan inside the equipment and heated to about800 °C with the increasing rate of 10 °C/min. The weight loss of the sample for theperiod of the heating time is automatically recorded and plotted as a function oftemperature.

The techniques used for nanostructures characterization are not limited to thosepresented here, and readers interested to get more information are encouraged torefer the books like Burda et al. (2000), Xu (2001), Cao (2004), Wang (2000a, b).Table 2.1 presents summary of those techniques addressed in this book for betterunderstanding and comparison.

2.2 Transforming Nanotechnology into NanoengineeringThrough Chemical Engineering Principles

Commercial production and applications of nanostructure materials (nanomaterials)have not yet been completely developed. A great majority of scientists and engi-neers are attempting to resolve the challenges posed by synthesis, processing,

Fig. 2.28 Thermogravimetric analysis device


application, purification and characterization of these new materials. The innova-tions and developments in these aspects of the nanomaterials are fueled by theprogress in all fields of engineering, science and technology. However, develop-ment and improvement in the nanomaterials production in large scale lie onengineering principles, in a specific manner on chemical engineering. For instance,design of new manufacturing processes of effective catalysis and improved sepa-ration and purification methods will pave the road for commercial production ofnanostructures. In addition, simulation and modeling of processes help to under-stand and hence to optimize the process including chemical reactions and regen-eration cycle systems. Although processes applied for synthesis of nanostructuresare somehow complicated, these complex processes can be dynamically simulatedand optimized with the aid of computer. The key parameters of processes deter-mined from models can be applied in practice to control the process, for instance, toproduce even more quality in the end products. Accordingly, development oftheoretical framework as well as advanced engineering knowledge to increaseunderstanding of structures and behaviors of nanostructures is strongly required.Besides, a practical framework capable of new process design and improvement inthe performance or controlling the existing processes are vital. In this respect, anintegrated program comprised of theoretical model accompanied with numerical oranalytic solution of the model equations, and comparison with experimental data onboth dynamics and structure of such systems is essential. In conclusion, it is fair tonote that nanotechnology should be transferred to nanoengineering that meansengineering-based effort like transport phenomena must be considered.

Fig. 2.29 TGA illustrations for product obtained from chemical vapor deposition of ethanol usingat T: 600 °C

2.2 Transforming Nanotechnology into Nanoengineering Through … 123

2.2.1 Nanotechnology in Support of General Scienceand Engineering

It is essential to explain interrelation of nanotechnology with the other conventionaltechnologies and sciences at first. There are differences between nanosciences andnanotechnology. These differences are related to how the nanoconcepts and

Table 2.1 Summary of common techniques employed for nanostructure analysis

Technique Characterization parameter Type of information

AES Auger Electron energy spectrum Elemental composition, layer thickness

AFM Local van der Waals force Surface topography map, elasticity,magnetic and electrostatic properties

BET Gas adsorption desorption Surface area and porosity

EDS Electron scattering Elemental composition

EELS Electron scattering Elemental composition, chemical bonding,electronic properties

FESEM Electron scattering Morphology, size, defects

FTIR Atom vibration Elemental composition, chemical bonding,functional groups, molecular composition

HRTEM Electron transmission Morphology, crystalline structure, size, defects

ISS Scattered ion energy and angle Elemental composition, surface structure,atomic distances

Mossbauerspectroscopy

Resonant absorption of y-ray Elemental composition, chemical bonding,crystallinity, magnetic properties

Ramanspectroscopy

Atom vibration Chemical bonding

RBS Energy spectrum of scatteredMeV-ions

Elemental composition, layer thickness

SIMS Mass of sputter-ejected ions Elemental composition, mass spectradetermination of trace elements in ppb

SEM Distribution and energy of scatteredelectrons

Morphology, constituent phases

STM Spatial variation of electrontunneling current

Surface topography, surface electronicstructure, constituent phases

TEM Electron transmission Morphology, size, defects

TGA Weight loss of a sample due totemperature increasing inpredetermined atmosphere

Thermal stability, volatile component

UV-visspectroscopy

Absorption or reflectance of light inthe ultraviolet-visible wavelength

Elemental composition, particle size, opticalproperty

XPS Photoelectron energy Elemental composition, chemical bonding,layer thickness

XRD Scattered and diffracted X-raydistributions

Crystalline structure, space groups, atomicpositions and profiles, layer thickness


nanoproducts are used. In other words, nanoscience discusses basic concepts anddescription of related phenomena to change of materials structure and behaviorwhen one of their characteristic dimensions is in nanosize. For example, in thephysics view point, it is important to understand why quantum properties of asemiconductive materials change when its size is around the nanorange. As anotherexample, variation of suspension color (such as gold particles in the water) is ananophenomenon when the size of particle situates in the nanosize. This variationof size of a material can also cause variation of chemical properties (such ascatalytic effect), and this represents an attractive nanophenomenon for manychemical applications.

In contrast, nanotechnology is different than nanoengineering from the viewpoint of nanophenomena and their process description. In fact, rule of technologyword in the nanotechnology slang causes that description of nanophenomena givesapplied and processed results and therefore gives new products with new propertieswith optimum consideration, but in s descriptive manner in contradistinction tonanoengineering which is more quantitative. As an example, one of the mostimportant and commercial production methods of carbon nano-tubes (CNTs) isCVD. General description of this method is thermal or catalytic decomposition of ahydrocarbon to hydrogen and carbon molecules and then control of depositionarrangement of carbon atoms to a regular structure named tube. Nanoengineersinterest to produce this product in the large scale and also to find new materials withimproved properties using CNT, while distinguish and determination of effectivemolecular scale mechanisms in synthesis of CNT as a nanostructures material tofind its kinetic are interesting for chemists.

There is an increasing emphasis on the development of systems for nanostruc-tures production, which require different aspects of chemical engineering such astransport phenomena, kinetic, thermodynamic process modeling and simulation andso on. In a system or process related to nanomaterials synthesis or production,distribution of sizes, shapes and their purity relates to their capacity to executediverse functions in specific applications (Rolando 2007). Fundamentals associatedwith production of nanomaterials are complicated, and conventional principles arenot able to predict and optimize the systems at nanoscale. As mentioned, whensystems scale down to the nanoscale, the fundamental theory used in larger-scalesystems breaks down because of fundamental differences in the physics. Thebehavior of materials and thus their properties at nanoscale are difficult to predict.

It is well known that almost all physical and chemical properties of systems inthe nanometer region become size dependent. For example, color of a piece of goldremains golden until it has size of microns. However, this golden color changes tored when the size of gold reaches regime of nanometers. Interparticle forces such asvan der Waals (attraction) and electrostatic (repulsion) also affect the behavior ofsuspension consisting nanoparticles. For example, the attractive van der Waalsforce may cause particles to aggregate. At the small scale, forces such as frictionand surface tension often dominate over forces such as gravity. Another example is


in selecting a material for energy conversion. The suitability of a thermoelectricmaterial for energy conversion is based on the figure of merit Z, defined as (Faghriand Zhang 2006)

Z ¼ a2

Rekð2:4Þ

whereα is the Seebeck coefficientRe is electrical resistivity andk is thermal conductivity

Materials with a high value of “Z” are difficult to find in bulk form, and ther-moelectric materials at nanoscale indicate good energy conversion. However, tomanipulate the nanostructures of certain materials, the electron and phonon ther-moelectric transport must be first understood. In this respect, Boltzmann transporttheory can be applied to describe transport of electrons and electron–lattice inter-action to yield the two-step heat conduction model at nanoscale and heat transfer(Chen et al. 2004). Quantum Boltzmann equation explains how particles are kickedinto or out of phase-space elements due to collisions. According to the examplesmentioned, classical physics alone fails in describing the phenomena at nanoscale topredict the behavior of nanostructures, and thus, classical physics begins to giveway to quantum physics in terms of description of physical phenomena. Whenelectrons are confined to nanosized objects, their energy levels change that causealtering the electronic and optical properties of the material (Shong et al. 2010).Nanostructures properties, such as their conductivity, magnetism and so on, can beexplained through properties of electrons. To describe the properties of electrons,the quantum mechanical wave description of matter is applied.

Another imperative difference of dealing with nanostructures compared to otherconventional materials is the strong coupling between structure and flow thatnecessitates a more microscopic view of transport phenomena in such systems. Innanotechnology, systems mostly deal with mixing and dispersion of nanoparticlesand/or nanoparticle growth in fluids that show interaction between nanostructureand flow that determines the final properties and also behavior of the system. Fullunderstanding of relation between nanostructure and flow is a key issue tomanipulate and control ultimate properties of such systems. Nanofluids, dilutecolloidal suspensions of nanoparticles, can be addressed as one of the most popularnanoscale systems. Nanofluids can be transparent, semitransparent and opaquedepending on the properties and concentration of the dispersed particles, and theymay contain a certain amount of surfactants or dispersants to enhance their stability.The popularity of the topic of nanofluids is pertained to observations of enhancedproperties and behavior in heat transfer (Choi and Eastman 2001), mass transfer(Krishnamurthy et al. 2006; Olle et al. 2006) and antimicrobial activities (Zhanget al. 2007). To interpret the experimental observations, a number of mechanisms


have been proposed (Yu and Choi 2003; Wang et al. 2007; Prasher et al. 2006a, b;Keblinski et al. 2002).

The laws of physics are certainly legitimate at the nanoscale, and many physicalphenomena can be adequately described with classical physics (Shong et al. 2010).However, it is vital to obtain structural information of nanoparticle, which can befed to the conventional effective medium theories to give predictive models fornanofluids behavior. The reasons are: first, new phenomena observed when novelmaterials and devices at the nanoscale are made, and second, nanoparticle struc-turing is a dominant mechanism for explanation and prediction of their behaviors ina fluid. For example, theory of Brownian motion has been examined for explanationand prediction of nanofluid behavior. The random movement of a small particle(about one micron in diameter) suspended in a fluid is called Brownian motion.Based on classical physics, Brownian motion is described in terms of diffusionprocesses. Since diffusion depends on temperature, it is fair to note that Brownianmotion is related to the thermal motion of molecules. When a particle is fallingdown in a fluid (Fig. 2.30), terminal velocity for a particle is defined as

vt ¼ 23mgplD

ð2:5Þ

wherem is mass of particle (kg)D is its diameter (m) andµ is fluid viscosity (kg/m s) andg is gravitational acceleration (9.8 m/s2)

Fig. 2.30 A particle is falling through a fluid


mass of particle can be written in terms of its density and volume, then

m ¼ qpD2

6

� �ð2:6Þ

By substituting m in Eq. 2.5:

vt ¼ 19qgD2

lð2:7Þ

vt / D2 ð2:8Þ

Note that the above treatment is only valid for small particles that flow in streamlinewith low velocities. This condition is met when the Reynolds number (Re) is lessthan about 2000, where Re is a nondimensional quantity that describes the type offlow in a fluid defined by:

Re ¼ qvDl

¼ qvl=d� � ¼ Inertial forces

Viscous forcesð2:9Þ

whereρ is fluid densityμ is fluid viscosityv is relative velocity of particle and solid andD is particle diameter

Since the terminal velocity is proportional to the diameter squared (Eq. 2.7), it isclear that small particles fall very much more slowly. As size decreases, the ratio ofinertia forces to viscous forces within the fluid decreases and viscosity dominates.Hence, nanoscale objects moving through fluids are dominated by viscous forces,and their motion is characterized by a low Reynolds number. This means thatnanoparticles sense the viscosity of the fluid much more than particles having fairlylarge diameter. To give a quantitative example, consider a gold sphere (density19,300 kg m−3) with diameter of 1 mm is falling through water (viscos-ity = 0.01 kg/m s). It has a terminal velocity calculated from Eq. 2.7 of about2 ms−1. If the sphere is now 1 μm in diameter, its terminal velocity becomes about2 μms−1. If its radius is further reduced to 1 nm, its terminal velocity drops to2 pm/s.1 Furthermore, at the nanoscale, the effects of individual molecules in thefluid impact significantly the Brownian motion of nanoparticle.

Time evolution of the position of a Brownian particle is best described using theLangevin equation. Langevin equation is a stochastic differential equation in whichtwo force terms have been added to Newton’s second law (Langevin 1908). One

1Pico meter per second.


term represents a frictional force due to viscosity, and the other one is random forceassociated with the thermal motion of the fluid molecules.

In general, Newton’s law of motion in one direction, considering presence of anexternal force (FEXT), Brownian diffusive force (FB(t)) and viscosity (μ) of thefluid, can be written as Eq. 2.10 below. Since friction opposes motion, the firstadditional force is proportional to the particle’s velocity (v) and is opposite indirection. This equation needs to be solved to describe the complete motion of anobject in a fluid.

FEXT þ FB tð Þ � 3plDv ¼ mdvdt

ð2:10Þ

Substituting m as a mass of a spherical particle with diameter D and velocity asv = dx/dt, gives:

FEXT þ FB tð Þ � 3plDdxdt

¼ p6qD3

� � d2xdt2

ð2:11Þ

When particles of nanosize are dispersed in a solution, Brownian motion ensuresthat the particles will move about constantly colliding with each other. Due to thesecollisions and intense affinity of nanoparticles, aggregation is expected in systemconsisting nanoparticles. To obtain a stable dispersion, DLVO theory, named afterDerjaguin and Landau, Verwey and Overbeek, can be found useful (Derjaguin andLandau 1993; Verwey and Overbeek 1999). This theory explains the aggregation ofaqueous dispersions quantitatively and describes the force between charged sur-faces interacting through a liquid medium. When two particles move close to eachother, their electrical double layers overlap, and thus, a repulsive electrostatic forcedevelops. According to DLVO theory, if there is a balance between the repulsiveinteractions of the double layers on neighboring particles and the attractive inter-actions arising from van der Waals forces between the molecules in the particles,there is a well dispersion of particles in a fluid. In case of having a well dispersionof nanoparticles, Brownian model has been used for explanation and prediction ofenhancement in thermal properties of nanofluids (Keblinski et al. 2002; Evans et al.2006). Brownian motion of particles contribute to the enhancement of thermalconvection in a system through two ways, the first is direct contribution due tomotion of particles that transports heat, and the second is indirect contribution dueto micro-convection of fluid surrounding individual particles. However, both directand indirect contributions of Brownian motion for nanoparticles were theoreticallyproven to be negligible. Besides, nanoparticles are often in the form of agglom-erates and/or aggregates, so the Brownian motion is not expected to play a sig-nificant role. Moreover, the enhancement of thermal conductivity in nanoscalesystems was found independent of temperature. Weak dependence of nanoparticlesthermal conductivity on temperature and base liquid viscosity suggests that theBrownian motion of nanoparticles cannot be a dominant mechanism for theenhanced thermal conductivity of nanofluids (Ding et al. 2007).


The relationship between convective heat transfer behavior of nanofluids andtheir rheological behavior has been extensively studied (Kwak and Kim 2005;Prasher et al. 2006a, b; Ding et al. 2006; He et al. 2007; Namburu et al. 2007; Chenet al. 2007, 2009a, b; Saeedinia et al. 2012; Yang et al. 2012; Hachey et al. 2014).In these studies, nanofluids exhibited either Newtonian or both Newtonian andnon-Newtonian behavior depending on particle size and shape, particle concen-tration, base liquid viscosity, solution chemistry-related surface layer andelectro-viscous effects.

For shear-thinning nanofluids (pseudo-plastic), the shear viscosity approaches aconstant at high shear rates. Such a constant is termed the high shear viscosity andis very relevant to the convective heat transfer applications where heat transferfluids are often in vigorous motion and subjected to very high shear.

In high shear rate regions where the shear viscosity approaches a constant value(called high shear viscosity), the viscosity scales with temperature in a similarfashion to that of the base liquid. The experimental results also show that both thehigh shear viscosity of nanofluids and the base liquids follow well the classicalVogel–Tammann–Fulcher (VTF) Eq. 2.12, which describes temperature depen-dence of viscosity (Bird et al. 2002). However, the relative increment in high shearviscosity at a given particle concentrations is almost independent of temperature(Chen et al. 2007, 2009a, b).

log l ¼ Aþ B=ðT � T0Þ ð2:12Þ

whereT is temperature (°C or K)μ is fluid viscosityA, B, and T0 are constants

The presence of nanoparticles in fluids increases the high shear viscosity, and theextent of increment depends on nanoparticle shape and volume fraction. Givenother conditions, the high shear viscosity of nanofluids containing rod-like particlesis much higher than those containing spherical nanoparticles. The experimentallyobserved rheological behavior of nanofluids containing rod-like particles cannot beexplained purely by the shape effect, and particle aggregation should be considered.Nanofluids containing spherical nanoparticles are less complicated than rod-likenanoparticles. Regression of the measured high shear viscosity of nanofluids con-taining spherical nanoparticles gives a binomial relationship (Chen et al. 2007):

l ¼ l0 1þ 10uþ ð10uÞ2� �

ð2:13Þ


where µ and µ0 are the shear viscosity of nanofluids and base liquids, respectively.These authors classified nanofluids as:

Dilute nanofluids (0 < φ < *0.001) containing well-dispersed nanoparticleswith no discernible shear-thinning behavior. Shear viscosity can be described bythe Einstein equation.Semi-dilute nanofluids (*0.001 < φ < *0.05) containing aggregates of nano-particles with no obvious shear-thinning behavior.Semi-concentrated nanofluids (*0.05 < φ < *0.10) containing aggregates ofnanoparticles with clear shear-thinning behavior.Concentrated nanofluids (φ > *0.10) have interpenetration of aggregates.

The concentrated nanofluids are out of the normal range of nanofluids for heattransfer applications.

Comparing the results of these studies shows that convective heat transfercoefficient was enhanced or deteriorated depending on the nanofluids characteristicsincluding the shape and size of nanoparticles and viscosity of fluid. Nanofluidscontaining tubular or rod-like nanoparticles often give a higher enhancement ofconvective heat transfer coefficient in comparison with spherical or disk-likenanoparticles. Nanofluids made of less viscous liquids give a higher heat transfercoefficient in comparison with those made of highly viscous liquids. For caseswhere heat transfer enhancement is observed, the convective heat transfer coeffi-cient generally increases with increasing flow rate or increasing particle concen-tration, and the enhancement may exceed the extent of the thermal conductionenhancement. No clear trend has been found in the effect of particle size on theconvective heat transfer coefficient of nanofluids. If nanofluids are non-Newtonian,then the analysis that leads to the constant Nusselt number is invalid. Nusseltnumber, a dimensionless quantity, describes the ratio of conductive thermal resis-tance to the convective thermal resistance of the fluid (Eq. 2.14).

Nu lð Þ ¼ Convective Heat TransferConductive Heat Transfer

¼ hxk

ð2:14Þ

wherex is the characteristic length (m)k is the thermal conductivity of the fluid (W/m K)h is the convective heat transfer coefficient of the fluid (W/m2 K)

As discussed earlier, even very dilute nanofluids can be non-Newtonian, par-ticularly for the water-based nanofluids. Given a nanofluid and pipe geometry, heattransfer enhancement in the turbulent flow regime is less significant than that in thelaminar flow regime. No sufficient quantitative information, however, is available inthe literature that can be used to infer the dominant mechanisms for heat transferenhancement/deterioration under convective and boiling heat transfer conditions,where many controversies remain and require further research. The effect of tem-perature depends on the relative importance of the Brownian diffusion and the shear


flow convection. At low shear rates and high temperatures, the Brownian diffusioncan be stronger in comparison with the convection; a stronger shear-thinningbehavior is expected (Chen et al. 2007).

Current approach for modeling transport phenomena is based on continuumstates, the materials are assumed to be continuous, and the fact that matter is madeof atoms is ignored (Bird et al. 2002). However, traditional Navier–Stokes equationand the energy equation based on the continuum assumption have failed as thespatial scale of flows approaches the molecular mean free path, and hence, thisapproach is not valid for systems at nanoscales (Faghri and Zhang 2006). Thediscreteness of those systems that involve flow in micrometer- and nanometer-scalechannels is important (Nie et al. 2004). For example, continuum approaches candescribe fluid motion in nanoporous solids using hydrodynamic models andapplying the appropriate boundary conditions; however, the precise relationbetween the fluids motion in nanopores and the details of the interactions of thefluid with the pore wall is still an open problem. A key parameter characterizing theapplicability of the continuum equations, Navier–Stokes equation, is the Knudsennumber, defined as the ratio of the molecular free path to the transverse dimensionsof the system (Quirke 2006). A thorough understanding and modeling of moleculartransport through nanostructures are essential to the logical design of new materialsand devices for various purposes like separation processes, nanofluidics, andhigh-throughput characterization, analysis and sequencing. Non-continuum mod-eling of transport phenomena, based on atomistic descriptions, such as moleculardynamics (MD) simulations, is an appropriate approach for describing moleculartransport through nanostructures. Nie et al. (2004) developed a hybrid method, inwhich continuum fluid dynamics and molecular dynamics are combined, fordescribing fluid dynamic at nanoscale. The continuum Navier–Stokes equation wasused in one flow region and atomistic molecular dynamics in another. The spatialcoupling between continuum equations and molecular dynamics is achievedthrough constrained dynamics in an overlap region (Fig. 2.31).

In a practical application of using nanosystems in blood, the motion of thesesystems in a fluid is complex and difficult to control that makes design of thepropulsion system a major engineering challenge. Brownian motion would cause aconstant random shaking, and surface forces at the nanoscale are significant,resulting in sticking of the system to any surface that it comes into contact with.Nevertheless, these effects can be useful by getting insight into the nature. Livingorganisms with nanodimension, like viruses, are able to find their way into cells. Ifmolecules with sticky and non-sticky areas are designed, then the agitation causedby Brownian motion will eventually lead to molecules sticking together in verywell-defined ways to form rather complex macromolecular structures (Shong et al.2010). This mode of assembly is known as self-assembly, as discussed before inSect. 2.1.1.2. The complications of interactions between nanostructures and flow, asaddressed in this section, offer difficult challenges that necessitate knowledge andexperiences of who is familiar with the material processing and engineering prin-ciples through fluid mechanics, i.e., chemical engineer. In general, the laws oftraditional physics describe our macroscopic world. The laws of quantum physics


give a universal image of our world. These laws are essential to our understandingof the nanoworld.

2.2.2 An Industrial Perspective Research Challengesin Nanotechnology

Nanotechnology is not a single industry or one a new phenomenon, like theInternet, but offers new possibilities which will likely take place at industrial giants.Any activity in the field of nanotechnology falls in one of the areas addressedbelow,

1. Synthesis of nanoscale building blocks, named nanostructures like nanoparti-cles, nanotubes and nanofibers

2. Fabrication and/or processing of nanoscale building blocks for a desired purposelike nanofluids

3. Incorporating nanoscale building blocks into final product, like nanocomposites,nanodevices and nanosensors.

The first two types can be viewed as individual industries where the productscoming from them are the substrates for the industries in the third group. In thefollowing, products obtained from each of these three groups are callednanotechnology-based products. Each process involved in producingnanotechnology-based products is strongly required to be commercialized regardingproducing commodity with unique performance and customer benefits at a rea-sonable cost. For example, CNTs have been found a wide range of applications as

Fig. 2.31 Schematic of the hybrid method. The continuum description is used in the shadowedregion and the atomistic description is used in the dotted region. In C → P, continuum solutionsprovide boundary conditions for MD simulations, and in P → C, atomistic solutions provideboundary conditions for continuum simulations (Nie et al. 2004)


dressed in Sect. 2.1.1. Nevertheless, its commercial synthesis and its using asbuilding blocks to produce new materials are still an open problem (Homma 2014;Oueiny et al. 2014; Mehra et al. 2014). CNTs can be produced via different routes;however, CVD has been found as a commercial method. Although large number ofresearch works conducted on synthesis of CNTs using CVD method, the control-lable procedure is still under observation. Synthesis of CNTs is categorized in thefirst group, and when CNTs are dispersed in a liquid, a nanofluid is produced(second group). Nanofluids containing CNTs have been indicated enhanced thermalheat transfer; however, the effects of parameters, including viscosity of fluid, dis-persion pattern of nanotubes, etc., are still under study (Aravind et al. 2011;Walvekar et al. 2012; Leong et al. 2014). The nanofluid then can be used inpreparing nanocomposites applicable in various applications, for example, in highheat-generating system (Jajja et al. 2013) or as a reinforcement of cement slurry(Nasiri et al. 2013). Preparing nanocomposites is categorized in third group andaccording to the recent articles, it still needs improvement toward full commer-cialization (Liu and Kumar 2014). An essential key to commercial production ofCNT itself and the products made from CNTs, it is vital to reduce the operationcosts as well as control the basic properties of the end products during the manu-facturing processes. Academic scientists, engineers and industry visionaries havebeen keen to prospect for novel strategies and fabrication methods ofnanotechnology-based products, in the hope to transform the strategies from alaboratory-scale approach into a mainstream process. However, for realization ofwide industrial applications, much research work is required to achieve controlledand large-scale synthesis of nanostructures with reasonable price. The most chal-lenging issues for commercialization of nanotechnology-based products include:

A. Development of robust methods for industrial manufacturing of certainnanostructures

Chemical engineering unit processes have significantly contributed to successfulcommercialization of many discoveries in the chemical industry. Examples includewell-known industrial petrochemical and polymers like Nylon, Teflon®, and manyother industrial materials improved peoples’ lifestyle around the world. Unitoperations will be equally important for the commercial production ofnanotechnology-based products. Unit operations mostly used in production ofnanotechnology-based products include milling, mixing and reaction. For example,gold nanoparticle is commonly synthesized in mixing-reaction system (Yang et al.2010; Zhao et al. 2013), FePt nanoparticles were produced by high-energy ballmilling (Velasco et al. 2012) and nanosilicon carbide-reinforced aluminum wasproduced by high-energy milling (Kollo et al. 2011). In industrial manufacturing, itis basically impractical to get a product from raw materials by just carrying out asingle unit process. It is essential to make the raw materials ready for the mainprocess through upstream processes (pre-treatment), and the materials after mainprocess generally need to go through downstream processes (post-treatment) for thedesired product(s) to be separated from unwanted materials. The pure main productin required form is obtained after downstream processes. Process design, including


all upstream, main and downstream process, describing how a product, includingnanotechnology-based products, is produced, can be addressed as the first steptoward industrialization of the process. The research activities conducted in thelaboratories are prerequisites, but the scale is small to obtain appropriate parametersfor the design of commercial units. Promising laboratory works will take time toscale up and bring costs down; however, scale-up requires a leap in manufacturingtechnology. For example, in commercial production of nanostructures, like CNTs,nanogold, nanosilver, as a building block of other products, it is essential to controlthe size and morphology of these nanostructures. In this regard, size-controlledsynthesis of nanoparticles has been developing (Yang et al. 2010; Wei and Liu2010).

Other limitations pertaining to large production of nanotechnology-basedproducts include low efficiency of process and extensive energy consuming. Forexample, synthesis of nanostructures using CVD needs temperature above 600 °C(Danafar et al. 2009). Development of micro-unit processes and then advancedtechnologies for obtaining reliable predictive information are under intense inves-tigation to reduce unwarranted development costs (Ehrfeld et al. 2000; Hessel et al.2006).

B. Active control of the structure and composition of the product with desiredmacroscopic properties

As mentioned above, the great challenge in synthesis of nanotechnology-basedproducts is how to produce nanostructures with the properties that can be predicted,tailored and tuned beforehand. For example, there is an explosive growth in pro-duction of nanocomposites with enhanced or novel properties like high thermalcomposites, scratch-resistant coating and self-cleaning coating; however, theend-use property of composite strongly depends on the nanostructure propertiesused as well as processing conditions applied for producing the nanocomposite. Toget deeper insight, let’s mention the famous example of CNTs, which can haveeither conductive or semiconductive property according to the configuration ofcarbon atoms. Depending on the chirality of single-wall CNTs, they may exhibiteither metallic or semiconducting properties. Generally, the percent fraction of eachtype can be affected by growth conductions. However, the electrical characteristicsof nanotubes are neither controllable nor selectable on a manufacturing scale. Smallchanges of the experimental parameters, specifically nature and concentration of thereactants and temperature, lead to considerable alteration of the resulting molecularassemblies. In the next step, nanocomposite preparing well dispersion of nanotubesis critical to reach the desired properties. Proper control of the properties andresponse of nanotechnology-based products need a comprehensive understandingof mechanisms involved in product synthesis. From an industrial standpoint, it issignificant to be able to describe the systems/operations used in the chemicalindustries and predict their behavior if any change is applied to the system. Thediversity of nanotechnology-based products means that there is not going to be asingle answer. However, developments of simulation methods at the molecularand atomic level may allow a better understanding of phenomena at nanoscale.


They provide powerful tools to interpret interparticle forces as well as predictnanostructures behaviors, as described in Sect. 2.2.2. Although all results obtainedfrom modeling and simulation have to be experimentally validated, with theincrease in computing power and software development, they are providing realreduction in time and cost in development and commercialization of nanotech-nology (Uddin et al. 2012). Accordingly, the processes involved in producingnanostructures need sensing and actuation of matter at the nanoscale. Such a del-icate control requires advanced smart sensing and characterization systems.

C. Development of process instrumentation

Online measurement and control of process parameters are a prerequisite forobtaining high production yields of the desired product properties. This taskbecomes more complicated due to complexity of processes involved in the for-mation of nanostructures and dependence of end-use properties of nanostructurewith their size, atomic configuration and chirality. For example, spherical nano-particles in a fluid indicated different thermal conductivity and thermal diffusivitycompared to the presence of cylindrical nanoparticles in the same fluid (Zhang et al.2007). A nanostructured filler with different size influences the morphologies andproperties of membrane (Yang et al. 2007). Different size of nanoparticles inFe/Cu/La catalyst showed different product distribution and kinetic parameters inFischer–Tropsch synthesis (Pour et al. 2010). In this regard, knowing processkinetics and the related issue of mechanisms are a necessary prelude to model andcontrol changes in progress (Texter and Tirrell 2001). Observing the evolution ofprocesses necessitates acquisition of accurate kinetic data in real time, and hence,researchers put their attention toward developing in situ measurement of nano-structure during their formation process to observe and control changes in progress(Zhao et al. 2003; Puretzky et al. 2005; Li-Pook-Than et al. 2010). The acquisitionof such data will impel development of a sophisticated instrumentation that pro-vides informative, online measurement to observe and control changes in progress.

D. Characterization of nanostructures

Bulk characterization of processing materials, in which length scale of meters tomicrometer is measured, is not sufficient for processes involved in production ofnanotechnology-based products. Properties and characteristics of nanostructures,specially, cannot be based on averaged bulk properties, and in-depth character-ization of the materials at the atomic and molecular scale configurations is vital.According to the type of nanotechnology-based products, characterization can beclassified into three main levels:

1. Primary characteristics of nanostructures like particle size distribution, defectsand impurity in nanostructure, morphology and surface of nanostructure

2. Interfaces and boundaries such as interparticle forces, boundary defects3. End-use properties like electrical, mechanical, thermal, antimicrobial.


Without truthful structural information, products comprising nanostructurescannot be engineered. Properties of nanoscale building blocks determine end-useperformance of the product. For instance, controlling and manipulating confor-mation and position of polymer molecules with nanometric resolution on surfacerepresent a major industrial challenge in sensors or controlled molecular assemblies.Fabrication of reliable, high-performance product requires an increased under-standing of the role played by the microstructure and the surface. Presence ofdefects and impurity is particularly important for the envisioned applications of thenanotechnology-based products. Therefore, there is a strong demand to developnew tools of characterization that provide accurate and reliable structural infor-mation. The use of online size measurement and control of today’s particulateproducts manufacturing processes will be essential to control nanoparticle forma-tion. Online nanocharacterization tools must be used, and therefore, their devel-opment needs to be actively pursued. In this framework, instrumentation in all itsforms (laser, molecular beams, NMR, TAP, X-ray diffraction, mass spectrometry,tomography, trajectography, computers, etc.) should be of great help in obtaining anaccurate picture of chemical transformations and in understanding cellularmechanism.

E. Sustainability, environmental, health and safety aspects

It is important for environmental health and safety to keep up with the rapid growthof the nanotechnology industry. A major contemporary industrial challenge ismanufacturing products while eliminating or substantially reducing the detrimentalenvironmental consequences of the processes adopted. Nanotechnology has greatpotential to transform science and industry in the fields of energy, material, envi-ronment and medicine. However, in socioeconomic viewpoints, a great challengefor nanorelated industries is to ensure that their new products are safe in the humanbody and in the environment. Accordingly, novel and sustainable approaches forcommercial production of nanostructures should be taken into account. In thisregard, economic and environmental impacts of the process are well examined andunderstood. The transparency of health, safety and environmental impacts ofnanotechnology in living systems should be at the forefront. Toxicity of nano-structures is in doubt and reliable and standardized methods for rapid assessment oftheir toxicity under various exposure, dosing and biological conditions are stronglyrequired. Effective methods must be established among researchers, developers andregulatory bodies to facilitate transfer of research results that reliably assess toxicityof a nanostructure to ensure product safety for industrial and medical users.Regulatory standards should be developed which ensure that precautions are takenin any commercialization development to provide a consistent, quality product tothe market with no disadvantage to any group of people. Pure research is moresuitable to universities or governmental organizations that the results of researchactivities are open to the public. On the other hand, improved activities may bemore efficiently performed by the sectors motivated by profit and based on theprinciple of nondisclosure of results. However, the real outcomes of industry mustbe tested in real society through trial and error.


Our surroundings are rapidly changing with the developments in science andtechnology. Already, safety and a sense of security cannot be obtained withoutconsciousness, that is, consciousness and investment have become indispensable.The expectations from science and technology for constructing a peaceful societyare very huge. In order to solve the problem related to the peace of mind at a higherlevel, it is necessary to consider the safety provided by science and technology andthe need for a sense of security together. Therefore, many sciences must worktogether. The inclusion of psychology, which regards the sense of security and theprobability terms, could be one such joint initiative.

2.2.2.1 Chemical Engineering Approaches to NanostructuredMaterials Manufacturing

Chemical engineers are playing impressive roles in developing a new science basedon phenomena, structures and potential applications where understanding atnanometer length scales is crucial. The areas in nanotechnology that chemicalengineers are making major contributions are addressed as:

• Development of new and novel products like

– Catalysts– Multicomponent composites– Natural nanostructures– Sorbents for environmental contaminants

• Innovation in processes for synthesis of nanotechnology-based products(Recall that nanotechnology-based products include both synthesis of nano-structures and incorporation of these nanostructures to produce new productslike nanocomposites and nanosensors)

• Process optimization for nanotechnology-based products• Mathematical modeling, kinetic study and thermodynamic study of processes

involved in synthesis of nanostructures• Transport phenomena investigation

Whether this new science base leads to important technological developmentsdepends even more on chemical engineers. Many laboratories have demonstratedcontrollable methods based on bottom-up approaches as a manufacturing procedurefor nanomaterials. The high level of scientific activity in this field has createdseveral kinds of major opportunities for chemical engineering research. The overallopportunity is a shift from laboratory techniques and provocative demonstrations ofstructure to practical, larger-scale, applied processing methods. This move requiresseveral achievements in chemical engineering, including the precision manufac-turing of precursors, expansion of chemical engineering ideas of molecularchemistry to supermolecular chemistry from a bonding, kinetics and mechanisticpoint of view and a focus on self-assembly processes that can be scaled up to


interesting levels. This leads to a rich array of research and process developmentissues for chemical engineers.

Understanding and controlling nanotechnology unit operations will be equallyimportant for the commercialization of nanotechnology. Platforms and processtechnologies in engineering nanotechnology laboratories consist of synthesis ofnanostructures and incorporate these nanostructures as building blocks into finalproducts like coatings, sensors and drug delivery systems (Roco 2004).

Engineering research and education, including chemical engineering, play a keyrole in nanomaterials manufacturing, and this role will even expand in the futurebecause of its integrative, system approach-oriented and transforming characteris-tics. As the degree of complexity of systems increases at the nanoscale, and variousdisciplines of science and engineering converge, chemical engineers role will bemore critical. The rudimentary capabilities of nanotechnology today for systematiccontrol and manufacture at the nanoscale are envisioned to evolve in four over-lapping generations of new nanotechnology products with different areas of R&Dfocus (Roco 2004).

2.2.2.2 Kinetic Approaches of the Reaction

Kinetics of chemical reaction is a quantitative description for the rate of chemicalreaction occurred and the rules of affecting parameters on this rate. The kinetics of achemical reaction is presented as mathematical correlations that show fundamentalaspects of the reaction pathways. This concept increases our ability to develop newand better ways of achieving desired chemical reactions, improve the yield ofdesired products or even develop a better catalyst for a specific reaction. From thechemical engineering viewpoint, the kinetic of chemical reaction is an essential toolto design chemical reactors as help the chemical engineer to optimize the reactorsize by adjusting the reaction rate. Generally, the size of reactor is related to theamount of reactant consumption and the rate of this consumption (or amount of theproduct and the rate of product generation via the chemical reaction). The quan-titative description of the kinetic is provided by several theoretical manners such asphysical chemistry rules and even quantum mechanics or molecular dynamics. Withtoday’s powerful computers, however, the kinetic studies have to be performedexperimentally to find a simple description of the rate and/or verifying the theo-retical reaction rate (Missen et al. 1999).

In experimental activities for kinetics study of a reaction, a chemical reactor isused to carry out the reaction and to obtain the rate of the chemical reaction.Mathematical description of the chemical reaction rate is usually equation thatdescribes functionality of the rate with the various factors such as the componentsconcentration and temperature. In chemical reaction engineering (CRE), theinformation obtained from kinetics is a means to determine size of reactor as well asother aspects such as equilibrium, product distribution and thermal effect. Kinetics,however, does not provide all the information required for this purpose, and heat,mass and momentum balances are also needed to describe the effects of other


factors and problems such as mixing, heat treatment, diffusion and mass transferdiffusion.

A system may be considered in three levels of size to compare the nature ofkinetics. These levels are as follows:

1. Microscopic or molecularA large set of reactant molecules characterized by a definite value of concen-tration, temperature, pressure and density at any time;

2. Local macroscopic—for example, one solid particle reacting with a fluid, inwhich gradients of component concentration, temperature, etc., within the par-ticle are detectable.

3. Global macroscopic—for example, in reaction of a fluid with a collection ofparticles as a bed, in addition to local gradients within each particle, there maybe global gradients throughout a bed of the reactor, from particle to particle andfrom point to point within the fluid.

These levels are illustrated in Fig. 2.32. Levels (1) and (2) are domains ofkinetics in the sense of focusing on the mechanism and the rate of reaction con-nected to the stoichiometric and equilibrium constraints. The level (3) providessufficient information about overall behavior of the reactor in this level. Thisinformation is required to make decisions about design and operation of reactorsfrom the economical view point. In spite of this matter, it is possible under certainideal conditions at level (3) to make the required decisions based on informationavailable at level (l), or at levels (1) and (2) combined.

At the molecular or microscopic level (Fig. 2.32), chemical change involves onlychemical reaction. At the local and global macroscopic levels, other processes may

Fig. 2.32 Different levels inkinetic analyses of a system(Missen et al. 1999)


be involved in change of composition. These are diffusion and mass transfer ofspecies as a result of differences in chemical potential between points or regions,either within a phase or between phases. The term “chemical engineering kinetics”includes all of these processes, as may be required for the purpose of describing theoverall rate of reaction. Yet another process that may lead to change in compositionat the global level is the mixing of fluid elements as a consequence of irregularitiesof flow (nonideal flow) or forced convection. Still other rate processes occur that arenot necessarily associated with change in composition: heat transfer and fluid flow.Consideration of heat transfer introduces contributions to the energy of a systemthat are not associated with material flow, and helps to determine T. Considerationof fluid flow for our purpose is mainly confined to the need to take frictionalpressure drop into account in reactor performance. Further details for quantitativedescriptions of these processes are introduced as required.

The rate of reaction is defined for a component involved in a reaction either as areactant or as a product. The situation of a reaction may vary from a system toanother system. The situation of a reaction system is defined as the reaction phase(single or multiple) and/or constant of variable properties (e.g., component con-centration, pressure, temperature and density) is changed with respect to position atany given time.

The reaction rate is negative if it is described based on consumption of a reactantcomponent, and it is positive if it is based on a product. The rate of reaction withrespect to a component (for example, A), denoted as RA, is an extensive factor rate offormation of A. The intensive form of the reaction rate (denoted as rA) is the ratereferred to a specified normalizing quantity such as the reactor volume or the mass(or surface) of catalyst. Since the rate rA does not depend on the size of system, it maybe considered to be the “point” or “intrinsic” rate at the molecular level (level 1).The two rates are related by the system size (such as the reactor volume as nor-malizing quantity) where for a uniform system (such as well-stirred tank)RA = rAV and for a nonuniform system (such as tubular plug flow) dRA = rA dV.

The rate of chemical reaction with respect to any other component involved inthe reaction is related to rA directly through reaction stoichiometry for a simple,single-phase system, or it may require additional kinetics information of a complexsystem.

General Rules

The rate law of chemical reaction is a function of a number of parameters. The mostimportant ones are:

1. The nature of the species involved in the reaction2. The component concentrations (reactant or product), the reaction rate is

improved with increasing the reactant concentration and decreasing the productconcentration usually.


3. Temperature, usually temperature increase causes faster reaction rate. Althoughthere is exception such as the oxidation of nitric oxide to produce nitric acid, inthis case, the rate decreases as T increases.

4. Catalytic activity, the rates of many industrial reactions are improved with usingthe desirable catalyst.

5. Nature of contact of reactants: The mixing of the reaction system causes thebetter contact of the components. Sometimes, the intimacy of contact of reac-tants can affect the rate of reaction oppositely.

6. Presence of foreign exciting factor: Today using the foreign source of energy toexcite the reaction is an interesting subject in many researches. The rate ofchemical reaction is affected with irradiation of the reaction media withmacro-wave, ultrasonic wave, infrared or ultra-violet light, etc.

Dependency of the chemical reaction rate to the mentioned factors is expressedmathematically in the form of a rate law, as following general form of rA = rA(concentration, temperature, catalyst activity, etc.)

The exact derivation of this relation is too complex and is a multidisciplineactivity which is related to different fields such as thermodynamic, transport phe-nomena, physical chemistry and kinetic. This causes to use approximate expres-sions which have to valid with the experimental data. Fortunately, rate of most ofthe reactions obeys simple mathematical functionality which is semi-experimental–semi-theoretical relation (such as power law functionality), but derivation ofparameters of this simple relation needs specific experimental data.

Kinetic Model for a New Process

The concept of kinetic in a process indicates rate of the process as a function of thestate variables. When a reaction takes place in a process (a system), generally themain step defining the process rate is the chemical reaction. Therefore, kinetic ofchemical reaction is an important topic in the study of any system. In a processinvolved for the synthesis of a nanomaterial via chemical reaction in both gas andliquid phase, multiple reactions, as main and side reactions, usually take place. Inthese systems, beside the chemical reactions, there are other steps affecting the rateof nanoproduct synthesis. This means that synthesize rate of the nanoproduct is toocomplex to be discovered and determined in single step. Several scientists andresearchers have tried to find the mechanisms involved in the nanomaterial syn-thesize. Although several variables affect the kinetics of nanomaterial formation,they can be divided into three types such as the three levels mentioned earlier(Fig. 2.32):

• Type 1: reactant transfer from bulk to surface, mechanism involves the materialto come from bulk to surface by a mechanism such as diffusion, adsorption andfilm transfer. These mechanisms are observed on transfer of the product (andalso the reactants) from the surface to the bulk phase. There is a global


macroscopic viewpoint to consider these mechanisms. Generally, the masstransfer phenomena can define these mechanisms and their rates.

• Type 2: conversion of reactant to product, mechanism of chemical reaction.A local macroscopic level describes this type of mechanism.

• Type 3: atom/molecules of product move to produce nanostructure, mechanismsof surface phenomena such as nucleation, growth, agglomeration and sinteringtake place. A microscopic level of consideration is involved in thesemechanisms.Figure 2.33 shows these mechanisms schematically.

Many investigations have been focused on the determination of kinetic of eachsteps and influence of the conditions on the rate steps. Comparison of the inves-tigation results shows that many parameters have effect on the rate functionality andthe rate constants. In other word, it is impossible to relate a definite kinetic law to ananostructure synthesis system without giving special attention to its specificconditions. The main parameters affecting the kinetic law can be classified asfollowing groups:

• Type, size and structure of nanostructure• The operating conditions: such as temperature, pressure, time and concentra-

tions of reactants• The catalyst properties such as composition, surface chemistry and structure• The synthesis process and type of reactor• Type and concentration of the raw materials.

