β-Cyclodextrin coated CdSe/ZnS quantum dots for vanillin sensoring in food samples
Transcript of β-Cyclodextrin coated CdSe/ZnS quantum dots for vanillin sensoring in food samples
Author's Accepted Manuscript
β-Cyclodextrin coated CdSe/ZnS quantumdots FOR VANILLIN SENSORING IN FOODSAMPLES
Gema M. Durán, Ana M. Contento, Ángel Ríos
PII: S0039-9140(14)00668-7DOI: http://dx.doi.org/10.1016/j.talanta.2014.07.100Reference: TAL15009
To appear in: Talanta
Received date: 21 May 2014Revised date: 28 July 2014Accepted date: 31 July 2014
Cite this article as: Gema M. Durán, Ana M. Contento, Ángel Ríos, β-Cyclodextrin coated CdSe/ZnS quantum dots FOR VANILLIN SENSORING INFOOD SAMPLES, Talanta, http://dx.doi.org/10.1016/j.talanta.2014.07.100
This is a PDF file of an unedited manuscript that has been accepted forpublication. As a service to our customers we are providing this early version ofthe manuscript. The manuscript will undergo copyediting, typesetting, andreview of the resulting galley proof before it is published in its final citable form.Please note that during the production process errors may be discovered whichcould affect the content, and all legal disclaimers that apply to the journalpertain.
www.elsevier.com/locate/talanta
1��
�-CYCLODEXTRIN COATED CdSe/ZnS QUANTUM DOTS
FOR VANILLIN SENSORING IN FOOD SAMPLES
�Gema�M.�Durán1,�2,�Ana�M.�Contento1,�Ángel�Ríos1*�
�1Department�of�Analytical�Chemistry�and�Food�Technology,�University�of�Castilla��La�Mancha.��2IRICA�(Regional�Institute�of�Applied�Scientific�Research).�Avenida�Camilo�José�Cela,�s/n.�13071,�Ciudad�Real,�Spain�*E�mail:�[email protected]�
�
Abstract
An� optical� sensor� for� vanillin� in� food� samples� using� CdSe/ZnS� quantum� dots� (QDs)�
modified�with���cyclodextrin�(��CD)�was�developed.�This�vanillin�sensor�is�based�on�the�
selective�host�guest�interaction�between�vanillin�and���cyclodextrin.�The�procedure�for�
the� synthesis� of� ��cyclodextrin�CdSe/ZnS� (��CD�CdSe/ZnS�QDs)� complex� was�
optimized,� and� its� fluorescent� characteristics� are� reported.� It� was� found� that� the�
interaction�between�vanillin�and���CD�CdSe/ZnS�QDs�complex�produced�the�quenching�
of� the� original� fluorescence� of� ��CD�CdSe/ZnS�QDs� according� to� the� Stern�Volmer�
equation.� The� mechanism� of� the� interaction� is� discussed.� The� analytical� potential� of�
this�sensoring�system�was�demonstrated�by�the�determination�of�vanillin� in�synthetic�
and� food�samples.�The�method�was�selective� for�vanillin,�with�a� limit�of�detection�of�
0.99� μg� mL�1,� and� a� reproducibility� of� 4.1%� in� terms� of� relative� standard� deviation�
(1.2%�under�repeatability�conditions).�Recovery�values�were�in�the�90�105%�range�for�
food�samples.�
�
Keywords:� CdSe/ZnS� quantum� dots,� ��cyclodextrin;� Functionalization;� Fluorescence;�
Vanillin�sensoring;�Food�samples.�
�
�
�
�
�
2��
�
�
1. Introduction
The�use�of�quantum�dots�(QDs)�for�the�development�of�sensors�is�one�of�the�most�
developing�fields�of�nanotechnology�so�far.�Their�fluorescence�efficiency�is�sensitive�to�
different�compounds�on�their�surface.�Therefore,�molecular�recognition�at�the�surface�
of� QDs� can� be� utilized� in� the� development� of� fluorescent�based� sensors.� For� this�
purpose,� several� strategies� for� surface�modified� QDs� have� been� employed.� Thiol�
ligands� were� used� to� modify� quantum� dots� such� as� L�cysteine� or� D�cysteine� for�
carnitine�enantiomers�determination�[1],�mercaptoacetic�acid�for�L�cysteine�detection�
[2],�3�mercaptopropionic�acid�for�detection�and�quantification�of�paraquat�[3],�among�
other�applications.�Other�surface�modified�quantum�dots,�such�as�ionic�liquid�modified�
CdSe/ZnS� QDs� for� trimethylamine� fluorimetric� determination� [4],� silica�coated�
CdSe/ZnS�nanoparticles�for�Cu2+�detection�[5],�calix[8]arene�coated�CdSe/ZnS�quantum�
dots�as�C60�nanosensor�[6],�have�also�been�used.��
Cyclodextrins�(CDs)�are�considered�one�of�the�best�host�molecules.���Cyclodextrins�
are� cyclic� receptors� consisting� of� seven� glucose� units� linked� one� to� another� by� 1�4�
glycoside� bonds.� Their� cavity�shaped� cyclic� phenol� molecules� are� capable� to� forming�
host�guest�complexes�with�a�variety�of�organic�molecules.�The�hydrophobic�cavities�of�
cyclodextrins� were� used� to� develop� different� sensors� [7,� 8]� and� separation� matrices�
[9].�Thus,�CDs�have�attracted�great�interest�in�supramolecular�chemistry.�Cyclodextrins�
coating�ensures�the�high�emission�efficiency�and�the�smaller�size�of�QDs�and�provides�
selectivity.� Therefore,� several� methods� for� the� preparation� of� highly� fluorescent� and�
stable� CdSe/ZnS� quantum� dots,� using� cyclodextrins� as� surface� coating� agents,� have�
