Oxygen Reduction Reaction with Manganese Oxide Nanospheres in Microbial Fuel Cells.

Operating microbial fuel cells (MFCs) under extreme pH conditions offers a substantial benefit. Acidic conditions suppress the growth of undesirable methanogens and increase redox potential for oxygen reduction reactions (ORRs), and alkaline conditions increase the electrocatalytic activity. However, operating any fuel cells, including MFCs, is difficult under such extreme pH conditions. Here, we demonstrate a pH-universal ORR ink based on hollow nanospheres of manganese oxide (h-Mn3O4) anchored with multiwalled carbon nanotubes (MWCNTs) on planar and porous forms of carbon electrodes in MFCs (pH = 3-11). Nanospheres of h-Mn3O4 (diameter ∼ 31 nm, shell thickness ∼ 7 nm) on a glassy carbon electrode yielded a highly reproducible ORR activity at pH 3 and 10, based on rotating disk electrode (RDE) tests. A phenomenal ORR performance and long-term stability (∼106 days) of the ink were also observed with four different porous cathodes (carbon cloth, carbon nanofoam paper, reticulated vitreous carbon, and graphite felt) in MFCs. The ink reduced the charge transfer resistance (R ct) to the ORR by 100-fold and 45-fold under the alkaline and acidic conditions, respectively. The current study promotes ORR activity and subsequently the MFC operations under a wide range of pH conditions, including acidic and basic conditions.

Introduction

Oxygen reduction reactions (ORRs) sustain performances of living systems (e.g., plants, algal cells) and energy harvesting fuel cell devices including microbial fuel cells (MFCs). MFCs use electrochemically active microorganisms (exoelectrogens) for converting organic matter in waste streams into direct current(DC) electricity. Although advanced designs involve a stack of rectangular or vertical modules with interchangeable cassette electrodes, a basic MFC unit consists of an anode (e– donor: organic matter), cathode (e– acceptor: oxygen), and optional membrane. A bottleneck to MFC operation includes high overpotential and sluggish ORR kinetics on the cathode surfaces. Catalysts based on platinum-group metals can effectively overcome these challenges. However, their high costs restrict their use to automobiles and space mission applications.

Operating MFCs under extreme pH conditions offers a substantial benefit. Such MFCs treat acidic wastes from mining, anaerobic digestors, and distillery wastewater, as well as alkaline wastes from manufacturing processes (e.g., animal hide, paper, and cement products) and electroplating industry.[1−5] Acidic conditions (pH < 4.0) suppress the growth of methanogens, expand the pH gradient, and accelerate the charge transport (proton) through the cation exchange membranes (e.g., pKa for protonated perfluorinated sulfonic acid in Nafion is less than one). Acidic conditions increase redox potential for the ORRs (+59 mV for each pH unit reduction). On the other hand, alkaline conditions increase electrocatalytic activity and enable the use of a range of affordable materials for fabricating MFC components (e.g., bipolar plates, membranes).

It is difficult to operate any fuel cells, including MFCs, under extreme pH conditions. The anode and cathode compartments are ionically connected through the ion exchange membranes. Thus, the extreme pH conditions in the anode can alter pH conditions in the cathode. The resulting membrane pH gradient, caused by the acidification of the anode and alkalization of the cathode, imposes operational challenges. Alkaline conditions create OH– that adsorbs onto the cathode surfaces, blocking the O2 adsorption as well as promoting the outer-sphere electron-transfer mechanism, resulting in an undesirable 2e– peroxide as an intermediate.[6] CO2 dissolves in alkaline environments, including those in hydrated anion exchange membranes (AEMs), causing cell carbonation and voltage losses (100–400 mV).[7] There is a clear need to develop a pH-universal catalyst for sustaining ORR reactions under extreme pH conditions for MFC applications (pH = 3–10).