This variety of parameters affecting the synthesis kinetics is because of multipleand complex mechanisms involved in the synthesis. It is commonly accepted thatthe thermogravimetric analysis combined with mass spectroscopy (TGA-MS) is apowerful coupling technique to study the growth kinetic of CNTs in situ, becausevariation of the sample weight, temperature and enthalpy with time during thereaction can be collected as the experimental measurement results online in TGA.

Fig. 2.33 Different mechanisms affecting on the nanomaterial synthesis


Then, the MS instrument indicates variation of type and composition of the gaseousoutlet stream with time online. These data can be analyzed using models to find thekinetic law of the reaction (Zhang et al. 2014).

Some Examples on Kinetic Study

To find the explanation, let us discuss this concept using an example. The CNTssynthesis process is one of the most important processes in nanotechnology thatattracted a great number of attentions to discover facts about its synthesis. Torealize various applications of CNT, it is necessary to control the structure of theCNTs. CVD is considered as one of the techniques that enable control of structureof CNTs. Experimental data suggest that different parameters affect the morphologyof the resulting carbon products as well as parameters of the CNTs. Thoughnumerous attempts to control CNT structure have been made, the effect of theseparameters on the yield and structure of CNTs are still not clear, and the observeddependences need to be explained. Many attempts are focused on finding thekinetics of CNT synthesis to find effect of the different parameters on the CNTproperties.

There are several main steps from carbon of hydrocarbon precursor in the bulkphase to carbon of CNT on the catalyst surface. These steps are mentioned below,respectively.

1. Mass transfer of the hydrocarbon gas from bulk to the external surface ofcatalyst support

2. Intraparticle diffusion of the hydrocarbon gas within the catalyst3. Adsorption of the hydrocarbon gas onto the catalyst surface (active site)4. Reaction on the catalyst surface to produce carbon molecules5. Dissolution of carbon molecules in the catalyst6. Supersaturation of the catalyst with carbon7. Nucleation and growth of CNT8. Desorption of hydrogen and other gaseous products from the catalyst surface9. Diffusion of the gaseous products from the catalyst surface to the external

surface of catalyst support10. Mass transfer of gaseous products from support to the bulk phase

Each step mentioned above has individual rate affecting the overall formationrate of CNT. Comparison of rate of these steps helps us to neglect some of steps.The generally accepted main steps of CNT synthesis in catalytic CVD processinclude the following steps:

(a) Adsorption/desorption, diffusion and gas-phase transformation of precursormolecules on the catalyst surface,

(b) Decomposition of precursor on the catalyst surface or in the gas phase,(c) Diffusion of carbon through the catalyst surface,(d) Nucleation and growth of CNTs.


Theoretically, any hydrocarbon is able to be used for CNT synthesis. This mayhave been show generally as following overall relation:

CxHy ! xCCNT þ y=2H2g

Since different hydrocarbons are characterized by different sticking coefficients anddecomposition energies, the carbon supply rate onto the catalyst surface can bechanged (Lebedeva et al. 2011). It is generally accepted that the growth is moreimportant to control and predetermine the CNT structure. It is not clear whether thegrowth kinetics can be modeled as transport limited or reaction limited. Transportlimitations can be further classified into factors such as the supply of carbon to thecatalyst from the gas phase, the diffusion of carbon through catalyst particles andthe diffusion of carbon over the nanotube surface. Extensive efforts have been madeto study the dependence of CNT growth on various parameters such as temperature,pressure, time, carbon source and catalyst system. Numerous aspects of growth andnucleation are not well understood, such as critical diameter, growth terminationand root-growth versus tip-growth mechanisms (Kwok et al. 2010). The later factoris shown graphically in Fig. 2.34.

For example, decomposition of ethanol on catalyst active site (S) of F2O3/MgOto form CNT is represented as (Kwok et al. 2010):

Reaction 1: C2H5OHþ 2S ! 2C � Sþ H2Oþ 2H2

As a simple model of CNT growth, in the first possible reaction, C·S reacts with theroot carbon of the CNT to increase the number of crystalline carbons on the nanotubechain by one (reaction 2). This reaction shows mechanism of CNT growth on thecatalyst surface. In the second, C.S may react with any carbon along the chain toincrease the mass of amorphous carbon (reaction 3). These reactions are shown as:

Fig. 2.34 Comparison of tip-growth (a) and root-growth (b) mechanisms of CNT growth


Reaction 2: gC � S ! CgðCNTÞ � Sþ g� 1ð ÞS

Reaction 3: dC � S ! CdðAmorÞ � Sþ d � 1ð ÞS

In these equations, an arbitrary g number of nanotube carbons and d number ofamorphous carbons are shown as Cg

CNTð Þ CdAmorð Þ and , respectively. In the etching

step, water from the decomposition of ethanol can react with any form of carbon inthe system to produce H2 and CO:

Reaction 4a: CgðCNTÞ � Sþ H2O ! Cg�1

ðCNTÞ � Sþ H2 þ CO

Reaction 4b: CdðAmorÞ � Sþ H2O ! Cd�1

ðAmorÞ � Sþ E2 þ CO

The chemical reaction rates of these reactions are defined as the power law kinetic.

R1 ¼ K1CC2H5OHC2S ð2:15Þ

R2 ¼ K2C2C�S ð2:16Þ

R3 ¼ K3CC�S CC�S þ CCNT þ CAMORð Þ ð2:17Þ

R4 ¼ K4CH2O CCNT þ CAMORð Þ ð2:18Þ

The kinetic parameters of these reactions are calculated and reported in Table 2.2using the Arrhenius functionality (K = A exp−E/RT).

To define variation of the components concentration during the reactions, we canwrite the mole balance on the components and obtain following relations:

dCC:S=dt ¼ 2k1CC2H5OHC2S � K2C2

C�S � K3CC�S CC�S þ CCNT þ CAMORð Þ ð2:19Þ

dCCNT=dt ¼ K2C2C�S � K4CH2OCCNT ð2:20Þ

dCAMOR=dt ¼ K3CC�S CC�S þ CCNT þ CAMORð Þ � K4CH2OCAMOR ð2:21Þ

where CS0 = CS + CC.S + CCNT + CAMOR and CS0 is constant. The concentration ofwater and ethanol ðCC2H5OH and CH2OÞ is controllable by adjusting concentration of

Table 2.2 Kinetic parameters of the reactions 1–4 (Kwok et al. 2010)

Reaction Activation energy (E) Ki/mol Pre-exponential factor (A) M6/mol min

1 128 ± 12 102.0±0.6

2 243 ± 3 1012.7±0.6

3 254 ± 5 1012.1±0.6

4 260 ± 80 1014.8±2.0


these components in feed stream of the reactor. It is important that the number ofcatalyst active sites (CS) is decreased because of amorphous/structured carboncoverage. This is a concept of catalyst deactivation. Basically, the concentration ofcatalyst active site (CS) can be defined as:

CS ¼ CS0 � a tð Þ ð2:22Þ

where a is activity of the catalyst surface. A simple model can be defined as onlyconsiders catalyst deactivation by formation of the coke that covered the activesites. This coke can be partially removed from the catalyst surface by gasificationwith the hydrogen present in the reaction atmosphere. Consequently, the rate ofcatalyst deactivation is expressed as (Pérez-Cabero et al. 2004)

�da=dt ¼ kda� krð1� aÞ ð2:23Þ

where kd and kr are the deactivation and regeneration kinetic constants, respectively.These constants are function of temperature with Arrhenius functionality. TheArrhenius parameters of these constants are presented in Table 2.3. Solution of thisequation will give:

a tð Þ ¼ Ks þ 1� Ksð Þexp�kGt ð2:24Þ

where kG = (kd + kr) and Ks = kr/kG. Using this expression and replacing in Eq. 2.22,the variable concentration of catalyst will drive with time as following:

CS ¼ CS0ðKs þ 1� Ksð Þexp�kGtÞ ð2:25Þ

Now, this relation can be used to define effect of catalyst deactivation on the CNTsynthesis (Eq. 2.20).

To discuss the rate law of diffusion of carbon through the catalyst surface, theFick’s first law is used correctly as (carbon atoms) (Klinke et al. 2005):

Jp ¼ �DDC ð2:26Þ

where C the concentration and D is the diffusion constant which is given byArrhenius equation as

D ¼ Do � e�E=RT ð2:27Þ

Table 2.3 Arrhenius parameters of the activity rate constants (kd and kr) (Pérez-Cabero et al.2004)

Rate constant Activation energy (E) kj/mol Pre-exponential factor (A) 1/min

kd 69.1 4.46 × 102

kr 187.2 2.8 × 106


where E is the activation energy and Do is diffusion factor. The typical amount ofthese parameter are Do = 2.2 cm2/s and E = 1.27 eV in the case of carbon diffusionin Fe (fcc). Thus, the diffusion coefficient at temperature 923 K will beD = 2.533 × 10−11 m2/s. Following the iron–carbon diagram, the maximal solubilityof carbon in iron at 923 K is S = 65 ppm (weight). Exceeding this amount leads tothe formation of iron carbide Fe3C. This limit determines the maximal concentra-tion gradient ΔC. Therefore, the amount of ΔC can be calculated as:

DCj j ¼ Sddiff

MWFe

MWC

1VFe

ð2:28Þ

where ddiff is the diffusion distance (usually ddiff ¼ 0:5dparticle), MWFe and MWC aremolecular weight of Fe and carbon, respectively, and VFe is molar volume of Fe(=7.093 × 10−6 m3/mol). One obtains:

DCj j ¼ 42:631

ddiff

molm3 ð2:29Þ

In this case, we have:

JP ¼ 1:079� 10�9 1ddiff

molm s

ð2:30Þ

This will get the maximum flux of carbon diffusion into Fe catalyst structure atT = 923 K.

2.2.2.3 Chemical Reactors

A chemical reactor is the heart of chemical processes, and it is a device in which thefeedstock (reactants) is converted to the desired product by chemical transformationsor chemical reactions (Fig. 2.35). Reactors are not only involved in producingchemical products but also in energy production like combustor and in certain elec-trochemical cells like fuel cells (Missen et al. 1999). There are different types ofreactors and various factors that need to be taken into account in selecting chemicalreactors for specific task. In addition to economic considerations, the appropriateselection of reactor will give the highest yields and purity for desired product, whileminimize pollution (Coker 2001). Three important parameters, conversion (Eq. 1.10),

ReactorFeedUn-reacted Feed Desired ProductBy-product

Fig. 2.35 Reactor transforms feed into desired product and by-product


http://dx.doi.org/10.1007/978-981-287-496-2_1

yield (Eq. 1.12) and selectivity (Eq. 1.14), describe the performance of a chemicalreactor.

Reactor design embodies many different facets and disciplines like mechanicaldesign of equipment, instrumentation and process control, economic and socio-economic (environmental and safe operation). However, reactor design in chemicalengineering discipline means is analysis of performance of an existing reactor. Theterm “reactor performance” may refer to the operating results achieved by a reactor,particularly with respect to fraction of reactant converted or product distribution fora given size and configuration, or the size and configuration for a given conversionor distribution (Missen et al. 1999). In any case, reactor performance depends onrates of processes involved and fluid characteristics of the system. The rates ofprocesses depend on the reaction occurred, heat and mass transfer characteristics ofthe system and equilibrium limitations. Fluid characteristics of the system correlatewith the motion and relative motion of fluid elements (both single-phase andmultiphase situations) and solid particles. Flow characteristics includeresidence-time distribution (RTD), mixing characteristics for elements of fluid inthe reactor and the level of segregation (Missen et al. 1999). Lack of sufficientinformation on any of these characteristics is a major impediment for completedesign of a reactor.

Information about the composition and temperature at each point of the reactorenables the designer to describe the behavior of a chemical reactor. Concentrationsof species at any point may change due to either chemical reaction or mass transfer.The temperature at any point of the reactor may also change because of the heatabsorbed or released by chemical reaction or heat transfer. The rate of the chemicalreaction as well as the rate of mass and heat transfer influences the concentrationand temperature of a given section of the system. Concentration, temperature andmolecular properties determine the reaction rate (Missen et al. 1999). This processoccurs through kinetic properties that affect the outcome of the process through thekinetics of heat and mass transfer. The rates of mass and heat transfer depend on theproperties relative to the reactor, such as size of the reactor, size and speed of theimpeller, and the area of heat exchanging surfaces (Coker 2001).

Temperature is the most important parameter that influences kinetics and qual-itative characteristics of the reaction products. The deviation from optimal reactiontemperature involves uncontrollable change of reaction rate, which negativelyaffects selectivity of chemical processes. The exact control of temperature, which isa driving force for heat exchange, is the central factor for determining otherparameters in a process. In order to provide an optimal progress of a chemicalreaction, different conditions must be achieved in the reactor: First, a nearly idealmixing of the reactants should be ensured, linked with the generation of anextended phase interface in multiphase reactions. Afterward, the required responsetime must be guaranteed by a residence time with preferentially narrow RTD.Finally, the reactor heat necessary for the reaction must be supplied or carried off. Inthis connection, control of temperature, pressure, time of reaction and flow velocityis important (Reschetilowski 2013).


http://dx.doi.org/10.1007/978-981-287-496-2_1

http://dx.doi.org/10.1007/978-981-287-496-2_1

Reactor design and selection are among the most important task in the wholeprocess design. Reactors produce both wanted products and unwanted by-products.The unwanted by-products lead to loss of revenue and environmental problems.The chemical reactor is designed and dimensioned to get the required yield andconversion of the raw material to desired product. Good performance of a reactor isthen imperative in both determining the economic viability of the overall processand its environmental impact. Reactor design that intends to predict the perfor-mance of a reactor for specified requirements necessitates both experimental andtheoretical endeavors (Prud’homme 2010). The first step in reactor design for acertain product is to make a decision regarding the reaction path, as there aresometimes different routes to produce a specific material. Besides commercial andtechnical factors that must be considered here, the preferred reaction path is the onethat use the cheapest raw materials and produce the smallest quantities ofby-products. It is worth mentioning that, at this early stage of design, all conse-quences of the selected reaction path are not predictable; however, some of them areclear (Smith 2005). The next step will be to choose the catalyst, if catalyst isneeded, and then determination of ideal characteristics (like type, size configura-tion) and operating conditions (like temperature and pressure in the reactor, feedingcondition) for the reaction system (Missen et al. 1999). Some of the parametersconsidered when selecting a reactor for chemical reactions are the number of phasesinvolved, the differences in the physical properties of the participating phases, thepost-reaction separation, the inherent reaction nature (stoichiometry of reactants, theintrinsic reaction rate, isothermal/adiabatic conditions, etc.), the residence timerequired, and the mass and heat transfer characteristics of the reactor(Reschetilowski 2013). To get through the best selection for reactor, there areseveral questions that should be first answered (Coker 2001; Prud’homme 2010;Smith 2005):

• What is expected from the reactor?• What are requirements imposed by the reaction mechanisms? For example, what

is the rate expression, and the required production capacity? What are reactionheat and reaction rate constant?

• What is the nature of chemical reaction (or reactions) that occur in reactor? Forexample, is the reaction homogeneous or heterogeneous? or is the reaction iscatalytic or non-catalytic?

• What are the reaction conditions including temperature of the heat transfermedium, temperature of the inlet reaction mixture, inlet composition andinstantaneous temperature of the reaction mixture?

• What is the thermodynamic state and deviation from thermodynamicequilibrium?

• What are the governed principles of transport phenomena in the reactor? Whatare the values of heat and mass transfer coefficient?


Besides these factors related to the nature and kinetics of the reactions andtransport phenomenon in the reactor, the optimum design relies on review of someinformation about type of reactor and its mode of processing (Smith 2005; Coker2001). For example:

Is process preferred to be batch or continuous?Is combination of reactors in series or parallel required?What is the mode of operation is it an isothermal (i.e., constant temperature) oran adiabatic (i.e., heat does not exchange with the surroundings) condition?Can the desired degree of conversion of the raw feedstock be obtained by asingle pass operation best, or recycling is needed? Incomplete conversion in thereactor requires a recycle for unconverted feed material, as depicted in Fig. 2.36(Smith 2005).

After answering to the questions addressed above, a good approximation isapproaching as nearly as possible the ideal condition. Ideal reactors are often a goodapproximation to real cases, and their studying can be used as a good approach forreal reactor design (Prud’homme 2010). Ideal reactors are either perfectly insulatedfrom the viewpoint of mass and heat transfer (adiabatic) or at equilibrium with thesurroundings (Coker 2001). More details for reactor selection according to thereactions can be found in Smith 2005.

Batch Reactors

A batch reactor is a vessel that has no input or output during the time reactionproceeds. In batch reactors, the reactants are loaded into the reactor at the beginningof the operation and the reaction is initiated by heating the contents to reactiontemperature, adding a catalyst and so on. The reaction is allowed continuing for apredetermined time, and finally, the products are discharged. The time that reaction

Fig. 2.36 Feed recycling to send the feedstock conversion


let to be proceeded is the required time to achieve a given conversion. A goodexample of batch reactors is small flasks and beakers usually used inlaboratory-scale setup. Here, reactants are added and brought to reaction temper-ature. A batch reactor is addressed by the following characteristics (Missen et al.1999; Coker 2001).

1. Batch reactor is a closed system that means the total mass is fixed.Although the total mass in the reactor is constant, the volume or density of eachbatch may vary as reaction proceeds. If a batch reactor is used for a liquid-phasereaction that the volume and density can be assumed constant, but thisassumption is not valid when there is a gas-phase reaction.

2. The energy of each batch may vary (as reaction proceeds). Heat exchangers areusually used to control temperature.

3. The operation of the reactor is inherently unsteady state. The extent of reactionand properties of the reaction mixture (for example, batch composition) changewith time

4. The reaction (residence) time for all elements of fluid is the same.

Batch reactors have several advantages, which make them attractive for manycommercial operations. The most important ones are being economical forsmall-volume production and being flexible. These parameters make them multi-purpose equipment for production of a variety of products at different rate ofproduction. Batch reactors are also preferred for those industrial purposes thatequipment needs regular cleaning and sterilization (Smith 2005), like pharmaceu-tical and biotechnology products. Besides the application of batch reactors inmanufacturing processes, they are extensively applied in the laboratory research.A general illustration of a batch reactor for liquid-phase reaction is depicted inFig. 2.37. Feed streams are introduced through input lines, and then, valves fitted inthe input lines are closed. To make the liquid-phase homogenous and provide bettercontact among reactants, a stirrer is used. Depending on the type of reaction (beingexothermic or endothermic), reactor needs to be cool down or warm up using thejacket. Temperature and level of materials are controlled by temperature indicatorcontroller (TIC) and level controller (LC). When the reaction is accomplished, thevalve fitted on the output line is opened to collect the materials including mainproduct, by-products and reactants not to be participated in the reaction.

The performance of batch reactors is influenced by several factors. The impor-tant ones are contacting pattern and operating condition (Smith 2005). There aredifferent contacting patterns to enhance the mass transfer. Agitated tank and movingbed are good example of contacting mode for batch reactors. In the ideal-batchmodel, the content of the reactor is subjected to perfect mixing. Concentrationschange with time, but the perfect mixing ensures that at any instant, the compositionand temperature throughout the reactor are uniform (Missen et al. 1999; Smith2005). Operating conditions of batch reactors are described by both fixed anddynamic variables. Batch cycle time and total amount of reactants are optimizedvalues fixed for a given batch reactor system. Nevertheless, values for temperature,


pressure, rate of feeding and product takeoff change through the batch cycle time. Ifthe profile of these dynamic variables (temperature, pressure and rates for feedingand product takeoff) is known, a simulation of the reactor can be carried out in thetime interval considered.

One of the main drawbacks for batch reactor is its limitation in mass production,because the characteristics and properties of product may vary from a batch toanother batch. Nevertheless, they are particularly useful for investigation of thekinetics of a chemical reaction (Prud’homme 2010) since the time is the mainvariable in this system and sampling in different times can provide informationabout concentration variation with time which is necessary for kinetic study. Inaddition, when a researcher aims to study an unknown reaction and tries to find theeffects of parameters such as temperature, pressure and concentration on the

Fig. 2.37 Batch reactor (LC level controller, TIC temperature indicator controller)


reaction characteristics, this type of reactor is usually preferred. These capabilitiesof batch reactor are important from the nanotechnology viewpoint, and therefore,the batch reactors are generally selected. However, a challenge in using a batchreactor for rate determination is the ability to obtain good conversion data as afunction of time (Denn 2011).

For example, synthesis of nanoparticles in a hydrothermal system is a batchreactor. In this system, the reactor vessel is loaded with the reactant. Then, the vessel(which is a pressure vessel) is heated to the high temperature that causes high-pressurecondition in the vessel volume. In this condition, a specific reaction takes place andthe monodispersed nanoparticles are formed with full control of the particle size.

Continuous Reactors

In continuous reactors, feeding and product takeoff are both continuous, but notnecessarily at a constant rate. Referring to Fig. 2.37, both input and output lines areopened during reaction. Not the total mass inside the vessel or the density of thestream is constant with time, and the system may operate at steady state or atunsteady state (Missen et al. 1999). An ideal model for continuous reactors iscontinuous stirred tank reactor (CSTR). In this model, a stirrer is installed in thevessel that perfectly mixes the reactor contents (Fig. 2.37). As a result of this wellmixing, the composition and temperature of the reaction are homogeneous in allparts of the vessel (Coker 2001). CSTRs are used both in a laboratory and on a largescale. They are preferred for the laboratory investigation of gas-phase reactions,particularly when solid catalysts are involved (Missen et al. 1999). They are alsoemployed in a series arrangement. A good example of this application is for thecontinuous copolymerization of styrene and butadiene to make synthetic rubber. Animportant disadvantage of the CSTR is that for a given conversion, it requires alarge inventory of material, which is not desirable if the reactants or products arehazardous (Nauman 2008). The residence time of individual fluid elements in theCSTR reactor is not constant and varies (Smith 2005).

Plug-flow reactors (PFR) are another type of ideal continuous reactor. A PFR issimilar to a CSTR in being a flow reactor, but is different in its mixing charac-teristics. In PFR, the flow in the vessel is plug flow, i.e., there is no mixing in thedirection of flow. An example for PFR reactors is large cylindrical tubes used in thepetrochemical industry for the cracking of hydrocarbons. This process is continuouswith reactants in the tubes and the products obtained from the exit. The extent ofreaction and properties, such as composition and temperature, depends on theposition along the tube and does not depend on the time (Coker 2001). The formeris illustrated in Fig. 2.38, in which concentration profiles are also shown withrespect to position in the vessel. Each element of fluid has the same residence time tas any other, that is, there is IZO spread in t.

Due to complete mixing in the plane perpendicular to the direction of flow, theproperties of the fluid, including its velocity, are uniform in this plane (Missen et al.1999). In the plug-flow model, due to steady movement of flow only in one


direction, a steady uniform movement of the reactants is assumed, with no attemptto induce mixing along the direction of flow. Like the ideal-batch reactor, theresidence time in a plug-flow reactor is the same for all fluid elements. Plug-flowoperation can be approached by using a number of mixed-flow reactors in series.The greater the number of mixed-flow reactors in series, the closer is the approachto plug-flow operation (Smith 2005). PFRs are used for both laboratory-scaleinvestigations of kinetics and large-scale production. They are preferred when(Nauman 2008):

• Careful control of residence time is important, as is the case where there aremultiple reactions in series.

• High rates of heat transfer are required as PFRs offer a high ratio of heat transfersurface area to volume.

• High pressure is required. Under high-pressure conditions, a small-diametercylinder requires a thinner wall than a large-diameter cylinder.

PFRs are not useful for multiphase reactions since it is often difficult to achievegood mixing between phases, unless static mixer tube inserts are used (Nauman2008). The most important results obtained for ideal reactor models are summarizedin Table 2.4

Fig. 2.38 Plug-flow models

Table 2.4 Summary of model for ideal reactors

Batch reactor PFR CSTR

Reactor model dCA

dt ¼ �rAdCA

ds ¼ �rADCADs ¼ �rAf

Reactor parameter t (residence time) s ¼ Vq (space time) s ¼ V

q (space time)

Concentrationdetermination

n = 1 CA ¼ CA0e�kt CAf ¼ CA0e�ksCAf ¼ CA0

1þks

n ≠ 1 C1�nA0 � C1�n

A ¼ ðn� 1Þkt C1�nA0 � C1�n

Af ¼ ðn� 1Þks ksCnAf þ CAf ¼ CA0


Many industrial reactors operate in the continuous mode. The main reason ofcommon usage of these reactors is their ability in mass production of many productsin any condition. These reactors are designed in the no-mixing status (PFR) orperfect mixing status (CSTR) at any conditions such as isothermal ornon-isothermal, homogeneous or heterogeneous, steady state or unsteady state,constant volume or variable volume, or any other conditions. As an appliedexample, in the catalytic reactions, contact of the reactants with the catalyst surfaceis a key parameter of the reactor operation. Tank vessel is an example for CSTR,where fixed bed is an example for PFR in heterogenous reactors. An improvedcontact can be achieved in reactors like fluidized bed, moving bed, trickle bed,rotary drum. In the nanotechnology field, many nanostructure materials mainly inthe gas phase are produced via catalytic reactions. However, controlling thesereactions to achieve the best conditions is too complex and necessitates specialarrangement of the phases contacting. For example, in CNT synthesis via CVD, thefluidized bed reactor is a suitable choice in which mass production and presetproperties of the product are achievable. In such system, the fluid phase continu-ously flows in the reactor as plug and the solid (catalyst) is perfectly mixed.

Semi-batch reactor

A semi-batch reactor is an intermediate of the batch and continuous conditions. Thistype of reactor is a vessel where a feed stream inlets to the reactor continuously,while there is no output stream (rarely with a continuous product stream withoutinput stream, such as decomposition reaction of a liquid and gas releasing which thegas product has to exit from the reactor environment during the reaction). Whenmultiple reactant components are involved in a complex reaction, the semi-batchreactor is appropriate choice to control the concentrations and therefore to controlthe reaction progress. Consider the reaction aA + bB → cC takes place in a batchreactor. The components A and B initially are in contact with each other at apredetermined concentration, but after a while, the concentration of each compo-nent varies uncontrollably. On the contrary, the reaction in progress can be simplycontrolled if one of the components A or B is added to the other gradually. This isthe principle of semi-batch state where an input stream (containing one of thereactant component) inlets to the reactor (containing another one of the reactant)while there is no output stream. Semi-batch reactors can operate in the steady-statecondition with adjusted rate of the input stream. It should be noticed that thecontrolling of the reactant concentration during the reaction is one of the advantagesof the semi-batch system. This system facilitates different purposes such as tem-perature controlling, pH controlling, diluter or additive adding.

The semi-batch state generates excellent condition for synthesis of a number ofnanomaterials. As an example, consider reduction in Ti cations in a liquid phasethat the component TiO2 is produced and precipitated as a solid based on precip-itation method (the precipitation method will be explained in the next section). Thereaction is:


TiOCl4liquid þ NH4OHliquid ! TiO2solid þ NH4Clliquid þ HClliquid

A vessel is first loaded with one of the reactants (for example, TiOCl4), and then,the other reactant (for example, NH4OH) is gradually added to the vessel. As thereaction proceeds, no stream leaves the vessel. Rate of the reaction and thus rate ofthe solid formation can be simply controlled by controlling the input rate of thereactant. This ability is imperative in synthesis of solid particle with nanosize(Namin et al. 2008).

Micro- and Nano-reactors

Micro-reactors are reactors with the small dimensions, which do not exceed 1 mm(Ehrfeld et al. 2000). They are miniaturized chemical reaction systems, whichcontain micro-channels, in which fluid flows continuously and chemical reactionstake place (Jensen 2001). The main difference of micro-reactors from the commoncontinuous-flow reactors consists in a laminar flow regime of the fluids (Hetsroniet al. 2005). Since the ratios of viscous force to inertial force and of interfacial forceto inertial force in micro-channels are several orders of magnitude higher than thatin regular equipment, laminar flow always happens in such systems. The smalldimensions of channels also lead to relatively large surface area-to-volume ratios(10,000–50,000 m2/m3) and shorter diffusion paths compared to conventionalreactors (100 m2/m3), and thus, transport process is improved (Ehrfeld et al. 2000;Capretto et al. 2011). As a result of laminar flow, molecular diffusion dominates thetransport phenomena, and hence, time taken to enable mixing across amicro-channel can be approximated according to molecular diffusion theory, inwhich the rate of transfer is directly proportional to the surface area (Reschetilowski2013). Accordingly, heat and mass transfer in micro-reactors is some orders ofmagnitude higher than that in usual reactors. For example, the mass transfercoefficient for micro-reactor could be KLa ≈ 5–15 s−1, which is two orders ofmagnitude larger than those for macroscopic reactors, KLa ≈ 0.01–0.08 s−1 (Jensen2001). Thermal diffusivities are high enough (up to 41,000 W/(m2 K)) that themicro-reactors will be approximately isothermal (Kockmann 2006; Watts and Wiles2007). In comparison with traditional reactors on the macroscale, higher rates ofheat and mass transfer facilitate fast and accurate control of temperature and con-centration in the reactor and therefore improve selectivity and productivity ofreaction, and also allow reactions to be performed under more uniform conditions(You-qi 2008; Jensen 2001).

Another outstanding advantage of micro-reactors is their capability to beemployed at high pressure. In the cylindrical vessels, the most allowable pressure isinversely proportional to the diameter of a capillary. Therefore, micro-channelsprovide the chance to use micro-reactors at high pressure, about 400 bar and above.Despite such high pressure in the micro-reactor, it can be carried out more safelycompared to conventional large-scale reactors (Reschetilowski 2013).


Consequently, micro-reactors are supreme for conducting operations even withhighly exothermic reactions or under supercritical conditions (Nauman 2008;Benito-Lopez et al. 2008; Marre et al. 2012). Besides all processing benefits formicro-reactor, the axial dispersion effects associated with the parabolic flow profileof typical micro-channel contribute to a broader RTD (Chang 2013). Designtechniques for micro-reactors are generally identical to those for macroscale reac-tors with laminar flow (Nauman 2008). Figure 2.39 shows timescale of differentreactor scale and the related phenomena.

Keeping in mind all the benefits presented, the main motivations for employingmicro-reactors are the gain in economy, safety and ecology impacts. Micro-reactors,due to their small dimension, contribute to the minimization of raw material andenergy consumption as well as waste production and thus improve its economics(Jensen 2001). This feature is significant both in the laboratory research activity andduring scale up to the pilot plant or large-scale production. The small quantity ofsubstances in the reactor also minimizes the potential of thermal explosion bydangerous reactions. This feature is particularly important for strongly exothermicreactions and reactions dealing with toxic or explosive materials.

Despite all exceptional advantageous of micro-reactors, there are some practicalconsiderations that need to be taken into account. For example, in liquid phase, therate of diffusion may not be as high to prevent lateral diffusion if liquid reactionsproceed very fast. In this case, mixing can be used to reduce diffusive barriers(Hoffmann et al. 2010). It is worth mentioning here that turbulence is not induced onthe microscale even if mechanical or magnetic stirrers are used (Reschetilowski2013). Another obstacle can be the pressure loss with higher throughputs, which canlead to a restriction of the flow in a micro-reactor. This problem can be also avoidedby using a series of continuous micro-reactors with equal dimensions that operate inparallel or in series (Kashid and Kiwi-Minsker 2009). Micro-reactors can be scaledto meet demands for large-scale production by numbering up the micro-reactors andrunning them for longer times. This strategy makes it feasible to go from microgram

Fig. 2.39 Timescale of chemical and physical processes


to kilogram quantities without additional chemistry modifications or reactor engi-neering (Jensen 2001).

Micro-reactors have been finding various applications in either conventional ormodern technology. For example, in the field of biotechnology, they are applied fordiagnostic applications, sometimes called lab-on-a-chip and production of designermolecules. They are employed in combinatorial chemistry and kinetic studies foreffective and fast optimization of investigated reactions (Reschetilowski 2013). Forexample, they are promising candidates for characterization of catalyst perfor-mance, such as kinetics, selectivity and deactivation, with the purpose ofhigh-throughput screening of catalysts (Jensen 2001). Due to their outstandingpractical benefits, micro-reactors are preferable for up to 70 % of all chemicalreactions and in particular for heterogeneous reaction systems (Roberge et al. 2005;Nauman 2008). Fast reactions with a large heat effect, which is not possible inmacroscale reactors, can be carry out in micro-reactor since they allow for nearlyisothermal conditions at high reactant concentrations (Chang 2013). In this respect,they have been extensively employed in synthesis of polymers, nanoparticles, aswell as bio-, electro- and photocatalysis process.

Since the emphasis of this book is on the nanoengineering, the following sen-tences will emphasis on this subject. Chemical synthesis of nanostructures usingmicro-reactors has offered unique capabilities in the advancement of nanotech-nology (Yen et al. 2005; Song et al. 2006; Karnik et al. 2008; Duraiswamy andKhan 2009; Marre et al. 2012; Jin et al. 2010; Lazarus et al. 2010; Zhao et al. 2011;Ishizaka et al. 2012; Eluri and Paul 2012). Micro-reactors have indicted thepotential to overcome some of the technical challenges in nanomaterial production.The important challenge in synthesis of nanostructure is having control over thesize and size distribution of nanostructures that these characteristics depend onseveral processing parameters, most importantly the residence time and temperatureand their distribution. In conventional large-scale reactors, fluctuations in temper-ature and concentration are difficult to correct because of large response time; e.g.,any alteration requires time to have an effect on the whole system. In comparison,changes on the microscale are observed almost immediately (Reschetilowski 2013).The possibility of rapid mixing of reactants and fast heating and cooling of thesystem enables precise control of the reaction parameters, thus improving theprocess yield as well as quality of nanostructure, e.g., composition, size and shapeof nanostructures. It becomes necessary to directly address the particulate nature ofmatter for nanoreactor design. On the nanoscale, the diffusion times and Reynoldsnumbers shown in Table 2.5 become rather meaningless. Instead, the models mustaddress the behavior of individual molecules. Thus, an alternative definition of thenanoscale is the scale at which continuum models must be replaced by molecularmodels (Smith 2005). The miniaturization of length scale causes significant effectson transport properties and acting forces, as shown in Table 2.5 (Mae 2007).


2.2.2.4 Health, Safety and Environmental Issues

As explained in the previous sections, nanotechnology has great applications in thefields of energy, material, environment and medicine. The quantity of engineerednanostructures is expected to grow significantly in the next several years (Roco2011). On the other hand, as nanotechnology progresses toward manufacturing andcommercial stages, more concerns are being raised about the potential risky effectsof nanostructures on human health and environmental media. For example, CNTs,due to their superior mechanical, electronic and thermal conductive properties, arecurrently used in consumer and industrial products like sports tool or flexibledisplays and touch screens. On the other hand, CNTs are likely to be releasedduring the life cycle of CNT-incorporated products through mechanical abrasionand degradation. Some researchers asserted that CNTs are toxic to aquatic organ-isms, and they can damage the lungs if inhaled. Similar reports have been alsopublished for other nanostructures that they can potentially lead to new hazards orincrease risks to the environment (Maynard et al. 2006; Maynard and Pui 2007;Oberdörster et al. 2007). However, all these reports are based on preliminarystudies, and exact environmental health and hazard risks associated with thenanostructures productions and their application are not fully known yet. Sincelarge numbers of novel nanotechnology-based products are continuously beingintroduced, it is vital to develop a robust scientific platform to understand health,safety and environments (HSE) outcomes of these products. Consequently, HSEstudies need to keep up with the rapid growth of the nanotechnology. Diversity ofnanostructures means that there is not going to be a single answer for all of thesesystems, and multidisciplinary exercises, which move beyond traditional hazard,exposure and risk assessment models, are critical. A comprehensive study on HSEissues associated with nanostructures will answer and resolve questions anduncertainties asked by toxicologists, the community, regulators. Development ofregulations and guidelines for occupational health and safety in the workplacethrough using data and information acquired by HSE studies will eventually have a

Table 2.5 Scaling effect of transport properties (Mae 2007)

Property nm μm mm m

Length (L) 10−9 10−6 10−3 1

Surface area (L2) 10−18 10−12 10−16 1

Volume (L3) 10−27 10−18 10−9 1

Specific surface area (L−1) 109 106 103 1

Rate (∝L) 10-9 10−6 10−3 1

Inertial force (∝L4) 10−36 10−24 10−12 1

Viscous force (∝L2) 10−18 10−12 10−6 1

Interfacial tension (∝L) 10−9 10−6 10−3 1

Viscous force/inertial force (∝L−2) 1018 1012 106 1

Interfacial tension/inertial force (∝L−3) 1027 1018 109 1


strong influence on public acceptance and the implications of nanotechnology andits sustainability.

The HSE impacts of nanostructures occur when these materials are released intothe environment or taken up by human being. To assess the toxicology impact ofnanotechnology-based products, a fundamental requirement is to understand whichphysico-chemical properties of nanostructures are important in toxicity and hazardgeneration (Maynard et al. 2006; Nel et al. 2006; Oberdörster et al. 2007; Seatonet al. 2010). Accordingly, the ability to measure and characterize nanostructures in arange of media (air, soil, water and all living systems) and to examine their toxicityis critical. In this regard, nanostructures properties, like their morphology, size andsize distribution, chemical composition, crystalline structure, surface area, surfacechemistry, agglomeration and aggregation, should be well evaluated for theirundesired HSE impacts in vivo and in vitro. There are usually two general pro-cedures for data collection including toxicity studies in animals and epidemiologicstudies in humans. However, uncertainties are quit large, and it is more importantthat the nanostructures to be judged on a case-by-case basis. For each nanostructure,short-term and long-term mode of exposure, such as skin absorption, ingestion andinhalation among others, must be systematically studied to determine any potentialfor organ or tissue damage, inflammation, a triggering of autoimmune diseases andother health-related consequences.

Nanostructures may enter human body through various pathways, and they mayact by reacting with surface receptors or by passing into cells and reacting withintracellular receptors. A general statement notes that nanostructures are toxic andthey are too small to be prevented by the cells that normally resist other air con-taminant. The size of nanostructures is an important characteristic in governing theirinter- and intracellular distributions. Agglomerated nanostructures with dimensionabout few microns can be only taken up by cells such as macrophages.Nanostructures with size of more than about 50 nm can penetrate cells but notcellular organelles such as the nucleus or mitochondria. Those nanostructures thathave dimensions below 20 nm can even enter the latter organelles. Consequently,any interactions between a nanostructure and living system will be size dependent.

Airborne exposure with inhalable uptake is currently viewed as the most criticalexposure route. On the other hand, nanostructure agglomerates are pervasive inatmospheric sciences, air pollution and material manufacturing, and it is thereforeimperative to study the aerosol behavior of nanostructures. Wang and Pui (2013)studied the relationship between morphology of nanostructures and their aerosolbehaviors. Based on modeling and experimental results, they elucidated that thepenetration of nanotubes, like CNTs, is less than the penetration for a nanospherewith the same mobility diameter, which is mainly due to the larger interceptionlength of the CNTs. CNTs tend to form bundles due to their geometry and van derWalls forces. Accordingly, when CNTs are provided by manufacturers in thepowder form, it consist large agglomerates in the order of microns. In fact, largeagglomerates have limited mobility, and they are expected to be handled andtransported easier and safer comparing to the nanostructures that exist in form ofindividual nanoparticles. Agglomeration, however, is a particularly complex issue,


and agglomerates may break down to individual nanoparticles. How tightly thenanoparticles are stuck together in the agglomerates is thus a key factor. CNTs havealso tendency to form agglomerates with a bundle-like form in aqueous mediabecause of their geometry and hydro-phobic surface. The mobility of CNTs insurface waters and subsurface environments is dependent on the deposition andrelease behavior of these nanostructures (Petosa et al. 2010; Chen et al. 2010; Yiand Chen 2014). It has been indicated that, in the aqueous systems, CNTs havelimited mobility and can deposit on naturally occurring surfaces, such as sand,rocks and sediments (Jaisi and Elimelech 2009). However, the deposited CNTs maybe released from the solid surfaces and reenter the aqueous phase when the solutionchemistry changes (Tian et al. 2012; Yi and Chen 2014). Besides understanding therelease kinetics of CNTs from a product to various ecosystems, it is crucial todetermine their behavior, in term of absorption, desorption, biotic uptake andaccumulation in living system. In the next step, the physico-chemical interactions,kinetics and thermodynamic exchanges between CNTs and the biological compo-nents must be well recognized. Since any nanostructure has different characteristicsand hence different behaviors, all the mentioned issues must be well investigatedseparately for any nanostructure. Entities engaged in nanotechnology must considerpractical and innovative steps to minimize identified risks while managing forunknowns and uncertainties. The first step of prudent practices related to thenanoengineering safety involves creating a comprehensive framework to summa-rize current scientific knowledge and applications of nanotechnology and to identifypossible HSE benefits and HSE risks of nanotechnology. In the next step, recom-mended methodology for risk characterization of nanostructures, assessment andimplementation of reasonably practicable combination of HSE risk managementapproaches will minimize risks to the workers and the environment. Knowing thesource of nanostructures release, a set of “reference” or representative materialsmust be identified for testing that requires separate consideration for occupationalexposure as well as exposure from consumer products and via the environment. It isalso necessary to prepare technical documents that assess the appropriateness ofrecommended risk assessment methodologies for nanostructures and suggestimprovements in these methodologies. To guide safe implementation of nano-technology, generation of an accurate informative database is then critical. Creatingsuch database necessitates time and consensus building, rational decision-makingimplementation of high-throughput and rapid screening platforms. Exploitingcomputational methods can assist in risk modeling and hazard assessment. Theresult of such inclusive study will be a regulatory program for the responsibledevelopment, production, use and end-of-life disposal of nanostructures employedby companies and other organizations. It also provides value to thenanotechnology-related businesses by helping focusing on decreasing the risk ofproducts while keeping costs down.

To acquire a practical and suitable guideline for nanostructures, one importantduty of engineers and scientists is now to reach an appropriate answer to thefollowing questions:


• What are the social and economic dimensions of nanotechnology-basedproducts?

• What is the rate of degradation for nanotechnology-based products?• What is the rate of nanostructures release and emission to the ecosystems?• What is the rate of nanostructures transport through the systems?• How the nanostructures may damage the ecosystems?• What are the risky and undesired impacts on HSE?• What is the rate of nanostructure absorption and accumulation in living systems?• How the nanostructures that enter living systems may damage the system?• What is the level of exposure that creates unacceptable hazards for

nanostructures?• What is the interaction of nanostructures with other hazardous?• How the hazardous structures can be controlled?• What is the quantity of production waste goes to the environment from pro-

cesses involved in synthesis of nanotechnology-based product?• What are the reaction intermediates and by-products formed through the process

for synthesis of nanotechnology-based product?

In order to get an appropriate answer to these questions, engineers must

• Develop methods for risk characterization and detection of nanostructures• Develop methods for evaluating the nanostructures properties in the context of

their ignition and explosion potential• Develop technologies that enable the measurement of exposure to nanostruc-

tures in different ecosystems (soil and water)• Develop instruments capable of fast and online measurement of nanostructures• Develop methods and instruments that able to differentiate materials in the

different environment• Develop appropriate techniques capable of measuring nanostructures in bio-

logical systems• Develop suitable metrics and methods for data collection and interpretation• Develop scientific procedures to determine the potential toxicity of nanostruc-

tures to human, in terms of all potential modes of exposure to hazardousmaterials

• Develop standardized, well-characterized reference nanostructures• Develop MSDS guidelines for nanostructures• Develop a laboratory safety guideline and manufacturing workplace guideline• Develop methods for hazard control and abatement of nanostructures released

Exercises

1. What is difference between nanotechnology and nanoengineering?2. Give an example for each field that comes at the below and state how

knowledge of chemical engineers can help:


Design of sensor/biosensor using nanostructuresDesign of smart carriers for controlled release of active componentDesign of nanofluids for water shut-off and improved/enhanced oil recoveryDesign of nanostructures for reduction of the viscosity of heavy oilDesign of nanocatalyst and usage of these catalystDesign and application of membranesWastewater treatmentDesign and application of new energy generator devices

3. What are the two synthesis strategies in nanotechnology? Define each strategyand provide an example for each one.

4. What are the three methods for synthesis of nanostructures using top-downstrategy? Find an example for everyone.

5. Explain principles for synthesis of nanoparticles using sol–gel process? Howprinciples of chemical engineering can help in controlling the process?

6. Consider synthesis of carbon nanotubes using CVD process and state how achemical engineer can develop the process for commercial production?

