Analytica Chimica ActaC. Frigerio et al. / Analytica Chimica Acta 735 (2012) 9–22 11 João L.M....

R

AA

CJR

h

�t�tctrEg

a

ARRAA

KNQAFCEFC

C

0h

Analytica Chimica Acta 735 (2012) 9– 22

Contents lists available at SciVerse ScienceDirect

Analytica Chimica Acta

j ourna l ho me page: www.elsev ier .com/ locate /aca

eview

pplication of quantum dots as analytical tools in automated chemical analysis: review

hristian Frigerio, David S.M. Ribeiro, S. Sofia M. Rodrigues, Vera L.R.G. Abreu, João A.C. Barbosa,oão A.V. Prior, Karine L. Marques, João L.M. Santos ∗

EQUIMTE, Laboratory of Applied Chemistry, Department of Chemical Sciences, Faculty of Pharmacy of Porto University, Rua Jorge Viterbo Ferreira, 228, 4050-313 Porto, Portugal

i g h l i g h t s

Review on quantum dots applica-ion in automated chemical analysis.

Automation by using flow-basedechniques. � Quantum dots in liquidhromatography and capillary elec-rophoresis. � Detection by fluo-escence and chemiluminescence. �lectrochemiluminescence and radicaleneration.

g r a p h i c a l a b s t r a c t

r t i c l e i n f o

rticle history:eceived 22 February 2012eceived in revised form 24 April 2012ccepted 27 April 2012vailable online 10 May 2012

eywords:anocrystalsuantum dots

a b s t r a c t

Colloidal semiconductor nanocrystals or quantum dots (QDs) are one of the most relevant developmentsin the fast-growing world of nanotechnology. Initially proposed as luminescent biological labels, they arefinding new important fields of application in analytical chemistry, where their photoluminescent prop-erties have been exploited in environmental monitoring, pharmaceutical and clinical analysis and foodquality control. Despite the enormous variety of applications that have been developed, the automationof QDs-based analytical methodologies by resorting to automation tools such as continuous flow analysisand related techniques, which would allow to take advantage of particular features of the nanocrystalssuch as the versatile surface chemistry and ligand binding ability, the aptitude to generate reactive species,

nalytical chemistrylow analysishromatographylectrophoresisluorescence

the possibility of encapsulation in different materials while retaining native luminescence providing themeans for the implementation of renewable chemosensors or even the utilisation of more drastic andeven stability impairing reaction conditions, is hitherto very limited. In this review, we provide insightsinto the analytical potential of quantum dots focusing on prospects of their utilisation in automatedflow-based and flow-related approaches and the future outlook of QDs applications in chemical analysis.

hemiluminescence

© 2012 Elsevier B.V. All rights reserved.

ontents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Optical properties and surface chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3. Application of quantum dots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.1. Fluorescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1. Direct fluorescence measurements . . . . . . . . . . . . . . . . . . . . . . .

∗ Corresponding author. Tel.: +351 220428668; fax: +351 226093483.E-mail address: [email protected] (J.L.M. Santos).

003-2670/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.aca.2012.04.042

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

dx.doi.org/10.1016/j.aca.2012.04.042http://www.sciencedirect.com/science/journal/00032670http://www.elsevier.com/locate/acamailto:[email protected]/10.1016/j.aca.2012.04.042

10 C. Frigerio et al. / Analytica Chimica Acta 735 (2012) 9– 22

3.1.2. FRET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133.2. Chemiluminescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

3.2.1. Direct chemiluminescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143.2.2. Chemiluminescence catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153.2.3. CRET. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163.2.4. Electrogenerated chemiluminescence (ECL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

3.3. Liquid chromatography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183.4. Capillary electrophoresis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

4. Prospects and trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204.1. Solid phase reactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204.2. Magnetic nanocomposites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204.3. Bead injection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

João A.C. Barbosa is an MSc student at Faculty of Phar-macy of the University of Porto. He is involved in aproject concerning the synthesis and characterisationof CdTe Quantum Dots and their utilisation in auto-mated analytical flow systems.

João A.V. Prior is appointed as Auxiliary Professorat the Faculty of Pharmacy at University of Porto.He obtained a PhD degree in Analytical Chemistry in2005. His research interests involve the developmentof new analytical methodologies and implementationin automatic analytical systems based in flow anal-ysis concepts. The research is aimed for applicationin pharmaceutical, environmental, forensic and tox-icological chemistry. His research work makes useof different spectrometric techniques for detectionand quantification of several drugs upon chemicalderivatisation.

Karine L. Marques is a Researcher at REQUIMTE Asso-ciate Laboratory, Faculty of Pharmacy, University ofPorto. She obtained her PhD degree in AnalyticalChemistry (2006) from Faculty of Pharmacy, Univer-sity of Porto. Her research is related with automatedflow methodologies with emphasis on automationof chemiluminometric detection. Her recent researchinterests are now focused on studying and applyinggold nanoparticles to the development of green ana-lytical methodologies.

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Christian Frigerio is a PhD student at the Faculty ofPharmacy of University of Porto. He obtained his MScdegree in Analytical Chemistry – Quality Control on2009. He is pursuing studies in the area of synthesisand characterisation of quantum dots for the develop-ment of automatic flow analysis systems. His actualwork is exploiting the surface chemistry propertiesof quantum dots with analytical purposes, in partic-ular reactions involving quantum dots fluorescenceenhancing.

David S.M. Ribeiro is finishing his PhD thesis in Phar-maceutical Analysis at the Faculty of Pharmacy ofPorto University. He holds a degree in Chemistry andhas obtained an MSc in Analytical Chemistry – Qual-ity Control, in 2008. He is carrying out research onthe development of automated methods of analysisfor forensic and toxicological chemistry, solid-phaseextraction and photocatalysis.

S. Sofia M. Rodrigues is a PhD student at Faculty ofPharmacy of Oporto University. She obtained a Mas-ter Degree in Analytical Chemistry – Quality Controlin 2009 and a degree in Physics in 2002. Her researchinterests focus on development of automated micro-flow analytical systems based on CdTe quantum dotsphotoluminescent properties and reactivity, for food,environmental and biomedical detection and moni-toring.

Vera L.R.G. Abreu is a Researcher at REQUIMTEAssociate Laboratory, Faculty of Pharmacy of OportoUniversity. She obtained her PhD degree, in Chem-istry of Materials (2009) at the Faculty of Technology,Tomas Bata University in Zlín. Her research inter-ests focus the automation of chemical analysis basedmainly on the development and implementation ofnew methodologies for applications in pharmaceuti-

cal, clinic and environmental products. Recently, herresearch interests include the exploitation of the opti-cal properties of quantum dots, in particular theirapplication in flow systems.

Chim

1

aseos[ptidc

doettca[

fr[tacifdMbtopsotss

dsvsrca

C. Frigerio et al. / Analytica

João L.M. Santos is Auxiliary Professor at the Facultyof Pharmacy, University of Porto, and research scien-tist at REQUIMTE Associate Laboratory. He receivedhis PhD in Analytical Chemistry in 2000. His researchactivities have been focused in the development ofautomated flow-based methodologies, mainly mul-ticommutation, multipumping and single reactioninterface flow systems. His current interests includenanotechnology and nanomaterials, namely quantumdots and metallic nanoparticles for photoluminescentdetection and nanodiagnostics.

. Introduction

Colloidal semiconductor nanocrystals or quantum dots (QDs)re monodisperse crystalline clusters with physical dimensionsmaller than the bulk-exciton Bohr radius, i.e. the distance in anlectron-hole pair [1]. This very small size has a definite effectn the QDs photoluminescent properties, which could be under-tood through the all spatial dimensions confinement of the exciton2,3]. Since the spacing of the highest occupied and lowest unoccu-ied quantum confined orbitals of QDs is determined by the size,he energy levels can be tuned in a controlled way by synthesis-ng nanocrystals of different diameter, assuring both specificallyesigned optical properties and the valuable feature of wide rangeontinuous band gap tunability [4].

QDs have attracted great interest during the last decades mostlyue to the immense and promising prospects of QDs-bioconjugatesr the bio-applications that they have made possible. With uniquelectro-optical properties, arising from the size-dependent andunable photoluminescence (PL) and long-term photostability,hese nanomaterials emerged as advantageous alternatives to theommonly used molecular probes in biological and biomedicalpplications including bio-labelling, bio-imaging and bio-targeting5–7].

A variety of QDs have been prepared, usually composed of atomsrom groups II–VI, III–V, or IV–VI, the emphasis been given to mate-ials such as CdSe [2], CdTe [2], ZnSe [8], CdS [2], CdSe/ZnS [9], InP10], etc. These were obtained by using distinct synthesis routeshat ensured a strict control of the constituent material, size, shape,nd surface chemistry. Among the different physical and chemi-al processes for QDs preparation, the colloidal chemistry methods the best route to synthesise nanocrystals with proper surfaceunctionality providing high luminescence efficiency, narrow sizeistribution and the ability to interact with selected species [2].oreover, since the QDs are dispersed in a solvent, they should

e stabilised in a way that prevents agglomeration. The ability toailor QDs surface with appropriate ligands is therefore an aspectf paramount importance affecting not only nanocrystals solutionroperties and solubility but also their potential use as chemosen-ors. In effect, selective binding on the surface of the nanocrystals,r the ligand template effect, has been frequently accountable forhe nanocrystals shape. In addition, ligand effects also control theize and size distribution during the QDs synthesis as well as crystaltructure and nanocrystals stability [11].