Earth-abundant materials, M–N–C materials, where M = earth-abundant transition metals (e.g., nickel and manganese), N = nitrogen, C = carbon, can yield viable ORR catalysts for MFC applications. Catalysts based on the MnO nanoparticles have emerged as attractive choices for electrochemical devices.[8−13] The MnO nanoparticles can be obtained in various shapes and sizes, as needed by the devices.[14−16] We present a first study that explores the ORR activity of hollow h-Mn3O4 nanospheres[17,18] under extreme conditions in MFCs. We present a scalable route for synthesizing the ink based on nanospheres, anchor them with MWCNTs, and deposit them on diverse electrodes, including a planar glassy carbon and four porous carbon electrodes. A series of electrochemistry tests based on RDE and MFCs was used to assess the ORR performances. Microscopy and spectroscopy tests, rotating disk electrode (RDE) tests, electrochemical impedance spectroscopy analysis, and MFC tests were used to assess the long-term stability and the ORR performance of the ink under extreme pH conditions.

Results and Discussion

We synthesized nanospheres of h-Mn3O4 using the Kirkendall growth process and anchored onto the MWCNTs in the presence of Nafion (herein referred to as catalyst ink). The commercially available carbon electrodes, including glassy carbon, graphite felt (GF), reticulated vitreous carbon (RVC), carbon cloth (CC), and carbon paper (CP), were modified with the catalyst ink and vacuum-dried at 200 °C. Details of the synthesis of MnO, h-Mn3O4 nanocrystals, ink, and coated electrodes are discussed in the later sections. As seen in the powder X-ray diffraction (PXRD) data (Figure a), the as-synthesized MnO particles were converted effectively into the h-Mn3O4 particles (hausmannite, JCPDS card no. 24-0734). Our earlier studies demonstrated the hollow morphology of the h-Mn3O4 nanospheres based on the high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) and TEM image analyses (Figure b). These nanospheres were characterized by an inner diameter of ∼20 nm, outer diameter of 30–40 nm, and shell thickness of ∼7 nm.[17,18] An immediate application of as-synthesized ink is its use as a noninvasive coating to promote the ORR activity of carbon electrodes in MFCs. As seen in Figure , both the planar and porous forms of electrodes were amenable to the modification by the ink. Considering the diverse surface properties of these electrodes (see Table for key surface properties), these findings indicate the use of ink for a diverse range of electrodes. As shown in the optical images for the CP, CC, h-Mn3O4/CP, and h-Mn3O4/CC, the coated and uncoated electrodes displayed distinct differences in terms of light interferences (Figure c). However, they displayed a varying degree of adsorption of nanoparticles and subsequently different ORR performances. These differences are due to the differences in their key surface properties. For instance, the light interference for h-Mn3O4/CP was more noticeable than that for h-Mn3O4/CC, which can be attributed to the greater adsorption in the CP electrode. The percentage of Mn on the surfaces of CC, RVC, and CP electrodes was 0.297, 0.036, and ∼2%, respectively, based on the elemental analysis using the ICP-AES techniques (Figure S3, Supporting Information). The adsorption by the CP electrode was 6.7 times and 55.5 times greater than CC and RVC, respectively, as corroborated by the Raman studies (Figure c). The CP electrode displayed a broader Raman band with higher intensity compared with CC electrode (Figure c). Both the CC and CP electrodes displayed a vibrational Raman band at 661 cm–1 that corresponds to h-Mn3O4.[19,20] The pristine CP and CC electrodes did not display these peaks (Figure d). The greater coverage on the surface of the CP and CC, compared with the RVC, is due to their lower void space (measured in terms of pore size), higher contact angle, and lower specific area (Table ). For example, the pore size of the CP electrode (0.7 nm) is 15-fold higher than that of the CC electrode (10 nm).

Figure 1

(a) PXRD shows the relevant peaks for Mn3O4, (b) TEM image of h-Mn3O4 crystal, (c) optical spectroscopy images at 10× magnification, and (d) Raman spectroscopy.