7. A researcher aims to synthesis catalyst comprising nanoparticles of metalsupported on alumina. The prepared catalyst will be used in fluidized bed wherethe size distribution range of the particles is important. What other character-istics of the catalyst needs to be considered and what are the related charac-terization methods?

8. A researcher aims to develop a nanocomposite for biosensing. The layerthickness, chemical bonding, and electronic properties of the nanocompositeneed to be observed. Which method/s do you recommend?

9. A dilute nanofluid containing spherical nanoparticles of Al2O3 in water isprepared in a flask with depth of 10 cm. Calculate high shear viscosity ofnanofluids.How will be the viscosity of nanofluid changes if the nanofluid getsconcentrated?

10. A semi-batch reactor is a perfect system for reaction controlling that makesthem suitable for nanostructures synthesis. The following figure shows asemi-batch reactor. Consider reaction A + B → C with reaction rate—rA(mol/L s) which is the main reaction in synthesis of nanoparticle “C”. In orderto control the reaction rate for component A, the reactor is first fed by reactantA with initial concentration CA0. Then, the reactant B is gradually added to thematerial in the reactor with constant flow rate v0 and concentration CB0.Calculate variation of the component concentrations A and B with time.Calculate the concentration of component C if the precipitation rate of com-ponent C is Rpre (mol/L).


11. In the kinetic example of section “Some Examples on Kinetic Study” assumethat synthesis of CNT via ethanol decomposition as αC2H5OH + S →Cα(CNT)·S + 2αH2O + αH2 with the kinetic:

RCNT ¼ dmCNTdt ¼ KCethCS

where ethanol concentration (Ceth) is constant. The amount of active sites varywith time because of catalyst deactivation (Eq. 2.25). Calculate weight ofdecomposed CNT (mCNT) after time τ.

12. Consider parallel reactions R2 and R3 in example of section “Some Exampleson Kinetic Study”:

R2: aC � S ! Ca CNTð Þ þ a� 1ð ÞS RR2 ¼ K1CaC:S

R3: bC � S ! Cb AMORð Þ þ b� 1ð ÞS RR3 ¼ K2CbC:S

If carbon nanotube (CNT) is the desirable product and amorphous carbon(amor) is undesirable one, discuss about methods to increase formation rate ofCNT compared to the formation rate of amorphous carbon (RR2/RR3).

References

Akia, M., Yazdani, F., Motaee, E., Han, D., & Arandiyan, H. (2014). A review on conversion ofbiomass to biofuel by nanocatalysts. Biofuel Research Journal, 1(1), 16–25.

Almao, P. P. (2012). In situ upgrading of bitumen and heavy oils via nanocatalysis. The CanadianJournal of Chemical Engineering, 90(2), 320–329.

Anandan, S., Pitchumani, S., Muthuraaman, B., & Maruthamuthu, P. (2006).Heteropolyacid-impregnated PVDF as a solid polymer electrolyte for dye-sensitized solarcells. Solar Energy Materials and Solar Cells, 90(12), 1715–1720.

Antonovič, V., Pundiene, I., Stonys, R., Česniene, J., & Keriene, J. (2010). A review of thepossible applications of nanotechnology in refractory concrete. Journal of Civil Engineeringand Management, 16(4), 595–602.

Exercises 165

Aouada, F. A., Mattoso, L. H., & Longo, E. (2011). New strategies in the preparation of exfoliatedthermoplastic starch–montmorillonite nanocomposites. Industrial Crops and Products, 34(3),1502–1508.

Aravind, S. J., Baskar, P., Baby, T. T., Sabareesh, R. K., Das, S., & Ramaprabhu, S. (2011).Investigation of structural stability, dispersion, viscosity, and conductive heat transferproperties of functionalized carbon nanotube based nanofluids. The Journal of PhysicalChemistry C, 115(34), 16737–16744.

Asefa, T., Duncan, C. T., & Sharma, K. K. (2009). Recent advances in nanostructuredchemosensors and biosensors. Analyst, 134(10), 1980–1990.

Belin, T., & Epron, F. (2005). Characterization methods of carbon nanotubes: A review. MaterialsScience and Engineering B, 119(2), 105–118.

Benito-Lopez, F., Egberink, R. J. M., Reinhoudt, D. N., & Verboom, W. (2008). High pressure inorganic chemistry on the way to miniaturization. Tetrahedron, 64(43), 10023–10040.

Benkstein, K. D., Montgomery, C. B., & Semancik, S. (2014, June). Trace chemical detection inair using electronic noses and nano-structured sensing materials. In SPIE Defense + Security(pp. 90830X–90830X). International Society for Optics and Photonics.

Benninghoven, A., Rüdenauer, F. G., & Werner, H. W. (1987). Secondary ion mass spectrometry:basic concepts, instrumental aspects, applications, and trends. New York: Wiley.

Ben-Nissan, B., & Choi, A. H. (2006). Sol-gel production of bioactive nanocoatings for medicalapplications. Part 1: An introduction.

Bhushan, B. (2010). Springer handbook of nanotechnology. Berlin: Springer.Bianco, A., & Prato, M. (2003). Can carbon nanotubes be considered useful tools for biological

applications? Advanced Materials, 15(20), 1765–1768.Binnig, G., Rohrer, H., Gerber, C., & Weibel, E. (1982). Surface studies by scanning tunneling

microscopy. Physical Review Letters, 49(1), 57.Bird, R. B., Stewart, W. E., & Lightfoot, E. N. (2002). Transport Phenomena (2nd edn). New

York: Wiley.Bonnell, D. A. (Ed.). (2001). Scanning probe microscopy and spectroscopy: Theory, techniques,

and applications (Vol. 2). New York: Wiley-VCH.Bora, T., & Dutta, J. (2014). Applications of nanotechnology in wastewater treatment—a review.

Journal of Nanoscience and Nanotechnology, 14(1), 613–626.Brunauer, S., Emmett, P. H., & Teller, E. (1938). Adsorption of gases in multimolecular layers.

Journal of the American Chemical Society, 60(2), 309–319.Brune, D., Hellborg, R., Whitlow, H. J., & Hunderi, O. (1997). Surface characterization—a user’s

source book. New York: Wiley.Burda, C., Green, T., Landes, C., Link, S., Little, R., Petroski, J., & El-Sayed, M. A. (2000).

Optical spectroscopy of nanophase material. In Wang, Z. L., Characterization of nanophasematerials (pp. 197–241). Wiley-VCH Verlag GmbH, USA

Caineng, Z., Shizhen, T., Zhi, Y., Xuanjun, Y., Rukai, Z., Lianhua, H., & Chun, Y. (2012). Newadvance in unconventional petroleum exploration and research in China. Bulletin ofMineralogy and Geochemistry, 31(4), 313–320.

Cao, G. (2004). Nanostructures and nanomaterials, synthesis, properties and applications.London: Imperial College Press.

Capretto, L., Cheng, W., Hill, M., & Zhang, X. (2011). Micromixing within microfluidic devices.In Microfluidics (pp. 27–68). Berlin, Heidelberg: Springer.

Chandra, A. K., & Bhandari, V. (2013). Nanocomposites for tyre applications. In Advances inelastomers II (pp. 183–203). Berlin, Heidelberg: Springer.

Chang, C. H. (2013). Synthesis of nanomaterials using continuous‐flow microreactors. InMicroreactors in preparative chemistry: Practical aspects in bioprocessing, nanotechnology,catalysis and more (pp. 165–195). Weinheim, Germany: Wiley-VCH Verlag GmbH & Co.KGaA.

Chang, H. W., Lee, J., Hofmann, S., Kim, Y. H., Müller-Meskamp, L., Lüssem, B., & Gather, M.C. (2013). Nano-particle based scattering layers for optical efficiency enhancement of organiclight-emitting diodes and organic solar cells. Journal of Applied Physics, 113(20), 204502.


Chatterjee, S., Gadad, S. S., & Kundu, T. K. (2010). Atomic force microscopy. Resonance, 15(7),622–642.

Chaudhry, Q., Scotter, M., Blackburn, J., Ross, B., Boxall, A., Castle, L., & Watkins, R. (2008).Applications and implications of nanotechnologies for the food sector. Food Additives &Contaminants, 25(3), 241–258.

Chen, H., Ding, Y., & Tan, C. (2007). Rheological behaviour of nanofluids. New journal ofphysics, 9(10), 367, 1–25.

Chen, K. L., Smith, B. A., Ball, W. P., & Fairbrother, D. H. (2010). Assessing the colloidalproperties of engineered nanoparticles in water: Case studies from fullerene C60 nanoparticlesand carbon nanotubes. Environmental Chemistry, 7(1), 10–27.

Chen, L., & Wang, Y. Z. (2010). A review on flame retardant technology in China. Part I:Development of flame retardants. Polymers for Advanced Technologies, 21(1), 1–26.

Chen, Y., Wang, Y., Lu, J., & Wu, C. (2009a). The viscosity reduction ofnano-keggin-K3PMo12O40 in catalytic aquathermolysis of heavy oil. Fuel, 88(8), 1426–1434.

Chen, H., Witharana, S., Jin, Y., Kim, C., & Ding, Y. (2009b). Predicting thermal conductivity ofliquid suspensions of nanoparticles (nanofluids) based on rheology.Particuology, 7(2), 151–157.

Chen, G., Yang, B., & Liu, W. (2004). Nanostructures for thermoelectric energy conversion. In B.S. M. Faghri (Ed. ), Heat transfer and fluid flow in microscale and nanoscale structures(pp. 45–91). Southampton: WIT Press.

Cheng, Q., Tang, J., Ma, J., Zhang, H., Shinya, N., & Qin, L. C. (2011). Graphene and carbonnanotube composite electrodes for supercapacitors with ultra-high energy density. PhysicalChemistry Chemical Physics, 13(39), 17615–17624.

Chevalier, J., & Gremillard, L. (2009). Ceramics for medical applications: A picture for the next 20years. Journal of the European Ceramic Society, 29(7), 1245–1255.

Chi, L., & Röthig, C. (2000). 5 scanning probe microscopy of nanoclusters. In Characterization ofnanophase materials (pp. 133–163).

Choi, S. U., & Eastman, J. A. (2001). Enhanced heat transfer using nanofluids (No.US 6221275).Argonne National Laboratory (ANL), Argonne, IL.

Choloupla, K., & Malam, Y. (2010). Nanosilver as a new generation of nano products inbiomedical application. Trends in Biotechnology, 11, 580–588.

Choudalakis, G., & Gotsis, A. D. (2009). Permeability of polymer/clay nanocomposites: A review.European Polymer Journal, 45(4), 967–984.

Chourasia, A. R., & Chopra, D. R. (1997). Auger electron spectroscopy. Handbook ofInstrumental Techniques for Analytical Chemistry, 42, 791–808.

Chow, L., Lupan, O., Chai, G., Khallaf, H., Ono, L. K., Roldan Cuenya, B., & Schulte, A. (2013).Synthesis and characterization of Cu-doped ZnO one-dimensional structures for miniaturizedsensor applications with faster response. Sensors and Actuators, A: Physical, 189, 399–408.

Chung, C. J., Lin, H. I., Tsou, H. K., Shi, Z. Y., & He, J. L. (2008). An antimicrobial TiO2 coatingfor reducing hospital-acquired infection. Journal of Biomedical Materials Research. Part B,Applied Biomaterials, 85(1), 220–224.

Coker, A. K. (2001). Modeling of chemical kinetics and reactor design (Vol. 1). Gulf ProfessionalPublishing.

Comini, E. (2006). Metal oxide nano-crystals for gas sensing. Analytica Chimica Acta, 568(1),28–40.

Condon, J. B. (2006). Surface area and porosity determinations by physisorption: Measurementsand theory. Amsterdam: Elsevier.

Cullity, B. D., & Stock, S. R. (2001). Elements of X-ray diffraction (3rd ed.). Upper Saddle River,New Jersy: Prentice Hall.

Cyras, V. P., Manfredi, L. B., Ton-That, M. T., & Vázquez, A. (2008). Physical and mechanicalproperties of thermoplastic starch/montmorillonite nanocomposite films. CarbohydratePolymers, 73(1), 55–63.

Danafar, F., Fakhru’l-Razi, A., Salleh, M. A. M., & Biak, D. R. A. (2009). Fluidized bed catalyticchemical vapor deposition synthesis of carbon nanotubes—a review. Chemical EngineeringJournal, 155(1), 37–48.

References 167

Dastjerdi, R., & Montazer, M. (2010). A review on the application of inorganic nano-structuredmaterials in the modification of textiles: focus on anti-microbial properties. Colloids andSurfaces B: Biointerfaces, 79(1), 5–18.

Davis, J. J., Coleman, K. S., Azamian, B. R., Bagshaw, C. B., & Green, M. L. H. (2003). Chemicaland biochemical sensing with modified single walled carbon nanotubes. Chemical EuropeanJournal, 9, 3733–3739.

De, D., Panda, P. K., Roy, M., Bhunia, S., & Jaman, A. I. (2013). Reinforcing effect of nanosilicaon the properties of natural rubber/reclaimed ground rubber tire vulcanizates. PolymerEngineering & Science, 53(2), 227–237.

Denn, M. (2011). Chemical engineering: An introduction. Cambridge: Cambridge UniversityPress.

Derjaguin, B., & Landau, L. (1993). Theory of the stability of strongly charged lyophobic sols andof the adhesion of strongly charged particles in solutions of electrolytes. Progress in SurfaceScience, 43(1), 30–59.

Diallo, M. S. (2009). 11 water treatment by dendrimer-enhanced filtration: Principles andapplications. In Nanotechnology applications for clean water: Solutions for improving waterquality (p. 143).

Ding, Y. L., Alias, H., Wen, D. S., & Williams, R. A. (2006). Heat transfer of aqueous suspensionsof carbon nanotubes (CNT nanofluids). International Journal of Heat and Mass Transfer, 49,240–250.

Ding, Y. L., Chen, H., Wang, L., Yang, C. Y., He, Y., Yang, W., et al. (2007). Heat transferintensification using nanofluids. KONA Powder and Particle, 25, 23–38.

Downard, K. (2004). Mass spectrometry: A foundation course. Cambridge: Royal Society ofChemistry.

Dresselhaus, M. S., Dresselhaus, G., Saito, R., & Jorio, A. (2005). Raman spectroscopy of carbonnanotubes. Physics Reports, 409(2), 47–99.

Dresselhaus, M. S., Jorio, A., Souza Filho, A. G., Dresselhaus, G., & Saito, R. (2002). Ramanspectroscopy on one isolated carbon nanotube. Physica B: Condensed Matter, 323(1), 15–20.

Drexler, K. E., & Minsky, M. (1990). Engines of creation (pp. 171–190). London: Fourth Estate.Du, M., Guo, B., & Jia, D. (2010). Newly emerging applications of halloysite nanotubes: A

review. Polymer International, 59(5), 574–582.Duncan, T. V. (2011). Applications of nanotechnology in food packaging and food safety: Barrier

materials, antimicrobials and sensors. Journal of Colloid and Interface Science, 363(1), 1–24.Duraiswamy, S., & Khan, S. A. (2009). Droplet-based microfluidic synthesis of anisotropic metal

nanocrystals. Small (Weinheim an der Bergstrasse, Germany), 5(24), 2828–2834.Egerton, R. F. (2005). Physical principles of electron microscopy. New York: Springer.Ehrfeld, W., Hessel, V., & Lowe, H. (2000). Microreactors. Weinheim, Germany: Wiley‐VCH

Verlag GmbH & Co. KGaA.Eluri, R., & Paul, B. (2012). Synthesis of nickel nanoparticles by hydrazine reduction: Mechanistic

study and continuous flow synthesis. Journal of Nanoparticle Research, 14(4), 1–14.Endo, M., Strano, M. S., & Ajayan, P. M. (2008). Potential applications of carbon nanotubes.

In Carbon nanotubes (pp. 13–61). Berlin, Heidelberg: Springer.Esashi, M. (2009, September). Micro/Nano electro mechanical systems for practical applications.

Journal of Physics: Conference Series, 187(1), 012001.Esawi, A. M. K. & Farag, M. M. (2007). Carbon nanotube reinforced composites: Potential and

current challenges. Materials and Design, 28, 2394–2401.Evans, W., Fish, J., & Keblinski, P. (2006). Role of Brownian motion hydrodynamics on nanofluid

thermal conductivity. Applied Physics Letters, 88(9), 093116.Faghri, A., & Zhang, Y. (2006). Transport phenomena in multiphase systems. New York:

Academic Press.Farokhzad, O. C., & Langer, R. (2009). Impact of nanotechnology on drug delivery. ACS Nano, 3

(1), 16–20.Feynman, R. P. (1961). There’s plenty of room at the bottom. In H. D. Gilbert (ed.),

Miniaturization. New York: Reinhold.


Fortina, P., Kricka, L. J., Graves, D. J., Park, J., Hyslop, T., Tam, F., & Waldman, S. A. (2007).Applications of nanoparticles to diagnostics and therapeutics in colorectal cancer. Trends inBiotechnology, 25(4), 145–152.

Frackowiak, E., Khomenko, V., Jurewicz, K., Lota, K., & Beguin, F. (2006). Supercapacitors basedon conducting polymers/nanotubes composites. Journal of Power Sources, 153(2), 413–418.

Funakoshi, K., & Nonami, T. (2007). Photocatalytic treatments on dental mirror surfaces usinghydrolysis of titanium alkoxide. Journal of Coatings Technology and Research, 4(3), 327–333.

Gaboriaud, F., & Dufrêne, Y. F. (2007). Atomic force microscopy of microbial cells: Applicationto nanomechanical properties, surface forces and molecular recognition forces. Colloids andSurfaces B: Biointerfaces, 54(1), 10–19.

Gan, N., Yang, X., Xie, D., Wu, Y., & Wen, W. (2010). A disposable organophosphoruspesticides enzyme biosensor based on magnetic composite nano-particles modified screenprinted carbon electrode. Sensors, 10(1), 625–638.

Gancs, L., Kobayashi, T., Debe, M. K., Atanasoski, R., & Wieckowski, A. (2008).Crystallographic characteristics of nanostructured thin-film fuel cell electrocatalysts: AHRTEM study. Chemistry of Materials, 20(7), 2444–2454.

Gao, L., Zhou, X., & Ding, Y. (2007). Effective thermal and electrical conductivity of carbonnanotube composites. Chemical Physics Letters, 434(4), 297–300.

Gemili, S., Yemenicioğlu, A., & Altınkaya, S. A. (2010). Development of antioxidant foodpackaging materials with controlled release properties. Journal of Food Engineering, 96(3),325–332.

Gharibi, M., Zangeneh, F. T., Yaripour, F., & Sahebdelfar, S. (2012). Nanocatalysts for conversionof natural gas to liquid fuels and petrochemical feedstocks. Applied Catalysis, A: General, 443,8–26.

Gommes, C., Blacher, S., Dupont-Pavlovsky, N., Bossuot, C., Lamy, M., Brasseur, A., & Pirard,J. P. (2004). Comparison of different methods for characterizing multi-walled carbon nanotubes.Colloids and Surfaces A: Physicochemical and Engineering Aspects, 241(1), 155–164.

Goodhew, P. J., Humphreys, J., & Beanland, R. (2000). Electron microscopy and analysis. BocaRaton: CRC Press.

Hachey, M. A., Nguyen, C. T., Galanis, N., & Popa, C. V. (2014). Experimental investigation ofAl2O3 nanofluids thermal properties and rheology—effects of transient and steady-state heatexposure. International Journal of Thermal Sciences, 76, 155–167.

Hagen, J. (1999). Industrial catalysis: A practical approach (pp. 131–140). Weinheim, Germany:Wiley-Vch Verlag GmbH & Co.

Hahm, M. G., Leela Mohana Reddy, A., Cole, D. P., Rivera, M., Vento, J. A., Nam, J., et al.(2012). Carbon nanotube–nanocup hybrid structures for high power supercapacitor applica-tions. Nano letters, 12(11), 5616–5621.

Hashemi, R., Nassar, N. N., & Pereira Almao, P. (2014). Nanoparticle technology for heavy oilin-situ upgrading and recovery enhancement: Opportunities and challenges. Applied Energy,133, 374–387.

Hashiguchi, G., Goda, T., Hosogi, M., Hirano, K., Kaji, N., Baba, Y., & Fujita, H. (2003). DNAmanipulation and retrieval from an aqueous solution with micromachined nanotweezers.Analytical Chemistry, 75(17), 4347–4350.

Hawkins, D. T. (1977). Auger electron spectroscopy—a bibliography–1925–1975 (297 p.).IFI/Plenum Data Company, Plenum Publishing Corp., New York, NY.

He, Y. R., Jin, Y., Chen, H. S., Ding, Y. L., Cang, D. Q., & Lu, H. L. (2007). Heat transfer andflow behaviour of aqueous suspensions of TiO2 nanoparticles (nanofluids) flowing upwardthrough a vertical pipe. International Journal of Heat and Mass Transfer, 50, 2272–2281.

He, B., Tan, J. J., Liew, K. Y., & Liu, H. J. (2004). Synthesis of size controlled Ag nanoparticles.Journal of Molecule Catalysis, 221, 121–126.

Hema, K. (2013). Mems pressure sensor in automotive industry. International Journal.Hessel, V., Hardt, S., & Löwe, H. (2006). Chemical micro process engineering: Fundamentals,

modelling and reactions. Weinheim, Germany: Wiley.

References 169

Hetsroni, G., Mosyak, A., Pogrebnyak, E., & Yarin, L. P. (2005). Fluid flow in micro-channels.International Journal of Heat and Mass Transfer, 48(10), 1982–1998.

Hoffmann, M., Schlüter, M., & Räbiger, N. (2010). Experimental analysis and modeling ofmicromixing in microreactors. In Micro and macro mixing (pp. 287–303). Berlin, Heidelberg:Springer.

Hölscher, H., Allers, W., Schwarz, U. D., Schwarz, A., & Wiesendanger, R. (2000). Interpretationof “true atomic resolution” images of graphite (0001) in noncontact atomic force microscopy.Physical Review B, 62(11), 6967.

Homma, Y. (2014). Gold nanoparticles as the catalyst of single-walled carbon nanotube synthesis.Catalysts, 4(1), 38–48.

Hong, W., Bai, H., Xu, Y., Yao, Z., Gu, Z., & Shi, G. (2010). Preparation of goldnanoparticle/graphene composites with controlled weight contents and their application inbiosensors. The Journal of Physical Chemistry C, 114(4), 1822–1826.

Hoogers, G. (Ed.). (2002). Fuel cell technology handbook. Boca Raton: CRC press.Hsieh, C. T., Lin, Y. T., Chen, W. Y., & Wei, J. L. (2009). Parameter setting on growth of carbon

nanotubes over transition metal/alumina catalysts in a fluidized bed reactor. PowderTechnology, 192(1), 16–22.

Hu, J., Chen, C., & Zhu, X. (2009). Removal of chromium from aqueous solution by usingoxidized multiwalled car bon nanotubes. Journal of Hazard Materials, 162, 1542–1550.

Ioannatos, G. E., & Verykios, X. E. (2010). H2 storage on single-and multi-walled carbonnanotubes. International Journal of Hydrogen Energy, 35(2), 622–628.

Ishizaka, T., Ishigaki, A., Kawanami, H., Suzuki, A., & Suzuki, T. M. (2012). Dynamic control ofgold nanoparticle morphology in a microchannel flow reactor by glucose reduction in aqueoussodium hydroxide solution. Journal of Colloid and Interface Science, 367(1), 135–138.

Izadi-Najafabadi, A., Yasuda, S., Kobashi, K., Yamada, T., Futaba, D. N., Hatori, H., & Hata, K.(2010). Extracting the full potential of single-walled carbon nanotubes as durable superca-pacitor electrodes operable at 4 V with high power and energy density. Advanced Materials, 22(35), E235–E241.

Jaisi, D. P., & Elimelech, M. (2009). Single-walled carbon nanotubes exhibit limited transport insoil columns. Environmental Science & Technology, 43(24), 9161–9166.

Jajja, S. A., Ali, W., & Ali, H. M. (2013). Multiwalled carbon nanotube nanofluid for thermalmanagement of high heat generating computer processor. Heat Transfer—Asian Research.

Jensen, K. F. (2001). Microreaction engineering—is small better?. Chemical Engineering Science,56(2), 293–303.

Jin, H. D., Garrison, A., Tseng, T., Paul, B. K., & Chang, C. H. (2010). High-rate synthesis ofphosphine-stabilized undecagold nanoclusters using a multilayered micromixer.Nanotechnology, 21(44), 445604.

Jurewicz, K., Delpeux, S., Bertagna, V., Beguin, F., & Frackowiak, E. (2001). Supercapacitorsfrom nanotubes/polypyrrole composites. Chemical Physics Letters, 347(1), 36–40.

Kar, S., Bindal, R. C., & Tewari, P. K. (2012). Carbon nanotube membranes for desalination andwater purification: Challenges and opportunities. Nano Today, 7(5), 385–389.

Karnik, R., Gu, F., Basto, P., Cannizzaro, C., Dean, L., Kyei-Manu, W., & Farokhzad, O. C.(2008). Microfluidic platform for controlled synthesis of polymeric nanoparticles. NanoLetters, 8(9), 2906–2912.

Kashid, M. N., & Kiwi-Minsker, L. (2009). Microstructured reactors for multiphase reactions:state of the art. Industrial & Engineering Chemistry Research, 48(14), 6465–6485

Keblinski, P., Phillpot, S. R., Choi, S. U. S., & Eastman, J. A. (2002). Mechanisms of heat flow insuspensions of nano-sized particles (nanofluids). International Journal of Heat and MassTransfer, 45, 855–863.

Kiliaris, P., & Papaspyrides, C. D. (2010). Polymer/Layered silicate (clay) nanocomposites: Anoverview of flame retardancy. Progress in Polymer Science, 35(7), 902–958.

Kim, B., Chung, H., & Kim, W. (2012). High-performance supercapacitors based on verticallyaligned carbon nanotubes and nonaqueous electrolytes. Nanotechnology, 23(15), 155401.


Kim, J., & Grate, J. W. (2003). Single-enzyme nanoparticles armoured by a nanometer-scaleorganic/inorganic network. Nano Letters, 3, 1219–1222.

Kim, J., Grate, J. W., & Wang, P. (2006a). Nanostructures for enzyme stabilization. ChemicalEngineering Science, 61(3), 1017–1026.

Kim, Y. A., Hayashi, T., Endo, M., Gotoh, Y., Wada, N., & Seiyama, J. (2006b). Fabrication ofaligned carbon nanotube-filled rubber composite. Scripta Materialia, 54(1), 31–35.

Kirubakaran, A., Jain, S., & Nema, R. K. (2009). A review on fuel cell technologies and powerelectronic interface. Renewable and Sustainable Energy Reviews, 13(9), 2430–2440.

Klemm, D., Schumann, D., Kramer, F., Heßler, N., Koth, D., & Sultanova, B. (2009, June).Nanocellulose materials–Different cellulose, different functionality. In Macromolecularsymposia (Vol. 280, No. 1, pp. 60–71). Weinheim, Germany: Wiley‐VCH Verlag.

Klinke, C., Bonard, J. M., & Kern, K. (2005). Thermodynamic calculations on the catalytic growthof multiwall carbon nanotubes. Physical Review B, 71(3), 035403.

Kockmann, N. (2006). Transport processes and exchange equipment. In Micro processengineering: Fundamentals, devices, fabrication, and applications (pp. 71–113).

Kollo, L., Bradbury, C. R., Veinthal, R., Jäggi, C., Carreno-Morelli, E., & Leparoux, M. (2011).Nano-silicon carbide reinforced aluminium produced by high-energy milling and hotconsolidation. Materials Science and Engineering A, 528(21), 6606–6615.

Kong, C., Qian, W., Zheng, C., Yu, Y., Cui, C., & Wei, F. (2013). Raising the performance of a4 V supercapacitor based on an EMIBF 4–single walled carbon nanotube nanofluid electrolyte.Chemical Communications, 49(91), 10727–10729.

Krishnamoorti, R. (2006). Extracting the benefits of nanotechnology for the oil industry. Journalof petroleum technology, 58(11).

Krishnamurthy, S., Bhattacharya, P., Phelan, P. E., & Prasher, R. S. (2006). Enhanced masstransport in nanofluids. Nano Letters, 6(3), 419–423.

Kuila, T., Bose, S., Khanra, P., Mishra, A. K., Kim, N. H., & Lee, J. H. (2011). Recent advances ingraphene-based biosensors. Biosensors & Bioelectronics, 26(12), 4637–4648.

Kumar, M. N. V. R. (2000). Nano and microparticles as controlled drug delivery devices. Journalof Pharmacy and Pharmaceutical Sciences, 3(2), 234–258.

Kuo, C. Y. (2009). Water purification of removal aqueous copper (II) by as-grown and modifiedmulti-walled carbon nanotubes. Desalination, 249(2), 781–785.

Kuznetsov, D. V., Lysov, D. V., Nemtinov, A. A., Shaleiko, A. S., & Korol’kov, V. A. (2010).Nanomaterials in refractory technology. Refractories and Industrial Ceramics, 51(2), 61–63.

Kwak, K., & Kim, C. (2005). Viscosity and thermal conductivity of copper oxide nanofluiddispersed in ethylene glycol. Korea-Australia Rheology Journal, 17, 35–40.

Kwok, C., Reizman, B. J., Agnew, D. E., Sandhu, G. S., Weistroffer, J., Strano, M. S., & Seebauer,E. G. (2010). Temperature and time dependence study of single-walled carbon nanotubegrowth by catalytic chemical vapor deposition. Carbon, 48(4), 1279–1288.

Kwon, S., Fan, M., Cooper, A. T., & Yang, H. (2008). Photocatalytic applications of micro-andnano-TiO2 in environmental engineering. Critical Reviews in Environmental Science andTechnology, 38(3), 197–226.

Lagaron, J. M., & Lopez-Rubio, A. (2011). Nanotechnology for bioplastics: opportunities,challenges and strategies. Trends in Food Science & Technology, 22(11), 611–617.

Lahiff, E., Lynam, C., Gilmartin, N., O’Kennedy, R., & Diamond, D. (2010). The increasingimportance of carbon nanotubes and nanostructured conducting polymers in biosensors.Analytical and Bioanalytical Chemistry, 398(4), 1575–1589.

Landi, B. J., Ganter, M. J., Cress, C. D., DiLeo, R. A., & Raffaelle, R. P. (2009). Carbon nanotubesfor lithium ion batteries. Energy & Environmental Science, 2(6), 638–654.

Langevin, P. (1908). Sur la théorie du mouvement brownien (On the Theory of Brownian Motion).C. R. Acad. Sci. (Paris), 146, 530–533. |accessdate = requires |url = (help); reviewed by D.S. Lemons & A. Gythiel: Paul Langevin’s 1908 paper “On the Theory of Brownian Motion”[…]. Am. J. Phys, 65.

References 171

Lazarus, L. L., Yang, A. S. J., Chu, S., Brutchey, R. L., & Malmstadt, N. (2010). Flow-focusedsynthesis of monodisperse gold nanoparticles using ionic liquids on a microfluidic platform.Lab on a Chip, 10(24), 3377–3379.

Le Corre, D., Bras, J., & Dufresne, A. (2010). Starch nanoparticles: A review. Biomacromolecules,11(5), 1139–1153.

Lebedeva, I. V., Knizhnik, A. A., Gavrikov, A. V., Baranov, A. E., Potapkin, B. V., Aceto, S. J., &Sommerer, T. J. (2011). First-principles based kinetic modeling of effect of hydrogen ongrowth of carbon nanotubes. Carbon, 49(7), 2508–2521.

Lee, S. M., Park, K. S., Choi, Y. C., Park, Y. S., Bok, J. M., Bae, D. J., & Lee, Y. H. (2000).Hydrogen adsorption and storage in carbon nanotubes. Synthetic Metals, 113(3), 209–216.

Lee, S. W., Yabuuchi, N., Gallant, B. M., Chen, S., Kim, B. S., Hammond, P. T., & Shao-Horn, Y.(2010). High-power lithium batteries from functionalized carbon-nanotube electrodes. NatureNanotechnology, 5(7), 531–537.

Lefrant, S. (2002). Raman and SERS studies of carbon nanotube systems. Current AppliedPhysics, 2(6), 479–482.

Leong, K. Y., Mohd, H. N., Mohd, S. R., & Amer, N. H. (2014). The effect of surfactant on stabilityand thermal conductivity of carbon nanotube based nanofluids. Thermal Science, 00, 78.

Li, Z. (2002). Industrial applications of electron microscopy. Boca Raton: CRC Press.Li, T. R., Li, Y.W., Bai, Z. Y., Peng, X. C., Zhong, Z. Q., &He, T. (2013). Application of nano-sized

TiO2 in environmental protection. Applied Mechanics and Materials, 295, 2227–2232.Liang, M., Fan, K., Pan, Y., Jiang, H., Wang, F., Yang, D., & Yan, X. (2012). Fe3O4 magnetic

nanoparticle peroxidase mimetic-based colorimetric assay for the rapid detection of organo-phosphorus pesticide and nerve agent. Analytical Chemistry, 85(1), 308–312.

Liao, Q., Sun, J., & Gao, L. (2008). The adsorption of resorcinol from water using multi-walledcarbon nanotubes. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 312(2), 160–165.

Li-Pook-Than, A., Lefebvre, J., & Finnie, P. (2010). Phases of carbon nanotube growth andpopulation evolution from in situ Raman spectroscopy during chemical vapor deposition. TheJournal of Physical Chemistry C, 114(25), 11018–11025.

Liu, Y., & Kumar, S. (2014). Polymer/Carbon nanotube nano composite fibers–a review. ACSapplied materials & interfaces.

Liu, J., Lai, L., Sahoo, N. G., Zhou, W., Shen, Z., & Chan, S. H. (2012). Carbon nanotube-basedmaterials for fuel cell applications. Australian Journal of Chemistry, 65(9), 1213–1222.

Liu, X., Sun, H., Chen, Y., Lau, R., & Yang, Y. (2008). Preparation of large particle MCM-41 andinvestigation on its fluidization behavior and application in single-walled carbon nanotubeproduction in a fluidized-bed reactor. Chemical Engineering Journal, 142(3), 331–336.

Lota, G., Fic, K., & Frackowiak, E. (2011). Carbon nanotubes and their composites inelectrochemical applications. Energy & Environmental Science, 4(5), 1592–1605.

Lowell, S. (Ed.). (2004). Characterization of porous solids and powders: Surface area, pore sizeand density (Vol. 16). Berlin: Springer.

Lu, C., Chung, Y. L., & Chang, K. F. (2005). Adsorption of trihalomethanes from water withcarbon nanotubes. Water Research, 39(6), 1183–1189.

Lupan, O., Ursaki, V. V., Chai, G., Chow, L., Emelchenko, G. A., Tiginyanu, I. M., & Redkin, A.N. (2010). Selective hydrogen gas nanosensor using individual ZnO nanowire with fastresponse at room temperature. Sensors and Actuators B: Chemical, 144(1), 56–66.

Mae, K. (2007). Advanced chemical processing using microspace. Chemical Engineering Science,62(18), 4842–4851.

Magonov, S., & Alexander, J. (2008). Advanced atomic force microscopy: Exploring measure-ments of local electric properties. Application Note, Agilent Technologies, Inc.

Magonov, S., Alexander, J., & Wu, S. (2011). Advancing characterization of materials with atomicforce microscopy-based electric techniques. In Scanning probe microscopy of functionalmaterials (pp. 233–300). New York: Springer.

Mansoori, G. A., Bastami, T. R., & Ahmadpour, A. (2008). Environmental application ofnanotechnology. In Annual review of nano research. World Scientific Publishing Company.


Marre, S., Roig, Y., & Aymonier, C. (2012). Supercritical microfluidics: Opportunities inflow-through chemistry and materials science. Journal of Supercritical Fluids, 66, 251–264.

Mastromatteo, M., Mastromatteo, M., Conte, A., & Del Nobile, M. A. (2010). Advances incontrolled release devices for food packaging applications. Trends in Food Science &Technology, 21(12), 591–598.

Matteo, C., Candido, P., Vera, R. R., & Francesca, V. (2012). Current and future nanotechapplications in the oil industry. American Journal of Applied Sciences, 9(6), 784–793.

Maynard, A. D., & Pui, D. Y. (2007). Nanotechnology and occupational health: new technologies—new challenges. In Nanotechnology and Occupational Health (pp. 1–3). SpringerNetherlands.

Maynard, A. D., Aitken, R. J., Butz, T., Colvin, V., Donaldson, K., Oberdörster, G. et al. (2006).Safe handling of nanotechnology. Nature, 444(7117), 267–269.

McCarthy, J. R., & Weissleder, R. (2008). Multifunctional magnetic nanoparticles for targetedimaging and therapy. Advanced Drug Delivery Reviews, 60(11), 1241–1251.

McKee, G. S., & Vecchio, K. S. (2006). Thermogravimetric analysis of synthesis variation effectson CVD generated multiwalled carbon nanotubes. The Journal of Physical Chemistry B, 110(3), 1179–1186.

Mehra, N. K., Mishra, V., & Jain, N. K. (2014). A review of ligand tethered surface engineeredcarbon nanotubes. Biomaterials, 35(4), 1267–1283.

Meneghetti, P., & Qutubuddin, S. (2006). Synthesis, thermal properties and applications ofpolymer-clay nanocomposites. Thermochimica Acta, 442(1), 74–77.

Meyer, E., Hug, H. J., & Bennewitz, R. (2004). Scanning probe microscopy: The lab on a tip.Berlin: Springer.

Minagar, S., Berndt, C. C., Wang, J., Ivanova, E., & Wen, C. (2012). A review of the applicationof anodization for the fabrication of nanotubes on metal implant surfaces. Acta biomaterialia,8(8), 2875–2888.

Mishra, A. K., Arockiadoss, T., & Ramaprabhu, S. (2010). Study of removal of azo dyeby functionalized multi walled carbon nanotubes. Chemical Engineering Journal, 162(3),1026–1034.

Missen, R. W., Mims, C. A., & Saville, B. A. (1999). Introduction to chemical reactionengineering and kinetics. (Vol.1). John Wiley & Sons Inc.

Moloney, P. G., & Barrera, E. V. (2013). Application of conductive carbon nanotube fibers andcomposites: Gas sensor. Department of Mechanical Engineering and Materials Science, RiceUniversity, Houston TX.

Murty, B. S., Shankar, P., Raj, B., Rath, B. B., & Murday, J. (2013). Tools to characterizenanomaterials. In Textbook of nanoscience and nanotechnology (pp. 149–175). Berlin,Heidelberg: Springer.

Namburu, P. K., Kulkarni, D. P., Misra, D., & Das, D. K. (2007). Viscosity of copper oxidenanoparticles dispersed in ethylene glycol and water mixture. Experimental Thermal and FluidScience, 32(2), 397–402.

Namin, H. E., Hashemipour, H., & Ranjbar, M. (2008). Effect of aging and calcination onmorphology and properties of synthesized nanocrystalline TiO2. International Journal ofModern Physics B, 22(18n19), 3210–3215.

Nasiri, A., Rashidi, A., Shariaty-Niasar, M., & Soltanian, H. (2013). Preparation and application ofcarbon nanotube nanofluid as a reinforcement of cement slurry.

Nassar, N. N., Husein, M. M., & Pereira-Almao, P. (2011). In-situ prepared nanoparticles insupport of oilsands industry meeting future environmental challenges. Exploration andProduction: Oil and Gas Review, 9(1), 46–48.

Nauman, E. B. (2008). Chemical reactor design, optimization, and scaleup (2nd ed.). Hoboken,NJ: Wiley.

Nel, A. E., Xia, T., Madler, L., & Li, N. (2006). Toxic potential of materials at the nanolevel.Science, 311(5761), 622–627.

Nentwich, M., & Greßler, S. (2012, March). Nano and the environment–Part I: Potentialenvironmental benefits and sustainability effects. NanoTrust Dossiers.

References 173

Nie, X. B., Chen, S. Y., & Robbins, M. O. (2004). A continuum and molecular dynamics hybridmethod for micro-and nano-fluid flow. Journal of Fluid Mechanics, 500, 55–64.

Nowack, B., Krug, H. F., & Height, M. (2011). 120 years of nanosilver history: Implications forpolicy makers. Environmental Science and Technology, 45(4), 1177–1183.

Oberdörster, G., Oberdörster, E., & Oberdörster, J. (2007). Concepts of nanoparticle dose metricand response metric. Environmental Health Perspectives 115, A290.

Olle, B., Bucak, S., Holmes, T. C., Bromberg, L., Hatton, T. A., & Wang, D. I. (2006).Enhancement of oxygen mass transfer using functionalized magnetic nanoparticles. Industrial& Engineering Chemistry Research, 45(12), 4355–4363.

Olsson, C. O., Hörnström, S. E., & Hogmark, S. (1997). Auger electron spectroscopy (pp. 272–288). Weinheim, Germany: Wiley-VCH Verlag GmbH.

Oueiny, C., Berlioz, S., & Perrin, F. X. (2014). Carbon nanotube–polyaniline composites.Progress in Polymer Science, 39(4), 707–748.

Panella, B., Hirscher, M., & Roth, S. (2005). Hydrogen adsorption in different carbonnanostructures. Carbon, 43(10), 2209–2214.

Pankhurst, Q. A., Connolly, J., Jones, S. K., & Dobson, J. (2003). Applications of magneticnanoparticles in biomedicine. Journal of Physics. D. Applied Physics, 36(13), R167.

Papazoglou, E. S., & Parthasarathy, A. (2007). Bionanotechnology. Synthesis Lectures onBioMedical Engineering, 2(1), 1–139.

Petosa, A. R., Jaisi, D. P., Quevedo, I. R., Elimelech, M., & Tufenkji, N. (2010). Aggregation anddeposition of engineered nanomaterials in aquatic environments: Role of physicochemicalinteractions. Environmental Science & Technology, 44(17), 6532–6549.

Paul, D. R., & Robeson, L. M. (2008). Polymer nanotechnology: Nanocomposites. Polymer, 49(15), 3187–3204.

Pérez-Cabero, M., Romeo, E., Royo, C., Monzón, A., Guerrero-Ruız, A., & Rodrıguez-Ramos, I.(2004). Growing mechanism of CNTs: A kinetic approach. Journal of Catalysis, 224(1),197–205.

Perfézou, M., Turner, A., & Merkoçi, A. (2012). Cancer detection using nanoparticle-basedsensors. Chemical Society Reviews, 41(7), 2606–2622.

Pezzin, S. H., Amico, S. C., Coelho, L. A. F., & de Andrade, M. J. (2011). Nanoreinforcements fornanocomposite materials. In Nanostructured materials for engineering applications (pp. 119–131). Berlin, Heidelberg: Springer.

Philippe, R., Morançais, A., Corrias, M., Caussat, B., Kihn, Y., Kalck, P., & Serp, P. (2007).Catalytic production of carbon nanotubes by fluidized-bed CVD. Chemical Vapor Deposition,13(9), 447–457.

Philippe, R., Serp, P., Kalck, P., Kihn, Y., Bordère, S., Plee, D., & Caussat, B. (2009). Kineticstudy of carbon nanotubes synthesis by fluidized bed chemical vapor deposition. AIChEJournal, 55(2), 450–464.

Popov, A. P., Priezzhev, A. V., Lademann, J., & Myllylä, R. (2005). TiO2 nanoparticles as aneffective UV-B radiation skin-protective compound in sunscreens. Journal of Physics.D. Applied Physics, 38(15), 2564.

Pour, A. N., Housaindokht, M. R., Tayyari, S. F., & Zarkesh, J. (2010). Effect of nano-particle sizeon product distribution and kinetic parameters of Fe/Cu/La catalyst in Fischer-Tropschsynthesis. Journal of Natural Gas Chemistry, 19(2), 107–116.

Powell, C. E., & Beall, G. W. (2007). Physical properties of polymer/clay nanocomposites. InPhysical properties of polymers handbook (pp. 561–575). New York: Springer.

Pradeep, T. (2007). Nano: The essentials, understanding nanoscience and nanotechnology. NewYork: McGraw-Hill Publishing.

Prasher, R., Phelan, P. E., & Bhattacharya, P. (2006a). Effect of aggregation kinetics on the thermalconductivity of nanoscale colloidal solutions (nanofluid). Nano Letters, 6(7), 1529–1534.

Prasher, R., Song, D., Wang, J., & Phelan, P. (2006b). Measurements of nanofluid viscosity and itsimplications for thermal applications. Applied Physics Letters, 89(13), 133108.

Prud’homme, R. (2010). Flows of reactive fluids (Vol. 94). Berlin: Springer.


