Bioelectrochemistry 101 (2015) 5865

Contents lists available at ScienceDirect

Bioelectrochemistry
journal homepage: www.elsevier.com/locate/bioelechem

Graphitized carbon nanoberPt nanoparticle hybrids as sensitive tool

for preparation of screen printing biosensors. Detection of lactate in
wines and ciders
Oscar A. Loaiza a,, Pedro J. Lamas-Ardisana a, Larraitz Aorga a, Elena Jubete a, Virginia Ruiz a, Maryam Borghei b,
Germn Cabaero a, Hans J. Grande a
a
Sensors Unit, Materials Division, IK4-CIDETEC, Parque Tecnolgico de San Sebastin, 20009 Donostia-San Sebastin, Spain
b
Department of Applied Physics, Aalto University School of Science, P.O. Box 15100, 00076 Aalto, Finland

a r t i c l e i n f o a b s t r a c t

Article history: This work describes the fabrication of a new lactate biosensor. The strategy is based on the use of a novel hybrid
Received 8 May 2014 nanomaterial for amperometric biosensors i.e. platinum nanoparticles (PtNps) supported on graphitized carbon
Received in revised form 1 July 2014 nanobers (PtNps/GCNF) prepared by chemical reduction of the Pt precursor at GCNF surfaces.
Accepted 6 July 2014
The biosensors were constructed by covalent immobilization of lactate oxidase (LOx) onto screen printed carbon
Available online 27 July 2014
electrodes (SPCEs) modied with PtNps (PtNps/GCNFSPCEs) using polyethyleneimine (PEI) and glutaraldehyde
Keywords:
(GA). Experimental variables concerning both the biosensor design and the detection process were investigated
PtNps for an optimal analytical performance. Lactate biosensors show good reproducibility (RSD 4.9%, n = 10) and
GCNF sensitivity (41,302 546) A/M cm2, with a good limit of detection (6.9 M). Covalent immobilization of the
Lactate biosensor enzyme allows the reuse of the biosensor for several measurements, converting them in a cheap alternative to
Screen printing electrodes the solid electrodes. The long-term stability of the biosensors was also evaluated. 90% of the signal was kept
Beverages after 3 months of storage at room temperature (RT), while 95% was retained after 18 months at 20 C.
These results demonstrate that the method provides sensitive electrochemical lactate biosensors where the
stability of the enzymatic activity can be preserved for a long period of time in adequate storage conditions.
2014 Elsevier B.V. All rights reserved.

1. Introduction elevated lactate level in blood is a major indicator of ischemic conditions

Lactic acid or lactate detection has an important role in several elds.
D- and L-lactic acid are found in many foods, beverages and fermented
products such as yoghurt, butter, cheese wine and ciders [17]. An
increase in L-lactate concentration in eggs is an indicator of spoilage
by contamination or incubation. Contamination of fruit juices with lactic
acid producing bacteria often remains unnoticed for a long time,
allowing the bacteria to spread and infect huge volumes of juice [8].
On the other hand, alcoholic beverage production includes alcoholic
fermentation conducted by yeast and a secondary fermentation
performed by lactic acid bacteria, called malolactic fermentation. During
this process, L-malic acid is transformed into L-lactic acid and CO2. This
transformation inuences the quality and taste of the beverage. In
addition to the deacidication, the malolactic fermentation is con-
sidered to contribute in the complexity of the avor, the nal taste
and the microbiological stability of the beverage.
L-lactate concentration in blood is essential for the diagnosis of
patient conditions in intensive care and during surgery [911]. An
Corresponding author. Tel.: +34 943309022; fax: +34 943309136.
E-mail address: oloaiza@cidetec.es (O.A. Loaiza).
of the respective tissue. The ischemic situation can be caused by all
types of shock, suffocation and respiratory insufciency. Another reason
for an altered lactate level is a disturbed lactate metabolism which may
be caused by diabetes. In sport medicine, blood lactate levels during
exercise are an indicator for training status and tness [1214]. During
the demand of high intensity exercise, the cell utilizes a substantial
amount of glucose and glycogen. The product of the anaerobic glucose
breakdown is lactate. This increase of lactate coincides with an increase
in blood and muscle acidosis. Therefore lactate is an excellent indirect
marker of cellular fatigue. Lactate testing is considered to be the single
most important determinant of success in endurance related activities.
The importance of lactate determination has generated much inter-
est in developing lactate biosensors. Some techniques including auto-
mated lactate analyzers, HPLC or UV methods [10,12,15] have been
developed for the detection method of lactate. Additionally, screen
printed amperometric biosensors based on enzymes have been general-
ly considered a simple, inexpensive and sensitive detection of lactate in
aqueous solution [1,1621]. Screen printed electrodes (SPEs) have over-
come serious drawbacks of conventional disc electrodes such as the
need of surface regeneration after each measurement [22,23], which
prevents them from being used in routine analysis. In the last decade,

