Simultaneous voltammetric detection of six biomolecules using a nanocomposite of titanium dioxide nanorods with multi-walled carbon nanotubes

Article abstract of DOI:10.1016/j.electacta.2020.137094

In this proof-of-concept study, a novel nanocomposite of titanium dioxide nanorods with multi-walled carbon nanotubes (TiO₂NRs-MWCNTs) was synthesized using a solvo-thermal method and characterized by transmission electron microscopy (TEM) and Fourier-transform infrared spectroscopy (FTIR). The nanocomposite of TiO₂NRs-MWCNTs was utilized to modify the surface of a glassy carbon electrode (GCE) using 1.5% (v/v) Nafion solution for a proof-of-concept study to detect uric acid (UA), xanthine (XA), theophylline (TP) and theobromine (TB) in the presence of ascorbic acid (AA) and dopamine (DA). Simultaneous detection of these six biomolecules was performed using differential pulse voltammetry (DPV) in a wide potential window from -0.3 V to -1.6 V (vs. Ag/AgCl) at pH 4.0. The TiO₂NRs-MWCNTs/GCE showed strong, stable and six simultaneous well-separated anodic peaks at 0.13, 0.35, 0.50, 0.85, 1.10 and 1.28 V for AA, DA, UA, XA, TP and TB, respectively. The calibration curves showed linearity to 191.0, 147.0, 537.0, 586.0, 893.0, 1653.0 μM with detection limits of 0.51, 0.06, 0.05, 0.09, 0.56 and 0.75 μM for AA, DA, UA, XA, TP and TB, respectively. The electrochemical performance of the TiO₂NRs-MWCNTs/GCE displayed good reproducibility for simultaneous determination of six analytes. Finally, our preliminary results suggested that the TiO₂NRs-MWCNTs/GCE can provide a promising platform for rapid quantitative detection of AA, DA, UA, XA, TP and TB in quality control studies of real samples such as pharmaceuticals and food products.

Full text of DOI:10.1016/j.electacta.2020.137094

Electrochimica Acta 362 (2020) 137094
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Research Article
Simultaneous voltammetric detection of six biomolecules using a
nanocomposite of titanium dioxide nanorods with multi-walled
carbon nanotubes
Bhargav R. Patel, Saim Imran, Wanying Ye, Hanyi Weng, Meissam Noroozifar,
Kagan Kerman∗
Department of Physical and Environmental Sciences, University of Toronto Scarborough, 1265 Military Trail, Toronto, ON, M1C 1A4, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
In this proof-of-concept study, a novel nanocomposite of titanium dioxide nanorods with multi-walled
carbon nanotubes (TiO2NRs-MWCNTs) was synthesized using a solvo-thermal method and character-
ized by transmission electron microscopy (TEM) and Fourier-transform infrared spectroscopy (FTIR). The
nanocomposite of TiO2NRs-MWCNTs was utilized to modify the surface of a glassy carbon electrode (GCE)
using 1.5% (v/v) NafionTM solution for a proof-of-concept study to detect uric acid (UA), xanthine (XA),
theophylline (TP) and theobromine (TB) in the presence of ascorbic acid (AA) and dopamine (DA). Simul-
taneous detection of these six biomolecules was performed using differential pulse voltammetry (DPV)
in a wide potential window from -0.3 V to -1.6 V (vs. Ag/AgCl) at pH 4.0. The TiO2NRs-MWCNTs/GCE
showed strong, stable and six simultaneous well-separated anodic peaks at 0.13, 0.35, 0.50, 0.85, 1.10 and
1.28 V for AA, DA, UA, XA, TP and TB, respectively. The calibration curves showed linearity to 191.0, 147.0,
537.0, 586.0, 893.0, 1653.0 μM with detection limits of 0.51, 0.06, 0.05, 0.09, 0.56 and 0.75 μM for AA, DA,
UA, XA, TP and TB, respectively. The electrochemical performance of the TiO2NRs-MWCNTs/GCE displayed
good reproducibility for simultaneous determination of six analytes. Finally, our preliminary results sug-
gested that the TiO2NRs-MWCNTs/GCE can provide a promising platform for rapid quantitative detection
of AA, DA, UA, XA, TP and TB in quality control studies of real samples such as pharmaceuticals and food
products.
Received 25 June 2020
Revised 13 August 2020
Accepted 9 September 2020
Available online 14 September 2020
Keywords:
Electrochemical sensor
Titanium dioxide nanorods
Multi-walled carbon nanotubes
Nanocomposite
Simultaneous determination
Differential pulse voltammetry
Cyclic voltammetry
© 2020 Elsevier Ltd. All rights reserved.
1. Introduction
cholamine neurotransmitter found in the central nervous system.
It is involved in the regulation of voluntary movement and reward
Simultaneous detection using electrochemical sensors are
emerging as a powerful technique for quality control assays, mass
production and on-field rapid tests for environmental studies due
to their high accuracy, selectivity, reproducibility, low cost and
simple operation [1]. Hence, they have been widely employed in
quantitative measurement of the biomolecules and drugs in in-
dustrial settings [2]. The sensitive and rapid quantitative detection
of these biomolecules also benefits pathological research, clinical
diagnosis and pharmaceutical quality control [3–5]. Ascorbic acid
(AA) and dopamine (DA) are related to cognitive wellness, while
uric acid (UA), xanthine (XA), theophylline (TP), and theobromine
(TB) are products in the purine metabolism pathway; hence all
the mentioned compounds represent important roles in the body.
Lack of AA increases the risk of scurvy [6], while DA is a cate-
value [7,8]. Past research has shown that inadequate dopamine lev-
els can cause Parkinson’s diseases and attention-deficit hyperac-
tivity disorder (ADHD) [9–11]. Purines are a class of heterocyclic
aromatic organic compounds including UA, XA, TP, and TB that
contain a pyrimidine ring fused with the imidazole ring struc-
ture [12]. TB and TP are neuro-stimulants, which can be formed
from xanthine (XA) as the precursor (Scheme 1) [13,14]. High UA
concentration in urine is reported to be a sign of hyperuricemia
which is known to progress into gout [15]. Scheme 1 illustrates
how the purine compounds such as TB, TP, and XA can be pro-
duced from degradation of the precursor molecule, xanthosine, a
xanthine-based nucleoside, through different metabolic pathways.
It also shows that UA is the final oxidized product of the hypoth-
esized purine catabolism pathway [16]. Depending on which en-
zyme is used, either nucleosidase or 7-methylxanthosine synthase
with S-adenosyl-l-methionine (SAM), the formation of XA or 7-
methylxanthine can occur, respectively (Scheme 1). Furthermore,
∗
Corresponding author:
E-mail address: kagan.kerman@utoronto.ca (K. Kerman).
https://doi.org/10.1016/j.electacta.2020.137094
0013-4686/© 2020 Elsevier Ltd. All rights reserved.

