Appl Biochem Biotechnol DOI 10.1007/s12010-014-0999-7

Immobilization of Horseradish Peroxidase Enzyme on Nanoporous Titanium Dioxide Electrodes and Its Structural and Electrochemical Characterizations E. T. Deva Kumar & V. Ganesh

Received: 24 January 2014 / Accepted: 23 May 2014 # Springer Science+Business Media New York 2014

Abstract Hierarchically ordered, honeycomb-like nanoporous TiO2 electrodes are prepared by a simple electrochemical anodization process using ammonium fluoride dissolved in ethylene glycol as an electrolytic medium. Formation of hexagonally arranged nanopores along with the tubular structure and anatase crystalline phase of TiO2 is confirmed by field emission scanning electron microscope (FESEM) and X-ray diffraction (XRD) studies. Further, these nanoporous TiO2 electrodes are employed as a substrate for enzyme (horseradish peroxidase, HRP) immobilization in an attempt to enhance the electron transport across the semiconductor electrode–electrolyte interface. Two different strategies, namely, physical entrapment and covalent linking, are used for anchoring the enzyme. Various parameters such as conductivity, stability, enzyme loading, enzymatic activity, sensitivity, linear range, etc., are investigated by using electrochemical techniques. Structural and morphological analyses of enzyme-modified electrodes are carried out using spectroscopic (UV−vis) and microscopic (AFM) methods. In the case of physical entrapment, a simple drop casting method of HRP solution on the nanoporous TiO2 electrodes is used in contrast to chemical linking method where a monolayer of 3-aminopropyltrimethoxy silane (APTMS) is formed initially on TiO2 followed by HRP immobilization using an amide coupling reaction. Interestingly, both of these methods result in anchoring of HRP enzyme, but the amount of enzyme loading and the stability are found to be higher in the covalent linking method. Cyclic voltammetric studies reveal the formation of a well-defined reversible peak for HRP enzyme. Dependence of peak current with the scan rate suggests that HRP enzyme is immobilized and stable and that the overall electron transfer process is predominantly controlled by a diffusion process. Enzymatic activity of HRP is investigated by monitoring the reduction process of hydrogen peroxide by incremental addition using cyclic voltammetry and amperometry techniques, from which several kinetic parameters are determined.

E. T. Deva Kumar : V. Ganesh (*) Electrodics and Electrocatalysis (EEC) Division, CSIR—Central Electrochemical Research Institute (CSIR–CECRI), Karaikudi 630006 Tamilnadu, India e-mail: [email protected] V. Ganesh e-mail: [email protected]

Appl Biochem Biotechnol

Keywords Electrochemical sensor . Enzymes . Hydrogen peroxide . Horseradish peroxidase . Nanomaterials . Semiconductor electrodes . Titanium dioxide

Introduction Enzymes are considered as one of the most important biological compounds owing to their enhanced catalytic activity and specificity towards many biological reactions. It is fundamentally interesting to analyze the enzymatic reactions because of their ability to provide a mechanism of electron exchange between different structures. [1–6] This also provides big hope to invent new drugs and new catalysts for various industrial applications based on the established mechanisms. In general, these studies can be performed via protein solution techniques and surface-confined techniques. In the case of surface-confined techniques, basically the enzymes are anchored over some suitable substrates like biopolymers [7, 8] and their reactions are mostly followed by using either spectroscopic or electrochemical techniques. [9–13] These surface immobilization techniques have their own merits and demerits. Some of the merits include fast response time, cost-effective method, need of a smaller amount of analyte for the study, and speedy way of detection, and the demerits are that usually the active sites of the enzyme are deeply buried into the polymer layers, loss of conduction (flow of electrons) between the active site and the electrode surface, and so on. During the early stages of development of biosensors, these limitations are overcome by the use of redox mediators such as methylene violet, ferrocene, and potassium ferrocyanide in order to enhance electron shuttling between the enzymes and the substrates. [14] In addition, electron transfer is also facilitated by using various promoters, namely, amino acids [15], carbohydrates [16], inorganic compounds [17], biological macromolecules, polymers etc. In the third generation of biosensors, these redox mediators are eliminated with the aid of direct electron transfer between the enzyme and the electrode surface. The surface area of the electrodes is increased by creating a suitable environment for protecting the enzymatic activity and to increase the enzyme loading. [18] Micro/nano-environment is normally crafted by forming polymer films and by sol−gel methods [19] in order to create a platform for enzyme immobilization, to load a higher amount of enzymes, and for preferable orientation of the enzyme. All of these factors ultimately result in appropriate conditions for facilitation of electron transfer process. [20–23] It is well known that direct electron transfer between active sites present within the enzyme and the substrate is generally viewed as a flow of current in the external circuit. Hence, to assist this further, porous electrodes such as indium oxide, [6, 24, 25] aluminium oxide, [26] copper, and nickel are employed as the electrode surfaces. Among these fascinating electrode materials, titanium dioxide (TiO2) is generally ignored for biosensing applications especially for investigating enzymatic kinetics, yet highly ordered porous electrodes of TiO2 can easily be obtained. These electrodes provide a better microenvironment for the enzymes, and because of its bio-compatibility it can hold the enzyme as such without undergoing denaturation. Out of various enzymes known, horseradish peroxidase (HRP), a representative of heme-containing enzyme [27], is important since it is associated with many vital biological functions. [28, 29] Therefore, it has been used as a biological recognizing element in several biosensor kits and bio-electroanalytical devices by monitoring the process of hydrogen peroxide (H2O2) reduction. [30] Many different materials including conducting polymers [31, 32], carbon nanotubes [33, 34], graphene [35–37], and nanostructures of gold [38, 39] are mostly used for the development of H2O2 biosensors. Though TiO2 is employed for a variety of applications including solar cells, photocatalysis, photovoltaics, photoelectrochemical water splitting,