http://dx.doi.org/10.1016/j.bioelechem.2014.07.005
1567-5394/ 2014 Elsevier B.V. All rights reserved.
 O.A. Loaiza et al. / Bioelectrochemistry 101 (2015) 5865
there have been a great number of reports on electrochemical lactate
biosensors based on screen printed technologies [2426].
Lactate biosensors can be based on two enzymes, either lactate oxi-
dase (LOx) [1,1618,20,21] or lactate dehydrogenase (LDH) [3,19,27].
However, the latter requires the presence of cofactor NAD+ which com-
plicates analysis and increases the cost. Moreover, a supplementary en-
zyme, glutamate pyruvate transaminase, is necessary for an efcient
reaction [28], thus increasing the analysis price. LOx, the most widely
used enzyme, catalyzes the conversion of lactate to pyruvate and
H2O2, which is subsequently detected by amperometry. However,
these detections are hindered in analytical applications by slow elec-
trode kinetics and high overpotentials which may result in large inter-
ferences from other electroactive species in real samples. The current
research on H2O2 detection is mainly focused on electrode modica-
tions in order to improve those overpotentials and electron transfer
kinetics [2931].
It is well established that electro-catalytic properties of platinum
nanoparticles (PtNps) are better than those of bulk platinum, due to
their large specic surface area, which results in several electro-
analytical advantages [32]. This characteristic is strongly dependent of
the size, shape and distribution of PtNps on the electrode [33,34]. In
addition to their higher mass specic activity, PtNps deposited onto
SPE electrodes offer several advantages compared to bulk Pt electrodes
such as mass production and low cost (disposability).
Carbon nanostructures, particularly carbon nanotubes (CNTs) and
carbon nanobers (CNFs) have become very popular electrode mate-
rials for electrochemical sensing due to their unique properties such as
high conductivity, very large surface area, easy functionalization, bio-
compatibility, fouling resistance and high electrocatalytic activity to-
wards many important electron transfer reactions. In fact, decoration
of these carbon nanostructures with metal nanoparticles has attracted
huge interest in biosensor research. The capability to promote electron
transfer reactions at low overpotentials has motivated extensive cou-
pling of these hybrid nanomaterials [35] with enzymes in biosensors
[31,3639]. In particular, Pt is widely employed in electrochemical bio-
sensors due to its excellent electrocatalytic activity towards H2O2 oxida-
tion/reduction which enables the amperometric detection of H2O2 at
low overpotentials. The use of CNFs in biosensors is less common than
CNTs although the usefulness of this family of nanomaterials has been
recently reviewed [40]. CNFs decorated with PtNp (PtNpCNF) have
been used for biosensing applications [41] although these nanohybrids
have been more extensively studied in fuel-cell research due to the
high electrochemical stability of the CNFs [42,43].
In this paper a stable and sensitive lactate biosensor, based on the
use of PtNps/GCNFSPCE, is reported. The biosensor was constructed
by covalent immobilization of LOx onto the PtNps/GCNFSPCEs using
polyethyleneimine (PEI) and glutaraldehyde (GA). The results concer-
ning the structural and electrochemical characterization of the PtNps/
GCNF are presented and discussed. In addition, the effect of each com-
ponent in the biosensor design is studied. Furthermore, the analytical
performance of the developed lactate biosensor is determined in
terms of sensitivity, reproducibility and stability. Finally, the behavior
of the sensor for lactate determination in Basque ciders and certied
wines was evaluated. Data were compared with results of a reference
method, obtaining an excellent agreement.
2. Experimental
2.1. Reagents and solutions
Ethylene glycol, ethanol absolute, hydrogen peroxide 30% (w/v),
ortho-phosphoric acid 85% (w/w) and glycerol were purchased from
Scharlab (Sentmenat, Spain). Glacial acetic acid was purchased from
Panreac. Sodium L-lactate, poly(diallyldimethyl ammonium chloride)
solution (PDDA) 20% (w/w) in water (Mw 100,000200,000),
polyethyleneimine branched (PEI) (Mw ~ 25,000), glutaraldehyde
59
(GA) 50% (w/w) in water, chloroplatinic acid hydrate 99.9% (w/w),
sulphuric acid, succinic acid, D-()-fructose, (+)-catechin, caffeic acid,
L-()-malic acid, D-sorbitol, naon peruorinated resin solution 20%
(w/w) in lower aliphatic alcohols and chitosan (Mw 110,000150,
000) were purchased from Sigma-Aldrich (Madrid, Spain) and used as
received.
Carbon nanobers were kindly supplied by Showa Denko (Tokyo,
Japan). L-lactate oxidase (LOx) 91.2 units/mg from Pediococcus sp. ly-
ophilized powder was purchased from Sorachim S.A. (Lausanne,
Switzerland). 20 l LOx stock solutions (nominally 1 U/l) were pre-
pared in 0.1 M phosphate buffer pH = 7.0 solution (0.1 M PB, pH =
7.0) and kept frozen until used. Diluted solutions of the enzyme were
prepared in the same buffer solution.
Unless stated otherwise, all solutions were prepared with ultrapure
water of Synthesis A10 from Millipore (18.2 cm) (Billerica, MA, USA).
Wine samples with certied content of lactate were supplied from the
Centre Oenologique de Bourgogne. A L-lactic acid spectrophotometric-
enzymatic kit (Megazyme) for the determination of lactate in cider
was also used.
2.2. Apparatus and electrodes
All electrochemical measurements were carried out with an Eco
Chemie Autolab PGSTAT 128N potentiostatgalvanostat (KM Utrecht,
The Netherlands), using the software package NOVA 1.9. SPCEs were
printed on a plastic substrate using a Thieme 110E screen printing ma-
chine, following a three electrode conguration. The SPCEs are com-