recently� reported.� Optical� sensing� and� chiroselective� sensing� of� different� substrates�
were� reported� using� ��CD� functionalized� CdSe/ZnS� QDs� based� on� a� fluorescence�
resonance� transfer� (FRET)� or� an� energy� transfer� mechanism� [10].� ��CD�coated�
CdSe/ZnS� QDs� was� also� applied� as� enantioselective� fluorescent� sensors� for� amino�
acids,�such�as�tyrosine�and�a�significant�fluorescence�enhancement�was�observed�[11],�
which�can�be�used�for�the�optical�detection�of�phenol�pollutants�in�water�samples�[12].�
Other���CD�modified�CdTe�QDs�were�also�used�as�a�nanosensor�for�acetylsalicylic�acid�
3��
and� its� metabolites� [13],� as� fluorescent� probes� for� polycyclic� aromatic� hydrocarbons�
(PHAs)� [14],� and� ��CD� modified� CdSe� QDs� as� a� recognition� system� for� tyrosine�
enantiomers�[15].�
Vanillin� is� one� of� the� most� popular� flavoring� substances� and� it� is� widely� used� in�
food,�beverages,�perfumery�and�pharmaceutical� industry.�Natural�vanillin� is�obtained�
from�vanilla�pods,�through�a�long�and�expensive�process.�Furthermore,�natural�vanillin�
obtained�in�this�way�can�supply�less�than�1%�of�the�market�demand.�Therefore,�most�
of� the�vanillin�employed� is� synthesised� through�chemical�processes� from�eugenol� (4�
allyl�2�methoxyphenol),� guaiacol� (2�methoxyphenol)� or� lignin.� The�chemical� synthesis�
leads�to�a�cheaper�vanillin,�but�of�lower�quality�with�a�wide�variety�of�complex�matrices�
that�need�selective�and�sensitive�clean�up�procedures�for�its�extraction�and/or�analysis�
[16].�The�yield�and�purification�of�vanillin�(a�biomolecule�relevant�for�several�purposes)�
are� still� of� major� interest.� Different� methods� for� determination� of� vanillin� in� several�
samples� have� been� developed.� Many� of� these� methods� involve� electrochemical�
detection�with�several�types�of�electrodes�[17�20].�Other�methods� include�the�use�of�
supported� liquid�membranes�with�amperometric� [21]� or�piezoelectric� [22]�detection.�
Spectrophotometric� [23]� detection,� liquid� cromatography� with� mass� spectrometry�
detection�[24],�capillary�electrophoresis�[25]�or�gas�chromatography�[26]�has�also�been�
used� for� determining� vanillin.� However,� these� techniques� usually� need� complicated�
sample� pretreatment.� Nowadays,� as� a� useful� analytical� technique,� fluorescent�
detection�has�been�extensively�employed�with�high�sensitivity�and�selectivity.�To�our�
knowledge,� the� use� of� CDs� functionalized� QDs� as� selective� probes� for� fluorescent�
determination� of� flavoring� is� almost� unexplored.� In� this� paper,� it� is� reported� the�
synthesis� of� water� soluble� and� stable� semiconductor� CdSe/ZnS� QDs� using� ��CD� as�
surface�coating�agent�by�a�very�simple�sonochemical�method.�Its�potential�application�
as� a� selective� fluorescent� sensor� for� vanillin� in� several� samples� has� also� been�
investigated,�obtaining�satisfactory�results�for�its�determination�in�food�samples.�
�
�
�
4��
2. Experimental 2.1. Reagents
All�chemical�reagents�were�obtained�from�commercial�sources�of�analytical�grade�
and�were�used�as�received�without�further�purification.�Cadmium�oxide�(CdO,��99.99%�
metal� basis),� trioctylphosphine� oxide� (TOPO,� 99%),� trioctylphosphine� (TOP,� 90.0%),�
selenium�(Se�powder,�100�mesh,�99.99%�metals�basis),�diethylzinc�solution�(ZnEt2,�1�M�
in� hexane),� bis(trimethylsilyl)� sulfide� ((TMS)2S),� anhydrous� methanol,� ethanol� and�
acetonitrile� were� purchased� from� Sigma�Aldrich� (Steinheim,� Germany).�
Hexylphosphonic�acid�(HPA)�was�obtained�from�Alfa�Aesar�(Karlsruhe,�Germany).�These�
reagents� were� used� to� prepare� CdSe(ZnS)� QDs.� 4�hydroxy�3methoxybenzaldehyde�
(Vanillin,� �98%)� and� �� cyclodextrin� (�98%)� were� obtained� from� Fluka� (Steinheim,�
Germany).���cyclodextrin�was�purchased�from�Sigma�Aldrich�(Steinheim,�Germany)�and�
��cyclodextrin�(>98%)�was�purchased�from�Tokyo�Chemical�Industry�America�(Portland,�
U.S.A).�
Di�sodium� hydrogen� phosphate� anhydrous� buffer� was� purchased� from� Panreac�
(Barcelona,� Spain).� 4�hydroxy�3�methoxybenzyl� alcohol� (Vanillin� alcohol,� 98%),� 4�
hydroxybenzaldehyde�(98%)�and�4�hydroxybenzyl�alcohol�(99%)�were�purchased�from�
Sigma�Aldrich�(Steinheim,�Germany).�
Analytical�standard�stock�solutions�of�vanillin�at�1�mg�mL�1�were�prepared�in�water.�The�
stock� solutions� were� stored� under� refrigerator� conditions� (4� °C)� and� protected� from�
the� light.� The� stock� solution� of� Se/TOP� was� prepared� using� 0.051� g� of� Se� in� 3� mL� of�
TOP.�Buffer�solutions�was�prepared�using�di�sodium�hydrogen�phosphate�buffer�fixing�
the�pH�to�8.�
