A graphene modified biocathode for enhancing hydrogen production

Abstract

Biocathodes have shown great promise for developing low-cost cathodes for hydrogen production in microbial electrolysis cells (MEC). To promote the performance of hydrogen production with a biocathode, we constructed a graphene modified biocathode and assessed the performance of the modified biocathode by setting different cathode potentials. The results indicated that it was feasible to promote the current density, electron recovery efficiency (ERE) and hydrogen production rate by a modified biocathode using graphene. At −1.1 V (vs. Ag/AgCl), the hydrogen production rate of the graphene modified biocathode even achieved 2.49 ± 0.23 m³ per m³ per day with 89.12 ± 6.03% of ERE at a current density of 14.07 ± 0.06 A m⁻², which were about 2.83 times, 1.38 times and 2.06 times that of the unmodified biocathode, respectively. The hydrogen production performance of the graphene modified biocathode was close to that of the platinum catalyzed cathode and superior to that of the stainless steel mesh cathode at −1.1 V.

---

1. Introduction

Hydrogen is a fuel that produces water as the only combustion product with an amount of entropy released in the form of heat,¹ therefore, the production of hydrogen is a key technology related to our energy future. Among most of the technologies for hydrogen production, Microbial Electrolysis Cell (MEC) is a promising technology for sustainable production of hydrogen from biodegradable carbon sources.^2–4 So far, metal catalysts (i.e. platinum, nickel or stainless steel) were employed well in MEC due to their low overpotential and high catalytic efficiency,^5–8 but they are costly and easily corroded or deactivated in widespread use.

Using microorganisms as the cathode catalyst has been a novel alternative for hydrogen production since they are inexpensive, not susceptible to corrosion and self-renewable. Rozendal et al. developed the first biocathode for hydrogen production, and this biocathode produced 0.63 m³ per m³ per day (m³ H₂ m⁻³ cathode liquid volume per day) with cathodic hydrogen recovery efficiency of 49% at a cathode potential of −0.7 V vs. SHE.⁹ The hydrogen production rate even obtained 2.2 m³ per m³ per day when using acetate instead of bicarbonate with a cathodic hydrogen recovery efficiency of 50 ± 2.3%.¹⁰ When the applied voltage was 0.5 V, the hydrogen production rate achieved 0.24 m³ per m³ per day with 21% of cathodic hydrogen recovery efficiency in the MEC.¹¹ Although these studies have obtained considerable hydrogen production rates, the cathodic hydrogen recoveries of the biocatalyst were still lower than metal catalysts. Therefore, further efforts focusing on the improvement of the biocatalyst performance for hydrogen production were needed.

Recently, graphene has attracted tremendous attention in modifying bioanodes to improve the performance of Microbial Fuel Cells (MFCs), for its high surface area, excellent electronic conductivity and good biocompatibility.^12–15 The graphene modified stainless steel mesh bioanode of MFC exhibited high surface area and excellent biocompatibility, whose maximum power density was 18 times larger than that of unmodified MFC.¹³ The crumpled graphene modified carbon cloth bioanode produced the highest maximum power density of 3.6 W m⁻³, which was significantly higher than that of the activated carbon modified anode (1.7 W m⁻³).¹⁴ Assembled graphene oxide nanoribbons network on carbon paper anode facilitated the extracellular electron transfer (EET) process of bioanode.¹⁵ Although graphene has increased the performance of bioanodes in MFCs, there are no reports using graphene to enhance the performance of biocathodes in MEC for hydrogen production.

In this study, we attempt to facilitate the EET process from electrode to microorganisms for enhancing the hydrogen production. For this purpose, we fabricated an Electrochemically Reducing Graphene Oxide (ERGNO) modified biocathode and evaluated the performance of the graphene modified biocathode at different cathodic potentials. To minimize the possible effects of bioanode, the biocathodes were studied in bioelectrochemical cells with a chemical anode (hexacyanoferrate).

