Talanta 158 (2016) 21–29

Contents lists available at ScienceDirect

Talanta
journal homepage: www.elsevier.com/locate/talanta

Graphene/Nafion composite film modified glassy carbon electrode for

simultaneous determination of paracetamol, aspirin and caffeine in
pharmaceutical formulations
Aydın Yiğit a, Yavuz Yardım a,n, Metin Çelebi b, Abdulkadir Levent c, Zühre Şentürk d,nn
a
Yüzüncü Yıl University, Faculty of Pharmacy, Department of Analytical Chemistry, 65080 Van, Turkey
b
Yüzüncü Yıl University, Faculty of Science, Department of Inorganic Chemistry, 65080 Van, Turkey
c
Batman University, Health Services Vocational College, 72100 Batman, Turkey
d
Yüzüncü Yıl University, Faculty of Science, Department of Analytical Chemistry, 65080 Van, Turkey

art ic l e i nf o a b s t r a c t

Article history: A graphene-Nafion composite film was fabricated on the glassy carbon electrode (GR-NF/GCE), and used
Received 6 March 2016 for simultaneous determination of paracetamol (PAR), aspirin (ASA) and caffeine (CAF). The electro-
Received in revised form chemical behaviors of PAR, ASA and CAF were investigated by cyclic voltammetry and square-wave
11 May 2016
adsorptive anodic stripping voltammetry. By using stripping one for simultaneous determination of PAR,
Accepted 14 May 2016
Available online 16 May 2016
ASA and CAF, their electrochemical oxidation peaks appeared at þ0.64, 1.04 and 1.44 V, and good linear
current responses were obtained with the detection limits of 18 ng mL  1 (1.2  10  9 M), 11.7 ng mL  1
Keywords: (6.5  10  8 M) and 7.3 ng mL  1 (3.8  10  8 M), respectively. Finally, the proposed electrochemical sensor
Paracetamol was successfully applied for quantifying PAR, ASA and CAF in commercial tablet formulations.
Aspirin
& 2016 Elsevier B.V. All rights reserved.
Caffeine
Graphene-Nafion composite film modified
glassy carbon electrode
Pharmaceutical formulation

1. Introduction spectrophotometry [10–13], and voltammetry [14–27]. However, a

Paracetamol (also known as acetaminophen, PAR) and aspirin
(also known as acetylsalicylic acid, ASA) are effective and im-
portant analgesic and antipyretic agents. These compounds are
often combined with caffeine (CAF) in multidrug pharmaceutical
formulations to improve the pharmacological value of these pre-
parations. The fixed combination of PAR, ASA and CAF is widely
used for the treatment of different types of pain caused by tension
headaches, migraine headaches, muscle aches, menstrual cramps,
arthritis, toothaches, the common cold, or nasal congestion [1–4].
Owing to the therapeutic use of this drug combination, it is still
highly needed to develop a fast, simple and reliable method for
simultaneous monitoring of their quantities in pharmaceutical
formulations.
The most common analytical methods used for the determi-
nation of PAR, ASA and CAF, alone or in their mixtures with dif-
ferent compounds include liquid chromatography [5–10],
n
Corresponding author.
nn
Corresponding author.
E-mail addresses: yavuzyardim2002@yahoo.com (Y. Yardım),
zuhresenturk@hotmail.com (Z. Şentürk).
survey of the literature revealed that only three voltammetric
procedures have been performed for the simultaneous determi-
nation of PAR, ASA and CAF in their ternary mixtures [28–30]. All
studies were published in recent years using carbon-based elec-
trodes. Sanghavi and Srivastava reported an adsorptive stripping
voltammetry at in situ surfactant-modified multiwalled carbon
nanotube paste electrode for the determination of the mixture of
these active ingredients in drug formulations [28]. In that study
the detection limits were observed as 2.6  10  8, 8.5  10  8, and
8.8  10  8 M for PAR, ASA and CAF, respectively. Goyal et al. pre-
pared a poly-1,5-diaminonapthalene modified pyrolytic graphite
electrode for the square-wave voltammetric determination of PAR,
ASA and CAF in their mixture, obtaining detection limits of
0.57  10  10, 0.93  10  10 and 0.64  10  10 M, respectively [29].
Previous work of our research group has also demonstrated the
effectiveness of a cathodically pretreated boron-doped diamond
electrode to determine PAR, ASA and CAF simultaneously, with
detection limits of 3.9  10  6, 7.3  10  6 and 1.4  10  6 M, re-
spectively [30]. In spite of these works, the exploration of novel
electrode materials for the simultaneous determination of PAR,
ASA and CAF is still a challenge.
Recently, graphene-based materials have received world-wide

