A polyaniline wrapped aminated graphene composite on nickel foam as three-dimensional electrodes for enzymatic microfuel cells

Abstract

Three dimensional flow-through electrodes made of Ni foam coated with aminated graphene and polyaniline composites are studied for enzymatic glucose based microfuel cells. The composites of aminated graphene and polyaniline were prepared by in situ aniline polymerization with aminated graphene in different ratios. The prepared composites were characterized using XRD, FTIR and SEM. Pristine aminated graphene, polyaniline and various ratios of both were dip-coated on Ni foam and tested as a three-dimensional electrode in enzymatic microfuel cells. The enzyme was physically adsorbed and covalently bonded using a cross-linker on the electrode surface and in some cases; the electrode surface was functionalized before covalent attachment of enzyme and tested for an anode half-cell. The electrochemical studies demonstrated that the composite performed better than the pristine and was more promising when it was functionalized and enzymes where covalently bonded to the electrode surface. The maximum power density was observed as 118 μW cm⁻² for the composite.

---

Introduction

Three dimensional flow-through electrodes are in high demand as these can substantially improve the performance of microfuel cells by utilizing maximum fuel.^1–3 Microfuel cells are based on laminar flow and can operate without any membrane, in a single microfluidic channel with appropriate separation of fuel and oxidant stream.^2–9 Being functional with a single channel, inter-diffusion is restricted at the center to an interfacial width with electrodes placed sufficiently far to prevent crossover effects at the inter-diffusion zone. In the recent times, the major challenge in microfuel cells remains their capacity of utilizing maximum fuel in a single pass.¹ Although various cross-sectional channel geometry of electrodes and different fuel cell configurations have been tried in past, but improvising electrode architecture is envisaged to be one of the active areas of research to achieve higher fuel utilization.^1,10 For example, in a vanadium redox system, a high aspect ratio allowed fuel consumption of ∼50% per single pass with a single and multiple inlets.^11,12

According to electrode architecture, microfuel cells are classified as flow-through design with three-dimensional porous electrodes, flow-over design with planar electrodes and microfuel cells with air-breathing cathode.^12–14 Among the different architectures, flow-through design is pursued using conducting porous structures that enabled the flow of fuel and oxidant stream through the electrode itself and hence improvising efficiency of microfuel cell.¹ In a recent work, power density was reported to increase by 72% by incorporation of porous electrodes in place of planar electrodes. This enhancement was attributed to the increase in active area and improved species transport characteristics through flow penetration within the porous electrode. Electrode material also plays an important role; its electrical conductivity and stability will influence efficiency of the cell.^15–17 Traditional materials that are used as electrodes are carbon based due to its low cost and large potential window. Some of the materials used are graphite rod, carbon paper, carbon cloth, graphite fiber brush, carbon felt and reticulated vitreous carbon (RVC).^1,16–19 Recently, carbon based nanomaterials are pursued as electrode material because of their unique properties such as small size, large surface area, high electrical conductivity, high chemical and thermal stability. Apart from carbon nanotubes, graphene has attracted considerable attention as electrode material. Graphene possesses fast charge carrier properties that are superior to that of carbon nanotubes.²⁰ There are several methods of preparing graphene; one is physical method wherein graphene sheets are mechanical exfoliated from the graphite block using scotch tape and others include chemical methods such as chemical vapor deposition, exfoliation by chemical routes and reduction of graphene oxide. Graphene prepared from chemical route is considered fascinating due to the presence of several functional groups on its surface that provides more opportunities for further modification and functionalization for electrochemical studies.

There are a few reports on use of carbon based materials as flow-through electrodes. Kjeang et al.¹ used the commercially available porous carbon sheets to demonstrate the role of porous electrode in improving performance of the microfuel cell. For a diffusion-free glucose oxidase anode, a micro-patterned carbon electrode by carbon-MEMS process was fabricated.²¹ Polyacrylonitrile based electrospun nanofibres followed by carbonization in inert atmosphere was tested for electrochemical activities of the vanadium redox reaction.²² Zhang et al.²³ used 3-D graphene network for the immobilization of dopamine and laccase for glucose/O₂ biofuel cell application. Apart from carbon based electrodes, metal foams are also studied as electrode material. For example, Lu et al.²⁴ reported on use of 3-D porous Ni foam for non-enzymatic glucose detection using electrochemical sensing method. Karimi et al.²⁵ reviewed the current state-of-the-art research in graphene based enzymatic bioelectrodes and found that graphene based electrodes that are reported are on planar electrode geometry. Devadas et al.²⁶ fabricated anode with electrochemically reduced graphene oxide – multi-walled carbon nanotubes for glucose/O₂ biofuel cell and achieved the maximum power density of 46 μW cm⁻² whereas with nanographene platelets based electrode, it was reported as 57.8 μW cm⁻² by Zheng et al.²⁷ Recently, flexible electrodes for a miniaturized glucose biofuel employing graphene-coated carbon fiber cloth as electrode was reported with a maximum power density as 34.3 μW cm⁻² at 0.43 V.²⁸ The anode of biofuel cell fabricated by Liu et al.²⁹ consisted of a gold electrode on that graphene–glucose oxidase was co-immobilized using silica sol–gel matrix. In another study, the maximum power density was 24.3 ± 4 μW cm⁻² for graphene based electrodes that is nearly two times greater than that of the SWCNTs biofuel cell.³⁰ Song et al.³¹ developed 3D graphene/enzyme network-based anode and generated a maximum power density of 136.3 μW cm⁻² at 0.59 V, which is about seven times higher than the maximum power density of bare 3D carbon micropillar array. There are several other reports on membraneless enzymatic microfuel cells based on glucose oxidase^32–35 but in all these cases, electrode geometry has been planar. All these reports aimed to improve the efficiency through a better attachment of the enzyme to the electrodes via covalent bonding. In these reports either pyrolyzed photoresist film made on Si³⁴ or carbon nanotubes on graphite was used as electrodes.^33,35 Beneyton et al.³⁵ were first to report on microfluidic fuel cell with enzymes covalently attached to the electrodes and they reported the maximum power density of 1.65 μW cm⁻² at 0.24 mV at a flow rate of approximately 30 mL min⁻¹.