Appl Biochem Biotechnol

biomedical field, etc., until now there are only very few reports on using TiO2 electrodes for enzymatic studies. [40–43] Curulli et al. initially reported the immobilization of glucose oxidase (GOD) and HRP on nanostructured substrates, namely, TiO2 film prepared by sol− gel method and gold nanotubes grown on Si wafers. They demonstrated that these enzymes can be immobilized and exhibit a quasi-reversible surface-confined electrochemical processes [40], but they did not really investigate the enzymatic reactions for sensing an analyte and did not prove the enzymatic activity by determining their sensitivity, stability, limit of detection, etc. Very recently, Tang et al. showed a photoelectrochemical biosensor for glucose detection using TiO2 nanowires grown on fluorine-doped tin oxide (FTO) glass by chemically immobilizing GOD onto them. [43] They reported an impressive sensitivity of ~0.9 nm, but this procedure requires tedious instrumentation and laborious work in addition to employing a complicated technique for monitoring the sensing events. Keeping this in mind, in this work, a simple methodology to prepare hierarchically ordered, hexagonally arranged porous nanotubes of TiO2 electrodes is developed. This method involves electrochemical anodization of Ti plates in an electrolyte consisting of diethylene glycol and ammonium fluoride by passing a constant voltage over a fixed duration. FESEM and XRD are used for their structural and morphological analyses. Further, these porous electrodes are employed as substrates for enzyme immobilization performed by two ways, viz., by physical entrapment method and by chemical functionalization (covalent bonding). A simple drop casting method is adopted for physical entrapment and a self-assembled monolayer film of APTMS is used further for anchoring of enzymes through coupling reaction in the case of covalent bond formation. Electrochemical techniques, namely, cyclic voltammetry (CV) and UV−vis spectroscopy, are employed for investigation of enzymatic reactions. Several parameters associated with the electrochemical biosensor are analyzed. Here we demonstrate HRP as an example and a similar strategy can easily be extended for other biomolecules such as proteins, DNA, RNA, etc.

Materials and Methods Chemicals Ethylene glycol (SRL chemicals), ammonium fluoride (Merck), hydrogen peroxide (Aldrich), 3-aminopropyltrimethoxysilane (Aldrich), HRP—peroxidase from horseradish as a type VI lyophilized powder (Aldrich), and toluene (Merck) were purchased and used as such without further purification unless otherwise stated. Millipore water having a resistivity of 18.2 MΩ cm obtained from a quartz distillation unit was used for all the experiments and analysis. Titanium plates (0.127 mm in thickness) possessing 99.7 % metal purity were purchased from Aldrich and used for electrochemical anodization process. These plates were cut into small pieces with a pre-defined geometric area of 1×1 cm2 for further studies. Electrode Processing Ti plates of prescribed geometric area were cleaned thoroughly by sonication in water and acetone sequentially for about 5 mins each. Further, these plates were cut into small pieces and sonicated with water and acetone simultaneously to clean the surface. Then, the pre-cleaned plates were made into electrodes by connecting with a copper wire using Teflon and parafilm to cover the rest of the area. Finally, these home-made electrodes were used as working electrodes. Platinum wire was used as a counter electrode; this was cleaned by dipping in

Appl Biochem Biotechnol

concentrated HNO3 for a few minutes and thoroughly cleaned with water. Saturated calomel electrode (SCE) was used as a reference electrode and it was cleaned by rinsing with millipore water. Electrochemical Anodization Pre-cleaned Ti plates were converted into nanoporous TiO2 electrodes by using a simple electrochemical anodization process. A two-electrode system consisting of Ti plate as an anode and Pt foil as a cathode was used. Anodization was performed in ethylene glycol medium containing 0.3 wt.% ammonium fluoride and 2 vol.% millipore water for about 24 h with a continuous supply of constant DC voltage of 30 V. After the anodization process, these plates were taken out, washed well with plenty of water, and sonicated for about 1–3 min to remove any loosely bound TiO2 tubes and other impurities, so-called debris. Then, the anodized plate was annealed at 450 °C in the presence of air for about 6 h in order to obtain pure crystalline TiO2 nanotubes. Formation of these tubes along with the porous structure was confirmed using FESEM and XRD studies. Immobilization of HRP on TiO2 Electrodes HRP was immobilized onto nanoporous TiO2 electrodes using two different strategies, namely, physical entrapment of the enzyme and chemical functionalization methods. In the case of physical adsorption, 2.5 mg of HRP was dissolved in 1 ml of phosphate (PBS) buffer (pH= 7.0) aqueous solution which was drop-casted over TiO2 electrodes and kept at 4 °C for about 14 h. After that, the enzyme-modified electrodes were taken out, washed thoroughly with phosphate buffer and millipore water, and used for further analysis and characterization. In the case of chemical functionalization method, initially, a self-assembled monolayer (SAM) of APTMS was formed on TiO2 electrodes by dipping them into 10 mM APTMS solution in toluene for about 14–15 h. Cyclic voltammetry (CV) was used to analyze the formation of monolayer on TiO2 electrodes. SAM-modified TiO2 electrodes were activated by dipping in 25 mM EDC and 30 mM NHS solution, and this mixture was stirred for 1 h in order to activate the surface functional groups present within the monolayer. About 2.5 mg of HRP was dissolved in 1 ml of PBS buffer, and this solution was added to the said mixture. Further, the resultant solution was stirred at 4 °C for about 14 h to chemically anchor HRP onto nanoporous TiO2 electrodes, and immobilization was confirmed by diffuse reflectance UV–vis spectroscopy, AFM, and electrochemical techniques. Electrochemical Investigation of HRP on TiO2 Electrodes A three-electrode setup was used for the electrochemical characterization. Cyclic voltammetry (CV) was primarily used for the analysis. A Pt wire and saturated calomel electrode (SCE) were used as counter and reference electrodes, respectively. HRP-immobilized nanoporous TiO2 electrode was employed as a working electrode. Prior to analysis, the counter electrode Pt wire is cleaned by dipping in concentrated HNO3 for about 2 min, and the reference electrode was thoroughly washed with millipore water as mentioned earlier. Formation of APTMS monolayer on TiO2 electrode was confirmed by recording CV for a well-known redox system, namely, potassium ferro/ferri cyanide, and change in redox current was monitored. In the case of HRP-modified electrodes, CV was recorded in an aqueous solution of phosphate buffer (PBS) of pH=7.0 under N2 atmosphere at a sweep rate of 50 mV/s over a potential range from −0.8 to 0.8 V. Initially, the redox behavior of HRP enzyme was investigated. In addition, scan

