Investigating substrate impact on electroactive biofilm performance in low-cost, single-chamber microbial electrolysis cells for biosensing†

Connor E. Sauceda and Adam L. Smith*
Astani Department of Civil and Environmental Engineering, University of Southern California, 920 Downey Way, Los Angeles, CA 90089, USA. E-mail: smithada@usc.edu

First published on 3rd July 2025

---

---

Water impact Real-time water quality monitoring is imperative to advance the sustainability of both engineered and natural water resources. Low-cost and structurally simple MECs have high potential to serve as cost-effective and easy to operate early warning sensors for organic pollution in various settings and have proven potential to predict concentrations of organic constituents in representative wastewater matrices.

1. Introduction

There is an urgent need for real-time, online monitoring of engineered wastewater treatment systems and the natural water resources they protect.^1–3 Due to the complex composition of water matrices, aggregate measurements of organic constituents, such as five-day biochemical oxygen demand (BOD₅), chemical oxygen demand (COD), and total organic carbon (TOC), are imperative to water quality monitoring and regulatory compliance.⁴ However, aggregate organic measurements can be time-consuming, costly, and unsustainable due to generation of hazardous waste.⁵ BOD₅, for example, requires a labor- and time-intensive workflow across 5 days, is prone to error, and only provides retrospective analysis.^6,7 On the other hand, because of its simplicity and consistency, COD is often used as a proxy for BOD, however analysis still requires at least 2 hours and uses hazardous chemicals such as potassium dichromate, sulfuric acid, and mercury.⁸ Additionally, COD tests can be prohibitively expensive at scale. In this study, for example, the cost of COD testing was approximately $3 USD per sample without accounting for equipment. Similarly, the cost of each acetate measurement in this study was approximately $2 USD without accounting for equipment. Simplifying and reducing costs of quantification methods remains imperative for effective water quality monitoring. Bioelectrochemical system (BES) based biosensors have garnered significant interest for their potential to provide substantially faster approximations of these parameters while remaining easy to operate, adaptable, and cost-effective.^9–11 Although real-time monitoring of BOD₅ in wastewater treatment plant effluents is an obvious application, biosensors could also be employed to monitor anaerobic digester stability via real-time volatile fatty acid (VFA) monitoring, or in advanced water purification facilities for potable reuse.¹² Additionally, real-time environmental water quality monitoring could provide crucial data to help enforce environmental regulations and monitor sensitive aquatic ecosystems.^10,11

Bioelectrochemical systems are a broad and diverse set of biotechnologies that incorporate microorganisms into electrochemical reactors.¹³ BESs vary extensively, including the size, shape, and material of electrodes, system configuration, membranes, and the electrochemically active microorganisms.^13,14 While many BES configurations have been investigated for biosensing applications, they can be broadly categorized as microbial fuel cells (MFCs) or microbial electrolysis cells (MECs).¹¹ In MFCs, both the electric current and potential are generated by the microbial oxidation of reduced compounds with no external power supply, while the electric potential of an MEC is externally controlled by a potentiostat, providing operational flexibility of the system. MFCs have received significant interest in recent years for their potential for energy recovery during wastewater treatment and subsequently have received the most attention for biosensing.^9,10 However, while MFCs are generally the simplest BES to operate, they offer limited information and operational flexibility, which impedes their application outside of power generation.¹⁴ MFCs notoriously suffer from high internal resistance and low voltage output, limiting the current flow and thus their capacity as biosensors.^11,15,16 Applying an external voltage with a potentiostat fixes the potential of the bioanode, thereby accelerating microbial oxidation, lowering resistance, and allowing measurement of biogenic current as the single signal in MECs.^11,17,18 Furthermore, single-chamber MEC (SCMEC) designs reduce internal resistance and increase current density by eliminating the chamber-separating membrane, thus extending the biosensing capacity of the system.^11,19–21 At the same time, the reduced complexity and simplified structure of a SCMEC reduces capital and operating cost, further facilitating practical application of the biosensor.^19–22 Additionally, potentiostats that provide external voltage control and amperometric monitoring are no longer cost prohibitive as several studies have documented reproducible fabrication of potentiostats for as little as $10 USD.²³ Therefore, SCMECs appear to have the highest potential for use as biosensors, yet the technology remains significantly understudied.¹⁹

