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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 Available online at www.sciencedirect.com ScienceDirect journal homepage: www.elsevier.com/locate/he Graphene functionalization with metallic Pt nanoparticles: A path to cost-efficient H2 production in microbial electrolysis cells  nchez-Pen Pilar Sa ~ a a, Jordi Rodriguez b, David Gabriel a, Juan Antonio Baeza a, Albert Guisasola a,*, Mireia Baeza b a GENOCOV, Departament d’Enginyeria Quı́mica, Biologica i Ambiental, Escola d’Enginyeria, Universitat Autonoma de Barcelona, 08193 Bellaterra, Spain b GENOCOV, Department de Quı́mica, Facultat de Ciencies, Edifici C-Nord, Universitat Autonoma de Barcelona, 08193 Bellaterra, Spain highlights graphical abstract  Reduced graphene oxide (Pt@rGO) and novel graphitene (Pt@Graphi- tene) were functionalized with Pt.  Pt-functionalized graphene-based materials tested as cathodes for MECs.  Higher surface (24.40 m2/g) and electroactive (1.14 cm2) area for Pt@rGO.  Both materials showed 47% more platinum functionalization than commercial inks.  Functionalized coating MECs had higher performance than those using commercial inks. article info abstract Article history: Platinum is one of the most widely used catalysts in the cathode of Microbial Electrolysis Received 10 November 2021 Cells (MECs) to overcome the relatively slow kinetics of hydrogen evolution, even though it Received in revised form is not economically feasible on a large scale. This work aims at developing, applying, 4 February 2022 characterizing, and optimizing two novel Pt-functionalized inks with promising charac- Accepted 9 March 2022 teristics: Pt@rGO based on reduced graphene oxide and Pt@Graphitene based on a home- Available online 7 April 2022 made material named Graphitene, which showed improved performance at a lower cost. The Pt-functionalized materials were deposited on carbon cloth and used as cathode Keywords: electrode in a single chamber MEC. These materials provided 47% increase in Pt func- Graphitene tionalization over commercial inks. Moreover, surface areas of 10.76 m2/g and 24.40 m2/g Microbial electrolysis cell and electroactive areas of 0.10 cm2/cm2 and 0.16 cm2/cm2 were determined for * Corresponding author. E-mail address: albert.guisasola@uab.cat (A. Guisasola). https://doi.org/10.1016/j.ijhydene.2022.03.078 0360-3199/© 2022 The Author(s). Published by Elsevier Ltd on behalf of Hydrogen Energy Publications LLC. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). 15398 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 Novel materials Pt@Graphitene and Pt@rGO, respectively, a difference caused by structural defects in the Pt-functionalization case of the Pt@rGO, which slightly improved its performance compared to Pt@Graphitene. Reduced graphene oxide Thus, the experimental results reached ca. 0.8 mA/cm2, a 43% higher intensity than that obtained using conventional commercial inks. © 2022 The Author(s). Published by Elsevier Ltd on behalf of Hydrogen Energy Publications LLC. This is an open access article under the CC BY license (http://creativecommons.org/ licenses/by/4.0/). alternatives to ensure the economic viability of full-scale Introduction MECs [9,18]. Pt-coated carbon-based materials are the most common materials for cathodes, but there are also promising Hydrogen (H2) is one promising energy vector to overcome the alternatives such as those based on nickel [19], titanium [20] or current dependence on fossil fuels, since it is a clean energy stainless steel [21], which have demonstrated good perfor- carrier with a high heat of combustion (122 kJ/g) when mance. These metals have similar chemical and physical compared to other possible fuels (coal 40 kJ/g, ethanol 26.5 kJ/ properties to Pt and can provide high catalytic activity for H2 g, methane 50.1 kJ/g or petroleum 44 kJ/g [1]). Furthermore, H2 production [20]. However, a replacement of Pt that provides a has a large number of industrial applications as it is employed similar performance at a lower price has not been reported for hydrogenation in the production of petrol, food and fer- yet. In this context, an alternative option is to minimize the tilizers. However, almost 90% of the H2 produced comes from amount of Pt added to the cathode without hindering its ac- steam reforming [2], a fossil-based, non-sustainable technol- tivity. For instance, decreasing the Pt particle size into the ogy that hinders future H2 utilization. Thus, research is nanoscale (2e3 nm) and functionalizing the cathode with Pt focused nowadays on developing novel CO2-free technologies nanoparticles would lead to great benefits due to the increase for H2 production. Among them, biological approaches stand of the Pt surface area while minimizing the amount of Pt as some of the most sustainable. Currently, there are three required on the cathode surface and increasing Pt availability alternatives for biological H2 production [3]: photosynthesis, [22]. This functionalization has been reported to be successful dark fermentation, and microbial electrolysis cells (MECs). on carbon based materials such as graphene or carbon MECs are bioelectrochemical devices that raise a promising nanotubes [23,24], even though to the best of our knowledge new approach to H2 production from the organic compounds they have not been used for MECs. contained in wastewater [4], thus with a double contribution On the other hand, graphene is a carbon-based novel ma- to a sustainable economy. A MEC operates as a potentiostatic terial that has been widely studied since the report of its cell consisting of an anode and a cathode connected to a physical properties back in 2004 and 2005 [25,26]. Graphene is a power source, i.e. with a constant cell potential. The anode is single layer of sp2-hybridized carbon atoms arranged in a colonized with anode-respiring bacteria (ARB), which act as a honeycomb crystal structure. Its large surface area and the 2-D catalyst to oxidize organic matter into electrons, CO2 and Hþ. carbon network of flat basal planes makes graphene effort- ARB have the unique characteristic of being able to transfer lessly modifiable [27]. Besides that, its synthesis and function- electrons extracellularly to the anode, which subsequently are alization are cheaper than buying a commercial ink. The best circulated to the cathode [4]. Several reductive processes are known modification of graphene is reduced Graphene Oxide reported to occur at the cathode [5e7]. However, in MEC, the (rGO) [28]. rGO is a single-atom-thick, two-dimensional mate- conventional process is the reduction of Hþ or H2O to H2 gas. rial with carbon ring domains, defects, and oxygen-containing MECs can be operated in single or two-chamber mode. In a groups (-OH, eCOOH, etc.) on the surface [29]. Furthermore, rGO two-chamber MEC, the anodic and cathodic chambers are has a higher conductivity compared to other carbonaceous usually separated by an ion-exchange membrane (IEM) materials, thus becoming a perfect candidate for cathodic whereas the anode and cathode coexist in a sole chamber in material of MECs in combination with Pt. In fact, graphene- single-chamber MECs. The lack of membrane reduces trans- based electrodes have already been used in a different type of port resistance at expenses of promoting the growth of bioelectrochemical systems: microbial fuel cells [30e32]. For hydrogen scavengers [8] such as hydrogenotrophic metha- this purpose, rGO could be functionalized with Pt metal nano- nogens [9], hydrogen oxidizing ARB [10] and homoacetogens particles. According to Georgakilas et al. [33], two approaches [11], which can limit considerably the observed H2 production for an organic covalent Pt functionalization reaction are basi- yield [12]. cally available: (i) the formation of covalent bonds between a MECs have provided very promising results under lab- free radical and a C]C bond of pure graphene; (ii) the formation conditions with H2 production raets up to 1e3 m3 H2/m3 of covalent bonds between an organic functional group and an reactor/d [13e15]. However, a successful scale-up has not oxygenated aliphatic domain [34]. been reported yet [16,17]. One of the hurdles to overcome is This work aims at reporting the novel utilization of gra- the actual need of an expensive catalyst in the cathode to phene functionalized with Pt nanoparticles as cathode for drive H2 evolution. The most common catalyst used is plat- efficient bioelectrochemical H2 production. Two different inum (Pt) and, thus, there is a need of finding cheaper types of graphene were synthesized: i) rGO derived from 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 15399 graphene oxide prepared according to the traditional Hum- medium and, afterwards, all the solution was heated up to mers’ method [35,36] and reduced with 2 mM ascorbic acid 45  C for 30 min to promote oxidation. When the material was and ii) a low-cost material, namely graphitene, composed of oxidized, the colour changed from black to dark brown and graphene particles with a broader distribution in the number the solution was allowed to cool down to room temperature. of layers, which was home-made with a simple and highly Deionized water (46 mL) was added to reduce the acidity of the scalable ball-milling procedure. medium and the mixture was shaken magnetically for 15 min. The solution was later treated with 140 mL of deionized water for Milli-Q system (Millipore, Billerica, MA, USA) and 2.5 mL of Materials and methods H2O2 to remove the excess of MnO4 and MnO2 as MnSO4. The solution was washed with 10% of HCl solution and centrifuged MEC design at 3500 rpm as many times as necessary to remove a yellow viscous liquid (undesired reaction by-products). The solid The MEC configuration used in this work was a 28 mL meth- product, graphene oxide, was washed with deionized water acrylate cylindrical vessel with a glass cylinder on top (16 mL), and centrifuged to reach a neutral pH and was dried at 80  C and a total working volume of 35 mL (Fig. 1). The glass cylinder for 24 h. was hermetically closed with a PTFE rubber cover and Then, the resulting solid was mixed with deionized water aluminium crimp, which allowed gas to be collected [37]. The and was sonicated for 1 h to separate the different sheets of anode was a graphite fibre brush (20 mm diameter x 30 mm graphene oxide. The pH was increased up to 9~10 with an length; 0.21 m2) made with type PANEX33 160 K fibres (ZOL- ammonium chloride solution (25% wt) to stabilize the gra- TEK) of 7.2 mm of diameter, thermally treated and connected phene oxide sheets. Then, ascorbic acid (2 mM) was added and with a titanium wire core. The cathode was prepared with the the solution was heated up to 95  C. Finally, the final product, different Pt-functionalized graphene materials. rGO, was centrifuged at 3500 rpm and washed with deionized The gas produced was recovered with a gas-tight bag (Rit- water until neutral pH was obtained. The resulting solid was ter, Cali-5-bond) connected with a PVC tube to the glass cyl- dried at 80  C for 24 h. inder. The cathode and the anode were connected to a power Pt was functionalized with a novel methodology based on supply (Velleman Energy, LABPS3005DN) that applied a con- the addition of Pt to the carbon materials through a covalent stant potential of 0.80 V. Current production was calculated by bond between Pt and the free radical or C]C bond [39]. measuring the voltage drop across a 10 U external resistance 140 mg rGO were mixed with 100 mL of deionized water and that was connected in series to the circuit and applying Ohm's were sonicated for 1 h. 104.1 mg of K2PtCl4 were added to the law. The cells worked in batch mode with sodium acetate mixture and mixed for 30 min. Meanwhile, a solution of 0.1 M (initial concentration in the batch: 1.5 g/L) as electron donor. of NaBH4 was added drop by drop. Then, the solution was mixed for 1 h. The mixture was taken to neutral pH by Functionalization of rGO with Pt washing with deionized water and centrifugation. Then, the product dried overnight at 100  C is referred as Pt@rGO rGO was produced following the Hummers' method [38] and, herein. afterwards, it was functionalized with K2PtCl4. For the rGO Ascorbic acid (99.5%), DMF (99%), hydrochloric acid production, 1 g of graphite flakes (particle size þ100 mesh, (30e35%), hydrogen peroxide (30%), nitric acid (65%), potas- Sigma-Aldrich, Spain), 23 mL of H2SO4 and 0.5 g of NaNO3 were sium permanganate (99.99%), potassium tetrachloroplatinate added in a stirred 500 mL Erlenmeyer, and later cooled down (II) (>99.99%), sodium nitrate (99.0%), sodium borohydride in an ice bath as the reaction is exothermic and needs to be (98%), sulphuric acid (95e98%) and Nafion perfluorinated cooled. Then, 3 g of KMnO4 were slowly dosed in the reaction Fig. 1 e Schematic (a) and picture (b) of the MEC (1: anode, 2: cathode, 3: H2 collection). 15400 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 resin solution were purchased from Sigma-Aldrich (St. Louis, Some materials were studied and morphologically char- MO, USA). acterized using Scanning Electron Microscopy (SEM) and The catalyst ink was obtained by mixing the rGO func- Transmission Electron Microscopy (TEM). SEM measurements tionalized with Pt (0.5 mg Pt/cm2), and Milli-Q water (4.25 mL/ were carried out in a Merlin Zeiss microscope operated at 5 kV cm2) during 20 s. Later, Nafion (33.4 mL/cm2) (Figs. S1 and S2) and with an EDX detector (SEM-EDX) analysis system and Jeol and, N-dimethylformamide (16.7 mL/cm2) were added to JSM 6010 (JEOL, Ltd, Tokyo, Japan) [41]. TEM images were ob- obtain a bonding paste and the solution was stirred again tained with the JEOL 1400 microscope operated a 120 kV. The (Fig. S3). The resulting paste was coated over the carbon cloth Gatan Microscopy Suite Software was used for the analysis of using a thick brush. The final layer was air-dried for a period of electron diffraction. When necessary, particularly for the 24 h (Fig. S4). Finally, a titanium wire was assembled to the external cathode side, a thin carbon film was used to coat the carbon cloth so the cathode could be electrically connected. samples, by thermal evaporation of carbon, to avoid any The same procedure was carried out with non-functionalized misleading charging effect [42]. rGO for comparison purposes. The percentage of Pt was analysed using plasma mass spectroscopy (ICP-OES) by an inductively coupled plasma op- Functionalization of graphitene with Pt tical emission spectrometer (Perkin, Elmer, model Optima 4300DC) with a Milestone microwave digester (model Ultra- Graphitene is the name we have given to a carbon material wave). Between 1.5 and 2 mg of each sample were weighed on with between 4 and 20 layers of graphene, being the material a a microbalance (MX5, Mettler Toledo) and digested, in dupli- mix of graphene and graphite. Graphitene was obtained from cate, with water in a microwave oven. Finally, the amount of GraphCat Community (Graphene community of Catalonia). Pt was determined by ICP-OES spectrometry. This material combines the benefits of graphene (large active The Brunauer, Emmett and Teller (BET) technique was used surface and high conductivity due to the lack of defects) with a to estimate the surface area of the cathodes. A Micromeritics low cost and easy scalability. ASAP 2000 unit was used, using N2 as adsorption/desorption gas. The first step in the Graphitene functionalization was the The oxidation state of Pt was analysed with photoemission activation of its surface with carboxylic groups by dispersing spectroscopy (XPS) at room temperature in a SPECS PHOIBOS Graphitene in a 2.5 M nitric acid and placing the solution in an 150 hemispherical analyser (SPECS GmbH, Berlin, Germany). ultrasound bath for 2 h. Deionized water followed by centrifu- The base pressure was 5$1010 mbar using monochromatic Al gation was used to wash the mixture to reach pH 7.0. 