Molybdenum disulphide and graphene quantum dots as electrode [601298]

Molybdenum disulphide and graphene quantum dots as electrode
modi fiers for laccase biosensor
Ioana Vasilescua, Sandra A.V. Eremiaa,n, Mihaela Kuskob, Antonio Radoib,nn,
Eugeniu Vasilec, Gabriel-Lucian Radua
aCentre of Bioanalysis, National Institute of Research and Development for Biological Sciences, Bucharest, 296 Splaiul Independentei, 060031 Buch arest,
Romania
bNational Institute for Research and Development in Microtechnologies (IMT-Bucharest), 126A Erou Iancu Nicolae, 077190 Bucharest, Romania
cDepartment of Oxide Materials and Nanomaterials, Faculty of Applied Chemistry and Material Science, University Politehnica of Bucharest, No. 1-7 G h.
Polizu Street, 011061 Bucharest, Romania
article info
Article history:Received 1 July 2015Received in revised form21 August 2015Accepted 22 August 2015
Available online 22 August 2015
Keywords:
MoS
2
Graphene quantum dotsLaccaseBiosensorPolyphenol indexabstract
A nanocomposite formed from molybdenum disulphide (MoS 2) and graphene quantum dots (GQDs) was
proposed as a novel and suitable support for enzyme immobilisation displaying interesting electro-chemical properties. The conductivity of the carbon based screen-printed electrodes was highly im-
proved after modi fication with MoS
2nano flakes and GQDs, the nanocomposite also providing compa-
tible matrix for laccase immobilisation. The in fluence of different modi fication steps on the final elec-
troanalytical performances of the modi fied electrode were evaluated by UV –vis absorption and fluor-
escence spectroscopy, scanning electron microscopy, transmission electron microscopy, X ray diffraction,
electrochemical impedance spectroscopy and cyclic voltammetry. The developed laccase biosensor has
responded ef ficiently to caffeic acid over a concentration range of 0.38 –100mM, had a detection limit of
0.32 mM and a sensitivity of 17.92 nA mM/C01. The proposed analytical tool was successfully applied for the
determination of total polyphenolic content from red wine samples.
&2015 Elsevier B.V. All rights reserved.
1. Introduction
Laccases are multicopper-oxidase enzymes that catalyse the
one-electron oxidation processes of various phenolic compounds
and aromatic amines with the concomitant reduction of oxygen(Kudanga et al., 2011 ). They occur widely in fungi but are also
found in higher plants, prokaryotes and bacteria. Besides their use
in analytical applications as biosensors and biofuel cells, laccasesplay an important role in food, paper, textile and pharmaceutical
industry, or in cosmetics and environmental bioremediation
(Shraddha et al., 2011 ). In recent years, laccase based biosensors
are becoming relevant for areas like food analysis and environ-
mental monitoring due to their properties such as fast responsetime, low-cost, low reagents consumption and the most important
one, the possibility to use them in field. The ability of laccase to
catalyse the one-electron oxidation of different polyphenols hasbeen used to develop laccase based electrochemical biosensors for
the determination of polyphenolic index in wines ( García-Guzmánet al., 2015 ;Lanzellotto et al., 2014 ), in tea infusions ( Cortina-Puig
et al., 2010 ;Eremia et al., 2013 ) and tea leaves extracts ( Rawal
et al., 2012 ) as well as for the determination of environmental
pollutants ( Brondani et al., 2013 ;Nazari et al., 2015 ;Oliveira et al.,
2013 ).
The tendency in the field of electrochemical biosensors is
substituting the organic redox mediators with different nanoma-
terials such as noble metal nanoparticles ( Radoi et al., 2011 ), car-
bon derived materials ( Eremia et al., 2013 ;Karimi-Maleh et al.,
2013 ;Karimi-Maleh et al., 2014 ) and transition-metal dichalco-
genides and thus immobilising the enzyme at the electrode sur-
face promoting the development of third-generation biosensors.
Therefore attention has to be paid in order to obtain an ef ficient
electron transport between the electrode and the redox active
sites of the enzyme, as enzymes may exhibit reluctant electrontransfer at the surface of conventional electrodes.
Molybdenum disulphide (MoS
2) is a two-dimensional (2D)
layered material having a hexagonal crystalline lattice, the Mo
layer being sandwiched between two S layers, the layers being
stacked by weak van der Waals interactions ( Song et al., 2014 ;
Wang, 2014 ). Graphene on the other hand, is a revolutionary 2D
material which has been intensively studied by many researchers
and has potential applications in many areas. The main differenceContents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/biosBiosensors and Bioelectronics
http://dx.doi.org/10.1016/j.bios.2015.08.051
0956-5663/ &2015 Elsevier B.V. All rights reserved.
nCorresponding author. Fax: ț40 21 220 09 00.
nnCorresponding author. Fax: ț40 21 269 07 72 76.