Recent advances in QDs nanotechnology have slowly intro-uced these nanomaterials in analytical areas mostly as chemicalensors in fluorescence-based measurements [12–14]. Due to theery small size and high surface-to-volume ratio of QDs their

urface become of utmost importance as any modification of sur-ounding medium or interaction of given chemical species, whichould be modulated at distinct levels, would result in significantlteration of the photoluminescent properties, namely in terms

ica Acta 735 (2012) 9– 22 11

of emission intensity. In most of the circumstances it could pre-dominantly affect the QDs quantum yield, defined as the rationbetween emitted and absorbed photons, by influencing the effi-ciency of electron-hole recombination, but could also promote aspectral shift or a change in the PL decay time. From an analyticalpoint of view, the modulation of QDs luminescence as a selec-tive response to a given analyte recognition could be perceived bydifferent means: fluorescence intensity quenching or enhancing,fluorescence resonance energy transfer (FRET), direct or indirectchemiluminescence enhancing or quenching, chemiluminescenceresonance energy transfer (CRET) and annihilation or coreactantelectroluminescence [15]. These could be used alone or coupledto separation techniques like liquid chromatography or capillaryelectrophoresis.

The aim of this article is to overview and highlight recentwork on analytical applications involving the utilisation of QDs,not directly related with a biological perspective and dealing witha certain degree of automation in terms of sample and reagentshandling, mostly by means of continuous flow approaches and/orrelated techniques. As this work will show up, this is a field scarcelyresearched but with an immense potential due to the combina-tion of three main features: high reactivity of the QDs, strict andreproducible control of the QDs/analyte reaction conditions andversatile monitoring of reaction progression with distinct detectiontechniques.

2. Optical properties and surface chemistry

With a size typically between 1 and 10 nm quantum dots showoptical, electronic and mechanical properties quite different fromthose exhibited by bulk materials [16]. In bulk semiconductors thelarge number of atoms leads to the formation of almost a continuumof energy levels. The valence band, comprising the lower energylevels, is filled with electrons and separated from the unoccu-pied conductance band, corresponding to the higher energy levels,by a fixed energy gap. In quantum dots, due to their small size,the energy levels are discrete and the energy gap between thevalence and conductance band depends on the nanocrystal size.The absorption of a photon, leading to the excitation of an elec-tron from the valence band to the conductance band, is relatedwith the band gap energy. Its dependence on nanocrystal size canbe understood through the three dimensions confinement of theexciton, i.e. the excited electron-hole pair whose recombinationis responsible for the fluorescence emission [1]. Similarly to theparticle-in-a-box model a stronger confinement leads to a largerseparation of the energy levels and therefore the band-gap of semi-conductor QDs increases as size decreases resulting in the emissionof light at shorter wavelengths (blue shift). Oppositely, the band-gap decreases as more atoms are added to the nanocrystal withthe lower limit of the band-gap corresponding to the bulk mate-rial. Moreover, increasing the nanoparticles size leads not onlyto an increase in the emission wavelength (red shift) but alsoto a decrease in the molar extinction coefficient [16]. Since QDsemission is size-dependent they can be tailored in a controlledway by adjusting the synthesis conditions to assure a fluorescenceemission matching virtually any wavelength of the visible region.Apart from the band-gap tunability, QDs exhibit other relevantphotoluminescent properties, namely an intense and highly sta-ble against photobleaching fluorescence, potentially high quantumyield, broad absorption and narrow, symmetric emission spectra

and long excited-state decay lifetimes. Moreover, the broad absorp-tion bandwidth, due to the presence of multiple electronic states athigher energy levels, allows simultaneous excitation of multicolourQDs by using a single light source.

1 Chim

fitpstcCibflptbrtdswoahmllimotcsrtirpalafl

stcoTonQboeorrtcaall

saT

2 C. Frigerio et al. / Analytica

Distinct synthetic methods for the preparation of quantum dotsrom a variety of materials have been reported. The colloidal chem-stry route enables not only controlling nanocrystals size but alsoheir surface ligands, providing nanomaterials with distinct surfaceolarity either prepared in aqueous or organic environment. Theolvent selection is crucial at various levels as it determines not onlyhe nanocrystals synthesis and solubility conditions but also theirhemical functionality and applicability [2]. Highly luminescentdSe QDs with low size dispersion were synthesised by the rapid

njection of organometallic precursors into a high-temperatureoiling coordinating organic solvent [17,18]. Since adequate QDsuorescence is only verified when the nanocrystals exhibited aroper surface passivation, the occurrence of surface imperfectionshat act as charge carriers traps could prevent electron-hole recom-ination favouring deactivation by non-radiative processes, thuseducing the QDs quantum yield. The synthesis of core/shell quan-um dots assured an improvement of passivation by means of theeposition of surface-capping shell of a secondary semiconductor,uch as ZnS or CdS, significantly increasing the quantum yield asell as the stability against photo-oxidation [19,20]. Since most

f the analytical, catalytic or bio-targeting applications demandqueous environments as reaction media QDs capped with organicydrophobic ligands are inadequate. Therefore, suitable surfaceodifications strategies to replace the non-polar encapsulating

ayer by more polar species, while preserving the stability and theuminescence properties have been reported. These include util-sation of protective layer of thiolated or dithiolated functional

onomers, such as thioglycolic (TGA) [21] and mercaptopropi-nic acids (MPA) [21], cysteine terminated peptides [22], etc., orhe coating of the QDs with polymeric materials [23,24] or sili-on oxide films [25,26]. In alternative, several methods to directlyynthesised water-soluble QDs were developed. These methodsesort to hydrophilic capping ligands that formed a thin-layer athe semiconductors surface by covalent or ligand-ion electrostaticnteractions and are characterised by both high simplicity andeproducibility [1] being also easily scaled-up for large amountroduction. In this regard, short-chain thiols (exhibiting as wellnother functional group such as carboxylic, amino, hydroxilic, etc.)ike TGA, MPA, cysteamine, have gained a noteworthy relevances solution passivating agents as they provide bright QDs with aexible surface chemistry that remain stable for years.

In the application of quantum dots, controlling the nanocrystalize, morphology and surface-ligands is of primordial importanceo control the optical properties, and therefore the photolumines-ence response [15]. Depending on the analyte-targeting strategyther specific characteristics should be also taken into account.hese include, but are not necessarily limited to: the occurrencef surface imperfections (traps), capping reactivity, capping thick-ess, surface charges, and solution stability. The influence on theDs photoluminescence of ligands, ions, small molecules, adsor-ates or other surface interactions that could affect the efficiencyf electron-hole recombination could be used as a detection strat-gy. In effect, QDs quantum yield is a direct consequence of theccurrence of radiative (higher QY) or non-radiative (lower QY)ecombination mechanisms. An analyte inducing a concentrationelated luminescence enhancement or quenching upon chargeransfer, mechanical adsorption, ion chelating or ligand exchangeould be most likely determined by using QDs. In this regard, annalyte promoting a luminescence enhancement could be moredequately determined by using QDs with lower QY, while when auminescence quenching is intended QDs with initially high QY areikely to provide improved results.

Aside from their well known photoluminescent properties QDshow catalytic properties for redox reactions, size dependent cat-lytic action and controllable charge and electron transfer events.hese properties could be exploited to maximise the response

ica Acta 735 (2012) 9– 22

signals of instrumental methods of analysis, enhancing detec-tion and improving sensitivity, to increase the selectivity or thekinetics of a given chemical reaction or even to implement newreactional schemes. Chemiluminescence (CL) and electrochemi-luminescence (ECL) measurements are based in redox reactionsinvolving suitable luminogenic compounds or precursors and/oradequate electrochemical conditions. The reactional process usu-ally entails the formation of short-lived intermediate radicals thatresult in unstable products that decompose to form electronicallyexcited molecules that relax upon light emission [27]

Quantum dots can be applied as luminescent probes formany species either biological or chemical. The sensitivity of theinvolved photophysical processes related to the formation of thefluorophore-target analyte pair depends markedly on the outerstabilizing capping layer surrounding the nanocrystals. Therefore,an important balance between the solution stability of the QDsassociated with the capping and the target recognition event affect-ing the photoluminescent properties of the nanoparticle should bemaintained during measurements. This is a noteworthy dilemmawhen batch-procedures are involved. The need for stable read-outs associated with chemical equilibrium could be a hindrancebecause surface interactions not only change the photolumines-cence response but also the quantum dots solution stability.Consequently, more pronounced signal intensity could be achievedonly at the expenses of a poorer stability, which could ultimatelyresult in no reading at all. Automation of all reactional processes,including sample and reagents (QDs) mixing, reaction develop-ment and analytical signal measurement and/or acquisition couldallow overcoming this hindrance by means of reproducible solu-tions insertion and timed reaction development without the needfor attaining equilibrium conditions. Flow-based techniques, suchas flow injection analysis (FIA) [28], sequential injection analysis(SIA) [29], multicommutation (MCFIA) [30], multisyringe (MSFIA)[31], multipumping (MPFS) [32] and single interface flow analy-sis (SIFA) [33] flow approaches are expeditious automation toolsthat could assure all the requisites for a more expeditious and effi-cient utilisation of quantum dots as chemosensors in combinationwith distinct detection modes. Similar results could be obtainedwith flow-related techniques such as liquid chromatography andcapillary electrophoresis.

3. Application of quantum dots

3.1. Fluorescence

As it was previously referred, in recent years QDs have emergedas remarkable fluorescent materials in numerous biological appli-cations with noteworthy advantages regarding comparable organicmolecular probes. The photoluminescence properties of QDs areindeed their most relevant feature in a chemical analysis perspec-tive. The ability to absorb light at a broad bandwidth and to emitat a narrow spectrum, being the emission intensity dependent onthe quantum yield, which in turn is conditioned by the type of sur-face interactions that the QDs established with the target analyte,has been exploited to implement assorted fluorescent analyticalmethodologies.