2. Materials and methods

2.1. Electrode fabrication

The method of electrode modified by ERGNO referred to Hou et al.¹⁶ Briefly, a carbon cloth (CC, 3.8 cm diameter disks, HCP331N, Hesen, China) was soaked in acetone for 4 h, then immersed in a 1 : 3 mixture of 70% nitric acid and 97% sulfuric acid aqueous solutions for 4 h, followed by rinsing in ultrapure water until its pH value equals to 7, finally dried under vacuum at 60 °C. GNO suspension (graphene oxide, 2 mg mL⁻¹, Institute of coal chemistry, Chinese academy of science) was dispersed by sonication for 3 h, then dripped onto the CC surface using layer by layer method. The GNO dripped CC was rinsed with the ultrapure water before subjected to air-drying at room temperature. The loading of the GNO was about 0.7 mg. The ERGNO/CC was finally obtained by cyclic voltammetry (−0.2 V to −1.8 V, 10 mV s⁻¹) in 100 mM PBS (pH = 7) in a three-electrode cell with titanium mesh counter electrode and Ag/AgCl reference electrode. All potentials are reported here versus Ag/AgCl reference electrode (218, Shanghai REX Instrument Factory, sat KCl, 197 mV vs. SHE). The ERGNO/CC was removed from the cell, rinsed with water and dried naturally in the air.

2.2. Bioelectrochemical cell setup

In this study, two-chamber bioelectrochemical cells were constructed. The anode chamber (25 mL) and cathode chamber (35 mL) were separated by a proton exchange membrane (PEM, Nafion 117, DuPont Co., USA). The anode electrode was a carbon fiber brush (2.5 cm in diameter and 3 cm long) made by twisting carbon fibers between two titanium wires. The cathode electrodes were ERGNO/CC and unmodified CC (3.8 cm diameter disks, projected cross sectional area of 7 cm²). Meanwhile, stainless steel woven mesh (SSM, 316L, mesh: 80) and carbon cloth with platinum catalyst (Pt/CC, 0.5 mg cm⁻², 40% Pt Hesen, China) were used as metal cathodes for comparison. An anaerobic glass tube (total headspace volume of 80 mL) was glued to the top of the cathode chamber for hydrogen collection. An external bottle (250 mL) filled with catholyte was connected to the cathode chamber to recycle the catholyte by a dosing pump.

2.3. Enrichment of electroactive mixed culture biofilm

To develop a biofilm on the cathode with good electrochemical activity, the bacteria of the cathode was inoculated from a bioanode of microbial fuel cell that had operated for a long period in our lab.¹⁷ 10 mL of bacteria suspension and 25 mL of medium were introduced into the cathode chamber. The medium contained sodium bicarbonate (0.84 gL⁻¹) and 100 mM phosphate buffer solution (PBS) containing (per liter): K₂HPO₄·12H₂O, 12.542 g; KH₂PO₄·2H₂O, 7.956 g; NH₄Cl, 0.31 g; KCl, 0.13 g; trace minerals (12.5 mL) and vitamins (5 mL).¹⁸ The minerals contains (per liter): nitrilotriacetic acid, 1.5 g; MgSO₄·7H₂O, 3.0 g; MnSO₄·2H₂O, 0.5 g; NaCl, 1.0 g; FeSO₄·7H₂O, 0.1 g; CaCl₂·2H₂O, 0.1 g; CoCl₂, 0.1 g; ZnSO₄, 0.10 g; CuSO₄·5H₂O, 0.01 g; AlK(SO₄)₂, 0.01 g; H₃BO₃, 0.01 g; Na₂MoO₄·2H₂O, 0.01 g and adjust pH to 7.0 with KOH. The vitamins contains (per liter): biotin, 0.2 g; folic acid, 0.2 g; pyridoxine hydrochloride, 1.0 g; thiamine hydrochloride, 0.5 g; riboflavin, 0.5 g; nicotinic acid, 0.5 g; vitamin B₁₂, 0.01 g; aminobenzoic acid, 0.5 g; lipoic acid, 0.5 g. The anode chamber was filled 25 mL of 100 mM K₄Fe(CN)₆ and 100 mM PBS. The cathode chamber was flushed with ultrapure N₂ (purity > 99.999%) for 15 min. The biofilm enrichment process consisted of: (a) to obtain a biocathode in a short time, 1 g L⁻¹ sucrose was added into cathode medium as an organic carbon source and electron donor for the enrichment process. Meanwhile, cathode electrode served as another electron donor for biocathode culture by setting the cathode potential at −1.0 V and operated for 4 batch runs. (b) At the start of the batch run 5, the sucrose was omitted and the cathode electrode served as a sole electron donor for microorganisms of cathode and operated 4 batch runs. Throughout this biofilm enrichment process, each batch run lasted 2 days and at the end of each batch run, 100% of the medium was replaced with fresh medium, the CC and ERGNO/CC were used as cathode electrode to enrich the microorganisms at the same condition, and the temperature was controlled at 35 °C.