Figure 2

SEM images of the carbon electrodes modified with the catalyst ink (a) CC at lower magnification; scale bar: 100 μm (b,c) at higher magnification; scale bar: 1 μm and 100 nm; (d) graphite felt at lower magnification; scale bar: 100 μm (e,f) at higher magnification; scale bar: 1 μm and 100 nm; (g) RVC at lower magnification; scale bar: 100 μm (h,i) at higher magnification; scale bar: 1 μm and 100 nm. (j) Carbon nanofoam paper at lower magnification; scale bar: 100 μm (k,l) at higher magnification; scale bar: 1 μm and 100 nm.

Table 1

Summary of the Key Electrode Surface Properties

electrode	pore size	specific surface area (m2/g)	contact angle	electrical conductivity (S/m)	ref
carbon cloth (CC)	10–31 mm	>1000	∼127°	∼623	(22,23)
graphite felt (GF)	3.5–15 nm	>0.7	∼120°	∼200 to 250	(24−27)
reticulated vitreous carbon (RVC)	5–100 pores/inch	>300	∼74°	∼370	(28−30)
carbon nanofoam paper (CP)	0.7–0.8 nm	400–600	∼90°	5000–23,000	(31)

Their interfacial properties influence heterogeneous catalytic reactions between reactants and catalyst surfaces. The wettability affects the molecular diffusion rates of O2 and the subsequent ORR performance of the electrode. To examine the influence of the ink on the wettability of the CP and CC electrodes, we measured their static contact angles using a sessile drop technique (Figure S1, Supporting Information). Upon coating with the ink, the contact angle of the CP electrode decreased from 90° to 0°, yielding a super hydrophilic surface. The contact angle for the pristine CC electrode was ∼122° (hydrophobic), which reduced only slightly after coating it with the ink. Generally, hydrophilic surfaces enhance the adsorption of O2 and overall ORR kinetics.[21] Based on the above results, we have selected h-Mn3O4/CC and h-Mn3O4/CP electrodes as promising candidates for aqueous air-cathodes in our MFC studies.

ORR Performance of Coated Glassy Carbon (GC) Electrodes (Abiotic Conditions)

The bare GC electrode did not show any ORR peaks under the N2-O2-saturated conditions (Figure S4, Supporting Information). The coated GC electrode did not register any ORR peaks under the N2-saturated conditions (Figure S5, Supporting Information). However, it yielded a distinct O2 reduction peak (0.45 V vs Ag/AgCl) under the O2-saturated conditions (Figure ), suggesting the conspicuous catalytic activity of the ink toward ORR. The reduction peak lay within −0.4 to −0.45 V (vs Ag/Ag/Cl) for all the scan rates. The limiting currents increased with the increasing rotating speeds at all three pH conditions. For example, at pH-1, the limiting current at 1600 rpm (563 μA) was 1.9-fold higher compared with 400 rpm (514 μA) (Figure a). The Koutecky–Levich (K–L) plots (J–1 vs ω–1) were used to determine the number of electrons (n) involved in the ORRs (Figure d). The linearity of the K–L plots indicated first-order rate kinetics toward the O2 concentration (Figure d). Based on the analyses of the K–L plots, the values of the average electron-transfer number were found to be 3.97, 3.9, and 2.2. at pH 1, pH 7, and pH 13, respectively. Unlike the 2e– pathway, the 4e– pathway minimizes the formation of hydrogen peroxide and yields higher energy conversion efficiency. These results indicate that the h-Mn3O4/MWCNT ink works effectively under acidic and neutral conditions.

Figure 3

The ORR performance of glassy carbon electrodes modified with h-Mn3O4/MWCNTs ink. Rotating disk cyclic voltammograms under O2-saturated conditions: (a) 0.1 M H2SO4 (pH 1), (b) 0.01 M KCl (pH 7), (c) 0.1 M KOH (pH 13) with a sweep rate of 50 mV/s, and (d) Koutecky–Levich plots at pH 1,7 and 13 with a linear fit.

