Self-assembled oil palm biomass-derived modified graphene oxide anode: An efficient medium for energy transportation and bioremediating Cd (II) via microbial fuel cells

3.2.1 Energy generation and polarization behavior

When the electricity-generating capacity of the MFCs using the L-GO and L-GO–ZnO anodes was investigated during an operational period of 80 days (Fig. 5a), both anodes revealed a similar electricity-generating ability, especially initially, despite a considerable difference in the maximum stable voltage achieved. On Day 25, the L-GO anode demonstrated the greatest generation of electricity, at 111 ± 3 mV, with a PD and CD at Point 1 of 16.872 × 10⁻⁴ mW/m² and 14.605 mA/m², respectively, compared with 191 ± 4 mV achieved by the L-GO–ZnO anode. Thus, Points 1 and 4 seemed to exhibit the highest voltage stability with an external resistance of 1 kΩ, whereas using less than 1 kΩ of external resistance achieved less stability in generating electric current. After the third cycle, both electrodes, at Points 3 and 6, showed sharp declines in voltage beginning on Day 72 until Day 80. On Day 25, the L-GO–ZnO composite anode generated the highest voltage, 191 ± 4 mV, which is approximately 1.72 times greater than that obtained with the unmodified L-GO anode. That difference stemmed from not only the presence of a metal oxide, ZnO, in the modified anode, which significantly enhanced the electron transfer from the anode to cathode, but also the anode’s high stability and biocompatibility with bacteria.

Close

Fig. 5

(a) Time–voltage curve of MFCs in synthetic wastewater of the L-GO and modified L-GO–ZnO composite anodes (i.e., 1 kΩ external resistance) and polarization plot of (b) the L-GO anode and (c) L-GO–ZnO anode.

Export to PPT

To determine the maximum power density (PD_max) and cell voltage behavior after stable performance was achieved, the polarization behavior of the MFCs was examined within 150 Ω to 10 kΩ of external resistance. The analysis also helped to evaluate the cells’ design point (CDP), which is critical for assessing the performance of MFCs. As shown by the polarization curves in Fig. 5b and c, the L-GO–ZnO anode exhibited a significantly enhanced CDP, with a PD_max of 2.55 mW/m², a CD of 135.96 mA/m², and 155 mV generated at an external resistance of 150 Ω, compared with that of L-GO, with a PD_max of 0.1076 mW/m², a CD of 15.65 mA/m², and 119 mV generated at an external resistance of 1 kΩ. With less external resistance, the voltage destabilization rate increased, whereas with more external resistance (i.e., 10 kΩ) both electrodes demonstrated less current efficiency. The potential drop and slow stabilization at low amounts of external resistance may have derived from the effective discharge of electrons. After all, higher external resistance prevents electrons from moving as rapidly and easily toward the cathode, which diminished the power generated.

Among other results, the values of internal resistance observed were 2700 Ω and 710.8 Ω with the L-GO and L-GO–ZnO anodic MFC operations, respectively, and the L-GO–ZnO anode showed less external loss, less internal resistance, and less activation energy than the L-GO anode. The setup with less internal resistance demonstrated a high rate of electron movement, which led to a greater generation of electric current than internal resistance (Abazarian et al., 2016). In general, the internal resistance of MFCs depends upon (1) the electrode’s nature and surface area, (2) its conductivity, (3) the distance between the anode and cathode, (4) the reaction on the electrode’s surface, and (5) the electrolyte’s concentration and/or nature. An external oxygen supply was also provided to enhance the reaction’s cathodic rate and, in turn, stabilize the voltage at a typically high level of resistance (Sajana et al., 2017). Most researchers consider the oxygen supply to the cathode chamber in differentiating the aerated and non-aerated MFC processes under different conditions.