2.4. Bioelectrochemical hydrogen evolution reaction experiments

The bioelectrochemical experiments were operated in three-electrode configuration. To evaluate the feasibility of biocathode of ERGNO/CC for enhancing hydrogen production, the bioelectrochemical cell was connected to the potentiostat and the cathode potential was set at −1.0 V. In parallel, biocathode without ERGNO modified, abiotic CC cathode and abiotic ERGNO/CC cathode as control tests were operated under the same operating conditions. At the next stage, the cathode potential was set in the range from −0.95 V to −1.1 V to test the effect of different potentials on the performance of the biocathode with ERGNO modified. The third stage, the metal catalystic cathodes including Pt/CC and SSM cathodes were tested for a comparison to the graphene modified biocathode at −1.1 V of cathode potential. All tests were operated in batch mode and each test lasted 24 h. The hydrogen gas was detected every 4 h. All batch tests were operated for at least three consecutive batch runs. The performance of the biocathode (modified and unmodified by graphene) was evaluated in terms of current density, hydrogen production rate and electron recovery efficiency of hydrogen.

2.5. Analysis and calculation

Electrochemical potentiostatic measurements and monitoring were performed with potentiostat (CHI 660 E, ChenHua Instruments Co., Ltd., Shanghai, China). Hydrogen gas was detected by gas chromatography using the same procedures as our previous work.¹⁹ The cathode and bacterial morphologies were observed using a scanning electron microscope (SEM; Quanta FEG 450 Oxford Instruments, Netherlands). The microbial biomass was determined by Phospholipid analysis.²⁰ Cyclic voltammetry (CV) and linear sweep voltammograms (LSV) were operated using an electrochemical working station (CHI 604E, ChenHua Instruments Co., Ltd., Shanghai, China), (CV: the potential range from 0.6 V to −1.2 V at a scan rate of 10 mV s⁻¹. LSV: the potential range from −0.6 V to −1.2 V at a scan rate of 10 mV s⁻¹). The current density (j, A m⁻²) was calculated from the measured current (I, A) by electrochemical potentiostatic measurements and cathode area (A_cat, m²). The formula was showed as follows:

The ERE for the formation of cathodic hydrogen was calculated as follows:

where f represents the molar conversion factor (2 eq. mol⁻¹ for hydrogen); n is the moles of hydrogen harvested; F is Faraday's constant (96 485 C mol⁻¹ of electrons); and I is the current over the period of electrode polarization.

3. Results and discussion

3.1. Electrochemical reduction of GNO/CC

The electrochemical reduction of GNO/CC was conducted by cyclic voltammetry (CV) in PBS (100 mM, pH = 7) at a potential range from −0.2 V to −1.8 V (Fig. 1). The GNO/CC showed a large cathodic current peak at around −1.8 V with a starting potential of −0.8 V in the first cycle. This reduction peak could be ascribed to the reduction of the surface-oxygenated species at GNO/CC, demonstrating that the GNO/CC could be reduced by electrochemistry to form ERGNO/CC, which was consistent with the previous reports.^16,21,22 However, the reduction current decrease considerably and disappeared in the second to the fourth cycle, which indicated that the reduction of GNO/CC by electrochemistry occurred quickly and irreversibly. Therefore, the ERGNO/CC was obtained by electrochemistry reduction.