Puretzky, A. A., Geohegan, D. B., Jesse, S., Ivanov, I. N., & Eres, G. (2005). In situ measurementsand modeling of carbon nanotube array growth kinetics during chemical vapor deposition.Applied Physics A, 81(2), 223–240.

Qiao, S. Z., Liu, J., & Lu, G. Q. M. (2011). Synthetic chemistry of nanomaterials in moderninorganic synthetic chemistry. Amsterdam: Elsevier.

Quirke, N. (2006). Adsorption and Transport at the Nanoscale. Taylor & Francis Group, LLC.Reschetilowski, W. (2013). Principles of microprocess technology. In Microreactors in

preparative chemistry: Practical aspects in bioprocessing, nanotechnology, catalysis andmore (pp. 1–12).

Roberge, D. M., Ducry, L., Bieler, N., Cretton, P., & Zimmermann, B. (2005). Microreactortechnology: A revolution for the fine chemical and pharmaceutical industries? ChemicalEngineering and Technology, 28(3), 318–323.

Roco, M. C. (2004). Nanoscale science and engineering: Unifying and transforming tools. AIChEJournal, 50(5), 890–897.

Roco, M. C. (2007). National nanotechnology initiative-past, present, future. In Handbook onnanoscience, engineering and technology (Vol. 2).

Roco, M. C. (2011). The long view of nanotechnology development: The NationalNanotechnology Initiative at 10 years. In Nanotechnology research directions for societalneeds in 2020 (pp. 1–28). Amsterdam: Springer Netherlands.

Roxworthy, B. J., & Toussaint, K. C. (2013, April). Plasmonic nanotweezers for applications inlife sciences. In Optical trapping applications (pp. TW4D-4). Optical Society of America.

Sadrieh, N., Wokovich, A. M., Gopee, N. V., Zheng, J., Haines, D., Parmiter, D., et al. (2010).Lack of significant dermal penetration of titanium dioxide (TiO2) from sunscreen formulationscontaining nano-and sub-micron-size TiO2 particles. Toxicological Sciences, kfq041.

Saeedinia, M., Akhavan-Behabadi, M. A., & Razi, P. (2012). Thermal and rheologicalcharacteristics of CuO–base oil nanofluid flow inside a circular tube. InternationalCommunications in Heat and Mass Transfer, 39(1), 152–159.

Sakintuna, B., Lamari-Darkrim, F., & Hirscher, M. (2007). Metal hydride materials for solidhydrogen storage: A review. International Journal of Hydrogen Energy, 32(9), 1121–1140.

Salah, N., Habib, S. S., Khan, Z. H., Memic, A., Azam, A., Alarfaj, E., & Al-Hamedi, S. (2011).High-energy ball milling technique for ZnO nanoparticles as antibacterial material.International Journal of Nanomedicine, 6, 863.

Sanchez-Garcia, M. D., Lopez-Rubio, A., & Lagaron, J. M. (2010). Natural micro andnanobiocomposites with enhanced barrier properties and novel functionalities for foodbiopackaging applications. Trends in Food Science & Technology, 21(11), 528–536.

Savage, N., & Diallo, M. S. (2005). Nanomaterials and water purification: Opportunities andchallenges. Journal of Nanoparticle Research, 7(4–5), 331–342.

Schlemmer, D., Angélica, R. S., & Sales, M. J. A. (2010). Morphological and thermomechanicalcharacterization of thermoplastic starch/montmorillonite nanocomposites. CompositeStructures, 92(9), 2066–2070.

Scholz, M. S., Blanchfield, J. P., Bloom, L. D., Coburn, B. H., Elkington, M., Fuller, J. D., &Bond, I. P. (2011). The use of composite materials in modern orthopaedic medicine andprosthetic devices: A review. Composites Science and Technology, 71(16), 1791–1803.

Seaton, A., Tran, L., Aitken, R., & Donaldson, K. (2010). Nanoparticles, human health hazard andregulation. Journal of The Royal Society Interface, 7, S119–S129.

Settle, F. A. (1997). Handbook of instrumental techniques for analytical chemistry. EnglewoodCliffs, New Jersey: Prentice Hall PTR.

Sharma, A., Tomar, M., & Gupta, V. (2013). Enhanced response characteristics of SnO2 thin filmbased NO2 gas sensor integrated with nanoscaled metal oxide clusters. Sensors and ActuatorsB: Chemical, 181, 735–742.

Shen, Y., & Yoshikawa, K. (2013). Recent progresses in catalytic tar elimination during biomassgasification or pyrolysis-A review. Renewable and Sustainable Energy Reviews, 21, 371–392.

Shong, C. W., Haur, S. C., & Wee, A. T. S. (2010). Science at the nanoscale: An introductorytextbook. Singapore: Pan Stanford Publishing Pte. Ltd.

References 175

Singh, S. K., Ahmed, R. M., & Growcock, F. (2010, January). Vital role of nanopolymers indrilling and stimulations fluid applications. In SPE Annual Technical Conference andExhibition. Society of Petroleum Engineers.

Siró, I., & Plackett, D. (2010). Microfibrillated cellulose and new nanocomposite materials: Areview. Cellulose, 17(3), 459–494.

Smith, R. (2005). Chemical process design and integration. England: Wiley.Song, Y., Modrow, H., Henry, L. L., Saw, C. K., Doomes, E. E., Palshin, V., & Kumar, C. S. (2006).

Microfluidic synthesis of cobalt nanoparticles. Chemistry of Materials, 18(12), 2817–2827.Souza Filho, A. G., & Fagan, S. B. (2011). Nanomaterials properties. In Nanostructured materials

for engineering applications (pp. 5–22). Berlin, Heidelberg: Springer.Stephanopoulos, N., Solis, E. O., & Stephanopoulos, G. (2005). Nanoscale process systems

engineering: Toward molecular factories, synthetic cells, and adaptive devices. AIChE Journal,51(7), 1858–1869.

Sun, S. S., Jia, F. W., & Li, Z. J. (2012a). Investigation on the dispersion of SiC nanoparticles inrubber matrix. Applied Mechanics and Materials, 182, 139–143.

Sun, Y. F., Liu, S. B., Meng, F. L., Liu, J. Y., Jin, Z., Kong, L. T., & Liu, J. H. (2012b). Metaloxide nanostructures and their gas sensing properties: A review. Sensors, 12(3), 2610–2631.

Sung, W. P., Tsai, T. T., Wu, M. J., Wang, H. J., & Surampalli, R. Y. (2010). Removal of indoorairborne bacteria by nano-Ag/TiO2 as photocatalyst: Feasibility study in museum and nursinginstitutions. Journal of Environmental Engineering, 137(3), 163–170.

Taglauer, E. (1991). Ion scattering spectroscopy. In Ion spectroscopies for surface analy-sis (pp. 363–416). New York, USA: Springer.

Takekawa, T., Nakagawa, K., & Hashiguchi, G. (2005, June). The AFM tweezers: Integration of atweezers function with an AFM probe. In Solid-state sensors, actuators and microsystems,2005. The 13th International Conference on Digest of Technical Papers. TRANSDUCERS’05(Vol. 1, pp. 621–624). IEEE.

Tendeloo, G. V., & Amelinckx, S. (2000). 12 electron microscopy of fullerenes and relatedmaterials. In Characterization of nanophase materials (p. 353).

Texter, J., & Tirrell, M. (2001). Chemical processing by self-assembly. AIChE Journal, 47(8),1706–1710.

Tian, X., Li, T., Yang, K., Xu, Y., Lu, H., & Lin, D. (2012). Effect of humic acids onphysicochemical property and Cd (II) sorption of multiwalled carbon nanotubes. Chemosphere,89(11), 1316–1322.

Thostenson, E. T., Li, C., & Chou, T. W. (2005). Nanocomposites in context. Composites Scienceand Technology, 65(3), 491–516.

Tran, K. Y., Heinrichs, B., Colomer, J. F., Pirard, J. P., & Lambert, S. (2007). Carbon nanotubessynthesis by the ethylene chemical catalytic vapor deposition (CCVD) process on Fe Co, andFe–Co/Al2O3 sol–gel catalysts. Applied Catalysis, A: General, 318, 63–69.

Uddin, N. M., Capaldi, F. M., & Farouk, B. (2012). Molecular dynamics simulations of carbonnanotube dispersions in water: Effects of nanotube length, diameter, chirality and surfactantstructures. Computational Materials Science, 53(1), 133–144.

Velasco, V., Martinez, A., Recio, J., Hernando, A., & Crespo, P. (2012). Synthesis andcharacterization of FePt nanoparticles by high energy ball milling with and without surfactant.Journal of Alloys and Compounds, 536, S13–S16.

Verma, A., & Stellacci, F. (2010). Effect of surface properties on nanoparticle–cell interactions.Small (Weinheim an der Bergstrasse, Germany), 6(1), 12–21.

Veronovski, N., Rudolf, A., Smole, M. S., Kreže, T., & Geršak, J. (2009). Self-cleaning andhandle properties of TiO2-modified textiles. Fibers and Polymers, 10(4), 551–556.

Verwey, E. J. W., & Overbeek, J. T. G. (1999). Theory of the stability of lyophobic colloids.Courier Dover Publications.

Vo-Dinh, T. (2004). Nanobiosensor. In H. S. Nalwa (Ed.), Encyclopedia of nanoscience andnanotechnology (Vol. 6). American Scientific Publishers.

Wang, Z. L. (Ed.). (2000a). Characterization of nanophase materials (pp. 13–14). New York:Wiley-VCH.


Wang, Z. L. (2000b). Transmission electron microscopy and spectroscopy of nanoparticles. InCharacterization of nanophase materials (Vol. 1, pp. 37–80).

Wang, J., & Pui, D. Y. (2013). Dispersion and filtration of carbon nanotubes (CNTs) andmeasurement of nanoparticle agglomerates in diesel exhaust. Chemical Engineering Science,85, 69–76.

Wang, Z. L., Tang, D. W., Liu, S., Zheng, X. H., & Araki, N. (2007). Thermal conductivity andthermal diffusivity measurements of nanofluids by 3ω method and mechanism analysis of heattransport. International Journal of Thermophysics, 28(4), 1255–1268.

Watts, P., & Wiles, C. (2007). Recent advances in synthetic micro reaction technology. ChemicalCommunications, 5, 443–467.

Wei, D., & Liu, Y. (2010). Controllable synthesis of graphene and its applications. AdvancedMaterials, 22(30), 3225–3241.

Wilkening, G., & Koenders, L. (2006). Nanoscale calibration standards and methods: Dimensionaland related measurements in the micro and nanometer range. New York: Wiley.

Williams, D. B., & Carter, C. B. (2009). High-resolution TEM. In Transmission electronmicroscopy (pp. 483–509). New York, USA: Springer.

Xu, R. (2001). Particle characterization: Light scattering methods (Vol. 13). New York: Springer.Yang, S., Ceylan, H., Gopalakrishnan, K., & Kim, S. (2014). Smart airport pavement

instrumentation and health monitoring.Yang, S. Y., Cheng, F. Y., Yeh, C. S., & Lee, G. B. (2010). Size-controlled synthesis of gold

nanoparticles using a micro-mixing system. Microfluidics and Nanofluidics, 8(3), 303–311.Yang, J. C., Li, F. C., Zhou, W. W., He, Y. R., & Jiang, B. C. (2012). Experimental investigation

on the thermal conductivity and shear viscosity of viscoelastic-fluid-based nanofluids.International Journal of Heat and Mass Transfer, 55(11), 3160–3166.

Yang, Y., Zhang, H., Wang, P., Zheng, Q., & Li, J. (2007). The influence of nano-sized TiO2

fillers on the morphologies and properties of PSF UF membrane. Journal of MembraneScience, 288(1), 231–238.

Yao, Y., Xu, F., Chen, M., Xu, Z., & Zhu, Z. (2010). Adsorption behavior of methylene blue oncarbon nanotubes. Bioresource technology, 101(9), 3040–3046.

Yen, B. K., Günther, A., Schmidt, M. A., Jensen, K. F., & Bawendi, M. G. (2005).A microfabricated gas-liquid segmented flow reactor for high-temperature synthesis: The caseof cdse quantum dots. Angewandte Chemie, 117(34), 5583–5587.

YeonáLee, H., KunáPark, H., MiáLee, Y., & BumáPark, S. (2007). A practical procedure forproducing silver nanocoated fabric and its antibacterial evaluation for biomedical applications.Chemical Communications, 28, 2959–2961.

Yi, P., & Chen, K. L. (2014). Release kinetics of multiwalled carbon nanotubes deposited on silicasurfaces: Quartz crystal microbalance with dissipation (QCM-D) measurements and modeling.Environmental Science & Technology, 48(8), 4406–4413.

You-qi, Y. A. N. G. (2008). Microscale and nanoscale process systems engineering: Challenge andprogress. The Chinese Journal of Process Engineering, 8(3), 616–624.

Yu, W., & Choi, S. U. S. (2003). The role of interfacial layers in the enhanced thermalconductivity of nanofluids: A renovated Maxwell model. Journal of Nanoparticle Research, 5(1–2), 167–171.

Yu, G., Gong, J., Wang, S., Zhu, D., He, S., & Zhu, Z. (2006). Etching effects of ethanol onmulti-walled carbon nanotubes. Carbon, 44(7), 1218–1224.

Yuan, G. (2004). Natural and modified nanomaterials as sorbents of environmental contaminants.Journal of Environmental Science and Health, Part A, 39(10), 2661–2670.

Zanchet, D., Hall, B. D., & Ugarte, D. (2000). X-ray characterization of nanoparticles. InCharacterization of nanophase materials (pp. 13–36). New York, USA: Wiley-VCHVerlagGmbH.

Zhang, M., & Dai, L. (2012). Carbon nanomaterials as metal-free catalysts in next generation fuelcells. Nano Energy, 1(4), 514–517.

References 177

Zhang, X., Gu, H., & Fujii, M. (2007). Effective thermal conductivity and thermal diffusivity ofnanofluids containing spherical and cylindrical nanoparticles. Experimental Thermal and FluidScience, 31(6), 593–599.

Zhang, Y., Yu, Q., Wang, X., & Tian, Y. (2014). A new understanding of carbon nanotubegrowth: Activation and deactivation of a catalyst. Applied Surface Science, 298, 221–224.

Zhao, Q. Q., Boxman, A., & Chowdhry, U. (2003). Nanotechnology in the chemical industry–opportunities and challenges. Journal of Nanoparticle Research, 5(5–6), 567–572.

Zhao, C. X., He, L., Qiao, S. Z., & Middelberg, A. P. (2011). Nanoparticle synthesis inmicroreactors. Chemical Engineering Science, 66(7), 1463–1479.

Zhao, P., Li, N., & Astruc, D. (2013). State of the art in gold nanoparticle synthesis. CoordinationChemistry Reviews, 257(3), 638–665.

Zubizarreta, L., Arenillas, A., & Pis, J. J. (2009). Carbon materials for H2 storage. InternationalJournal of Hydrogen Energy, 34(10), 4575–4581.


Chapter 3Learning Synergism in Nanotechnologyand Chemical Engineering by Case Study

3.1 Chemical Vapor Deposition (CVD) Techniquesin Nanotechnology

Chemical vapor deposition (CVD) is a versatile process applied to producehigh-purity, high-performance solid materials by a chemical reaction ofvapor-phase precursors (Vahlas et al. 2006). In typical CVD, a heated substrate isexposed to one or more volatile precursors, which decompose near or on the surfaceof substrate to form a solid deposit. As a result of reaction that takes place, volatileby-products are also released (Pierson 1999). The chemical reactions of precursorspecies occur both in the gas phase and in the solid phase, where is surface ofdeposition. In general, CVD reactions are thermodynamically endothermic, andhence, energy has to be supplied to the reactor. Early example of using CVD is forthe electron industry to produce ultraclean silicon, semiconductors and otherelectronic components (Jones and Hitchman 2009). Traditionally, reactions werepromoted or initiated by heat (thermal CVD). However, elevated temperatures fordeposition put some restrictions on the desired type of substrates to be coated,which is not favorable. The high temperature also leads to stresses in the filmdeposited on materials and causes mechanical instabilities in it (Park and Sudarshan2001). One way of reducing growth temperatures is to use plasma-assisted orplasma-enhanced CVD (PECVD) (Hess and Graves 1989). In this technique,electrical energy rather than thermal energy is used to initiate reactions anddeposition can occur at low temperatures, even close to ambient. The ability ofPECVD to achieve low-temperature deposition is often critical in manufacturing ofsemiconductors and organic coatings. Although PECVD usually allowslower-temperature deposition than thermal CVD, it has some drawbacks. Forexample, the plasma bombardment of a surface often causes damage of the catalystsubstrate and hence the growing film. PECVD has also strong process dependencyon several parameters such as power and frequency of the source, gas pressure,reagent flow rate, reactor geometry, etc. (Jones and Hitchman 2009). Other methods


179

of introducing energy to the CVD process are laser-enhanced CVD (LECVD) alsocalled photo-assisted CVD (PCVD) (McCaulley et al. 1989). In this type of CVD,laser is used to enhance the surface reactions. Two processes simultaneously occur:pyrolytic process, in which substrate is heated to enhance reactions, and photolyticprocess by use of ultraviolet (UV) radiation, in which molecules of gas phase isdissociated to enhance reactivity. Other types of CVD are atomic layer deposition(ALD) and the specialist version atomic layer epitaxy (ALE); however, most ofthese techniques are too expensive and are rarely utilized (Leskelä and Ritala 2002).

Besides the need of energy to initiate the reaction in the CVD process, theefficiency of CVD process principally depends on appropriate contact between thereactive gas phase and the solid particles treated. Based on this requirement, CVDtechnology, with regard to both the type of reactor and source of energy used, hasgone through a wide range of developments over the years (Park and Sudarshan2001; Vahlas et al. 2006; Jones and Hitchman 2009). The flow into a reactor can bein a vertical or horizontal manner, and schematic illustrations of some commonreactor types are shown in Fig. 3.1. The choice of process and reactor is determinedby the type of reactant used and the products and their applications in addition toeconomic considerations.

Irrespective of the different configurations and chemistries employed in CVDprocesses, they have common features:

Fig. 3.1 Contacting modesfor CVD on powders in gas–solid reactors. Adopted fromVahlas et al. (2006)

180 3 Learning Synergism in Nanotechnology and Chemical Engineering by Case Study

• Precursor sources or reactant gasesIt is the precursors of deposition that must be injected into the reactor andundergo desirable and undesirable reactions that take place both in the gas phaseand in the solid phase; surface of deposition.

• A reaction zoneThe reaction zone is the part of an enclosed chamber that is heated by a sur-rounding oven or furnace or by external radiofrequency or infrared radiation.

• Controlling systemsGas handling system controls the input rate of precursor gases or vapors to thereaction zone.Temperature controlling system, like thermocouple, is required to accuratelycontrol the temperature of reaction zone.

• An exhaust systemBy-products of the reactions, usually in form of gas, must be removed from thechamber. The outlet of reactor may be equipped with a vacuum pump forlow-pressure operation and remove the gases from the chamber.

Another category of CVD process is based on the type of precursor and chemicalreaction used. This category introduces many derivatives of the CVD terminology,such as metal–organic chemical vapor deposition (MOCVD), organo-metallicchemical vapor deposition (OMCVD). In these techniques where metallo-organicprecursors are used, the temperature process is considerably lowered.

According to the conditions used, deposits can be in various forms includingfilms, powders and composites. The most promising characteristic of CVD reactionis when gas decomposes an ultrapure solid deposit. Accordingly, CVD is broadlyemployed in the electronic industries to produce ultraclean silicon from silicongases (Bhattacharya and Tummala 2001; Habuka et al. 2010). Beyond pioneeringapplications of CVD in the electronic industry, this technique is widely used formanufacturing many industrial products including heterogeneous catalysis, semi-conductors, optoelectronics, coatings, fibers, monolithic components and manyother products (Pierson 1999; Caussat and Vahlas 2007). The most commerciallyimportant products produced by CVD include polycrystalline silicon, silicondioxide, silicon nitride and diamond. The recent promising application of CVDprocess is in bulk production of nanotechnology-based products like carbonnanotubes, carbon nanofibers, nanocoating and nanotitanuim (Caussat and Vahlas2007). In addition to the wide variety of materials that can be deposited, theexcellent advantage of CVD is that the solid can be deposited with very high purity.The advantage of CVD for depositing thin films is that a quite conformal film canbe produced; i.e., the film thickness on the sidewalls of features is comparable to thethickness on the top (Park and Sudarshan 2001). Synthesis of nanostructures andtheir treatment is playing an imperative role for industrial development ofnanotechnology-based products.

3.1 Chemical Vapor Deposition (CVD) Techniques in Nanotechnology 181

3.1.1 Basic Principles of CVD

CVD processes are quite complex and involve a series of reactions that occur bothin the gas phase and in the solid phase at elevated temperatures (200–1600 °C).These chemical reactions include thermal decomposition (pyrolysis), reduction,hydrolysis, disproportionation, oxidation, carburization and nitridation, which canbe used either singly or in combination (Pierson 1999). The overall result of thesereactions is deposition of solid and release of volatile gases (Park and Sudarshan2001):

The basic physicochemical steps hypothesized to occur during CVD processinclude:

1. Transport of precursors in the bulk gas flow region into the reactor (fluiddynamics);

2. Evaporation of reactant gases;3. Chemical reactions of reactant gases to produce intermediate reactants and

gaseous by-products (gas-phase chemistry or gas-phase reaction);4. Diffusion of reactant gases through the gaseous boundary layer to the substrate

surface (mass transfer);5. Adsorption of the reactants on the substrate surface;6. Surface diffusion to growth sites, nucleation and formation of solid deposit at

the surface (surface reaction or surface chemistry);7. Desorption of volatile by-products from the surface;8. Mass transport of gaseous by-products away from the reactor.

These eight mechanistic steps are illustrated in Fig. 3.2. It is worth mentioningthat all these steps are not essentially consecutive and some steps may occursimultaneously.

Fig. 3.2 Schematic illustrations of the physicochemical steps that occur during CVD process


The first step in CVD process is evaporation of precursors to reactant gases,which can be converted into the desired solid deposits through some reaction.Precursor for CVD process can be gases, volatile liquids, sublimable solids orcombinations; however, it is suitable for CVD process if it possesses the charac-teristics described as follows:

• Adequate volatilityPrecursor needs to have high enough partial pressure to get good growth rates.

• Adequate stabilityPrecursor should not be decomposed during evaporation.

• A sufficiently large temperature ‘‘window’’ between evaporation anddecomposition

• Chemical purityContaminants will affect the volatility and stability of the precursor. They alsomay lead to the formation of unwanted by-products.

• Readily manufactured in high yield at low cost; in other words, to producedesired deposits with easily removable by-products

• Lower reaction temperature than melting point of substrate• Low toxicity, non-hazardous or a low hazard risk.

Although volatility of precursor is prerequisite of the multistepped CVD process,it is not sufficient and efficiency of process depends on the successful accom-plishment of the other steps, addressed above. Availability of the volatile chemical,in other words, the way the reactant gases can be transported to the surface, wherereaction happens, is an essential key. It is also imperative to know how the vaporscan react in the gas phase or on the surfaces to be converted into the deposits(product). To sum up, the rate of CVD process depends on

• transport of reactant gas to deposition surface,• adsorption of reactant gas on deposition surface,• reaction rates,• transport of products away from the surface.

CVD process takes place at the molecular level, driven by a variety of physicaland chemical fundamentals that control the deposition rate and quality of the soliddeposit. These fundamentals involve thermodynamics, kinetics, transport phe-nomena (gas flow dynamics, mass and heat transfer), physics of surface adsorptionand crystal growth. Figure 3.3 shows an illustration of the fundamentals that finallycontrol the CVD process. The control strategies for CVD are developed throughexperimentation and modeling, with feedback from material analysis that requires acomprehensible understanding of the CVD process.

Any reaction including CVD reaction is governed by thermodynamics thatindicates the direction of the reaction and by kinetics that indicate the rate of thereactions. The kinetics of the reaction accompanied with transport phenomenadetermines the mechanism that controls the rate of the reaction. Besides, to getinsight of these principles that govern CVD processes, it is essential to obtain deep


understanding of the constraints involved in this technique. These subjects areexplained in the following sections. In case of using energy source rather thanthermal sources like PECVD, understanding of the plasma and quantum physics isalso necessary. However, these subjects are out of the concentration of this book.Theoretical aspects for modeling of CVD process can be categorized as thermo-dynamic analysis and overall reaction rate analysis, which involves kinetics andtransport phenomena investigation (Fig. 3.4).

Before embarking on the development of any new CVD process, feasibility ofthe system should be determined through thermodynamic analysis (Pierson 1999).Chemical thermodynamics expresses the interrelation of different forms of energy.It is also involved in the energy transfer from one chemical system to another basedon the first and second laws of thermodynamics. In CVD process, energy transferoccurs when the gaseous compounds, introduced to the reactor, react to form thesolid deposit and by-product gases. Thermodynamic analysis of CVD providesimportant information about whether the phases of interest are thermochemicallyallowed to form from a proposed precursor system and whether secondary phasescan form (Jones and Hitchman 2009). It also provides valuable assistance in thechoice of process parameters including reactant concentrations, pressures andtemperatures to use for a given chemical system (Park and Sudarshan 2001).

The most comprehensive thermochemical approach for assessing a CVD systemis to determine the Gibbs free-energy change in a deposition reaction ðDG�

rxnÞ forthe system as the precursors are computationally allowed to react and reach equi-librium. Any reaction will happen if the Gibbs free-energy change in the reaction is

Fig. 3.3 Illustration of various fundamental aspects involved in a CVD process


negative. To determine DG�rxn requires a summation of the Gibbs free energies of

formation ðDG�f Þ for constituents at the temperature of interest, defined as:

DG�rxn ¼

XDG�

f product �X

DG�f reactant ð3:1Þ

DG�f = Standard Gibbs free energy of formation of species “i” at standard condition

(T = 298 K and P = 1 atm).The free energy of formation is a function of several parameters including the

type of reactants, the molar ratio of these reactants, the process temperature and theprocess pressure (Pierson 1999). This relationship is represented by the followingequation:

DG�f ¼ DH�

f þZT298

DCPdT � TDS� �ZT298

ðCP=TÞdT ð3:2Þ

Heat transport

Fluid dynamics

Mass transport

Surface kineticsTransport Phenomena

Overall reaction rate analysisThermodynamic analysis

Chemical vapor deposition modeling

Reaction rateDiffusion

GrowthNucleation

Fig. 3.4 Theoretical aspects for modeling of CVD process


The most accurate approach to determining whether desired phases will formrequires computing the minimum total Gibbs free energy (G) for the system. GlobalGibbs free-energy minimization that considers all possible gaseous species andcondensed phases, as well as potential complex solid solution–defect structures inthe deposited phases, is expressed as (Jones and Hitchman 2009).

G ¼Xj

Xi

n ji G

j ð3:3Þ

where n is the number of moles of species i in phase j.According to the explanation above, the fundamental assumption of thermo-

dynamic analysis of chemical processes relies on reaching chemical equilibrium inthe system. A system is in equilibrium when the Gibbs free energy is at a minimum.Although the presumption of chemical equilibrium is not realistic, given the rela-tively short residence time of precursors in CVD reactors, reactions will proceedtoward equilibrium to a sufficient extent that thermodynamic modeling is still veryuseful for gaining process insights (Jones and Hitchman 2009). Calculation of thethermodynamic equilibrium of a CVD system provides useful information on thecharacteristics and behavior of the reaction, including (Pierson 1999)

• The composition and amount of deposited material (the maximum efficiency foruse of reactants) that is theoretically possible under any given set of depositionconditions that is at a given temperature, a given pressure and given inputconcentration of reactants,

• The existence of gaseous species and their equilibrium partial pressures,• The possibility of multiple reactions and the number and composition of pos-

sible solid phases, with the inclusion of the substrate as a possible reactant,• The likelihood of a reaction between the substrate and the gaseous or solid

species,• The optimum range of deposition conditions.

Despite its valuable role in the development of CVD process, thermodynamicanalysis is subject to some limitations (Park and Sudarshan 2001; Jones andHitchman 2009):

• Thermodynamic analysis only gives information on the theoretically possibleresults, which may not actually be achievable. CVD systems are generally notoperated at chemical equilibrium.

• The analysis requires data on the enthalpy, entropy and heat capacity for allcomponents. Such data are not always available, particularly for the newerprecursors.

• The results of thermodynamic analysis for many CVD reactions must be treatedwith caution because of their complexity that influences the reliability of theresult.

• Constraints can be also imposed on equilibrium calculations as a way of startingto incorporate kinetic limitations. For example, if nitrogen is used as a carrier


gas and has been experimentally shown not to participate in the process, speciessuch as N atoms and solid nitride species would be omitted from the equilibriumcalculation to account for the kinetic inaccessibility of these species.

• Thermodynamic modeling, which is based on equilibrium, is much less suc-cessful when applied to low-temperature processes. There are no firm guidelineswith regard to temperatures or other conditions that govern whether depositedsystems are near or far from equilibrium. A rough rule of thumb is thoseprocesses at high temperatures, more than 1000 °C, are governed by equilibriumthermochemistry, whereas those at temperatures lower than 500 °C are poten-tially far from equilibrium.

• Thermodynamic analysis indicates that a chemical reaction can occur, but thekinetics of a chemical reaction may prevent a reaction from occurring.

• It does not provide the answer to the questions like how do the reactant gasesreach the deposition surface and how do they react to form solid deposits?Answer to these questions that determine the phenomena and their ratesinvolved in CVD is essential for design and optimization of the CVD reactor. Tooptimize the deposition reactions and, to some extent, control the nature of thedeposit, it is critical to determine the factor or factors that control the rate ofdeposition.

Mass-transport processes in the vapor phase carry the input CVD precursors,often in a carrier gas, from the injection point to the surface being reacted. Toachieve a uniform deposition, it is necessary to deliver gas uniformly to the surfaceand optimize flow for maximum deposition rate.

The behavior of the gas as it flows is controlled by fluid mechanics. Masstransport on the surface or within the solid can also influence CVD processes, withregard to the morphology and composition of the solid deposit. Diffusion of atomson the surface plays an important role in the initial steps of nucleation and, incompetition with the chemical reactions depositing the atoms on the surface, candetermine the morphology and composition of the product (Park and Sudarshan2001). It should be first emphasize that CVD process is subject to complicated fluiddynamics due to its complexity. Reactant gases enter the reactor by forced flow, andthen, they have to diffuse through the boundary layer to come in contact withdeposition surface, where reaction takes place. The gaseous by-products of thereaction are then diffused away from the surface, through the boundary layer. Theboundary layer is that region in which the flow velocity changes from zero at thesurface to that of the bulk gas away from the surface. This boundary layer starts atthe inlet of the surface, and its thickness increases until the flow is stabilized asshown in Fig. 3.5. Thickness of boundary layer, δ, can be written as (Jones andHitchman 2009):

d �ffiffiffiffiffiffilxqu

rð3:4Þ


where “d” is chamber dimension (m), “µ” is gas viscosity, “ρ” is gas density, “u” isgas velocity, and “x” is distance from inlet in flow direction.

According to the equation, for any gaseous flow, the thickness of the boundarylayer increases with lower gas flow velocity and with increased distance from thetube inlet (Pierson 1999). Reynolds number (Re), which is a dimensionlessparameter, characterizes the regime of the flow of a fluid. This parameter is definedas:

Re ¼ quxl

ð3:5Þ

Then, thickness of the boundary layer can be written as:

d �ffiffiffiffiffiffilxqu

r� xffiffiffiffiffiffi

Rep ð3:6Þ

This correlation indicates that the thickness of the boundary layer is inverselyproportional to the square root of the Reynolds number, and hence, it depends ongas flow pattern.

Basically, there are two types of flow, namely “viscose flow” where intermo-lecular collisions dominate and “molecular flow” where intermolecular collisionsare rare. There is a region between these two called transition regimes. In the bulkgas flow, the gas flow above the boundary layer, there is a viscose flow, wheregaseous reactants are being convected through the reactor by intermolecular col-lisions in the carrier gas flow. The gaseous reactants then diffuse from the bulk gasflow, through boundary layer, to deposition surface due to a concentration gradient.When molecules move from higher concentration area to the lower concentrationarea, there is a molecular flow. Both viscose and molecular flow play important rolein CVD process. The uniformity of the deposition depends on the local diffusionflux to the surface, and that flux depends on the local boundary layer thickness andthe local precursor concentration in the bulk flow (Jones and Hitchman 2009).

In viscose flow, then, there is convection mass transport, whereas in molecularflow there is diffusion mass transfer. Accordingly, these regions, viscose and

Fig. 3.5 Boundary layer


molecular flows, are distinguished by Knudson number, a mass-transport charac-teristic, defined as

Kn ¼ k=L ð3:7Þ

whereλ mean free molecular pathL Characteristic dimension of a system perpendicular to the flow direction.

The “mean free path,” is a well-known parameter in gas dynamics, defined as theaverage distance traveled by a gas molecule between collisions with anothermolecule.

If Kn < 0.01, there is a viscous flow, and if Kn > 1, there is a molecular flow.Transition flow is the region, where 0.01 < Kn < 1.

In case of viscose flow, two regimes are defined. Low flow rates produce laminarflow, and high flow rates produce turbulent flow. For Re less than 100, the flow isconsidered as laminar, and for Re more than 2500, the flow is considered asturbulent.

The rules of the boundary layer apply in most CVD depositions in the viscousflow range where pressure is relatively high. In cases where very low pressure isused (i.e., in the mTorr range), the rules are no longer applicable (Pierson 1999).

Recall that CVD is a multistep process that can be classified into two categories,mass transport and surface reaction steps. Referring to Fig. 3.2, steps 1, 3 and 7, 8are mass-transport processes and surface reaction steps include 2, 4, 5, and 6. Theslowest step determines whether the process is mass transport or surface reactionlimited. Consider a reactant (gaseous precursor) with the molecular formula of “AB”is converted via CVD to two components of “A,” which is a solid deposited, and“B,” which is gas released.

AB gð Þ ! A sð Þ þ BðgÞ

In this process, there are two reactant fluxes, with unit kg/m2.s (or mole/m2 s),one is the flux across boundary to surface (F1) and the other is flux consumed inreaction at the surface (F2); see Fig. 3.6.

According to Fick’s law, the reactant flux (F1) can be written as

F1 ¼ �DdCdx

ð3:8Þ

whereD Is the diffusivity of the reactants (m2/s),C is the concentration of reactant (kg/m3 or mol/m3),x is the direction perpendicular to the substrate surface (m).


An approximation for the concentration gradient is,

dCdx

� DCDx

¼ CG � CS

dð3:9Þ

whereCG concentration of reactant in bulk stream (kg/m3 or mol/m3)CS concentration of reactant at surface (kg/m3 or mol/m3)Δ is the boundary layer thickness (m) that depends on gas flow pattern.

Thus, the reactant diffusion through the boundary layer is described as,

F1 ¼ �DCG � CS

d

� �ð3:10Þ

F1 ¼ Dd

CG � CSð Þ ¼ hg CG � CSð Þ ð3:11Þ

hg ¼ D=d ð3:12Þ

hg gas diffusion rate constant (m/s).

therefore,

F1 ¼ hgðCG � CSÞ ð3:13Þ

F2, which is flux consumed in reaction at the surface, is defined as

F2 ¼ kSCS ð3:14Þ

Fig. 3.6 Illustration of fluxes in CVD; F1 diffusion of reactant species through the boundary layer,CG concentration of reactant in bulk stream (kg/m3 or mol/m3), CS concentration of reactant atsurface (kg/m3 or mol/m3)


where kS = surface reaction parameter (m/s)

kS ¼ koexpð�EA=RTÞ ð3:15Þ

EA Apparent activation energy (j\mol)R gas constant (j/mol K)T temperature (K).

At steady state: F1 = F2 = FThen,

ksCs ¼ hg CG � CSð Þ ð3:16Þ

CS can be derived as

CS ¼ hgks þ hg

CG

F ¼ ksCs ¼ kshgks þ hg

CG ð3:17Þ

Finally,

F ¼ CG1ksþ 1

hg

ð3:18Þ

Therefore, growth rate of film is proportional to F.As described above, the reaction kinetics determines the rate at which a phase

will form and whether its formation is limited by any step in the process. Thereaction will be the rate limiting if kS is small, and thus, growth is controlled by thekinetics of chemical reactions occurring either in the gas-phase or on the substratesurface, including adsorption, decomposition, surface migration, chemical reactionand desorption of products. kS is highly temperature dependent, and it increasesexponentially with temperature rise according to the Arrhenius equation. As aresult, the surface reaction rate rises exponentially, resulting in a mass-transportlimited because transport becomes the slowest step in the series of deposition steps.At lower temperatures, typically less than 600 °C, the deposition rate is generallysurface reaction limited. Reaction resistances are often used to predict rate-limitingsteps in CVD process (Jones and Hitchman 2009; Park and Sudarshan 2001).

If hg is small, hence growth is controlled by transfer to the surface and then theCVD process becomes mass-transfer limited. Since hg is weakly temperaturedependent, the growth rate becomes nearly independent of temperature and iscontrolled by the mass transport of reagents through the boundary layer to depo-sition surface. Consequently, at high temperatures (higher that 600 °C), masstransfer is the common rate-limiting process and the rate of CVD is controlled bymass transport (Jones and Hitchman 2009). If the process is mass-transport limited,


atypical rate-limiting step can be diffusion of reactant species through the boundarylayer. Mass transport depends on parameters, like reactant concentration, diffusivityand boundary layer thickness, pressure, gas velocity, temperature distribution,reactant geometry and gas properties (viscosity). The effect of gas flow rate will besignificant if deposition rate is mass-transfer limited (Fig. 3.7). According toEqs. (3.4)–(3.6), rate of deposition is proportional to the inverse of velocity square.

R / hg / 1d/ 1ffiffiffiffiffiffi

Rep / 1ffiffiffi

up

To achieve a uniform deposition, it is necessary to deliver gas uniformly to thesurface and optimize flow for maximum deposition rate (Park and Sudarshan 2001).

Mass transport has a much weaker dependence on temperature than chemicalreaction kinetics as illustrated in Fig. 3.8. Mass-transfer coefficient (D) is

Fig. 3.7 Effect of gasvelocity on the rate ofdeposition

Fig. 3.8 Deposition rateversus temperature for CVD


proportional to T3/2/P from kinetic theory of gases. Accordingly, reducing pressurewill result in higher D and hence higher deposition rate. This fact is the reason todevelop low-pressure CVD (LPCVD), where gas pressures in the reactor is around1 m Torr–1 Torr (rather than 1 atm) (Jones and Hitchman 2009). In LPCVD due tohigh D, there is often adequate initial gas concentrations in the surface, and thus, thesurface reaction becomes rate limiting. At lower pressures, diffusional transport canbe more important than convective transport. Accordingly, the pressure of the CVDreactor is a crucial factor that determines the relative importance of mass transfer orreaction kinetics. From atmospheric pressure to intermediate pressures (e.g.,10 Torr), gas-phase reactions are important and, in addition, a significant boundarylayer is present. Kinetics and mass transport both can play a significant role indeposition process. As the pressure falls, gas-phase reactions tend to become lessimportant, and particularly at pressures below 1 Torr, layer growth is often con-trolled by surface reactions. At very low pressures (<10−4 Torr), mass transport iscompletely absent and layer growth is primarily controlled by the gas and substratetemperature and by desorption of precursor fragments and matrix elements from thegrowth surface (Jones and Hitchman 2009).

The effect of pressure on deposition rate at fixed temperature is depicted inFig. 3.9. According to the simplified model presented here, process variables can bevaried so that the deposition process is limited either by gas-phase diffusion to thesubstrate surface or by reaction at the substrate. Such control of the process isvaluable because, for example, geometric surface irregularities in the substrate(grooves and corners) are coated uniformly in a kinetically controlled process, butin a process controlled by diffusion a protrusion receives a thicker coating while adepression is thinly coated (Park and Sudarshan 2001).

Fig. 3.9 Effect oftemperature and pressure onthe rate of deposition.Adopted from Park andSudarshan (2001)


3.1.2 Constraints on CVD Processes

The most traditional CVD operations have been readily optimized experimentallyby changing the reaction chemistry, the activation method or the deposition vari-ables until a satisfactory deposit was achieved (Pierson 1999). However, the newCVD processes are complicated with much more exacting requirements, whichwould make the empirical approach too cumbersome. The main problem in theCVD kinetic studies is the complexity of the deposition process and differentphysical and chemical principles that control the rate and quality of deposition(Jones and Hitchman 2009). As mentioned before, deposition depends on severalparameters, the primary ones are precursor, geometry of the deposition surface andprocess conditions (Ci, pressure, velocity, temperature and temperature distribu-tion). There are several limiting factors on deposition rate, but mass transport andsurface kinetic control are the most predominant. In reality, transport and chemicalreactions are closely coupled, with their relative importance varying with the detailsof the operating conditions. The difficulty in kinetic analysis of CVD process arisesnot only from the various steps of the CVD process but also from the processparameters effects, geometric effects (geometry of deposition surface and reactor)and the effects of gas flow patterns in the reaction zones (Park and Sudarshan 2001).Deposition rates can also be supply limited, i.e., determined by the rate at whichreactants are fed into the system, rather than the rate at which they are transportedthrough the reactor to deposition surface. This can be manifested as loading effects,where the deposition rate varies strongly with the area of the deposition surface.

For a particular CVD process, it is often sufficient to identify a few reaction stepsthat are slow enough to be rate determining. However, developing CVD toward novelprocesses increases the complexity of the process andmore stringent requirements areplaced on them; thus, traditional approaches no longer work well enough in opti-mization of such processes (Park and Sudarshan 2001). When treating powders,especially nanoparticles, surface reactions tend to be very intense, due to the highsurface-to-volume ratio of these particles. Surface-to-volume ratios for CVD on flatsubstrates like wafers do not exceed 100 m2/m3 of heated volume, whereas those forCVD on powders vary between 104 and 3 × 107 m2/m3 (Allouche and Monthioux2005). An efficient contact between particles and the reactive gas phase will result inuniform deposit. Fluidized beds are the best-known and efficient gas–solid contac-tors.Mass transfers in fluidized beds are very high that guarantee a good uniformity ofCVD treatment as well as a high conversion rate of the gaseous precursors. Intensethermal transfers also exist between the solid and the gas phase, and the reactor wallsin fluidized beds. Accordingly, they are mostly considered isothermal andmake thesetypes of reactors a suited medium for performing thermal CVD. Nevertheless, there isa main limitation for performing CVD in fluidized beds that only a limited size rangeof powder can be fluidized, roughly those with mean diameter between 50 and800 µm and with grain density between 1500 and 4000 kg/m3 (Geldart 1973). Thetreatment of individual particles of micrometer or nanometer size is difficult becausefor these fine powders, the interparticle forces, especially the van derWaals force, are


higher than the forces exerted by gas flow. Consequently, either particles do notfluidize at all or they can fluidize under the form of agglomerates of millimeter size.Formation of such agglomerates prevents uniform deposition around each individualparticle, and CVD consolidates the agglomerates both externally and in the voids ofthe intergrain porosity. To overcome these problems, external energy, for instancevibrating, can be applied to a fluidized bed to somewhat overcome the interparticleforces (Alavi and Caussat 2005).

Production of nanotechnology-based products in laboratory scale has beengoverned but upscaling toward industrial exploitation seems difficult. The reason islarge number of parameters needs to be considered. In order to take advantage ofusing CVD process, it is necessary to deal with its constraints such as side depo-sition, precursor handling or homogeneous nucleation. This can only be achievedthrough an integrated, interdisciplinary approach with inputs from chemistry,chemical engineering and materials science. There are some approaches to mod-eling the formation of particle in CVD process, but they are not mature enough tomake them applied for the formation of nanostructures by CVD process. Predictionof particle nucleation and growth by surface reaction are the major challenges formodeling of such systems. The prediction of particle sintering is also a significantchallenge for intentional synthesis of nanostructures (Caussat and Vahlas 2007).

3.1.3 CVD Synthesis of Carbon Nanotubes (CNTs)

Carbon nanotube growth by the chemical vapor deposition method is accomplishedin a reaction furnace with flowing gaseous form of carbon feedstock in the presenceof catalyst. Generally, the yield and features of produced CNTs, includingdimensions, wall number, chirality and graphitization, are determined by the growthmechanism and process conditions. The mechanism developed for CNT growth viaCVD basically stems from vapor–liquid–solid (VLS) mechanism (Wagner et al.1964; Peigney 2001; Little et al. 2003). The model consists of three relativelyindependent steps, which lead to the formation of tubular carbon solid hybridized insp2 configuration, combination of an “s” orbital and two “p” orbitals from the samevalence shell. These three steps include

1. Decomposition of carbon source (any hydrocarbon, alcohol, CO2, CO) on thesurface of metallic catalyst.Carbon source is decomposed on the surface of a metallic catalyst. The gasesreleased in this step come back into the gas phase, and the carbon diffuses overand throughout the catalyst particle.