2.2. Apparatus
Fluorescence� emission� spectra� were� measured� on� a� Photon� Technology�
International�(PTI)�Inc.�QuantaMaster�40�spectrofluorometer�that�was�equipped�with�a�
75�W� continuous� xenon� arc� lamp.� An� ASOC�10� USB� interface� FeliXGX� software� was�
used�for�fluorescence�data�acquisition�and�also�controlled�the�hardware�for�all�system�
configurations.�The�slits�for�excitation�and�emission�widths�were�both�5�and�3�nm.�All�
5��
optical� measurements� were� performed� in� a� 10� mm� quartz� cell� at� room� temperature�
under�ambient�conditions.�UV�vis�spectra�were�obtained�on�a�SECOMAM�UVI�Light�XS�2�
spectrophotometer�equipped�with�a�LabPower�V3�50�for�absorbance�data�acquisition�
using� 10� mm� quartz� cuvettes.� QDs� were� precipitated� and� purified� using� a� centrifuge�
Centrofriger� BL�II� model� 7001669,� J.P� Selecta� (Barcelona,� Spain).� The� pH�
measurements� were� achieved� in� a� Crison� Basic� 20� pH�meter� with� a� combined� glass�
electrode� (Barcelona,� Spain).� An� ultrasonic� cleaning� bath� Ultrasons,� J.P.� Selecta�
(Barcelona,� Spain)� and� a� 254/365� nm� UV� lamp� 230� V,� E2107� model,� Consort� nv�
(Turnhout,�Belgium)�were�also�used.�
2.3. Preparation of hydrophobic CdSe/Zn QDs
CdSe�core�nanocrystals�were�prepared�via�a�modified�process�reported�Peng�et�al�
[27].�Typically,�0.06�g�of�CdO,�0.22�g�HPA,�and�7�g�of�TOPO�were� loaded�in�a�250�mL�
three�neck� flask� clamped� in� a� heating� mantle� and� air� in� the� system� was� pumped� off�
and�replaced�with�N2.�The�mixture�was�stirred�and�heated�at�300�310� °C� for�15�min,�
and�CdO�was�dissolved�in�HPA�and�TOPO.�The�solution�was�cooled�down�to�270�°C�and�
2.5� mL� of� the� solution� of� Se/TOP� was� swiftly� injected.� After� the� injection,� the�
temperature�was�adjusted�to�250�°C�for�nucleus�growth�during�20�min�and�a�change�in�
the�color�of�the�solution�to�red�was�observed.�To�make�ZnS�shell�on�the�CdSe,�3�mL�of�a�
solution� of� Zn/S/TOP� (0.58� g� of� ZnEt2,� 0.087� mL� of� (TMS)2S� and� 3.4� mL� of� TOP)� was�
added�dropwise�to�the�mixture�under�vigorous�stirring.�The�mixture�was�kept�to�90��C�
for� 4� h� to� improve� the� crystallinity� of� ZnS� shell.� After� cooling� the� solution� down� to�
room�temperature,�QDs�were�diluted�with�10�mL�of�anhydrous�chloroform.�Finally,�the�
synthesized� QDs� were� purified� by� adding� 10� mL� of� methanol� to� 10� mL� of� the� QD�
solution.�Then,�QDs�were�precipitated,�collected�by�ultracentrifugation�(at�13�000�rpm�
during�15�min),�and�washed�with�methanol� four� times.�The�purified�QD�nanocrystals�
were�finally�dispersed�in�10�mL�of�anhydrous�chloroform�and�stored�in�darkness.��
�
2.4. Preparation of n-cyclodextrin capped CdSe/ZnS QDs
Different� cyclodextrins� were� studied� (n=�,� �,� �).� The� n�CD�CdSe/ZnS�QDs� were�
prepared�by�using�a�modified�procedure�previously�reported�[28].�Thus,�a�0.5�mL�(200�
6��
mg�L�1)�of�TOPO�capped�CdSe/ZnS�QDs�in�chloroform�(0.1�mg)�were�added�into�a�2�mL�
polypropylene�vial�and�the�chloroform�was�dried�by�nitrogen�atmosphere.�Then,�nCD�
powder�(3.1,�3.6�or�4.2�mg�for��,��or���CD,�respectively)�was�added�to�dried�QDs�and�
the� mixture� was� dispersed� in� acetonitrile� (2� mL).� The� mixture� was� placed� in� a� high�
intensity�ultrasound�bath� for�about�45�min�at� room�temperature.�When�the�reaction�
was� finished,� a� rosy� precipitate� was� obtained.� The� precipitate� was� separated� by�
centrifuging�at�13500�rpm.�The�resulting�supernatant�was�eliminated�and�the�remained�
acetonitrile�was�evaporated.�Finally,�it�was�purified�by�further�cycles�of�centrifugation�
in�water.�The�resulting�precipitate�of� the�n�CD�CdSe/ZnS�QDs�was�dispersed� in�water�
(10�mL)�and�stored�at�room�temperature�in�the�dark�for�further�investigations.�
2.5. Preparation of samples and analytical procedure
Several�commercial� food�samples,�such�as�sugar,�milk�or�custard,�were�purchased�
from�a�local�market.�These�samples�were�prepared�as�follows:�
Sugar�samples�were�ground�to�a�fine�powder.�Then,�0.5�g�of�this�powder�and�2�mL�
of�absolute�ethanol�were�place�into�a�tube�and�shacked�by�a�laboratory�shaker�for�10�
min.�This�mixture�was�centrifuged�at�10000�rpm.�The�clear�part�of�the�solution�in�the�
tube� was� used� for� analysis.� Ethanol� was� evaporated� and� the� resulting� residue� was�
dissolved�in�water.�
Milk�samples.�1�mL�of�milk�sample�and�2�mL�of�absolute�ethanol�were�place�into�a�
tube� and� shacked� by� a� laboratory� shaker� at� 40� �C� for� 10� min.� This� mixture� was�
centrifuged�at�12000�rpm�for�15�min�for�precipitate�the�proteins.�The�supernatant�was�
evaporated�and�the�resulting�sample�residue�was�dissolved�in�water.�
Custard� samples.� 0.05� g� of� custard� powder� and� 2� mL� of� absolute� ethanol� were�
place�into�a�tube�and�shacked�by�a�laboratory�shaker�at�40��C�for�10�min.�This�mixture�
was�centrifuged�at�6000�rpm�for�15�min.�The�supernatant�was�used�for�analysis.�The�
absolute� ethanol� was� evaporated� and� the� resulting� sample� residue� was� dissolved� in�
water.�
For�vanillin�determination,� suitable�amount�of� these�samples�and�0.4�mL�of���CD�
modified�CdSe/ZnS�fixed�at�pH�8�were�transferred�into�a�volumetric�flask.�The�mixture�
was�stirred�at�room�temperature�and�stored�at�ambient�light�in�the�dark�for�30�min�for�
7��
reaction.� Then,� this� mixture� was� transferred� into� a� 10� mm� quartz� cuvette� and� the�
emission�fluorescent�spectrum�was�measured,�at�an�excitation�wavelength�of�450�nm,�
between�500�and�670�nm.� I0/I�was�used�as�analytical� signal,�where� I0�and� I�were� the�
fluorescence� intensity� at� 590� nm� of� the� systems� in� the� absence� and� presence� of�
vanillin,�respectively.�
3. Results and Discussion
Studies�for�QDs�modification�was�carried�out�using�different�cyclodextrins�(n�=��,���
or��).�The�solubilization�procedure�of�the�QDs�was�conducted�by�ultrasonic�irradiation�
of� a� mixture� of� TOPO�coated� CdSe/ZnS� QDs� and� n�CD.� The� assayed� strategy� for�
creating� nCD�QDs� was�a� chemical�procedure� based� on� the� formation� of� a� host�guest�
complex� between� the� passivized� ligand� (TOPO)� and� nCD� by� hydrophobic� interaction�
(Figure�1A).� In�order� to�obtain�the�optimal�conditions� in� the�modification�procedure,�
several�parameters�were�studied.�
3.1. Influence�of�experimental�factors�in�the�modification�of�n�CD�CdSe/ZnS�QDs�
The�effect�of�solvent�reaction�on�the�fluorescence�intensity�and�stability�of�QDs�was�
tested�using�absolute�ethanol,�anhydrous�methanol�and�acetonitrile.�It�was�found�that�
fluorescence� intensity� of� n�CD�CdSe/ZnS�QDs� when� it� was� used� acetonitrile� as� a�
reaction�solvent�was�dramatically�higher� than�when�ethanol�or�methanol�were�used.�
Therefore,�this�solvent�was�used�for�further�experiments.�Figure�2A�shows�the�effect�of�
these�solvents�using���CD�CdSe/ZnS�QDs.�
The� concentration� of� n�CD� was� varied� between� 0.5� and� 4.5� mM,� maintaining�
constant�the�other�parameters.�It�was�observed�that�the�complexation�of�TOPO�with�n�
CD�is�essential�to�produce�the�solubilization�of�CdSe/ZnS�QDs�in�an�aqueous�medium.�
When�the�concentration�of�n�CD�is�too�low,�only�small�portions�of�surface�bound�TOPO�
molecules� on� the� surfaces� of� CdSe/ZnS� QDs� form� complexes� with� n�CD,� which� is� not�
enough�to�give�a�hydrophilic�property�to�the�nanoparticles,�and�hence�to�produce�their�
stabilization� in� the� aqueous� phase.� At� relatively� high� concentrations� of� n�CD,�
substantial�amounts�of�TOPO�molecules�are�able�to�form�host�guest�complexes�with�n�
8��
CD,� which� led� to� an� increase� in� the� stability� of� the� QDs� in� the� hydrophilic� media.�
However,�when�the�dosage�of�n�CD�is�very�high,�although�phase�transfer�was�efficiently�
achieved,�the�resulting�complex�was�found�to�be�unstable�in�water,�and�the�nCD�excess�
could�mask�the�determination�of�the�analyte.�Therefore,�1.6�mM�of�n�CD�was�chosen�
for� further�experiments,�as�the�maximum�fluorescence� intensity�was�obtained�at�this�
concentration.�Finally,� time�dependent�experiments�were�performed�by�exposing�the�
reaction�of�modification�of�QDs�at�several�times�between�15�and�180�min�to�ultrasonic�
irradiation.�The�best�results�were�obtained�when�45�min�of�ultrasonic� irradiation�was�
used.�
The� resulting� n�CD�CdSe/ZnS�QDs� thus� obtained� were� highly� fluorescent� and�
stable.� Figure� 2A� shows� the� emission� spectra� of� n�CD�CdSe/ZnS�QDs� in� water� and�
TOPO�CdSe/ZnS�QDs� in� chloroform.� As� it� can� be� seen� the� maximum� emission� band�
around� 590� nm� (�exc� =� 450� nm)� were� obtained� in� all� cases.� The� line� width� of� the�
fluorescence�spectrum�is�relatively�narrow�(with�the�fullwidth�at�half�maximum�of�44�
nm),� indicating� that� the� n�CD�CdSe/ZnS�QDs� nanoparticles� have� a� narrow� size�
distribution.� Compared� to� TOPO�CdSe/ZnS� QDs� in� chloroform,� n�CDCdSe/ZnS�QDs�
increased� the� fluorescence� intensity.� It� was� also� observed� no� change� in� emission�
wavelength� and� the� spectral� width� regarding� to� TOPO�QDs.� The� UV/Vis� spectrum� of�
TOPO�CdSe/ZnS�QDs,� ��CD�CdSe/ZnS�QDs,� ��CD�CdSe/ZnS�QDs� and� ��CD�CdSe/ZnS�
QDs�are�also�illustrated�in�Figure�2B.�As�it�can�be�seen,�absorption�bands�at�ca.255�and�
ca.585� nm� were� obtained� in� all� cases.� Therefore,� no� significant� differences� were�
observed� in� the� modification� procedure� of� QDs.� The� stability� of� obtained� for� n�CD�
CdSe/ZnS�QDs� in�water�was�estimated�by�measurements�of� the�emission� intensity�at�
room�temperature�at�several�times.�From�the�results�obtained�it�was�concluded�that�n�
CD�CdSe/ZnS�QDs�were�stable�at� least� for�two�weeks,�with�any�significant�changes� in�
their�fluorescence�spectra.�
�
3.2. Effect of vanillin on the luminescence response of modified QDs
In� the� n�cyclodextrin� modified� QDs,� the� hydrophobic� pockets� of� the� cyclodextrin�
molecules� interact�with�the�aliphatic�chains�of�the�TOPO�present�on�the�nanoparticle�
9��
surface� from� the� QDs� synthesis� (Figure� 1A).� Nevertheless,� the� immobilized�
cyclodextrins� retain� their� capability�of�engaging�molecular� recognition.� In� this�way,� it�
was�studied�the�use�of�n�CD�CdSe/ZnS�QDs�nanoparticles�as�a�selective� luminescence�
sensors� of� vanillin.� First,� preliminary� studies� were� made� in� order� to� know� the� best�
recognition�of�vanillin�when��,�� or���CD�CdSe/ZnS�QDs�were�used.� It�was� found� that�
vanillin�affects�luminescence�of���CD�CdSe/ZnS�QDs�in�a�more�drastic�way,�producing�a�
quenching�effect�on�the�QDs�emission�band,�as�it�can�be�seen�in�Figure�3.�Therefore,���
CD�CdSe/ZnS�QDs�was�selected�to�develop�the�vanillin�sensors.��
The� selective� host�guest� interaction� between� vanillin� and� ��cyclodextrin,� through�
the� vanillin� binds� to� the� receptor� sites,� can� act� as� an� electron� transfer� quencher� of�
luminescence� of� the� particles� (Figure� 1B).� Under� these� conditions� the� association� of�
the� vanillin� to� the� ��CD� cavities� concentrates� the� analyte� on� the� semiconductor� QDs�
surface.� The� time�dependent� luminescence� changes� of� the� ��CD�QDs� upon� their�
interaction�with�4.2�mg�L�1�of�vanillin�were�investigated.�From�the�results�obtained,� it�
can� be� concluded� that� luminescence� of� the� ��CD�QDs� in� the� presence� of� vanillin�
decreased� until� 30� min,� and� remained� constant� after� this� time� value.� The� same�
behavior�was�also�observed�by�recording�of�absorbance�time�spectra�of�vanillin���CD�
CdSe/ZnS�QDs�complex.���
The�pH�effect�in�the�recognition�of�vanillin�using���CD�CdSe/ZnS�QDs�nanoparticles�
was� studied� by� measurements� of� fluorescence� intensity� of� the� ��CD�CdSe/ZnS�QDs�
without�(I0)�and�at�given�vanillin�concentration�(I).�It�was�found�that�the�pH�significantly�
influenced� the� fluorescence� intensity� of� the� vanillin���CD�CdSe/ZnS�QDs� system.� The�
maximum�value�of� I/I0�was�obtained�when�the�pH�was�8.0.�Therefore,�this�value�was�
chosen�as�optimum.�As�a�second�part�of�this�study,�the�influence�of�the�concentration�
of�buffer�solution,�fixed�with�Na2HPO4�pH=8,�was�carried�in�order�to�evaluate�the�effect�
of�this�parameter�in�the�fluorescence�intensity�of�vanillin���CD�CdSe/ZnS�QDs�system.�
The� results� showed� that� the� maximum� value� I/I0� was� obtained� when� the� buffer�
solution� concentration� was� 1.2� mM.� Therefore,� this� value� was� chosen� in� all�
experiments.�
10��
According� to� the� literature� [29],� the� surface� of� the� n�CD�CdSe/ZnS�QDs� affords� a�
finite� number� of� binding� sites.� Each� of� the� binding� sites� could� absorb� one� vanillin�
molecule�from�the�solution.�Therefore,�according�to�Langmuir�equation�[28]:�
�1 1 [1]max max
C CI BI I
� � � �� �� � � �� �
�
where�C� is� the�concentration�of�vanillin�and� I� the� fluorescence� intensity�obtained�for�
this�concentration�level.�According�to�the�literature,�if�the�Langmuir�description�of�the�
binding�of�vanillin�on�the�surface�of���CD�CdSe/ZnS�QD�is�correct,�a�linear�plot�of�c/I�as�
a� function� of� c� must� be� obtained.� In� this� case,� a� good� linearity� was� observed�
throughout� the� entire� range� of� vanillin� concentration� (2� to� 20� mg� L�1).� The� binding�
constant�B�of���CD�CdSe/ZnS�QDs�with�vanillin�is�found�to�be�0.99.�
3.3. Analytical features for vanillin determination
In� order� to� develop� an� analytical� method� to� determine� vanillin� in� foods,� several�
analytical� performance� characteristics� were� evaluated� under� the� optimized�
experimental� conditions.� The� quenching� effect� of� the� vanillin� can� be� described� using�
the�following�Stern�Volmer�equation:�
� �0 1 [2]svI K QI
� � �
where� I0� and� I� are� the� fluorescence� intensity� of� ��CD�CdSe/ZnS�QDs� in� absence� and�
presence�of�vanillin.�A�linear�relationship�between�I0/I�and�vanillin�concentration�in�the�
range� of� 2�20� mg� L�1� with� a� correlation� coefficient� of� 0.9963� was� obtained.� Figure� 4�
shows� the� fluorescence� spectra� of� ��CD�CdSe/ZnS�QDs� at�different�concentrations� of�
vanillin�between�2�and�20�mg�L�1.�The�calibration�equation�was:�
� �0 1.094 0.051 [3]I vanillinI
� � �
The� precision� of� the� methodology� was� evaluated� in� terms� of� repeatability� and�
reproducibility.�To�determine�the�repeatability�of�the�method,�10�analysis�of�samples�
containing�4.12�mg�L�1�of�vanillin�were�carried�out�and�the�obtained�relative�standard�
deviation� (R.S.D.)� was� 1.2%.� Then,� the� reproducibility� was� estimated� for� three�
11��
replicates�of�4.12�mg�L�1�vanillin�under�inter�day�conditions�(for�two�consecutive�days),�
obtaining�a�R.S.D.�of�4.7%.�A� limit�of�detection�(LOD)�of�0.99�mg�L�1�was�obtained�for�
vanillin� determination,� based� on� the� IUPAC� method� (blank� signal� plus� 3� times� its�
standard� deviation).� 10� measurements� were� used� to� obtain� the� LOD.� According� to�
these�results,�it�can�be�concluded�that�this�approach�opens�the�possibilities�to�develop�
an�analytical�method�using���CD�CdSe/ZnS�QDs�for�the�determination�of�vanillin.�
In�order�to�apply�the�method�to�food�samples�a�selectivity�study�was�also�carried�
out.� For� vanillin� determination� in� food� samples,� the� main� interferences� come� from�
several� colorant� additives,� such� as� curcumine� or� riboflavine.� However,� following� the�
procedure� reported� in� the� Section� 2.5,� the� interferences� of� these� compounds� were�
eliminated.�By�contrast,�sucrose�(glucose�and�lactose)�could�be�present�in�the�sample�
as� principal� interfering� compound.� For� this� purpose,� two� different� level� of�
concentration�were�used�in�combination�with�vanillin.�From�the�results�obtained�it�can�
concluded�that�when�the�concentration�of�interferences�were�increase,�not�difference�
were�observed�in�the�vanillin�signal,�given�values�of�R.S.D�of�3.9�and�4.5%,�respectively.�
The�coexisting�compounds�caused�a�relative�error�of�less�than�±�5%�in�the�fluorescence�
intensity�of�the�vanillin���CD�CdSe/ZnS�QDs.�Therefore�it�can�be�considered�to�have�no�
interference� with� the� detection� of� vanillin.� The� data� revealed� that� the� proposed�
method�might�be�applied�to�the�detection�of�vanillin�in�food�samples.�
On�the�other�hand,�several�similar�structures�to�vanillin,�such�as�vanillin�alcohol,�4�
hydroxybenzaldehyde�or�4�hydroxybenzyl�alcohol�were�tested.�For�this�purpose,�three�
different� levels� of� concentration� were� used� in� combination� with� vanillin.� From� the�
results�obtained�it�can�be�concluded�that�when�the�concentration�of�interference�was�
increased,�no�difference�was�observed�in�the�vanillin�signal,�with�less�of�±�5%�of�R.S.D.�
when� vanillin� alcohol� and� 4�hydroxybenzyl� alcohol� were� used.� However,� the� results�
obtained�in�the�presence�of�4�hydroxybenzaldehyde�indicated�that�this�compound�can�
produce� interference� in� the� vanillin� determination.� This� fact� demonstrated� that� the�
structure�of�the�analyte�and�pH�of�solution�system�could�play�an�important�role�in�the�
selectivity� of� vanillin� determination� with� ��CD�CdSe/ZnS�QDs.� Table� 1� shows� the�
obtained� change� of� fluorescence� intensity� (%)� at� the� three� concentration� levels�
studied.�
12��
4. Application
To� demonstrate� the� applicability� of� the� proposed� method,� it� was� applied� to�
determine� vanillin� in� several� commercial� sugar� and� milk� samples,� purchased� in�
different� supermarkets.� Each� one� of� these� samples� was� spiked� with� several�
concentrations� of� vanillin� and� were� prepared� according� to� the� steps� described� in�
Section� 2.5.� The� summary� of� these� results� are� shown� in� Table� 2.� The� obtained�
recoveries�indicated�an�acceptable�agreement�between�the�amounts�added�and�those�
found�for�all�types�of�samples.�
The�proposed�method�was�also�used�for�the�quantitation�of�vanillin�in�custard�with�
vanilla� flavor.� These� samples� contained� vanillin� as� flavor� additive.� This� product� was�
analyzed� by� triplicate,� according� the� procedure�described� in� Section� 2.5.� To� evaluate�
the�matrix�effect,�the�standard�addition�method�was�also�used�for�the�determination�
of�vanillin�in�the�studied�product.�The�obtained�results�were�75.6�±�0.6�and�76.3�±�0.9�
mg�L�1�with�and�without�standard�addition,�respectively,�corresponding�to�the�original�
sample� (samples�were�diluted�before�analyses).�The�application�of�Student� statistical�
test� for� a� confidence� level� of� 95%� demonstrated� the� statistical� coincidence� between�
the�concentration�found�with�those�found�by�the�standard�addition�method�(n�=�6,�tcrit�
=�2.92�>�texp=�0.48).��
�
5. Conclusions
In� this� work,� an� optical� sensor� for� vanillin� determination� based� on� the� selective�
supramolecular�recognition�of�vanillin�with���cyclodextrin�modified�CdSe/ZnS�QDs�was�
developed.� The� procedure� for� the� synthesis� of� n�CD�CdSe/ZnS�QDs� complex� was�
simple� and� very� effective.� Different� coating� agents,� such� as� �,� �� and� ��CD,� were�
studied,� � � and� the� effect� of� several� experimental� parameters� was� optimized.� The�
proposed� methodology� presents� some� advantages.� Thus,� the� solubilization� of� TOPO�
CdSe/ZnS�QD,�initially�in�organic�media,�was�allowed�in�aqueous�media.�This�fact�allows�
its� compatibility� with� biological� samples,� and� aqueous� media� in� general,� and� in�
addition�the�conservation�in�organic�media�over�long�periods�of�time.�In�this�way,�it�is�
possible� to� modify� only� the� necessary� amount� of� QDs� when� required.� On� the� other�
13��
hand,�the�methodology�demonstrated�that�the�surface�coating�of�QDs�with�different�n�
cyclodextrins�keeps�the�emission�intensity�of�the�quantum�dots�and�their�diameter.�In�
addition,� the� immobilized� ��cyclodextrins� on� the� surface� of� the� QDs� retain� their�
capability�of�engaging�molecular�recognition.�Therefore,�the�use�of���cyclodextrin�for�
QDs� surface� modification� showed� selectivity� in� vanillin� recognition� to� alpha� and�
gamma�cyclodextrins.�Thus,�the�potential�of�the�sensor�for�the�analysis�of�food�samples�
was� demonstrated,� opening� other� possible� alternatives� for� the� selective� fluorimetric�
sensing�of�other�compounds�through�the�appropriated�modification�of�quantum�dots�
surface.��
�
Acknowledgements
This�research�was�supported�by�Project�CTQ2013�48411�P�(MINECO).�Gema�M.�Durán�
thanks�the�Spanish�Ministry�of�Economy�and�Competitiveness�for�a�Predoctoral�Grant.�
� �
14��
References
[1]� C.� Carrillo�Carrion,� S.� Cardenas,� B.� M.� Simonet,� M.� Valcarcel,� Selective�Quantification� of� Carnitine� Enantiomers� Using� Chiral� Cysteine�Capped�CdSe(ZnS)�Quantum�Dots,�Anal.�Chem.�81�(2009)�4730�4733.�
[2]� S.�Huang,�Q.�Xiao,�R.�Li,�H�L.�Guan,�J.�Liu,�X�R.�Liu,�Z�K.�He,�Y.�Liu,�A�simple�and�sensitive� method� for� L�cysteine� detection� based� on� the� fluorescence� intensity�increment�of�quantum�dots,�Anal.�Chim.�Acta�645�(2009)�73�78.�
[3]� G.� M.� Duran,� A.� M.� Contento,� A.� Ríos,� Use� of� Cdse/ZnS� quantum� dots� for�sensitive� detection� and� quantification� of� paraquat� in� water� samples,� Anal.�Chim.�Acta�801�(2013)�84�90.�
[4]� C.� Carrillo�Carrion,� B.M.� Simonet,� M.� Valcarcel,� (CdSe/ZnS� QDs)�ionic� liquid�based�headspace�single�drop�microextraction�for�the�fluorimetric�determination�of�trimethylamine�in�fish,�Analyst�137�(2012)�1152�1159.�
[5]� T�W.�Sung,�Y�L.�Lo,�Highly�sensitive�and�selective�sensor�based�on�silica�coated�CdSe/ZnS�nanoparticles�for�Cu2+�ion�detection,�Sens.�Actuators,�B:�Chemical�165�(2012)�119�125.�
[6]� C.�Carrillo�Carrion,�B.�Lendl,�B.�M.�Simonet,�M.�Valcarcel,�Calix[8]arene�Coated�CdSe/ZnS�Quantum�Dots�as�C60�Nanosensor,�Anal.�Chem.�83�(2011)�8093�8100.�
[7]� P.�Ncube,�R.�W.�Krause,�B.� B�Mamba,�Fluorescent� sensing�of� chlorophenols� in�water� using� an� azo� dye� modified� ��cyclodextrin� polymer,� Sensors� 11� (2011)�4598�4608.�
[8]� A.� T.� Ogoshi,� A.� Harada,� Chemical� Sensors� Based� on� Cyclodextrin� Derivatives�Sensors�8�(2008)�4961�4982.��
[9]� M.� Li,� X.� Liu,� F.� Y.� Jiang,� L.P.� Guo,� L.� Yang,� Enantioselective� open�tubular�capillary�electrochromatography�using�cyclodextrin�modified�gold�nanoparticles�as�stationary�phase,�J.�Chromatogr.�A�1218�(2011)�3725�3729.�
[10]� R.� Freeman,� T.� Finder,� L.� Bahshi,� I.� Willner,� ��Cyclodextrin�Modified� CdSe/ZnS�Quantum� Dots� for� Sensing� and� Chiroselective� Analysis,� Nano� Lett.� 9� (2009)�2073�2076.�
[11]� C.� P.� Han,� H.� B.� Li,� Chiral� Recognition� of� Amino� Acids� Based� on� Cyclodextrin�Capped�Quantum�Dots,�Small�4�(2008)�1344�1350.�
[12]� H.� B.� Li,� C.� P.� Han,� Sonochemical� Synthesis� of� Cyclodextrin�Coated� Quantum�Dots� for� Optical� Detection� of� Pollutant� Phenols� in� Water,� Chem.� Mater.� 20�(2008)�6053�6059.�
[13]� M.�Algarra,�B.�B.�Campos,�F.�R.�Aguiar,�J.�E.�Rodriguez�Borges,�J.�C.�G.�Esteves�da�Silva,� Novel� ��cyclodextrin� modified� CdTe� quantum� dots� as� fluorescence�nanosensor�for�acetylsalicylic�acid�and�metabolites,�Mater.�Sci.�Eng.,�C�32�(2012)�799�803.�
15��
[14]� C.�P.�Han,�H.�B.�Li,�Novel���cyclodextrin�modified�quantum�dots�as�fluorescent�probes� for� polycyclic� aromatic� hydrocarbons� (PAHs),� Chin.� Chem.� Lett.� 19�(2008)�215�218.�
[15]� Y.� Cao,� S.� Wu,� Y.� Liang,� Y.� Yu,� The� molecular� recognition� of� ��cyclodextrin�modified�CdSe�quantum�dots�with�tyrosine�enantiomers:�Theoretical�calculation�and�experimental�study,�J.�Mol.�Struct.�1031�(2013)�9�13.�
[16]� S.� Ramachandra� Rao,� G.� A.� Ravishankar,� Vanilla� flavour:� production� by�conventional�and�biotechnological�routes,�J.�Sci.�Food�Agric.�80�(2000)�289–304.�
[17]� Y.�Yardm,�M.�Gülcan,�Z.�entürk,�Determination�of�vanillin�in�commercial�food�product� by� adsorptive� stripping� voltammetry� using� a� boron�doped� diamond�electrode,�Food�Chem.�141�(2013)�1821–1827.��
[18]� D.�Zheng,�C.�Hu,�T.�Gan,�X.�Dang,�S.�Hu,�Preparation�and�application�of�a�novel�vanillin� sensor� based� on� biosynthesis� of� Au–Ag� alloy� nanoparticles,� Sens.�Actuators,�B�148�(2010)�247–252.�
[19]� F.� Bettazzi,� I.� Palchetti.,� S.� Sisalli,� M.� Mascini,� A� disposable� electrochemical�sensor�for�vanillin�detection,�Anal.�Chim.�Acta�555�(2006)�134–138.�
[20]� J.� L.� Hardcastle,� C.� J.� Paterson,� R.� G.� Compton,� Biphasic� Sonoelectroanalysis:�Simultaneous�Extraction�from,�and�Determination�of�Vanillin�in�Food�Flavoring,�Electroanal.�13�(2001)�899–905.�
[21]� M.� Luque,� E.� Luque�Pérez,� A.� Ríos,� M.� Valcárcel,� Supported� liquid� membranes�for�the�determination�of�vanillin�in�food�samples�with�amperometric�detection,�Anal.�Chim.�Acta�410�(2000)�127–134.�
[22]� M.�Avila,�M.�Zougagh,�A.�Escarpa,�A.�Rios,�Supported�liquid�membrane�modified�piezoelectric� flow� sensor� with� molecularly� imprinted� polymer� for� the�determination�of�vanillin�in�food�samples,�Talanta�72�(2007)�1362–1369.�
[23]� Y.� Ni,� G.� Zhang,� S.� Kokot,� Simultaneous� spectrophotometric� determination� of�maltol,� ethyl� maltol,� vanillin� and� ethyl� vanillin� in� foods� by� multivariate�calibration�and�artificial�neural�networks,�Food�Chem.�89�(2005)�465–473.��
[24]� U.�Pyell,�B.�Pletsch�Viehmann,�U.�Ramus,�Component�analysis�of�vanilla�extracts�and� vanilla� containing� commercial� preparations� by� micellar� electrokinetic�chromatography� or� high�performance� liquid� chromatography�A� method�comparison,�J.�Sep.�Sci.�25�(2002)�1035–1042.��
[25]� M.�Ohashi,�H.�Omae,�M.�Hashida,�Y.�Sowa,�S.�Imai,�Determination�of�vanillin�and�related� flavor� compounds� in� cocoa� drink� by� capillary� electrophoresis,� J.�Chromatogr.,�A,�1138�(2007)�262–267.�
[26]� A.� Perez�Silva,� E.� Odoux,� P.� Brat,� F.� Ribeyre,� G.� Rodriguez�Jimenes,� V.� Robles�Olvera,� GC–MS� and� GC–olfactometry� analysis� of� aroma� compounds� in� a�representative� organic� aroma� extract� from� cured� vanilla� (Vanilla� planifolia� G.�Jackson)�beans,�Food�Chem.�99�(2006)�728–735.�
[27]� Z.A.�Peng�,�X.�Peng,�Formation�of�high�quality�CdTe,�CdSe,�and�CdS�nanocrystals�using�CdO�as�precursor,�J.�Am.�Chem.�Soc.�123�(2001)�183�184.�
16��
[28]� C.�Han,�H.�Li,�Chiral�Recognition�of�Amino�Acids�Based�on�Cyclodextrin�Capped�Quantum�Dots,�Small�4�(2008)�1344�1350.�
[29]� Y.Chen,�Z.�Rosenzweig,�Luminescent�CdS�Quantum�Dots�as�Selective�Ion�Probes,�Anal.�Chem.�74�(2002)�5132–5138.�
17��
FIGURES�CAPTION��Figure�1.�Schematic�illustration�of�surface�modification�of�TOPO�CdSe/ZnS�QDs�with�n�cyclodextrins�(A).�Host�guest�interaction�between���CD�CdSe/ZnS�QDs�and�vanillin�(B).��Figure� 2.� Emission� (A)� � and� absorption� (B)� spectra� of� TOPO�CdSe/ZnS�QDs� (a),� ��CD�CdSe/ZnS�QDs�(b),���CD�CdSe/ZnS�QDs�(c)�and���CD�CdSe/ZnS�QDs�(d).��Figure� 3.� Effect� of� 4.2� mg� L�1� vanillin� concentration� over� luminescence� of� ��CD�CdSe/ZnS�QDs�(A),���CD�CdSe/ZnS�QDs�(B)�and���CD�CdSe/ZnS�QDs�(C).��Figure�4.�Fluorescence�spectra�of���CD�CdSe/ZnS�QDs�with�different�concentrations�of�vanillin�between�2�and�20�mg�L�1.����
18��
HIGHLIGHTS�
�
��Cyclodextrin�CdSe/ZnS�quantum�dots�were�synthesized.�
�
��Compatibility�with�aqueous�media.�
�
��The�new�materials�were�used�as�selective�sensor�for�vanillin.��
�
����cyclodextrin�CdSe/ZnS�was�used�for�analysis�of�food�samples.�
�
�
19��
Table�1.�Effect�of�coexisting�foreign�substances�at�three�different�concentration�levels�with�respect�to�vanillin�concentration.�
Foreign�substances� Foreign�species�ratio��Error�in�vanillin�determination�(%)�
Sucrose�1:10 +1.91:20 +2.5
Vanillin�alcohol� 1:0.5 �0.11:1 +0.31:2 +0.2
4�hydroxybenzyl�alcohol� 1:0.5 �2.51:1 +0.41:2 +2.9
4�hydroxyaldehyde� 1:0.5 �6.51:1 �8.21:2 �9.7
Conditions:�pH�8;�[vanillin]=�6.65�μg�mL�1.�
20��
Table�2.�Determination�of�vanillin�in�several�food�samples�(n=5).�
Sample� Added�(mg�L�1) Found�(mg�L�1) Recovery�(%)�
Sugar�1�5� 5.26�±0.2� 105�±4�10� 9.96 ±0.5 100�±5�20� 19.2�±1 96�±5�
Sugar�2�5� 4.9�±0.2� 99�±3�10� 10.0 ±0.4 100 ±4�20� 20.0�±0.3� 99�±1.5�
Sugar�3�5� 5.0 ±0.2 101 ±4�10� 9.6�±0.4� 96�±4�20� 18.2�±0.9� 91�±4.5�
Milk�1�5� 5.2�±0.2� 104�±4�10� 9.0�±0.4� 90�±4�20� 20.0�±1.0� 99�±5�
Milk�2�5� 5.1�±0.2� 103�±4�10� 9.0�±0.5� 90�±5�20� 19.6�±0.8� 98�±4�
�
Figu
re
Figu
re
Figu
re
Figu
re
GRA
PHIC
AL
ABS
TRAC
T
*Gra
phic
al A
bstr
act (
for r
evie
w)