Fig. 1 CV of GNO/CC in PBS (100 mM, pH = 7.0) at a scan rate of 10 mV s⁻¹. Arrow 1 indicates the first cycle of CV test; arrows 2, 3, 4 indicate the second, third and fourth cycles of CV test.

3.2. Feasibility of ERGNO modified biocathode for enhancing hydrogen production

To evaluate the feasibility of ERGNO/CC biocathode for improving hydrogen production, the cathode potential was first set at −1.0 V and the three control tests including ERGNO/CC, unmodified biocathode and CC cathode were conducted in separate cells at the same time. The current densities of one of batch assays were shown in Fig. 2A. The current densities were dramatically enhanced by modified CC and modified biocathode. In the abiotic control test, the current density of ERGNO/CC cathode was 1.42 times that of CC cathode (0.3 A m⁻² vs. 0.21 A m⁻²), which might be ascribed to an increase in specific surface area and conductivity as well as a decrease of overpotential. With the biocatalyst, the current density of ERGNO/CC biocathode was about 2.2 times that of the CC biocathode (4.8 A m⁻² vs. 2.2 A m⁻²), attributing to the synergy of graphene and biofilm. In addition, the hydrogen production rate and the ERE of ERGNO/CC biocathode were significantly higher than that of CC biocathode (Fig. 2B). After 24 h reaction, the hydrogen of ERGNO/CC biocathode accumulated 0.71 ± 0.03 m³ per m³ per day (m³ H₂ m⁻³ cathode liquid volume per day), which was over 3.3 times that of CC biocathode (0.21 ± 0.02 m³ per m³ per day). The ERE of ERGNO/CC biocathode also enhanced to 76.94 ± 2.65%, which was 1.63 times that of the CC biocathode (47.22 ± 2.64%). The abiotic cathode only produced little hydrogen (below 0.03 m³ per m³ per day) due to the lack of catalyst and the low current density. According to the above results, it is feasible to improve the EET process from solid electrode to microorganisms by using graphene modified carbon cloth to increase the hydrogen production.

Fig. 2 (A) The current densities at −1.0 V (vs. Ag/AgCl) in 24 h, (a) ERGNO/CC biocathode, (b) CC biocathode, (c) ERGNO/CC cathode, (d) CC cathode. (B) Hydrogen production rates and EREs at −1.0 V in 24 h.

3.3. Influence of different cathode potentials on hydrogen production

The influence of the cathode potential on hydrogen production was assessed by varying the cathode potentials in the range from −0.95 V to −1.1 V. All experiments were done by comparing the ERGNO/CC biocathode and the unmodified CC biocathode. The hydrogen production rates, current densities and EREs of biocathode with modified and unmodified all presented the same potential-dependent manners, which increased as the cathode potential became more negative (Fig. 3A and B). At −0.95 V, ERGNO/CC biocathode produced hydrogen at a rate of 0.25 ± 0.09 m³ per m³ per day with ERE of 64.05 ± 2.95% and a current density of 1.89 ± 0.13 A m⁻², which were 3.57 times, 1.83 times and 2.0 times that of CC biocathode, respectively. Moreover, at −1.1 V, ERGNO/CC biocathode produced hydrogen at a rate of 2.49 ± 0.23 m³ per m³ per day with ERE of 89.12 ± 6.03% and a current density of 14.07 ± 0.06 A m⁻², which were about 2.83 times, 1.38 times and 2.06 times of CC biocathode, respectively. Remarkably, the graphene can dramatically improve the performance of biocathode for hydrogen production, and the superiority of modified biocathode became more obvious at higher cathode potential.

Fig. 3 (A) The current densities of biocathode with and without graphene at different cathode potentials after 24 h. (B) Hydrogen production rates and EREs with and without graphene at different cathode potentials after 24 h.