posed of a carbon counter and disc working electrode ( = 4.4 mm)
and a silver/silver chloride (Ag/AgCl) pseudo-reference electrode. The
potential values were referred to the screen printed Ag/AgCl pseudo-
reference electrode unless otherwise stated. Sensitivity of the calibra-
tion plots was related to the geometrical area of the working electrodes
i.e. 15.2 mm2.
Field Emission Scanning Electron Microscopy (FESEM) images were
acquired with a JEOL JSM-5500LV. The diameters of the PtNps in the
SEM images were determined with Digital Micrograph (TM) 3.7.0
(Gatan Inc.). Transmission Electron Microscopy (TEM) images were
acquired from a Tecnai 12 Bio Twin with a 120 kV LaB6 gun (Oregon,
USA). Thermogravimetric balance model Q500-TGA TA Instruments
(Delaware, USA), a UNE 200 oven from Memmert (Wisconsin, USA), a
pH meter GLP 2 from Crison (Barcelona, Spain), a microltration vacu-
um system from Scharlab (Sentmenat, Spain), polytetrauoroethylene
(PTFE) membrane discs with mean pore size of 0.22 mm from Millipore
(Massachusetts, USA), a microwave model LG 700 W19 L Touch Con-
trol from LG (Madrid, Spain) and an ultrasonic bath model Ultrasons
HD-5L from J.P. Selecta (Barcelona, Spain) were also employed.
2.3. Procedures
2.3.1. Preparation of PtNps/GCNF
Hybrid nanomaterial was prepared according to a previously report-
ed procedure [30]. Briey, highly graphitized nanobers with a diame-
ter of ~ 150 nm and 13 m2/g surface area were oxidized by reuxing
them in a 1:1 mixture of HNO3 2 M and H2SO4 1 M at 120 C for 6 h.
The oxidized GCNF was ltered through a PTFE membrane disc and
washed with water. PtNps were then deposited on GCNFs by the
microwave-assisted heating polyol reduction of the metal precursor.
Briey, appropriate amount of H2PtCl6, NaOH, ethylene glycol and
GCNFs was stirred for 20 min and sonicated for 20 min. After heating
in microwave for 1 min, PtNp/GCNF hybrids were cooled down, ltered,
washed and dried.
2.3.2. Characterization of PtNps/GCNF and PtNps/GCNFSPCEs
The morphological characterization of PtNps/GCNFs was carried out
by SEM, TEM and TGA. PtNp size distribution on GCNFs was examined
by TEM. The samples were prepared by ultrasonic dispersion of PtNps/
60 O.A. Loaiza et al. / Bioelectrochemistry 101 (2015) 5865
Fig. 1. (a, b) FESEM and (c, d) TEM micrographs of the PtNp/GCNF nanoparticles.
GCNFs in isopropanol and drop casting on copper TEM grids. The metal-
lic content was determined thermogravimetrically by TGA and with
SEMEDAX analysis.
2.3.3. Preparation of PtNps/GCNFSPCEs
The PtNps/GCNFs were weighed in a glass vial and dispersed in 1%
PDDA (prepared in 0.1 M PB, pH = 7.0). The suspension was immersed
in an ultrasonic bath until a homogeneous dispersion was obtained.
Then 4 l of this dispersion was dropped on the SPCE surface, previously
casted with a 1.5 l drop of a 1:1 ethanol:water solution. The drop was
left to dry for 1 h at room temperature (RT) (24 C unless stated other-
wise). Finally the PtNps/GCNFSPCEs were heated at 80 C for 5 min to
enhance the adhesion of the nanobers on the electrodic surface.
2.3.4. Enzyme immobilization
Immobilization of LOx enzyme was carried out following a reported
method with slight modications [2]. Firstly, 20 l of PEI 1% (w/v) in 0.1
M PB, pH = 7.0 was placed on the active area of the PtNps/GCNFSPCEs
for 1 min (PtNps/GCNFPEI). Then the modied electrode was rinsed
with 0.1 M carbonate buffer, pH = 10.2 and dried. After that, 20 l of
GA 1% (w/v) in 0.1 M carbonate buffer, pH = 10.2, was added for
5 min (PtNps/GCNFPEIGA). Subsequently, the electrode was washed
with 0.1 M PB, pH = 7.0. The working electrode was covered with a 2 l
drop of LOx solution 0.5 (U/l) and dried for 30 min at RT (PtNps/GCNF
PEIGALOx). Finally, the biosensor was washed with 0.1 M glycine
(Gly) in water (PtNps/GCNFPEIGALOxGly) and stored at 20 C
or RT until used.
2.3.5. Electrochemical measurements
All amperometric measurements were carried out by adding a 60 l
drop of the corresponding solution to the biosensor. For amperometric
measurements the potential was xed at + 300 mV and the signals
were recorder for 60 s. The supporting electrolyte consisted of 0.1
M PB, pH = 7.0. Cyclic voltammograms were recorded at 100 mV s1.
All measurements were carried out at 20 C unless stated otherwise
2.3.6. Real sample analysis
In order to evaluate the applicability of the developed biosensors to
real samples, several ciders and wines were analyzed. The set consisted
of a homemade cider (Olaziregi), 2 commercial ciders (Mina and Zelaia)
and 4 wines with certied content of lactate: 2 red, 1 white and 1 ros.
The sample treatment consisted only of an appropriate dilution in order
to t the concentration of lactate in the ciders and wines to the specied
concentration range. Next, 60 L of the diluted sample was added to the
biosensor and the amperometric measurement was carried out by
applying the desired potential. Lactate concentration was calculated
by interpolation of the corresponding amperometric signal into a cali-
bration graph constructed with standard solutions of sodium lactate.
3. Results and discussion
3.1. Characterization of the PtNps/GCNFSPCEs
The morphology and composition of the PtNps/GCNFSPCEs were
characterized by FESEM, TEM and TGA. Fig. 1 shows FESEM and TEM mi-
crographs from the PtNps/GCNF nanohybrid material. No agglomera-
tions are observed and the PtNps/GCNFs used contain 17 wt.% of Pt
(calculated by TGA). FESEM images (Fig. 1a and b) presented well dis-
tributed PtNps of very uniform size. The average particle diameter was
2.8 nm, with around 95% of the particles being between 2 and 4 nm
(see Fig. S1). The use of ethylene glycol as dispersing agent allows a uni-
form distribution of PtNps with small particle size.
 O.A. Loaiza et al. / Bioelectrochemistry 101 (2015) 5865 61