http://dx.doi.org/10.1016/j.talanta.2016.05.046
0039-9140/& 2016 Elsevier B.V. All rights reserved.
22 A. Yiğit et al. / Talanta 158 (2016) 21–29
attention in electrochemical sensors [31–36]. This is because gra-
phene, a two-dimensional (2D) carbon material with one-atom-
thickness has excellent electrical conductivity, enormous electro-
active area, high mechanical and thermal resistance, and low cost.
More recently, it is attractive to develop graphene-based nano-
composite films for enhancing the sensitivity of the electro-
chemical sensors. Nafion, an ionic polymer, has been compre-
hensively employed for preparation of modified electrodes with
excellent properties such as antifouling capacity, chemically in-
ertness, high conductivity and catalytic activity. In corporation of
graphene, Nafion has also improved its dispersity in aqueous so-
lution and prevented its aggregation [19,37–40].
In keeping with relatively new and limited data on voltam-
metric determination of PAR, ASA and CAF simultaneously, this
work aims to present a highly sensitive and selective electro-
chemical nanosensor based on the graphene/Nafion composite
film modified glassy carbon electrode for their simultaneous de-
termination in pharmaceutical dosage forms.
2. Experimental
2.1. Chemicals and solutions
PAR, ASA, CAF, potassium permanganate, sodium nitrate and
hydrazine hydrate were purchased from Sigma. PAR and CAF stock
solutions (0.5 mg mL  1) were prepared in water. ASA stock solu-
tions (0.5 mg mL  1) were prepared in 0.01 M sodium hydroxide.
Graphite powder, sulfuric acid (98%) and hydrogen peroxide so-
lution (30%) were received from Merck. Nafions was purchased
from Fluka. Nafion solution (0.5%, w/v) was prepared in methanol-
water (50:50 v/v) using a magnetic stirring bar at about 3 h in
50 °C. The other chemicals were of analytical grade and used
without further purification. Three different supporting electro-
lytes, namely Britton-Robinson (BR, 0.1 M, pH 2–8), acetate (ABS,
0.1 M, pH 4.7) and phosphate (PBS, 0.1 M, pH 2.5 and 7.4) buffers
were used. All stock solutions were preserved at 4 °C when not in
use and protected from daylight during use in the laboratory.
Aqueous solutions were prepared with deionized water further
purified via a Milli-Q unit (Millipore). All the electrochemical
measurements were operated in ambient conditions.
2.2. Apparatus and measurements
FT-IR spectra were recorded on an IR-Affinity-1 Fourier Trans-
form Infrared Spectrophotometer (Shimadzu, Japan). UV–vis ab-
sorption spectra were obtained using a UV-2600 spectro-
photometer (Shimadzu, Japan). The Raman spectra were carried
out using Horiba Jobin-Yvon HR800 Raman spectrometer (Horiba,
Japan). The surface morphologies of the electrodes were examined
using a scanning electron microscope (SEM, Zeiss EVO MA15)
(Zeiss, Germany).
The whole electrochemical measurements were performed
with an Autolab PGSTAT204 electrochemical analyzer with GPES
4.9 software package (EcoChemie, The Netherlands). A three-
electrode-system in a 10-mL one-compartment voltammetric cell
included bare glassy carbon electrode (GCE, 3 mm in diameter,
Model MF-2012, BAS) or modified GCE as working electrode, an
Ag/AgCl (3 M NaCl, Model RE-1, BAS) as reference electrode, and a
platinum wire as auxiliary electrode, respectively.
Cyclic voltammetry (CV) was employed for preliminary studies
on the electrochemical behavior of PAR, ASA and CAF. Square-wave
adsorptive anodic stripping voltammetry (SW-AdASV) was used
for the development of electroanalytical methodology and the si-
multaneous determination of these drugs in commercial drug
samples.
Unless otherwise stated, 0.1 M PBS (pH 2.5) was used as the
supporting electrolyte. The optimized experimental conditions for
stripping voltammetric analysis of samples were as follows: Fre-
quency, 75 Hz, pulse amplitude, 50 mV, scan increment, 14 mV,
accumulation time 60 s, and accumulation potential 0.1 V. Suc-
cessive measurements were carried out by repeating the above
assay protocol on the modified GCE.
2.3. Graphene synthesis
High-quality graphene (GR) was prepared by an effective che-
mical method consisting two steps. First, graphene oxide (GO) was
synthesized from graphite powder using Hummers' method [41].
GR was then obtained by reduction of GO according to the lit-
erature procedure [42]. The general procedure for this purpose
was as follows:
In the first step, 5.0 g of graphite flakes and 2.5 g of sodium
nitrate were mixed together followed by the addition of 115 mL of
concentrated sulfuric acid, and the mixture was cooled using an
ice bath to 0 °C. 15.0 g of potassium permanganate was added
slowly to the above solution while keeping the temperature below
20 °C to avoid overheating and explosion, and then the mixture
was stirred at 35 °C for 30 min Afterwards, the resulting solution
was diluted by adding 230 mL of water which produced a large
exotherm to 98 °C. To fully oxidize graphite, the reaction tem-
perature was maintained at 98 °C for 15 min After cooling with
water-bath for 10 min, the reaction mixture was diluted with
600 mL of water, and 5 mL of 30% hydrogen peroxide to produce
another exotherm. After air cooling, the mixture was purified by
sifting, filtration, multiple washings with 5% hydrochloric acid and
water, centrifugations, and decanting followed by a final vacuum
drying to give 2.5 g of solid product.
As a continuous step for the preparation of reduced graphene
oxide (GR or rGO), 100 mg of GO was added to 50 mL water, which
yielded an unhomogeneous yellow-brown dispersion. This dis-
persion was then sonicated until it became clear without visible
particles. Next, this solution was transferred into a mixture that
contained 200 mL of dimethylformamide and 17 mL of hydrazine
hydrate. The final mixture was heated at 100 °C under reflux
conditions for 24 h, in which the GR was gradually precipitated out
as a black powder. This product was filtrated, washed with water
(5  50 mL) and ethanol (5  50 mL), and dried under vacuum to
give 0.70 g of black solid.
2.4. Preparation of the modified electrodes
Prior to modification, the bare GCE was polished manually with
slurries prepared from 0.05 mm alumina powder on a microcloth
polishing pad (BAS), and then rinsed with water thoroughly. After
successive sonication in 1:1 nitric acid, ethanol, and water, the
electrode was rinsed with water. The cleaned GCE was dried at
room temperature. 3.0 mg of GR was added into 3.0 mL of Nafion-
alcohol solution (0.5%, w/v), followed by ultrasonication for 2 h to
form a homogeneous solution. Next, 10 mL of the homogeneous
dispersion of graphene-Nafion (labeled as GR-NF) solution was
applied to the clean GCE and allowed to dry in air at room tem-
perature for 8 h. The obtained electrode was noted as GR-NF/GCE.
For comparison, the graphene modified GCE (GR/GCE) was treated
by the same way as the GR-NF/GCE. Prior to use, the GR/GCE and
GR-NF/GCE were pretreated in the selected buffer solution by
cyclic scans between potentials of þ0.2 and þ 1.65 V (10 cycles) at
scan rate of 100 mV s  1.
2.5. Sample preparation for assay
Pharmaceutical formulations were obtained from local drug
 A. Yiğit et al. / Talanta 158 (2016) 21–29
store. Individual analysis of PAR and ASA was first carried out in
pharmaceutical formulations such as Parols tablet (Atabay Co.,
Turkey), and Aspirins tablet (Bayer Turk Co., Turkey). For si-
multaneous determination of PAR, CAF and ASA, Thomapyrins
tablet (Boehringer Ingelheim Co. Germany) samples were em-
ployed. For all the samples, ten tablets were weighed, finely
ground to a powder and thoroughly mixed. Aspirin and Thoma-
pyrin tablet powders which contain ASA were boiled with 0.01 M
sodium hydroxide for 10 min to hydrolyze it completely into sal-
icylic acid [28]. The obtained amount of salicylic acid was equal to
the original amount of ASA in the samples. In the case of Parol
tablet, sample was prepared in water. The samples were further
diluted in 0.1 M PBS at pH 2.5 to obtain the desired concentration
in the cell. Then the results were evaluated using the standard
addition method.
3. Results and discussion
3.1. Characterization of GO and GR
FT-IR spectroscopy was performed to investigate the functional
groups in the structure of graphite, GO and GR (rGO) (Fig. 1A).
Fig. 1. (A) FT-IR spectra of graphite, GO and rGO samples, (B) Raman spectra of GO and rGO samples, and (C) UV–vis spectra of 0.05 mg mL  1 GO and 0.05 mg mL  1 rGO in
the aqueous dispersion.
23
From the FT-IR pattern, it was observed that GO shows a broad and
intense peak between 3452–3402 cm  1 assigned to O–H stretch-
ing vibrations mode of intercalated water. In the spectrum, the
expected bands were also observed C ¼ O stretching vibration
(1728–1708 cm  1), C ¼C from un-oxidized sp2 CC bonds
(1577–1616 cm  1), and C–O vibrations (1188–1250 cm  1). The
results confirmed the successful oxidation of graphite. The FT-IR
spectrum of rGO differs from that of GO by the significant decrease
of the intensities of C ¼O functional groups bands. Furthermore,
the intensity of the sharp absorption band between
3452–3402 cm  1 decreases in the rGO. These results agree with
those reported in the literature [43] and confirm that most of the
oxygenated functionalities have been removed and a substantial
reduction of GO to rGO has been achieved.
Fig. 1B displays Raman characterization of GrO and GR (rGO). In
the Raman spectrum of GO, two characteristic bands (D and G
band) were observed at 1333 and 1584 cm  1, respectively. The
Raman spectrum of rGO has also two apparent bands [D
(1332 cm  1) and G (1576 cm  1)], while an increased intensity
ratio of D band to G band was observed after the chemical re-
duction of GO because of the sp2 carbon orientation. This change
indicate the successful reduction of GO [44].
The electrochemical reduction of GO was also monitored by
24 A. Yiğit et al. / Talanta 158 (2016) 21–29

