C H E M I C A L R O U T E S T O M A T E R I AL S
Effect of hexaammineruthenium chloride and/
or horseradish peroxidase on the performance
of hydrogen peroxide (bio)sensors: a comparative
study
Dilek So¨g˘u¨t O¨ zdemir1 , Ceren Kac¸ar2 , Berna Dalkıran2 , Semahat Ku¨c¸u¨kkolbas¸ı1,* , Pınar Esra Erden2,3 , and Esma Kılıc¸2
1
Department of Chemistry, Faculty of Science, Selçuk University, Konya, Turkey
2Department of Chemistry, Faculty of Science, Ankara University, Ankara, Turkey 3
Department of Chemistry, Polatlı Faculty of Science and Arts, Ankara Hacı Bayram Veli University, Ankara, Turkey
Received:11 July 2018 Accepted:9 December 2018 Published online:
19 December 2018
Ó
Springer Science+Business Media, LLC, part of Springer Nature 2018ABSTRACT
A comparison of the performances of three hydrogen peroxide (bio)sensors, based on the use of modified glassy carbon electrodes (GCE), is reported. GCE was modified with carboxylated carbon nanotubes (c-MWCNT), graphene (GR), titanium dioxide nanoparticles (TiO2) and hexaammineruthenium chloride (RUT) for the sensor design. In biosensor construction, coupling agents N-ethyl-N0-(3-dimethylaminopropyl) carbodiimide and N-hydroxyl succinimide were
used for the immobilization of horseradish peroxidase (HRP) onto TiO2 –c-MWCNT–GR–RUT and TiO2–c-MWCNT–GR modified GCEs. The modified electrodes were characterized by scanning electron microscopy, atomic force microscopy, cyclic voltammetry and electrochemical impedance spectroscopy methods. Electrode composition and critical working conditions such as pH and applied potential were optimized. TiO2–c-MWCNT–GR–RUT/GCE and HRP/ TiO2–c-MWCNT–GR–RUT/GCE exhibited better analytical performance than HRP/TiO2–c-MWCNT–GR/GCE in terms of detection limit and sensitivity. Moreover, TiO2–c-MWCNT–GR–RUT/GCE sensor showed a sensitivity for H2O2reduction 1.96 times higher than achieved with HRP/TiO2–c-MWCNT– GR/GCE configuration. The (bio)sensors were also applied to the determination of H2O2in a disinfector sample, and satisfied results were obtained.
Address correspondence toE-mail: ksemahat@gmail.com
https://doi.org/10.1007/s10853-018-03243-4
Introduction
Hydrogen peroxide (H2O2) is extensively used in various industrial processes such as paper bleaching, disinfection and sterilization, food processing and pharmaceutical applications [1]. It is one of the major members of reactive oxygen species and is a marker for oxidative stress which has been implicated in the progression of Alzheimer’s disease, Parkinson’s dis-ease and other neurodegenerative disorders [2]. Furthermore, it is a by-product of a large number of oxidases such as glucose oxidase, lactate oxidase, galactose oxidase and xanthine oxidase [3]. There-fore, reliable and rapid analysis of H2O2 is of para-mount importance in many fields including food industry, pharmaceutical analyses, clinical control and environmental monitoring [4]. A range of ana-lytical techniques such as chromatography [5], fluo-rescence [6], chemiluminescence [7], titrimetry [8], spectrophotometry [9] and electrochemistry [10] have been proposed for H2O2 detection. Among these techniques, electrochemical sensors and biosensors have been shown to have inherent benefits of high sensitivity, good selectivity, operational simplicity, suitability for real-time detection, ease-of-usage by non-specialist personnel, low cost and insensitivity to optical drawbacks, like turbidity and quenching [11,12].
Electrochemical H2O2sensors can be classified as enzymatic and nonenzymatic according to the way the electrode is modified [13]. The former ones have shown good performances for H2O2 detection with low detection limits, high sensitivity and good selectivity [14]. However, there are several draw-backs associated with enzyme-based sensors such as the high cost of enzymes and complexity of enzyme immobilization procedure. Moreover, the enzyme activity is highly dependent on the temperature and the pH of the sensing medium [15,16]. On the other hand, nonenzymatic sensors have the advantages of simplicity, fast response time, low cost and good operation stability [13,17]. One of the major problems in determination of H2O2is its high overpotential on the ordinary electrodes where interference effects originating from common electroactive species can-not be avoided [14]. Thus, the construction of H2O2 sensors operating at lower potentials has been intensively investigated [18–21].
Recent developments in nanomaterials have greatly enhanced the sensitivity and selectivity of
electrochemical (bio)sensors [22, 23]. Carbon nan-otubes (CNTs) have been extensively used in the fabrication of (bio)sensors owing to their large sur-face area, good mechanical strength, high electrical conductivity and chemical stability [24]. Metal oxide nanoparticles (MONPs) represent another critical group of nanomaterials. As electrode material, MONPs present large surface-to-volume ratio, high surface reaction activity and good catalytic efficiency [25]. Among different metal oxides, TiO2 nanostruc-tures have been widely used in (bio)sensor con-struction as they have advantages such as good biocompatibility, low cost and large numbers of active reaction sites for chemical reactions [26]. GR is a monolayer of sp2-bonded carbon atoms closely packed into a two-dimensional honeycomb arrange-ment. The rapid electron transfer ability, biocompat-ibility, good thermal conductivity and strong mechanical strength make the GR and its derivatives another promising material for sensing applications [27]. To date, various MONPs, carbon-based materi-als and novel nanostructures have been utilized to fabricate electrochemical H2O2 (bio)sensors [28–31]. Nevertheless, using only a single material makes it difficult to achieve eximious performance in (bio)sensors. Recently, various composites have become the focus of research in the field of modified electrodes to improve the analytical performance. For this purpose, nanomaterial-based composites such as CNTs-GR [32], CNTs-MONPs [33], GR-MONPs [34] and CNTs-GR-MONPs [35] have been investigated.
Electron transfer mediators such as Prussian Blue [36], ferrocene or its derivatives [37], methylene blue [38] and 1,4-benzoquinone [39] have been widely applied in amperometric biosensor development in order to obtain better selectivity and sensitivity. Incorporation of a suitable mediator in biosensing system can reduce the operating potential, resulting in a minimal risk of interference of coexisting elec-troactive species [40]. Hexaammineruthenium chlo-ride (RUT) has been used as mediator in the construction of biosensors [41]. It is easily available, soluble in water, sufficiently stable in both oxidized (Ru III) and reduced (Ru II) forms, positively charged in both forms and has a relatively low redox potential which is desired to minimize the effect of common interferences [42].
It is critical to eliminate the effect of interfering substances in H2O2 determination. Many recent
studies on electrochemical H2O2 sensing have focused on the fabrication of modified electrodes onto the surface of which reduction or oxidation of H2O2is achievable by operating at lower potentials, thus improving the sensor’s selectivity [19,43]. Tak-ing into consideration the drawbacks of HRP-based biosensors, fabrication of simple, selective and reli-able nonenzymatic H2O2 sensing platforms is also needed. The aim of this study is to develop H2O2 sensor and biosensors operating at low potentials using RUT and/or HRP in the TiO2–c-MWCNT–GR matrix and to compare the performance of these devices in terms of linear working range, sensitivity, detection limit and stability. Moreover, the effect of coexistence of RUT and HRP in the TiO2–c-MWCNT– GR matrix on the performance of the biosensor was investigated. The analytical applicability of the (bio)sensors for the determination of H2O2 in disin-fection solution was also addressed.
Materials and methods
Reagents
Peroxidase from horseradish (type II, 311 units/mg solid, EC 1.11.1.7), TiO2 nanoparticles (\ 100 nm particle size), K3Fe(CN)6, K4Fe(CN)6.3H2O, dopa-mine, sodium monohydrogen phosphate, sodium dihydrogen phosphate, ascorbic acid, oxalic acid, gelatin (type A, porcine skin), Nafion (5 wt% in lower aliphatic alcohols), uric acid, urea, 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide, hexaam-mineruthenium(III) chloride and N-hydroxy-succin-imide were purchased from Sigma-Aldrich (St. Louis, MO, USA). Graphene solution (in N,N-dimethylfor-mamide and deionized water 2 mg mL-1) was bought from DropSens (Llanera, Spain). H2O2 and glucose were obtained from Fluka (Buchs, Switzer-land). Carboxylated multiwall carbon nanotubes (length: 10–30 lm, O.D. \ 8 nm) were from Cheap Tubes Inc. (Brattleboro, USA). Double-distilled water supplied from Milli-Q System (Millipore, Bedford, MA, USA) was used throughout the experimental studies. The substrate solution was diluted freshly before every experiment.