80 30
(2)
(2) (b)
(a)
60
20

I/ A
I/ A

40

10

20
(1)
(1)
0
0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0.0 0.2 0.4 0.6 0.8
E/ V (vs. Ag/AgCl pseudo-reference) E/ V (vs. Ag/AgCl pseudo-reference)

8 (c)
(7)
(6)
6 (5)
(4)
I/ A

(3)
4
(2)

0 (1)

-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6

E/ V (vs. Ag/AgCl pseudo-reference)

Fig. 2. (a) Cyclic voltammograms of bare SPCEs and (b) PtNPs/GCNFSPCEs. (1). (0.1 M PB, pH = 7.0), (2). (1 mM H2O2 in 0.1 M PB, pH = 7.0). (c) Amperometric signals at different
potentials obtained for PtNPs/GCNFSPCEs. 1 mM H2O2 in 0.1 M PB, pH = 7.0. (1) 0 g PtNPs/GCNF; (2) 1 g PtNPs/GCNF; (3) 2 g PtNPs/GCNF (4) 3 g PtNPs/GCNF (5) 4 g PtNPs/
GCNF (6) 5 g PtNPs/GCNF (7) 8 g PtNPs/GCNF.

3.2. PtNps/GCNFSPCE as H2O2 sensor the different polymer/PtNps/GCNF solutions. The highest sensitivity
(data not shown) was obtained with the PPDA and this dispersant
Sensitive and selective detection of H2O2 is of huge interest in many was chosen for further work.
elds including analytical and bioanalytical chemistry. Besides being an
important analyte by itself, it is one of the most important products of 3.2.2. Characterization of PtNps/GCNFSPCEs
enzyme-catalyzed oxidation reactions. Hence peroxide determination Electrochemical characterization of PtNps/GCNFSPCEs was carried
could be used to monitor several biomolecules of interest in clinical out by CV. Fig. 2a and b shows the CVs recorded for a bare SPCE and a
and food elds such as amines, uric acid, glucose, glutamate, lactate, PtNps/GCNFSPCE in a 0.1 M PB, pH = 7.0 containing 1 mM H2O2 re-
cholesterol and alcohols [44]. In this sense, the detection of H2O2 with spectively. As it can be seen, lower overpotential for H2O2 oxidation
the PtNps/GCNFSPCEs was evaluated as a previous step for the devel- was obtained when a PtNps/GCNFSPCE was used. Moreover, the anod-
opment of lactate biosensors. ic current corresponding to the oxidation of H2O2 begins at approxi-
mately + 0.1 V (versus Ag/AgCl pseudoreference electrode), which
allows working at lower potentials. In addition, control experiments
3.2.1. Optimization of PtNps/GCNFSPCE lm were carried out using SPCEs modied with oxidized GCNF/PDDA and
Polymer wrapping was chosen as simple and effective method for SPCEs modied with PDDA. The oxidation signals of H2O2 were similar
the homogeneous dispersion of the PtNps/GCNF. Different polymers: in both cases.
chitosan 0.1% (w/v) in acetic acid 1% (v/v), naon 1% (w/v) in ethanol: The shapes of the CVs were similar to those obtained with the bare
water (1:1) and PDDA 1% (w/v) in 0.1 M PB, pH = 7.0 were evaluated SPCE, although the background currents increased signicantly and
to obtain optimum analytical performance. Thus H2O2 calibration plots the potential shifted 0.5 V in the cathodic direction. From these
in the 0.010.75 mM range were constructed for SPCE modied with results, it could be concluded that the main electrocatalytic effect for
H2O2 detection could be attributed to the Pt present in the PtNps/
Table 1 GCNF nanohybrid material. Fig. 2c shows the response for the oxidation
Optimized working conditions and variables range tested with the PtNps/GCNFPEIGA of H2O2 at the PtNps/GCNFSPCE modied with different amounts of
LOxGlySPCE biosensor.
PtNps/GCNF. As it can be seen, the oxidation current increases with
Variable Range tested Selected value the amount of PtNps/GCNF until 4 g, leveling off for higher PtNp/
PEI concentration (04) % (w/v) 1% (w/v) GCNF amounts. Moreover, the current rises with the applied potential
PEI accumulation time (130) min 1 min being leveled off from +0.3 V. Higher potentials showed no signicant
GA concentration (0.24) % (w/v) 1% (w/v) increase in the amperometric signal, indicating that this was the poten-
LOx amount (0.15) U 1U
tial to reach an optimal signal. Hence, +0.3 V was chosen as the poten-
Gly concentration (0.050.5) M 0.1 M
tial applied for further measurements.
62 O.A. Loaiza et al. / Bioelectrochemistry 101 (2015) 5865

6
(a) (b)
6
5

4 4
I/ A

I/ A
3
2
2

0
1

0
1 2 3 4 5 0.0 0.2 0.4 0.6
Configuration E/ V (vs. Ag/AgCl pseudo-reference)

Fig. 3. (a) Currents obtained for different biosensor congurations: PtNPs/GCNFPEIGALOxGly (1), PtNPs/GCNFPEIGALOx (2), PtNPs/GCNFPEILOxGly (3), PtNPs/GCNFGA
LOxGly (4), PtNPs/GCNFLOxGly (5). Eapp: +0.3 V. (Error bars n = 10). (b) Amperometric signals at different potentials for 1 mM lactate in 0.1 M PB, pH = 7.0 with the PtNPs/
GCNFPEIGALOxGly biosensor. (Error bars n = 5).

3.2.3. Analytical characteristics of PtNps/GCNFSPCEs for H2O2 detection
Different aspects were evaluated regarding the use of PtNps/GCNF
SPCEs as sensors for H2O2 detection. Firstly, repeatability of the electro-
chemical response was checked. For this purpose, 10 successive amper-
ometric measurements for 2.5 mM H2O2 solution were performed with
the same electrode. The electrode was cleaned in between measure-
ments with 0.1 M PB, pH = 7.0. Moreover, the experiments were
repeated with ten different sensors. In these conditions, relative stan-
dard deviation (R.S.D.) of 4.5% was obtained indicating a good stability
and reusability of the PtNps/GCNFSPCEs.
Secondly, the reproducibility of the PtNps/GCNFSPCEs was tested.
For this, a linear calibration plot was constructed for H2O2 in the 0.01
0.75 mM range. The R.S.D. of the slopes obtained for ten different
PtNps/GCNFSPCEs constructed in the same manner was used as crite-
rion. The obtained value (0.4%) demonstrates the reliability of the sen-
sor construction procedure, allowing high reproducible amperometric
responses with different sensors.
Finally, a linear calibration plot (Fig. S2) was constructed for the
PtNps/GCNFSPCEs in the 125,000 M H2O2 range with a slope value
of (45,928 2105) A/M cm2 and an intercept of 0.5 A (r2 = 0.999).
A low detection limit of 0.5 M was calculated according to the 3 s/m
criterion, where m is the slope of the calibration plot and s was estimat-
ed as the standard deviation (n = 10) of the amperometric signals
recorded for 1 M H2O2 concentration.
3.3. PtNps/GCNFPEIGALOxGlySPCE biosensor
3.3.1. Optimization of the biosensor conguration
In the proposed conguration, GA cross links the absorbed PEI layer
via the method of mediated covalent layer-by-layer (LbL) assembly
[45]. This bound introduces free aldehyde group on the surface for the
next covalent LOx adsorption. Glycine was used to block any unreacted
aldehyde group from GA. Different experimental variables were opti-
mized in order to achieve the highest sensitivity of the biosensor.
Obtained results are summarized in Table 1. On one hand, different con-
gurations of lactate biosensors were tested in order to select the best
conguration in terms of signal stability. Fig. 3a shows the currents ob-
tained from the signals registered at + 0.3 V for 1 mM lactate in each
checked conguration. Error bars correspond to 10 consecutive mea-
surements. As it can be seen, the highest and most stable response is ob-
tained for the PtNps/GCNFPEIGALOxGlySPCE biosensor (1) and
the lowest ones for the PtNps/GCNFLOxGly (5), followed by PtNps/
GCNFGALOxGly (4) and PtNps/GCNFPEILOxGly (3) respectively.
These results indicate that the PEI and GA layers are necessary to retain
the enzyme and get more sensitive and stable signals. Moreover, only
the PtNps/GCNFPEIGALOxGly conguration retains 90% of the
amperometric signal after 50 measurements.
The inuence of the applied potential on the amperometric signal
was studied in order to get the best response at the lowest applied