2. Saturation of carbon atoms on catalyst particles.Saturation occurs either by reaching the carbon solubility limit in the metal atgiven temperature or by lowering the solubility limit by a relative temperaturedecrease.

3. Precipitation of solid carbon.


Once carbon levels reach supersaturation, solid carbon is precipitated. Theformation of CNT results from carbon precipitation, and its morphology dependsupon the precise system conditions such as, temperature, strength of the metal–substrate interaction. During CVD, the actual carbon source derives primarily fromthe catalytic cracking of the introduced carbon feedstock on the catalyst, whichsubsequently diffuses to the site of CNT growth. It has been proposed that therate-limiting step in CNT formation is the growth of CNTs themselves and not thenucleation step (Lamouroux et al. 2007). Design and/or operating parameters whichaffect the CVD should be well considered in order to synthesize the desired cleanCNTs. The key parameters affecting CVD synthesis of CNTs, including catalyst,temperature, carbon feedstock, carrier gas and reaction time, are discussed in fol-lowing sections.

Catalyst

Catalysts play a crucial role in the CVD synthesis of CNTs, and therefore,improving the desired characteristics of catalyst will enhance the obtained CNTquality as well as the process yield. Materials with capability of decomposinghydrocarbon and CNT formation are employed as catalyst in CVD processes.However, it has been found that the ability of catalyst for decomposing ofhydrocarbon molecules is not sufficient for CNT formation. The catalyst also needsto be in a form to nucleate the CNT (Yao et al. 2004; Kukovecz et al. 2000; Tranet al. 2007). Transition metals in the form of nanoparticles are considered as themost effective catalysts for CNT production. Transition metals can promote CNTgrowth due to their ability for (Teo et al. 2003; Sinnott et al. 1999; Perez-Caberoet al. 2004; Lee et al. 2003; Ortega-Cervantez et al. 2005; Zheng et al. 2006):

(a) Catalytic decomposition of volatile carbon compounds,(b) Meta-stable carbides formation,(c) Carbon diffusion through and over the metallic particles.

Nanoparticles of transition metal have high surface to volume ratios, and hence,the surface effects dominate over bulk properties. As a consequence, they possesssignificantly depressed melting points and much larger carbon solubility than thebulk, which has noteworthy implications for CNT growth. The better solubility ofcarbon in the catalyst particle influences nucleation and growth of CNT in a numberof ways, including (MacKenzie et al. 2010):

1. Increasing the carbon available for growth;2. Creating a larger concentration driving force to accelerate CNT diffusion;3. Influencing nucleation of CNT caps; and4. Determining the type of carbon product.

The CVD process for CNT growth generally utilizes heterogeneous catalysts,which are the transition metal nanoparticles, typically with a diameter of less than5 nm anchored on a high surface inert area, like silica, alumina. CVD synthesis ofCNTs using heterogeneous catalysts is essentially a two-step process consisting ofan initial catalyst preparation step followed by the real reaction for which the


presence of catalyst is vital. It is worth re-iterating that careful selection of thecatalyst and support improves the process yield significantly. Interactions, eitherchemical or physical, between support and metal nanoparticles are significant forthe catalytic properties of the nanoparticles. Size limitation and determination ofmetal particles by support porosity are physical interaction, whereas chemicalinteraction affects the electronic structure of the nanoparticles and thus their cata-lytic properties. Both physical and chemical interactions are contingent upon, notonly, both support and catalyst materials but also on their crystallographic orien-tations, surface roughness and porosity of the support. As a rule of thumb, weakinteractions yield tip-growth mode, whereas strong interactions lead to base growth,which are illustrated in Fig. 3.10 (Sinnott et al. 1999). Stronger metal–supportinteractions will improve dispersion, narrow size distribution, and reduce sinteringand agglomeration of active metal species. On the other hand, stronger metal–support interactions will hinder the reduction in the oxide precursors on the activecatalytic metal species.

The results of CNT growth with the same catalyst but on different types ofsupports suggest that substrates with larger surface areas, such as alumina andsilica, promote CNT nucleation and growth (Fan et al. 1999; Liao et al. 2003;Serquis et al. 2003; Su et al. 2000; Dai et al. 2001; Herrera et al. 2001; Kong et al.1998; Hernadi 2002; Reshetenko et al. 2004; Dupuis 2005; Ciambelli et al. 2007;Nagaraju et al. 2002a, b; Van der Wal et al. 2001). High surface area allows thecarbon source atoms to diffuse readily to the metal catalyst particles. However,

Fig. 3.10 Schematic diagrams of CNT growth mechanisms: Root growth and tip growth (Sinnottet al. 1999)


maximizing the surface area is not the only reason for using nanoparticles asheterogeneous catalysts, and they need to have desired thermal and mechanicalstrength. Basically, the role of the supports can be summarized as:

(a) To disperse the active phase,(b) To prevent sintering of catalyst, and(c) To improve mechanical strength.

The pore structure of the support, though important in heterogeneous catalysis ingeneral, does not play an important role in CNT formation by CVD. Fine pores arenot active if they do not contain active metal catalyst species. If an active catalystmetal species is present within a fine pore, it will quickly form enough carbon to fillup the pore and block the entrance of additional reactant hydrocarbons.

There is a consensus in the scientific literature that transition metals in the formof nanoparticles can produce CNTs, and the outer diameter of formed nanotubes isdirectly correlated with the catalyst particle size. Another factor that influences thesupported catalyst activity is the level of nanometals dispersion on the support.Large particles and aggregates as compared to fine and well-dispersed particles areinactive for nanotube formation. Optimization of the nanometal particles and theirdispersion in support lead to a maximum number of active points for hydrocarbondecomposition (Baker et al. 1989; Dai et al. 1996; Yudasaka et al. 1997; Kitiyananet al. 2000; Wei et al. 2001; Alvarez et al. 2001; Kukovitsky et al. 2002; Yamadaiet al. 2006; Aghababazadeh et al. 2006; Aslam et al. 2006). Accordingly, one of thescientific and technological challenges associated with heterogeneous catalyticsystems is to find the synthesis methods that fulfill the strong requirements in termsof their composition and structure in order to maximize their productivity of CNTs.There are numerous strategies for synthesizing supported metal catalyst for theproduction of CNTs; however, the most common methods are sol–gel, impregna-tion and co-precipitation. The effectiveness of the method mostly depends on thesurface properties of the support through interaction with the active metal phases(Yao et al. 2004; Fan et al. 1999; Liao et al. 2003; Su et al. 2000; Aslam et al.2006). The sol–gel synthesis method has been reported to ensure a highly homo-geneous distribution of transition metal in the matrix on which the aligned nano-tubes grow (Yudasaka et al. 1997; Kathyayini et al. 2006). However, every singleapplied condition affects the supported catalyst activity, and it should be controlledmeticulously to make the method reproducible. Dispersion of metal particles on thesupport also depends on the metal–support content ratio besides the preparationprocedure. There is an optimal metal content ratio between catalyst and support,which leads to the CNT synthesis with the desired properties at maximum yield.Increasing the metal percentage is fruitful only if it enhances the available activesites for CNT growth, instead of increasing the mean particle size of the metal (Yaoet al. 2004; Zheng et al. 2006; Liao et al. 2003; Herrera et al. 2001; Kouravelou andSotirchos 2005). At first, by increasing catalyst metal amount on support, the rateand yield of CNT formation is ramped up since more active catalytic sites are


available for carbon deposition. After the optimum point, the increased metalcontent makes the metal particles bigger resulting in lower specific area.

Iron (Fe), cobalt (Co) and nickel (Ni) have been enumerated as the most effectivecatalysts for CNT growth. However, the challenge is that which metal is moreactive and provides better quality of CNTs. Generally on the basis of the results inrelevant studies, iron produces lower-quality CNTs compared to Co and Ni(Kukovecz et al. 2000; Tran et al. 2007; Yamadai et al. 2006; Smajda et al. 2007;Hernadi 2002; Corrias et al. 2005; Arena et al. 2006; Smith et al. 2006; Luo et al.2002; Qian et al. 2004; Avdeeva et al. 1999; Klinke et al. 2001; Xia et al. 2007). Analloy of two transition metals with each other or with other non-transition metalshas been indicated to improve the CVD synthesis of CNTs in terms of CNT quality(less defective and well crystallized) and lowering the reaction temperature(Nagaraju et al. 2002a, b; Garcıa-Garcıa et al. 2008; Van Steen and Prinsloo 2002;Ermakov et al. 2001). These results may stem from the effect of alloying on carbonsupply, decomposition and diffusion as explained by Moisala et al. (2003) andHeight et al. (2005) as alloy catalysts typically have lower melting temperaturesthan single metals, which lead to higher carbon solubility in the catalyst. Alloycatalysts can also lower the binding energy of the metal with carbon atoms andhence improve the decomposition of carbon source and its diffusivity. Metal latticestructure is re-organized to improve diffusive transport of carbon atoms through thebulk. However, precise source of the synergistic benefit(s) and how the catalyst canbe optimized to exploit variation in solubility and diffusion to tune growth of CNTwith desired morphology remain unclear. It is worth mentioning here that, in mostof the application, the catalysts need to be separated from CNTs using an appro-priate purification process.

Although catalyst selection and preparation are the most significant factor inCNT synthesis, there are a number of mechanisms that can lead to catalyst deac-tivation, and hence loss of reactor performance. These mechanisms are poisoning,fouling and sintering of the catalyst due to elevated temperature or long reactiontime. Phase change in catalyst support also sometimes occurs due to the hightemperature, and this can lead to a surface area reduction, thereby reducing thereaction rate on account of impact on the access to active sites. Consequently, otherprocess parameter effects on CNT synthesis should be taken into account, which arediscussed in following sections.

Temperature

Each stage of CNT growth is a thermally activated process, and therefore, theremust be a characteristic threshold temperature for each step (Teo et al. 2003). Thestarting temperature of CVD synthesis of CNTs has been mostly reported to behigher than 500 °C, but there are some conflicting results for maximum temperature(Danafar et al. 2009; MacKenzie, 2010). Early on, it was believed that a very hightemperature (e.g., 1000–2000 °C) favors the formation of SWCNTs over MWCNTs(Ajayan 1999); however, with the advent of new CVD methods, especiallyFBCVD, this opinion is changing as the SWCNT growth temperature keeps getting


lower (Mizuno et al. 2005; Mora et al. 2007; Maruyama et al. 2002; Son et al. 2008;Kim et al. 2005).

Temperature has been mostly referred as the key variable in the CVD synthesisof CNTs, and the effect of this parameter on the morphology and yield of producthas been widely studied (MacKenzie et al. 2010). For example, Muataz et al. (2006)studied the role of reaction temperature on CVD synthesis of CNTs when benzeneand ferrocene were utilized as carbon precursor and catalyst, respectively. Theypointed out that MWCNT synthesis occurs at temperatures greater than 500 °C andmaximum wall numbers with less impurity are obtained at 850 °C (Fig. 3.11). Thehigher the reaction temperature, the more pronounced the formation of non-tubularcarbon-like nanofibers. They also revealed that there is a positive correlationbetween the average diameter and the length of CNTs and temperature. As a result,they proposed that temperature is a dominating factor for CNT diameter control.The same conclusions have been also reported by other researchers (Herrera et al.2001; Xiong et al. 2005; Nourbakhsh et al. 2007; Kim et al. 2005).

The main effects of increasing the reaction temperature are to increase themetallic particle size during the CVD growth of CNTs and, consequently, thenanotube diameter. However, there are some inconsistencies about the exact effectof the temperature on CNT growth mechanism. Kim et al. (2005) failed to observeany significant effects of temperature on average diameter of MWCNTs (Fig. 3.12),but observed that elevating the temperature had increased the length as well as thecrystallinity of the nanotubes. Son et al. (2008) concluded that the CNT diameterssynthesized from methane in fluidized bed decrease as the reaction temperatureincreases. Nevertheless, the effect of temperature was evident on the growth rate,purity and the crystallinity of CNTs in all mentioned studies.

The higher is the growth temperature, the more energy is available to form lessdefective, well-crystallized CNT. However, there is a great paradox between thequality of CNTs and other parameters of interest like yield and purity. Studiesachieving CNTs of high quality (less defective, well-crystallized CNTs) havegenerally reported a low yield (Ouyang et al. 2008; Liu et al. 2008). Temperatures

200 nm 200 nm 200 nm

(a) (b) (c)

Fig. 3.11 TEM images of CNTs at a 550 °C, b 650 °C, c 850 °C (Muataz et al. 2006)


higher than 1000 °C mostly result in non-tubular form of carbon (Kukovecz et al.2007; Kathyayini et al. 2006; Smajda et al. 2007; Muataz et al. 2006).

Temperature has the direct effect both on the diffusion coefficient and on thesolubility of carbon in the metallic nanoparticle. Consequently, high temperaturesimprove CNT growth by increasing the rate-limiting step of carbon diffusion;however, the temperature cannot be arbitrarily high due to several reasons that areunfavorable (Teo et al. 2003; Andrews et al. 2002; Kathyayini et al. 2006):

1. Deactivation of the catalyst.High temperatures (more than 1000) cause deformation of catalyst by sinteringof the supported catalysts and formation of alloy in bimetallic catalyst. Indeed, itis likely that the size and shape of the catalytic particles, which are in nanosize,are more stable at lower temperatures, leading to better control of the size andchirality of the nanotubes (Dupuis 2005; Zhao et al. 2005).

2. Formation of pyrolytic amorphous carbon.Thermal decomposition of carbon feedstock is accelerated. More carbon pre-sents than the amount can be dissolved and transported. This excess amount ofcarbon precipitates as unwanted amorphous carbon. Besides that pyrolysis,CNTs formed occur at temperatures around 1000 °C.

Carbon Feedstock

Besides catalyst and temperature, the carbon feedstocks also play an important rolein the growth, characteristics and properties of CNTs, because of their own bindingenergy, type and role of reactive groups and thermodynamic properties (Oncel andYurum 2006). A comparison of produced CNT characterizations showed that thereis a relationship between chemical structures of hydrocarbons and the CNT for-mation (Ajayan 1999; Teo et al. 2003; Lee et al. 2003; Hernadi 2002; Mizuno et al.2005; Maruyama et al. 2002; Quingwen et al. 2004; Qian et al. 2002; Melechkoet al. 2005). Hernadi and co-researchers (2002) showed that unsaturated hydro-carbons, like ethylene or acetylene, have much higher yield and deposition rate than

(a) (b) (c)

Fig. 3.12 TEM images of carbon nanotubes at a 600 °C, b 700 °C, c 800 °C (Kim et al. 2005)


saturated gases, like methane. In addition, saturated carbon gases tend to producehighly graphitized filaments with fewer walls compared to unsaturated gases.Consequently, they suggested that saturated hydrocarbons are favored for SWCNTgrowth and unsaturated hydrocarbons for MWCNTs. However, SWCNTs havebeen obtained from a highly diluted unsaturated hydrocarbon (Melechko et al.2005; Li et al. 2004a; Kathyayini et al. 2006; Alvarez et al. 2001; Yamadai et al.2006; Su et al. 2000; Zheng et al. 2006; Andrews et al. 2002). Li et al. (2004b)demonstrated that the chemical structure of hydrocarbons, i.e., straight-chained ringor benzene-like structures, is significantly more influential than the thermodynamicproperties (e.g., enthalpy) of the carbon source on the type of CNT formation.Besides configuration, functional groups of hydrocarbons have a decisive influenceon the quality of the produced material. The growth of clean SWCNTs wasobserved at relatively low temperatures using alcohols with various catalysts(Zheng et al. 2006; Maruyama et al. 2002; Yu et al. 2006; Bachmatiuk et al. 2008).The authors concluded that alcohols are much better carbon sources for SWNTsthan hydrocarbons and this is likely due to the ability attributed to OH− radicals toetch away amorphous carbon deposits. However, Hernadi (2002) and Qian et al.(2002) suggested that CNT morphology is independent of the carbon feedstockwhen a specific catalyst is used.

A kinetically stable carbon feedstock that undergoes the least pyrolyticdecomposition at process temperature is desirable for CNT synthesis. The mostcommonly used carbon sources are ethylene (Qian et al. 2004; Corrias et al. 2005;Ciambelli et al. 2007; Tran 2007; Melechko 2005), acetylene (Perez-Cabero et al.2004; Nagaraju et al. 2002a, b; Kathyayini et al. 2006; Garcıa-Garcıa et al. 2008;Nourbakhsh et al. 2007), methane (Ermakova et al. 2001; Avdeeva et al. 1999;Mora et al. 2007; Qian et al. 2004; Aslam et al. 2006, Aghababazadeh et al. 2006;Reshetenko et al. 2004), ethanol (Kouravelou and Sotirchos 2005; Zheng et al.2006; Maruyama et al. 2002; Yu et al. 2006; Ortega-Cervantez et al. 2005) and CO(Liao et al. 2003; Herrera et al. 2001; Kitiyanan et al. 2000; Serquis et al. 2003;Chiang et al. 2001). Ethanol has been recently reported as a suitable carbon sourcefor the CVD techniques due to several advantages, namely lower reaction tem-peratures compared to other hydrocarbons (Zheng et al. 2006, Fakhru’l‐Razi et al.2009) and its etching effect of amorphous carbon on the catalyst surface (Yu et al.2006; Liu and Fang 2006; Zheng et al. 2006; Kouravelou and Sotirchos 2005;Maruyama et al. 2002; Ortega-Cervantez et al. 2005).

Recently, it was reported that CNTs could be synthesized from solid-statepolymers (Yen et al. 2008; Arena et al. 2006). This is a good alternative to purecarbon source like hydrocarbons, since they are very expensive and have limitedsupply. However, the most organic gases that come from the solid-state polymersafter it produces CNTs are toxic, dangerous and difficult to store or transport. Coal,liquefied petroleum gas and natural gas have been also used successfully for CNTsynthesis (Qian et al. 2002; Qiu et al. 2004, 2006). However, there are somedrawbacks due to the presence of impurities in natural sources as they affect thereaction and process and also damage the equipment used for the production ofCNTs. It is worth mentioning that more work is needed to improve the yield and


quality of the CNTs from natural sources. To achieve commercial CNT synthesis,attempts are conducting toward finding low-cost carbon source with high conver-sion ability to less-defect CNTs. In this respect, using sustainable source, likebioethanol, will be imperative.

The CVD synthesis of CNT is a supply-limited process that means feeding rateof carbon feedstock influences the reaction. There is an optimal feeding rate for thecarbon feedstock; at flow rates lower than this value, there is not enough reactant tobe decomposed, and thus, the concentration of carbon source controls the rate ofdecomposition and consequently the rate of CNT formation. Increasing the feedingrate will increase the decomposition rate; however, after a critical point, increasingthe feeding rate of carbon does not significantly affect the decomposition rate orCNT growth, since it is controlled by the availability of active sites or better to notecatalyst particles (Kukovitsky et al. 2002; Herrera et al. 2001; Liao et al. 2003). Thegrowth of single- or multiwall carbon nanotubes, in fact, could be achieved bycontrolling the feeding rate of carbon source as well (Bachmatiuk et al. 2008).

Carrier Gas

In CNT synthesis process, a specific amount of carbon precursor is continuously fedinto the reactor in gaseous form. To reduce the formation of amorphous carbon anddecrease the contact time between carbon feedstock and catalysts, the carbonfeedstock is diluted by a carrier gas (Piedigrosso et al. 2000; Teo et al. 2003). Thechamber is also kept free of oxygen by using the carrier gas during the process toavoid carbon oxidation. It is worthy to stress that the carrier gas should be anon-reactive gas. Argon and nitrogen are the most common carrier gases becausethey easily form inert atmosphere; however, other kinds of gases like helium(Yamadai et al. 2006; Kitiyanan et al. 2000) or NH3 (Piedigrosso et al. 2000) havealso been used. It is generally considered that the carrier gases could affect thegrowth of CNTs and hence the structure and properties of the resulting assemblies(Venegoni et al. 2002; Kukovitsky et al. 2002; Zhao et al. 2005; Xiong et al. 2005;Malgas et al. 2008). Li et al. (2004a) investigated the influences of type of gas usedon the CVD process of cyclohexane. They reported that in case of using Argon,MWNTs were produced, whereas when hydrogen was used, some SWNTs wereformed. In a relevant study, Miet al. (2005) compared the effects of NH3 andnitrogen as carrier gases on the structure and morphology of CNTs. In the case ofusing ferrocene as catalyst and acetylene as a carbon source, the growth ofbamboo-like structures was observed in NH3 with larger diameters compared toCNTs obtained in nitrogen. The above findings are in fair agreement withKukovitsky et al. (2002) results. Authors noted that the nature of gaseous envi-ronment has a profound influence on the mobility and sintering process of thecatalyst particles, and therefore, the strength of the metal–support interaction isaltered. Besides the type of carrier gas, its flow rate also was indicated as a keyparameter in the CVD synthesis of the CNTs The gas flow rates influences the mass


transfer and hence the reaction steps as well as catalyst stability and activity (Teoet al. 2003). Accordingly, carrier gas has impact on the morphological and struc-tural characteristics of the grown CNTs and their uniformity.

Considering the CVD synthesis of CNTs, the sequence of events taking place is:reactant gases enter the reactor by forced flow, gases diffuse through the boundarylayer by mass transfer, gases come into contact with the surfaces of substrate,deposition reaction takes place on the surfaces of substrate, and gaseousby-products of the reaction diffuse a way from the surface through the boundarylayer by mass transfer. Various transport phenomena, either diffusive or convective,and reaction processes underlying CVD involve many intermolecular collisionsbetween existing molecules in the reactor. These collisions transfer momentum andenergy between the collision partners or lead to net transport of mass from one partof the system to another. Transport properties of multicomponent gases, includingviscosity and thermal conductivity as well as heat and mass diffusion coefficients,are directly influenced by temperature, pressure and mixture composition (Vahlaset al. 2006). To make it more clear, CNT growth process occurs in a diffusiveregime where hot gases surrounding the catalyst force a viscous media and slowdown the CNT growth (Vinciguerra1 et al. 2003). Thus, it can be inferred thatgaseous environment nature has intense effects on all steps of the CVD synthesis ofCNTs.

Hydrogen is sometimes referred to as a reactive gas rather than carrier (Xionget al. 2005; Teo et al. 2003). These researchers believed that hydrogen can provide areducing environment for the catalytic metals and hence prevents poisoning of thecatalytic surface by carbon deposition which lessens the formation of undesirablecarbon deposits, like amorphous carbon. However, their explanation became con-troversial because CNTs of good quality (less defective and well crystallized) havebeen obtained by using just argon as a carrier inert gas, where no hydrogen wasused (Herrera et al. 2001; Li et al. 2004a, b; Son et al. 2008) nor nitrogen (Hao et al.2003; Kukovecz et al. 2007; Yen et al. 2008). Even Son et al. (2008) achievedCNTs of small diameter and higher crystallinity in cases where no hydrogen wasutilized. Accordingly, more studies were used to investigate whether hydrogenplays a different role rather than being a carrier. For instance, Qian et al. (2004)adopted two modes of reaction to examine the effects of unreduced catalyst andreduced catalyst on CNT formation. In the first mode, the catalytic particles weredirectly used to decompose methane at 823–1123 K, using unreduced catalyst.In the second mode, the catalytic particles were first reduced by hydrogen(at 973 K, for 3 h) and then used to decompose methane in the same temperaturerange. Results of this study illustrated that there is no any obvious difference in themorphology of CNTs using the unreduced catalyst or reduced catalyst. They alsoreported that when the unreduced catalysts were used, the conversion of methane asa carbon feedstock to produce CNTs was significantly increased. Further studiesconducted to examine the effects of hydrogen concentration on CNT morphology(Liu et al. 2009; Xiong et al. 2005) have also shown that increase in hydrogenconcentration leads to larger CNTs in terms of wall number and diameter. It wasproved that high concentration of hydrogen accelerates the sintering of the catalyst


particles and, consequently, enlarges their diameters. Conclusive evidence on thedependence of catalyst size on the formation of CNTs was explained in the previoussection about the role of catalyst in CVD synthesis of CNTs. Consequently, the ideaof positive role of hydrogen in CNT formation was rejected. However, as theprofound influences of the gaseous environment on the CVD synthesis of CNTshave been established, a more meticulous and detailed investigation is stronglyrecommended.

Reaction Time and Space Velocity

The best operation period determination with respect to product quality and processyield is vital from the economic point of view for any process. In the CVD process,there is also an optimum reaction time when beyond it, a progressive deactivationof the catalytic particles starts to occur due to covering of the active sites by CNTsformed on them. However, there is a great disparity in the proposed optimum timefor CVD reaction with respect to process parameters and nature of catalyst as wellas carbon source (Andrews et al. 2002; Zheng et al. 2006; Garcıa-Garcıa et al. 2008;Son et al. 2008). To nucleate CNTs, accumulated carbon atoms must reach a criticalconcentration on the surface and inside the catalyst particles, which is related to thedeposition rate. Kouravelou and Sotirchos (2005) in a systematic study proved thatthe rate of deposition of CNTs depends on carbon precursor nature and the catalystcomposition as well as temperature. Meanwhile, it has been reported that CNTspossess different morphologies for different reaction times. Lamouroux andco-researchers (2007) believed that short reaction time is adapted to SWCNTgrowth. Kim et al. (2005) reported that CNT diameter is determined by duration ofreaction time, generally changed between 20 min and 1 h. In accordance with theirstatement, Brukh and Mitra (2006) showed that the mean diameter of MWCNTsincreases with the CVD durations, possibly due to sintering and agglomeration ofparticles, which is not desired if the formation of SWCNTs is the objective.Furthermore, CNT length is controlled by the length of the deposition time (Teoet al. 2003; Kim and Grate 2003; Kathyayini et al. 2006). Physical characteristics ofCNTs sampled at different periods may vary due to differences in their morphology.In view of that, better understanding of the reaction kinetics is needed for a con-trollable CNT synthesis.

Throughout the Brukh and Mitra work (2006), it has been shown that residencetime of the flow, the time spent by the fluid in a reactor, also plays an important rolein CVD of CNTs. According to their results obtained in a fixed bed reactor, at verylow residence times only CNTs with practically no non-tubular form of carbon wereformed. As the residence time increased, the concentrations of active radicals andintermediates in the gas phase increased; therefore, they began to recombine andformed larger molecules. Under these conditions, active radicals consumed theavailable C that would have otherwise formed CNTs and other forms of carbon.The available specific surface area for CVD reaction and hence its severity aremainly influenced by synthesis duration as well as space velocity. The spacevelocity, which is the inverse of the residence time, is the volumetric flow raterelative to the catalyst mass. This parameter is important for any comparative


measurements in catalytic process such as catalyst screening, determination ofprocess parameters, optimization of catalyst production conditions and deactivationstudies (Blanchard et al. 2008; Louis et al. 2005; Hagen 2006).

Basically, space velocity must be determined in a pilot plant (Walas 1990). Thespace velocity in fluidized-bed reactors is relatively high compared to fixed bedreactors and hence provides selective synthesis of CNTs especially. High spacevelocity avoids external diffusional effects in the catalyst particle, and it can beobtained by using high gas flow rate or not very dense bed of catalyst. A constraintmay arise from high flow rates that it cools down the bed environment and thereforeaffects the deposition quality (Vahlas et al. 2006). Moreover, high gas velocitycreates some difficulties in process scale-up as it influences the stable fluidization ofcatalyst and CNTs of low bulk density in the reactor (Mora et al. 2007; Liu et al.2009). Liu et al. (2009) demonstrated a practical method to increase the spacevelocity for selective synthesis of DWCNTs. A very high space velocity is achievedby feeding fresh catalyst in small amounts to the fluidized-bed reactor.Nevertheless, this prominent parameter has been almost systematically occulted inthe majority of studies regarding CVD synthesis of CNTs in both fixed and flui-dized beds.

3.1.3.1 CVD Synthesis of CNT Using Fixed Bed

There are mainly two processing system configurations for CVD synthesis ofCNTs, i.e., horizontal and vertical. The illustration of a typical horizontal system,which is used in the fixed bed and floating catalyst technique, is depicted inFig. 3.13. In the fixed bed process, the solid-phase catalyst is placed in boats insidethe reactor and the gas-phase reactant is introduced when operational temperature isattained. The efficiency of CNT growth in this process is limited severely byinhomogeneous gas–solid contact and temperature gradients. In a horizontal fixedbed reactor, the diffusion of the carbon source to the catalyst particles becomes ratelimiting because as more and more nanotubes are grown, it will cover the surface,thus reducing the effectiveness of the catalyst particles on the surface.

Floating catalyst technique utilizes a mixture of catalyst and reactants introducedin the gas phase to the reactor maintained at an elevated temperature where theCVD reaction takes place (Ajayan 1999). The gas-phase catalyst undergoestransformation in the reactor and forms nanosized solid-phase active catalyst par-ticles in situ. One of the drawbacks of this method is the difficulty in preventingparticle coalescence. Only when these solid nanocatalysts adhere to the reactorsurfaces, will they have sufficient residence time to grow CNTs (Ajayan 1999;Andrews et al. 2002; Oncel and Yurum 2006). Any unreacted catalyst and solidparticles which have not been able to adhere to the reactor walls at sufficiently hightemperatures are swept away with the unused reactants and carrier gases or reactionproduct gas (hydrogen), thereby dramatically reducing the process efficiency.


3.1.3.2 CVD Synthesis of CNT Using Fluidized Bed

Fluidized-bed reactors are widely applied for several industrial purposes, such asdifferent types of chemical reactors, fluid catalytic cracking, fluidized-bed com-bustion, fluidized-bed biofilter, steam reforming of hydrocarbon or applying acoating on solid items (Vahlas et al. 2006). Fluidized bed reactors offer excellentgas–solid contacting, and particle mixing provides good gas-to-particle andbed-to-wall heat transfer. Fluidized-bed reactors were found to be equally effectivefor mass production of CNTs (Vahlas et al. 2006; Danafar et al. 2009; MacKenzieet al. 2010). Fluidized-bed reactors utilized for CVD synthesis of CNTs are typi-cally a vertical reactor enclosed by an electrical furnace. An upward flow of carbonsource and carrier gas mixture fluidizes the supported catalytic particles in wellcontact by the reactant. Figures 3.14 and 3.15 demonstrate two different types ofschematic representation of fluidized-bed configuration employed for CVD syn-thesis of CNTs. Fluidized-bed CVD process do not only draws upon the advantagesof both the floating and fixed bed catalyst systems, described in the previoussection, but also has its own advantages. It has excellent heat- and mass-transfercharacteristics compared to fixed bed method because of the continuous solidmovement. In addition, the catalyst nanoparticles anchored to the support surfaceare not swept away by gas stream. Fluidized beds provide a larger contact areaamong the reactants and catalyst powder that improve the mass- and heat-transfercharacteristics and thus chemical reactions (Vahlas et al. 2006). Consequently, asthe CNT formation rate is directly associated with the availability of the active

Fig. 3.13 a Fixed bed technique: (1) furnace, (2) thermocouple, (3) carrier gas, (4) carbon source,(5) flow meter, (6) outlet, (7) boat, (8) catalyst. b Floating technique: (1) furnace, (2) thermocouple,(3) carrier gas, (4) catalyst dissolved in volatile carbon source, (5) flow meter, (6) outlet, (7) emptyboat for CNT deposition


Fig. 3.14 A fluidized-bedreactor for CNT growth. Thereactor is a 2.8-cm-diameterquartz tube with a sealedporous quartz distributor.Adopted from Venegoni et al.(2002)

Fig. 3.15 A fluidized-bedreactor for CNT growth. Thereactor is made of quartz glasswith an ID of 250 mm and aheight of 1 m. There is asintered porous plate used asthe gas distributor at thebottom of the reactor (Wanget al. 2002)


catalyst sites, FBCVD compared to the other types of CVD is more efficient for thesynthesis of large quantities of CNTs.

In the mass production of CNTs, fluidized-bed chemical vapor deposition(FBCVD) techniques offer numerous advantages over fixed beds and floating cat-alyst thanks to the technology robustness, flexibility and high productivity.Moreover, the FBCVD is easily scaled up and can be operated continuously whichis important for cost-effective large-scale production of CNTs. Fluidization pro-vides higher space velocity, which leads to efficient gas–solid contact, and hencehigh mass and heat transfer. In accordance with that high process yield, producthomogeneity, purity and selectivity are attained. Moreover, FBCVD is a flexibleprocess in terms of operating conditions such as parameters like gas mixture andtemperature can be finely tuned according to the definition of the desired product(Philippe et al. 2007). Furthermore, space is available for growing CNTs and theirresidence times can be controlled accurately and the activity of the catalyst isutilized sufficiently which favors the selective mass production of CNTs withuniform properties. Despite all mentioned advantageous, improper design and/oroperation of fluidized-bed reactors can lead to conversions that fall well below thetheoretical lower limit of perfectly mixed flow.

Xu and Zhu (2004) demonstrated another facility provided by fluidized-bedreactor which emphasis the superiority of these kinds of reactors for mass productionof CNTs. They developed a new technique of fluidized-bed metal–organic chemicalvapor deposition (MOCVD) as a one-step preparation of highly dispersedmetal-supported catalysts followed by FBCVD to synthesis CNTs. This method hassome advantageous over the conventional methods (such as impregnation, ionexchange, co-precipitation and co-crystallization) since it eliminates the drying andthe subsequent calcinations/reduction operations and hence minimizes the aggre-gation or growth of crystalline size of the supported metal particles caused by theseoperations. Prepared supported catalyst particle activities were tested for CNTsynthesis through FBCVD using acetylene. Results proved that for all themetal-supported catalysts, the deposited metals were highly dispersed on the surfaceof the support particles. In a relevant attempt regarding the development of FBCVDtoward the mass production of CNTs, See and Harris (2008) explored the technicalviability of using CaCO3 as a soluble support material for the synthesis of CNTs viaFBCVD. All models and experimental results presented showed the versatility ofFBCVD which fulfills the priority of this method for large-scale production.

In FBCVD systems, the catalytic particles act as turbulence promoters to reducethe hydrodynamic boundary layer and enhance mass transport. Basically, thedeposition of reactants on the catalyst surface involves a balance between diffusionof the reactant molecules to the catalyst and the kinetics of deposition reaction. Thedeposition reaction taking place on the catalyst surface itself is composed of varioussteps, such as adsorption of the reactants on the active sites, chemical reaction at theactive sites and desorption of the products from the active sites. It is necessary tonote that in the FBCVD process the heterogeneous catalytic particles, which are thecatalytically active metal particles anchored on a high surface inert area, are uti-lized. With regard to the reaction within the heterogeneous catalytic particles, the


reactant must first diffuse into the particles, leading to a lowering of its concen-tration in the inner regions of the particle. Obviously, as the reactant diffusesinward, it react to form the product, but at a progressively diminishing rate. Theactual rate of the reaction is a function of the diffusion rate as well as the pelletshape and size. Correspondingly, several factors influence the actual rate of thereaction, namely pore shape and constriction, particle-size distribution, micropore–macropore structure and surface area of the catalytic particles as well as bed volumechange upon reaction (Cassell et al. 1999). Catalytic particles in FBCVD process,more specifically, determine both the CVD reaction characteristics and the fluidi-zation quality. Fluidization quality is closely related to the intrinsic properties of thebed, e.g., density, particle size and size distribution, and also their surface char-acteristics (Gupta and Sathiyamoorthy 1998). A systematic study of the effects ofcatalytic particle size on the formation of CNTs using FBCVD process has beenreported by Danafar et al. (2011). From a heat transfer point of view, however, theparticles perform an extra function by carrying the heat which is very importantduring CNT formation. It is well known that the conversion efficiency of the CVDreaction depends on the extent of heat and mass transfer occurring within andbetween the gas–solid phases. Exothermic CVD reaction results in hot spots thatfrequently lead to catalyst deactivation, and hence adverse changes in conversionand/or selectivity of the process. The high heat-transfer efficiency in fluidized-bedreactors leads to rapid dissipation of local hot spots, resulting in a uniform bedtemperature. Moreover, significant heat transfer results in lower required reactiontemperature for CNT formation which is economically favored. However, theperformance of a gas–solid fluidized system is strongly determined by many basicfeatures, including

1. Design of fluidized-bed reactor and its dimensionalityThe most important design parameters include reactor diameter, reactor heightand gas distributor. Small diameter beds, i.e., less than 5 cm in diameter, will notprovide true fluidization due to the significant wall effects (MacKenzie et al.2010).Gas distributor is the key fluidized-bed components, and its design is likely toresult in a homogeneously distributed gas flow.

2. Operating conditionsOperating conditions involve all parameters that have effects on the CVDreaction and fluidization quality such as temperature, carbon source, flow rates,characteristics of catalytic particles. For example, it is critical to use sufficientamount of catalyst material not only because of the reaction but also because ofproviding suitable fluidization. Insufficient bed material does not permit fluidi-zation to occur; in other words, no bubbles can form. Employing catalystmaterial less than 5 g will not provide true fluidized beds because wall effectsare dominant (MacKenzie et al. 2010). If the ratio of bed diameter (d) to theparticle diameter (dp) is low, d/dp ≪ 100 (Saxena et al. 1989), and hence, thecharacteristic pressure drop versus fluid velocity profile of a fluidized bed is notfollowed.


Accordingly, despite all advantageous, improper design and/or operation offluidized-bed reactors can lead to conversions that fall well below the theoreticallower limit of perfectly mixed flow. The entire optimization of the FBCVD processis clearly a complicated task, with many unanswered questions. In practice, to getthe general awareness of the process, it is necessary to proceed carefully byfocusing on one parameter at a time. Accordingly, the discussion will now benarrowed down to the study of the effects of the bed particle size on FBCVDperformance.

As far as FBCVD synthesis of CNTs is concerned, all parameters, whichinfluence the reaction and fluidization quality, should be considered. Unfortunately,the scattering of data and the diversity of the experimental equipment and proce-dures made a reliable comparison of results difficult (Danafar et al. 2009;MacKenzie et al. 2010). Moreover, there is a lack of process parameters and largegaps exist between reported information, especially factors associated with fluidi-zation criteria. Despite conducting a number of studies on the role of processparameters and their interactions in FBCVD, this technique is in its infancy andfurther in-depth understanding of the impact of the operating parameters on productquality and quantity is strongly required for commercial production of CNT.

The chemical conversion in a fluidized reactor depends on the basic features thattogether determine the fluidization quality. The term “fluidization quality” isapplied to describe the various fluid-dynamic conditions brought about by thefluidization process itself. Quantities, which determine the fluidization mode andcharacteristics, include intrinsic properties of particles, reactor geometry, superficialfluidizing velocity (U) and minimum fluidization velocity (Umf). Fluidizationquality is closely related to the intrinsic properties of particles, e.g., particle density,particles size and shape as well as their size distribution and surface characteristics.Geldart (1973) classified powders into four groups according to their fluidizationbehaviors at ambient conditions. Type “A” are particles with sizes of about 20–100 µm, form a slightly cohesive structure. Type “B” particles are dense materialslike glass and sand with diameters of around 150 µm. Type “C” particles are evensmaller and lighter than type “A,” usually less than 20 µm in diameter. Type “D”particles are large, on the order of one or more millimeters. Nanosized particles,corresponding to Geldart classification, fall in group “C,” but their fluidizationbehavior differs from conventional “C” particles because of extremely smalldimensions and bulk density (Geldart 1973; Jung and Gidaspow 2002). The Geldart“C” particles are difficult to fluidize due to the large cohesive forces which lead tocrack formations and channeling in the bed. Nanoparticles despite the anticipatedlarge surface charge can be easily fluidized due to the formation of light agglom-erates (Valverde 2008). This type of fluidization has been termed as agglomerateparticulate fluidization.

The key element on differentiators between fixed and fluidized beds and flui-dization quality is the ratio of U/Umf. This value is essential for the reactor scale-upbecause it determines reactant feed rates and bed hydrodynamics (MacKenzie et al.2010). Figure 3.17 illustrates the range of flow regimes possible in a vertical flui-dized bed. At low values of U, the gas flows through the interstitial spaces between


the particles. The pressure drop across the bed increases linearly with U but is notlarge enough to balance the bed weight. This fixed bed is illustrated in Fig. 3.17.Raising the superficial gas velocity until the pressure drop equals the bed weightyields minimum fluidization, exhibited in Fig. 3.16 as particulate fluidization.Minimum fluidization velocity is the superficial fluid velocity at which the upwarddrag force exerted by the fluid is equal to the apparent weight of the particles in thebed; the pressure drop of the gas across the bed becomes constant with increasinggas velocity. Minimum fluidization velocity depends on particle size and particledensity, Geldart group, and fluid properties as well. The value of Umf is usuallydetermined experimentally by measuring the pressure drop as a function of gasvelocity. It also may be predicted by some empirical or semiempirical correlations(Davidson and Harrison 1963; Wen and Yu 1966; Kunii and Levenspiel 1991).

Fig. 3.16 Basic vertical fluidized-bed schematic depicting possible flow regimes. Adopted fromGrace (1986)

Fig. 3.17 Three-dimensional structures of CNTs produced in FBCVD (Corrias et al. 2003;Philippe et al. 2007)


However, the result from these kinds of equation is not accurate due to differentbehaviors of particles in practice.

Further increasing U causes the bed to expand and yields various flow regimes ofincreasing intensity, noted as aggregative fluidization. Initially, bubbling fluidiza-tion occurs either at the onset of fluidization or shortly thereafter, depending onparticle and gas properties. The velocity at which bubbles are first present is Umb,the minimum bubbling velocity. For sufficiently narrow beds at higher gas veloc-ities, slugs may begin to form. This occurs when the diameters of the bubblesapproach that of the confining tube, as illustrated in Fig. 3.17. Large enough slugsmay lift portions of the bed, and the next regime described by turbulent motion ofthe bed is obtained. Here, the clearly defined upper surface of the bed visible in theprevious regimes disappears. With a high enough superficial gas velocity, the dragforce from the fluid will match the weight of each particle, known as pneumatictransport.

There is a prominent issue for FBCVD of CNTs related to its fluidization qualitydetermination, because the bed characteristics change on account of reaction andelevated temperature. Catalyst particles initially provide fluidization, and theirfluidization behavior can be easily predicted at ambient condition according toGeldart classification. However, their fluidization characteristics at process tem-perature, which is above 500 °C, appear different from that which it occupies atambient conditions. This is due to the effect of gas properties on the grouping andmay have serious implications as far as the operation of the fluidized bed is con-cerned. Moreover, Umf differs significantly from the size distribution in the freshfeed due to elutriation of fines, attrition, agglomeration of particles, bed solidcharacteristics change due to reaction, density and viscosity of fluids alteration dueto reaction and temperature.

Venegonia et al. (2002) in their parametric study for CNT growth by FBCVDmeasured the Umf of SiO2 particles, belonging to B group of Geldart classification,both at 25 and at 550 °C, and found that Umf = 517 cm min−1 decreases to57 cm min−1, respectively. In accord with this observation, Morancais et al. (2007)pointed out during the synthesis of CNTs, Umf is roughly divided by a factor 6 after120 min of running the process, since the grain density sharply decreases; as aconsequence, the fluidization ratio increases. They also noted that carbon yieldtends to increase when the U/Umf ratio decreases for a similar amount of carbonfeedstock introduced, because of the increase in residence time of the gaseousprecursors into the bed. They proposed at U/Umf = 0.87 for a 5.3-cm-diameterreactor, optimum conditions of run in terms of fluidization quality and MWCNTformation are obtained. Note that the ratio 0.87 means the applied velocity waslower than minimum fluidization and in this case there was not any fluidization. Sonand co-workers (2007) reported that increasing the gas flow rates from 2 to 4 Umf

enhances the carbon yield and results CNTs with smaller diameter.In FBCVD synthesis of CNTs, the initial amount of catalyst corresponds to the

static bed height, Hs, and plays a critical role in characteristics of fluidized beds.Therefore, the challenge associated with the catalyst mass placed in the reactor is itseffects not only on CNT formation but also on fluidization condition. First, the


heterogeneous catalyst should carefully be prepared with respect to its activity forCNT formation. Then, appropriate catalyst amount should be chosen to obtainrequired fluidization properties. Son et al. (2007) comment that if static bed height(Hs) is smaller than the inside diameter of the reactor, (i.e., a shallow bed), flui-dization quality is poor. Adding 114.3 μm alumina powders as inert particles, intheir study, improved CNT production, and the mean CNT diameter and agglom-erate size of synthesized CNTs decreased as a result of attrition of the inert particles.In accordance with their comments, See and Harris (2008) also applied 40 g pureCaCO3 in addition to the 70 g Fe–Co catalyst supported on CaCo3.