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
methylation with SAM reduction to S-adenosyl-l-homocysteine
(SAH) can produce TP or TB, respectively [17]. TB and TP can be
converted back to XA upon the action of N-demethylase in the
liver, while XA can then be oxidized to UA by xanthine oxidase
[18,19]. Since, these biomolecules usually co-exist in real-life sam-
ples, such as urine, simultaneous detection can be of paramount
significance. Therefore, it is physiologically relevant and essential
to perform the simultaneous determinations of these compounds
with high sensitivity and selectivity.
TiO2 nanomaterials have been routinely used in electrode sur-
face modifications, as they present large surface area and en-
hanced electrocatalytic properties, while requiring only mild tem-
perature conditions for synthesis [33]. Among the previously syn-
thesized TiO2-based nanomaterials, TiO2NRs have been character-
ized to show properties of one-dimensional nanostructures with
large surface-to-volume area as excellent electron transfer medi-
ators [34,35]. MWCNTs are also widely used in the synthesis of
nanocomposites due their high chemical stability, electrical con-
ductivity and large surface area [36]. Previous studies [25,37] have
shown that the electrode modification with MWCNTs can largely
increase electron and proton transferring rate, giving rise to bet-
ter peak separation, and improved electrode sensitivity than other
modified electrodes. Nafion (NF) is a polymer extensively used
in sensors due to its electrocatalytic properties [38], as it pro-
vides a connection for proton transfer from its sulfonic groups
to the perfluorinated hydrophobic backbone [39]. Therefore, NF
was used for binding the TiO2NRs-MWCNTs nanocomposite onto
the activated surface of GCE. Furthermore, researchers have found
that GCE modified with both nanomaterials and NF has a better
electrochemical performance than GCE modified with solely nano-
materials [40]. In this study, the TiO2NRs-MWCNTs nanocompos-
ite was synthesized by a solvo-thermal method, which employed
an ultrasonic bath to pre-disperse the solid-liquid matrix instead
of stirring the matrix during overnight autoclaving. A solution of
TiO2NRs and MWCNTs underwent a solvo-thermal treatment form-
ing the TiO2NRs-MWCNTs as a nanocomposite. Finally, a mixture
of TiO2NRs-MWCNTs with NF, as the binding agent, was used
for preparation of nanocomposite modified GCEs. The TiO2NRs-
MWCNTs/GCE was used for the simultaneous electrochemical mea-
surements using DPV. To the best of our knowledge, there is no
study that has reported a nanocomposite-based electrochemical
sensor capable of simultaneously detecting TB and TP in the pres-
ence of AA, DA, XA and UA. In addition, recovery tests of the
six analytes were completed using real samples of cocoa products
and physiologically relevant matrices such as urine samples. The
TiO2NRs-MWCNTs/GCE presented enhanced stability, reproducibil-
ity, selectivity and sensitivity during detection of all six analytes.
However, the challenge of simultaneous detection is the fusion
of electrooxidation peaks of different biomolecules due to the ox-
idative potentials of them being close to each other, which re-
sults in a single indistinguishable peak. Subsequently, nanocom-
posite modified electrodes have been used to produce better sepa-
ration and resolution of oxidation peaks during simultaneous de-
tection of up to six different redox-active compounds [20–25].
AA, DA, UA and XA have been simultaneously detected by mul-
tiple electrodes. Ganesh et al. [20] successfully developed epigal-
locatechin gallate-modified graphite paste electrode for simultane-
ous detection of AA, DA and UA. Shastan et al. [21] detected four
compounds AA, DA, UA and Trp (Trptophan) simultaneously by
buckyball-modified carbon-ceramic microelectrodes (CCMEs). Jesny
and Kumar [22] used modified GCE with an electropolymerized
layer of para-amino benzene sulfonic acid to achieve simultane-
ous detection of XA, TP, and caffeine (CA). Furthermore, Hossieny
and Yassien [23] also fabricated a nano-boron doped ceria modi-
fied glassy carbon paste electrode to detect AA, DA, UA and XA,
simultaneously. However, only a very few studies have been re-
ported about the simultaneous electrochemical detection of more
than four redox-active biomolecules. Recently, Wang et al., [24] de-
veloped chemical vapor deposition (CVD) graphene-based electro-
chemical sensor for simultaneous detection of five biomolecules
(AA, DA, UA, Trp, and NO2−). In addition, Li et al. [25] used
ferricyanide-doped chitosan and multi-walled carbon nanotubes
(MWCNTs) nanocomposite-modified graphite paste electrode for
simultaneous detection of AA, DA, UA, Trp, XA and CA. To the
best of our knowledge, the work of Li et al. [25] has been the
only reported study so far to achieve the simultaneous detection
of six redox-active biomolecules with well-separated and well-
defined oxidation peaks. In this study, TB and TP were detected us-
ing voltammetry, whereas these biomolecules would generally re-
quire bench-top HPLC and UV array systems for quantitative anal-
ysis [26,27]. To the best of our knowledge, despite having simi-
lar structures and co-existing in real samples, simultaneous elec-
trochemical detection of both TB and TP has been reported by
Spãtaru et al. [28]. Recently, Švorc et al. [29], used miniaturized
boron-doped diamond electrodes for voltammetric detection of TB
in chocolate products. Moreover, TP has been electrochemically de-
tected in the presence of XA and CA by Jesny and Girish [22].
Therefore, this work would be only the second one to achieve si-
multaneous detection of six biomolecules and the first one which
can simultaneously detect AA, DA, UA, XA, TP and TB.
2. Experimental
2.1. Materials
L-Ascorbic acid (AA, ≥ 99% crystalline), dopamine hydrochlo-
ride (DA, ≥ 98%), uric acid (UA, ≥ 99% crystalline), xanthine (XA,
≥ 99%), anhydrous theophylline (TP, ≥ 99% powder), theobromine
(TB, ≥ 98%), multi-walled carbon nanotubes (MWCNTs, OD of 20–
30 nm with a wall thickness of 1–2 nm and length ranging from
0.5 to 2 μm with purity of ≥ 95%), and TiO2 nanoparticles (21 nm
with purity ≥ 99.5%) and Nafion-117 were purchased from Sigma-
Aldrich (Oakville, ON).
The glassy carbon electrode (GCE) surface can be modified by
casting of novel nanocomposite materials to improve the over-
all selectivity and sensitivity of the electrode to target analytes
[30]. Furthermore, GCE presents great electrocatalysis value due to
low reactivity, high temperature tolerance, strong hardness, imper-
meability and high electrical conductivity [31]. A nanocomposite
of titanium dioxide nanorods (TiO2NRs) and multi-walled carbon
nanotubes (MWCNTs) was used to modify the surface of a GCE
(TiO2NRs-MWCNTs/GCE). TiO2-based nanomaterials present great
potential for use in electrochemical detection of biomolecules due
to its stability, non-toxicity, and biocompatibility [32]. Also, their
mesoporous internal structure gives promising electron-accepting
property that enables TiO2 to collect electrons produced by oxi-
dation of biomolecules during electrochemical measurements [32].
2.2. Instruments
All electrochemical measurements were performed using μAu-
tolab PGSTAT 128 N (Metrohm-EcoChemie, Utrecht, The Nether-
lands) potentiostat/galvanostat, that was controlled by NOVATM
2.1 (Metrohm-EcoChemie, Utrecht, The Netherlands) software con-
nected to a three-electrode cell. The TiO2NRs-MWCNTs/GCE was
utilized as the working electrode, while platinum wire acted as the
auxiliary electrode, and a saturated Ag/AgCl electrode was the ref-
erence electrode. All measurements were conducted at room tem-
perature. The DPV potential window was optimized and set from
−0.3 to 1.6 V at a step potential of 5 mV and modulation am-
plitude of 0.025 V with a modulation time of 0.05 s as well as
an interval time of 0.5 s. The morphology of the nanocomposites
2

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
was characterized through TEM using Hitachi H-7500 transmission
electron microscope (Hitachi, Japan). Bruker alpha-P FT-IR spec-
trometer (Billerica, MA) having deuterated lanthanum α alanine
doped triglycine sulfate (DLaTGS) detector was used to obtain the
FT-IR spectra. Powdered MWCNTs, TiO2NPs, TiO2NRs, and TiO2NRs-
MWCNTs nanocomposite were placed on the crystal to generate
the FT-IR spectrum. FT-IR spectra were obtained at a resolution of
2 cm−1 at a scan rate of 16, between 4000 cm−1 to 360 cm−1. FT-
IR spectra were baseline corrected and smoothened once with an
mixed with known concentrations of AA, DA, XA, TP and TB using
the standard addition method to prepare the spiked urine sample
solutions. A commercially available chocolate powder was used to
prepare a stock concentration of 0.133 g/mL in distilled water. The
solution was heated while stirring continuously on a hotplate for
24 h. After cooling down, the solution was diluted five-fold using
0.2 M PBS (pH 4.0) followed by the standard addition of six target
biomolecules as described above.
ꢀR
average of 17 data points. Opus 6.5 software was used to perform
3. Results and discussion
all the FT-IR spectral data analysis.
3.1. Characterization of the nanocomposite
2.3. Preparation of reagents
Particle morphology for TiO2NPs, TiO2NRs and TiO2 NRs-
MWCNTs are shown as TEM images in Fig. 1. The rod-like struc-
tures of TiO2NRs (as shown in Fig. 1b) indicated the success-
ful formation of TiO2NRs from the initial sphere-like structures
of TiO2NPs observed in Fig. 1a. Additionally, the TEM images of
TiO2NRs-MWCNTs in Fig. 1c and 1d showed the combination of the
MWCNTs with the TiO2NRs.
The stock solutions (5.0 mL) of 0.1 M AA and DA were prepared
directly using sterile 18.2 μΩ ultra-pure water obtained from Cas-
cada LS water purification system (Pall Co., Mississauga, ON). In
addition, the stock solutions (5.0 mL) of 0.01 M UA, XA, TP, and
TB were also prepared using ultra-pure water and the pH was ad-
justed with 15 μL of 10 M NaOH. Furthermore, 0.2 M phosphate
electrolyte solution was prepared from H3PO4 (14.8 M), and the
desired pH range (pH 2.0–5.0) was adjusted using 10 M NaOH,
while the buffer solutions with higher pH range were prepared by
equimolar mixing of K2HPO4 with KH2PO4 solutions, then the de-
sired pH range (pH 6.0–8.0) was adjusted using 4.0 M HCl solution.
The FTIR spectrum of MWCNTs showed peaks at 1544, 1951,
2118, 2328 and 2665 cm−1 indicates C = C stretching, O H de-
–
–
–
formation vibrations, C = O stretching, C OH stretching and C H
stretching, respectively [41,42]. The peaks at 437 and 734 cm−1
in FTIR spectrum of TiO2NPs are the contributions of the anatase
phase of TiO2NPs as reported by Karakitsou & Verykios [43,44]. In
the FTIR spectrum of TiO2NRs, the intense broad peak at 3251 was
2.4. Synthesis of TiO2NRs and TiO2NRs-MWCNTs
–
characteristic of the O H stretching vibrations, the band around
TiO2NRs-MWCNTs were prepared by mixing 200 mg TiO2
nanoparticles and 50 mg MWCNTs with 10.0 mL of concentrated
NaOH solution in a scintillation vial. The mixture was placed in
an ultrasonic bath for 30 min at 60 °C and was then transferred to
a Teflon-lined stainless-steel autoclave container and kept at 130 °C
for 24 h. The resulting nanocomposite was washed with 1.0 L ultra-
pure water and neutralized to pH 7.0 with 0.1 M HCl, followed
by a thorough washing with 0.5 L ultra-pure water. Finally, the
nanocomposite was dried at 80 °C for 24 h. TiO2NRs were similarly
prepared with the exception that initially step did not include the
addition of 50 mg MWCNTs.
1633 cm−1 represented the deformative vibration of Ti-OH stretch-
ing mode, and the band around 930 cm−1 indicated the stretch-
ing vibration of short Ti-O bonds incorporating non-bridging oxy-
gen. The weak peak at 694 cm−1 was again a contribution of
the anatase phase of TiO2, while the sharp and intense peak at
445 cm−1 represented Ti-O bending mode [35,45]. Finally, for the
TiO2NRs-MWCNTs, the characteristic sharp peak of Ti-O bending
mode was split into two peaks at 444 and 415 cm–1, in addition,
peaks at 939, 2117, and the broad peak at 3211 cm−1 were indica-
tive of the stretching vibration of short Ti-O bonds incorporating
–
non-bridging oxygen, C = O stretching, and O H stretching vibra-
tions, respectively.
2.5. Preparation of the modified GCEs with TiO2NRs-MWCNTs
3.2. Comparison of the electrochemical behavior of GCE,
MWCNTs/GCE, TiO2NRs/GCE and TiO2NRs-MWCNTs/GCE
The surface of GCE was cleaned by polishing with micro-polish
Gamma Alumina powder for 15 mins and activated in 0.5 M
H2SO4 using cyclic voltammetry (CV). A desired amount (5 mg) of
TiO2NRs-MWCNTs nanocomposite was mixed with an aliquot (100
μL) of 1.5% (v/v) NafionTM solution in a scintillation vial, which
was then dispersed for 30 mins in an ultrasonic bath at 60 °C. An
aliquot (2 μL) of the resulting solution was drop-cast on the acti-
vated GCE surface and left to dry under IR radiation for 30 min. For
the purpose of electrochemical performance comparison studies to
check the effectiveness of the modified electrodes, TiO2NRs/GCE
and MWCNTs/GCE were also prepared with the same methods de-
scribed above and compared with each other as well as with the
bare GCE surfaces.
The electrochemical responses of the modified electrodes and
the bare GCE were studied in a mixture of AA, DA, UA, XA, TP,
and TB. As shown in Fig. 2, the bare GCE was able to detect only
UA, TP and TB with the oxidation peak potentials at 0.50, 1.13
and 1.31 V, however, the oxidation peaks were low and broad,
which suggested slow electron transfer kinetics. In addition, both
MWCNTs/GCE and TiO2NRs/GCE showed very broad and low oxi-
dation peaks for all six biomolecules at specified concentrations.
However, under the same experimental conditions, DPV of the
TiO2NRs-MWCNTs/GCE provided the detection of all six analytes
and showed substantial increase in the anodic peak current at 0.13,
0.35, 0.50, 0.85, 1.10 and 1.28 V, for AA, DA, UA, XA, TP and TB, re-
spectively. Moreover, with TiO2NRs-MWCNTs/GCE displayed well-
defined oxidation peaks, demonstrating the remarkable sensitivity
of this novel sensor.
2.5. Preparation of real samples
Urine samples were collected from a healthy individual and
used following the ethical guidelines. All procedures using real
samples were approved for the project no. 950–231116 by the Re-
search Ethics Committee of the University of Toronto and were in
accordance with the guidelines and codes established by the Inter-
agency Advisory Panel on Research Ethics (PRE). An aliquot (1 mL)
of urine solution was added to a scintillation vial and diluted five-
fold using 0.2 M PBS (pH 4.0). The diluted urine samples were then
3.3. Redox-active surface area of TiO2NRs-MWCNTs/GCE
The redox-active surface area (A) was calculated using the
TiO2NRs-MWCNTs/GCE and bare GCE. All the measurements were
performed using CV between potentials of −0.2 to 0.8 V at vary-
ing scan rates. The anodic peak current (Ipa) from the respective
3