The aim of this work is to demonstrate how the commercially available material can be tailored to function as flow-through electrodes. In this work, we report on use of metal Ni foam coated with carbon-based material such as conducting polymers and functionalized graphene to enhance the performance of electrodes in enzymatic fuel cell. In the present work, the focus is on development of 3-D porous flow-through electrodes with enzyme immobilized on the electrode surface. Ni foam will provide the necessary three-dimensional porous structure for flow-through mechanism. The use of amine functionalized graphene (NH₂-G) with a redox conducting polymer such as polyaniline (PAni) will circumvent the polymer conductivity limitations along with enhancing the stability of electrode. The presence of functionalized graphene will facilitate better immobilization of enzymes through covalent bonding onto electrode surfaces and provide good electrical contact between enzyme and electrode surfaces. In this study, metal foam supported with pristine NH₂-G, PAni and various ratio of NH₂-G/PAni composite as electrode material for glucose based enzymatic microfuel cell has been validated. Enzyme has been immobilized onto the electrode surface by adopting three different protocols namely physical adsorption, direct covalent attachment with and without functionalization of electrode and following which anode half-cell performance of each electrode has been studied. We have focused our work on covalently bonded enzyme as for physically adsorbed enzyme, the stability and the sensitivity of the electrode decay with time. To circumvent this issue, choice of an excellent electrode material that can favorably bind with enzyme through a strong covalent bond is important. Functionalized graphene possesses a number of reactive functional groups that can easily bind with the free –NH₂ terminals of enzymes to form a strong amide covalent linkage and thus resulting to a better stability and sensitivity.

Experimental

Materials

Natural graphite flakes (45 μm) were bought from Alfa Aesar. Sodium nitrate, potassium permanganate, concentrated sulphuric acid, potassium dichromate, aniline, potassium ferricyanide, ethylene glycol, phosphate-buffered saline (PBS) buffer tablet, and ammonia were purchased from Fisher Scientific. Glucose oxidase (GOx), ethylene diamine (EDA), poly(ethylene glycol)diglycidyl ether (PEGDGE), Nafion (5 wt% solution), fluorescein isothiocyanate (FITC) and GAGO20 kit were purchased from Sigma-Aldrich. Nickel foam was obtained from MTI Corporation.

Synthesis of graphene oxide (GO) and amine functionalized graphene (NH₂-G)

Modified Hummer's method was used to synthesize graphene oxide (GO).^36,37 Amine functionalized graphene was prepared by hydrothermal method. Briefly, 5 mg mL⁻¹ GO dispersion (in ethylene glycol) and 1 mL of liquor ammonia was mixed and then transferred in to 50 mL Teflon lined stainless steel hydrothermal reactor at 180 °C for 10 h.³⁸ After cooling, the solution was centrifuged, washed several times and dried.

Synthesis of polyaniline (PAni) and NH₂-G/PAni composites

Homogeneous composites of NH₂-G and PAni were synthesized by in situ aniline polymerization in a dispersion of NH₂-G in an acidic solution. Typically, 50 mg of NH₂-G was dispersed in 10 mL of DI water followed by 5 mL of concentrated sulphuric acid and required amount of aniline. 0.5 mL of 0.3 M potassium dichromate solution was added dropwise in the solution mixture. The polymerization started instantaneously as the solution color turned green. The precipitated graphene/polymer composite was collected by filtration and repetitively washed with ethanol and water. The weight ratio of aniline to NH₂-G was varied as 1 : 5, 1 : 10, and 1 : 20 and the resulting composites were named as NH₂-G/PAni-1 NH₂-G/PAni-2 NH₂-G/PAni-3 respectively. PAni was synthesized by the similar procedure without the addition of NH₂-G.

Fabrication of electrodes

The electrodes were tested for anode half-cell using glucose oxidase enzyme. PAni, NH₂-G and NH₂-G/PAni composites were dip-coated on Ni foam. Prior to coating, the solution was prepared by dispersing of 100 mg of material in 4 mL of isopropanol and 0.5 mL of Nafion solution. Before coating, Ni foam was cut into 12 mm diameter discs and was washed with 0.1 M hydrochloric acid followed by DI water and isopropanol. Further, it was dried at 100 °C for 1 h in oven. For comparison, bare Ni foam without any coating was also tested.