E-mail addresses: sandraeremia@gmail.com (S.A.V. Eremia),
radoiantonio@yahoo.com (A. Radoi).Biosensors and Bioelectronics 75 (2016) 232 –237

between the mentioned 2D materials is the presence of an indirect
bandgap of 1.2 eV for bulk MoS 2(or 1.8 eV for a MoS 2monolayer)
(Mak et al., 2010 ) in contrast to the zero-gap semiconductor gra-
phene ( Wang, 2014 ) that leads to an attractive electrocatalytic
activity ( Mao et al., 2015 ). The first scienti fic papers describing the
use of MoS 2nanomaterial in the construction of biosensors ap-
peared in 2013; the most cited work is describing the develop-
ment of a sensing platform for DNA detection ( Zhu et al., 2013 ),
while the other two articles describing biosensors based on MoS 2
nanosheets for H 2O2determination ( T. Wang et al., 2013 ;G.-X
Wang et al., 2013 ).
The tremendous interest in graphene is related to its potential
electronic applications such as transistors ( Schwierz, 2010 ),
transparent electrodes ( Pang et al., 2011 ) or photodetectors ( Liu
et al., 2014 ) despite of its zero-energy gap ( Güçlü et al., 2014 ). The
obstacle regarding graphene zero-bandgap can be eliminated by
reducing the lateral size of graphene and thus opening an energy
gap that is characteristic to graphene ribbons and graphene
quantum dots ( Güçlü et al., 2014 ). Graphene quantum dots (GQDs)
are graphene fragments having their diameter below 20 nm that
attracted considerable attention for the last years. One of the most
important properties of GQDs is that they are carbon derived
materials, showing low toxicity, high solubility in many solvents
and the possibility of functionalization at their edges ( Bacon et al.,
2014 ). The main use of GQDs is in photovoltaic devices due to their
ability to absorb a high percentage of incident light ( Mihalache
et al., 2015 ), followed by fuel cells, bio-imaging and biosensing
(Bacon et al., 2014 ;Shen et al., 2012 ). Literature data revealed few
papers regarding GQDs-based electrochemical biosensors. Zhao
et al. have designed an electrochemical biosensor based on the
interaction between GQDs and single-stranded DNA ( Zhao et al.,
2011 ), Muthurasu and Ganesh have also developed an enzymatic
biosensor by anchoring horseradish peroxidase on GQDs for thedetection of H
2O2(Muthurasu and Ganesh, 2014 ), Razmi and
Mohammad-Rezaei have immobilised glucose oxidase on GQDs
and have obtained a performant platform for sensitive detection of
glucose ( Razmi and Mohammad-Rezaei, 2013 ).
The present work wants to bring together for the first time in
thefield of biosensors MoS 2and GQDs as electrode modi fiers for
constructing a sensitive and reusable laccase based biosensor. The
conductivity of the carbon based screen-printed electrodes was
highly improved when both MoS 2nano flakes and GQDs were
added providing thus the host matrix for enzyme immobilisation.
The biosensor construction was firstly realised by drop casting the
carbon screen-printed electrodes (CSPEs) with successive addi-
tions of MoS 2and GQDs dispersions, and finally Trametes versicolor
laccase (TvL) was immobilised. It was observed that the used
materials enabled a better adsorption of the enzyme due to their
high surface-to-volume ratio and thus the developed CSPE
modi fied with MoS 2, GQDs and TvL based biosensor
(CSPE-MoS 2-GQDs-TvL) was successfully applied for the determi-
nation of polyphenol index in red wines.
2. Experimental section
2.1. Reagents and solutions
Laccase (benzenediol: oxygen oxidoreductase EC 1.10.3.2, ac-
tivity provided on the bottle Z10 U mg/C01) from Trametes versi-
color was purchased from Sigma-Aldrich (Germany) and stored at
–18°C. The exact laccase activity was determined according to the
data from the Supplementary Material. Caffeic acid, chlorogenic
acid, ( /C0) epicatechin, ABTS, potassium ferricyanide, Folin –Cio-
calteu reagent, D-( ț)-glucosamine hydrochloride, sodium carbo-
nate and potassium chloride were also purchased from Sigma-Aldrich (Germany). Polyphenols stock solutions were prepared in
0.1 M acetate buffer containing 10% ethanol (v/v), pH 5.00 and
more diluted standard solutions were prepared by suitable dilu-
tion in 0.1 M acetate buffer pH 5.00. All the reagents were of
analytical grade. Poly(ethyleneimine) solution 50% ( w/v)i nH 2O
was from Fluka and molybdenum disulphide (MoS 2) was supplied,
as nanoscale crystals dispersed in ethanol solution, by GrapheneSupermarket.
The supporting electrolyte consisted of 0.1 M acetate buffer pH
5.00 in 0.1 M KCl. Acetate buffer solutions (0.1 M) at various pH
values (ranging between 3.76 –6.00 as acetate p K
ais 4.75) were
used as supporting electrolytes during the biosensor optimisation
studies. High purity deionized water obtained from a Milli-Q
system (Millipore, France) has been used to prepare all the aqu-eous solutions.
Some of the tested red wine samples were kindly supplied by
the Institute of Research and Development for Viticulture and
Oenology Valea Calugareasca (ICDVV), Romania while others were
purchased from a local hypermarket in Bucharest (Romania). The
wine samples pretreatment consisted of 200 times dilution in
acetate buffer before biosensor analysis and 100 times in water forFolin –Ciocalteu assessment.