Fig. 2. SEM images of the GR/GCE (A) and GR-NF/GCE (B) surfaces.

UV–vis spectroscopy measurements. The UV–vis spectrum (Fig. 1C)

of GO in water shows strong absorption centered at 232 nm and
306 nm corresponding to the π-π* transitions of aromatic C-C
ring and n-π* transitions of C ¼O bonds, respectively. When GO
was reduced to GR (rGO), the peak at 306 nm disappears and the
peak at 232 nm redshifts to 266 nm, indicating the removal of
most oxygen groups and restoration of electronic conjugation
within graphene sheets [45,46].
The morphology of GR/GCE (A) and GR-NF/GCE (B) were illu-
strated using SEM as presented in Fig. 2. For SEM study, the head
of GCE was easily removed. The GR/GCE and GR-NF/GCE were
prepared as it was described in Section 2.4. As shown from the
figure, the surface of the GR-NF/GCE is rougher and crumpled than
that of GR/GCE, which is indicated that the Nafion film can be
homogeneously distributed on the surface of GR/GCE with larger
surface area.

3.2. Independent CV behavior of PAR, ASA and CAF

A CV was used to investigate the electrochemical behaviors of

single PAR (4 mg mL  1), ASA (40 mg mL  1) and CAF (50 mg mL  1).
Fig. 3 shows the electrochemical responses obtained at the surface
of bare GCE, GR/GCE and GR-NF/GCE at scan rate of 50 mV s  1 in
0.1 M PBS (pH 2.5). As shown in Fig. 3A, the anodic peak potential
of PAR was observed at about þ0.605 V at bare GCE and the
modified GCEs. The anodic peak currents were 0.99, 1.10 and
2.34 mA at bare GCE, GR/GCE and GR-NF/GCE, respectively, show-
ing the increased current response at the GR-NF/GCE. In the re-
verse scan, weak reduction peak was observed at each electrode.
In the case of ASA as seen in Fig. 3B, the anodic peak potentials and
Fig. 3. Cyclic voltammograms of bare GCE (a), GR/GCE (b) and GR-NF/GCE (c) in
0.1 M PBS (pH 2.5) containing 4 μg mL  1 PAR (A), 40 μg mL  1 ASA (B) and 50 μ
g mL  1 CAF (C). Scan rate: 50 mV s  1.
currents were obtained as þ1.072 V and 1.37 mA, þ1.014 V and
1.45 mA, and þ0.984 V and 1.81 mA at bare GCE, GR/GCE and GR-
NF/GCE, respectively. On the other hand, for CAF, the anodic peak
potentials were þ1.45, þ1.40 and þ1.34 V at bare GCE, GR/GCE
and GR-NF/GCE, while the peak currents were observed as 4.62,
 A. Yiğit et al. / Talanta 158 (2016) 21–29 25

5.24, and 8.99 mA, respectively (Fig. 3C). No cathodic peaks of ASA
and CAF were observed at the surface of three types of electrodes.
These observations indicate that both GR and NF are responsible
for the changes of the voltammetric behavior of these species. The
difference in the shifts of the oxidation peak potential for these
three compounds may be explained by their different sensitivity
towards the electronic properties, surface chemistry and struc-
tures of the electrodes. Consequently, GR/NF composites have high
electrocatalytic effects on the oxidation of these three compounds.
The effect of scan rate was investigated in PAR, ASA and CAF by
CV at the surface of GR-NF/GCE under the above experimental
conditions (data not shown). The oxidation peak currents (ip) of
PAR, ASA and CAF were proportional to the scan rate (v) in the
range of 50–700 mV s  1, and the linear regression equations can
be expressed as ip (μA)¼0.047 v (mV s  1)  0.808 (PAR, n ¼9,
r ¼0.999), ip (μA) ¼0.004 v (mV  1) þ1.057 (ASA, n¼9, r ¼0.981),
and ip (μA)¼0.055 v (mV s  1) þ9.269 (CAF, n ¼9, r ¼0.988).
The findings above indicated that the electrocatalytic oxidation
processes for all the three species were adsorption-controlled at
the electrode surface, suggesting the possibility of setting up
method to determine the three compounds by AdSV.