140 mg of K-alpha radiation (1486.74 eV) as excitation source operated at Graphitene were mixed with 100 mL of deionized water and 300 W. The energy resolution as measured by the Full-Width sonicated for 1 h. 104.1 mg of K2PtCl4 were added to the mixture Half-Maximum of the Ag 3d5/2 peak for a sputtered silver and stirred for 30 min. Afterwards, 100 mL of 0.1 M of NaBH4 foil was 0.62 eV. were added drop by drop. Then, the mixture was stirred for 1 h. Both electrodes were connected to a power supply (Velle- Finally, it was rinsed with deionized water followed by centri- man LABPS3005DN) with 0.8 V as applied potential. The fugation until reaching pH 7. The product dried overnight in an voltage across external resistances was measured with a data oven at 100  C is referred as Pt@Graphitene herein. acquisition card with 16-bit resolution (Advantech PCI-1716) Nitric acid (65%), sodium borohydride (98%), potassium installed in a computer where the AddControl programm, a tetrachloroplatinate (II) (>99.99%), DMF (99%) and Nafion per- propietary software developed in LabWindows/CVI2019 by the fluorinated resin solution were purchased from Sigma-Aldrich research group, was used for data management and storage. (St. Louis, MO, USA). H2, CH4, and CO2 were measured with a gas chromatograph The catalyst ink was obtained by mixing the Pt@Graphi- (Agilent Technologies, 7820-A) equipped with two columns, a tene (0.5 mg Pt/cm2) and Milli-Q water (4.25 mL/cm2) during packed column Porapaq Q 80/100 3 ft G3591-81136 20 s. Then, Nafion (33.4 mL/cm2) and N,N-dimethylformamide (1.38 m  2 mm) and a second column MolSieve 5 A 80/100 3 ft. (16.7 mL/cm2) were added and stirred to produce the binder G3591-80017 (1.83 m  2 mm), from Agilent Technologies. N2 was paste, which was coated over the carbon cloth using a thick used as carrier gas. The temperature of the oven was initially set brush. The resultant layer was air-dried for 24 h (Fig. S4). at 70  C for 2 min, followed by a ramp of 20  C$ min1 until Finally, a titanium wire was assembled to the carbon cloth so reaching a temperature of 140  C. Dionex Chromeleon 6.8 the cathode could be electrically connected. The same pro- (ThermoFisher Scientific) was used for data acquisition and cedure was carried out with non-functionalized Graphitene processing to estimate the concentrations of each gas. for comparison purposes. Electrochemical techniques Analytical methods and instrumentation The MEC performance was assessed using electrochemical Samples for chemical analysis were taken at the beginning techniques, in which an abiotic anode was used to assess the and at the end of each cycle, representing a maximum of global cell and cathode performance without being biased by 10% of the complete reactor volume. Acetate in selected the activity of a biofilm on the anode. These techniques were samples was analysed with a gas chromatograph (Agilent applied in a three-electrode configuration and single meth- Technologies, 7820-A), employing a DB-FFAB column acrylate cell in all cases, employing the same abiotic anode for (30 m of length, 250 mm of internal diameter and 0.25 mm of all the cathodes. A commercial Ag/AgCl electrode (BioLogic) film thickness), a flame ionization detector and He as carrier with KCl 3 M internal solution was used as reference gas [40]. electrode. 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 15401 Electrochemical impedance spectroscopy being Ip (A) the peak intensity, n the electrons transferred in Electrochemical impedance spectroscopy (EIS) measurements the reaction, A (cm2) the electroactive surface area, D are helpful to gather performance information of electrodes (6.4$106 cm2/s) the Fe(CN)64 diffusion coefficient in water, C (anode or cathode) [43,44]. EIS experiments were performed (mol/cm3) the concentration of the reaction species in the using a potentiostat/galvanostat with a frequency analyser solution and v (V/s) the scan rate (0.005 V/s). (Autolab PGSTAT302 N, Methrom Inc.). A three-electrodes configuration was used to analyse the cathode performance: Calculations the cathode as the working electrode, the anode as the counter electrode, and an Ag/AgCl electrode as the reference The intensity was calculated with the voltage measured electrode. EIS analyses were recorded in the frequency range across the external resistance and the Ohm law (equation (3)): 100 kHz - 10 mHz at open-circuit potential. The amplitude was I ¼ E=Rext (3) set to 1 mV at AC to avoid detachment of the biofilm and to reduce perturbations of steady state conditions of the cell [41]. being I the current (A), E the voltage (V) and Rext the external In addition, acetate was added to prevent it from being the resistance (U). limiting factor and nitrogen was purged for 30 min to simulate The Coulombic Efficiency (CE), equation (4), is a common real conditions in the cell. indicator of the MEC performance and it corresponds to the The circuit used for fitting the results was a simplified ratio of electrons recovered as current vs those contained in version of the Randles circuit without the Warburg element the organic matter degraded in the anode. [45]. Fig. 2 shows the circuit and corresponding Nyquist plot, Z tF where Rs is the ohmic resistance (related to the resistance of I dt the electrolyte and components, such as electrodes or mem- CE ¼ t0 (4) branes), Rct is the charge resistance (related to the difficulty of F$bS $VL $Dc$M1 S the electrons to flow) and Cdl is the capacitance (related to the being t0 and tF (s) initial and final time of the batch, I (A) the double electrolytic layer). current, F (96,485C mol1$e1) the Faraday's constant, bS the e transferred per mole of substrate, VL (L) the volume of liquid in Cyclic voltammetry the reactor, DC (g$L1) the concentration of substrate The amount of electroactive surface area was tested by cyclic consumed in a batch and MS (g$mol1) the substrate molecular voltammetry (CV), employing a potentiostat/galvanostat weight. (Autolab PGSTAT302 N, Methrom Inc.). CV measurements The performance of the MEC was also evaluated using the were executed using the three-electrode configuration as in cathode gas recovery (rCAT), equation (5), which corresponds section: Electrochemical impedance spectroscopy. The redox to the ratio of electrons recovered as H2 to the electrons pair Fe(CN)64/Fe(CN)63 was the redox marker in a 100 mL of flowing from the anode to the cathode as current: 0.1 M KCl solution containing 0.01 M K4Fe(CN)6/K3Fe(CN)6 under quiescent condition. The oxidation of Fe(CN)64 and the bH2 $VH2 $F$V1 rCAT ¼ Z tF m (5) reduction of Fe(CN)63 on the surface of the working electrode I dt created anodic and cathodic currents, respectively: t0 being bH2 the moles of e transferred per mole of H2 (2 mol [Fe(CN)6]4 4 [Fe(CN)6]3 þ 1 e- (1) e$mol1$H2), VH2 the volume of hydrogen produced, and Vm the molar gas volume (24.03 L mol1) at 20  C. The intensity (Ip) was proportional to the electroactive The H2 relative composition was evaluated using equation  c surface area of the cathode (equation (2), Randles-Sev ik (6): equation [46]). VH2 Ip ¼ 268600$n3=2 $A$D1=2 $C$v1=2 (2) Relative compositionH2 ¼ (6) VH2 þ VCH4 where VH2 and VCH4 represent the volume of H2 and CH4 produced, respectively. The amount of gas was estimated according to the “Gas Bag Method” presented by Ambler and Logan [47]. The method starts with a first analysis of the gas composition in the gas collection bag. Then, a known volume of a reference gas, in this case, carbon dioxide, is added to the gas collection bag and the composition of the resulting mixture is analysed again. From these two analyses and proper mass balances, the total initial volume in the bag (Vtotal, initial) can be calculated using equation (7).     