3.1.1. Direct fluorescence measurementsPioneering studies focusing the interactions between quantum

dots and metal ions showed that the luminescence response wasmarkedly affected by the nanocrystals surface capping ligands

[34,35]. As a result, by suitable selection within a variety of QDs sur-face ligands it would be possible to establish specific chemosensors.Amid the multiple organic capping agents that have been subse-quently exploited for QDs surface modification and to implement

C. Frigerio et al. / Analytica Chim

scocicemebaodsq

meiasTtf

Fig. 1. Schematic quenching process for QDs particle.

elective analytical applications, mercapto-acids [21,36], and mer-aptoamines [37] were those providing the best results. The typef modification on the QDs luminescence response depends on theapping and also on the chemical nature of the target analyte beingts magnitude ideally concentration related. Distinct QDs fluores-ence modification strategies have been exploited such as intensitynhancing and quenching and emission wavelength shifting. Theost exploited mechanism was, with no doubt, the quenching

ffect (Fig. 1), where a given analytes reduces the QDs PL emissiony interacting with his surface or capping. Depending on the inter-ction between quencher and QDs, quenching could be either staticr dynamic, and these could be distinguished by their differingependence on temperature. For static quenching the rate con-tants decreased with increasing temperature while for dynamicuenching the opposite phenomenon occurred [38].

A literature survey reveals that a great number of analyticalethods have been already developed with analytical purposes,

specially for metals [39–48] and organic molecules monitor-ng [49–58]. All of these methods, except one, relied on batchpproaches involving the manual handling of sample and reagents

olutions for reaction implementation and signal measurement.he only exception resorted to an automated continuous flow sys-em. In this work, Butwong et al. [59] developed a methodologyor the determination of arsenic in water based on the fluorescence

Fig. 2. FRET proce

ica Acta 735 (2012) 9– 22 13

quenching of CdS-MAA quantum dots using a sequential injectionanalysis (SIA) system coupled to an online gas-diffusion unit forarsine (AsH3) generation. The generated arsine diffused throughthe PTFE membrane and interacted with CdS-MAA QDs promotingthe fluorescence quenching. The main parameters that influencedthe fluorescence intensity such as CdS-MAA QDs concentration,pH, the nature and the concentration of the buffer were evalu-ated. pH strongly affected the quenching ability of CdS-MAA QDsfluorescence intensity by AsH3. The highest relative fluorescenceintensity (I0/I) was observed when using a pH of 4.6, indicatinga maximum quenching. The use of acetate buffer (10 mmol L−1)provided a higher fluorescence intensity when compared with thephosphate buffer. The proposed interaction mechanism was basedboth on the fact that there was no significant shift of the full width athalf-maximum (FWHM) of the absorption and fluorescence spec-tra when increasing the concentration of arsenic and also in theappearance, in the IR spectra, of new peaks at 1160, 1050 and1020 cm−1. These evidences, in combination with the ability ofAsH3 to reduce covalent bonds, supported the authors’ assump-tion that the quenching of CdS-MAA QDs fluorescence intensityoccurred as a consequence of the formation of new bonds As–Son MAA functional group on CdS–MAA QDs surface. The devel-oped analytical system combining the sensitivity and selectivityprovided by the QDs fluorescence quenching with the simplicity,ease of operation and automation level of the automated flow sys-tem revealed to be a successful and expeditious strategy for thedetermination of arsenic in spiked ground water samples, showingthat this synergy has a huge potential that could be advantageouslyapplied in countless analytical circumstances.

3.1.2. FRETIn the last decades Förster resonance energy transfer, also

known as FRET, has gained a noteworthy relevance as a power-ful tool in physical chemistry and biophysics. Perrin and Försterfirst theoretically described the process of non-radiative trans-fer of energy from a donor (or sensitizer) chromophore, in anexcited electron state, to an acceptor chromophore over distancesof up to 10 nm. Along with the orientation factor, the spectraldonor emission/acceptor absorption overlap and the distance (r)between the donor and the acceptor are parameters that affectthe FRET effectiveness. In this regard, the distinctive QDs size-

dependent photoluminescent properties make them superlativedonors (Fig. 2). In effect, the tunable size achieved during the QDssynthesis makes it possible to vary this parameter (r) as well as itenables matching the QDs emission wavelength with the acceptor

ss for QDs.

1 Chimica Acta 735 (2012) 9– 22

ae

raottcmiwf

tumcosstb6opt6pc6st6tw

efl(tqsmm

tbctblptdwh

3

lybo


bsorption maximum, creating and adjusting sensing systems withnhanced sensitivity.

In literature, many assays using QDs and FRET have beeneported in which QD-based nanosensors with QDs as FRET donorsnd organic dyes as acceptors have been used [60]. However, thepposite situation, in which the organic dyes acted as donors andhe QDs as acceptors, has been seldom exploited. On the other hand,he combined utilisation of automated methods of analysis, such asontinuous flow analysis systems, and quantum dots for the imple-entation of FRET based assays has been only vaguely examined

n chemical and biological studies. Indeed, only a few examples inhich QDs-based FRET is combined with microfluidic devices are

ound in literature.In 2006, Zhang and Johnson [61] described a novel approach

o improve the detection sensitivity of QD-based nanosensorssing single molecule detection in a capillary flow. By using com-ercially available streptavidin-coated QDs as FRET donors and

yanine dye (Cy5) as acceptor, QD-based biosensors were devel-ped for the rapid and sensitive detection of DNA. To this end,ingle-stranded Cy5-labelled 25-mer DNA (ssDNA) and double-tranded Cy5-labelled 25-mer DNA (dsDNA) were assembled ontohe 605QD surface by specific streptavidin-biotin binding. Theinding of dsDNA/ssDNA to 605QD resulted in the formation of05QD/DNA/Cy5 complexes. Upon excitation with a wavelengthf 488 nm, FRET occurred between 605QD and Cy5s in the com-lexes. The authors verified that the FRET efficiency improved withhe increasing DNA-to-605QD ratio in both 605QD/dsDNA/Cy5 and05QD/ssDNA/Cy5 complexes since a single 605QD had multi-le Cy5-labelled dsDNA/ssDNA assembled on its surface. Whenomparing the two complexes, the FRET efficiency was higher in05QD/ssDNA/Cy5 complexes because of the higher flexibility ofsDNA in solution, which could form a random coiled conformationhat consequently brought the Cy5 acceptor spatially closer to the05QD. The deformation of DNA in the capillary stream increasedhe FRET efficiency of 605QD/DNA/Cy5 complexes comparativelyith the one that was achieved in bulk measurements.

The same conjugates QDs (605QD) were used to develop anxpeditious convergence of microfluidics and QD-FRET techniqueor monitoring the self-assembly of chitosan/DNA polyplexes underaminar flow in real time [62]. The 605QD, used as FRET donorexcited at 488 nm), and Cy5, used as acceptor, were labelledo plasmid DNA (pDNA) and chitosan, respectively, and subse-uently the pDNA/chitosan interactions were monitored via FRETignals. According to the authors, the convergence of QD-FRET andicrofluidics enables an innovative platform to monitor funda-ental reactions with high sensitivity and milliseconds resolution.More recently a solid-phase nucleic acid assay within an elec-

rokinetically driven microfluidic system using immobilised QDsiosensors was presented [63]. A green-emitting CdSe/ZnS QDs wasonjugated with two different oligonucleotide sequences, one ofhe oligonucleotide sequences being available as a linker for immo-ilisation via hybridisation with complementary oligonucleotides

ocated on a microfluidic channel while the second one served as arobe to transduce hybridisation with the target nucleic acid inhe sample solution. This way, CdSe/ZnS QDs was used as FRETonor and a Cy3 label, on the target oligonucleotide sequences,as used as acceptor providing an analytical signal to monitor theybridisation detection.

.2. Chemiluminescence

Chemiluminescence (CL) is typically defined as the emission of

ight by a molecule as consequence of a chemical reaction. In recentears, however, research has been focused not only in moleculesut also in the CL of nanomaterials systems, to expand the fieldf applications of this mode of detection and to attain improved

Fig. 3. Chemiluminescence emission process for QDs.

sensitivity and stability, mainly resulting from the high surface areaand special structure of nanoparticles. Effectively, most CL reactionsshow low quantum efficiencies and therefore weak luminescence.This hindrance can be overcome by resorting to fluorescent com-pounds with high quantum efficiencies acting as sensitizers. In thisparticular QDs, due to their reactivity, brightness and continuousband gap tunability, have attracted great attention. Indeed, the useof QDs for instance as chemiluminescence emitters in combina-tion with CL equipment capable of wavelength discrimination canprofit from multicolour labelling at a wide range of wavelengthswithout excitation light source [64]. Moreover, the combined utili-sation of differently sized nanocrystals targeting distinct moleculescould enable multi-parametric monitoring.

There are three possible mechanisms that could be used toexplain the participation of QDs in a CL reaction: (i) as emitterspecies, after direct oxidation; (ii) as catalysts of a reaction involv-ing others luminophores; (iii) or as emitter species, after CRET.

The identification of the QDs role in a CL emission could be achallenging task. In fact, when for instance the QDs are the finalemitters, it is particularly difficult to find out if the CL generationmechanism is associated with a direct QDs oxidation or a CRET phe-nomenon has occurred. Sometimes, it is also possible that the twomechanisms take place simultaneously. Nevertheless, in a broadperspective, when in a CL system the QDs are the only lumines-cent compounds one can definitely presume that a direct oxidationtook place. However, if an additional luminophore is also includedin the reactional scheme, the mechanism can be either CRET, if thefinal emitters are the QDs, or CL catalysis by the QDs, when the finalemitter is the luminophore.