3.2.2 Cyclic voltammetry and electrochemical impedance spectroscopy characterization

Next, CV and EIS were applied to study the rate of electron transfer and the electrochemical behavior of the fabricated anodes. As shown in Fig. 6a, the highest currents of the L-GO anode in the forward scan were 2.2 × 10⁻⁶ mA on Day 20, 3.0 × 10⁻⁶ mA on Day 40, 1 × 10⁻⁶ mA on Day 60, and 0.8 × 10⁻⁶ mA on Day 80. In the reverse scan, by contrast, the values were −6 × 10⁻⁶ mA on Day 20, −8.5 × 10⁻⁶ mA on Day 40, −1.2 × 10⁻⁵ mA on Day 60, and −2.3 × 10⁻⁵ mA on Day 80. Similarly, for the L-GO–ZnO anode, the values attained during the forward scan were 4.3 × 10⁻⁶ mA on Day 20, 4.0 × 10⁻⁶ mA on Day 40, 3.8 × 10⁻⁶ mA on Day 60, and 0.7 × 10⁻⁵ mA on Day 80 and in the reverse scan were −0.9 × 10⁻⁵ mA on Day 20, −1.9 × 10⁻⁵ mA on Day 40, −2.3 × 10⁻⁵ mA on Day 60, and −2.7 × 10⁻⁵ mA on Day 80 (Fig. 6b). The electrodes had the same geometrical area despite significantly different specific surface areas, of which the L-GO anodes was approximately 21.9 times greater. The maximum current registered during the forward and reverse scans indicate the highest oxidation and reduction rates during operation. For the L-GO–ZnO electrode, the current for oxidation and reduction processes peaked after Day 80 of operation, possibly due to the external supply of oxygen to the cathodic chamber, which may have enhanced the rate of electron discharge and neutralized the electrons before they approached the anode. The oxidation peak was also high due to the oil palm trunk sap, which has been recognized as an excellent source for bacteria (Liu et al., 2016).

Close

Fig. 6

Cyclic voltammograms recorded at a scan rate of 20 mV s⁻¹ on the (a) L-GO and (b) L-GO–ZnO electrodes using a phosphate buffer solution (pH = 7.0).

Export to PPT

Using the data, the specific capacitance (C_s) was determined for both electrodes. As shown in Table 3, the specific capacitance increased along with the operation time until peaking after Day 60 and thereafter decreasing. That increment during the first 60 days indicates the growth of healthy biofilm on the anode’s surface. The anode’s high capacitance occurred due to enhanced exoelectrogenic accumulation on the anode’s surface, as also observed by Hong et al. (2011).

EIS Nyquist plots (Fig. 7) were investigated within the frequency range of 100 kHz to 100 mHz at an amplitude of 1.00 mV AC, which prevented the biofilm’s detachment and decreased disturbances in the stable state. The fabricated L-GO and L-GO–ZnO anodes were presumably affected by diffusion and a kinetic reaction with the solution’s resistance (R_s) and charge transfer resistance (R_ct) (Hung et al., 2019). The EIS plot revealed bent lines or irregular semicircles at higher frequencies, which correspond to a high rate of electron transfer, and straight or irregular lines at lower frequencies, which indicate good capacitive behavior. L-GO–ZnO demonstrated low charge transfer resistance in the electrolyte and a higher electronic mobility than the unmodified L-GO. The presence of high internal resistance in the L-GO anode indicated poor electronic mobility between the redox solution and electrodes. By comparison, the internal resistance of the L-GO–ZnO was lower, which suggested a high electron transfer to the redox solution and electrodes. The low R_ct in the case of L-GO–ZnO promoted the rate of kinetic reactions and concurrently reduced the electron transfer resistance. The fitted results were determined by using the Zview software. The L-GO-ZnO anode showed a lower R_ct value (0.02 Ω) than the L-GO anode (1.00 Ω). It indicated that L-GO-ZnO anode with active biofilms shows improved electron transfer capability. That aspect made the L-GO-ZnO anode in MFCs more prolific than the L-GO anode, which facilitated better energy efficiency.

Close

Fig. 7

Nyquist plots (EIS) for the (a) L-GO and (b) L-GO–ZnO electrodes.