16 (a) (b)
16
2.0 mM

12
12
I/ A
I/ A

8 8

4 4

0 mM
0
0
10 20 30 40 50 60 70 80 0 2 4 6 8 10
t/ s Clactate/ mM

Fig. 4. (a) Currenttime plot of lactate obtained with PtNPs/GCNFPEIGALOxGlySPCE biosensor. (b) Calibration curve of lactate obtained with the biosensor. (Error bars, n = 5 dif-
ferent electrodes). Eapp = +0.3 V, t = 60 s.
 O.A. Loaiza et al. / Bioelectrochemistry 101 (2015) 5865 63

56 56

48
(a) (b)
48

Slope/ (mA/Mcm )
2 40 40

32 32

24 24

16 16

8 8

0 0
0 20 40 60 80 100 120 0 100 200 300 400 500 600 700
t/ days t/ days

Fig. 5. (a) Control chart obtained for the PtNPs/GCNFPEIGALOxGlySPCE biosensors at RT, (b) At 20 C. Upper and lower limits are 3 times the standard deviation of the slope values
(n = 10). Supporting electrolyte: 0.1 M PB, pH = 7.0. Eapp = +0.3 V. (Error bars, n = 5).

potential. The applied potential was checked from 0.1 to +0.6 V in a
0.1 M PB, pH = 7.0 containing 1 mM lactate. Fig. 3b shows the results for
the PtNps/GCNFPEIGALOxGlySPCE biosensor. As it can be seen,
the current signal increased with the applied potential between 0.1
and + 0.3 V, reaching a plateau at + 0.3 V. No signicant differences
were found in applying higher potentials. Therefore, +0.3 V was chosen
as the optimal working potential for amperometric detection of lactate.
3.3.2. Analytical characteristics
Repeatability of the amperometric signals and reproducibility of the
fabrication procedure have also been checked. A repeatability value of
1.8% (R.S.D.) was obtained for ten successive amperometric measure-
ments of lactate 1 mM. The reproducibility in the fabrication procedure
was checked by collecting the calibration curves of ten different biosen-
sors prepared in the same way. A R.S.D. of 4.9% was obtained, pointing
out the good reproducibility of the developed biosensor.
On the other hand, Fig. 4a shows the currenttime plots obtained at
+ 0.3 V for different concentrations of L-lactate and Fig. 4b shows the
corresponding calibration curve for lactate. The response of the lactate
biosensor showed a linear relationship between current and lactate
concentration in the range from 0.01 to 2.0 mM with a slope of
(41,302 546) A/M cm2, an intercept of 0.17 0.02 A and a correla-
tion coefcient of 0.999. A detection limit of 6.9 M was obtained and
calculated according to the 3 s/m criterion.
3.3.3. Storage stability
One of the major problems of the lactate biosensors is the long term
stability. Here, two batches of biosensors (n = 100) were fabricated and
stored at two different temperatures: RT (20 C unless stated other-
wise) and 20 C, with the aim of evaluating the long-term stability
of the developed nanostructured lactate biosensor. Calibration curves
in the 0.21.0 mM lactate range (n = 5 biosensors) were collected
every 2 days over a period of 1 month. After this period, measurements
were collected every two weeks. Fig. 5 displays the control charts
obtained. The mean value for the rst day is the average of the slope
obtained with 10 biosensors. Upper and lower control limits are 3
times the standard deviation from the slope value obtained on the
rst day. As it can be seen, the slope remains almost constant (90%)
after 3 months at RT and after more than 18 months at 20 C (95%).

Table 2
Comparison of different parameters and data performance of different amperometric lactate biosensors.

Mediator/ Enzyme Electrode Stirring Slope Linear range LOD RSD Storage Reference
catalyst (mA/Mcm2) (M) (M) (%)

Co(II)Pc LOx SPCE No 10.5 5500 3.4 5.3 30% of activity after 20 days at 4 C [16]
Co(II)Pc LOx SPCE No 4.53 181500 18.3 5.3 98% of response after 9 months at 4 C [21]
Co(II)Pc LOx SPCE Yes 6.9 10006000 289 9 [20]
MB LDH SPCE Yes 4.2a 55010000 550 3.1 [19]
MB LDH SPCE Yes 9.8a 112 2 [27]
3, 4DHB 5.5a 120 6
DCPIP 1.9a 118 7
0-PDA 11.1a 19.9 5
Prussian blue LOx SPCE No, FIA analysis 180 0.51000 0.1 6 months at 4 C [48]
Fc-HRP LOx SPCE Yes 116.9 1.1560 0.5 2.7 40% of response after 15 days [18]
Fc polymer LOx Pt Yes 2.14 0600 2.1 65% of response after 3 months at 4 C [49]
Ptcarbon LOx SPCE Yes 1.41 05000 [17]
PtNpMWCNTs LOx GCE Yes 6.36a 2002000 300 90% of the initial response after 1 month at RT [46]
PtNpMWCNTs LOx GCE Yes 424 0100 0.25 8 Few days [50]
Ptcarbon LOx SPCE Yes 0.87a 0300 2 weeks at 4 C [1]
PVI-Os/MWCNT LOx AuE Yes 19.7 Up to 1000 5 [31]
LOx Pt Yes 230 0350 0.5 3 [51]
PtNps/GCNF LOx SPCE No 41.3 102000 6.9 4.9 3 months at RT and 18 months at 20 C This work