Appl Biochem Biotechnol

rate dependence on the redox peak was also analyzed by varying the scan rate from 10 to 100 mV/s. Enzymatic kinetics was studied using CV, and this experiment was carried out in an aqueous PBS buffer solution under N2 atmosphere at a fixed scan rate of 50 mV/s for incremental additions of hydrogen peroxide (H2O2) with a potential ranging from 0.8 to −0.8 V. The concentration of H2O2 was varied from 100 μM to 2 mM by preparing a stock solution of 10 mM concentration with an incremental addition of 100 μL. For comparison, similar experiments were carried out using TiO2 electrodes without HRP immobilization and Ti electrodes before anodization under similar conditions. HRP-anchored TiO2 electrodes were stored at 4 °C whenever not in use. All these experiments were carried out at room temperature. Instrumentation FESEM studies were performed using Hitachi model S3000-H which also has an EDAX facility attached. AFM analysis was carried out using a scanning probe microscope obtained from Agilent Technologies (Model 5500). The images were recorded using non-contact tapping mode with the help of a cantilever made of n+–silicon type (NSC-18) by employing Picoview software. These images were corrected for a plane tilt and analyzed using PicoImage software. XRD studies were carried out using XPERT-PRO multipurpose X-ray diffractometer procured from The Netherlands using Cu Kα radiation with a wavelength of 1.540 Å. UV−vis analysis in diffuse reflectance mode was performed using Perkin Elmer Lambda 650 with Infinite M200MPC model UV−vis spectrophotometer attached with 60-mn integrated sphere detector module. These spectra were recorded over a wavelength range of 250 to 800 nm. Electrochemical studies were carried out using Electrochemical Impedance Analyzer model 6310, EG&G instruments obtained from Princeton Applied Research, USA, and echem software provided by them was used for the data collection and analysis. All the other parameters are shown in the respective diagrams.

Results and Discussion Formation and Characterization of TiO2 Nanotubes Using SEM Analysis Figure 1 shows FESEM images of bare Ti (Fig. 1a) and anodized Ti (Fig. 1b–d) plates using ethylene glycol electrolytic medium containing ammonium fluoride as an additive, leading to the formation of nanoporous TiO2 tubular structure. Anodization was carried out at an optimized voltage of 30 V for a duration of 24 h. It can be seen from the images that bare Ti (Fig. 1a) shows no distinguishable structural features, and the entire surface looks very smooth and homogeneous. On the contrary, anodized Ti plate (Fig. 1b–d) depicts the formation of uniform porous structures with an array of nanotubes. These images display the structure of anodized Ti plates at different scales. The top surface of as-prepared anodized plates was found to be covered with broken nanotubes and other debris (Fig. 1b, d). A simple sonication process in acetone or ethanol is employed to remove those loosely bound broken nanotubes and results in the formation of more uniform and homogeneous array of nanotubes (Fig. 1c). Nevertheless, the formation of a honeycomb structure with ordered hexagonal arrangement of porous nanotubes is clearly seen (Fig. 1b). These images clearly show the formation of beautiful ripples across the growth of the nanotubes (Fig. 1c). These nanotubes are about 14–15 μm in length, and the outer diameter of the tubes is ~150±5 nm, with a wall thickness of about 15– 20 nm. Detailed analysis of these images indicates that some of the pores seem to be a

Appl Biochem Biotechnol

Fig. 1 Field emission scanning electron microscopic (FESEM) images of a bare Ti and b–d anodized Ti electrodes at various scales obtained using ethylene glycol consisting of ammonium fluoride as an electrolytic medium at an applied voltage of 30 V for 24 h

combination of two or three nanotubes (Fig. 1d), suggesting the variation of current density distribution in the course of anodization. In fact, during the initial stage of anodization, formation of hairy, fiber-like structures was observed. Due to the fluctuations in current density arising out of conductivity associated with the electrolyte employed, thickness of the oxide layer formed, and potential drop across the electrode–electrolyte interface, these eventually result in the formation of a homogeneous, highly ordered hexagonal arrangement of arrays of porous nanotubes leading to a honeycomb-like structure possessing an average pore diameter of ~100–120 nm, which is clearly evident from these images. Further elemental mapping analysis of these anodized samples indicates the presence of Ti and oxygen, suggesting the formation of pure TiO2 nanotubes during anodization. XRD Studies Phase purity and crystalline structure of the anodized Ti plates were characterized using X-ray diffraction (XRD) studies. Figure 2 shows the XRD pattern of bare Ti (before anodization; Fig. 2a) and anodized Ti (TiO2; Fig. 2b) plates. It can be noted from the pattern that bare Ti displays a disticnt prominent peak at 38.87° corresponding to (002) plane and other peaks at 35.65°, 40.65°, 53.39°, and 71.05° representing the existence of (100), (101), (102), and (103) planes of Ti, respectively. [44–47] On the other hand, anodized plates after annealing show the