2.2. Instrumentation and electrodes
Spectrophotometric measurements were performed on a
Thermo Evolution 260 Bio (Thermo Fisher Scienti fic Inc., Rockford,
IL, USA). The micrographs were obtained using a Nova NanoSEM
630 (FEI Company, USA) Scanning Electron Microscope (SEM).
Transmission electron microscopy (TEM) images were obtained on
a Tecnai ™G2F30 S-TWIN transmission electron microscope op-
erated at 300 kV. Diffraction patterns were collected using the
9 kW rotating anode Rigaku SmartLab thin film triple-axis dif-
fraction system operated at U¼45 kV, I¼200 mA with a CuK
α
radiation ( λ¼1.541867 Å), in parallel beam mode. Fluorescence
emission spectra were recorded at room temperature using theFLS920 spectrometer equipped with a 450 W Xenon lamp as ex-
citation source. The absolute fluorescence quantum yield (QY) of
the GQDs solution was determined using the integrating sphere
setup. Zeta potential of the MoS
2and GQDs dispersions was
measured using the Delsa ™Nano analyser from Beckman Coulter,
USA.
The cyclic voltammetry, electrochemical impedance spectro-
scopy and chronoamperometry measurements were carried out
using a potentiostat AutoLab PGSTAT 302N (Utrecht, Netherlands)
that was equipped with GPES (4.9) Eco Chemie (Utrecht, Nether-
lands) software. The electrochemical cell consisted of carbon
screen-printed electrodes (DRP-C150, DropSens, Spain) the work-
ing electrode having 4 mm in diameter, the counter electrode is
made of Pt, while the pseudo-reference electrode and the electriccontacts are made of Ag, all printed onto a ceramic support
(3.4 cm /C21.0 cm). The CSPEs were connected to the potentiostat
using the speci fic connector (DRP-DSC) from DropSens.
2.3. Synthesis of graphene quantum dots (GQDs)
Graphene quantum dots (GQDs) were synthesized by a micro-
wave assisted single step hydrothermal route ( Mihalache et al.,
2014 ;Mihalache et al., 2015 ). In a typical procedure, 2.0 g of D-
(ț)-glucosamine hydrochloride, 3.0 g of poly(ethyleneimine) so-
lution and 2.5 mL of distilled water were thoroughly mixed, al-lowed to react during 4 h, at room temperature (24 °C). A volume
of 5 mL was introduced in the reaction vessel and during 60 s,
700 W were applied; the reaction mixture was allowed to cool
down at room temperature for 1 h. The reaction mixture was
further centrifuged at 20.000 rpm, during 1 h, at 10 °C, using aI. Vasilescu et al. / Biosensors and Bioelectronics 75 (2016) 232 –237 233

Beckman Coulter Avanti J-30I centrifuge. The supernatant was
collected and used for further experiments or stored at 4 °C. A
microwave acid digestion vessel (4781 type) provided by Parr In-
strument Company and a commercial microwave oven were used
during synthesis.
2.4. Preparation of CSPE-MoS2-GQDs-TvL biosensor
Carbon screen-printed electrodes from DropSens were used as
biosensor substrate due to their good conductivity, reproducibilityand low-cost. Firstly, 8 mL of MoS
2nano flakes suspension
(18 mg L/C01) were casted on the surface of the CSPE and allowed to
dry, followed by drop-casting of 6 mL of GQDs water based sus-
pension (1:10 v:v dilution from the collected supernatant). Finally,
TvL immobilisation was performed by casting 5 mL from a
58.23 U mL/C01laccase solution onto the modi fied working elec-
trode and allowed to dry in a dehumidi fied room at 24 °C. The
resulting CSPE-MoS 2-GQDs-TvL biosensor was stored at 4 °C.Fig. 1
describes the construction approach and the working principle of
the CSPE-MoS 2-GQDs-TvL biosensor, the nanocomposite formed
by MoS 2and GQDs providing an appropriate environment for
laccase immobilisation at the surface of the electrode, im-
mobilisation that occurred through electrostatic interaction be-
tween the negatively charged laccase and the positively chargedGQDs. Caffeic acid is oxidised to the corresponding quinone in the
presence of laccase, under acidic conditions and in the presence of
molecular oxygen, the quinone being further reduced (electro-
chemically) at the biosensor surface. The resulting current ob-
tained by quinone reduction is proportional to the caffeic acidconcentration.
2.5. Electrochemical measurements
All cyclic voltammetry (CV) and chronoamperometry (CA)
measurements were performed at room temperature (24 °C70.5),
in drop mode, the supporting electrolyte consisting of 0.1 M
acetate buffer pH 5.00 in 0.1 M KCl, and polyphenols or red winesamples were added in the electrochemical cell using a micro-
pipette. The CV experiments were recorded by cycling the poten-
tial between ț0.65 and /C00.30 V at a scan rate of 50 mV s
/C01. TheCA measurements were realized at 50 mV vs.Ag pseudo-reference
electrode, the calibration curves being performed by washing theelectrode between the additions of different substrateconcentrations.