3.3. Stripping analysis of PAR, ASA and CAF at the GR-NF/GCE

After the results given in previous section, the attention was

then turned to the effect of stripping conditions on the oxidation
process of these three compounds by application GR-NF/GCE.

3.3.1. Effect of pH
The effect of solution pH on the oxidation response of GR-NF/
GCE for 0.05 mg mL  1 PAR, 1 mg mL  1 ASA and 1 mg mL  1 CAF in
the mixed solution was carefully investigated by SW-AdASV. In
Fig. 4A, this parameter was established in a series of BR buffer
solutions with pH 2.0–8.0 by carrying out stripping measurement
on the three species solution, with an open-circuit accumulation at
60 s It can be easily concluded that the oxidation peak current of
PAR was very high at pH 5.0, while the maximum oxidation peak
currents of ASA and CAF were observed at pH 2.0. The relation-
Fig. 4. SW stripping voltammograms of 0.05 μg mL  1 PAR, 1.0 μg mL  1 ASA and
ships between the peak potential of these compounds and pH 1.0 μg mL  1 CAF solutions in BR buffer at different pH values between pH 2–8 (A),
were also investigated, respectively. The linear regression equa- and in various supporting electrolytes (B) at the GR-NF/GCE. tacc ¼60 s at open
tions for peak potentials (Ep) and pH could be expressed as Ep circuit condition; SWV parameters: frequency, 50 Hz; scan increment, 8 mV; pulse
(V) ¼ 0.053 pH þ 0.729 (PAR, r ¼0.999), Ep (V) ¼  amplitude, 30 mV.