Equipments and measurements
Electrochemical studies were conducted on a Com-pactSoft portable electrochemical analyzer (Ivium Technologies, Netherlands). Electrochemical cell composed of a modified GCE (3.0 mm diameter, BASi MF 2012) as working electrode, Ag/AgCl elec-trode (BASi MF 2052) as reference elecelec-trode and Pt wire (BASi MW 1032) as counter electrode was uti-lized for the electrochemical experiments. The amperometric measurements were carried out at -0.30 V vs. Ag/AgCl electrode in phosphate buffer solution (PBS) (0.05 M, pH 7.0). After the working electrode reached the steady state (background cur-rent), aliquots of H2O2 stock solution were succes-sively added to the stirred PBS and the steady-state current values recorded. In amperometric measure-ments, the variation (Di) between the steady-state current and background current was marked as the response current. EVO LS model scanning electron microscope (Carl Zeiss SMT Ltd. 511 Coldhams Lane, Cambridge, UK) was utilized to observe the mor-phology of different electrodes. A Bruker detector was used for the energy dispersive X-ray spec-troscopy (EDX). The morphological analysis of modified electrode surfaces was performed using an NT-MDT NTEGRA Solaris atomic force microscope. Electrochemical impedance spectroscopy (EIS) and cyclic voltammetry (CV) experiments were carried out in the presence of 5.0 mM K3[Fe(CN)6]/K 4-[Fe(CN)6] (1:1) and 0.10 M KCl to investigate the electrochemical characteristics of modified electrodes at different steps of electrode fabrication. Alternating voltage of 10 mV and frequency range of 105 Hz– 0.05 Hz were used in EIS experiments. All electro-chemical measurements were conducted at room temperature.
Biosensor preparation
Before every modification, GCE surface was polished first with 0.05 lm alumina slurry, then rinsed thor-oughly with distilled water and ultrasonically agi-tated successively in ethanol and distilled water for 5 min each. 50.0 mg of gelatin (GEL) was dissolved in 5.0 mL distilled water by magnetic stirring. This mixture was sonicated for 1 h in order to obtain a clear solution. Three different types of modified electrodes were prepared to construct H2O2 (bio)sensors.
1. TiO2–c-MWCNT–GR–RUT/GCE: RUT (6 mg), TiO2 (4 mg) and c-MWCNT (4 mg) were dis-persed into 1 mL of GEL solution, and this mixture was sonicated for 2 h. TiO2–RUT–GEL mixture was further mixed with GR solution (2 mg mL-1) in the ratio of 1:2 and ultrasonically stirred for 2 h. 9 lL of the resulting mixture was pipetted onto GCE surface and allowed to dry. Then, 5 lL of 0.25% Nafion solution was coated on the modified electrode.
2. HRP/TiO2–c-MWCNT–GR/GCE: TiO2 (4 mg) and c-MWCNT (4 mg) were dispersed into 1 mL of GEL solution, and this mixture was kept under mechanical stirring for 2 h. TiO2–c-MWCNT– GEL mixture was further mixed with GR solution in the ratio of 1:2 and ultrasonically stirred for 2 h. 9 lL of the resulting mixture was casted onto GCE surface and allowed to dry. 6 lL of N-ethyl-N0-(3-dimethyaminopropyl) carbodiimide (EDC)
(50 mM)-N-hydroxyl succinimide (NHS) (200 mM) mixture was dropped onto the TiO2– c-MWCNT–GR/GCE surface to activate carboxyl groups of the carbon nanotubes into active carbodiimide esters. 8 lL of HRP solution (3250 U mL-1) was drop coated onto the modified sensing interface (TiO2–c-MWCNT–GR/GCE) and allowed to dry at ? 4 °C to construct the biosensor. Finally, 5 lL of 0.25% Nafion solution was coated on the modified electrode and the resulting biosensor was stored at ? 4 °C when not in use.
3. HRP/TiO2–c-MWCNT–GR - RUT/GCE: HRP solution was immobilized onto the modified electrode prepared in step 1 (TiO2–c-MWCNT– GR–RUT/GCE) according to the procedure explained in step 2 to construct the HRP/TiO2– c-MWCNT–GR-RUT/GCE. Scheme1 illustrates the fabrication steps of the (bio)sensors.
Results and discussion
Optimization of electrode surface
composition
Effects of c-MWCNT, GR, TiO2, RUT amounts and HRP loading on the current response of (bio)sensors were studied to optimize the surface composition of GCE. The electrode response was measured by CV in
0.10 M KCl solution containing 5.0 mM Fe(CN)6 3-/4-to optimize the c-MWCNT, GR, TiO2 amounts. Optimization studies of RUT and HRP amounts were carried out in 0.05 M PBS (pH 7.0) containing 0.01 mM H2O2, and response currents of the modi-fied electrodes were recorded at -0.30 V vs. Ag/ AgCl electrode.
The effect of GR amount on the electrode response was tested by changing its amount on the modified electrode as 6.0, 9.0, 12.0 and 15.0 lg. Optimum elec-trode response was obtained with 12.0 lg GR, and this amount was used for the fabrication of the (bio)sen-sors. c-MWCNT amounts ranging from 6.0 to 15.0 lg in 3.0 lg increments were investigated to select its optimum amount on the GCE surface, while GR amount kept constant. The highest peak current was obtained with the electrode containing 12.0 lg c-MWCNT, and 15 lg c-MWCNT did not increase the electrode response. Thus, optimum c-MWCNT amount was chosen as 12.0 lg for the fabrication of all (bio)sensors. A series of modified electrodes was pre-pared with four different TiO2amounts (6.0 lg, 9.0 lg, 12.0 lg and 15.0 lg), while the amounts of GR and c-MWCNT kept constant. Similar to the results obtained for GR and c-MWCNT amounts, the highest electrode response was obtained with 12.0 lg TiO2 and this amount was used for all further modifications. The decrease in response current beyond 12 lg GR, c-MWCNT or TiO2may be attributed to the limited diffusion rate of the substrate in the thicker films.
The amount of RUT in the TiO2–c-MWCNT–GR– RUT composite was varied between 6.0 and 24.0 lg, and response currents of TiO2–c-MWCNT–GR–RUT/ GCE were recorded. The response of the electrode increased with the RUT amount up to 18 lg and then decreased with increasing RUT amount (Fig.1a). The modified electrode prepared with 18 lg RUT showed the highest response, and this value was selected as the optimum RUT amount.
The effect of HRP loading (13; 20; 26 and 33 U) onto TiO2–c-MWCNT–GR/GCE on the response current of H2O2biosensor was also investigated (Fig.1b). The biosensor response increased with HRP loading, reaching a maximum for 26 U, following which the response decreased remarkably. Therefore, 26 U of HRP was used for all further modifications. The decrease in response current for loadings higher than 26 U HRP is probably due to the increased hindrance in electron transfer [44]. Meanwhile, the low responses below 26 U HRP loading can be attributed
to the amount of the HRP on the biosensor was not enough to convert all the H2O2[45].
Surface characterization of modified
electrodes
Scanning electron microscopy (SEM), atomic force microscopy (AFM) and EDX analysis were performed to study the surface morphology and structure of the TiO2–c-MWCNT–GR–RUT/GCE. The SEM images of GCEs modified with c-MWCNT–GR, TiO2 –c-MWCNT–GR and TiO2–c-MWCNT–GR–RUT dis-persed in GEL are illustrated at Fig.2together with the EDX spectrum and elemental analysis data tables. It can be clearly seen that the modification materials distributed widely in the GEL matrix. SEM images of TiO2–c-MWCNT–GR/GCE and TiO2–c-MWCNT–
GR-RUT/GCE (Fig.2A-b and c) exhibit a porous structure which provides a significant increase in the effective surface for HRP loading. Elemental compo-sition of the TiO2–c-MWCNT–GR and TiO2 –c-MWCNT–GR-RUT composites examined by EDX analysis confirmed the presence of TiO2 (Fig.2B-b) and RUT (Fig.2B-c) on the surface of the electrodes.