Many studies have investigated biosensing capacity of various BES design and operational configurations, however several problems are left unaddressed. First, clean, buffered matrices with simple substrates, such as phosphate buffered saline (PBS) with acetate as the sole carbon and electron source, are most often used to assess biosensing performance.^{18,22,24–35} By providing high buffering capacity, however, PBS is not representative of real wastewaters and therefore provides limited utility in advancing BES technologies towards practical biosensing applications.^36,37 Second, some studies claim to have demonstrated a BOD biosensor, but in reality have used acetate or similar, low molecular weight and readily biodegradable organic compounds, which is further unrepresentative of real wastewaters.^{22,28,29,32,34,35,38} Little research has used complex, fermentable substrates in representative matrices to investigate biosensing performance, especially in MECs. Finally, many studies have achieved a high range of detection but do not provide real-time monitoring, instead operating only in batch mode or taking discrete, coulometric measurements, each of which can take up to hours or days.^{18,22,26–29,31,34,35,39} Additionally, while few studies report the cost of their designs, many include expensive components such as commercial Ag/AgCl reference electrodes, working electrodes containing platinum or carbon nanotubes, or ion exchange membranes for two-chamber systems.^{24–28,32–35,39} Therefore, research assessing the real-time, amperometric biosensing capacity of continuous-flow MECs in complex, representative matrices while maintaining low-cost designs is certainly needed to advance BES biosensors towards practical applications.

This study seeks to investigate the performance of low-cost SCMECs in a continuous system using unbuffered synthetic wastewater for real-time water quality monitoring. This is the first study to our knowledge that directly compares the impact of complex and simple substrates on the performance and microbial activity of electroactive biofilms (EABs) in MECs under both batch and continuous-flow operation in representative wastewater conditions. We compared the exoelectrogenic current and microbial ecology of wastewater-derived, anaerobic, mixed-culture MECs under increasing substrate conditions, and additionally investigated the effect of drastic shifts in substrate composition. Acetate and COD concentrations were compared with MEC current production to assess MEC biosensing capacities by quantifying their relationship and range of detection. Further, DNA- and RNA-based 16S rRNA sequencing was used to investigate microbial community structure and activity of the MEC EABs to further assess differences in performance due to substrate conditions.

2. Materials and methods

2.1. MEC setup

Eight single-chamber MECs were constructed using 473 mL Ball® glass jars (Ball Corporation, Westminster, Colorado) fit with silicone gasket sealant rings and polypropylene caps (Fig. 1). Holes were drilled in each cap to accommodate three working electrodes, one counter electrode, one Ag/AgCl reference electrode (3 M KCl), an influent port, and an effluent port. All working and counter electrodes were graphite rods approximately 0.64 cm in diameter with a working surface area of 16.6 cm² (McMaster-Carr, Elmhurst, Illinois). The three working electrodes in each MEC were electrically connected via 24 AWG stranded wired and conductive silver epoxy. Each electrode and port were secured with epoxy adhesive and subsequently sealed with three coats of waterproof liquid rubber sealant to ensure an airtight seal and maintain anaerobic conditions. The MEC design in this experiment was chosen due to its proven efficacy, simple design, ease of operation, and associated low cost.^24,40 The cost estimate of the materials required to make one of the MECs in this study is <$35 USD (Tables S1 and S2†). This is approximately the cost of seven days of substrate analysis (COD & acetate) per MEC, indicating the MEC could pay for itself within a week. This cost can also further be reduced by minimizing the cell size and using higher surface area electrodes for biofilm growth.