Export to PPT

3.2.3 Stability studies of anodes

The thermal, chemical, and mechanical stabilities of electrodes are critical factors for MFCs that determine long-term versus lack of operation. Currently, long-term stability is an emerging challenge in fabricating anodes, especially for MFCs, because some conditions can prolong the treatment of wastewater and production of energy. According to the literature, GO-based anodes usually demonstrate high mechanical, chemical, and thermal forms of stability (Mishra and Jain, 2016).

To examine the stability of the synthesized L-GO and L-GO–ZnO, thermal gravimetric analysis was first performed after Day 80 of operation in order to evaluate the thermal stability of the L-GO and L-GO–ZnO electrodes. As shown in Fig. S3a, after Day 80 at room temperature (25 °C) the L-GO and L-GO–ZnO anodes showed 5% and 1% weight loss, respectively. From 25 to 100 °C, 24% and 6% total weight loss was observed for L-GO and L-GO–ZnO, also respectively, primarily because water adhered to the anode’s surface. However, from 25 to 900 °C, L-GO and L-GO–ZnO respectively exhibited 56% and 34% total weight loss. L-GO–ZnO exhibited excellent thermal stability, thereby suggesting its eligibility to operate at moderate and relatively high temperatures.

The chemical stability of the anodes was explored via compositional and morphological analyses after Day 80 of operation. Although the full results are discussed in Section 3.2.5, it can be concluded that the chemical stability of L-GO and L-GO–ZnO was relatively high, because no chemical hazards or toxic elements appeared in the electrochemical media, and significant morphological changes were detected on the surface of the anodes after such long-term operation.

A mechanical test was additionally performed to measure the mechanical strength of the anodes, because continuous, long-term operation, especially at an industrial scale, can damage electrodes. Graphene derivatives usually possess considerable mechanical strength (e.g., 130 GPa reported by Papageorgiou et al. (2017), and as shown in Fig. S3b, L-GO could sustain a maximum pressure of 20 psi, whereas L-GO–ZnO could sustain a pressure of 30 psi. At higher pressures, both electrodes were broken or damaged. Such results suggest that both electrodes exhibited relatively high mechanical strength.

3.2.4 Removal efficiency and bacterial identification studies

As various studies have shown, MFCs are ideal for remediating wastewater, and graphene-based electrodes can enhance the generation of energy and the efficient remediation of toxic metals from wastewater (Kumar et al., 2019; Zhang et al., 2018; Yaqoob et al., 2020e). Considering such evidence, the potential of two GO-based anodes to remove 100 ppm Cd (II) from synthetic wastewater was investigated in our study.

As shown in Table 4, the removal efficiencies of the L-GO and L-GO–ZnO anodes up to Day 80 of operation were 83.5% and 90.0%, respectively. The high performance of L-GO–ZnO could be attributed to its high conductivity, which offers exceptional stability for healthy bacterial respiration due to the improved transfer of electrons and protons. By contrast, the unmodified L-GO anode’s surface failed to allow the bacteria to breath rapidly enough to release the electrons and protons. In both cases, the adsorptive properties of graphene-based material attracted Cd (II) ions to the surface, where exoelectrogenic biofilms removed the Cd (II) ions during bacterial respiration. In the first 20 days of operation, the unmodified GO anode’s removal efficiency was 1.3 times greater than the L-GO anodes. In general, removal efficiency depends heavily upon the operating conditions, including pH level, temperature, and water matrix effects. In our study, 90% of Cd (II) was removed in an oxic environment at 25 °C and a pH of 6.41 using L-GO–ZnO as the anode and oil palm trunk sap as the organic substrate. That moderate temperature and pH level were selected to capture their performance under realistic, green, competitive conditions. The result was comparable, if not superior, to the outcomes of other Cd (II) removals (i.e., 100 ppm) from synthetic wastewater by using MFCs reported in the literature.