Son and co-researchers (2008b) studied the effects of catalytic particle loading(2.5, 5, 10 and 20 g) on the CNT formation using the FBCVD process in a reactorwith an internal diameter of 0.05 m and height of 1.0 m. Based on this study, whena 5 g of catalytic particles was used, well-structured CNTs with high surface area(338.8 m2/g) were formed. However, by increasing the amount of catalytic parti-cles, no significant improvement on the amount of deposited CNTs was observed.Their observations are in agreement with other researchers like Philippe et al.(2007); See and Harris (2008); Jeong et al. (2009). This can be attributed to therapid bed expansion during CNT formation and a significant decrease in beddensity. Moreover, a large amount of catalytic particles demand a very high flowrate of gas to ensure a homogeneous condition in the reactor and for transportationinto the reactor. High gas velocities will lead to complexities in the processscaling-up (Mora et al. 2007; Liu et al. 2009). Accordingly, such peculiaritiesrequire appropriate reactor design and tuning the operating conditions.

Considering the CNT structure, it possesses both nanometric (related to itsradius) and microscopic (related to its typical length) dimensions. The apparentfeature of CNTs as an end product of FBCVD is a black spongy powder with lowbulk density. In FBCVD process, like other types of CVD, CNTs grow from thecarbon atoms by the action of metal nanoparticles. The most accepted model forCNT growth is similar to the vapor–liquid–solid mechanism (Vahlas et al. 2006), aspresented earlier. In this model, the catalytic decomposition of a carbon feedstockinto carbon atoms and hydrogen is initiated on the surface of active transitionmetals and followed by diffusion of carbon into the metal nanoparticles until thesolution of metal–carbon becomes saturated. When supersaturation occurs, theprecipitation of graphite carbon from the metal surface starts and hence forms acylinder, e.g., CNT. Accordingly, observing the microscopic scale in the FBCVprocesses, the carbon source self-assembles into a one-dimensional tubular CNTstructure with the help of a catalyst. However, a single-carbon nanotube cannot befluidized according to the conventional fluidization knowledge (Geldart 1973),because it is a linear nanometer material. Researchers who elaborated to produceCNTs using FBCVD process demonstrated that the CNTs are in the form ofagglomerates (Wang et al. 2002; Hao et al. 2003; Corrias et al. 2003; Philippe et al.2007). In other word, the final product of the FBCVD is a three-dimensionalnetwork structure from large amounts of CNTs which are rather linear structure ofindividual CNT (Fig. 3.17).


Hao et al. (2003) studied the evolution of CNT agglomerates during FBCVDprocess by observing the macroscopic properties of agglomerates includingagglomerate sizes and bulk density. A model, which is illustrated in Fig. 3.18, wasproposed for agglomerate formation of CNTs in FBCVD process. According to themodel proposed by Hao et al. (2003), CNT growth crushes the catalytic particles,disrupts their initial structure and forms separated catalytic sites. With the increasein carbon deposition, the catalysts are crushed as much as they can be and the CNTsgrow around the catalyst sites. The growing CNTs will push away and separate thesites from each other, leading to increase in the agglomerate size and decrease in thedensity until fragmentation by ablation dominates the growing process.

CNTs produced by FBCVD are in the form of tangled and loose agglomerates.During the process, larger agglomerates may break into smaller ones and theaverage diameter of agglomerate remains nearly constant. Agglomerate morphol-ogy of CNTs provides good fluidization during the growth process and causeshigh-quality CNT production in large scale at low cost. In case of low fluidizationquality condition, large agglomerates are formed due to less mixing of powders thatresult in decrease in carbon conversion and CNT growth (Morancais et al. 2007).

As mentioned earlier, FBCVD does not allow the production of aligned CNTsthat might be considered as an obstacle for this process (Vahlas et al. 2006).However, the comparisons of CNTs grown on a same catalyst in a fixed andfluidized beds showed that CNTs produced by FBCVD has a higher specific surfacearea (Liu et al. 2009) and better thermal stability (Yu et al. 2006). Moreover, due tothe effective mass and heat transfer, longer CNTs with smaller diameter and fewerlattice defects are formed in an FBCVD compared to fixed bed.

(b)(a)

(c)(d)

Fig. 3.18 Schematic mechanism of a model of CNT agglomerate formation: a original catalyticparticle; b the catalytic particle structure is crushed by CNT growth; c catalytic sites are adequatelyseparated and subagglomerates form; d fully developed agglomerates (Hao et al. 2003)


3.1.3.3 Modeling of Fluidized-Bed CVD Synthesis of CNTs

Recall that fluidization occurs when small solid particles are suspended in anupward flowing stream of fluid, as shown in Fig. 3.19. The fluid velocity is suffi-cient to suspend the particles, but it is not large enough to carry them out of thevessel. The best model of the fluidized-bed reactor developed thus far is thebubbling-bed model of Kunii and Levenspiel (1991). In this model, fluidized bed istreated as a two-phase system—an emulsion phase (often called the dense) and abubble phase.

Let us consider a vertical bed of solid particles supported by a porous or per-forated distributor plate, as in Fig. 3.19. The direction of gas flow is upward throughthis bed. There is a drag exerted on the solid particles by the flowing gas, and at lowgas velocities, the pressure drop resulting from this drag will follow the Ergunequation (Ergun 1952), just as for any other type of packed bed. When the gasvelocity is increased to a certain value, however, the total drag on the particles willequal the weight of the bed, and the particles will begin to lift and barely fluidize.The range of velocities over which the Ergun equation can be applied is fairly large.On the other hand, the difference between the velocity at which the bed starts toexpand and the velocity at which the bubbles start to appear can be very small andeven sometimes this difference does not exist. Accordingly, if the gas flow rate issteadily increased, the first evidence of the bed expansion may be due to theappearance of gas bubbles in the bed and the movement of solids. When the gasvelocity is low, the rising bubbles may contain a small amount of solid particles.The remainder of the bed has a much higher concentration of solids in it and isknown as the “emulsion phase” of the fluidized bed. An intermediate phase betweenthe bubble and emulsion phases is addressed as cloud phase that contains a low

Fig. 3.19 An illustration offluidized bed


concentration of solid (Fig. 3.20). In real situation, however, the bubbles containvery small amounts of solids. Bubbles are not spherical, rather they have anapproximately hemispherical top and a pushed-in bottom. As the bubble rises, itpulls up the cloud and the wake with their solid. The solids are then going up anddown in the emulsion phase and circulating. The gas within a particular bubbleremains largely within that bubble, only penetrating a short distance into the sur-rounding emulsion phase. There is mass transfer between the bubble phase andemulsion phase: reactants from the bubble phase to the emulsion phase and prod-ucts in the opposite direction.

Kunii and Levenspiel (1991) with some simplifying assumptions proposed apractical, usable model of fluidized-bed behavior. This model, named two-phasemodel, assumes the bed to be divided into dense phase and bubble phase with masstransfer between the phases. The reactants diffuse from bubble to dense phase andproducts diffuse from dense to bubble phase. The rate of mass transfer in and out ofthe bubble phase affects the conversion, as does the time it takes for the bubble topass through the bed. In this context, it is also postulated that almost all the gas inexcess of that necessary for minimum fluidization will pass through the bed in theform of bubbles (Fig. 3.21). Other assumptions to develop the model include:

1. The bubbles are spherical, of uniform size2. The bubble phase does not contain solids and is in plug flow. The extent of

reaction in the bubble–cloud phase is negligible.3. The dense phase is assumed to be perfectly mixed.4. Dense phase exists at minimum fluidizing conditions. The gas occupies the

same void fraction in this phase that it had in the entire bed at minimumfluidization point.

5. All the particles are homogeneous in character, spherical in shape and uniformin size during reaction; all particles within the bed are at the same temperature.

6. In the wake, the concentration of solids equals to the solid concentration in theemulsion phase, and thus, the gaseous void fraction in wake is the same as in

Fig. 3.20 Schematic ofbubble, cloud and wake


emulsion phase. The average velocities of both solid and gas in the wake areassumed to be the same and equal to the upward velocity of the bubbles.

7. The ideal gas law applies to the gas phase in both phases.8. The mass-transfer resistance between the particles and the dense-phase gas is

negligible.

Using the two-phase model, a fluidized-bed catalytic reactor can be divided intotwo regions, one for the dense phase, i.e., the emulsion phase, and another for thebubble phase, with associated mass and heat transfer between the two regions andphases (Fig. 3.22). In the example presented below, we will consider modeling ofFBCVD for CNT synthesis.

Fig. 3.21 Bubbles in fluidized bed (Elnashaie et al. 2007)

qF

qDFqBF

Bubble Phase Dense Phase

qDoutqBout

qFig. 3.22 Illustration oftwo-phase model


Example

CNTs are grown using fluidized-bed chemical vapor deposition (FBCVD) proce-dures. Figure 3.23 illustrates the schematic of the system used. The main body ofthe reactor is a vertical 316 stainless steel cylinder with dimensions of 5.3 cm ininternal diameter (DR) and 100 cm in height (HR). The gas distributor is a stainlesssteel grid with 20-µm meshes. The reactor is enclosed by an electrical furnace withlength of about 40 cm (model: LBF-0.5-1000, AC, 240 V, 1Φ, 1.2 kW, maximumtemperature: 1000 °C), which allows monitoring the different parts of the reactorvia type K thermocouples.

To produce CNTs, the procedure presented here has been used. A mass of 5 gcatalytic particles, comprising nanoparticles of iron and cobalt supported on alu-mina, is loaded into the fluidized-bed reactor through an access port in the top of thereactor. These catalytic particles involve 3 % transition metals (1 % cobalt and 2 %iron), which act as active sites for CNT growth were prepared by a known methodof liquid impregnation. Iron nitrate (Fe (NO3)3.8H2O) and cobalt acetate (Co(CH3CO2)2.4H2O), in amount to obtain a desired percentage, were dissolved in30 ml of distilled water. The mixture then was sonicated for about 10 min. Aluminawas mixed with 100 ml of distilled water in a different flask and then stirred usingmagnet stirrer at room temperature for 15 min. Well-mixed aqueous solution ofsalts was added gently drop by drop to the alumina solution while stirring. Thesuspension was magnetically stirred at temperature around 70 °C for about 24 h.Then, the resulted slurry was dried in the oven at 220 °C and finally ground intofine powder in a mortar. The final product in the form of powder was sievedmanually to the desired particle-size distributions. These particles with average

Fig. 3.23 Schematic of thesystem used


diameter of 100 μm form the bulk of the fluidized bed. The catalytic particles arethen fluidized and heated under constant flow rate of argon supplied to the bedthrough a rotameter. The overall flow rate of argon is fixed at 2000 cc/min usingDwyer flow meter. Typically, 2–3 h are required to preheat the reactor to the desiredoperating temperatures, which is 600 °C. Once the desired reaction temperature isreached and stabilized, the carbon source is pumped into the reactor at a constant flowrate (2 cc/min) using a peristaltic pump. The carbon source used in the present workis ethanol (liquid, C2H5OH, boiling point *78.5 °C, decomposition at *300 °C)held in a measuring cylinder sealed tightly with para film tape. After about 40 min,the ethanol feeding by pump is stopped. During this period of time, the crackingreaction for the production of CNTs takes place; ethanol is pumped to the lineconnected to the path of carrier gas, argon, to the reactor.

Considering the FBCVD process, large flow of carrier gas, i.e., 2000 cc/min, isused, and if carrier gas enters the reactor at ambient temperature, it will causesignificant temperature drop inside the reactor. The temperature drop then willdeteriorate the carbon deposition process, which takes place at temperature higherthan 600 °C (Danafar et al. 2009). Warming the inlet flow of gas may prevent thetemperature drop in the reactor and improve the process production toward CNTselectivity. Besides that if the carbon source is warmed to the temperature near itsthermal decomposition, it will easily go to the reaction. A configuration, as depictedin Fig. 3.24, is to warm up the argon and ethanol using the heat of main furnace.The liquid form of ethanol converted to its vapor is then introduced to the system bythe carrier gas. A thermocouple fixed about 2 cm below the distributor indicates thetemperature of the inlet gas. Gas inlet temperature was controlled to be in the rangeof 200 ± 10 °C, below the thermal decomposition of ethanol. If temperature reachesto temperature for ethanol cracking, it will be physically decomposed, which is notdesired. Since the chemical vapor deposition is endothermic, the gases includingethanol vapor will reach to the cracking temperature in situ by the furnace in thepresence of catalytic particles. The furnace can be tuned to control the reactor

Fig. 3.24 Catalyst activityversus time


temperature, and thus, the reactor can be isothermal. The fixed point for furnace was600 °C.

The reaction for ethanol decomposition is:

C2H5OH ! 2Cþ H2Oþ 2H2 ð3:19Þ

The rate of cracking reaction to CNT is taken as first order with respect to theconcentration of ethanol (Kwok et al. 2010).

Units for “r” is molm3 s

rETH ¼ �kETHðCETHÞD ð3:20Þ

kETH ¼ k0we� E

RTD ð3:21Þ

E activation energy (j/mol)TD temperature (K)R ideal gas constant (j/mol K)k0 reaction constant (1/s)

rETH ¼ �k0we� E

RTDðCETHÞD ð3:22Þ

Since CNT formation directly depends on the catalyst activity, this variable needs tobe considered in the kinetic relation. It is assumed here that each active site willresult in a single CNT. Accordingly, catalytic activity (ψ) is defined as:

w ¼ a0 � ca0

ð3:23Þ

α0 initial numbers of active sitesγ number of active sites covered as the reaction proceeds.

According to the assumption, each active site will result in a single CNT; thenfor fresh catalyst, the number of active sites covered is zero and it increases ascarbon is deposited on the active sites, and at the end of reaction, when ψ goes tozero, it will be equal to α0.

at t ¼ 0 c ¼ 0

w 0ð Þ ¼ 1 and w t1ð Þ ¼ 0 it means that c 1ð Þ ¼ a0ð Þ

It is now clear that γ is a function of time, and hence, catalyst activity (ψ) is timedependent too. Therefore, Eq. 3.24 can be written as

dwdt

¼ � 1a0

� �dcdt

ð3:24Þ


Equation (3.24) expresses deactivation rate of catalyst that relates to the ratecarbon formed or the rate ethanol consumed. To use Eq. 3.24, the initial numbers ofactive sites (α0) must be known. It is worth mentioning that it is not practical todetermine α0 exactly, and using some reasonable approximations, α0 can be esti-mated. According to the experimental test, some characteristics of catalytic particlesare known as listed in Table 3.1

Mass of 5 g catalytic particles was used for FBCVD synthesis of CNT, and only3 % of these particles are active sites, and thus, mass of active sites (mas) will be

mas ¼ 0:03mcat ¼ 1:5� 10�4 kg

Density of catalytic particles, according to the percentage of each element, canbe determined as:

qP ¼ qFe þ qCo2

¼ 8365kgm3 ð3:25Þ

where ρFe and ρCo are density of iron and cobalt in kg/m3, respectively.Then, total volume of active sites (Vtas) can be obtained as:

Vtas ¼ mas

qP¼ 1:8� 10�8 m3 ð3:26Þ

TEM observation of catalytic particles revealed that active sites have an averagediameter of 2 nm. Assuming they have spherical shape, the total number of activesites (Nas) will be then:

Nas ¼ Vtas

Vas¼ Vtas

p6 � D3

as¼ 3:4� 1020 ð3:27Þ

If only one percent of these active sites participate in CNT formation, then α0will be about 1018.

To develop a mathematical model for FBCVD synthesis of CNTs, the followingsimplifying and reasonable assumptions are made:

1. The dense (emulsion) phase is perfectly mixed and is at incipient fluidizationconditions with constant voidage. The gas flow rate through it is that necessaryfor minimum fluidization.

Table 3.1 Some characteristics of catalytic particles

Constituents Diameter(dp)

Active sitespercentage

Active sitesdiameter (das)

Mass of catalyticparticles used (mcat)

Alumina,cobalt, iron

100 µm 3 2 nm 5 g


2. The flow of gas in excess of the minimum fluidization requirement passesthrough the bed in the form of bubbles.

3. The bubbles are spherical, of uniform size, and in plug flow.4. The bubble phase is always at quasi-steady state and also the dense phase

except for change associated with catalyst activity.5. Reaction occurs only in the dense (emulsion) phase; no reaction occurs in the

bubble phase.6. Mass and heat transfer between the bubble and emulsion phases occur at

uniform rates along the height of the bed.7. An average value of the bubble size and thus average values of the heat and

mass exchange parameters are used for computations.8. Mass- and heat-transfer resistances between the particles and the emulsion

(dense) phase are negligible.9. No elutriation of solids occurs.

10. Reactor is isothermal.

The first step includes determining the flow rates in phases. It is assumed thatdense phase exists at minimum fluidizing conditions (assumption number 2), then

qD ¼ qmf ð3:28Þ

qD is dense-phase flow rate (m3/s)qmf is minimum fluidization flow rate (m3/s)

qmf ¼ Umf � AR ð3:29Þ

Umf is minimum fluidization velocity (m/s) and AR is cross-sectional area of thereactor (m2)

AR ¼ p4ðDRÞ2 ð3:30Þ

where DR is reactor diameter (m).The minimum fluidization velocity (Umf) for catalytic particles determined

experimentally was 0.3 m/min (0.005 m/s). Putting 0.05 for DR, cross-sectional areaof the reactor equals 0.0022 m, and then, the minimum flow rate (qmf) will be1323 cc/min.

The minimum fluidization velocity can be also determined by empirical equa-tions like (Kunii and Levenspiel 1991),

Umf ¼ lfqfdP

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi25:252 þ 0:0651Ar

p� 25:25

� �ð3:31Þ


where Ar is Archimedes number defined as

Ar ¼ qf qP � qfð Þg d3P

l2fð3:32Þ

ρP (kg/m3) and dP(m) are density and mean diameter of particlesμf (kg/m s) and ρf (kg/m

3) are viscosity and density of fluid, respectively.

The fluid here contains argon and ethanol. In this work, the volume flow rate ofargon is 2000 cc/min and ethanol is pumped into the reactor at a constant flow rate(2 cc/min) using a peristaltic pump. To obtain volumetric flow rate of ethanol,(qETH)F, entering the reactor in gaseous form at 200 °C, correlation 3.33 is used.

ðqETHÞF ¼ ðqETHÞL � ðqETHÞLðqETHÞg

ð3:33Þ

(ρETH)L density of ethanol in liquid form (kg/m3)(ρETH)g density of ethanol in gas form (kg/m3)(qETH)L flow rate of liquid ethanol (m3/s).

Flow rate of ethanol in gaseous form, calculated from Eq. 3.33, is about40 cc/min. The volumetric ratio of argon to ethanol will be then 50, and it isacceptable that the viscosity and density of the fluid to be assumed equal argon’sproperties. Argon viscosity was extracted from Perry’s hand book, and its density iscalculated using ideal gas correlation.

qf ¼PMAr

RTð3:34Þ

P is process pressure, which equals to 100 kPas.

Flow rate for bubble phase (qB) will be

qB ¼ qF � qmf ð3:35Þ

qF is feed flow rate (m3/s).

The feed consists of argon and ethanol with known feed flow rates for both. Theflow rate known for ethanol is when it is at ambient temperature and in liquid form.The temperature at which ethanol enters the reactor is about 200 °C, in whichethanol is in gaseous form.

qF ¼ ðqETHÞF þ ðqArÞF ð3:36Þ


(qETH)F ethanol flow rate in feed (m3/s)(qAr)F argon flow rate in feed (m3/s).

Putting these values in Eq. 3.36, qF is determined.In the next step, the FBCVD model equations for CNT synthesis can be given as

follows using kinetic relation (Eq. 3.19) and the simple two-phase model forbubbling fluidized beds.

Mass Balance for Ethanol in Dense Phase

Assumption: Dense phase is a well-mixed reactor in which the concentration ofeach component including ethanol is homogeneous with height, and thus, it equalsthe exit of the dense phase.

qDðCETHÞD� �

in� qDðCETHÞD� �

out

þ ZH

0

ðkbdÞb � d � AR � aððCETHÞB � CETHÞD� �

dy

� 1� dð Þ 1� edð ÞrETHARH ¼ 0

ð3:37Þ

The variable “y” corresponds to the distance from the reactor entrance to anypoint along the height of the reactor varying from 0 to H, the height of fluidized bed.

qDð Þin ¼ qDð Þout¼ qDCETHð ÞD

� �out ¼ CETHð ÞD

CETHð ÞD� �

in ¼ CETHð ÞF

qDðCETHÞF � qDðCETHÞD þ ZH

0

ðkbdÞb � d � AR � aððCETHÞB� CETHÞD� �

dy� 1� dð Þ 1� edð ÞrETHARH ¼ 0

ð3:38Þ

(CETH)B ethanol concentration in bubble phase (mol/m3)(CETH)D ethanol concentration in dense phase (mol/m3)(CETH)F ethanol concentration in feed (mol/m3)

ðCETHÞF ¼ ðnETHÞFqF

¼ qETHðqETHÞLðMETHÞqF ð3:39Þ

ρETH (kg/m3) andMETH (g/mol)

are density and molecular weight of ethanol, respectively.

Kbd is the overall mass-transfer coefficient between the bubble phase and the densephase based on bubble volume (m/s) (Lu et al. 2004).


1ðKbdÞb

¼ 1ðKbcÞb

þ 1ðKcdÞb

ð3:40Þ

ðKbcÞb ¼ 4:5Umf

db

� �þ 5:85

D0:5im g0:25

d1:25b

� �ð3:41Þ

ðKcdÞb ¼ 6:78emfDimUb

d3b

� �ð3:42Þ

Dim ¼ 1� xiPnj¼1

xiDij

� � ð3:43Þ

Dij ¼ 0:04357T1:5

P v13i þ v

13j

� �2ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi1Mi

þ 1Mj

sð3:44Þ

Dim diffusivity of component “i” in gas mixture, m2/sDij diffusivity of component “i” in component “j”, m2/sMi, Mj molecular weights of “i” and “j”, respectivelyT temperature, KP pressure, pascalxi mole fraction of component “i”, dimensionlessi here is ethanol and “j” is argonv is specific molar volumes (m3/gmol)δ is fractional volume of bubbles, dimensionless, expressed as

d ¼ qBUb � AR

ð3:45Þ

Ub rising velocity of bubble, predicted using the equation of Davidson andHarrison (1963)

Ub ¼ UF � Umf þ 0:711ðgdbÞ0:5 ð3:46Þ

UF feed superficial gas velocity (m/s)

UF ¼ qFAR

ð3:47Þ

The bubble diameter (db) is predicted using the equation of Mori and Wen(1975),


db ¼ dbm � ðdbm � dboÞexp � 0:3zDR

� �ð3:48Þ

where

dbo ¼ 0:347� 7:85� 10�5ðUF � UmfÞ0:4

dbm ¼ 0:652ARðUF � UmfÞ0:4

εd dense-phase voidage, dimensionless

ed ¼ 0:4 Kunii and Levenspiel 1991ð Þ

The height of active part of the reactor or height of fluidized bed is defined as(Kunii and Levenspiel 1991)

H ¼ Hmf

1� dð3:49Þ

where Hmf corresponds to the height of fluidized bed at minimum fluidizationvelocity.

Hmf ¼ Qmf

ARð3:50Þ

where α (1/m) is defined as

a ¼ Bubble surface areaBubble volume

¼ 6db

ð3:51Þ

and db is bubble diameter.

Mass Balance for Ethanol in Bubble Phase

qBdðCETHÞB

dy¼ �ðkbdÞbððCETHÞB � CETHÞD

� �ARd ð3:52Þ

qB ¼ UbARd ð3:53Þ

Defining β as Eq. (3.54), Eq. (3.52) can be written as Eq. (3.55).

b ¼ kbdUb

ð3:54Þ

dðCETHÞBdy

¼ �bððCETHÞB � CETHÞD� � ð3:55Þ


at y ¼ 0 CETHð ÞB¼ CETHð ÞF ð3:56Þ

Solving Eq. (3.55) analytically with its initial condition (3.56) gives Eq. (3.57):

ððCETHÞB � CETHÞD� � ¼ ððCETHÞF � CETHÞD

� �e �bð Þy ð3:57Þ

As dynamic term depends upon the chemisorption on the catalyst not the gasphase, its capacitance is negligible, and therefore, it is pseudo steady state.Accordingly, Eq. (3.57) can be used to evaluate analytically the integral in the massbalance of A in the dense phase (Eq. 3.38) to obtain

qDðCETHÞF � qDðCETHÞD þZH0

ðkbdÞb � d � ARððCETHÞB

� CETHÞD� �

dy� 1� dð Þ 1� edð ÞrETHARH ¼ 0

qDðCETHÞF � qDðCETHÞD þ UbdARð1� e �bð ÞHÞððCETHÞF� CETHÞD� �� 1� dð Þ 1� edð ÞrETHARH ¼ 0

ð3:58Þ

According to the overall mass balance for ethanol, moles of ethanol convertedwith time is

nETH ¼ 1� dð Þ 1� edð ÞrETHARH ð3:59Þ

According to the stoichiometric of the reaction (Eq. 2.1), moles of carbon (mc)formed with time is

nC ¼ 2nETH ¼ 2 1� dð Þ 1� edð ÞrETHARH ð3:60Þ

Then, number of carbon atoms formed with time is

NuCatm ¼ nC � NAvo ð3:61Þ

NuCatm ¼ 2NAvo 1� dð Þ 1� edð ÞrETHARH ð3:62Þ

Using number of carbon atoms (Eq. 3.62), number of atoms of carbon involvedin one CNT with specified characteristics (length and outer and inner diameter) iscalculated using Wrapping software1 (NC-CNT). The characteristics of CNTs areaverage values obtained using electron microscopic observation. It is assumed thatthe catalyst activity (ψ) is directly proportional to the number of active sites coveredas the reaction proceeds (γ) (Eqs. 3.23 and 3.24). It is assumed that the each active

1www.photon.t.u-tokyo.ac.ir.


http://dx.doi.org/10.1007/978-981-287-496-2_2

http://www.photon.t.u-tokyo.ac.ir

site is covered by one CNT formed. Then, number of CNTs formed (NuCNT) withtime will be as:

NuCNT ¼ NuCatmNC�CNT

¼ dcdt

ð3:63Þ

Referring to Eq. (3.62), number of CNTs formed by time will be

NuCatm ¼ 2NAvo 1� dð Þ 1� edð ÞrETHARH ð3:64Þdcdt

¼ 2NAvo

NC�CNT1� dð Þ 1� edð ÞrETHARH ð3:65Þ

dwdt

¼ 2NAvo

NC�CNT� 1a0

� �1� dð Þ 1� edð ÞrETHARH ð3:66Þ

Substituting rETH from Eq. (3.22), Eq. (3.66) will be as:

dwdt

¼ 2NAvo

NC�CNT� 1a0

� �1� dð Þ 1� edð ÞARHk0we

� ERTDðCETHÞD ð3:67Þ

Accordingly, there is an algebraic equation at t = 0, when the catalyst activity isequal to 1.0, for the dense phase that can be used to get the exit ethanol concen-tration. Using difference between this ethanol concentration and the feed concen-tration and the stoichiometry of the reaction, we can calculate the amount of carbonformed or CNTs formed as it is assumed that all carbons are in the form of CNTs.The catalyst activity in the next time step will be determined from the ethanolconcentration obtained and so on. As a result, the rate of catalyst deactivation,ethanol consumed and carbon formed are obtained.

The values used are listed below:

1. The kinetic parameters were estimated as k0 = 35 1/s and E = 30 kJ/mol.2. The characteristics of materials were extracted from Perry’s chemical engi-

neering handbook as listed in Table 3.2.3. Values calculated based on the correlations are presented in Table 3.3.

As mentioned before, to collect the product (carbon formed) from the reactor, thefurnace must be shut down and the reactor must be opened. Accordingly, eachresult for grams of carbon deposited corresponds to one of the experimental runs.

Table 3.2 Characteristics of materials

Ethanol METH (ρETH)g (ρETH)L vETH49 g/gmol 40 kg/m3 789 kg/m3 59 m3/gmol

Argon MAr ρAr μf vAr39.5 g/gmol 0.9 kg/m3 2.3564 pas s 18.5 m3/gmol


Different runs under the same conditions indicated that grams of CNTs obtained donot change considerably if the reaction takes place for longer than 40 min. Thisobservation can be justified by saying that the catalyst lost its effective activity afterthis time (40 min). The amount of carbon deposited after 20, 30 and 40 min wasdetermined to be about 7.5, 8.3 and 8.5 g, respectively. The mathematical modelingof this process also showed that catalytic activity reaches near zero after about40 min (Fig. 3.24) and grams of carbon formed after 20, 30 and 40 min werepredicted relatively equal to the measured amount of carbon formed (Fig. 3.25).Consequently, it is acceptable to note that the proposed model can predict FBCVDsynthesis of CNTs, because the final results of the model agrees with the experi-mental result for the amount of CNTs formed after a of certain reaction time.

Accordingly, this model will help to obtain useful information that cannot bedetermined through experimental results. The important information includes notonly the CNTs formed but also the change in ethanol concentration in dense phaseas well as bubble phase. It is worth mentioning that ethanol concentration at the exitof the dense phase changes with time as reaction proceeds (Fig. 3.26). This figurereveal that at the beginning of the process (time = 0 and catalyst activity = 1), theethanol concentration at the exit of the dense phase equals to 4.92 mol/m3 (in thefigure, it is shown close to 5 mol/m3), and then, it increases to reach finally tothe ethanol concentration in the feed (12 mol/m3). This means that the choice ofparameters are such that for fresh catalyst there is 59 % conversion of ethanol in thedense phase and as catalyst deactivates the outlet concentration of ethanol from the

Table 3.3 Values calculatedbased on the correlations

QF (CETH)F Ub

4.6 × 10−5 (m3/s) 12 mol/m3 0.1453 m/s

UF db Di

0.0209 m/s 0.0034 m 5.69 × 10−5 m/s

Fig. 3.25 Gram of carbonaccumulated versus time


dense phase increases, because catalyst activity decreases and hence conversion indense phase decreases.

Since dense phase is assumed perfectly mixed (or CSTR), there is no change inethanol concentration with height. However, there is variation of ethanol concen-tration with height in the bubble phase, because it is plug flow. In this model, thecloud and wake are taken as parts of the dense phase, and thus, there is no reactionin the bubble phase. A variation of ethanol concentration with time observed inbubble phase is due to mass transfer between the phases (bubble and dense), and thedriving force is the difference in concentration(s). In other words, the change in theconcentration profile in the bubble phase with time is related to the change withtime in the dense phase, which is the dominating dynamics and the bubble phase isconsidered pseudo steady state. Therefore, the bubble phase is described byordinary differential equation with the independent variable as the height coordinateand solved analytically (3.47 and 3.48). Figure 3.27 shows the concentrationchange in ethanol in bubble phase with time and height. At the beginning of theprocess (time = 0), at height equals zero, ethanol concentration in the bubble phaseequals to ethanol concentration in the feed. By moving upward from the reactorentrance, this concentration decreases. It is observed in Figs. 3.24 and 3.26, asreaction proceeds, that catalyst activity decreases, and hence, the ethanol concen-tration in dense phase increases. Accordingly, the driving force for mass transferfrom bubble phase to dense phase decreases and as a result ethanol concentration inbubble phase at a certain height of reactor (except the entrance where ethanolconcentration is constant and equals to the feed concentration at all time) increaseswith time to reach the feed ethanol concentration. To make it clearer, Fig. 3.28illustrates ethanol concentration change with time at certain heights and Fig. 3.29illustrates profiles of ethanol concentration in bubble phase at a certain time. Theseprofiles indicate that the mass transfer between the bubble phase and the dense

Fig. 3.26 Concentration change with time of ethanol in dense-phase exit


phase is large so that the concentrations between them get equal after a small heightand the ethanol concentration in the bubble phase remains constant for the rest ofthe height.

The effect of process temperature on reaction kinetic is illustrated in Fig. 3.30. Itis assumed that the increase in T does not affect any other parameters than the rateconstant of the reaction k according to its exponential Arrhenius dependence upontemperature (Eq. 3.21). Accordingly, increasing process temperature increases therate of reaction due to increasing k (3.20 and 3.22). As a result, ethanol concen-tration in dense-phase exit at t = 0 decreases, and thus, conversion in dense-phase

Fig. 3.27 Three-dimensional concentration change in ethanol in bubble phase with time andheight

Fig. 3.28 Concentrationchange with time of ethanol inbubble phase versus time for acertain bed heights


exit increases. Since the catalyst has certain number of active sites and based onassumptions of this model each active site is covered by one CNT, the catalystactivity will lessen faster, if the rate of reaction increases. The final amount of CNTsformed, however, remained constant since it depends on the number of active sitesand characteristics of CNTs including their diameters, and length and increasingtemperature just enhance the rate of reaction that causes a certain amount of CNTsto produce in a shorter time. It is worth mentioning here that there are somelimitations in increasing temperature as explained in the section of CVD synthesisof CNTs. Although increasing T will increase the rate of reaction, it may deterioratethe CNT quality by moving toward the formation of amorphous carbon and carbonnanofibers, which are not the desired products of the process. This section verifiedthat experimental results and observations accompanied with mathematical mod-eling will assist a chemical engineer to predict the process behaviors and eventuallyoptimize the process.

Fig. 3.29 Concentrationchange in ethanol in bubblephase with height for certaintimes

Fig. 3.30 Profile of ethanolconcentration in dense-phaseexit for different processtemperatures


3.2 Chemical Vapor Synthesis (CVS) of Nanostructures

3.2.1 Basic Principles of CVS

Fundamentally, the syntheses of nanostructures with determined characteristicsusing bottom to up procedure need to be performed in a fluid phase. The reason isthat the mobility of the atoms and/or molecules which control their arrangement inthe liquid and gas phases is achievable. The general methods to synthesize nano-structures in liquid phase include precipitation, sol–gel, micro-emulsion, hydro-thermal techniques. Besides CVD described in the previous chapter, there are othermethods of nanostructures synthesis in the gas (vapor) phase, which includechemical vapor synthesis (CVS), inert gas condensation (IGC), physical vapordecomposition (PVD). The important issue is the ability of controlling final prop-erties and characteristics of product during the synthesis process. This ability in theliquid phase is relatively lower than in the gas-phase processes, since the synthesisin the gas phase allows capacity to control operating conditions, product particlesize, purity and structure (Kruis et al. 1998).

Developing methods for synthesis of nanostructures is centered on using theliquid-phase processes more than the gas-phase processes. Nevertheless, thegas-phase processing methods may be preferred in some cases because of theirfollowing inherent advantages (Kruis et al. 1998).

• Gas-phase processes are generally purer than liquid-based processes since eventhe most ultrapure water contains traces of minerals, detrimental forelectronic-grade semiconductors. These impurities seem to be avoidable todayonly in vacuum and gas-phase systems.

• Aerosol processes have the potential to create complex chemical structureswhich are useful in producing multicomponent materials, such ashigh-temperature superconductors.

• The process and product control is usually very good in aerosol processes.Particle size, crystallinity, degree of agglomeration, porosity, chemical homo-geneity, stoichiometry, all these properties can be controlled with relative easeby either adjusting the process parameters or adding an extra processing step,e.g., sintering or size fractionation.

• Being a non-vacuum technique, aerosol synthesis provides a cheap alternative toexpensive vacuum synthesis techniques in thin- or thick-film synthesis.Furthermore, the much higher deposition rate as compared to vacuum tech-niques may enable mass production.

• An aerosol droplet resembles a very small reactor in which chemical segregationis minimized, as any phases formed cannot leave the particle.

• Gas-phase processes for particle synthesis are usually continuous processes,while liquid-based synthesis processes or milling processes are often performedin a batch mode. Batch processes can result in product characteristics whichvary from one batch to another.


Chemical vapor synthesis (CVS) method is an effective and most popularmethod in synthesis of some nanostructure materials such as metal and metal oxidenanoparticles, thin films and nanocapsules with perfect controllable properties andstructure. The fundamental concepts of nanotechnology and its relation to transportphenomena concepts in chemical engineering are well demonstrated in the syn-thesis of a nanostructure in this method. Mechanism and steps of nanoparticleformation in the CVS method are similar to the precipitation method regardless ofthe reaction phase. In fact, the CVS method is carried out in the gas (or vapor)phase, while the precipitation method is carried out in liquid phase.

In the CVS method, the gaseous or vapor reactants enter into the reaction zonewith adjusted concentration, temperature and pressure. The reactants react chemi-cally to form a solid-phase product. Mechanisms of solid particle generation fromthe solid product molecules are too complex and are subjects of many experimentaland simulation investigation today. Briefly, when the vapor-phase mixture of theprecursors is in the chemical supersaturation situation, the reaction–condensationtakes place, and this means the product particles will nucleate without seeding(homogeneous nucleation) and/or with seeding (heterogeneous nucleation). Oncenucleation occurs and simultaneously the reaction progresses on the resultingparticles, the particle growth will occur rather than further nucleation. The nucle-ation and growth rate are functions of the process conditions such as temperature,concentration, mixing. Therefore, the rates can be modified to produce desirableparticle size.

It should be noticed that the amount of produced nanostructure particles arelimited by the flow rate of the materials in the reactor, because of the nucleation andgrowth rate of the particles in a coagulation controlled system. A high reaction–evaporation rate results in a higher number concentration of nuclei formed.Therefore, the rate of coagulation step can be increased, since the residence time ofthe nuclei in the nucleation and growth zone is unchanged. Thus, the enhancedcoagulation yields large submicronized (or even micronized) particles rather thannanosized particles. The properties of nanoparticles depend on the solid volumefraction and on the particle-size distribution. Hence, it would be advantageous tocontrol the particle-size distribution within broad limits in order to tailor propertiesfor specific applications (Haas et al. 1997).

3.2.2 Similarity Between CVS, CVD and Inert GasCondensation

Chemical synthesis of nanostructures is a rapidly growing field with a greatpotential for making useful materials. In the last section, the CVD process wasdescribed and the steps of the nanostructure formation on the catalyst wereexplained. As it is mentioned, the reaction takes place on the solid surface ofcatalyst and then the nucleation, diffusion and growth have their effects on the

3.2 Chemical Vapor Synthesis (CVS) of Nanostructures 235

nanostructure properties. Precursors both in CVD and in CVS can be metalorganics, carbonyls, hydrides, chlorides and other volatile compounds in gaseous,liquid or solid state. However, the availability of appropriate precursor is a majorlimitation of CVS process. In both processes, the required energy for conversion ofthe reactants into nanostructures can be supplied by external furnace, flame, plasma(microwave or radiofrequency) or laser (photolysis or pyrolysis) sources. The mostimportant process parameters that determine the quality and usability of thenanostructures include

• total pressure (typical range from 102 to 105 Pa),• precursor material (decomposition kinetics and ligands determining the impurity

level),• partial pressure of the precursor (determining the production rate and particle

size),• temperature or power of the energy source,• carrier gas (mass flow determining the residence time),• reactor geometry.

The product, which is a nanostructure, is extracted from the aerosol by means offilters, thermophoretic collectors and electrostatic precipitators or scrubbing by aliquid. Generally, a typical laboratory reactor consists of a precursor deliverysystem, a reaction zone, a particle collector and a pumping system. Modifications ofthe precursor delivery system and the reaction zone allow the synthesis of pureoxides, doped oxides, coated nanoparticles, functionalized nanoparticles andgranular films (Hahn 1997).

In the CVS, the effective steps take place in the gas–vapor phase as well as inertgas condensation (IGC) process. In addition, in both of them the driving force of thenanostructures formation is supersaturation. Perhaps the most straightforwardmethod of achieving supersaturation is to heat precursor to evaporate into an inertcareer gas and then mix the vapor with a cold gas to reduce the temperature. Thismethod is well suited for the production of metal nanostructures, since many metalsevaporate at reasonable rates at attainable temperatures. By including a reactive gas,such as oxygen, in the cold gas stream, oxides or other compounds of the evapo-rated material can be prepared. Controlling the mixing of the cold gas with the hotgas and the gases flow rates carrying the evaporated metal could control theparticle-size distribution. Other advances in this method have been in preparingnanocomposites and in controlling the morphology of nanostructures by controlledsintering after particle formation (Swihart 2003).

The inert gas evaporation technique (formation of particles by evaporation of thesource in an inert gas) has been used for the synthesis of ultrafine metal particles inadjusted pressure. One of the methods for the production of nanostructures is anevaporation–condensation process where a vacuum chamber is used to evaporate asubstance like Pd at reduced pressure into a static inert gas atmosphere. In theprocess condition, Pd atoms quickly lost their kinetic energy through collisions withthe gas atoms and then a highly supersaturated vapor is produced leading tohomogeneous nucleation of Pd particles and subsequent growth by condensation


and coagulation. This phenomenon takes place in a thin zone directly after thevapor source, and the nanometer-sized particles, after leaving this zone, are trans-ported by the inert career gas to the separation zone. In the separation zone, thenanostructures are collected using different methods such as nanofiltration orthermophoresis. It has been shown that increasing carrier gas pressure, evaporationtemperature and the distance between the evaporation source and the collectingposition increased the size of the product aluminum particles formed by evaporationof aluminum pellets in helium. The higher the evaporation temperature, inert gaspressure and molecular weight of the inert gas favored larger particles. In allevaporation–condensation processes, metal vapor is rapidly cooled forming clustersand, subsequently, nanostructures by nucleation, condensation and coagulation(Panda and Pratsinis 1995).

The primary advantage of IGC processes is their simple and flexible experimentalsetup. For example, mixtures of nanostructures can be synthesized or coated by justinstallation of a second evaporation source. Usually, the precursor is evaporatedthrough a thermal evaporation source such as pulse laser ablation, spark dischargegeneration, hot filament and ion sputtering (Swihart 2003). Nonetheless, the IGCmethod has several limitations with respect to the process itself and the properties ofthe product powders. The two important ones include IGC is a batch process and hasgenerally a low efficiency (Haas et al. 1997). Its low efficiency is due to powderlosses to the reactor walls. The reactor wall is colder than the evaporation source(approximately about 1000 K), and this large temperature difference between thewall and the gas causes the reactor wall to act as an effective sink to which a majorportion of the nanoparticles is transported by the free convection and/or thermo-phoresis (Mädler and Friedlander 2007). The amount of nanostructures produced isrestricted by the free convection in the vacuum chamber. This limitation can beexplained considering mechanisms like nucleation and growth of small particles in acoagulation controlled system. An enhanced rate of evaporation results in a largerconcentration of nuclei formed (number of particles per unit volume). Consequently,the number of potential coagulation partners increases while residence time of thenuclei in the nucleation and growth zone is unchanged. In other words, the coagu-lation rate is not matched by an enhanced convective transport out of the growthzone above the evaporation source. As a result, the enhanced coagulation yieldslarger primary particles, micrometer rather than nanometer particles (Haas et al.1997). The properties of nanostructures depend on the volume fraction of grainboundaries and thus on the particle-size distribution. Accordingly, it would be usefulto control the size distribution of particles within broad limits in order to tailor thenanostructures properties for specific applications. With the IGC process, however,only a small bandwidth of particle sizes seems to be accessible for a specific materialbecause of the free convection. Production of powders with smaller particle sizes andsharp distribution is particularly desirable to facilitate the fabrication of compacts bysintering these powders. Basically, the formation of nanostructures in IGC processencounters the effective steps of nucleation, growth and coagulation on the productparticle-size distribution as well as CVD and CVS processes. Modeling IGC processto predict the size of nanostructures formed as well as process yield in entire reactors


of different configurations has been subject of some studies (Panda and Pratsinis1995; Haas et al. 1997).

However, the wide range of IGC using indicates that despite its limitations thisprocess has a great potential for the synthesis of nanostructures and majorimprovements toward resolving the limitations are possible.

The IGC method certainly pioneered the field of nanostructured solids. Thewidespread use of IGC shows that despite its limitations this process has a greatpotential in nanotechnology, and hence, further development could be rewarding.