3.2.1. Direct chemiluminescenceDirect QDs chemiluminescence (Fig. 3) QDs happens when an

electron is injected in the conduction band (Eq. (1)), a hole isinjected in the valence band (Eq. (2)). The chemically generatedexciton then relaxed and upon electron-hole recombination (Eq.(3)) radiant energy is emitted as showed in Eq. (4);

O2− + CdTe → CdTe(e−1Se) + O2 (1)

OH• + CdTe → CdTe(h+1Sh) + OH− (2)CdTe(e−1Se) + CdTe(h

+1Sh) → (CdTeNCs)∗ (3)

(CdTeNCs)∗ → (CdTeNCs) + hv (4)Wang et al. [64] were the first to directly oxidise CdTe TGAcapped nanocrystals in solution with a corresponding CL emis-sion. In this work, the evaluation of the influence of theoxidants used in the reaction showed that the intensity ofthe CL generated was dependent on the oxidant nature, asH2O2 > KMnO4 > KIO4 > Ce4+ > K3Fe(CN)6 > (NH4)2S2O8, even con-

sidering that H2O2 was 1 mol L−1 and KMnO4 was 1 × 10−4 mol L−1and that the CL reaction was faster in the case of KMnO4. Followingthese results, the authors have used a flow-injection chemilumi-nometric system and the CdTe–TGA/H2O2 system to study other


Table 1Reactions for direct chemiluminescence.

Wang et al. [64] RSH + O2 + OH− → O2− + RS + H2OO2

− + CdTe → CdTe(e−1Se) + O2O2− + H2O2 → •OH + 1O2OH• + CdTe → CdTe(h+1Sh)CdTe(e−1Se) + CdTe(h

+1Sh) → (CdTeNCs)∗

(CdTeNCs)* → (CdTeNCs) + h�Li et al. [67] H2O2 + OH− → HO2− + H2O

H2O2 + HO2− → O2− + OH− + H2OO2− + H2O2 → •OH + OH− + 1O2•OH + CdS → CdS(h+1Sh)•OH + H2O2 → H2O + HO2•HO2• + OH− → O2− + H2OO2

− + CdS → CdS(e−1Se) + O2CdS(h+1Sh) + CdS(e−1Se) → (CdS NCs)

∗ → hv(CdS NCs)* → CdS NCs + h�

Chen et al. [68] HCO3− + H2O2 ↔ HCO4− + H2OHCO4− → •CO3− + •OHH2O2 + •CO3− → HCO3− + HO2•HO2• → H+ + •O2−•O2− + •OH → 1O2 + OH−1O2 → 3O2 + h�•OH + HCO3− → OH− + •HCO32•HCO3 → (CO2)2* + H2O2(CO2)2* → CO2 + h�CdSe/CdS+•O2

− → CdSe/CdS(e−1Se) + O2CdSe/CdS+•OH → CdSe/CdS(h+ )+−OH

fsfiatNttlntbremTi

tiHfipftPctwtai

yapt

1ShCdSe/CdS(e−1Se)+CdSe/CdS(h

+1Sh) → CdSe/CdS∗

CdSe/CdS* → CdSe/CdS + h�

undamental parameters affecting light emission such as pH, QDsize and concentration and the presence of different kinds of sur-actants. The obtained results demonstrated that the CL intensityncreased linearly with the H2O2 concentration (until 1 mol L−1)nd the QDs concentration (until 1 mmol L−1). In alkaline mediumhe reaction was strongly enhanced and the optimal value for theaOH concentration was 0.2 mol L−1. An important finding was

hat QDs CL emission was dependent of the QDs size, accordingo the CL energy matching theory [65]. Surfactants also promotedight emission, especially CTAB and �-CD. CTAB, due to his cationicature, interacted with TGA in the QDs surface and promotedhe formation of nanocrystals aggregates, reducing the distanceetween nanocrystals and minimizing the energy loss by non-adiative processes, two aspects that favoured the formation ofxcited states. �-CD did not promote aggregation but formed aicellar nanoenvironment that reduced non-radiative processes.

he reaction mechanism proposed for the CdTe–TGA/H2O2 systems presented in Table 1.

By employing the same CL reaction system Li et al. [66] studiedhe influence of metal ions on the generated CL using a steady-njection system. The effect of the main variables, including pH,

2O2 concentration and size and concentration of QDs, were con-rmed. Of the metal ions studied Ba2+, Ca2+, Cu2+, Fe2+ and Pb2+

romoted a CL enhancing. Under the optimal reaction conditionsor Fe2+ a linear relationship between CL intensity and Fe2+ concen-ration in the range from 0.1 to 10 mmol L−1 was verified, while forb2+ CL a ternary function response to the logarithm of Pb2+ con-entration in a range from 0.01 to 10 mmol L−1 was observed underhe same circumstances. Cr3+, Ni2+, Zn2+ and Ag+, on the other hand,ere responsible for a decrease in the CL intensity proportional

o the logarithm of metal concentration. The authors proposed thepplication of the developed system for the detection of these metalons.

In another work, Li et al. [67] used a similar flow-injection anal-

sis (FIA) system to investigate the CdS–MA/H2O2 and apply it tossess the effects of selected biological molecules as model com-ounds for future applications. The general principles regardinghe influence of pH and NaOH and H2O2 concentrations were once

ica Acta 735 (2012) 9– 22 15

again confirmed, as it was the enhancing of CL response for increas-ing QDs size and concentration. The influence of cationic surfactantCTAB was similar to the previously referred, but differently fromCdTe–TGA/H2O2 system. �-CD did not promote a signal enhancing,maybe because of the different capping used. The most importantdifference resided in the reaction mechanism that was proposed(Table 1). They authors sustained that peroxide and superoxideradicals were formed by the reaction of a base (OH−) with H2O2,and that O2− can form OH• by Haber–Weiss reaction. Nonethe-less, the core reaction that led to the formation of the exciton wasidentical. The developed CL system was tested with some reducermolecules and ions that reveal to inhibit, with no exception, the CLsignals. This happened because compounds with reducing activitycan compete with QDs for the reactive oxygen species, while themetal ions can interact with the QDs surface changing the CL prop-erties. The obtained results showed that the CL system exhibited avery good sensitivity for some compounds allowing the determi-nation at concentration levels between 10−6 and 10−9 mol L−1 andwide dynamic concentration ranges of 2–3 orders of magnitude.

The HCO3−–H2O2–CdSe/CdS–MAA CL system was implementedin FIA by Chen et al. [68] and applied in the determination ofascorbic acid in human serum. An improved analytical system per-formance was verified in alkaline pH (11.16), NaHCO3 0.5 mol L−1

and H2O2 0.5 mol L−1. It was also detected the CL dependenceof QDs size with an enhancement of the analytical signal up to2.14 nm. For bigger nanoparticles a CL decrease was observed.The CL intensity also increased until a QDs concentration of6 × 10−6 mol L−1 decreasing for higher values probably because thechemical energy of the system was not enough to excite all QDs.

The proposed reaction mechanism (Table 1) suggested the pres-ence in the system of 3 emitters; 1O2 that recombines into 3O2and emits light, (CO2)2* converted into CO2 emitting light and thealready mentioned excited form of the QDs (CdSe/CdS*) formed byOH• direct hole injection into the 1Sh quantum confined orbital andby •O2− electron injection into the 1Se orbital.

The determination of ascorbic acid showed a linear rangebetween 1 × 10−4 and 1 × 10−7 mol L−1, and a detection limit of6.7 × 10−9 mol L−1. The developed method was applied to theanalysis of human serum with low interference and acceptableperformance.

A similar HCO3−–H2O2–CdTe system was tested [69] with thepurpose of attaining a better comprehension of the CL reaction. Theconclusion reached upon the utilisation of different selective scav-engers for 1O2, OH• and •O2− was in agreement with the generalmechanism proposed in the previous work, putting an emphasison the role of the excited form of CdTe as the emitter.

Recently, carbon dots were successfully applied to promote anenhancement of the CL signal obtained with a NaNO2–H2O2 system[70]. This reaction was applied in a FIA system for the determina-tion of nitrite in milk, river water and pond water samples. The CLenhancement was found sensitive to carbon dots, H2SO4, nitrite andH2O2 concentrations. CL intensity linearly increased for nitrite con-centrations between 1 × 10−7 and 1 × 10−5 mol L−1 with a detectionlimit of 5.3 × 10−8 mol L−1. As in previous examples, the CL emissionfor carbon dots was also due to the recombination of the alreadyinjected hole and electron (Table 1).

3.2.2. Chemiluminescence catalystsQDs can be used in CL systems also as catalyst, due to the redox

properties of both the conduction and valence bands. A strong oxi-diser can inject a hole in the valence band of the nanocrystal or areducer can inject an electron in the conduction band (Fig. 4). In this

manner QDs can either promote the formation of short-lived radi-cal species (for instance hydroxyl and superoxide radicals) able tooxidise CL organic dyes, such as luminol, leading to a CL emission, oract themselves individually as radicals. Besides, in the generation

16 C. Frigerio et al. / Analytica Chimica Acta 735 (2012) 9– 22

emis

ota

casbttsarIpS“pitddirswt

TC

Fig. 4. Chemiluminescence

f oxidizing species by photoirradiation of solutions of QDs [10],hese can be considered as catalysts because the formed excitoncts itself as a redox system, producing radical species.

