Abstract
Microbial electrosynthesis (MES) has emerged as a sustainable energy platform capable of simultaneous wastewater treatment and valuable chemical production. The performance of MES, like other bioelectrochemical systems, largely depends on its electrode (cathode), providing the platform for microbial growth as well as electron transfer. However, most of the electrodes are expensive, and their nonrenewable characteristics, cost, and poisoning nature are major bottlenecks in MES commercialization. Thus, several efforts have been made to explore the potential of waste carbon-based electrodes to reduce carbon footprints as well as electrode manufacturing costs. In this study, the feasibility of using spent tea leaves (STL) and spent coffee grounds (SCG) as MES biocathode was tested. Different bioelectrochemical tests suggested improved MES performance with STL and SCG biocathode along with reduced electrode resistance and improved current density. A 1.5- and 2.0-fold increase in cyclic voltammetry (CV) current output was observed for SCG and STL, respectively, with substantial mediator peaks of high intensity indicating enhanced electrocatalytic activity. Enrichment of some fermentative and exoelectrogenic microbial classes such as Clostridia, Bacteroidia, and Deltaproteobacteria led to a 1.3- and 1.4-fold increase in butyrate production for SCG and STL cathode, respectively. These results demonstrate the potential of STL and SCG as MES cathode for improved energy and chemical production.
1. Introduction
Continuous exhaustion and depletion of natural energy resources combined with increased emissions of greenhouse gases such as carbon dioxide (CO2) has been a major problem faced by mankind today [1]. To overcome this alarming situation, microbial electrosynthesis (MES) has emerged as an efficient and multifaceted technology capable of consuming CO2 and converting it into useful chemicals such as biofuels, organic acids, and other chemicals via microbial catalysts [2]. This technology not only provides a green method of chemical synthesis, but it also reduces the carbon footprint, which makes it a competent and viable platform for sustained and simultaneous production of energy and chemical [3].
Similar to other bioelectrochemical systems, MES performance largely depends on its electrodes for microbial attachment as well as its platform for electron transfer [4, 5]. In this context, different electrode materials including carbon based electrodes decorated with nanomaterials, metals, polymers, and composites have been used [6, 7]. However, the complex and rigorous preparation processes along with the high cost hinder scaling up MESs [8]. Therefore, it is imperative to explore inexpensive, readily available, and highly efficient electrode materials. In this context, carbon-rich waste materials have become promising materials for fabricating electrodes owing to their naturally porous structure, availability, high mineral content, biocompatibility, and low cost [8]. Several electrodes prepared from carbon-rich waste materials, such as those from biochar and organic waste [9, 10], kenaf and bamboo [11], and cotton textile and chestnut shell [12], have been tested. All these materials exhibit improved bioelectrochemical performance in terms of biofilm formation, electrode-microbe interaction, and current density performance than conventional electrodes [13].
Among different carbon-rich waste materials, spent tea leaves (STL) and spent coffee grounds (SCG) are one of the most abundant by-products generated worldwide due to high global consumption of tea and coffee, which reached up to 6 and 10 million tons, respectively, in 2020 [14, 15]. Huge quantities of solid residues (~90%) of STL and SCG are produced during their brewing for different types of products such as canned, bottled, and instant products [14]. Continuous disposal of these waste products in open field is problematic and may increase anthropogenic gas emission and water and soil pollution owing to the toxic nature of its components, including caffeine, polyphenols, and tannins [15]. Notably, STL and SCG are potential energy sources enriched with carbon-derivatives such as alkaloids and polyphenols (STL) and organic compounds including fatty acids, hemicelluloses, lignin, cellulose, and other polysaccharides (SCG) [14, 15]. Recent studies have highlighted the potential of SCG and STL as a source of sugars [16], bioethanol [17], and biodiesel [18], and some studies have exploring the use of STL and SCG for fabricating electrodes [19–21]. In this context, STL and SCG are promising options for use in MESs to reduce the carbon footprint and energy consumption associated with electrode manufacturing. In this study, the bioelectrochemical performance and bicarbonate reduction to different volatile fatty acids (VFAs) of STL and SCG were monitored to evaluate their feasibility as MES cathodes (Figure 1). A mixed microbial culture was used instead of pure strains owing to its resilience to environmental disturbances and capacity to produce high biomass [22]. At the end of the experiment, the transformation and enrichment of relatively abundant microbes were evaluated to identify the microbial communities active on STL and SCG biocathode.

