Fabrication of Ni nanoparticles-dispersed carbon ...

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 1 1 4 5 e1 1 5 3

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/he

Fabrication of Ni nanoparticles-dispersed carbon micro-nanofibers as the electrodes of a microbial fuel cell for bio-energy production Shiv Singh a, Nishith Verma a,b,* a b

Department of Chemical Engineering, Indian Institute of Technology Kanpur, Kanpur 208016, India Center for Environmental Science and Engineering, Indian Institute of Technology Kanpur, Kanpur 208016, India

article info

abstract

Article history:

A nickel (Ni) nanoparticles- (NPs) dispersed web of carbon micro-nanofibers (ACFs/CNFs)

Received 5 May 2014

was fabricated as the electrode of a microbial fuel cell (MFC) for bio-energy production

Received in revised form

using Escherichia coli as a microbial catalyst. The multiscale web of ACFs/CNFs was prepared

19 October 2014

using chemical vapor deposition, with the Ni NPs as the catalyst and benzene as the carbon

Accepted 13 November 2014

source to grow the CNFs on the ACF substrate. The Ni NPs attached at the tips of graphitic

Available online 6 December 2014

CNFs facilitated the transfer of E. coli-released electrons to the anode of MFCs. Linear sweep voltammetry was performed to determine power density and open circuit potential (OCP)

Keywords:

of the prepared MFCs. The power density and OCP were experimentally measured as

Microbial fuel cells

710 ± 5 mV and 1145 ± 20 mW/m2, respectively, which were approximately 9 times greater

Power density

than those of the ACF substrate-based MFCs. The transition metal-CNFs-based electrodes

Escherichia coli

prepared in this study may be a potentially alternative to the expensive noble metals-based

Carbon nanofiber

electrodes presently used in MFCs.

Metal nanoparticles

Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

Introduction Presently, the enormous amount of fossil fuels is burnt to generate electricity. This has caused significant increase in the concentration levels of carbon dioxide in the atmosphere, leading to global warming and also, the gradual depletion of fossil fuels. Therefore, there is a need to explore clean source of energy as an alternative to fossil fuels. In this context, microbial fuel cells (MFCs) have drawn considerable attention of researchers. MFCs are novel electrochemical devices which can convert microbial metabolic power into electricity.

Considering that the municipal or domestic wastewater contains significant amounts of biodegradable organic wastes, such devices have potential of producing electricity from the wastewater treatment plants [1,2]. These devices may also be potentially used for the recovery of metals and nutrients from industrial effluents [3,4]. One of the challenges in developing efficient MFCs is the fabrication of high-quality electrodes. The electrodes should be relatively inexpensive, highly conductive and should have large surface area. Anode is responsible for the electron transfer, whereas the preferred reaction on cathode is oxygen reduction. Studies have shown that biocompatibility,

* Corresponding author. Department of Chemical Engineering, Indian Institute of Technology Kanpur, Kanpur 208016, India. Tel.: þ91 512 2596352; fax: þ91 512 2590104. E-mail addresses: [email protected], [email protected] (N. Verma). http://dx.doi.org/10.1016/j.ijhydene.2014.11.073 0360-3199/Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

1146


electronic conductivity and high electro-catalytic activity towards the oxidation of metabolites are vital parameters for the selection of anode to enhance power density [5,6]. Cathode is also a restrictive component of MFCs [7]. In the cathode chamber, atmospheric oxygen is usually used as the electron acceptor. Despite oxygen being readily available and having high oxidative potential, an efficient catalyst is required to overcome the large over-potential of oxygen. In the past decade, several noble metals-based carbon electrodes have been fabricated for MFCs, because of their superior activity, biocompatibility, and high electrical conductivity. Platinum (Pt) is the dominant metal used as a catalyst in the electrode for the oxidation of metabolites and oxygen reduction [8]. The other noble metals-based catalysts used for MFCs are Ru/Pt loaded on carbon nanotubes (CNTs) [9,10], Au nanoparticles (NPs) sputtered in carbon-paper [11], and TieTiO2/Pt composite [8]. Different forms of carbon without metals have also been developed as electrodes for MFCs, for examples, the graphene deposited carbon cloth [12], reticulated carbon foam [13], the electrospun polyacrylonitrile-based carbon nanofibers (CNFs) [14] and graphitic carbon felt [15,16]. The present study describes the development of the nickel (Ni) NPs-dispersed and chemical vapor deposition- (CVD) grown multiscale forming web of carbon micro-nanofibers (ACFs/CNFs) as the electrodes (anode and cathode) of a double chambered MFC. The development of the metal NPsdispersed ACFs/CNFs is relatively newer. Such materials have been applied as efficient adsorbents for environmental remediation [17e19]. In the most recent application the bimetals-dispersed ACF/CNF was applied as the electrode of a glucose biosensor [20]. Ni-based materials have been used in other bioelectrochemical systems, in particular, microbial electrolysis cells for hydrogen evolution [21e25]. For the first time, Ni-ACF/CNF has been fabicated and demonstrated as an effective electrode for MFCs in the present study. The Ni NPs had the following multiple roles: (1) as the CVD catalyst for the growth of the CNFs on the ACF substrate, (2) as the facilitator of electron transfer from Escherichia coli (E. coli) to anode, and (3) as the catalyst for oxygen reduction at cathode. The CNFs provided biocompatibility to the growth of biofilms and increased the electroconductivity of the electrodes.

prepared in Milli-Q water. The MFC was fabricated from the acrylic pipe procured from a local market. Proton exchange membrane (PEM, nafion 117) was purchased from Sinsil International, India.