3.2.3 Nanoparticle Synthesis via CVS

Generally in the gas-phase synthesis processes, the particle is generated frommolecular precursors via chemical reactions. The powder is qualified with differentparameters such as particle microstructure, morphology, size distribution andcrystallinity. Inception of monomer formation due to the conversion of the pre-cursor into monomers (nucleation), formation of primary particles from monomers(growth species), coagulation of primary particles (formation of agglomerates) andsintering of the primary particles within the agglomerates are affected on theparticle-size distribution. The sequence of these steps and their effects on the par-ticle size are shown in Fig. 3.31 schematically (Sander et al. 2009).

At sufficiently high temperatures (before melting point of particle), particlescoalesce (sinter) faster than they coagulate, while at lower temperatures, wherecoalescence is negligibly slow, loose agglomerates are formed. At intermediateconditions, partially sintered particles are produced. In contrast to the synthesis inthe liquid phase, where an agglomeration of nanoparticles can be prevented byusing appropriate ligands, nanostructures synthesized in the gas phase will alwaysagglomerate. The agglomerate (and/or coagulate) particles can be re-dispersedsimply because of loose interaction as compared to sintered particles (hard

Fig. 3.31 Effect of inception and coalescence on the particle size and number (Sander et al. 2009)


agglomeration) that cannot be re-dispersed. These main steps cause two basicparameters of nanostructures characteristic which are particle size and particlenumber concentration, and in other words, the particle-size distribution (PSD) vary.Figure 3.32 shows PSD of nanostructures synthesized in a CVS process schemat-ically. At it is shown, the increasing time or temperature during the process allowsthe particle to change the PSD as the pattern of Fig. 3.33.

Fig. 3.32 Variation ofparticle-size distribution ofnanoparticle with time in aCVS process

Fig. 3.33 Different mechanisms of coalescence in the CVS process (Nakaso et al. 2003)


These particles are usually synthesized by the evaporation–condensation processor the gas to particle conversion routes. The two processes employ high tempera-tures, and after nucleation of the species, growth occurs by coagulation and/orsurface reaction. In case particle coalescence (sintering) is faster than collision, thenspherical particles are formed. However, as particles grow, the rate of sinteringlessens, and thus, the time for particles coalescence increases. As coalescence ofparticles cannot happen quickly with respect to particle collision, irregularly shapedaggregates are formed. The degree of aggregation depends on the relative rates ofcollision and coalescence of particles. Fairly fast coalescence results in low degreesof aggregation and large primary particles in the aggregates. The coalescence rate isdetermined by process conditions including temperature and residence time andsintering mechanisms. The coalescence rate is, however, determined by the char-acteristic sintering time. Depending on the type of material, different mechanismscontrol the coalescence rate, i.e., viscous flow, lattice diffusion, surface diffusion orgrain boundary diffusion (Nakaso et al. 2003). Silica, which sinters by a viscousflow mechanism, exhibits an extremely high degree of aggregation and small pri-mary particles, a few nanometers in size. On the contrary, boron carbide, whichsinters by solid-state diffusion, has a coalescence rate several orders of magnitudehigher than silica at the same process conditions and exhibits low degree ofaggregation and relatively large primary particles (Akhtar et al. 1994).

The PSD parameter as the most important characteristic of the process can becontrolled using operating conditions of the reaction such as temperature, pressure,dwell time, carrier gas flow and type, precursor concentration, heating and coolingrates. These operating parameters have strong effects on the mechanisms of particleprocessing. Here, it is possible to understand rules of basic concepts of chemicalengineering such as reaction kinetic, thermodynamic, heat and mass-transfer rates,velocity distribution of the components, on a completely nanosystems. Figure 3.34shows a schematic illustration of the CVS setup used for the synthesis of TiO2

particles from titanium tetra isopropoxide (TTIP) (Lorke et al. 2012).An attractive application of gas-phase synthesis of solid product is the synthesis

of core-shell nanoparticles (i.e., nanocapsulation). In this system, two steps ofcondensation and reaction take place that first nanoparticles are produced and thenthese nanoparticles are coated with another solid phase. For instance, through theinstallation of a second evaporation source into the first CVS process, it is possibleto produce particle mixtures or coat the particles with a different material. As anexample, combining the CVS process with CVD process (sputtering or evapora-tion) produces two metal composites (such as tungsten–gallium composite) made ofa metal nanocrystallites (W) surrounded by several layers of another one (Ga).Tungsten nanocrystallites are generated by high-pressure sputtering and Ga evap-oration through a thermal evaporation source (Haas et al. 1997). The setup ofcore-shell particle synthesis by a two-step CVS–CVD process at atmosphericpressure is shown in Fig. 3.35.


3.2.4 The CVS Process Simulation

As it is mentioned before, there are several phenomena in the gas-phase processwhich affect the size of nanostructures and their concentration. The mechanismsinvolved in the process should be modeled and simulated to understand the effect ofoperating conditions on the process rates. Different models with different views tothe process can be used, but let us start with a simple and monodispersity model.Titania powders are made from TiCl4 oxidation at a high production rate. Thegas-phase synthesis is a major process making more than half of the annualworldwide consumption aimed for different applications. The properties of titaniaare improved if its particle size decreases. The overall oxidation reaction for TiCl4is (Pratsinis and Spicers 1998)

TiCl4 þ 2O2 ! TiO2 þ 2Cl2 ð3:68Þ

Total rate of this reaction (R) carried out in the gas phase is demonstrated as afirst-order kinetic as follows:

Fig. 3.34 Schematic diagram of a CVS process (Lorke et al. 2012)

Fig. 3.35 Schematic diagram of a nanostructure core shell in a CVS process (Weis et al. 2013)


R ¼ kC ð3:69Þ

where C is the precursor concentration (TiCl4) in the vapor (mol/cm3) and K is thereaction rate constant (s−1). Total consumption of TiCl4 can be carried out in the gasphase or on a solid surface. Hence, two distinct chemical pathways for reaction ofTiCl4 vapor can be identified: First, vapor of TiCl4 may react with oxygen in the gasphase forming titania or precursor particles at rate Rg. Second, TiCl4 vapor mayreact with oxygen on the surface of titania particles coating them with titania orprecursor films at rate Rs. As a result, the overall reaction rate of TiCl4 consumption(R) can be written as summation of its consumption through both surface andgas-phase reactions:

R ¼ Rg þ Rs� � ð3:70Þ

Rg ¼ kgC ð3:71Þ

Rs ¼ ksAC ð3:72Þ

where the kg and ks are rate constant of the gas-phase and surface reactions,respectively, and A is total aerosol surface area over which the TiCl4 oxidation takesplace (cm2/cm3). This area may be constant during film growth, while it is rapidlychanging during powder manufacture since it is the area of the newly formed titaniaparticles.

When the surface growth reaction component is neglected (case without surfacegrowth), kg = k and ks = 0, while in the general case (with gas phase forming andsurface growth of nanoparticles) the gas-phase reaction constant is calculated by:kg = k − Aks. At high precursor concentrations though, the rate of TiCl4 con-sumption by surface reaction employed here may exceed the overall rate. When thishappens, the surface reaction rate is set as ks = k/A and kg is set equal to zero so thatthe mass balance is preserved and the overall precursor conversion remainsunchanged (Tsantilis and Pratsinis 2004).

Product molecules are formed by instantaneous chemical reaction and grow tomolecular clusters and macroscopic particles by successive collisions. Initially, theparticle coalescence rate is much larger than the collision rate so spherical particlesare formed. As the particle size increases, however, aggregates of spherical primaryparticles are formed. Assuming that all aggregates contain the same number ofequally sized primary particles, the evolution of the aggregate particle-size distri-bution is (Kruis et al. 1993):

dNa=dt ¼ �0:5bN2 ð3:73Þ

where Na is total aggregate number concentration per unit gas volume (#/cm3), N istotal particle concentration N, (#/cm3), and β is the collision frequency function(m3/s) of equally sized particles from free molecule to continuum particle-sizeregime and is defined in the literature (Kruis et al. 1993; Panda and Pratsinis 1995;


Pratsinis and Spicers 1998). On the other side, nucleation increases the particlenumber with forming new nuclease. The nucleation rate In is equal to the rate ofnew particle (here, molecule) formation by gas-phase chemical reaction over a widerange of conditions (Pratsinis and Spicers 1998):

dNn=dt ¼ In ¼ RgNA ð3:74Þ

where NA is the Avogadro number. Thus, neglecting the spread of the aerosol sizedistribution and assuming perfect coalescence upon particle collision, rate of thetotal particle concentration change (dN/dt) is equal to summation of nucleation andcoalescence rates. This is described as following equation:

dN=dt ¼ dNn=dt þ dNa=dt ¼ RgNA � 0:5bN2 ð3:75Þ

Equation (3.75) provides a simple and commonly accepted description ofnucleation and growth effects on the nanoparticle number concentration. Thisequation is modified in the literature to synthesis model to describe the variation ofnumber found. According to this model, the number concentration of the con-densing solid monomers in molecules, nm, per unit mass of gas decreases bynucleation and condensation neglecting the subcritical cluster distribution (Tsantiliset al. 1999):

dnmdt

¼ � g�

qgv1

ffiffiffiffiffiffiffiffiffi2rpm1

rn2msS exp h� 4h3

27ðln SÞ2 !

� pd2pnmsðS� 1ÞffiffiffiffiffiffiffiffiffiffiffikBT2pm1

rNf ðKnÞ

ð3:76Þ

where g� is the number of monomers in the particle of critical size, dp is the averageparticle diameter, nms is the monomer concentration at saturation in molecules, S isthe saturation ratio (nmqgn

�1ms ), kB the Boltzmann constant, m1 is the nanoparticle

molecular mass, v1 is the equivalent sphere volume of nanoparticle molecules basedon its atomic diameter (d), N is the number concentration of particles, r is thesurface tension of the solid, h is the dimensionless surface tension (=rs1=KBT), andKn the Knudsen number (=2λ/dp where λ is mean free path of the gas). The firstright-hand side term in Eq. (3.76) accounts for the particle molecules lost bynucleation. The second RHS term accounts for the loss of the particle molecules(monomers) by condensation. The effect of particle curvature (Kelvin effect) can betaken into account by replacing (S − 1) on the second RHS term of Eq. (3.76) with

(S� Sd�=dp ). d* is critical or Kelvin diameter, and dp (¼ð6VNpÞ

13) is average particle

size. Assuming spherical particles, these characteristic diameters are determined bythe following equations (Panda and Pratsinis 1995):


d�p ¼ 4rv1kBTðln SÞ ð3:77Þ

A mass balance requires that the total particle and monomer mass should alwaysbe equal to the initial mass of the monomers (nm0 m1 = V ρp + nm m1), where m1 ismass of monomer and ρp is the solid density. Thus, the total volume of particles perunit mass of gas (V) is given by (Tsantilis et al. 1999)

dV=dt ¼ � dnm=dtð Þ m1= qp� � ¼ v1 dnm=dtð Þ: ð3:78Þ

dVdt

¼ � g�

qgv21

ffiffiffiffiffiffiffiffiffi2rpm1

rn2msS exp h� 4h3

27ðln SÞ2 !

� pd2pnmsv1 S� 1ð ÞffiffiffiffiffiffiffiffiffiffiffikBT2pm1

rNf Knð Þ

ð3:79Þ

The first term on the RHS of Eq. (3.79) accounts for the volume contribution bynucleation, and the second RHS term accounts for the contribution by condensa-tion–surface reaction. This equation describes the variation of volume of thenanoparticle (nm nanoparticles number with the same volume) as a function of theprocess conditions.

3.2.5 Sintering Effect

The formation of nanostructures from gaseous precursors includes several stepswith complex mechanisms. During these processes, product molecules generated bychemical reactions that form molecular clusters become particles by surface reac-tions. Nucleation, growth, coagulation and sintering are the important steps to formthese particles. The nucleation of a particle is induced by a collision of two pre-cursors from the gas phase. These first particles grow due to condensation ofmolecules from the surrounding gas phase or further collisions with other particles.Condensation leads to growth and increases each particle volume and enhances thesphericity.

Sintering reduces the surface of the particles and makes them more sphericity,whereas the coagulation of particles decreases the sphericity of the particles. Therelative timescale of the sintering and the coagulation processes determines theshape and the size of the particles. The faster the sintering is, the more sphericity theparticles become. Sintering of nanoparticles in the gas phase involves onlyintra-aggregate densification and restructuring, and no interaggregate mass transfer.In the earliest stages of growth, restructuring and sintering occur by an atomisticdiffusion mechanism. Early in these processes high temperatures prevail, particlecoalescence (sintering) is faster than collisions resulting in spherical particles.Downstream, however, the temperature drops, particles grow further and theirsintering energy decreases substantially so particle coalescence can no longer be


regarded instantaneous with respect to particle collision. Consequently, irregularlyshaped particles (agglomerates or aggregates) are formed (Gutzow et al. 1998).Figure 3.36 illustrates sintering to change particles morphology where two separatenanoparticles merge together because of sintering after a definite time (character-istic coalescence time). In the assumed state of infinite time, the sphericity of thesintered particle will be perfect.

The theory of sintering of solid particles or highly viscous fluids is based on theinteraction of two spheres with equal diameter (for large different size of twoparticles, another phenomenon such as Ostwald ripening takes place). For the initialstage of the sintering process, a kinetic model was formulated for the viscous flowsintering mechanism. At the contact point of the two spheres, a neck is formedwhose size x grows proportional to t0.5 (t is the sintering time). This model has beengeneralized for other sintering processes such as solid-state surface or volumediffusion. In this case, the neck size is related to the surface area of an aggregate “a”(m2), and therefore, the kinetic equation is written in the following form (Koch andFriedlander 1990)

dadt

� 1sf

1� aa0

� �c�1

ð3:80Þ

where a0 as the initial condition of Eq. 3.80 is the initial surface area of the twospheres and c and sf are constants whose values depend on the underlying sinteringmechanism. For example, for the viscous flow mechanism, c ¼ 1, whereas forsolid-state diffusion, c ¼ 2:6. The characteristic coalescence time sf is a strongfunction of the temperature. For the coalescence of two equal-sized liquid spheressf given by (Koch and Friedlander 1990) is:

sf ¼ ldr

ð3:81Þ

where l is the viscosity (N s/m2), d the diameter of the final sphere (m), and r thesurface tension (N/m).

A detailed numerical analysis of the process of viscous flow coalescence wasfound that the neck radius x (Fig. 3.43) approaches its final value xfinal (the radius ofthe resulting sphere) exponentially:

Fig. 3.36 Schematic diagram of sintering between two spherical particles


dxdt

� � 1sfðx� xfinalÞ ð3:82Þ

This equation holds beyond a small initial timescale where 10–15 % of the necksize evolution takes place. From the above relation, the surface area “a” can becalculated if it is assumed that the remaining parts of the coalescing spheres keeptheir spherical shape. For long times (t sf ), this leads to a similar behavior for thesurface area (Nakaso et al. 2002):

dadt

¼ � 1sfða� asÞ ð3:83Þ

where as (m2) is the surface area of the completely fused (spherical) particle which

is calculated as Eq. (3.85):

as ¼ vvm

� �23

am ð3:84Þ

where am and vm denote monomer (TiO2 molecule) surface area and volume,respectively.

As an example, consider a chemical reaction which produces titania from tita-nium precursor by either combustion or hydrolysis as shown below (Heine andPratsinis 2007):

Titanium precursor TiCl4ð Þ ! Titania TiO2ð Þ ð3:85Þ

The rate of reactant consumption (�RTiCl4 ) and product production (RTiO2) is thesame which follows the first-order kinetic, but of course with opposite signsbecause one is consumption and the other is production:

�RTiCl4 ¼ RTiO2 ¼ KCTiCl4 ð3:86Þ

The formed titania particles are followed by coagulation and sintering.Neglecting primary particle polydispersity and surface growth, the evolution of thetotal particle number concentration “N” (#/m3) is given by the monodispersepopulation balance (PB):

dNdt

¼ NARTiO2 �12bN2 ð3:87Þ

where NA is Avogadro’s number and b is the appropriate collision frequencyfunction. Molecules of product are formed by instant chemical reaction and thengrow to molecular clusters by following collisions. At first, the rate of particlecoalescence is much larger than the rate of collision, and hence, spherical particles


are formed. As the size of particles increase, aggregates of spherical particles areformed.

The surface area of an aggregate particle increases by coagulation and decreasesby sintering; therefore, the rates of change in the total particle surface area (a) are:

dadt

¼ NARTiO2am � 1sfða� asÞ ð3:88Þ

The variation rates of the volume concentrations of an aggregate, V (m3/m3 gas),affected by coagulation are

dVdt

¼ NARTiO2vm: ð3:89Þ

3.3 Precipitation of Nanostructures

3.3.1 Definition of the Process

An insoluble solid that forms during a reaction is called a precipitate. A reactionwhich forms a precipitate is called a precipitation reaction. The limewater test forcarbon dioxide is a precipitation reaction. Limewater is actually a dilute solution ofcalcium hydroxide. The calcium hydroxide reacts with carbon dioxide to formcalcium carbonate, which is insoluble in water:

Ca OHð Þ2 þ CO2 ! CaCO3 þ H2O ð3:90Þ

As another example, reaction of silver nitrate (AgNO3) and hydrochloride acid(HCl) in aqueous solution generates silver chloride which has low solubility in thesolution and then precipitate in the liquid phase. This reaction is:

AgNO3 þ HCl ! AgClþ HNO3 ð3:91Þ

Most precipitation reactions are very fast reactions that occur between ions. Thismakes them very useful for identifying specific ions based on the type of precipitateformed.

Precipitation reactions have a number of other uses such as production of col-ored pigments for paints and dyes, removal of toxic chemicals from water andseparation of reaction products.

For two components MmR1n and NnR2 m which are oxidizer and reducer (or acidand alkaline) dissolved in a solvent, a reaction takes place and a material, MmNn,produces which dissolves and precipitates according to the overall equation:


MmR1n solnð Þ þ NnR2m solnð Þ $ MmNn solnð Þ $ MmNn solidð Þ ð3:92Þ

This overall equation is carried out as following intermediate reactions:

MmR1n solnð Þ $ mMþn solnð Þ þ n R1ð Þ�m solnð Þ ð3:93Þ

NnR2m solnð Þ $ nN�m solnð Þ þm R2ð Þþn solnð Þ ð3:94Þ

mMþn solnð Þ þ nN�m solnð Þ $ MmNn solnð Þ $ MmNn solidð Þ ð3:95Þ

The dissociation rate of the reactants MmR1n and NnR2m (reactions 3.93 and3.94) is fast enough, and therefore, the main step in the precipitation process is thereaction 3.87. Thus, in the equilibrium conditions of this reaction, it can be:

aMmNn ¼ aMmNnðsolnÞ ¼ amMþnðsolnÞanN�mðsolnÞ ð3:96Þ

The value ai is the activity of the component i in solution, a measure of the molarconcentration (moles/L) of an ion in solution under ideal conditions of infinitedilution. Expressing the activities in terms of the product of molar concentrationsand activity coefficients, γ (a measure of the extent the ion deviates from idealbehavior in solution), thus,

ai ¼ ci � Ci ð3:97Þ

where ci 1, and this equation becomes:

Mþn½ �mcmMþn N�m½ �ncnN�m ¼ Ksp ð3:98Þ

where the constant Ksp is the solubility product constant. For dilute solutions ofelectrolytes (concentration ≤10−2 molar), the activity coefficient is approximatelyone (γ = 1; it approaches one as the solution becomes more dilute, becoming oneunder the ideal conditions of infinite dilution). In this case, following equation isderived:

Ksp ¼ Mþn½ �m N�m½ �n ð3:99Þ

The value for the ion product is calculated from the expression above. Becausethe ion product Q is less than Ksp, no precipitate will form. Only when the ionproduct is greater than Ksp, a precipitate forms. Basically, the precipitation processincludes several steps depicted schematically in Fig. 3.37. Reaction is the main andkey step for synthesis of the nanostructures. The post-precipitation steps includingfiltration, washing, drying and calcination treat the nanostructures to be separatedand purified.


3.3.2 Effect of the Process Parameters on the ProducedNanoparticles

In the precipitation process, the chemical reaction generates the precipitate, but sizeof precipitate is a function of a number of physical events as well as the chemicalreactions. When the reaction starts, the generated precipitate molecules interacttogether. The results of these interactions are different events named nucleation andgrowth. In the nucleation, the product crystals are created from a supersaturatedsolution, while in the crystal growth step, the crystal size increases up to the desiredsize by growth from supersaturated solution. In addition of these events, there areother events affecting the particle size of the product such as Ostwald ripening andagglomeration. In the Ostwald ripening, a large particle grows at the expense of thesmaller ones until the latter disappears completely. This event occurs at relativelylow temperature (Ostwald ripening phenomenon will be explained in Sect. 3.3.3.3).In the agglomeration, formation of particle clusters linked by crystalline bridges (insupersaturated solution) occurs. Agglomeration of individual nanostructuresthrough chemical bonds and physical attraction forces is created at interfaceswithout altering the individual nanostructures. The smaller the particles, the greater

Fig. 3.37 Schematic diagram of precipitation process steps

3.3 Precipitation of Nanostructures 249

the bonding forces. All of these events directly affect the particle size of precipitate.Main factors affecting nanoparticles can be mentioned as follows (MARLAP 2004):

• Rate of precipitation: Formation of large, well-shaped crystals is encouragedthrough slow precipitation because fewer nuclei form and they have time togrow into larger crystals. Larger crystals have less solubility than smallercrystals because smaller crystals possess more exposed surface area to thesolution. Larger crystals also provide less surface area for the absorption offoreign ions. Slow precipitation can be accomplished by adding a very dilutesolution of the precipitant gradually, with stirring, to a medium in which theresulting precipitate initially has a moderate solubility.

• Concentration of ions and solubility of solids: The rate of precipitationdepends on the concentration of ions in solution and the solubility of the solidsformed during the process. A solution containing a low concentration of ions,but sufficient concentration to form a precipitate will slow the process, resultingin larger crystal formation. At the same time, increasing the solubility of thesolid will also slow precipitation.

• Temperature: Precipitation at higher temperature slows nucleation and crystalgrowth because of the increased thermal motion of the particles in solution.Therefore, larger crystals form, reducing the amount of adsorption and occlu-sion. However, most solids are more soluble at elevated temperatures, effec-tively reducing precipitate yield; an optimum temperature balances theseopposing factors.

• Digestion: Extremely small particles, with a radius on the order of one micron,are more soluble than larger particles because of their larger surface areacompared to their volume (weight). Therefore, when a precipitate is heated overtime (digestion), the small crystals dissolve and larger crystals grow (Ostwaldripening). Effectively, the small crystals are recrystallized, allowing the escapeof impurities (occluded ions) and growth of larger crystals. This process reducesthe surface area for adsorption of foreign ions and, at the same time, replaces theimpurities with common ions that properly fit the crystal lattice.

• Degree of supersaturation: A relatively high degree of supersaturation isrequired for spontaneous nucleation, and degree of supersaturation is the mainfactor in determining the physical character of a precipitate. Generally, thehigher the supersaturation used, the more aggregated colloid will precipitatebecause more nuclei form under conditions of higher supersaturation and crystalgrowth is faster. In contrast, the lower the supersaturation used, the more likely acrystalline precipitate will form because fewer nuclei form under these condi-tions and crystal growth is slower. Most perfect crystals are formed, therefore,from supersaturated solutions that require lower ion concentrations to reach thenecessary degree of supersaturation and, as a result, inhibit the rate of nucleationand crystal growth.

• Solvent: The nature of the solvent affects the solubility of an ionic solid (pre-cipitate) in the solvent. The polarity of water can be reduced by the addition ofother miscible solvents such as alcohols, thereby reducing the solubility of


precipitates. For example, strontium chromate (SrCrO4) is soluble in water, butit is insoluble in a methanol–water mixture and can be effectively precipitatedfrom the solution. In some procedures, precipitation is achieved by addingalcohol to an aqueous solution, but the dilution effect might reduce the yieldbecause it lowers the concentration of ions in solution.

• Stirring: Stirring the solution during precipitation increases the motion ofparticles in solution and decreases the localized buildup of concentration of ionsby keeping the solution thoroughly mixed. Both of these properties slownucleation and crystal growth, thus promoting larger and purer crystals. Thisapproach also promotes recrystallization because the smaller crystals, with theirnet larger surface area, are more soluble under these conditions.

• pH: Altering the pH of aqueous solutions affects the concentration of ions in theprecipitation equilibrium by the common-ion effect. For example, calciumoxalate can be precipitated or dissolved, depending on the pH of the solution.The pH can also influence selective formation of precipitates. Barium chromateprecipitates in the presence of strontium at pH 4–8, leaving strontium in solu-tion. Sodium carbonate is added and strontium precipitates after ammonia isadded to make the solution more alkaline.

The purity of nanostructures precipitated as an important characteristic param-eter of product is controllable in the precipitation process. The purity can increaseusing re-precipitation technique, and adding second component to the nanostructurewith purposed function is possible with co-precipitation technique.

Re-precipitation increases the purity of precipitates. During the initial precipi-tation, crystals collected contain only a small amount of foreign ions relative to thecommon ions of the crystals. When the precipitate is re-dissolved in pure solvent,the foreign ions are released into solution, producing a concentration of impuritiesmuch lower than that in the original precipitating solution. On re-precipitation, asmall fraction of impurities are carried down with the precipitate, but the relativeamount is much less than the original. Nevertheless, foreign ions are not eliminatedbecause absorption is greater at lower, rather than at higher, concentrations. Onbalance, re-precipitation increases the purity of the crystals (MARLAP, 2004).

Co-precipitation, on the contrary, control impurities present in the nanostructure.The common definition of co-precipitation is the contamination of a precipitate bysubstances that are normally soluble under the conditions of precipitation. In otherwords, co-precipitation is alternately defined as the precipitation of one compoundsimultaneously with one or more other compounds to form mixed crystals. Each ispresent in macro-concentrations (i.e., sufficient concentrations to exceed the solu-bility product of each). Co-precipitation introduces foreign ions into a precipitate asdesirable impurities that would normally be expected to remain in solution; pre-cipitation techniques are normally used to maximize this effect while minimizingthe introduction of undesirable impurities. Three processes are responsible forco-precipitation, although the distinction between these processes is not alwaysclear. They are: inclusion, surface adsorption and occlusion (MARLAP 2004).


3.3.3 General Description of the Nucleationand Growth Steps

As it is mentioned before, a number of physical phenomena occur during theprecipitation (or crystallization) process which depends upon particle size. Thesephenomena are nucleation, crystal growth, Ostwald ripening and agglomeration.These phenomena (steps) are discussed briefly below.

3.3.3.1 Nucleation

To start the nucleation step, the concentration of precipitate should be in thesupersaturation region. Consider a solution with precipitate concentration Ci attemperature Ti. It is possible to supersaturate the system via two distinct methods:cooling the solution (path a in Fig. 3.45) and increasing the precipitate concen-tration at constant temperature (path b in Fig. 3.38). The later is possible with orwithout reaction. Supersaturation creates the required driving force to generate theprecipitate molecules.

When molecules of the precipitate are generated in the solution, they “see”solvent molecules around them frequently and other precipitate molecules occa-sionally. There will be some attractive forces between these molecules. Sometimes,two molecules stay together long enough to meet up with a third one, and then afourth (and fifth, etc.) solute molecules. Result of this multiple meeting is nucle-ation. Most of the time when there are just a few molecules joined together, theymay break apart because the attractive force of solvent-precipitate molecules ishigher than precipitate–precipitate attractive force. When a certain number ofprecipitate molecules are reached (a so-called critical size of cluster), the combinedattractive forces between the precipitate molecules become stronger than the otherforces in the solution which tend to disrupt the formation of these nuclei (nucleationsite). At this situation, the other precipitate molecules feel the attractive force fromthe nuclei and were bound to it. That is how the crystal begins to grow. At the

Fig. 3.38 Changing thesolution situation fromundersaturation tosupersaturation


metastable zone (Fig. 3.45), the supersaturation is not high enough to producenuclei of the critical size.

Nucleation occurs in a gas, liquid or solid phases. Most nucleation processes arephysical events, rather than chemical events. Nucleation is a complex phenomenonwhich involves different mechanisms. Nucleation induced by foreign surface nor-mally occurs at preferential sites such as phase boundaries of contacting liquid orvapor or impurities. Suspended particles or minute bubbles also provide nucleationsites. This is called heterogeneous nucleation which requires seeding (adding for-eign nucleuses to initiate nucleation). Compared to the heterogeneous nucleation,homogeneous nucleation occurs with much more difficulty in the interior of auniform substance. The creation of a nucleus implies the formation of an interfaceat the boundaries of a new phase.

In the homogeneous nucleation, the change in Gibbs free energy per unit volumeof the solid phase, ΔGv, is dependent on the concentration of the solute (Cao 2004):

DGv ¼ � kBTX

lnð1þ SÞ ð3:100Þ

where S is the supersaturation defined by (C − Co)/Co, C is the concentration of thesolute, Co is the equilibrium concentration or solubility, and Ω is the atomic volume(also called the van der Waals volume or molecular volume). Without supersatu-ration (i.e., S = 0), ΔGv is zero, and no nucleation would occur. In the supersatu-ration condition (C > Co or σ > 0), ΔGv is negative and nucleation occursspontaneously. In this state, for total Gibbs free energy, the following relation isapplied (Thanh et al. 2014)

DG ¼ �4=3pr3DGv þ 4pr2r ð3:101Þ

where the first term (volume free energy) shows the energy gain of creating a newvolume and the second term (interfacial energy) shows the energy loss due tosurface tension (σ) of the new interface. It costs free energy to add molecules to this

cluster (because dDGdr [ 0 until the radius reaches to critical size r*). Figure 3.39

shows the relation of total Gibbs free energy with the particle size. As it is shown inthis figure, trend of Gibbs free energy follows a maximum point in the curve. Thispoint introduces critical particle size (r*) (Rao et al. 2007).

At the critical size, dDGdr jr¼r� ¼ 0. This relation gives the critical size of particles:

r� ¼ � 2rDGv

ð3:102Þ

The free energy needed to form this critical radius can be found by:

DG� ¼ 16pr3

3ðDGvÞ2ð3:103Þ


ΔG* is the energy barrier that a nucleation process must overcome, and r* repre-sents the minimum size of a stable spherical nucleus. The above discussion is basedon a supersaturated solution; however, all the concepts can be generalized for asupersaturated vapor and a supercooled vapor or liquid. In the synthesis andpreparation of nanostructures by nucleation from supersaturated solution or vapor,this critical size represents the limit of how small nanostructure can be synthesized.To reduce the critical size and free energy, it is required to increase the change inGibbs free energy, ΔGv, and reduce the surface energy of the new phase (r). ΔGv

can be significantly increased by increasing the supersaturation (S) (Koetz 2007).Figure 3.40 compares Gibbs free energy of two precipitate systems at two

temperatures. The critical sizes and critical free energy of the spherical nuclei arevaried with temperature. Other affecting parameters include supersaturation, type ofsolvent, existence of additives in solution and impurities in solid phase (Fig. 3.41).

Fig. 3.39 Gibbs free energyof nucleation as a function ofnanoparticle size

Fig. 3.40 Variation of Gibbsfree energy of nucleation withtemperature


The rate of nucleation can be calculated by the following equation (Vekilov2010):

I ¼ dNdt

¼ N CP ð3:104Þ

where I is nucleation rate (#/sec), N is the number of growth species per unitvolume (#/cm3), which can be used as nucleation centers (in homogeneousnucleation, it equals the initial concentration, Co), P is probability that a thermo-dynamic fluctuation of critical free energy (P ¼ exp �DG�=kTð Þ), Γ is the suc-cessful jump frequency of growth species from one site to another (C ¼ E

3pk3l),

where λ is the diameter of the growth species and μ is the viscosity of the solution.So the rate of nucleation described in Eq. (3.104) can be rewritten with thereplacement of its parameters (which is mentioned). This result is presented inEq. (3.105) which is an Arrhenius-type functionality of temperature:

I ¼ C0kT

3pk3gexp �DG�

kBT

� �ð3:105Þ

Finally, the rate of nucleation of particles during time can be described ascombining Eqs. 3.105 and 3.103 to give Eq. 3.106.

I ¼ C0kT

3pk3gA exp � 16pr3X2

3k3BT3ðlnð1þ sÞÞ2 !

ð3:106Þ

In this equation, three experimental parameters can be varied: supersaturation(S), temperature (T) and the surface free energy “r” (surface free energy of the solidis equivalent to surface tension of the liquid, and the unit is the same N/m) which isthe variation caused by different surfactants. Effect of these three parameters on thenucleation rate is shown in Fig. 3.41. The largest effect on nucleation rate comesfrom supersaturation, where a change from S = 2 to S = 4 causes an increase in thenucleation rate about ∼1070.

(a) (b) (c)

Fig. 3.41 Order of magnitude effect of a supersaturation, b temperature and c surface tension onthe nucleation rate (Thanh et al. 2014)


Heterogeneous nucleation occurs much more often than homogeneous nucle-ation because heterogeneous nucleation requires less energy than homogeneousnucleation (ΔGheter. < ΔGhomo.) as shown in Fig. 3.42. At such preferential sites, theeffective surface energy is lower, thus diminishing the free-energy barrier andfacilitating nucleation (Kuni et al. 2001). The barrier energy needed for heteroge-neous nucleation is reduced. This decreasing is shown in Fig. 3.42. Comparison ofDG� in this figure shows lower limitation of heterogeneous nucleation related tohomogeneous nucleation.

The free energy needed for heterogeneous nucleation is equal to the product ofhomogeneous nucleation and a heterogeneity factor (f) as a function of the contactangle (θ). The definition of contact angel and its rule on the particle formation areshown in Fig. 3.43. The heterogeneity factor which is in the range 0–1 is calculatedas follows (Sear 2007):

DGhete ¼ DGhomo:f hð Þ ð3:107Þ

f hð Þ ¼ 2� 3coshþ cos3h� �

=4 ð3:108Þ

In practice, the nucleation rate is a function of supersaturation (S) and nature ofprecipitate (presented with θ), and therefore, it can be calculated using the followingsemiexperimental equation (Sear 2007):

Fig. 3.42 Comparison of free Gibbs energy for homogeneous and heterogeneous nucleation

Fig. 3.43 Definition of contact angel in nucleation


I exp � 4h3

27 ln2ðSþ 1Þ

� �ð3:109Þ

3.3.3.2 Growth

The size distribution of nanostructures depends on the subsequent growth processof the nuclei, which is a multistep process, and the major steps include:

• Generation of growth species,• Diffusion of the growth species from bulk to the growth surface,• Adsorption of the growth species onto the growth surface,• Growth through irreversible incorporation of growth species onto the solid

surface.

These steps can be generally classified to “diffusion” and “surface growth”processes. Supplying the growth species to the growth surface is termed diffusion,which includes the generation, diffusion and adsorption of growth species onto thegrowth surface. Incorporation of growth species adsorbed on the growth surfaceinto solid structure is then addressed as surface growth.

Diffusion regime: In the diffusion regime, the diffusion rate is the only mechanismaffecting the growth. In this case, total diffusion flux through surface of nuclei is(Wen et al. 2014)

J ¼ 4pr2DdCdx

ð3:110Þ

where r, J, D and C are, respectively, the particle radius, the total flux of monomerspassing through a spherical plane with radius x, the diffusion coefficient and theconcentration at a distance x. Assuming linear dependency of concentration in thediffusion layer, as shown in Fig. 3.44, Eq. 3.110 is reformed as Eq. 3.111:

J ¼ 4pr r þ dð ÞDDCd

� 4prDDC ðr � dÞ ð3:111Þ

where δ (m) is the distance from the particle surface to the bulk concentration ofmonomers within solution, and DC ¼ Cb � Ci, where Cb is the bulk concentrationof monomers within the solution and Ci is the concentration of monomers at thesolid–liquid interface. The concentration Cr in Fig. 3.44 is the solubility of theparticle.

The molar content of nuclei (N) varies with time because of diffusion. The molarcontent of nuclei is defined as its volume (V) to its molar volume (Vm). As theresult, it is clear that


dNdt

¼ dðV=VmÞdt

¼ 1Vm

dð4=3pr3Þdt

¼ 4pr2

Vm

drEt

ð3:112Þ

where r0 is the initial particle radius which formed in the nucleation step.In the diffusion regime, the molar content variation of the nuclei with time is

because of diffusion. Therefore, with equality of Eqs. 3.111 and 3.112, the rate ofsize variation of nanoparticle with time (growth rate) follows the following equation(Kwon and Hyeon 2011):

drdt

¼ DVmDCr

ð3:113Þ

Initial condition:

at t ¼ 0 r ¼ ro ð3:114Þ

Assuming constant amount of Ci (which is approximated as equilibrium con-centration, Cr), the DVmDC in Eq. 3.113 can be consider as a constant. In this state,the growth of the particle is derived with time as:

r2 ¼ KDt þ r2o ð3:115Þ

where KD ¼ 2DVmDC and ro is the initial radius of nuclei. For two particles withinitial radius difference, dro, the radius difference, dr, decreases as time increases orparticles grow bigger, according to:

dr ¼ rodro=r ð3:116Þ

Fig. 3.44 Concentration variation over surface of a spherical nucleus in the diffusion regime


Combining Eqs. 3.115 and 3.116, the following relation is obtained:

dr ¼ rodro= kDt þ r2o� �1=2 ð3:117Þ

Both indicate that the radius difference decreases with increase in nuclear radiusand prolonged growth time. The diffusion-controlled growth promotes the forma-tion of uniformly sized particles.

Surface growth regime: When diffusion of growth species from bulk to thegrowth surface is sufficiently fast, the concentration of growth species on the sur-face will be the same as that in the bulk, and thus, the growth rate is controlled bythe surface process. Figure 3.45 shows the concentration variation over surface of aspherical nucleus in the surface growth regime. As it is observed, there is noconcentration variation in the diffusion region because of negligible diffusionresistance (and therefore fast diffusion rate), while there is a sharp decrease in theconcentration near the nanostructures surface because of controlling effect of sur-face growth.

There are two mechanisms, mononuclear and polynuclear growth, for the surfaceprocesses. For the mononuclear growth, the growth proceeds layer by layer; thegrowth species is incorporated into one layer and proceeds to another layer onlyafter the growth of the previous layer is completed. In this mechanism, the growthof particle and increase in its radius is uniform. The rate of nuclei mass transfer intothe surface of particle (nA) can be described as a film mass transfer:

nA ¼ ApkgDC0 ð3:118Þ

where Ap is the nanoparticle surface, kg is film mass-transfer coefficient andDC0 ¼ Ci � Cr, where Ci ¼ Cb. The Ap is assumed constant during growth whenthe particle radius is in the range of nanometer. In this regime, the rate of particle

Fig. 3.45 Concentration variation over surface of a spherical nucleus in the surface growth regime


growth is presented by equality of the nuclei mass-transfer rate (Eq. 3.118) and thegrowth rate (3.113) which gives the following equation:

drdt

¼ 4pr2kmDC0 ð3:119Þ

Initial condition:

at t ¼ 0 r ¼ ro ð3:120Þ

where km (=Ap kg Vm) is a proportionality constant. The growth of the nanoparticlesize is given by integration of Eq. 3.119 as follows:

1r¼ 1

ro� Kmt ð3:121Þ

where Km ¼ 4pkmDC0 is constant. The radius difference increases with theincreasing radius of the nuclei as follows which is derived by differentiation ofEq. (3.121).

dr ¼ dror2=r2o ð3:122Þ

Combining Eqs. 3.122 and 3.121 gives:

dr ¼ dro= 1� Kmrotð Þ2 ð3:123Þ

where Kmrot\1, and it means that the radius of nucleus is not large.Equation (3.123) indicates that the radius difference increases with a prolongedgrowth time. This growth mechanism does not favor the synthesis of uniformparticles.

When the surface concentration is very high, the surface process will be so fastand polynuclear growth occurs. In polynuclear growth, second layer growth pro-ceeds before the first layer growth is completed and the growth rate of particles doesnot depend on the size of particles or time. Consequently, the growth rate is con-stant and can be defined by a simple differential equation as:

drdt

¼ kp ð3:124Þ

Initial condition:

at t ¼ 0 r ¼ ro ð3:125Þ


where kp is a constant and depends just on temperature. Hence, the particles growlinearly with time:

r ¼ kpt þ ro ð3:126Þ

Differentiation of Eq. (3.126) gives Eq. (3.127) to describe particle radiusdifference:

dr ¼ dro ð3:127Þ

The absolute radius difference remains constant regardless of the growth timeand the absolute particle size. It is worth noting that although the absolute radiusdifference remains unchanged, the relative radius difference would be inverselyproportional to the particle radius and the growth time. As particles get bigger, theradius difference becomes smaller; so this growth mechanism also favors thesynthesis of uniform particles. Figure 3.46 schematically illustrates the radius dif-ference as functions of particle size and growth time for all three mechanisms ofsubsequent growth discussed above. It is clear that a growth mechanism controlledby diffusion is required for homogeneous nucleation and synthesis of uniformparticles (Cao 2004).

Different mechanisms of growth can become predominant when favorablegrowth conditions are established. For example, the rate of chemical reaction is low,the growth species is supplied slowly, and thus, the growth of nuclei would bepredominant by the diffusion-controlled process. Diffusion-limited growth isdesired for the formation of uniform nanostructures; however, relatively largeparticles are generated. There are several ways to achieve diffusion-limited growth:

Fig. 3.46 Comparison of the different mechanism of growth in the growth size of precipitate (Cao2004)


• Keeping the concentration of growth species extremely low.In this case, diffusion distance is very large, and consequently, diffusionbecomes the limiting step.

• Increasing the viscosity of solution.• Introduction of a diffusion barrier such as a monolayer on the surface of a

growing particle.• Controlling supply of growth species.

When growth species is generated through chemical reactions, the rate ofreaction can be manipulated through the control of the concentration of by-product,reactant and catalyst.

3.3.3.3 Ostwald Ripening

Combining individual nanostructures together to form large structures so as toreduce the overall surface area, this includes:

• Sintering: individual structures merge together replacing solid–vapor interfaceby solid–solid interface and becomes polycrystalline.

• Ostwald ripening: A large particle grows at the expense of the smaller one untilthe latter disappears completely and becomes a single crystal. It occurs at rel-atively low temperature.

• Agglomeration of individual nanostructures through chemical bonds andphysical attraction forces at interfaces without altering the individual nano-structures. The smaller the particles, the greater the bonding forces.

Conversion of small particles into larger particles is enhanced by agglomerationof particles to form larger particles, which is the continual growth until equilibriumis reached. The changes in crystal structure that take place over time are often calledaging. A phenomenon called ripening may also take place where by the crystal sizeof the precipitate increases (Hale 2005).

Many small crystals form in a system initially but slowly disappear except for afew that grow larger, at the expense of the small crystals. The smaller crystals act as“nutrients” for the bigger crystals. As the larger crystals grow, the area around themis depleted of smaller crystals. This thermodynamically driven spontaneous processoccurs because larger particles are more energetically favored than smaller particles.This stems from the fact that molecules on the surface of a particle are energeticallyless stable than the ones in the interior. Figure 3.47 shows the effect of Ostwaldripening and growth steps on the particle number and size.

In other words, above some critical radius, the particles form and grow whereasbelow this radius, the particles will re-dissolve. This, however, does not explain thedifferences in the sizes of the particles during growth. There is a size-focusing effectcalled Ostwald ripening. Ostwald ripening was first described in 1900. The mecha-nism of growth is caused by the change in the solubility of NPs dependent on theirsize which is described by the Gibbs–Thomson relation Cr ¼ Cbexpð2cv=rkBTÞð Þ.


Due to the high solubility and the surface energy of smaller particles within solution,the smaller particles re-dissolve and in turn allow the larger particles to grow evenmore. Digestive ripening is effectively the inverse of Ostwald ripening. Within thiscase, smaller particles grow at expense of the larger ones and have been described byan applicable form of the Gibbs–Thomson equation. This process of formation iscontrolled once again by the surface energy of the particle within solution where thelarger particles re-dissolve and in turn smaller particles grow (Voorhees 1985).

An everyday example of Ostwald ripening is the re-crystallization of water withinice creamwhich gives old ice cream a gritty, crunchy texture. Larger ice crystals growat the expense of smaller ones within the ice cream, creating a coarser texture. Inchemistry, Ostwald ripening refers to the growth of larger crystals from those ofsmaller size ones which have a higher solubility than the larger ones. In the process,many small crystals formed initially slowly disappear, except for a few that growlarger, at the expense of the small crystals. The smaller crystals act as fuel for thegrowth of bigger crystals. Limiting Ostwald ripening is fundamental in moderntechnology for the solution synthesis of quantum dots. Ostwald ripening is alsoobserved in liquid–liquid systems. For example, in an oil-in-water emulsion poly-merization, Ostwald ripening causes the diffusion of monomers from smaller to largerdroplets due to greater solubility of the single monomer molecules in the largermonomer droplets. The rate of this diffusion process is linked to the solubility of themonomer in the continuous (water) phase of the emulsion. This can lead to thedestabilization of emulsions (e.g., by creaming and sedimentation). Ostwald ripeningcan also occur in emulsion systems, with molecules diffusing from small droplets tolarge ones through the continuous phase. When a miniemulsion is desired, anextremely hydrophobic compound is added to stop this process from taking place.