Wang et al. [71] verified that the combined utilisation of TAG-apped CdTe QDs and luminol in the presence of KMnO4 broughtbout a great sensitizing effect on the CL emission, the enhancedignals being a consequence of the accelerated luminol CL inducedy the oxidised species of CdTe. Two interesting applications wereherefore implemented by using both a 3-channel FIA CL sandwich-ype immunoassay to detect human IgG and a 4-channel FIA CLystem to detect catecholamines and other phenols compoundss well as ascorbic acid. For the detection of immunoglobulin aeactor with anti-IgG conjugate with Au nanoparticles was used.n a first step an IgG solution was passed trough the system,ermitting the binding with the anti-IgG/Au conjugate reactor.ubsequently a QDs/anti-IgG solution was inserted forming thesandwich” Au/anti-IgG-IgG-QDs/anti-IgG. Finally the reactor waslaced near the CL detector and a luminol–KMnO4 solution was

nserted and subsequently sensitized by CdTe QDs containing probeo generate enhanced CL in alkaline medium, at a relative intensityirectly related to the bound QDs/anti-IgG. The phenol compoundsetermination was carried out simply by measuring the CL quench-

ng produced when these compounds were mixed in-line with theeagent solutions used to establish the luminol–KMnO4–QDs CLystem. The CL radiation spectra showed that the emitter specie

as luminol, and that QDs played a promoting role, according to

he mechanism displayed in Table 2.

able 2hemiluminescence reactions for QDs as catalyst.

Wang et al. [71] Lum− + −OH + KMnO4 → Lum•−Lum•− + KMnO4 + (CdTe)•+ → Lum*Lum* → Lum + h�

Li et al. [72] IO4− + O2 + 3OH− + ZnS → 2•O2− + •OH + IO3− + H2OLum + −OH → Lum− + H2OZnS + FQs → ZnS–FQsZnS–FQs + IO4− → (ZnS–FQs(•OH)2)(ZnS–FQs(•OH)2) + Lum− → Lum•−Lum•− + •O2 + (ZnS − FQs)ox → Lum*Lum* → Lum + h�

Silvestre et al. [74] CdTe + h� → CdTe(e−Cb

+ h+Vb)e−

Cb+ h+Vb → h�PL

−OH + h+Vb → •OHO2 + e−Cb → O2•−•OH/O2•− + lum → lum*Lum* → lum + h�

sion with QDs as catalysts.

Enhancement by ZnS–MPA QDs of the CL intensity of theluminol–KIO4 system was demonstrated by Li et al. [72] by meansof a 3-channel FIA system. The CL enhancement was attributedto the catalytic effect of ZnS QDs, which presumably fostered theelectron transfer process facilitating radical generation. Using a2 × 10−5 mol L−1 KIO4 solution in alkaline medium, the system wasapplied in the determination of lemofloxacine, a fluoroquinolone,which further enhanced the CL signal. It was proposed that uponKIO4 oxidation and catalysis by ZnS, lemofloxacine was oxidisedand promoted to an excited electronic state, which when relaxingfavoured the formation of more excited 3-aminophthalate, the CLemitting specie (Table 2). The method was applied in the determi-nation of lemofloxacin in capsules.

Another widely used CL system, pyrogallol–H2O2, was stud-ied by Kanwal et al. [73] and coupled with MSA-capped CdTeto accomplish enhanced CL generation. The QDs effect was sizeand concentration dependent. The proposed mechanism was quiteunclear, mainly for what concerned the QDs role, although it wasmanifest that the final emitter was pyrogallol. The system wasapply to Cr3+ determination with good results in terms of sensitivity(6 pmol L−1) and selectivity.

An approach that put in evidence the role of QDs as catalystswas proposed by Silvestre et al. [74]. In this work the capac-ity of QDs to generate oxidizing species in alkaline medium byphotoirradiation (Table 2) was used to degrade organic matterand to develop an indirect chemiluminometric approach for thedetermination of chemical oxygen demand in wastewaters. Byimplementing an automated single interface reaction flow system(SIFA) the authors irradiated MPA-capped CdTe quantum dots gen-erating radical species that degrade distinct organic compoundsused as model compounds. The radicals also oxidised luminol usedas luminogenic probe. The developed approach was applied to theanalysis of certified materials (wastewaters) with good results.

3.2.3. CRETCRET is the acronym of chemiluminescence resonance energy

transfer and could be defined as the non-radiative energy transferbetween a chemiluminescent (CL) donor and a fluorescent acceptorthat, like in FRET, are in close proximity (


Qfl

setsbovie

si5adtQo

flcdqapt

sw

TC

Fig. 5. Chemiluminescence emission in a CRET system.

Ds, exhibiting broad excitation spectra, are therefore well suiteduorescent acceptors to be used in CRET processes (Fig. 5).

In CRET, and differently from what was referred in the previousection, QDs are the finals emitters and not the sensitizers being thisvidence clearly perceptible when monitoring the emitted spec-ra. Also, with CRET it is difficult to understand if the QDs excitedpecies are formed by a resonance energy transmission process ory a redox process, because in all reactions various reducing andxidizing species are present. Another important issue is that inarious situations QDs play a double role in the reaction: actingnitially as sensitizers and after that as final emitters, sometimesven co-emitting with a CL organic dye.

TGA-capped CdTe QDs were applied in a CL Ce(IV)–SO3− FIAystem by Sun et al. [75] for the determination of various reduc-ng organic compounds. The best CL signal was obtained using

× 10−4 mol L−1 Ce(IV) in 0.02 mol L−1 H2SO4 solution. The sizend concentration dependent CL emission increase was alreadyescribed. For the reaction mechanism (Table 3), it was assumedhat a resonance energy transfer occurred between SO2* and CdTeDs that were the final emitters as confirmed by the CL spectra. Allf the tested compounds produced a decrease on the CL emission.

Fortes et al. [76] implemented an automated multi-pumpingow system combining Ce(IV)–SO3−–QDs CL system, using MPA-apped CdTe QDs for the determination of two anti-diabeticrugs in pharmaceutical formulations. Both glipizide and gliclazideuenched the CL emission, probably due to radical scavengingctivity. The results obtained in the analysis of 7 commercial sam-les were satisfactory and in agreement with those obtained withhe reference method.

Effect of the capping-ligand in a K3Fe(CN)6–CdTe system wastudied by Zhang et al. [77] using a FIA system. The reactionas found to be favoured in alkaline solution, like in all other

able 3hemiluminescence reactions for CRET.

Sun et al. [75] Ce(IV) + HSO3− → HSO3• + Ce(III)2HSO3• → S2O62− + 2H+S2O62− → SO42− + SO2*SO2* + CdTe QDs → SO2 + (CdTe QDs)*(CdTe QDs)* → CdTe QDs + h�

Zhang et al. [77] K3Fe(CN)6 + ligand → K4Fe(CN)6 + M*M* + CdTe QDs → M + (CdTe QDs)*(CdTe QDs)* → CdTe QDs + h�

Liu et al. [79] IO4− + H2O2 → •O2− + H2O•O2− → (O2)2*(O2)2* + QD → O2 + (QD)*(QD)* → QD + h�

Zhou et al. [80] ClO− + H2O2 → ClO• + •OH + −OH−OH + H2O2 → HO2− + H2OH2O2 + QDs(catalyst) → 2•OH2•OH + HO2− → •O2−•O2− + •OH → −OH + 1O2ClO• + •OH → HO ClO + −OHHO ClO + −OH → Cl− + H2O + 1O21O2 + 1O2 → (1O2)2* → 2O2 + h�(1O2)2* + QDs → QDs* + 2O2QDs* → QDs + h�

ica Acta 735 (2012) 9– 22 17

circumstances in which a direct QDs oxidation took place. Amidthe three studied capping-ligands mercaptoacetic acid (MAA), glu-tathione (GSH) and cysteine (CYS), GSH provided the highest CLemission. This work demonstrated that the CL reaction could bestrongly affected by the capping nature. Another important find-ing of this study was that the capacity of CdTe QDs to generate CLemission was inversely proportional to the QDs fluorescence. Thisfact was probably due to the low capping efficiency that allowedeasier oxidation or electron injection (also explaining the low quan-tum yield). Again in this case, a direct proportionality between CL,QDs size and concentration, was observed. The proposed reactionmechanism is showed in Table 3. K3Fe(CN)6–CdTe–GSH system wassuccessfully apply to detect 9 different ions (Ca2+, Co2+, Mn2+, Hg2+,Mg2+, Cu2+, Ni2+, Cr3+ and Fe3+).

An interesting example of CRET is the system studied by Kan-wal et al. [78]. MSA-capped CdTe QDs conjugated with IgG wereused to successfully enhance the CL emission of luminol–H2O2and to determine some reducing compound and metal ions. Thereaction showed a higher CL intensity in alkaline pH with a H2O2concentration of 0.1 mol L−1. The mechanism proposed for the CLis a resonance energy transfer between CdTe QDs and luminol that,in this specific case, is the main emitter. A particularly interestingobservation of this work is that if CRET occurs and the QDs are notthe final emitters, the CL intensity enhancement is inversely pro-portional to the QDs size, which is exactly the opposite of whathappens in direct QDs chemiluminescence.

The fist application of MIPs capped QDs in a CL-FIA system wascarried out by Liu et al. [79] aiming at determining 4-nitrophenolin water. In this work the enhancing effect of MIP-capped Mn-doped ZnS QDs in a H2O2–NaIO4 system was assayed revealingan enhancement of the CL emission. The proposed mechanism isdescribed in Table 3 and the emitters were the QDs. Determinationof 4-nitrophenol was accomplished in a first stage by fluorimetryand subsequently, and more extensively, by chemiluminescenceassay. The utilisation of MIPs as capping agents allowed overcom-ing the hindrance of lack of selectivity inherent to QDs, and themethod showed a good sensitivity and broad working analyticalrange.