Next, the bacterial communities’ composition in both the L-GO and L-GO–ZnO anodes was determined by studying clone libraries (Table S1). Both electrodes hosted a diversity of exoelectrogenic bacterial species, which shows promise for MFCs as a material for remediating toxic pollutants as well as generating energy. A particularly efficient species was Lysinibacillus fusiformis, detected on the surface of both anodes, which exhibits exceptional resistance to Cd (II) and other metallic ions such as lead, copper, and zinc (Mathivanan et al., 2016; Nandy et al., 2013; Páez-Vélez et al., 2019). Beyond that, Lysinibacillus sphaericus, Escherichia coli, Enterobacter spp., Chryseobacterium gleum, Bacillus spp., Citrobacter youngae, and Klebsiella pneumoniae are other well-known exoelectrogenic bacterial species that excel in remediating metals and generating energy via MFCs (Feng et al., 2014; Tkach et al., 2015). Table S1 presents an exhaustive list of the principal species responsible for Cd (II)’s removal from wastewater and generation of energy in the MFC-based approach.

3.2.5 Biofilm morphology and its biocompatibility

Also known as a “city of bacteria,” biofilm is produced in natural circumstances during bioelectrochemical process. The biofilm’s matrix consisted of water (97%), bacterial cells (2–5%), extracellular polymeric substance (EPS, 3–6%), and several trace mineral ions (Di Martino, 2018). The most important component was EPS, which aggregates the cells on biofilms due to its composition of proteins (1–60%), polysaccharides (40–95%), lipids (40%), and nucleic acids (10%)(Kumar et al., 2011). EPS was extraordinarily hydrated such that it can hold water molecules via hydrogen bonding and its quantity in biofilms determined their age. Of course, bacteria consortia and different environmental parameters, including pH and temperature, also affected the formation of biofilms. For that reason, it was necessary to evaluate the biofilm’s morphology to gauge the performance of energy generation and Cd (II) removal.

As shown in Fig. 5a, in both L-GO and L-GO–ZnO the initial voltage was rather small (i.e., 1 mV and 3 mV, respectively). Over time, however, the daily intake of organic substrate by the bacterial community produced a biofilm and enabled the bacterial colonies to work as electrochemically active agents in releasing electrons and protons. During the reaction, the capacity of bacterial growth on the anodes was examined by running the reaction in three cycles. The greatest energy was observed in the first cycle (i.e., 111 mV and 191 mV for L-GO and L-GO–ZnO, respectively) up to Day 25. Again, the removal rate increased with the passage of time and peaked at 83.5% and 90% for L-GO and L-GO–ZnO, respectively, because the bacteria present on the biofilm effectively reduced the metal’s toxic effect.

SEM analysis was conducted to characterize the bacterial colonies on the surface of the anodes (Fig. 8). The treated cathode surface was also analyzed to determine the bacterial presence on a graphite rod (i.e., cathode electrode). As shown in Fig. 8, the untreated anode did not exhibit any type of biofilm on its surface. However, after Day 80, several specific bacteria colonies were clearly identifiable on the surface of the treated anode and were considered to be responsible for the generation of energy and Cd (II)’s removal. Overall, the results indicated that a limited formation of biofilm promotes the generation of energy. Sasaki et al. (2019) also demonstrated that the limited formation of biofilm on an anode’s surface facilitates the generation of energy and the efficiency of removal. The high biomass on the anode’s surface affected the EPS’s composition, which translated into less energy generation. MFCs thus seem to support ecofriendly approaches to generating energy without producing hazardous gases or by-products. Last, SEM and EDX, performed to analyze the biofilm’s composition (Fig. S4), revealed no toxic elements on the anode’s surface, which would have hindered the growth and production of bacteria and, in turn, the formation of biofilm.

Close

Fig. 8

SEM images of the untreated (a) L-GO anode, (b) cathode (i.e., graphite rod), and (c) L-GO–ZnO anode; and of the treated (d) L-GO anode, (e) cathode (i.e., graphite rod), and (f) L-GO–ZnO anode. Scale bar: 30 µm.