0.032 pH þ1.123 (ASA, r ¼0.965). The slopes of the two regression

equations for PAR and ASA were 53 and 32 mV/pH, respectively,
indicating that the redox reactions of two compounds at the sur-
face of GR-NF/GCE were accompanied by proton transfer. However,
in the case of CAF, the influence of pH on the peak potential can
almost be neglected since only slight negative potential shift was
observed by raising the solution pH. Fig. 4B depicts the stripping
voltammograms in various supporting electrolytes in the various
concentrations of the three species mixed solution. Using PBS pH
2.5, oxidation peak potentials of þ0.61, 1.03 and 1.40 V were ob-
tained, with the peak currents of 22.88, 10.49 and 9.60 μA for PAR,
ASA and CAF, respectively. In the case of ABS pH 4.7, the oxidation
peak potentials were þ0.49, 0.99 and 1.40 V for PAR, ASA and CAF,
with the peak currents of 16.29, 10.18 and 6.13 μA, respectively. On
the other hand, in PBS pH 7.4, these values were þ 0.34 V (26.04
μA) for PAR, 0.86 V (3.652 μA) for ASA, and 1.37 V (7.33 μA) for
CAF. As seen clearly from the figures, the highest peak current of
PAR was obtained in PBS at pH 7.4 whereas the significant increase
of the peak currents of ASA and CAF were observed in PBS at pH
2.5. Therefore, pH 2.5 employing 0.1 M PBS was chosen for the
subsequent analytical experiments. It should be noted at this point
that one additional peak appeared at less positive potentials than
the main oxidation peak in strongly acidic solutions for PAR.
However, this oxidation step was less distinct and well separated
from the main one. Thus, the optimum potential range for analy-
tical studies was selected as þ 0.45 to þ1.75 V.
3.3.2. Effect of accumulation time and accumulation potential
The effect of accumulation time (tacc) and accumulation po-
tential (Eacc) (data not presented) at the surface of GR-NF/GCE
were studied in the mixed solution of PAR (0.05 mg mL  1), ASA
(1 mg mL  1), and CAF (1 mg mL  1) under the optimum experi-
mental conditions. The influence of the tacc upon the analytical
signal was examined in the range 0–300 s at open-circuit condi-
tion. From 0 to 60 s, there was a linear increase in the peak current
of PAR, ASA and CAF. Beyond this range, the current increased
slightly for all three compounds. The tacc of 60 s is short and
doubtlessly advantageous for practical use of this electrode.
Therefore, this value was selected for all the SW-AdASV experi-
ments. Keeping the tacc as 60 s, the dependence of the stripping
peak current on the Eacc was evaluated either at open-circuit
condition or over the potential range þ0.1 to þ 0.5 V. The strip-
ping peak currents of PAR and ASA reached their maximum at an
Eacc of þ0.1 V, whereas that for CAF occurred at open-circuit
condition. For all further studies, þ 0.1 V was selected as accumu-
lation potential.
26 A. Yiğit et al. / Talanta 158 (2016) 21–29
Fig. 5. SW stripping voltammograms obtained at different concentrations of
(A) PAR (0.00125, 0.0025, 0.005, 0.01, 0.04, 0.06, 0.08, 0.1 and 0.25 μg mL  1) in the
presence of 1.0 μg mL  1 ASA and 2.5 μg mL  1 CAF; (B) ASA (0.05, 0.1, 0.15, 0.25, 0.5,
1.0, 2.5 and 5.0 μg mL  1) in the presence 0.04 μg mL  1 PAR and 2.5 μg mL  1 CAF;
(C) CAF (0.05, 0.1, 0.25, 0.5, 1.0, 2.5, 5.0 and 7.5 μg mL  1) in the presence 0.04 μ
g mL  1 PAR and 1.0 μg mL  1 ASA, in PBS (pH 2.5) at the GR-NF/GCE. Inset: Cor-
responding calibration curves. tacc ¼60 s at 0.1 V; SWV parameters: frequency,
75 Hz; scan increment, 14 mV; pulse amplitude, 50 mV.
3.3.3. Effect of SWV parameters
In order to optimize the experimental set-up for PAR, ASA and
CAF determination, the dependence of stripping responses on
SWV parameters was finally analyzed. The ranges studied were
25–100 Hz for SW frequency (f), 8–18 mV for scan increment
(ΔEs), and 25–75 mV for pulse amplitude (ΔEsw). For entire ana-
lysis the optimized values were: f, 75 Hz; ΔEs, 14 mV; and ΔEsw,
50 mV.
3.4. Determination of PAR, ASA and CAF at GR-NF/GCE
Based on the results described above, an analytical method was
proposed for determination of PAR, ASA and CAF individually, se-
lectively and simultaneously at GR-NF/GCE. For this purpose, three
cases were studied.
For the first case, individual detection of PAR, ASA and CAF was
implemented. Fig. S1(A), (B) and (C) illustrates the SW-AdASV
curves of individual detection of PAR, ASA and CAF, respectively.
In the second study, in a mixture of PAR, ASA and CAF, the
concentration of one species changed while the concentrations of
the other two remained constant. The results are shown in Fig. 5.
The SW-AdASV results show three well distinguished stripping
peaks at potentials of þ0.64, þ1.05 and þ1.44 V, corresponding to
the oxidation of PAR, ASA and CAF, respectively. The detected
potential difference of 0.41 V between PAR and ASA and 0.39 V
between ASA and CAF is large enough for separate determination
of these three substances in their mixture. As shown in Fig. 5A, the
current response of PAR increases regularly as it concentration is
increased at fixed concentrations of ASA and CAF (their oxidation
currents remain fairly constant). Similarly, Fig. 5B and C show that
the peak current responses of ASA and CAF increase with the in-
crease of their contents, while the peak currents of the other two
compounds remain fairly stable.
After the second case, to study the possibility of GR-NF/GCE for
the simultaneous determination of PAR, ASA and CAF, the stripping
peak currents of these species by simultaneously changing their
concentrations in a mixture were measured. As can be seen in
Fig. 6, the respective SW-AdASV curves for PAR, ASA, and CAF at
þ0.62, 1.06 and 1.44 V present good linearities in the investigated
concentration ranges.
Analytical parameters for three cases are presented in Table 1
(A)–(C). The sensitivity of the proposed method was checked in
terms of the limit of detection (LOD) values. LOD was calculated
using the following equations: LOD ¼3 s/m, where s is the standard
deviation of the peak current (three runs) of the lowest con-
centration of the related linearity range, and m the slope of the
related calibration plot. It is worthy remarking that the sensitivity
of the analytical curves for PAR, ASA, and CAF observed in all of the
aforementioned cases did not present significant differences.
The sensitivity in terms of LOD clearly demonstrates the su-
periority of this electroanalytical approach on GR-NF/GCE over the
voltammetric methods on BDD electrode [30]. More improved LOD
values have been found than that reached at in situ surfactant-
modified multiwalled carbon nanotube paste electrode [28].
However, it presents higher LOD values with poly-1,5-diamino-
napthalene modified pyrolytic graphite electrode [29].
3.5. Repeatability, stability and interferences
The repeatability of GR-NF/GCE was examined by measuring
the stripping responses in a mixed solution of PAR, ASA and CAF
using their lowest concentrations (0.00125, 0.15, and 0.05 mg mL  1
for PAR, ASA, and CAF, respectively) at various times within 24 h.
The relative standard deviations (RSD) in peak currents were
found to be 3.49%, 4.57% and 5.72% for detecting PAR, ASA and CAF,
respectively. For evaluating the electrode fabrication repeatability,
 A. Yiğit et al. / Talanta 158 (2016) 21–29 27

Fig. 6. SW stripping voltammograms of GR-NF/GCE in PBS (pH 2.5) containing different concentrations of PAR þ ASAþ CAF in mg mL  1, from inner to outer:
0.00125þ 0.15þ 0.05; 0.005þ 0.2 þ 0.1; 0.0075þ 0.4 þ 0.2; 0.02 þ 1.0 þ 0.5; 0.04þ 1.25 þ 0.75; 0.06 þ2.0 þ 1.5; and 0.075þ 3.0 þ 2.0, respectively. Insets: calibration plots of PAR,
ASA and CAF. Other operating conditions as indicated in Fig. 5.

Table 1.
The calibration characteristics of PAR, ASA and CAF obtained at GR-NF/GCE.

Compound Linear working range Linear regression equation r LOD RSD %

(A)
PAR 0.001–0.225 μg mL  1 ip (μA) ¼798.70 C (μg mL  1)  1.931 0.999 0.13 ng mL  1 2.88
6.6  10  9–1.5  10  6 M 8.6  10  10 M
ASA 0.025–5.0 mg mL  1 ip (μA) ¼13.89 C (μg mL  1) þ 1.259 0.996 7.4 ng mL  1 3.72
1.4  10  7–2.9  10  5 M 4.1  10  8 M
CAF 0.025–7.5 mg mL  1 ip (μA) ¼8.83 C (μg mL 1
)þ 0.859 0.999 6.1 ng mL  1 4.61
1.3  10  7–3.9  10  5 M 3.1  10  8 M

(B)
PAR 0.00125–0.25 μg mL  1 ip (μA) ¼778.50 C (μg mL  1) þ 0.196 0.999 0.16 ng mL  1 3.14
8.3  10  9–1.7  10  6 M 1.1  10  9 M
ASA 0.05–5.0 mg mL  1 ip (μA) ¼12.50 C (μg mL  1)  0.680 0.999 7.7 ng mL  1 4.65
2.9  10  7–2.9  10  5 M 4.3  10  8 M
CAF 0.05–7.5 mg mL  1 ip (μA) ¼8.26 C (μg mL  1)þ 1.433 0.996 7.0 ng mL  1 5.22
2.6  10  7–3.9  10  5 M 3.6  10  8 M

(C)
PAR 0.00125–0.075 μg mL  1 ip (μA) ¼460.90 C (μg mL  1)þ 0.575 0.998 0.18 ng mL  1 4.08
8.3  10  9–5.1  10  7 M 1.2  10  9 M
ASA 0.15–3.0 μg mL  1 ip (μA) ¼8.69 C (μg mL 1
)  0.348 0.999 11.7 ng mL  1 5.43
8.7  10  7–1.7  10  5 M 6.5  10  8 M
CAF 0.05–2.0 μg mL  1 ip (μA) ¼10.10 C (μg mL  1) þ 0.663 0.997 7.3 ng mL  1 6.31
2.6  10  7–1.0  10  5 M 3.8  10  8 M

(A) Statistical data for individual compounds.