Vadded; CO2 1  xrun 2; CO2 þ Vrun 1 xrun 2; CO2  xrun 1; CO2 Vtotal; initial ¼ xrun 2; CO2  xrun 1; CO2 Fig. 2 e Nyquist plot of EIS with the corresponding (7) elements. 15402 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 where Vadded,CO2 is the known volume of carbon dioxide Besides that, the Pt@rGO (Fig. S5) showed finer crystals, which added, Vrun 1 is the volume obtained for the first GC analysis, is a preferred scenario since the electrode coating is more and xrun i, CO2 is the molar fraction of carbon dioxide in the efficient. The ink with finer crystals should be less prone to analysis number i. come off when entering in contact with the mineral medium and, thus, should be more stable. Moreover, the Pt@Graphi- tene (Fig. S6) covered less cathodic area because of the larger Results and discussion crystals. Also, Pt detachment was qualitatively observed with the naked eye since the medium became darker. Finally, Assessment of Pt functionalization larger crystals resulted in a more scattered global Pt distri- bution on the cathodic surface. According to the SEM obser- In a first step, both carbonaceous materials, i.e. Graphitene vations, Pt@rGO would be preferred as it is easier to disperse and rGO, were functionalized with Pt. The functionalization and less prone to detachment. yield, i.e. the total amount of Pt adhered over each material, Furthermore, the electron diffraction and the crystallinity was quantified by quintuplicate through ICP-OES. The per- of the different materials were studied through TEM (Fig. 5). centage of Pt functionalization (i.e. the amount of Pt in the The Pt@rGO electron diffraction showed a well-ordered ma- nanomaterial) was 18.4 ± 2.4% for Pt@rGO and 19.7 ± 5.0% for terial with a perfect hexagonal structure (Fig. 5a), which Pt@Graphitene. Thus, Pt@Graphitene had a slightly higher indicated the presence of only a few layers of material. On the average content but also a higher standard deviation. The other hand, Pt@Graphitene showed a crystal structure (Fig. 5b) success of this Pt-based functionalization is already a relevant but the hexagon was not visible. Pt@Graphitene showed many outcome of this work since the obtained Pt@rGO and messy layers and the TEM electron beam could not be focused Pt@Graphitene showed a higher proportion of Pt than other properly, showing a more amorphous material. Pt@Graphi- commercial inks used as catalyst in MECs (z10% Pt content) tene showed a circular figure composed of many hexagons in [48]. In addition, the functionalization percentages are up to different positions, each one corresponding to a crystalline three-fold higher than those reported for the functionaliza- plate and, therefore, to a sheet of material. Thus, the higher tion of similar materials (Table 1). the number of material layers, the less superficial area, and The morphological characteristics of the internal side of thus, the lower current density. Furthermore, the distance the cathode, i.e. the side in contact with the solution, were between planes was extremely similar (around 1.11 nm in studied using SEM. Fig. 3 shows the SEM images of the Pt both cases), even though Pt@rGO had more defects than distribution of both synthesized powder materials before they Pt@Graphitene. were deposited in the carbon cloth. Pt@rGO are tiny crystals The size of the Pt nanoparticles was also studied by TEM. with white dots (corresponding to Pt). Contrarily, Pt@Graphi- The nanoparticles size varied between 8 and 38 nm for Pt@rGO tene presents larger crystals also displaying the Pt-related (Fig. S7) and 12 and 87 nm for Pt@Graphitene (Fig. S8) with an white dots. Pt@rGO crystals are more fragmented than average size of 23 ± 5 nm and 35 ± 12 nm for Pt@rGO (n ¼ 100) Pt@Graphitene crystals and, therefore, they would be a priori and Pt@Graphitene (n ¼ 100), respectively. This average size easier to disperse. was higher than that obtained in previously reported gra- Both carbonaceous materials functionalized with Pt, i.e. phene functionalization [22] but lower than the Pt size in a Pt@Graphitene and Pt@rGO, were used to prepare two commercial ink [57]. The lower the size of the nanoparticles, different cathodes. Fig. 4 shows the SEM images of the func- the higher the superficial area and, thus, the higher avail- tionalized materials when deposited on the carbon cloth ability of Pt (where the electrochemical reaction takes place). surface. The Pt and C distribution was evaluated through SEM- A BET analysis was conducted to evaluate the surface area and EDX through a mapping of the surface (Figs. S5 and S6). Pt was pore size distribution of the different inks before and after homogeneously distributed in both cases and, thus, it can their Pt functionalization. The surface area results were effectively be used for H2 evolution reaction as a catalyst. 12.39 m2/g and 24.40 m2/g for rGO and Pt@rGO, respectively, Table 1 e Comparison of the yield of different functionalization in carbon materials based on graphene. Carbon material Functionalized with % (wt) of functionalization Ref. Pt@Graphitene Pt 19.7 ± 5.0 This work Pt@rGO Pt 18.4 ± 2.4 This work Graphene Pt 16.61 [23] rGO Pt 7.41 [49] GO Pt 3.61 [39] Graphene Fe@Au 13 [50] Graphene Nitrogen 9.5 [51] Graphene Nitrogen 5.74 [52] Graphene Nitrogen <2 (9% atomic percentage) [53] Graphene Nitrogen 8.9 [54] Graphene Porphyrin 5 [55] Graphene Poly (acrylonitrile) 1 [56] Poly (methylmethacrylate) 0.05 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 15403 Fig. 3 e SEM images of both graphenes functionalized: Pt@rGO (a) and Pt@Graphitene (b). Fig. 4 e SEM images of the cathodes with ink based on Pt@rGO (a) and Pt@Graphitene (b). Fig. 5 e Electron diffraction of Pt@rGO and Pt@Graphitene. and 5.80 m2/g and 10.76 m2/g for Graphitene and Pt@Graphi- content in Pt@rGO. Besides, Pt nanoparticles were smaller in tene, respectively. The superficial area increased when the Pt@rGO leading to a more homogenous distribution and a materials were functionalized because Pt induced a separa- higher electroactive area. tion of the layers, thereby raising the active area. In addition, Finally, the oxidation state of Pt in both materials was rGO had a higher initial surface area than Graphitene since analysed using XPS. Both Pt@rGO and Pt@Graphitene pre- rGO had more defects in its structure, such as carbonyl, hy- sented peaks in 71.00 and 74.95 eV with similar intensities, droxyl, or phenol groups. Moreover, rGO had a lower amount which corresponds to metallic Pt and PtO2 [58] (see Fig. S9). of graphene per gram of Pt@Graphitene while a higher surface The Pt functionalization process was reproducible and both area was found for Pt@rGO compared to that of Pt@Graphi- materials had a similar amount of Pt and a similar proportion tene. Comparing Figs. S5 and S6, a large number of defects can of metallic Pt and PtO2. Metallic Pt is the catalyst of the be observed in the Pt@rGO structure compared to that of hydrogen evolution reaction and, therefore, 50% of the Pt on Pt@Graphitene, which was related with a higher oxygen the cathode could be used as catalyst. This value is in the high- 15404 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 end of the range of other reported Pt-based cathodic mate- Table 2 e Resistances obtained in the simplified Randles rials. For instance, the commonly-used commercial ink pre- equivalent circuit in cells