Abbreviations: Co(II)Pc: Cobalt (II) phthalocyanine; Fc: Ferrocene; LDH: lactate dehydrogenase; LOx: lactate oxidase; MWCNTs: multiwall carbon nanotubes; PVI-Os: polyvinylimidazole-
osmium; SPCE: Screen printed carbon electrode; GCE: glassy carbon electrode; AuE: gold electrode; FIA: ow injection amperometry; GCNF: graphitized carbon nanobers; RT: room
temperature.
a
Slope in mA/M because the area of the electrode is not provided.
64 O.A. Loaiza et al. / Bioelectrochemistry 101 (2015) 5865

lactate and 3 Basque ciders. Dilution factors of 1:20 for white; 1:50 for
100 ros; 1:100 for red wines and 1:100 for ciders were applied for the
analysis.
As indicated in Section 3.3.4., no matrix effects were observed in any
80
of these samples and, therefore, the determination was accomplished
by interpolation of the corresponding amperometric signals obtained
(n = 5 replicates for each one) into calibration plots done with stan-
% Signal

60
dards. Basque ciders were also analyzed using a commercially available
spectrophotometric kit for L-lactate determination. Condence inter-
40 vals with a condence level of 95% were calculated according to the
formulae:

t
20 confidence interval X  p 1
n

0 where X : Mean value; t: Student's t distribution with n-1 degrees of

freedom; : standard deviation; and n: number of replicates
te

ol
tol
.

A.