Synthesis of Ni-ACF/CNF electrodes The preparation of CNFs on the ACF substrate has been previously described [19]. Briefly, the pretreated ACF samples were impregnated with the 0.4 M-Ni(NO3)2.6H2O salt dispersed in water, using the wet incipient method. Approximately 0.3% (w/w) SDS was used as an anionic surfactant in the impregnating solution to prevent the agglomeration of Ni(NO3)2 crystals and facilitate their uniform dispersion in the ACF during impregnation. After the impregnation, the ACF samples were dried for 6 h at room temperature (30 ± 5 C) and then for another 12 h at 120 C. Next, calcinations, reduction and CVD were performed on the impregnated ACF sample in a vertical reactor. The detailed configuration of the reactor and the arrangement for holding samples are described in the previous study [26]. The calcination and reduction of the impregnated sample were performed in a N2 atmosphere at 400 C for 4 h and in a H2 atmosphere at 550 C for 2 h, respectively. During the calcination and reduction, flow rates of the gases were maintained constant at 200 standard cc per min (sccm). Ni(NO3)2 dispersed in the ACFs were converted to the metallic state (Ni NPs) after calcination and reduction. The produced Ni NPs decomposed C6H6 during the CVD, forming a multiscale web of ACFs/CNFs. The CVD was performed at 800 C for 2 h. A few samples of Ni-ACFs/CNFs were ultrasonicated in a 0.05 M-HNO3 solution for 5 min to dislodge the Ni NPs from the tips of the CNFs. The prepared Ni-ACFs/CNFs were directly used as anode and cathode in MFCs, without requiring any further treatment. For the reference purposes, the three types of materials used as the electrodes for MFCs in this study are termed as (1) ACF, (2) Ni-ACF/CNF, and (3) ACF/ CNF. ACFs/CNFs were produced by sonicating Ni-ACFs/CNFs. Fig. 1 schematically describes the steps involved in the preparation of electrode materials for MFCs developed in this study.

Construction and operation of MFCs

Materials and methods Materials Ni(NO3)2.6H2O (purity > 97%), sodium dodecyl sulfate (SDS, purity > 99%), C6H6, potassium hexacyanoferrate (K3Fe(CN)6, purity > 99%), sodium acetate (CH3COONa, purity > 98%), sucrose and other reagents, used for preparing buffers of NaCl, KCl, Na2HPO4 and KH2PO4 were procured from Merck, Germany Hydrogen (purity > 99.999%) and nitrogen (purity > 99.999%) gases were purchased from Sigma Gases, India. The phenolic resin precursor-based ACFs were purchased from Gun Ei Chemical Industry Co. Ltd., Japan. The E. coli (K12) culture was indigenously purchased. The reagents used to prepare Luria Bertani (LB) medium, including tryptone and yeast extract, were purchased from Thomas Baker Laboratory Reagent, India. All aqueous solutions used in this study were

The supplementary figure S1 shows the schematic illustration of an MFC. The H-shaped MFC consisted of two chambers (effective volume ¼ 200 ml) fabricated from an acrylic pipe (inner diameter ¼ 67.0 mm). They were connected using a 40 mm-diameter pipe. A PEM (nafion 117) sheet (exposed area ¼ 2800 mm2) was used in the connecting pipe to separate the chambers. The electrodes were held using a Teflon-framesuspended 3 mm-brass wire in the respective chambers. The anode and cathode (working area ¼ 8 cm2) were electrically connected via an external circuit, made of a 3 mm-thick brass wire, to a resistor (5000 U). An electronic multimeter (Model 2000, Keithley) was used to measure voltage across the resistor. The distance between the anode and cathode was 100 mm. The anode chamber was filled with approximately 50 ml volume of 10 mM-PBS (prepared using 8.0 g-NaCl, 0.2 g-KCl, 1.44 g-Na2HPO4, and 0.24 g-KH2PO4 at pH ¼ 7), 50 ml volume of


1147

Fig. 1 e Schematic illustration of the fabrication of the Ni-ACF/CNF electrodes for MFCs.

100 mM-sucrose solution made up in 10 mM-PBS, 50 ml volume of E. coli-LB broth (108e109 CFU/ml at pH ¼ 7) and 50 ml volume of 12 mM-CH3COONa. N2 was continuously purged through the anolyte solution for 30 min to establish an anaerobic condition. The cathode chamber was filled with 200 ml of 100 mM-K3Fe(CN)6 solution prepared in 10 mM-PBS. The initial pH of the catholyte solution was 7.0. The base of the cathode chamber was fitted with a bubble disperser externally connected to a peristaltic pump (speed ¼ 100 rpm) which continuously dispensed the atmospheric air (~200 sccm) in the cathode chamber during the test. Both electrodes (anode and cathode) were fabricated from the same material: the ACF substrate, or Ni-ACF/CNF, or ACF/CNF. The tests were performed in the batch mode at room temperature.