ORR Performance of the Ink on Cathodes in MFCS under Neutral pH Conditions

After assessing the performance of the ink under well-controlled RDE conditions on a bench scale, we tested its performance on the porous cathode materials (CP, CC, GF, and RVC) in MFC prototypes (Figure ). We demonstrate the long-term performance of the ink under neutral conditions and latter under extreme pH conditions. The net performance of the MFCs was evaluated by monitoring temporal profiles of the operating voltage, peak power density, and impedance. The h-Mn3O4/CC cathode (Figure a) yielded 0.45 V in cathodic potential (vs Ag/AgCl), resulting in an overall MFC potential of 0.225 V in the full cell configuration. These values were significantly greater than the control based on a bare CC cathode (∼0.00 V). The current density profiles with h-Mn3O4/CC increased over time. The current density was relatively low for the first 22 days, peaked on day 23, and then remained stable for the entire test duration of 105 days (Figure a). This confirms the stability of the ink over the entire test duration (105 days). The peak power density (PD) with the h-Mn3O4/CC electrode (75 mW·m–2) was 20-fold higher compared with the control (3.75 mW·m–2) (Figure b).

Figure 4

Performance of MFC equipped with the h-Mn3O4/CC and h-Mn3O4/carbon paper. The controls are devoid of the catalyst ink. (a) Temporal variation of cathodic and full cell current density. (b) Current density vs power density (day 92) and (c) Nyquist curve of EIS test for MFC with respect to cathode (day 92) (d) Nyquist curve of EIS test for MFC with respect to full cell (day 92) (total duration = 106 days).

Next, we assessed if the other coated electrodes, including h-Mn3O4/CP, h-Mn3O4/RVC, and h-Mn3O4/felt, display similar outstanding ORR performances. The PD in the MFC with h-Mn3O4/CP electrode (674.58 mW·m–2) was 180-fold higher than the bare CC control (3.75 mW·m–2) (Figure b). The h-Mn3O4/RVC electrode yielded a PD of 58.2 mW·m–2, which is 15.5-fold higher than that of bare CC. h-Mn3O4/felt also yielded a PD (50.2 mW·m–2) that is 13.4-fold higher. As shown in Figure b and Figure S6 (Supporting Information) under identical polarization conditions, the MFCs with the h-Mn3O4/CC, h-Mn3O4/RVC, h-Mn3O4/felt, and h-Mn3O4/CP electrodes also yielded a stable performance throughout the test duration.

The PD with the h-Mn3O4/CP electrode (674.58 mW m–2) was ninefold higher compared with h-Mn3O4/CC (75 mW·m–2) (Figure b). We attribute this to the greater catalytic activity of the h-Mn3O4/CP electrode, as evident from the lower charge transfer resistance (Rct) compared to control, based on the EIS tests (Figure c,d). The Rct value, as indicated by the diameter of the semicircle in the Nyquist curve, was lower for h-Mn3O4/CP compared to h-Mn3O4/CC. The Rct values for h-Mn3O4/CP (5.323 Ω cm2) were 13- and 100-fold lower compared with h-Mn3O4/CC (66.56 Ω cm2) and control (536.8 Ω cm2) (see Table ). These lower Rct values suggest that the ink greatly enhances the ORR rate activity with higher electron-transfer efficiency, primarily due to greater catalyst surface area and conductivity. All the Rct values were obtained by fitting the electrical equivalent circuit (Figure S2, Supporting Information) to the data from the Nyquist plots.

