photochemical transformations in continuous flow devices

Research Collection

Doctoral Thesis

Photochemical transformations in continuous flow devices

Author(s): Laurino, Paola

Publication Date: 2011

Permanent Link: https://doi.org/10.3929/ethz-a-006682166

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.

ETH Library

https://doi.org/10.3929/ethz-a-006682166

http://rightsstatements.org/page/InC-NC/1.0/

https://www.research-collection.ethz.ch

https://www.research-collection.ethz.ch/terms-of-use

DISS. ETH NO. 19754

PHOTOCHEMICAL TRANSFORMATIONS IN

CONTINUOUS FLOW DEVICES

A dissertation submitted to

ETH ZURICH

for the degree of

Doctor of Sciences

presented by

PAOLA LAURINO

Dipl. Pharm. Chem., Universitá degli Studi di Milano

Date of birth

1.4.1981

citizen of

Italy

accepted on the recommendation of

Professor Dr. Hansjörg Grützmacher

Professor Dr. Peter H. Seeberger

Professor Dr. Dieter A. Schlüter

2011

I

A Maria Emilia, Silvia e Antonio

II

III

“The fun of chemistry is in its unexpectedness. There are times when you come to face-to-

face with an unexpected phenomenon while carrying out experiments. You simply have to

be sufficiently aware and open to accept the seemingly unbelievable. There are still many

more valuable ideas remaining to be discovered. The question is how to find them and how

to develop them into new possibilities.” (Mukaiyama, Angew. Chem. Int. Ed. 2004, 43,

5590)

IV

V

Acknowledgements

I would first like to thank Professor Peter H. Seeberger for his support and encouragement

during my entire stay in his group. The enthusiasm and scientific freedom that he provided

were strong driving forces for me to develop new ideas, and his belief in my abilities

helped me to be more confident and successfully carry on my research.

I am grateful to Professor Hansjörg Grützmacher, his discussions which helped me to

better understand the problems encountered in the polymerization project and improved the

quality of my work and our collaboration.

Thanks to Professor Dieter A. Schlüter who accepted to be my co-examiner.

I thank Dr. Klaus Tauer, who patiently explained and taught me a lot about polymerization,

helped me evaluate my data and supported me to develop the polymerization project.

I especially want to thank Dr. Kikkeri Raghavendra to help me develop the QD projects.

His creativity and imagination taught me a lot about a different way of pursuing science.

Thanks to Dr Nahid Azzouz to carry most of the bioassays of my projects.

I am grateful to Dr. Riccardo Castelli who was always available to help me with chemical

mechanisms and reactions.

Thanks to Dr. Hugo Hernandez who helped me understand the polymerization mechanism

and scientifically supported me in the first period in Berlin.

Thanks to Dr. Bastien Castagner who initially contributed to develop the flow reactor.

I would like to thank the microreactor team, the former members who introduced me in the

micro world and also the present members.

VI

Thanks to Dr. Farhan Bou Hamdan and Dr. Alexander O´Brien to proof read the first

versions of the thesis. I am grateful to Dr. Xiao Yin Mak, Dr. Mattan Hurevich, Dr. Hugo

Hernandez and Dr. Klaus Tauer to correct and help me organize different parts of this

thesis. Thanks to Mr. Bopanna Monnanda Ponnapa to proof read my thesis. Finally a

special acknowledgement to Dr. Claney Lebev Pereira for going carefully through my

entire thesis and helping me improve it with critical suggestions and corrections.

I would like to thank Herr Mario Kessinger and Frau Dorothee Böhme their help in the

administrative issues was fundamental.

Thanks to

My parents for supporting me consistently during these years.

Silvia my sister who taught me a lot about life, without her advices I would have often felt

lost.

My Italian friends who were always there for me, and the friends that I met in Leiden,

Zürich and Berlin, even if spread around the world, I bring part of them always with me.

Professor Ad IJzerman and Dr. Laura Fumagalli without them I would have never been

able to start my PhD in ETH.

My boyfriend who endured me during this last period, his support was endless.

VII

TABLE OF CONTENTS

Abstract 1

Riassunto 4

1. Microreactors for Continuous Flow Organic Synthesis 7

1.1 Concepts and Definition 8

1.2 Continuous Flow Microreactor Design 10

1.3 Particle Synthesis in Continuous Flow 16

1.4 Continuous Flow Photochemistry 22

1.5 Advantages and Limitations 31

1.6 Conclusions 33

1.7 References 34

2. Synthesis and Functionalization of Quantum Dots 41

2.1 Introduction 42

2.2 QDs Batch Synthesis 48

2.3 Continuous Flow QD Synthesis and Functionalization 53

2.4 Photophysical Properties of QDs Produced in Continuous Flow 58

2.5 Biological Studies Using Carbohydrate-Capped QDs 63

2.6 Conclusions 67

2.7 Experimental Part 68

2.8 References 83

3. Continuous Flow Microreactors and Phosphine Oxide Initiators:

an Unique Combination 86

3.1 Introduction 87

3.2 Continuous Flow Microreactor Design 95

3.3 Emulsification 100

3.4 Extraordinarily High Molecular Weight Polymer in Small Particles 104

3.5 Ultrafast Polymerization of Styrene 110

3.6 Investigations on BAPO-AA 113

3.7 Studies on Different Phosphine Oxide Initiators 121

VIII

3.8 Molecular Weight Distribution and Average Size of Polymer

Particles 123

3.9 Snowballing Radical Generation Mechanism 125

3.10 Snowballing Radical Generation Simulation 133

3.11 Seed Polymerization of Polymer Particles 138

3.12 Applications of SRG Mechanism 140

3.13 Conclusions 145


3.15 References 151

4. Final Conclusions and Outlook 155

Appendix 159

5. Glyco-dendronized Polylysine for Bacteria Detection 161

5.1 Indroduction 162

5.2 Synthesis of Glyco-dendrons 165

5.3 Photofunctionalization of Polylysine 166

5.4 Bacteria Detection with Glyco-dendronized Polylysine 168

5.5 Conclusions 170


5.7 References 183

List of Abbreviations 185

Curriculum Vitae 187

IX

1

Abstract

Microreactor enables precise and automated manipulation of reaction mixture in a

defined volume of space. In recent times there has been an increasing interest in the

miniaturization of chemical and biochemical reactors. The microreactor is an efficient and

versatile alternative to the traditional batch reactor techniques and has facilitated the

development of new methods for organic and polymer chemists. Microreactors, when

used in continuous flow permit the precise control and screening of reaction parameters.

This thesis emphasizes the benefits of using microreactor technology for the

synthesis of nanoparticles and the study of photochemical reactions. Herein two successful

applications were studied and described:

1) The use of continuous flow device for the synthesis and functionalization of quantum

dots (QDs).

2) Discovery of a new mechanism for the reaction of phosphine oxide initiators during the

emulsion polymerization in a continuous flow microreactor.

Preliminary results regarding the synthesis of QDs in a round bottomed flask

resulted in major limitations in the control of particle nucleation and growth, which were

primarily due to the non-homogeneous mixing and heating of the reaction mixture. To

overcome these limitations, a versatile new strategy was developed and applied for

producing carbohydrate- or dihydrolipoic acid-capped CdSe/ZnS and CdTe/ZnS quantum

2

dots (QDs) for biological applications. This method involved a three-step flow synthesis

in single-phase by using a continuous flow microreactor.

The first step involved screening different reaction time and temperature conditions to

produce QDs of different size with precise photophysical properties. The second and third

step involved the shell formation (ZnS) to obtain QDs and functionalization of the same

with thiol linked carbohydrate or dihydrolipic acid.

The method described has the following advantages: 1) Production of QDs in the range of

2–4 nm with narrow particle size distribution. 2) Precise control of the nucleation and

growth of QD cores in a single-phase flow and lower temperature than previous syntheses.

3) Easy screening of reaction parameters. 4) The procedure can be scaled up to produce

large amounts of QDs.

In the second part of this thesis a photo-initiated phosphine oxide mediated

polymerization reaction accelerated by snowballing radical generation (SRG) in a

continuous flow microreactor was described.

This method afforded narrow size distributed particle in a scalable and reproducible

manner. Surprisingly, high molecular weight polymer chains were detected in small

particles generated at short residence times (36.5 s). The use of microreactor permitted an

in-depth investigation of the reaction mechanism, screening parameters and conditions not

obtainable in a batch reactor. The classical emulsion polymerization could not justify the

3

observed screening results. A new mechanism for the reaction was proposed whereby

upon irradiation phosphine oxide was able to generate more radicals for particles than

expected. The discovery of this new mechanism of reaction (Snowballing Radicals

Generation) permitted to produce materials with defined morphologies that were difficult

to synthesize previously.

4

Riassunto

Il microreattore permette di manipolare in modo preciso e automatizzato miscele di

reazione in un volume di spazio definito. Recentemente línteresse per la

miniaturizzazione di reattori chimici e biochimici è aumentato. Lúso del microreattore si

presenta come unálternativa efficiente e versatile ai tradizionali reattori e ha facilitato lo

sviluppo di nuovi metodi in chimica organica e polimerica. I microreattori quando usati in

flusso continuo, permettono il controllo preciso e lo screening di parametri di reazione.

Questa tesi enfatizza i vantaggi nellúso della tecnologia del microreattore per la

sintesi di nanoparticelle e lo studio di reazioni di fotochimica. Due esempi sono stati

studiati con successo e saranno riportati:

1) Lúso di un microreattore in flusso continuo per la sintesi e funzionalizzazione di

quantum dots (QDs).

2) La scoperta del meccanismo di reazione degli ossidi di fosfina, usati come iniziatori

della polimerizzazione in emulsione in un microreattore in flusso continuo.

I risultati preliminari della sintesi di QDs in un reattore tradizionale hanno mostrato

delle limitazioni nel controllo della crescita e nucleazione delle particelle, le quali erano

principalmente dovute a una mancanza di riscaldamento e miscelazione omogenei della

5

miscela di reazione. Per superare queste limitazioni, una nuova e versatile strategia è

stata sviluppata e poi utilizzata per produrre QDs (CdSe/ZnS e CdTe/ZnS)

funzionalizzati con diversi carboidrati o con lácido diidrossilipoico per studi biologici.

Questo metodo ha coinvolto tre reazioni in fase omogenea usando un microreattore in

flusso continuo.

La prima reazione è stata ottimizzata utilizzando diverse temperature e tempi di

reazione per produrre QDs con diverse dimensioni e precise proprietà fotofisiche. Il

secondo e terzo passaggio hanno generato lo strato esterno (ZnS) dei QDs e la loro

funzionalizzazione con carboidrati o con lácido diidrossilipoico.

Il metodo descritto presenta i seguenti vantaggi: 1) La produzione di QDs monodispersi

con una dimensione di 2–4 nm. 2) Il controllo preciso della nucleazione e crescita del

nucleo dei QDs in fase omogenea e a temperature piú basse di quelle riportate

precedentemente. 3) Uno screening facile dei parametri di reazione. 4) La procedura

può essere facilmente riprodotta su vasta scala.

Nella seconda parte di questa tesi una reazione di polimerizzazione inizializzata

dagli ossidi di fosfina e accelerata da unínaspettata generazione di radicali (SRG) è

stata studiata usando un microreattore in flusso continuo.

Questo metodo ha permesso di sintetizzare particelle con grandezza uniforme in un

modo riproducibile anche in vasta scala. Il peso molecolare delle catene dei polimeri

6

che costituivano le particelle si è rivelato sorprendentemente elevato per la grandezza

delle particelle e per la velocità con cui erano sintetizzate (36.5 s.). Lúso del

microreattore ha permesso unínvestigazione dettagliata del meccanismo di reazione,

uno screening di parametri e di condizioni di reazione difficilmente ottenibili in un

reattore tradizionale. Un nuovo meccanismo di reazione è stato proposto sulla base del

quale se irradiati gli ossidi di fosfina sono in grado di generare in una particella di

polimero più radicali di quelli previsti. La scoperta di questo nuovo meccanismo di

reazione (Snowballing Radical Generation) ha permesso di produrre materiali con

morfologia definita che prima erano difficili da sintetizzare.

1. Continuous Flow Microreactor for Organic Synthesis

7

1. Microreactor for Continuous Flow Organic Synthesis


8

1.1 Concepts and Definitions

A microreactor or a microstructured reactor is a device where a chemical reaction

takes place within a confined space.1 The devices consist generally, but not exclusively,2 of

a microtubular system that permits chemical reactions to take place in channels of µm

diameter size.1 In contrast to the round-bottomed flask, microreactors can be operated in

continuous flow by constantly feeding in reagents and harvesting the products. Continuous

flow reactors can be classified based on the reaction volume and thus the size. Continuous

flow microreactors and microfluidic devices are miniaturized systems where reactions

occur in a volume scale of µL to few mL. Continuous flow reactors of 5 mL scale and up

are called mesoscale reactors (Figure 1.1).1 In particular, continuous flow microfluidic

device controls and manipulates continuous liquid flow in microchannels.

Figure 1.1 Examples of different volume continuous flow reactors: (a) 6 µL borosilicate glass

reactor; 3 (b) 2.4 mL borosilicate glass reactor;3 (c) 20 mL stainless steel coil reactor.4 Pictures not

to scale.

Continuous flow reactors or microreactors can be classified according to the flow

generated into the channels either in single-phase or segmented-phase (Figure 1.2).

Homogeneous-liquid phase reactions are often carried out in single-phase flow mode

where reagent streams are brought into contact. Heterogeneous phase reactions (liquid-

liquid or gas-liquid) often exploit the two phase segmented flow approach characterized by

the formation of reaction aliquots containing different phases.5


9

Figure 1.2 Schematic representations of single phase and segmented gas–liquid flow microreactor.

The flow process depends on three important parameters: flow rate, residence time,

and reaction throughput. The flow rate is the volume of the reaction mixture that passes

through the device per unit time and is measured in [mL min-1]. The residence time is the

average amount of time that a reaction mixture spends in a continuous flow microreactor

measured in [min]. The residence time is calculated based on the reactor volume and the

flow rate using (eq. 1).

(1) τ = V/q

τ (residence time); V (reactor volume); q (flow rate).

To achieve longer residence times, reagents can be purged more slowly or a larger volume

reactor used. Production rates can vary from nL min-1 to L min-1. The throughput of a

reaction in a continuous flow microreactor defines the productivity. The throughput

measured in [mol min-1] is directly proportional to the flow rate and concentration of the

desired product.

In the past decades, continuous flow microreactors have emerged as an alternative

to the traditional round bottom flask in chemical laboratories.6 Particle synthesis and

photochemical transformations in continuous flow have been explored and will be the

focus of this thesis.


10

1.2 Continuous Flow Microreactor Design

Microreactors are often custom-designed for the desired chemical process, and the

materials chosen for fabrication may depend on the reaction conditions. Commonly used

materials, that each offer advantages and disadvantages include polymers, glass, silicon,

ceramics and metals (Figure 1.3).

Figure 1.3 Examples of microreactors fabricated from: (a) glass;7 (b) ceramic;8 (c) metal9 and (d)

silicon.10

Polymers are generally the cheapest solution, but can swell during the fabrication

process.11 Recently, perfluorinated polymers have been used, most commonly in the form

of tubing and despite temperature limitations, are inert to most reagents and solvents.12

Glass is difficult to handle during fabrication but is the most chemically inert material

available.13 Ceramics are another chemically inert option and stable at high temperature,

but difficult to handle, during the fabrication process.14 Other choices include steel and

silicon, both of which are resistant to high temperature, pressure and have good thermal

conductivity. Silicon is not compatible with bases and the manufacturing costs are very

high.15 The reactor is usually integrated into a system composed of many components like

an injection system, mixer, purification apparatus and analytical system that are described

in detail below (Figure 1.4).


11

Figure 1.4 Components of continuous flow microreactor.

1.2.1 Injection System

Continuous flow microreactors are typically driven by an injection system, that can

be operated by a mechanical or a non-mechanical pump. Mechanical pumps depend on

moving parts such as valves or pistons to pump liquid with a precise flow rate. Three

examples of common mechanical pumps are syringe pumps, peristaltic pumps and high-

performance-liquid chromatography (HPLC) pumps (Figure 1.5).

Figure 1.5 Schematic representation of mechanical pumps: (a) syringe pump; (b) peristaltic pump;

(c) HPLC pump.

Syringe pumps are small infusion pumps where the liquid is stored in syringes. In a

peristaltic pump the rotation of a rotor compresses a flexible tube, which forces the liquid

to move through the tubing.16 HPLC pumps use the action of small pistons to displace the

volume of liquid through the tubing. A non-mechanical pump, on the other hand, utilizes

an activation force such as electro-hydrodynamic, electro-osmotic or ultrasonic flow


12

generation to move the fluids. 17 Electro-hydrodynamic pumps use an electric field

generated between two electrodes to create a flow. Operating at a defined voltage, the

electrodes can accelerate charged molecules thereby generating a fluidic stream. The

absence of moving elements in non-mechanical pumps renders them easier to integrate in

the continuous flow microreactor.

The connections between the microreactor and other devices in the system are

created by polymer, silicon, glass or metal tubing. The tubing as well as the connections

must often withstand high pressures of up to 10 MPa. Connection can be established via

compression sealing with rubber O-rings, that allows for the connection to be easily

interchanged.18 Another option is a permanent connection with epoxy-glue, glass brazing

or metal soldering while these connections are more robust, they require elaborate

fabrication techniques. 15b, 19

1.2.2 Mixing Systems

Reagents used in continuous flow microreactors can be premixed or simultaneously

injected by passing through a mixing system (for example, via T or V connections,

micromixers, etc) before reaching the microreactor. 20 The mixing system can be designed

whereby the fluid flow is constrained by a laminar regime, as a result of small-scale

tubing.1c Flow is defined as laminar when the Reynolds number that represents the ratio of

convective to viscous forces, is below 2100 for cylindrical channels (eq. 2).

(2) Re = Ud/ ρ

Re (Reynolds number); U (average velocity); d (hydraulic diameter of the channel); ρ (viscosity).


13

In a laminar regime, the mixing is mainly based on the diffusion between fluids. The Peclét

number,1c which represents mass transfer as the ratio of convection and diffusion describes

this situation (eq. 3).

(3) Pe = Ud/D

Pe (Peclét number); U (average velocity); d (hydraulic diameter of the channel); D (diffusion

coefficient).

A flow microreactor can have an integrated mixer or an external mixer, depending on

whether the microdevice has multiple inlet channels that combine into the mixer or

whether it is made of just a single inlet. The mixer can be either passive or active. In the

first case, mixing is due to the passive diffusion of the fluids.21 The laminar effect between

the two fluids increases the surface of contact of fluids and consequently improves the

diffusion of reactants between the phases. With multi-laminar flow the above effect can be

further enhanced (Figure 1.6).

Figure 1.6 Schematic representation of (a) laminar flow, or (b) multi-laminar flow generated by Y

connection.

In the case of active mixing, forces such as ultrasonic waves, thermal gradient, electric or

magnetic field fluctuations are applied to enhance the diffusion process.22 Active mixing

requires more elaborate fabrication skills and the chemicals used during the reaction must

be compatible with the forces applied.23 Active mixing and the geometry of the channel

networks of the device can give rise to turbulent flow generated by fluctuations of flow

velocity and pressure, and characterized by a chaotic and stochastic regime (Figure 1.7).24


14

Figure 1.7 (a) Laminar versus (b) turbulent flow.

1.2.3 Purification Apparatus and Analytical Systems

The ability to incorporate purification and analytical devices makes continuous

flow microreactors a very versatile system. According to the needs and nature of the

microreactor, other tools can be integrated downstream of the device such as an

microextractor,25 a distillatory apparatus,26 a column containing a scavenger27 or even

other reactors for multistep reactions.28 Liquid-liquid extraction in continuous flow

microreactors can be achieved by using a thin porous fluoropolymer membrane that

selectively wets organic solvent. To retain the reaction byproduct, a scavenger supported

on a solid support and packed into a column can be used in continuous flow as a

purification apparatus (Figure 1.8).

Figure 1.8 Schematic representation of the principle for (a) liquid-liquid microextractor and (b)

scavenger column.

The continuous flow reactor can be used as a platform for rapid reaction screening when

combined with an analytical system. An automated system to measure changes in

temperature and pressure can be integrated in-line with the flow device. For example, an


15

internal thermocouple allows for the real time monitoring of reaction temperatures.29

Incorporation of analytical systems such as HPLC, FTIR, IR etc to facilitate the control

and screening of parameters is also possible.30


16

1.3 Particle Synthesis in Continuous Flow

Continuous flow microreactor can be exploited for the efficient synthesis of nano-

and micro-particles.31 The major drawback of synthesizing these kinds of particles in

round-bottomed flasks is the general lack of reproducibility and difficulty in scaling up,

due to non-uniform mixing and heating of the reaction mixing. These limitations

significantly affect particle nucleation and growth, both of which are important features in

obtaining high quality material. The main advantage of continuous flow microreactor

compared to batch processes is the easy scale up by simply prolonging the run time while

keeping the other parameters and conditions consistent throughout the process. Excellent

composition and morphology of the final material can result from the high performance

intrinsic property of the microreactor system (better heating and pressure control, mixing,

etc).32 Furthermore, the synthesis of complex heterogeneous particles, multi-shell particles,

novel materials and structures can be achieved using the microreactor device due to the

ease and safety with which high temperatures and pressures can be attained in the

device.18a, 33 However, one major challenge encountered in the synthesis of particles by

continuous flow microreactor is to maintain the controlled growth of these particles so as

to avoid clogging of the flow system. To highlight the contributions of microreactors to the

synthesis of particles, some examples are discussed below.

1.3.1 Organic Particles

The polymerization of monomer droplets in a continuous flow microreactor can

result in the formation of particles of different shapes and sizes.34 The polymerization can

also occur at the interface of the polymer droplets to generate capsules.35 Furthermore, the


17

formation of different sized particles can be achieved by using immiscible solvents, taking

advantage of droplet formation in a heterogeneous system.36

Kumacheva et al. have reported the effects of changing the channel dimensions to achieve

the required polymer particle characteristics (discoid or spherical shapes). 37 A solution of

monomer (dimethylacrylate oxypropyldimethylsiloxane, DMOS) containing a

photoinitiator (1-hydroxycyclohexyl phenyl ketone, HCPK), was dispersed in water and

stabilized by a surfactant (sodium dodecyl sulfate, SDS) (Figure 1.9).

Figure 1.9 (a) A schematic bidimensional increase of the reaction zone where the particles are

generated and irradiated (UV lamp with an output of 400W at a wavelength of 330-380 nm). The

monomer phase is injected as core flow and the water phase as sheath flow. Depending to the

height of the channel, disk or spherical particles can be generated (b) Image of the solidified disk

(scale bar = 200 nm).38

Different particle shapes were generated using this method by forcing the monomer

droplets at different heights of the channel and by controlling the flow rate into the reactor

and varying the viscosity of the solution according to the capillary number (eq. 4).

(4) Ca = µυ/γ

Ca (capillary number); µ (dynamic viscosity of the aqueous phase); υ (velocity of the aqueous

phase); γ (interfacial tension between the oil and aqueous phase).


18

After generating circular or discoid shapes, a lattice structure was trapped by UV

irradiation for 30–180 s (Figure 1.7). Depending on whether the droplets relax faster or

slower after entering the inlet and on their viscosity before the irradiation, the shape of the

disk (spherical or hexagonal or pentagonal) was controlled. The resulting particles had

controlled size distribution and shape. 38


19

1.3.2 Inorganic Particles

The synthesis of inorganic particles using microreactors has been widely

investigated and syntheses of silver,39 gold,40 palladium41 or copper42 nanoparticles have

been reported. Köhler et al. have described the synthesis of gold nanoparticles in a

continuous flow glass–silicon microreactor.40 Gold nanoparticles with mean diameters of

5–50 nm were obtained directly from the gold salt (HAuCl4) and a reducing agent

(ascorbic acid). Optimization and rapid screening of flow rate, pH, amount of reducing

agent and concentration of poly(vynilpyrrolidone) (PVP) resulted in gold nanoparticles that

were twice as narrow in size distribution than those synthesized using a round-bottomed

flask (Figure 1.10).

Figure 1.10 Picture of microreactor used for the synthesis of gold nanoparticles [PVP =

poly(vynilpyrrolidone)].40

The synthesis of inorganic particles often requires the addition of a ligand or surfactant in

order to avoid clogging of microreactor channels, and can be carried out under homo- or

heterogeneous conditions. The main advantage of using the continuous flow reactor for

homogenous synthesis of particles is the possibility to add reagents consistently at different

time points to the reaction stream. Furthermore, the segmented flow of a heterophasic

system can be a valid option to control nucleation and particles growth.43

The use of metal oxides to generate particles containing iron,44 silica45 or titanium

dioxide has also been reported.46 Tadaki et al. prepared titanium dioxide nanoparticles on

the insoluble interface of a continuous flow ceramic microreactor (Figure 1.11).46


20

Figure 1.11 Ceramic microreactor used for the synthesis of TiO2 nanoparticles. Two inlets and two

outlets are shown.46

The nanoparticles with a diameter below 10 nm were prepared by hydrolysis of titanium

alkoxide. The selection of proper solvents and optimum flow rate allowed for a stable

interface between two insoluble solvents and this interface was exploited as the reaction

chamber where particle nucleation and growth occurred.

Coalescence of droplets containing reagents is another approach for the synthesis of

metal oxide nanoparticles. In the continuous flow microreactor two electrodes can be

placed to generate droplet currents. The coalescence of these droplets can be fast enough

that the mixing of the reagents is optimal for the preparation of narrow size distribution

particles.47 Other examples reported in the literature use small seed particles as nucleus-

precursors for the growing particles.48

1.3.3 Miscellaneous Particles

Other types of particles that combine two or more organic and/or inorganic

materials, such as shell particles, can also be synthesized in continuous flow reactors. One

example are the particles made of different semiconducting materials that we have

synthesized and will be discussed in the chapter 2 of this thesis.49. These semiconducting

particles can be covered by a shell of polyethylene glycol, where the optical properties of


21

the core particles combined with the biocompatible characteristics of PEG yielded particles

suitable for in vitro or in vivo imaging studies.50


22

1.4 Continuous Flow Photochemistry

Since the beginning of the 20th century, the importance of photochemical reactions

has been recognized, and much effort has been made to improve the efficiency of these

reactions.51 The advantages of photochemical transformations are well documented. 52

Reactions can be carried out in mild and environmental friendly conditions. Light is used

to activate the starting materials, minimizing or avoiding the use of other reagents which

can often result in the generation of side products.53 Photochemical reactions are key

synthetic organic chemistry.54

The disadvantages associated with photochemical reactions are usually related to

the light source and reactor design. The light source is usually a lamp with a specific life

time (e. g. xenon lamp, halogen lamp, medium, low or high pressure mercury lamps). The

emission of the lamp typically changes over time, and usually a cooling system is required

to dissipate the heat that is generated. Drawbacks related to the traditional batch

photochemical reactions include a lack of reproducibility, due to the difficulty in

maintaining the ratio between the irradiated area and the reaction volume consistent

especially during reaction scaling up.

Continuous flow reactors permit excellent control of the irradiation time as it is

directly related to the flow rate. According to the Beer-Lambert law, the penetration of

light decreases exponentially with the distance from the light source (eq. 5).

(5) I/I0 = e-εlc

I (Intensity transmitted light); I0 (Intensity incident light); ɛ (extinction coefficient); l (path length

that the light has to pass through); c (concentration of the substance in the solution).

In a continuous flow device, the penetration of light is maximized and consistent

throughout the system (Figure 1.12).


23

Figure 1.12 Schematic representation of a vertical section of batch photoreactor (a) and continuous

flow photoreactor (b). (l is the path length that light has to go through in the reaction solution)

Reactions can be scaled up without changing the parameters. Scaling down allows

for screening various reaction parameters. Finally the compactness of the system facilitates

efficient cooling of the system. 17c, 55 For smaller flow systems, a good alternative light

source that was recently exploited, are light emitting diodes (LEDs). LEDs are small,

efficient and do not require powerful cooling system because they do not warm up

excessively. LEDs emit light at a specific wavelength and avoid side reactions that can

occur at different wavelengths thereby simplifying product purification.56


24

1.4.1 Miniaturized Systems for Photochemical

Transformations

Different miniaturized systems have been used for photochemical reactions,

ranging from commercially available systems to in-house designs. One of the systems

initially introduced for photochemical reaction is the FFMR-cyl, which is a thin falling-

film reactor, where the channels are parallel and engraved around a metal cylinder (Figure

1.13).57

Figure 1.13 Cylindrical falling film micro reactor (FFMR-cyl) for scale-up and its standard version

FFMR.

The type of material used for the fabrication of the microreactor is of critical

importance. Jensen and coworkers have fabricated two reactor prototypes, one made of

glass for reactions in the wavelength of 365 nm and the other made of quartz for shorter

wavelength reactions. In both cases, the microreactors are serpentine shaped in order to fit

the longest channel in the smallest space possible, with one inlet and one outlet (Figure

1.14). The mask has been designed according to the channel shape and then covered by a

Pyrex or Quartz wafer.58


25

Figure 1.14 Serpentine shaped microreactor fabricated by Jensen et al.59

De Mello et al. reported the fabrication of a glass microreactor that is serpentine

shaped and has two inlets, one for a gaseous reagent and other for liquids.59 Fukuyama et

al. have used a pre-fabricated reactor Mikroglas Dwell Device, FOTURAN®60 which is

made of glass and integrated with a cooling/heating system (Figure 1.15).6c

Figure 1.15 A glass microreactor with integrated cooling system (1000 µm width and 500 µm

depth). View of microreactor (a) from the top; (b) from lateral. (c) Picture of the system.

Another approach is a tubular reactor, coiled around the lamp with a cooling system

placed in-between the lamp and the tubing reactor. In this case, the channel is made of

flexible Teflon tubing and not of glass or quartz (Figure 1.16) 2, 61


26

Figure 1.16 FEP (fluorinated ethylene propylene) flow reactor (a) and picture of the system (b).61a

The most commonly used types of Teflon tubing are polytetrafluoroethylene (PTFE) and

fluorinated ethylene propylene (FEP) because of high light transmittance, high pressure

resistance and inertness to most organic compounds.62


27

1.4.2 Photochemical Reactions in Continuous Flow

Photochemical cycloaddition reactions have been investigated with different

substrates from different groups. Booker-Milburn et al. initially reported the cycloaddition

of malemide with 1-hexyne to yield substituted cyclobutene (Scheme 1.1).61a The reaction

was carried out in a FEP tubing coiled around a medium pressure Hg lamp (Figure 1.16).

This system produced more than 500 g of 3 in a continuous 24 h processing time.

Scheme 1.1 [2 + 2] cycloaddition between maleimide 1 and 1-hexyne 2.

Fukuyama et al. reported the [2 + 2] addition of vinyl acetates with various

cyclohexenone derivatives using a microflow system (Scheme 1.2).51c, 63

Scheme 1.2 [2 + 2] Cycloaddition of cyclohexenone 4 and vinyl acetate 5.

Yields were improved and reaction times significantly reduced, as compared to the same

reaction run in batch using a pyrex flask. Two serially-connected micro-photoreactors used

in tandem at twice the flow-rate used for a single reactor provided a similar result (Table

1.2).


28

Table 1.2 Comparison between microreactor and flask conditions.

In addition to their studies of [2 + 2] cycloadditions, Fukuyama et al. investigated

the use of a Barton reaction to construct steroid 8, an important intermediate in the

synthesis of an endothelial receptor antagonist (Scheme 1.3). Initial optimization studies of

this reaction involved the comparison of various glass covers (quartz, pyrex, and soda lime

glass) for the stainless steel continuous flow microreactor, as well as examination of the

light source (300 W high-pressure Hg lamp, or 15 W black light lamp) and its distance

from the reactor. A scale up synthesis was achieved (up to 20 h of run) under optimal

conditions using two serially-connected reactors (total volume of 8 mL, a residence time of

32 min and irradiation by eight 20 W black light lamps, 60 % yield).55c

Scheme 1.3 Barton nitrite photolysis of steroid 8.

Booker-Milburn et al. used the same flow system for an intramolecular [5 + 2]

cycloaddition to form the key bicyclic azepine intermediate for the synthesis of the

Stemona alkaloid (±)-neostenine (Scheme 1.4). The flow reaction was performed

continuously for 24 h, demonstrating that scale-up was possible. The yields were improved

to 63% in microreactor compared to batch reactor results (20% in flask) with shorter

residence time.64

Device Residence time (h) Yield (%)

Microreactor 2 88

Two serially connected

microreactors 2 85

Pyrex flask 2 8

Pyrex flask 4 22


29

Scheme 1.4 Key step in the synthesis in (±)-neostenine

Another relevant example was reported by Aggarwal et al. The thermal [4 + 2]

cycloaddition involved a thioaldehyde 13 generated by photolysis of a corresponding

thioester 12, and reaction with cyclopentadiene yielded intermediate 14 for the synthesis of

a chiral sulphide, a synthetically useful reagent for organo-catalytic reactions (Scheme

1.5). After optimizing the reaction conditions in a continuous flow reactor they reported a

large scale synthesis of the intermediate (38 g substrate, 75% yield). The superior

performance of a continuous flow photoreactor compared to a batch photoreactor was

demonstrated clearly.

Scheme 1.5 Photolysis of thioester 12.

Photochemical reactions have been also used in polymer synthesis. Kumacheva et

al. investigated a multistep polymerization in continuous flow microreactor combining a

polyaddition of acrylate monomer [tri(propylene glycol) diacrylate, TPGDA]

photoinitiated by 2-diethoxyacetophenone (DEAP) and a thermal polycondensation of

urethane oligomer (Scheme 1.6). The first step was a free radical photoinitiated

polymerization reaction that was able to generate the necessary heat to activate a second


30

polycondensation. The result is the formation of an interpenetrating polymer network

structure. The resulting cross-linked polymer particles were of narrow size distribution and

controlled morphology. Furthermore, the microfluidic device provided for an easy

screening of monomers.65

Scheme 1.6 Step 1, Photopolyaddition of tri(propylene glycol) diacrylate, TPGDA initiated by 2-

diethoxyacetophenone (DEAP) under UV irradiation (Dr Hönle UVA Print 40C, F-lamp, 400 W, λ

= 330–380 nm); step 2, thermal polycondensation of urethane oligomer catalyzed by dibutyltin

dilaurate (DBTDL).


31

1.5 Advantages and Limitations

The benefits of continuous flow microreactor technology has been widely reviewed

and several examples have demonstrated the utility of microreactor and their influence on

chemistry.1f, 15b, 66 Drug discovery and process development can be improved by the use of

a continuous flow system by shortening the screening process and optimizing reaction

conditions and reagents.27, 67 Compared to the traditional round bottom flask, the geometry

of the channels of the continuous flow microdevice permits a larger surface-to-volume

ratio thereby enhancing the mass-heat transfer that influences the time,68 yield and

selectivity of the reactions.69 The continuous flow microreactors can be used for the

synthesis of hazardous compounds due to the highly controlled nature of the system and

isolation of the reaction from air and moisture. In principle, reactions in a continuous flow

microdevice can be performed easily under inert conditions.70 The minimal use of solvent

and energy helps to conserve resources and helps practice environmentally friendly

chemistry. Scaling up can be done by extending the run time, or numbering up the number

of reactor (Figure 1.17).51c

Figure 1.17 Continuous flow approaches to large scale production numbering up system.