(B) Statistical data for one compound when the concentrations of the other two compounds are kept constant.
(C) Statistical data for all the three compounds simultaneously.
28 A. Yiğit et al. / Talanta 158 (2016) 21–29
Table 2.
Results obtained in the determination of PAR, ASA and CAF in pharmaceutical formulations using the SW-AdASV (proposed) and HPLC (comparative) methods.
Sample Compound Tablet label value (mg)
Parols tablet PAR 500
Aspirins tablet ASA 500
Thomapyrins tablet PAR 200
ASA 250
CAF 50
a
Mean 7 s (n ¼3).
b
Relative error 1 (%)¼ [(voltammetric value-label value)/label value]  100.
c
Relative error 2 (%) ¼[(voltammetric value-HPLC value)/HPLC value]  100.
standard PAR, ASA and CAF solutions was measured by three in-
dependent modified electrodes. The RSD (n ¼3) values for PAR
(0.00125 mg mL  1), ASA (0.15 mg mL  1) and CAF (0.05 mg mL  1)
were 3.92%, 5.44% and 7.33%, respectively. The experimental re-
sults showed that the proposed electrode has a satisfactory re-
peatability performance for the determination of these com-
pounds. After the modified electrode was stored for 21 days, a
decrease of the oxidation peak current of PAR, ASA and CAF was
observed with the RSD (n ¼3) of 3.58%, 4.75% and 6.22%, respec-
tively, which could be attributed to the acceptable storage stability
of the proposed electrode.
To investigate the selectivity of the GR-NF/GCE, various sub-
stances as compounds potentially interfering with the determi-
nation of PAR, ASA and CAF was studied under optimum condi-
tions. The potentially interfering substances were chosen from the
group of substances commonly found with PAR, ASA and CAF in
pharmaceuticals and/or in biological fluids. The tolerance limit
was taken as the maximum concentration of the interfering sub-
stances, which caused an approximately 7 5% relative error in the
determination of the three compounds. According to the results, it
was found that K þ , Na þ , Ca2 þ , Zn2 þ , Mg2 þ , SO42  , NO3  , glucose,
fructose, sucrose, saturated starch solution, ascorbic acid (50-fold),
and dopamine (50-fold) did not interfere with the determination
of these compounds. However, uric acid with equal molar showed
interference with PAR, this interference could be solved, if neces-
sary (e.g., in urine samples), by using the HPLC technique.
3.6. Analytical applications
Finally, the practical application of the proposed modified
electrode was illustrated by simultaneously determining the con-
centrations of PAR, ASA, and CAF in commercial pharmaceutical
samples (tablets) by means of the standard addition method. The
data obtained employing the herein proposed voltammetric pro-
cedure for tablet formulations were also compared to those ob-
tained by the reverse-phase high-performance liquid chromato-
graphic method (HPLC) with diode-array detection (DAD) which
was reported recently in our previous paper [30].
As it can be seen in Table 2, it can be concluded that the results
obtained by the proposed voltammetric method show a satisfac-
tory agreement with those obtained by HPLC. Applying the Stu-
dent′s t-test [47] to the results obtained both methods, the cal-
culated t-values (1.84, 1.32 and 2.15 for PAR, ASA and CAF, re-
spectively) were less than the critical value (2.78, α ¼0.05) for the
mean of three experiments. The results indicate that there are no
significant differences between the data obtained using the two
methods at the 95% confidence level. Besides, such results are
within an acceptable relative error. On the other hand, the pro-
posed voltammetric method has the advantage of being less time-
consuming and less expensive than HPLC [30].
SW-AdASVa (mg) HPLCa (mg) E1b (%) E2c (%)
491.7 7 7 504 72  1.7  2.4
517.0 7 11 508.7 73 3.4 1.6
1887 8 197 71  6.0  4.6
255 7 9 248 71 2.0 2.8
477 3 51 72  6.0  7.8
4. Conclusions
In the present study, a novel GR-NF/GCE was fabricated to in-
vestigate the electrochemical oxidation behavior of PAR, ASA and
CAF. The obtained results allow concluding that the GR-NF/GCE
could be promising alternative with a great potential for the
quantitative determination of PAR, ASA and CAF in their mixture
as commonly found in pharmaceutical formulations.
Acknowledgements
The authors gratefully acknowledge financial support from The
Scientific and Technological Research Council of Turkey (TUBITAK)
(Project number: KBAG114Z405).
Appendix A. Supplementary material
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.talanta.2016.05.
046.
References
[1] M. Jozviak-Bebenista, J.Z. Nowak, Paracetamol: mechanism of action, appli-
cations and safety concern, Acta Pol. Pharm. Drug. Res. 71 (2014) 11–23.
[2] R. Amann, B.A. Peskar, Anti-inflammatory effects of aspirin and sodium sali-
cylate, Eur. J. Pharmacol. 447 (2002) 1–9.
[3] N.L. Benowitz, Clinical pharmacology of caffeine, Annu. Rev. Med. 41 (1990)
277–288.
[4] R.B. Lipton, W.F. Stewart, R.E. Ryan, J. Saper, S. Silberstein, F. Sheftell, Efficacy
and safety of acetaminophen, aspirin, and caffeine in alleviating migraine