Enzyme immobilization

Enzyme solution was prepared by dissolving GOx enzyme in PBS solution in a concentration of 0.5 mg mL⁻¹ (64 U mL⁻¹). Three different protocols were adopted for immobilization of enzyme onto electrode surface. In the first protocol, enzyme was physically adsorbed on the electrode surface. In this method, electrodes were incubated with 1 mL of GOx in PBS solution (pH 7.0) for 24 h for the adsorption of enzyme. Excess enzyme was removed by rinsing the electrode in PBS solution. In the second protocol, enzyme was covalently bonded to the electrode surface. As these electrodes are amine-terminated, 1 mL of PBS solution containing PEGDGE : GOx (2 : 1 w/w) was used for incubation.³⁵ After incubating for 24 h, electrodes were thoroughly rinsed with PBS solution. In the third protocol, electrodes were subjected to electro-oxidation in ethylene diamine (EDA) to increase the surface functionalization on the surface. For this purpose, EDA electro-oxidation was performed following a reported procedure.²¹ Briefly, the process was performed in anhydrous ethanol with 50 mM EDA and 100 mM LiClO₄ as supporting electrolyte. The scan was carried between 0.0 and 1.3 V at a scan rate of 10 mV s⁻¹ for 3 cycles. After electro-oxidation, electrodes were incubated in PEGDGE : GOx solution and rinsed thoroughly.

Enzyme doping and its stability

Immobilization of enzyme was analyzed by measuring the FITC-labelled glucose oxidase enzyme present on the electrodes. GOx (10 mg) was dispersed in PBS and was mixed with 1 mg mL⁻¹ of FITC solution and kept for stirring for 24 h for FITC labelling. The resulting solution was ultra-centrifuged in a PD-10 column and unconjugated FITC was removed by washing in excess PBS solution for multiple times. FITC–GOx was used for enzyme attachment on the electrodes. Absorption at 488 nm was monitored using UV-Visible spectrophotometer (Thermo Scientific Multiskan Spectrum) for the starting labeled enzyme solution and after immobilization, the resulting enzyme solution to measure the amount of FITC–GOx present in solution to estimate the enzyme loading on the electrode surface.

The stability and activity of the enzyme was determined by using a commercial spectrophotometric glucose assay kit. Briefly, all electrodes were dipped with 25 μg mL⁻¹ of glucose solution and assay reagent was added to analyze the change in glucose concentration for different electrodes. The color change observed on addition of assay was determined using a spectrophotometer. The amount of glucose oxidation caused by the enzyme present on the electrodes and amount of stable enzyme was calculated using a standard curve plotted for optical density versus enzyme concentration.

General characterization

The morphological characterization of pristine NH₂-G, PAni and NH₂-G/PAni composites were visualized by field emission scanning electron microscopy (FESEM, ZEISS Supra 40VP, Germany). X-ray diffraction (XRD) (PANalytical, Germany) was performed to determine the crystalline nature of the material. Raman spectra (WITec alpha300 R, Germany) were recorded to determine graphitic nature of these composites. Atomic force microscopy (AFM) was performed on Agilent PicoView 1.14.2, USA in non-contact mode. Fourier transform infrared (FTIR) spectroscopy was performed (Tensor 27, Bruker, Germany) to determine functional groups present on the electrode material. UV-Visible spectra were recorded using Varian Cary 50 bio spectrophotometer, USA. Chemical fluorescence and imaging was recorded on Olympus 1X81 fluorescence microscope.

Electrochemical characterization of electrodes

For anode half-cell, cyclic voltammetry was performed in a three electrode system between the voltage range of −0.2 V and +0.6 V, vs. Ag/AgCl reference electrode at room temperature for all electrodes after enzyme attachment. The testing was done in PBS solution with potassium ferricyanide as a redox mediator. The glucose solution was prepared at a concentration of 20 mM in PBS solution with redox mediator (10 mM Fe(CN)₆³⁻) and used after 24 h to test the anodic response of GOx immobilized electrodes. A platinum wire was used as the counter electrode. An Autolab potentiostat/galvanostat model 302N was used for data acquisition. Scan rate was varied from 10–50 mV s⁻¹. In cyclic voltammogram measurements, electrodes size were kept uniform and the measured current was normalized with respect to unit weight of composite material.

Results and discussion

Characterization of NH₂G-PAni composite

Polyaniline wrapped aminated graphene (NH₂-G/PAni) was synthesized by in situ polymerization of aniline on the surface of aminated graphene sheets in acidic condition.³⁸ Starting from graphene oxide aminated graphene was synthesized and further polymerized with aniline to obtain NH₂-G/PAni. Detailed characterization was performed using AFM, FTIR, Raman and XRD (Fig. S1, see ESI†). AFM image in Fig. S1a† showed the size variation in GO nanosheets. The lateral dimension of nanosheets was found to vary between 1 and 4 μm with uniform thickness in the range of 1 nm. The presence of functional group on GO, NH₂-G and NH₂-G/PAni composites was confirmed by FTIR spectra (Fig. S1b, see ESI†). Oxidation degree of GO nanosheets was 78%, this was estimated through the relative intensity of the FTIR bands related to oxygen containing groups using the quotient between the C–O related bands area to the total area under the spectra.³⁹ The peaks corresponding to amine groups were identified after functionalization and presence of PAni was also confirmed in composites. Further degree of amination for functionalized graphene was estimated to be 55.5% from all N–H related bands. Raman spectroscopy (Fig. S1c†) performed on these samples revealed intensity of G band is lower than D band in NH₂-G that confirmed increase in disordering of graphite with amine functionalization.⁴⁰ For NH₂-G/PAni composites, peak observed at 1590 cm⁻¹ were wider than that of NH₂-G, that confirmed successful incorporation of polyaniline in between graphene sheets.⁴¹ XRD peak observed at 2θ = 23.64° with interlayer spacing of ∼0.37 nm confirmed formation of NH₂-G (Fig. S1d†). There was a slight peak shift for NH₂-G/PAni composite that confirmed wrapping of PAni over graphene with interlayer spacing of ∼0.39 nm.³⁸