A comprehensive study on the enhancing effect on NaClO–H2O2CL system promoted by l-cysteine capped Mn-doped ZnS QDs wascarried out by Zhou et al. [80]. The enhancing effect was studiedin a batch assay and by using a FIA system. The CL increment wasfound dependent on the QDs concentration and more effective forMn doped QDs, probably because the incorporated metal acted ascatalyst in the CL reaction. Of great significance was also the depen-dence of CL on the capping “density”. In fact, up to 4% (by mass) ofl-cysteine capping the CL intensity showed an increase. However,for higher values it decreased probably because a high capping den-sity can turn difficult the interaction of active species with the QDscore. The reaction (Table 3) was studied in detail by EPR using NaN3as quencher. The obtained results revealed that QDs participated inthe reaction in two distinct ways: initially they acted as catalysts inthe formation of •OH and •O2− and in a second stage they acceptedenergy from (1O2−)2* and acted as final emitters.

QDs have a great potential of application in CRET. One of thenon-explored possibilities will, for instance, take advantage of thecombined utilisation of multiple differently-sized QDs, emitting atdifferent wavelengths and labelling different analytes, which uponreceiving energy from the same donor could be detected simulta-neously.

3.2.4. Electrogenerated chemiluminescence (ECL)

Electrochemiluminescence or electrogenerated chemilumines-

cence is a form of chemiluminescence in which the light emittingreaction is preceded by an electrochemical reaction. This involvesthe generation of reactive species, at an electrode surface, that

18 C. Frigerio et al. / Analytica Chim

uwdtrenibdepQidQsEaliofaeltearcotso

ioa

Fig. 6. Electrochemiluminescence of QDs.

ndergo electron-transfer (redox) reactions to form excited stateshich emit light upon an energy-relaxation process (Fig. 6). ECLoes not only exhibit the advantages of chemiluminescence detec-ion, like the high sensitivity and wide concentrations workingange, but it assures alongside specific advantages, such as annhanced selectivity, due to the absence of a background sig-al arising from the emission of excited species not specifically

nvolved in the detection process, extended analytical applicationy means of analyte electrochemical modification, and improvedetection spatial resolution as the time and position of the lightmitting reaction can be controlled. Apart from their ability toarticipate in photoluminescent and chemiluminescent processes,Ds were found to generate light under an applied potential,

nitially in non-aqueous [81], and more recently in aqueous con-itions [82]. In spite of the great success already achieved inDs-based ECL systems, the fundamental understanding of these

ystems has not been, until now, entirely clarified. Generally, theCL mechanisms of QDs are considered as one of three main types:nnihilation, coreactant ECL and cathodic luminescence. Annihi-ation ECL does not need additional reagents for emission andnvolves the formation of cation and anion QDs radicals through-ut electron transfer between oxidised and reduced species. Theormed oppositely charged radicals should be chemically stablend maintain they charged states until they interact, annihilatingach other, to generate the excited species that emit light. Cathodicuminescence is a cathodic light emission observed during elec-rolysis of aqueous electrolyte solutions that originates from hotlectron injection. Coreactant ECL is the most common approachnd relies on reaction between a luminophore and an additionaleagent (coreactant), which usually enhances luminescence effi-iency, under a unidirectional applied potential. Electron transferccurs only between electrochemically-generated QDs species andhe coreactant. Several compounds have been used as coreactantsuch as tri-n-propylamine, oxalate, peroxydisulfate, hydrogen per-xide, etc.

Despite the great number of ECL analytical applications resort-ng to QDs nanotechnology [83,84], only a very restrict numberr works, all published during 2011, are based on automatedpproaches. Wan et al. [85] have proposed a flow injection

ica Acta 735 (2012) 9– 22

electrochemiluminescence (FI-ECL) sensor, by using TGA-cappedCdTe quantum dots, for the determination of durabolin in foodsamples. The QDs were layer-by-layer self-assembled on ITO glassand the slide was packed in a homemade flow cell for ECLmeasurements. The number of assembled layers affected systemperformance being the highest ECL intensity observed for 4 layers.Durabolin acted as coreactant increasing the ECL emission linearlyfor concentration values between 1.0 × 10−8 and 1.0 × 10−5 g mL−1.The detection limit was 2.5 × 10−9 mL−1. The developed FI-ECLwas cost-effective, and provided high sensitivity and high sam-ple throughput. Another flow injection analysis ECL approach wasdeveloped by Jinjin et al. [86] for the high sensitive determina-tion of dopamine in cerebro-spinal fluid. The authors constructed anano-litre (300 nL) sized flow-cell containing the ECL working elec-trode obtained by immobilisation of a composite of TGA-cappedCdTe QDs, carbon nanotubes and chitosan on ITO glass slides. Tri-ethylamine was used as coreactant, increasing the ECL emission upto a concentration of 40 mM. During sample measurement an ECLquenching effect was verified due to the energy transfer from theexcited CdTe QDs to the oxidised products of dopamine. The work-ing analytical concentrations ranged from 10 pM to 4 nM with adetection limit of 3.6 pM. The developed flow-cell is expected to beused as a component of microfluidic devices.

3.3. Liquid chromatography

The well known liquid chromatography is in essence a physicalseparation technique for mixtures resolution relying on the distri-bution of the analyte between a stationary phase, typically a solid,and a liquid mobile phase. The type of interactions between sam-ple and stationary/mobile phases allows a deeper differentiation(ion-exchange, adsorption, partition, affinity, size exclusion, etc.).Nowadays, high performance liquid chromatography is the premiertechnique for modern chemical analysis and related applications.In a single step process it can separate a mixture into its individualcomponents and simultaneously provide a quantitative estimate ofeach constituent. The analytical efficiency of a liquid chromatogra-phy method depends not only on adequate separation but also onsuitable inline detection and quantification. The most common LCdetectors are based on UV/vis or fluorescence spectroscopy, elec-trochemical measurements or mass spectrometry.

An ideal LC detector should provide high sensitivity, fastresponse, wide linear dynamic range and low noise level. In thisregard, fluorescence has been shown to be extremely useful asa detection process offering not only selectivity but also some ofthe highest sensitivities available in LC along with low instrumentinstability due to the very low light background. The greatest weak-ness is that relatively few compounds fluoresce in a practical rangeof wavelengths. In this context, QDs can be a valuable tool in LC asthey could act as labels enabling the signalling and identification ofthe analyte similarly to what happens in capillary electrophoresis[105–110]. Moreover, upon bioconjugation of QDs with a varietyof biomolecules such as proteins and immunoglobulins it is pos-sible to attain the specificity and high sensitivity needed for thetargeted analyte detection being as well an expeditious strategyfor multiparametric array analysis by means of multiplexed analytedetections.

Despite the great analytical potential, in the past no typicalliquid chromatographic method has used QDs as labels for com-pound detection although a few size exclusion chromatography(SEC) approaches have been proposed to separate QDs and QDs-conjugates from precursors. In the first work published in this

field [87] HPSEC was used to separate the products of conju-gation of cohesin/dockerin protein polymers with CdSe/ZnS. Itwas possible to separate two major fractions of bioconjugatespecies: a first eluting peak representing the highest molecular

Chim

mtcad

factmt

itmeiadaiI

3

otBstedmdtp

arstfl

tQ

--

--

nsCcaia

mee


ass cohesin/dockerin-QDs clusters and a later eluting peak con-aining single QDs containing protein conjugate. This work alsoonfirmed the occurrence of strong interactions between self-ssembled, recombinant protein polymers and CdSe/ZnS quantumots.

Wang et al. [88] proposed a SEC system for isolation of QDsrom free polymers after conjugation reactions and to quantify themount of polymer bound to the QDs surface. Separation of QDsonjugated from free conjugation polymer was easily achieved ashe former were eluted much faster, while the number of the poly-

eric ligands bound to each QDs was quantified by determininghe linked polymer using the nanocrystals as fluorescent tracers.

A SEC system for separation of different size QDs was alsomplemented by applying a QDs repelling surface of a concen-rated poly(methyl methacrylate) (PMMA) brush onto a silica

onolithic column [89]. In what could be considered the onlyffective analytical application, Chouhan et al. [90], without resort-ng to conventional chromatographic equipment, implemented

fluoroimmunochromatographic technique for methylparathionetermination at picogram level by using thiol-stabilised CdTe QDss fluorescent probes in a flow-injection analysis system compris-ng an immunoreactor column packed with immobilised anti-MPgY antibodies.

.4. Capillary electrophoresis

Capillary electrophoresis (CE) is a separation technique basedn the different analytes mobility under the influence of a spa-ially uniform electric field due to their different charge/size ratios.riefly, in CE a narrow fused capillary filled with an electrolyteolution is used to separate the charged analytes. The ends ofhe capillary are immersed into vials containing the BGE and thelectrodes. The sample solution is inserted in the capillary withifferent strategies: by pressure, gravity or an applied voltage. Byeans of the application of a high voltage the analyte forms zones

ue to the different electrophoretic mobility and migrate towardhe outlet side. The use of an in-line detector, prior to the outlet,ermitted the detection of the analyte.

Due to their electric surface charge QDs are well suited forpplication in CE explaining the motifs behind the emergence, inecent years, of CE as a powerful tool for characterizing [91–94] andeparating QDs [95–100], and also to take advantage of the pho-oluminescent properties of QDs by using these nanoparticles asuorescent labels in distinct analytical methodologies [105–109].

Among the different CE operation modes, characterised by dis-inct separation mechanism that were used in combination withDs nanotechnology, the most relevant are:

capillary zone electrophoresis (CZE) [92–96,98,104,107,109,111]. micellar electrokinetic chromatography (MEKC)[97,102,103,108,110].

capillary gel electrophoresis (CGE) [91]. microchip electrophoresis (MCE) [105,106].