To better understand the electrochemical interfacial behaviour
of the electrode, in the presence and in the absence of the modi-fiers, i.e. MoS
2, GQDs and laccase, electrochemical impedance
spectroscopy (EIS) studies were also performed, at frequenciesranging between 300 kHz and 0.01 Hz, with a voltage perturbationof 10 mV rms, and ț130 mV applied dc bias as formal redox po-
tential. The supporting electrolyte during EIS investigations was0.1 M KCl solution containing 2 mM [Fe(CN)
6]3/C0/4/C0.
3. Results and discussion
3.1. UV –vis,fluorescence, SEM, HR-TEM and XRD characterisation
Although absorption was recorded between 200 and 1100 nm,
strong optical absorption appeared only in the UV region with atail reaching the visible range. Fig. 2 (inset) shows the GQDs ab-
sorption spectrum, the main peak at 267 nm being assigned to
π–
π* transitions of C –O and aromatic C ¼C bonds present in the sp2
hybridisation region. Typically, graphene quantum dots display
broad bandwidths and excitation dependant photoluminescenceemissions as their main spectroscopic features, photo-
luminescence emission being the result of electron –hole pair re-
combination. Fig. 2 shows a broad bandwidth covering wave-
lengths mainly from visible domain (350 nm up to 650 nm).Photoluminescence emission had a constant bathochromic shiftwith a pronounced decrease in intensity as excitation was ex-tended to longer wavelengths. GQDs displayed a quantum yield of11%, measured at 340 nm excitation.
The alteration of the bare CSPE surface morphology ( Fig. S1A )
was achieved using MoS
2, GQDs and TvL. Scanning electron mi-
croscopy (SEM) indicated that MoS 2was uniformly distributed
(Fig. S1B ) on the surface of the electrode, while the presence of
graphene quantum dots was indicated by the appearance of fuz-ziness and darker areas ( Fig. S1C ). Laccase layer was evidenced in
Fig. S1D as a transparent membrane covering the modi fied elec-
trode. Graphene quantum dots HR-TEM ( Fig. S2 ) analysis provided
the lattice parameter of 3.3 Å corresponding to the [002] facet, in
agreement with the interlay distance of graphene sheets,
Fig. 1. Schematic representation of components used for CSPE-MoS 2-GQDs-TvL
biosensor construction and the detection principle.
Fig. 2. Fluorescence and UV –vis absorption (inset) spectra of GQDs.I. Vasilescu et al. / Biosensors and Bioelectronics 75 (2016) 232 –237 234

demonstrating the graphitic nature of the crystalline dots. Also the
lattice spacing of 2.4 Å was consistent with the (1120) lattice
fringes of graphene. GQDs diameters ranged between 2.8 and
4.3 nm, having an average diameter of 3.4 nm. As for the nano-
composite ( Fig. S3 ), beside graphene characteristic lattice para-
meters, an interlayer distance of 6.4 Å was evidenced, being as-
signed to (002) orientation in MoS 2. The XRD analysis ( Supple-
mentary Material, Fig. S4 ) indicated for MoS 2nano flakes an in-
creased interlayer distance ( Δc/co¼ț2.82%) and a decrease in
crystallinity, based on lower peak intensity at (002) orientation.
3.2. Electrochemical behaviour of CSPE-MoS2-GQDs-TvL biosensor
Zeta potential measurements have shown that MoS 2nano-
flakes were negatively charged with a zeta potential of /C030.4 mV,
while GQDs were positively charged with a zeta potential of
ț4.6 mV. Therefore, laccase that is negatively charged with an
isoelectric point around 3.5 ( Wu et al., 2014 ) can be immobilised
onto the GQDs surface by electrostatic interaction.
The recorded EIS spectra ( Fig. S5 ) were analysed by means of
three equivalent electrical circuits ( Fig. S6, Supplementary Mate-
rial). The characteristic semicircle ( Fig.S5A ) that appeared in the
high frequency range was ascribed to the charge-transfer re-
sistance ( Rct1) and the double layer capacity (CPE 1) occurring at
the interface between the electrode and the electrolyte. The Rct1
values were inversely proportional to the rate of electron transfer
and it can be observed that while a 1.5 increase was observed
when the MoS 2sheets were casted on the bare screen printed
electrode (SPE), a signi ficant dropping (approximately 30 times
less) was achieved when GQDs were used to modify the bare
electrode ( Table S1 ). The first behaviour indicates that there are
electrostatic repulsive interactions between negatively charged
MoS 2sheets and [Fe(CN) 6]3/C0/4/C0anions. On the contrary, albeit the
previous reported impedance analyses of the electrode interfacemodi fied with GQDs obtained via graphene oxide revealed their
insulating nature ( Mazloum-Ardakani et al., 2015 ), our results
confirmed that the network of GQDs obtained by microwave as-
sisted hydrothermal route supplies excellent electronic con-ductivity. A further decrease was achieved when both MoS
2and
GQDs were dropped on the SPE ( Fig. S5A , inset), the Rct1change
being more than two orders of magnitude, illustrating the sy-nergistic interaction between GQDs and MoS
2sheets.
Due to the electrode porosity, surface disorder, and adsorption
processes ( Jorcin et al., 2006 ), the capacitor element is replaced by
a constant phase element (CPE 1) to represent the non-ideal ca-
pacitor behaviour of the system. For GQDs modi fied SPE, the basic
circuit was modi fied introducing a second parallel C/T element,
related to an additional capacitance of the nanostructures –Fig. S6.