Appl Biochem Biotechnol Fig. 2 X-ray diffraction spectra of a bare Ti (before anodization) and b TiO2 prepared by electrochemical anodization in ethylene glycol containing ammonium fluoride as the electrolytic medium (30 V for 24 h). Here asterisk denotes the peaks corresponding to the formation of anatase TiO2 crystalline phase

peaks at 25.89° and 63.41° representing (101) and (110) planes corresponding to the anatase phase of TiO2, respectively, [44–47] in addition to the peaks of underlying Ti substrate. Since the TiO2 film formed after anodization is so thin when compared to base Ti metal, the peaks due to Ti were apparent and clearly visible. Nevertheless, these studies clearly demonstrate the formation of closely packed hexagonal arrangement of porous TiO2 nanotubes arrays that are crystalline anatase phase. Characterization of Porous TiO2 Nanotubes Using AFM Measurements and Diffuse Reflectance UV−Vis Spectroscopic Analysis Growth and surface profile of the anodized Ti plates leading to the formation of porous TiO2 nanotubes were further analyzed using AFM measurements and diffuse reflectance UV−vis spectroscopy. Figure 3a shows the representative 3D image of TiO2 plates at 2-μm scale obtained by anodization at 30 V for 24 h using ethylene glycol containing ammonium fluoride

Fig. 3 Surface topography image (3D representation, a) and surface height profile (c) of TiO2 electrodes obtained using AFM studies. b UV−vis spectrum of porous TiO2 nanotubular electrodes recorded in diffuse reflectance mode

Appl Biochem Biotechnol

as the electrolytic medium. Furthermore, the surface topography and surface profile of this sample were also analyzed and the corresponding graph is shown in Fig. 3c. These images were basically taken using non-contact mode, and they are highly reproducible. Moreover, the images were obtained at various locations of the TiO2 sample in order to investigate the exact representation of the resultant anodized sample. The formation of different patches of ~300 nm in size can be clearly seen from the image, and the surface height of these structural features varies from ~10 to 20 nm and vertically extends by the formation of pores within these domains. In-depth analysis of this image indicates the formation of nanotube arrays (Fig. 3a), and small surface corrugations may be corresponding to the formation of individual nanotubes. Figure 3b displays the UV−vis spectra of TiO2 obtained in diffuse reflectance mode. It showed a broad reflectance covering a complete visible region wavelength from 300 to 800 nm rather than any peaks, but small kinks were oberseved at 250 and 297 nm that were assigned to the reflection of the underlying Ti metal. Deeper analysis of this spectra suggests that the relative reflectance value is very much lower, indicating the maximum absorption of light over a wide wavelength region. This peculiar phenomenon can be employed for biosensing applications where the molecules which absorb in the visible region could be easily detected with enhanced absorption without the addition of any dopants or sensitizers. Band gap values of these porous TiO2 nanotubes were determined from this spectrum by converting the absolute reflection values to Kubelka–Munk function [48–52], and the value is determined to be 3.2 eV, which is very similar to the semiconductive TiO2 material. Overall, these studies clearly demonstrate the formation of hexagonally ordered arrays of porous TiO2 nanotubes as a result of anodization of Ti plates. Characterization of HRP-Immobilized TiO2 Electrodes by Using SEM, AFM, and Diffuse Reflectance UV−Vis Spectroscopic Analyses As mentioned earlier, two different methods were followed for HRP enzyme immobilization onto nanoporous TiO2 electrodes. In the first method, a simple physical adsorption of enzyme was carried out by drop-casting the HRP solution on these electrodes. On the contrary, the second method involves the chemical functionalization of HRP on TiO2 electrodes. In this case, initially self-assembled monolayer of APTMS was formed on TiO2 which was further functionalized with HRP using a well-known EDC coupling reaction. Formation of SAM was confirmed from the reduction of peak current corresponding to Fe(CN)63−/4− redox couple using CV studies. Subsequently, HRP-immobilized TiO2 electrodes were characterized by using microscopic (SEM and AFM), spectroscopic, and electrochemical studies. Figure 4 shows the SEM images of HRP-anchored TiO2 electrodes using physical (Fig. 4a) and chemical (Fig. 4b) methods, respectively. It can be noted from these images that the physical method results in a poor assembly of enzymes on TiO2 in contrast to the chemical method where the formation of dense areas of enzymes is clearly seen. From Fig. 4b, it is evident that some pores are blocked by HRP immobilization and the aggregated structures are fairly distinguishable. Similarly, Fig. 5a, b displays the representative 3D images of HRPimmobilized TiO2 electrodes by following the physical and chemical methods obtained using AFM studies. These images are clearly different from nanoporous TiO2 electrodes before HRP immobilization (Fig. 3a). The surface appeared to be smooth because of the complete coverage by enzyme in both cases. Further surface profile analysis of these images indicates the difference in surface heights/thickness, suggesting the different structural arrangements of enzymes. Figure 5a shows mostly the formation of domains of ~80–120 nm in size. Figure 5b displays the uniform coverage of individual TiO2 nanotubes by HRP using chemical functionalization, and the pores appeared to be blocked due to the presence of enzymes.

Appl Biochem Biotechnol

Fig. 4 SEM pictures of HRP-immobilized TiO2 electrodes prepared by following physical adsorption (a) and chemical functionalization (b) methods

Furthermore, immobilization of HRP on these electrodes was confirmed using diffuse reflectance UV−vis spectroscopic studies, and the corresponding spectrum is shown in Fig. 5c. Formation of a peak at 365 nm, which is a characteristic soret absorption of HRP enzyme, arises from the presence of a heme prosthetic group, and a small kink at 263 nm in the case of

Fig. 5 AFM images shown in 3D representation of HRP-anchored nanoporous TiO2 electrodes obtained by simple physical adsorption (a) and by a chemical modification procedure (b) followed for enzyme immobilization. c UV−vis diffuse reflectance spectrum of HRP-immobilized TiO2 electrode using chemical methods and d UV−vis spectrum of free HRP enzyme (before immobilization) dissolved in phosphate (PBS) buffer solution (pH=7.0) by monitoring its absorbance

