Biosynthesized α-MnO₂-based polyaniline binary composite as efficient bioanode catalyst for high-performance microbial fuel cell

Abstract

Microbial fuel cell (MFC) has novel technological advances in simultaneous power generation and wastewater treatment applications. In this study, low-cost biosynthesized α-MnO₂ nanoparticles integration with conducting polyaniline (PANI) matrix to form α-MnO₂/PANI hybrid nanocomposite was fabricated by in situ polymerization method. The prepared material was characterized through UV-Vis spectroscopy, XRD, FTIR, TGA-DTA, DSC, SEM, cyclic voltammetry, and impedance spectroscopy. MFC performance study was done by using an external resistance in the range of 100 Ω–100 kΩ. The continuous test on bare pencil graphite electrode (PGE), α-MnO₂/PGE, PANI/PGE, and α-MnO₂/PANI/PGE were evaluated in glucose-fed-Escherichia coli-based MFC. It was found that α-MnO₂/PANI/PGE produces a maximum power and current densities of 426.26 ± 38.89 mW m⁻² and 2485.51 ± 397.31 mA m⁻², respectively. This was 6.5 and 5.7-fold higher in power and current densities than unmodified PGE. The maximum chemical oxygen demand produced by hybrid composite modified anode during closed circuit voltage or with external resistance and open circuit voltage (OCV) (circuit without connecting external resistance) measurements were found to be 88.19% and 92.27%, respectively. A maximum of 650.61 ± 10.11 mV OCV was obtained by α-MnO₂/PANI/PGE while 222.36 ± 8.16 mV of OCV was generated by PGE.

KEYWORDS:

• Manganese dioxide nanoparticle
• polyaniline
• bioanode
• microbial fuel cell
• wastewater treatment

Introduction

Environmental pollution along with energy crisis is causing a determinant effect on human health and lives on ecosystems. Producing alternative energy in the form of bioelectricity and wastewater treatment from organic liquid effluents is thus the key strategy to reduce energy crises and to keep the environment healthy (Dessie et al. 2020a). To address such problems, emphasis has been given to microbial fuel cell (MFC) device system due to its technological novelty (Malik et al. 2014 ). MFC therefore produces energy by transforming organic matter into electricity through the metabolic activities of inoculated bacteria (Li et al. 2018). As an advantage, MFC is better in application than other conventional energy conversion and pollutant removal methods due to the less activated sludge production, no need for additional energy investment for aeration, simple operation at mils condition, and environmental friendly (Yaqoob et al. 2020). Due to this reasons, various research advancements have been focused up on it to produce such clean energy with simultaneous waste treatments from various waste discharges (Li et al. 2018; Dessie et al. 2019). From all components of MFC, the anode electrode plays a primary role in energy conversion and pollutant removal. Conventional carbons, metal/metal oxides, and conducting polymers are the most common electrode materials to build anode in MFC operation. However, the poor chemical and mechanical stability, small surface area, large pore size, non-biocompatibility, corrosion, and high cost across such ranges of materials limits both power output and wastewater treatment efficiencies in MFC (Wu et al. 2018; Yaqoob et al. 2020). Therefore, to solve these challenges, modifying a conventional electrode with most conductive and high-quality materials is required. These materials are functional nanocomposites (Evelyn et al. 2019).

Anode materials selection in MFC device system is still a critical issue to achieve a successful electrochemical efficiency, electron transfer rate, andbacterial adhesion. This happening is due to low electrical conductivity which enables the flow of electron and increases speed followed by decreasing bulk electrolyte resistance, low surface area which increases resistance power, biocompatibility which directly defines bacterial contact and increasing their respiration process, low stability with low durability, and low availability and high cost (Yaqoob et al. 2020). To address those factors, building a functional metal organic framework strategies from novel Co-based N-doped carbon (Li et al. 2018), Co₉S₈ nanoparticles (NPs) coupled (Li et al. 2019) and Co₃O₄ NPs decorated (Tan et al. 2017) with nitrogen-doped hollow carbon sphere, porous Co₃O₄ decorated nitrogen-doped grapheme (Tan et al. 2018), and gallic acid-assisted Pd anchored on porous nitrogen-doped reduced graphene oxide (Wu et al. 2018) are the most advanced way to enhance MFC performance by increasing the stability and electrocatalytic activity. In addition, the surface degree of surface roughness of anode should be high to detach water molecule and to offer more active sites for bacterial attachments (Hindatu et al. 2017). Conventional carbon-based material are commonly utilized to fabricate anode electrodes due to the fulfillment of least requirements stated above make them very suitable anode in MFC performance. Apart from these, pencil graphite electrodes (PGEs) is one of the most suitable anode material in MFC due to its biocompatibility, availability, renewability, reproducibility, and disposability during its operation (Kariuki et al. 2015; David et al. 2017; Tsuji et al. 2019). However, PGE and other conventional carbon electrodes like graphite felt (Ringeisen et al. 2006), carbon fiber (Salman et al. 2019), graphene foil (Mink et al. 2014), and toray carbon paper (Abrevaya et al. 2010) still have insufficient biocompatibility, low conductivity and limited surface reaction sites. The presence of these factors would limit bacterial biofilm production, decreases electron transfer rate, and slow electro kinetic mechanism for electrochemical process in MFC operation (Wu et al. 2018).

To solve and improve such drawbacks, anode modification by surface treatment (either thermal or electrochemical), coating, and compositing from different material combinations are a very important strategy to enhance bacteria biofilm performance for good electrical contact between anode and bacterial species (Yaqoob et al. 2020). Currently, the low conductivity PGE anode can be addressed by modifying PGE surface with transition metal oxide nanomaterials (NMs) (Goswami et al. 2018) like manganese dioxide (MnO₂) (Mahmudi et al. 2018) and conducting polymers like PAN (Singh and Shukla 2020) in form of composites. Therefore, MnO_2-based composites, i.e. MnO₂/polyaniline (PANI) is expected to improve the power and treatment efficiencies in MFC due to the bacterial adhesion ability and these criteria be able to increase electron transfer rate towards anode surface. The main objective of using conducting polymer in general and PANI in particular in anode modifications is to improve surface area and surface roughness for superior electron transfer rate. So, the chemical and mechanical instability in metal oxide is improved by better mechanical strengthening PANI by forming composites. But the inefficient conductivity of PANI in neutral aqueous solution often limits its activity in the MFC performance (Zhao et al. 2013). Because the optimum condition to operate MFC is a neutral or weak basic solution but PANI at this experimental condition doesn’t compete its work properly (i.e. which is unstable) (Jia et al. 2018).

As an advantage, the presence of positive charge on PANI and the negatively charge characteristics on bacteria could increase interactions between bacteria and PANI which helps to form an active biofilm surface (Geetanjali and Kumar 2019). Such criteria enhance the surface areas of the modified electrodes that enable to increase electron transfer process (Jia et al. 2018). To get high bioelectricity and treatment efficiency, PANI is combined with novel carbon nanotubes (Wu et al. 2018), carbon fibre-embedded bacterial cellulose (Trindade et al. 2020), and carbon felt (Rajesh et al. 2020). In addition to anode modification by different NMs combinations, Configuration of MFC has also an impact on its performance. Commonly, double and single-chambered MFC configurations have been operated at the laboratory scale (Heyang et al. 2016). Since double-chambered MFC has an advantage over single-chambered MFC configuration. For example, purging of oxygen in a single-chambered configuration might cause direct oxygen interference due to cathode insertion in a common electrolytic medium. The sluggish kinetics during oxygen reduction reactions causes slow electrocataytic activities of the catalyst, resulting in the poor performance of MFC (Huang et al. 2015). To solve this problem, a double-chambered MFC configuration has had preferred over it. As a result, choosing a flexible dual-chambered MFC setup has been given a broad emphasis (Wanget al. 2013).

Therefore, modification of MnO₂ by PANI to form a hybrid composite and coated on PGE would make the best opportunity to addresses all problems which were mentioned above. The cost effectiveness, abundance, and environmental friendly nature of MnO₂ (Haoran et al. 2016) have an advantage to modify PANI. Even though, the low-cost MnO₂ lucks in electrical conductivity in protonic media (Dessie et al. 2020a). At these conditions, it may have a chance to change its phase and then could have the possibility to decrease its performance. The necessity, in the +4 oxidation states, an α-type chain-like tunnel structure (α-MnO₂, 2 × 2 tunnel (Julien and Mauger 2017)) exhibits fast electro kinetics and could be considered as an excellent catalytic activity in the neutral state due to having a large surface area.

Based on those detailed justifications, we prepared the first biocompatible hybrid composite material from Vernonia amygdalina (V. amygdalina) assisted α-MnO₂ NPs and PANI matrix by chemical polymerization method to increase the PGE anode catalytic activity by proposing double-chambered MFC bioreactor system. The test was done by Escherichia coli (E.coli) as a microbe and glucose as a carbon source for wastewater and bioelectricity production applications. Up-to-date, the biosynthesized α-MnO₂ NPs integrated PANI composite modified PGE in glucose-based MFC is not studied elsewhere.