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
Fig. 1. TEM images of (a) TiO2NPs, (b) TiO2NRs, (c) TiO2NRs-MWCNTs (low magnification) and (d) c with higher magnification. Scale bars indicate 0.2 μm for (a), (b), (d),
and 0.5 μm for (c). e) FTIR of MWCNTs, TiO2NPs, TiO2NRs, and TiO2NRs-MWCNTs nanocomposite.
CVs were obtained in the presence of 1 mM of [Fe(CN)6]3−/4− in
the bare GCE and TiO2NRs-MWCNTs/GCE, the active surface ar-
eas were calculated to be 0.05 and 0.31 cm2, respectively. In or-
0.1 M KCl as the supporting electrolyte (Fig. S3) and were plot-
der to better assess the effect of nanocomposite modification on
the redox-active surface area, an estimation of the roughness fac-
tor (fr) was performed. The roughness factors of the bare GCE and
TiO2NRs-MWCNTs/GCE were determined on the basis of the cur-
rent of [Fe(CN)6]3−/4− redox as a reversible process (Fig. S3). The
measurements were carried out by means of CV in the same so-
lution. Based on Randles–Sevcik equation for the reversible pro-
cesses, all parameters are similar except active surface areas, A1
and A2, for bare GCE and TiO2NRs-MWCNTs/GCE, respectively. Also,
ted against the square root of scan rate (v). The linear relation-
ship between the increasing Ipa vs v1/2 indicated the presence of a
diffusion-controlled process. Furthermore, in a reversible process,
the Randles–Sevcik equation can be used as follows [46]:
ꢀ
ꢁ
Ipa
=
2.69 x 105 n3/2D1/2CAv1/2
(1)
where Ipa refers to the anodic peak current,
n
is the elec-
tron transfer number, A is the active surface area of the elec-
trode, D is the diffusion coefficient, C is the concentration of
under the same conditions, Ip1 and Ip2 were peak currents of
Ip2
[Fe(CN)6]3−/4− and v is the scan rate. For [Fe(CN)6]3−/4−, n = 1
the unmodified and modified electrodes, respectively. The
A2
A1
is
−
and D = 7.6 × 10−6 cm s
1; the active surface areas of the
Ip1
electrodes were calculated using the slope of the Ipa–v1/2 plot. For
equal to
and describes the redox-active surfaces by rough-
4

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
Fig. 2. DPV at the surface of i) TiO2NRs-MWCNTs/GCE (red curve), ii) TiO2NRs/GCE, iii) MWCNTs/GCE and iv) bare GCE, tested in a mixture of AA (97.0 μM), DA (2.5 μM),
UA (2.5 μM), XA (5.0 μM), TP (20.0 μM), and TB (35.0 μM) in 0.2 M PBS at pH 4.0. (For interpretation of the references to colour in this figure legend, the reader is referred
to the web version of this article.)
A2
A1
in the pH range from 2.0 to 8.0 using the TiO2−MWCNTs/GCE. The
results in Fig. 4a indicated that as pH increased from pH 2.0 to
8.0, the peak potentials of AA, DA, UA, XA, TP, and TB moved to-
wards less positive values. When the anodic peak potential vs. pH
was plotted in Fig. 4b, the slopes of AA, DA, UA, XA, and TP were
0.0482, 0.0521, 0.0591, 0.0571, 0.0493 V/pH, respectively. These val-
ues were close to Nernstian slope of 0.059 V/pH and this indicated
an equal proton-electron (i.e. two-proton and two-electron) trans-
fer mechanism for the oxidation of these compounds, which was
in good agreement with the literature [25,27,28,49,50]. The anodic
peak potential shift with increasing pH for AA, DA, UA, XA, TP, and
TB displayed a linear trend as shown in the Eqs. (2)–7:
ness factor (fr
=
=
IIpp21 ) [47,48]. As a result, an average fr
was estimated at 6.2
0.05, which indicated that the modifi-
cation of GCE with the TiO2NRs-MWCNTs nanocomposite signifi-
cantly increased the redox-active surface area. An EIS study was
also performed to determine the capacitance of nanocomposite-
modified GCEs. Fig. S3d represents the Nyquist plots of the bare
GCE and TiO2NRs-MWCNTs/GCE. The Nyquist plots showed that
the charge-transfer resistance (Rct) values of GCE and TiO2NRs-
MWCNTs/GCE were 993.7
2.1 and 352.5
1.9 Ω, respectively.
Upon fitting the Nyquist plots with the modified Randles equiv-
alent circuit shown in the inset of Fig. S3d, the Rct of TiO2NRs-
MWCNTs/GCE was determined to be lower than that of the bare
GCE. An estimated exchange current (I0) can be used for compar-
ꢀ
ꢁ
Ep AA = − 0.0482
0.012 x + 0.3392
0.017 R2 = 0.9910
(2)
ison of the catalytic activity. The exchange current was calculated
(
)
(
)
(
)
R×T
using I0
=
where R, T, n and F are gas constant, absolute
n×F×Rct
temperature, number of transferred electrons (n = 1) and Faraday
constant, respectively. Based on this equation, the I0 values for the
bare GCE and TiO2NRs-MWCNTs/GCE were calculated as 2.6 0.10
ꢀ
ꢁ
Ep DA = − 0.0521
0.009 x + 0.5327
0.010 R2 = 0.9962
(3)
(
)
(
)
(
)
and 7.3
0.14 mA, respectively.
3.4. Simultaneous and individual determinations of six biomolecules
ꢀ
ꢁ
Ep UA = − 0.0591
0.013 x + 0.7010
0.013 R2 = 0.9962
(4)
(
)
(
)
(
)
The electrochemical detection of six biomolecules was achieved
by conducting simultaneous and individual DPV measurements us-
ing the TiO2NRs-MWCNTs/GCE. The electrooxidation peak poten-
tials of AA, DA, UA, XA, TP and TB were found to be at 0.13,
0.35, 0.50, 0.85, 1.10 and 1.28 V, respectively. The separation be-
tween each anodic peak were significant enough to provide the
simultaneous detection of six biomolecules in a mixed solution
(Fig. 3-red curve). Importantly, the oxidation peak potentials of AA
(Fig. 3-brown line), DA (Fig. 3-black line), UA (Fig. 3-green line), XA
(Fig. 3-pink line), TP (Fig. 3-purple line) and TB (Fig. 3-blue line)
observed in individual test were detected at similar anodic peaks
under same conditions.
ꢀ
ꢁ
Ep XA = − 0.0571
0.023 x + 1.0576
0.021 R2 = 0.9907
(5)
(
)
(
)
(
)
ꢀ
ꢁ
Ep TP = − 0.0493
0.014 x + 1.3088
0.032 R2 = 0.9962
(6)
(
)
(
)
(
)
ꢀ
ꢁ
Ep TB = − 0.0216
0.011 x + 1.4374
0.035 R2 = 0.9962
(
)
(
)
(
)
(7)
3.5. pH effect and the hypothesized electrooxidation mechanisms of
six biomolecules
As shown in Fig. 4a, pH 4.0 displayed higher anodic peak cur-
rent signals and the best peak separations, therefore, it was se-
lected as the optimal pH for further studies.
We hypothesized that the acidity of electrolyte affected the an-
odic peak potentials of biomolecules significantly, since the pro-
tons took part in the surface reactions. The pH effect was studied
The hypothesized mechanisms are shown in Eqs. (8)–13 for AA,
DA, UA, XA, TP, and TB respectively. However, TB seemed to be
5