Table 2

Resistance Values Obtained by Fitting Electrical Equivalent Circuit to Nyquist Plot of MFC Operated in Half-Cell Mode (Cathode as Working Electrode and Ag/AgCl as Counter and Reference Electrode)

cathode	Rs (Ω·cm2)	Rct (Ω·cm2)
bare carbon cloth (control)	6.757	536.8
h-Mn3O4/CC	8.743	66.56
h-Mn3O4/CP (pH 4)	20.73	12.18
h-Mn3O4/CP (pH 7)	6.63	5.323
h-Mn3O4/CP (pH 13)	5.423	4.837

The higher performance with h-Mn3O4/CP was due to the higher surface loading rate of the h-Mn3O4/MWCNT ink (∼2%) on the CP due to its nonporous surface. Also, the loading of h-Mn3O4/MWCNT ink onto the CP made the electrode super hydrophilic. The hydrophilic surfaces are known to enhance the ORR rate by increasing the adsorption, enriching the reactants onto the electrode surface.[21]Table demonstrates that none of the prior MFC studies based on the MnO catalysts were based on extreme pH conditions. The performance of h-Mn3O4/MWCNT ink was far superior to other catalysts under extreme pH conditions and equivalent or better at the neutral pH conditions.

Table 3

Performance of h-Mn3O4/MWCNT Ink Compared to Other MnO Catalysts in MFC Applications

cathode material	catalyst	synthesis method	application	pH for ORR	peak power density	electron transfer	ref
carbon cloth	MnOx/C	chemical oxidation	aqueous air-cathode	13	161 mW·m–2a	4	(32)
carbon cloth	α-MnO2	hydrothermal process	air-cathode	7	125 mW·m–2a	NA	(12)
	β-MnO2				172 mW·m–2a
	γ-MnO2				88 mW·m–2a
polyacrylonitrile carbon felt	MnO2/CNT	sonochemical-coprecipitation	air-cathode	NA	215 mW·m–3a	NA	(33)
carbon paper	MnO2-graphene nanosheet	redox reactions and microwave irradiation	air-cathode	7	2083 mW·m–2	NA	(34)
carbon cloth	Co-doped octahedral molecular sieve MnO2	hydrothermal process	air-cathode	NA	180 mW·m–2a	NA	(35)
carbon paper	nano-structured MnOx	electrochemical deposition	air-cathode	NA	772 mW·m–3	NA	(36)
carbon cloth	MnO2/polypyrrole	hydrothermal process	air-cathode	13	721 mW·m–2	4	(37)
carbon paper	MnO2/functionalized CNT	redox between functionalized CNT KmnO4	aqueous air-cathode	NA	520 mW·m–2a	NA	(11)
carbon nanofoam paper	h-Mn3O4/MWCNT	Kirkendall effect	aqueous air-cathode	4	462 mW·m–2a	4	this study
carbon nanofoam paper	h-Mn3O4/MWCNT	Kirkendall effect	aqueous air-cathode	7	675 mW·m–2a	4	this study
carbon nanofoam paper	h-Mn3O4/MWCNT	Kirkendall effect	aqueous air-cathode	13	392 mW·m–2a	2	this study
carbon cloth	h-Mn3O4/MWCNT	Kirkendall effect	aqueous air-cathode	7	75 mW·m–2a	4	this study

Normalized with anode surface area and others based on the cathode surface area.

Performance of the Ink under Extreme pH Conditions in MFCs

Here, we demonstrate the ORR performance of h-Mn3O4/CP cathode under extreme acidic and alkaline conditions. In the RDE tests, the h-Mn3O4/MWCNT ink displayed its affinity to promote a four-electron-transfer mechanism. Here, we analyze the performance of h-Mn3O4/CP electrodes under extreme pH conditions (Figure ). The coated electrodes displayed stable ORR performance at three diverse pH conditions. Their performance was higher than the uncoated electrodes, which is evident from the higher values of current density and PD (Figure a). We observed slight performance differences between the acidic and alkaline conditions. The OCV values at pH of 4.0 and 13 were 893 and 621.1 mV, respectively, which are different from neutral conditions (857.6 mV). Thermodynamically, the electrode potential changes by 0.059 V per unit pH change, evident from the greater OCV values at lower pH conditions (Figure d).[38] The potential losses due to an increase in pH affected the overall electrochemical performance of the MFCs.