2.2. MEC operation

Each MEC was maintained in a 37 °C water bath and continuously mixed using a magnetic stir bar and stir plate running at approximately 100 rpm (Multistirrer 15, Velp Scientific, Deer Park, New York). Upon startup, each MEC was filled with 390 mL of synthetic wastewater and monitored for 1 h using chronoamperometry to establish an abiotic electrochemical baseline, which was found to be <1 μA cm⁻² in each MEC. The synthetic wastewater was modeled after SYNTHES medium-strength domestic wastewater, and a list of base ingredients can be found in Table S3.†⁴¹ Resazurin sodium salt was also added to a final concentration of 1 mg L⁻¹ to visually confirm that anaerobic conditions were maintained in the MECs. Resazurin is electrochemically active, however the abiotic electrochemical baseline of the media in the MECs was negligible compared to the biotic electrochemical current production and was not found to interfere with the signal. After the abiotic baseline was established, each MEC was inoculated with 10 mL of municipal wastewater sludge collected from York River Treatment Plant (Seaford, Virginia) and were subsequently run in batch-fed mode with a working volume of 400 mL. After two initial batches running for 5 d each (10 d total), the MECs were fed 400 mL of fresh, nitrogen-sparged synthetic wastewater every 3 d.

To compare the effect of organic carbon substrate on biofilm growth and MEC performance, four MEC replicates included 820 mg L⁻¹ (10 mM) sodium acetate as the sole electron donor (AC-MECs), while the other four MEC replicates were supplemented with 41 mg L⁻¹ sodium acetate, 290 mg L⁻¹ of yeast extract, and 270 mg L⁻¹ of solubilized potato starch (WW-MECs). Each MEC was provided approximately 630 mg L⁻¹ of soluble COD according to empirical COD measurements of each substrate (Table S5†). All eight MECs were operated in batch-fed mode (Batch) for 37 d. After 37 d, MEC operation was switched to continuous flow (CSTR) mode using a peristaltic pump (BT100-1L, Langer Instruments, Tucson, Arizona) maintaining a hydraulic retention time (HRT) of 24 h. During continuous flow mode, the influent was stored in a 5 L reservoir at 4 °C and fresh influent was prepared every 2 d. The MECs remained under continuous flow mode for 76 d. For the first 28 d of continuous flow operation the MECs were fed with the same synthetic wastewater used in batch-fed mode. The influent strength for all MECs was subsequently stepwise increased by 50% on day 28, and again by 100% on day 42. On day 56 the influent substrate composition was switched for three replicates of each MEC type, leaving one replicate as a control. MECs were operated under this condition for 21 d until the end of the experiment. A detailed outline of operational conditions can be found in Table 1.

---

All electrochemical measurements were performed using a 12-channel multipotentiostat (MultiEmStat4, BASi, West Lafayette, Indiana). Chronoamperometry was used to continuously monitor the bioelectrogenic current production from the set of working electrodes in each MEC. During chronoamperometry, the working electrodes were constantly polarized at +300 mV (vs. Ag/AgCl) and current measurements were taken every 60 s for each MEC.

2.3. Substrate analysis

MEC performance was monitored by evaluating influent and MEC soluble COD and acetate concentrations. During batch-fed mode, 10 mL of influent and effluent were sampled at the end of each batch, except during the last three batches where samples were collected every 12 h for batch 9 & 10, and every 6 h for batch 11. During continuous flow mode, 10 mL samples were taken approximately every 24 h, 5–6 times per week. All samples were immediately filtered with a 0.2 μm nylon membrane filter (Tisch Scientific, Cleves, Ohio) and subsequently stored at −20 °C until further processing. Two mL of each sample was used for soluble COD measurements via digestion for two hours at 150 °C in high-range COD digestion vials using a colorimetric dichromate method and then read on a visible spectrophotometer (HI801-01, Hanna Instruments, Smithfield, Rhode Island). Acetate concentrations were determined using an ICS-2100 ion chromatography system with separation achieved via a 2 × 250 mm AS11-HC column (Dionex, Sunnyvale, California). A gradient separation method was used starting with 1 mM KOH linearly increasing to 6 mM after a run duration of 7 min with a flowrate of 0.38 mL min⁻¹.