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
Fig. 3. DPV of the background (blank 0.2 M PBS, yellow line), individual solutions of AA (97.0 μM, brown line), DA (2.5 μM, black line), UA (5.0 μM, green line), XA (7.5
μM, pink line), TP (20.0 μM, purple line), TB (35.0 μM, blue line), and for mixture of all of them (red line) in 0.2 M PBS (pH 4.0). (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 4. (a) DPV of TiO2NRs-MWCNTs/GCE recorded in the presence of AA (125.0 μM), DA (12.0 μM), UA (17.0 μM), XA (17.0 μM), TP (58.0 μM), and TB (160.0 μM) simulta-
neously in 0.2 M PBS at pH 2.0, 3.0, 4.0, 5.0 6.0, 7.0 and 8.0. (b) Linear regression plots of anodic peak potential vs. pH for six biomolecules.
an exception as it displayed a slope of 0.0216 V/pH, which sug-
gested that the electrooxidation mechanism included a ratio of 1
proton to 3 electrons during the redox process. This initially refutes
the expected result for a mildly acidic buffer (0.2 M PBS, pH 4),
however, compared to previous literature, there is strong evidence
showing that TB did not follow the direct relationship between
pH and proton involvement as observed with AA, DA, UA, XA, or
TP [22,25,28,29,51]. Moreover, it was reported to follow a similar
mechanism as caffeine, where the oxidation would take place over
two steps with the first being the rate determining step [28,52,53].
(8)
(9)
6

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
(10)
The limit of determination (LOD) of the six biomolecules was
calculated using the equation: LOD = 3 × Sblank/m, where Sblank
represents the standard deviation of background signals (n = 10)
of the TiO2NRs-MWCNTs/GCE, and m is the slope of the calibration
curve from the first segment. The theoretical LOD for AA, DA, UA,
XA, TP and TB were calculated to be 0.51, 0.06, 0.05, 0.09, 0.56
and 0.75 μM, respectively. The unique redox properties and the
high electron transfer kinetics due to the increased surface area
obtained from the TiO2NRs-MWCNTs nanocomposite enabled the
enhanced analytical performance of the sensor According to the
prior literature, the pKa values of DA, UA, AA, XA, TP and TB are
8.86, 5.8, 4.19, 7.4, 8.8, 10.0, respectively [55-58]. In an environ-
ment where pH is lower than pKa values of the biomolecules, the
biomolecules will be in their protonated form, therefore, at the
pH 4.0, DA, UA, AA, XA, TP, and TB will mostly exist as cations
[55,59]. This facilitates the electrostatic interactions of the cationic
forms of DA, UA and AA with the negatively charged sulfonic acid
(SO3−) functional groups of Nafion and COO− groups of MWCNTs
at pH 4.0, which is consistent with previous literature [60,61]. In
addition, according to Cheng et al. [61], specifically for DA, the
(11)
(12)
(13)
presence of phenyl groups enables the π-π interactions with
MWCNTs, which would assist in the peak separation between AA,
DA, and UA. Finally, the presence of TiO2NRs-MWCNTs nanocom-
posite provides excellent electrocatalytic activity which also facil-
itates the detection of DA, UA and XA. However, DA forms an
electro-polymerised film due to electrooxidation at the electrode
surface that leads to electrode fouling or poisoning [62,63]. This
is commonly observed due the requirement of high overpoten-
tial [61,64]. Moreover, indistinguishable and overlapping oxidation
peaks for target analytes can be observed in the absence of mod-
ifications at the electrode surfaces [65,66]. Past studies have in-
dicated that one of the ways to overcome electrode fouling from
a biological substrate is by modifying the surfaces using chem-
ical compounds (i.e. polyethylene glycol, PEG) and nanocompos-
ites [62]. Kumar et al. [64] have demonstrated that the pres-
ence of nano-TiO2 in the nanocomposite played an important
role in improving the antifouling properties. Therefore, modifying
the surfaces using Nafion, MWCNTs and TiO2NRs significantly im-
proved the selectivity and electrocatalytic properties of the sen-
sors [64,67–70]. A comparison of our nanocomposite-modified sen-
sor with the ones in literature is shown in Table 1. Based on
the comparison shown in the table, the TiO2NRs-MWCNTs/GCE
showed a relatively larger linear dynamic range especially for AA,
XA, TP and TB, while it also had comparable or better LOD for all
3.6. Calibration
DPV measurements were performed to explore the relationship
between the anodic peak currents and concentrations of AA, DA,
UA, XA, TP and TB. The measurements continued until the satura-
tion was reached, and the peaks began to fuse with each other
or did not increase as the concentrations increased. The results
of the concentration-dependence study for simultaneous detection
are shown in the Fig. 5, as six well-distinguished anodic peaks.
For each biomolecule, the anodic peak currents were sub-
tracted by the background current (measured in blank PBS) of
GCE-TiO2NRs-MWCNTs to calculate ꢀI, and ꢀI (μA) was plotted
against the concentration (μM) of the biomolecules. The plot of
anodic peak current vs the increasing concentrations of AA, DA,
UA, XA, TP and TB are shown in Fig. 6. All six calibration curves
showed two segments of linear relationship, with a slope for low
concentrations and a slope for higher concentrations. This was at-
tributed to the difference in the availability of redox-active area on
the electrode surface. We hypothesized that a large redox-active
area was accessible on the electrode surface at lower concentra-
tions of the biomolecules. However, at higher concentrations of the
biomolecules, the accessible redox-active areas were rapidly satu-
rated, and this resulted in the reduced sensitivity of the slope in
the second linear segments for all six biomolecules [21,25,54].
7

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
Fig. 5. DPV measurements for the simultaneous detection of AA, DA, UA, XA, TP, and TB at varying concentrations using the TiO2NRs-MWCNTs/GCE in 0.2 M PBS (pH 4.0).
Fig. 6. The plots for the concentration dependence of increasing anodic peak current signals for (a) AA, (b) DA, (c) UA, (d) XA, (e) TP, and (f) TB with the linear range marked
in blue and orange. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
six biomolecules. Moreover, TiO2NRs-MWCNTs/GCE is the second
modified electrode that is reported to simultaneously detect up to
six biomolecules, to the best of our knowledge.
oxidation peak currents of DA, UA, XA, TP and TB remained con-
stant. Similarly, Fig. S1(b–f) shows the DPV signals corresponding
to the oxidation of DA, UA, XA, TP, and TB, respectively, which in-
creased linearly in response to the increasing concentration, while
the anodic peak currents for all other biomolecules remained rel-
atively constant. This indicated that the oxidation of AA, DA, UA,
XA, TP, and TB at TiO2NRs−MWCNTs/GCE took place independently
and did not cause any interference.
3.7. Interference
AA, DA, UA, XA, TP and TB can coexist at varying concentra-
tions in real-life samples. Hence the interference study of each
biomolecule is important to determine whether the biomolecules
can disrupt the simultaneous detection of other species in the
same solution. In all experiments conducted, the concentration of
one species was gradually increased while the others were kept at
a constant concentration in the solution throughout the study. Fig.
S1a-f shows the results of interference studies performed for each
of the six biomolecules. As shown in Fig. S1a, the peak current of
AA increased linearly with increasing concentrations, however, the
3.8. Chronoamperometry
To determine the diffusion coefficient (D) of the six
biomolecules, chronoamperometric measurements were conducted
using TiO2NRs-MWCNTs/GCE by applying a constant potential of
0.13, 0.33, 0.47, 0.83, 1.12, and 1.36 V for AA, DA, UA, XA, TP, and
TB, respectively. As shown in Fig. S2(a-f), chronoamperograms
8