The continuous flow device also permits to optimize many reactions in a microscale by

reducing the cost and limiting the output of hazardous byproducts. The above factors allow

for process optimization and later scaling up for production.71


32

Continuous flow microreactor nevertheless have some limitations arising from the

limited solubility of the reactants and products, incompatibility of solid reactants and the

necessity of using of a unique solvent or miscible solvent for multistep reaction when an

extractor cannot be integrated into the system. The combination of different operation units

is not always efficient due to mixing problems, the handling of multi-step reactions or

different phase reactions and finally the separation of these phases.72 The identification of

products in a very small scale reaction is another issue, which however if feasible can be

overcome by the incorporation of an on-line analytical system.73 In addition, the cost of

construction and technical complexity of such systems has to be considered.74 The decision

of running a reaction whether in continuous flow or in a batch reactor must be evaluated on

a case-by-case basis considering the purpose of the study (rapid screening or optimization

of conditions, scale up, high throughput etc.) and important factors such as mixing, heat

transfer, or reaction time.75


33

1.6 Conclusions

Continuous flow techniques have been used as alternative to batch reactions.

Applications have been shown in diverse fields of research ranging from biology,

biotechnology, polymer chemistry to traditional organic chemistry. Typical uses of the

continuous flow microreactors require that these systems be inexpensive and simple to

operate. Eventually these devices can become powerful tools for chemical reactions,

analysis, cells sorting, high-throughput screening and particle production. Despite

successful examples in many research areas on how the continuous flow microreactors can

facilitate and improve various transformations, their potential has not been realized

entirely. One main limitation of these technologies is related to the fact that even if

modular systems are available, there are no flexible systems that allow for a wide set of

chemical transformation. Still each system remains quite specific for one particular

chemical reaction. Another main disadvantage that chemists have to face is its

incompatibility with solids and the clogging of the system due to precipitation of

components of the reaction mixture.

The aim of this dissertation is two-fold: first, to investigate and improve upon the synthesis

of particles, benefiting from the highly controlled parameters and reaction conditions of the

microreactor. The second part of the thesis focuses on photo-initiated emulsion

polymerization in continuous flow microreactor and reports on the reaction mechanism of

phosphine oxide initiators.


34

1.7 References

1 For reviews and books on microreactor technology, see: (a) Ehrfeld, W.; Hessel, V.;

Löwe, H. Microreactors: New Technology for Modern Chemistry, Wiley-VCH, Weinheim,

2000. (b) Fletcher, P. D. I.; Haswell, S. J.; Pombo-Villar, E.; Warrington, B. H.; Watts, P.;

Wong, S. Y. F.; Zhang, X. L. Tetrahedron 2002, 58, 4735–4757. (c) Hessel, V.; Hardt, S.;

Löwe, H. Chemical Micro Process Engineering, Wiley-VCH, Weinheim, 2004. (d)

Thayer, A. M. Chem. Eng. News 2005, 83, 43–87. (e) Geyer, K.; Codée, J. D. C.;

Seeberger, P. H. Chem. Eur. J. 2006, 12, 8434–8442. (f) Brivio, M.; Verboom, W.;

Reinhoudt, D. N. Lab Chip 2006, 6, 329–344. (g) Mason, B. P.; Price, K. E.; Steinbacher,

J. L.; Bogdan, A. R.; McQuade, D. T. Chem. Rev. 2007, 107, 2300–2318. (h) Wiles, C.;

Watts, P. Eur. J. Org. Chem. 2008, 10, 1655–1671. (i) Microreactors in Organic Synthesis

and Catalysis ed. T., Wirth; Wiley–VCH, Weinheim, 2008; (f) Luis, S. V.; Garcia-

Verdugo, E. Chemical Reactions and Processes under Flow Conditions RSC Green

Chemistry Series 2010. 2 Anzenbacher, P.; Palacios, M. A. Nat. Chem. 2009, 1, 80–86. 3 www.micronit.com 4 www.uniqsis.com 5 Köhler, J.M.; Henkel, Th.; Grodrian, A.; Kirner, Th.; Roth, M.; Martin; Metze, K. J.

Chem. Eng. J. 2004, 201, 201–216. 6 Hook, B. D. A.; Dohle, W.; Hirst, P. R.; Pickworth, M.; Berry, M. B.; Booker-Milburn,

K. I. J. Org. Chem. 2005, 70, 7558–7564. 7 www.pumpen-ventile.de 8 www.emeraldinsight.com 9 www.nanobits.com 10 Snyder, D. A.; Noti, C.; Seeberger, P. H.; Schael, F.; Bieber, T.; Rimmel, G.; Ehrfeld,

W. Helv. Chim. Acta 2005, 88, 1-‐9. 11 Holger, B. H.; Gärtner, C. Rev. Mol. Biotech. 2001, 82, 89–99.


35

12 (a) Willis, P. A.; Hunt, B. D.; White, V. E.; Lee, M. C.; Ikeda, M.; Bae, S.; Pelletier, M.

J.; Grunthaner, F. J. Lab Chip 2007, 7, 1467–1474; (b) Grover, W. H.; von Muhlen, M. G.;

Manalis, S. R. Lab Chip 2008, 8, 913–918; (c) Yoon, T. H.; Park, S: H.; Min, K. I.; Zhang,

X. L.; Hanswell, S. J.; Kim, D. P. Lab Chip 2008, 8, 1454–1459. 13 Kikutami, Y.; Hibara, A.; Uchiyama, K.; Hisamoto, H.; Tokeshi, M.; Kitamori, T. Lab

Chip 2002, 2, 193–196. 14 Knitter, R.; Gohring, D.; Risthaus, P.; Hausselt, J. Microsyst. Technol. 2001, 7, 85–90. 15 (a) Jahnisch, K.; Hessel, V.; Lowe, H.; Bearns, M. Angew. Chem. Int. Ed. 2004, 43, 406–

446. (b) Murphy, E. R.; Inoue, T.; Sahoo. H. R.; Zaborenko, N.; Jensen, K. F. Lab Chip,

2007, 7,1309–1314. 16 Strike, D. J.; Thiébaud, P.; van der Sluis, A. C.; Koudelka-Hep, M.; de Rooij, N. F.;

Microsys. Tech. 1994, 1, 48–50. 17 (a) Watts, P.; Haswell, S. J. Chem. Soc. Rev. 2005, 34, 235–246. (b) Okamoto, H.;

Ushijima, T.; Kitoh, O. Chem. Eng. J. 2004, 101, 57–63. (c) Hasegawa, T.; Koyama, D.;

Nakamura, K.; Ueha S. J. Electroceram. 2008, 20, 145–151. 18 (a) Marre, S.; Park, J.; Rempel, J.; Guan, J.; Bawendi, M. G.; Jensen, K. J. Adv. Mater.

2008, 20, 4830–4834. (b) Marre, S.; Adamo, A.; Basak, S.; Jensen, K. Lab Chip 2010,

DOI: 10.1039/c000156b. (c) Muller, A.; Cominos, V.; Hessel, V.; Horn, B.; Schurer, J.;

Ziogas, A.; Jahnisch, K.; Hillmann, V.; Grosser, V.; Jam, K. A.; Bazzanella, A.; Rinke, G.;

Kraute, A.; Chem. Eng. J. 2005, 107, 205–214. 19 (a) Yen, B. K. H.; Gunther, A.; Schmidt, M. A.; Jensen, K. F.; Bawendi, M. G. Angew.

Chem. Int. Ed. 2005, 44, 5447–5451. (b) Trachsel, F.; Hutter, C.; von Rohr, P. R. Chem.

Ing. J. 2008, 135, S309–S316. 20 Jähnisch, K.; Hessel, V.; Löwe, H.; Baerns, M. Angew. Chem., Int. Ed. 2004, 43, 406. 21 (a) Hessel, H.; Lowe, H.; Schonfeld, F. AIChE J. 2003, 49, 599–577. (b) Nguyen, N. T.;

Wu, Z. G. J Micromech. Microeng. 2005, 15, R1–R16. 22 (a) Hessel, H.; Lowe, H.; Schonfeld, F.; Chem. Eng. Sci. 2005, 60, 2479–2501. (b) Liu,

R. H.; Yang, J. N.; Pindera, M. Z.; Athavale, M.; Grodzinski, P. Lab Chip 2002, 2,151–

157. 23 El Moctar, A. O.; Aubry N.; Batton J. Lab Chip 2003, 3, 273–280.


36

24 Strrock, A. D.; Dertinger, A. A.; Mezic, I.; Stone, H. A.; Whitesides G. M. Science 2002,

295, 647–651. 25 (a) Kralij, J. G.; Sahoo, H. R.; Jensen, K. F. Lab Chip 2007, 7, 256–263. (b) Aota, A.;

Nonaka, M.; Hibara, A.; Kitamori, T. Angew. Chem. Int. Ed., 2007, 46, 878–880. 26 Hartman, R. L.; Sahoo, H. R.; Yen, Be. C.; Jensen, K. F. Lab Chip 2009, 9, 1843–1849. 27 (a) Baxendale, I. R.; Lee, A.-L.; Ley, S. V. J. Chem. Soc., Perkin Trans. 1 2002, 1850–

1857. (b) Baxendale, I. R.; Ley, S. V.; Smith, C. D.; Tranmer, G. K. Org. Lett. 2006, 8,

5231–5234. (c) Iglesias-Sigüenza, F. J.; Smith, C. J.; Baxendale, I. R; Ley, S. V. Org.

Biomol. Chem. 2007, 5, 2758–2761. (d) Ley, S. V.; Baxendale, I. R. Chimia 2008, 62,

162–168. 28 (a) Sahoo, H. R.; Kralj, J. G.; Jensen, K. F. Angew. Chem. Int. Ed. 2007, 46, 566–577.

(b) Baxendale, I. R.; Griffiths-Jones, C. M.; Ley, S. V.; Tranmer, G. K. Synlett 2006, 427–

430. 29 Losey, M. W.; Schmidt, M. A.; Jensen, K. F. Ind. Eng. Chem. Res. 2001, 40, 2555–2562. 30(a) Löbbecke, S.; Ferstl, W.; Pani, S.; Türcke, T. Chem. Eng. Tech. 2005, 28, 484–493.

(b) Zhang, X. L.; Wiles, C.; Painter, S. L.; Watts, P.; Haswell, S. J. Chim. Oggi-Chem.

Today 2006, 24, 43–45. (b) Hartman, R. L.; Naber, J. R.; Buchwald, S. L.; Jensen, K.F.;

Angew. Chem. Int. Ed. 2010, 49, 899–903. 31 (a) Leng, J.; Salmon, J. B. Lab Chip 2009, 9, 24–34. (b) Engl, W.; Backov, R.; Panizza,

P. Curr Opin. Colloid Interface Sci. 2008, 13, 206–216. (c) Dendukuri, D.; Doyle, P. S.

Adv. Mater. 2009, 21, 4071–4086. (d) Marre, S.; Jense, K. F. Chem Soc Rev. 2010, 39,

1183–1202. 32 (a) Tumarkin, E.; Kumacheva, E. Chem. Soc. Rev. 2009, 38, 2161–2168. (b) LoPresti,

C.; Lomas, H.; Massignani, M.; Smart, T.; Battaglia, G. J. Mater. Chem. 2009, 19, 3567–

3590. 33 (a) Sue, K.; Kimura, K.; Arai, K. Mater. Lett. 2004, 58, 3229–3231. (b) Sue, K.; Hattori,

H.; Satp, T.; Komoriya, T.; Kawai-Nakamura, A.; Tanaka, S.; Hiaki, T.; Kawasaki, S.;

Takebayashi, Y.; Yoda, S.; Furuya, T. Chem Lett. 2009, 38, 792–793. 34 Shah, R. K.; Kim, J. W.; Agresti, J. J.; Weitz, D. A.; Chu, L. Y. Soft Matter 2008, 4,

2303–2309.


37

35 Zhang, H.; Ju, X. J.; Xie, R.; Cheng, C. J.; Ren, P. W.; Chu, L. Y. J. Colloid Interface

Sci. 2009, 336, 235–243. 36 Xu, S.; Nie, Z.; Seo, M.; Lewis, P.; Kumacheva, E.; Stone, H. A.; Garstecki, P.; Weibel,

D. B.; Gitlin, I.; Whitesides G. M. Angew. Chem. Int. Ed. 2005, 44, 724–728. 37 Seo, M.; Nie, Z.; Xu, S.; Lewis, P. C.; Kumacheva, E. Langmuir 2005, 21, 4773–4775. 38 (a) Hayward, R. C.; Utada, A. S.; Dan, N.; Weitz, D.A. Langmuir 2006, 22, 4457–4461.

(b) Kim, J. W.; Utada, A. S.; Fernandez-Nieves, A.; Hu, Z. B.; Weitz, D. A. Angew. Chem.

Int. Ed. 2007, 46, 1819–1822. (c) Karnik, R.; Gu, F.; Basto, P.; Cannizarro, C.; Dean, L.;

Kyei, M. W.; Langer, R.; Farokhzad, O. C. Nano Lett. 2008, 8, 2906–2912. 39 Lin, X. Z.; Terepka, A. D.; Yang, H. Nano Lett. 2004, 4, 2227–2232. 40 Köhler, J. M.; Wagner, J. Nano Lett. 2005, 5, 685–691. 41 Song, Y.; Cumar, C. S.; Hormes, J. Nanosci. Nanotechnol. 2004, 7, 788–793. 42 Hung, L. H.; Lee, A. P. J. Med. Biol. Eng. 2007, 27, 1–6. 43 (a) Kumar, D. V. R.; Kasture, M.; Prabhune, A. A.; Ramana, C. V.; Prasad, B. L. V.;

Kulkarni, A. A. Green Chem. 2010, 12, 609–615. (b) Hartlieb, K. J.; Saunders, M.;

Jachuck, R. J. J.; Raston, C. L. Green Chem. 2010, 12, 1012–1017. (c) Iyer, K. S.; Raston,

C. L.; Saunders, M. Lab Chip 2007, 7, 1800–1805. (d) Nishikawa, H.; Morita, T.;

Sugiyama, J.; Kimura, S. J. Colloid Interface Sci. 2004, 280, 506–510. 44 Salazar-Alvarez, G.; Muhammed, M.; Zagorodni, A. A. Chem. Eng. Sci. 2006, 61,

4625–4633. 45 (a) He, P.; Greenway, G.; Haswell, S. J. Chem. Eng. J. 2011, 167, 694–699. (b) Corradi,

A. B.; Bondioli, F.; Ferrari, A. M.; Focher, B.; Leonelli, C. Pow. Tech. 2006, 167, 45–48. 46 Yonemoto, T.; Kubo, M.; Doi, T.; Tadaki, T. Chem. Eng. Res. Des. 1997, 75, 413–419. 47 (a) Abou-Hassan, A.; Bazzi, R.; Cabuil, V. Angew. Chem. Int. Ed. 2009, 48, 7180–7183.

(b) Wang, H.; Nakamura, H.; Uehara, M.; Miyazaki, M.; Maeda, H. Chem. Comm. 2002,

1462–1463. 48 Wagner, J.; Köhler, J. M. Nano Lett. 2005, 5, 685–691. 49 Kikkeri, R.; Laurino, P.; Odedra, A.; Seeberger, P. H. Angew. Chem. Int. Ed. 2010, 49,

2054–2057. 50 (a) Marre, S.; Park, J.; Rempel, J.; Guan, J.; Bawendi, M. G.; Jensen, K. F. Adv. Mater.

2008, 20, 4830–4834. (b) Sounart, T. L.; Safier, P. A.; Voigt, J. A.; Hoyt, J.; Tallant, D. R.;


38

Matzke, C. M.; Michalske, T. A. Lab Chip 2007, 7, 908–915. (c) Yen, B. K. H.; Günther,

A.; Schmidt, M. A.; Jensen, K. F.; Bawendi, M. G. Angew. Chem. Int. Ed. 2005, 44, 5447–

5451. (d) Shestopalov, I.; Tice, J. D.; Ismagilov, R. F. Lab Chip 2004, 4, 316–321. (e)

Edel, J. B.; Fortt, R.; deMello, J. C.; deMello, A. J. Chem. Commun. 2002, 1136–1137. (f)

Dideko, Y. T.; Suslick, K. S. J. Am Chem Soc. 2005, 127, 12196–12197. 51 (a) Paternó, E. Gazz. Chim. It. 1909, 39, 237–250. (b) Ciamacian, G. Science 1912, 36,

385–394. 52 (a) Protti, S.; Dindi, D.; Fagnoni, M.; Albini, A. Green Chem. 2009, 11, 239–249. (b)

Protti, S.; Fagnoni, M. Photochem. Photobiol. Sci. 2009, 8, 1499–1516. (c) Coyle, E. E.;

Oelgemöller, M. Photochem. Photobiol. Sci. 2008, 7, 1313–1322. 53 Cook, T. R.; Dogutan, D. K.; Reece, S. Y.; Surendranath, Y.; Teets, T. S.; Nocera, D. G.

Chem Rev. 2010, 110, 6474–6502. 54 (a) Photochemical Key Steps in Organic Synthesis, ed. Mattay J., Griesbeck A., VCH-

Wiley, Weinheim, 1994. (b) Synthetic Organic Chemistry, ed. Griesbeck A., Mattay J.,

Dekker M., New York, 2005. (c) CRC Handbook of Organic Photochemistry and

Photobiology, Horspool W., Lenci F., 2nd Ed, CRC, Orlando, 2004. (d) Ciana, C. L.;

Bochet, C. Chimia 2007, 61, 650–654. (f) Hoffmann, N. Chem. Rev. 2008, 108, 1052–

1103. (g) Esser, P.; Pohlmann, B.; Scharf, H.–D. Angew. Chem. Int. Ed. 1994, 33, 2009–

2023. (h) Oelgemoller, M.; Jung, C.; Mattay, J. Pure Appl. Chem. 2007, 79, 1939–1948. (i)

Pohlmann, B.; Scharf, H. D.; Jarolimek, U.; Mauermann, P. Solar Energy 1997, 61, 159–

168. (l) Oelgemoller, M.; Schiel, C.; Mattay, J.; Frohlich, R. Eur. J. Org. Chem. 2002,

2465–2474. (m) Schiel, C.; Oelgemoller, M.; Ortner, J.; Mattay, J.; Zimmermann, E.

Green Chem. 2005, 7, 35–38. (n) Schiel, C.; Oelgemoller, M.; Ortner, J.; Mattay, J. Green

Chem. 2001, 3, 224–228. (o) Oelgemoller, M.; Healy, N.; de Oliveira, L.; Jung, C.; Mattay,

J. Green Chem. 2006, 8, 831–834. (p) Wilkins, F. W.; Blake, D. M., Chem. Eng. Progr.

1994, 90, 41–49. 55 (a) Chin, P.; Barney, W. S.; Pindzola, B. A. Curr. Opin. Drug Discov. Dev. 2009, 12,

848–861. (b) Hartman, R. L.; Jensen, K. F. Lab Chip 2009, 9, 2495–2507. (c) Fukuyama,

T.; Rahman, M. T.; Sato, M.; Ryu, I. Synlett 2008, 151–163. 56 (a) Lapkin, A. A.; Boddu, V. M.; Aliev, G.N.; Goller, B.; Polisski, S.; Kovalev, D. Chem

Eng. J. 2008, 136, 331–336. (b) Chen, H. W.; Ku, Y.; Irawan, A. Chemosphere 2007, 69,


39

184–190. (c) Chen, D. H.; Ye, X. J.; Li, K. Y. Chem Eng Technol. 2005, 28, 95–97. (d)

Kriesel, G.; Meyer, S.; Tietze, D.; Fidler, T.; Gorges, R.; Kirsch, A.; Schäfer, B.; Rau, S.;

Chem. Ing. Tech. 2007, 79, 153–159. (e) Landgraf, S. Spectrochim. Acta, Part A, 2001, 57,

2029–2048. (f) Sugimoto, A.; Sumino, Y.; Takagi, M.; Fukuya, T.; Ryu, I. Tetrahedron

Lett. 2006, 47, 6197–6200. 57 (a) Hessel, V.; Löwe, H.; Stange, T. Lab Chip 2002, 2, 14N–21N. (b) He, Z.; Li, Y.;

Zhang, Q.; Wanga, H. App. Cat. B: Env. 2010, 93, 376–382. 58 Lu, H.; Schimdt, M. A.; Jensen, K. F. Lab. Chip 2001, 1, 22–28. 59 Wootton, R. C. R.; Fortt, R.; de Mello, A. J. Org. Process Res. Dev. 2002, 6, 187–189. 60 www.mikroglas.com 61 (a) Hook, B. D. A.; Dohle, W; Hirst P., L.; Pickworth, M.; Berry, M. B.; Booker-

Milburn, K.M J. Org. Chem. 2005, 70, 7558–7564. (b) Laurino, P.; Kikkeri, R.; Azzouz

N., Seeberger P. H. Nano Lett 2011, 11, 73–78. 62 Funayama, H.; Sugarawa, T. Bull Chem Soc. Jpn 1987, 70, 2245–2249. 63 Tsutsumi, K.; Terao, K.; Yamaguchi, H.; Yoshimura, S.; Morimoto, T.; Kakiuchi, K.;

Fukuyama, T.; Ryu I. Chem. Lett. 2010, 39, 828–829. 64 Booker-Milburn, K.; Anson, C. E.; Clissold, C.; Costin, N. J.; Dainty, R. F.; Murray, M.;

Patel, D.; Sharpe, A. Eur. J. Org. Chem. 2001, 1473–1482. 65 Li, W.; Pham, H. H.; Nie, Z.; MacDonald, B.; Güenther, A.; Kumacheva, E. J. Am.

Chem. Soc. 2008, 130, 9935–9941. 66 (a) Geyer, K.; Gustafsson, T.; Seeberger, P. H. Synlett 2009, 15, 2382–2391. (b) Watts,

P.; Haswell, S. J. Chem. Soc. Rev. 2005, 34, 235–346. 67 (a) Bogdan, A. R. Poe, S. L. Kubis, D. C. Broadwater, S. J. McQuade, D. T. Angew.

Chem. Int. Ed. 2009, 48, 8547–8550. (b) Kawaguchi, T. Miyata, H. Ataka, K. Mae, K.

Yoshida, J. Angew. Chem. Int. Ed. 2005, 44, 2413–2416. (c) Suga, S. Nagaki, A. Yoshida,

J.-I. Chem. Commun. 2003, 354–355. 68 Fukuyama, T., Chino, Y.; Katama, N.; Ryu, I. Chem. Lett. 2004, 33, 1430. 69 (a) Schwalbe, T.; Autze, V.; Hohmann, M.; Stirner, W. Org. Process. Res. Dev., 2004, 8,

440–454. (b) Roberge, D. M.; Ducry, L.; Bieler, N.; Cretton, P.; Zimmermann, B. Chem.

Eng. Technol. 2005, 28, 318–323.


40

70 (a) Gustafsson, T.; Gilmour, R.; Seeberger, P. H. Chem. Comm. 2008, 3022–3024. (b)

O'Brien, A. G.; Lévesque, F.; Seeberger, P. H. Chem. Comm. 2011, 2688–2690. 71 Schwalbe, T.; Kadzimirsz, D.; Jas, G. QSAR Comb. Sci. 2005, 24, 758–768. 72 Watts, P.; Wiles, C. Org. Biomol. Chem. 2007, 5, 727–732. 73 Martha, C. T.; Heemskerk, A.; Hoogendoorn, J.-C.; Elders, N.; Niessen, W. M. A.; Orru

R., V. A.; Irth H. Chem. Eur. J. 2009, 15, 7368–7375. 74 Brandt, J. C.; Wirth, T.; Beilstein J. Org. Chem. 2007, 5, No.30. 75 Valera, F. E.; Quaranta, M.; Moran, A.; Blacker, J.; Armstrong, A.; Cabral, J. T.;

Blackmond, D. G. Angew. Chem. Int. Ed. 2010, 49, 2478–2485.

2. Synthesis and Functionalization of Quantum Dots

41

2. Synthesis and Functionalization of

Quantum Dots

This part of the thesis was performed with Dr. R. Kikkeri (collaboration on project design

and data evaluation), Dr. A. Odedra. (attempts for the first QDs synthesis in continuous

flow microreactor) and Dr. B. Lepenies (studies on in vitro and in vivo imaging).

Parts of this chapter have been published in:

• “In Vitro Imaging and in Vivo Liver Targeting with Carbohydrate Capped Quantum

Dots” Kikkeri, R.; Lepenies, B.; Adibekian, A.; Laurino, P.; Seeberger, P. H. J. Am.

Chem. Soc., 2009, 131, 2110–2112.

• “Synthesis of Carbohydrate-Functionalized Quantum Dots in Microreactors”

Kikkeri, R.; Laurino, P.; Odedra, A.; Seeberger, P. H: Angew. Chem. Int. Ed. 2010, 49,

2054–2057.

• “Continuous Flow Reactor Based Synthesis of Carbohydrate and Hydrolipoic Acid

Capped Quantum Dots” Laurino, P.; Kikkeri, R.; Seeberger, P. H. Nat. Protoc. 2011, in

press.


42

2.1 Introduction

Nanomaterials have been applied broadly fields raging from electronics to life

sciences.1 Semiconducting nanoparticles also called “quantum dots” (QDs) are of

significant importance owing to their tunable emission wavelength and long luminescence

lifetimes (compared to organic dyes).2

QDs are semiconducting crystalline solids composed of an inorganic core stabilized

by an organic shell of surfactant (ligand). In solid state physics semiconducting materials

are characterized by a defined energy bandgap where no electron states can exist, between

the top of the valence band and the bottom of the conduction band. The electrons can

absorb photon (light) or phonon (heat) and gain enough energy to overcome this bandgap.

In QDs, the bandgap and size inversely proportional, smaller the size of the particle bigger

will be the bandgap.3

Figure 2.1 Relationship between bandgap energy and size of QDs.4

Due to the strong relationship between the size and bandgap energy, QDs with a high

controlled size and very precise photo-physical properties can be produced.5 The

tuneability of QDs in the visible and near-infrared (NIR) spectral region confers a broad


43

absorption and a narrow emission wavelength for these particles (Figure 2.2).6 These

optical properties of QDs find very important applications in various fields.

Figure 2.2 Absorbance and fluorescence spectra of CdSe/ZnS, CdTe and CdTe/CdSe quantum dots

of various sizes.

The synthesis of QDs can be accomplished by the pyrolysis of organometallic

precursors of Cd/S or Se or Te.7 QDs are commonly prepared by the reaction of dimethyl

cadmium (CdMe2),8 with trioctylphosphine selenide (TOPSe) in a mixture of coordinating

solvents such as trioctylphosphine (TOP) and trioctylphosphine oxide (TOPO) at higher

temperatures (300 °C). Improvements of the above process included the addition of non-

volatile Cd precursors (cadmium chloride, cadmium oxide, cadmium acetate or cadmium

carbonate) to a mixture of alkylphosphonic acid or alkyl amines at higher temperatures.

These changes allowed for better control of particle nucleation and growth, and

consequentially yielded QDs of different size.9, 1c

The oxidation of QDs or leaching of the core metal into the solution necessitates

the core protection with a suitable semiconductor such as ZnS, CdS, CdSe, ZnSe or

ZnTe.10 The shell can also have more specific functions depending on the bandgaps and

the relative position of the electronic energy state of the involved semiconductors. The

core/shell particles can be divided into three categories depending on the differences

between the bandgap energies of the core and shell (Figure 2.3).


44

Figure 2.3 Schematic representation of the energy-level alignment in type I and type II core/shell

QDs.11

The core material of type I QDs are characterized by a smaller bandgap energy than the

shell material (e.g. CdSe/ZnS).12 The emission peak of type I QDs depends on the size of

the core, and the shell has the function of improving the optical properties of the core and

stabilizing it from photodegradation. In type II QDs the energy level of the shell is upper or

lower compared to the bandgap of the core (e.g. CdTe/CdSe or CdSe/CdTe) and in these

particles the emission wavelength can be changed by increasing or decreasing the

thickness of the shell (700–1000 nm). The quantum yield of the type II QDs particles is

lower than to type I QDs.13 The third class, reverse type I QDs has a bandgap larger for the

core than that of the shell. The emission spectra of these particles can be varied in the

range of 450–700 nm by varying the thickness of the shell without substantial decrease in

the quantum yield of the particles (e.g. CdS/HgS, CdS/CdSe or ZnSe/CdSe).14

To further improve the stability and biocompatibility of QDs, the shell can be

capped with polymers, bio-organic compounds or silica. The properties of the shell can

improve the solubility of the nanoparticles and can act as a handle for bioconjugation while

retaining the optical properties of the QDs.15 The shell can be functionalized by

exchanging the ligand on the surface of the particle with organic componds. In the case of

polymers or dendrimers having hydrocarbon chains, the functional group can be bound to

the external shell through hydrophobic or electrostatic interactions.16 Bio-organic

compounds such as proteins, peptides, sugars,17 can also be functionalized by a thiol group

that forms a dative bond with the external surface of the shell.2a In case of capped silica

QDs a base catalyzed hydrolysis of tetraethoxysilane (TES) is carried out on the surface of

QDs to encapsulate the QDs shell.18


45

2.1.1 Previous work on the Continuous Flow Synthesis of

QDs

The increasing use of QDs for different applications necessitates the development

of a scalable procedure for their preparation.19 The scale up of conventional batch QDs

syntheses is limited, due to the lack of precise control of nucleation and QD growth. This

limitation arises in batch syntheses due to the poor homogenous heating and mixing of the

reactants. To overcome these difficulties, some research groups have reported that the high

surface-to-volume ratio of the microreactor channels enable precise temperature control as

well as efficient mixing, resulting in the preparation of QDs with high control of particle

nucleation and growth.20 In addition, injecting large volumes of precursor in the system

facilitates the scale up in a straightforward manner in the continuous flow microreactor

providing enough material for applications.

The preparation of high quality CdSe QDs in a continuous flow microreactor

requires a solution of selenium and cadmium precursors containing surfactant stabilizers.

The selenium precursor is generally tri-n-octylphosphine selenide (TOPSe) formed from

elemental Se dissolved in trioctylphosphine (TOP). To afford cadmium precursor,

cadmium sources (CdO, CdMe2, and Cd(Ac)2) and a surfactant (stearic acid or oleic acid)

are typically dissolved in an high-boiling solvent like oleylamine, squalene, or toluene.

The reactants are introduced into the system by two syringes that are filled with selenium

and cadmium precursor respectively and then simultaneously injected into the microreactor

or pre-mixing the precursors and then injecting using one single syringe. The synthesis of

QDs occurs in the heated section of the microreactor followed by the quenching of the

reaction at room temperature in the later part of the microreactor. The approach mainly

used for the synthesis of QDs in continuous flow is the two phase flow (gas-liquid phase).

In the gas-liquid phase a premixed solution of the precursors is first injected followed by

addition of an inert gas (e.g. Argon)21 and in the liquid phase the two precursor solutions

are delivered into the microreactor in two separated flows.22


46

Figure 2.4 Schematic representation of QDs synthesis in a continuous flow reactor using gas-liquid

or liquid phases.

The first two-phase synthesis of QDs in continuous flow was reported by Bawendi

et al.21b who used a gas-liquid segmented flow microreactor for the synthesis of QDs at

high temperature. The advantage of using a two phase system is that every droplet

undergoes a re-circulatory motion which results in better mixing of the reactants (Figure

2.5).

Figure 2.5 Schematic representation of recirculatory motion in a two phase reaction (segmented

flow).

Bawendi et al.21b achieved the synthesis of monodisperse CdSe QDs exploiting the two

phase system at 280 °C. The inert gas used in the study was argon and the liquid phase

consisted in a mixture of cadmium 2,4-pentanedionate in oleic acid, squalane, oleylamine

and SeTOP. A systematic study for screening residence times and different ratio between

gas and liquid phase was facilitated by the use of a continuous flow microreactor. This

investigation illustrated how a microreactor synthesis facilitated the systematic study of the


47

mechanism and control of QD formation. Later studies used this two phase continuous

flow microreactor system for the synthesis of QDs.21a, b, f

Compared to the synthesis of QDs under homogenous condition the two phase

approach requires complex experimental setup and involves longer residence time as major

disadvantages. To achieve narrow residence time distribution for two-phase systems higher

temperature are involved. Therefore, the synthesis and functionalization of QDs under

homogeneous condition is very challenging. In this chapter, we present the synthesis of

type-I core/shell crystals (CdSe/ZnS and CdTe/ZnS) and their functionalization in batch

and an improved synthesis in continuous flow in homogenous phase at lower temperature.

The obtained QDs were later functionalized and used for binding assays and imaging

studies.


48

2.2 Quantum Dots Batch Synthesis

Our initial studies were directed toward the synthesis of QDs using a batch process.

CdSe/ZnS quantum dots with an emission maximum of 635 nm were chosen for the

synthesis.6 The QDs were prepared by heating a mixture of cadmium oxide,

trioctylphosphine oxide (TOPO), lauric acid, and hexadecylamine under nitrogen at 310

°C. A solution of tri-n-octylphosphine selenide (TOPSe) was rapidly injected into the

reaction mixture at 310 °C which was immediately cooled down to 270 °C (Scheme 2.1).

Scheme 2.1 QDs synthesis in batch reactor. (a) 310 °C to 270 °C 5 min; cooled to 60 °C; (b) 160

°C, 1h (TOP: tri-n-octylphosphine; TOPO: tri-n-octylphosphine oxide).

The reaction was stirred for five minutes at 270 °C and then cooled to 60 °C to stop

particle growth (Figure 2.6).2b After purification of the CdSe QDs, the ZnS precursor was

added to a solution of CdSe at 160 °C to afford CdSe/ZnS QDs. 2b The resulting CdSe/ZnS

QDs were stabilized by a layer of surfactant (TOPO/TOP).

The carbohydrate–PEGylated linker and PEGylated linker (polyethylenglycol) that

were required for the functionalization of CdSe/ZnS QDs, were then synthesized. The

synthesis of the PEG derivatives begin from PEG2000 1, which was coupled to thioctic

acid to give ester 2 (Scheme 2.2).


49

Scheme 2.2 Synthesis of DHLA-PEG2000 (3) and DHLA-PEG2000-NH2 (6). Reagents and

conditions: (a) DL-Thioctic acid, DCC, DMAP, rt, 12 h (12%); (b) NaBH4, MeOH/H2O, 0°C, 3h

(36%); (c) MsCl, TEA, NaN3, rt, 12 h (84%); Ph3P, H2O, rt, 12 h (63%); (d) Thioctic acid, DIC,

NHS, rt, 12 h (48%); (e) NaBH4, MeOH/H2O, rt, 3h (47%).

Subsequent reductive opening of 2 afforded 3, which was used as control during imaging

studies. On the other hand reaction of 1 with methanesulfonyl chloride, sodium azide and

triphenylphosphine afforded diamino-PEG2000 4. Monoacylation of 4 by treatment with

thioctic acid N-hydroxysuccinimide (NHS) ester afforded 5, which underwent reductive

ring opening to give 6. The DHLA-PEG2000-NH2 6 was used for functionalization of QDs

and further modification.