cid

rol
id
se

ls

er
ic A

no

an
ac
cta

cid
rbi
cto

ce
inic

ca
he

Eth
ikim

The results obtained with the biosensors, together with those from
lic

So
Gly
La

Fru

ial
eti
cc

lyp

Ma

ific
Ac
Sh

Su

the enzymatic kit and the reference values in the wine samples are sum-
Po

Art
marized in Table 3. As it can be seen, there is a good agreement between
results obtained by the different methods. Our biosensor was able to de-
Fig. 6. Effect of interfering compounds in the response of lactate at the PtNps/GCNFPEI
GALOxGlySPCE biosensor. Percentage signals are related to 0.75 mM lactate in 0.1 tect low L-lactate concentration values, as can be seen for example in the
M PB, pH = 7.0. Concentrations of interfering compounds and preparation of articial white wine sample. The results indicate that the methods have no sys-
cider are described in Section 3.3.4. Eapp +0.3 V, t = 60 s. (Error bars n = 5). tematic errors and that the developed biosensors can be successfully
used for determination of L-lactate in wines and ciders with just an ad-
equate sample dilution for the analysis.
In order to evaluate the behavior of our biosensor, a comparison of
different performance parameters with other reported amperometric 4. Conclusions
L-lactate biosensors has been carried out. Results are shown in Table 2.
In the case of LDH biosensors, the incorporation of the cofactor Construction of a novel lactate biosensor using covalent immobiliza-
(NAD+) is needed to carry out the L-lactate detection [19,27]. In addi- tion of LOx on PtNps/GCNFSPCEs has been demonstrated. The ability of
tion, the reduction of NADH takes place through intermediate radicals, GCNFs to promote electron-transfer reactions and the high catalytic ac-
giving rise to electrode fouling and lack of stability. Our lactate biosensor tivity of PtNps towards hydrogen peroxide, allow the development of a
is comparable in terms of linear range and detection limits to other sensitivity-enhanced electrochemical lactate biosensor. Moreover, the
biosensors based on non-stirring measurements [17,21]. The developed developed biosensor exhibits good reproducibility (R.S.D 4.9%) and
biosensor presents an excellent sensitivity compared to other bio- long term stability (maintaining at least 90% of the original signal after
sensors [1,31,46]. Furthermore, PtNps/GCNFPEIGALOxGlySPCE storing for 3 months at RT, or 18 months at 20 C). These excellent
biosensor is among of the most stable ones, maintaining 95% of the analytical characteristics together with its simplicity and low-cost
initial response after more than 18 months stored at 20 C. fabrication turn this kind of biosensors into a very attractive alternative
to conventional methods for lactate determination in wine industry.
3.3.4. Selectivity against interferences Furthermore, the results obtained with the biosensors were in good
The selectivity of the PtNps/GCNFPEIGALOxPEIGlySPCE agreement both with data provided by commercial enzymatic kits
biosensor towards L-lactate detection was evaluated in the presence of and also with reported values supplied by Centre Oenologique de
some compounds found common in cider and wine. The potentially Bourgogne, thus demonstrating the suitability of the developed bio-
interfering compounds were selected from the constituents reported sensors for lactate detection in this type of beverages.
for the cider [47]. In order to obtain an accurate value, concentrations
assayed were at least twice the maximum reported value i.e. shikimic Acknowledgments
acid (6.7 g/l), succinic acid (0.9 g/l), fructose (0.94 g/l), polyphenols
(caffeic acid (0.5 g/l), catechin (0.5 g/l), acetic acid (1.53 g/l), malic Financial support from ENIAC JOINT UNDERTAKINGNanoelectronics
acid (0.1 g/l), glycerol (3.2 g/l), sorbitol (6.0 g/l) and ethanol (70 g/l). for Mobile Ambient Assisted Living (AAL) Systems Project Grant Agree-
Additionally, an articial cider was prepared by mixing the compounds ment N.: 12 02 28 is gratefully acknowledged. L. Aorga acknowledges
before mentioned in the concentrations reported as standard constitu-
ents [47]. All solutions were prepared in 0.1 M PB, pH = 7.0 and
contained in 0.75 mM lactate. Table 3
Concentrations of L-lactate in various beverage samples determined using amperometric
Fig. 6 shows the percentage of signals obtained for 0.75 mM lactate
and spectrophotometric methods.
without and with each interfering compound. There were no signicant
changes in the amperometric signals due to the presence of interfering Sample Lactate concentration g/L
compounds, since the differences remained always lower than 5%. This work Megazyme kit COBb
Therefore, the PtNps/GCNFPEIGALOxGlySPCE biosensor can be Wines Ref. 1 red .30 a
0.1 2.31 0.08
considered selective towards the L-lactate determination in standard Ref. 2 red 0.88 0.20 1.04 0.06
ciders and beverages with similar characteristics. Ref. 3 ros 2.77 0.27 2.80 0.10
Ref. 4 white 0.37 0.07 0.35 0.04
Ciders Mina 3.78 0.13 3.94 0.25
3.3.5. Real sample analysis Olaziregi 2.44 0.11 2.24 0.07
Finally, in order to evaluate the biosensor applicability in food anal- Zelaia 3.08 0.11 3.29 0.10
ysis, specically in the analysis of lactic acid in alcoholic beverages, a
Condence interval calculated with a condence level of 95% n = 5 and ttab = 2.78.
it was tested for the analysis of 4 real wines with certied content of b
Centre Oenologique de Bourgogne.
 O.A. Loaiza et al. / Bioelectrochemistry 101 (2015) 5865
nancial support from the DGI of MICINN (Torres Quevedo Program;
ref: PTQ-11-04404). Authors do also want to acknowledge the technical
collaboration of E. Begiristain.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.bioelechem.2014.07.005.
References
[1] M. Albareda-Sirvent, A.L. Hart, Preliminary estimates of lactic and malic acid in wine
using electrodes printed from inks containing solgel precursors, Sensors Actuators
B Chem. 87 (1) (2002) 7381.
[2] M. Esti, et al., Electrochemical biosensors for monitoring malolactic fermentation in
red wine using two strains of Oenococcus oeni, Anal. Chim. Acta. 513 (1) (2004)
357364.
[3] M. Gamella, et al., Integrated multienzyme electrochemical biosensors for monitor-
ing malolactic fermentation in wines, Talanta 81 (3) (2010) 925933.
[4] M.M.F. Choi, Application of a long shelf-life biosensor for the analysis of L-lactate in
dairy products and serum samples, Food Chem. 92 (3) (2005) 575581.
[5] B. Serra, et al., Graphiteteon composite bienzyme electrodes for the determina-
tion of L-lactate: application to food samples, Biosens. Bioelectron. 14 (5) (1999)
505513.
[6] F. Shapiro, N. Silanikove, Rapid and accurate determination of D- and L-lactate, lactose
and galactose by enzymatic reactions coupled to formation of a uorochromophore:
applications in food quality control, Food Chem. 119 (2) (2009) 829833.
[7] C. Soukoulis, et al., Industrial yogurt manufacture: monitoring of fermentation pro-
cess and improvement of nal product quality, J. Dairy Sci. 90 (6) (2007)
26412654.
[8] W.H. Sperber, et al., Microbiological spoilage of dairy products, Compendium of the
Microbiological Spoilage of Foods and Beverages, Springer, New York, 2010, pp.
4167.
[9] C.R. Ispas, G. Crivat, S. Andreescu, Review: recent developments in enzyme-based
biosensors for biomedical analysis, Anal. Lett. 45 (23) (2012) 168186.