2.4. Biofilm microbial community analysis

Biofilm samples were taken at day 10 (10 days) and 37 (batch) of batch-fed mode, and day 55 (CSTR) and 76 (substrate switch; SS) under continuous flow mode to understand the microbial community structure and activity of the EABs throughout the experiment. All samples were stabilized using DNA/RNA Shield (Zymo Research, Tustin, CA) and preserved at −20 °C until further processing. DNA and RNA were extracted, with RNA further treated with Ambion™ DNA-free DNA Removal Kit (Invitrogen, Carlsbad, CA) and subsequently reverse transcribed to cDNA using iScript™ cDNA Synthesis Kit, according to the manufacturer's recommendations (Bio-Rad, Hercules, CA). Extracted DNA and cDNA were subsequently sent to the University of Michigan's Microbiome Core for Illumina 16S rRNA gene sequencing using primers targeting the V4 region.⁴² Resulting FASTQ files were processed using mothur software and sequence reads were identified by referencing the SILVA database v138.2.^43,44

2.5. Statistical analysis

All statistical analysis was done in R Statistical software (v4.3.2). Simple linear regression was performed to analyze the relationship between acetate or COD concentration and the current signal produced by the MECs.⁴⁵ Using this analysis, best fit lines, along with their coefficients of determination, and associated p-values were determined to analyze the predictive power of electrical current in determining acetate and COD concentrations. Student's t-tests and non-metric multidimensional scaling were performed to assess significance in differences of microbial communities between the MEC types and over the duration of the experiment.

3. Results and discussion

3.1. Batch operation revealed a strong relationship between MEC current production and acetate concentration

Performance of the AC-MECs and WW-MECs were similar during batch 1 but began to diverge during batch 2, with AC-MECs producing substantially higher current on average than the WW-MECs (up to 220 μA cm⁻² vs. 145 μA cm⁻²; Fig. 2A). Throughout each batch, AC-MECs produced current that rapidly increased within the first few hours, then increased linearly over two days before reaching a turning point where current declined exponentially until the next batch feed (Fig. 2D). The WW-MECs exhibited a significantly different current production pattern, remaining steady for around 36 h before reaching a more gradual turning point and declined slowly compared to the AC-MECs (Fig. 2D). The two MECs also stabilized at different rates; AC-MEC current production was consistent after batch 4 (day 16), whereas WW-MEC current production declined slowly from batch 3 to 8, only stabilizing after day 27. These distinct patterns indicate that substrate impacted the EAB differently. Similarly, Tardy et al. (2020) observed different voltage evolution patterns when feeding MFCs batches of domestic wastewater (particulate and slowly biodegradable) versus brewery wastewater (soluble and highly biodegradable).³⁹ The brewery wastewater peaked at higher voltages but declined faster than the domestic wastewater, which resulted in a lower, steady voltage that took longer to decline towards zero.³⁹ Additionally, when filtered, the domestic wastewater produced voltage evolution more similar to that of the brewery wastewater, further supporting the notion that slowly biodegradable substrates result in longer sustained exoelectrogenesis.³⁹

COD consumption differed between the AC-MECs and WW-MECs, averaging 92 ± 5% versus 78 ± 5%, respectively. However, WW-MECs exhibited consistent COD consumption starting with batch 2 (after 10 d), whereas AC-MEC COD consumption only stabilized after batch 4 (16 d). Coulombic efficiency was substantially higher on average in the AC-MECs (86 ± 10%) than the WW-MECs (53 ± 10%) (Fig. S3†). The model exoelectrogen Geobacter spp. preferentially metabolizes acetate as an electron source, and the relative activity of Geobacter spp. was substantially higher in the AC-MEC biofilms (91 ± 4%) than the WW-MEC biofilms (25 ± 16%) (p < 0.01) (Fig. 3B), corroborating differences in Coulombic efficiency.⁴⁶ The complex organic substrate supported a more diverse microbial community, where fermentative microbial populations scavenged complex organics for electrons and produced acetate as a waste product, some of which was metabolized by Geobacter spp. and produced current as a result. This hypothesis is supported by the greater diversity of the WW-MEC biofilm (inverse Simpson diversity of 6.36 ± 1.98 for WW-MECs compared to 1.22 ± 0.08 for AC-MECs, p < 0.05) and the high relative activity of fermentative bacteria like Clostridiaceae spp., Acetobacteroides spp., and Bacteroides spp. (Fig. 3B).