The thiol linked monosaccharide mannose 10 and galactose 14 that were required

as sugar moiety for the functionalization of QDs, were prepared. The corresponding

commercial available monosaccharides 7 and 11 were acetylated and then glycosylated

using 2-[2-(2-chloroethoxy)ethoxy]ethanol and BF3·Et2O (Scheme 2.3).


50

Scheme 2.3 Synthesis of mannose-SH (10) and galactose-SH (14) derivatives. Reagents and

conditions: (a) Ac2O/Py, rt, 1 h (quant.); (b) 2-[2-(2-chloroethoxy)ethoxy]ethanol, BF3·Et2O, rt,

overnight (64%); (c) KSAc, DMF, rt, 3 h (84%); (d) NaOMe, MeOH Amberlite 120 (H+), rt, 1 h

(87%); (e) Ac2O/Py, rt, 1 h (quant.); (f) 2-[2-(2-chloroethoxy)ethoxy]ethanol, BF3·Et2O, rt,

overnight (68%); (g) KSAc, DMF, rt, 3 h (82%); (h) NaOMe, MeOH Amberlite 120 (H+), rt, 1 h

(89%)

Substitution of the chloride group with the thio-acetyl residue followed by basic

deacetylation under standard conditions afforded compounds 9 and 13 respectively.

The synthesis of thiol linked galactosamine 19 was achieved by treatment of

peracetylated galactal 15 with trimethylsilyl azide, bis(acetoxy)iodobenzene and diphenyl

diselenide (Scheme 2.4).

Scheme 2.4 Synthesis of galactosamine-SH (19). Reagents and conditions: (a) TMSN3, PhI(AcO)2,

diphenyl diselenide, -30 °C, 12 h (97%); (b) 2-[2-(2-chloroethoxy)ethoxy]ethanol, N-

Iodosuccinimide, TfOH, rt, overnight (61%); (c) KSAc, DMF, rt, 12 h (78%); (d) NaOMe, MeOH,

Amberlite 120 (H+), rt, 2 h (80%); (e) PMe3, H2O, rt, 12 h (81%).


51

Glycosylation with 2-[2-(2-chloroethoxy)ethanol followed by substitution gave the thiol

derivative 18, before deprotection and reduction of which provided access to 19.

The prepared CdSe/ZnS nanoparticles were functionalized by the exchange of the

ligand (TOP/TOPO) with 6 to give 22 (Scheme 2.5).

Scheme 2.5 QD functionalization. Reagents and conditions: (a) Chloroform/ethanol, 60 °C, 12 h

CdSe/ZnS; (b) borate buffer (pH = 8.5), overnight; (c) borate buffer (pH = 7.3), 3 h, rt,

monosaccaride linked thiols.

The PEGylated particles 22 were coupled with 4-maleimidopropanoic acid NHS ester 21

to give 23, which upon treatment with thiol linked monosaccharide gave functionalized

QDs 24.


52

Figure 2.6 Carbohydrate PEGylated QDs. Mannose-PEG2000-QD 24a, Galactosamine-PEG2000-

QD 24b, Galactose-PEG2000-QD 24c.

CdSe QDs exhibited absorption at 400 nm and emission at 635 nm, which remained

constant for the CdSe/ZnS and the coated particles 24. The quantum yield (Φ), calculated

according to equation 4 using a standard sample (Fluorescein) was 0.33 for core particles

and coated particles.

(4)

Φref = 0.93 (quantum yield of the reference in this case fluorescence is used as a reference sample);

S = area under the emission peak.

The particle size was calculated from the fluorescence spectra by extrapolating from the

full width at half maximum (FWHM) values, which showed the formation of particles with

15–20 nm diameters.

The main problem during the synthesis in batch was limited control over mixing and the

temperature of the reaction mixture that did not allow for a precise control of the particle

size. The difficulties encountered in cooling down the batch system allowed only for the

synthesis of particles with long emission wavelengths (635 nm).


53

2.3 Continuous Flow QD Synthesis and

Functionalization

The problems associated with the synthesis of QDs in a batch process mainly due

to temperature control, homogenous mixing and reproducibility were overcome by

synthesizing of QDs in continuous flow microreactor. In this section the three steps

synthesis in homogenous phase of CdSe or CdTe/ZnS quantum dots capped with

monosaccharides or dihydrolipoic acid has been reported. In the first approach every

intermediate of the synthesis step was purified, isolated and analyzed and in the second

approach only the final product was isolated and analyzed (Figure 2.7).23 Furthermore,

different residence time and temperature for the core formation were screened and resulted

in the formation of narrow distributed QDs with different sizes and optical properties of

the core.24

Figure 2.7 Continuous microreactor setup for the synthesis of functionalized QDs (OA: oleic

acid; TOP: tri-n-octylphosphine).


54

2.3.1 System Set-up

A commercially available 1 mL glass microreactor (Syrris FRX volcano) was used

for QD synthesis.25 The glass microreactor is composed of a micromixer, which allows

controlled and homogeneous mixing of different starting materials and a tubular reactor for

that enables to reach laminar flow and homogenous heating of the reaction mixture (Figure

2.8).

Figure 2.8 Schematic overview of continuous flow system used for QD synthesis. (1) Syringe

pump; (2) Gas-tight syringes; (3) and (3’) PTFE tubing; (4) and (4’) Inlets; (5) Hot plate; (6)

Thermocouple; (7) Mixing system; (8) Microreactor ; (9) Outlet; (10) PTFE tubing with back

pressure regulator; (11) Collection vessels.

The injection system consists of a syringe pump that controls the flow rate and injection

volume according to the residence time required. Using two gas-tight syringes the starting

materials were simultaneously injected.

The microreactor was washed with the reaction solvent to equilibrate the system and

warmed to the required temperature prior to performing the reaction. It is very important to

reach a stable temperature before starting the run as any oscillations of the reaction

temperature can produce broader and irreproducible distribution particle sizes. In addition,


55

the first 0.1 mL and the last 0.1 mL of each run were discarded to avoid dilution effects

that can reduce the quality and the consistency of the QDs.

2.3.2 QD Synthesis

Initially, CdSe and CdTe QDs with different emission maxima were prepared by

injection of a 1:1 ratio of cadmium precursor and selenium or tellerium precursor

(Figure 2.9).

Figure 2.9 First step of the QDs synthesis: the core particle (OA: oleic Acid; TOP: tri-n-

octylphosphine).

The cadmium precursor was prepared by the addition of oleic acid (OA) and

dedecanoic acid to a solution of cadmium oxide dissolved in lauric acid at 150 °C. The

selenium and tellerium precursors were prepared by dissolving elemental Se or Te

powder in tri-n-octylphosphine (TOP). Reaction times varied from three to thirty

minutes (3, 10, 20 and 30 min). In case of CdTe, three residence times (3, 10 and 20

min) were investigated, while for CdSe four residence times (3, 10, 20 and 30 min) and

four different temperatures at 3 min residence time (160, 180, 200 and 220 °C) were

investigated.

The produced CdSe or CdTe core QDs were used for further modification via of

two approches. In the first approach, the core was directly injected into the second

reactor, whereas in the second approach the core was precipitated using a mixture of

methanol/chlorofom/hexane, characterized and then injected into the second reactor.


56

A freshly prepared mixture of hexamethyl disilathiane, TOP, diethylzinc in toluene and

zinc sulfide was injected simultaneously with the QDs core precursor dissolved in toluene

and TOP into the microreactor. The temperature was maintained at 90 °C with a residence

time of 30 min (Figure 2.10).

Figure 2.10 Second step of the QD synthesis: the core/shell particle (OA: oleic Acid; TOP: tri-n-

octylphosphine).

The resulting ZnS coated CdTe or CdSe QDs were purified by precipitation using

methanol/chloroform or used directly as it was in the second approach of synthesis

process.

2.3.3 QD Functionalization

The ZnS capped CdTe or CdSe QDs were functionalized by ligand exchange with

pyridine in continuous flow. Freshly prepared oleic acid capped CdTe or CdSe /ZnS QDs

were dissolved in pyridine and injected into the microreactor at 60 °C for 30 min. The

resulting pyridine capped QDs served to install carbohydrates on the QD nanoparticles. A

mixture of freshly prepared dihydrolipoic acid, mercapto-PEG α-mannose 10 or mercapto-

PEG β-galactose 14 or dihydrolipoic acid in an ethanol/ethylenedichloride mixture (1:1)

and a solution of pyridine-coated QDs in dichloroethane were simultaneously injected into

the microreactor at 60 °C with 30 min residence time.


57

Figure 2.11 Third step of the QD synthesis: capping of core shell particles (EDC:

Ethylenedichloride).

The resulting QDs were purified by precipitation with hexane/chloroform/methanol and

redispersed in water for further characterization. All mannose, galactose and dihydrolipoic

acid-modified QDs had a narrow size distribution, a spherical shape and emission spectra

as the CdSe/ZnS QDs.


58

2.4 Photophysical Properties of QDs

Produced in Continuous Flow

In order to understand how different parameters could influence the synthesis and

subsequently the properties of the final product, the optical properties of the QDs

generated using the microreactor were determinated. The CdSe or CdTe core of QDs

particles were isolated and characterized by UV-Vis spectroscopy. The UV-Vis spectra of

the CdSe and CdTe core particles prepared at different residence time at 160 °C showed a

shift of the curves at higher wavelength and the absorbance peaks (λmax) increased

proportionally with the residence times which were directly related to the size, shape and

composition of the QDs (Figure 2.12).26

Figure 2.12 UV-visible absorption spectra of (a) CdSe QDs and (b) CdTe QDs in chloroform.

The optical properties of QDs showed a time dependent bathochromic shift in the band-

edge emission and enhanced intensity. The fluorescence emission peaks of CdTe QDs

with maximum emission at 521, 575, and 598 nm were sharp, thus indicating a very

narrow size distribution. (Figure 2.13). More importantly, an increased quantum yield

was observed with residence times increasing from 3 to 20 min (Table 2.1).


59

Figure 2.13 Fluorescence spectra of CdTe QDs in chloroform prepared at different residence

times: 3, 10 and 20 min.

Table 2.1 Photophysical properties of CdTe QDs.

Entry Time (min) Temperature (°C) QDs diameter (nm) Quantum Yield (Φ)

1 3 160 1.67 ± 0.27 14 ± 1

2 10 160 3.01 ± 0.36 21 ± 1

3 20 160 3.24 ± 0.49 23 ± 1

The photoluminescence peaks of CdSe QDs were sharp indicating the narrow size

distribution of the QDs, as evident from the observed 40 to 50 nm FWHM (full width at

half maximum) of the band-edge luminescence (Figure 2.14).


60

Figure 2.14 Fluorescence spectra of CdSe QDs prepared at different residence times at 160 °C

with different residence times: 3, 10, 20 and 30 min.

An attempt to increase the reaction time to 30 min resulted in an increase in the FWHM

(90 nm) and a decrease in quantum yield. It became obvious that longer reaction times

allowed for saturated nucleation to occur after 20–30 min at 160 °C leading to a

alterated dispersity and subsequent decrease in quantum yield (Table 2.2).


61

Table 2.2 Photophysical properties and yields of CdSe QDs prepared at different

temperature and residence times. FWHM values calculated according to the fluorescent

spectra. Quantum yield was compared to fluorescein at 470 nm (0.93).

Entry Time

(min)

Temperature

(°C)

Particle

diameter (nm)

FWHM

(nm)

Quantum

Yield (Φ)

Yield

(%)

1 3 160 1.45 ± 0.22 40–50 8 ± 1 72 2 10 160 1.83 ± 0.26 40–50 11 ± 1 71 3 20 160 2.64 ± 0.43 40–60 19 ± 1 67 4 30 160 3.36 ± 0.34 40–90 15 ± 1 66 5 3 180 1.87 ± 0.39 40–45 12 ± 1 72 6 3 200 2.61 ± 0.24 35–40 16 ± 1 79 7 3 220 3.06 ± 0.41 30–35 21 ± 1 88

Constant shift of the absorption and emission curves of the CdSe or CdTe particles

according to the residence time was a clear indication for the formation of bigger sized

QDs with increased reaction times.

It was also evident that higher temperature resulted in the formation of QDs with

narrower size distribution, as shown by the fluorescence spectra (Figure 2.15). The

FWHM values decreased with an increase in temperature and the diameter increased

thus indicating a complete saturation of the particles nucleation (Table 2.2)


62

Figure 2.15 Fluorescence spectra of CdSe QDs prepared at different temperatures (--- 160 °C, ---

180 °C, ---200 °C and ---220 °C) with 3 min residence time.

The yield of the reaction was calculated according to equation 5 using a known

concentration of the starting materials, in this case the concentration of Cd precursor was

used as a reference sample (eq. 5)

(5)

The efficiency of the QDs preparation process, determined by the yield of QDs, were

satisfying in most of the cases (ca. 70%). A direct correlation with temperature was

observed wherein higher yields were obtained at higher temperatures (Table 2.2). The UV-

Vis spectra also demonstrated that the core/shell QDs exhibited similar trends to the core

QDs. More importantly, the quantum yield increased from 23% to 31% due to the

stabilization of the core by the ZnS shell.

TEM analysis of the CdTe, CdTe/ZnS and CdTe/ZnS/mannose QDs showed that

the QDs did not change in shape and geometry even if covered by a layer of ZnS or ZnS

and mannose linker. The QDs maintained their monodispersity and spherical shape (Figure

2.16).

Figure 2.16 TEM images of CdTe core QDs (a); CdTe/ZnS QDs (b); CdTe/ZnS/mannose QDs (c);

(scale bar 50 nm).


63

2.5 Biological Studies Using

Carbohydrate-Capped QDs

The development and use of quantum dots for biological and biomedical

applications is a fast emerging field of nanotechnology. 27 The unique optical properties of

these nanometer-sized semiconducting crystals make them an interesting fluorescent tool

for in vivo and in vitro imaging as well as for sensor applications.28 We used sugar-capped

PEGylated QDs for targeting hepatic carcinoma cells and as fluorescent biosensor for the

detection of carbohydrate-lectin interactions.

The asialoglycoprotein (ASGP-R) that is only expressed in hepatic cells,29 is a

receptor that specifically binds to galactosyl-terminal proteins. We utilized the galactose

capped QDs to target ASGP-R mediated endocytosis using the carcinoma cell line

HepG2.2f, 30 The CdSe/ZnS QDs prepared in batch and functionalized with galactosamine

(GalN)-PEG2000 24b and PEG2000 linker 24d, were exploited in this study (Figure 2.17).

Figure 2.17 PEGylated QDs involved in the biological studies: Galactosamine-PEG2000-QD 24b,

PEG2000-QD 24d.

HepG2 cells were incubated overnight with 20 nmol of PEG2000 QDs 24d or 20 nmol of

GalN-PEG2000 QDs 24b. As negative control, PBS was added to the cells. After washing,

cells were collected and uptake of QDs was measured by flow cytometry. Compared to the

negative control (gray, Figure 2.18), PEG2000 QDs 24d were taken up only to a minor

extent (dashed line, Figure 2.18). In contrast, GalN-capped QDs 24b were taken up to a


64

higher extent by HepG2 cells (solid line, Figure 2.18). Further studies demonstrate the

specific up–take of QDs by liver cells in vivo.2f

Figure 2.18 Specific uptake of D-GalN capped QDs by HepG2 cells.

As a proof-of-principle, QDs prepared and functionalized in continuous flow were

used as fluorescent biosensors for specific carbohydrate-protein interaction. The lectin

concanavalin A (ConA), a protein well known to specifically bind mannose, was incubated

with α-mannose-functionalized QDs 25 and with β-galactose QDs 26.


65

Figure 2.19 Mannose-QDs 25 and Galactose-QDs 26.

Immediately after incubation, the binding between CoA and α-mannose-QDs resulted in

turbidity whereas β-galactose-QDs did not (Figure 2.20). The disappearance of the

turbidity upon the addition of a concentrated solution of mannose (0.1 M), a competitive

inhibitor, demonstrated that the turbidity was caused by specific carbohydrate-protein

interactions.

Figure 2.20 Kinetics of turbidity by α-mannose (black line) and β-galactose (red line) CdTe/ZnS

capped QDs.

In addition, the turbidity caused by the specific interaction between α-mannose

capped-QDs and ConA resulted in a decreased intensity in the fluorescence spectra (Figure

2.21). The specific agglutination of α-mannose coupled QDs clearly shows the

bioavailibity for protein binding of the monosaccharide on the surface of QDs.


66

Figure 2.21 Fluorescence spectra of CdTe/ZnS α-mannose capped QD before (black line) and after

addition of ConA lectin (pink line).


67

2.6 Conclusions

The continuous flow three step syntheses of CdTe and CdSe/ ZnS/ monosaccaride or

dihydrolipoic acid capped QDs was successfully performed in a continuous flow

microreactor to overcome the limitations of QD syntheses in batch. The microreactor

further allowed the syntheses of different sized QDs in efficient, reproducible and scalable

fashion, and in an inert atmosphere and safe working conditions. The first step syntheses

was carried out under homogeneous conditions and at lower temperature compared to the

previously reported syntheses (160–220 °C vs 300 °C). To the best of our knowledge we

believe that the lower temperatures suppressed the continuous particle nucleation through

the high control of the precursor consume. After screening different residence times and

temperatures the CdSe or CdTe QDs were capped with a shell of ZnS at 60 °C in

continuous flow microreactor.

The QDs were then conjugated with thiol linked monosaccharides, and used as

sensors to detect the specific interactions with carbohydrates binding proteins. These

studies showed that QDs are very potent tools for biological studies because of their

extraordinary optical properties and minimal photodegradation. However the major

drawback of QDs is the toxicity, and many efforts have been made to synthesize good

alternative QD candidates (e.g. InP QDs31) but these particles are not comparable to the

optical properties of CdSe and CdTe.


68

2.7 Experimental Part

2.7.1 General Information

All the reagents were purchased from Aldrich and used as received. Dichloromethane

(CH2Cl2) was purified by a Cycle-Tainer Solvent Delivery System. Triethylamine was

distilled over CaH2 prior to use. Analytical thin layer chromatography (TLC) was

performed on Merck silica gel 60 F254 plates (0.25 mm). Compounds were visualized by

UV irradiation or dipping the plate in CAN solution followed by heating. Flash column

chromatography was carried out on Fluka Kieselgel 60(230-400 mesh). Absorption spectra

were recorded using a Varian CARY 50 spectrophotometer fitted with Hellma optical

fibers (Hellma, 041.002-UV) and an immersion probe made of quartz suprazil (Hellma,

661.500-QX). Fluorescence emission spectra were recorded on a Perkin-Elmer LS-50B

spectrofluorometer in Chloroform. IR spectra were recorded on a Perkin-Elmer 1600 FTIR

spectrometer. Optical rotation measurements were conducted using a Perkin-Elmer 241

polarimeter. TEM images were performed with EFTEM Omega 912, Zeiss AG apparatus.

1H and 13C NMR spectra were recorded on a Varian VXR-300 (300 MHz) orBruker

DRX500 (500 MHz) spectrometer. High-resolution mass spectra (HR MALDI MS) were

recorded using mass Spectrometry service at the Laboratory for Organic Chemistry (ETH

Zurich).


69

2.7.2 Description of the Experimental Set Up

The flow reactor setup consists of commercially available parts. A syringe pump injection

system (Harvard PHD 2000, cat. # 702012) is used for reagent delivery and a glass

microreactor (Syrris FRX reactor, cat. # 2100146) as the reaction device, which is held on

a metal support (Syrris FRX Volcano, cat. # 2101346) mounted on a hotplate (Syrris FRX

hotplate, cat. # 2101310). The dimensions of the glass device are 2.50 x 8.5 cm (width x

length), with a glass rectangular micromixing zone of 161 x 1240 x 536 x103 µm (width x

depth x length) and inner channel dimensions of microreactor are 391 x 1240 x 1844 x 103

µm (width x depth x length). The total volume of the continuous flow device is 1 mL. The

reactor is connected with two inlets to the gastight syringes (Hamilton syringes 0.5 mL) via

PolyTetraFluoroEthylene (PTFE) tubing (1/16″ OD, 0.030 ID) and the outlet is connected

to the collecting flask through PTFE tubing (1/16″ OD, 0.030 ID) as well. Injection of

reagents is controlled by adjusting flow rate of in the syringe pump. The syringes are

refilled manually and the scale up is due to the use of bigger syringes. For every 1-mL run

the first 0.1 mL and the last 0.1 mL were discavered and only the middle fraction (0.8 mL)

was collected.

2.7.3 Batch Synthesis of CdSe/ZnS QDs

CdSe/ZnS quantum dots with an emission maximum centered at 635 nm were synthesized

via pyrolysis of the organometallic precursors.2b Cadmium oxide (0.05 g, 4.0 mmol),

trioctylphosphine oxide (TOPO) (2 g, 5.2 mmol), lauric acid (0.22 g, 1.1 mmol), and

hexadecylamine (2 g, 8.3 mmol) were mixed and heated to 310 °C, under nitrogen flow,

for 30 min. A selenium stock solution, prepared from selenium (0.03 g, 4.0 mmol) in

trioctylphosphine (TOP) (2 mL, 5.4 mmol), was rapidly injected into the reaction mixture

at 310 °C. The reaction mixture was then immediately cooled to 270 °C, the stirring was

stopped after 5 min and the reaction vessel cooled to 60 °C. CdSe-QD cores were purified


70

by precipitation in methanol/CHCl3. The CdSe particles were subsequently coated with

ZnS layers by dropwise addition of a mixture of TOP (3 mL), diethylzinc in hexane (2.0

mL), and hexamethyldisilathiane (0.2 mL) over 1 h at 160 °C. Following cooling to room

temperature, the 3 concentration of the resulting CdSe/ZnS QDs solution was then

estimated.32

2.7.4 Synthesis of the PEG-Linker

TA-PEG2000-OH (2).

DL-thioctic acid (0.13 g, 0.63 mmol) and PEG2000 1 (12.0 g, 6.0 mmol) were added to a

solution of DCC (0.17 g, 0.83 mmol) and N,N dimethylaminopyridine (0.10 g, 0.80 mmol)

in dichloromethane at 0 °C, stirred at room temperature for 12 h and concentrated in vacuo.

The crude residue was purified over silica gel using EtOAc/MeOH (2-5%) to afford ( TA-

PEG2000-OH 2 1.5 g, 12%). 1H NMR (300 MHz, CD3OD): δ 1.42-1.51 (m, 2H); 1.61-

1.70 (m, 4H); 1.87-1.95 (m, 2H); 2.35 (t, J = 7.5 Hz, 2H); 2.42–2.48 (m, 1H); 2.58–2.62 (t,

J = 6.0 Hz, 1H); 3.12–3.18 (m, 2H); 3.41 (t, J = 6.0 Hz, 1H); 3.55–3.72 (m, 164H); 3.87 (t,

J = 6.0 Hz, 1H); 4.24 (t, J = 5.4 Hz, 2H); 13C-NMR (75 MHz, CD3OD); δ 22.3, 24.6, 26.6,

34.1, 88.7, 39.4, 42.7, 53.4, 61.7, 69.3, 70.5, 71.2, 71.3, 72.2, 72.7; HRMS-MALDI (m/z):

[M+Na]+ Calculated for C80H158O38S2Na + (C2H4O)n 1813.9770 + (44.026)n; Found:

1813.972 + (44.032)n.


71

DHLA-PEG2000-OH (3). TA-PEG2000-OH 2 (1.0 g, 0.5 mmol) was dissolved in an ethanol/water (1:1) mixture and

cooled to 0 °C. Sodium borohydride (20 mg, 0.54 mmol) was added and stirred at 0 °C for

3 h. The solution was neutralized with 1 N HCl solution, extracted with chloroform, dried

over sodium sulfate, concentrated and purified over silica gel using CH2Cl2/CH3OH (2-

5%) to afford 0.45 g (46%) of DHLA-PEG2000-OH 3. 1H NMR (300 MHz, CD3OD): δ

1.28–137 (m, 2H); 1.48-1.75 (m, 4H); 1.82–1.92 (m, 1H); 2.34 (t, J = 7.5 Hz, 2H); 2.42–

2.48 (m, 2H); 2.68–2.672 (m, 2H); 2.84–2.92 (m, 1H); 3.55-3.72 (m, 164H); 4.21 (t, J =

6.0 Hz, 1H); 13C NMR (75 MHz, CD3OD); δ 21.8, 24.1, 26.1, 33.5, 38.3, 38.9, 42.4, 61.2,

63.1, 68.7, 69.9, 70.2, 71.3, 71.8, 72.2, 72.5; HRMS-MALDI (m/z): [M+Na]+ Calculated

for C80H160O38S2Na + (C2H4O)n: 1815.9770 + (44.026)n; Found: 1815.975 +

(44.031)n.

Diazido-PEG2000 (4a).

PEG2000 1 (10 g, 5.0 mmol) was coevaporated with 50 mL toluene to remove water, and

then 50 mL of THF and methane sulfonyl chloride (1.22 mL. 15.0 mmol) were added. The

solution was stirred at 0 °C, and a solution of triethylamine (2.29 mL, 16.5 mmol) in THF

(20 mL) was added dropwise over 30 min. After 1 h the ice bath was removed and the

mixture was stirred overnight. The formed solid was dissolved by the addition of water (50

mL), forming two layers, that were cooled on an ice bath. Then, 1 N solution of sodium

bicarbonate (5 mL) was added followed by sodium azide (1.3 g, 20.0 mmol). THF was

removed and the water layer was refluxed for 24 h. The aqueous layer was extracted with

dichloromethane (4 x 20 mL). Each dichloromethane layer was washed with an additional

100 mL saturated NaCl solution. The combined organic extracts, were dried (Na2SO4) and

concentrated to afford (8.4 g, 84%) diazido-PEG2000 4a as white solid flakes. 1H NMR

(300 MHz, CDCl3): δ 3.37 (t, J = 5.1 Hz, 4H); 3.63–3.68 (m, 168H). 13C NMR (75 MHz,

CDCl3); δ 50.7, 67.8, 68.1, 68.9, 69.6, 70.5, 70.6, 71.6, 72.6, 3.0, 73.2.


72

Diamino-PEG2000 (4). Diazido-PEG2000 4a (8 g, 4.0 mmol) was dissolved in 50 mL THF and

triphenylphosphine (3.14 g, 12.0 mmol) was added. The solution was stirred at room

temperature for 10 h, water (3 mL) was added and stirred overnight. THF was removed in

vacuo, and additional 100 mL of water was added. The triphenyl phosphine oxide

precipitate was removed by filtration and the water was removed in vacuo to afford

(diamino- PEG2000 4 5.1 g, 63%) as white solid. 1H NMR (300 MHz, CD3OD): δ 2.77 (t,

J = 5.4 Hz, 4H); 3.52 (t, J = 5.4 Hz, 4H); 3.63–3.68 (m, 164H). 13C NMR (75 MHz,

CD3OD); δ 39.7, 68.1, 68.6, 68.8, 69.6, 70.7.

TA-PEG2000-NH2 (5).

DL-thioctic acid (0.3 g, 1.45 mmol) and N-hydroxysuccinimide (0.16 g, 1.6 mmol) were

added to diisopropyl carbodiimide (0.27 g, 1.74 mmol) in CH2Cl2 and stirred at 0 °C for 2

h. This mixture was added dropwise to diamino-PEG2000 4 (4.0 g, 2.0 mmol) in 30 mL of

THF. The reaction mixture was stirred at room temperature for 12 h and concentrated in

vacuo. The crude residue was purified over silica gel using CH2Cl2/CH3OH (10-15%) as

eluent to afford 2.25 g (48%) of TA-PEG2000-NH2 5. 1H NMR (300 MHz, CDCl3): δ

1.42–1.51 (m, 2H); 1.62–1.70 (m, 4H); 1.89–1.93 (m, 2H); 2.17–2.2 (br.s, 2H); 2.42–2.48

(m, 1H); 2.58–2.62 (m, 1H); 3.12–3.18 (m, 2H); 3.41–3.46 (m 4H); 3.55–3.72 (m, 164H); 13C NMR (75 MHz, CDCl3); δ 24.5, 24.9, 28.6, 33.8, 34.5, 38.4, 40.1, 61.56, 63.4, 68.5,

68.6, 69.1, 69.7, 70.5, 71.2, 71.3, 72.5, 73.3; HRMS-MALDI (m/z): [M+Na]+ Calculated

for C82H164O37N2S2Na +(C2H4O)n 1856.03466 + (44.026)n; Found: 1856.0346 +

(44.032)n.

DHLA-PEG2000-NH2 (6).

TA-PEG2000-NH2 5 (2.0 g, 0.5 mmol) was dissolved in an ethanol/water (1:1) mixture,

cooled to 0 °C and sodium borohydride (20 mg, 0.54 mmol) was added and stirred at 0 ºC

for 3 h and overnight stirring at room temperature. The solution was extracted with

chloroform, dried over sodium sulfate and concentrated to afford (0.94 g, 47%) of DHLA-

PEG2000-NH2 6. 1H NMR (300 MHz, CD3OD): δ 1.28-137 (m, 2H); 1.48–1.75 (m, 4H);

1.82-1.92 (m, 1H); 2.34 (t, J = 7.5 Hz, 2H); 2.42-2.48 (m, 2H); 2.68-2.672 (m, 2H); 2.84–

2.925 (m, 1H); 3.55–3.72 (m, 164H); 13C-NMR (75 MHz, CD3OD); δ 21.6, 24.2, 26.1,

33.5, 38.2, 38.9, 42.5, 61.2, 63.1, 68.9, 69.9, 70.2, 71.2, 71.8, 72.1, 72.5, HRMS-MALDI


73

(m/z): [M+Na]+ Calculated for C82H166O37N2S2Na +(C2H4O)n: 1858.03466 + (44.026)n

Found: 1858.0342 + (44.022)n.

2.7.5 Synthesis of Thiol Linked Monosaccharide

General procedure A. The acetylated sugar (1 equiv) and bis(2-chloroethoxy)ethanol (3

equiv) were dissolved in 10 mL CH2Cl2 and cooled to 0 ºC. BF3.Et2O (4 equiv) was added

and stirred at room temperature for 48 h. The reaction was quenched with triethylamine

and extracted with dichloromethane/water mixture. The organic layer was concentrated

and purified over silica gel to afford sugar-PEG-Cl.

General procedure B. Sugar-PEG-Cl (1 equiv) and potassium thioacetate (3 equiv) were

dissolved in 10 mL of anhydrous DMF and stirred at room temperature for 12 h. Ethyl

acetate was added and the organic layer was washed five times with water. Purification by

flash silica column chromatography afforded sugar-PEG-SAc.

General Procedure C. Sugar-PEG-Sac (1 equiv) and sodium methoxide (10 mol%) were

dissolved in methanol (10 mL) and stirred at room temperature for 15 min. The reaction

mixture was neutralized with Amberlite-H+ and solvent was evaporated to afford the free

thiol.

2-(2-(2-Chloroethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-α-D-mannopyranoside

(9a).

General procedure A with with 1,2,3,4,6-penta-O-acetyl-D-mannopyranoside 8 (1.5 g, 3.8

mmol), 2-[2-(2-Chloroethoxy)ethoxy]ethanol (1.92 mL, 11.4 mmol) and BF3 Et2O (2.43

mL, 15.2 mmol), followed by purification over silica gel using hexane/ethyl acetate

(40:60) afforded 1.23 g (64%) of 2-(2-(2-Chloroethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-

acetyl-α-D-mannopyranoside 9a [α]D24 = +0.17 (c = 1.0, CHCl3); 1H NMR (300 MHz,

CDCl3): δ 1.91 (s, 3H); 1.96 (s, 3H); 2.01 (s, 3H); 2.07 (s 3H); 3.53–3.72 (m, 11H); 4.02-


74

4.04 (m, 2H); 4.18 (dd, J =5.4, 4.2 Hz, 1H); 4.79 (d, J = 2.1 Hz, 1H); 5.17–5.26 (m, 4H); 13CNMR (75 MHz, CDCl3): δ 20.5, 42.7, 61.6, 66.4, 66.2, 67.3, 68.3, 68.5, 69.1, 69.4,

69.6, 70.3, 71.3, 72.5, 97.6, 97.7, 169.6, 169.8, 169.9, 170.5; HRMS-MALDI (m/z): [M]+

Calculated for C20H31O12Cl: 498.1504; Found: 498.1501.

2-(2-(2-Thioacetylethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-α-D-mannopyranoside

(9).

General procedure B with 2-(2-(2-chloroethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-α-D

galactopyranoside 9a (0.5 g, 1.0 mmol) and potassium thioacetate (0.29 g, 3.0 mmol),

followed by purification over silica gel using hexane/ethylacetate (40:60) afforded 0.51 g

(84%) of 2-(2-(2-Thioacetylethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-α-D-

mannopyranoside 9 [α]D24 = +0.19 (c = 1.0, CHCl3); 1H NMR (300 MHz, CDCl3): δ 1.98

(s, 3H); 2.03 (s, 3H); 2.10 (s, 3H); 2.15 (s 3H); 2.33 (s, 3H); 3.11 (t, J = 6.6 Hz, 2H); 3.58–

3.68 (m, 9H); 4.08–4.15 (m, 2H); 4.31 (dd, J = 5.4, 4.2 Hz, 1H); 4.87 (d, J = 2.1 Hz, 1H);

5.25–5.38 (m, 4H); 13C NMR (75 MHz, CDCl3): δ 20.7, 21.1, 28.8, 30.5, 30.7, 60, 62.2,

62.4, 66.2, 67.4, 68.3, 68.5, 68.9, 69.5, 69.7, 87.0, 70.3, 70.6, 97.7, 169.5, 169.7, 169.8,

170.5. HRMS-MALDI (m/z): [M]+ Calculated for C22H34O13S: 538.1723; Found:

538.1721.

2-(2-(2-Thioethoxy)ethoxy)ethoxy-α-D-mannopyranoside (10).

General procedure C with 2-(2-(2-thioacetylethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-

α-D galactopyranoside 9 (0.2 g, 0.33 mmol) and sodium methoxide (20 mg, 10%) afforded

0.11 g (87%) of 2-(2-(2-Thioethoxy)ethoxy)ethoxy-α-D-mannopyranoside 10 [α]Dr.t =

+7.82 (c = 1.0, MeOH); 1H NMR (300 MHz, CD3OD): δ 2.67 (t, J = 6.6 Hz, 2H); 3.58–-

3.71 (m, 12H); 3.81–3.85 (m, 4H), 4.92 (s, 1H); 13C NMR (75 MHz, CD3OD); δ 21.9,

60.2, 65.1, 65.8, 68.5, 68.6, 68.8, 69.3, 69.4, 69.8, 71.4, 71.8, 98.9; HRMS-MALDI (m/z):

[M]+ Calculated for C12H24O8S 328.1192; Found: 328.1192.


75

2-(2-(2-Chloroethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside (13a).