Figure 5

Performance of MFC at pH 4, pH 7, and pH 13 with h-Mn3O4/CP as cathode: (a) EIS of the full cell, (b) EIS of cathode, (c) power density, and (d) open-circuit potential.

Considering the higher OCV at acidic conditions, one can expect the enhanced performance of MFC; however, the overall electrochemical performance was lower at acidic conditions. The PD under the acidic conditions (462.34 mW m–2) reduced slightly when compared with the neutral condition (675 mW m–2) (Figure c). The peak current density at the acidic conditions (1854.4 mA m–2) was lower than neutral (2397.5 mA m–2) and alkaline conditions (2489 mA m–2). The lower performance under the acidic conditions corroborates the findings of the RDE tests (Figure ). This decreased performance was due to the higher impedance under the acidic conditions. The charge transfer resistance (12.18 ohm·cm2) increased by ∼10-fold higher compared with the neutral conditions (2.323 ohms·cm2) (Table ). The higher impedance can be attributed to the catalyst fouling, likely by the intermediate products, as indicated by the inductance loop in the lower frequency at pH 4.0 (Figure b).[39,40] The pourbaix diagram for manganese also shows the dissolution of MnO into Mn2+ under acidic conditions.[41]

Under the alkaline conditions (391.68 mW m–2 and 2489 mA m–2), the PD and current density were only slightly lower than neutral conditions (675 mW·m–2, 2398 mA·m–2), respectively. The alkaline conditions in the catholyte (pH 13) resulted in a greater flux of alkali cations (K+), combined with greater consumption of the protons for ORR, decreasing the current proton availability to the cathode.[42]

Underlying ORR Mechanisms of h-Mn3O4/MWCNT Catalyst Ink

Having established the outstanding ORR activity of the ink on both planar and porous forms of carbon electrodes (Figures and 4), we turn our attention onto the mechanistic possibilities of ORR using the following steps: (a) diffusion and adsorption of O2 molecules onto the h-Mn3O4/MWCNT catalytic sites at the cathode (eq ), (b) electron transport from the MFC circuit onto the adsorbed O2 on the cathode (eq ), (c) weakening and splitting of O=O binding by h-Mn3O4, (eq ), and (d) removal of the as-produced OH– ions (eqs −6).[43−45] The overall four-electron ORR that we observed in the RDE tests (eq ) occurs by a direct pathway or a series pathway involving a peroxide intermediate. The direct pathway proceeds through a sequence of eqs −5 in such a way that all the four electrons from eq occur on the same catalytic site. Based on the series pathway in eqs −6,[46−51]eq can occur on the carbon electrode support onto which h-Mn3O4 is deposited or on Mn(IV) sites of h-Mn3O4. The catalytic activity for eq is attributed to h-Mn3O4,[46,47,51−58] and primarily to the higher valence state of Mn(IV) species.[59−61] The hollow structure increases the catalyst site area for the ORR reactions to take place.

The stable performance of the ink for ∼106 days and on all the four porous carbon electrodes demonstrates the following unique roles of the ink. They resolve the issues with the significant volume changes, poor electrical conductivity, and low ORR current of h-Mn3O4 nanoparticles under ambient conditions[62] that prevail in MFCs. The MWCNT supports anchor the h-Mn3O4 nanoparticle while enhancing the accessible surface area for their loading (Figure S4, Supporting Information) and promoting the charge transfer (i.e., electrons and protons) to meet and reduce the terminal electron acceptors (i.e., oxygen) onto the catalyst surface. Nafion served as a proton-conducting binder to tightly adhere the h-Mn3O4 with MWCNT, as shown in the SEM images (Figure S4, Supporting Information).[17] The high surface area, thin walls, and hollow interior of h-Mn3O4 imparted the electrode materials with a larger electrode–electrolyte contact area, reduced diffusion path, as well as free interior space for alleviating structural strain and volume expansion. Altogether, we achieved a high surface coverage of h-Mn3O4 by deeply integrating them within the hierarchical pores of the network covered with the MWCNT on the 3D carbon electrodes. For example, as shown in the SEM and optical microscopy images (Figure S4, Supporting Information), the MWCNTs anchor h-Mn3O4 onto the carbon electrode surfaces, and the h-Mn3O4/MWCNT particles are uniformly distributed without any signs of aggregates.