2. Materials and Method
2.1. Preparation of STL and SCG Electrode
The SCG was collected from a local coffee shop in Korea while STL was collected from a local tea brand. Both materials were washed and air dried. To assemble STL and SCG as electrodes, an empty tea bag was used. Five gram of each of STL and SCG was filled in empty tea bags (), and an empty tea bag was used as control. The anode used was carbon cloth, infused with 20 wt% Pt (Fuel Cell Earth, Wakefield, USA) with an effective area of .
2.2. MES System Assembly and Operation
A dual Plexiglas box () with 0.2 L capacity comprised the MES setup, as described in our previous studies [23, 24]. The cathode and anode chambers were separated by a proton exchange membrane PEM (Nafion® 117, DuPont Co., USA) pre-treated with 0.5 M H2SO4 and 3% H2O2. The experiment was run with three separate MES systems each installed with a different cathode (control/STL/SCG) while other experimental parameters were kept constant. For starting MES, each reactor was fed with growth media (0.18 L) and microbial culture (0.02 L) for four weeks before the start of experiment. To ensure smooth operation and microbial activity, exhausted growth media was replaced with fresh one at the end of each cycle. The experiment was carried for three cycles, each lasting for 120 h at constant temperature () and 180 rpm; 0.1 M phosphate buffer (pH 7.0) was used as the anolyte.
2.3. Electrochemical Measurements
A multichannel potentiostat comprising a three-electrode setup (wizECM-8100 Premium, Korea) was used for electrochemical measurements. In brief, cathode, anode, and saturated Ag/AgCl functioned as working, counter, and reference electrode, respectively. The chronoamperometry (CA) evaluated the current performance in a current density-time plot, with cathode poised at a fixed –800 mV vs. Ag/AgCl potential. The microbial exoelectrogenic activity was analyzed by cyclic voltammetry (CV) under a wide range of applied potential (–800 mV to +200 mV) and scan rate (1 mV/s to 200 mV/s). Fresh growth medium was added prior to CV analysis to ensure equilibrium between electrode and electrolyte. Electrochemical impedance spectroscopy (EIS) was used to measure the electrode resistance in an electrolyte solution mixture of potassium ferricyanide (5 mM) and phosphate buffer (5 mM). The electrode was poised at a fixed potential (+200 mV vs. Ag/AgCl) and evaluated under a broad frequency range of 100 kHz to 0.01 Hz.
2.4. Volatile Fatty Acid Detection and Microbial Community Analysis
VFA production was monitored by collecting samples every 24 h and subjecting them to a high-performance liquid chromatography (HPLC, Agilent 1200, USA). A refractive index detector (G1362A) and ion exchange column (Aminex HPX-87H; ) were used to analyze the samples at 60°C using 4 mM H2SO4 (mobile phase) under isocratic elution mode. VFAs were identified by comparing the observed peaks with available standards. The microbial communities were analyzed at the end of the experiment. Briefly, the supernatant was removed by centrifuging samples at 12,000 rpm and storing the microbial pellet for further analysis. The detailed microbial community analysis using bacterial 16S ribosomal ribonucleic acid along with the list of primers and templates (Table S1) used for the polymerase chain reaction is provided in supplementary information (available here).
3. Result and Discussion
3.1. Current Performance
The current performance was evaluated for MES cycle 1, 2, and 3 (Figures 2(a)–2(c)), each lasting 5 days with MES system poised at fixed cathodic potential of −800 mV (vs. Ag/AgCl). The negative potential enabled the electron transfer from cathode to outer cell membrane proteins/cytochromes, depicting its electrochemical activity [25]. The current output was documented every 60 s, and we noted a drop in current during MES operation, which possibly indicated exhaustion of microbial feed with time [26]. Hence, nutrient availability (supplementary data) was ensured by using fresh growth media at the start of each MES cycle. The system initiation gradually stabilized current density performance as current output increased for STL and SCG systems in subsequent MES cycles. The average current density was evaluated from 24 h of each cycle, once current became increasingly stabilized. Comparison of the average current densities revealed an increase in current output (mA/m2) from –40.6 to –48.1 (SCG) and –44.9 to –128.1 (STL) from cycle 1 to cycle 3, respectively (Figures 2(a) and 2(c)). Several studies have reported increased current output with time [27, 28]. In our study, STL showed higher current output (–128 mA) compared to SCG (–48 mA), indicating its improved bioelectrochemical performance. At the end of the third MES cycle, STL exhibited 2.4- and 2.7-fold increased current output compared to control and SCG, respectively, which was consistent with the results discussed above (Figure 2(c)). Recent studies have also reported improved current performance for STL and reported mild antimicrobial properties of SCG that could affect microbial metabolism [19, 21, 29].

(a)

(b)

(c)
3.2. Cyclic Voltammetry and Microbial Electrosynthesis of VFAs
The bioelectrochemical performance was further evaluated by CV at the end of the experiment. The electron uptake was monitored by conducting the CV analysis under multiple scan rates (1–200 mV s−1) for control (Figure 3(a)), STL (Figure 3(b)), and SCG (Figure 3(c)). A considerable increase in current output (mA) was observed from −2.2 to −3.2 and−4.3 for control, SGC, and STL electrode system, respectively (Figure 3(d)). The 1.5- and 2.0-fold increase in the current output using SGC and STL electrode system, respectively (compared to control), indicated improved electron transfer. To further confirm the effect of STL and SGC on MES performance, VFA production was evaluated every 24 h using HPLC. Butyrate was observed for all MES systems (Figure 4(a)), while minimal concentration of acetate and propionate were noted (Figure 4(b)). Furthermore, butyrate production (mg/L) increased from 294, 256, 308 to 376, 473, 533 for control, STL and SCG, from first to third cycle, respectively, and the increase was significantly higher for STL (Figure 4(a)). This improved performance could be attributed to the enhanced electron utilization, which is key for improving VFA [30]. These results are consistent with the results showing enhanced current and CV output for STL.