The morphological characterization of NH₂-G, PAni and NH₂-G/PAni composites were performed using FESEM as shown in Fig. 1. Sheet like structures of NH₂-G is clearly seen in Fig. 1a whereas Fig. 1b indicated dense irregular nanoparticles of PAni. In case of NH₂-G/PAni composite, after wrapping of PAni over NH₂-G sheets during polymerization process, NH₂-G sheets were roughened and decorated with PAni nanoparticles. Increase in roughness with increase in aniline concentration confirmed successful wrapping of PAni on the surface of graphene.³⁸ Fig. 1c showed FESEM image of NH₂-G/PAni-1 composite that showed presence of flake like structure of graphene as well as grainy texture of PAni. FESEM images of other composites, namely, NH₂-G/PAni-2 and NH₂-G/PAni-3 are provided in ESI as Fig. S2(a and b).†

Fig. 1 FESEM image of (a) NH₂-G (b) PAni and (c) NH₂-G/PAni-1 composite.

Fabrication of electrode for electrochemical testing

For the fabrication of electrode, Ni foam was used as a template, as it has a porous structure that will facilitate fluid flow and is ideal for flow-through electrodes in microfuel cells. Instead of using Ni foam directly, the foam was dip-coated with NH₂-G, PAni and NH₂-G/PAni composites to carry out electrochemical testing. Schematic of fabrication technique of the electrode is shown in Scheme 1, synthesis of composite material starting from functionalization of graphene oxide (GO) to dip-coating Ni foam with the composite material. We envisaged working with aminated graphene as it has functional group present that will facilitate enzyme attachment onto the electrode surface using the cross-linker molecule better than non-functionalized graphene. A composite with polyaniline, which is a conducting polymer, will help in binding of functionalized graphene onto the electrode surface without compromising on electrical conductivity.³⁸ Electrodes with coating of pristine NH₂-G, PAni and their composites were compared with bare Ni foam for electrochemical testing. In Fig. 2, FESEM images of Ni foam before and after coating with different materials are shown. The 3-D porous structure present in Ni foam in Fig. 2a is expected to facilitate fluid flow and also provide larger active surface area for the electrode in comparison to planar thin films. With pore sizes ranging between 200 and 300 μm, these are ideal for 3-D electrode media in microfuel cells. The coating of NH₂-G/PAni-1 composite on Ni foam is shown in FESEM image in Fig. 2b. The coating on electrode is uniform and is further confirmed by high magnification images shown as inset. After coating composite material, enzyme immobilization was carried out for electrochemical testing of electrodes. Generally, enzymatic fuel cells are fabricated in two different methods: first method wherein electrodes are bare and the enzyme flows along with fuel and oxidant in the solution and second wherein enzymes are immobilized on electrodes. The important issue to be addressed for enzyme based microfuel cells are short lifetime of enzymes and their poor electron transfer rate. Immobilization of enzymes onto the electrode surface effectively increases the lifetime of enzymes with some compromise with their activity than they being dissolved in solution.⁴² Our work is focused entirely on second method that involved immobilization of enzymes onto electrode surface and we have adopted three different protocols as detailed in Experimental section and results of each protocol are discussed below.

Scheme 1 Synthesis of composite material and fabrication of electrode.

Fig. 2 FESEM images of (a) bare and (b) NH₂-G/PAni-1 composite coated Ni foam with inset showing high magnification images (scale – 10 μm).