Appl Biochem Biotechnol

chemical functionalization method suggests the anchoring of HRP enzyme on the porous TiO2 electrodes. In order to make sure that these absorption peaks arise from the enzyme, UV−vis absorption spectrum of HRP dissolved in PBS buffer solution (pH=7.0) was recorded, and the respective spectrum is shown in Fig. 5d. This clearly depicts the formation of two distinct peaks at 270 and 400 nm (soret absoption) due to the presence of a heme prosthetic group within HRP enzyme [51, 52], which is very similar to the spectra shown in Fig. 5c. In-depth analysis indicates the shift in soret absorption peak in the case of HRP immobilized on TiO2 to lower wavelength, suggesting the compact structural arrangement of enzymes within the pores of TiO2 although the other peak was weak due to massive absorption of light by the porous TiO2 nanotubes. Nevertheless, the formation of hexagonally ordered, porous nanotubular structures of TiO2 with the subsequent HRP enzyme immobilization is evident from all these characterization techniques, which is supplemented further by electrochemical studies. Electrochemical Characterization of HRP-Anchored Nanoporous TiO2 Electrodes Cyclic voltammetry (CV) is primarily used for the investigation of redox property and electron transfer process of HRP-immobilized porous nanotubular TiO2 electrodes. CV studies were carried out in an aqueous PBS buffer solution of pH=7.0 at a fixed scan rate of 50 mV/s over the potential range from −0.8 to 0.8 V vs. SCE under N2 atmosphere. Both physical and chemical functionalized enzyme-anchored TiO2 electrodes were analyzed, and the corresponding cyclic voltammograms are shown in Fig. 6a, b, respectively. Formation of redox peaks in CV for both electrodes corresponding to HRP enzyme suggests the direct electron transfer process in the case of HRP-immobilized TiO2 electrodes. This process is highly significant especially in the case of semiconductor electrodes without the addition of any external redox mediators. This redox peak is formed due to Fe(III)/Fe(II) redox reaction occurring in the active site of heme cofactor, which is a crucial component in oxidoreductase enzyme family [53–55] and HRP is one such kind. The formal potential of this redox process is −0.22 V with HRP enzyme in its native structure. [56] In our case, formal potential for physically immobilized HRP is determined to be −0.285 V, and in the case of chemical functionalization method this value is found to be −0.432 V. These values are very close to the formal potential of HRP in its native structure. Small difference in the potential values arises due to the structural arrangement, orientation, and accessibility of the redox groups present within the

Fig. 6 Cyclic voltammograms of HRP-functionalized TiO2 electrodes using physical adsorption (a) and chemical modification (b) methods recorded in phosphate buffer (PBS, pH=7.0) aqueous solution under N2 atmosphere at a fixed scan rate of 50 mV/s

Appl Biochem Biotechnol

immobilized HRP on TiO2 electrodes. Surface coverage of HRP enzyme on these electrodes is determined using the following equation by measuring the area under these redox peaks: Γ ¼ Q=nFA

ð1Þ

where Γ is the surface coverage in terms of concentration, Q is the area (charge) under the redox peak, n is the number of electrons involved in the redox reaction, F is the Faraday’s constant, and A is the geometric area of the electrode, respectively. Using Eq. (1), the surface concentration of HRP is calculated to be 1.408×10−9 moles in the case of physical adsorption method and 8.072×10−9 moles for the chemical functionalization method, respectively. On comparison, about five to six times higher enzyme loading is obtained in the case of chemical functionalization method compared to physical adsorption of HRP on TiO2 electrodes. This observation is also supported by microscopic studies discussed earlier. Similarly, the active surface area of HRP-immobilized TiO2 electrodes is calculated using the modified Randles– Sevcik equation provided as follows. Ip ¼ 0:4463nFðnF=RTÞ1=2 ACD1=2 v1=2

ð2Þ

where Ip is the peak current density, n is the number of electrons involved in the redox reaction, F is the Faraday’s constant, R is the universal gas constant, T is the temperature, A is the active surface area of the electrode, C is the concentration of the redox probe, D is the diffusion coefficient, and ν is the sweep rate, respectively. By using a standard redox probe, namely, Fe(CN)63−/Fe(CN)64− couple, and by assuming a diffusion coefficient value of 6.20× 10−6 cm2/s, the active surface area was determined to be 0.44 and 2.85 cm2 for physical and chemical methods of enzyme functionalization, respectively. Based on these values, the surface coverage of HRP enzyme on these electrodes was estimated to be 2.8 and 3.2 nmol per unit active surface area of electrodes for physical and chemical methods, respectively. It is also worth mentioning here that, for comparison, Ti electrodes before anodization was also used for HRP immobilization under similar conditions using the same methods. CV studies performed on those electrodes displayed no characteristic redox peak, indicating that no enzyme was immobilized on those electrodes. These results provide evidence that TiO2 nanotubes provide a good microenvironment favoring enzyme immobilization and orientation. Moreover, the effect of scan rate on this redox behavior of HRP on TiO2 electrodes was investigated over a wide range of scan rate varying from 10 to 100 mV/s. The respective CVs recorded for both cases of physical and chemical immobilization methods are displayed in Fig. 7a, b, respectively. In these figures, the arrow indicates the direction of increasing scan rate values. It can be observed that the redox current increases systematically with the increasing scan rate in both cases. In addition, a shift in the redox potential values was also noted. In fact, in the case of physical adsorption of HRP, increasing scan rate results in the large separation of forward and reverse peaks, leading to a major shift in the redox potential values. In contrast, chemically immobilized electrodes show a very small shift in the redox potentials, but the current increases quite significantly with the increase in scan rate. Analysis of change in peak currents with respect to square root of scan rate suggests a linear relation especially in the case of chemically anchored HRP on TiO2 electrodes, indicating that the direct electron transfer process is predominantly controlled by a diffusion process. This also recommends that HRP is surface-confined, immobilized onto TiO2 electrodes and retained its redox property, involved in direct electron transfer process, and did not undergo any structural deformation while potential cycling in both cases. This also supports our observation from microscopic and spectroscopic characterizations that HRP has been anchored onto TiO2 and retains its structure and activity through proper orientation.