B. The tangent hyperbolic element (T element) describes the im-
pedance of finite-length diffusion processes with a re flective
boundary involving thin films with electroactive substances ( Soon
and Loh, 2007 ). It completes the classic Warburg diffusion beha-
viour modelling the diffusion process inside the porous nano-
composite film.
The presence of the laccase layer requested the insertion of a
new charge – transfer resistance ( Rct2) in parallel with a second
constant phase element (CPE 2) in the fitting circuit in order to
compensate the charge transfer involved in the polarisation pro-cess –Fig. S6C .
In order to get a better view about the nature of the mechanism
involved at the electrode, Bode impedance and Bode phase angle
plots were investigated. Fig. S5B and Cshows the frequency de-
pendence of the impedance module and the phase angle,respectively.
Fig. S5B showed a decrease in the impedance in the middle
range frequencies, which is an indication of an improvedconductivity of the modi fied electrodes. Furthermore, Bode phase
angle diagrams exhibited a more pronounced variation: while the
bare CSPE presents two time constants, with a phase angle max-imum of 32 °at high frequency of around 1000 Hz, and one at
0.01 Hz, when it was modi fied the maximum shifted to lower
frequencies ( Fig. S5C ).
The electrochemical characterisation of the electrode, in its
various states such as CSPE (a), CSPE-MoS
2(b), CSPE-GQDs (c) and
CSPE-MoS 2-GQDs (d) was accomplished by cyclic voltammetry
studies using [Fe(CN) 6]3/C0/4/C0(1 mM) in 0.1 M KCl solution at a scan
rate of 50 mV s/C01(Fig. 3 ). The presence of a couple of redox peaks
with a peak-to-peak separation ( ΔEp) of 100 mV was observed at
the surface of CSPE. After modi fication with MoS 2nano flakes the
ΔEpbecome 120 mV and the redox peak currents intensities
slightly decreased due to the repulsion existing between the ne-
gatively charged MoS 2and negatively charged redox probe, re-
spectively. When CSPE was modi fied with GQDs the ΔEpde-
creased to 66 mV and the redox peak current intensities also in-creased that is clearly related to the good conductivity of GQDs.Finally, at the CSPE-MoS
2-GQDs surface a ΔEpof 71 mV was ob-
tained and the redox peak currents were 2.2 higher than the redoxpeak currents obtained for CSPE. This behaviour may be attributed
to the good electrocatalytic activity, large surface area and high
conductivity of the MoS
2nano flakes and GQDs nanoassembly.
The interesting results obtained for MoS 2-GQDs modi fied
electrode could be ascribed to a large increase in the surface area,
resulting in an increase in real electro-active area (roughness
factor) and high conductivity of the modi fied electrode ( Table S2,
Supplementary Material ). The real electro-active area, Arealwas
calculated according to the following equation:
AA
A 1realRS
G=()−
where A R–Sis the area Randles –Sevčik and it was calculated from
cyclic voltammetry experiments in 1 mM [Fe(CN)]3/C0/4/C0solution
containing 0.1 M KCl, according to Randles –Sevčik equation ( Bard
and Faulkner, 2001 ) and AGis the geometric area of the DropSens
carbon working electrode, namely 12.56 mm2. The high real elec-
tro-active area obtained in the case of CSPE-MoS 2-GQDs modi fied
electrode makes it a good electrochemical support for TvL im-
mobilisation.
⎡
⎣⎢⎤
⎦⎥⎡
⎣⎢
⎢⎛
⎝⎜⎞
⎠⎟⎤
⎦⎥
⎥kDn v F
RTnF
RTEp Ep 2. 18 exp
201/2 2
ox red()()αα=− −
()
Fig. 3. Cyclic voltammograms of 1 mM [Fe(CN) 6]3/C0/4/C0in 0.1 M KCl on CSPE,
CSPE-MoS 2, CSPE-GQDs and CSPE-MoS 2-GQDs; scan rate 50 mV s/C01.I. Vasilescu et al. / Biosensors and Bioelectronics 75 (2016) 232 –237 235

Table S2 also reports the electron transfer rate constants (k0)
values for 1 mM [Fe(CN)]3/C0/4/C0calculated from cyclic voltammo-
grams applying Kochi's method ( Klingler and Kochi, 1981 ), see Eq.
(2) considering Dthe diffusion coef ficient of [Fe(CN) 6]3/C0/4/C0
(7.6/C210/C06cm2s/C01),αthe electron transfer coef ficient (0.5 for
reversible reaction), Ris the universal gas constant, Tis the ab-
solute temperature (K), nis the number of transferred electrons
(1) and Fis the Faraday constant. It is obvious that the electro-
chemical performances were improved with the modi fication of
the electrode surface, the highest k0values being obtained for
CSPE-MoS 2-GQDs.