Electrochemical measurement and analysis Power density and open circuit potential (OCP) were obtained at the scan rate of 1.0 mVs1 from 1 V to 0 V, using the potentiostat (AUTOLAB-PGSTAT302N) and the linear sweep voltammetry (LSV) method [27]. Nova software was used to draw the polarization curves. Tests were replicated 5 times for a single MFC set-up. The area-power densities (P mW/m2) of the prepared MFC were calculated using P ¼ V I/A, where V is the measured voltage, I is the current with respect to the applied voltage and A is the area (8 cm2) of either anode or cathode. Before measurements, the MFCs were left idle for 5 days, connected to a 5000 U-resistor, for the stabilization of cell voltage, because it was found to slowly increase during the initial period of the tests. After 2 days of initialization, the voltage nearly stabilized, although a slight fluctuation was observed. Therefore, the electrochemical analysis was performed after 5 days when the cell voltage had completely stabilized. The electrochemical impedance spectroscopy (EIS) measurements were performed to determine the internal resistances of the MFC, using the FRA software and the potentiostate. EIS measurements were performed for the cell from 100,000 to 0.01 Hz at the potential amplitude of 0.01 V. The EIS

measurements of the MFC were obtained at the OCP of the respective electrodes. The counter and reference electrodes of the potentiostate were connected to the cathode, whereas the working electrode was connected to the anode.

Surface characterization of electrocatalyst ACFs, Ni-ACFs/CNFs and the ACFs/CNFs were characterized using several techniques, including atomic absorption spectroscopy (AAS), scanning electron microscopy (SEM), energy dispersive X-ray (EDX) spectroscopy, X-Ray diffraction (XRD), Raman spectroscopy and Brunauer-Emmet-Teller (BET) surface area and pore-size distribution (PSD) measurements. The AAS (Varian AA-240, USA) analysis of the impregnating solution was performed to calculate the Ni loading in the ACFs. The surface topographies of the ACFs, Ni-ACFs/CNFs and the ACFs/CNFs were examined using field emission SEM (Supra 40VP, Zeiss, Germany). XRD analysis was performed to determine the crystal size and patterns of the Ni NPs dispersed in the ACFs/CNFs. A Cu-Ka radiation (k ¼ 1.54178 A ) was used for determining the XRD patterns in a 2q range of 10 e120 and scan rate of 3 per min. Raman analysis (Model: Alpha, Make: Witec, Germany) was performed to determine the relative disorder or graphitization (ID/IG) in the prepared electrode samples. The data were collected using the Ar-ion laser (l ¼ 532 nm) as an excitation source and a CCD as the detector in the range of 500e4000 cm1 in air at 25 C. The SBET, total pore volume (Vtotal) and PSD of the prepared electrodes were determined using nitrogen at 77 K and the Autosorb-1C Quantachrome instrument.

Result and discussion Ni loading on prepared electrode materials The Ni loading in the impregnated ACF samples was determined to be ~450 mg/g. The loading was ~30% higher than that

1148


Fig. 2 e SEM images of (AeA′) ACF, (BeB′) Ni-ACF, (CeC′) Ni-ACF/CNF, (DeD′) ACF/CNF, and EDX spectra of (E) Ni-ACF and (E′) Ni-ACF/CNF samples. Inset shows the focus area for EDX.

in the samples impregnated without using SDS. The surfactant enhanced the metal loading and dispersion in the ACFs by preventing the agglomeration of the salt crystals in the impregnating solution. As shown later, the uniform Ni loading facilitated the dense and uniform growth of CNFs on ACFs.

Surface topography The surface morphologies of the fabricated materials were examined using SEM. Fig. 2 describes the SEM images at low (5e15 KX) and high magnifications (100e200 KX). Fig. 2(AeA0 ) shows the SEM images of the ACF substrate. The ACF surface was smooth. Macro-pores were visible on the external surface. Fig. 2(BeB0 ) shows the SEM images of the Ni NPs-dispersed ACF (Ni-ACF) samples. The Ni NPs were uniformly dispersed on the ACF surface. The Ni NPs were also observed inside the pores. Fig. 2(CeC0 ) shows the SEM images of Ni-ACFs/CNFs. The growth of the CNFs on the ACF was uniform and dense.

As shown later, the CNFs were graphitic and enhanced the conductivity of the prepared electrodes. The shiny Ni NPs present at the tips of the CNFs confirmed the tip-growth mechanism. Fig. 2(DeD0 ) shows the SEM images of the ACF/ CNF samples. As observed, most of the Ni NPs were dislodged from the tips during the ultrasonication. Fig. 2(EeE0 ) describes the EDX spectra of the Ni-ACF and NiACF/CNF samples. The area-spectrum (Fig. 2E) showed the presence of carbon and Ni, confirming that NiO was completely reduced to metallic (Ni) state during the H2reduction at 550 C. Fig. 2E0 shows the point-EDX spectra of NiACFs/CNFs. The spectra confirmed the shiny NPs as Ni. No other foreign impurities were detected.