The AFM images of three modified electrode sur-faces are illustrated in Fig.3as 2D and 3D plots. The coexistence of TiO2and RUT with c-MWCNT and GR changed the morphology of the surface (Fig. 3c), and a relatively lower surface roughness was observed.
Electrochemical characterizations
Electrochemical characterizations of bare GCE, GR/ GCE, c-MWCNT–GR/GCE, TiO2–c-MWCNT–GR/
EDC-NHS + HRP GCE &2 1 + &2 1+ 2& 1+ & 21+ &2 1 + &2 1 + 2& 1+ & 21+ HRP/TiO 2‒c-MWCNT‒GR−RUT/GCE &2 1 + &2 1 + & 2&1+ 21+ &2 1+ &2 1+ 2& 1+ & 2 1+ EDC-NHS + HRP Nafion TiO2‒c-MWCNT‒GR−RUT/GCE RUT-GR-TiO2 c-MWCNT HRP/TiO 2‒c-MWCNT‒GR/GCE Nafion GR-TiO2 c-MWCNT
Scheme 1 The stepwise fabrication of the (bio)sensors.
0.0 0.3 0.6 0.9 0 6 12 18 24 30 ΔI, μA RUT amount, μg (a) 0.05 0.15 0.25 10 20 30 40 Δ I, μ A HRP, Unit (b) Figure 1 Effect ofa RUT and
b HRP amounts on TiO2
–c-MWCNT–GR/GCE (in 0.05 M pH 7.0 PBS containing 0.01 mM H2O2,
(A)
(B)
Element Series unn. C [wt.%] norm.C [wt.%] Atom.C [wt.%] Error [%] C K-series 67.76 67.76 73.68 21.1 O K-series 32.24 32.24 26.32 11.0
Element Series unn. C [wt.%] norm.C [wt.%] Atom.C [wt.%] Error [%] C K-series 56.31 56.31 64.72 17.6 O K-series 39.47 39.74 34.06 54.1 Ti K-series 4.22 4.22 1.22 0.1
Element Series unn. C [wt.%] norm.C [wt.%] Atom.C [wt.%] Error [%] C K-series 29.53 41.44 62.63 3.4 O K-series 18.74 26.30 29.84 21.8 Ti K-series 6.21 8.72 3.31 0.2 Ru L-series 16.78 23.54 4.23
GCE and TiO2–c-MWCNT–GR–RUT/GCE were car-ried out by CV in the potential range of (- 0.50) to (? 0.80) V at 50 mV s-1 scan rate, as shown in Fig.4A. The response of the ferri/ferrocyanide redox couple at the TiO2–c-MWCNT–GR–RUT/GCE (curve e) is much higher than that at bare GCE, GR/GCE, c-MWCNT–GR/GCE and TiO2–c-MWCNT–GR/ GCE (curves a, b, c and d, respectively). Results show that accelerating electron transfer on the electrode surface has been occurred in following sequence: bare GCE, GR/GCE, c-MWCNT–GR/GCE, TiO2 –c-MWCNT–GR/GCE and TiO2–c-MWCNT–GR–RUT/ GCE. The DEp values obtained with four electrodes (curves b–e) were found to be lower than the DEp obtained with bare GCE (curve a) which confirms that electron transfer at these electrodes is faster than the bare GCE [46]. Figure4B shows the cyclic voltammogram of TiO2–c-MWCNT–GR–RUT/GCE recorded at different scan rates ranging from 5 to 1000 mVs-1in 0.05 M PBS solution containing 0.10 M KCl. The voltammogram recorded at 50 mV s-1 exhibits an anodic peak (Epa =-0.15 V) at forward scan of potential and a cathodic peak (Epc =-0.23 V) in backward scan of the potential corresponding to the redox behavior of RUT. Quite similar voltam-mograms for RUT were previously reported at dif-ferent electrode surfaces [47, 48]. Moreover, these voltammograms show that the mediator has been immobilized on the modified electrode surface [49]. The redox peak currents were proportional to the scan rate in the range less than 200 mV/s (inset of Fig.4B), indicating a surface-controlled process [50]. When the scan rate was larger than 200 mV/s, peak current was found to be linearly proportional to the square root of scan rates (inset of Fig.4B), which indicated that the electrochemical kinetics of the modified electrode was mainly controlled by diffu-sion [51].
The electron transfer rate constants (ks) for HRP/ TiO2–c-MWCNT–GR/GCE, TiO2–c-MWCNT–GR– RUT/GCE and HRP/TiO2–c-MWCNT–GR–RUT/ GCE were calculated using Laviron equation [52].
Log ks ¼ a Log 1 að Þ þ 1 að Þ Log a
Log RT=nFmð Þ a 1 að ÞnFDEp=2:303RT
where R is the gas constant (8.314 J mol-1K-1), T is the room temperature (298.15 K), n is the number of electrons transferred, and DEp is the peak separation. Taking a charge transfer coefficient a of 0.5 with a scan rate of 100 mV s-1, the electron transfer rate constant of HRP at the HRP/TiO2–c-MWCNT–GR/ GCE was found to be 0.82 s-1, which is larger than that at the solid graphite electrode (0.66 s-1) [53]. This indicates that the TiO2–c-MWCNT–GR–RUT modified electrode facilitates the electron transfer between the redox-active sites of enzyme and the electrode. The value of ks was calculated as 0.69 s-1 and 0.65 s-1for TiO2–c-MWCNT–GR–RUT/GCE and HRP/TiO2–c-MWCNT–GR–RUT/GCE, respectively. The similar rate constants obtained at TiO2 –c-MWCNT–GR–RUT/GCE and HRP/TiO2 –c-MWCNT–GR/GCE show that RUT can be good alternative to HRP for the development of a nonen-zymatic H2O2sensor.
The cyclic voltammograms of TiO2–c-MWCNT– GR–RUT/GCE during continuous potential cycling between (- 0.50) and (? 0.80 V) in 0.05 M PBS at 50 mV s-1 (data not shown) were stable, indicating that the GEL matrix and Nafion layer could prevent RUT from leakage. In the cyclic voltammogram of TiO2–c-MWCNT–GR–RUT/GCE recorded in ferri/ferrocyanide redox probe (Fig. 4A-curve e), the redox peaks of RUT were also observed in addition to redox peaks of Fe(CN)63-/4-.
Voltammetric response of H2O2 at (a) bare GCE, (b) GR/GCE, (c) c-MWCNT–GR/GCE, (d) TiO2 –c-MWCNT–GR/GCE and (e) TiO2–c-MWCNT–GR– RUT/GCE was investigated by means of CV. The cyclic voltammograms of the above mentioned elec-trodes were recorded from - 0.5 to 0.8 V (vs. Ag/ AgCl) at a scan rate of 50 mV/s in 0.05 M PBS con-taining 0.1 M KCl in the absence and presence of 2.0 mM H2O2. Figure 4C, given as an example, shows the cyclic voltammogram of TiO2–c-MWCNT–GR/ GCE. From curve b in Fig.4C, we could see that TiO2–c-MWCNT–GR modified electrode exhibited good electrocatalytic activity toward the reduction of H2O2. CV curves of c-MWCNT–GR/GCE and TiO2– c-MWCNT–GR–RUT/GCE also exhibited a high reduction current at a relatively low potential, and the response currents of c-MWCNT–GR/GCE and
bFigure 2 A SEM images (10 lm; EHT = 20.00 kV; Mag = 1.00
KX; I Probe = 50 pA) and B EDX spectrum and elemental analysis data tables of (a) c-MWCNT–GR/GCE, (b) TiO2
–c-MWCNT–GR/GCE and (c) TiO2–c-MWCNT–GR–RUT/GCE
Figure 3 2D and 3D AFM images ofa c-MWCNT–GR/GCE, b TiO2–c-MWCNT–GR/GCE andc TiO2–c-MWCNT–GR–RUT/GCE
TiO2–c-MWCNT–GR–RUT/GCE were higher than that of bare GCE and GR/GCE indicating the elec-trocatalytic activity of these electrodes toward the H2O2reduction (data not shown).