Enzyme immobilization by physical adsorption

Ni foam after thoroughly washed was electrochemically tested as electrode (data not shown), in absence and presence of glucose and no catalytic current was observed. Ni foam was dip-coated with different polymer and aminated graphene compositions following that GOx enzyme were physically adsorbed onto surface and catalytic activity of enzyme was monitored in absence and presence of glucose. The background current was measured by scanning the electrode cell without addition of glucose and then in presence of glucose, to the study of enzyme activity. The cyclic voltammogram was performed in the voltage range of −0.2 to +0.6 V vs. Ag/AgCl as reference electrode and platinum wire as counter electrode at 10 mV s⁻¹ and the results are shown in Fig. 3. After immobilization of electrodes with enzyme, an increase in anodic peak current was observed. In cyclic voltammogram for bare Ni foam adsorbed with GOx enzyme, Fig. 3a, black curve was obtained in absence of glucose, the peak at 0.26 V indicated oxidation of Ni foam. In presence of glucose (see red curve), a peak appeared at 0.42 V that is due to oxidation of glucose, in presence of GOx enzyme along with the peak corresponding to oxidation of Ni foam. Thus, the peak provided strong evidence for adsorption of enzyme that is catalytically active. Fig. 3b–d corresponded to the catalytic activity of enzyme adsorbed on NH₂-G, PAni and NH₂-G/PAni-1 composite coated Ni foam, respectively. Cyclic voltammogram measured in glucose solution showed increase in anodic current and confirmed effective immobilization of enzyme onto the electrode surface. The anodic current difference or relative anodic current in presence and absence of glucose, under similar testing conditions was found to increase by ∼15–18% after coating. The increase is found to be maximum (18%) for composite coating, indicating that combination of conducting polymer and functionalized graphene is better compared to pristine PAni or NH₂-G coated electrodes. Another point to note, in case of NH₂-G coating a higher anodic current was observed along with oxidation peak of Ni foam indicating the coating was not uniform. By introducing PAni, though there is some compromise with relative anodic current in comparison to NH₂-G coated electrodes but it seem to help in providing a uniform coating onto the electrode surface. Other ratios showed similar behavior (data not shown) as NH₂-G/PAni-1 and anodic current was in similar range hence 1 : 5 ratio was chosen for further studies. Lower ratios such as 1 : 2 showed more of NH₂-G characteristics with oxidation peak of Ni foam in cyclic voltammogram hence this ratio was also not considered.

Fig. 3 Cyclic voltammogram for GOx immobilized by adsorption on (a) bare (b) NH₂-G (c) PAni and (d) NH₂-G/PAni-1 composite coated Ni foam in absence (black) and presence (red) of glucose at 10 mV s⁻¹.

Enzyme immobilization by covalent attachment

In the second protocol, wherein enzyme was covalently bonded to the surface of electrode using a cross-linker molecule namely PEGDGE was studied. As the coating material were amine-terminated, enzyme immobilization was facilitated through a layer using a PEGDGE cross-linker, whose epoxide groups reacted with NH₂ groups.³⁵ In terms of attachment, covalently bonded enzymes onto electrodes are much better than physically adsorbed as later tends to get washed away in flow-through electrode setup and thus reducing overall efficiency of microfuel cell. Cyclic voltammogram performed on NH₂-G, PAni and NH₂-G/PAni-1 coated electrodes in absence and presence of glucose is shown in Fig. 4a–c and that of bare Ni foam is shown in Fig. S3a (see ESI†). The increase in anodic current was observed for all electrodes in glucose solution though slightly less (10–12%) then the first protocol wherein enzymes were physically adsorbed. This result is still encouraging as chances of enzyme detachment from electrode surface is circumvented by strong covalent bonding. This result is in line with our expectations, as in this case enzyme attachment is site specific and strong in comparison to physically adsorbed enzymes that are expected to get washed away after few cycles.

Fig. 4 Cyclic voltammogram for PEGDGE : GOx (1 : 2) immobilized on (a) NH₂-G (b) PAni and (c) NH₂-G/PAni-1 composite coated Ni foam in absence (black) and presence (red) of glucose at 10 mV s⁻¹.

Enzyme immobilization by covalent attachment on functionalized electrode

In the third protocol, the electrode surface was functionalized with amine and then cross-linker molecules were introduced for the covalent bonding of the enzyme onto the electrode surface. For this purpose, electrodes were first subjected to electro-oxidation by ethylene diamine (EDA) to increase the surface functionalization. EDA is an aliphatic and small flexible linker and thus can maximize the functionalization yield. The cyclic voltammogram for electro-oxidation of NH₂-G coated electrode is shown in Fig. 5. The scan was carried between 0.0 and 1.3 V (vs. Ag/AgCl non-aqueous reference electrode) at a scan rate of 10 mV s⁻¹ for 3 cycles on Ni foam (Fig. 5a). A single irreversible peak appeared due to the attachment and oxidation of amine group. Further, the drop in current was observed during the cycling from 1 to 3, demonstrated saturation of EDA molecules on electrode surface. Similar cycles were carried out for other compositions and the results are summarized in Fig. 5b. The functionalization process was found maximum for amine-terminated graphene, as lot of vacant sites were available on the surface whereas in case of PAni, sites were saturated. This trend continued with different composite ratios as well, compositions with higher content of amine-terminated graphene, showed more current, indicating more functionalization in comparison to other compositions. This validated NH₂-G/PAni-1 to be the optimized ratio as this ratio provided uniform coating on Ni foam as well as maximum functionalization.

Fig. 5 (a) Functionalization of NH₂-G electrode with ethylene diamine (EDA). The irreversible peak at 0.8 V corresponded to EDA amino-oxidation. Scan rate: 10 mV s⁻¹. (b) Peak current vs. ratio of PAni for EDA oxidation.