with Pt@rGO, Pt@Graphitene sented peaks in 71.93 and 74.99 eV, indicating that the Pt and standard cathodes. distribution was approximately, 50% of PtO2, 25% of PtO and RS (U) Rct (U) 25% of metallic Pt. Thereby, both Pt@rGO and Pt@Graphitene had around two-fold proportion of metallic Pt than the com- Pt@rGO 2.9 151.0 Pt@Graphitene 2.8 266.1 mercial ink and, thus, higher catalytic activity. Pt/C Commercial 10.5 788.0 Electrochemical characterization: CV and EIS measurements transducer capacity of rGO and Graphitene improved CVs were conducted with both materials (Fig. S10) to deter- compared with other carbon materials leading to a reduction mine the electroactive area using the intensity of the oxida- on cell resistance. Resistance for both Pt-functionalized tion peaks (positive intensity values) and equation (2), the coatings was lower than that reported for other carbon ma- current peak. The CVs showed two peaks related to the terials, i.e. graphene oxide modified with non-imprinted oxidation and reduction of iron. The electroactive area of polymer provided 3974 U [59] while a non-functionalized Pt@rGO and Pt@Graphitene corresponded to 1.14 and 0.69 cm2, graphene surface showed a Rct of around 6000 U [60]. Conse- respectively, confirming the higher availability of Pt in quently, such reduction on cell resistance in the materials Pt@rGO, which correspond to 0.16 cm2/cm2 and 0.1 cm2/cm2, studied herein caused such increase in performance of the respectively, once normalized with respect to the surface area cells. of the electrode. These values are lower compared with the These novel nanomaterials reported herein facilitated the geometric area (7 cm2) but a 13% higher compared with the electrochemical reaction to take place on the electrode surface commercial ink (0.60 cm2). due to a higher Pt availability and an improved electronic Fig. 6 shows the Nyquist diagrams for the EIS measurements transfer of material. Furthermore, compared with other BES, for the three Pt-based cathodes tested in this work. Table 2 our cells showed an order of magnitude lower cell resistances shows the characteristic EIS parameters obtained for each than other previously reported [61]. experiment. Rs (related to the ohmic resistance) were similar in both functionalized cells because both used the same solvent Performance assessment of the cells and electrode, the major contributors. Regarding Rct (related to the charge resistance), Pt@rGO showed less than half the Rct of Eight MECs were built using cathodes coated with the Pt@Graphitene because the Pt distribution on the cathode was different carbonaceous materials manufactured in this work more homogeneous and the ink was less scattered throughout (rGO and Graphitene, and Pt@rGO and Pt@Graphitene) and its domain. Consequently, the electronic transfer was favoured. were operated for more than two months under cyclic, re- Furthermore, Pt@rGO showed a slightly higher surface area for petitive conditions. Fig. 7 shows the current density evolution the electrochemical reaction to take place. obtained for these MECs during the last 18 days of operation Over a 250% higher resistance was found for the cell with (i.e. under pseudo steady-state conditions). The cells with Pt/C commercial because of the higher carbon content of the cathodes containing Pt@rGO and Pt@Graphitene provided an cathode coated with the Pt/C ink and because of its lower Pt average current density of 0.78 and 0.71 mA/cm2, respectively, availability, as the amount of ink needed to obtain the same whereas these values decreased to 0.25 and 0.22 mA/cm2 for ratio of Pt/cm2 was around 50% higher. Besides that, the rGO and Graphitene, respectively, due to the lack of Pt functionalization. Current densities obtained remained relatively stable during the operation of the cells (more than two months). No catalyst detachment or corrosion was detected that could have affected the performance of the material during the period of operation and, therefore, the material was not only structurally stable, as indicated in section: Assessment of Pt functionalization (Appendix A), but also had stable electrical and chemical properties. Regarding the 10% higher average current density in Pt@rGO with respect to Pt@Graphitene cathode, differences are explained by several factors: (i) a higher Pt@rGO surface area (12.39 m2/g vs. 10.76 m2/g) as the powder-like material has a larger specific surface area than crystals, (ii) a higher electroactive area of Pt@rGO (1.14 vs. 0.69 cm2), (iii) the Pt in Pt@rGO is less prone to be detached (Fig. 3), and (iv) Pt@rGO crystals are more fragmented, so, the ink and, subsequently, Pt, are more homogeneously distributed (Fig. 4). Fig. 6 e EIS of different carbon materials functionalized The cells with cathodes coated with commercial ink with Pt, obtained using a frequency range of 100 kHz to showed an intermediate performance (0.34 mA/cm2). Previous 10 mHz. Randles circuit: Rs·(Rct·Cdl) (inset). 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 15405 Fig. 8 e The experimental value of CE, rCAT, rE, rS, rE þ S, and H2 production of Pt@rGO and Pt@Graphitene. Error bar is the standard deviation for n ¼ 3 measurements. Fig. 7 e Evolution of the current intensity during the last 18 cathodes perform better [62e64] than sole carbon-based days of stable operation for the different MEC tested with cathodes [66]. The materials tested in this work were Pt- rGO, Graphitene, Pt@rGO and Pt@Graphitene, and standard coated and carbon-based, which increased the cathode per- cathode with commercial ink. formance. Remarkably, Pt@rGO provided the highest values reported (0.78 mA/cm2) and Pt@Graphitene also reached a high value (0.71 mA/cm2), similar to the peak value reported results with the same cells and the commercial ink reached for NieP coated Ni-foam (0.73 mA/cm2) [69]. intensities lower than 0.50 mA/cm2 [62]. The commercial ink Fig. 8 shows the conventional MEC performance parame- contains 10% Pt/90% C and the amount of commercial ink ters for the cells with the Pt-functionalized cathode. Both cells added was calculated so that the commercial and the manu- resulted in an unexpected excessive CE (more than 100%) and factured materials had the same amount of Pt. Thus, both too low rCAT (less than 60%). Unrealistic CE and rCAT results Pt@rGO and Pt@Graphitene cells reached, for instance, in- indicate that some of the H2 generated was scavenged by tensities twice higher than these in other works (ranging from microorganisms: CE higher than 100% indicates H2-recycling, 0.03 to 0.64 mA/cm2) [63,64]. whereas rCAT around 50% or less suggests H2 losses probably Furthermore, Fig. 7 shows as rGO and Graphitene without because of methanogenesis or homoacetogenesis. This Pt also produced current density meaning that these materials observation is very common in single-chamber cells [8]. In can catalyse the reaction in spite of having worse behavior these cases, CE and rCAT are not the most realistic option to than the functionalized graphenes since Pt is one of the best evaluate MEC performance. and most widely used H2 catalyst. CE was 114 ± 1% for Pt@rGO and 106 ± 6% for Pt@Graphi- Table 3 compares current densities found in the literature tene. Hence, the Pt@rGO cell showed a slightly higher H2- for H2 production in similar cells but using different cathodic recycling. Such CEs are higher than those reported in previous materials. These results clearly show that the metal-coated works [70], which had an average CE between 50 and 74%. For Table 3 e Comparison of the current density obtained with different cathode materials. Cathode material Current density (mA/cm2) Ref. Carbon cloth with Pt@rGO 0.78 This work Carbon cloth with Pt@Graphitene 0.71 This work Granular graphite 0.48 [65] Carbon mesh with activated carbon <0.1 [66] Carbon cloth with Pt 0.35 [62] 0.64 [63] RuO2 coated TiO metal 0.35 [64] Molybdenum disulfide coated stainless steel mesh 0.26 [67] Nickel foam 0.24 [67] Stainless steel wool 0.17 [67] Stainless steel woven mesh 0.48 [68] NieP coated Ni foam 0.73 [69] Platinum coated stainless steel mesh 0.34 [67] 0.62 [69] 15406 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 instance, high CEs have been reported in low-scale lab sys- cathodes and with cathodes coated with the commonly used tems. For example, the CE reached 82% in an 18 mL MEC [71]. commercial ink. Pt@rGO and Pt@Graphitene had an average On the other hand, a pilot-scale MEC fed with domestic current density of 0.78 and 0.71 mA/cm2 vs. 0.25 and 0.22 mA/ wastewater reached a CE of 41% [72] working with a 100 L MEC cm2 for rGO and Graphitene because of the higher proportion or 55% working with a 120 L MEC [73]. of metallic Pt (50%) and lower internal resistances of Pt- Regarding the rCAT, 56 ± 12% and 58 ± 9% were obtained for functionalized cathodes. Pt@rGO and Pt@Graphitene, respectively, indicating potential Pt@rGO and Pt@Graphitene provided a higher conductivity H2 consumption. Despite that, the H2 production rate (HPR) and a higher active area, thus, a better performance than was 1.1 ± 0.2 m3 m3 d1 and 0.9 ± 0.3 m3 m3 d1 in the cells conventional inks used in coated cathodes. These function- with a Pt@rGO-coated cathode and with a Pt@Graphitene- alized conductive materials open new opportunities for on- coated cathode, respectively. Since the cell with the Pt/C demand modification of more efficient catalysts, being able commercial coating did not have H2-recycling, only a 28% to modify both the ink composition and the catalyst itself. lower HPR than that of Pt@rGO or Pt@Graphitene was measured. The rCAT in the case of the Pt/C commercial coating was higher because the H2 losses increase when the HPR is Declaration of competing interest improved. If no H2 losses were present (i.e. if rCAT was 100%), the HPR obtained for the Pt@rGO-coated cathode would have The authors declare that they have no known competing been around 2.25 m3 m3$day1, which is much higher than financial interests or personal relationships that could have that obtained with the commercial ink coating in our previous appeared to influence the work reported in this paper. works with the same configuration [43]. Other works with a similar single-chamber MEC configuration and a carbon cloth cathode treated with a commercial ink to reach 0.5 mg Pt/cm2 Acknowledgments obtained an HPR of 1.99 m3 m3$day1 [74]. Furthermore, other authors obtained these low HPRs using a different substrate This work has been supported by the Spanish Ministry of [70] like pig slurry (0.079 m3 m3$day1) or anaerobic sludge Economy and Competitiveness (CTQ2017-82404-R) with funds (0.88 m3 m3$day1). Contrarily, since acetate corresponds to nchez from the European Regional Development Fund. Pilar Sa the substrate that generates the highest current density, ~ a is grateful for the predoctoral grant FI (2018FI_B01161) Pen similar HPRs to those found in our work were found in pre- from the Generalitat de Catalunya. All authors are members of vious works that used acetate as substrate [75]. With respect the GENOCOV research group (Grup de Recerca Consolidat de to the gas composition, Pt@rGO contained 76 ± 3% of H2, la Generalitat de Catalunya, 2017 SGR 1175, www.genocov. 17 ± 8% of CO2, and 7 ± 3% of CH4 whereas Pt@Graphitene  mez- com). We acknowledge the collaboration of Pedro Go composition was 81 ± 5% of H2, 16 ± 8% of CO2, and 3 ± 2% of Romero, Raul Benages and Carlos Marchante (who provided CH4. Thus, the relative composition H2/CH4 was 0.92 and 0.96 the graphitene for this study) and GraphCAT project (001-P- for Pt@rGO and Pt@Graphitene, respectively. Part of the H2 001702), cofunded by the ERDF allocated to the ERDF Opera- losses were attributed to methanogenesis. tional Program of Catalonia 2014-2020, with the support of the Generalitat de Catalunya. Conclusions Appendix A. Supplementary data This work reports the use of two promising nanomaterials, Pt@rGO and Pt@Graphitene, as coating materials for MECs Supplementary data to this article can be found online at cathodes. Both of them have similar performances regardless https://doi.org/10.1016/j.ijhydene.2022.03.078. of their synthesis procedure. The Pt-functionalization of these materials was optimized and the percentage of functionalization obtained by ICP-OES references (18.4 ± 2.4% for Pt@rGO and 19.7 ± 5.0% for Pt@Graphitene) were higher than metallic functionalization of similar carbo- naceous materials. The optical analyses of the different ma- [1] Sewsynker Y, Kana EBG, Lateef A. Modelling of biohydrogen terials showed that Pt@rGO crystals were finer and, thus, generation in microbial electrolysis cells (MECs) using a easier to disperse and less prone to detach when entering in committee of artificial neural networks (ANNs). Biotechnol Biotechnol Equip 2015;29:1208e15. https://doi.org/10.1080/ contact with the mineral medium. TEM images revealed a 13102818.2015.1062732. crystalline structure for both materials and showed that the [2] Haryanto A, Fernando S, Murali N, Adhikari S. Current status size of nanoparticles was lower in the case of Pt@rGO of hydrogen production techniques by steam reforming of (23 ± 5 nm) with respect to that of Pt@Graphitene (32 ± 11 nm). ethanol: a review. Energy Fuel 2005;19:2098e106. https:// Pt@rGO also showed better results than Pt@Graphitene in doi.org/10.1021/ef0500538. terms of surface area (12.39 m2/g vs. 10.76 m2/g) and electro- [3] Lee H-S, Vermaas WFJ, Rittmann BE. Biological hydrogen active area (1.14 vs. 0.69 cm2). As consequence, Pt@rGO production: prospects and challenges. Trends Biotechnol 2010;28:262e71. https://doi.org/10.1016/j.tibtech.2010.01.007. behavior was slightly better than that of Pt@Graphitene. Be- [4] Logan BE, Call D, Cheng S, Hamelers HVM, Sleutels THJA, sides, the cells with functionalized coatings had higher per- Jeremiasse AW, et al. Microbial electrolysis cells for high formance in MECs with respect to non-functionalized 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 15407 yield hydrogen gas production from organic matter. Environ catalyst-free cathode to produce hydrogen. Environ Sci Sci Technol 2008;42:8630e40. https://doi.org/10.1021/ Technol 2009;43:9525e30. https://doi.org/10.1021/es9022317. es801553z. [19] Selembo PA, Merrill MD, Logan BE. Hydrogen production [5] Gao Y, Ryu H, Santo Domingo JW, Lee H-S. Syntrophic with nickel powder cathode catalysts in microbial interactions between H2-scavenging and anode-respiring electrolysis cells. Int J Hydrogen Energy 2010;35:428e37. bacteria can improve current density in microbial https://doi.org/10.1016/j.ijhydene.2009.11.014. electrochemical cells. Bioresour Technol 2014;153:245e53. [20] Farhangi S, Ebrahimi S, Niasar MS. Commercial materials as https://doi.org/10.1016/j.biortech.2013.11.077. cathode for hydrogen production in microbial electrolysis [6] Torres CI, Marcus AK, Lee H-S, Parameswaran P, Krajmalnik- cell. Biotechnol Lett 2014;36:1987e92. https://doi.org/10.1007/ Brown R, Rittmann BE. A kinetic perspective on extracellular s10529-014-1565-7. electron transfer by anode-respiring bacteria. FEMS [21] Selembo PA, Merrill MD, Logan BE. The use of stainless steel Microbiol Rev 2010;34:3e17. https://doi.org/10.1111/j.1574- and nickel alloys as low-cost cathodes in microbial 6976.2009.00191.x. electrolysis cells. J Power Sources 2009;190:271e8. https:// [7] Sa nchez-Pen~ a P, Rodriguez J, Montes R, Baeza JA, Gabriel D, doi.org/10.1016/j.jpowsour.2008.12.144. Baeza M, et al. Less is more: a comprehensive study on the [22] Xin L, Yang F, Rasouli S, Qiu Y, Li ZF, Uzunoglu A, et al. effects of the number of gas diffusion layers on airecathode Understanding Pt nanoparticle anchoring on graphene microbial fuel cells. Chemelectrochem 2021;8:3416e26. supports through surface functionalization. ACS Catal https://doi.org/10.1002/celc.202100908. 