Materials and methods

Materials

KMnO₄ (99%, Alpha Chemika, Mumbai, India), aniline (99.5%, Merck), ammonium per sulfate (APS; ((NH₄)₂S₂O₈) 98%, Sigma-Aldrich), ethanol (98%, Merck), and HCl (37%, Merck) were used as received. All chemicals, nutrients, and substrates used for MFC test were analytical grades and also used as received. Solutions for these experiments were prepared from distilled water.

Biosynthesis of MnO₂ NPs

An optimized α-MnO₂ NPs was synthesized based on our previous work using Vernonia amygdalina (V. amygdalina) leaf extract as a reducing agent (Dessie et al. 2020b). Briefly, 56.28% in volume of aqueous solution of 1.81 mM potassium permanganate (KMnO₄) and 43.72% by volume of V. amygdalina leaf extract were mixed followed by constant stirring at room temperature for 103.42 min until black brown suspension was formed (Dessie et al. 2020b). The pH of the performed condition was proceeding at 6.02. The obtained solution was centrifuged at 4000 rpm for 15 min and the precipitate was collected and washed with both distilled water and ethanol several times to remove impurities. The formed precipitate was collected and dried overnight at 80°C. The sample was calcined in muffle furnace at 450°C for 3 h.

Synthesis of PANI

Synthesis of PANI was based on mixing aqueous solution of aniline hydrochloride and APS at room temperature, followed by the separation of PANI hydrochloride precipitate by filtration and drying adopted from the literature (Mohammad Shafiee et al. 2019). Briefly, 0.5 mL aniline monomer was dissolved in 1 M (100 mL) HCl aqueous solution in a round-bottom flask and stirred for 30 min. Next, 5.71 g of APS was dissolved in 50 mL of distilled water. Both solutions were kept each separately for 1 h at room temperature. Then, the two solutions were mixed in a beaker and stirred with a magnetic stirring bar until the color change from colorless to dark green, which was indicated PANI formation, and waited for 4 h to polymerize. The formed PANI precipitate was collected by filtration and washed with 0.2 M HCl solution until the filtrate becomes clear, then with ethanol several times to remove monomer, oligomer as well as with distilled water several times to remove the acid. Finally, PANI powder was dried in an oven at 60°C for 24 h.

Synthesis of α-MnO₂/PANI nanocomposite

A typical in situ oxidative polymerization method was used to synthesis α-MnO_2-based PANI nanocomposites by following previously reported methods with little modification (Inamdar et al. 2016). Firstly (protonation step), 0.5 mL aniline (0.1 M) monomer was dissolved in 1 M HCl (100 mL) aqueous solution in a round-bottom flask and stirred for 30 min to form aniline hydrochloride. Secondly (homogenization of oxidant), dissolving 0.5 g of green synthesized α-MnO₂ NPs in 25 mL distilled water using ultrasonication for 5 min in order to keep the α-MnO₂ homogeneously suspended in the solution. Finally, a homogeneously suspended α-MnO₂ NPs was added drop wise to doped aniline hydrochloride solution with 2.86 g APS (co-dopant) at room temperature, and the mixture was stirred continuously at room temperature for 4 h to complete the polymerization process. The reaction mixture was turned into dark green precipitate, the obtained materials were then collected by filtration and washed with 0.2 M HCl solution until the filtrate become clear, then with ethanol several times to remove monomer, oligomer as well as with distilled water several times to remove the acid. Finally, the product was dried in the oven at 60°C for 24 h to give α-MnO₂/PANI nanocomposite (Scheme 1). The composition of α-MnO₂-based composite is given in Table 1.

Scheme 1. Synthesis protocol for α-MnO₂/PANI nanocomposite preparation.

Table 1. Composition of α-MnO₂ composite.

Sample	α-MnO2 (wt%)	Aniline (v%)	Oxidant/reductant
α-MnO2	100	0	V. amygdalina *
PANI	0	100	APS^+
α-MnO2/PANI	50	50	α-MnO2^+

*Reductant.

+Oxidant.

Anode catalyst fabrication

Nanocatalyst coatings were modified on PGE with HB pencil grade (Camlin, India). Before coating four equal HB PGE were handmade with pencil lead having a diameter of 0.2 cm with 1 cm in length. The lead was soaked in distilled water for 24 h and rinsed with 0.2 M phosphate buffer solution (PBS) at the time of setting up the fuel cell (Patade et al. 2016). Afterwards, synthesized α-MnO₂, PANI, and α-MnO₂/PANI nanocatalysts and polyvinyl alcohol (PVOH, M.W. = 44.05 g mol⁻¹) polymer binder solution were prepared by drop-cast coating method. The purpose of the binder is to increase the bonding force between the prepared catalyst and bare PGE. A required amount of NPs was taken in the beaker and deionized water was added; then it was kept in an ultrasonic water bath for 15 min under ambient condition to form a uniform sol since the binder step was adopted from the literature (Harshiny et al. 2017). The PVOH was added in the ratio of 1:5 (PVOH to NPs wt/wt %) and dissolved in 10 mL deionized water, followed by mixing for 1 h by magnetic stirring at 70°C. Before coating on PGE surface, the mixture was kept for sonication to make uniform distribution of nanocatalyst throughout the solution. Then, a 100 µL of uniformly distributed nanocatalyst-based PVOH solution (suspension) was coated on PGE surface by drop-cast method with the help of microsyringe (Madhusudhana et al. 2020), and finally it was dried for 12 h at 60°C using oven drying to improve its interface stability and bonding forces of the modified layer.

Characterization of nanocatalysts

Formation of nanocatalysts was analyzed by double beam UV-Vis spectroscopy (Azzota SM-1600 SPECTROPHOTOMETER, USA) to elucidate and measure the optical properties in the range of 200–800 nm. Thermogravimetric-differential thermal analyses (TGA-DTA) were carried out on a DTG-60H detector (Shimadzu, Japan), with the sample flow rate at 50 mL min⁻¹ under nitrogen (N₂) atmosphere. Differential scanning calorimeter (PerkinElmer, DSC 4000, USA) was carried out under N₂ atmosphere at a flow and heating rates of 20 mL min⁻¹ and 10°C min⁻¹, respectively. The crystalline structure of α-MnO₂, PANI, and α-MnO₂/PANI nanocatalysts were performed using Shimadzu diffractometer (XRD-7000, Tokyo, Japan) with a voltage of 40 kV and a current of 30 mA using at (Cu Kα = 1.5406 Å) radiation as an X-ray source in the 2θ range from 10° to 80°. Fourier transforms infrared (FTIR) spectra of all samples were recorded with a Perkin Elmer FT-IR BX spectrophotometer in the range 4000–400 cm⁻¹ with samples prepared using KBr pellets. Scanning electron microscopy (SEM) was performed using a JEOL Ltd. instrument with a model of JCM-6000Plus. The surface roughness, pore structure, and pore size distributions of α-MnO₂, PANI, and α-MnO₂/PANI surfaces were studied by Gwyddion software (http://gwyddion.net).

Electrochemical measurements were performed in a one compartment three electrode system connected to an electrochemical workstation (Biologic SP-300, Canada) using EC-Lab for windows v10.36 (software), consisting of bare/modified PGE, Ag/AgCl and Pt wire as a working, reference, and counter electrodes, respectively. Cyclic voltammetry (CV) measurements were performed in the potential window between −0.8 and +0.8 V at a scan rate of 50 mV/s in 100 mM PBS (pH = 7.4) as the supporting electrolyte. Electrochemical impedance spectroscopy (EIS) measurements were recorded over the frequency range 200 kHz to 10 mHz with amplitude of 10 mV superimposed on working potential (E_we = +0.252 V vs. Ag/AgCl electrode).

The Nyquist diagrams in EIS were employed to estimate quantities of important elements, such as solution resistance/or ohmic loss (R_s), charge transfer resistance/or polarization loss (R_ct), double-layer capacitance (C_dl), and Warburg impedance/activation or concentration loss (W) of the equivalent circuit by using EC-Lab V10.40 Zsim demonstration software. For this study, a representation of Randles type circuit model shown in the Nyquist plots was fitted for all EIS experimental data. The model defined that semicircle region found at the high frequencies in the diagram provides information about electron transfer limited process while the linear region at the low frequencies region associated with Warburg diffusion process (Tahtaisleyen et al. 2020). The length of the diameter from the semicircle showed surface/or interface charge transfer resistance that defines the dielectric and insulating properties of the electrode/electrolyte solution interface (Çakir et al. 2015).

Microorganism and inoculum preparation

Effluents of wastewater sample were collected from Adama Science and Technology University wastewater treatment plant, Adama, Ethiopia (latitude 8°33′43.56″ North and longitude 39°17′23.28″ East), using sterile glass bottle tied up with light weight steel round stick. The sample was immediately transferred to the biology laboratory room and was kept overnight in a refrigerator at 4°C. Areas which was used for E.coli culturing were cleaned with 95% ethanol during sample container opening. Before culturing, all the necessary facilities were also sterilized in an autoclave at 120°C for 30 min. After room temperature was kept, 3 mL of settled waste sewage effluent sample was cultured in 400 mL autoclaved conical flask using Lactose broth as a culture media. After 15 min, 1 mL supernatant was taken from the conical flask with the help of a micropipette and added the supernatant into another sterile test tube in order to make dilution. The cultured sample was anaerobically incubated at 37°C for 24 h. After 24 h, the morpho-biochemical characterizations of the grown microbial colonies after isolation were done (Figure S1).