Furthermore the voltammetric response of 2.5 mM caffeic acid
was measured at the differently modi fied electrodes having im-
mobilised laccase (300 mIU/electrode) in 0.1 M acetate buffer so-lution pH 5.00 at a scan rate of 50 mV s
/C01(Fig. 4 ). At the CSPE-TvL
modi fied electrode a pair of quasi-reversible peaks was observed,
the reduction peak being located at approximately 0 V with an Ipc
of 24.88 mA. When CSPE was modi fied with MoS 2nano flakes and
TvL, the reduction peak had an Ipcof 27.98 mA and it was present at
the same potential as in the case of CSPE-TvL. When the electrodewas modi fied with GQDs and TvL, the reduction peak position has
shifted to more positive values (0.11 V) and has considerably in-creased in current intensity, I
pcbeing 73.3 mA. And finally, when
the developed CSPE-MoS 2-GQDs-TvL biosensor was tested, the
reduction peak had an Ipcof 87.59 mA, 3.5 times higher than the
current intensity of CSPE-TvL electrode and also kept its position at0.11 V as in the case of CSPE-GQDs-TvL. The obtained results em-phasise once more that the optimum base for laccase based bio-sensor is the CSPE-MoS
2-GQDs con figuration, as con firmed by EIS
investigations.
The effect of the scan rate on the electrochemical response of
the CSPE-MoS 2-GQDs-TvL biosensor was also studied in the pre-
sence of 2.5 mM caffeic acid ( Fig. S7A ).Epashifted to more positive
values and Epcshifted to more negative values with the increase in
the scan rate. The anodic and cathodic potentials were in-dependent of the logarithm of the scan rate (log v)(Fig. S7B ) while
anodic and cathodic peak current intensities increased linearlywith the square root of the scan rate from 10 to 300 mV s
/C01(Fig.
S7C). The latter property indicated that the process occurring at
the modi fied electrode is a diffusion controlled process ( Bard and
Faulkner, 2001 ). However, the process is slightly in fluenced by the
weak adsorption of caffeic acid, the effect being seen in the heightof the anodic current intensity that is smaller when compared tothe cathodic current intensity ( Bard and Faulkner, 2001 ).3.3. Optimisation of the CSPE-MoS
2-GQDs-TvL biosensor experi-
mental parameters
The developed biosensor was optimised in terms of pH of the
used buffer and applied potential. Different buffer solutions having
pH values ranging between 3.76 and 6.00 were used and the cyclic
voltammograms were registered at a scan rate of 50 mV s/C01(Fig.
S8) using 2.5 mM caffeic acid. The cathodic peak potentials have
shifted to negative values when buffer solutions pH increases from
3.76 to 6.00. The obtained slope was /C065.7 mV/pH unit con firm-
ing the proton dependence of the substrate ( Fig. S8 inset ). The
highest current intensity value was achieved at pH 5.00. The
highest sensitivity of the biosensor in 0.1 M acetate buffer pH 5.00containing 5 mM caffeic acid was obtained for an applied potential
value of 50 mV vs.Ag pseudo-reference electrode, the enzyme
loading being previously optimised (300 mIU) ( Eremia et al.,
2013 ).
3.4. Polyphenols determination
The electrochemical response of the optimised
CSPE-MoS
2-GQDs-TvL biosensor toward caffeic acid was in-
vestigated using chronoamperometry at 50 mV vs.Ag pseudo-re-
ference electrode, the investigated concentrations ranging be-
tween 0.38 and 100 mM in 0.1 M acetate buffer pH 5.00 ( Fig. S9.A ).
As it can be observed from Fig. S9B , caffeic acid had two linearity
domains. Chlorogenic acid and ( /C0) epicatechin were also studied
using the developed biosensor as they are commonly found in red
wines together with caffeic acid.
The analytical parameters of the developed biosensor in the
determination of polyphenolic compounds, namely the linear re-
sponse ranges, sensitivity, correlation coef ficients, limits of de-
tection and calculated Kmappare shown in Table 1 . The proposed
biosensor showed linear response in the ranges of 0.38 –10.00 mM
and 10.00 –100.00 mM for caffeic acid, 0.38 –8.26 mM and 8.26 –
100.00 mM for chlorogenic acid and 2.86 –100.00 mM for ( /C0) epi-
catechin. The lowest limit of detection was obtained for chloro-genic acid while the highest for ( /C0) epicatechin. By comparing the
calculated K
mappfor the different substrates it was observed that
the best af finity was obtained for chlorogenic acid. The analytical
parameters of the proposed biosensor with respect to caffeic acid
were compared to those found in literature ( Table S3, Supple-
mentary Material ) and one can notice that the proposed biosensor
worked at positive applied potentials, had a wide linearity domain
and a very good limit of detection.
The reproducibility of the developed biosensor was checked
and it was observed that for seven consecutive measurements of
30mM caffeic acid solution the relative standard deviation was
710.06%. In the same time, when five different electrodes were
tested in the same conditions the RSD for electrode-to-electrode
Fig. 4. Cyclic voltammograms of 2.5 mM caffeic acid in 0.1 M acetate buffer pH 5.0
(with 0.1 M KCl) at CSPE-TvL, CSPE-MoS 2-TvL, CSPE-GQDs-TvL and
CSPE-MoS 2-GQDs-TvL; scan rate 50 mV s/C01.
Table 1
Analytical parameters of the CSPE-MoS 2-GQDs-TvL biosensor in the determination
of polyphenolic compounds.