3.4. Comparison with metal catalytic cathodes

Table 1 showed the results of different cathodes at −1.1 V of cathode potential. The performance of graphene modified biocathode outperformed the stainless steel mesh in terms of current densities (14.07 ± 0.06 A m⁻² vs. 10.99 ± 0.27 A m⁻²), hydrogen production rates (2.49 ± 0.23 vs. 1.55 ± 0.09 m³ per m³ per day), and EREs (89.12 ± 6.03% vs. 72.26 ± 2.04%). Moreover, the performance of graphene modified biocathode was close to those of Pt catalytic cathode (2.71 ± 0.04 m³ per m³ per day with ERE of 91.23 ± 1.69% and a current density of 15.31 ± 0.32 A m⁻²). These results demonstrate that biocatalyst can be an alternative method for metal catalysts after the cathode is modified by graphene.

Table 1 The comparison of different cathodes at −1.1 V of cathode potential

Cathode	Current densities (A m^−2)	Hydrogen production rates (m^3 per m^3 per day)	Electron recovery efficiencies (%)
Biocathode	6.83 ± 0.57	0.88 ± 0.13	64.51 ± 7.51
ERGNO/biocathode	14.07 ± 0.06	2.49 ± 0.23	89.12 ± 6.03
Stainless steel mesh	10.99 ± 0.27	1.55 ± 0.09	72.26 ± 2.04
Pt/CC	15.31 ± 0.32	2.71 ± 0.04	91.23 ± 1.69

---

3.5. Electrochemical and morphological characteristics of biocathode

The electrochemical catalytic activity of modified biocathode and unmodified biocathode were measured by CV (Fig. 4A) and LSV (Fig. 4B). The CV showed that the current of modified biocathode was higher than that of unmodified biocathode which could be attributed to the increase of faradic charge capacity by using graphene biocathode that has a higher electrode surface area,²³ indicating that the catalytic activity of ERGNO/CC biocathode outperformed the CC biocathode. The LSV (Fig. 4B) showed that the current of graphene modified biocathode increased significantly when the cathode potential was more negative than −0.9 V, while the current of unmodified biocathode increased significantly when the cathode potential was more negative than −0.95 V, which indicating the overpotential of biocatalyst was decreased by graphene modification. The theoretical cathode potential for hydrogen production was −0.62 V, therefore, the overpotential of modified biocathode was −0.28 V and the overpotential of unmodified biocathode was −0.33 V.

Fig. 4 (A) CV of biocathode with and without graphene. (B) LSV of biocathode with and without graphene. (C) SEM of CC biocathode without graphene. (D) SEM of biocathode with graphene.

After the batch tests, the electrochemical cells were disassembled, the samples of graphene modified biocathode and unmodified biocathode were investigated by SEM (Fig. 4C and D). The CC electrode surface was smooth (Fig. 4C), but the ERGNO/CC electrode possessed a thin wrinkled and crumpled surface structure (Fig. 4D), which was typical of ERGNO sheets, consisted with the previous results.^16,22 The bacteria grown on the electrodes exhibited a rod-shaped structure, but the number of bacteria on the modified electrode was not obviously increased as expected. The amount of microorganism with and without graphene modification were monitored by means of phospholipid analysis after batch tests, the biomass concentration of graphene unmodified biocathode was 0.13 μg P cm⁻², which was slight lower than graphene modified biocathode of 0.14 μg P cm⁻². The excellent catalytic activity of graphene modified biocathode might be attributed to the change of bacterial community cultured in cathode, which we would have an further study in the future.

4. Conclusions

In this present study, graphene was used to enhance the performance of biocathode for hydrogen production in MEC by modifying the biocathode, the current density, ERE and hydrogen production rate were significantly improved compared to the unmodified biocathode. It is believed that graphene modified biocathode is an appropriate solution to enhance the performance of biocathodes for hydrogen production, which enables it to be a viable alternative for metal catalytic cathodes in MEC.