MFC setup and operation

Dual-chambered MFC was designed (Figure S2). Pencil graphite (1 cm length, 0.2 cm diameter) was used as both anode and cathode electrode. The artificial wastewater as the anolyte was prepared in the anode reactor contained: 5840 mg L⁻¹ of NaCl, 100 mg L⁻¹ of KCl, 250 mg L⁻¹ of NH₄Cl, 12000 mg L⁻¹ of Na₂HPO₄·12H₂O, 2570 mg L⁻¹ of NaH₂PO₄·2H₂O, 840 mg L⁻¹ of MgCl₂.6H₂O, 500 mg L⁻¹ of FeCl₃, 2 mg L⁻¹ of MnSO₄.H₂O, 5 mg L⁻¹ of CaCl₂, and 340 mg L⁻¹ of yeast extract to fulfill the micronutrient deficiency (Danish Khan et al. 2015). The anolyte with pH 7.40 could act as the microbial growth medium. The anode chamber was kept airtight and filled completely with 166 mg L⁻¹ glucose substrate as a carbon source with the inoculum added to achieve anaerobic experimental conditions (Pandit et al. 2014). The cathode chamber was filled with 50 mM KMnO₄ which was dissolved in a neutral medium (100 mM, pH 7.40) PBS. The volumes of both chambers were contained from 250 mL. The salt bridge was prepared from 3 g of agar and 12 g of KCl (see supplementary materials). Finally, the MFC operation was operated at room temperature (28 ± 2°C) for four days.

Electrochemical calculation

Open circuit voltage (OCV) was recorded by using digital multimeter (DT830D, Haoyue, China) for 3 h/day after 5 min stabilization for every 5 min continuous interval. The accuracy of the multimeter for voltage measurement was ±1.2% of indicated values ±5 least-significant digits. Then it was directly connected between an anode and a cathode of MFC to measure the generated voltage. The polarization curves were measured from closed circuit voltage (CCV) against external resistors for 15-min interval using Decade Resistance Box (ANSHUMAN, Model: DRB-7 T) by varying resistances from 100 kΩ to 100 Ω.

The flow of current (I) across each resistor was calculated via Ohm’s law: I = V/R, V as a cell voltage and R as the external resistance (R_ext). Similarly, power (P) is the multiplication of voltage (V) and I. Current (j) and power densities (p) were calculated from the ratio of current and power to surface area of anode (PGE, area = 0.6914 cm²). The area of PGE is determined based on the right circular cylinder calculation formula (Figure S3).

For electrode biofilm topography observation using SEM (performed using a JEOL Ltd. instrument with a model of JCM-6000Plus to observe the structural morphology of the formed biofilm), four anodic samples including control were fixed in 2.5% glutaraldehyde for 2 h which is used as a disinfectant, rinsed three times using PBS (pH of 7.4, 100 mM), dehydrated by different alcoholic series (60%, 70%, 80%, 90%, 95% and 100%) and then air dried. Finally, samples were subjected to SEM observations.

COD and CE measurements

Chemical Oxygen Demand (COD) measurements were conducted using the acidic permanganate (COD_Mn) method under the protocol of Standard Methods (Goh and Lim 2008). Briefly, 10 mL of aqueous sample solution was transferred into a 250 mL ground-glass neck conical flask; the sample was added with 1 mL of 9 M H₂SO₄ followed by 10 mL of 0.01 M KMnO₄ digestion solution. The sample was heated for 30 min in a water bath at 96°C–98°C. After digestion, the sample was cooled immediately in an icebox to quench the reaction. The remaining KMnO₄ in the sample solution was determined by spectrophotometer in the range of 200–800 nm. A spectrophotometer is needed to make a standard calibration curve (Figure S4 and S5) by measuring concentration of the remaining permanganate and their absorbance which was adopted from dichromate method (Beer et al. 2017). If the sample's actual COD was higher than the range, the sample was diluted but the real results were multiplied by dilution multiple (Oljira et al. 2018). The degradation efficiency was evaluated by calculating the COD removal with the following formula: (1) $\text{COD}=\left(\frac{\text{CO}{\text{D}}_{\text{in}}-\text{CO}{\text{D}}_{\text{out}}}{\text{CO}{\text{D}}_{\text{in}}}\right)\times 100\text{\%,}$(1) where COD_in (mg L⁻¹) is the influent COD and COD_out (mg L⁻¹) is the effluent COD.

The columbic efficiency (CE) was measured as the ratio between the electrons transferred to the anode from the microorganism to the theoretical one that can be achieved if all organic substrates were digested by the microorganisms to generate electrons. Therefore, the CE can be calculated for batch reactor conditions as follows: (2) $\text{CE}=\frac{M{\int }_0^{t}I\text{dt}}{Fb{V}_{\text{an}}\mathrm{\Delta }\mathrm{C}\mathrm{O}\mathrm{D}},$(2) where M is the oxygen molecular weight (32), I is the current and t is the time at maximum power density. F is Faraday’s constant, b = 4 indicates the number of electrons exchanged per mole of oxygen, V_an is the anodic solution volume, and ΔCOD is the chemical oxygen demand change over time.

Results and discussion

UV-Visible analysis

The incorporation through the insertion of α-MnO₂ in PANI matrix to form α-MnO₂/PANI composite were evidenced from UV-Visible absorptionspectroscopy and their spectra are given in Figure 1. The synthesized and optimized α-MnO₂ exhibited an absorption band at about 285 nm due to electronic transitions from triply degenerate, t₂g to doubly degenerate, e_g in the d-orbital due to having the tetrahedral shape of α-MnO₂ (Figure 1(a)) (Dessie et al. 2020b). The absorption spectrum of polymerized PANI is shown in Figure 1(b) indicated a change in position and intensity at about 225 nm (PANI doped with APS). Band at 267 nm indicating π → π* transition of C = C in the benzenoid (–(C₆H₄)–) rings while band near 405 nm formed in the range of 400–450 nm corresponds to the transition of π → polaron (Mohammad Shafiee et al. 2019). The electrical conductivity of PANI is improved due to the faster delocalization of charges by forming a radical cation (polaron) or dication (bipolaron) (Rajesh et al. 2020). So, it is evidenced that from the formed polaron, synthesized PANI is an emaraldine form (Mathew and Thomas 2020). Earlier studies also evidenced that an absorption band shifted from a higher wavelength to the lowest excitation peak closest to 227 nm might indicate the high oxidation state PANI existence with different degree of protonation of the imine nitrogen atoms in the polymer chain (Tanwar and Ho 2015). The absence of an absorption band in the range of 400–430 nm and blue shifted in the UV region with intense and broad peak than α-MnO₂ and PANI shown in Figure 1(c), indicated the clear incorporation of α-MnO₂ inPANI conducting matrix to form α-MnO₂/PANI hybrid composite.

Figure 1. UV-Vis absorption spectra of α-MnO₂ (a, black curve); PANI (b, red curve); and α-MnO₂/PANI (c, blue curve).

XRD analysis

The crystalline structure of pure α-MnO₂, PANI, and α-MnO₂/PANI hybrid composite was performed by X-ray diffraction (XRD) test as shown in Figure 2. The XRD pattern of pure α-MnO₂, which was indexed with standard Crystallography Open Database (COD, Entry # 96-151-4117) (Dessie et al. 2020b), showed three intense peaks found at 2θ of 12.479°, 25.228°, and 37.218° with lattice plane of (110), (220), and (211), respectively (Figure 2(a)). The XRD pattern of PANI exhibits two main broad peaks with 2θ of 20.494° and 25.413°, which are indexed to a diffraction planes at (100) and (110), respectively, due to the characteristic chain of PANI (Figure 2(b)) (Zhou et al. 2018). For the α-MnO₂/PANI hybrid composite, its diffraction peaks are almost similar with pure PANI in all planes except diffraction angle at 21.151° (Figure 2(c)), this change in peak symmetry on α-MnO₂/PANI than PANI alone showed that a distortion in crystal structure of α-MnO₂ has occurred during the polymerization reaction, which shows the transformation of the crystalline α-MnO₂ into the amorphous phase of PANI. This confirms a successful integration of α-MnO₂ could form in PANI matrix. A similar literature work has been studied from β-MnO₂/PANI hybrid composite (Zhou et al. 2018). Generally, PANI incorporated α-MnO₂ showed broad peaks and this suggested that the particle size is so fine (Table 2). The interplanar spacing in the lattice and their crystallite size of the prepared particles formed by each of the NMs sample is presented in Table 2.

Figure 2. XRD patterns of (a) α-MnO₂ nanosphere flower, (b) PANI powders, and (c) α-MnO₂/PANI/ nanocomposite.

Table 2. XRD data used for calculation interlayer spacing, crystalline size of different sample.