(a)

(b)

(c)

(d)

(a)

(b)
3.3. First Derivative and EIS Analysis
The electrochemical activities of STL and SCG and the subsequent microbial electron uptake were examined by performing CV first-derivative analysis [25]. The control exhibited limited redox peaks, while the sizes/intensities of the STL and SCG peaks were visibly dominant indicating their improved electrochemical activity [31]. The STL and SCG biocathode exhibited a number of redox peaks (Figure 5(a)) such as +79 mV, −172 mV, −421 mV, −673 mV, and −800 mV (STL) and +145 mV, −207 mV, −287 mV, −446 mV −612 mV, and −713 mV (SCG), respectively. The presence of these peaks signifies the presence of different mediators assisting the electron transfer in STL and SCG [32, 33]. The redox peak of −185 mV (close to −172 mV) and−201 mV (close to −207 mV) represented by STL and SCG, respectively, could correspond to OmcA [34] species, whereas flavoproteins are likely related to the −436 mV redox potential (SCG) [35]. The potential values of −261 mV, −459 mV, and −800 mV that are close to those of our MES system have also been reported in literature with minor differences which could be attributed to system configuration, experimental conditions, microbial source, and the existence of free or biofilm-bound mediators [36–38]. Since we used a mixed microbial culture, the observed peaks could not be ascribed to any specific bacteria. Furthermore, the electrocatalytic activity was evaluated using EIS analysis in a typical three-electrode system (Figure 5(b)). The charge transfer resistance (Rct) illustrated the charge transport efficiency and electrode/electrolyte interfacial phenomenon [39]. The circuit component parameters were evaluated across a large frequency range using an equivalent circuit of R(QR)(QR), and the equivalent circuit model fitting was performed using ZSimpWin software (AMETEK, USA) for plotting the Nyquist curve. Nyquist curves showed a lower charge transfer resistance for STL (335 Ω) and SCG (282 Ω) compared to control (1378 Ω). A lower Rct reportedly enhances the electronic delivery, thus improving the current density performance and electrocatalytic activity toward the charge transfer [40, 41]. The EIS results supported the current density and CV, further confirming that STL and SCG improved the bioelectrochemical system performance.

(a)

(b)
3.4. Microbial Community Analysis
The microbial community and biofilm generation were evaluated for control, STL, and SCG at the end of the experiment. The microbial samples were classified at the phylum, class, and genus levels based on the results of 16S rRNA gene pyrosequencing and assigned operational taxonomic units. At the phylum level, Firmicutes, Proteobacteria, and Bacteroidetes showed the highest relative abundance for control (95%), STL (79%), and SCG (75%); however, microbial shift and transformation were evident for STL and SCG in terms of Chloroflexi, Actinobacteria, Thermotogae, and Ignavibacteriae (Figure 6(a)). These phyla comprise excellent fermenters that show better VFA production in MES [42, 43]. This was further confirmed in our study where Clostridia (54% and 43%), Bacteroidia (14% and 8%), and Deltaproteobacteria (6% and 4%), shared the highest relative abundance for STL and SCG biocathode, respectively (Figure 6(b)). The Clostridia, Bacteroidia, and Deltaproteobacteria are exoelectrogenic and fermentative bacteria which may be linked to the improved current output and VFA [44–46]. Genus level analysis indicated enrichment of Acidaminobacter, Lentimicrobium, Mesotoga, and Ignavibacterium, which can improve VFA production (Figure 6(c)) [47–49]. Moreover, the collective abundance of these microbes enriched on STL and SCG may have further favored MES performance as evident from improved current density and VFA results of STL and SCG.

(a)

(b)

(c)
4. Conclusion
The SCG and STL biocathodes exhibited improved bioelectrochemical performance and current density with 1.5- and 2.0-fold increase in CV current compared to that in control. Both MESs exhibited lower electrode resistance and redox peaks of significantly high intensity suggesting improved electrocatalytic activity. Butyrate production is improved by 1.3- and 1.4-folds for SCG and STL, respectively, which could be associated with the enrichment of different fermentative and exoelectrogenic class of bacteria such as Clostridia, Bacteroidia, and Deltaproteobacteria. These results highlight the feasibility of using STL and SCG cathodes for improving MES performance and VFA production.
Data Availability
Data is available on request.
Conflicts of Interest
The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper.
Acknowledgments
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF2018R1A6A1A03024962) and the Ministry of Science and ICT (NRF-2020R1A2C2100746).
Supplementary Materials
Microbial culture/sludge: information regarding the initial inoculum source and pretreatment. Media and culture conditions: composition of growth media and trace element solution along with microbial enrichment conditions. Microbial community analysis: detailed description of microbial community analysis process. Table S1: primer and template details for microbial community analysis. (Supplementary Materials)