2016;6:2642e53. https://doi.org/10.1021/acscatal.5b02722. [8] Ruiz Y, Baeza JA, Guisasola A. Revealing the proliferation of [23] Zhao S, Li Y, Yin H, Liu Z, Luan E, Zhao F, et al. Energy hydrogen scavengers in a single-chamber microbial Resources: three-dimensional graphene/Pt nanoparticle electrolysis cell using electron balances. Int J Hydrogen composites as freestanding anode for enhancing Energy 2013;38:15917e27. https://doi.org/10.1016/ performance of microbial fuel cells. Sci Adv 2015;1:e1500372. j.ijhydene.2013.10.034. https://doi.org/10.1126/sciadv.1500372. [9] Lee H-S, Torres CI, Parameswaran P, Rittmann BE. Fate of H2 [24] Li X, Liu X, Wang W, Li L, Lu X. High loading Pt nanoparticles in an upflow single-chamber microbial electrolysis cell using on functionalization of carbon nanotubes for fabricating a metal-catalyst-free cathode. Environ Sci Technol nonenzyme hydrogen peroxide sensor. Biosens Bioelectron 2009;43:7971e6. https://doi.org/10.1021/es900204j. 2014;59:221e6. https://doi.org/10.1016/j.bios.2014.03.046. [10] Lee H-S, Rittmann BE. Significance of biological hydrogen [25] Novoselov KS, Jiang D, Schedin F, Booth TJ, Khotkevich VV, oxidation in a continuous single-chamber microbial Morozov SV, et al. Two-dimensional atomic crystals. Proc electrolysis cell. Environ Sci Technol 2010;44:948e54. https:// Natl Acad Sci Unit States Am 2005;102:10451e3. https:// doi.org/10.1021/es9025358. doi.org/10.1073/pnas.0502848102. [11] Parameswaran P, Torres CI, Lee H-S, Rittmann BE, [26] Novoselov KS, Geim AK, Morozov SV, Jiang D, Zhang Y, Krajmalnik-Brown R. Hydrogen consumption in microbial Dubonos SV, et al. Electric field effect in atomically thin electrochemical systems (MXCs): the role of homo- carbon films. Science 2004;306:666e9. https://doi.org/ acetogenic bacteria. Bioresour Technol 2011;102:263e71. 10.1126/science.1102896. https://doi.org/10.1016/j.biortech.2010.03.133. [27] Rowley-Neale SJ, Randviir EP, Abo Dena AS, Banks CE. An [12] Rago L, Ruiz Y, Baeza JA, Guisasola A, Corte s P. Microbial overview of recent applications of reduced graphene oxide as community analysis in a long-term membrane-less a basis of electroanalytical sensing platforms. Appl Mater microbial electrolysis cell with hydrogen and methane Today 2018;10:218e26. https://doi.org/10.1016/ production. Bioelectrochemistry 2015;106. https://doi.org/ j.apmt.2017.11.010. 10.1016/j.bioelechem.2015.06.003. [28] Yu W, Sisi L, Haiyan Y, Jie L. Progress in the functional [13] Yang E, Omar Mohamed H, Park S-G, Obaid M, Al- modification of graphene/graphene oxide: a review. RSC Adv Qaradawi SY, Castan ~ o P, et al. A review on self-sustainable 2020;10:15328e45. https://doi.org/10.1039/d0ra01068e. microbial electrolysis cells for electro-biohydrogen [29]  M, Vla Ionit‚a  sceanu GM, Watzlawek AA, Voicu SI, Burns JS, production via coupling with carbon-neutral renewable Iovu H. Graphene and functionalized graphene: energy technologies. Bioresour Technol 2021;320:124363. extraordinary prospects for nanobiocomposite materials. https://doi.org/10.1016/j.biortech.2020.124363. Compos B Eng 2017;121:34e57. https://doi.org/10.1016/ [14] Dange P, Pandit S, Jadhav D, Shanmugam P, Gupta PK, j.compositesb.2017.03.031. Kumar S, et al. Recent developments in microbial electrolysis [30] Li M, Zhou S, Xu M. Graphene oxide supported magnesium cell-based biohydrogen production utilizing wastewater as a oxide as an efficient cathode catalyst for power generation feedstock. Sustainability 2021;13:8796. https://doi.org/ and wastewater treatment in single chamber microbial fuel 10.3390/su13168796. cells. Chem Eng J 2017;328:106e16. https://doi.org/10.1016/ [15] Rousseau R, Ketep SF, Etcheverry L, De lia M-L, Bergel A. j.cej.2017.07.031. Microbial electrolysis cell (MEC): a step ahead towards [31] Yaqoob AA, Ibrahim MNM, Yaakop AS, Umar K, Ahmad A. hydrogen-evolving cathode operated at high current density. Modified graphene oxide anode: a bioinspired waste material Bioresour Technol Rep 2020;9:100399. https://doi.org/10.1016/ for bioremediation of Pb2þ with energy generation through j.biteb.2020.100399. microbial fuel cells. Chem Eng J 2021;417:128052. https:// [16] Guo K, Prevoteau A, Rabaey K. A novel tubular microbial doi.org/10.1016/j.cej.2020.128052. electrolysis cell for high rate hydrogen production. J Power [32] Xin S, Shen J, Liu G, Chen Q, Xiao Z, Zhang G, et al. Electricity Sources 2017;356:484e90. https://doi.org/10.1016/ generation and microbial community of single-chamber j.jpowsour.2017.03.029. microbial fuel cells in response to Cu2O nanoparticles/ [17] Guisasola A, Baeza JA, Marone A, Trably E,  Bernet N. reduced graphene oxide as cathode catalyst. Chem Eng J Opportunities for hydrogen production from urban/ 2020;380:122446. https://doi.org/10.1016/j.cej.2019.122446. industrial wastewater in bioelectrochemical systems. [33] Georgakilas V, Otyepka M, Bourlinos AB, Chandra V, Kim N, Microb. Electrochem. Technol. Boca Raton: CRC Press; 2020. Kemp KC, et al. Functionalization of graphene: covalent and p. 225e43. https://doi.org/10.1201/9780429487118-15. non-covalent approaches, derivatives and applications. [18] Chae K-J, Choi M-J, Kim K-Y, Ajayi FF, Chang I-S, Kim IS. A Chem Rev 2012;112:6156e214. https://doi.org/10.1021/ solar-powered microbial electrolysis cell with a platinum cr3000412. 15408 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 [34] Xu M, Liang T, Shi M, Chen H. Graphene-like two- [50] Kuila T, Bose S, Mishra AK, Khanra P, Kim NH, Lee JH. dimensional materials. Chem Rev 2013;113:3766e98. https:// Chemical functionalization of graphene and its applications. doi.org/10.1021/cr300263a. Prog Mater Sci 2012;57:1061e105. https://doi.org/10.1016/ [35] Hou Y, Lv S, Liu L, Liu X. High-quality preparation of j.pmatsci.2012.03.002. graphene oxide via the Hummers' method: understanding [51] Chen Y, Xie B, Ren Y, Yu M, Qu Y, Xie T, et al. Designed the roles of the intercalator, oxidant, and graphite particle nitrogen doping of few-layer graphene functionalized by size. Ceram Int 2020;46:2392e402. https://doi.org/10.1016/ selective oxygenic groups. Nanoscale Res Lett 2014;9:646. j.ceramint.2019.09.231. https://doi.org/10.1186/1556-276X-9-646. [36] Mun ~ oz J, Brennan LJ, Ce spedes F, Gun’ko YK, Baeza M. [52] Wen Q, Wang S, Yan J, Cong L, Chen Y, Xi H. Porous nitrogen- Characterization protocol to improve the electroanalytical doped carbon nanosheet on graphene as metal-free catalyst response of graphene-polymer nanocomposite sensors. for oxygen reduction reaction in air-cathode microbial fuel Compos Sci Technol 2016;125:71e9. https://doi.org/10.1016/ cells. Bioelectrochemistry 2014;95:23e8. https://doi.org/ j.compscitech.2016.01.018. 10.1016/j.bioelechem.2013.10.007. [37] Montpart N, Rago L, Baeza JA, Guisasola A. Hydrogen [53] Leela A, Reddy M, Srivastava A, Gowda SR, Gullapalli H, production in single chamber microbial electrolysis cells Dubey M, et al. Synthesis of nitrogen-doped graphene films with different complex substrates. Water Res For Lithium Battery Application. ACS Nano 2020;4:18. https:// 2015;68:601e15. https://doi.org/10.1016/j.watres.2014.10.026. doi.org/10.1021/nn101926g. [38] Paredes JI, Villar-Rodil S, Martı́nez-Alonso A, Tasco  n JMD. [54] Wei D, Liu Y, Wang Y, Zhang H, Huang L, Yu G. Synthesis of Graphene oxide dispersions in organic solvents. Langmuir N-doped graphene by chemical vapor deposition and its 2008;24:10560e4. https://doi.org/10.1021/la801744a. electrical properties. Nano Lett 2009;9:1752e8. https:// [39] Liu R, Hu J, Zhu S, Lu J, Zhu H. Synergistically enhanced doi.org/10.1021/nl803279t. optical limiting property of graphene oxide hybrid materials [55] Zhu M, Li Z, Xiao B, Lu Y, Du Y, Yang P, et al. Surfactant functionalized with Pt complexes. ACS Appl Mater Interfaces assistance in improvement of photocatalytic hydrogen 2017;9:33029e40. https://doi.org/10.1021/acsami.7b10585. production with the porphyrin noncovalently [40] Baeza JA, Martı́nez-Miro  Guerrero J, Ruiz Y, Guisasola A.  