headache painthree double-blind, randomized, placebo-controlled trials, Arch.
Neurol. 55 (1998) 210–217.
[5] A.W. Abu-Qare, M.B. Abou-Donia, A validated HPLC method for the determi-
nation of pyridostigmine bromide, acetaminophen, acetylsalicylic acid and
caffeine in rat plasma and urine, J. Pharm. Biomed. Anal. 26 (2001) 939–947.
[6] J.T. Franeta, D. Agbaba, S. Eric, S. Pavkov, M. Aleksic, S. Vladimirov, HPLC assay
of acetylsalicylic acid, paracetamol, caffeine and phenobarbital in tablets, Il,
Farmaco 57 (2002) 709–713.
[7] J.K. Mbinze, P. Lebrun, B. Debrus, A. Dispas, N. Kalenda, J.M.T. Mbay,
T. Schofield, B. Boulanger, E. Rozet, pH Hubert, R.D. Marini, Application of an
innovative design space optimization strategy to the development of liquid
chromatographic methods to combat potentially counterfeit nonsteroidal anti-
inflammatory drugs, J. Chromatogr. A 1263 (2012) 113–124.
[8] H.A.-A. Hashem, Chromatographic application on calixarene bonded sta-
tionary phases: A stability indicating method for simultaneous determination
of paracetamol, caffeine and acetylsalicylic acid in excedrin tablets, Chroma-
tographia 71 (2010) 31–35.
[9] M.L. Castillo-García, M.P. Aguilar-Caballos, A. Gómez-Hens, Determination of
acetylsalicylic acid and its major metabolites inbovine urine using ultra per-
formance liquid chromatography, J. Chromatogr. B 985 (2015) 85–90.
[10] F.F. Belal, M.K.S. El-Din, M.M. Tolba, H. Elmansi, Derivative spectrophotometric
and liquid chromatographic methods for the simultaneous determination of
metoclopramide hydrochloride and aspirin in pharmaceuticals, J. Chromatogr.
Sci. 52 (2014) 1224–1232.
[11] R. Hajian, A. Soltaninezhad, The spectrophotometric multicomponent analysis
of a ternary mixture of paracetamol, aspirin, and caffeine by the double di-
visor-ratio spectra derivative method, J. Spectrosc. 2013 (2013) 1–7.
 A. Yiğit et al. / Talanta 158 (2016) 21–29
[12] A.C. Moţ, F. Soponar, A. Medvedovici, C. Sârbu, Simultaneous spectro-
photometric determination of aspirin, paracetamol, caffeine, and chlorphe-
namine from pharmaceutical formulations using multivariate regression
methods, Anal. Lett. 43 (2010) 804–813.
[13] L.F. Sallum, F.L.F. Soares, J.A. Ardila, R.L. Carneiro, Determination of acet-
ylsalicylic acid in commercial tablets by SERS using silver nanoparticle-coated
filter paper, Spectrochim. Acta A Mol. Biomol. Spectrosc. 133 (2014) 107–111.
[14] B.C. Lourenção, R.A. Medeiros, R.C. Rocha-Filho, L.H. Mazo, O. Fatibello-Filho,
Simultaneous voltammetric determination of paracetamol and caffeine in
pharmaceutical formulations using a boron-doped diamond electrode, Talanta
78 (2009) 748–752.
[15] E.R. Sartori, R.A. Medeiros, R.C. Rocha, O. Fatibello, Square-wave voltammetric
determination of acetylsalicylic acid in pharmaceutical formulations using a
boron-doped diamond electrode without the need of previous alkaline hy-
drolysis step, J. Braz. Chem. Soc. 20 (2009) 360–366.
[16] M.H. Pournaghi-Azar, A. Saadatirad, Simultaneous determination of para-
cetamol, ascorbic acid and codeine by differential pulse voltammetry on the
aluminum electrode modified by thin layer of palladium, Electroanalysis 22
(2010) 1592–1598.
[17] M. Mazloum-Ardakani, H. Beitollahi, M.K. Amini, F. Mirkhalaf, M. Abdollahi-
Alibeik, New strategy for simultaneous and selective voltammetric determi-
nation of norepinephrine, acetaminophen and folic acid using ZrO2 nano-
particles-modified carbon paste electrode, Sens. Actuators B: Chemical 151
(2010) 243–249.
[18] X. Kang, J. Wang, H. Wua, J. Liua, I.A. Aksay, Y. Lin, A graphene-based elec-
trochemical sensor for sensitive detection of paracetamol, Talanta 81 (2010)
754–759.
[19] J.-Y. Sun, K.-J. Huang, S.-Y. Wei, Z.-W. Wu, F.-P. Ren, A graphene-based elec-
trochemical sensor for sensitive determination of caffeine, Coll. Surf. B:
Biointerfaces 84 (2011) 421–426.
[20] E.O. Faria, A.C.V. Lopes Jr., D.E.P. Souto, F.R.F. Leite, F.S. Damos, R.C.S. Luz, A.
S. dos Santos, D.L. Franco, W.T.P. dos Santos, Simultaneous determination of
caffeine and acetylsalicylic acid in pharmaceutical formulations using a boron-
doped diamond film electrode by differential pulse voltammetry, Electro-
analysis 24 (2012) 1141–1146.
[21] A. Kutluay, M. Aslanoglu, Modification of electrodes using conductive porous
layers to confer selectivity for the voltammetric detection of paracetamol in
the presence of ascorbic acid, dopamine and uric acid, Sens. Actuators B:
Chemical 185 (2013) 398–404.
[22] A.P.P. Eisele, D.N. Clausen, C.R.T. Tarley, L.H.D. Antonia, E.R. Sartori, Simulta-
neous square-wave voltammetric determination of paracetamol, caffeine and
orphenadrine in pharmaceutical formulations using a cathodically pretreated
boron-doped diamond electrode, Electroanalysis 25 (2013) 1734–1741.
[23] S.M. Patil, S.R. Sataraddi, A.M. Bagoji, R.M. Pathan, S.T. Nandibewoor, Elec-
trochemical behavior of graphene-based sensors on the redox mechanism of
aspirin, Electroanalysis 26 (2014) 831–839.