Conclusions

This study demonstrated a proof of concept regarding the stable ORR performances of h-Mn3O4/MWCNT ink under extreme pH conditions in MFCs. We attribute this outstanding performance to the unique electronic and morphological properties of the h-Mn3O4 nanospheres. Their amenability to functionalization with MWCNTs and the ease of drop-casting the ink demonstrate the scalability prospects. We have demonstrated the long-term and stability and longevity of the ink on cathodes in real MFC for ∼105 days. The ink yielded extremely high open-circuit voltage (OCV) (893 mV, pH 4; 857.6 mV, pH 7; and 621.1 mV, pH-13), equivalent or better than platinum-based materials. The closed-circuit performance was also superior compared with the catalysts based on earth-abundant metals. Overall, the current study indicates promise for using the ink to sustain ORRs in large-scale MFC applications, including photosynthetic MFCs. The current study opens a new research direction for designing ORR reactions in MFCs to treat waste streams (e.g., mining wastes) that emerge at extreme pH conditions.

Methods

Synthesis of the Catalyst Ink

Figure a–d depicts an overview of the synthesis of MnO, h-Mn3O4 nanocrystals, catalyst ink, and coated electrodes (discussed below). A detailed synthesis of MnO particles is described in the supplementary section.

Figure 6

Synthesis of (a) MnO, (b) h-Mn3O4, (c) h-Mn3O4/MWCNT catalyst ink, and (d) electrodes modified with the catalyst ink.

Synthesis of h-Mn3O4 Nanoparticles

The h-Mn3O4 nanoparticles were prepared as described in the literature.[17] Briefly, MnO (180 mg, 2.53 mmol) was introduced into a three-neck round-bottom flask, and the hexane (6 mL) was evaporated under a vacuum. To this, 20 mL of degassed 1-octadecene and CuCl2·2H2O were added. Degassed oleylamine was added to achieve an N:Cu ratio of 12:1. The flask was fitted with a reflux condenser, and N2 was passed into the flask to blanket the solvent. The reaction mixture was heated to 150 °C at a rate of 10 °C/min and kept at that temperature for 3 h and then cooled to room temperature. Then, 25 mL of isopropanol was added to the cooled mixture, and the product was centrifuged at 3500 rpm for 8 min. The precipitated nanoparticles, typically dark brown, were collected, and the supernatant was discarded (Figure b).

Synthesis of the Ink

Details on combining h-Mn3O4, MWCNT and Nafion binder for obtaining a stock solution of the ink has been described in our earlier works (ref (18)) (Figure (referred to as ink or catalyst ink).

Deposition of the h-Mn3O4/MWCNT Ink on Carbon Electrodes

GC Electrode for RDE Tests

Prior to the application of the ink, a GC electrode was polished sequentially using three different alumina slurries (0.05, 0.3, and 5 μm suspensions, respectively) and sonicated in distilled water for 5 min (40 kHz, Branson 2510 water bath). The ink was drop cast onto the treated GC surface and air-dried for 5 min. The modified GC electrode was used in the RDE studies (see the RDE Tests section).