EIS is widely used to obtain detailed information on changes of the surface electron transfer behavior of surface modified electrodes. Figure5 shows the impedance spectra of different modified electrodes presented in the form of Nyquist plots. In Nyquist plot, the semicircle portion observed at higher fre-quencies corresponds to the electron transfer-limited process and the linear part observed at lower fre-quency range represents the diffusion-limited pro-cess. The diameter of the semicircle equals to the
electron transfer resistance (Rct) [54]. Figure5 indi-cates that TiO2–c-MWCNT–GR–RUT/GCE and TiO2–c-MWCNT–GR/GCE showed lower electron transfer resistance compared to RUT/GCE and TiO2/ GCE. This result can be attributed to the existence of GR and c-MWCNT in the composites. However, the electron transfer resistances of TiO2–c-MWCNT–GR– RUT/GCE and TiO2–c-MWCNT–GR/GCE were very close to the electron transfer resistances of GR/GCE and c-MWCNT/GCE which suggests that the exis-tence of TiO2and/or RUT in the composite did not decrease the conductivity of the modified electrodes.
-100 -40 20 80 -0.6 -0.2 0.2 0.6 1 I, μA Potential, V (vs. AgCl) (a) (b) (c) (d) (e) (A) -8 -3 2 -0.8 -0.4 0 0.4 I, μA Potential, V (vs. AgCl) -0.08 -0.04 0 0.04 0.08 0.12 -0.6 -0.3 0 0.3 0.6 0.9 I, μA Potential, V (vs. AgCl) y = 0.18x - 0.44 R² = 0.999 y = -0.18x + 0.38 R² = 0.9975 -8 0 8 16 24 32 I, μA Scan rate1/2(mV s-1)1/2 y = 0.009x + 0.456 R² = 0.9802 y = -0.007x - 0.854 R² = 0.9853 -4 0 4 0 50 100 150 200 I, μA Scan rate (mV s-1) (B) (C) (a) (b)
Figure 4 A Cyclic voltammograms of (a) bare GCE, (b) GR/ GCE, (c) c-MWCNT–GR/GCE, (d) TiO2–c-MWCNT–GR/GCE
and (e) TiO2–c-MWCNT–GR–RUT/GCE in 0.10 M KCl solution
containing 5.0 mM [Fe(CN)6]3-/4-at a scan rate of 50 mV s-1.
B Cyclic voltammograms of TiO2–c-MWCNT–GR–RUT/GCE in
0.05 M PBS containing 0.10 M KCl recorded at 5, 10, 25, 50, 75,
100, 150, 200, 250, 300, 400, 500, 750 and 1000 mV s-1(from internal to external). (Inset: plots of peaks currents vs. square root of scan rate and plots of peaks currents vs. scan rate.). C CV curves in the (a) absence and (b) presence of 2.0 mM H2O2for the
Optimization of experimental variables
It is widely known that applied potential is one of the critical parameters affecting the selectivity and sen-sitivity of (bio)sensors. Thus, the influence of the applied potential on the current response of the HRP/TiO2–c-MWCNT–GR–RUT/GCE was investi-gated. The reduction of H2O2on this electrode starts at - 0.20 V (data not shown). The optimization study was performed over the (- 0.10) to (- 0.40) V range due to the oxidation and reduction peaks of RUT (Fig.4B) and the working range of the electrode (data not shown). The most stable and reproducible H2O2 response was obtained at -0.30 V, and this potential was selected as the working potential. The effect of applied potential on electrode response was also studied for TiO2–c-MWCNT–GR–RUT/GCE, and optimum potential was found to be -0.30 V. This low working potential is very important to reduce the effects of interfering species commonly coexist in real samples and to fabricate a reliable sensor for H2O2 determination.
pH of the working buffer solution has a strong influence on enzyme activity and its stability [55]. The effect of pH on the H2O2response of the HRP/ TiO2–c-MWCNT–GR/GCE was studied by measur-ing the current response of the electrode to 0.01 mM H2O2in the pH range of 6.0–8.5 (Fig.6a). The highest amperometric response was obtained at pH 7.0, which is in good agreement with what is reported for the soluble HRP [56]. So, pH 7.0 was chosen as the optimized condition for the HRP/TiO2–c-MWCNT– GR/GCE. When the pH value is higher than 7.0, a
decrease in amperometric response was observed, which may be due to the denaturing of the lized enzyme. It can be concluded that the immobi-lization technique has no significant influence on the optimum pH of HRP. The operation mechanism of the H2O2 sensor proposed in performance charac-teristics section shows that the hydronium ions take a part in the reaction between RUT and H2O2. There-fore, the effect of pH on the response of this sensor was investigated. Figure6b shows that the highest electrode response was recorded at pH 7.0. The decrease in response current observed below and above pH 7.0 may be attributed to the change in the structure of RUT complex. At low pH values NH3 molecules in the coordination complex may be pro-tonated, and at high pH values OH-ions may behave as ligand instead of NH3molecules.
These optimized electrode composition and experimental conditions were used in all experiments conducted with sensors and biosensors presented in this study.
Performance parameters
Amperometric responses of TiO2–c-MWCNT–GR– RUT/GCE and HRP/TiO2–c-MWCNT–GR/GCE after addition of H2O2 aliquots were recorded and compared. Figure7 depicts the i - t curves (A) and calibration plots (B) of these modified electrodes obtained under optimal experimental conditions. TiO2–c-MWCNT–GR–RUT/GCE response to H2O2 (Fig.7B-b) was linear between 2.0 9 10-6 and 1.2 9 10-4 M (R2= 0.9998) which is wider than the sensors based on silver nanoparticles/oxidized poly-2,20:50,200-terthiophene-3-p-benzoic acid/MWCNT
modified GCE (1.0 9 10-5–2.6 9 10-4 M) [57] and HRP/MWCNTs/chitosan modified GCE (1.67 9 10-5–7.4 9 10-4 M) [29]. A low detection limit of 4.7 9 10-7 M was obtained at a signal-to-noise ratio of 3 and the sensitivity of the sensor was calculated to be 9.74 lA mM-1. The detection limit of TiO2–c-MWCNT–GR–RUT/GCE is lower than 1.0 9 10-6 M of hematite nanoparticles/reduced graphene oxide modified GCE [58], 1.7 9 10-6 M of silver nanoparticles/manganese dioxide/MWCNT modified GCE [21], 1.03 9 10-5 M of HRP/ MWCNTs/chitosan modified GCE [29] and 1.2 9 10-5 M of Fe3O4–Au magnetic nanoparticles coated HRP and GR sheets–Nafion film modified screen-printed carbon electrode [59]. The linear
0 1000 2000 3000 4000 5000 6000 0 2000 4000 6000 -Z'', o hm Z', ohm (a) ■ (b) ● (c) ▲ (d) ▬ (e) * (f) ◊
Figure 5 The Nyquist curves of (a) RUT/GCE, (b) TiO2/GCE,
(c) GR/GCE, (d) c-MWCNT/GCE, (e) TiO2–c-MWCNT–GR/
GCE and (f) TiO2–c-MWCNT–GR–RUT/GCE (in 0.10 M KCl
working range of the HRP/TiO2–c-MWCNT–GR/ GCE was 6.0 9 10-6–2.0 9 10-4 M (R2= 0.9976) (Fig.7B–c) with a detection limit of 1.3 9 10-6M. The sensitivity of this biosensor was calculated to be 4.98 lA mM-1which was 1.96-fold lower than that of TiO2–c-MWCNT–GR–RUT/GCE. These results indi-cate that using TiO2–c-MWCNT–GR–RUT composite increases sensitivity when compared to electrode which is only modified with HRP/TiO2–c-MWCNT– GR. It should be mentioned that the higher sensitivity obtained with the TiO2–c-MWCNT–GR–RUT/GCE was due to the presence of RUT in the modification matrix, which result in improved electrocatalytic activity for the reduction of H2O2. It can be concluded that RUT can catalyze the reduction of H2O2 with higher efficiency than HRP. Furthermore, this result indicates that TiO2–c-MWCNT–GR–RUT composite can be used toward H2O2determination in oxidase-based biosensors without the incorporation of HRP.