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
Scheme 1. Hypothesized degradation pathways for xanthosine to uric acid via different routes of Theophylline, Xanthine, and Theobromine.
Table 1
Analytical performance comparison between the reported modified electrochemical sensors and the TiO2NRs-MWCNTs/GCE sen-
sor in this work.
Analyte
AA
Electrode
LOD (μM)
Linear range (μM)
Ref
FCdCH-MWCNTs
5.26
9.6
10.0–2056.8
10–800
[25]
AgNPs-rGO/GCE
[71]
[Ni(phen)2]2+/SWCNTs/GCE
TiO2NRs-MWCNTs/GCE
FCdCH-MWCNTs
12.0
0.51
0.0011
5.4
30–1546
[72]
1.5–51.0, 51.0–191.0
0.99–94.1
10–800
This work
[25]
DA
UA
AgNPs-rGO/GCE
[71]
[Ni(phen)2]2+/SWCNTs/GCE
TiO2NRs-MWCNTs/GCE
pPABSA/GCE
1.0
1–780
[72]
0.06
0.1
0.45–30.0, 30.0–147.0
1–100
This work
[22]
AgNPs-rGO/GCE
8.2
10–800
[71]
[Ni(phen)2]2+/SWCNTs/GCE
0.8
1–1407
[72]
TiO2NRs-MWCNTs/GCE
0.05
0.40–61.0,
61.0–537.0
0.5–100
This work
XA
pPABSA/GCE
0.4
[22]
FCdCH-MWCNTs/GPE
BDDE
0.0073
–
1–191.27
[25]
1–100
[28]
GCE/Nf–{RuDMSO–Cl–H2O}-MME
TiO2NRs-MWCNTs/GCE
pPABSA/GCE
2.35
0.09
7.0
50–500
[73]
0.5–97.0, 97.0–586.0
1- 100
This work
[22]
TP
TB
BDDE
–
1–400
[28]
ED-GO/GCE
0.1
0.8- 60
[74]
TiO2NRs-MWCNTs/GCE
BDDE
0.56
–
1.0–203.0, 203.0–893.0
1–400
This work
[28]
BDDE
0.42
0.75
0.99–54.5
1.5–368.0,
368.0–1653.0
[29]
TiO2NRs-MWCNTs/GCE
This work
Abbreviations: FCdCH: Ferricyanide-doped chitosan;; AgNPs: Silver nanoparticles; rGO: Reduced graphene oxide; [Ni(phen)2]2+
:
Nickel(II)-bis(1,10-phenanthroline); SWCNTs: Single-walled carbon nanotubes; pPABSA: poly (para-amino benzene sulfonic
acid); BDDE: Boron-doped diamond electrode; ED-GO: Electrodeposited graphene oxide.
−
1/2
display the current response for an electrochemical reaction which
is under diffusion control as described by Cottrell’s equation
(Eq. (14)) [75]:
process, the plot of I vs. t
would be linear and the slope
(nFAD1/2Cbπ −1/2) from the best-fit of the linear regions can be
used to calculate D for AA, DA, UA, XA, TP and TB. D values were
calculated to be (6.33
0.10) × 10−6, (7.97
0.12) × 10−7, (5.60
0.21) × 10−6
0.19) × 10−8
,
,
I = nFACD1/2π−1/2t−1/2
(14)
(2.65
and (2.97
0.17) × 10−7, (1.24
0.15) × 10−8cm2
s
for AA, DA, UA, XA, TP and TB,
−
1
where, n represents the number of electrons transferred, F repre-
sents Faraday’s constant (96,485 C mol−1), A represents the work-
ing electrode’s surface area, C is the analyte concentration (mol
cm−3), and D is the diffusion coefficient. For a diffusion-controlled
respectively. Furthermore, several studies in the literature have re-
ported D values of AA, DA, UA, XA, and TP. Table S1 displays the
comparison of the D values calculated in this study with the pre-
9

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
viously reported values. To the best of our knowledge, an electro-
chemical method has been used for the first time in this report to
determine the experimental value of D for TB.
Laviron’s equation [83]:
ꢄ
ꢅ
ꢂ
ꢃ
ꢂ
ꢃ
RT
RTk0
RT
Epa = E0 +
ln
+
ln v
(16)
αnF
αnF
αnF
where E0 is the formal redox potential, R is gas constant, T is tem-
perature, F is Faraday’s constant, k0 is the standard heterogeneous
rate constant, α is the electron transfer coefficient, n is the number
of electrons transferred, v is the scan rate. The slope from the plot
of Epa versus ln v was used to determine the value of αn. The αn
for AA, DA, UA, XA, TP and TB were calculated to be 1.17, 1.08, 1.20,
1.06, 0.98, and 1.14, respectively. Generally, for an irreversible pro-
cess, based on previous studies reported in literature, the α value
is assumed to be 0.5 [81,82]. Therefore, the value of n was calcu-
lated to be 2.34, 2.16, 2.40, 1.96, and 2.28 for AA, DA, UA, XA, TP
and TB, respectively. All the obtained values could be rounded to
2 electrons, which would be in good agreement with the literature
for the estimated number of electrons hypothesized during the ox-
idation of the target biomolecules.
3.9. Repeatability and stability
The stability and repeatability of TiO2NRs-MWCNTs/GCE was
tested by performing multiple DPV measurements (n = 10) for si-
multaneous detection of AA (727 μM), DA (34 μM), UA (24 μM),
XA (24 μM), TP (58 μM) and TB (97 μM). The results (Fig. S4)
indicated that the standard deviations resulting from the anodic
peak currents of six biomolecules were calculated to be 3.22% (AA),
2.54% (DA), 2.82% (UA), 5.19% (XA), 3.99% (TP), and 4.74% (TB). Fur-
thermore, the stability of the TiO2NRs-MWCNTs/GCE was exam-
ined by measuring the DPV signals for simultaneous detection of
AA, DA, UA, XA, TP, and TB for a period of 1 month in 0.2 M
PBS (pH 4.0). DPV measurements were performed using the same
concentration of the six biomolecules. The stability of TiO2NRs-
MWCNTs/GCE was highlighted by the retained peak current (ꢀIpa
)
3.11. Standard heterogeneous rate constant (ks) for the
electrochemical reactions
values for AA, DA, UA, XA, TP, and TB at 97.23, 95.38, 93.55, 98.99,
94.17, and 95.46%, respectively. These results indicated an excellent
stability and repeatability of the TiO2NRs-MWCNTs/GCE for simul-
taneous detection of six target biomolecules.
The standard heterogenous rate constants (ks) of the electro-
chemical oxidation of AA, DA, UA, XA, TP and TB at the surface
of TiO2NRs-MWCNTs/GCE was determined using CV (Fig. S7) based
on the Velasco equation (Eq. (17)) [84]:
3.10. Scan rate
ks = 1.11D1/2(Ep − Ep/2
where, D is the apparent diffusion coefficient, Ep and Ep/2 are the
anodic peak potential and half-wave anodic peak potential, respec-
tively, and v is the scan rate.
)
−1/2v1/2
(17)
Linear sweep voltammetry (LSV) measurements were per-
formed at varying scan rates using the TiO2NRs-MWCNTs/GCE in
a mixture of all six target biomolecules. The dependence of the
peak current with different scan rates was used to determine if
the electrochemical processes at the surface of the modified elec-
trodes were diffusion- or adsorption-controlled [25,76]. Fig. S5 in-
dicates that the anodic peak current was directly proportional to
the square root of the scan rate for all six biomolecules, suggest-
ing a diffusion-controlled process was occurring at the surface of
the TiO2NRs-MWCNTs/GCE. As shown in Fig. S5a, it was observed
for each of the target biomolecule that the anodic peak current in-
creased and the oxidation peak potential shifted to more positive
values with increasing scan rates during the LSV measurements.
In addition, a plot of logarithm of anodic peak current (Ipa) ver-
sus logarithm of scan rate (v) was analyzed to confirm whether
the electrode process was diffusion- or adsorption-controlled (Fig.
S5c). Prior literature studies have indicated that the electrocataly-
sis of the biomolecules at the electrode surface was significantly
impacted with the slope value of log (Ipa) vs. log (v) [77]. The
slope value closer to 0.5 indicated that the electrode reaction is
diffusion-controlled, while the slope value closer to 1.0 indicated
an adsorption-controlled process [77-80]. The relationship for log
(Ip) vs. log (v) was found to be linear with a slope of 0.46, 0.51,
0.49, 0.52, 0.52, and 0.57 for AA, DA, UA, XA, TP and TB, respec-
tively that was near to the theoretical value of 0.5 indicating a
diffusion-controlled process [77-80].
The standard heterogenous rate constant values for the
six biomolecules provided quantitative information about the
electrode-transfer redox reactions for AA, DA, UA, XA, TP and TB
at TiO2NRs-MWCNTs/GCE. The experimental ks values of AA, DA,
UA, XA, TP and TB were determined to be (3.24
0.13) × 10−3
,
(4.89
(1.48
0.23) × 10−4, (9.00
0.68) × 10−3, (7.97− 0.50) × 10−3
,
0.11) × 10−2, (1.80
0.10) × 10−2 cm. s 1, respectively.
3.12. Real sample analysis
The practical applicability of sensor was tested for simultaneous
detection of six target biomolecules in complex matrices such as
human urine and chocolate powder. A urine sample was obtained
from a healthy individual and was used for the simultaneous de-
tection of AA, DA, UA, XA, TP and TB using DPV. The stock urine
sample was diluted five times with 0.2 M PBS (pH 4.0) and a back-
ground DPV measurement was performed. An anodic peak at 0.5 V
indicated the presence of UA in the urine sample. For spiking stud-
ies, the randomized concentrations of AA, DA, UA, XA, TP, and TB
were added to determine the recovery rate of the six biomolecules
as shown in Table 2. Similarly, a stock solution of chocolate pow-
der was prepared and spiked with the known concentrations of six
biomolecules. The background DPV measurements were performed
using the five-fold diluted solution of chocolate powder in 0.2 M
PBS (pH 4.0). The voltammograms did not indicate the presence of
any target analytes in the sample. As described earlier, the known
concentrations of AA, DA, UA, XA, TP, and TB were added to de-
termine the recovery of the six target biomolecules as shown in
Table 2. These results demonstrated the ability of the sensor to de-
tect all six analytes in both matrices, showing potential to use this
electrode in real-life samples.
Moreover, LSV signals indicated that AA, DA, UA, XA, TP and TB
underwent an irreversible electrochemical charge transfer process
[81,82]. For all the six biomolecules, since only the oxidation peak
was observed, a graph showing the dependence of the anodic peak
potential (Epa) and the natural logarithm of the scan rate (ln v) was
plotted. As shown in Fig. S6, the plot followed a linear regression
based on the following equation:
ꢀ
ꢁ
Epa
=
0.0424
0.002 ln v mV s−1
(
)
ꢀ
ꢁ
+ 0.6434
(
0.017 R2 = 0.9966
(15)
)
4. Conclusions
For totally irreversible and diffusion-controlled electrode pro-
cesses as observed for the six biomolecules, the dependence of the
In this proof-of-concept study,
a novel nanocomposite of
Epa versus the ln v can be represented by the following a modified
TiO2NRs-MWCNTs was synthesized using ultrasonication and a
10