XRD patterns The XRD analysis of the ACF and Ni-ACF samples was performed to determine the structure of the crystalline phase,

1149


using JCPDS software (No- 04-0850). Fig. 3a shows the XRD spectra of ACFs (shown in black) and Ni-ACFs (shown in blue). There was a common characteristic peak at 2q angle of ~26 in both samples, attributed to the crystallographic index of (0 0 2) in the amorphous carbon. The spectra of the Ni-ACFs exhibited five distinctive peaks at 2qs ¼ 44.5, 51.8, 76.3, 92.9 and 98.44 , corresponding to (1 1 1), (2 0 0), (2 2 0), (3 1 1) and (2 2 2) indices, respectively. The spectra of the Ni-ACFs also revealed that the prepared samples contained Ni in its pure metallic FCC phase (previously confirmed using the EDX analysis). The crystallite size of the Ni NPs, calculated using the Scherrer formula, was found between 10 nm (corresponding to 44.5 ) and 31 nm (corresponding to 76.3 ).

Raman spectra Raman spectroscopy was used to investigate the presence of ordered and disordered phases in ACFs/CNFs. Fig. 3b describes the Raman spectrum of the ACF and Ni-ACF/CNF samples. The first vibrational peak observed at 1342.6 cm-1 was attributed to the disordered phase (D-band), whereas the second vibrational peak at 1580.4 cm-1 was attributed to the graphitic phase (G-band) of carbon. The D-band represents the vibrations of carbon atoms having dangling bond with the unsaturated sp3 hybridized valence in the planar terminations of the disordered graphite structure, whereas the G-band represents the planar geometry (sp2 hybridized) of carbon atoms.

The ID/IG ratio reflects the degree of disorder, as expressed in the sp3/sp2 carbon ratio [28]. The ID/IG ratio of ACFs was determined to be 1.17. The ratio decreased to 1.06 in Ni-ACFs/ CNFs. Therefore, Ni-ACFs/CNFs contained higher degree of graphitization than ACFs. In other words, ACFs had more disordered structure than ACFs/CNFs. The graphitic phase enhanced the conductivity of the ACF/CNF electrode, resulting in the relatively superior performance of Ni-ACFs/CNFs, shown later.

SBET, Vtotal and PSD analysis The BET surface area of the prepared electrodes was determined using the linearized BET equation fitted to 20 data points corresponding to the amounts of nitrogen adsorbed vs relative pressures (P/P0) ranging from 0.05 to 0.35. The microporosity and meso-porosity of the prepared materials were determined using density functional theory (DFT) and the Barrett-Joyner-Halenda (BJH) method, respectively. Before starting the adsorption/desorption analysis, the prepared samples were heated at 150 C for 12 h in vacuum. Table 1 presents the data of the analysis. The SBET and Vtotal of the ACFs were determined to be 1140 m2/g and 0.609 cc/g, respectively. The micro-porosity contents of the ACFs were significantly high (~86% of the total pore volume). The SBET and Vtotal values significantly decreased in the Ni(NO3)2-impregnated samples because of the blockage of the pores with the salts. However, the calcination increased SBET and pore volume, as the pores were opened up. The SBET and Vtotal values increased to ~805 m2/g and 0.450 cc/g, respectively, in the NiOACF samples. The reduction of NiO to Ni NPs further increased the SBET and Vtotal values to 959 m2/g and 0.558 cc/g, respectively, in Ni-ACFs. Both NiO- and Ni-ACF samples were largely microporous. Such materials are not considered to be conducive for the adsorption or attachment of large size molecules such as E. coli. Interestingly, although Ni-ACF/CNF had relatively lesser SBET (~597 m2/g) and VTotal (0.376 cc/g), the material was considerably mesoporous (~47%), and therefore, was suitable for the growth or multiplications of E. coli. As shown later, the performance of the electrodes fabricated using the multiscale web of Ni-ACFs/CNFs was superior to that of Ni-ACFs.

Performance of MFCs using different electrode materials Fig. 4 describes variation in polarization and power density with increasing current densities, for the ACF, Ni-ACF/CNF, and ACF/CNF samples. The bioelectricity was successfully generated from the MFCs constructed using the prepared

Table 1 e SBET, Vtotal and PSD of the prepared materials. Sample

Fig. 3 e (a) XRD patterns and (b) Raman spectra of the ACF and Ni-ACF samples. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

ACF Ni(NO3)2-ACF NiO-ACF Ni-ACF Ni-ACF/CNF

SBET (m2/g)

Vtotal (cc/g)