To determine whether the use of RUT and HRP together in the modification composite improve the sensitivity, we measured the H2O2response of HRP/ TiO2–c-MWCNT–GR–RUT/GCE under optimal
experimental conditions. Compared with the sensi-tivity achieved with the TiO2–c-MWCNT–GR–RUT/ GCE (9.74 lA mM-1), a slightly higher sensitivity (10.27 lA mM-1) was obtained using the HRP/TiO2– c-MWCNT–GR–RUT/GCE (Fig. 7B-a). However, the linear working range of this biosensor (2.0 9 10-6– 1.2 9 10-4 M) was similar to those values obtained with TiO2–c-MWCNT–GR–RUT/GCE and HRP/ TiO2–c-MWCNT–GR/GCE. The detection limit of HRP/TiO2–c-MWCNT–GR–RUT/GCE was 4.4 9 10-7 M. In comparison with HRP/TiO2 –c-MWCNT–GR/GCE configuration, HRP/TiO2 –c-MWCNT–GR–RUT/GCE biosensor has a sensitivity 2.06 times higher for H2O2. These results also pro-moted the efficiency of RUT toward H2O2 determination.
To clarify the role of RUT and HRP as modification materials in the improvement of the sensitivity toward H2O2, we have also constructed a TiO2 –GR–c-MWCNT modified GCE. The electrochemical response of TiO2–GR–c-MWCNT to H2O2 (data not shown) was very low compared to those of presented (bio)sensors in this study. Table1 shows the 0.03 0.06 0.09 5.0 6.0 7.0 8.0 9.0 ΔI, μA pH (a) 0 0.008 0.016 5.0 6.0 7.0 8.0 9.0 ΔI, μA pH (b) Figure 6 Effect of pH on the
response ofa HRP/TiO2 –c-MWCNT–GR–RUT/GCE and b TiO2–c-MWCNT–GR– RUT/GCE (in 0.05 M PBS containing 0.01 mM H2O2, Eapp= - 0.30 V). -6.5 -2.5 1.5 0 300 600 900 1200 I, μA Time,s 0 0.8 1.6 0 0.1 0.2 ΔI, μA H2O2, mM (a) (b) (c) (a) (b) (c) (A) (B)
Figure 7 Ai-t curves and B calibration graphs of (a) HRP/TiO2–c-MWCNT– GR–RUT/GCE (b) TiO2 –c-MWCNT–GR–RUT/GCE and (c) HRP/TiO2–c-MWCNT– GR/GCE on successive additions of H2O2into a stirred
0.05 M PBS (pH 7.0), Eapp= - 0.30 V.
analytical performance characteristics of the (bio)sensors developed in this study.
The apparent Michaelis–Menten constant (KMapp), a reflection of the enzymatic affinity, was calculated to be 0.03 mM for HRP/TiO2–c-MWCNT–GR–RUT/ GCE and 0.12 mM for HRP/TiO2–c-MWCNT–GR/ GCE from the Lineweaver–Burk equation [60]. According to the characteristics of KMapp, the lesser the value of KMapp, the stronger will be the affinity between HRP and substrate [61]. The KMapp value of HRP/TiO2–c-MWCNT–GR–RUT/GCE biosensor (0.03 mM) is lower than that of HRP/TiO2 –c-MWCNT–GR/GCE-based biosensor (0.12 mM) sug-gesting a higher affinity toward H2O2 and a good bioactivity. These KMapp values are lower than those previously reported [62–64]. The catalytic effects of RUT and HRP on the reduction of H2O2appear to be beneficial to improve the biosensor’s performance.
According to the results discussed previously, the operation mechanism of TiO2–c-MWCNT–GR–RUT/ GCE can be expressed by the following equations [65].
RUT (II) catalyzes the reduction of H2O2to water, and RUT(III) is produced. Then, RUT(III) is reduced to RUT(II) on the surface of the electrode at 2 0.30 V. 2Ru NHð 3Þ2þ6 þH2O2þ 2Hþ! 2Ru NHð 3Þ3þ6 þ2H2O
Ru NHð 3Þ3þ6 þe
! Ru NH 3
ð Þ2þ6 ðElectrode; at 0:30 VÞ
On the other hand, the following equations can be written for the response mechanism of the HRP/ TiO2–c-MWCNT–GR–RUT/GCE. In this mechanism,
RUT(red) gives electrons to the oxidized HRP and
RUT(ox) is produced. RUT(ox) is reduced on the sur-face of the electrode at the applied potential of -0.30 V. The final response current is directly pro-portional to H2O2concentration. H2O2þ 2Hþþ HRPðredÞ ! 2H2O þ HRPðoxÞ Ru NHð 3Þ2þ6 þHRPðoxÞ! Ru NHð 3Þ3þ6 þHRPðredÞ Ru NHð 3Þ3þ6 þe ! Ru NH 3 ð Þ2þ6 ðElectrode; at 0:30 VÞ Response time, repeatability, reproducibility, selec-tivity and operational stability are the other param-eters required for full (bio)sensor characterization. The response time of the (bio)sensors was quite short, reaching 95% of their maximum response less than 5 s. In order to investigate the repeatability of the (bio)sensors, five calibration curves were plotted by the use of the same modified electrode. The slopes of these calibrations were used to evaluate the repeata-bility in terms of relative standard deviation (RSD), yielding values of 3.3%, 3.0% and 4.1% for TiO2 –c-MWCNT–GR–RUT/GCE, HRP/TiO2–c-MWCNT– GR/GCE and HRP/TiO2–c-MWCNT–GR–RUT/ GCE, respectively. Five modified electrodes were prepared independently using the same construction technique to study the reproducibility of the pur-posed (bio)sensors. To check the response current of 0.01 mM H2O2, the results revealed a RSD of 2.1%, 3.0% and 4.0% for TiO2–c-MWCNT–GR–RUT/GCE, HRP/TiO2–c-MWCNT–GR/GCE and HRP/TiO2 –c-MWCNT–GR–RUT/GCE, respectively. These RSD values demonstrated the good repeatability and reproducibility of the (bio)sensors. Operational sta-bility is an important parameter of biosensors per-formance testing. Therefore, this parameter was investigated by successive measurements of the modified electrode response to 0.01 mM H2O2. All the purposed (bio)sensors exhibited excellent opera-tional stability, and no significant decrease was observed after 100 measurements. This excellent sta-bility observed for the presented biosensors indicates
Table 1 The analytical performance characteristics of the presented (bio)sensors obtained at- 0.30 V and pH 7.0 Performance parameters TiO2–c-MWCNT–GR–
RUT/GCE HRP/TiO2–c-MWCNT– GR/GCE HRP/TiO2–c-MWCNT–GR– RUT/GCE Linear range, M 2.0 9 10-6–1.2 9 10-4 6.0 9 10-6–2.0 9 10-4 2.0 9 10-6–1.2 9 10-4
Linear equation y(lA) = 9.74 ? 0.1x (mM) y(lA) = 4.98 ? 0.1x (mM) y(lA) = 10.27 ? 0.2x (mM)
Regression coefficient,R2 0.9998 0.9976 0.9939
Sensitivity, lAmM-1 9.74 4.98 10.27
LOD, M 4.7 9 10-7 1.3 9 10-6 4.4 9 10-7
T able 2 Characteristics of various nanomaterial-based amperometric H2 O2 (bio)sensors reported in recent years Modified electrode W orking potential (V) Sensitivity Detection limit Linear range Operational stability Repeatability/ reproducibility (RSD) Response time (s) References Ag–MnO 2 – MWCNT s/GCE -0.30 82.5 l AmM -1 cm -2 1.7 l M 0.005–10.4 mM No obvious decrease after 50 cycle -/\ 4% 2 [ 21 ] HRP–MWCNT s/ Chit/GCE -0.20 4.995 l Am M -1 1.03 9 10 -5 M 1.67 9 10 -5 – 7.40 9 10 -4 M – 3.3%/ -–[ 29 ] Co 3 O4 /MWCNT s/ HRP/GCE -0.30 15.245 l Am M -1 0.74 l M 7.4 9 10 -7 – 1.9 9 10 -5 M – 3.4%/– 5 [ 33 ] PtAu/G-CNT s/GCE -0.47 313.4 l AmM -1 cm -2 0.6 l M 2.0–8561 l M – –/4.37% \ 4[ 20 ] Nafion/Gr -CSS-AgNPs/GCE -0.20 0.0149 mA mM -1 2.49 l M 2.0 9 10 -5 – 5.02 9 10 -3 M – - /-1[ 18 ] 0.0176 mA mM -1 9.51 l M 5.02 9 10 -3 – 3.41 9 10 -2 M AgNPs/Ox-pTTBA/ MWCNT/GCE -0.60 2.42 l A l M -1 0.24 l M 10–260 l M 2.5% loss after 20 measurements -/3.8% \ 5[ 57 ] SPCE|GS–Nafion/ Fe 3 O4 –Au-HRP -0.30 4.8017 9 10 -6 l AmM -1 1.2 9 10 -5 M 2.0 9 10 -5 – 2.5 9 10 -3 M – 4.4/3.6% 3 [ 59 ] a -Fe 2 O3 /rGO/GCE -0.22 126.9 l Ac m -2 mM -1 1.0 l M 5.0–4495.0 l M – 3.2/2.6% 2 [ 58 ]