[10] J.D. Artiss, et al., A liquid-stable reagent for lactic acid levels, Am. J. Clin. Pathol. 114
(1) (2000) 139143.
[11] M. Frost, M.E. Meyerhoff, In vivo chemical sensors: tackling biocompatibility, Anal.
Chem. 78 (21) (2006) 73707377.
[12] J.D. Buckley, P.C. Bourdon, S.M. Woolford, Effect of measuring blood lactate concen-
trations using different automated lactate analysers on blood lactate transition
thresholds, J. Sci. Med. Sport 6 (4) (2003) 408421.
[13] W. Meng, et al., A lactate needle-type biosensor for in vivo detection in muscular tis-
sues, Sensors Actuators B Chem. B66 (13) (2000).
[14] O. Faude, W. Kindermann, T. Meyer, Lactate threshold concepts, Sports Med. 39 (6)
(2009) 469490.
[15] W.S. Simonides, et al., A nonenzymatic method for the determination of picomole
amounts of lactate using HPLC: its application to single muscle bers, Anal. Biochem.
169 (2) (1988) 268273.
[16] S. Perez, S. Sanchez, E. Fabregas, Enzymatic strategies to construct L-lactate biosen-
sors based on polysulfone/carbon nanotubes membranes, Electroanalysis 24 (4)
(2012) 967974.
[17] W.A. Collier, D. Janssen, A.L. Hart, Measurement of soluble L-lactate in dairy products
using screen-printed sensors in batch mode, Biosens. Bioelectron. 11 (10) (1996)
10411049.
[18] S. Perez, E. Fabregas, Amperometric bienzymatic biosensor for L-lactate analysis in
wine and beer samples, Analyst 137 (16) (2012) 38543861.
[19] M. Piano, et al., Amperometric lactate biosensor for ow injection analysis based on
a screen-printed carbon electrode containing Meldola's blue-Reinecke salt, coated
with lactate dehydrogenase and NAD+, Talanta 82 (1) (2010) 3437.
[20] F.J. Rawson, et al., A microband lactate biosensor fabricated using a water-based
screen-printed carbon ink, Talanta 77 (3) (2009) 11491154.
[21] T. Shimomura, et al., Amperometric L-lactate biosensor based on screen-printed car-
bon electrode containing cobalt phthalocyanine, coated with lactate oxidase-
mesoporous silica conjugate layer, Anal. Chim. Acta. 714 (2012) 114120.
[22] F. Lucarelli, et al., Carbon and gold electrodes as electrochemical transducers for
DNA hybridisation sensors, Biosens. Bioelectron. 19 (16) (2004) 515530.
[23] O.A. Loaiza, et al., Disposable magnetic dna sensors for the determination at the
attomolar level of a specic enterobacteriaceae family gene, Anal. Chem. 80
(2008) 82398245.
65
[24] F. Ghamouss, et al., Bulk-modied modied screen-printing carbon electrodes with
both lactate oxidase (LOD) and horseradish peroxide (HRP) for the determination of
L-lactate in ow injection analysis mode, Anal. Chim. Acta. 570 (2) (2006) 158164.
[25] D. Lowinsohn, M. Bertotti, Flow injection analysis of blood L-lactate by using a Prus-
sian blue-based biosensor as amperometric detector, Anal. Biochem. 365 (2) (2007)
260265.
[26] N. Sato, H. Okuma, Amperometric simultaneous sensing system for D-glucose and L-
lactate based on enzyme-modied bilayer electrodes, Anal. Chim. Acta. 565 (2)
(2006) 250254.
[27] B. Prieto-Simn, E. Fbregas, A. Hart, Evaluation of different strategies for the devel-
opment of amperometric biosensors for L-lactate, Biosens. Bioelectron. 22 (11)
(2007) 26632668.
[28] L. Gorton, A. Hedlund, A ow-injection method for the amperometric determination
of L-lactate with immobilized enzymes and a chemically modied electrode, Anal.
Chim. Acta. 213 (1988) 91100.
[29] W. Chen, et al., Recent advances in electrochemical sensing for hydrogen peroxide: a
review, Analyst 137 (1) (2012) 4958.
[30] W.-X. Chen, J.Y. Lee, Z. Liu, Preparation of Pt and PtRu nanoparticles supported on
carbon nanotubes by microwave-assisted heating polyol process, Mater. Lett. 58
(25) (2004) 31663169.
[31] X. Cui, et al., Highly sensitive lactate biosensor by engineering chitosan/PVI-Os/CNT/
LOD network nanocomposite, Biosens. Bioelectron. 22 (12) (2007) 32883292.
[32] Y. Tianyan, et al., Characterization of platinum nanoparticle-embedded carbon lm
electrode and its detection of hydrogen peroxide, Anal. Chem. 75 (2003)
20802085.
[33] Z. Peng, H. Yang, Designer platinum nanoparticles: control of shape, composition in
alloy, nanostructure and electrocatalytic property, Nano Today 4 (2) (2009)
143164.
[34] X. Li, et al., High loading Pt nanoparticles on functionalization of carbon nanotubes
for fabricating nonenzyme hydrogen peroxide sensor, Biosens. Bioelectron. 59
(2014) 221226.
[35] Y. Xiao, C.M. Li, Nanocomposites: from fabrications to electrochemical
bioapplications, Electroanalysis 20 (6) (2008) 648662.
[36] W. Wang, et al., Amperometric bienzyme glucose biosensor based on carbon nano-
tube modied electrode with electropolymerized poly(toluidine blue O) lm,
Electrochim. Acta 55 (23) (2010) 70557060.
[37] Y.-C. Tsai, S.-Y. Chen, H.-W. Liaw, Immobilization of lactate dehydrogenase within
multiwalled carbon nanotube-chitosan nanocomposite for application to lactate
biosensors, Sensors Actuators B Chem. 125 (2) (2007) 474481.
[38] L. Ag, P. Yez-Sedeo, J.M. Pingarrn, Role of carbon nanotubes in electroanalyt-
ical chemistry: a review, Anal. Chim. Acta. 622 (12) (2008) 1147.
[39] J. Wang, Carbon-nanotube based electrochemical biosensors: a review, Electroanal-
ysis 17 (2007) 717.
[40] J.S. Huang, Y. Liu, T.Y. You, Carbon nanober based electrochemical biosensors: a re-
view, Anal. Methods 2 (3) (2010) 202211.
[41] Y. Liu, et al., A novel and simple route to prepare a Pt nanoparticle-loaded carbon
nanober electrode for hydrogen peroxide sensing, Biosens. Bioelectron. 26 (11)
(2011) 45854590.
[42] S.M. Andersen, et al., Durability of carbon nanober (CNF) & carbon nanotube (CNT)
as catalyst support for proton exchange membrane fuel cells, Solid State Ionics 231
(2013) 94101.
[43] D. Wang, et al., In situ synthesis of Pt/carbon nanober nanocomposites with en-
hanced electrocatalytic activity toward methanol oxidation, J. Colloid Interface Sci.
367 (2012) 199203.
[44] J.F. Botero-Cadavid, et al., Detection of hydrogen peroxide using an optical ber-
based sensing probeJuan, Sensors Actuators B Chem. 185 (2013) 166173.
[45] T. Weijun, G. Changyou, M. Helmuth, Poly(ethyleneimine) microcapsules:
glutaraldehyde-mediated assembly and the inuence of molecular weight on
their properties, Polym. Adv. Technol. 19 (2008) 817823.
[46] J. Huang, et al., Development of an amperometric L-lactate biosensor based on L-
lactate oxidase immobilized through silica solgel lm on multi-walled carbon
nanotubes/platinum nanoparticle modied glassy carbon electrode, Mater. Sci.
Eng. C Biomim. Supramol. Syst. 28 (7) (2008) 10701075.
[47] A. Picinelli, et al., Chemical characterization of Asturian cider, J. Agric. Food Chem. 48
(9) (2000) 3997.
[48] E.I. Yashina, et al., Solgel immobilization of lactate oxidase from organic solvent:
toward the advanced lactate biosensor, Anal. Chem. 82 (5) (2010) 16011604.
[49] M.P.G. Armada, et al., A siloxane homopolymer with interacting ferrocenes as a new
material for the preparation of sensors based on the detection of hydrogen peroxide,
Electroanalysis 15 (13) (2003) 11091114.
[50] K.B. Male, S. Hrapovic, J.H.T. Luong, Electrochemically-assisted deposition of oxi-
dases on platinum nanoparticle/multi-walled carbon nanotube-modied elec-
trodes, Analyst 132 (12) (2007) 12541261.
[51] X. We, M.G. Zhang, W. Gorski, Coupling the lactate oxidase to electrodes by
ionotropic gelation of biopolymer, Anal. Chem. 75 (9) (2003) 20602064.