Appl Biochem Biotechnol

Fig. 7 Cyclic voltammograms of HRP-immobilized TiO2 electrodes prepared by following simple physical adsorption (a) and by a chemical functionalization method (b) in PBS buffer (pH=7.0) aqueous solution under N2 atmosphere at various scan rates ranging from 10 to 100 mV/s to investigate the effect of scan rate on the redox behavior of HRP. In these figures, arrows indicate the direction of increasing scan rate

Electrochemical Activity of HRP-Immobilized TiO2 Electrodes Towards H2O2 Reduction HRP-immobilized TiO2 electrodes were further explored for the possibility of fabrication of an electrochemical biosensor for H2O2 detection. These studies were carried out by monitoring the enzymatic activity of reduction of H2O2 in aqueous phosphate buffer (PBS, pH=7.0) solution using CV. Generally, colorimetric techniques are employed for H2O2 detection without HRP enzyme, which is primarily non-selective, requires additional reagents, and is user dependent. Herein, enzymatic detection is employed by functionalizing TiO2 with HRP, which is very selective; redox groups present within the enzyme are accessible and involved direct electron transfer process without redox mediators for the reduction of H2O2 that can be monitored easily using electrochemical techniques. Figure 8 shows the CVs of HRP-anchored electrodes using a chemical method towards the addition of various concentrations of H2O2 in aqueous phosphate buffer solution (pH=7.0) under N2 atmosphere at a scan rate of 50 mV/s.

Fig. 8 Representative cyclic voltammograms corresponding to HRP-anchored TiO2 electrode prepared by chemical functionalization method in aqueous phosphate buffer solution (pH=7.0) at a sweep rate of 50 mV/s showing the bio-electrocatalytic response towards the addition of various concentrations of H2O2. The concentration was varied from 100 μM to 2 mM. Arrow indicates the direction of increasing concentration of H2O2. Inset shows a plot of current vs. concentration of H2O2 corresponding to various additions. These data points were collected from CV measurements

Appl Biochem Biotechnol

In this figure, the arrow indicates the direction of increasing H2O2 concentration. Upon addition of H2O2, a significant enhancement in the reduction current was noticed. A systematic increase in H2O2 concentration results in a systematic increase of the reduction current. On adding 100 μM H2O2, a threefold enhancement in the reduction current was observed. A wide concentration of H2O2 ranging from 100 μM to 2 mM was used for the study. A plot of reduction current vs. concentration is shown as inset. A linear variation of reduction current with increasing concentration of H2O2 is noted. The onset potential of H2O2 reduction is determined to be −0.21 V, which is close to the formal potential of HRP enzyme, indicating the involvement of enzyme in the reduction process. It can be noticed from these CVs that, up to a concentration of below 1 mM, addition of H2O2 results in a systematic increase in the reduction current at similar potential values. Surprisingly, from 1 mM H2O2, the reduction potential is shifted towards a more negative value by ~200 mV. This shift in potential may arise due to the accessibility of HRP enzyme present within the pores of nanotubes and on the surface of TiO2 electrodes. Nevertheless, the reduction current increases for each and every addition. This significant increase in the reduction current suggests the bioelectrocatalytic activity of HRP enzyme, where a fast direct electron transfer between the electrode and heme groups of the enzyme occurs because of the favorable orientation of HRP enzyme. Similar experiments of H2O2 reduction without HRP enzyme are also performed and no significant increase in the reduction current was observed in that case. These results clearly indicate the facilitation of direct electron transfer of HRP using TiO2 electrodes. It can be observed from the inset of Fig. 8 that each incremental addition of H2O2 results in an increase of its reduction current, and a linear variation with respect to concentration change is noted. Based on these data, sensitivity and limit of detection are determined using the following formulae: SensitivityðσÞ ¼ √

X

—2

ðI− IÞ =n

Detection limit ¼ 3σ=slope of I vs: concentration plot

ð3Þ ð4Þ

where σ is sensitivity of the electrode, I is current measured due to addition of increasing I is the average current, and n is the number additions carried out for concentration of H2O2,‾ the experiment. Using these formulae, a sensitivity value of 2.864 mA/mM, detection limit of 5.5 μM, and a linear concentration range of 100 μM to 1.5 mM were determined for the bioelectrocatalytic reduction of H2O2 in the case of chemical modification method. Similarly for the case of physical entrapment method, a sensitivity of 0.1586 mA/mM and a detection limit of 0.52 mM were estimated. On comparing these values, ~20 times increase in sensitivity and 100-fold enhancement in the detection limit were observed for chemical methods when compared to simple physical adsorption employed for enzyme immobilization. The main reason for this behavior is enhancement in enzyme loading and its reaction with the aid of facilitated electron transfer through TiO2 and favorable orientation of HRP on these electrodes. These sensitivity and detection limit values are compared with some of the other values reported earlier for H2O2 sensor in the literature, and the comparison is shown in Table 1. It can be noted that the chemical functionalization method yielded the highest sensitivity and a wide linear range of detection for H2O2 sensor, while the limit of detection is almost comparable with that of the other reports. [24, 25, 38, 57–60] In fact, some of the sensors required a redox mediator for the detection purpose, which is not a critical factor in this proposed sensor. Even the sensitivity values obtained for the physical entrapment method in the present case are much better than the few other enzymatic H2O2 sensors reported, although

Appl Biochem Biotechnol Table 1 Comparison of sensitivity, linear concentration range, and detection limit for the proposed H2O2 sensor based on HRP-immobilized nanoporous TiO2 electrodes with the other electrode materials reported for enzymatic H2O2 detection Electrode

Mediator

Linear concentration Sensitivity range

GCE/Au NP/HRP Methylene blue/MWCNT/HRP

Hydroquinone 6.1 μM–1.8 mM Methylene blue 4.0 μM–3.8 mM

TiO2 nanotube/chitosan/HRP Methylene blue 5.0 μM–40.0 mM

0.25 mA/mM 22.5 μA/mM

Detection Reference limit 6.1 μM 1.0 μM

[38] [57]