5. Nomenclature

Acat	Cathode area (m^2)
CC	Carbon cloth
CC-biocathode	Carbon cloth with biofilm
EET	Extracellular electron transfer
ERE	Electron recovery efficiency (%)
ERGNO	Electrochemically reducing graphene oxide
ERGNO/CC	Carbon cloth with electrochemically reducing graphene oxide modification
ERGNO/CC-biocathode	Electrochemically reducing graphene oxide modified carbon cloth cathode with biofilm
f	Molar conversion factor (2 eq. mol^−1 for hydrogen)
F	Faraday's constant (96 485 C mol^−1 of electrons)
GNO	Graphene oxide
I	Current (A)
j	Current density (A m^−2)
MEC	Microbial electrolysis cell
MFC	Microbial fuel cell
n	The moles of hydrogen harvested (mol)
SHE	Standard hydrogen electrode

Acknowledgements

This work was supported by the National Natural Science Fund of China (no. 91127012 and no. 21403251) and the Chinese Academy of Sciences for financial support (no. KJCX2-YW-H21).

References

1. U. Eberle, B. Müller and R. von Helmolt, Energy Environ. Sci., 2012, 5, 8780.
2. H. Liu, S. Grot and B. E. Logan, Environ. Sci. Technol., 2005, 39, 4317–4320.
3. B. E. Logan, D. Call, S. Cheng, H. V. Hamelers, T. H. Sleutels, A. W. Jeremiasse and R. A. Rozendal, Environ. Sci. Technol., 2008, 42, 8630–8640.
4. D. Kavak, Desalin. Water Treat., 2013, 51, 1720–1726.
5. D. Call and B. E. Logan, Environ. Sci. Technol., 2008, 42, 3401–3406.
6. A. W. Jeremiasse, H. V. M. Hamelers, M. Saakes and C. J. N. Buisman, Int. J. Hydrogen Energy, 2010, 35, 12716–12723.
7. D. F. Call, M. D. Merrill and B. E. Logan, Environ. Sci. Technol., 2009, 43, 2179–2183.
8. Y. Zhang, M. D. Merrill and B. E. Logan, Int. J. Hydrogen Energy, 2010, 35, 12020–12028.
9. R. A. Rozendal, A. W. Jeremiasse, H. V. Hamelers and C. J. Buisman, Environ. Sci. Technol., 2007, 42, 629–634.
10. A. W. Jeremiasse, H. V. Hamelers, E. Croese and C. J. Buisman, Biotechnol. Bioeng., 2012, 109, 657–664.
11. A. W. Jeremiasse, H. V. Hamelers and C. J. Buisman, Bioelectrochemistry, 2010, 78, 39–43.
12. Y. Liu, X. Dong and P. Chen, Chem. Soc. Rev., 2012, 41, 2283–2307.
13. Y. Zhang, G. Mo, X. Li, W. Zhang, J. Zhang, J. Ye, X. Huang and C. Yu, J. Power Sources, 2011, 196, 5402–5407.
14. L. Xiao, J. Damien, J. Luo, H. D. Jang, J. Huang and Z. He, J. Power Sources, 2012, 208, 187–192.
15. Y.-X. Huang, X.-W. Liu, J.-F. Xie, G.-P. Sheng, G.-Y. Wang, Y.-Y. Zhang, A.-W. Xu and H.-Q. Yu, Chem. Commun., 2011, 47, 5795–5797.
16. J. Hou, Z. Liu and P. Zhang, J. Power Sources, 2013, 224, 139–144.
17. L. Wei, H. Han and J. Shen, Int. J. Hydrogen Energy, 2012, 37, 12980–12986.
18. D. R. Lovley, E. J. Phillips and D. J. Lonergan, Appl. Environ. Microbiol., 1989, 55, 700–706.
19. H. Han, M. Cui, L. Wei, H. Yang and J. Shen, Bioresour. Technol., 2011, 102, 7903–7909.
20. P. Aelterman, S. Freguia, J. Keller, W. Verstraete and K. Rabaey, Appl. Microbiol. Biotechnol., 2008, 78, 409–418.
21. M. Du, T. Yang, X. Li and K. Jiao, Talanta, 2012, 88, 439–444.
22. H. Liang, X. Miao and J. Gong, Electrochem. Commun., 2012, 20, 149–152.
23. M. Sun, F. Zhang, Z. H. Tong, G. P. Sheng, Y. Z. Chen, Y. Zhao, Y. P. Chen, S. Y. Zhou, G. Liu, Y. C. Tian and H. Q. Yu, Biosens. Bioelectron., 2010, 26, 338–343.

---