The driving force for the ripening process is the well-known curvature depen-dence of the chemical potential (l) which, assuming isotropic surface energy, isgiven by (Voorhees 1985):

l ¼ l0 þ Vmck ð3:128Þ

where k is the mean interfacial curvature, l0 is the chemical potential of an atom ata flat interface, Vm is the molar volume, and γ is the surface energy. From thisequation, it is clear that atoms will flow from regions of high to low curvature. This

Fig. 3.47 Schematic processof Ostwald ripening


results in the disappearance of surfaces possessing high curvature and an increase inthe size scale of dispersed second phase, which is consistent with the necessarydecrease in total energy of the two-phase system.

Exercises

1. A researcher aims to produce a nanoparticels of ZnO through CVD process.What are parameters need to be well considered?

2. Calculate Gibbs free-energy change for carbon nanotube deposition in a CVDreaction, where ethylene is the carbon source.

3. A tube flow reactor is used for CVD synthesis of CNT. The gas flow in thereactor is considered weakly compressible and as a steady state two-dimensionalaxis symmetric flow. Derive the mathematical correlations for transport phe-nomena in this reactor. (The equations involve Navier–Stokes flow, convectionand conduction heat transfer, and the Maxwell-Stefan diffusion and convectionmass transfer.)

4. Develop the FBCVD model presented for the case C2H2 is used as carbonsource.

5. How the model presented for FBCVD synthesis of CNT will be changed if theprocess is considered non-isothermal and non adiabatic?

6. Carbon nanotubes are produced via CVD of benzene using Ferrocene as cat-alyst. Discuss if the surface kinetic controls CNT formation or mass transport.

7. What are the special constraints in selecting variables for production ofnanoparticles through CVD process?

8. What are differences and similarities in nanoparticle synthesis throughco-precipitation and CVS?

9. Comparing growth rate in the diffusion regime and surface growth regime in theprecipitation process, explain which parameters affect the formation of nano-particles uniformly.

10. Discuss about effects of nucleation, growth, coalescence, sintering, and Oswaldripening phenomena on the particle size distribution of the nanoparticles syn-thesized via CVS.

11. What do you think about advantageous and disadvantageous of fluidized bedreactor relative the other gas-catalyst contacting systems (such as fixed-bed,rotary drum, and spray) in synthesis of nanoparticles through CVD.

References

Aghababazadeh, R., Mirhabibi, A. R., Ghanbari, H., Chizari, K., Brydson, R. M., & Brown, A.P. (2006). Synthesis of carbon nanotubes on alumina-based supports with different gas flowrates by CCVD method. Journal of Physics: Conference Series, 26(1), 135.

Ajayan, P. M. (1999). Nanotubes from carbon. Chemical Reviews, 99(7), 1787–1800.


Akhtar, M. K., Lipscomb, G. G., & Pratsinis, S. E. (1994). Monte Carlo simulation of particlecoagulation and sintering. Aerosol Science and Technology, 21(1), 83–93.

Allouche, H., & Monthioux, M. (2005). Chemical vapor deposition of pyrolytic carbon on carbonnanotubes. Part2. Texture and structure. Carbon, 43(6), 1265–1278.

Alavi, S., & Caussat, B. (2005). Experimental study on fluidization of micronic powders. PowderTechnology, 157(1), 114–120.

Alvarez, W. E., Kitiyanan, B., Borgna, A., & Resasco, D. E. (2001). Synergism of Co and Mo inthe catalytic production of single-wall carbon nanotubes by decomposition of CO. Carbon,39(4), 547–558.

Andrews, R., Jacques, D., Qian, D., & Rantell, T. (2002). Multiwall carbon nanotubes: Synthesisand application. Accounts of Chemical Research, 35(12), 1008–1017.

Arena, U., Mastellone, M. L., Camino, G., & Boccaleri, E. (2006). An innovative process for massproduction of multi-wall carbon nanotubes by means of low-cost pyrolysis of polyolefins.Polymer Degradation and Stability, 91(4), 763–768.

Aslam, Z., Li, X., Brydson, R., Rand, B., Falke, U., & Bleloch, A. (2006).Supported catalyticgrowth of SWCNTs using the CVD method. Journal of Physics: Conference Series, 26(1),139–142.

Avdeeva, L. B., Kochubey, D. I., & Shaikhutdinov, S. K. (1999). Cobalt catalysts of methanedecomposition: Accumulation of the filamentous carbon. Applied Catalysis, A: General,177(1), 43–51.

Bachmatiuk, A., Borowiak-Palen, E., & Kalenczuk, R. J. (2008). Advances in engineering ofdiameter and distribution of the number of walls of carbon nanotubes in alcohol CVD.Nanotechnology, 19(36), 365605.

Baker, R. T. K. (1989). Catalytic growth of carbon filaments. Carbon, 27(3), 315–323.Bhattacharya, S. K., & Tummala, R. R. (2001). Integral passives for next generation of electronic

packaging: Application of epoxy/ceramic nanocomposites as integral capacitors.Microelectronics Journal, 32(1), 11–19.

Blanchard, J., Oudghiri-Hassani, H., Abatzoglou, N., Jankhah, S., & Gitzhofer, F. (2008).Synthesis of nanocarbons via ethanol dry reforming over a carbon steel catalyst. ChemicalEngineering Journal, 143(1), 186–194.

Brukh, R., & Mitra, S. (2006). Mechanism of carbon nanotube growth by CVD. Chemical PhysicsLetters, 424(1), 126–132.

Cao, G. (2004). Nanostructures and nanomaterials. Synthesis properties and applications.London: Imperial College Press.

Cassell, A. M., Raymakers, J. A., Kong, J., & Dai, H. (1999). Large scale CVD synthesis ofsingle-walled carbon nanotubes. The Journal of Physical Chemistry B, 103(31), 6484–6492.

Caussat, B., & Vahlas, C. (2007). CVD and powders: A great potential to create new materials.Chemical Vapor Deposition, 13(9), 443–445.

Ciambelli, P., Sannino, D., Sarno, M., Leone, C., & Lafont, U. (2007). Effects of alumina phasesand process parameters on the multiwalled carbon nanotubes growth. Diamond and RelatedMaterials, 16(4), 1144–1149.

Ci, L., Xie, S., Tang, D., Yan, X., Li, Y., Liu, Z., & Wang, G. (2001). Controllable growth ofsingle wall carbon nanotubes by pyrolizing acetylene on the floating iron catalysts. ChemicalPhysics Letters, 349(3), 191–195.

Corrias, M., Caussat, B., Ayral, A., Durand, J., Kihn, Y., Kalck, P., & Serp, P. (2003). Carbonnanotubes produced by fluidized bed catalytic CVD: First approach of the process. ChemicalEngineering Science, 58(19), 4475–4482.

Corrias, M., Kihn, Y., Kalck, P., & Serp, P. (2005). CVD from ethylene on cobalt ferrite catalysts:The effect of the support. Carbon, 43(13), 2820–2823.

Dai, H. (2001). Nanotube growth and characterization. In Carbon Nanotubes (pp. 29–53). Berlin:Springer.

Dai, H., Rinzler, A. G., Nikolaev, P., Thess, A., Colbert, D. T., & Smalley, R. E. (1996).Single-wall nanotubes produced by metal-catalyzed disproportionation of carbon monoxide.Chemical Physics Letters, 260(3), 471–475.

References 265

Danafar, F., Fakhru’l-Razi, A., Salleh, M. A. M., & Biak, D. R. A. (2009). Fluidized bed catalyticchemical vapor deposition synthesis of carbon nanotubes—A review. Chemical EngineeringJournal, 155(1), 37–48.

Danafar, F., Fakhru’l-Razi, A., Salleh, M. A. M., & Biak, D. R. A. (2011). Influence of catalyticparticle size on the performance of fluidized-bed chemical vapor deposition synthesis of carbonnanotubes. Chemical Engineering Research and Design, 89(2), 214–223.

Davidson, J. F., & Harrison, D. (1963). Fluidised particles. Cambridge, UK: CambridgeUniversity Press.

Dupuis, A. C. (2005). The catalyst in the CCVD of carbon nanotubes—A review. Progress inMaterials Science, 50(8), 929–961.

Elnashaie, S. S., Affane, C., & Uhlig, F. (2007). Numerical techniques for chemical and biologicalengineers using MATLAB®: A simple bifurcation approach. Berlin: Springer.

Ergun, S. (1952). Fluid flow through packed columns. Chemical Engineering Progress, 48, 89–94.Ermakova, M. A., Ermakov, D. Y., Chuvilin, A. L., & Kuvshinov, G. G. (2001). Decomposition

of methane over iron catalysts at the range of moderate temperatures: The influence of structureof the catalytic systems and the reaction conditions on the yield of carbon and morphology ofcarbon filaments. Journal of Catalysis, 201(2), 183–197.

Fakhru’l‐Razi, A., Danafar, F., Dayang Radiah, A. B., &Mohd Salleh, M. A. (2009). An innovativeprocedure for large‐scale synthesis of carbon nanotubes by fluidized bed catalytic vapordeposition technique. Fullerenes, Nanotubes and Carbon Nanostructures, 17(6), 652–663.

Fan, S., Chapline, M. G., Franklin, N. R., Tombler, T. W., Cassell, A. M., & Dai, H. (1999).Self-oriented regular arrays of carbon nanotubes and their field emission properties. Science,283(5401), 512–514.

García-García, F. R., Pérez-Cabero, M., Nevskaia, D. M., Rodríguez-Ramos, I., & Guerrero-Ruiz,A. (2008). Improving the synthesis of high purity carbon nanotubes in a catalytic fluidized bedreactor and their comparative test for hydrogen adsorption capacity. Catalysis Today, 133,815–821.

Geldart, D. (1973). Types of gas fluidization. Powder Technology, 7(5), 285–292.Grace, J. R. (1986). Modelling and simulation of two-phase fluidized bed reactors. In Chemical

reactor design and technology (pp. 245–289). Netherlands: Springer.Gupta, C. K., & Sathiyamoorthy, D. (1998). Fluid bed technology in materials processing.

New York: CRC Press.Gutzow, I., Pascova, R., Karamanov, A., & Schmelzer, J. (1998). The kinetics of surface induced

sinter crystallization and the formation of glass-ceramic materials. Journal of MaterialsScience, 33(21), 5265–5273.

Haas, V., Birringer, R., Gleiter, H., & Pratsinis, S. E. (1997). Synthesis of nanostructured powdersin an aerosol flow condenser. Journal of aerosol science, 28(8), 1443–1453.

Habuka, H., Ohmori, H., & Ando, Y. (2010). Silicon carbide film deposition at low temperaturesusing monomethylsilane gas. Surface and Coatings Technology, 204(9), 1432–1437.

Hagen, J. (2006). Industrial catalysis: A practical approach (2nd ed.). Germany: WILEY-VCHVerlag GmbH & Co.

Hahn, H. (1997). Gas phase synthesis of nanocrystalline materials. Nanostructured Materials,9(1), 3–12.

Hale P. S., Maddox L. M., Shapter J. G., Voelcker N. H., Ford M. J. & Waclawik E. R. (2005).Growth kinetics and modeling of ZnO nanoparticles. Journal of Chemical Education, 82(5).

Hao, Y., Qunfeng, Z., Fei, W., Weizhong, Q., & Guohua, L. (2003). Agglomerated CNTssynthesized in a fluidized bed reactor: Agglomerate structure and formation mechanism.Carbon, 41(14), 2855–2863.

Height, M. J., Howard, J. B., Tester, J. W., & Vander Sande, J. B. (2005). Carbon nanotubeformation and growth via particle-particle interaction. The Journal of Physical Chemistry B,109(25), 12337–12346.

Heine, M. C., & Pratsinis, S. E. (2007). Agglomerate TiO2 aerosol dynamics at highconcentrations. Particle & Particle Systems Characterization, 24(1), 56–65.


Hernadi, K. (2002). Catalytic synthesis of multiwall carbon nanotubes from methylacetylene.Chemical Physics Letters, 363(1), 169–174.

Herrera, J. E., Balzano, L., Borgna, A., Alvarez, W. E., & Resasco, D. E. (2001). Relationshipbetween the structure/composition of Co–Mo catalysts and their ability to producesingle-walled carbon nanotubes by CO disproportionation. Journal of Catalysis, 204(1),129–145.

Hess, D. W., & Graves, D. B. (1989). Plasma-enhanced etching and deposition. Microelectronicsprocessing–a chemical engineering aspects. American Chemical Society, 382.

Jones, A. C., & Hitchman, M. L. (Eds.). (2009). Chemical vapour deposition: Precursors,processes and applications. Royal Society of Chemistry.

Jung, J., & Gidaspow, D. (2002). Fluidization of nano-size particles. Journal of NanoparticleResearch, 4(6), 483–497.

Kathyayini, H., Willems, I., Fonseca, A., Nagy, J. B., & Nagaraju, N. (2006). Catalytic materialsbased on aluminium hydroxide, for the large scale production of bundles of multi-walled(MWNT) carbon nanotubes. Catalysis Communications, 7(3), 140–147.

Kim, J., & Grate, J. W. (2003). Single-enzyme nanoparticles armoured by a nanometer-scaleorganic/inorganic network. Nano Letters, 3, 1219–1222.

Kim, K. E., Kim, K. J., Jung, W. S., Bae, S. Y., Park, J., Choi, J., & Choo, J. (2005). Investigationon the temperature-dependent growth rate of carbon nanotubes using chemical vapordeposition of ferrocene and acetylene. Chemical Physics Letters, 401(4), 459–464.

Kitiyanan, B., Alvarez, W. E., Harwell, J. H., & Resasco, D. E. (2000). Controlled production ofsingle-wall carbon nanotubes by catalytic decomposition of CO on bimetallic Co–Mo catalysts.Chemical Physics Letters, 317(3), 497–503.

Klinke, C., Bonard, J. M., & Kern, K. (2001). Comparative study of the catalytic growth ofpatterned carbon nanotube films. Surface Science, 492(1), 195–201.

Koch, W., & Friedlander, S. K. (1990). The effect of particle coalescence on the surface area of acoagulating aerosol. Journal of Colloid and Interface Science, 140(2), 419–427.

Koetz, J., & Kosmella, S. (2007). Polyelectrolytes and nanoparticles. Berlin: Springer.Kong, J., Cassell, A. M., & Dai, H. (1998). Chemical vapor deposition of methane for

single-walled carbon nanotubes. Chemical Physics Letters, 292(4), 567–574.Kouravelou, K. B., & Sotirchos, S. V. (2005). Dynamic study of carbon nanotubes production by

chemical vapor deposition of alcohol. Advanced Material Science, 10, 243–248.Kruis, F. E., Fissan, H., & Peled, A. (1998). Synthesis of nanoparticles in the gas phase for

electronic, optical and magnetic applications—A review. Journal of Aerosol Science, 29(5),511–535.

Kruis, F. E., Kusters, K. A., Pratsinis, S. E., & Scarlett, B. (1993). A simple model for theevolution of the characteristics of aggregate particles undergoing coagulation and sintering.Aerosol Science and Technology, 19(4), 514–526.

Kukovecz, A., Konya, Z., Nagaraju, N., Willems, I., Tamasi, A., Fonseca, A., & Kiricsi, I. (2000).Catalytic synthesis of carbon nanotubes over Co, Fe and Ni containing conventional and sol–gel silica–aluminas. Physical Chemistry Chemical Physics, 2(13), 3071–3076.

Kukovecz, Á., Smajda, R., Kónya, Z., & Kiricsi, I. (2007). Controlling the pore diameterdistribution of multi-wall carbon nanotube buckypapers. Carbon, 45(8), 1696–1698.

Kukovitsky, E. F., L’vov, S. G., Sainov, N. A., Shustov, V. A., & Chernozatonskii, L. A. (2002).Correlation between metal catalyst particle size and carbon nanotube growth. Chemical PhysicsLetters, 355(5), 497–503.

Kuni, F. M., Shchekin, A. K., & Grinin, A. P. (2001). Theory of heterogeneous nucleation forvapor undergoing a gradual metastable state formation. Physics-Uspekhi, 44(4), 331–370.

Kunii, D., & Levenspiel, O. (1991). Fluidization engineering. Boston: Butterworth.Kwok, C., Reizman, B. J., Agnew, D. E., Sandhu, G. S., Weistroffer, J., Strano, M. S., & Seebauer,

E. G. (2010). Temperature and time dependence study of single-walled carbon nanotubegrowth by catalytic chemical vapor deposition. Carbon, 48(4), 1279–1288.

Kwon, S. G., & Hyeon, T. (2011). Formation mechanisms of uniform nanocrystals viahot-injection and heat-up methods. Small, 7(19), 2685–2702.

References 267

Lamouroux, E., Serp, P., & Kalck, P. (2007). Catalytic routes towards single wall carbonnanotubes. Catalysis Reviews, 49(3), 341–405.

Lee, T. Y., Han, J. H., Choi, S. H., Yoo, J. B., Park, C. Y., Jung, T., & Kim, J. M. (2003).Comparison of source gases and catalyst metals for growth of carbon nanotube. Surface &Coatings Technology, 169, 348–352.

Leskelä, M., & Ritala, M. (2002). Atomic layer deposition (ALD): From precursors to thin filmstructures. Thin Solid Films, 409(1), 138–146.

Li, Q., Yan, H., Zhang, J., & Liu, Z. (2004a).Effect of hydrocarbons precursors on the formation ofcarbon nanotubes in chemical vapor deposition. Carbon, 42(4), 829–835.

Li, Y. L., Kinloch, I. A., Shaffer, M. S., Geng, J., Johnson, B., & Windle, A. H. (2004b). Synthesisof single-walled carbon nanotubes by a fluidized-bed method. Chemical physics letters, 384(1),98–102.

Liao, X. Z., Serquis, A., Jia, Q. X., Peterson, D. E., Zhu, Y. T., & Xu, H. F. (2003). Effect ofcatalyst composition on carbon nanotube growth. Applied Physics Letters, 82(16), 2694–2696.

Little, R. B. (2003). Mechanistic aspects of carbon nanotube nucleation and growth. Journal ofCluster Science, 14(2), 135–185.

Liu, Q., & Fang, Y. (2006). New technique of synthesizing single-walled carbon nanotubes fromethanol using fluidized-bed over Fe–Mo/MgO catalyst. Spectrochimica Acta Part A: Molecularand Biomolecular Spectroscopy, 64(2), 296–300.

Liu, Q., Ouyang, Y., Zhang, L., Xu, Y., & Fang, Y. (2009). Effects of argon flow rate and reactiontemperature on synthesizing single-walled carbon nanotubes from ethanol. Physica E:Low-dimensional Systems and Nanostructures, 41(7), 1204–1209.

Liu, X., Sun, H., Chen, Y., Lau, R., & Yang, Y. (2008). Preparation of large particle MCM-41 andinvestigation on its fluidization behavior and application in single-walled carbon nanotubeproduction in a fluidized-bed reactor. Chemical Engineering Journal, 142(3), 331–336.

Lorke, A., Winterer, M., Schmechel, R., & Schulz, C. (2012). Nanoparticles from the gasphase:Formation, structure, properties. Properties: Springer.

Louis, B., Gulino, G., Vieira, R., Amadou, J., Dintzer, T., Galvagno, S., & Pham-Huu, C. (2005).High yield synthesis of multi-walled carbon nanotubes by catalytic decomposition of ethaneover iron supported on alumina catalyst. Catalysis Today, 102, 23–28.

Lu, W. Z., Teng, L. H., & Xiao, W. D. (2004). Simulation and experiment study of dimethyl ethersynthesis from syngas in a fluidized-bed reactor. Chemical Engineering Science, 59(22),5455–5464.

Luo, G., Li, Z., Wei, F., Xiang, L., Deng, X., & Jin, Y. (2002). Catalysts effect on morphology ofcarbon nanotubes prepared by catalytic chemical vapor deposition in a nano-agglomerate bed.Physica B: Condensed Matter, 323(1), 314–317.

MacKenzie, K. J., Dunens, O. M., & Harris, A. T. (2010). An updated review of synthesisparameters and growth mechanisms for carbon nanotubes in fluidized beds. Industrial andEngineering Chemistry Research, 49(11), 5323–5338.

Mädler, L., & Friedlander, S. K. (2007). Transport of nanoparticles in gases: Overview and recentadvances. Aerosol and Air Quality Research, 7, 304–342.

Malgas, G. F., Arendse, C. J., Cele, N. P., & Cummings, F. R. (2008). Effect of mixture ratios andnitrogen carrier gas flow rates on the morphology of carbon nanotube structures grown byCVD. Journal of Materials Science, 43(3), 1020–1025.

MARLAP (Multi-Agency Radiological Laboratory Analytical Protocols Manual). (2004). U.S.Nuclear Regulatory Commission, Washington, D.C.

Maruyama, S., Kojima, R., Miyauchi, Y., Chiashi, S., & Kohno, M. (2002). Low-temperaturesynthesis of high-purity single-walled carbon nanotubes from alcohol. Chemical PhysicsLetters, 360(3), 229–234.

McCaulley, J. A., McCrary, V. R., & Donnelly, V. M. (1989). Laser-induced decomposition oftriethylgallium and trimethylgallium adsorbed on gallium arsenide (100). The Journal ofPhysical Chemistry, 93(3), 1148–1158.


Melechko, A. V., Merkulov, V. I., McKnight, T. E., Guillorn, M. A., Klein, K. L., Lowndes, D.H., & Simpson, M. L. (2005). Vertically aligned carbon nanofibers and related structures:Controlled synthesis and directed assembly. Journal of Applied Physics, 97(4), 041301.

Mi, W., Lin, J. Y., Mao, Q., Li, Y., & Zhang, B. (2005). A study on the effects of carrier gases onthe structure and morphology of carbon nanotubes prepared by pyrolysis of ferrocene andC*2H*2 Mixture. Journal of Natural Gas Chemistry, 14(3), 151–155.

Mizuno, K., Hata, K., Saito, T., Ohshima, S., Yumura, M., & Iijima, S. (2005). Selective matchingof catalyst element and carbon source in single-walled carbon nanotube synthesis on siliconsubstrates. The Journal of Physical Chemistry B, 109(7), 2632–2637.

Moisala, A., Nasibulin, A. G., & Kauppinen, E. I. (2003). The role of metal nanoparticles in thecatalytic production of single-walled carbon nanotubes-a review. Journal of Physics:Condensed Matter, 15(42), S3011.

Mora, E., Tokune, T., & Harutyunyan, A. R. (2007). Continuous production of single-walledcarbon nanotubes using a supported floating catalyst. Carbon, 45(5), 971–977.

Morançais, A., Caussat, B., Kihn, Y., Kalck, P., Plee, D., Gaillard, P., et al. (2007).A parametricstudy of the large scale production of multi-walled carbon nanotubes by fluidized bed catalyticchemical vapor deposition. Carbon, 45(3), 624–635.

Mori, S., & Wen, C. Y. (1975). Estimation of bubble diameter in gaseous fluidized beds. AIChEJournal, 21(1), 109–115.

Muataz, A. A., Ahmadun, F., Guan, C., Mahdi, E., & Rinaldi, A. (2006). Effect of reactiontemperature on the production of carbon nanotubes. NANO, 1(03), 251–257.

Nagaraju, N., Fonseca, A., Konya, Z., & Nagy, J. B. (2002a). Alumina and silica supported metalcatalysts for the production of carbon nanotubes. Journal of Molecular Catalysis A: Chemical,181(1), 57–62.

Nagaraju, N., Fonseca, A., Konya, Z., & Nagy, J. B. (2002b). Alumina and silica supported metalcatalysts for the production of carbon nanotubes. Journal of Molecular Catalysis A: Chemical,181(1), 57–62.

Nakaso, K., Okuyama, K., Shimada, M., & Pratsinis, S. E. (2003). Effect of reaction temperature onCVD-made TiO2 primary particle diameter.Chemical Engineering Science, 58(15), 3327–3335.

Nakaso, K., Shimada, M., Okuyama, K., & Deppert, K. (2002). Evaluation of the change in themorphology of gold nanoparticles during sintering. Journal of Aerosol Science, 33(7),1061–1074.

Nourbakhsh, A., Ganjipour, B., Zahedifar, M., & Arzi, E. (2007). Morphology optimization ofCCVD-synthesized multiwall carbon nanotubes, using statistical design of experiments.Nanotechnology, 18(11), 115715.

Öncel, Ç., & Yürüm, Y. (2006). Carbon nanotube synthesis via the catalytic CVD method: Areview on the effect of reaction parameters. Fullerenes, Nanotubes, and Carbon Nonstructures,14(1), 17–37.

Ortega-Cervantez, G., Rueda-Morales, G., & Ortiz-Lopez, J. (2005). Catalytic CVD production ofcarbon nanotubes using ethanol. Microelectronics Journal, 36(3), 495–498.

Ouyang, Y., Chen, L., Liu, Q. X., & Fang, Y. (2008). A temperature window for the synthesis ofsingle-walled carbon nanotubes by catalytic chemical vapor deposition of CH4 overMo-Fe/MgO catalyst. Spectrochimica Acta Part A: Molecular and BiomolecularSpectroscopy, 71(2), 317–320.

Panda, S., & Pratsinis, S. E. (1995). Modeling the synthesis of aluminum particles byevaporation-condensation in an aerosol flow reactor. Nanostructured Materials, 5(7), 755–767.

Park, J. H., & Sudarshan, T. S. (Eds.). (2001). Chemical vapor deposition (Vol. 2). ASMInternational.

Peigney, A., Coquay, P., Flahaut, E., Vandenberghe, R. E., De Grave, E., & Laurent, C. (2001).A study of the formation of single-and double-walled carbon nanotubes by a CVD method. TheJournal of Physical Chemistry B, 105(40), 9699–9710.

Pérez-Cabero, M., Monzón, A., Rodrıguez-Ramos, I., & Guerrero-Ruı́z, A. (2004). Syntheses ofCNTs over several iron-supported catalysts: Influence of the metallic precursors. CatalysisToday, 93, 681–687.

References 269

Philippe, R., Morançais, A., Corrias, M., Caussat, B., Kihn, Y., Kalck, P., & Serp, P. (2007).Catalytic production of carbon nanotubes by fluidized-bed CVD. Chemical Vapor Deposition,13(9), 447–457.

Piedigrosso, P., Konya, Z., Colomer, J. F., Fonseca, A., Van Tendeloo, G., & Nagy, J. B. (2000).Production of differently shaped multi-wall carbon nanotubes using various cobalt supportedcatalysts. Physical Chemistry Chemical Physics, 2(1), 163–170.

Pierson, H. O. (1999). Handbook of chemical vapor deposition: Principles, technology andapplications. London: William Andrew.

Pratsinis, S. E., & Spicer, P. T. (1998). Competition between gas phase and surface oxidation ofTiCl4 during synthesis of TiO2 particles. Chemical Engineering Science, 53(10), 1861–1868.

Qian, W., Liu, T., Wei, F., Wang, Z., & Li, Y. (2004). Enhanced production of carbon nanotubes:Combination of catalyst reduction and methane decomposition. Applied Catalysis, A: General,258(1), 121–124.

Qian, W., Yu, H., Wei, F., Zhang, Q., & Wang, Z. (2002). Synthesis of carbon nanotubes fromliquefied petroleum gas containing sulfur. Carbon, 40(15), 2968–2970.

Qiu, J., An, Y., Zhao, Z., Li, Y., & Zhou, Y. (2004). Catalytic synthesis of single-walled carbonnanotubes from coalgas by chemical vapor deposition method. Fuel Processing Technology,85(8), 913–920.

Qiu, J., Li, Q., Wang, Z., Sun, Y., & Zhang, H. (2006). CVD synthesis of coal-gas-derived carbonnanotubes and nanocapsules containing magnetic iron carbide and oxide. Carbon, 44(12),2565–2568.

Quingwen, L., Hao, Y., Jin, Z. & Zhongfan, L. (2004). Effect of hydrocarbons precursors on theformation of carbon nanotubes in chemical vapor deposition. Carbon, 42, 829–835

Rao, C. N. R., Müller, A., & Cheetham, A. K. (Eds.). (2007). Nanomaterials chemistry: Recentdevelopments and new directions. New York: Wiley.

Reshetenko, T. V., Avdeeva, L. B., Khassin, A. A., Kustova, G. N., Ushakov, V. A., Moroz, E.M., et al. (2004). Coprecipitated iron-containing catalysts (Fe-Al2O3, Fe-Co-Al2O3,Fe-Ni-Al2O3) for methane decomposition at moderate temperatures I. Genesis of calcinedand reduced catalysts. Applied Catalysis A: General, 268(1–2), 127–138.

Sander, M., West, R. H., Celnik, M. S., & Kraft, M. (2009). A detailed model for the sintering ofpolydispersed nanoparticle agglomerates. Aerosol Science and Technology, 43(10), 978–989.

Saxena, S. C., & Vadivel, R. (1988). Heat transfer from a tube bundle in a bubble column.International Communications in Heat and Mass Transfer, 15(5), 657–667.

Saxena, S. C., Vadivel, R., & Saxena, A. C. (1989). Hydrodynamic and heat transfercharacteristics of bubble columns involving fine powders. Powder Technology, 59(1), 25–35.

Sear, R. P. (2007). Nucleation: Theory and applications to protein solutions and colloidalsuspensions. Journal of Physics: Condensed Matter, 19(3), 033101.

See, C. H., & Harris, A. T. (2008). CaCo3 supported Co-Fe catalysts for carbon nanotubesynthesis in fluidized bed reactors. AIChE Journal, 54, 657–664.

Serquis, A., Liao, X. Z., Huang, J. Y., Jia, Q. X., Peterson, D. E., & Zhu, Y. T. (2003). Co–Mocatalyzed growth of multi-wall carbon nanotubes from CO decomposition. Carbon, 41(13),2635–2641.

Sinnott, S. B., Andrews, R., Qian, D., Rao, A. M., Mao, Z., Dickey, E. C., & Derbyshire, F.(1999). Model of carbon nanotube growth through chemical vapor deposition. ChemicalPhysics Letters, 315(1), 25–30.

Smajda, R., Kukovecz, Á., Kónya, Z., & Kiricsi, I. (2007). Structure and gas permeability ofmulti-wall carbon nanotube buckypapers. Carbon, 45(6), 1176–1184.

Smith, D. K., Lee, D. C., & Korgel, B. A. (2006). High yield multiwall carbon nanotube synthesisin supercritical fluids. Chemistry of Materials, 18(14), 3356–3364.

Son, S., Lee, D. H., Kim, S. D., & Sung, S. W. (2007). Effect of inert particles on the synthesis ofcarbon nanotubes in a gas-solid fluidized bed reactor. Journal of Industrial and EngineeringChemistry, 13(2), 257–264.


Son, S. Y., Lee, Y., Won, S., Lee, D. H., Kim, S. D., & Sung, S. W. (2008). High-qualitymultiwalled carbon nanotubes from catalytic decomposition of carboneous materials ingas-solid fluidized beds. Industrial and Engineering Chemistry Research, 47(7), 2166–2175.

Su, M., Zheng, B., & Liu, J. (2000). A scalable CVD method for the synthesis of single-walledcarbon nanotubes with high catalyst productivity. Chemical Physics Letters, 322(5), 321–326.

Swihart, M. T. (2003). Vapor-phase synthesis of nanoparticles. Current Opinion in Colloid &Interface Science, 8(1), 127–133.

Teo, K. B., Singh, C., Chhowalla, M., & Milne, W. I. (2003).Catalytic synthesis of carbonnanotubes and nanofibers. Encyclopedia of Nanoscience and Nanotechnology, 10(1).

Thanh, N. T., Maclean, N., & Mahiddine, S. (2014). Mechanisms of nucleation and growth ofnanoparticles in solution. Chemical Reviews, 114(15), 7610–7630.

Tran, K. Y., Heinrichs, B., Colomer, J. F., Pirard, J. P., & Lambert, S. (2007). Carbon nanotubessynthesis by the ethylene chemical catalytic vapor deposition (CCVD) process on Fe Co, andFe–Co/Al2O3 sol–gel catalysts. Applied Catalysis, A: General, 318, 63–69.

Tsantilis, S., & Pratsinis, S. E. (2004). Narrowing the size distribution of aerosol-made titania bysurface growth and coagulation. Journal of aerosol science, 35(3), 405–420.

Tsantilis, S., Pratsinis, S. E., & Haas, V. (1999). Simulation of synthesis of palladiumnanoparticles in a jet aerosol flow condenser. Journal of Aerosol Science, 30(6), 785–803.

Vahlas, C., Caussat, B., Serp, P., & Angelopoulos, G. N. (2006). Principles and applications ofCVD powder technology. Materials Science and Engineering: R: Reports, 53(1), 1–72.

Valverde, J. M., & Castellanos, A. (2008). Fluidization of nanoparticles: A simple equation forestimating the size of agglomerates. Chemical Engineering Journal, 140(1), 296–304.

Van Steen, E., & Prinsloo, F. F. (2002). Comparison of preparation methods for carbon nanotubessupported iron Fischer-Tropsch catalysts. Catalysis Today, 71(3), 327–334.

Vander Wal, R. L., Ticich, T. M., & Curtis, V. E. (2001). Substrate–support interactions inmetal-catalyzed carbon nanofiber growth. Carbon, 39(15), 2277–2289.

Vekilov, P. G. (2010). Nucleation. Crystal Growth & Design, 10(12), 5007–5019.Venegoni, D., Serp, P., Feurer, R., Kihn, Y., Vahlas, C., & Kalck, P. (2002). Parametric study for

the growth of carbon nanotubes by catalytic chemical vapor deposition in a fluidized bedreactor. Carbon, 40(10), 1799–1807.

Vinciguerra, V., Buonocore, F., Panzera, G., & Occhipinti, L. (2003). Growth mechanisms inchemical vapour deposited carbon nanotubes. Nanotechnology, 14(6), 655–660.

Voorhees, P. W. (1985). The theory of Ostwald ripening. Journal of Statistical Physics, 38(1–2),231–252.

Wagner, R. S., & Ellis, W. C. (1964). Vapor-liquid-solid mechanism of single crystal growth.Applied Physics Letters, 4(5), 89–90.

Walas, S. M. (1990).Chemical process equipment selection and design.(2nd Ed.).New York:Butterworth-Heinemann Reed Publishing.

Wang, Y., Wei, F., Luo, G., Yu, H., & Gu, G. (2002). The large-scale production of carbonnanotubes in a nano-agglomerate fluidized-bed reactor. Chemical Physics Letters, 364(5),568–572.

Wei, Y. Y., Eres, G., Merkulov, V. I., & Lowndes, D. H. (2001). Effect of catalyst film thicknesson carbon nanotube growth by selective area chemical vapor deposition. Applied PhysicsLetters, 78(10), 1394–1396.

Weis, F., Schneider, R., Seipenbusch, M., & Kasper, G. (2013). Synthesis of Bi2O3/SiO2 core–shell nanoparticles by an atmospheric CVS/CVD process and their modification by hydrogenor electron-beam induced reduction. Surface and Coatings Technology, 230, 93–100.

Wen, C. Y., & Yu, Y. H. (1966). A generalized method for predicting the minimum fluidizationvelocity. AIChE Journal, 12(3), 610–612.

Wen, T., Brush, L. N., & Krishnan, K. M. (2014). A generalized diffusion model for growth ofnanoparticles synthesized by colloidal methods. Journal of Colloid and Interface Science, 419,79–85.

References 271

Xia, W., Chen, X., Kundu, S., Wang, X., Grundmeier, G., Wang, Y., & Muhler, M. (2007).Chemical vapor synthesis of secondary carbon nanotubes catalyzed by iron nanoparticleselectrodeposited on primary carbon nanotubes. Surface & Coatings Technology, 201(22),9232–9237.

Xiong, G. Y., Suda, Y., Wang, D. Z., Huang, J. Y., & Ren, Z. F. (2005). Effect of temperature,pressure, and gas ratio of methane to hydrogen on the synthesis of double-walled carbonnanotubes by chemical vapour deposition. Nanotechnology, 16(4), 532–535.

Xu, C., & Zhu, J. (2004). One-step preparation of highly dispersed metal-supported catalysts byfluidized-bed MOCVD for carbon nanotube synthesis. Nanotechnology, 15(11), 1671–1681.

Yamada, T., Namai, T., Hata, K., Futaba, D. N., Mizuno, K., Fan, J., & Iijima, S. (2006).Size-selective growth of double-walled carbon nanotube forests from engineered iron catalysts.Nature Nanotechnology, 1(2), 131–136.

Yao, Y., Falk, L. K. L., Morjan, R. E., Nerushev, O. A., & Campbell, E. E. B. (2004). Synthesis ofcarbon nanotube films by thermal CVD in presence of supported catalyst particle. Part II: Thenanotube film. Journal of Materials Science: Materials in Electronics, 15(9), 583–594.

Yen, Y. W., Huang, M. D., & Lin, F. J. (2008). Synthesize carbon nanotubes by a novel methodusing chemical vapor deposition-fluidized bed reactor from solid-stated polymers. Diamondand Related Materials, 17(4), 567–570.

Yu, G., Gong, J., Wang, S., Zhu, D., He, S., & Zhu, Z. (2006). Etching effects of ethanol onmulti-walled carbon nanotubes. Carbon, 44(7), 1218–1224.

Yudasaka, M., Kikuchi, R., Ohki, Y., Ota, E., & Yoshimura, S. (1997). Behavior of Ni in carbonnanotube nucleation. Applied Physics Letters, 70(14), 1817–1818.

Zhao, J., Martinez-Limia, A., & Balbuena, P. B. (2005). Understanding catalysed growth ofsingle-wall carbon nanotubes. Nanotechnology, 16(7), S575–S581.

Zheng, L., Liao, X., & Zhu, Y. T. (2006). Parametric study of carbon nanotube growth viacobalt-catalyzed ethanol decomposition. Materials Letters, 60(16), 1968–1972.


Chapter 4Conclusions and Outlook

This is a relatively early book addressing the modern field of nanotechnology froma chemical engineering point of view. It tries to follow the route adopted in the lastfew decades from chemical technology to chemical engineering. This is takingspecifically the descriptive principle of technology to the quantitative principles ofengineering. This is achieved using mainly material and energy balances coupled toquantitative rates of processes such as rates of reactions: mass transfer, heat transfer,etc., in terms of state variables, e.g., concentrations, temperatures.

An integrated system approach (ISA) based on system theory (ST) is used whichis the best approach organizing the optimal route for the design, analysis andresearch of such complicated, sensitive and high-quality systems. Mathematicalmodeling coupled to experimental results is coupled to address these relativelycomplex heterogeneous systems. Special examples are addressed such as the pro-duction of carbon nanotubes (CNTs) in catalytic bubbling fluidized beds forchemical vapor deposition (CVD).

This book should be useful for chemical engineers wanting to get into the fieldof nanotechnology using chemical engineering principles and to develop it intonanoengineering. It can also be useful for nanotechnologists wanting to developinto nanoengineers and learning chemical engineering principles to do that. Itshould also be useful to support multi-disciplinary (MD) research and applicationsin this important, critical and vital modern field.

Last but not least, this book should be useful for expanding and completing thisprocess of transforming nanotechnology to nanoengineering and assisting in thedevelopment of more advanced and comprehensive work in this crucial field,coupling in an optimal manner, mathematical modeling and experimental explo-ration and verification of the models to be reliable design and research tools.


273

About the Authors

Prof. Dr. Said Salaheldeen Elnashaie

Born 1947, Cairo/Egypt; grew up in Egypt; married, two children, sixgrandchildren.

1963–1968: Chemical Engineering undergraduate student, Cairo University, Egypt1968–1969: Teaching Assistant, Chemical Engineering Department, Cairo

University, Egypt1969–1970: Master student, Chemical Engineering Department, University of

Waterloo, Ontario, Canada1970–1973: PhD student, Chemical Engineering Department, University of

Edinburgh, UK1973–1974: Postdoctorate fellow, Chemical Engineering Departments, Universities

of McGill and Toronto, Canada1974–1979: Assistant Professor, Chemical Engineering Department, Cairo

University, Egypt1979–1984: Associate Professor, Chemical Engineering Department, Cairo

University, Egypt1974–1993: Full Professor, Chemical Engineering Department, Cairo University,

Egypt

© Springer Science+Business Media Singapore 2015S. Salaheldeen Elnashaie et al., Nanotechnology for Chemical Engineers,DOI 10.1007/978-981-287-496-2

275

1986–1996: Full Professor, Chemical Engineering Department, King SaudUniversity (KSU), Riyadh, Kingdom of Saudi Arabia

1996–1999: Vice President, Environmental Energy Systems and Services (EESS),Egypt

1999–2005: Full Professor, Chemical Engineering Department, Auburn University,Alabama, USA

2005–2006: Full Professor, Chemical and Biological Engineering Department(CBED), University of British Columbia (UBC), Vancouver

2006–2009: Quentin Berg Chair Professor of Sustainable DevelopmentEngineering (SDE), Pennsylvania State University, Harrisburg, USA

2009–2010: Dean of Engineering & IT, Sinai University, Egypt2006–date: Adjunct Professor, Chemical and Biological Engineering Department

(CBED), University of British Columbia (UBC), Vancouver2012–date: Full Professor, Chemical and Environmental Engineering Department,

University Putra Malaysia (UPM), Serdang, MalaysiaResearch Areas: Modeling, simulation and optimization of chemical and bio-

chemical processes, biofuels and integrated bio-refineries (IBRs); nonlineardynamics, bifurcation and chaos in chemical and biochemical engineering;nanotechnology; fixed and fluidized bed catalytic reactors, sustainable devel-opment engineering; hydrogen clean energy

Publications: Papers in International Journals and Conferences: 400; Books: 4;Chapters: 3; Patents: 3

Dr Firoozeh Danafar

Born 1976, Kerman/Iran; grew up in Kerman/Iran; married, one child.1995–1999: Chemical Engineering undergraduate student, Amirkabir University of

Technology (Tehran Poly-Technique), Tehran, Iran2000–2003: Master student, Chemical Engineering Department, Sharif University

of Technology, Tehran, Iran

276 About the Authors

2006–2011: PhD student, Chemical Engineering Department, University PutraMalaysia, Malaysia

2011–2013: Postdoctorate fellow, Chemical Engineering Department, UniversityPutra Malaysia, Malaysia

2013–date: Assistant Professor, Chemical Engineering Department, ShahidBahonar University of Kerman, Kerman

Research Areas: Nanostructures production and their applications in bio- and foodtechnologies; Sustainable Development Engineering

Publications: More than 10 papers in International Journals and Conferences and 1Patent

Dr. Hassan Hashemipour Rafsanjani

Born 1971, Rafsanjan/Iran; grew up in Iran; married, two children.1989–1993: Chemical Engineering undergraduate student, Tehran University, Iran1993–1995: Master student, Chemical Engineering Department, Amirkabir

University of Technology (Tehran Poly-Technique), Tehran, Iran1995–2002: PhD student, Chemical Engineering Department, Amirkabir

University of Technology (Tehran Poly-Technique), Tehran, Iran1999–2000: Research Opportunity, University of Illinois at Urbana-Champaign,

USA2002–2010: Assistant Professor, Chemical Engineering Department, Shahid

Bahonar University of Kerman, Kerman, Iran2010–date: Associate Professor, Chemical Engineering Department, Shahid

Bahonar University of Kerman, Kerman, Iran2007–2009: Head of Production and Separation Research Group, Mineral

Industries Research Center, Shahid Bahonar University of Kerman, Kerman,Iran

2010–2014: Head of Department, Chemical Engineering Department, ShahidBahonar University of Kerman, Kerman, Iran

About the Authors 277

Research Areas: Modeling, simulation and experimental investigations in theheterogeneous systems including adsorption, catalytic and non-catalytic fluid–solid reaction, and nanotechnology focused on the nanoparticles synthesis in thegas and/or liquid phase and application of these materials in the separationprocesses specially in the environmental cleanup process

Publications: Papers in International Journals: 50; Books: 2; Patents: 1

278 About the Authors

Date post:	17-Jul-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Said Salaheldeen Elnashaie Firoozeh Danafar Hassan...

Documents