0.542 mA/mM

2.0 μM

[58]

GCE/PAMAM/HRP

Hydroquinone

3.1 μM–2.0 mM

0.36 mA/mM

0.8 μM

[59]

ITO/APTMS/Au NPs/HRP ITO/AuNPs/HRP

No mediator No mediator

20 μM–8.0 μM 8.0 μM–3.0 mM

– –

8.0 μM 2.0 μM

[24] [25]

GCE/PANI/HRP

No mediator

1.0 μM–2.0 mM

–

0.6 μM

[60]

TiO2/APTMS/HRP (chemical method)

No mediator

100 μM–1.5 mM

2.864 mA/mM

5.5 μM

This work

TiO2/HRP (physical adsorption)

No mediator

100 μM–1.0 mM

0.1586 mA/mM 520 μM

This work

the linear concentration range and detection limit values are not impressive. [24, 25, 57, 60] Nevertheless, successful fabrication of H2O2 biosensor based on HRP-immobilized TiO2 nanoporous electrode is demonstrated. Among the two methods explored for enzymatic immobilization, the chemical functionalization provided a better H2O2 sensor with a higher sensitivity, lower detection limit, and a wide linear concentration range when compared to the physical entrapment method.

Conclusions Hexagonally ordered, honeycomb-like arrangement of porous, nanotubular structures of TiO2 are prepared easily by simple electrochemical anodization in ethylene glycol containing ammonium fluoride as an additive at a fixed voltage for a fixed duration. Structural and morphological analyses of such TiO2 electrodes are carried out using SEM, AFM, and XRD studies. Interestingly, the possibility of employing such a highly nanoporous, tubular, semiconductive TiO2 electrode for enzyme immobilization and subsequent use as an electrochemical biosensor is explored. Basically, two different strategies, viz., physical entrapment and covalent linking through chemical functionalization, are adopted for HRP immobilization on TiO2 electrodes. Microscopic (SEM and AFM), spectroscopic (UV–vis in diffuse reflectance mode) and electrochemical techniques are utilized for the characterization and analysis of those HRP enzyme-modified electrodes, and their application for H2O2 biosensor is demonstrated by monitoring its reduction process using CV. Several parameters such as stability, enzyme loading, linear concentration range, sensitivity, etc., are determined from those studies. Overall, a couple of significant points are highlighted in the present work: (a) Development of electrochemical biosensor using nanoporous, tubular TiO2 semiconductor electrodes by tuning the fundamental aspects of electron transfer across this semiconductor interface and (b) wide wavelength region absorption of light (from 300 to 800 nm), almost covering the entire visible region, by these nanoporous TiO2 electrodes suggest the possibility of creating optical (bio)sensors for biomolecules and species that are active in this region using these TiO2 electrodes. Finally, here we demonstrate HRP as an example for monitoring enzymatic

Appl Biochem Biotechnol

reactions, and a similar strategy can easily be extended to other important biomolecules such as proteins, DNA, RNA, etc., using these nanoporous TiO2 electrodes. Acknowledgments The authors acknowledge the funding from the Department of Science and Technology (DST), India, through Fast Track Scheme for Young Scientists with project number GAP 16/10 for carrying out this research work. Central Instrumentation Facility (CIF) of CSIR–CECRI, Karaikudi is also acknowledged for providing necessary characterization facilities.

References 1. Srinivasan, S., & Chizmadzhev, Y. A. (1985). Comprehensive treatise of electrochemistry, 10. New York: Plenum. 2. Armstrong, F. A., Heering, H. A., & Hirst, J. (1997). Chemical Society Reviews, 26, 169–179. 3. Song, S., Clark, R. A., Bowden, E. F., & Tarlov, M. J. (1993). Journal of Physical Chemistry, 97, 6564– 6572. 4. Rusling, J. F. (1998). Accounts of Chemical Research, 31, 363–369. 5. Boussaad, S., & Tao, N. J. (1999). Journal of the American Chemical Society, 121, 4510–4515. 6. Willit, J. L., & Bowden, E. F. (1987). Journal of Electroanalytical Chemistry, 221, 265–274. 7. Huang, H., Hu, N., Zeng, Y., & Zhou, G. (2002). Analytical Biochemistry, 308, 141–151. 8. Liu, H., & Hu, N. (2003). Analytica Chimica Acta, 481, 91–99. 9. Qiao, Z., & Dong, S. (1993). Chinese University Chemistry, 14, 1377–1379. 10. Zhao, Q., Gan, Z., & Zhuang, Q. (2002). Electroanalysis, 14, 1609–1613. 11. Ahammad, A. J. S., Sarker, S., Rahman, M. A., & Lee, J. J. (2010). Electroanalysis, 22, 694–700. 12. Ahammad, A. J. S., Rahman, M. A., Xu, G. R., Kim, S., & Lee, J. J. (2011). Electrochimica Acta, 56, 5266– 5271. 13. Ahamad, A. J. S., Choi, Y. H., Koh, K., Kim, J. H., Lee, J. J., & Lee, M. (2011). Journal of the Electrochemical Society, 6, 1906–1916. 14. Xuan, J., Jiang, L. P., & Zhu, J. J. (2010). Chinese Journal of Analytical Chemistry, 38, 513–516. 15. Cuendet, P., & Gratzel, M. (1986). Bioelectrochemistry and Bioenergetics, 16, 125–133. 16. Wang, B., Zhang, J. J., Pan, Z. Y., Tao, X. Q., & Wang, H. S. (2009). Biosensors and Bioelectronics, 24, 1141–1145. 17. Dave, B. C., Dunn, B., Valentine, J. S., & Zink, J. I. (1994). Analytical Chemistry, 66, 1120–1127. 18. Bowden, E. F., Hawkridge, F. M., & Blount, H. N. (1980). Bioelectrochemistry and Bioenergetics, 7, 447– 457. 19. Ru, J., Du, J., He, H. X., & Lu, X. Q. (2013). Chinese Journal of Analytical Chemistry, 41, 1249–1253. 20. Chaplin, M. F., & Bucke, C. (1990). Enzyme technology. Cambridge: Cambridge University Press. 21. Armstrong, F. A., & Wilson, G. S. (2000). Electrochimica Acta, 45, 2623–2645. 22. Liu, S. Q., & Ju, H. X. (2002). Analytical Biochemistry, 307, 110–116. 23. Qian, L., & Yang, X. (2006). Talanta, 68, 721–727. 24. Wang, L., & Wang, E. (2004). Electrochemistry Communications, 6, 225–229. 25. Wang, J., Wang, L., Di, J., & Tu, Y. (2009). Talanta, 77, 1454–1459. 26. Hiroyuki, O. (1998). Electrochimica Acta, 43, 1581–1587. 27. Dunford, H. B. (1991). In J. Eerse, K. E. Everse, & M. B. Grisham (Eds.), Peroxidases in chemistry and biology (p. 1). Boca Raton: CRC. 28. Creighton, T. E. (1984). Proteins—structure and molecular properties. New York: Freeman. 29. Lehninger, A. I., Nelson, D. L., & Cox, M. M. (1993). Principles of biochemistry (2nd ed.). New York: Worth. 30. Ruzgas, T., Csoregi, E., Emnes, J., Gorton, L., & Marko-Varga, G. (1996). Analytica Chimica Acta, 330, 123–138. 31. Ekanayake, E. M. I. M., Preethichandra, D. G. M., & Kaneto, K. (2008). Sensors and Actuators B Chemical, 132, 166–171. 32. Che, X., Yuan, R., Chai, Y., Ma, L., Li, W., & Li, J. (2009). Microchimica Acta, 167, 159–165. 33. Huang, J. L., & Tsai, Y. C. (2009). Sensors and Actuators B Chemical, 140, 267–272. 34. Tripathi, V. S., Kandimalla, V. B., & Ju, H. (2006). Biosensors and Bioelectronics, 21, 1529–1535. 35. Zhou, K., Zhu, Y., Yang, X., Luo, J., Li, C., & Luan, S. (2010). Electrochimica Acta, 55, 3055–3060. 36. Zhou, K., Zhu, Y., Yang, X., & Li, C. (2011). Electroanalysis, 23, 862–869.