MWCNT/Ag nanohybrids/gold electrode
-0.20 1.42 l Am M -1 5 9 10 -7 M 0.05–17 mM – -/3.3% 5 [ 66 ] Ag@T iO 2 /GCE -0.73 65.2328 ± 0.01 l A l M -1 cm -2 0.83 l M 0.83 l M– 30.0 mM More than 200 cycles Excellent reproducibility –[ 67 ] Ti O2 –c-MWCNT– GR–RUT/GCE -0.30 9.74 l Am M -1 (137.18 l Ac m -2 mM -1 ) 4.7 9 10 -7 M 2.0 9 10 -6 – 1.2 9 10 -4 M No obvious decrease after 100 measurements 3.3/2.1% \ 5 This work HRP/T iO 2 –c-MWCNT–GR– RUT/GCE -0.30 10.27 l Am M -1 (144.65 l Ac m -2 mM -1 ) 4.4 9 10 -7 M 2.0 9 10 -6 – 1.2 9 10 -4 M No obvious decrease after 100 measurements 4.1/4.0% \ 5 This work Ag Silver nanoparticles, MnO 2 manganese dioxide, Co 3 O4 cobalt oxide nanoparticles, PtAu PtAu bimetallic nanoparticles, G-CNTs graphene sheets-multi walled carbon nanotubes, AgNPs silver nanoparticles, Gr-CSS graphene-colloidal carbon sphere, Ox-pTTBA/MWCNT oxidized poly-2,2 0:5 0,2 00-terthiophene-3-p-benzoic acid/multi wall carbon nanotube, SPCE screen-printed carbon electrode, GS graphene sheets, a-Fe 2 O3 Hematite nanoparticles, rGO reduced graphene oxide, Ag nanoparticle-decorated T iO 2 nanowires
the suitability of the TiO2–c-MWCNT–GR–RUT and TiO2–c-MWCNT–GR matrix to conserve the activity of HRP for long-term operation. Moreover, the Nafion layer used to modify the (bio)sensors con-tributed to the stability of the electrodes by prevent-ing the HRP and/or RUT from leakprevent-ing out of the membrane into the test solution.
In order to determine the selectivity of the (bio)sensors the effect of six substances (urea, glu-cose, oxalic acid, ascorbic acid, uric acid and dopa-mine) on the amperometric response current of the (bio)sensors was examined chronoamperometrically by successive addition of 0.10 mM H2O2 and 0.10 mM interfering species in a stirred PBS at pH 7.0. The results indicated that urea, glucose, oxalic acid, ascorbic acid, uric acid had no effect on the H2O2 response of TiO2–c-MWCNT–GR–RUT/GCE, HRP/ TiO2–c-MWCNT–GR/GCE and HRP/TiO2 –c-MWCNT–GR–RUT/GCE probably due to the low applied potential of -0.30 V and the barrier effect of Nafion layer. Dopamine interfered significantly and increased the responses of HRP/TiO2–c-MWCNT– GR/GCE and HRP/TiO2–c-MWCNT–GR–RUT/GCE by about 30% and 10%, respectively. In contrast, dopamine did not interfere in H2O2 detection with TiO2–c-MWCNT–GR–RUT/GCE sensor. It should be mentioned that the interference effect of dopamine decreased dramatically in the presence of RUT.
Table2 presents the analytical performances of published data about H2O2 (bio)sensors based on nanomaterials. From these data, we can see that the reported TiO2–c-MWCNT–GR–RUT/GCE and HRP/ TiO2–c-MWCNT–GR–RUT/GCE have similar or better performances in response time, linear working range, sensitivity and operational stability compared to most of those recently reported in the literature.
Real-sample measurements
The analytical reliability and application potential of the presented (bio)sensors were evaluated for the determination of the concentration of H2O2in com-mercial disinfector solution. The determination of H2O2content in this solution was performed by using standard addition method. In this technique, addi-tions of standard H2O2solution were made to disin-fector solution, and a multiple addition calibration curve was obtained. It was shown that the calibration curves are linear and H2O2 concentrations in the sample obtained with the presented (bio)sensors are given in Table3. It can be seen from this table that the mean recoveries are almost 100%. It can be concluded that there is no difference between the results obtained with presented (bio)sensors at a confidence level of 95%. The high analytical recovery shows the reliability of the method. The satisfying results indi-cated that the (bio)sensors have a great potential for practical application in real samples.
Conclusion
Amperometric H2O2sensor and biosensors have been successfully fabricated using TiO2–c-MWCNT–GR– RUT, HRP/TiO2–c-MWCNT–GR and HRP/TiO2 –c-MWCNT–GR–RUT modified GCEs. From the com-parison of the analytical performance of the three (bio)sensors, it can be concluded that all of them can be efficiently used for the monitoring of the H2O2. These (bio)sensors have lower detection limit [21,29,58,59], higher reproducibility [20] and show better operational stability [57] when compared with other H2O2 (bio)sensors previously reported in the literature. Among the three (bio)sensors proposed,
Table 3 Determination of H2O2content in disinfector sample using the presented (bio)sensors
H2O2amount (g/100 mL) HRP/TiO2–c-MWCNT–GR/GCE TiO2–c-MWCNT–GR–RUT/GCE HRP/TiO2–c-MWCNT–GR–RUT/
GCE
Found (g/100 mL) Recovery (%) Found (g/100 mL) Recovery (%) Found (g/100 mL) Recovery (%)
3.00 3.01 100.4 3.06 101.8 3.00 100.0 3.00 2.99 99.6 3.02 100.5 3.02 100.6 3.00 3.00 100.0 3.02 100.7 3.01 100.5 3.00 2.98 99.2 3.02 100.6 3.01 100.4 3.00 3.02 100.7 3.01 100.2 2.98 99.3 Recovery, % 100.0 ± 0.75 100.8 ± 0.78 100.2 ± 0.67
the HRP/TiO2–c-MWCNT–GR–RUT/GCE showed the highest sensitivity. On the other hand, the sensi-tivity obtained with TiO2–c-MWCNT–GR–RUT was about twofold higher than that obtained with HRP/ TiO2–c-MWCNT–GR/GCE which was due to the presence of RUT in the modification matrix. The results showed that RUT had a favorable catalytic ability for the reduction of H2O2and is a promising modification material for fabricating nonenzymatic H2O2sensors. Moreover, the excellent performance of the TiO2–c-MWCNT–GR–RUT composite toward detection of H2O2makes it attractive for the fabrica-tion of oxidase-based biosensors. To lower the working potential is critical especially in interference study since low working potential helps to minimize the effect of some common electroactive species which can easily interfere in current response at high potentials. In our study, the resulting (bio)sensors exhibited good ability of anti-interference to most common interferants, which was mainly attributed to the low working potential of -0.30 V. However, the existence of dopamine had seriously interfered with the determination of H2O2at HRP/TiO2–c-MWCNT– GR–RUT/GCE (10%) and HRP/TiO2–c-MWCNT– GR/GCE (30%), whereas TiO2–c-MWCNT–GR– RUT/GCE sensor had no response to dopamine. It can be concluded that the presence of RUT on the modified electrode decreased the effect of dopamine to a great extent.
Acknowledgements
This work was supported by the Research Founda-tion of Selcuk University (No: 16201004).
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of interest.