Sample type	d-spacing (nm)	2θ (°)	FWHM (°)	Miller indices (h k l)	Crystallite size (nm)
α-MnO2	0.69792	12.479	0.3875	1 1 0	21.55
	0.35008	25.228	0.4050	2 2 0	21.00
	0.23830	37.218	0.4453	2 1 1	19.67
					Average = 20.74
PANI	0.43275	20.494	2.245	1 0 0	3.56
	0.38851	25.413	10.024	1 1 0	0.79
					Average = 2.18
α-MnO2/ PANI	0.41971	21.151	1.545	1 0 0	12.48
	0.35021	25.413	14.831	1 1 0	0.54
					Average = 6.51

The interlayer spacing (d) and crystallite size (D) were determined using Braggs law and Debye Scherer formula, respectively, from Equations (3) and (4(4) $D=k\lambda /\text{β}\text{Cos}\theta ,$(4) ) given below (Ajeel and Kareem 2019): (3) $n\lambda =2d\text{ sin}\theta ,$(3) where n is an integer, λ is the XRD wavelength (0.15406 nm) and θ is the angle between the incident ray and the scattering planes. (4) $D=k\lambda /\text{β}\text{Cos}\theta ,$(4) where D is the crystallite size (nm), k is shape factor for an average crystallite (0.89) depending on the miller index of the reflecting plane and the shape of the crystal, λ is the XRD wavelength (nm), β is the full width at half maximum (FWHM) of crystallite peak (measured in radians) and θ is Bragg’s angle (°) of the XRD peak.

FTIR analysis

The characteristic peak observed at a wavenumber of 3422 and 3304 cm⁻¹ might be assigned as –OH stretching and H–O–H bending vibrations of surface adsorbed water found on pure α-MnO₂ NPs (Figure 3(a)). The peak observed in the band between 600 and 475 cm⁻¹ at about 596 and 495 cm⁻¹ corresponding to the stretching collision of O–Mn–O and Mn–O bond stretching vibration which justifies the formation of α-MnO₂ NPs (Dessie et al. 2020b). In Figure 3(b), the typical FTIR spectra of pure PANI, the characteristic band at 3448 and 2924 cm⁻¹ corresponding to N–H stretching and aromatic C–H stretching, respectively, which can also observed in α-MnO₂/PANI composite (Figure 3(c)). The characteristic peak found at 1636 and 1445 cm⁻¹ related to the stretching vibration of quinoid (–N = (C₆H₄) = N–) ring and benzenoid rings, respectively (Mahajan et al. 2015), while the peak appearing at 1098 and 969 cm⁻¹ corresponding to the stretching of C–H in plane and C–H out of plane bending (Shah et al. 2018). The weaker band found in the finger print region shown in Figure 3(c), suggests that the doping degree of PANI is changed by α-MnO₂ NPs. So, the characteristic peak appearing at about 508 cm⁻¹ with better intense peak than pure PANI indicated the incorporation of α-MnO₂ on PANI conducting matrix.

Figure 3. FTIR spectra of (a) α-MnO₂ (blue line), (b) PANI (black line), and (c) α-MnO₂/PANI (red line) in the range from 4000 to 400 cm⁻¹ with KBr pellet sampling.

Thermal analysis

TGA-DTA

The thermal stability of α-MnO₂, PANI, andα-MnO₂/PANI composite were investigated by TGA (Figure 4(a)) and DTA (Figure 4(b)). As shown in Figure 1(a), the α-MnO₂ has a weight loss of −0.660 mg (−5.024%) due to existing water removal from the surface and lattice of the nanostructure before 250°C at about 146.13°C. The weight loss of −1.101 mg (−8.381%) at 888.16°C is due to the transformation of α-MnO₂ to Mn₂O₃ (Hu et al. 2015). The four major stages of weight loss have been observed on PANI thermal behavior. The first loss observed at about 37.67°C is due to the loss of volatile impurities while the second loss obtained at about 205.62°C was attributed to the loss of the dopant, the surfactant and deprotonation of the PANI backbone. At 319.78°C temperature was due to exothermic decomposition of PANI whereas the residual product weight formed after 600°C at about 692.81°C may different oxidation products after decomposition (Ansari et al. 2016). In contrast, the α-MnO₂/PANI hybrid composite showed a better smoother curve with four minimum weight loss levels, it shows its better stability. The high residual weight loss of α-MnO₂/PANI obtained after 509.59°C was −6.917 mg (−82.502%) which suggests that PANI and α-MnO₂ do not undergo complete oxidation and their composite is highly stable. Figure 4(b) attributed that the DTA curve for α-MnO₂ shows two endothermic curves at about 146.13 and 888.16°C might be the loss of physically adsorbed water molecules and phase change from α-MnO₂ to Mn₂O₃ NPs, respectively. The DTA curve for PANI at 210.46°C endothermic peak may either dehydration or dopant loss of PANI and exothermic peak at about 524.31°C is due to and PANI chain decomposition followed by heat loss (Bhanvase et al. 2015). The domination of exothermic peak at about 503.44°C, is due to the high cooling rate capacity of the α-MnO₂/PANI composite with smooth weight loss after 600°C. The decreasing of endothermic peak on the composite than PANI and absence of any other band above 600°C indicated that hybrid composite is more stable. As a support, a similar fashion has been done between PANI and Mesoporous MnO₂ by using H₂SO₄ as a dopant (Hu et al. 2015). Finally, the whole summaries of weight loss from TGA and their peak analysis from DTA are given in Tables S1 and S2.

Figure 4. Thermogravimetric analysis (TGA, curve (a)) and differential thermal analysis (DTA, curve (b)) of α-MnO₂ (blue line), PANI (black line), and α-MnO₂/PANI (red line) NMs with flow rate 50 mL min⁻¹ under N₂ atmosphere.

DSC analysis

The thermal effect on phase behavior was done by DSC analysis shown in Figure 5. Occurrence of endothermic peak at a temperature of 181.13°C indicated the glass transition (T_g) temperature and its phase change in to Mn₂O₃ under N₂ atmosphere (Figure 5(a)) (Dessie et al. 2020b). The thermographs of PANI show a broad endothermic peak at 116.37°C, which attributed to the T_g (Figure 5(b)). It is reported that, depending up on the terms of synthesis conditions PANI might be extended and occurred near 100°C (Tikish et al. 2018). The falling of broad endothermic peaks on each pure component composed from α-MnO₂ and PANI showed the clear integration of α-MnO₂ in PANI matrix to form a new and stable nanocomposite formation with loss of surface absorbed water occurrence at about 83.41°C while an exothermic temperature after 376.61°C shows the beginning of decomposition of PANI Chain degradation (Figure 5(c)).

Figure 5. DSC thermogram of α-MnO₂ (a), PANI (b), and α-MnO₂/PANI (c) in N₂ atmosphere (heating rate: 10°C min⁻¹, flow rate: 20 mL min⁻¹).

Morphology analysis

The surface topography of α-MnO₂/PANI composite as well as synthesized PANI and optimized α-MnO₂ NPs were examined by SEM analysis. Figure 6(a) shows the SEM micrograph of low aggregated with ball-shaped flower-like architecture for biosynthesized α-MnO₂ NPs using V. amygdalina leaf extract (Dessie et al. 2020b). Figure 6(b) illustrates agglomerated, non-uniform, rough, and closely packed micrographs for pure PANI with an average crystallite particle size of 2.18 nm. Careful examination attributed that the type and shape of PANI SEM micrograph highly depends on oxidant type and aniline monomer concentration. So, when PANI is synthesized using strong oxidant and high aniline concentrations under strongly acidic conditions (pH < 2.5), globular (granular) type of PANI morphology is synthesized (Mazzeu et al. 2017). The clear distribution of α-MnO₂ NPs over PANI granular is shown in Figure 6(c). This surface modification and binary composite formations (α-MnO₂/PANI) is proved from the elemental compositions shown in Figure 6(d) and confirms the presence of main elements, 89.51% (Mn, C, O); dopant elements, 7.74% (S and Cl); and trace, 2.75% (K). Such composites modified with PANI due to its electronic (10⁰–10¹ S cm⁻¹) and proton conductivities (10⁻⁹–10¹ S cm⁻¹) nature (Sapurina and Stejskal 2009), and uniformly distribution of α-MnO₂ in PANI matrix can be considered as a best electrode material in fuelcells.

Figure 6. SEM images of (a) α-MnO₂ NPs, (b) PANI, (c) α-MnO₂/PANI composite, and (d) EDX spectrum of α-MnO₂/PANI composite.

Surface roughness analysis

The surface topographic properties of MnO₂, PANI and their nanocomposite were analyzed using Gwyddion software from SEM input image shown in Figure 7. The height of all catalyst surfaces was represented using 2-D map found in Figure 7(a1–c1) while the 3-D map visualize the degree of porous and pore occurrences that might show the highest vertical compact surface height followed by visualizing the active surface sites for each catalyst (Figure 7(a2–c2)) (Sun et al. 2013). The degree of porosity and pore occurrence from 2-D and 3D maps directly showed the degree of surface roughness information. These roughness quantifications were explained in terms of mean roughness (R_a), mean square roughness (R_q), surface skewness (R_sk), and kurtosis coefficient (R_ku) (Zhao et al. 2019).