[24] A.B. Lima, L.M.F.C. Torres, C.F.R.C. Guimarães, R.M. Verly, L.M. da Silva, Á.D.
C. Júniorb, W.T.P. dos Santos, Simultaneous determination of paracetamol and
ibuprofen in pharmaceutical samples by differential pulse voltammetry using
a boron-doped diamond electrode, J. Braz. Chem. Soc. 25 (2014) 478–483.
[25] M. Purushotham, P. Gupta, R.N. Goyal, Graphene modified glassy carbon
sensor for the determination of aspirin metabolite in human biological sam-
ples, Talanta 143 (2015) 328–334.
[26] K. Tyszczuk-Rotko, I. Beczkowska, Nafion covered lead film electrode for the
voltammetric determination of caffeine in beverage samples and pharma-
ceutical formulations, Food Chem. 172 (2015) 24–29.
[27] D.M. Fernandes, N. Silva, C. Pereira, C. Mourac, J.M.C.S. Magalhães, B. Bachiller-
Baeza, I. Rodríguez-Ramos, A. Guerrero-Ruiz, C. Delerue-Matos, C. Freire,
MnFe2O4@CNT-N as novel electrochemical nanosensor for determination of
caffeine, acetaminophen and ascorbic acid, Sens. Actuators B: Chemical 218
29
(2015) 128–136.
[28] B.J. Sanghavi, A.K. Srivastava, Simultaneous voltammetric determination of
acetaminophen, aspirin and caffeine using an in situ surfactant-modified
multiwalled carbon nanotube paste electrode, Electrochim. Acta 55 (2010)
8638–8648.
[29] R.N. Goyal, A.R.S. Rana, H. Chasta, Simultaneous monitoring of aspirin, para-
cetamol and caffeine in human urine at poly-1,5-diaminonapthalene modified
pyrolytic graphite sensor, J. Electrochem. Soc. 160 (2013) G3014–G3019.
[30] A. Yiğit, Y. Yardım, Z. Şentürk, Voltammetric sensor based on boron-doped
diamond electrode for simultaneous determination of paracetamol, caffeine
and aspirin in pharmaceutical formulations, IEEE Sens. J. 16 (2016) 1674–1680.
[31] X. Huang, Z. Zeng, Z. Fan, J. Liu, H. Zhang, Graphene-based electrodes, Adv.
Mater. 24 (2012) 5979–6004.
[32] X. Kang, J. Wang, H. Wua, J. Liua, I.A. Aksay, Y. Lin, A graphene-based elec-
trochemical sensor for sensitive detection of paracetamol, Talanta 81 (2010)
754–759.
[33] Y. Li, K. Li, G. Song, J. Liu, K. Zhang, B. Ye, Electrochemical behavior of codeine
and its sensitive determination on graphene-based modified electrode, Sens.
Actuators B: Chemical 182 (2013) 401–407.
[34] F. Gao, X. Cai, X. Wang, C. Gao, S. Liu, F. Gao, Q. Wang, Highly sensitive and
selective detection of dopamine in the presence of ascorbic acid at graphene
oxide modified electrode, Sens. Actuators B: Chemical 186 (2013) 380–387.
[35] X.X. Jiao, H.Q. Luo, N.B. Li, Fabrication of graphene-gold nanocomposites by
electrochemical co-reduction and their electrocatalytic activity toward 4-ni-
trophenol oxidation, J. Electroanal. Chem. 691 (2013) 83–89.
[36] M. Purushotham, P. Gupta, R.N. Goyal, Graphene modified glassy carbon
sensor for the determination of aspirin metabolites in human biological
samples, Talanta 143 (2015) 328–334.
[37] Y. Wu, S. Zhai, K. Lu, L. Gao, Electrochemical oxidation behavior of 8-azagua-
nine at graphene-Nafion composite film-modified glassy carbon electrode, J.
Solid State Electrochem 18 (2014) 1593–1600.
[38] X. Lin, Y. Ni, S. Kokot, Voltammetric analysis with the use of a novel electro-
polymerised graphene-nafion film modified glassy carbon electrode: Si-
multaneous analysis of noxious nitroaniline isomers, J. Hazard. Mater. 243
(2012) 232–241.
[39] S. Wu, F. Huang, X. Lan, X. Wang, J. Wang, C. Meng, Electrochemically reduced
graphene oxide and Nafion nanocomposite for ultralow potential detection of
organophosphate pesticide, Sens. Actuators B: Chemical 177 (2013) 724–729.
[40] E. Er, H. Çelikkan, N. Erk, Highly sensitive and selective electrochemical sensor
based on high-quality graphene/nafion nanocomposite for voltammetric de-
termination of nebivolol, Sens. Actuators B: Chemical 224 (2016) 170–177.
[41] D.C. Marcano, D.V. Kosynkin, J.M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L.
B. Alemany, W. Lu, J.M. Tour, Improved synthesis of graphene oxide, ACS Nano
8 (2010) 4806–4814.
[42] S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia,
Y. Wu, S.T. Nguyen, R.S. Ruoff, Synthesis of graphene-based nanosheets via
chemical reduction of exfoliated graphite oxide, Carbon 45 (2007) 1558–1565.
[43] C.D. Zangmeister, Preparation and evaluation of graphite oxide reduced at
220 °C, Chem. Mater. 22 (2010) 5625–5629.
[44] A. Gasnier, M.L. Pedano, M.D. Rubianes, G.A. Rivas, Graphene paste electrode:
electrochemical behavior and analytical applications for the quantification of
NADH, Sens. Actuators B: Chemical 176 (2013) 921–926.
[45] D. Li, M.B. Mueller, S. Gilje, R.B. Kaner, G.G. Wallace, Processable aqueous
dispersions of graphene nanosheets, Nat. Nanotechnol. 3 (2008) 101–105.
[46] X. Gao, J. Jang, H. Nagase, Hydrazine and thermal reduction of graphene oxide:
Reaction mechanisms, product structures, and reaction design, J. Phys. Chem.
C 114 (2010) 832–842.
[47] J.C. Miller, J.N. Miller, Statistics for Analytical Chemistry, Van Nostrand Rein-
hold, 3rd ed., Ellis Horwood & Prentice Hall, New York 1993, pp. 53–80.