The high-resolution sampling series employed in the last three batch feedings (batch 9, 10, and 11) gave further insight into the substrate evolution and EAB metabolism. The COD consumption over the course of each batch was comparable between both MECs, however COD declined at a slower rate towards the end of each batch in the WW-MECs due to slower biodegradability of the substrate (Fig. 2D). Although WW-MECs were only fed with approximately 29 mg L⁻¹ of acetate, the complex biopolymers in the potato starch and yeast extract were fermented to VFAs, including acetate, as seen by the elevated acetate concentrations in the WW-MECs (approximately 100–120 mg L⁻¹). Notably, the acetate concentration evolution in the WW-MECs closely resembled the current production, with a steady sustained concentration for approximately 36 h followed by a gradual decline (Fig. 2D). This suggests that the MEC current production is closely tied to acetate concentration rather than aggregate organic carbon measurements such as COD.

Although Geobacter spp. can grow on a variety of fermentation end products, acetate supports optimal biofilm growth and current production.^46,47 Kretzschmar et al. (2017) only observed current production when MECs were fed acetate, or a mixture of VFAs including acetate (propionate or butyrate alone only produced baseline current).⁴⁸ Higher relative abundance of Geobacter spp. was also observed in biofilms grown solely on acetate relative to mixtures of VFAs.⁴⁸ Similarly, Atci et al. (2016) observed that Geobacter sulfurreducens biofilms produced less than 20% of the current response using lactate, pyruvate, or hydrogen as compared to acetate.⁴⁹ Formate was the only alternative electron donor that produced a more substantial current response (39% compared to acetate).⁴⁹ In the present study, the lower relative activity of Geobacter spp. in the WW-MEC biofilms likely resulted in a weaker relationship between current production and acetate concentration than the AC-MECs during batch-fed mode (r² = 0.68 vs. 0.93, Table 2). Other active bacteria likely scavenged a larger proportion of acetate that was then unavailable to Geobacter spp.³¹

---

3.2. AC-MEC current declined while WW-MEC current remained stable despite increased influent strength during continuous operation

After switching the operational mode of the MECs from batch to continuous, the MECs produced a consistently elevated current, but at lower magnitudes compared to the peaks during batch mode. AC-MECs initially averaged between 125–165 μA cm⁻² of current in the first two weeks of CSTR operation, although exhibited a downward trend on average (Fig. 4A). A short upset in CSTR operation occurred on day 14 when the influent pump was unintentionally powered off overnight for around 10 h. Once the influent pump resumed operation, the AC-MEC current production recovered. However, AC-MEC current production continued to decline, and in the following two weeks eventually dropped to below 33 μA cm⁻² on average. Over the same period, the WW-MECs took approximately two weeks to reach a stable current production of approximately 70 μA cm⁻² (Fig. 4A). The WW-MEC current in the initial two weeks showed cyclical variability on a 48 h interval, coinciding with the refreshment of the influent media, which was made every two days. These bumps, apparent in the WW-MEC current production, were found to be a result of the rapid onset of biodegradation in the influent reservoir, even while stored at 4 °C. To combat premature biodegradation, starting on day 10 of CSTR operation, the tubing immersed in the influent reservoirs was washed with bleach and replaced every 48 h on the same interval as the media replacement. After this operational change was made, the WW current production stabilized and converged among replicates for the rest of the low-strength influent operation.