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
Table 2
The simultaneous detection of AA, DA, UA, XA, TP and TB in human urine and breakfast
essentials powder samples using TiO2NRs-MWCNTs/GCE.
Sample matrix
Analyte
Detected∗
Spiked
Found∗
Recovery (%)
Human
urine
AA
DA
UA
XA
TP
–
86.27
3.27
83.43
3.13
3.30
0.42
3.40
96.71
95.71
101.35
101.44
101.52
98.74
95.95
103.45
103.73
96.95
96.03
99.34
–
134.73
3.1
4.42
139.22
4.92
–
–
–
–
–
–
–
–
–
4.85
0.78
0.71
0.88
6.57
6.67
TB
AA
DA
UA
XA
TP
8.75
8.64
Chocolate
powder
66.68
5.17
63.98
5.35
3.13
0.43
0.33
0.47
3.72
3.86
4.97
4.81
16.62
134.29
15.96 0.79
133.40 4.30
TB
t(n−1, α)
∗
√
Confidence interval for 95% probability calculated as [x¯
]; t2 ; 0.05 = 2.92.
n
hydrothermal procedure. To the best of our knowledge, this
nanocomposite was used for the first time to prepare an electro-
chemical sensor that facilitated the simultaneous detection of six
analytes - AA, DA, UA, XA, TP, and TB. GCE surfaces modified with
TiO2NRs-MWCNTs exhibited an improved electrochemical perfor-
mance in peak separation and current intensity in comparison with
bare GCE, MWCNTs/GCE, and TiO2NRs/GCE. The optimal pH value
for the sensor was determined to be pH 4.0, which yielded good
detection limits and a wide linear range for all six analytes. More-
over, the application of the TiO2NRs-MWCNTs/GCE was tested in
real-life complex matrices by successfully detecting the six target
analytes in the human urine sample and chocolate powder solu-
tions with acceptable recovery values. With further development,
TiO2NRs-MWCNTs/GCE can become a promising tool for the simul-
taneous detection of redox-active biomolecules in environmental
on-field pre-screening studies and quality control assays in food
industry.
[5] A. Agus, J. Planchais, H. Sokol, Gut Microbiota Regulation of Tryptophan
Metabolism in Health and Disease, Cell Host Microbe 23 (2018) 716–724.
[6] M. Kobayashi, Y. Hoshinaga, N. Miura, Y. Tokuda, S. Shigeoka, A. Murai,
F. Horio, Ascorbic acid deficiency decreases hepatic cytochrome P-450, espe-
cially CYP2B1/2B2, and simultaneously induces heme oxygenase-1 gene ex-
pression in scurvy-prone ODS rats, Biosci. Biotechnol. Biochem. 78 (2014)
1060–1066.
[7] G.W. Arbuthnott, J. Wickens, Space, time and dopamine, Trends Neurosci 30
(2007) 62–69.
[8] M.A. Cenci, Dopamine dysregulation of movement control in l-DOPA-induced
dyskinesia, Trends Neurosci 30 (2007) 236–243.
[9] H.W. Berendse, J. Booij, C.M.J.E. Francot, P.L.M. Bergmans, R. Hijman, J.C. Stoof,
E.C. Wolters, Subclinical dopaminergic dysfunction in asymptomatic Parkin-
son’s disease patients’ relatives with a decreased sense of smell, Ann. Neurol.
50 (2001) 34–41.
[10] G.U. Höglinger, P. Rizk, M.P. Muriel, C. Duyckaerts, W.H. Oertel, I. Caille,
E.C. Hirsch, Dopamine depletion impairs precursor cell proliferation in Parkin-
son disease, Nat. Neurosci. 7 (2004) 726–735.
[11] N.D. Volkow, G.J. Wang, S.H. Kollins, T.L. Wigal, J.H. Newcorn, F. Telang,
J.S. Fowler, W. Zhu, J. Logan, Y. Ma, K. Pradhan, C. Wong, J.M. Swanson, Evalu-
ating dopamine reward pathway in ADHD: Clinical Implications, JAMA - J. Am.
Med. Assoc. 302 (2009) 1084–1091.
[12] H. Rosemeyer, The chemodiversity of purine as a constituent of natural prod-
ucts, Chem. Biodivers. 1 (2004) 361–401.
Declaration of Competing Interest
None.
[13] D.H. Murgida, G.M. Bilmes, R. Erra-Balsells, A Photophysical Study of Purines
and Theophylline by Using Laser-Induced Optoacoustic Spectroscopy, Pho-
tochem. Photobiol. 64 (1996) 777–784.
[14] E. Fischer, Ueber Caffeïn, Theobromin, Xanthin und Guanin, Justus Liebig’s,
Ann. Der Chemie 215 (1882) 253–320.
Acknowledgements
[15] H.K. Choi, D.B. Mount, A.M. Reginato, Pathogenesis of gout, Ann. Intern. Med.
143 (2005) 499–516.
[16] S. Watanabe, Y. Kounosu, H. Shimada, A. Sakamoto, Arabidopsis xanthine dehy-
drogenase mutants defective in purine degradation show a compromised pro-
tective response to drought and oxidative stress, Plant Biotechnol 31 (2014)
173–178.
K. K. acknowledges financial support from the Canada Research
Chair Tier-2 award for “Bioelectrochemistry of Proteins” (Project
no. 950–231116), Ontario Ministry of Research and Innovation
(Project no. 35272), Discovery Grant (Project no. 3655) from Nat-
ural Sciences and Engineering Research Council of Canada (NSERC),
and Canada Foundation for Innovation (Project no. 35272).B. R. P.
is thankful for the PhD award from TD Graduate Research Schol-
arship in Environmental Science and Ontario Graduate Scholarship
(OGS).
[17] H. Ashihara, T. Yokota, A. Crozier, Biosynthesis and catabolism of purine alka-
loids, in: Adv. Bot. Res., Academic Press Inc, 2013, pp. 111–138.
[18] K.D. Georgiev, M. Radeva-Ilieva, S. Stoeva, I. Zhelev, Isolation, analysis and
in vitro assessment of CYP3A4 inhibition by methylxanthines extracted from
Pu-erh and Bancha tea leaves, Sci. Rep. (2019) 9.
[19] A. Onãtibia-Astibia, E. Martínez-Pinilla, R. Franco, The potential of methylxan-
thine-based therapies in pediatric respiratory tract diseases, Respir. Med. 112
(2016) 1–9.
[20] H.V.S. Ganesh, M. Noroozifar, K. Kerman, Epigallocatechin gallate-modi-
fied graphite paste electrode for simultaneous detection of redox-active
biomolecules, Sensors 18 (2018) 23.
Supplementary materials
[21] Z.O. Shastan, H.V.S. Ganesh, M. Noroozifar, K. Kerman, Carbon ceramic mi-
croelectrodes modified with buckyballs for simultaneous determination of re-
dox-active biomolecules, RSC Adv. 8 (2018) 5960–5966.
Supplementary material associated with this article can be
found, in the online version, at doi:10.1016/j.electacta.2020.137094.
[22] S. Jesny, K. Girish Kumar, Non-enzymatic Electrochemical Sensor for the Si-
multaneous Determination of Xanthine, its Methyl Derivatives Theophylline
and Caffeine as well as its Metabolite Uric Acid, Electroanalysis 29 (2017)
1828–1837.
References
[1] N. Jadon, R. Jain, S. Sharma, K. Singh, Recent trends in electrochemical sensors
for multianalyte detection-A review crossmark, Talanta 161 (2016) 894–916.
[2] S. Shahrokhian, M. Ghalkhani, Simultaneous voltammetric detection of ascor-
[23] H. Ibrahim, Y. Temerk, Sensitive electrochemical sensor for simultaneous de-
termination of uric acid and xanthine in human biological fluids based on the
nano-boron doped ceria modified glassy carbon paste electrode, J. Electroanal.
Chem. 780 (2016) 176–186.
[24] X. Wang, D. Gao, M. Li, H. Li, C. Li, X. Wu, B. Yang, CVD graphene as an elec-
trochemical sensing platform for simultaneous detection of biomolecules, Sci.
Rep. 7 (2017) 7044.
bic acid and uric acid at
thionine-nafion ion-pair as an electron mediator, Electrochim. Acta 51 (2006)
2599–2606.
a carbon-paste modified electrode incorporating
[3] A. Covarrubias-Pinto, A. Acuña, F. Beltrán, L. Torres-Díaz, M. Castro, Old Things
New View, Ascorbic Acid Protects the Brain in Neurodegenerative Disorders,
Int. J. Mol. Sci. 16 (2015) 28194–28217.
[4] D.I. Feig, D.-.H. Kang, R.J. Johnson, Uric Acid and Cardiovascular Risk, N. Engl.
J. Med 359 (2008) 1811–1821.
[25] S. Li, M. Noroozifar, K. Kerman, Nanocomposite of ferricyanide-doped chitosan
with multi-walled carbon nanotubes for simultaneous senary detection of re-
dox-active biomolecules, J, Electroanal. Chem. 849 (2019) 113376.
11