General procedure A with 1,2,3,4,6-penta-O-acetyl-D-galactopyranoside 12 (1g, 2.5

mmol), 2-[2-(2-Chloroethoxy)ethoxy]ethanol (1.28 mL, 7.6 mmol) and BF3 .Et2O (1.6 mL,

10.1 mmol), followed by purification over silica gel using hexane/ethyl acetate (40:60)

afforded 0.88 g (68%) of 2-(2-(2-chloroethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-β-D-

galactopyranoside 13a. [α]Dr.t

= +6.8 (c = 1.0, CHCl3); 1H NMR (300 MHz, CDCl3): δ 1.96

(s, 9H); 2.02 (s, 9H); 2.04 (s, 9H); 2.14 (s, 9H); 3.58–3.84 (m, 25H); 4.08 -4,12 (m, 6H);

4.49 (d, J = 8.0 Hz, 1H); 4.99 (dd, J = 10.9, 3.4 Hz, 1H); 5.37 (d, J = 3.5 Hz, 1H); 5.29-

5.43 (m, 4H); 13C NMR (75 MHz, CDCl3): δ 20.6, 28.3, 36.4, 37.0, 39.1, 41.7, 45.6, 59.7,

61.2, 66.9, 67.0, 67.2, 68.7, 69.1, 70.6, 101.1, 169.5, 169.6, 169.7, 169.9, 170.1, 170.2;

HRMS-MALDI (m/z): [M]+ Calculated for C20H31O12Cl: 498.1504; Found: 498.1501.

2-(2-(2-Thioacetylethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-β-D galactopyranoside

(13).

General procedure B with 2-(2-(2-chloroethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-α-D-

galactopyranoside 13a (0.5 g, 1.0 mmol) and potassium thioacetate (0.29 g, 3.0 mmol),

followed by purification over silica gel using hexane/ethylacetate (40:60) afforded 0.51 g

(84%) of 2-(2-(2-thioacetylethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-β-D-

galactopyranoside 13. [α]Dr.t = +11.5 (c = 1.0, CHCl3); 1H NMR (300 MHz, CDCl3): δ 1.96

(s, 9H); 2.02 (s, 9H); 2.04 (s, 9H); 2.14 (s, 9H); 3.58–3.84 (m, 25H); 4.08–4.12 (m, 6H);

4.49 (d, J = 8.0 Hz, 1H); 4.99 (dd, J = 10.9, 3.4 Hz, 1H); 5.37 (d, J = 3.5 Hz, 1H); 5.29–

5.43 (m, 4H); 13C NMR (75 MHz, CDCl3): δ 20.6, 28.3, 36.4, 37.0, 39.1, 41.67, 45.65,

59.7, 61.2, 66.9, 67.0, 67.2, 68.7, 69.1, 70.6, 101.1, 169.5, 169.6, 169.7, 169.9, 170.1,

170.2; HRMS-MALDI (m/z): [M]+ Calculated for C22H34O13S: 538.1723; Found: 538.1721.

2-(2-(2-Thioethoxy)ethoxy)ethoxy-β-D-galactopyranoside (14).

General procedure C with 2-(2-(2-thioacetylethoxy)ethoxy)ethoxy-2,3,4,6-tetra-O-acetyl-

α-D galactopyranoside 13 (0.2 g, 0.33 mmol) and sodium methoxide (20 mg, 10%)

afforded 0.11 g (87%) of 2-(2-(2- thioethoxy)ethoxy)ethoxy-β-D-galactopyranoside 14.

[α]Dr.t

= +0.22 (c = 1.0, MeOH); 1H NMR (300 MHz, CD3OD): δ 2.67 (t, J = 6.6 Hz, 2H);

3.56–3.67 (m, 12H); 3.81–3.88 (m, 4H); 4.26 (d, J = 7.2 Hz, 1H); 13C NMR (75 MHz,


76

CD3OD); δ 21.9, 60.2, 65.1, 65.8, 68.5, 68.6, 68.8, 69.3, 69.7, 71.4, 72.3, 72.9, 102.5;

HRMS-MALDI (m/z): [M]+ Calculated for C12H24O8S 328.1192; Found: 328.1192.

Phenyl seleno-2-azido-3,4,6-tri-O-acetyl-α-D-galactopyranoside (16).

Tri-O-acetyl-D galactal 15 (1.5 g, 5.51 mmol) was dissolved in 30 mL CH2Cl2 and

diphenyl diselenide was added (1.72 g, 5.51 mmol). The solution was cooled to -30°C and

bis(acetoxy)iodobenzene was added (1.77 g, 5.51 mmol). After addition of trimethylsilyl

azide (1.45 ml, 11.02 mmol) the reaction was stirred for 12 h while warming to rt. The

solvent was removed and the residue was purified over silica gel (10% to 30% EtOAc in

cyclohexanes) to give galactosamine 17 as α-anomer (2.51 g, 5.35 mmol, 97%). The

analytical data were in agreement with those reported in the literature.33

2-(2-(2-chloroethoxy)ethoxy)ethoxy-3,4,6-tri-O-acetyl-α-D-galactosamine (17).

D-galactosamine selenoglycoside 16 (522 mg, 1.1 mmol) was dissolved in 3 mL CH2Cl2

and 3 mL Et2O and cooled to 0 °C. (2-Chloroethoxy)bis ethanol (561 mg, 3.3 mmol) was

added, followed by addition of N-iodosuccinimide (1.24 g, 5.5 mmol) and triflic acid (18

PL, 0.2 mmol). The mixture was stirred overnight while warming to r.t. The reaction was

quenched with NEt3 and washed with sat. aq. sodium thiosulfate solution. The organic

layer was separated and dried over Mg2SO4. The solvent was evaporated and the residue

was purified over silica gel (30% to 60% EtOAc in cyclohexanes) to give galactosamine

17 as a separable α/β-mixture (1:1) (376 mg, 0.78 mmol, 71%). Rf 0.3 (cyclohexanes/ethyl

acetate = 1:1); [α]Drt

= 42.1 (c = 1, CHCl3); IR (thin film on NaCl): ν = 2874, 2112, 1747,

1434, 1371, 1228, 1127, 1046 10 cm-1; 1H NMR (300 MHz, CDCl3): δ 5.43-5.29 (m, 4H),

5.06 (d, J = 3.5 Hz, 1H), 4.74 (dd, J = 10.9, 3.4 Hz, 1H), 4.48 (d, J = 8.0 Hz, 1H), 4.30–

3.98 (m, 6H), 3.84–3.58 (m, 25H), 2.12 (s, 3H), 2.11 (s, 3H), 2.03 (s, 3H), 2.02 (s, 3H),

2.01 (s, 3H), 2.00 (s, 3H); 13C NMR (75 MHz, CDCl3): δ 170.2, 170.1, 169.9, 169.7,

169.6, 169.5, 102.4, 98.1, 71.3, 70.9, 70.8, 70.7, 70.6, 70.5, 70.4, 70.1, 69.3, 68.1, 67.7,

67.5, 66.5, 66.3, 61.5, 61.2, 60.7, 57.3, 42.8, 42.7, 21.0, 20.9, 20.8, 20.7, 20.6 20.5;

MALDI-HRMS (m/z): Calculated for C18H28ClN3O10Na+: 504.1355, Found: 504.1361

[M+Na]+.


77

2-(2-(2-Thioacetylethoxy)ethoxy)ethoxy-2-azido-3,4,6-tetra-O-acetyl-α-D

galactopyranoside (18a).

General procedure A with 2-(2-(2-chloroethoxy)ethoxy)ethoxy-2-azido-3,4,6-tetra- O-

acetyl-α-D-galactopyranoside 17 (0.1 g, 0.21 mmol) and potassium thioacetate (60 mg,

0.63 mmol), followed by purification over silica gel using hexane/ethyl acetate (20:80)

afforded 85 mg (78%) of 2-(2-(2-thioacetylethoxy)ethoxy)ethoxy-2-azido-3,4,6-tetra-

Oacetyl- α-D-galactopyranoside 18a. [α]Dr.t = +12.3 (c = 1.0, CHCl3); 1H NMR (300 MHz,

CDCl3): δ 2.02 (s, 9H); 2.12 (s, 3H); 2.32 (s, 3H); 3.07 (t, J = 6.6 Hz, 2H); 3.58–3.69 (m,

9H); 3.99–4.16 (m, 1H); 4,46 (d, J = 7.5 Hz, 1H); 4.74 (dd, J = 3.3, 7.5 Hz, 1H); 5.07 (d, J

= 3.6 Hz, 1H); 5.28–5.41 ( m, 1H); 13C-NMR (75 MHz, CDCl3); δ 20.5, 28.7, 30.5, 42.6,

7.3, 60.6, 61.4, 66.3, 67.6, 68.1, 69.3, 70.0, 70.2, 70.4, 70.8, 71.2, 98.1, 169.7, 169.9,

170.2; HRMS-MALDI (m/z): [M]+ Calculated for C20H31O11N3S: 521.1679; Found:

521.1678.

2-(2-(2-Thioethoxy)ethoxy)ethoxy-2-azido-α-D-galactopyranoside (18).

General procedure B with 2-(2-(2-thioacetylethoxy)ethoxy)ethoxy-2-azido-3,4,6-tetra-O-

acetyl-α-D-galactopyranoside 18a (80 mg, 0.17 mmol) and sodium methoxide (10 mg,

0.17 mmol) afforded (45 mg, 80%) of 2-(2-(2-thioethoxy)ethoxy)ethoxy-2-azido-α-D-

galactopyranoside 18. [α]Dr.t

= +0.96 (c = 1.0, MeOH);1H NMR (300 MHz, CD3OD): δ

2.67 (t, J = 6.6 Hz, 2H); 3.58–3.71 (m,12H); 3.81-3.85 (m, 4H); 4.34 (d, J = 7.2 Hz, 1H);

4.96 (d, J = 3.6 Hz, 1H); 13C NMR (75 MHz, CD3OD); δ 21.9, 59.7, 66.9, 67.5, 67.6, 68.,

6.7, 69.7, 71.3, 72.1, 73.9, 102.3; HRMS-MALDI (m/z): [M]+ Calculated for C12H23O7N3S:

353.1257; Found: 353.1255.


78

2-(2-(2-Thioethoxy)ethoxy)ethoxy-2-amino-α-D-galactopyranoside (19). Trimethylphosphine (0.1 mL of a 10 M solution) was added to 2-(2-(2-

thioethoxy)ethoxy)ethoxy-2-azido- α-D-galactopyranoside 18 (40 mg, 0.11 mmol) in

anhydrous THF. The solution was stirred at room temperature for 3 h before addition of

water (1 mL) and stirring overnight. THF was removed in vacuo, and the residue was

dissolved in 10 mL water. The water layer was washed with chloroform (4 x5 mL). Water

was lyophilized to afford 30 mg (81%) of 2-(2-(2- thioethoxy)ethoxy)ethoxy-2-amino-α-

D-galactopyranoside 19. [α]Dr.t

= +1.25 (c = 1.0, MeOH); 1H NMR (300 MHz, CD3OD): δ

2.66 (t, J = 6.6 Hz, 2H); 3.45–3.75 (m, 14H); 3.99–4.04 (m, 1H); 4.26 (d, J = 7.2 Hz, 1H); 13C NMR (75 MHz, CD3OD); δ 21.9, 59.7, 66.9, 67.5, 67.6, 68.4, 69.7, 69.7, 71.3, 72.1,

73.9, 102.3; HRMS-MALDI (m/z): [M]+ Calculated for C12H25O7NS: 327.1352; Found:

327.1351.

2.7.6 Batch Functionalization of CdSe/ZnS QDs

QD635-DHLA-PEG2000-NH2 (22).

To 0.1 mL of QDs in chloroform (O.D > 50 at 400 nm) 50 µL of DHLA-PEG2000-NH2 6

in 50 µL of ethanol were added. The mixture was stirred at 60 °C for 12 h. In between,

another 50 µL of ethanol were added. PEGylated-QDs were separated by 12 precipitation

using ethanol/chloroform/hexane (1:0.5:5). Centrifugation at 3000 rpm for 3 min afforded

a clear pellet-like suspension. This pellet was separated and dissolved in distilled water (1

mL), centrifuged at 3000 rpm for 3 min to afford a clear solution. This solution was

lyophilized and pure PEGylated-QD was obtained from NAP-1 column chromatography

using 1.0 µM borate-buffer pH 7.5 as the eluent. Concentration of the QDs was estimated

using a previously published procedure.32


79

QD635-DHLA-PEG2000-maleimide (23). QD635-DHLA-PEG2000-NH2 22 (0.05 µM in 1 mL of 1.0 µM borate-buffer pH 8.5) was

added to 3-maleimido propanoic acid N-hydroxysuccinimide ester (13 mg, 49.0 µM) and

stirred at room temperature for 2 h. The aqueous layer was washed with chloroform (3 x 1

mL) to remove unreacted maleimido derivative. The aqueous layer was taken as such for

the next step without further characterization.

QD635-DHLA-PEG2000-mannose (24a).

QD635-DHLA-PEG2000-maleimide 23 (0.02 µM in 1 mL of 1.0 µM borate-buffer pH 7.3)

was added to 2-(2-(2 thioethoxy)ethoxy)ethoxy-α-D-mannopyranoside 10 (10 mg, 30.5

µM) and stirred at room temperature for 3 h. The crude product was purified by NAP-1

column using distilled water as the eluent. The final concentration of the sample was

estimated by using a previously published procedure.32

QD635-DHLA-PEG2000-galactosamine (24b).

QD635-DHLA-PEG2000-maleimide (0.02 µM in 1 mL of 1.0 µM borate-buffer pH 7.3)

was added to 2-(2-(2 thioethoxy)ethoxy)ethoxy-2-amino-β-Dgalactopyranoside 19 ( 10

mg, 30.5 µM) and stirred at room temperature for 3 h. The crude product was purified by

NAP-1 column using distilled water as the eluent. The final concentration of the sample

was estimated by using a previously published procedure.32

QD635-DHLA-PEG2000-galactose (24c).

QD635-DHLA-PEG2000-NH2 (0.02 µM in 1 mL of 1.0 µM borate-buffer pH 7.3) was

added to 2-(2-(2 thioethoxy)ethoxy)ethoxy-2-amino-β-Dgalactopyranoside 14 (10 mg, 30.5

µM) and stirred at room temperature for 3 h. The crude product was purified by NAP-1

column using distilled water as the eluent. The final concentration of the sample was

estimated by using a previously published procedure.32


80

QD635-DHLA-PEG2000-OH (24d). To 0.1 mL of QDs in chloroform (O.D > 50 at 400 nm) 50 µL of DHLA-PEG2000-OH 3

in 50 µL of ethanol were added. The mixture was stirred at 60 °C for 3 h. The crude

product was purified by NAP-1 column using distilled water as the eluent. The final

concentration of the sample was estimated by using a previously published procedure.32

2.7.7 Continuous Flow Synthesis of CdSe and CdTe QDs

Cadmium precursor was prepared by heating 100 mg (0.75 mmol) cadmium oxide with

600 mg (3.1 mmol) lauric acid at 150 °C until a clear solution was obtained. This solution

was cooled to room temperature and 1.5 mL each of oleic acid and oleylamine were added

to the flask. A solution containing 80 mg (1.0 mmol) selenium in 2 mL (6.76 mmol) of

trioctylphosphine or 120 mg (0.97 mmol) tellurium in 2 mL (6.76 mmol) of

trioctylphosphine was prepared. A solution of cadmium precursor (0.097 mmol) in 0.5 mL

of squalene and 0.097 mmol of Se or Te precursor in 0.5 mL of squalene were pushed into

the microreactor using two syringe pumps. CdSe or CdTe QDs were prepared at reaction

times of 3, 10, 20 and 30 min at flow rates of 333 µL/min, 100 µL/min, 50 µL/min, 33.33

µL/min. The QDs were purified by precipitation with anhydrous

CHCl3/MeOH/hexane/isopropanol to afford 13 mg (72%, after 3 min), 15 mg (71%, after

10 min), 18 mg (67%, after 20 min) and 17 mg (66%, after 30 min) product.

2.7.8 Continuous Flow Synthesis of CdTe or CdSe/ZnS

QDs

A solution of 20 mg of the CdTe or CdSe QDs in 1 mL of toluene and 2 mL of

trioctylphosphine was used as nanocrystal precursor. Trioctylphosphine (2 mL), 50 µL


81

(0.28 mmol) of hexamethyldisilathiane and 400 µL of 10% diethylzinc in toluene were

mixed. The two solutions were injected into the microreactor at 90-110 °C at a flow rate of

33.33 µL/min (residence time of 30 min) The resulting QDs were purified by precipitation

with chloroform/methanol mixture to afford 22 mg (68%)32 product. Surface exchange of

oleic acid–capped CdTe/ZnS QDs with pyridine was performed by dissolving 20 mg of the

QDs in pyridine and passing the solution through the microreactor at 60 °C at a flow rate

of 33.33 µL/min (30 min residence time). Precipitation with hexane followed by

centrifugation afforded 12 mg (74%)32 of pyridine capped QDs.

2.7.9 Continuous Flow Functionalization of QDs

Preparation of mannose capped QDs (25) or galactose capped QDs (26).

A solution of 10 mg of pyridine-capped CdTe-ZnS QDs in 1 mL ethylene dichloride and a

freshly prepared solution of 35 mg 2-(2-(2-thioethoxy) ethoxy) ethoxy-α-D-

mannopyranoside 10 (0.11 mmol) or 2-(2-(2-thioethoxy)ethoxy)ethoxy-β-D-

galactopyranoside 14 (35 mg, 0.11 mmol) in 1 mL ethylenedichloride/ethanol (1:1) was

prepared. The solutions (0.5 mL each) were simultaneously injected into the microreactor

that was preheated at 50 °C at a flow rate of 33 µL/min (30 min residence time). The

solvent was evaporated and carbohydrate coupled QDs were precipitated with

hexane/chloroform/methanol (9:1:1). The final concentration of the sample was estimated

using a published procedure.32

Preparation of dihydrolipoic acid capped QDs (27).

A solution of 10 mg pyridine capped CdTe-ZnS QDs in 1 mL of ethylene dichloride and

20 mg (0.10 mmol) of dihydrolipoic acid in 1 mL of ethylene dichloride/ethanol (1:1) was

prepared. The solutions (0.5 mL each) were simultaneously injected into the microreactor

at 50 °C at a flow rate of 33.33 µL/min (residence time 30 min). The solvent was

evaporated and dihydrolipoic acid coupled QDs were precipitated by addition of

tetramethylammonium hydroxide. The final sample concentration was estimated using a

published procedure.32


82

2.7.10 Three Steps Continuous Flow Synthesis of

Mannose Capped CdSe/ZnS QDs

Preparation of mannose capped CdSe/ZnS QDs.

Cadmium precursor was prepared by heating 100 mg (0.75 mmol) cadmium oxide with

600 mg (3.1 mmol) lauric acid at 150 °C until a clear solution was obtained. This solution

was cooled to room temperature and 1.5 mL each of oleic acid and oleylamine were added

to the flask. A solution containing 80 mg (1.0 mmol) selenium in 2 mL (6.76 mmol) of

trioctylphosphine in 2 mL (6.76 mmol) of trioctylphosphine was prepared. A solution of

cadmium precursor (0.097 mmol) in 0.5 mL of squalene and 0.097 mmol of Se precursor

in 0.5 mL of squalene were pushed into the microreactor using two syringe pumps (15 min

of residence time, flow rates of 66.66 µL/min). CdSe QDs were then flushed directly into

another microreactor at 90-110°C. A solution of trioctylphosphine (2 mL), 50 µL (0.28

mmol) of hexamethyldisilathiane and 400 µL of 10% diethylzinc in squelene was prepared

and injected separately. Finally CdSe-ZnS QDs in 1 mL ethylene dichloride and a freshly

prepared solution of 35 mg 2-(2-(2-thioethoxy) ethoxy) ethoxy-α-D-mannopyranoside 10

(0.11 mmol) were flushed into a third microreactor at 60 °C o afforded 12 mg 52% final

compound.


83

2.8 References

1 (a) Xie, R.; Kolb, U.; Li, J.; Basch, T.; Mews, A. J. Am. Chem. Soc. 2005, 127, 7480–

7488. (b) Susumu, K.; Mei, B. C.; Mattoussi, H. Nat. Prot. 2009, 4, 424–436. (c)

Giepmans, B. N. G.; Adams, S. R.; Ellisman, M. H.; Tsien, R. Y. Science 2006, 312, 217–

224; (c) George, N.; Pick, H.; Vogel, H.; Johnsson, N.; Johnsson, K. J. Am. Chem. Soc.

2004, 126, 8896–8897. (d) Somers, R. C.; Bawendi, M. G.; Nocera, D. G. Chem. Soc. Rev.

2007, 36, 579–591. (e) Carion, O.; Mahler, B.; Pons, T.; Dubertret, B. Nat. Prot. 2007, 2,

2383–2390. (f) Zhong, H.; Scholes, G. D. J. Am. Chem. Soc. 2009, 131, 9170–9171. 2 (a) Mattoussi, H.; Mauro, J. M.; Goldman, E. R.; Anderson, G. P.; Sundar, V. C.;

Mikulec, F. V.; Bawendi, M. G. J. Am. Chem. Soc., 2000, 122, 12142–12150. (b) Peng, Z.

A.; Peng, X. J. Am. Chem. Soc., 2001, 123, 183–184. (c) Liu, W.; Choi, H. S.; Zimmer, J.

P.; Tanaka, E.; Frangioni, J. V.; Bawendi, M. J. Am. Chem. Soc., 2007, 129, 14530–1453.

(d) Somers, R. C.; Bawendi, M. G.; Nocera, D. G. Chem. Soc. Rev., 2007, 36, 579–591. (e)

Liu, W.; Howarth, M.; Greytak, A.; Zheng, B. Y.; Nocera, D. G.; Ting, A. Y.; Bawendi,

M. G. J. Am. Chem. Soc., 2008, 130, 1274–1284. (f) Kikkeri, R.; Lepenies, B.; Adibekian,

A.; Laurino P.; Seeberger, P. H. J. Am. Chem. Soc., 2009, 131, 2110–2112. 3 Cohen, H.; Sarkar, S. K.; Hodes G. J. Phys. Chem. B 2006, 110, 25508–25513. 4 Ioannou, D.; Griffin, D. K. Nano Reviews 2010, 1, 5117–5131. 5 Murray, C. B.; Kanagan, C. R.; Bawendi, M. G. Annu. Rev. Mater. Sci. 2000. 30, 545–

610. 6 Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc., 1993, 115, 8706–8715. 7 Yang, Y. A.; Wu, H.; Williams, K. W.; Cao, Y. C. Angew. Chem. Int. Ed. 2005, 44,

6712–6715. 8 (a) Ekimov, A. I.; Onushchenko, A. A. JETP Lett. 1981, 34, 345–349. (b) Murray, C. B.;

Norris, D. J.; Bawendi, M. G., J. Am. Chem. Soc. 1993, 115, 8706–8715. 9 (a) Qu, L.; Peng, Z. A.; Peng, X. Nano Lett., 2001, 1, 333–337. (b) Talapin, D. V.;

Rogach, A. L.; Kornowski, A.; Haase, A.; Weller, H. Nano Lett. 2001, 1, 207–211. (c)

Rempel, J. Y.; Bawendi, M. G.; Jensen, K. F. J. Am. Chem. Soc. 2009, 131, 4479–4489.


84

10 Reiss, P.; Protiere M.; Li, L. Small, 2009, 5, 154–168. 11 Altınoglu, E. I.; Adair, J. H. WIREs Nanomed Nanobiotechnol, 2010, 2, 461–477. 12 Piryatinski, A.; Ivanov, S. A.; Tretiak, S.; Klimov, V. I. Nano Lett. 2007, 7, 108–115. 13 Kim, S.; Fisher, B.; Eisler, H. J.; Bawendi, M. J. Am. Chem. Soc. 2003, 125, 11466–

11467 14 (a) Mews A., Eychmüller A., Giersig M., Schooss D., Weller H., J. Phys. Chem. 1994,

98, 934–941; (b) Battaglia D., Li J. J., Wang Y. J., Peng X. G., Angew. Chem. Int. Ed.

2003, 42, 5035–5039; (c) Zhong X. H., Xie R. G., Zhang Y., Bansche T.; Knoll W. Chem.

Mater. 2005, 17, 4038–4042. 15 Biju, V., Itoh, T. Ishikawa, M. Chem Soc. Rev, 2010, 39, 3031–3056. 16 (a) Gao, X. H.; Cui, Y. Y.; Levenson, R. M.; Chung, L. W. K., Nie, S. M. Nat.

Biotechnol. 2004, 22, 969–976; (b) Hezinger, A.F.E.; Teßmar J.; Goepferich, A. Eur. J.

Pharm. Biopharm. 2008, 68, 138–152. 17 (a) Sun, X.–L.; Cui, W.; Haller, C.; Chaikof, E. L. ChemBioChem, 2004, 5, 1593–1596.

(b) Babu, P.; Sinha, S.; Surolia, A. Bioconj. Chem. 2007, 18, 146–151. 18 Rogach, A. L.; Nagesha, D.; Ostrander, J. W.; Giersig, M.; Kotov, N. A. Chem. Mater.

2000, 12, 2676–2685. 19 (a) Li, J. J.; Wang, Y. A.; Guo, W.; Keay, J. C.; Mishima, T. D.; Johnson, M. B.; Peng,

X. J. Am. Chem. Soc. 2003, 125, 12567–12575. (b) Sanderson, K. Nature 2009, 459, 760–

761. 20 (a) Ehrfeld, W.; Hessel, V.; Löwe, H. Microreactors: New Technology for Modern

Chemistry, Wiley-VCH, Weinheim, 2000. (b) Hessel, V.; Hardt, S.; Löwe, H. Chemical

Micro Process Engineering, Wiley-VCH, Weinheim, 2004. (c) Geyer, K;. Codée, J. D. C.;

Seeberger, P. H. Chem. Eur. J. 2006, 12, 8434–8442. (d) Mason, B. P.; Price, K. E.;

Steinbacher, J. L.; Bogdan, A. R.; McQuade, D. T. Chem. Rev. 2007, 107, 2300–2318. (e)

Wiles, C.; Watts, P. Eur. J. Org. Chem. 2008, 10, 1655–1671. (f) Brivio, M.; Verboom,

W.; Reinhoudt, D. N. Lab Chip, 2006, 6, 329–344. (g) Jähnisch, K.; Hessel, V.; Löwe, H.;

Baerns, M. Angew. Chem. Int. Ed. 2004, 43, 406–446. 21 (a) Marre, S., Park, J., Rempel, J., Guan, J., Bawendi, M. G. Jensen, K. F., Adv. Mater.

2008, 20, 4830–4834. (b) Yen B. K. H., Günther A., Schmidt M. A., Jensen K. F.,

Bawendi M. G., Angew. Chem. Int. Ed., 2005, 44, 5447–5451. (c) Shestopalov I., Tice J.


85

D., Ismagilov R. F., Lab Chip 2004, 4, 316–321. (d) Dideko Y. T., Suslick K. S. J. Am.

Chem. Soc. 2005, 127, 12196–12197. 22 (a) Edel J. B., Fortt R., deMello J. C., deMello A. J., Chem. Commun. 2002, 1136–1137.

(b) Sounart T. L., Safier P. A., Voigt J. A., Hoyt J., Tallant D. R., Matzke C. M.,

Michalske T. A., Lab Chip 2007, 7, 908–915. (c) Yen, B. K. H.; Stott, N. E. S.; Jensen, K.

F.; Bawendi, M. G. Adv. Mater. 2003, 15, 1858–1862. 23 Kikkeri, R.; Laurino, P.; Odedra, A.; Seeberger, P. H. Angew. Chem. Int. Ed. 2010, 49,

2054–2057. 24 Laurino, P.; Kikkeri, R.; Seeberger, P. H. Nat. Protoc., 2011, in press. 25 Syrris FRX, http://www.syrris.com. 26 (a) Rong, H.; Xiaogang, Y.; Hongye, T.; Feng, G.; Daxiang ,C.; Hongchen, G. Front.

Chem. China 2006, 4, 378–383. (b) Zhong, H.; Scholes, G. D. J. Am. Chem. Soc. 2009, 31,

9170–9171. (c) Peng, X.; Manna, L.; Yang, W.; Wickham, J.; Scher, E.; Kadavanich, A.;

Alivisatos, A. P. Nature, 2000, 404, 59–61. (d) Yu, W. W.; Wang, Y. A.; Peng, X. Chem.

Mater. 2003, 15, 4300–4308. 27 Yu, W.W. Chang, E. Drezeck, R. Colvin V.L. Biochem. Biophys. Res. Commun. 2006,

384, 781–786. 28 (a) Michalet, X.; Pinaud, F.F.; Bentolila, L.A.; Tsay, J.M.; Li, J.J.; Sundaresan, G.; Wu,

A.M.; Gambhir, S.S.; Weiss, S. Science 2005, 307, 538–544. (b) Hild, W.A.; Breunig, M.;

Goepferich, A. Eur. J. Pharm. Biopharm. 2007, 68, 112–128. 29 (a) Lunney, J.; Ashwell, G. Proc. Natl. Acad. Sci. U.S.A. 1976, 73, 341–343. (b)

Baenziger, J. U.; Fiete, D. Cell 1980, 22, 611–620. (c) Braun, J. R.; Willnow, T. E.;

Ishibashi, S.; Ashwell, G.; Herz, J. J. Biol. Chem. 1996, 271, 21160–21166. 30 Higuchi, Y.; Oka, M.; Kawakami, S.; Hashida, M. J. Controlled Release 2008, 125,

131–136. 31 Xie, R.; Battaglia, D.; Peng, X. J. Am. Chem. Soc. 2007, 129, 15432–15433. 32 Leutherdale, C. A.; Woo, W.-K.; Mikulec, F. V.; Bawendi, M. G. J. Phys. Chem. 2002,

106, 7619–7622. 33 Mironov, Y. V.; Medintz, I. L.; Jaiswal, J. K.; Simon, S. M.; Mattoussi, H. J. Am. Chem.

Soc. 2004, 45, 9107–9110.

3. Continuous Flow Microreactor and Phosphine Oxide Initiator: an Unique Combination

86

3. Continuous Flow Microreactors and Phosphine Oxide Initiators: an Unique Combination

This work was performed in collaboration with Dr. K. Tauer (discussions on the

experiments and evaluation of data), the group of Prof. H. Grützmacher (synthesis of

photoinitiator) and Dr. H. Hernandez (simulation studies).

Part of this chapter has been submitted for publication titled “Snowballing Radical

Generation Leads to Ultrahigh Molecular Weight Polymers” Laurino, P.; Hernandez, H.

F.; Bräuer, J.; Krüger, K.; Grützmacher, H.; Tauer, K.; Seeberger, P. H. 2011.


87

3.1 Introduction

Polymers are large molecules made of repeating units of smaller molecules of

defined nature called monomers. The monomers are connected by covalent chemical bonds

to form a polymer chain. Polymers can occur naturally or can be produced for specific

purposes by chemical synthesis. The synthetic process of monomer compounds reacting

together to generate a polymer chain is called polymerization. Polymerization is a well

established technique for the fabrication of different polymers and material composites for

various applications ranging from paper producing to well defined nanoparticles for drug

delivery or specific biomedical applications.1 Of the many forms of polymerization, the

most common type is the free radical mediated polymerization where the polymer is

formed by the successive addition of free radical monomers. Three significant reactions for

the generation of polymer by the free radical mediated polymerization are: initiation,

propagation and termination (Figure 3.1).

Figure 3.1 Three key reactions in free radical polymerization (R = initiator; M = monomer)

In the first step of the polymerization process the free radical is generated by a molecule

called initiator that upon thermal decomposition, photolysis, redox reaction or sonication

can generate radical species that promote radical reactions. During the initiation, an active

center is generated from the initiator and the radical is transferred to the monomer

molecule. After the radical is transferred to a monomer molecule the polymer chain starts


88

to grow and the propagation occurs until no monomer is available or a termination reaction

occurs. Termination can occur either via combination whereby two chains couple to form

one single long chain or by disproportionation where a hydrogen is abstracted from a

growing polymer chain giving an unsaturated terminal chain and a saturated terminal

chain. The polymerization rate can be affected by a decrease of the initiator efficiency due

to various side reactions. Common side reactions involving the initiator are: 1) Primary

decomposition, when two radicals recombine before initiating a chain. 2) Other reaction

pathways involving the recombination of the radicals with impurities.

The most frequently used methods for radical polymerization are:

1) Bulk polymerization: reaction mixture comprised of initiator and monomer.

2) Solution polymerization where the reaction mixture is comprised of an initiator,

monomer and solvent.

3) Emulsion polymerization: where the reaction mixture is comprised of an aqueous phase,

water-insoluble monomer, initiator and emulsifying agent (surfactant).

Emulsion polymerization in particular is a well investigated and characterized

technique which in most cases is the method of choice for the synthesis of composite

polymer particles with special morphologies to explore their potential as new materials for

novel technologies.2 The most common type of emulsion polymerization is an oil-in-water

emulsion where droplets of the hydrophobic monomer are emulsified with surfactants in an

aqueous phase. It has to be noted that in emulsion polymerization, the polymerization

events do not occur in the monomer droplets but occur in the polymer particles also called

latex particles that are generated spontaneously in the early stages of the polymerization

reaction. A key factor in radical-mediated emulsion polymerization is the average number

of radicals per polymer particle that is defined by . This value depends on particle size,

concentration of the monomer, rate of initiator decomposition and radical flux into and out

of particles.3 Classical emulsion polymerization of polystyrene produces particles of less

than 100 nm diameter, that contain, on average, one half radical per particle ( = 0.5).

Such reactions are conformed to so-called zero–one kinetics, where each particle contains

either one or no growing radical. In emulsion polymerizations, the latex particles obtained

vary in diameter from less than 100 nm to above 1000 nm and contain many polymer

chains. The latex particles are stabilized in many cases by the presence of the charged

surfactant that generates a repulsion effect on the particle surface avoiding coagulation

events. Polymerization parameters like temperature, reaction times, monomer, solvent,


89

surfactant and mixing have been investigated,4 and have an effect on the physical

properties of the polymer latex including molecular weight, particle size and other

properties.5

This chapter mainly focuses on the investigation of an emulsion polymerization

mechanism mediated by phosphine oxide photoinitiators in a continuous flow

microreactor.

3.1.1 Photo-initiated Polymerization in a Continuous

Flow Microreactor

Continuous flow microreactors are widely used as alternative reactor systems6 for

the formation of polymer particles or emulsions to achieve better control procedures. 7

Emulsion polymerization in continuous flow microreactors can generate particles with

predetermined sizes and well defined shape in the nanometer range.8 The continuous flow

devices can be efficiently employed to screen conditions, to use precious reagents, and to

change reaction times based only on varying the flow rate.9

To emphasize the contribution of continuous flow on photo-polymerization a set of

examples where the technical aspects of the continuous flow microreactor influenced the

particle formation and facilitated the screening and optimization of polymerization

conditions were reviewed. Doyle et al. reported on the formation of plug- or disk-shaped

particles by the photo-polymerization of polyethylene glycol (PEG) gel using in situ UV-

irradiation.10 The PEG-polymer solution is injected perpendicular to the water phase

solution (Figure 3.2).


90

Figure 3.2 Schematic representation of the reaction zones where the particles are generated and

irradiated. Irradiation can occur in (a) a first stage to generate plugs or in (b) a second stage where

the droplet is able to relax to generate disks (100 W HBO mercury lamp, and UV filter to provide

the desired wavelength).11

Key determinants for the different particle shapes are the flow rate, the capillary number

which is related to viscosity and density of the fluids involved (eq. 6), and the channel

dimension.