1140 482 805 959 597

0.609 0.270 0.450 0.554 0.376

PSD (%) Micro

Meso

Macro

86.48 85.40 86.29 79.27 47.87

5.64 9.48 8.91 17.33 47.34

7.86 5.11 4.78 3.39 4.78

1150


electrodes. During the initial stage of the test, OCP was relatively lower, attributed to discontinuity in the formation of biofilm on the anode, possibly because of a relatively lesser number of E. coli produced, thereby resulting in a small electron transfer rate to the anode. As the metabolic activity progressed, the number of colonies of E. coli grew. The biofilm formed on the anode became stable and the electron transfer rate or OCP increased. E. coli is not electrochemically conductive. However, it is mentioned that there are two ways of transferring the electron to the anode surface by E. coli in the absence of mediator(s). (1) E. coli present in the biofilm can self-mediate the extracellular electron transfer through the electrochemically activated excretion of electron shuttling redox molecules which work as a mediator for the electron transfer [6,29,30], and (2) the pili of E. coli also facilitate the electron transfer to anode [31]. In addition to the redox molecules produced by E. coli and/or the pili of E. coli, the metal NPs or carbon-based nanomaterials can also facilitate the electron transfer from E. coli to electrodes in the mediator-less MFCs [32,33]. Here, it is also important to mention that Ni or ACFs/CNFs do not exhibit antibacterial activities and have been used as electrode materials in MFCs and also in microbial electrolysis cells [22,34]. To corroborate this aspect of the material, an antibacterial test was performed on the ACFs and Ni-ACFs/ CNFs using E. coli. The tests were performed using plate count method. The results are included in the supplementary figure S2, showing an insignificant antibacterial effect of the produced materials in this study. The colony counts of E. coli in the broths containing the ACFs and Ni-ACFs/CNFs were found to be approximately the same (~108 CFU/ml) as in the control even after 72 h, thus showing no sensitivity of E. coli to the electrode materials. Therefore, the electrochemical activity attributed to E. coli remained unchanged during the operation of the MFCs and the performance of MFCs was considered to be affected by the anodic reactions only. It is also mentioned that the Coulombic efficiency (CE) may be diminished by alternate electron acceptors present in the anolyte, for example, wastewater [35]. In the present study, synthetic

Fig. 5 e EIS analysis of the ACF, Ni-ACF/CNF and ACF/CNF samples at the respective OCP. Inset shows the magnified plots for the Ni-ACF/CNF and ACF/CNF samples.

water was used. Therefore, CE was not measured and the other parameters (power density, current density, OCP, and electrochemical impedance) were considered to be sufficient to study the performance of the MFCs, discussed below. The OCP and current density of the Ni-ACF/CNF-based MFCs were determined to be 710.0 ± 5 mV and ~4650 ± 20 mA/m2, respectively, which were significantly higher than those (380.0 ± 5 mV and 1400 mA/m2, respectively) of the MFCs constructed from the ACF substrate. The areapower density of the multiscale forming web-based MFCs was ~9 times greater than that of the ACF-based MFCs. The relative superior performance of Ni-ACF/CNF-based MFCs was attributed to the following three characteristics of the material. (1) It had large (~47%) mesopore-contents which were favorable for the attachment of E. coli and the growth of biofilms. Further, E. coli may have produced adhesins on pili and fimbriae, facilitating the attachment of the biofilm on the surface of ACFs/CNFs [31]. (2) The graphitic phase present in the CNFs facilitated the electron transfer from the c-type cytochrome of E. coli to the anode [32]. Therefore, the charge transfer resistance (shown later) was smaller in Ni-ACFs/ CNFs, which was the reason for the production of high current density (4650 ± 20 mA/m2), using the MFC based on the Ni-ACF/CNF electrodes. (3) The presence of the Ni NPs at the tips of the CNFs also facilitated the electron transfer. To further examine the role of the Ni NPs, polarization tests were performed on the MFCs constructed using the ACF/CNF samples. As shown in Fig. 4, the OCP (565 ± 5 mV) and power density (1005 ± 20 mW/m2) of the MFCs based on the ACF/CNF samples were smaller than that of the Ni-ACFs/ CNFs-based MFCs. All or most of the Ni NPs were removed

Table 2 e MFC parameters of ACF and Ni-ACF/CNF calculated using Nyquist plots. Sample Fig. 4 e Polarization and power density curves for different electrodes prepared in the study.

ACF Ni-ACF/CNF ACF/CNF

Rs (U)

Rct (U)

W (U)

220.4 44.6 49.5

90.8 1.8 3.6

14.28 9.17 11.23

1181 260 140 200e400 350e650 2350e7530 37 1150e1900 ~700 700 500 300e620 400e810 700e800 ~300 390e570 Au sputtered carbon paper Plain carbon cloth Reticulated vitreous carbon electrodes Polyacrylonitrile based carbon nanofiber Pt dispersed on carbon paper Bare carbon felt Carbon felt Pt-coated carbon cloth

462 52.5 11 19.31e61.3 10e110 166e1326 5.7 300e830

530 ~4000 5670 430 e ~700 ~1000e7000 480 ~720 749 434e867 e 329e759 481e941 Graphic sheets Porous carbon fiber Carbon cloth with gas diffusion layers Copper-phthalocyanine and Ni NPs Pt CNT/Pt RuePt/MWNT and SnePt/MWNT