References
[1] Chen S, Yuan R, Chai Y, Hu F (2013) Electrochemical sensing of hydrogen peroxide using metal nanoparticles: a review. Microchim Acta 180(1–2):15–32
[2] Chen X, Guo C, Kong J (2012) Oxidative stress in neu-rodegenerative diseases. Neural Regen Res 7(5):376–385 [3] Wang Q, Li M, Szunerits S, Boukherroub R (2014)
Envi-ronmentally friendly reduction of graphene oxide using
tyrosine for nonenzymatic amperometric H2O2 detection.
Electroanal 26(1):156–163
[4] C¸ elik Kazıcı H, Caglar A, Aydogmus T, Aktas N, Kivrak H (2018) Microstructured prealloyed Titanium–Nickel powder as a novel nonenzymatic hydrogen peroxide sensor. J Colloid Interface Sci 530:353–360
[5] Hong J, Maguhn J, Freitag D, Kettrup A (1998) Determi-nation of H2O2and organic peroxides by high-performance
liquid chromatography with post-column UV irradiation, derivatization and fluorescence detection. Fresenius J Anal Chem 361:124–128
[6] Towne V, Will M, Oswald B, Zhao Q (2014) Complexities in horseradish peroxidase-catalyzed oxidation of dihydrox-yphenoxazine derivatives: appropriate ranges for pH values and hydrogen peroxide concentrations in quantitative anal-ysis. Anal Biochem 334(2):290–296
[7] Chen W, Li B, Xu C, Wang L (2009) Chemiluminescence flow biosensor for hydrogen peroxide using DNAzyme immobilized on eggshell membrane as a thermally stable biocatalyst. Biosens Bioelectron 24(8):2534–2540 [8] Hurdis EC, Romeyn H (1954) Accuracy of determination of
hydrogen peroxide by cerate oxidimetry. Anal Chem 26(2):320–325
[9] Nogueira RFP, Oliveira MC, Paterlini WC (2005) Simple and fast spectrophotometric determination of H2O2in
photo-Fenton reactions using metavanadate. Talanta 66(1):86–91 [10] Chen W, Cai S, Ren QQ, Wen W, Zhao YD (2012) Recent
advances in electrochemical sensing for hydrogen peroxide: a review. Analyst 137:49–58
[11] Wang L, Xu M, Xie Y, Song Y, Wang L (2018) A nonen-zymatic electrochemical H2O2sensor based on macroporous
carbon/polymer foam/PtNPs electrode. J Mater Sci 53:10946–10954. https://doi.org/10.1007/s10853-018-2386-1
[12] Sarika C, Shivakumar MS, Krishnamurthy G, Murthy BN, Lekshmi IC (2017) A novel amperometric catechol biosen-sor based on a-Fe2O3 nanocrystals-modified carbon paste electrode. Artif Cell Nanomed Biotechnol 45(3):625–634 [13] Zhang R, Chen W (2017) Recent advances in
graphene-based nanomaterials for fabricating electrochemical hydro-gen peroxide sensors. Biosens Bioelectron 89:249–268 [14] Scandurra G, Arena A, Ciofi C, Saitta G (2013) Electrical
characterization and hydrogen peroxide sensing properties of gold/nafion: polypyrrole/MWCNTs electrochemical devices. Sensors 13:3878–3888
[15] Yang J, Xiang H, Shuai L, Gunasekaran S (2011) A sensitive enzymeless hydrogen-peroxide sensor based on epitaxially-grown Fe3O4thin film. Anal Chim Acta 708(1–2):44–51
[16] Zhang X, Li L, Peng X, Chen R, Huo K, Chu PK (2013) Non-enzymatic hydrogen peroxide photoelectrochemical
sensor based on WO3decorated core–shell TiC/C nanofibers
electrode. Electrochim Acta 108:491–496
[17] Chen X, Wu G, Cai Z, Oyama M, Chen X (2014) Advances in enzyme-free electrochemical sensors for hydrogen per-oxide, glucose, and uric acid. Microchim Acta 181(7–8):689–705
[18] Wang H, Wang H, Li T, Ma J, Li K, Zuo X (2017) Silver nanoparticles selectively deposited on graphene-colloidal carbon sphere composites and their application for hydrogen peroxide sensing. Sens Actuators B Chem 239:1205–1212 [19] Benedet J, Lu D, Cizek K, Belle JL, Wang J (2009)
Amperometric sensing of hydrogen peroxide vapor for security screening. Anal Bioanal Chem 395(2):371–376 [20] Lu D, Zhang Y, Lin S, Wang L, Wang C (2013) Synthesis of
PtAu bimetallic nanoparticles on graphene–carbon nanotube hybrid nanomaterials for nonenzymatic hydrogen peroxide sensor. Talanta 112:111–116
[21] Han Y, Zheng J, Dong S (2013) A novel nonenzymatic hydrogen peroxide sensor based on Ag–MnO2–MWCNTs
nanocomposites. Electrochim Acta 90:35–43
[22] Kac¸ar C, Erden PE, Pekyardımcı S¸, Kılıc¸ E (2013) An Fe3O4-nanoparticles-based amperometric biosensor for
cre-atine determination. Artif Cell Nanomed Biotechnol 41:2–7 [23] Li J, Hu H, Li H, Yao C (2017) Recent developments in electrochemical sensors based on nanomaterials for determining glucose and its byproduct H2O2. J Mater Sci
52:10455–10469. https://doi.org/10.1007/s10853-017-12 21-4
[24] Yang W, Ratinac KR, Ringer SP, Thordarson P, Gooding JJ, Braet F (2010) Carbon nanomaterials in biosensors: should you use nanotubes or graphene? Angew Chem Int Ed 49(12):2114–2138
[25] Shi X, Gu W, Li B, Chen N, Zhao K, Xian Y (2014) Enzymatic biosensors based on the use of metal oxide nanoparticles. Microchim Acta 181(1–2):1–22
[26] Zhu J, Liu X, Wang X, Huo X, Yan R (2015) Preparation of polyaniline–TiO2 nanotube composite for the development
of electrochemical biosensors. Sens Actuators B Chem 221:450–457
[27] Kac¸ar C, Erden PE, Kılıc¸ E (2017) Graphene/poly (vinyl-ferrocene) composite based amperometric biosensor for L -lysine determination. Electroanal 29(9):2114–2124
[28] Lu Q, Dong X, Li LJ, Hu X (2010) Direct electrochemistry-based hydrogen peroxide biosensor formed from single-layer graphene nanoplatelet–enzyme composite film. Talanta 82(4):1344–1348
[29] Qian L, Yang X (2006) Composite film of carbon nanotubes and chitosan for preparation of amperometric hydrogen peroxide biosensor. Talanta 68(3):721–727
[30] Salimi A, Hallaj R, Soltanian S, Mamkhezri H (2007) Nanomolar detection of hydrogen peroxide on glassy carbon electrode modified with electrodeposited cobalt oxide nanoparticles. Anal Chim Acta 594(1):24–31
[31] Wang L, Wang E (2004) A novel hydrogen peroxide sensor based on horseradish peroxidase immobilized on colloidal Au modified ITO electrode. Electrochem Commun 6(2):225–229
[32] Niu X, Yang W, Guo H, Ren J, Gao J (2013) Highly sen-sitive and selective dopamine biosensor based on 3, 4, 9, 10-perylene tetracarboxylic acid functionalized graphene sheets/multi-wall carbon nanotubes/ionic liquid composite film modified electrode. Biosens Bioelectron 41:225–231 [33] Kac¸ar C, Dalkiran B, Erden PE, Kılıc¸ E (2014) An
amper-ometric hydrogen peroxide biosensor based on Co3O4
nanoparticles and multiwalled carbon nanotube modified glassy carbon electrode. Appl Surf Sci 311:139–146 [34] Dalkıran B, Erden PE, Kılıc¸ E (2017) Graphene and tricobalt
tetraoxide nanoparticles based biosensor for electrochemical glutamate sensing. Artif Cell Nanomed Biotechnol 45(2):340–348
[35] Hwa KY, Subramani B (2014) Synthesis of zinc oxide nanoparticles on graphene–carbon nanotube hybrid for glu-cose biosensor applications. Biosens Bioelectron 62:127–133
[36] Jin E, Bian X, Lu X, Wang C (2012) Fabrication of multiwalled carbon nanotubes/polypyrrole/Prussian blue ternary composite nanofibers and their application for enzymeless hydrogen peroxide detection. J Mater Sci 47:4326–4331. https://doi.org/10.1007/s10853-012-6283-8
[37] Erden PE, Kac¸ar C, O¨ ztu¨rk F, Kılıc¸ E (2015) Amperometric uric acid biosensor based on poly (vinylferrocene)-gelatin-carboxylated multiwalled carbon nanotube modified glassy carbon electrode. Talanta 134:488–495
[38] Niyomdecha S, Limbut W, Numnuam A, Asawatreratanakul P, Kanatharana P, Thavarungkul P (2017) A novel BOD biosensor based on entrapped activated sludge in a porous chitosan-albumin cryogel incorporated with graphene and methylene blue. Sensor Actuators B Chem 241:473–481 [39] Erden PE, Zeybek B, Pekyardimcı S¸, Kilic¸ E (2013)
Amperometric carbon paste enzyme electrodes with Fe3O4
nanoparticles and 1, 4-benzoquinone for glucose determi-nation. Artif Cell Nanomed Biotechnol 41(3):165–171 [40] Chaubey A, Malhotra B (2002) Mediated biosensors.