Being one of the most powerful separation techniques withoteworthy advantages in terms of simplicity, low sample con-umption, high separation efficiency and rapidity, in early works,E techniques where mostly used for characterisation, regardingomposition and size distribution, and separation of QDs [91–98],s well as to typify their potential ecotoxicity as a consequence of anncreased environmental prevalence of these nanomaterials due to

more widespread utilisation and, hence, waste disposal [99,100].

The pioneering work on the separation of CdTe–MPA QDs by

eans of a CGE system was carried out by Song et al. [91] bymploying a laser induced fluorescence (LIF) detector. Using lin-ar polyacrylamide (PAA) as sieving medium, a good resolution

ica Acta 735 (2012) 9– 22 19

for different size QDs was achieved. Of great importance was thedescribed effect of peak broadening, a consequence of the highpolydispersivity that acted as a noteworthy limitation when deal-ing with close size QDs separation, and the enhanced PL emissionobserved at higher pH. CZE technique was used for the first timeby Pyell [94] to separate and determine the size of CdSe/ZnS/SiO2QDs with distinct electrophoretic mobilities depending not onlyon size but also on pH and ionic strength of the separation elec-trolyte [101]. The obtained results matched the size distributiondetermined by TEM and showed that the electrophoretic mobilitywas independent on the applied electric field force. Similar evi-dence was found by Li et al. [98] by using PEG (4%) as sievingmedium. The separation of three different sized core/shell CdSe/ZnSQds was achieved confirming that the electrophoretic mobility ofthe analysed nanocrystals was inversely proportional to their size.More recently micellar electrokinetic chromatography was used byCarrillo-Carrión et al. [102] achieving the highest resolution levelfor QDs (0.5 nm).

In the previous works the focus was mainly set on the sep-aration of QDs with the same composition but with differentsize. Since the electrophoretic mobility can also change with thesurface charge another important variable that should be takeninto consideration, along with the size, is the nature of the cap-ping/coating/bioconjugation of the QDs. Huang et al. [92] applieda CE method to separate CdTe QDs capped with MPA from thoseconjugated with bovine serum albumine (BSA). The importanceof the net charge due to the presence of different capping agentsin CZE was highlighted in the work of Pereira et al. [93] andmore extensively studied by Zhang et al. [97] for equally sizedCdTe QDs capped with MPA, MAA and GSH by MEKC. The micel-lar environment was created upon addiction of SDS and permittedan even higher selectivity due to the different interactions of theassayed organic cappings. This same principle is the basis of thework of Oszwałdowski et al. [103], which have assessed a micellarplug containing QDs and TX-100 or DOSS by CZE. The devel-oped methodology increased the separation ratio and permitted apreconcentration of the QDs. It also put in evidence that the distri-bution of the nanostructures between a micellar and micellar freezone is controlled by the QDs affinity and is the decisive factor inthe separation process.

Vicente and Colón [95] successfully developed a method for sep-aration of different CdSe/ZnS conjugates QDs using CE with PEO assieving agent, confirming the results of the previous works. Theystudied the effect of the reaction of two bioconjugated QDs thatresulted in the appearance of a third peak but of difficult resolution.

One of the biggest problems for CE resolution of bioconju-gates is the heterogeneity of the final products. Pereira and Lai[96] were the first describing the peak broadening arising fromQDs bioconjugation (selective or non-selective) with proteinsand immunoglobulins. They explained the higher polydisper-sity observed as a likely consequence of the variable number ofbiomolecules that could conjugate a single QDs and the existenceof multiple linkage sites on these molecules participating in the bio-conjugation process. These results were confirmed by Liskova et al.[104] in a systematic study of the products obtained by differentbioconjugation methods using CZE and achieving a great resolutionwithout sieving media.

In the last two years EC was coupled with QDs to developnews analytical methods combining the separation potentiality ofEC with the optical properties and reactivity of QDs. Chen et al.[105] proposed an ultrasensitive detection method for the deter-mination of 7-aminoclonazepam (7-ACZP) in urine by using a

microfluidic chip–based immunoassay with laser induced fluores-cence (LIF) detection. Denaturated bovine serum albumin (dBSA)coated CdTe–TGA QDs were conjugated with anti-7-ACZP anti-body. 7-ACZP and 7-ACZP-OVA (ovalbumin) compete to form

2 Chim

igdTr

wlwoct

mQeeact

ttsCi8

mDQatrwUgt

ntP

aCii

4

dfiialomrteabcu


mmunological complexes with the dBSA-coated CdTe QDs conju-ated antibody (Ab). Separation was achieved based on the mobilityifference between the antibody and antibody-antigen complex.he utilisation of QDs as biomarkers enabled the detection of drugesidues at a pictogram level.

A CRET-based method coupled to microchip electroforesisas the first work published with this purpose [106]. In a

uminol–NaBrO–QDs chemiluminescence system the nanocrystalsere pointed out as final acceptor and emitter specie. The devel-

ped approach was successfully applied in the detection of variousompounds and extensively studied for the simultaneous separa-ion and detection of amino acids in a single cell.

An EC-CL system was also used for the separation and deter-ination of dopamine and epinephrine (Zhao et al.) [107]. ADs–luminol–H2O2 system was added to the running buffernabling the attainment of a very high CL intensity. The two cat-cholamines, which are recognised radical scavengers, promoted

pronounced quenching of the CL signal due to inhibition of theatalytic process. The method was applied in the determination ofhe neurotransmitters in human urine samples.

Chen and Fung [108] proposed a method for the determina-ion of organophosphorus pesticides. By exploiting the capacity ofhese compounds to interact with QDs they optimise a MEKC–LIFystem for their separation and determination in vegetables. ThedTe/CdS QDs were immobilised onto the inside capillary surface

n the LIF window and showed good stability to borate buffer (pH–10), methanol 5% and ageing.

CdTe capped with l-cysteine or glutathione were linked toolecular beacons (MBs) to detect dual and single base mutation inNA by EC [109]. The linkage with MBs yielded a quenching of theDs emission due to a FRET mechanism between the nanocrystalnd the quencher present on the MBs. When the target DNA linkedhe MBs a rise in the QDs fluorescence was observed due to inter-uption of the FRET mechanism. A decrease in the migration timeas also recognised due to modification of the MBs conformation.sing two QDs–MBs conjugates it was possible to discriminate sin-le base from dual base mutations because the former do not linko the MBs and thus do not alter the migration time.

CdTe–MPA QDs were applied as buffer additives for the determi-ation of acrylamide by MECK with LIF detection [110]. In this casehe interaction with QDs produced a quenching of the backgroundL proportional to acrylamide concentration.

The fluorescence and mobility effect resulting from the inter-ction of fluvic acids with QDCOOH and QDNH2 were studied byeliz et al. [111]. A decreasing trend in the electrophoretic mobil-

ty and a decrease in the PL emission were observed in presence ofncreasing concentrations of fluvic acids.

. Prospects and trends

The analytical relevance of quantum dots, although pre-ominantly based on their utilisation as luminescent labels, isnding new fields of application involving not only the design-

ng of sensitive and selective chemosensors but also takingdvantage of their facility of immobilisation and/or encapsu-ation in solid supports and their high reactivity for assortednline reactional schemes implementation. Coupled to auto-ated analytical methodologies QDs could be tailored for analyte

ecognition and separation opening up the possibility of detec-ion of hazardous chemicals, food contaminants, xenobiotics,nvironmental pollutants, pharmaceuticals, etc., which are usu-

lly presented at very low levels or could not be determinedy conventional procedures, demanding, for instance, a pre-oncentration or an interference removal step, or could besed to promote fast redox reactions, or to yield metastable

ica Acta 735 (2012) 9– 22

or short-lived species that require prompt measurement.On the other hand, in terms of a more environmental friendlychemistry, automated methodologies not only minimise reagentsconsumption and waste generation, which could be a valuableasset when semiconductor materials involving heavy metalsare concerned, but could also prevent operators to come intocontact with toxic materials. Additionally, automated flow-basedmethodologies are particularly suitable to implement complexreactional schemes or multiparametric determinations, conceptsthat are also easily associated to quantum dots nanotechnology.

4.1. Solid phase reactors

In the prospect of green chemistry and environmentalsafety, reactions resorting to solid state reagents are gaining awidespread support [112]. Either due to chemical instability ofwet reagents, renewed utilisation, consumption reduction, sim-plicity of reaction implementation or online reagent derivatisation,solid phase reactors, most of the times applied in redox reac-tions, have demonstrated noteworthy advantageous performance.Solid-phase reactors can be easily accommodated in any of theavailable automated flow-based and related techniques for redoxderivatisation, for chromatographic separations, for sorptive pre-concentration and/or matrix removal, etc., facilitating sampleprocessing, integration of multi-parametric analysis, reduction oftime of analysis and of operators’ intervention, sample resolution,elimination of carry-over problems, etc. In this regard, the numer-ous possibilities of QDs conjugation, encapsulation or coatinginvolving either silica, polymeric matrices, glass spheres or sheetsand porous materials, provides a panoply of solid-phase strategiesthat could be used for instance to implement pre-concentrationand interference removal units, low-pressure chromatographiccolumns, catalytic reactors, photoreactors, photoinduced chemi-luminescence reactors and electrogenerated chemiluminescenceunits. QDs with different sizes could be embedded into polymericmicrobeads for multiplexed detection. Functionalised QDs–SiO2beads could be applied in-line solid-phase extraction (SPE) anddetection. QDs could be conjugated with molecularly imprintedpolymers for target recognition or selective heterogeneous cataly-sis mimicking enzyme reactors at a lower cost.