240 686e1487 4250 38e118 1313 36e170 151e2470

4650 1145 710 Ni-ACFs/CNFs Ni-ACFs/CNFs

Graphic sheets Graphite rods Carbon brush Plain carbon paper TieTiO2/Pt Plain carbon paper Graphite,SnePt/MWCNT anode with methylene blue and natural red Au sputtered carbon-paper Graphene deposited carbon cloth Reticulated vitreous carbon electrodes Plain carbon paper Plain carbon paper PPy/GO-modified graphite felt Carbon felt Fe3O4/CNT dispersed carbon paper

E. coli (K12)

Wastewater Geobacter sulfurreducens Sludge Sludge from industrial effluent Sludge Wastewater E. coli (DH5a)

Domestic wastewater Pseudomonas aeruginosa Activated sludge Sludge Anaerobic sludge Shewanella oneidensis MR-1 E. coli (K12) E. coli

Present study [1] [2] [4] [7] [8] [9] [10]

[11] [12] [13] [14] [16] [27] [31] [33]

Anode Biocatalyst References

Table 3 e A comparison of performance of double chambered MFCs.

Cathode

OCP (mV)

Power density (mW/m2)

Current density (mA/m2)


1151

from the Ni-ACF/CNF samples by sonication under the present sonication conditions. In our previous study, the identical sonication conditions were used to dislodge or remove the Ni NPs from the ACF/CNF [26]. A negligible amount of Ni NPs may have been left on the surface of the ACFs after sonication. Therefore, indeed, the graphitic structure of CNFs was responsible for power generation using the electrodes fabricated from the sonicated samples (ACF/CNF). Therefore, it may be concluded that the Ni NPs facilitated the generation of current via the electron transfer to the anode. A small crystal size (10e31 nm) of the Ni NPs in ACFs/CNF had the catalytic effect on the reduction of atmospheric oxygen in the cathode chamber [8]. In general OCPs should be the same for different electrodes. However, the OCPs have been measured to be different in many cases including this study, if the electrodes are different or modified [7, 9, 10, 14, 27, and 33]. The difference in the OCPs for the different electrode materials used in this study is attributed to the enhanced electron transfer at the anode prepared from the mesoporous CNFs, in comparison to the microporous ACFs. The former materials favor the biofilm formation or microbial activity via the increased bacterial adhesion, also discussed earlier in SBET, Vtotal and PSD analysis.

EIS analysis EIS analysis was performed to determine the different voltage losses in the MFCs, attributed to the ohmic (solution) resistance (Rs), charge transfer resistance (Rct) and the Warburg mass transfer diffusion resistance (W). Rs depends on the conductivity of electrolytes and electrodes, PEM, and different electrical interconnections between electrodes, whereas Rct depends on the conductivity of cathode. Fig. 5 shows the Nyquist plots and an electrical model of the MFCs. Table 2 summarizes the resistances of different electrodes-based MFCs, calculated using the Nyquist plots. The Rs, Rct and W values obtained for the Ni-ACF/CNF-based MFCs were lowest (44.6, 1.8 and 9.17 U, respectively) amongst the prepared MFCs, attributed to the favorable condition available for the growth of biofilm in the mesospores containing ACFs/CNFs and high conductivity of the CNFs, which facilitated electrons transfer to cathode. Further, the Rs, Rct and W values obtained for the MFCs based on the nonsonicated samples were lower than those (49.5, 3.6 and 11.23 U, respectively) for the sonicated samples, attributed to the Ni NPs present in the former samples. Therefore, the Ni NPs also facilitated the electron transfer from the electrolyte to the electrodes of the MFC. On the contrary, the Rs value was the highest for the ACF-based MFCs, because of the absence of graphitic contents and the metal (Ni) NPs. The Rct and W values obtained for the ACF substrate-based electrode MFCs were also significantly high. Therefore, the produced electrons were not easily transferred to cathode, adversely affecting the efficacy of the MFC. The solution resistance (Rs), the x-intercept of the Nyquist plot, should be the same for different anode materials. However, the different interfacial contact resistances between an anode and an electrolyte can produce different Rs [36e38], keeping the other parameters unchanged, including electrolyte, PEM, and distance between the electrodes. Here, the

1152


difference (~5 U) observed between the Rs value of the Ni-ACF/ CNF in comparison to that of the ACF/CNF is attributed to the presence of the conductive Ni NPs at the tips of the CNFs, which were in contact with the electrolyte solution, whereas the difference (~ 180 U) between that of the ACF and ACF/CNF or Ni-ACF/CNF anode is attributed to the conductive graphitic CNFs dispersed on the ACF substrate, also in contact with the solution. The performance of the double chambered MFC based on the Ni-ACF/CNF electrodes was compared to the different types of carbon-based electrodes discussed in the literature when possible. Table 3 describes the comparative data. The OCP and power density values of the MFCs prepared in this study are relatively higher in most cases and comparable in few cases. In general, the noble metal-based electrodes performed superior to the transition metals-based electrodes. Also, the metal-dispersed carbon electrodes performed superior to those without metal. This may be mentioned that if one were to determine the role of Ni NPs or graphitic carbon separately in anode and cathode, the tests should be performed using the same material for one of the electrodes and different materials for the other electrode, and vice-versa. Based on the data, it may be, however, inferred that Ni-ACF/CNF was the best material because of its effects (facilitating the electron transfer at the anode and on the catalytic oxygen reduction at the cathode), considered together.