After functionalization, the electrode surface was immobilized with enzyme using the cross-linker molecule. The cyclic voltammogram performed on NH₂-G, PAni and NH₂-G/PAni-1 electrodes in absence and presence of glucose is shown in Fig. 6a–c and that of bare Ni foam is shown in Fig. S3b (see ESI†). As observed in previous protocols, there is an increase in relative anodic current in glucose solution for all electrodes by 20–25%. Overall, anodic current for each electrode is much higher than physically adsorbed or covalently bonded enzyme without functionalization (Fig. 3 and 4). This relative increase in anodic current is a good indication that EDA attachment on the surface has successfully helped in improving enzyme attachment onto the electrode surface. Thus, this protocol will serve fabrication of electrodes with increased efficiency for microfuel cell in a flow-through setup in comparison to other protocols followed.

Fig. 6 Cyclic voltammogram for PEGDGE : GOx (1 : 2) immobilized on amine functionalized electrodes of (a) NH₂-G (b) PAni and (c) NH₂-G/PAni-1 composite coated Ni foam in absence (black) and presence (red) of glucose at 10 mV s⁻¹.

Given that covalently bonded enzymes on electrode surface are better than physically adsorbed; now we compare results obtained for enzyme immobilized by covalently bonded protocols. These electrodes were further studied using cyclic voltammogram at various scan rates as shown in Fig. 7. The increase in peak current with increasing scan rate for directly immobilized enzyme on electrode surface and after amine functionalization are shown in Fig. 7a and b. A plot of the anodic peak current with different scan rate is shown in Fig. 7c. A linear trend revealed non-diffusive redox performance and confirmed the peaks in cyclic voltammogram are because of enzyme immobilization onto the electrode. This linear trend also indicated the stability of the enzymes attached to the electrode surface and is sign of successful immobilization of enzymes which is key for any good electrode fabrication method. The kinetics rate of electron transfer in a redox reaction process that occurred on the electrode surface, which was calculated by using Laviron's analysis.⁴³ The relationship between ΔE_p and log of the potential scan rate was used to determine the electron-transfer rate constant (k). The ΔE_p was defined as E_p − E°, where E_p is the anodic peak potential value and E° is the average potential taken of the anodic (E_a) and cathodic (E_c) peak. From this correlation, the plot between (ΔE_p) and the log of the potential scan rate showed a linear trend as shown in Fig. 7d. This behavior clearly indicated diffusionless redox system. The k values were determined from Fig. 7d to be 0.0309 s⁻¹ and 0.0316 s⁻¹, for direct and amine-functionalized electrode surface, respectively. The values are almost similar indicating that both protocols are quite efficient except that amine functionalized electrode surface showed higher peak current for same amount to composite material. Thus proving EDA is behaving like a linker molecule and is instrumental in providing more functional groups in order to enable their interaction with enzymes.

Fig. 7 Cyclic voltammogram for PEGDGE : GOx (1 : 2) immobilized on NH₂-G/PAni-1 composite coated Ni foam (a) directly and (b) after amine-functionalized electrodes in glucose at different scan rate 10–50 mV s⁻¹ (black to green color). (c) Anodic peak current vs. scan rate. (d) Plot of peak separation vs. log of scan rate used for calculation of the electron transfer rate (k) using Laviron's method.

Since NH₂-G/PAni-1 composite, has shown promising results, we have estimated the power density for this composite and compared with the literature. For the first protocol wherein enzymes are physically adsorbed onto the electrode surface, the maximum power density is found to be 37 μW cm⁻² and that for covalently bonded enzymes was 29 μW cm⁻². For the functionalized electrode with covalently bonded enzymes, the value was found to be maximum as 118 μW cm⁻². This values are quite impressive and are comparable with the reported value from the literature.³¹ Compared to other reports, another point to appreciate is that our electrode fabrication protocol is simple and does not require microlithography techniques as employed by Song et al.³¹

Doping and stability of GOx enzyme on electrode surface

So far we have discussed about electrodes coated with different compositions and their response to the three different protocols adopted. However, higher enzyme loading is also critical for higher efficiency of microfuel cell. For an efficient operation of an enzymatic microfuel cell, the enzyme should have high catalytic activity as well as stability. It is to be noted that while immobilization of enzyme and performing the cyclic voltammogram, enzyme present on the electrodes should retain their stability and functional properties. We have monitored enzyme attachment onto the electrode surface using a fluorescent maker namely FITC and details of which are provided in Fig. S4 and Table S1 in ESI.† From the absorption measurements shown in Fig. S4,† it is observed that the enzyme attachment is very poor onto the amine functionalized Ni foam electrode as more than 85% enzyme has been discarded in solution. However, in the case of amine functionalized NH₂-G/PAni-1, almost 68% enzymes (43.52 U cm⁻²) are retained onto the electrode surface and they are intact after subsequent washes. Table S1† detailed of the amount of enzyme present on the electrode surface of Ni and NH₂-G/PAni-1 for the three protocols adopted. The results clearly indicate that enzyme attachment is better on the composite material and best results are obtained for amine functionalized NH₂-G/PAni-1. After physical adsorption, the enzyme found on electrode surface is high, however they get washed away during subsequent washes but in case of covalently bonded enzymes, they remain intact after many washes. This is an important observation as this would lead to degradation in performance of physically adsorbed electrodes. This aspect is very crucial in our study as we aim to fabricate flow-through electrodes wherein fuel and oxidant stream will be in circulation during the microfuel cell operation.