Compound Linear range
(mM)Sensitivity
(nAmM/C01)R2LoDa
(mM)Kmapp
(mM)
Caffeic acid 0.38 –10.00 17.92 70.21 0.9992 0.32 11.25
10.00 –100.00 7.32 70.14 0.9982 ––
Chlorogenic acid 0.38 –8.26 11.25 70.36 0.9990 0.19 1.09
8.26–100.00 5.63 70.12 0.9972 –
(/C0) Epicatechin 2.86 –100.00 16.42 70.09 0.9997 2.04 121.80
aLoD: 3.3 /C2intercept standard error/sensitivity.I. Vasilescu et al. / Biosensors and Bioelectronics 75 (2016) 232 –237 236

reproducibility was 15.36%. The stability of the biosensor was also
evaluated and the enzyme retained its activity over a period of
4 weeks, an 85.00% of the initial current intensity was obtained.
These results indicated that the developed and optimisedCSPE-MoS
2-GQDs-TvL biosensor is reproducible and stable for the
analysis of caffeic acid.
3.5. Application to red wine samples
The possibility of applying the developed and optimised
CSPE-MoS 2-GQDs-TvL biosensor for measurements in real samples
was investigated by determining total polyphenolic content fromred wine samples. Before starting the polyphenol analysis ethanol(1:200, v:v in buffer), the main wine matrix was tested as inter-ference but neither inhibition nor activation of laccase occurred.The results obtained using the biosensor were expressed as caffeic
acid equivalents and compared with the results obtained using
Folin –Ciocalteu method as reference method. As it can be seen in
Table 2 the values obtained for total polyphenolic content by the
two methods are in good agreement excepting the Pinot Noir andFeteasca Neagra from Beciul Domnesc for which smaller valueswere obtained when using the proposed biosensor. Howeverwhen performing Anova test, the p-value corresponding to the
F-statistic of one-way Anova was 0.38 that is higher than 0.05suggesting that the values obtained by the two different methods
for the two red wine samples were not signi ficantly different and
therefore the developed biosensor is reliable for testing red winesamples.
4. Conclusions
A laccase based biosensor using the interesting electrochemical
and catalytic properties of MoS
2nano flakes and GQDs has been
developed and characterised. Laccase was successfully im-mobilised by electrostatic interactions at the surface of the
CSPE-MoS
2-GQDs modi fied electrode that is a challenging feature
for enzyme based biosensors. The developed and optimisedCSPE-MoS
2-GQDs-TvL biosensor exhibited wide linear ranges,
0.38–10mM (low concentration range) and 10 –100mM (high con-
centration range) with a LoD of 0.32 mM and a sensitivity of
17.92 nA mM/C01for caffeic acid, 0.38 –8.26 mM (low concentration
range) and 8.26 –100.00 mM (high concentration range) with a LoD
of 0.19 mM and a sensitivity of 11.25 nA mM/C01for chlorogenic acid
and 2.86 –100.00 mM with a LoD of 2.04 mM and a sensitivity of
16.42 nA mM/C01for (/C0) epicatechin. Moreover, the developed
biosensor proved good stability and reproducibility in the de-termination of the polyphenolic compounds. Based on these ob-servations, the combination of MoS
2and GQDs may represent thephysical foundation for the successful development of novel
biosensors.
Acknowledgement
This work was financially supported by a grant of the Romanian
National Authority for Scienti fic Research, CNDI –UEFISCDI, Project
number PN-II-PT-PCCA-2011-3.1-1809 and COST Action FA1403.
Appendix A. Supplementary material
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.bios.2015.08.051 .
References
Bacon, M., Bradley, S.J., Nann, T., 2014. Part. Part. Syst. Charact. 31, 415 –428.
Bard, A.J., Faulkner, L.R., 2001. Electrochemical Methods: Fundamentals and Ap-
plications. John Wiley and Sons, New York .
Brondani, D., de Souza, B., Souza, S., Neves, B., Vieira, C.I., A., 2013. Biosens. Bioe-
lectron. 42, 242 –247.
Cortina-Puig, M., Muñoz-Berbel, X., Calas-Blanchard, C., Marty, J.-L., 2010. Micro-
chim. Acta 171, 187 –193.
Eremia, S.A.V., Vasilescu, I., Radoi, A., Litescu, S.-C., Radu, G.-L., 2013. Talanta 110,
164–170.
García-Guzmán, J.J., Hernández-Artiga, M.P., Palacios-Ponce De León, L., Bellido-
Milla, D., 2015. Food Chem. 182, 47 –54.
Güçlü, A.D., Potasz, P., Korkusinski, M., Hawrylak, P., 2014. Graphene Quantum Dots.
Springer, Berlin, Heidelberg .
Jorcin, J.-B., Orazem, M.E., Pébère, N., Tribollet, B., 2006. Electrochim. Acta 51,
1473 –1479 .
Karimi-Maleh, H., Biparva, P., Hatami, M., 2013. Biosens. Bioelectron. 48, 270 –275.
Karimi-Maleh, H., Tahernejad-Javazmi, F., Ensa fi, A.A., Moradi, R., Mallakpour, S.,
Beitollahi, H., 2014. Biosens. Bioelectron. 60, 1 –7.