(6) Ca = µV/γ

Ca, capillary number; µ, viscosity of the liquid; V characteristic velocity; γ surface or interfacial

tension between the two phases.

Injecting the PEG polymer solution into an aqueous continuous phase followed by the

immediate irradiation of the formed droplets resulted in plug-shape particles. When the

formed droplets were allowed to relax in a bigger channel before irradiation, disk shaped

particles were formed. Constraining of droplets into the microreactor channel was

exploited to define size and shape of the final particles.

Lee et al. have described the preparation of microparticles in flow.11 In situ UV

irradiation of 4-hydrobutyl acrylate (4-HBA) droplets in a sheath flow of poly(vinyl

acrylate) (PVA) resulted in the generation of small particles of 10 µm diameter without the

use of a surfactant, which is a key component for heterophase polymerization reaction


91

(Figure 3.3). Surfactant free particles are ideal candidates for various biomedical

applications.

Figure 3.3 Reaction zones where the particles are generated and irradiated. The flow is constituted

initially by the core flow (4-HBA) and then the sheath flow (PVA) is injected in diagonal. The

irradiation occurs where the particles have defined spherical shape (UV light EXPO OmniCure

S1000, Maritimes, Cadana, 324 nm, 300 mW/cm2).12

Lee et al. have described the formation of hydrogels via UV–initiated

polymerization of a monomer (80 wt % 4-hydroxybuthyl acrylate, 11 wt % acrylic acid, 1

wt % ethyleneglycol dimethyl acrylate, and 3 wt % 2,2-dimethoxy-2-phenyl-

acetonphenone) / enzyme solution in an immiscible non–polymerizable sheath fluid of

mineral oil (Figure 3.4).12


92

Figure 3.4 Reaction zones for particles generation and irradiation (365 nm, 1.2 mW/cm2,

Novacure, Photonic Solutions Inc.).16

The formed droplets were flashed by light (1 s irradiation time) in an unprotected part of

the channel producing particles below 100 µm in diameter and higher flow rates of the

flow sheath resulted in the formation of smaller particles.

3.1.2 Phosphine Oxide Photoinitiators

Acylphosphine oxides are well known and widely used photoinitiators for

polymerization processes. 13 The photolysis of monoacylphosphine oxides (MAPOs,

Scheme 3.1) is similar to that of arylalkylketone upon UV irradiation at 380 nm.

Scheme 3.1 Examples of monoacylphosphine oxides (MAPOs) as photoinitiators.

MAPOs undergo α-cleavage upon irradiation with high efficiency to produce

aroylphosphinoyl radical pair (quantum yield = 0.6, Scheme 3.2). 14, 15


93

Scheme 3.2 Photolysis of (2, 4, 6-trimethylbenzoyl)diphenylphosphine oxide.

The arylphosphinoyl radical 7 exhibits 10–100 fold higher reactivity toward unsaturated

compounds than the commonly used benzoyl radicals. This exceptional reactivity makes

MAPOs excellent candidates for radical polymerization process. The absorption of

MAPOs in the UV-visible region facilitates their use in the presence of pigments and

results in many more applications.16

Nevertheless, subsequent investigations have lead to the development of a second

generation of acylphosphine oxide initiators called bisacylphosphine oxide derivatives,

BAPOs which are superior photoinitiators because of the additional chromophore present

in them. (Scheme 3.3).17

Scheme 3.3 Examples of bisacylphosphine oxide (BAPOs) photoinitiators

The presence of a second chromophore shifts the absorption to a higher wavelength (ca.

420 nm) than MAPOs which absorb at 380 nm. This wavelength shift facilitates a photo-

cleavage of the first chromophore at 420 nm and subsequent cleavage of the second

chromophore at 380 nm.18 The enhanced absorption of BAPOs at higher wavelengths

enables better performance using visible light. This property is important for biological

applications as it allows for tissue curing without the extensive damage associated with the

use of UV light.


94

Despite their useful properties, the incompatibility of BAPOs with aqueous

formulations presents a major drawback. In order to overcome this drawback, Grützmacher

et al prepared two new water soluble bisacylphosphine oxide derivatives: (2-(bis(2, 4, 6-

trimethylbenzoyl)phosphoryl)acetic acid (BAPO-AA), and bis(2, 4, 6-

trimethylbenzoyl)phosphineoxide tris-oxyethylene (BAPO-PEG) (Scheme 3.4).

Scheme 3.4 Structures of (2-(bis(2, 4, 6-trimethylbenzoyl)phosphoryl)acetic acid 14 (BAPO-AA)

and bis(2, 4, 6-trimethylbenzoyl)phosphineoxide tris-oxyethylene 15 (BAPO-PEG).

The derivatives described above are the first examples of water-soluble initiators having

the photophysical properties of BAPOs.28 The availability of these initiators further

expands the study of radical polymerization to water–soluble photoinitiators. To overcome

difficulties associated with scale-up and reproducibility of emulsion polymerizations

initiated by water soluble BAPOs, we decided to investigate the polymerization process in

continuous flow microreactors.


95

3.2 Continuous Flow Microreactor Design

A detailed description of the continuous flow microreactor and its influence on the

polymerization reaction are necessary to fully understand the emulsion polymerization

investigation. The flow reactor set-up is comprised of two gas-tight syringes, a 2.5 mL

syringe containing the monomer and a 10 mL syringe containing the aqueous phase where

the photoinitiator and sodium dodecyl sulfate (SDS) were dissolved (Figure 3.5).

Figure 3.5 Continuous flow microreactor for photo-initiated emulsion polymerization. Syringe

pump injection system (1, Harvard PHD 2000), gas tight syringe (2, Hamilton), countercurrent

micromixer (3, Standard Slit Interdigital Micro Mixer), FEP tubing (4, fluorinated ethylene

polymer, Tub FEP Nat 1/16 in x .030 in), quartz immersion well (5) thermostat (10), Pyrex filter

(6), Hanovia medium pressure Hg lamp (7, 450 W), a power supply (9), and a collection vessel

(11).

A syringe pump was used for flushing the reagents into the reactor through the gas-tight

syringes. The emulsion was generated in a countercurrent micromixer19 (Standard Slit

Interdigital Micro Mixer) resulting in a mean droplet diameter of about 50–100 µm. The

mixing system was made of 15 lamellae, each with a height of 200 µm, width of 45 µm,

and outflow width of 30 µm. The system combines the use of fluorinated ethylene

propylene (FEP) tubing coiled around the cooling quartz jacket system containing a 450 W

Hanovia medium pressure Hg Lamp with an arc length (27.9 cm) and a pyrex filter (1 mm


96

of wall thickness). The part of the reactor involving the lamp was placed in a box to protect

the users from dangerous irradiation.

The total reaction was divided into reaction zones with different volumes (Figure

3.6).

Figure 3.6 Detailed view of the reaction volumes in the tubing.

The first part of the reaction was placed out of the protection box and included two zones:

an emulsification zone (E), followed by a relaxing and spontaneous emulsification zone

(SE) with a volume of 0.075 mL. Next to the E and SE zones was the irradiated zone, with

a volume of 2.432 mL including three parts: RZ1, MRZ and RZ2. RZ1 and RZ2 were

reaction zones, irradiated but not coiled around the cooling system and MRZ was the

reaction zone coiled around the cooling system. The last part of the reaction tubing outside

of the protective box was the outflow zone of a volume of 0.193 mL.

The initial stage of the investigations required some optimization of the continuous

flow microreactor. The use of a T-connection was explored but results achieved with the

micromixer were found to be better in terms of particle size and overall reproducibility.

Optimization of the light source and tubing system was also performed. Medium pressure

Hg lamps with different arc lengths (4.0 cm, 27.9 cm and 2.0 cm) and power (125 W, 450

W and 100W) were tested, as well as different kinds of Teflon tubing,

PolyTetraFluoroEthylene (PTFE) and Fluorinated Ethylene Propylene (FEP).20 The system

that afforded the best results for photon transmittance was FEP tubing and for light

emission a 450 W medium pressure Hg lamp.


97

3.2.1 Compartmentalization of the Emulsion

Polymerization Process in a Continuous Flow

Microreactor

The possibility to transfer the emulsion polymerization reaction from a batch

reactor to a continuous flow microreactor resulted in the change of certain parameters and

conditions according to the characteristics of the microreactor system. The emulsion

polymerization reaction in a continuous flow microreactor with a single micro-mixer and a

finite capillary length is different from a traditional batch reactor, and may drastically

influence the polymerization reaction. The differences between batch reactor and

continuous flow microreactor were thought to be dependent on the reaction volume that is

a defined volume at a precise reaction time. In a batch reactor, theoretically at any reaction

time the reaction volume is identical while in the continuous flow microreactor, the same

reaction time is characterized by different every reaction volumes that are unique in their

properties and parameters since the system consists of different parts (injection system,

mixing system, tubing, etc) (Figure 3.7).


98

Figure 3.7 Illustration of polymerization reaction volume in the continuous flow microreactor.

In a continuous flow microreactor the composition of the reaction mixture were changing

along the length of the tubing. In addition, every part of the tubing with the corresponding

reaction volume represented a different stage of the polymerization reaction. Assuming

that every reaction volume moved independently along the tubing, each of these reaction

volumes were unique with their conversion rate and kinetic situation and behaved as a

small batch reactor at different reaction times along the length of the tubing. 21

The reduced diameter of the tubing influenced the reaction volume as well as the

flow rate and the pressure generated by the injection system and the mixing system

partially controlled the diffusion of the reactants between the two phases and the volume of

the droplets at different reaction times. All these technical characteristics of the continuous

flow microreactor influenced the emulsion polymerization process and the features of the

resulting polymer product (particle size, particle size distribution, molecular weight of the

polymer chains, solid content, etc.). The influence of continuous flow microreactors on the

emulsion polymerization process had to be understood to evaluate the data and further it

has also to be considered that this device afforded experimental conditions not achievable

in batch reactor such as extremely short reaction times. In conclusion, a full understanding


99

and use of the continuous flow technique was essential to investigate the emulsion

polymerization mechanism initiated by phosphine oxide photoinitiators.


100

3.3 Emulsification

Emulsion polymerization in general involves two immiscible phases. In our

investigations an oil-in-water emulsion was used where the continuous phase is an aqueous

phase and the discontinuous phase is constituted by the monomer and stabilized by a

surfactant. The emulsification process plays a key role for emulsion polymerization since

the hydrophobic monomers have to be transported in a multi-step process from a monomer

droplet to a polymer particle as main reaction loci. The first step in this process is the

passage of the phase boundary between the monomer droplets and water. In the second

step the diffusion through the aqueous phase to a polymer particle occurs. The last step

involves passing of the monomer the phase boundary between water and the particles. It is

clear from these steps that larger the monomer droplet–water interface, faster the passage

into the water phase. The passage of monomer as smallest possible droplets is essential for

achieving high polymerization rates. In the continuous flow microreactor emulsification is

influenced by mechanical emulsification, spontaneous emulsification and flow rate.

3.3.1 Mechanical Emulsification

Mechanical emulsification is characterized by the constraining of the two phases in

the reduced diameter channels of a mixer at well defined pressures and flow rates (Figure

3.8).


101

Figure 3.8 Droplet formation in a mixing system.22

In our continuous flow microreactor the mechanical emulsification occured in the

interdigital micro mixer SSIMM within seconds. To investigate the effects of micro mixer

on the emulsion polymerization process, a light microscopy image was taken one minute

after the passage of a styrene solution and aqueous solution containing sodium dodecyl

sulfate (SDS) and BAPO-AA into the mixer at a flow rate of 4 mL/min (Figure 3.9a).

Figure 3.9 (a) Styrene emulsion generated in the SSIMM interdigital micromixer. (b) TEM image

of polystyrene particles stained with PTA-Ru and generated by the standard procedure (300 mg

SDS, 10 mL water, 10 mg BAPO-AA, 2.5 mL styrene).

The results were remarkable since the average droplet size observed was smaller than the

predicted (median droplet size of about 50–70 µm)19 and there was no correlation between

the droplet size (average of 10 µm) and the final particle size (40 nm).

The rate of polymerization of styrene using BAPO-AA as photoinitiator and

varying SDS concentration was investigated. The rate of polymerization dropped when the


102

polymerization experiments were carried out with SDS concentration below 100 mg per 10

mL of water and at concentrations of 50 mg per 10 mL no polymerization was observed. A

very efficient emulsification using 300 mg of SDS in 10 mL water for 2.5 mL styrene

resulted high polymerization rates and complete conversion (18% of solid content in 36.5

s). A large interfacial area between the droplets and the aqueous phase guaranteed fast

monomer transport to the growing polymer particles. In addition, high surfactant

concentration stabilized the emulsion along the entire tubular reactor and guaranteed fast

monomer transport along the entire reactor even far away from the micromixer. The

observations were in accordance with studies on the effect of the surfactant concentration

in the continuous flow microreactor that have been previously reported.9e

3.3.2 Spontaneous Emulsification

The spontaneous emulsification that could occur in the SE zone of the tubing in the

continuous flow microreactor has to be considered (Figure 3.6). Spontaneous

emulsification was facilitated by surfactant high concentration.23 A simple experiment

illustrates the phenomena where the styrene or methyl methacrylate monomer was

quiescently placed on the top of an aqueous solution containing surfactant (300 mg/10 mL

of SDS) and UV spectra of the aqueous solution were recorded at different time points

(Figure 3.10).


103

Figure 3.10 UV-vis spectra illustrating the uptake of A – styrene and B- methyl methacrylate via

spontaneous emulsification into an aqueous surfactant solution (curve a - 36 sec, b - 72 sec, c -

10 min, and d - 3 h).

The concentration of the monomer increases very fast reflecting the growing monomer

concentration in the aqueous phase without the application of mechanical shear. This

process is called spontaneous emulsification since not only single monomer molecules, but

also monomer aggregates and nano-droplets are transferred. 23 Increasing the concentration

of SDS to 500 mg/10 mL led to higher conversion and smaller particles leading to the

assumption that, the empty but still existing micelles contributed to stabilize the tiny

formed particles.

3.3.3 Influence of the Flow Rate on the Emulsion

Polymerization

The flow rate influenced the emulsification and consequentially the rate of

polymerization. Decreased flow rates result in increased residence times in the continuous

flow microreactor that generally induced the formation of more reactive radicals since the

irradiation exposure of the photoinitiator is longer and therefore a higher initiation rate,

polymerization rate, and number of active growing particles might be expected. A solution

containing SDS, water and BAPO-AA and a styrene solution were flushed through the

continuous flow microreactor at a flow rate of 1 mL/min and then at a flow rate of 4

mL/min. Decreasing the flow rate resulted in the formation of two distinct optically

recognizable phases, signifying instability of the emulsion and consequentially resulting in

a lower polymerization rate.

In conclusion, the photo-induced polymerization in the continuous flow

microreactor required a high flow rate and sequentially short residence time. Phosphine

oxide photoinitiators decompose and initiate radical polymerizations fast enough for

achieving an effective polymerization in the continuous flow microreactor.24


104

3.4 Extraordinarily High Molecular

Weight Polymer in Small Particles

In continuous flow microreactors the polymerization of styrene occurred in an

unusually short residence time (36.5 sec) resulting in small particles (< 50 nm) with a

narrow particle size distribution (Figure 3.11) and polymer chains with extremely high

molecular weight. Initially the emulsion polymerization of styrene initiated by BAPO-AA

14 in a continuous flow microreactor (Scheme 3.5) in presence of different concentration

of sodium dodecyl sulfate (SDS, 200–500 mg/10 mL) was investigated.

Figure 3.11 SEM image of polystyrene particle generated in a microfluidic device.

Scheme 3.5 Structures of (2-(bis(2, 4, 6-trimethylbenzoyl)phosphoryl)acetic acid 14 (BAPO-AA).


105

At different surfactant concentrations an emulsion polymerization thermally-initiated with

potassium peroxodisulfate (KPS) or photo-initiated with poly(ethylene glycol)-azo-initiator

(16, PEGA200) in batch was performed. The size of the latex particles detected by

dynamic light scattering (DLS), generated by the batch and continuous flow microreactor

methods were compared and found to be similar (Figure 3.12).

Scheme 3.6 PEGA200 structure 16.

Figure 3.12 Correlation between average particle size (D) of polystyrene latex and the

concentration of emulsifier (SDS) after thermal batch polymerization initiated by KPS (white

square) and PEGA200 (grey square), and continuous flow polymerization photo-initiated by

BAPO-AA (red square).

After emulsion polymerization of styrene photo-initiated with BAPO-AA in aqueous phase

containing SDS in the continuous flow microreactor the latex was directly collected in a

solution of tetrahydrofuran (THF) which is known to be able to break the colloidal

dispersion and quench the polymerization reaction. The molecular weight distribution of

the polystyrene collected was up to 30·106 g/mol with peak maximum at 3·106 g/mol

(Figure 3.13).


106

Figure 3.13 Molecular weight distribution of polystyrene chains produced via photo-initiated

emulsion polymerization in the continuous flow microreactor. GPC analysis of the polymer

molecular weight used set of column B having a pore size of 106 Å allowing resolution of

molecular weights of up to 107 g/mol (see experimental part, latex characterization).

The high solid content (18%) and ultrahigh molecular weight polymers (up to 30·106

g/mol) obtained within this short reaction time was unprecedented and triggered an in-

depth investigation of the mechanism of BAPO-AA initiated process. The average

molecular weight of polystyrene from the photo-initiated polymerizations was up to 30

times higher than that of the polystyrene product from the batch reaction photo-initiated by

PEGA200 (Figure 3.14).


107

Figure 3.14 Correlation between cumulative number average molecular weight (Mn) of

polystyrene latex and concentration of emulsifier (SDS) after thermal batch polymerization

initiated by KPS (white circle) and PEGA200 (grey circle) and continuous flow photo-

polymerization initiated by BAPO-AA (red circle). GPC analysis of the polymer molecular weight

used set of column A having a pore size of 103 and 105 Å (see experimental part, latex

characterization).

Emulsifier concentrations between 200–500 mg SDS/10 mL water, well above the critical

micelle concentration (CMC = 23 mg SDS/10 mL water), were need for all photo-initiated

continuous flow reactions. According to the classical kinetics, the degree of polymerization

(DPkin) of emulsion polymerization (eq 7) with a rate constant Kp of 107.3 L/mol·s at

30 °C9d at a styrene concentration inside the particles (CM) of 3.5 M (estimated based on

equilibrium swelling of the particles),25 was calculated.

(7)

(8)

Kp is the propagation rate constant; Ktr,m the rate constant of chain transfer to monomer; CM the

monomer concentration at the reaction site; treac the reaction time.

The calculation of the DPkin indicated that the polystyrene chains should not grow larger

than 1.5·106 g/mol, an order of magnitude smaller than the observed value of 3·107 g/mol


108

for emulsion polymerization of styrene photo-initiated with BAPO-AA in a continuous

flow microreactor (Figure 3.15).

Figure 3.15 Comparison of the molecular weight distribution of polystyrene in the final latexes

obtained in the continuous flow microreactor (curve a, solid red) and in a batch reactor (curve b,

solid blue); the vertical lines indicate the kinetically accessible molecular weights (solid lines a1,

b1) and the maximum molecular weights attainable in the polymerization time (dashed lines a2, b2)

by multiplying DPkin and DPtime with the molecular weight of styrene. Polymerization conditions

curve a: 2.5 mL of styrene, 10 mL of water, 300 mg of SDS, 10 mg of BAPO-AA, 36.5 seconds

irradiation (polymerization) time, polymerization conditions curve b: 20 g of styrene, 80 g of water,

625 mg of SDS per 10 mL of water, 640 mg of KPS, 80 °C, batch reactor, 25 minutes

polymerization time.

Upon comparison of the molecular weight distribution of both processes, it was also

observed that the DP values for this process were inverted compared to a classical thermal

emulsion polymerization. The kinetically accessible DPkin value for the thermal

polymerization was lower than DPtime while for the BAPO-AA initiated emulsion

polymerization in a continuous flow microreactor the situation was reversed (Figure 3.15,

line b1 and line b2 vs a1 and a2).

To understand the anomalous molecular weight of the polymer chain, an

explanation on the polymer growth mechanism is required. It was clear that the classical


109

mechanism could not explain the results obtained during the emulsion polymerization in a

continuous flow microreactor.


110

3.5 Ultrafast Polymerization of Styrene

The standard analysis of the molecular weight of the latex obtained by the emulsion

polymerization BAPO-initiated in continuous flow microreactor, disclosed a high

molecular weight of the polymer chain, which was surprisingly not due to an extended

chain growth after irradiation time (post effect). The data collected proved that the

emulsion polymerization of styrene using BAPO-AA as initiator occurred in 36.5 s to give

full conversion. The rate of emulsion polymerization of styrene using BAPO-AA as

phospine oxide photoinitiator has been investigated in detail in the microfluidic device and

compared to a well known emulsion polymerization initiated by an azo-initiator

(azodi[poly(ethylene glycol)] isobutyrate, PEGA200, 16).

To investigate the rate of BAPO initiated polymerization and the eventual extended

chain growth after irradiation time (post-effect), the polymer latex that was formed was

checked by solid content and size exclusion chromatography (SEC) at different time

intervals. Solid content of the samples were assessed few hours after sample collection,

after three, five and 20 days. These results showed the absence of any post-polymerization

reaction (Figure 3.16 a). On the contrary, the solid content for a standard emulsion

polymerization photo-initiated by PEGA200 increased dramatically in the first three days

(Figure 3.16 a). The molecular weight distribution increased as well from the initial value

detected few hours after collecting the sample (black curve) during three (red curve), five

(green curve) and 20 days (blue curve) for a standard polymerization initiated by

PEGA200 (Figure 3.16 b). The emulsion polymerization was carried out and the molecular

weight distribution checked few hours after polymerization (black curve), after three, five

and 20 days (respectively red, green and blue curve, Figure 3.16 c). Overlapping curves

indicates that the chain size of the polymer does not increase with time.


111

Figure 3.16 Time-dependent changes of properties of polystyrene latex particles following photo-

initiated emulsion polymerization in a continuous flow reactor using BAPO-AA or PEGA200 as

initiators; a) Solids content (spheres, %) and average molecular weight (squares, Mw), BAPO-AA

(black) and PEGA200 (red); b) MWD shift for latex prepared using PEGA200; c) MWD shift for

the latex prepared using BAPO-AA; d) MWD comparison of both latexes after 20 days (BAPO-

AA – black, PEG200 – red). Light source: medium pressure Hg lamp 100 W (HBO 100).

Finally, molecular weight distribution of BAPO-AA and PEGA200 can be compared

demonstrating that BAPO-AA polymerization can reach the same molecular weight

distribution in 36.5 s as PEGA200 in 20 days (black curve vs red curve, Figure 3.16 d). In

conclusion, the polymerization rate was extremely fast (36.5 s) and did not show any post-

effect.

Another direct confirmation that the polymerization terminated in 36.5 s was the

scanning electron microscopy analysis (Figure 3.17a).


112

Figure 3.17 (a) Cryo-SEM image of polystyrene particles frozen immediately after removal from

the reactor. (b) Standard SEM image of polystyrene particles 24 hours after polymerization.

The comparison between the cryo-SEM image of a polystyrene latex sample frozen

immediately after collection and the high resolution image of the dried latex sample

prepared 24 hours after the polymerization showed that the particle size in both cases was

in a similar range. This result was a further confirmation that the entire polymerization

process occurred in the continuous flow device.

We investigated the decomposition mechanism of the initiators to explain the

rationale behind this behavior. The absence of post-polymerization using BAPO-AA may

be due to its decomposition pathway as acyl phosphine oxide photo-initiators decompose

after excitation and intersystem crossing to the triplet state.24a, 26 The azo-group of

PEGA200 16 decomposed, and generated free radicals, upon thermal treatment or

irradiation, and surfactant micelles further enhanced the decomposition of azo-initiators.27

In conclusion, post-effect was observed only using PEGA200 as initiator but not using

BAPO-AA.

Scheme 3.7 Decomposition of PEGA200 16.


113

3.6 Investigation of Emulsion

Polymerization using BAPO-AA

The behavior of 2-(bis(2, 4, 6-trimethylbenzoyl)phosphoryl)acetic acid (BAPO-

AA) under different polymerization conditions has been investigated. 28 A quite

comprehensive study on BAPO-AA was necessary to understand its role as photoinitiator

in emulsion polymerization reactions in continuous flow microreactors. The rate of

decomposition of BAPO-AA, the influence of its concentration on the polymerization, its

behavior at different irradiation times and finally the effect of different monomers on

BAPO-AA have been studied in detail.

3.5.1 Decomposition Rate of BAPO-AA

It was important to understand the decomposition of BAPO-AA to establish the

rate of its decomposition and the amount of radicals generated during the polymerization

process (36.5 sec). The high reactivity of BAPO-AA was evaluated by taking UV-spectra

after irradiating it in aqueous solution. The UV spectra were monitored in the range where

BAPO´s chromophore groups absorb (350–450 nm).17 After 36 s irradiation in a quartz

cuvette the absorption curve dropped considerably compared to the reference solution

(curve b vs curve a, Figure 3.18). After 72 s the absorption curve overlapped curve b due

to the absence of further decomposition, and longer irradiation (144 s) led to a slight

decrease that could still be detected (curve c, Figure 3.18).


114

Figure 3.18 UV spectra of BAPO-AA (10 mg/10 mL of water) after different irradiation time; a –

irradiation time zero, b – irradiation time 36 s and 72 s, c – irradiation time 144 s.

The absence of monomer in the solution allowed us to assume that most likely all radicals

involved in the decomposition reaction of BAPO-AA recombined to different byproducts

other than BAPO-AA or reacted with water. 29 Based on these data it can be concluded that

the decomposition of BAPO is very fast during the polymerization.


115

3.5.2 Emulsion Polymerization at Different BAPO-AA

Concentrations

The concentration of the initiator directly influences the number of radicals

generated upon irradiation and has to be considered a fundamental aspect to elucidate the

polymerization events. It is well known for free radical polymerization that the average

molecular weight of the polymer chain depends on the square root of the initiator

concentration.30 The unusual effects observed by varying the concentration of BAPO-AA,

on the molecular weight of the final polymer are reported here. The molecular weight

distribution of polystyrene did not change even though the BAPO-AA concentration was

increased three times compared to the initial concentration (Figure 3.19).

Figure 3.19 Molecular weight distribution of polystyrene prepared in a continuous flow

microreactor varying concentration of BAPO-AA, 5 mg (curve a), 10 mg (curve b), 15 mg

(curve c), 50 mg (curve d) in 10 mL water solution containing 300 mg SDS.

Only when the photoinitiator concentration was ten times higher did the molecular weight

shift slightly towards the lower value. The collected data showed that a special kinetic

situation using BAPO photoinitiators in continuous flow emulsion polymerization can

cause an unusual correlation between the photoinitiator concentration and the average

molecular weight.


116

3.5.3 Emulsion Polymerization using BAPO-AA at

Different Residence Times

Further investigations of the irradiation times were performed in an effort to

understand the effect of this parameter on the average molecular weight distribution of the

final polystyrene chain. Experiments were carried out blocking different portions of the

irradiation zone with aluminum foil, but keeping all other parameters and condition the

same. The original size exclusion chromatography (SEC) profile clearly showed an

increase in the polymerization conversion by prolonging the irradiation time, expressed via

the increase in the refractive index signal (Figure 3.20A).

Figure 3.20 Original SEC elugrams (A) and normalized molecular weight distributions (B) of

polystyrene formed in a continuous flow microreactor with varying length irradiation zones.

Molecular weight distributions of the final polymer chain after 14.6 s (curve a), 21.9 s (curve b),

29.2 s (curve c) and 36.5 s (curve d) of irradiation time.

While this data exposes the significant influence of the irradiation time on the final

polymerization conversions, there was not obviously correlation between the irradiation

time and the molecular weight distributions of the polymer chains.


117

3.5.4 Screening of Different Monomers for Emulsion

Polymerization

The effect of the monomers on the emulsion polymerization process photo-initiated

by BAPO-AA in a continuous flow microreactor was also investigated. To understand the

emulsion polymerization process, the behavior of three different monomers (styrene, STY

17; butyl methacrylate, BMA 18; methyl methacrylate, MMA 19; Scheme 3.8) in the

continuous flow microreactor had to be analyzed.

Scheme 3.8 Styrene (STY) 17; butyl methacrylate (BMA) 18; methyl methacrylate (MMA) 19.

Initially a series of experiments were perfomed to optimize the emulsion

polymerization conditions for STY, BMA and MMA. The optimum residence time for

STY BMA and MMA was 36.5 s, 72 s and 144 s and this was determined by the screening

different residence time in the continuous flow microreactor (Table 3.1).


118

Table 3.1 Flow rates and residence times for different monomers in continuous flow microreactor during emulsion polymerization.

Monomer Photoinitiator Flow Rate

(mL/min)

Reaction

Time (s) Kp (L mol-1s-1)

Solubility in

Water

(mol/L)

STY BAPO-AA 4.0 36.5 88 0.003 BMA BAPO-AA 2.0 72 368 0.002 MMA BAPO-AA 1.0 144 318 0.160

Styrene (STY); Butyl methacrylate (BMA); Methyl-methacrylate (MMA); (2-(bis(2, 4, 6-

trimethylbenzoyl)phosphoryl)acetic acid (BAPO-AA), propagation rate constant at 25 °C (Kp), 31

Solubility in Water detected at 25 °C.

In order to explain the different residence time exhibited by STY, BMA and MMA, the

solubility in water and the propagation rate constant of the monomers could be considered.

Apparently, lower the water-solubility of the monomer, shorter is residence time for the

polymerization even though the water-solubility between styrene and BMA is only slightly

different. Interestingly, the optimum average residence time shows no correlation with the

value of the propagation rate constants of the monomers. All these data suggest that the

difference in the irradiation time for the emulsion polymerization of different monomers

could be due to the termination modes of the same monomers.

The maximum peak in the high molecular weight part of the molecular weight distribution

is inversely proportional to the residence time. It is at the highest molecular weight for

STY (Figure 3.21, curve c), shifted towards lower values for BMA (Figure 3.21, curve b),

and even further shifted for MMA (Figure 3.21, curve a).


119

Figure 3.21 Molecular weight distribution of polymer latexes prepared in a continuous flow

microreactor with various monomers: styrene (curve c), butyl methacrylate (curve b) and methyl

methacrylate (curve a).

In order to explain the molecular weight distribution results, the termination mode and

polarity of the monomer have to be taken into account. The different termination behavior

of styrene and methyl acrylate affected the reaction conditions and the polymer features.

Styrene terminates exclusively by combination which occurs when the polymer growth is

stopped by recombination of two growing radical to form a single unreactive polymer

chain. Methyl methacrylate on the other hand terminates mainly by disproportionation

event which halts the chain growth by the extraction of a proton leading to two unreactive

polymers as a consequence of hydrogen transfers (Scheme 3.9).

Scheme 3.9 Termination mode by (a) combination and (b) disproportionation.

To explain the highest value of molecular weight for STY the lowest polarity of styrene

should be considered which leads to the highest concentration in the latex particles and

then the termination mode which goes exclusively via combination.32 For methacrylate the

termination occurs via disproportionation or combination (generation of two chains or one


120

chain). Due to the different termination mode starting from the same reaction conditions,

shorter chains are more likely for methacrylate than for STY.


121

3.7 Studies on Different Phosphine Oxide

Initiators

In order to check the influence of the structure and reactivity of the

photoinitiators on the emulsion polymerization, a study with different phosphine oxide

photoinitiators varying their concentrations was carried out. In this regard three phosphine

oxides initiators have been considered: 2, 4, 6-trimethylbenzoyl-diphenylphosphineoxide

(MAPO 1), 2-(bis(2, 4, 6-trimethylbenzoyl)phosphoryl)acetic acid (BAPO-AA 14) and

bis(2, 4, 6-trimethylbenzoyl)phosphineoxide tris-oxyethylene (BAPO-PEG 15) (Scheme

3.10). 13a, 13b The first photoinitiator is a mono(acyl) phopshine oxide and is soluble only

in the monomer phase while the latter two are bis(acyl) phosphine oxide initiators that are

water–soluble initiators.

Scheme 3.10 Photoinitiator structures: 2,4,6-trimethylbenzoyl-diphenylphosphineoxide (Lucirin

TPO, MAPO) 1; 2-(bis(2, 4, 6-trimethylbenzoyl)phosphoryl)acetic acid (BAPO-AA) 14; bis(2, 4,

6-trimethylbenzoyl)phosphineoxide tris-oxyethylene (BAPO-PEG) 15.

The molecular weight distribution of the polystyrene chain obtained by emulsion

polymerization using BAPO-AA and BAPO-PEG were comparable (Figure 3.22).


122

Figure 3.22 Molecular weight distribution of polymer latexes prepared in the continuous flow

microreactor with different photoinitiators: BAPO-AA (curve a), BAPO-PEG (curve b), MAPO

(curve c).

As expected, the molecular weight distribution of the polymer generated using MAPO

shifted towards a lower value due to the hydrophobicity of the photoinitiator that results in

a higher concentration of radicals inside the particles. Despite these results, a significant

fraction of chains produced by MAPO had an extremely high molecular weight (> 1.65·106

g/mol).


123

3.8 Molecular Weight Distribution and

Average Sizes of Polymer Particles

Emulsion polymerization in a continuous flow microreactor initiated by phosphine

oxide influence in an unusual way the relationship between molecular weight distribution

of the polymer chain and the average particle size. The average particle size of polystyrene

particles obtained by emulsion polymerizations in continuous flow microreactors using

different concentrations of SDS (between 200–500 mg), 10 mL of water and 2.5 mL of the

STY were compared to the average particle size diameter of polystyrene particles produced

via thermal polymerization in a batch reactor reported previously (Figure 3.23 A).33

Figure 3.23 Experimental data comparing properties of latex particles prepared via thermal

polymerization in the stirred tank reactor and photopolymerization in the microfluidic device. Log

– log plot of the average particle size versus the surfactant concentration (A), black and grey circles

are data for polymerization in the batch stirred tank and microfluidic reactor, respectively. The

number average molecular weight versus the average particle size (B), black symbols are data for

polymerization in the batch stirred tank and grey symbols for polymerization from microreactor;

respectively, circles, squares, and triangles representing styrene, BMA, and MMA polymerizations,

respectively; the lines are linear regressions.

The relationship between the average particle size and the surfactant concentration did not

depend on either the polymerization method or the initiation conditions (black spheres


124

represent data from thermal reactions in batch reactors and grey spheres emulsion

polymerization photoinitiated in continuous flow microreactor; Figure 3.23 A).

In contrast, considering all three monomers (STY, BMA, MMA) the relationship

between the average molecular weight and the average particle size depended strongly on

the polymerization process (Figure 3.23 B). The emulsion polymerization was run in a

continuous flow microreactor using a solution of SDS, BAPO-AA in water and a solution

of STY, BMA or MMA with a residence time of 36.5 s, 72 s and 144 s respectively and the

obtained particles were analyzed. The data collected were compared with the data

extrapolated by analysis of polystyrene particles produced using thermal polymerization in

batch reactor and reported previously.33 The average molecular weight decreased with

increasing average particle size for the thermal polymerization in stirred batch reactors in

accordance with common emulsion polymerization mechanisms.34 For polymer particles

with a size below 70 nm in diameter equals 0.5 and entry of a second radical instantly

causes termination. As the entry rate increases with growing particle size, the average

molecular weight decreases. With increasing average particle size the

compartmentalization effect, which was responsible for the high molecular weight of

emulsion polymers, was more and more weakened and hence, the average molecular

weight decreased (black circles, line b, Figure 3.23 B). The photo-initiated emulsion

polymerization in continuous flow microreactor showed the opposite effect (line b, Figure

3.23 B). For all three monomers the average molecular weight increased with increasing

particle size (Figure 3.25 B, black triangles correspond to MMA, black squares to BMA,

black circles to styrene).