Bio-sens Bioelectron 17(6–7):441–456
[41] Astier Y, Canters GW, Davis JJ, Hill HA, Verbeet MP, Wijma HJ (2005) Sensing nitrite through a pseudoazurin– nitrite reductase electron transfer relay. Chem Phys Chem 26(6):1114–1120
[42] Loew N, Tsugawa W, Nagae D, Kojima K, Sode K (2017) Mediator preference of two different fad-dependent glucose dehydrogenases employed in disposable enzyme glucose sensors. Sensors 17(11):2636
[43] Kamyabi MA, Hajari NJ (2017) Low potential and non-enzymatic hydrogen peroxide sensor based on copper oxide nanoparticle on activated pencil graphite electrode. J Brazil Chem Soc 28(5):808–818
[44] Zhang Y, Yuan R, Chai Y, Xiang Y, Hong C, Ran X (2010) An amperometric hydrogen peroxide biosensor based on the immobilization of HRP on multi-walled carbon nan-otubes/electro-copolymerized nano-Pt-poly (neutral red) composite membrane. Biochem Eng J 51(3):102–109 [45] Campuzano S, Serra B, Pedrero M, de Villena FJM,
Pin-garro´n JM (2003) Amperometric flow-injection determina-tion of phenolic compounds at self-assembled monolayer-based tyrosinase biosensors. Anal Chim Acta 494(1–2):187–197
[46] Yang L, Ren X, Tang F, Zhang L (2009) A practical glucose biosensor based on Fe3O4nanoparticles and chitosan/nafion
composite film. Biosens Bioelectron 25:889–895
[47] Yao H, Hu NJ (2010) pH-switchable bioelectrocatalysis of hydrogen peroxide on layer-by-layer films assembled by concanavalin A and horseradish peroxidase with electroac-tive mediator in solution. J Phys Chem B 114(9):3380–3386 [48] Metzker G, de Aguiar I, Martins SC, Schultz MS, Vascon-cellos LCG, Franco DW (2014) Electrochemical and chemical aspects of ruthenium (II) and (III) ammines in basic solution: the role of the ruthenium (IV) species. Inorg Chim Acta 416:142–146
[49] Dervisevic M, Custiuc E, C¸ evik E, S¸enel M (2015) Con-struction of novel xanthine biosensor by using polymeric mediator/MWCNT nanocomposite layer for fish freshness detection. Food Chem 181:277–283
[50] Liu S, Ju H (2003) Reagentless glucose biosensor based on direct electron transfer of glucose oxidase immobilized on colloidal gold modified carbon paste electrode. Biosens Bioelectron 19(3):177–183
[51] C¸ elik AC, O¨ ztu¨rk F, Erden PE, Kac¸ar C, Kılıc¸ E (2015) Amperometric lactate biosensor based on carbon paste electrode modified with benzo [c] cinnoline and multiwalled carbon nanotubes. Electroanal 27(12):2820–2828
[52] Laviron E (1979) General expression of the linear potential sweep voltammogram in the case of diffusionless electro-chemical systems. J Electroanal Chem Interfacial Elec-trochem 101(1):19–28
[53] Ruzgas T, Gorton L, Emneus J, Marko-Varga G (1995) Kinetic models of horseradish peroxidase action on a gra-phite electrode. J Electroanal Chem 391(1–2):41–49
[54] Devi R, Batra B, Lata S, Yadav S, Pundir CS (2015) A method for determination of xanthine in meat by ampero-metric biosensor based on silver nanoparticles/cysteine modified Au electrode. Process Biochem 48(2):242–249 [55] Rocchitta G, Spanu A, Babudieri S, Latte G, Madeddu G,
Galleri G, Nuvoli S, Bagella P, Demartis MI, Fiore V, Manetti R, Serra PA (2016) Enzyme biosensors for biomedical applications: strategies for safeguarding analyti-cal performances in biologianalyti-cal fluids. Sens (Basel) 6:780 [56] Korkut S, Keskinler B, Erhan E (2008) An amperometric
biosensor based on multiwalled carbon nanotube-poly(-pyrrole)-horseradish peroxidase nanobiocomposite film for determination of phenol derivatives. Talanta 76:1147–1152 [57] Abdelwahab AA, Shim YB (2014) Nonenzymatic H2O2
sensing based on silver nanoparticles capped polyterthio-phene/MWCNT nanocomposite. Sens Actuators B Chem 201:51–58
[58] Wang MY, Shen T, Wang M, Zhang DE, Tong ZW, Chen J (2014) One-pot synthesis of a-Fe2O3
nanoparticles-deco-rated reduced graphene oxide for efficient nonenzymatic H2O2biosensor. Sens Actuators B Chem 190:645–650
[59] Xin Y, Fu-bing X, Hong-wei L, Feng W, Di-zhao C, Zhao-yang W (2013) A novel H2O2biosensor based on Fe3O4–Au
magnetic nanoparticles coated horseradish peroxidase and graphene sheets–Nafion film modified screen-printed carbon electrode. Electrochim Acta 109:750–755
[60] Guler M, Turkoglu V, Basi Z (2017) Determination of malation, methidathion, and chlorpyrifos ethyl pesticides using acetylcholinesterase biosensor based on Nafion/Ag@ rGO-NH2nanocomposites. Electrochim Acta 240:129–135
[61] Apetrei C, Rodrı´guez-Me´ndez ML, De Saja JA (2011) Amperometric tyrosinase based biosensor using an elec-tropolymerized phosphate-doped polypyrrole film as an immobilization support. Application for detection of phe-nolic compounds. Electrochim Acta 56(24):8919–8925 [62] Chen C, Hong X, Xu T, Chen A, Lu L, Gao Y (2016)
Hydrogen peroxide biosensor based on the immobilization of horseradish peroxidase onto a poly (aniline-co-N-methylthionine) film. Synth Metal 212:123–130
[63] Xiang C, Zou Y, Sun LX, Xu F (2009) Direct electro-chemistry and enhanced electrocatalysis of horseradish per-oxidase based on flowerlike ZnO–gold nanoparticle–Nafion nanocomposite. Sens Actuators B Chem 136(1):158–162 [64] Liu SQ, Ju HX (2002) Renewable reagentless hydrogen
peroxide sensor based on direct electron transfer of horse-radish peroxidase immobilized on colloidal gold-modified electrode. Anal Biochem 307(1):110–116
[65] Kristine FJ, Johnson CR, O’Donnell S, Shepherd RE (1980) Reduction of hydrogen peroxide by ruthenium (II) ammine complexes: the surprisingly identical mechanism for
hexaammineruthenium (2?), pentaammine (aquo) ruthenium (2?), and pentaammine (1-methylimidazole) ruthenium (2?). Inorg Chem 19(8):2280–2284
[66] Zhao W, Wang H, Qin X, Wang X, Zhao Z, Miao Z, Chen L, Shan M, Fang Y, Chen Q (2009) A novel nonenzymatic hydrogen peroxide sensor based on multi-wall carbon
nanotube/silver nanoparticle nanohybrids modified gold electrode. Talanta 80(2):1029–1033
[67] Khan MM, Ansari SA, Lee J, Cho MH (2013) Novel Ag@ TiO2nanocomposite synthesized by electrochemically active
biofilm for nonenzymatic hydrogen peroxide sensor. Mat Sci Engi C Mater 33(8):4692–4699