Porous Carbon Electrode for MFC Tests

GF, RVC, carbon cloth (CC), and carbon paper (CP) electrodes were coated with the ink to obtain ORR cathodes for the MFC studies. These electrodes were cleaned with acetone and ammonium peroxydisulfate and heated at 450 °C for 30 min.[63,64] The pretreated electrodes were washed with distilled water thrice and immersed in the ink. The mass of the ink was adjusted to be ∼5% of the equivalent mass of the electrode. The ink turned colorless after the nanoparticles were transferred onto the electrodes. The coated electrodes were dried in a vacuum oven at 200 °C (see Figure d). The morphology of nanoparticles coverage on the electrode surfaces and their morphology were assessed using scanning electron microscopy (SEM), Raman spectroscopy (ffTA Foram X3 module [Foster + Freeman Ltd., Evesham, UK] with 10× magnification, laser excitation wavelength of 638 nm and power of 10 mW), and optical spectroscopy.[65] Contact angle measurements were carried out using a Goniometer (Model 500, ramé-hart Instrument Co.) and DROP-image advanced v2.4 software. An ICP-AES technique was used to carry out the elemental analysis of the attached nanoparticles.

RDE Tests (Abiotic Conditions)

A Pine Research modulated speed rotator (MSR) that was integrated with a three-electrode electrochemical cell was used to carry out the RDE tests. The coated GC electrode (disk surface area = 0.196 cm2) served as the working electrode (WE), Ag/AgCl as a reference electrode, and platinum wire as a counter electrode. The RDE tests were carried out at pH 1, 7, and 13 using 0.1 M H2SO4, 0.01 M KCl, and 0.1 M KOH, respectively. An uncoated electrode served as a control. Cyclic voltammetry tests were carried between −0.8 and 0 V at a scan rate of 25 mV/s and rotation speeds of 400–1600 rpm. The electrolytes were saturated with O2 by bubbling O2 for 15 min using a dual-port gas inlet. The O2 flow was maintained over the electrolyte during the measurements.

MFC Tests (Biotic Conditions)

The test MFCs and controls that were based on two-compartment MFCs (see ref (63)) were operated under the cathode limited conditions. The volume of the cathode and anode compartments was 400 mL each. A graphite brush (4 cm × 5 cm) constituted the anode and four coated electrodes the cathodes (4 cm × 4 cm rectangles). The test MFCs were operated under the cathode limited conditions to assess the ORR performances of the coated electrodes. A cation exchange membrane (CMI-7000) (Membranes International Inc.) separates the anode and cathode compartments. This membrane was soaked in a 5% NaCl solution for 12 h at room temperature prior to its use. The control MFC used a bare CC as the cathode. A titanium wire was used to connect the electrodes with the external electrical circuit. The anolyte was based on the primary clarifier effluent from the Rapid City wastewater reclamation facility supplemented with sodium acetate (3 g/L). The catholyte was based on a phosphate buffer (100 mM). The catholyte and anolyte were continuously purged with oxygen (0.1 L/min) and ultrapure nitrogen (0.1 L/min), respectively (see Figure ). This arrangement ensured aerobic and anaerobic conditions in each of the compartments, respectively. The MFC studies were carried out at room temperature.

Figure 7

Two-compartment MFCs with wastewater as the anolyte, phosphate buffer as catholyte, and h-Mn3O4 ink as ORR catalyst on the cathode surface. The catholyte and anolyte were purged continuously with oxygen and nitrogen, respectively.

Data Acquisition and Electrochemical Analysis

Electrochemical impedance spectroscopy (EIS) tests were performed at an OCV using an alternating current (AC) signal (±10 mV) at a frequency range of 10 kHz to 0.01 Hz. A data acquisition module (DAQ/54; I/O Tech Inc., Cleveland, OH) was used to acquire the voltage data across an external load. The voltage at each resistance was measured every 30 min, and the voltage value was recorded under steady-state conditions. The EIS spectra were obtained using the cathode as the WE, anode as the counter electrode (CE), and Ag/AgCl as the reference electrode (RE). A Gamry reference 3000 workstations were used for all the EIS tests and Gamry Echem Analyst software for analyzing the impedance data. Individual resistances were obtained by external equivalent circuit (EEC) fitting analysis of the EIS data, limiting the maximum error tolerance to 2%. A modified Randle’s circuit used for the EEC analyses is shown in Figure S2, Supporting Information.