The relation between molecular weight distribution and average particle size was

completely reversed in emulsion polymerizations initiated by phosphine oxide in a

continuous flow microreactor compared to normal emulsion polymerizations in batch

reactor.33 In conclusion, the chain growth follows different rules and the molecular weight

is not determined by radical entry but by the monomer concentration inside the particles

which increases with increasing particle size.35


125

3.9 Snowballing Radical Generation

Mechanism

Established theories of the mechanism of emulsion polymerization do not explain

how phosphine oxide-photoinitiated polymerization can generate ultrahigh molecular

weight polymers in a continuous flow reactor in short residence time. Even though it is

usually possible to achieve high molecular weight product due to a post-polymerization

effect, (it is well known that the isolated polymeric radical can survive inside the latex

particles for many hours after the radical generation is terminated),36 no post-effects were

observed in BAPO-AA-initiated continuous flow emulsion polymerizations. The

explanation of this result required a reconsideration of the mechanism of the polymer

growth. Studies have shown that more than one polymer chain can grow per radical during

the duration of a classical emulsion polymerization. It was clear that the classical

mechanism does not explain the results obtained during the emulsion polymerization in a

continuous flow reactor. A different decomposition of the bis(acyl)phosphine oxide

photoinitiator, resulting in a new mechanism could provide a possible explanation for these

observations.

3.9.1 Average Number of Radicals per Polymer Particle

( )

To justify the generation of an ultra fast polymerization yielding an ultra high

molecular weight polymer chains, different parameters have to be considered compared to

the traditional batch polymerization. Two fundamental values determined the average

number of polymer chains per particle:


126

1) The number of chains that can grow to an average chain length during the propagation

time (np,kin). This depends on the time a growing polymer chain resides in the reaction zone

(tres), compared to the time (tadd) it takes for a radical to add DP monomer units at a

monomer concentration CM with propagation rate constant kp (eq. 9).

(9)

2) The geometric constraint (np,geo) the number of chains of density ρp with a given

molecular weight (DP times the molecular weight of the monomer Mmon) that fill the

volume of an average particle of diameter D (eq. 10).

(10)

The ratio of np,geo/np,kin constitutes an alternative definition of the average number of

radicals per particle and allows for the calculation of from experimental data (eq. 11),

where c1 is a geometric constant.

(11)

To understand if the results observed could be explained by a bigger number of

radicals per particle, equation 11 was used to calculate the theoretical values for styrene

and butyl methacrylate (BMA) polymerization initiated with BAPO-AA 1, BAPO-PEG 2,

and the commercially available mono-acyl phosphine oxide 2, 4, 6-trimethylbenzoyl-

diphenylphosphine oxide (Lucirin TPO or MAPO). Higher values were calculated for all

photo-initiated polymerization reactions as compared to the zero-one scenario and

confirmed the presence of many growing radicals per particles (Figure 3.24).


127

Figure 3.24 Average number of radicals per particle ( ) calculated using equation (5) for

photoinitiated emulsion polymerizations in a continuous flow microreactor with different

monomers, photoinitiators and residence times correlated with the average particle size.

The large number of radicals per particle can be best explained by the incorporation of the

phosphine oxide units into the polymer backbone, which upon photolysis, results in

“snowballing” radical generation (SRG) during the emulsion polymerization process.

3.9.2 Decomposition of BAPO-AA

For the purpose of the newly proposed mechanism, the following criteria were

considered: 1) Radical formation is faster than the aggregation of particles during the

nucleation process.37 2) Phosphorus-radicals are more reactive than the carbon radicals.24,

26, 38 3) Generation of diradical species during radical polymerization is also possible.39

Initially, decomposition of phosphine oxide initiator generates two primary radicals

(20 and 22) before these radicals propagate the chain length to form polymer via monomer

addition (21 and 23) (Scheme 3.11).


128

Scheme 3.11 Photochemical decomposition of BAPO-AA and subsequent polymerization of

styrene leading to polymer chains enabling snowballing radical generation

The polymer chain containing the phosphine oxide unit can then undergo decomposition in

four possible ways generating eight different kinds of radicals, a phosphorus centered

radical (22), a diradical chain presenting a phosphorus centered radical and a carbon


129

centered radical (24), three carbon centered radicals (20, 21 and 23) and three diradical

chain presenting two carbon centered radicals (25, 26 and 27), each of which can undergo

further propagation of the chain or further decomposition. For example, upon irradiation

radical 26 can generate intermediates 24 and 25 upon irradiation (Figure 3.11). Finally

radicals can combine: in case of monoradicals, this leads to termination (28) while for the

diradicals the polymer chains can still propagate (26 and 29).40

Scheme 3.12 Possible pathway of combination of radical polymer chain resulting in (a) termination

and (b) diradical polymer chains.

In the later diradical combination, if the phosphine oxide is already present in a diradical

molecule, two different diradicals are obtained as a result (29).41, 39b The diradical chains

are formed during decomposition when the phosphine oxide group is linked to at least one

radical-containing chain (23 and 26).13b The rate of radical combination is dependent on

the mobility of both the radicals involved in the reaction. 13C-NMR analysis of purified polystyrene polymers revealed a sharp signal at δ 13C

~31 ppm (Figure 3.25), an indication of tail-to-tail styrene additions (27). Polystyrene

samples prepared by conventional means show no signal in this region, but only the

expected aliphatic CH (40–41 ppm) and CH2 (42–48 ppm) resonances indicating the

presence of atactic polystyrene in both samples.42, 34


130

Figure 3.25 13C-NMR spectra of the polystyrene generated by emulsion polymerization in

continuous flow microreactor using BAPO-AA after purification (blue curve) and of thermally

polymerized polystyrene (black curve).

To confirm the presence of 23 and 26 in the long polymer chain and the fact that

this mechanism is universal for all the phosphine oxide initiators, 31P-NMR spectra of the

reprecipitated polymer generated by di(acyl)phosphane oxide (Irgacure® 819, scheme 3.13,

8) were measured.

Scheme 3.13 Chemical structure of bis(2, 4, 6-trimethylbenzoyl)-phenylphosphine oxide 8

(Irgacure® 819).

The signals at 25-27 ppm show the presence of a phosphorus unit at the end of the chain

(23, Figure 3.26 a). In this case the polymer was prepared at a 100 times higher

concentration of BAPOs, only one α-cleavage occurs and the other chromophoric center is

intact while the use of a lower concentration of BAPOs results in the α-cleavage also of the

second chromophore. The signals at 39–41 ppm demonstrate the presence of the phosphine

oxide residue in the middle of the polymer chain (26) (Figure 3.26 b).


131

Figure 3.26 (a) 31P-NMR spectra of polystyrene polymer containing compound 23; (b) 31P-NMR

spectra of polystyrene polymer 26.

In addition, emulsion polymerization experiments using 2,4,6-tribenzyl-

phosphineoxide (TBP, scheme 3.14, 30) as photoinitiator were carried out leading to cross-

linked products but with lower efficiency under longer irradiation times.


132

Scheme 3.14 Chemical structure of 2, 4, 6-tribenzyl-phosphinoxide (TBP) 30.

The observed behavior can be explained by the lower decomposition rate of the

phosphorus-benzyl bond as compared to the phosphorus-acyl bond. The isolated cross-

linked product could only have been because of the cleavage of all the three P-C bonds,

and the growth polymer of the polymer from these sites. In addition, the molecular weight

of the linear polymer increased with increasing residence time reaching up to 107 g/mol for

residence times above 10 minutes (Figure 3.27).

Figure 3.27 Molecular weight distribution of the linear chain (soluble fraction) polymer in

polystyrene particles prepared using 2, 4, 6-tribenzyl-phosphineoxide under varying residence

times in the irradiation zone of the microfluidic device.

In conclusion, to explain the observed results, the new mechanism proposed demonstrates

the concept of snowballing radical generation (SRG) by a repeated photo-induced cleavage

of the phosphine oxide initiators producing unprecedented large number of radicals.


133

3.10 Snowballing Radical Generation

Simulation

In order to understand the influence of snowballing radical generation on the

emulsion polymerization kinetics and to provide a mechanistic rationale for the

experimental results, a simulation based on a stochastic model was formulated.43 The

simulation assumes an idealized situation where a radical enters a 40 nm spherical droplet

of pure monomer and propagates for two seconds without additional radicals entering or

exiting.

The modeling frame considered the following kinetic events (represented as equations):

1. Primary BAPO decomposition:

2. Propagation of the polymer chain:

3. Decomposition of phosphine oxide-containing compounds:

4. Radical combination of monoradicals:

5. Radical combination of diradicals:

B= benzoyl group, PO= phosphine oxide group, h is Planck’s constant and v is the frequency of the

irradiating light. The rate constant of BAPO photodecomposition is kd. Radicals are indicated using


134

the symbol (●). R● is an arbitrary molecule or macromolecule containing a radical, M is a monomer

molecule and kp is the rate constant of radical propagation. R● represents all possible radicals and

diradicals present in the system, including: R’-B●, R’-PO●, R’-M●, ●B-R’-B●, ●B-R’-PO●, etc.; kd2

is the kinetic rate constant of photodecomposition of all other phosphine oxide-containing

molecules; kc is the rate constant of radical combination.

The simulation model was based on the the events occurring during the decomposition of

phosphine oxide initiators. In the simulation model, the polymer chains were represented

by a vector of varying length. Each vector contains information about length, structure and

composition of the polymer chain. Vector manipulation rules have been defined on the

basis of the events described above. In Figure 3.28, the results of the simulation are

represented and compared to those for a typical radical polymerization.

Figure 3.28 Simulation of the snowballing radical polymerization mechanism. Final chain length

distribution when a single radical with DP = 1 (green bar) starts to polymerize in a 40 nm sized

monomer droplet. Distribution based on a phosphine oxide radical snowballing mechanism (red

bars), growth of the starting radical based on normal radical polymerization kinetics (grey bar).

The average chain length based on a phosphine oxide radical snowballing mechanism is

represented in red bars, the maximum chain length achieved by a polymer radical based on

normal radical polymerization kinetics was almost one order of magnitude lower (grey

bar). The green bar represents the starting point for the growth of a single radical entering a

40 nm size droplet. The simulation predicts the formation of 35 growing radicals from a


135

single phosphine oxide radical, and justifies the unusually high polymerization rate. On the

other hand, the generation and further recombination of diradicals also explains the

formation of polymers with extremely high molecular weights.

The average chain length ( ) increases at similar rates for SRG (red line) and for

traditional polymerization (black line) because it was assumed that the rate of propagation

for both scenarios was the same (Figure 3.29).

Figure 3.29 Time dependent growth of the average chain length ( ) for snowballing (red line)

and ordinary radical (black line) polymerization. DPmax simulated for snowballing kinetics

(spheres).

The small differences in the average chain length observed were due to the stochastic

model used during the simulation. Maximum chain lengths (DPmax) were simulated for

polymers produced by snowballing kinetics (spheres), showing once more that the broad

distribution of molecular weight of the resulting polymer chains is a results of the SRG

mechanism.

3.10.1 Simulation of the Emulsion Polymerization at

Different Initiator Concentration


136

The simulation of the emulsion polymerization was expanded to underpin also

other experimental results collected during the investigation using BAPO-AA

photoinitiator (see Section 3.14.8 for simulation assumption for the data). The influence of

SRG on the chain length distribution in the final polymer using two different concentration

of photoinitiator to start the polymerization was considered during the simulation (Figure

3.30 A).

Figure 3.30 Simulated chain length distributions for radical polymerization inside styrene droplets

assuming SRG (graph A) and ordinary kinetics (graph B) for two different initial initiator

concentrations (I0); grey bars I0 = 10 mM and black bars I0 = 1mM. Reaction rate constants: kd =

100 s-1, kp = 240 lM-1s-1, kc = 300 lM-1s-1, kd2 = 100 s-1.

For both simulations there was an expected shift to higher values when the concentration

decreased. This effect was more evident for the ordinary kinetics than for the snowballing

radical generation (SRG). The minor influence of the change in the concentration of

BAPO-AA was consistent with the experimental data that was collected (Figure 3.19,

Section 3.5.2). The simulation showed that SRG leads to broader chain distribution length

of the polymer than the traditional radical kinetics in accordance with the results collected

during polymerization investigation.

Same simulation study allowed results on the development of the chain length.

These data were experimentally difficult to access due to the short residence time in the

irradiation zone (Figure 3.31, symbols).


137

Figure 3.31 Simulated conversion dependence of average (graph A) and maximum (graph B) chain

length in the chain length distribution for radical polymerization inside styrene droplets assuming

SRG (symbols) and ordinary kinetics (lines) for two different initial initiator concentrations (I0);

grey circles and solid lines I0 = 10 mM; grey triangles and dashed lines I0 = 1mM.

The average degree of polymerization (DPave, graph A) and the maximum chain length

achievable (DPmax, graph B) were plotted against the conversion at two concentrations of

photoinitiator. All the data were compared to the classical kinetics for emulsion

polymerization (Figure 3.31, lines). At low conversion the DPave and DPmax were almost

overlapping independently from the photoinitiator concentrations, while at higher

conversion the photoinitiator concentration played a clear role. From the simulation data it

is clear that DPmax was strongly dependent on the photoinitiator concentration only in the

classical polymerization (Figure 3.31 B). Moreover, DPmax increased smoothly with the

conversion in the classical polymerization but as a scatter in the phosphine oxide mediated

polymerization confirming the stochastic nature of the chain scission in SRG. These

simulations interestingly revealed again the fundamental difference between SRG and

common radical polymerization kinetics. The classical kinetic demonstrated higher DPave

values but lower DPmax. These data supported and explained the experimental data where

SRG lead to a much broader chain length distribution.


138

3.11 Seed Polymerization of Polymer

Particles

According to the SRG mechanism, the use of phosphine oxide photoinitiators allows

for the incorporation of phosphine oxide units in the polymer backbone, which enable,

after extensive dialysis of the polymer latex and addition of a monomer, the restarting of

the polymerization by irradiation without additional photoinitiator. Upon irradiation, the

incorporated phosphine oxide units go through further cleavage allowing for a seed

polymerization. Seed polymerization experiments are reported as support of the validity of

the SRG mechanism.

The purified polystyrene latex, generated in the continuous flow microreactor, was

swollen with butyl methacrylate (BMA) and re-irradiated without the addition of additional

photoinitiator. The black curve in figure 3.32 illustrates the molecular weight distribution

of cleaned polystyrene seed latex prepared with BAPO-AA in a continuous flow reactor

and the red line represents the molecular weight distribution of the latex after seed

polymerization.

Figure 3.32 Molecular weight distributions of the starting polystyrene polymer generated using

BAPO (black curve) and MAPO (black dashed curve) and after seed polymerization using BMA

(red curve for the seed latex generated by BAPO and red dashed curve for the seed latex generated

by MAPO).


139

The shift in the molecular weight distribution clearly shows the re-starting of the

polymerization process due to the presence of phosphine oxide groups. To confirm the

behavior of phosphine oxide initiators the same set of experiments were repeated using

MAPO as photoinitiator (dashed curve, Figure 3.32). The black dashed curve shows the

molecular weight distribution of purified polystyrene latex prepared with MAPO and the

red dashed curve, which is shifted to lower values, the molecular weight distribution of the

latex after seed polymerization using BMA.

FT-IR spectroscopic analyses of the “seed” polymer, and of the polymers obtained

after restarting the polymerization with BMA, showed that an IR absorption peak that

corresponds to the carbonyl stretching frequency appears only after seeded polymerization

(Figure 3.33).

Figure 3.33 FT-IR spectra of the seed polymer (black spectrum) and the copolymers after seeded

polymerizations (red spectra).

This data clearly confirmed that an incorporation of butyl methacrylate units occurred

during the seed polymerization. No further polymerization was detected when polystyrene

latex generated by a non-phosphine oxide type photoinitiator was swollen and irradiated.

Direct evidence through IR or 31P-NMR spectroscopy studies of the presence of the

phosphine oxide units incorporated in the polymer chain prepared in continuous flow

microreactor using BAPO-AA as photoinitiator, were difficult to obtain since the

concentration of phosphorus unit in the extremely long polymer chain is too low to be

detected.


140

3.12 Applications of SRG Mechanism

Emulsion polymerizations studies in continuous flow microreactor using phosphine

oxide photoinitiator have lead to the discovery of SRG. This knowledge on the SRG

mechanism was further used to show how the phosphine oxide units incorporated in the

polymeric chains could be employed to reinitiate the polymerization as a new method for

the synthesis of materials. The features of the emulsion polymerization process in

combination with SRG mechanism allowed the incorporation only of few phosphine oxide

units in the polymer chain, to obtain higher number of these active units in the polymer

chain a different process is required. As the SRG mechanism should be applicable to every

polymerization system we used a precursor polymer (PS-PO) prepared via bulk

polymerization of styrene with the commercially available hydrophobic initiator bis(2, 4,

6-trimethylbenzoyl)-phenylphosphine oxide (8, Irgacure® 819). The concentration of

photoinitiator was increased 100 times compared to the standard concentration because of

the necessity to have a higher concentration of phosphine oxide repeating units in the

polymer backbone. An analysis of the average molecular weight of the precursor polymer

revealed shorter chain length than the chain lengths of the polymer prepared with standard

concentration of BAPO (Figure 3.34).

Figure 3.34 Molecular weight distribution of PS-PO precursor polymers prepared via bulk

polymerization, standard concentration of photoinitiator (curve 1) and 100 times higher (curve 2).

These latter PS-PO precursor polymers, possessing a greater amount of photo-cleavable

bonds per unit mass, as characterized by 31P-NMR signals at 25–27 ppm (Figure 3.25 a),


141

were used to initiate different polymerization reactions photolytically. Some possible

applications using PS-PO precursor follow in this chapter.

The PS-PO polymer was used for the preparation of a thermoplastic–elastomeric

glass glue. 10 weight-% solution of PS-PO in butyl acrylate (BA) was prepared and spread

between two thin glass plates. After irradiation both the glass plates were firmly stuck

together by the completely transparent PS-PO-poly(butylacrylate) PBA block copolymer

(Figure 3.35 a).

Figure 3.35 (a) thermoplastic–elastomeric glass glue composed of polystyrene-b-poly(butyl

acrylate)(PBA) block copolymer generated by photo-initiated polymerization, and (b) glass slides

glued together by a mixture of PS and PBA homopolymers produced by photo-polymerization of a

PS solution using Irgacure® 819. (c) Light microscopy image of a portion of (a) and (d) Light

microscopy image of a portion of (b).

If a mixture of non–reactive polystyrene generated by traditional polymerization, and BA

was polymerized using Irgacure® 819, a strongly turbid layer was formed, due to the

macroscopic demixing of the two homopolymers (Figure 3.35 b). The phosphine oxide

units in the PS-PO precursor allowed for the re-initiation of the polymerization and the

subsequent formation of the block copolymer. The presence of the copolymer was

demostrated by the absence of macroscopic phase separation (Figure 3.35 c). The phase

separation was clearly visible at the same magnification for the mixture of two

homopolymers (Figure 3.35 d).


142

The PS-PO prepared by this method was employed in the synthesis of a different

block copolymer PS-b-poly(vinyl acetate), that is difficult to prepare by other

polymerization methods. PS-b-poly(vinyl acetate) (PVA) copolymer can be prepared

usually only by multi-step procedures based on controlled radical polymerization

techniques using copper, cobalt or sulfur compounds.44 Irradiation of a solution of PS-PO

precursor in vinyl acetate produced an elastic PS-PO-PVAc block copolymer that retains

the shape of the reactor used for the preparation. The resulting PS-PO-PVAc block

copolymer was dissolved in glacial acetic acid (Figure 3.36 a) and a light microscope

image shows the block copolymer dispersed in the solvent (Figure 3.36 c).

Figure 3.36 (a) PS-PO—PVAc block copolymer disperses in glacial acetic acid; (b) PS-PO—

PVAc block copolymer immediately after the polymerization; (c) light microscopy images of the

PS-PO—PVAc block copolymer dispersed in glacial acetic acid.

The comparison of the dispersion of the obtained PS-PO-PVAc block copolymer in glacial

acetic acid with a mixture of PS and PVA in the same acid provided further evidence for

the formation of the co-polymer. While the mixture of the two homopolymers (PS and

PVA) showed flocculated chunks in the solvent (Figure 3.37 a), the PS-PO-PVAc block

copolymer formed a homogenous dispersion.


143

Figure 3.37 (a) Solution of polystyrene (PS) and polyvinylacetate (PVA) homopolymers in

glacial acetic acid, and (b) of PS-PO-PVA block copolymer prepared using PS-PO as initiator.

The PS-PO precursor was also used as an initiator for the emulsion polymerization of

styrene. In a standard procedure BAPO-AA was replaced with 500 mg of PS-PO. The

resulting polymer was characterized by large particle size distribution, average particle size

of 140 nm. The molecular weight distribution was slightly shifted towards higher

molecular weights (Figure 3.38 and 3.39).

Figure 3.38 Transmission electron microscopy images of polystyrene latex particles obtained by

photoinitiated polymerization with PS-PO as initiator.


144

Figure 3.39 Molecular weight distribution of PS-PO precursor (red dashed curve) and polystyrene

obtained after photoinitiated polymerization with PS-PO as initiator (black curve).

As visual proof-of-concept, a 50 weight-% PS-PO solution in a styrene-butyl

acrylate mixture (3:2 g/g) containing a yellow dye (Hostasol Yellow 36, Clariant) was used

as ink to photo-type through a PTFE stencil “SRG” after irradiation with a visible light for

30 min (Figure 3.40).

Figure 3.40 (a) Snapshots of the preparation of the photo-typing of “SRG” on a microscopy glass

slide through a PTFE stencil (Osram, L18 W, light color 840, lumilux, cold white) and (b) of the

photo-typed letters.


145

3.13 Conclusions

The use of continuous flow devices to study the phosphine oxide initiated emulsion

polymerization process led to the discovery of the mechanism of these photoinitiators:

snowballing radical generation (SRG). Photolysis of BAPOs rapidly generates an

avalanche of radicals that produce polymer chains with ultra-high molecular weight up to

30·106 g/mol. The phosphine oxide photoinitiators decompose and integrate as units into

the polymer chains allowing for repeated radical generation upon irradiation. This property

was exploited to prepare materials with different morphologies that were difficult to

produce previously. A stochastic model was used to simulate snowballing kinetics and to

quantitatively rationalize the polymerization process. The discovery of this new

mechanism was only possible by a unique combination of heterophase polymerization,

phosphine oxide photoinitiators and the use of a continuous flow microreactor, will have a

significant impact on the application of bis-acylphosphine oxide initiators and underscore

the importance of using continuous flow reactor in reaction mechanism studies.


146


3.14.1 Chemicals

The water used throughout was purified using a Seral purification system

(PURELAB Plus) or an Integra UV plus (SG Reinstwassersysteme) system with a

conductivity of 0.06 µS cm-1. The monomers styrene (Aldrich), methyl methacrylate

(MMA, Aldrich), and butyl methacrylate (BMA, Aldrich) were purified by distillation

under reduced pressure and stored in a refrigerator. Before use the monomers were

checked regarding oligomers by instilling a drop into an excess of methanol. Only

oligomer-free monomers were applied in the polymerizations. Sodium dodecyl sulfate,

ultrapure, (SDS, Roth GmbH) and potassium peroxodisulfate (KPS, Aldrich) were

employed as received. 2-(bis(2,4,6-trimethylbenzoyl)phosphoryl)acetic acid (BAPO-AA),

bis(2,4,6-trimethylbenzoyl)phosphineoxide tris-oxyethylene (BAPO-PEG) were

synthesized as described previously28c and 2,4,6-trimethylbenzoyl-diphenylphosphineoxide

(MAPO) was a gift from BASF.

3.14.2 General Procedure for Emulsion Polymerizations

Emulsion polymerization in microfluidic device. A syringe was filled with monomers (2.5

mL as total volume) and a second syringe with a solution previously prepared of sodium

dodecyl sulphate (SDS), photoinitiator (10 mg) and ultrapure water (10 mL). All

operations involving the photoinitiator before and after the irradiation time were performed

in the dark. The solutions were flushed into the reactor and the resulting reaction mixture

was collected in a flask under stirring.


147

Emulsion polymerization in a stirred tank reactor. The polymerizations were carried out

according to standard procedures either in glass reactor or a CPA200 reaction calorimeter

(ChemiSens, Sweden).45

3.14.3 Latex Characterization

Solid content was determined using a HR73 Halogen Moisture Analyzer (Mettler

Toledo). Average particle size (intensity-weighted diameter) was determined with a

Nicomp particle sizer (PSS Santa Barbara, USA, model 370) at a fixed scattering angle of

90°.

Molecular weight distributions (MWD) were determined by gel permeation

chromatography (GPC) and were used to calculate weight and number average molecular

weight of the polymers (Mw, Mn). GPC was carried out by injecting 100 µL polymer

solutions (solvent tetrahydrofuran (THF)) through a Teflon-filter with a mesh size of

450 nm, into a Thermo Separation Products set-up equipped with ultra violet (UV) (TSP

UV1000), and refractive index (RI) (Shodex RI-71) detectors in THF at 30 °C with a flow

rate of 1 mL/min. Two column sets were employed. Column set A consists of two 300 x 8

mm columns filled with a MZ-SDplus spherical polystyrene gel (average particle size 5

µm) with a pore size of 103 and 105 Å, respectively. For column set B an additional

column with a pore size of 106 Å was used allowing resolution of molecular weights of up

to 107 g/mol. Average molecular weights, and number of average molecular weight

polymers (Mw and, Mn) were calculated based on polystyrene standards (between 500 and

2·106 g mol-1 from PSS, Mainz, Germany).

FT-IR spectra were recorded with a Varian 1000 FT-IR on an ATR Golden Gate (Pike)

cell. Broadband decoupled 13C-NMR spectra were recorded using a Bruker Spectrospin (DPX400) spectrometer. The

data acquisition time was 10 hours.


148

SEM images were performed with Omega 912, Zeiss AG apparatus. Light microscopy

images taken by Keyence VHX100 microscope under oblique illumination in transmission

mode.

3.14.4 Determination of Molecular Weight by Static

Light Scattering

Static light scattering measurements were carried out in tetrahydrofuran using an

ALV machine at 25 °C (ALV-7004 Multiple tau digital correlator equipped with CGS-3

Compact Goniometer system, 22 mW He-Ne laser, wavelength of 632.8 nm, and pair of

avalanche photodiodes operated in a pseudo-cross-correlation mode). The program ALV-

Stat was used for data evaluation. The Zimm plot resulted in apparent average molecular

weights of 3.428·106 and 3.464·106 g/mol for polystyrene samples prepared in the

microfluidic device under standard conditions with SDS concentrations of 300 and 500 mg

per 10 mL water.

3.14.5 Monomer Concentration Inside Latex Particles

The CM value is an optimistic guess for fast monomer consumption. Under these

conditions equilibrium thermodynamics are not suited to estimate the exact CM value as it

does not remain constant during the period of time for average chain growth. CM = 3.5 M

is likely too high, but constitutes the best approximation available today. Despite this

uncertainty, using the standard IUPAC numerical value for the propagation frequency and

a temperature of 30 °C, we were able to calculate an upper limit of the attainable molecular

weight (see primary article for details).


149

3.14.6 Seed Polymerization

Seed polymers were produced by styrene photo-initiated polymerization in

continuous flow device (10 g water, 2.25 g styrene, 0.01 g BAPO-AA or 0.01 g of MAPO,

25 °C, irradiation time 36.5 sec). For the second stage polymerization the BAOP-AA seed

latex was swollen with BMA after dialysis and irradiated (shaking, 2 h, rt). After

purification by repeated precipitation from a tetrahydrofuran solution in methanol the

MAPO seed polymer was then dissolved in BMA. This solution was subjected to

polymerization in the microfluidic device under standard conditions with an irradiation

time of 144 sec.

3.14.7 Synthesis of the Precursor Polymer

The polymerizations were carried out according to standard procedures in a glass

reactor using 18 g of styrene and 80 g or 8 g of Irgacure® 819. The sample was irradiated

for 72 h with a light source (Osram, L18 W, light color 840, lumilux, cold white). The

polymer was purified by repeated precipitation from a tetrahydrofuran solution in

methanol.

3.14.8 Assumption for Simulation Data for Different

Initial Initiator Concentrations

For the particular simulation of heterophase polymerization initiated by BAPO

photodecomposition, the reaction volume was defined as the volume of a monomer

droplet, containing initially only one single primary radical, either including or not, a

phosphine oxide group. The only difference between the kinetic model based on SRG and


150

the traditional model based on common radical polymerization kinetics is the consideration

of the decomposition of phosphine oxide groups inside the polymeric chain (reactions 14

a-c). In the traditional model, the kinetic rate coefficient for these reactions, kd2, is set to

zero.

Table 3.4 Kinetic scheme and reactions rate constants of radical polymerization based on SRG

used for the model simulations; the assignment of the particular chemical structures to the symbols

used in the kinetic equations is possible with Scheme 2, Y is the phosphine oxide acetic acid unit, Z

the trimethylbenzoyl unit, P and P’ are different polymer chains, and the open circle close to a

letter denotes a single radical function


151

3.15 References

1 (a) Elaissari, A; Fessi, H. Braz. J. Phys. 2009, 39, 146–149. (b) Seo, M.; Nie, Z.; Xu, S.;

Mok, M.; Lewis, P. C.; Graham, R.; Kumacheva, E. Langmuir 2005, 21, 11614–11622. 2 (a) Kitayama, Y.; Yorizane, M.; Kagawa, Y.; Minami, H.; Zetterlund, P. B.; Okubo, M.

Polymer 2009, 50, 3182–3187. (b) Konishi, N.; Fujibayashi, T.; Tanaka, T; Minami, H.;

Okubo, M. Polym. J. 2010, 42, 66–71. (c) Tanaka, T.; Komatsu, Y.; Fujibayashi, T.;

Minami, H.; Okubo, M. Langmuir 2010, 26, 3848–3853. (d) Tanaka, T.; Okayama, M.;

Kitayama, Y.; Kagawa, Y.; Okubo, M. Langmuir 2010, 26, 7843–7847. (e) Tanaka, T.;

Okayama, M.; Minami, H.; Okubo, M. Langmuir 2010, 26, 11732–11736. (f) Lu, Y.;

Wittemann, A.; Ballauff, M. Macromol. Rapid Commun. 2009, 30, 806–815. 3 Ugelstad, J.; Mork, P. C.; Aasen, J. O. J. Polym. Sci., A: Pol. Chem. 1967, 5, 2281–2288. 4 (a) Smith, W. V.; Ewart, R. H. J. Chem. Phys. 1948, 16, 592–599. (b) De Geest, B. G.;

Urbanski, J. P.; Thorsen, T.; Demeester, J.; De Smedt, Langmuir 2005, 21, 10275–10279.

(c) Xu, S.; Nie, Z;. Seo, M.; Lewis, P. C.; Kumacheva, E.; Stone, H. A.; Garstecki, P.;

Weibel, D. B.; Gitlin, I.; Whitesides, G. M. Angew. Chem. Int. Ed. 2005, 44, 724–728. (d)

Ugelstad, J.; Mork, P. C.; Aasen, J. O. J. Polym. Sci., A: Pol. Chem. 1967, 5, 2281–2288. 5 (a) Hubbell, J. A.; Thomas, S. N.; Swartz, M. A. Nature 2009, 462, 449–460. (b) Sharp,

F. A.; Lavelle, E. C. PNAS 2009, 106, 870–875. 6 (a) Esen, C.; Scweiger, G. J. Colloid Interface Sci 1995, 179, 276–280. (b) Whitesides, G.

M. Nature, 2006, 442, 368–373. (c) Ouzineb, K.; Graillat, C.; McKenna, T. J. Applied Pol.

Sci. 2004, 91, 2195–2207. (d) Paquet, D. A. Jr.; Ray, W. H. AIChE, 1994, 40, 73–87. (e)

Ganan-Calvo, A. M.; Gordillo, J. M. Phys. Rev. Lett. 2001, 87, 274501–274505. (f)

Garstecki, P.; Gitlin, I.; DiLuzio, W.; Whitesides, G. M.; Kumacheva, E.; Stone, Howard

A. Appl. Phys. Lett. 2004, 85, 2649–2651. (g) Thorsen, T.; Roberts, R. W.; Arnold, F. H.;

Quake, S. R. Phys. Rev. Lett. 2001, 86, 4163–4166. (h) Link, D. R.; Anna, S. L.; Weitz, D.

A.; Stone, H. A. Phys. Rev. Lett. 2004, 92, 054503. (i) Anna, S. L.; Bontoux, N.; Stone, H.

A. Appl. Phys. Lett. 2003, 82, 364–366. (l) Whitesides, G. M. Nature 2006, 442, 368–373.

(m) Feldon, M.; McCann, R. F.; Laundrie, R. W. India Rubber World 1953, 128, 51–53.


152

(n) Ouzineb, K.; Graillat, C.; McKenna, T. J. Applied Pol. Sci. 2004, 91, 2195–2207. (o)

Paquet, D. A. Jr.; Ray, W. H. AIChE 1994, 40, 73–87. 7 (a) Sounart, T. L.; Safier, P. A.; Voigt, J. A.; Hoyt, J.; Tallant, D. R.; Matzke, C. M.;

Michalske, A. Lab Chip 2007, 7, 908–915. (b) Ganan-Calvo, A. M.; Gordillo, J. M. Phys.

Rev. Lett. 2001, 87, 274501–274505. (c) Thorsen, T.; Roberts, R. W.; Arnold, F. H.;

Quake, S. R. Phys. Rev. Lett. 2001, 86, 4163–4166. (c) Link, D. R.; Anna, S. L.; Weitz, D.

A.; Stone, H. A. Phys. Rev. Lett. 2004, 92, 054503–1–054503–3. (d) Anna, S. L.; Bontoux,

N.; Stone, H. A. Appl. Phys. Lett. 2003, 82, 364–366. 8 (a) Okubo, Y.; Maki, T.; Nakanishi, F.; Hayashi, T.; Mae, K. Chem. Eng. Sci. 2010, 65,

386–391. (b) Tan, Y. C.; Fisher, J. S.; Lee, A. I.; Cristini, V.; Lee, A. P. Lab Chip 2004, 4:

292–298. (c) Xu, S.; Nie, Z.; Seo, M.; Lewis, P.; Kumacheva, E.; Stone, H. A.; Garstecki,

P.; Weibel, D. B.; Gitlin, I.; Whitesides, G. M. Angew. Chem. Int. Ed. 2005, 44, 3799–

3799. (d) Sheperd, R. F.; Conrad, J. C.; Rhodes, S. K.; Link, D. R.; Marquez, M.; Weitz,

D. A.; Lewis, J. A. Langmuir 2006, 22, 8618–8622. (e) Tan, Y.C.; Fisher, J. S.; Lee, A. I.;

Cristini, V.; Lee, A. P. Lab Chip 2004, 4, 292–298. 9 (a) Mozharov, S.; Nordon, A.; Littlejohn, D.; Wiles, C.; Watts, P.; Dallin, P.; Girkin, J.