We have performed fluorescence imaging on the electrodes after using FITC labelled GOx enzyme for visualization of immobilization and the results of same are shown in Fig. S5 (see ESI†). Essentially, FITC labeled GOx enzymes when attached to the electrode surface show green fluorescence under fluorescent microscope. In our studies, it is noticed that signal is quite feeble from Ni, PAni and NH₂-G based electrodes (see Fig. S5a–c respectively, ESI†) and signal is better from NH₂-G/PAni-1 composite coated electrode (see Fig. S5d–f†). The best results are obtained for amine functionalized NH₂-G/PAni-1 coated electrode indicating PEGDGE cross-linker molecule is most effective in immobilizing the enzyme onto the electrode surface. It should be noted that PEGDGE as cross-linker is known to be a mild method of enzyme immobilization and preserves the enzyme-substrate specificity which supports our results showing better performance in cyclic voltammetry studies.⁴⁴ Stability of electrode was studied by performing cyclic voltammogram on the electrode after enzyme immobilization using cross-linker molecule after amine functionalization. Initially, cyclic voltammogram on the electrode was recorded and the electrode was stored for six weeks at 4 °C in PBS solution and then again cyclic voltammogram was performed. The change in current was less than 9% from the initial run. Cyclic voltammogram data is provided in ESI as Fig. S6.† These results indicate that enzymes were quite stable and enzyme activity losses due to storage and several runs were quite small and negligible.

These results clearly demonstrated NH₂-G/PAni coated Ni foam as promising material for electrode material for enzymatic microfuel cells and since enzyme is covalently attached to the electrode surface which provided more stability, durability and high selectivity in comparison on non-enzymatic electrodes too and is amenable for mass production. This work opens up to the possibility to integrate Ni foam as electrodes in microsystems and provides new platform for enzyme based sensing applications.

Conclusion

We have successfully demonstrated fabrication of a novel 3-D porous electrode material for enzyme-based microfuel cells. For the first time, Ni foam has been tested as electrode material with composite coating of NH₂-G and PAni. Morphological characterization revealed that the coating is quite uniform for all composites. However, cyclic voltammogram studies showed performance of NH₂-G/PAni composite was better than the pristine NH₂-G and PAni by providing a uniform coating over the electrode surface and without much compromise on the relative anodic current as compared to pristine. An efficient protocol has also been developed to fabricate flow-through electrodes for glucose based enzymatic microfuel cells. Anode half-cell studies demonstrated that physically adsorbed enzymes on electrode surface were as promising as covalently bonded enzyme. But by further functionalizing the electrode surface with amine groups and covalently bonding the enzyme using cross-linker molecule, the relative anodic current was found to increase by 20–25%. Future work will focus on fabrication of microfuel cells using these electrodes. Following the third protocol, the maximum power density was observed as 118 μW cm⁻² for the NH₂-G/PAni composite which is at par with reported values.

Acknowledgements

T. B. thanks the Department of Science & Technology (DST), Government of India, for SERB-Start-up Research Grant (Young Scientist) (SB/FT/CS-060/2012) and DST-INSPIRE Fellowship (DST/INSPIRE/04/2014/002251) for the financial support. This work was supported by Centre of Nanosciences, Indian Institute of Technology, Kanpur.