Figure 7. 2-D and 3-D SEM images of (a1) and (a2) from catalyst α-MnO₂; (b1) and (b2) from catalyst PANI; (c1) and (c2) from catalyst α-MnO₂/PANI.

The quantitative roughness parameters were determined using row statistical quantity analysis approach and given in Table 3. The R_a value of α-MnO2/PANI catalyst was 22.92 ± 3.648 nm. It has a higher value than α-MnO₂ and PANI, showing that the surfaces of α-MnO₂/PANI hybrid composite being rougher than other catalysts. R_q values of α-MnO₂/PANI hybrid composite was 28.15 ± 3.645 nm, its value was higher than values determined for α-MnO₂ and PANI, indicating that the degree of surface roughness variation in the hybrid composite are higher. The R_sk values of α-MnO₂/PANI shows more positive value than α-MnO₂, meaning that composite has more regions of positive curvature than peaks in the surface but lower in positive value than PANI, while values of α-MnO₂ catalyst has less positive (more negative curvature), indicating that there are fewer troughs as shown in Table 3. Finally, The R_ku values of α-MnO₂/PANI catalyst was −0.2429 ± 0.709 and the R_ku values of PANI was −0.5079 ± 0.342. The results show that the shape of pore size distribution of MnO₂/PANI nanocomposite is scattered while the pore size distribution of α-MnO₂ catalyst is more concentrated with a value of 0.9627 ± 1.798. So, R_ku values supports that the pore size distribution and surface area to volume ratio of PANI surface resulted from Watershed pore statistics analysis method is intermediated between MnO₂ and MnO₂/PANI as shown in Figure 8 and (Table S3).

Figure 8. Dependence of mean pore size and surface area to volume ratio vs. number of pores for α-MnO₂, PANI, and α-MnO₂/PANI catalysts by the Watershed method.

Table 3. Surface roughness statistical quantity results for different samples acquired from SEM entire image analysis.

	Row method results
Catalyst	Ra (nm)	Rq (nm)	Rsk	Rku
α-MnO2	13.11 ± 3.163	17.21 ± 3.587	−0.8903 ± 0.5439	0.9627 ± 1.798
PANI	19.96 ± 2.264	24.22 ± 2.423	−0.3789 ± 0.1872	−0.5079 ± 0.342
α-MnO2/PANI	22.92 ± 3.648	28.15 ± 3.645	−0.6072 ± 0.3500	−0.2429 ± 0.709

Note: R_a, mean roughness; R_q, mean square roughness (RMS); R_sk, surface skewness; R_ku, kurtosis coefficient.

Pore structure and size distribution analysis

Watershed method is more realistic techniques for pore and their distribution determination in the range between 1 and 200 nm while the rest method has a large range of size with more complicated structural determination including pore with bigger in size (Zhao et al. 2019). So that is based on the principle of water flows to a region with a local minimum which represents a plane pore. The method has two specific processes, such as water drops to each point on the surface of the catalyst and water flows to the region of the local minimum. Then the pores are identified with the pore size, pore surface area, and pore volumes were determined by the volume of water converging to the local minimum region. Based on these prospective, all pores associated with all catalysts are marked in blue color (Figure S6(a1–c1)) and their detailed pore parameters are shown in Table S3. The distributions of pore structure and pore numbers as a function of pore sizes are shown in Figure S6(a2–c2). The maximum numbers of pores are formed in α-MnO₂/PANI nanocomposite with average maximum pore size of 104.3 ± 6.548 nm while lowest average maximum pore size of 91.19 ± 8.996 nm was for α-MnO₂ nanocatalyst. By comparison, MnO₂/PANI nanocomposite catalyst has more surface area to volume ratio than MnO₂ and PANI catalysts with a value of 24.08, 23.09, and 17.88 µm⁻¹, respectively. Summarizing of pore size distribution with other literature work is given in Table 4.

Table 4. Pore size distribution comparison between this work and other literature.

Synthesis method	Catalyst	Pore size distribution (nm)	Reference
CVD	CNT/MOF/PANI	50–100	(Gong et al. 2019)
Deposition and oxidative polymerization	rGO/MnO2/PANI	2–120	(Etana et al. 2019)
Solution-phase assembly	MnO2/Graphene	∼8.9	(Wu et al. 2010)
Hydrothermal synthesis	Co/MnO2	3.7	(Kim et al. 2014)
Redox reaction	HTC@MnO2	10.24–11.09	(Mu et al. 2015)
Breath figure technique	α–MnO2	150–70	(Shimamura et al. 2019)
Electrochemical co-deposition	PANI/MnO2	100–200	(Shah et al. 2018)
In situ reduction	MnO2/Carbon	3.5–3.6	(Dong et al. 2006)
One-step pyrolysis and in situ polymerization.	N-HPCN/PANI	1.0	(Yu et al. 2019)
Micelle template and oxidative polymerization	NiO/MnO2@PANI	3.6	(He et al. 2017)
Redox reaction	MnO2/CN	2–8	(Le et al. 2016)
Green synthesis and in situ polymerization	α–MnO2/PANI	45.81 ± 12.16	This study

CVD = Chemical vapor deposition, CNT = Carbon nano tube, rGO = Reduced graphene oxide, Co = Cobalt, HTC = Hollowed-out tubular carbon, N-HPCN = Nitrogen-doped hierarchical porous carbon nanospheres, CN = Carbon nanofibers.

Electrochemical characterization

Cyclic voltammetry (CV)

The electrocatalytic activities of the prepared materials were evaluated using CV under phosphate buffer electrolyte solution are given in Figure 9(a). The peak current observed using bare PGE was lower than the other electrodes, due to its limited electrical conductivity. The lower conductivity found on bare PGE stores lower charge capacity (Figure 9(b)). This electrode was improved by α-MnO₂, PANI, and α-MnO₂/PANI; their peak currents were increased shown in Figure 9(a). Pair of redox peak current for PANI/PGE and α-MnO₂/PANI/PGE was increased due to the better electrocatalytic activity of PANI. This activity increases the total charge storing performance of α-MnO₂/PANI/PGE shown in Figure 9(b). Therefore, among the electrodes, α-MnO₂/PANI/PGE exhibited maximum peak currents and this improves its electrical conductivity and conduction channels. The peak to peak potential separation (ΔE_p) of α-MnO₂/PANI/PGE has a minimum value (0.224 V) when compared with other electrodes, this directly indicated the best in electron transfer rate. For more understanding regarding peak current, peak potential, and charge comparison see (Table S4). Hence, α-MnO₂/PANI composite modified PGE promotes fast electron transfer movement between electroactive species and exposed surface area of electrode interfaces. This statement was supported by surface roughness and pore structure analysis. Besides, the shift in anodic peaks potential towards more negative scan showed that oxidation activity of analyte on α-MnO₂/PANI surface takes place at a faster rate than bare PGE. So, the degree of oxidizing ability increases from bare PGE to α-MnO₂/PANI (Table S4). Finally, the possible oxidation responses (oxidation reaction) at −0.570 V vs. Ag/AgCl and its reversible reduction responses at −0.794 V vs. Ag/AgCl might be given from the following Equations. This reaction was adopted from other literature under PBS as an electrolyte (He et al. 2017): (5) $\begin{array}{rl}& \text{Anode reaction}:\left[\text{M}\left(\text{s}\right)+{\text{HPO}}_4^{2-}\left(\text{aq}\right)\overrightarrow{\leftarrow \phantom{{}_{}}}{\text{M}}_3\text{P}{\text{O}}_4\left(\text{s}\right)\right.\\ & \left.+{\text{H}}^{+}\left(\text{aq}\right)+3{\text{e}}^{-}\right]\times 4\end{array}$(5) (6) $\begin{array}{rl}& \text{Cathodic reaction}:\left[{\text{O}}_2\left(\text{g}\right)+4{\text{H}}^{+}\left(\text{aq}\right)+4{\text{e}}^{-}\right.\\ & \left.\overrightarrow{\leftarrow \phantom{{}_{}}}2{\text{H}}_2\text{O}\left(\text{l}\right)\right]\times 3\end{array}$(6) (7) $\begin{array}{rl}& \text{Overall reaction}:\left[12\text{M}\left(\text{s}\right)+4{\text{HPO}}_4^{2-}\left(\text{aq}\right)+3{\text{O}}_2\left(\text{g}\right)\right.\\ & \left.+8{\text{H}}^{+}\left(\text{aq}\right)\overrightarrow{\leftarrow \phantom{{}_{}}}4{\text{M}}_3\text{P}{\text{O}}_4\left(\text{s}\right)+6{\text{H}}_2\text{O}\left(\text{l}\right)\right]\end{array}$(7)

Figure 9. Cyclic voltammogram (a) and the total charge exchanged from the beginning of the experiment, Q–Q_o (b) of different catalyst modified electrodes. (i) bare PGE; (ii) α-MnO₂/PGE; (iii) PANI/PGE; (iv) α-MnO₂/PANI/PGE in 100 mM PBS (pH = 7.4) at a scan rate of50 mV s⁻¹.

Where M stands for catalysts such as PGE, α-MnO₂, PANI, and α-MnO₂/PANI and the CV techniques was conducted in dissolved oxygen conditions.