Klingler, R.J., Kochi, J.K., 1981. J. Phys. Chem. 85, 1731 –1741 .
Kudanga, T., Nyanhongo, G.S., Guebitz, G.M., Burton, S., 2011. Enzyme Microb.
Technol. 48, 195 –208.
Lanzellotto, C., Favero, G., Antonelli, M.L., Tortolini, C., Cannistraro, S., Coppari, E.,
Mazzei, F., 2014. Biosens. Bioelectron. 55, 430 –437.
Liu, C.-H., Chang, Y.-C., Norris, T.B., Zhong, Z., 2014. Nat. Nanotechnol. 9, 273 –278.
Mak, K.F., Lee, C., Hone, J., Shan, J., Heinz, T.F., 2010. Phys. Rev. Lett. 105 (Article
number 136805) .
Mao, K., Wu, Z., Chen, Y., Zhou, X., Shen, A., Hu, J., 2015. Talanta 132, 658 –663.
Mazloum-Ardakani, M., Aghaei, R., Abdollahi-Alibeik, M., Moaddeli, A., 2015. J.
Electroanal. Chem. 738, 113 –122.
Mihalache, I., Radoi, A., Mihaila, M., Munteanu, C., Marin, A., Danila, M., Kusko, M.,
Kusko, C., 2015. Electrochim. Acta 15, 306 –315.
Mihalache, I., Radoi, A., Munteanu, C., Kusko, M., Kusko, C., 2014. Org. Electron. 15,
216–225.
Muthurasu, A., Ganesh, V., 2014. Appl. Biochem. Biotechnol. 174, 945 –959.
Nazari, M., Kashanian, S., Ra fipour, R., 2015. Spectrochim. Acta Part A 145, 130 –138.
Oliveira, T.M.B.F., Fátima Barroso, M., Morais, S., De Lima-Neto, P., Correia, A.N.,
Oliveira, M.B.P.P., Delerue-Matos, C., 2013. Talanta 106, 137 –143.
Pang, S., Hernandez, Y., Feng, X., Müllen, K., 2011. Adv. Mater. 23, 2779 –2795 .
Radoi, A., Litescu, S.-C., Eremia, S.A.V., Miu, M., Danila, M., Dinescu, A., Radu, G.-L.,
2011. Microchim. Acta 175, 97 –104.
Rawal, R., Chawla, S., Devender, Pundir, C.S., 2012. Enzyme Microb. Technol. 51,
179–185.
Razmi, H., Mohammad-Rezaei, R., 2013. Biosens. Bioelectron. 41, 498 –504.
Schwierz, F., 2010. Nat. Nanotechnol. 5, 487 –496.
Shen, J., Zhu, Y., Yang, X., Li, C., 2012. Chem. Commun. 48, 3686 –3699 .
Shraddha, Shekher, R., Sehgal, S., Kamthania, M., Kumar, A., 2011. Enzyme Res. 2011
(Article number 217861) .
Song, H., Ni, Y., Kokot, S., 2014. Biosens. Bioelectron. 56, 137 –143.
Soon, J.M., Loh, K.P., 2007. Electrochem. Solid State Lett. 10, 250 –254.
Wang, G.-X., Bao, W.-J., Wang, J., Lu, Q.-Q., Xia, X.-H., 2013a. Electrochem. Commun.
35, 146 –148.
Wang, T., Zhu, H., Zhuo, J., Zhu, Z., Papakonstantinou, P., Lubarsky, G., Lin, J., Li, M.,
2013b. Anal. Chem. 85, 10289 –10295 .
Wang, Zhiming M., 2014. In: Zhiming M. Wang (Ed.), Lecture Notes in Nanoscale
Science and Technology. Springer, vol. 21.
Wu, Y., Jiang, Y., Jiao, J., Liu, M., Hu, F., Grif fiths, B.S., Li, H., 2014. Colloids Surf. B 114,
342–348.
Zhao, J., Chen, G., Zhu, L., Li, G., 2011. Electrochem. Commun. 13, 31 –33.
Zhu, C., Zeng, Z., Li, H., Li, F., Fan, C., Zhang, H., 2013. J. Am. Chem. Soc. 135,
5998 –6001 .Table 2
Comparison of total polyphenolic content obtained for five red wines with the
CSPE-MoS 2-GQDs-TvL biosensor by chronoamperometry in 0.1 M acetate buffer pH
5.00 at a fixed potential of 50 mV vs.Ag pseudo-reference electrode and total
polyphenolic content obtained using Folin –Ciocalteu method.
Wine sample CSPE-MoS 2-GQDs-TvL
biosensor (mM)Folin –Ciocalteu
method (mM)
Negru Aromat ICDVV 2013 18.93 71.15 17.09 70.07
Feteasca Neagra ICDVV
201411.2570.88 12.52 70.75
Cabernet Sauvignon Beciul
Domnesc 200715.5071.93 16.69 70.17
Pinot Noir Beciul Domnesc
200910.0470.96 14.94 70.26
Feteasca Neagra Beciul
Domnesc 200915.9170.89 18.79 70.14I. Vasilescu et al. / Biosensors and Bioelectronics 75 (2016) 232 –237 237