Appl Biochem Biotechnol 37. Zhiguo, G., Shuping, Y., Zaijun, L., Xiulan, S., Guangli, W., Yinjun, F., et al. (2011). Analytica Chimica Acta, 701, 75–80. 38. Lei, C. X., Wang, H., Shen, G. L., & Yu, R. Q. (2004). Electroanalysis, 16, 736–740. 39. Zhong, H., Yuan, R., Chai, Y., Li, W., Zhang, Y., & Chengyan, W. (2011). Bioprocess and Biosystems Engineering, 34, 923–930. 40. Curulli, A., Cusma, A., Kaciulis, S., Padeletti, G., Pandolfi, L., Valentini, F., et al. (2006). Surface and Interface Analysis, 38, 478–481. 41. Si, P., Ding, S., Yuan, J., Lou, X. W., & Kim, D. H. (2011). ACS Nano, 5, 7617–7626. 42. Xu, M., Da, P., Wu, H., Zhao, D., & Zheng, G. (2012). Nano Letters, 12, 1503–1508. 43. Tang, J., Wang, Y., Li, J., Da, P., Geng, J., & Zheng, G. (2014). Journal of Materials Chemistry A, 2, 6153– 6157. 44. Paulose, M., Prakasam, H. E., Varghese, O. K., Peng, L., Popat, K. C., Mor, G. K., et al. (2007). Journal of Physical Chemistry C, 111, 14992–14997. 45. Yoriya, S., & Grimes, C. A. (2011). Journal of Materials Chemistry, 21, 102–108. 46. Hanzu, I., Djenizian, T., & Knauth, P. (2011). Journal of Physical Chemistry C, 115, 5989–5996. 47. Baram, N., & Ein-Eli, Y. (2010). Journal of Physical Chemistry C, 114, 9781–9790. 48. Murphy, A. B. (2007). Solar Energy Mater. Solar Cells, 91, 1326–1337. 49. Anderson, C., & Bard, A. J. (1997). Journal of Physical Chemistry B, 101, 2611–2616. 50. Yu, J., & Yu, X. (2008). Environmental Science and Technology, 42, 4902–4907. 51. Yi, X., Xian, J. H., & Yuan, C. H. (2000). Analytical Biochemistry, 278, 22–28. 52. Ferri, T., Poscia, A., & Santucci, R. (1998). Bioelectrochemistry and Bioenergetics, 45, 221–226. 53. Frew, J. E., Harmer, M. A., Hill, H. A. O., & Libor, S. I. (1986). Journal of Electroanalytical Chemistry, 201, 1–10. 54. Gu, B. X., Xu, C. X., Zhu, G. P., Liu, S. Q., Chen, L. Y., Wang, M. L., et al. (2009). Journal of Physical Chemistry B, 113, 6553–6557. 55. Dimcheva, N., & Horozova, E. (2005). Analytical and Bioanalytical Chemistry, 382, 1374–1379. 56. Polsky, R., Harper, J. C., Dirk, S. M., Arango, D. C., Wheeler, D. R., & Brozik, S. M. (2007). Langmuir, 23, 364–366. 57. Zhang, Y., Liu, L., Xi, F., Wu, T., & Lin, X. (2010). Electroanalysis, 22, 277–285. 58. Kafi, A. K. M., Wu, G., & Chen, A. (2008). Biosensors and Bioelectronics, 24, 566–571. 59. Zeng, Y. L., Huang, H. W., Jiang, J. H., Tian, M. N., & Li, C. X. (2007). Analytica Chimica Acta, 604, 170–176. 60. Xua, Q., Zhua, J. J., & Hu, X. Y. (2007). Analytica Chimica Acta, 597, 151–156.