Conclusions The multiscale forming web of ACFs/CNFs dispersed with Ni NPs was fabricated and applied as the efficient anode and cathode of a double chambered MFC. The MFC produced high OCP (710 mV), current density (4650 mA/m2) and power density (1145 mW/m2). The prepared electrode material having significant mesoporosity, electrical conductivity, and dispersed Ni NPs facilitated electron transfer to the anode and the catalytic reduction of oxygen at the cathode. The MFCs were stable and had small internal resistances. The novel and efficient Ni-ACF/CNF prepared as the electrodes of MFCs may be an alternative to the expensive noble metals-based electrodes.

Acknowledgments The authors gratefully acknowledge the Gun Ei Chemical Industry Co. Ltd., Japan for supplying ACFs. The authors are also thankful to the Center for Environmental Science and Engineering at IIT Kanpur for carrying out the research.

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ijhydene.2014.11.073

references

[1] Mansoorian HJ, Mahvi AH, Jafari AJ, Amin MM, Rajabizadeh A, Khanjani N. Bioelectricity generation using two chamber microbial fuel cell treating wastewater from food processing. Enzyme Microb Technol 2013;52:352e7. € der U. [2] Liu Y, Harnisch F, Fricke K, Sietmann R, Schro Improvement of the anodic bioelectrocatalytic activity of mixed culture biofilms by a simple consecutive electrochemical selection procedure. Biosens Bioelectron 2008;24:1006e11. [3] Ichihashi O, Hirooka K. Removal and recovery of phosphorus as struvite from swine wastewater using microbial fuel cell. Bioresour Technol 2012;114:303e7. [4] Choi C, Cui Y. Recovery of silver from wastewater coupled with power generation using a microbial fuel cell. Bioresour Technol 2012;107:522e5. € der U, [5] Rosenbaum M, Zhao F, Quaas M, Wulff H, Schro Scholz F. Evaluation of catalytic properties of tungsten carbide for the anode of microbial fuel cells. Appl Catal B Environ 2007;74:261e9. [6] Wang Y-F, Tsujimura S, Cheng S-S, Kano K. Self-excreted mediator from Escherichia coli K-12 for electron transfer to carbon electrodes. Appl Microbiol Biotechnol 2007;76:1439e46. [7] Ghasemi M, Daud WRW, Rahimnejad M, Rezayi M, Fatemi A, Jafari Y, et al. Copper-phthalocyanine and nickel nanoparticles as novel cathode catalysts in microbial fuel cells. Int J Hydrogen Energy 2013;38:9533e40. [8] Akman D, Cirik K, Ozdemir S, Ozkaya B, Cinar O. Bioelectricity generation in continuously-fed microbial fuel cell: effects of anode electrode material and hydraulic retention time. Bioresour Technol 2013;149:459e64. [9] Ghasemi M, Ismail M, Kamarudin SK, Saeedfar K, Daud WRW, Hassan SHA, et al. Carbon nanotube as an alternative cathode support and catalyst for microbial fuel cells. Appl Energy 2013;102:1050e6. [10] Sharma T, Mohana Reddy AL, Chandra TS, Ramaprabhu S. Development of carbon nanotubes and nanofluids based microbial fuel cell. Int J Hydrogen Energy 2008;33:6749e54. [11] Faa Alatraktchi, Zhang Y, Angelidaki I. Nanomodification of the electrodes in microbial fuel cell: Impact of nanoparticle density on electricity production and microbial community. Appl Energy 2014;116:216e22. [12] Liu J, Qiao Y, Guo CX, Lim S, Song H, Li CM. Graphene/carbon cloth anode for high-performance mediatorless microbial fuel cells. Bioresour Technol 2012;114:275e80. [13] Lepage G, Albernaz FO, Perrier G, Merlin G. Characterization of a microbial fuel cell with reticulated carbon foam electrodes. Bioresour Technol 2012;124:199e207. [14] Ghasemi M, Shahgaldi S, Ismail M, Kim BH, Yaakob Z, Wan Daud WR. Activated carbon nanofibers as an alternative cathode catalyst to platinum in a two-chamber microbial fuel cell. Int J Hydrogen Energy 2011;36:13746e52. [15] Du Z, Li H, Gu T. A state of the art review on microbial fuel cells: a promising technology for wastewater treatment and bioenergy. Biotechnol Adv 2007;25:464e82. [16] Ghasemi M, Wan Daud WR, Ismail M, Rahimnejad M, Ismail AF, Leong JX, et al. Effect of pre-treatment and biofouling of proton exchange membrane on microbial fuel cell performance. Int J Hydrogen Energy 2013;38:5480e4. [17] Bikshapathi M, Singh S, Bhaduri B, Mathur GN, Sharma A, Verma N. Fe-nanoparticles dispersed carbon micro and nanofibers: surfactant-mediated preparation and application to the removal of gaseous VOCs. Colloid Surfaces A: Physicochem Eng Asp 2012;399:46e55. [18] Chakraborty A, Deva D, Sharma A, Verma N. Adsorbents based on carbon microfibers and carbon nanofibers for the