A, functionalized graphene nanocomposite. ACS Appl Mater Bioelectrochemical hydrogen production from urban Interfaces 2013;5:1732e40. https://doi.org/10.1021/ wastewater on a pilot scale. J Power Sources 2017;356:500e9. am302912v. https://doi.org/10.1016/j.jpowsour.2017.02.087. [56] Ramanathan T, Abdala AA, Stankovich S, Dikin DA, Herrera- [41] Borole AP, Aaron D, Hamilton CY, Tsouris C. Understanding Alonso M, Piner RD, et al. Functionalized graphene sheets for long-term changes in microbial fuel cell performance using polymer nanocomposites. Nat Nanotechnol 2008;3:327e31. electrochemical impedance spectroscopy. Environ Sci https://doi.org/10.1038/nnano.2008.96. Technol 2010;44:2740e5. https://doi.org/10.1021/es9032937. [57] Kocha SS, Zack JW, Alia SM, Neyerlin KC, Pivovar BS. [42] Rodrı́guez J, Mais L, Campana R, Piroddi L, Mascia M, Influence of ink composition on the electrochemical Gurauskis J, et al. Comprehensive characterization of a cost- properties of Pt/C electrocatalysts. ECS Trans effective microbial fuel cell with Pt-free catalyst cathode and 2013;50:1475e85. https://doi.org/10.1149/05002.1475ecst. slip-casted ceramic membrane. Int J Hydrogen Energy 2021. [58] Sealy C. The problem with platinum. Mater Today https://doi.org/10.1016/j.ijhydene.2021.01.066. 2008;11:65e8. https://doi.org/10.1016/S1369-7021(08)70254-2. [43] He Z, Mansfeld F. Exploring the use of electrochemical [59] Liu W, Ma Y, Sun G, Wang S, Deng J, Wei H. Molecularly impedance spectroscopy (EIS) in microbial fuel cell studies. imprinted polymers on graphene oxide surface for EIS Energy Environ Sci 2009;2:215e9. https://doi.org/10.1039/ sensing of testosterone. Biosens Bioelectron 2017;92:305e12. B814914C. https://doi.org/10.1016/j.bios.2016.11.007. [44] Manohar AK, Bretschger O, Nealson KH, Mansfeld F. The [60] Bonanni A, Loo AH, Pumera M. Graphene for impedimetric polarization behavior of the anode in a microbial fuel cell. biosensing. TrAC Trends Anal Chem (Reference Ed) Electrochim Acta 2008;53:3508e13. https://doi.org/10.1016/ 2012;37:12e21. https://doi.org/10.1016/j.trac.2012.02.011. j.electacta.2007.12.002. [61] Kretzschmar J, Harnisch F. Electrochemical Impedance [45] Mun ~ oz J, Montes R, Baeza M. Trends in electrochemical Spectroscopy on biofilm electrodes e conclusive or impedance spectroscopy involving nanocomposite euphonious? Curr Opin Electrochem 2021:100757. https:// transducers: characterization, architecture surface and bio- doi.org/10.1016/j.coelec.2021.100757. sensing. TrAC Trends Anal Chem 2017;97:201e15. https:// [62] Ruiz Y, Baeza JA, Guisasola A. Microbial electrolysis cell doi.org/10.1016/j.trac.2017.08.012. performance using non-buffered and low conductivity [46] Zhu P, Zhao Y. Cyclic voltammetry measurements of wastewaters. Chem Eng J 2016;289:341e8. https://doi.org/ electroactive surface area of porous nickel: peak current and 10.1016/j.cej.2015.12.098. peak charge methods and diffusion layer effect. Mater Chem [63] Kiely PD, Cusick RD, Call DF, Selembo PA, Regan JM, Phys 2019;233:60e7. https://doi.org/10.1016/ Logan BE. Anode microbial communities produced by j.matchemphys.2019.05.034. changing from microbial fuel cell to microbial electrolysis [47] Ambler JR, Logan BE. Evaluation of stainless steel cathodes cell operation using two different wastewaters. Bioresour and a bicarbonate buffer for hydrogen production in Technol 2011;102:388e94. https://doi.org/10.1016/ microbial electrolysis cells using a new method for j.biortech.2010.05.019. measuring gas production. Int J Hydrogen Energy [64] Pasupuleti SB, Srikanth S, Venkata Mohan S, Pant D. 2011;36:160e6. https://doi.org/10.1016/j.ijhydene.2010.09.044. Development of exoelectrogenic bioanode and study on [48] Middaugh J, Cheng S, Liu W, Wagner R. How to make feasibility of hydrogen production using abiotic VITO- cathodes with a diffusion layer for single-chamber microbial CoRETM and VITO-CASETM electrodes in a single chamber fuel cells. 2008. microbial electrolysis cell (MEC) at low current densities. [49] Mirzaei A, Bang JH, Choi MS, Han S, Lee HY, Kim SS, et al. Bioresour Technol 2015;195:131e8. https://doi.org/10.1016/ Changes in characteristics of Pt-functionalized RGO j.biortech.2015.06.145. nanocomposites by electron beam irradiation for room [65] Cerrillo M, Vin~ as M, Bonmatı́ A. Anaerobic digestion and temperature NO2 sensing. Ceram Int 2020;46:21638e46. electromethanogenic microbial electrolysis cell integrated https://doi.org/10.1016/j.ceramint.2020.05.271. system: increased stability and recovery of ammonia and 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 7 ( 2 0 2 2 ) 1 5 3 9 7 e1 5 4 0 9 15409 methane. Renew Energy 2018;120:178e89. https://doi.org/ application, advance and challenge. J Chem Technol 10.1016/j.renene.2017.12.062. Biotechnol 2019;94:1697e711. https://doi.org/10.1002/ [66] Luo S, Liu F, Fu B, He K, Yang H, Zhang X, et al. Onset jctb.5898. investigation on dynamic change of biohythane generation [71] Call DF, Wagner RC, Logan BE. Hydrogen production by and microbial structure in dual-chamber versus single- geobacter species and a mixed consortium in a microbial chamber microbial electrolysis cells. Water Res electrolysis cell. Appl Environ Microbiol 2009;75:7579e87. 2021;201:117326. https://doi.org/10.1016/ https://doi.org/10.1128/AEM.01760-09. j.watres.2021.117326. [72] Heidrich ES, Edwards SR, Dolfing J, Cotterill SE, Curtis TP. [67] Ribot-Llobet E, Nam J-Y, Tokash JC, Guisasola A, Logan BE. Performance of a pilot scale microbial electrolysis cell fed on Assessment of four different cathode materials at different domestic wastewater at ambient temperatures for a 12 initial pHs using unbuffered catholytes in microbial month period. Bioresour Technol 2014;173:87e95. https:// electrolysis cells. Int J Hydrogen Energy 2013;38:2951e6. doi.org/10.1016/j.biortech.2014.09.083. https://doi.org/10.1016/j.ijhydene.2012.12.037. [73] Heidrich ES, Dolfing J, Scott K, Edwards SR, Jones C, Curtis TP. [68] Zhou R, Zhou S, He C. Quantitative evaluation of effects of Production of hydrogen from domestic wastewater in a pilot- different cathode materials on performance in Cd(II)- scale microbial electrolysis cell. Appl Microbiol Biotechnol reduced microbial electrolysis cells. Bioresour Technol 2013;97:6979e89. https://doi.org/10.1007/s00253-012-4456-7. 2020;307:123198. https://doi.org/10.1016/ [74] Call D, Logan BE. Hydrogen production in a single chamber j.biortech.2020.123198. microbial electrolysis cell lacking a membrane. Environ Sci [69] Li F, Liu W, Sun Y, Ding W, Cheng S. Enhancing hydrogen Technol 2008;42:3401e6. https://doi.org/10.1021/es8001822. production with NieP coated nickel foam as cathode catalyst [75] Kadier A, Kalil MS, Abdeshahian P, Chandrasekhar K, in single chamber microbial electrolysis cells. Int J Hydrogen Mohamed A, Azman NF, et al. Recent advances and emerging Energy 2017;42:3641e6. https://doi.org/10.1016/ challenges in microbial electrolysis cells (MECs) for microbial j.ijhydene.2016.10.163. production of hydrogen and value-added chemicals. Renew [70] Hua T, Li S, Li F, Zhou Q, Ondon BS. Microbial electrolysis cell Sustain Energy Rev 2016;61:501e25. https://doi.org/10.1016/ as an emerging versatile technology: a review on its potential j.rser.2016.04.017.