4.2. Magnetic nanocomposites

Multimodal probes combining luminescent and superparamag-netic properties have been recently proposed for in vivo imagingapplications facilitating disease diagnostic and surveillance by pro-viding both optical and magnetic monitoring. In terms of analyticalchemistry, by conjugating luminescent nanocrystals with magneticnanoparticles, such as Fe3O4, it is possible to combine event oranalyte detection with a rapid and efficient separation upon theaction of a magnetic field. Labelled molecules could be thereforeisolated or pre-concentrated providing enhanced detection and theremoval of interferences. Aside from the biological applicationsalready under development for instance for targeting drug-deliverysystems or cells isolation, magnetic luminescent nanocompositescould be assured appropriate surface multifunctionality and bemagnetically manipulated to be used as bifunctional probes able toselectively bind the target analyte and to retain it at the detector’sflow cell for measurement, for instance in terms of fluorescenceenhancement or quenching. Deactivation of the applied magneticfield would subsequently enable discharge to waste or to an in-lineautomated recycling unit that could renew the magnetic nanocom-

posite surface assuring their continuous utilisation at the same asthe flow cell is emptied for a new sample analysis. These multifunc-tional nanoparticles could be coated with a silica shell, a polymericcoating or a bilipid layer or be encapsulated within a silica or

Chim

asuaudidwa

4

ttoesaeawcspr

ubatfrtoceswftsopo(

idiarut

A

Pt

R


polymer matrix [113] for improved luminescence, robustness,tability and versatility of functionalisation. This would enable theirse as solid supports for instance in an in-line flow column, theirpplication in different reaction conditions or media, or even theirtilisation as in-line heating up mechanism when the reactionemands increased temperature. In a more futuristic perspective

t is even possible to envisage their utilisation as magnetic han-lers for positioning, manipulation and guidance of the sample zoneithin the flow manifold, the establishment of physical barriers,

ppliance of mechanical forces, etc., without direct contact.

.3. Bead injection

Bead injection is an automated flow technique that seems par-icularly suitable for the combined utilisation with quantum dotsechnology. In review, bead injection is a FIA technique relyingn the injection of microspheres (microbeads), made of differ-nt materials, into a conduit where functional groups on beadurfaces reacted with the sample analyte. Microbeads are usu-lly trapped within the detector’s flow cell or the analyte may beluted downstream for detection from the microbeads retained at

selected location [114,115]. This flow technique exhibited note-orthy advantages namely high sensitivity, since target molecules

an be accumulated on microbead surfaces from an initially largeample volume and detected in situ without need for elution, highrecision of bead delivery, absence of carryover and fast automatedenewal of the reactive solid phase.

A quick literature survey shows that some of the strategiessed to functionalise the polysaccharide microspheres used inead injection are also used to functionalise QDs surface, whichnticipates a complementary or a synergetic effect between thesewo approaches. Furthermore, QDs exhibit a more versatile sur-ace chemistry providing the means for implementation of a wideange of targeting strategies and analyte recognition functionali-ies which could expand the scope of flow analysis methods basedn bead injection widening as well the detection techniques thatould be applied to. As an example, polymer based fluorescence-ncoded microspheres have been prepared from distinct polymersuch as polystyrene, N-isopropylacrylamide and 4-vinylpyridine asell as silica colloidal crystal beads, which were tagged with QDs

or assorted analyte determination. On the other hand, the utilisa-ion of a single bead (either silica, polymeric or magnetic) as a solidupport for successive layers of the same QDs or the encapsulationf QDs into polymeric materials or by attaching QDs to beads couldrovide a signal enhancement (multi-layer), while the assemblagef different QDs on the same bead could enable multiplexed signalsmulti-detection).

In a very brief conclusion, quantum dots are incredibly promis-ng nanoparticles with a huge potential of application in the veryistinct fields of analytical chemistry, combining great functional-

sation simplicity with an enormous operational versatility. Theyre not perfect tools but some of the shortcomings that are usuallyelated with their conventional application could be overcome bysing some of the automation tools, mostly flow-based or relatedechniques, available to the analytical chemists.

cknowledgements

Authors are grateful for the financial support of the projectTDC/QUI-QUI/105514/2008 under COMPETE/QREN/FEDER. Chris-ian Frigerio thanks FCT for the PhD grant SFRH/BD/47651/2008.

eferences

[1] A.L. Rogach (Ed.), Semiconductor Nanocrystal Quantum Dots, Verlag, Wien,2008.

ica Acta 735 (2012) 9– 22 21

[2] C.B. Murray, D.J. Norris, M.G. Bawendi, J. Am. Chem. Soc. 115 (1993)8706–8715.

[3] A.P. Alivasatos, J. Phys. Chem. 100 (1996) 13226–13239.[4] S.K. Poznyak, D.V. Talapin, E.V. Schevchenko, H. Weller, Nano Lett. 4 (2004)

693–698.[5] M. Bruchez, M. Moronne, P. Gin, S. Weiss, A.P. Alivisatos, Science 281 (1998)

2013–2016.[6] U. Resch-Genger, M. Grabolle, S. Cavaliere-Jaricot, R. Nitschke, T. Nann, Nat.

Methods 9 (2008) 763–775.[7] W.C.W. Chan, S. Nie, Science 281 (1998) 2016–2018.[8] M.A. Hines, P. Guyot-Sionnest, J. Phys. Chem. B 102 (1998) 3655–3657.[9] M.A. Hines, P. Guyot-Sionnest, J. Phys. Chem. 100 (1996) 468–471.

[10] O.I. Mi ci c, C.J. Curtis, K.M. Jones, J.R. Sprague, A.J. Nozik, J. Phys. Chem. 98(1994) 4966–4969.

[11] W.W. Yu, Y.A. Wang, X. Peng, Chem. Mater. 15 (2003) 4300–4308.[12] H. Kuang, Y. Zhao, W. Ma, L.G. Xu, L.B. Wang, C.L. Xu, TRAC – Trends Anal.

Chem. 30 (2011) 1620–1636.[13] N. Gaponik, A.L. Rogach, Phys. Chem. Chem. Phys. 31 (2010) 8685–8693.[14] F.M. Raymo, I. Yildiz, Phys. Chem. Chem. Phys. 17 (2007) 2036–2043.[15] W.R. Algar, A.J. Tavares, U.J. Krull, Anal. Chim. Acta 673 (2008) 1–25.[16] T. Trindade, P. O’Brien, N.L. Pickett, Chem. Mater. 13 (2001) 3843–3858.[17] Z.A. Peng, X. Peng, J. Am. Chem. Soc. 123 (2001) 183–184.[18] B.O. Dabbousi, J. Rodriguez-Viejo, F.V. Mikulec, J.R. Heine, H. Mattoussi, R.

Ober, K.F. Jensen, M.G. Bawendi, J. Phys. Chem. B 101 (1997) 9463–9475.[19] X. Peng, M.C. Schlamp, A.V. Kadavanich, A.P. Alivisatos, J. Am. Chem. Soc. 119

(1997) 7019–7029.[20] D.V. Talapin, A.L. Rogach, A. Kornowski, M. Haase, H. Weller, Nano Lett. 1

(2001) 207–211.[21] M. Gao, S. Kirstein, H. Möhwald, A.L. Rogach, A. Kornowski, A. Eychmüller, H.

Weller, J. Phys. Chem. B 102 (1998) 8360–8363.[22] A.L. Rogach, L. Katsikas, A. Kornowski, D. Su, A. Eychmueller, H. Weller, Ber.

Bunsen-Ges Phys. Chem. 100 (1996) 1772–1778.[23] A.J. Nozik, F. Williams, M.T. Nenadovic, T. Rajh, O.I. Micic, J. Phys. Chem. 89

(1985) 397–399.[24] M.X. Wu, H. Liu, J. Liu, K.N. Haley, J.A. Treadway, J.P. Larson, N. Ge, F. Peale,

M.P. Bruchez, Nat. Biotechnol. 21 (2003) 41–46.[25] A. Henglein, Ber. Bunsen-Ges Phys. Chem. 86 (1982) 301–305.[26] D. Gerion, F. Pinaud, S.C. Williams, W.J. Parak, D. Zanchet, S. Weiss, A.P.

Alivisatos, J. Phys. Chem. B 105 (2001) 8861–8871.[27] D.L. Giokas, A.G. Vlessidis, G.Z. Tsogas, N.P. Evmiridis, TRAC – Trends Anal.

Chem. 29 (2010) 1113–1126.[28] J. Ruzicka, H. Hansen, Anal. Chim. Acta 78 (1975) 145–157.[29] J. Ruzicka, G.D. Marshall, Anal. Chim. Acta 237 (1990) 329–343.[30] B.F. Reis, M.F. Giné, E.A.G. Zagatto, J.L.F.C. Lima, R.A.S. Lapa, Anal. Chim. Acta

293 (1994) 129–138.[31] V. Cerda, J.M. Estela, R. Forteza, A. Cladera, E. Becerra, P. Altimira, P. Sitjar,

Talanta 50 (1999) 695–705.[32] R.A.S. Lapa, J.L.F.C. Lima, B.F. Reis, J.L.M. Santos, E.A.G. Zagatto, Anal. Chim.

Acta 466 (2002) 125–132.[33] M.F.T. Ribeiro, J.L.M. Santos, J.L.F.C. Lima, A.C.B. Dias, E.A.G. Zagatto, Talanta

68 (2005) 351–358.[34] M.T. Fernandez-Arguelles, W.J. Jin, J.M. Costa-Fernandez, R. Pereiro, A. Sanz-

Medel, Anal. Chim. Acta 549 (2005) 20–25.[35] J.H. Wang, H.Q. Wang, H.L. Zhang, X.Q. Li, X.F. Hua, Y.C. Cao, Z.L. Huang, Y.D.

Zhao, Anal. Bioanal. Chem.

Date post:	31-Jan-2021
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

Analytica Chimica ActaC. Frigerio et al. / Analytica Chimica Acta 735 (2012) 9–22 11 João L.M....

Documents