M. J. Am. Chem. Soc. 2011, 133, 3601–3608. (b) Miyazaki, M.; Maeda, H. Trends

Biotechnol. 2006, 24, 463–470. (c) Urban, P. L.; Goodall, D. M.; Bruce, N. C. Biotechnol.

Adv. 2006, 24, 42–57. (d) Whitesides, G. M. Nature 2006, 442, 368–373. (e) Feldon, M.;

McCann, R. F.; Laundrie, R. W. India Rubber World 1953, 128, 51–53. (f) Ouzineb, K.;

Graillat, C.; McKenna, T. J. Applied Pol. Sci. 2004, 91, 2195–2207. (d) Paquet, D. A. Jr.;

Ray, W. H. AIChE 1994, 40, 73–87. 10 Dendukuri, D.; Tsoi, K.; Hatton, T. A.; Doyle, P. S. Langmuir 2005, 21, 2113–2116. 11 Shin, S.; Hong, J. S.; Lee, K-H; Lee, S-H. Langmuir 2009, 25, 12361–12366. 12 Jeong, W. J.; Kim, J. Y.; Choo, J.; Lee, E. K.; Han, C. S.; Beebe D. J.; Seong, G. H.;

Lee, S. H. Langmuir 2005, 21, 3738–3741. 13 (a) Griesser, M.; Neshchadin, D.; Dietliker, K.; Moszner, N.; Liska, R.; Gescheidt, G.

Angew. Chem. Int. Ed. 2009, 48, 9359–9361. (b) Jockusch, S.; Turro, N. J. J. Am. Chem.

Soc. 1998, 120, 11773–11777. 14 Sluggett, G. W.; Turro, C.; George, M. W.; Koptyug, I. V.; Turro, N. J. J. Am. Chem.

Soc. 1995, 117, 5184–5187.


153

15 Urban, D.; Takamura, K. Polymer Dispersions and Their Industrial Applications, 2002,

Wiley-VCH Verlag GmbH, Weinheim. 16 Allen, N. S. J. Photochem. Photobiol. A: Chemistry 1996, 100, 101–107. 17 (a) Rutsh, W.; Dietliker, K.; Leppard, D.; Köhler, M.; Misev, L.; Kolczak, U.; Rist, G.

Prog. Org. Coatings 1996, 27, 227–239. (b) Kolczak, U.; Rist, G.; Dietliker, K.; Wirrz, J.

J. Am. Chem. Soc. 1996, 118, 6477–6489. (c) Jockusch, S.; Koptyug, I. V.; McGarry, P. F.;

Sliggett, G. W.; Turro, N. J.; Watkins, D. M. J. Am. Chem. Soc. 1997, 119, 11495–11501. 18 Günersel, E. D.; Hepuzer, Y.; Yusuf, Y. Angew. Makromol. Chem. 1999, 264, 88–91. 19 Löb, P.; Pennemann, H.; Hessel, V.; Men, Y. Chem. Eng. Science 2006, 61, 2959–2967. 20 Funayama, H.; Sugarawa, T. Bull. Chem. Soc. Jpn. 1987, 70, 2245–2249. 21 Hernandez, H. F.; Tauer, K. Macromol. Symp. 2008, 271, 64–74. 22 www.imtek.de 23 (a) Nishimi, T.; Miller, C. A. Langmuir 2000, 16, 9233–9241. (b) Tauer, K.; Kozempel,

S.; Rother, G. J. Coll. Interf. Sci. 2007, 312, 432–438. 24 (a) Ikemura, K; Ichizawa, K.; Yoshida, M.; Ito, S.; Endo, T. Dent. Mat. J. 2008, 27,

765–774. (b) Shergill, R.; Haberler, M.; Vink, C. B.; Patten, H. V.; Woodward, J. R. Phys.

Chem. 2009, 11, 7248–7256. (c) Sumiyoshi, T.; Schnabel, W.; Henne, A.; Lechtken, P.

Polymer 1985, 26, 141–146. 25 Ouzineb, K.; Graillat, C.; McKenna, T. J. App. Pol. Sci. 2004, 91, 2195–2207. 26 Keskin, S.; Jockusch, S.; Turro, N. J.; Arsu, N. Macromol. 2008, 41, 4631–4634. 27 Tauer, K.; Öz, N. Macromol. 2004, 37, 5880–5888. 28 Grutzmacher, H.; Geier, J.; Stein, D.; Ott, T.; Schonberg, H.; Sommerlade, R. H.;

Boulmaaz, S.; Wolf, J. P.; Murer, P.; Ulrich, T. Chimia 2008, 62, 18–22. 29 (a) Tauer, K. Pol. Adv. Tech., 1995, 6, 435–440. (b) Buxton, G. V.; Greenstock, C. L.;

Helman, W. P.; Ross, A. B. J. Phys. Chem. Ref. Data, 1988, 17, 513–886. 30 Tauer, K.; Yildiz, U. Macromol. 2003, 36, 8638–8647. 31 Brandrup, J.; Immergut, E. H.; Grulke, E. A Polymer Handbook 1999, 4th Ed, Wiley,

New York. 32 (a) Moad, G.; Solomon, D. H. The Chemistry of Free Radical Polymerization 1995

Pergamon-press, Oxford. (b) Buback, M.; Egorov, M.; Gilbert, R. G.; Kaminsky, V.; Olaj,

O. F.; Russell, G. T.; Vana, P.; Zifferer, G. Macromol. Chem. Phys. 2002, 203, 2570–2582. 33 Tauer, K.; Muller, H.; Colloid Polymer Sci. 2003, 281, 52–65.


154

34 Gilbert, R. G. Emulsion Polymerization a Mechanistic Approach. 1995, Academic Press,

London. 35 Tauer, K.; Kaspar, H.; Antonietti, M. Coll. Pol. Sci. 2000, 278: 814–820 36 Parker, H. Y.; Westmoreland, D. G.; Chang, H. R. Macromol. 1996, 29, 5119–5127. 37 Kühn, I.; Tauer, K. Macromol. 1995, 28, 8122–8128. 38 Sumiyoshi, T.; Schnabel, W.; Henne, A.; Lechtken, P. Polymer 1985, 26, 141–146. 39 (a) Wyzgoski, F.; Meng, H.; Rinaldi, P. L.; Harwood, H. J. Polymer Preprints (Am.

Chem. Soc., Division of Polymer Chemistry) 1999, 40, 1028–1029. (b) Li, T.; Willis, T. J.;

Padias, A. B.; Hall, H. K., Macromol. 1991, 24, 2485–2487. 40 (a) Russell, K. E.; Tobolsky, A. V. J. Am. Chem. Soc. 1954, 76, 395–399. (b) Zimm, B.

H.; Bragg, J. K. J. Polym. Sci. 1952, 9, 476–478. 41 (a) Breitenbach, J. W.; Kastell, A. Monatshefte Chemie 1953, 84, 645–647. (b) Hall, H.

K.; Padias, A. B.; Pandya, A.; Tanaka, H. A Macromol. 1987, 20, 247–254. (c) Percec, V.;

Kim, H. J.; Barboiu, B., Macromol. 1997, 30, 6702–6705. 42 (a) Jasse, B.; Laupretre, F.; Monnerie, L. Macromol. Chem. Phys. 1977, 178, 1987–

1992. 43 Gillespie, D. T. J. Computational Phys. 1976, 22, 403–434. 44 (a) Destarac, M.; Pees, B.; Boutevin, B. Macromol. Chem. Phys. 2000, 201, 1189–1199.

(b) Jeon, H. J.; Youk, J. H. Macromol. 2010, 43, 2184–2189. 45 Tauer, K.; Muller, H. Coll. Pol. Sci. 2003, 281, 52–65.

4. Conclusions and Outlook

155



156

This thesis has reported on the beneficial properties of continuous flow

microreactors for chemical synthesis and mechanistic studies of reactions which are

summarized below.

1. The synthesis of CdSe or CdTe quantum dots (QDs), covered by a shell of ZnS and

further functionalized with carbohydrates and hydrophilic acid in a commercially available

microreactor (1 mL glass microreactor) was described.

The generation of the QD core was the fundamental step for the synthesis since the shape

and size of the seed influenced the photophysical properties of the CdSe or CdTe QDs. The

first step afforded CdSe or CdTe QD core, followed by screening of different conditions in

a continuous flow microreactor to achieve different particles size distributions. Under these

conditions, QDs of different absorption, emission spectra, and quantum yields were

obtained. After covering the core with a shell of ZnS, the QDs were successfully

functionalized with hydrolipoic acid or thio-linked carbohydrates. The carbohydrate

functionalized QDs were used for in vitro and in vivo biological studies.

Figure 4.1 Schematic representation of carbohydrate capped QDs project.

The significance of this method can be summarized as follows: 1) A synthesis in single-

phase with high control of nucleation and growth of the particles, at lower temperature

than the previous synthesis reported in the literature was achieved. 2) Clogging of the

continuous flow microreactor has been overcome by screening the best solvent,

concentration of precursors, temperature and residence time. 3) The three step synthesis in

continuous flow microreactor was successfully achieved requiring one purification step for

the final QDs. 4) The functionalized QDs were ready to use for biological studies.

Enormous improvements in the synthesis of QDs and their accessibility have

further enhanced their applications in electronics and life sciences. Despite overcoming the


157

problem of scaling up of QDs, maintaining consistent morphology and properties of the

particles, the metals used in the synthesis of QDs are highly toxic and difficult to handle.

Efforts are now focused on the use of less toxic metals but with similar photo-physical

properties.

Several studies have been reported in general, for the synthesis of particles using

continuous flow microreactor. It is remarkable to notice how the highly controlled

synthesis in continuous flow has a direct impact on the morphology of particle and on the

final material. This area of research is now emerging and is already taking great advantage

of the microreactor technology.

2. A continuous flow microreactor was designed for the synthesis of polymer particles

via phosphine oxide photo-initiated emulsion polymerization reactions. The polymer

particles were generated under short residence time and resulted in a very high molecular

weight of the polymer chains. Microreactor technology facilitated the screening of

different conditions and to propose a mechanism for phosphine oxide initiators in the

polymerization reaction. Upon irradiation, the rapid generation of radicals and the repeated

breaking of the polymer backbone where the phosphine oxide units were incorporated

produced a snowballing generation of radicals (SRG). The SRG process was then exploited

to generate block copolymer or to restart the polymerization via the use of the polymer

precursor without additional initiators (Figure 4.2).

Figure 4.2 Schematic representation of emulsion polymerization in continuous flow microreactor

project


158

Two further developments can be seen on the basis of the knowledge gained on the

reaction mechanism of phosphine oxide initiators: 1) Design new materials with novel

morphologies. 2) The microreactor technology could be further improved to design a

system which permits the use of compact light source with specific emission such as LEDs

(light-emitting diode).

To the best of our knowledge few examples of photochemical reactions are

reported and the system can be further improved to achieve better results in this area using

different sources of light or designing systems which can maximize the irradiation surface

of microreactors. Alternatively kinetic studies of reactions using continuous flow

microreactors are still rare examples in the literature and probably these fundamental

investigations will find more space in this field in the future. Generally microreactor is the

perfect tool to explore safe chemistry using limit conditions as high temperature and

pressure. The design of a modular system easy to assemble and to use according to the

required reaction with integrated analytical systems to characterize products and

intermediates of reaction would further expand the use of microreactors.

In conclusion, the two investigations reported in this thesis highlight the

importance of the microreactor in addressing chemical phenomena and could encourage

the development of other chemical reactions which previously could not be achieved.

159

Appendix

160

5. Glyco-dendronized Polylysine for Bacteria Detection

161


This appendix summarizes the work performed in collaboration with Dr. R. Kikkeri

(project design and evaluation of the data) and Dr. N. Azzouz (bacteria detection

experiments). The E. Coli bacteria strain was provided by Professor P. E. Orndorff

(College of Veterinary Medicine, Raleigh, NC, United States).

Parts of this chapter have been published in:

• “Detection of Bacteria Using Glyco-dendronized Polylysine prepared by Continuous

Flow Photo-Functionalization” Laurino, P.; Kikkeri, R.; Azzouz, N.; Seeberger, P. H.

Nano Lett. 2011, 11, 73–78.


162

5.1 Introduction

Dendrimers are molecules with repeating branching units, symmetrically developed

around a core with nanometer dimensions.1 The core is characterized by its functionality

which is the number of chemical groups through which it can be connected to the external

parts of the molecule. The origin of the branched structure starts from the core and when

the structure of the dendrimer is developed in only one direction the dendrimer is called a

dendron (Figure 5.1).2

Figure 5.1 Dendritic Polymer structure.

Dendrons can be used as lateral functional groups of a polymer main-chain and the

resulting structure consists of a high density of functional groups and these are called

dendronized polymer.3 The major drawback in the synthesis of dendrimers or dendrons is

the tedious synthetic steps involved. An alternative to dendrimer and dendrons are the

hyperbranched polymers which consist of imperfectly branched or irregular structures.4

Hyperbranched polymers present often an easier synthetic route compared to dendrimer

structures but lack the regularity typical of the dendrimers which is the base for their

physical, photophysical and supramolecular properties. A well defined dendrimer

structures is characterized by a precise size and shape. All these features enable

dendrimers, dendrons and dendronied polymer to be candidates for applications in sensors,


163

catalysis, molecular electronics and photonics, and in recent times are considered one of

the most important research topics nowdays in Material and Biomedical sciences.5

Dendronized polymers in particular are a new class of compounds.6 In these

systems the polymer is used as a polyfunctional and polydisperse backbone functionalized

by dendrons. The dendronized polymers are sterically hindered and lead to the formation

of a rigid structure with different polymers compared to those of the polymer or dendrons

precursors.7 An important feature of the dendronized polymer is the secondary interactions

between the dendrons which can be attractive and lead to the generation of a self-assembly

phenomena.8 Dendronized polymers can be prepared via three different strategies (Figure

5.2):

1. The graft-to approach, where a preformed polymer is functionalized by dendrons

(convergent route).

2. The macromonomer approach, where the monomer are functionalized by a preformed

dendrons and then polymerized.

3. The graft-from approach, where dendrons of the first generation are functionalized to

the polymer backbone and then successively elongated (divergent route).

Figure 5.2 Schematic representation of the three synthetic strategies for dendronized polymer.

The graft-to approach has been choosen for this study because of the

straightforward strategy for the formation of dendronized polymer. The dendronized

polymer are multivalent and flexible tools that if functionalized with biologically active

compounds (carbohydrates, peptides, nucleotides, etc.) can be used as biosensors.9

Consequently, we explored the photochemical [2 + 2] cycloaddition reaction,10 as an


164

alternative graft-to approach for the synthesis of a dendronized polymer for biological

applications (Figure 5.3).

Figure 5.3 Schematic representation of [2 + 2] cycloaddition between glycol-dendron and a

polymer chain.

When functionalized with carbohydrate the dendronized polymer served as a

powerful platform for the detection of bacteria by providing a multivalent system which

was needed to enhance the generally weak protein-carbohydrate interactions. The glyco-

dendronized polymer could further bend and adapt to the shape and morphology of the

bacterial surface and optimize the binding to several bacterial carbohydrate receptors hence

displaying a multivalent interaction (Figure 5.4).11

Figure 5.4 Schematic representation of the glyco-dendronized polymer binding to the bacterium

surface.

Contamination of food with Escherichia coli, a pathogenic bacterium, results in frequent

outbreaks of infections with grave consequences. Rapid, reliable and straightforward

detection of pathogenic bacteria is a key step in avoiding the spread of contamination.12

Developing a detection system that meets these criteria has proven challenging and the

glycodendronized polylysine has been demostrated to be a powerful candidate for the

detection of bacteria.


165

5.2 Synthesis of Glyco-dendrons

Firstly two sets of dendrons with three and nine carbohydrates respectively, were

synthesized bearing either mannose or galactose (Scheme 5.2).13

Scheme 5.2 Reagents and conditions: (a) Acrylonitrile/NaOH (40%), Conc HCl/EtOH, 51%; 5-

Hexynoic acid/DIC/HOBT/DCM, 72%; (b) NaOH/MeOH; pentafluorophenol /DIC/ HOBT/DCM,

56%; (c) 2-(t-butoxycarbonylamino)ethoxy-2,3,4,6-tetra-O-acetyl-α-D-mannoside or 2-(t-

butoxycarbonylamino)ethoxy-2,3,4,6-tetra-O-acetyl-α-D-galactoside/TFA/DCM (1:3) (53–61%)

was added to 3/TEA/DCM, 67–74%; (d) tripod-mannose/tripod-galactose /DCM/TEA, 67–71%;

(e) MeOH/ NaOMe, quant.; (f) MeOH/ NaOMe, quant.

Starting from commercial 2-amino-2-(hydroxymethyl)propane-1,3-diol 1, treatment with

acrylonitrile under basic conditions, followed by hydrolysis and esterification, prior to

coupling with 5-hexynoic acid provided structure 2. Ester hydrolysis under basic

conditions, followed by coupling with pentafluorophenol yielded the active ester 3.

Reaction of 3 with peracetylated amino linked mannose or galactose yielded first

generation dendrons 4 or 5, while reaction with the peracetylated tri-pod mannose or

galactose gave second generation dendrons 8 or 9. Removal of the acetate groups under

basic conditions yielded 6 or 7 and 10 or 11.


166

5.3 Photo-functionalization of Polylysine

The [2+2] cycloaddition has been selected for the functionalization of a polymer

chain. [2+2] Cycloaddition reactions are attractive since they can be carried out in water

using inexpensive starting materials, and circumvent the use of heavy metals or other

reagents that contaminate the polymer product, while avoiding the formation of side

products. To improve the efficiency and to render easy scalability of the reaction a

continuous flow device has been exploited for the functionalization of the polymer. As the

polymer backbone, polylysine has been chosen due to its biocompatibility and

biodegradability14 and two generations of glycodendrimer have been synthesized for the

functionalization.

The first and second generation dendrons were subsequently used for the

functionalization of the PLL polymer backbone 12 (Scheme 5.2). The dendronized

polymers 13-16 were created by photochemical functionalization using the [2 + 2]

cycloaddition reaction in a continuous flow microreactor. Reactions were carried out at

room temperature and in water with 40 min of irradiation time using a medium pressure

Hg lamp (450 W). The first generation dendrons afforded the glyco-dendronized

polylysine with higher yield (ca 68%) then the second generation (ca 50%). The major

disadvantage of the functionalization was the steric hindrance of the second generation

dendron on the polymer chain which did not allow the attachment of a second dendron

because of proximity effect resulting in lower functionalization. Moreover, the self

assembly of the second generation dendron can shield its own active group which serves

for the attachment to the polymer chain and consequentially further decrease the

conversion of the functionalization compared to the first generation dendron.


167

Scheme 5.2 Reagents and conditions: (a) 3-maleimidopropanoic acid N-hydroxysuccinimide

ester, pH 8.0, 24 h; (b) or (c) Medium pressure Hg Lamp, (Hanovia ®), 450W, alkyn dendrimer

mannose or galactose (6 or 7, 10 or 11), 40 min, rt.

The samples were immobilized on a gelatin coated mica surface and the size and

shape of the dendronized polymers characterized by atomic force microscopy (AFM)

analysis.15 The nonapod-mannose-functionalized polymer 15 adopted a globular structure,

with a diameter of approximately 4 Å (Figure 5.5). The globular shape likely results from

polymer and dendrimer integration.

Figure 5.5 AFM images of mannose nonapod PLL 15 on gelatin coated mica surfaces.


168

5.4 Bacteria Detection with Glyco-

dendronized Polylysine

To demonstrate the clustering capability of the dendronized polymers, bacterial

detection assays were performed.16 The dendronized polymers were incubated with either

the mannose-binding, invasive strain of E. coli, ORN178 (positive control), or with the

mutant E. coli strain, ORN208 that does not bind mannose (negative control). Binding,

and the resultant clustering was detected by confocal microscopy using DAPI (4΄,6-

diamidino-2-phenylindole), a fluorescent dye that stains DNA. Different numbers of E.

coli bacteria were suspended in buffer and incubated with polymer 15 and single mannose

containing polymer (17, see section 5.6.6). After centrifugation and washing, the cells

were re-suspended and stained with DAPI (4΄, 6-diamidino-2-phenylindole). Confocal

images showed no aggregation in case of single mannose polymer 17 (see section 5.6.6)

whereas Figure 5.6 shows that, incubation of the nona-mannose functionalized polymer 15

with strain ORN178 resulted in large clusters of bacteria (Figure 5.6), while strain

ORN208 did not aggregate with the polymer (Figure 5.6 c and d).


169

Figure 5.6 Confocal laser scanning microscope images for the incubation of bacteria E. coli strain

ORN178 and dendronized polymer 15 (a); and ConA-FITC (b). Strain ORN208 (negative control)

in the presence of 15 (c) and ConA-FITC (d).

We confirmed specific multivalent interactions between specific E. coli strains and

polymer 15 by adding ConA-FITC (Concavanalin A-fluorescein isothiocyanate

conjugate). ConA is a lectin well known for its strong interaction with α-mannose.17

Solutions containing different numbers of E. coli bacteria (105–107) were incubated with

polymer 15. The cells were then incubated with ConA-FITC for approximately 10 min,

centrifuged and washed. ConA-FITC staining successfully localized polymer 15 within a

cluster of E. coli strain ORN178 (Figure 5.6b). Neither clustering nor significant ConA-

FITC staining were observed in samples of E. coli strain ORN208 incubated with

compound 15 (Figure 5.6d). Therefore, we conclude that the mannose-binding E. coli was

aggregated by a matrix of the nona-mannose dendronized polymer 15, which was

localized by fluorescently labeled ConA. E. coli bacteria that do not bind to mannose were

not bound by the flexible polymer resulting in no visible aggregation.


170

5.5 Conclusions

To afford a glyco-dendronized polymer, a [2+2] cycloaddition reaction was carried

out in a continuous flow microreactor. The [2+2] cycloaddition occurred under mild

conditions upon irradiation in aqueous solution. The continuous flow microreactor

allowed improved yields enhancing the photo-efficiency of the reaction compared to batch

reactor. The reaction can be scaled up by maintaining the same conditions and extending

the reaction time. Unfortunately the functionalization with the second generation glyco-

dendrons occurred at lower yield than the first generation glyco-dendrons probably due to

the steric hindrance of the dendron.

The glyco-dendronized polymers were later used for bacteria detection. The

multivalence of the system allowed for stronger carbohydrate-protein interactions and the

flexibility of the polymer enhanced the sensitivity of detection. The glyco-dendronized

polymer can be a powerful tool for efficient detection of bacteria in an early stage of

contamination and furthermore serve as a platform to study pathogen-carbohydrate

interactions.


171


5.6.1 Continuous Flow Reactor Set Up

Figure 5.7 Schematic representation of the continuous flow microreactor

The flow reactor setup consists of a syringe pump injection system (1, Harvard PHD

2000), gas tight syringe (2, Hamilton), multiple loops of narrow FEP tubing (3, fluorinated

ethylene polymer, Tub FEP Nat 1/16 in x .030 in)18 wrapped tightly around a quartz

immersion well (4) cooled by a thermostat (9), a Pyrex filter (5), a Hanovia medium

pressure Hg lamp (6, 450 W), a power supply (8), and a collection flask (10).


172

5.6.2 General Information

All chemicals were reagent grade and used as supplied except where noted.

Dichloromethane (CH2Cl2) was purified by a Cycle-Tainer Solvent Delivery System.

Triethylamine was distilled over CaH2 prior to use. Analytical thin layer chromatography

(TLC) was performed on Merck silica gel 60 F254 plates (0.25 mm). Compounds were

visualized by UV irradiation or dipping the plate in CAN solution followed by heating.

Flash column chromatography was carried out using force flow of the indicated solvent on

Fluka Kieselgel 60 (230-400 mesh).

The water was taken from a Seral purification system (PURELAB Plus) with a

conductivity of 0.06 µS/cm. 1H and 13C NMR spectra were recorded on a Varian VXR-300 (400 MHz) spectrometer.

High-resolution mass spectra (HR-MALDI MS) and ESI-MS were performed by the

Mass Spectrometry-service at the MPI Berlin. ESI-MS were run on an Agilent 1100

Series LC/MSD instrument. IR spectra were recorded on a Perkin-Elmer 1600 FTIR

spectrometer. Optical rotation measurements were conducted using a Perkin-Elmer 241

polarimeter.

Atomic Force Microscopic (AFM) was performed with multimode instrument with

Nanoscope IIIa controller (Vecco Instruments) operating in the tapping mode. The

samples were placed on a Mica disk and tips (Nano world TIP) with a spring constant of

42 N/m, and a resonance frequency of 285 kHz was used for air measurements and spring

constant of 0.02 or 0.06 N/m, and resonance frequency of 10-20 kHz were used for liquid

measurements.

Confocal laser scanning microscope images were taken with Carl Zeiss confocal laser

scanning microscope, LSM 700.


173

5.6.3 Synthesis of Dendrimers

General Procedure A: Synthesis of Sugar-nonapods.

The Boc-protected sugar-amine or sugar-tripod (4.0 eq) was dissolved in 10 mL

dichloromethane/trifluoroacetic acid (3:1) and stirred at room temperature for 1 h. The

solvent was evaporated under reduced pressure and the resulting oil was dissolved in

anhydrous dichloromethane (20 mL). To this mixture was added tert-butoxycarbonyl-3-

{N-{tris[3-[pentafluoro-phenyl-carboxyl-ethoxy) methyl]}methyl amine}-3-β-alanine or

N-{tris[3-[pentafluoro phenyl carboxyl-ethoxy)methyl]}methyl amide}-5-hexynylamide

(1.0 eq), adjusted to pH 8 with triethylamine (TEA) and the mixture was stirred at room

temperature for 12 h. The solvent was evaporated in vacuo and purified by flash silica

column chromatography.

General Procedure B: Synthesis of sugar dendrons. Tripod or nonapod (1.0 eq) and

sodium methoxide (10 eq) were dissolved in methanol (10 mL) and stirred at room

temperature for 2 h. The solvent was then evaporated in vacuo, the residue was

redissolved in water and dialyzed against water using 500 molecular weight cut-off resin.

After two days of dialysis, the sample was lyophilized.

N-{Tris[3-[ethylcarboxyl-ethoxy)methyl]}methylamide}-5-hexynylamide (2).

To a solution of N-{tris[(3-[ethylcarboxyl-ethoxy)methyl]}methylamine (2 g, 4.7 mmol)

and 5-hexynoic acid (0.53 g, 4.7 mmol) in dichloromethane (10 mL) at 0 °C, were added

diisopropylcarbodiimide (0.87 mL, 5.6 mmol) and 1-hydroxybenzotriazole (0.07 g, 0.56

mmol) The reaction mixture was stirred at room temperature for 12 h and concentrated in

vacuo. The crude residue was purified by flash silica column chromatography to yield N-

{tris[3-[ethylcarboxyl-ethoxy)methyl]}methylamide}-5-hexynylamide (1.75 g, 72%). Rf

= 0.5 (CH2Cl2/MeOH, 98:4); 1H NMR (400 MHz, CDCl3): δ 5.86 (br.s, 1H), 4.17 (q, J =

8.0 Hz, 6H), 3.94 (br. s, 11H), 2.88 (t, J = 4.0 Hz, 6H), 2.24-2.16 (m, 4H), 1.88 (t, J = 4.0

Hz, 1H), 1.78-1.74 (m, 2H), 13C NMR (100 MHz, CDCl3): δ 172.4, 171.6, 69.5, 66.5,

60.3, 45.2, 37.2, 35.4, 24.3, 14.3, FTIR(CHCl3): 3375, 2988, 2873, 1744, 1726, 1644,

1525 cm-1. HRMS (MALDI-ToF) (m/z) calcd. for C24H41NO10Na 538.2628, found:

538.2627.


174

N-{Tris[3-[pentafluoro phenyl carboxyl-ethoxy)methyl]}methyl amide}-5-

hexynylamide (3).

N-{Tris[3-[ethylcarboxyl-ethoxy)methyl]} methylamine}-5-azido pentamide (1.5 g, 2.9

mmol) was dissolved in ethanol (10 mL) and sodium hydroxide solution (aqueous, 1 N, 2

mL) was added and the mixture was stirred at room temperature for 1 h, concentrated in

vacuo, adjusted to pH 5 with hydrochloric acid (aqueous 1 N) and extracted with ethyl

acetate. The organic layer was dried with sodium sulfate and concentrated to dryness

under reduced pressure. The residue was dissolved in dichloromethane (10 mL) and

2,3,4,5,6-pentafluorophenol (2.1 g, 11.6 mmol) was added. After cooling to 0 °C,

diisopropyl carbadiazine (2.2 mL, 13.9 mmol) was added and the reaction mixture was

stirred for 12 h at room temperature. The reaction mixture was concentrated in vacuo and

purified by silica column flash chromatography to afford tert-butoxycarbonyl-3-{N-

{tris[3-[pentafluorophenylcarboxyl-ethoxy)methyl]}methylamide}-5-azido pentamide

(1.56 g, 56%). Rf = 0.6 (CH2Cl2/EtOAc, 88:12); 1H NMR (400 MHz, CDCl3): δ 3.83 (br.

s, 12H), 2.84 (t, J = 6.0 Hz, 6H), 2.22-2.17 (m, 4H), 1.71-1.62 (m, 3H), 13C NMR (100

MHz, CDCl3): δ 172.2, 171.1, 69.3, 66.5, 60.3, 45.2, 37.1, 35.4, 24.3, FTIR(CHCl3) :

3688, 3385, 1749, 1658, 1522, 1359 cm-1. HRMS (MALDI-ToF) (m/z) calcd. for

C37H28F15NO10Na 952.1215, found: 952.1218.

Tris[3-[2-ethoxy-2,3,4,6-tetra-O-acetyl-α-D-mannopyranoside-

ethoxy]methyl]methylamide }-5-hexynylamide (4).

General procedure A using 2-(tert-butoxycarbonylamino)ethoxy-2,3,4,6-tetra-O-acetyl-α-

D-mannopyranoside (0.97 g, 1.96 mmol), tris[3-[pentafluoro phenyl carboxyl-

ethoxy)methyl]}methyl amide}-5-hexynylamide 2 (0.5 g, 0.53 mmol) and purified by

flash chromatography to yield tris[3-[2-ethoxy-2,3,4,6-tetra-O-acetyl-α-D-

mannopyranoside-ethoxy]methyl]methylamide }-5-hexynylamide (0.39 g, 47%). Rf =

0.45 (CH2Cl2/MeOH, 93:7); [α]Dr.t = +23.4 (c = 1.0, CHCl3); 1H NMR (400 MHz,

CDCl3): δ 6.79 (br.s, 2H), 6.33 (br.s, 1H), 5.27-5.20 (m, 9H), 4.80 (d, J = 9.0 Hz, 3H),

4.25 (dd, J = 9.0, 5.1 Hz, 3H), 4.10 (dd, J = 2.1, 9.9 Hz, 3H), 3.90 (br.s, 3H), 3.76 (dd, J

= 4.5, 5.4 Hz, 3H), 3.68 (dd, J = 5.4, 6.0 Hz, 6H), 3.64 (m, 6H), 3.54-3.52 (m, 6H), 3.37

(br.s, 8H), 2.72 (t, J = 5.4 Hz, 6H), 2.39 (t, J = 9.0 Hz, 4H), 2.28 (t, J = 5.4 Hz, 3H),

2.12 (t, J = 5.4 Hz, 2H), 2.07 (s, 9H), 2.02 (s, 9H), 1.96 (s, 9H), 1.71-1.62 (m, 2H), 13C

NMR (100 MHz, CDCl3): δ 171.4, 170.5, 170.0, 169.5, 155.8, 97.6, 69.3, 69.2, 68.6,


175

67.3, 65.1, 62.2, 60.4, 53.8, 39.0, 35.9, 34.5, 23.5, 20.3; FTIR(CHCl3): 3376, 2918, 1751,

1663, 1515, 1457, 1250 cm-1; HRMS (MALDI-ToF) (m/z) calcd. for C67H98N9O37Na

1573.5808; found: 1573.5810.

Tris-[3-4-ethoxy-α-D-mannosepyranosyl-ethoxy}methyl]-5-hexynylamide (6).

General procedure B with tris[3-[2-ethoxy-2,3,4,6-tetra-O-acetyl-α-D-mannopyranoside-

ethoxy]methyl]methylamide }-5-hexynyl amide 3 (100 mg, 97.5 µmol) sodium

methoxide (10 mg, 0.18 mmol) yielded 45 mg, (65%) of cis-

ruthenium(II)bis(bipyridine){1,1’-(2,2’-bipyridine-4,4’-diyl)bis-3-β-beta-propane-{tris-

[3-4-ethoxy-β-D-galactopyranosyl-ethoxy}methyl] methylamide. [α]Dr.t = +1.7 (c = 1.0,

H2O); 1H NMR (400 MHz, D2O/MeOD): δ 4.72 (s, 3H), 3.96-3.94 (m, 4H), 3.90 (d, J =

2.7 Hz, 12H), 3.75-3.65 (m, 34H), 3.61 (br.s, 24H), 3.49-3.46 (m, 12H), 3.73-3.21 (m,

44H), 2.37 (t, J = 6.0 Hz, 6H), 2.35 (t, J = 6.0 Hz, 3H), 2.12 (t, J = 6.0 Hz, 2H), 1.71-

1.62 (m, 2H), 13C NMR (100 MHz, CD3OD): δ 173.2, 172.6, 101.3, 74.8, 71.8, 71.5,

68.5, 67.6, 66.6, 62.4, 61.4, 40.0, 37.3, 34.9,27.3, 23.1, 21.2, 18.5. MALDI-ToF (m/z):

[M-1]+ Calcd for C43H174N4O2 1046.4642; Found: 1046.4645.

Tris[3-[2-ethoxy-2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside-

ethoxy]methyl]methylamide}-5-hexynylamide (5).