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
[26] A. Meyer, T. Ngiruwonsanga, G. Henze, Determination of adenine, caffeine,
theophylline and theobromine by HPLC with amperometric detection, Frese-
nius. J. Anal. Chem. 356 (1996) 284–287.
[27] H. Sereshti, M. Khosraviani, S. Samadi, M.S. Amini-Fazl, Simultaneous determi-
nation of theophylline, theobromine and caffeine in different tea beverages by
graphene-oxide based ultrasonic-assisted dispersive micro solid-phase extrac-
tion combined with HPLC-UV, RSC Adv. 4 (2014) 47114–47120.
[28] N. Spãtaru, B.V. Sarada, D.A. Tryk, A. Fujishima, Anodic voltammetry of xan-
thine, theophylline, theobromine and caffeine at conductive diamond elec-
trodes and its analytical application, Electroanalysis 14 (2002) 721–728.
[29] Ľ. Švorc, M. Haššo, O. Sarakhman, K. Kianicˇková, D.M. Stankovic´, P. Otrˇísal,
A progressive electrochemical sensor for food quality control: Reliable deter-
mination of theobromine in chocolate products using a miniaturized boron–
doped diamond electrode, Microchem. J. 142 (2018) 297–304.
[30] N.S. Punde, A.S. Rajpurohit, A.K. Srivastava, Sensitive electrochemical platform
based on nano-cylindrical strontium titanate/N-doped graphene hybrid com-
posite for simultaneous detection of diphenhydramine and bromhexine, Elec-
trochim. Acta 319 (2019) 727–739.
[53] L. Švorc, Determination of caffeine: A comprehensive review on electrochemi-
cal methods, Int. J. Electrochem. Sci. 8 (2013) 5755–5773.
[54] B.R. Patel, M. Noroozifar, K. Kerman, Prussian blue-doped nanosized polyani-
line for electrochemical detection of benzenediol isomers, Anal. Bioanal. Chem.
412 (2020) 1769–1784.
[55] F. Shang, Y. Liu, S. Hrapovic, J.D. Glennon, J.H.T. Luong, Selective detection of
dopamine using a combined permselective film of electropolymerized (poly–
tyramine and poly-pyrrole-1-propionic acid) on a boron-doped diamond elec-
trode, Analyst 134 (2009) 519–527.
[56] E.Y. Andreeva, S.G. Dmitrienko, Y.A. Zolotov, Methylxanthines: Properties and
determination in various objects, Russ. Chem. Rev. 81 (2012) 397–414.
[57] M.C. Gennaro, C. Abrigo, P. Biglino, Quantification of theophylline in human
plasma by reversed-phase ion-interaction high-performance liquid chromatog-
raphy and comparison with the TDx fluorescence polarization immunoassay
procedure, Analyst 117 (1992) 1071–1074.
[58] R. Fu, C.J. Chen, H.A. Jinnah, Genotypic and phenotypic spectrum in attenuated
variants of lesch-nyhan disease, Mol. Genet. Metab. 112 (2014) 280–285.
[59] W. Sroysee, S. Chairam, M. Amatatongchai, P. Jarujamrus, S. Tamuang, S. Pim-
mongkol, L. Chaicharoenwimolkul, E. Somsook, Poly(m-ferrocenylaniline) mod-
ified carbon nanotubes-paste electrode encapsulated in nafion film for selec-
tive and sensitive determination of dopamine and uric acid in the presence of
ascorbic acid, J. Saudi Chem. Soc. 22 (2018) 173–182.
[31] P. Chen, R.L. McCreery, Control of Electron Transfer Kinetics at Glassy Carbon
Electrodes by Specific Surface Modification, Anal. Chem. 68 (1996) 3958–3965.
[32] J. Bai, B. Zhou, Titanium dioxide nanomaterials for sensor applications, Chem.
Rev. 114 (2014) 10131–10176.
[33] A. Ghicov, P. Schmuki, Self-ordering electrochemistry, A review on growth and
functionality of TiO2 nanotubes and other self-aligned MOx structures, Chem.
Commun. (2009) 2791–2808.
[60] A.A. Ensafi, M. Taei, T. Khayamian, Simultaneous determination of ascor-
bic acid, dopamine, and uric acid by differential pulse voltammetry using
tiron modified glassy carbon electrode, Int. J. Electrochem. Sci.
116–130.
5 (2010)
[34] Y.H. Jung, K.H. Park, J.S. Oh, D.H. Kim, C.K. Hong, Effect of TiO2 rutile nanorods
on the photoelectrodes of dye-sensitized solar cells, Nanoscale Res. Lett.
(2013) 37.
[35] N.B. Ashoka, B.E.K. Swamy, H. Jayadevappa, Nanorod TiO2 sensor for dopamine:
A voltammetric study, New J. Chem. 41 (2017) 11817–11827.
8
[61] J. Cheng, X. Wang, T. Nie, L. Yin, S. Wang, Y. Zhao, H. Wu, H. Mei, A novel
electrochemical sensing platform for detection of dopamine based on gold
nanobipyramid/multi-walled carbon nanotube hybrids, Anal. Bioanal. Chem.
412 (2020) 2433–2441.
[36] X.C. Lu, L. Song, T.T. Ding, Y.L. Lin, C.X. Xu, CuS–MWCNT based electrochemical
sensor for sensitive detection of bisphenol A, Russ. J. Electrochem 53 (2017)
366–373.
[37] Z. Zhu, L. Garcia-Gancedo, A.J. Flewitt, H. Xie, F. Moussy, W.I. Milne, A criti-
cal review of glucose biosensors based on carbon nanomaterials: Carbon nan-
otubes and graphene, Sensors 12 (2012) 5996–6022.
[62] M. Mazloum-Ardakani, H. Rajabi, H. Beitollahi, B.B.F. Mirjalili, A. Akbari,
N. Taghavinia, Voltammetric determination of dopamine at the surface of TiO2
nanoparticles modified carbon paste electrode, Int. J. Electrochem. Sci. 5 (2010)
147–157.
[63] M. Hosseini, M.M. Momeni, M. Faraji, An innovative approach to electro-oxida-
tion of dopamine on titanium dioxide nanotubes electrode modified by gold
particles, J. Appl. Electrochem. 40 (2010) 1421–1427.
[38] K. Zarei, H. Helli, Electrochemical determination of aminopyrene on glassy car-
bon electrode modified with multi-walled carbon nanotube-sodium dodecyl
sulfate/Nafion composite film, J. Electroanal. Chem. 749 (2015) 10–15.
[39] J. Chlistunoff, J.M. Sansiñena, On the use of NafionꢀR in electrochemical stud-
ies of carbon supported oxygen reduction catalysts in aqueous media, J. Elec-
troanal. Chem. 780 (2016) 134–146.
[40] S. Tarahomi, G. Hossein Rounaghi, M. Hossein Arbab Zavar, L. Daneshvar, Elec-
trochemical Sensor Based on TiO2 Nanoparticles/Nafion Biocompatible Film
Modified Glassy Carbon Electrode for Carbamazepine Determination in Phar-
maceutical and Urine Samples, J. Electrochem. Soc. 165 (2018) 946–952.
[41] V.B. Koli, A.G. Dhodamani, A.V. Raut, N.D. Thorat, S.H. Pawar, S.D. Delekar, Visi-
ble light photo-induced antibacterial activity of TiO2-MWCNTs nanocomposites
with varying the contents of MWCNTs, J. Photochem. Photobiol. A Chem. 328
(2016) 50–58.
[64] S.A. Kumar, C.F. Tang, S.M. Chen, Electroanalytical determination of ac-
etaminophen using nano-TiO2/polymer coated electrode in the presence of
dopamine, Talanta 76 (2008) 997–1005.
[65] S. Alwarappan, G. Liu, C.Z. Li, Simultaneous detection of dopamine, ascorbic
acid, and uric acid at electrochemically pretreated carbon nanotube biosensors,
Nanomedicine Nanotechnology, Biol. Med. 6 (2010) 52–57.
[66] N. Mohammed Modawe Alshik Edris, J. Abdullah, S. Kamaruzaman,
M.I. Saiman, Y. Sulaiman, Electrochemical reduced graphene ox-
ide-poly(eriochrome black T)/gold nanoparticles modified glassy carbon
electrode for simultaneous determination of ascorbic acid, dopamine and uric
acid, Arab. J. Chem. 11 (2018) 1301–1312.
[67] X. Zhou, N. Zheng, S. Hou, X. Li, Z. Yuan, Selective determination of
dopamine in the presence of ascorbic acid at
a multi-wall carbon nan-
[42] S.D. Delekar, A.G. Dhodamani, K.V. More, T.D. Dongale, R.K. Kamat, S.F.A. Ac-
quah, N.S. Dalal, D.K. Panda, Structural and Optical Properties of Nanocrys-
talline TiO2 with Multiwalled Carbon Nanotubes and Its Photovoltaic Studies
Using Ru(II) Sensitizers, ACS Omega 3 (2018) 2743–2756.
[43] K.E. Karakitsou, X.E. Verykios, Effects of altervalent cation doping of TiO2 on
its performance as a photocatalyst for water cleavage, J. Phys. Chem. 97 (1993)
1184–1189.
otube-poly(3,5-dihydroxy benzoic acid) film modified electrode, J. Electroanal.
Chem. 642 (2010) 30–34.
[68] A.R. Rajamani, R. Kannan, S. Krishnan, S. Ramakrishnan, S.M. Raj, D. Kumare-
san, N. Kothurkar, M. Rangarajan, Electrochemical sensing of dopamine, uric
acid and ascorbic acid using tRGO-TiO2 nanocomposites, J. Nanosci. Nanotech-
nol. 15 (2015) 5042–5047.
[69] B.L. Hanssen, S. Siraj, D.K.Y. Wong, Recent strategies to minimise fouling in
electrochemical detection systems, Rev. Anal. Chem. 35 (2016) 1–28.
[70] G. Li, S. Yang, L. Qu, R. Yang, J. Li, Simultaneous voltammetric determina-
[44] S. Bagheri, K. Shameli, S.B. Abd Hamid, Synthesis and characterization of
anatase titanium dioxide nanoparticles using egg white solution via Sol-Gel
method, J. Chem 2013 (2013) 848205.
tion of ascorbic acid and uric acid using
a Nafion/multi-wall carbon nan-
[45] L.S. Chougala, M.S. Yatnatti, R.K. Linganagoudar, R.R. Kamble, J.S. Kadadevar-
math, A simple approach on synthesis of TiO2 nanoparticles and its application
in dye sensitized solar cells, J. Nano- Electron. Phys. 9 (2017) 04005.
[46] B.R. Patel, M. Noroozifar, K. Kerman, Review — Nanocomposite-Based Sensors
for Voltammetric Detection of Hazardous Phenolic Pollutants in Water, J. Elec-
trochem. Soc. 167 (2020) 037568.
otubes composite film-modified electrode, J. Solid State Electrochem 15 (2011)
161–166.
[71] B. Kaur, T. Pandiyan, B. Satpati, R. Srivastava, Simultaneous and sensitive de-
termination of ascorbic acid, dopamine, uric acid, and tryptophan with silver
nanoparticles-decorated reduced graphene oxide modified electrode, Coll. Surf.
B Biointerfaces 111 (2013) 97–106.
[47] P. Krzyczmonik, E. Socha, S. Skrzypek, K. Soliwoda, G. Celichowski, J. Gro-
belny, Honeycomb-structured porous poly(3,4-ethylenedioxythiophene) com-
posite layers on a gold electrode, Thin Solid Films 565 (2014) 54–61.
[48] F. Kaedi, Z. Yavari, M. Asmaei, A.R. Abbasian, M. Noroozifar, Ethanol elec-
trooxidation on high-performance mesoporous ZnFe2O4-supported palladium
nanoparticles, New J. Chem. 43 (2019) 3884–3890.
[49] K. Cinková, K. Kianicˇková, D.M. Stankovic´, M. Vojs, M. Marton, L. Švorc, The
doping level of boron-doped diamond electrodes affects the voltammetric
sensing of uric acid, Anal. Methods 10 (2018) 991–996.
[50] K. Cinková, N. Zbojeková, M. Vojs, M. Marton, A. Samphao, Švorc, Electroana-
lytical application of a boron-doped diamond electrode for sensitive voltam-
metric determination of theophylline in pharmaceutical dosages and human
urine, Anal. Methods 7 (2015) 6755–6763.
[72] S. Yan, X. Li, Y. Xiong, M. Wang, L. Yang, X. Liu, X. Li, L.A.M. Al-
shahrani, P. Liu, C. Zhang, Simultaneous determination of ascorbic acid,
dopamine and uric acid using
a glassy carbon electrode modified with
the nickel(II)-bis(1,10-phenanthroline) complex and single-walled carbon nan-
otubes, Microchim. Acta 183 (2016) 1401–1408.
[73] A.S. Kumar, P. Swetha, Ru(DMSO)4Cl2 nano-aggregated Nafion membrane mod-
ified electrode for simultaneous electrochemical detection of hypoxanthine,
xanthine and uric acid, J, Electroanal. Chem. 642 (2010) 135–142.
[74] F. Cui, X. Zhang, A method based on electrodeposition of reduced graphene
oxide on glassy carbon electrode for sensitive detection of theophylline, J. Solid
State Electrochem 17 (2013) 167–173.
[75] A.J. Bard, L.R. Faulkner, Kinetics of electrode reactions, Electrochem. Methods
Fundam. Appl. (2001) 91–92 2nd ed., New York.
[51] M. Shahbakhsh, M. Noroozifar, Copper polydopamine complex/multiwalled
carbon nanotubes as novel modifier for simultaneous electrochemical deter-
mination of ascorbic acid, dopamine, acetaminophen, nitrite and xanthine, J.
Solid State Electrochem 22 (2018) 3049–3057.
[76] N. Sabbaghi, M. Noroozifar, Nanoraspberry-like copper/ reduced graphene ox-
ide as new modifier for simultaneous determination of benzenediols isomers
and nitrite, Anal. Chim. Acta 1056 (2019) 16–25.
[77] Y. Zhou, W. Tang, J. Wang, G. Zhang, S. Chai, L. Zhang, T. Liu, Selective de-
termination of dopamine and uric acid using electrochemical sensor based
[52] A. Carolina Torres, M.M. Barsan, C.M.A. Brett, Simple electrochemical sensor for
caffeine based on carbon and Nafion-modified carbon electrodes, Food Chem
149 (2014) 215–220.
on poly(alizarin yellow R) film-modified electrode, Anal. Methods
3474–3481.
6 (2014)
12