The effect of scan rate on the produced peak current by α-MnO₂/PANI hybrid composite modified PGE was studied using CV in 100 mM PBS (Figure 10(a)). It was performed at scan rates of 5, 10, 20, 50, 100, and 200 mV s⁻¹. The obtained voltammograms show that the produced peak current was enhanced with increasing scan rate. The dependence of plot of anodic and cathodic peak currents over the scan rates (Figure 10(b)), provides a nonlinear relationship with R² equal to 0.84117. This result shows that both oxidation and reduction reactions on α-MnO₂/PANI/PGE surface are not under surface adsorption controlled process (Dessie and Admassie 2013). On the other hand, the plots of the anodic and cathodic peak currents were linearly dependent on the square root of the scan rate (Figure 10(c)). This indicates that the redox process occurring on the electrode surface is nearly reversible (Mishra and Jain 2016) and clearly diffusion controlled (Tahtaisleyen et al. 2020). The result demonstrated that a linear relationship with the oxidation of buffer solutions by producing H⁺ ion to the solution on MnO₂/PANI/PGE has controlled by the diffusion process (Tahtaisleyen et al. 2020). It is concluded that, if the redox reaction is diffusion controlled, the peak current should be linearly proportional to square root of scan rate (Ates et al. 2009), and negative intercept = −0.37604 with R² = 0.97419 from logarithm of the scan rates vs. logarithms of anodic peak current are shown in Figure 10(d) which suggests there might be some adsorption intermediate in the process (Mphuthi et al. 2017).

Figure 10. Cyclic voltammograms of α-MnO₂/PANI/PGE electrode recorded at scan rates of 5–200 mVs⁻¹ (inner to outer) (a), plot of peak currents vs. scan rates (b), variation of peak currents vs. square root of scan rate (c), and variation of log v vs. log I_pa (d) in 100 mM PBS (pH = 7.4).

Impedance spectroscopy

The obtained parameters from EIS experiment results were presented in Table 5. The Nyquist plots of bare PGE, α-MnO₂/PGE, PANI/PGE, and α-MnO₂/PANI/PGE electrodes are shown in Figure 11(a). The simulation equivalent circuit analysis of R₂ or R_ct values for PGE, α-MnO₂/PGE, PANI/PGE, and α-MnO₂/PANI/PGE were found to be 1519, 1119, 1071, 958 Ω, respectively (Table 5). The decrease in charge transfer resistance is due to the effective modification of bare PGE evidenced from a successful integration between α-MnO₂ and PANI, and played a fast ET rate. Thus, the low R_ct value of α-MnO₂/PANI composites demonstrates a high reaction rate due to having high reactive surface site. The electrolytic ohmic resistance (R_s) obtained by α-MnO₂/PANI/PGE has a lower values (8.86 Ω) than other electrodes due to its effective surface roughness modification results for lower intrinsic resistance between composites and bare PGE interfaces. This provides a fast diffusion of the electrolyte from its bulk region to the surfaces of electroactive materials that leads R_s become small. The statement is clearly supported from Warburg impedance (Z_W) results as shown in Table 5. The highest double-layer capacitance obtained by hybrid composite modified PGE increases due to the faster electron transfer rate properties resulting from having lower polarization and ohmic resistances.

Figure 11. EIS plots Nyquist (a) and the inset displayed the equivalent circuit (Randles model with R.E is reference electrode and W.E is working electrode), Bode-magnitude (b), Bode-phase (c), and Admittance (d) plots of bare PGE, α-MnO₂/PGE, PANI/PGE, and α-MnO₂/PANI/PGE electrodes performed in 100 mM PBS (pH = 7.4).

Table 5. Electrochemical parameters of the bare PGE, α-MnO₂/PGE, PANI/PGE, and α-MnO₂/PANI/PGE electrodes in 100 mM PBS (pH = 7.4) supporting electrolyte.

Electrode	R1 (Rs)/Ω	R2 (Rct)/Ω	C2 (Cdl)/F	W2 (Zw)/Ω s^−1/2	|Y| (mS)	Log Z/Ω	θ (°)	I/mA	f/kHz
PGE	11.91	1519	5.12 × 10^−5	25,283	92	1.07	−72.851	0.0029	19.15
α-MnO2/PGE	11.54	1119	7.42 × 10^−5	13,283	95	1.06	−70.039	0.0111	28.32
PANI/PGE	10.47	1071	8.50 × 10^−5	736.1	106	1.01	−72.854	0.0119	19.15
α-MnO2/PANI/PGE	8.86	958	8.01 × 10^−4	448.5	126	0.94	−73.182	0.0124	19.15

The impedance magnitude (|Z|) and phase angle (Φ) dependence on frequencies are shown in Figure 11(b,c). From the frequency dependence of log |Z| versus frequency, it can be seen that α-MnO₂/PANI composite coated PGE exhibited with lowest log |Z| value of 0.94 Ω at 19.15 kHz, indicating that its electron transfer process is best (Table 5). A typical capacitive and resistive behaviors can also obtained from Φ dependence versus frequency plot as shown in Figure 11(c). The capacitive behavior of the material have been found in the phase angle between 90° and 45° while the resistive behavior were governed between 45° and 0°, and finally −90° phase angle conforms the ideal supercapacitor properties of the prepared material (Sankar et al. 2018). All modified electrodes including bare PGE showing a capacitive behavior. Their optimum phase angles are found in the range between −45° and −90°, while the capacitive behavior of composite modified PGE exhibited with highest value of −73.182°. This result revealed that α-MnO₂/PANI/PGE has a pseudo-capacitive behavior. Therefore, the hybrid composite could have a benefit for energy conversion and storage performance applications due to its pseudo-capacitive behavior contribution of α-MnO₂ on PANI (Relekar et al. 2019).

The admittance plots of bare PGE, α-MnO₂/PGE, PANI/PGE, and α-MnO₂/PANI/PGE nanocomposites were done to understand the conductivity of materials. The highest conductivity obtained by α-MnO₂/PANI modified PGE exhibited the highest magnitude of conductance (|Y|) (126 mS) (Figure 11(d)). Bode-phase plots from the frequency dependence current and the current dependence of Phase angle are shown in Figure 12(b,a)). From these figures, the highest current generation on α-MnO₂/PANI/PGE throughout the frequency is shown in Figure 12(b), indicating that the lower ohmic resistance, charge transfer, and diffusion resistance associated with the modified electrode surface. General performances done by impedances for all materials are comparative and consistent with results obtained by cyclic voltammetric results. So, for more data summary, impedemetric parameters are given in Table 5.

Figure 12. Plots of current vs. frequency (a) and Phase angle vs. current (b) for PGE (i), PANI/PGE (ii), α-MnO₂/PGE (iii), and α-MnO₂/PANI/PGE (iv).

MFC activity test

Electricity generation

An OCV is one of the most important criteria to investigate the catalytic activity of the prepared materials for better MFC performance. The OCV responses obtained by α-MnO₂, PANI, α-MnO₂/PANI modified PGE and bare PGE were operated for four continuous days each with a total of 16 days (Figure 13(a)). On each day a continuous 3 h measurement per day was operated. Throughout the time of operations, maximum OCV of 471.17 ± 89.82 mV (day 1), 348.58 ± 42.04 mV (day 2), 650.61 ± 10.11 mV (day 3), 222.36 ± 8.16 mV (day 4) were produced by PANI/PGE, α-MnO₂/PGE, α-MnO₂/PANI/PGE, and PGE, respectively. The selected time is considered as the best time to operate MFC at steady state. From the investigation, higher OCV using PANI/PGE and α-MnO₂/PANI/PGE anode catalyst during the first and third day operating time was due to the effective electrogenic E.coli adhesion with modified anode to form a conductive biofilm as shown in Figure 13(b). The maintained OCV after 3 days of operation was found to be higher with α-MnO₂/PANI hybrid composite modified PGE that leads to improving fast electron transfer rate on anode surface in MFC. From Figure 13(b), throughout the days of operations, the hybrid composite enhances MFC performance due to an increase in active surface area formation. (Table S5) compares the OCV produced by four anodic electrodes under relatively identical conditions. The unusually high OCV may be attributed to the high charge density developed on the modified electrode forms a bioanode. Thus, the fabricated nanocomposite could be suggested as a good and effective anode material to increase performance in MFC operation system.

Figure 13. Variation of open circuit voltage (a) and variation of average open circuit voltage (b) with time.