[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

removal of phenol and lead from water. J Colloid Interface Sci 2011;359:228e39. Singhal RM, Sharma A, Verma N. Micro-nano hierarchal web of activated carbon fibers for catalytic gas adsorption and reaction. Industrial Eng Chem Res 2008;47:3700e7. Hood AR, Saurakhiya N, Deva D, Sharma A, Verma N. Development of bimetal-grown multi-scale carbon micronanofibers as an immobilizing matrix for enzymes in biosensor applications. Mater Sci Eng C 2013;33:4313e22. Rausch S, Wendt H. Morphology and utilization of smooth hydrogen-evolving raney nickel cathode coatings and porous sintered-nickel cathodes. J Electrochem Soc 1996;143:2852e62. Hrapovic S, Manuel MF, Luong JHT, Guiot SR, Tartakovsky B. Electrodeposition of nickel particles on a gas diffusion cathode for hydrogen production in a microbial electrolysis cell. Int J Hydrogen Energy 2010;35:7313e20. Selembo PA, Merrill MD, Logan BE. Hydrogen production with nickel powder cathode catalysts in microbial electrolysis cells. Int J Hydrogen Energy 2010;35:428e37. Jeremiasse AW, Hamelers HVM, Saakes M, Buisman CJN. Ni foam cathode enables high volumetric H2 production in a microbial electrolysis cell. Int J Hydrogen Energy 2010;35:12716e23. Selembo PA, Merrill MD, Logan BE. The use of stainless steel and nickel alloys as low-cost cathodes in microbial electrolysis cells. J Power Sour 2009;190:271e8. Singh S, Singh A, Bais VSS, Prakash B, Verma N. Multi-scale carbon micro/nanofibers-based adsorbents for protein immobilization. Mater Sci Eng C 2014;38:46e54. Lv Z, Chen Y, Wei H, Li F, Hu Y, Wei C, et al. One-step electrosynthesis of polypyrrole/graphene oxide composites for microbial fuel cell application. Electrochim Acta 2013;111:366e73. Battumur T, Mujawar SH, Truong QT, Ambade SB, Lee DS, Lee W, et al. Graphene/carbon nanotubes composites as a counter electrode for dye-sensitized solar cells. Curr Appl Phys 2012;12(Suppl. 1):e49e53.

1153

[29] Zhang T, Cui C, Chen S, Yang H, Shen P. The direct electrocatalysis of Escherichia coli through electroactivated excretion in microbial fuel cell. Electrochem Commun 2008;10:293e7. [30] Qiao Y, Bao S-J, Li CM. Electrocatalysis in microbial fuel cellsfrom electrode material to direct electrochemistry. Energy & Environ Sci 2010;3:544e53. [31] Sugnaux M, Mermoud S, da Costa AF, Happe M, Fischer F. Probing electron transfer with Escherichia coli: a method to examine exoelectronics in microbial fuel cell type systems. Bioresour Technol 2013;148:567e73. [32] Wang Y, Li B, Cui D, Xiang X, Li W. Nano-molybdenum carbide/carbon nanotubes composite as bifunctional anode catalyst for high-performance Escherichia coli-based microbial fuel cell. Biosens Bioelectron 2014;51:349e55. [33] Park IH, Christy M, Kim P, Nahm KS. Enhanced electrical contact of microbes using Fe3O4/CNT nanocomposite anode in mediator-less microbial fuel cell. Biosens Bioelectron 2014;58:75e80. [34] Singh S, Joshi HC, Srivastava A, Sharma A, Verma N. An efficient antibacterial multi-scale web of carbon fibers with asymmetrically dispersed AgeCu bimetal nanoparticles. Colloid Surfaces A Physicochem Eng Asp 2014;443:311e9. € der U, Keller J, [35] Logan BE, Hamelers B, Rozendal R, Schro Freguia S, et al. Microbial fuel cells: methodology and technology. Environ Sci Technol 2006;40:5181e92. [36] Lee LTL, He J, Wang BH, Ma YP, Wong KY, Li Q, et al. Fewlayer MoSe2 possessing high catalytic activity towards iodide/tri-iodide redox shuttles. Sci Rep 2014;4:1e7. [37] An J, Kim B, Jang JK, Lee HS, Chang IS. New architecture for modulization of membraneless and single-chambered microbial fuel cell using a bipolar plate-electrode assembly (BEA). Biosens Bioelectron 2014;59C:28e34. [38] Taek Lee K, Yoon HS, Ahn JS, Wachsman ED. Bimodally integrated anode functional layer for lower temperature solid oxide fuel cells. J Mater Chem 2012;22:17113e20.