B.R. Patel, S. Imran, W. Ye et al.
Electrochimica Acta 362 (2020) 137094
[78] A.K. Timbola, C.D. Souza, C. Soldi, M.G. Pizzolatti, A. Spinelli, Electro-oxida-
tion of rutin in the presence of p-toluenesulfinic acid, J. Appl. Electrochem.
37 (2007) 617–624.
[79] J. Wang, Analytical Electrochemistry, 2 ed., John Wiley & Sons, Inc., Hoboken,
NJ, 2006.
[82] S.A. Shahamirifard, M. Ghaedi, Z. Razmi, S. Hajati, A simple ultrasensitive
electrochemical sensor for simultaneous determination of gallic acid and uric
acid in human urine and fruit juices based on zirconia-choline chloride-gold
nanoparticles-modified carbon paste electrode, Biosens. Bioelectron. 114 (2018)
30–36.
[80] D. Thomas, Z. Rasheed, J.S. Jagan, K.G. Kumar, Study of kinetic parameters and
development of a voltammetric sensor for the determination of butylated hy-
droxyanisole (BHA) in oil samples, J. Food Sci. Technol. 52 (2015) 6719–6726.
[83] E. Laviron, General expression of the linear potential sweep voltammogram
in the case of diffusionless electrochemical systems, J. Electroanal. Chem. 101
(1979) 19–28.
[81] X. Yan, C. Zhou, Y. Yan, Y. Zhu,
A
Simple and Renewable Nanoporous
[84] J.G. Velasco, Determination of standard rate constants for electrochemical irre-
versible processes from linear sweep voltammograms, Electroanalysis 9 (1997)
880–882.
Gold-based Electrochemical Sensor for Bisphenol A Detection, Electroanalysis
27 (2015) 2718–2724.
13