General procedure A using 2-(tert-butoxycarbonylamino)ethoxy-2,3,4,6-tetra-O-acetyl-β-

D-galactopyranoside (0.42 g, 1.06 mmol), tert-butoxycarbonyl-3-{N-{tris[3-

[pentafluoro-phenyl-carboxyl-ethoxy)methyl]}methyl amine}-5-hexynylamide 2 (0.25 g,

0.26 mmol) and purified by flash chromatography to yield tert-butoxycarbonyl-3-{tris[3-

[2-ethoxy-2,3,4,6-tetra-O-acetyl-α-D-mannopyranoside-ethoxy]methyl]methylamide}-3-

β-alanine (0.27 g, 57%). Rf = 0.45 (CH2Cl2/MeOH, 93:7); [α]Dr.t = +17.4 (c = 1.0,

CHCl3); 1H NMR (400 MHz, CDCl3): δ 6.47 (br.s, 2H), 5.37 (d, J = 4.0 Hz, 4H), 5.14 (t,

J = 8.0 Hz, 3H), 4.99 (dd, J = 4.0 Hz, 3H), 4.49 (d, J = 8.0 Hz, 3H), 4.12-4.10 (m, 6H),

3.92 (t, J = 6.0 Hz, 3H), 3.84-3.83 (m, 3H), 3.70 (t, J = 6.0 Hz, 8H), 3.65 (s, 9H), 3.42 (t,

J = 6.0 Hz, 6H), 3.33 (q, J = 6.0 Hz, 2H), 2.39 (t, J = 9.0 Hz, 4H), 2.28 (t, J = 5.4 Hz,

3H), 2.12 (t, J = 6.0 Hz, 2H), 2.07 (s, 9H), 2.02 (s, 9H), 1.96 (s, 9H), 1.71-1.62 (m, 2H); 13C NMR (100 MHz, CDCl3): δ 171.1, 170.0, 169.8, 169.7, 155.7, 101.1, 70.6, 69.1, 68.7,

67.2, 67.0, 66.9, 61.2, 59.7, 45.65, 41.67, 39.1, 37.0, 36.4, 28.3, 20.6; FTIR(CHCl3):


176

3376, 2918, 1751, 1663, 1515, 1457, 1250 cm-1; HRMS (MALDI-ToF) (m/z) calcd. for

C67H98N9O37Na 1573.5808; found: 1573.5810.

Tris[3-4-ethoxy- β-D-galactopyranoside -ethoxy}methyl]-5-hexynylamide (7).

General procedure D with cis-ruthenium(II)bis(bipyridine){1,1’-(2,2’-bipyridine-4,4’-

diyl)bis-3-β-propane-{tris-[3-4-ethoxy-2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside-

ethoxy}methyl]methylamide 5 (100 mg, 97.7 µmol) sodium methoxide (10 mg, 2.2

µmol) yielded 51 mg, (76%) of tris[3-4-ethoxy- β-D-galactopyranoside -ethoxy}methyl]-

5-hexynylamide . [α]Dr.t = +1.8 (c = 1.0, H2O); 1H NMR (400 MHz, D2O/MeOD): δ 4.36

(d, J = 8.0 Hz, 3H), 3.96-3.94 (m, 4H), 3.90 (d, J = 4.0 Hz, 12H), 3.75-3.65 (m, 17H),

3.61 (br.s, 12H), 3.49-3.46 (m, 6H), 3.43-3.41 (m, 2H), 2.37 (t, J = 6.0 Hz, 6H), 2.35 (t, J

= 6.0 Hz, 3H), 2.12 (t, J = 6.0 Hz, 2H), 1.71-1.62 (m, 2H), 13C NMR (100 MHz,

CD3OD): δ 174.9, 173.2, 104.3, 77.8, 75.3, 73.5, 71.8, 71.8, 68.5, 67.6, 66.6, 62.4, 61.4,

40.0, 37.3, 30.1, 26.4. MALDI-ToF (m/z): [M-1]+ Calcd for C43H174N4O2 1046.4642;

Found: 1046.4645.

3-{Tris[3-carboxylethoxy]methyl]3’-{tris-[2-ethoxy-2,3,4,6-tetra-O-acetyl-α-D-manno

pyranoside -ethoxy]methyl] methylamide}-5-hexynylamide (8).

General procedure A using tert-butoxycarbonyl-3-{tris[3-[2-ethoxy-2,3,4,6-tetra-O-acetyl-

α-D-mannopyranoside-ethoxy]methylamide}-3-β-alanine (0.34 g, 0.35 mmol), {N-{tris[3-

[pentafluoro phenyl carboxyl-ethoxy)methyl]}methyl amide}-5-hexynlamide (50 mg,

0.053 mmol) and purified by flash chromatography to yield 3-{tris[3-

carboxylethoxy]methyl]3’-{tris-[2-ethoxy-2,3,4,6-tetra-O-acetyl-α-D-mannopyranoside-

ethoxy] methyl] methylamide}-5-azido pentamide (0.22g, 57 %). [α]Dr.t = +19.4 (c =1.0,

CHCl3); 1H NMR (400 MHz, CDCl3): δ 6.84 (br.s, 3H), 6.13 (br.s, 1H), 4.87-5.8 (m, 27H),

4.25 (s, 9H), 3.81 (dd, J = 6.0, 4.0 Hz, 9H), 3.77 (d, J = 6.0 Hz, 9H), 3.55 (br.s, 9H), 3.31-

3.01 (m, 57H), 3.06 (m, J = 4.0 Hz, 34H), 2.47 (br, 32H), 2.32 (br.s, 2H), 2.21 (br.s, 2H),

2.13 (s, 27H), 2.06 (s, 27H), 2.04 (s, 27H), 1.96 (s, 27H), 1.38-1.32 (m, 2H); 13C-NMR

(100 MHz, CDCl3): δ 173.92, 172.2, 171.4, 102.7, 70.7, 69.6, 67.3, 63.4, 61.8, 54.5, 45.2,

39.7, 35.6, 20.6, 9.6. FTIR(CHCl3): 3332, 2734, 1745,1365,cm-1;HRMS-MALDI (m/z):

[M+ Na]+ Calcd for C211H314N16O118 4989.9061; Found : 4989.9071.


177

Tris-[3-carboxyl-ethoxy]methyl]3’-{tris[2’-ethoxy-β-D-mannosepyranosyl-

ethoxy]methyl]-5-hexynylamide (10).

General procedure D with 3-{tris[3-carboxylethoxy]methyl]3’-{tris-[2-ethoxy-2,3,4,6-

tetra-O-acetyl-α-D-manno pyranoside -ethoxy]methyl] methylamide}-5-hexynylamide 7

(200 mg, 40.0 µmol) and sodium methoxide (22 mg, 0.4 mmol) gave 108 mg (78%) of

tris-[3-carboxyl-ethoxy]methyl]3’-{tris[2’-ethoxy-β-D-mannosepyranosyl-ethoxy]methyl]-

5-hexynylamide. [α]Dr.t = +22.3 (c =1.0, H2O); 1H NMR (400MHz, MeOD): δ 4.79 (s, 9H),

3.78-3.6 (m, 98H), 3.54 (m, 24H), 3.41 (m, 24H), 2.46 (t, J = 4.0 Hz, 32H), 2.25 (t, J = 6.0

Hz, 3H), 2.16 (t, J = 6.0 Hz, 2H), 1.71-1.62 (m, 2H); 13C NMR (125MHz, CD3OD): δ

173.4, 104.3, 73.8, 71.5, 70.8, 68.8, 67.5, 65.6, 62.6, 60.4, 37.4, 35.1, 25.3, 16.4. MALDI-

HRMS (m/z): [M+1]+ Calcd for C139H243N16O82 3448.5258; Found: 3448.5251.

3-{Tris[3-carboxylethoxy]methyl]3’-{tris-[2-ethoxy-2,3,4,6-tetra-O-acetyl-β-D-

galactopyranoside-ethoxy]methyl] methylamide}-5-hexynylamide (9).

General procedure A using tert-butoxycarbonyl-3-{tris[3-[2-ethoxy-2,3,4,6-tetra-O-

acetyl- β-D-galactopyranoside-ethoxy]methylamide}-3-β-alanine (0.15 g, 0.09 mmol),

{N-{tris[3-[pentafluoro phenyl carboxyl-ethoxy)methyl]}methyl amide}-5-hexynlamide

(21mg, 0.021 mmol) and purified by flash chromatography to yield 3-{tris[3-

carboxylethoxy]methyl]3’-{tris-[2-ethoxy-2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside-

ethoxy] methyl] methylamide}-5-azido pentamide (95 g, 54%). [α]Dr.t = +2.4 (c =1.0,

CHCl3); 1H NMR (400 MHz, CDCl3): δ 6.64 (br.s, 3H), 5.37 (d, J = 3.3 Hz, 9H), 5.12 (t,

J = 8.0 Hz, 9H), 4.49 (d, J = 4.0 Hz, 9H), 4.31-4.15 (m, 9H), 4.14-4.08 (m, 18H), 3.95-

3.92 (m, 9H), 3.84-3.78 (m, 9H), 3.66-3.56 (m, 64H), 3.41-3.36 (m, 34H), 2.41 (t, J = 5.4

Hz, 32H), 2.32 (br.s, 2H), 2.21 (br.s, 2H), 2.13 (s, 27H), 2.06 (s, 27H), 2.04 (s, 27H), 1.96

(s, 27H), 1.38-1.32 (m, 2H), 13C-NMR (100 MHz, CDCl3): δ 173.92, 172.2, 171.4, 102.7,

70.7, 69.6, 67.3, 63.4, 61.8, 40.1, 35.6, 20.6, FTIR(CHCl3): 3332, 2734, 1745,1365,cm-

1;HRMS-MALDI (m/z): [M+ Na]+ Calcd for C211H314N16O118 4989.9061; Found :

4989.9071.


178

Tris-[3-carboxyl-ethoxy]methyl]3’-{tris[2’-ethoxy-β-D-galactopyranoside-ethoxy]methyl]-5-hexynylamide (11).

General procedure D with 3-{tris[3-carboxylethoxy]methyl]3’-{tris-[2-ethoxy-2,3,4,6-

tetra-O-acetyl-β-D-galactopyranoside-ethoxy]methyl] methylamide}-5-hexynylamide 7

(90 mg, 18.0 µmol) and sodium methoxide (10 mg, 0.22 mmol) gave 41 mg (76%) of

tris-[3-carboxyl-ethoxy]methyl]3’-{tris[2’-ethoxy-β-D-galactopyranoside-

ethoxy]methyl]-5-hexynylamide. [α]Dr.t = + 1.3 (c =1.0, H2O); 1H NMR (400MHz,

MeOD): δ 4.77 (s, 9H), 4.39 (d, J = 7.8 Hz, 9H), 4.15 (q, J =4.0 Hz, 8H), 3.91-3.85 (m,

18H), 3.76-3.60 (m, 93H), 3.47-3.45 (m, 24H), 3.41-3.35 (m, 26H), 2.47-2.42 (m, 32H);

2.25 (t, J = 6.0 Hz, 3H), 2.16 (t, J = 6.0 Hz, 2H), 1.71-1.62 (m, 2H), 13C NMR (125

MHz, CD3OD): δ 173.4, 104.3, 73.8, 71.5, 70.8, 68.8, 67.5, 65.6, 62.6, 60.4, 37.4, 35.1,

25.3, 16.4. MALDI-HRMS (m/z): [M+1]+ Calcd for C139H243N16O82 3448.5258; Found:

3448.5251.

5.6.4 Functionalization of Polylysine

PLL-maleimido (12).

Poly-L-Lysine hydrobromide (10 mg, 0.5 pmol) was added to 3- maleimidopropionic acid-

N-hydrosuccinimide ester (133 mg, 0.5 mmol) in 6 mL of nano pure water. The solution

was basified with 1 N KOH solution and shaken at room temperature. The aqueous layer

was washed with dichloromethane (3 x 7 mL) to remove unreacted maleimido derivative

gave 19 mg of pure yellow oil (98%). [α]D 25= -1.2 (c=1.5, MeOH/H2O 1:1); 1H NMR

(300 MHz, MeOD) δ 2.2-2.4 (m, 10 H); 3.5 (t, J = 7 Hz, 2H); 5.8 (d, J = 13 Hz, 1H); 6.3

(d, J = 13 Hz, 1H); 13C NMR (300 MHz, MeOD) δ26.2. 37.8, 38.5, 125.5, 138.4, 167.5,

174.6, 178.6, 179.9.

General Procedure C: Synthesis of PLL-dendrone and nonapod. 3-

Maleimidopropionic acid-N-poly-L-Lysine (0.2 µmol) and tripod or nonapod (0.4 mmol or

0.2 mmol)) were dissolved in water (3 mL) and flushed in a continuous flow reactor for 40


179

min under irradiation (medium pressure Hg Lamp, Hanovia, 450W). The crude product

was further purified with sephadex G-10 and finally the sample was lyophilized.

PLL-tripodmannose (13).

General procedure C with tris-[3-4-ethoxy-α-D-mannosepyranosyl-ethoxy}methyl]-5-

hexynylamide 6 (30 mg, 0.4 mmol) and maleimido-PLL (2 mg, 0.2 µmol) to yield 20 mg

of PLL-tripodmannose (75%). 1H NMR (300 MHz, D2O) δ 0.97 (d, J = 6.6 Hz, 1H); 1.69 –

1.59 (m, 3H); 1.78 (s, 1H); 1.87 (s, 1H); 2.11 (m, 3H); 2.21 (m, 3H); 2.39 (m, 6H); 3.24-

3.29 (m, 2H); 3.31 – 3.28 (m, 2H); 3.33-3.36 (m, 2H); 3.37-3.39 (m, 1H); 3.48 – 3.45 (m,

2H); 3.51 – 3.47 (m, 6H); 3.52 (s, 2H); 3.52 (s, 2H); 3.64 – 3.59 (m, 10H); 3.68 – 3.64 (m,

7H); 3.73 (d, J = 1.8 Hz, 2H); 3.76 (d, J = 2.0 Hz, 1H); 3.81 (m, 3H); 4.74 (m, 3H); 13C

NMR (300 MHz, D2O) δ 175.90, 174.19, 99.63, 72.79, 70.48, 69.97, 68.42, 67.43, 66.65,

65.76, 60.85, 60.13, 38.95, 36.08, 35.19, 24.06, 21.80, 16.97.

PLL- tripodgalactose (14).

General procedure C with Tris[3-4-ethoxy- β-D-galactopyranoside -ethoxy}methyl]-5-

hexynylamide 7 (30 mg, 0.4 mmol) and PLL-maleimido (2 mg, 0.2 µmol) to yield 14 mg

of PLL-tripodgalactose (60%). 1H NMR (300 MHz, D2O) δ 0.93 (d, J = 6.4 Hz, 1H); 1.01

(t, J = 7.1 Hz, 1H); 1.11 (t, J = 7.2 Hz, 1H); 1.66 – 1.55 (m, 3H); 1.74 (s, 4H); 2.08 (t, J =

7.0 Hz, 3H); 2.18 (t, J = 7.4 Hz, 3H); 2.31-2.20 (m, 2H); 2.51 – 2.47 (m, 1H); 3.33 – 3.24

(m, 6H); 3.37-3.39 (m, 2H); 3.49 – 3.46 (m, 1H); 3.55 – 3.49 (m, 9H); 3.65 – 3.55 (m,

16H); 3.76 (d, J = 3.4 Hz, 3H); 3.85 – 3.78 (m, 4H); 3.91 (m, 1H); 4.24 (m, 3H); 4.84 (s,

1H); 13C NMR (300 MHz, D2O) δ 175.89, 174.22, 107.23, 102.94, 82.84, 80.92, 76.58,

75.10, 72.64, 70.70, 68.54, 68.32, 67.41, 66.17, 62.67, 60.91, 60.13, 39.33, 39.18, 36.08,

35.18, 24.05, 16.96.

PLL- nonapodmannose (15).

General procedure C with tris-[3-carboxyl-ethoxy]methyl]3’-{tris[2’-ethoxy-β-D-

mannosepyranosyl ethoxy]methyl]-5-hexynylamide 10 (30 mg, 0.2 mmol) and PLL-

maleimido (2 mg, 0.2 µmol) to yield 14 mg of PLL-nonapodmannose (50%). 1H NMR

(300 MHz, D2O) δ 2.12 – 2.00 (m, 1H); 2.20 (s, 1H); 2.53 (t, J = 6.6 Hz, 1H); 2.59-2.64

(m, H); 2.69 (s, 1H); 2.86 – 2.72 (m, 6H); 2.95-3.00 (m, H); 3.74 – 3.65 (m, 3H); 3.83 –

3.73 (m, 3H); 3.90-3.97 (m, H); 3.99-4.07 (m, H); 4.26 – 4.13 (m, 5H); 5.20 – 5.13 (m,


180

2H); 13C NMR (300 MHz, D2O) δ 174.06, 173.79, 173.39, 129.11, 125.26, 99.59, 72.77,

70.45, 69.95, 68.39, 67.44, 66.62, 65.72, 60.82, 60.17, 60.06, 38.90, 36.02, 35.79, 35.60,

17.01.

PLL-nonapodgalactose (16).

General procedure C with tris-[3-carboxyl-ethoxy]methyl]3’-{tris[2’-ethoxy-β-D-

galactopyranoside-ethoxy]methyl]-5-hexynylamide 11 (30 mg, 0.4 mmol) and PLL-

maleimido (2 mg, 0.2 µmol) to yield 18 mg of PLL-nonapodgalactose (50%). 1H NMR

(300 MHz, D2O) δ 1.70 (dt, J = 14.6, 7.3 Hz, 1H); 1.84 (s, 1H); 1.91 (s, 1H); 2.17 (t, J =

7.0 Hz, 1H); 2.29 – 2.23 (m, 1H); 2.49 – 2.32 (m, 8H); 2.64 – 2.57 (m, 1H); 2.87 – 2.80

(m, 1H); 3.28 (t, J = 5.6 Hz, 1H); 3.33 - 3.44 (m, H); 3.78 – 3.53 (m, 25H); 3.86 (d, J = 3.3

Hz, 2H); 3.96 – 3.87 (m, 3H); 3.96 – 3.87 (m, 3H); 4.34 (d, J = 7.8 Hz, 1H); 4.94 (s, H); 13C NMR (300 MHz, D2O) δ 174.10, 173.80, 173.39, 107.21, 104.98, 102.92, 82.82, 80.92,

76.57, 75.08, 72.62, 70.69, 68.52, 68.29, 67.43, 66.14, 62.64, 60.90, 60.19, 41.43, 39.31,

39.16, 36.04, 35.78, 35.57, 17.01.

PLL-mono-mannose (17).

3- Maleimidopropionic acid-N-poly-L-Lysine (10 mg, 0.2 µmol) and allyl-galactose (88

mg, 0.4 mmol) were dissolved in H2O and flushed in a continuous flow reactor for 20

min under irradiation (medium pressure Hg Lamp, Hanovia, 450W). The crude product

was further purified with sephadex G-10 to yield 10 mg of PLL-mannose (90%). 1H

NMR (300 MHz, D2O) δ 1.14 (t, J = 7.1 Hz, 1H); 1.24 (t, J = 7.5 Hz, 2H); 1.97-1.78 (m,

10H); 2.40-2.36 (m, 1H); 2.65-2.61 (m, 8H); 2.68-2.71 (m, 8H); 2.86 (t, J = 6 Hz, 5H);

3.19-3.13 (m, 1H); 3.33-3.37 (m, 2H); 3.52-3.63 (m, 23H); 3.65-3.78 (m, 29H); 3.79-3.95

(m, 28H); 4.63 (s, 1H); 4.83 (s, 1H); 5.29 (dd, J = 25.0, 13.9 Hz, 2H); 5.84-6.02 (m, 1H); 13C NMR (300 MHz, MeOD) δ 44.80, 45.50, 50.23, 56.83, 71.69, 76.06, 77.09, 83.62,

87.01, 116.44, 116.48.


181

5.6.5 Cell Growth and Incubation with Polymer

The strains used in this study were a kind gift donated by Prof. Orndorff (College of

Veterinary Medicine, Raleigh, NC United States), ORN178 for the mannose binding

strain and ORN208 for the mutant strain that does not bind mannose. Cells were grown in

LB media.19 The inoculated culture was incubated overnight at 37°C by shaking on an

incubator shaker at 250 rpm. Intermittently, aliquots of bacteria were removed from the

batch culture to monitor growth until they reached OD600 of 1.0. Aliquot of 2 ml cultures

were then centrifuged for 10 min at 1,600 x g. The resulting pellets were then washed

twice with PBS buffer and resuspended in 100 µL PBS. A 10 µg of the polymer was

added in each incubation mixture or omitted from the control incubations. Each

suspension was then incubated for 30 min at room temperature with gentle shaking and

centrifuged to pellet the cells. To remove excess of polymer pellets were softly washed

twice with PBS and centrifuged. This procedure was repeated twice for each culture.

5.6.6 Confocal Fluorescence Microscopy

The washed polymer-treated and non-treated bacteria were fixed with 100 µL with 4%

(w/v) paraformaldehyde for 15 min. Bacteria were then centrifuged and the pellet

resuspended in 100 µL of PBS containing the organic dye 4’,6-diamidino-2-phenylindole

dihydrochloride, (DAPI) at dilution of 1:1000 and incubated for 15 min. DAPI forms

fluorescent complexes with double-stranded DNA. Alternatively to image the polymers,

following DAPI treatment bacteria were centrifuged and the pellet incubated for

additional 15 min in the presence of 100 µL PBS containing fluorescein-labelled

Concanavalin A (ConA-FITC) at a dilution of 1:1000. Bacteria were then centrifuged,

washed twice as described above prior to analysis. Bacteria were imaged using a Carl

Zeiss confocal laser scanning microscope (LSM 700) using a 63 x oil immersion or a 40

x objectives.


182

The single mannose-PLL was incubated with bacteria ORN178 strain and treated as

described above for the dendronized polymer. Bacteria were then imaged with confocal

fluorescence microscope, Figure 5.8 shows no aggregation.

Diluted solution of bacteria ORN178 expressing mannose binding protein were treated as

described for the previous experiments to determine the detection limit of aggregation of

bacteria (Figure 5.9). Results show that aggregations are detectable to a limit of 105 of

bacteria.

Figure 5.8 Confocal fluorescence microscopy image after incubation of single mannose-PLL (17)

with E. coli ORN178 strain. Individual bacteria detected no aggregation.

Figure 5.9 Detection limit aggregation of bacteria ORN178 with nanopodmannose (15). The

number of bacteria incubated with PLL-containing solution is indicated above each imagine.


183

5.7 References 1 Astruc, D.; Boisselier, E.; Ornelas, C. Chem. Rev., 2010, 110, 1857–1859. 2 Marcos, M.; Martín-Rapún, R.; Omenat, A.; Serrano, J. L. Chem. Soc. Rev., 2007, 36,

1889–1901. 3 Frauenrath, H. Prog. Polym. Sci., 2005, 30, 325–384. 4 Gao, C.; Yan, D. Prog. Polym. Sci., 2004, 29, 183–275. 5 (a) Fréchet, J. M. J.; Tomalia, D. A. Dendrimers and other Dendritic Polymers, John

Wiley and Sons, Chichester, 2001. (b) Gingras, M.; Raimundo, J. M.; Chabre, Y. M.

Angew. Chem. Int. Ed., 2007, 46, 1010–1017. 6 Hawker, C. J.; Fréchet, J. M. J. Polymer, 1992, 33, 1507–1511. 7 Yin, R.; Zhu, Y.; Tomalia, D. A.; Ibuki, H. J. Am. Chem. Soc., 1998, 120, 2678–2679. 8 Schlüter, A. D.; Rabe, J. P. Angew. Chem. Int. Ed., 2000, 39, 864–883. 9 (a) Hein, C. D.; Liu, X. M.; Wang, D. Pharmaceutical Research, 2008, 25, 2216–2230.

(b) Fournier, D.; Hoogenboom, R.; Schubert, U. S. Chem. Soc. Rev. 2007, 36, 1369–1380.

(c) Soares, E. V.; Hebbelinck, K.; Soares, H. M. Can. J. Microbiol. 2003, 49, 336–343. (d)

Chen, Z.; Meng, H.; Xing, G.; Chen, C.; Zhao, Y.; Jia, G.; Wang, T.; Yuan, H.; Ye, C.;

Zhao, F.; Chai, Z.; Zhu, C., Fang, X.; Ma, B.; Wan, L. Toxicol. Lett., 2006, 163, 109–120.

(e) Campos, L. M.; Kilops, K. L.; Sakai, R.; Paulusse, J. M. J.; Damiron, D.;

Drockenmuller, E.; Messmore, B. W.; Hawker, C. J. Macromol. 2008, 41, 7063–7070. (f)

Connal, L. A.; Kinnane, C. R.; Zelikin, A. N.; Caruso, F. Chem. Mater. 2009, 21, 576–

578. 10 (a) Booker-Milburn, K.; Anson, C. E.; Clissold, C.; Costin, N. J.; Dainty, R. F.; Murray,

M.; Patel, D.; Sharpe, A. Eur. J. Org. Chem. 2001, 1473–1482. (b) Hook, B. D. A.; Dohle,

W; Hirst P., L.; Pickworth, M.; Berry M. B.; Booker-Milburn, K.M; J. Org. Chem. 2005,

70, 7558–7564. 11 Ting, S. R. S.; Chen, G.; Stenzel, M. H. Polym. Chem. 2010, 1, 1392–1412. 12 (a) Sadik, O. A.; Aluoch, A. O.; Zhou, A. Biosens. Bioelectron. 2009, 24, 2749–2765.

(b) Moldovan, C.; Mihailescu, C.; Stan D.; Ruta, L.; Iosub, R.; Gavrila, R.; Purica, M.;

Vasilica, S. Applied Surface Science 2009, 255, 8953–8959. (c) Zhao, X.; Hilliard, L. R.;

Mechery, S. J.; Wang, Y.; Bagwe, R., P.; Jin, S.; Tan, W. PNAS 2004, 101, 15027–15032.

(d) Disney, M. D.; Zheng, J.;Swager, T. M., Seeberger, P. H. J. Am. Chem. Soc. 2004,


184

126, 13343–13346. (e) So, H. M.; Park, D. W.; Jeon, E. K.; Kim, Y. H.; Kim, B. S.; Lee,

C. K.; Choi, S. Y., Kim, S. C.; Lee, J. O. Small 2008, 197–201. 13 (a) Kikkeri, R.; Liu, X.; Adibekian, A.; Tsai, Y.-H.; Seeberger, P. H. Chem. Commun.

2010, 2197–2199. (b) Kikkeri, R.; Garcia-Rubio, I.; Seeberger, P. H. Chem. Commun.

2009, 235–237. 14 Orive, G.; Tam, S. K.; Pedraz, J. L.; Halle, J. P. Biomaterials 2006, 27, 3691–3700. 15 Doktycz, M. J.; Sullivan, C. J.; Hoyt, P. R.; Pelletier, D. A.; Wu, S.; Allison D. P.

Ultramicroscopy 2003, 97, 209–216. 16 Altschuh, D.; Oncul, S.; Demchenko, P. A. J. Mol. Recognit. 2006, 19, 459–477. 17 (a) Kikkeri, R.; Kamena, F.; Gupta, T.; Hossain, L. H.; Boonyarattanakalin, S.;

Gorodyska, G.; Beurer, E.; Coullerez, G.; Textor, M.; Seeberger, P. H. Langmuir, 2010,

26, 1520–1523. (b) Evangelista, K.; Andrich F.; Rezende, F. F.; Niland, S.; Cordeiro, M.

N.; Horlacher, T.; Castelli, R.; Schmidt-Hederich, A.; Seeberger, P. H.; Sanchez, E. F.;

Richardson, M.; Figueiredo, S.; Eble, J. A. J. Biol. Chem., 2009, 284, 34747–34759. 18 Upchurch Scientific, USA 19 Harris, S. L.; Spears, P. A., Havell, E. A.; Hamrick, T. S.; Horton, J. R.; Orndorff, P. E.

J. Bacteriol. 2001, 183, 4099–4102.

List of Abbreviations

185


Ac Acetyl

aq Aqueous

arom Aromatic

BMA Butylmethacrylate

br Broad

brs Broad singlet

c Concentration

δ Chemical shift

d Doublet

dd Doublet of doublets

ddd Double douplet of

douplets

dddd Double douplet douplet

of douplets

ddt Double douplet of triplets

DHLA Dihydrolipoic acid

DMF N,N-‐Dimethylformamide

DMSO Dimethylsulfoxide

DP Degree of polymerization

dt Doublet of triplets

equiv Molar equivalents

ESI Electron spray ionization

Et Ethyl

FEP Fluorinated Ethylene

Propylene

FT Fourier Transformation

Gal galactose

Glc Glucose

GlcN Glucosamine

GPC Gel Permeation

Chromatography

h Hour(s)

HDA Hexadecylamine

HRMS High Resolution Mass

Spectroscopy

Hz Hertz

IR Infrared Spectroscopy

J Coupling constant

m Multiplet

M Molar

MALDI Matrix Assisted Laser

Desorption/Ionisation


186

Me Methyl

min Minute(s)

MMA methylmethacrylate

Mn Molecular weight

MS Mass spectroscopy

Mw Molecular weight

m/z Mass to charge ratio

NMR Nuclear Magnetic

Resonance

PEG Polyethylene Glycol

Ph Phenyl

ppm Parts per million

PTFE Poly(tetrafluoroethylene)

q Quartet

QDs Quantum Dots

quint. Quintett

rt Room temperature

s Singlet

s Second(s)

SEM Scanning Electron

Microscope

STY Stryrene

t Triplet

T Temperature (°C)

TEM Transmission Electronic

Microscope

THF Tetrahydrofuran

tlc Thin layer chromatography

TMS2S Hexamethyldisilathiane

TOF Time of flight

TOP Trioctyl phosphine

TOPO Trioctyl phosphine oxide

tt Triplet of triplets

UV Ultraviolet

VA Vinylacrylate

Vis Visible

ZnEt2 Diethylzinc

187

Curriculum Vitae

Name: Paola Laurino Contact address: Arnimallee 22

Date of Birth: 01.04.1981 14195 Berlin, Germany

Nationality: Italian Phone: +49-30-83859562

Email: [email protected]

Languages

Italian: Native language

English: Fluent speaking, reading and writing

German: Basic knowledge

Spanish: Basic knowledge

Education and Research Experience

Since 10/2007 PhD studies, Department of Organic Chemistry

ETH Zurich, Switzerland and

Max Planck Institute of Colloids and Interfaces,

Department of Biomolecular Systems, Germany

Research advisor: Professor P. H. Seeberger

06/2006-09/2007 MPhil, Department of Medicinal Chemistry

University of Leiden, The Netherlands

Research advisor: Professor A. IJzerman

Project: Photoaffinity probes for adenosine A1 receptor.

9/2000-03/2006 Degree (Laurea) in Pharmaceutical Chemistry and Technology

Department of Medicinal Chemistry, Pharmacy Faculty,

University of Milan, Italy

Research advisor: Professor E. Valoti

Thesis: Design, Synthesis and Characterization of antagonists for α1

adrenergic receptor.

188

Publications

1. Laurino, P.; Tauer, K.; Hernandez, H.; Grützmacher, H.; Seeberger, P. H.

“Ultrafast Photo-initiated Polymerization in Continuous Flow Reactor” Under

preparation.

2. Laurino, P.; Hernandez, H.; Bräuer, J.; Grützmacher, H.; Tauer, K.; Seeberger, P.

H. “Snowballing Radical Generation Leads to Ultrahigh Molecular Weight

Polymers” 2011, Submitted.

3. Azzouz, N.; Kamena, F.; Laurino, P.; Kikkeri, R.; Mercier, C.; Cesbron-Delauw,

M-F.; Dubremetz, J-F.; De Cola, L.; Seeberger, P. H. “Toxoplasma gondii

secretory proteins bind to sulfated heparin structures” 2011, Submitted.

4. Laurino, P.; Kikkeri, R.; Azzouz, N.; Seeberger, P. H. “Detection of Bacteria

Using Glyco-Dendronized Polylysine Prepared by Continuous Flow

Photofunctionalization” Nano Lett. 2011, 11, 73–78.

5. Laurino, P.; Kikkeri, R.; Seeberger, P. H. “Continuous Flow Reactor Based

Synthesis of Carbohydrate Capped Quantum Dots” Nat. Prot. 2011, In press.

6. Christina, D.; Laurino, P.; Azzouz, N.; Seeberger, P. H. “Accelerated Continuous

Flow RAFT polymerization” Macromol. 2010, 43, 10311–10314.

7. Bernardes, G. J. L.; Kikkeri, R.; Maglinao, M.; Laurino, P.; Collot, M.; Hong, S.

Y.; Lepenies, B.; Seeberger, P. H. “Design, synthesis and biological evaluation of

carbohydrate functionalized cyclodextrins and liposomes for hepatocyte-specific

targeting” Org. Biomol. Chem. 2010, 21, 4987–4996.

8. Kikkeri, R.; Laurino, P., Odedra, A.; Seeberger, P. H. “Synthesis of carbohydrate-

functionalized quantum dots in microreactors”. Angew. Chem. Int. Ed. Engl. 2010,

49, 2054–2057.

189

9. Laurino, P.; Odedra, A.; Mak, X. Y.; Gustafsson, T.; Geyer, K.; Seeberger, P. H.

“Microfluidic devices for organic processes”. Chemical Reactions and Processes

under Flow condition (RCS publishing 2010).

10. Mak, X. Y.; Laurino, P.; Seeberger, P. H. “Asimmetric reactions in continuous

flow”. Beilstein J. Org. Chem. 2009, 5, 1–11.

11. Mak, X. Y.; Laurino, P.; Seeberger, P. H. “Synthesis of pharmaceutical and bio-

active compounds in microreactors” Chemistry Today 2009, 27, 15–18.

12. Kikkeri, R.; Lepenies, B.; Adibekian, A.; Laurino, P.; Seeberger P. H. “In vitro

imaging and in vivo liver targeting with carbohydrate capped quantum dots”. J.

Am. Chem. Soc. 2009, 131, 2110–2112.

Patents

1. Seeberger, P. H.; Grützmacher, H.; Tauer, K.; Laurino, P.; Bräuer, J. “Ultra fast

process for the preparation of polymer nanoparticles” Submitted.

2. Seeberger, P. H.; Grützmacher, H.; Tauer, K.; Laurino, P.; Bräuer, J. “Process for

the modification of polymer nanoparticles” Submitted.

3. Seeberger, P. H.; Laurino, P.; Kikkeri, R. “Continuous Flow Photoconjugation” In

preparation.

Poster Presentations

1. “Synthesis of Monodisperse Latex Particles”, IPCG conference. Il Ciocco, Italy.

July 2009.

2. “Photochemical Reactions in Microreactor” at Industry Day “Micro and Nano

Science – From Ideas to Innovations” ETH Zurich, Switzerland. 8th May, 2008.

190

3. “Photochemical Reactions in Mircoreactor” at 4th Workshop “Microtechnology for

Chemistry and Biology Laboratories” Technische Universität Ilmenau, Germany.

February 2008

Courses Attended

1. “Enzymes”, Professor D. Hilvert. Autumn Semester 2008, ETH Zurich,

Switzerland.

2. “Advanced Methods and Strategies in Asymmetric Synthesis”, Professor E.

Carreira. Autumn Semester 2008, ETH Zurich, Switzerland.

3. “Biomaterial Surfaces: Properties and Characterization”, Professor M. Textor.

Autumn Semester 2008, ETH Zurich, Switzerland.

4. 2008 Annual course SPERU “Colloidal Chemistry for Materials Science”,

Competence Centre for Materials Science and Technology, Luzern, Switzerland.

30-31 October 2008.

5. Short Course “Photochemistry of Materials” von Stefan Berndhard, Princeton,

USA. Universität Zürich, Switzerland. November 2008

Teaching experiences

1. Spring Semester 2007/2008 Teaching assistant, Course of Organic Chemistry II.

ETH Zurich, Switzerland.

2. Autumn Semester 2008/2009 Lab assistant, Praktikum Organic Chemistry I. ETH

Zurich, Switzerland.

Date post:	11-Sep-2021
Category:	Documents
Upload:	others
View:	7 times
Download:	0 times

photochemical transformations in continuous flow devices

Documents