Polarization curve and power density

One of the most important parameter to characterize MFC performance is polarization curve which represents voltage as a function of current (Agarry 2017). The electrocatalytic activity of α-MnO₂, PANI, and α-MnO₂/PANI hybrid composite modified PGE towards MFC performances were shown in Figure 14(a,b)). Bare PGE exhibited the power and current densities of 65.74 ± 0.96 mW m⁻² and 436.52 ± 79.83 mA m⁻², respectively, which has the lowest values due to the inefficient electrical conductivity. The α-MnO₂/PGE exhibited an improved MFC performance due to its ability to motivate a fast electron transfer rate between bacterial colonies and the electrode surface. The α-MnO₂ NPs have a significant achievement to increase the specific capacitance and cycling stability on PGC anodic surface by forming composites under near and neutral solutions. It is very helpful to avoid the swelling and shrinkage behavior of PANI in the neutral medium of conditions. Thus, to solve these problems, α-MnO₂ provided a number of active surface sites for the polymerization growth of PANI. The presence of a number of active sites which was proved by surface roughness and pore size distribution analysis would help to increase the surface areas of anode; hence integrating α-MnO₂ in PANI matrix improves electrode activity by minimizing charge transfer resistance due to the effective interactions between E.coli outer membrane cell with the modified electrode. Such interactions promoted to produce a higher power output and current densities to 426.26 ± 38.89 mW m⁻² and 2485.51 ± 397.31 mA m⁻², respectively, using α-MnO₂/PANI composite modified PGE. Since the power density formed by PANI/PGE is also much higher in magnitude than bare PGE but is closer in power density with α-MnO₂/PANI/PGE composite at higher load resistance region. This is might happen when a peptide bond formation between E.coli and PGE surface which serve as a road to transfer electron rate effectively (Yaqoob et al. 2020). It is reported that an electrogenic bacteria can transfer electrons produced intracellularly by their central metabolism to an extracellular anode which serves as electron acceptor via either self-secreted electron mediators, direct cell membrane-anode coupling, or electrically conducting pili (Qian et al. 2009). Once these electrons arrive at the α-MnO₂/PANI modified PGE is elevated to the higher potential cathode through an external circuit, forming a current is the general basic principle for the current generation in an MFC. So, the possible electrochemical reactions occurred at the interface between anode/anolyte and cathode/catholyte can be described by the following chemical reactions: (8) $\begin{array}{rl}& \text{Anode half reaction}:{\text{C}}_6{\text{H}}_{12}{\text{O}}_6+6{\text{H}}_2\text{O }\\ & \stackrel{\text{Microorganism}}{\to }6\text{C}{\text{O}}_2+24{\text{H}}^{+}+24{\text{e}}^{-}\end{array}$(8) (9) $\begin{array}{rl}& \text{Cathode half reaction}:{\text{MnO}}_4^{-}+2{\text{H}}_2\text{O}+3{\text{e}}^{-}\\ & \to \text{ Mn}{\text{O}}_2+4\text{O}{\text{H}}^{-}\end{array}$(9) (10) $\begin{array}{rl}& \text{Overall cell reaction}:{\text{C}}_6{\text{H}}_{12}{\text{O}}_6+8{\text{MnO}}_4^{-}\\ & \to 6\text{C}{\text{O}}_2+\text{Mn}{\text{O}}_2+2{\text{H}}_2\text{O}+8\text{O}{\text{H}}^{-}\end{array}$(10)

Figure 14. Polarization (a) and power density (b) curves.

The generated proton from the metabolism of glucose to carbon dioxide is passed through salt bridge which acts as a proton exchange membrane, to balance the produced charge in the cathode compartment.

It is clear that the transfer of electrons between π orbitals of PANI and d orbitals of Mn⁴⁺ stimulated the better electrical conductivity in the α-MnO₂/PANI composite, which enable to increase glucose substrate oxidation during MFC operation. So, during glucose decomposition the produced power density was higher in α-MnO₂/PANI/PGE than other catalysts including bare electrode (Figure 15(a)). Again, the higher electrical conductivity and presence of rich surface active site helps to reduce using higher external load shown in Figure 15(b). A strong bond creation between the PGE layer and PANI backbone via π-π stacking, electrostatic interactions or hydrogen bonding forms an excellent conduction channels.

Figure 15. Dependence of power density with time (a) and external resistance (b), and profile of current discharge with time at different external resistance (c).

The current discharge is shown in Figure 15(c) represents the charge storage capacity by the modified anode during MFC operation. So, the charge storage capacity produced by α-MnO₂/PANI composite modified PGE becomes higher. The free flow of charge in the MFC is always hindered as a result of the presence of a different kind of anodic resistance (Mahadevan et al. 2014). Increasing the conductivity of anode materials could be able to minimize contact resistance, ohmic losses, and slow electron transfer rate. As time goes on at higher external resistance, the area under current vs. time slows down due to the charge loss from glucose concentration reduction. It is clear that the smaller α-MnO₂ NPs integration in PANI conductive matrix created a good synergetic effect than pure PANI and α-MnO₂ NPs results for stabilized nanocomposite catalyst formation in the MFC. Thus, the maximum current discharge observed using α-MnO₂/PANI/PGE was 0.1715 mA at 1000 Ω while the current discharge observed with bare PGE (0.0301 mA) at 5000 Ω. Finally, the optimum power density, operating voltage, stabilized time, and resistances are summarized in Table 6. Table 7 also summarizes the comparison of our experimental result with other previous literature work.

Table 6. Summary of optimum power density, operating voltage, stabilized time and resistance.

Electrode	Time (min)	External resistance (Ω)	Operating voltage (mV)	Current density (mA m^−2)	Power density (mW m^−2)
PGE	210	5000	150.6	436.52 ± 79.83	65.74 ± 0.96
α-MnO2/PGE	135	900	140.0	2254.43 ± 237.41	315.62 ± 4.87
PANI/PGE	150	1000	170.8	2475.36 ± 3.58	422.79 ± 21.77
α-MnO2/PANI/PGE	150	1000	171.5	2485.51 ± 397.31	426.26 ± 38.89

Table 7. Comparison of unmodified PGE, α-MnO₂, PANI, and α-MnO₂/PANI/PGE anode-based MFC with other literatures

Anode volume (mL)	Substrate (Fuel)	Inoculum (Electrogenic bacteria)	Anode material	Cathode material	Catholyte	Membrane/ saltbridge	Pmax (mW m^−2)	Imax (mA m^−2)	COD	Reference
100	Lactate	S. oneidensis	PANI/GNRs/CP	CP	K3Fe(CN)6 (50 mM)	Nafion 211	856	5800	−	(Zhao et al. 2013)
40	Acetate	Anaerobic sludge	PANI/rGO/CC	Carbon felt	K3Fe(CN)6 (50 mM)	Nafion 117	1390	2670	−	(Hou et al. 2013)
500	Glucose	E. coli	PANI/GCE	Platinum rod	K3Fe(CN)6 (100 mM)	Nafion 112,	820	1400	−	(Mehdinia et al. 2013)
−	Lactate	S. oneidensis	PANI-TA/CC	CC	K3Fe(CN)6 (50 mM)	Nafion 117	490 ± 12	∼1500	−	(Liao et al. 2015)
100	Acetate	−	MnO2/PANI/MnO2/CF	GR	K3Fe(CN)6 (10 g/L)	Nafion 117	1124.8	566,000	−	(Wang et al. 2017)
4	Lactate	S. loihica	CNTs/PANI/APTES/ITO	Platinum wire	Buffer	−	34.51	69.8	−	(Wu et al. 2018)
250	Glucose	E. coli	CF/BC/PANI	Carbon tissue	K3Fe(CN)6	Nafion 115	−	90	−	(Trindade et al. 2020)
0.8	Glucose	Sewage treatment plant	MoO2/PANI/CC	Pt/CC	Air	−	1101	−	−	(Geetanjali and Kumar 2019)
250	Acetate	Anaerobic sludge	PANI/Carbon felt	Carbon felt	−	−	216	3300	77.85 ± 2.32	(Rajesh et al. 2020)
250	Sewage waste water	Mixed bacterial	MWCNT-MnO2/PPy/CC	CC	K3Fe(CN)6 (40 mM)	PEM	1125.4	3.35 ± 0.6	92	(Mishra and Jain 2016)
−	Acetate	Anaerobic activated sludge	CNT/MnO2/CP	Pt/CP	Oxygen	Glass bridge & Nafion membrane	20 ± 1.7	262 ± 0.015	−	(Kalathil et al. 2013)
24	Acetate	Anaerobic activated sludge	NiO/MnO2/Carbon felt	Carbon felt	K3Fe(CN)6 (50 mM)	PEM	628	−	58.2	(Zhong et al. 2018)
28	Acetate	Pre-acclimated bacteria from other reactors	rGO/MnO2/ Carbon felt	Platinum wire	K3Fe(CN)6	PEM	2065	∼ 6500	−	(Zhang et al. 2016)
25	Glucose	Anaerobic sludge	CC	MnO2/GO/CC	Air	Nafion 117	3359	∼3600	−	(Gnana kumar et al. 2014)
−	Acetate	Mixed bacterial culture	PPy/MnO2/CC	CC	K3Fe(CN)6 (50 mM)	Nafion 117	2139.7 ± 17.5	8000	−	(Zhao et al. 2020)
150	Acetate	Anaerobic sludge	Carbon felt	MnO2/ Carbon felt	Air	Clayware wall material	27.12	217.5	62.8 ± 2.15	(Das et al. 2020)
−	Acetate	Anaerobic sludge	CC	MnO2/graphene/CC	Air	Ion exchange membranes	12.5	∼125	86.2 ± 4.9	(Elawwad et al. 2020)
500	Glucose	Bacillus subtilis	Sb-SnSb@rGO	MnO2/rGO	−	Nafion 117	7.6	44.77	49.23	(Divya Priya et al. 2020)
250	Glucose	E. coli	α-MnO2/NiO/PGE	PGE	KMnO4 (50 mM)	Agar and KCl	412.24 ± 5.98	2576.49 ± 18.88	−	(Dessie et al. 2021)
250	Glucose	E. coli	PGE	PGE	KMnO4 (50 mM)	Agar and KCl	65.74 ± 0.96	436.52 ± 79.83	67.00	This work
250	Glucose	E. coli	α-MnO2/PGE	PGE	KMnO4 (50 mM)	Agar and KCl	315.62 ± 4.87	2254.43 ± 237.41	75.08	This work
250	Glucose	E. coli	PANI/PGE	PGE	KMnO4 (50 mM)	Agar and KCl	422.79 ± 21.77	2475.36 ± 3.58	78.89	This work
250	Glucose	E. coli	α-MnO2/PANI/PGE	PGE	KMnO4 (50 mM)	Agar and KCl	426.26 ± 38.89	2485.51 ± 397.31	88.19	This work