Export to PPT

3.2.6 Energy generation, transportation, and metal reduction mechanism of the present study

Before metal reduction via bacterial species, energy generation and its transportation mechanism need to be understood. Bacterial species work as biocatalysts in MFCs approach, in which provided organic substrate is oxidized by bacterial species to produce electrons and protons. Several bacterial species are well known exoelectrogens and metal reducing species such as E. coli, Rhodoferax ferrireducens, Clostridium butyricum, Geobacter spp. Aeromonas hydrophila, Bacillus, Shewanella sp. and Klebsiella pneumoniae (Yaqoob et al., 2020c). However, in the present study we used oil palm trunk sap as an organic substrate that was oxidized by bacterial species to generate the electrons and protons. The oil palm trunk sap was converted into simple glucose, which was then further oxidized by bacterial species. Yamada et al. (2010) reported that oil palm sap contains more than 90% total sugar content. The oil palm trunk sap is easily and freely available in ASEAN countries and is already proven to be a good fuel source for bacterial activities (such as the Krebs cycle process) (Yaqoob et al., 2020b). The biochemical reaction of the present study can be written as: Anodic reaction: Oil palm trunck sap → C₆H₁₂O₆ + 6H₂O → 6CO₂ + 24H⁺ +24e⁻ Cathodic reaction: 24H⁺ + 24e⁻ + 6O₂ → 12H₂O Overall reaction: C₆H₁₂O₆ + 6O₂ → 6CO₂ + 6H₂O + Electricity + Biomass

The produced biofilm around the anode surface oxidized and generated electron and protons. The protons were transferred directly through PEM and the electrons were used on the outer circuit to reach the cathode for a reduction reaction. Two major mechanisms (direct and indirect electron transfer) were involved in the transportation of electrons from the bacterial cells to the anode surface before transferring toward the cathode through the outer circuit, as shown in Fig. 9.

• –

  Direct electron transfer is further classified into two processes: the electron transportation through conductive pili-type bacterial species or utilizing the redox-active proteins to transfer electrons. The Aeromonas hydrophila, Acinetobacter radioresistens strains, Klebsiella pneumoniae and many others contain conductive pili. Pili is part of the bacterium body that is conductive like metal and transfers electrons directly to any conductive material. According to the present study, the bacterial identification process showed the species that belong to the conductive pili-type bacterial class. This indicates that the present study followed the direct electron transfer mechanism. The presence of conductive pili-type species is also confirmed in the SEM images of the treated anode. Fig. 8 shows the tube/rod filamentous structures that belong to conductive pili-type species according to pervious literature (He et al., 2011; Ishii et al., 2008; Zhang et al., 2008). Further, in direct electron transfer the utilization of redox-active proteins are also commonly found in Geobacter sulfurreducens. This can produce self- redox-active proteins to use as a transferring medium. Some redox active proteins are OmcE, OmcZ, OmcS, OmcB, and OmcT and are mostly reported (Lebedev et al., 2015; Santos et al., 2015; Shang et al., 2015).

• –

  The indirect electron transfer mechanism uses oxidized/reduced soluble electron shuttles. Geobacteraceae and Desulfuromonadaceae mainly followed this mechanism to transfer electrons (Vargas et al., 2013).

Close

Fig. 9

Electron transfer mechanism from bacterial cell to the anode in MFCs.

Export to PPT

On the other side, the reduction mechanism of Cd (II) to Cd also occurred. As a reduction result, Cd (II) to Cd_(s), means the soluble state of metal ions is converted into the insoluble solid state of the Cd. It is less toxic as compared to the soluble state of Cd. The reduction of Cd (II) to Cd can be written as: Cd²⁺+ 2e⁻ Cd_(s) 2Cd²⁺ + 2H₂O 2CdO + 4H⁺ CdO + 2e⁻ +2H⁺ Cd_(s) + H₂O