References

1. E. Kjeang, R. Michel, D. A. Harrington, N. Djilali and D. Sinton, J. Am. Chem. Soc., 2008, 130, 4000–4006.
2. R. Ferrigno, A. D. Stroock, T. D. Clark, M. Mayer and G. M. Whitesides, J. Am. Chem. Soc., 2002, 124, 12930–12931.
3. J. Phirani and S. Basu, J. Power Sources, 2008, 175, 261–265.
4. S. Tominaka, S. Ohta, H. Obata, T. Momma and T. Osaka, J. Am. Chem. Soc., 2008, 130, 10456–10457.
5. R. F. Ismagilov, A. D. Stroock, P. J. A. Kenis, G. Whitesides and H. A. Stone, Appl. Phys. Lett., 2000, 76, 2376–2378.
6. E. Kjeang, N. Djilali and D. Sinton, J. Power Sources, 2009, 186, 353–369.
7. S. A. Mousavi Shaegh, N.-T. Nguyen and S. H. Chan, Int. J. Hydrogen Energy, 2011, 36, 5675–5694.
8. E. R. Choban, L. J. Markoski, A. Wieckowski and P. J. A. Kenis, J. Power Sources, 2004, 128, 54–60.
9. E. R. Choban, J. S. Spendelow, L. Gancs, A. Wieckowski and P. J. A. Kenis, Electrochim. Acta, 2005, 50, 5390–5398.
10. J. L. Cohen, D. A. Westly, A. Pechenik and H. c. D. Abruna, J. Power Sources, 2005, 139, 96–105.
11. E. Kjeang, J. McKechnie, D. Sinton and N. Djilali, J. Power Sources, 2007, 168, 379–390.
12. M.-A. Goulet and E. Kjeang, J. Power Sources, 2014, 260, 186–196.
13. R. S. Jayashree, L. Gancs, E. R. Choban, A. Primak, D. Natarajan, L. J. Markoski and P. J. A. Kenis, J. Am. Chem. Soc., 2005, 127, 16758–16759.
14. R. A. Rincon, C. Lau, H. R. Luckarift, K. E. Garcia, E. Adkins, G. R. Johnson and P. Atanassov, Biosens. Bioelectron., 2011, 27, 132–136.
15. M. Zhou, M. Chi, J. Luo, H. He and T. Jin, J. Power Sources, 2011, 196, 4427–4435.
16. B. E. Logan, B. Hamelers, R. Rozendal, U. Schroder, J. Keller, S. Freguia, P. Aelterman, W. Verstraete and K. Rabaey, Environ. Sci. Technol., 2006, 40, 5181–5192.
17. J. Wei, P. Liang and X. Huang, Bioresour. Technol., 2011, 102, 9335–9344.
18. B. Logan, S. Cheng, V. Watson and G. Estadt, Environ. Sci. Technol., 2007, 41, 3341–3346.
19. S. Gianmario, K. Petri, K. Tanja and F. Sami, J. Micromech. Microeng., 2012, 22, 094006.
20. P. T. Yin, T.-H. Kim, J.-W. Choi and K.-B. Lee, Phys. Chem. Chem. Phys., 2013, 15, 12785–12799.
21. G. S. Bisht, S. Holmberg, L. Kulinsky and M. Madou, Langmuir, 2012, 28, 14055–14064.
22. G. Wei, J. Liu, H. Zhao and C. Yan, J. Power Sources, 2013, 241, 709–717.
23. Y. Zhang, M. Chu, L. Yang, Y. Tan, W. Deng, M. Ma, X. Su and Q. Xie, ACS Appl. Mater. Interfaces, 2014, 6, 12808–12814.
24. W. Lu, X. Qin, A. M. Asiri, A. O. Al-Youbi and X. Sun, Analyst, 2013, 138, 417–420.
25. A. Karimi, A. Othman, A. Uzunoglu, L. Stanciu and S. Andreescu, Nanoscale, 2015, 7, 6909–6923.
26. B. Devadas, V. Mani and S.-M. Chen, Int. J. Electrochem. Sci., 2012, 7, 8064–8075.
27. W. Zheng, H. Y. Zhao, J. X. Zhang, H. M. Zhou, X. X. Xu, Y. F. Zheng, Y. B. Wang, Y. Cheng and B. Z. Jang, Electrochem. Commun., 2010, 12, 869–871.
28. H. Kazuki, M. Kazuo, S. Hisato and N. Yasushiro, Jpn. J. Appl. Phys., 2016, 55, 02BE05.
29. C. Liu, Z. Chen and C. Z. Li, IEEE Trans. Nanotechnol., 2011, 10, 59–62.
30. C. Liu, S. Alwarappan, Z. Chen, X. Kong and C.-Z. Li, Biosens. Bioelectron., 2010, 25, 1829–1833.
31. Y. Song, C. Chen and C. Wang, Nanoscale, 2015, 7, 7084–7090.
32. A. Zebda, L. Renaud, M. Cretin, C. Innocent, R. Ferrigno and S. Tingry, Sens. Actuators, B, 2010, 149, 44–50.
33. D. MacAodha, P. Ó. Conghaile, B. Egan, P. Kavanagh and D. Leech, ChemPhysChem, 2013, 14, 2302–2307.
34. M. J. Gonzalez-Guerrero, J. P. Esquivel, D. Sanchez-Molas, P. Godignon, F. X. Munoz, F. J. del Campo, F. Giroud, S. D. Minteer and N. Sabate, Lab Chip, 2013, 13, 2972–2979.
35. T. Beneyton, I. P. M. Wijaya, C. B. Salem, A. D. Griffiths and V. Taly, Chem. Commun., 2013, 49, 1094–1096.
36. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
37. R. Kumar, K. Jahan, R. K. Nagarale and A. Sharma, Ind. Eng. Chem. Res., 2015, 54, 10183–10189.
38. R. Kumar, K. Jahan, R. K. Nagarale and A. Sharma, ACS Appl. Mater. Interfaces, 2015, 7, 593–601.
39. J. Guerrero-Contreras and F. Caballero-Briones, Mater. Chem. Phys., 2015, 153, 209–220.
40. O. C. Compton, D. A. Dikin, K. W. Putz, L. C. Brinson and S. T. Nguyen, Adv. Mater., 2010, 22, 892–896.
41. W. Sun, T. Peng, Y. Liu, S. Xu, J. Yuan, S. Guo and X.-Z. Zhao, J. Mater. Chem. A, 2013, 1, 2762–2768.
42. A. Zebda, C. Gondran, A. Le Goff, M. Holzinger, P. Cinquin and S. Cosnier, Nat. Commun., 2011, 2, 370.
43. E. Laviron, J. Electroanal. Chem. Interfacial Electrochem., 1979, 101, 19–28.
44. N. Vasylieva, C. Maucler, A. Meiller, H. Viscogliosi, T. Lieutaud, D. Barbier and S. Marinesco, Anal. Chem., 2013, 85, 2507–2515.

---

Footnotes

† Electronic supplementary information (ESI) available: AFM, FTIR, Raman, XRD, FESEM image of composites, CV on Ni foam under different protocols and doping and stability of GOx enzyme on electrode surface. See DOI: 10.1039/c6ra08195a

‡ Contributed equally.

§ Major part of work was performed when the author was part of Department of Chemical Engineering, Indian Institute of Technology, Kanpur 208016, India.

---