CC = carbon cloth, CP = carbon paper, GNRs = Graphene nanoribbons, GCE = Glassy carbon electrode, TA = tartaric acid, CF = Carbon felt, GR = graphite rode, CF = Carbon fiber, BC = Bacterial cellulose, CNTs = Carbon nanotubes, APTES = Υ-aminopropyltriethoxysilane, ITO = Indium-tin oxide, PEM = Proton exchange membrane, Antimony-tin (Sb-Sn).

Organic matter removal and coulombic efficiencies

Organic matter removal concentration and removal efficiency in terms of COD during OCV measurements were shown in Figure 16(a,b). After 4 days with 3 h per day measurement, an average concentration removal of 59.05 ± 1.11 mg L⁻¹, 34.69 ± 0.60 mg L⁻¹, 33.23 ± 0.27 mg L⁻¹, and 12.84 ± 0.29 mg L⁻¹ was noted using PGE, α-MnO₂/PGE, PANI/PGE, and α-MnO₂/PANI/PGE, respectively (Table S6). The lower concentration removal with best COD removal efficiency of 92.27% was obtained by α-MnO₂/PANI/PGE was due to the better integration between α-MnO₂ and PANI, results for active biofilm formation that enhances the fast anaerobic respiration for E.coli. The image obtained from the anodic surface topography clearly showed the adhesion of the E.coli microorganism on the composites is much effective than the individual catalysts. This indicated that the fast electron transfer across the composite facilitates glucose decomposition which insured its excellent removal efficiencies. Similar results from PANI modified chitosan/titanium carbide composites anode electrode were reported with COD removal efficiency of 87–93% (Karthikeyan et al. 2016). A slight higher in both glucose concentration and its removal efficiency under OCV and CCV obtained in α-MnO₂/PANI modified PGE compared with other anode catalysts (Figure 16(a,b)) might be better due to the enhanced glucose decomposition caused by the enriched electroactive E.coli on its surface because of its improved electrochemical properties.

Figure 16. Concentration removal (a) and their efficiency removal (b) for different anodic electrodes under open and closed circuit conditions.

Closed circuit measurement (non-steady state conditions for 420 min measurement), an average concentration removal of 54.84 ± 6.54 mg L⁻¹, 41.42 ± 0.98 mg L⁻¹, 35.08 ± 3.67 mg L⁻¹, and 19.62 ± 0.92 mg L⁻¹ were recorded using PGE, α-MnO₂/PGE, PANI/PGE, and α-MnO₂/PANI/PGE, respectively (Table S7). Maximum COD and CE efficiencies of 88.19% and 26.09%, respectively, were obtained from α-MnO₂/PANI modified PGE anode electrode, whereas PANI modified PGE anode delivered a CE of 14.54% (Figure 17). The PGE electrode without any catalyst modification but under inoculum pretreatment noted a maximum CE of 2.29% only. The highest CE obtained in α-MnO₂/PANI/PGE followed by PANI/PGE anode electrodes showed that pretreatment of PGE with E.coli in the presence of α-MnO₂/PANI and PANI catalysts reduces the charge transfer resistance and enhanced the electron recovery from glucose substrate. A 14.54% increase in CE was observed in PANI modified PGE electrode compared with α-MnO₂ modified PGE and bare PGE, due to enhanced biocompatibility of the PANI modification which increases the charge recovery from glucose and resulted in an improvement in CE. On the other hand, the Mn⁴⁺ in the nanocomposite may enhance the ET between the microorganisms and the anode material which facilitates electron conduction (Kalathil et al. 2013). To support this study, modification of carbon felt anode with PANI was effective by increasing a CE from 17% to 42.45% efficiency under dual chamber MFC Construction and operation (Rajeshet al. 2020).

Figure 17. COD removal and coulombic efficiency for all anodic electrodes under glucose substrate during closed circuitconditions.

Anode electrode surface property tests

After MFC operation test, a piece of α-MnO₂, PANI, and α-MnO₂/PANI modified PGE with E.coli attachment to verify biofilm on anode were studied using SEM imaging. The surface of bare PGE before E.coli attachment shows porous topographic view, as shown in Figure 18(a). After inserting in an anodic chamber containing colonies of E.coli, it is easily observed that portions of E.coli cells are adhered to PGE (Figure 18(b)) due to its hydrophilicity nature and better surface active area. As shown in Figure 18(c,d), the PANI/PGE anode exhibits a higher bacterial colony loading than α-MnO₂/PGE, which can be attributed to the positively charged PANI backbone and its electrostatic interaction with the negatively charged E.coli outer cell membrane. It is interesting that after insertion and integration of α-MnO₂ in PANI, it is noted that E.coli adhere effectively to the α-MnO₂/PANI/PGE anode surface to form a thick biofilm (Figure 18(e)), it demonstrates that the E.coli loading is effectively improved and the synergistic effect between α-MnO₂ and PANI are remarkably active. The magnified images shown on the inset SEM topography of composites investigated that E.coli cells attached very tightly to the anode surface to form an effective bioanode electrode. All these modifications have a positive result in the direct electron transfer improvement between microbe outer membrane c-type cytochromes of E.coli and α-MnO₂/PANI/PGE anode electrode. Because the involvements of PANI in α-MnO₂ possess high electron mobility, stability, biocompatibility, anticorrosion nature, excellent electrokinetics which improves the anode performance via enhanced microbial adhesion. General observations from the SEM images clearly indicate that coating of PGE with PANI integrated α-MnO₂ resulted in surface area improvement of the anodes which provided more area of the biofilm growth and adhesion of the EAB than the unmodified bare anode electrode (Sonawane et al. 2018). So, from such architecture, the α-MnO₂/PANI/PGE anode surface provide a large surface area for enhanced electron collection from the fast glucose decomposition during E.coli respiration, indicated the best anode than other three electrodes. Therefore, α-MnO₂/PANI/PGE anode is a favorable electrode for biofilm growth results for improved MFC performance towards better power production and wastewater treatment efficiency.

Figure 18. SEM images of (a) bare PGE and the attachment of E. coli cells on (b) PGE, (c) α-MnO₂/ PGE, (d) PANI/PGE, and (e) α-MnO₂/PANI/PGE anodes.

Conclusions

Biosynthesized α-MnO₂ integrated PANI composite was proposed as an alternative bioanode electrocatalyst in double-chambered MFCs. The α-MnO₂ supported PANI nanocomposite was synthesized through simple in situ polymerization in which that α-MnO₂ NPs act as an oxidant and its composite formation was identified from the structural characterizations. The interfacial properties proved by EIS between the composite-based electrode surface and PBS supporting electrolyte were promoted by the increased EC of α-MnO₂/PANI/PGE composite. The lower charge transfer resistance, high electrical conductivity, and number of pore size which shows the catalytic active sites observed by α-MnO₂/PANI/PGE composite harvest excess number of electrons that helps to validate maximum MFC performance. All the characteristics physicochemical and electrical properties were enough to perform the required test. So, using these composite, an effective power output, current generation, wastewater treatment and coulombic efficiencies were demonstrated under moderate temperature (28 ± 2°C). The PGE support α-MnO₂ prevents PANI from its swelling at or near neutral aqueous solutions (PBS, pH of 7.4) that improved more than 6-fold power density efficiency than PGE alone throughout the cell operation. From the surface roughness and pore size distribution analysis, the structure of MnO₂/PANI nanocomposite has a better surface roughness, surface area, and pore size distribution than other pure NMs. This improvement is attributed to the better biocompatibility (facilitate bacterial adhesion) and stronger interaction of α-MnO₂/PANI nanocomposite with E.coli, improved electrochemical activity and lowered over potential than bare PGE. Therefore, in this study, we recommend that the constructed setup provides easy, efficient, and economical design for the future MFC upgrade.

Acknowledgements

The authors gratefully acknowledge Adama Science and Technology University for financial support to this research work. The authors also acknowledge Dr. Tadele Hunde Wondimu from Materials Science and Engineering department, Adama Science and Technology University for his valuable help during three electrode-based electrochemical measurements.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available in https://doi.org/10.6084/m9.figshare.13295507.v1