When the influent strength was stepwise increased by 50% to medium-strength influent on day 27, both MECs exhibited immediate increases in current production (Fig. 4A). AC-MECs showed a drastic spike in current production from an average of 33 μA cm⁻² to an average peak of 120 μA cm⁻² within 24 h. However, as was the case under low-strength influent, the current production for the AC-MECs started to decline rapidly again within a few days. Over the course of the two weeks of medium-strength influent operation, the average AC-MEC current signal declined and began to plateau below 20 μA cm⁻², even lower than was reached with low-strength influent. Upon the next stepwise increase to high-strength influent, AC-MEC current only spiked from 19 μA cm⁻² to 60 μA cm⁻² within 24 h, a much smaller change and total magnitude than the first stepwise increase, despite a 100% increase in concentration. Once again, following this peak, AC-MEC current decreased, but this time much more gradually and ending at around 27 μA cm⁻² on average after two weeks. Low current production after high-strength acetate operation also coincided with a drastic decrease in Geobacter spp. relative activity from 71 ± 12% to 9 ± 4% (p < 0.01). Comparatively, the WW-MECs exhibited relatively stable current across stepwise increases in influent strength. Stable current production despite substantial increases in influent strength suggests an exoelectrogenic saturation reached by the WW-MEC EABs, likely due to low acetate availability and the limited relative activity of Geobacter spp. (Fig. 3C).

The decreasing trend observed in AC-MEC current production corresponded to MEC acetate concentrations, rather than influent strength (Fig. 4B). Therefore, substrate concentration and current production followed typical Monod growth kinetics, which has been observed for Geobacter sulfurreducens.^50,51 This suggests that HRT may be used as an operational parameter to adjust substrate concentration in the MEC consistent with the linear range of detection. WW-MEC current production similarly aligned with acetate concentrations, except immediately following stepwise increases in influent strength (Fig. 4C). The high-strength influent, however, resulted in elevated acetate concentrations for the AC-MEC (200–250 mg L⁻¹). The WW-MEC COD increased up to 390 mg L⁻¹ and did not drop below 200 mg L⁻¹. The concentration profiles observed suggest that the metabolic capacities of the MECs were exceeded. Therefore, substrate consumption, and thus biosensing performance, were limited at such high influent strength. Future research is needed to explore design or operational modifications that could enable applications of MEC biosensing in this concentration range.²⁴

3.3. Adaptation of mature EABs is slow when subject to substantial change in substrate composition

At day 55 of CSTR operation, the substrates were switched: AC-MECs to high-strength WW, and WW-MECs to high-strength AC. One of the four replicates for each AC- and WW-MEC was unchanged and acted as a control. The control AC-MEC followed its previous trend as its current production continued to decline over the following three weeks (Fig. 5A). The altered AC-MEC current production was even lower than that of the control, likely due to the sudden, drastic decrease in acetate concentration (Fig. 5B). The current production did appear to increase slightly around day 69, but only to a maximum of around 44 μA cm⁻² (Fig. 5A). While the control WW-MEC also continued its trend and maintained a baseline current production around 68 μA cm⁻², the altered WW-MEC current production initially increased, but then declined gradually, consistent with the AC-MEC trend when fed acetate (Fig. 5A). The WW-MECs additionally exhibited greater variability in current with a standard deviation of 33 μA cm⁻².

The AC-MEC COD was relatively stable after the substrate switch (276 ± 54 mg COD L⁻¹ to 237 ± 65 mg COD L⁻¹; Fig. 5B). However, WW-MEC COD increased from 279 ± 54 mg COD L⁻¹ (Fig. 4C) to 536 ± 48 mg COD L⁻¹ (Fig. 5C) after the substrate switch. These results suggest that acetate selects for an EAB community that adapts well to different substrates, but EABs grown on complex substrate do not adapt well, contrary to a previous report that glucose-acclimated MFCs with a diverse EAB enabled wider substrate utilization as compared to acetate-acclimated MFCs.³¹

The substrate switch changed the selective pressure on the AC-MEC microbial community as seen by the increased inverse Simpson diversity from 1.94 ± 0.20 to 6.96 ± 2.44 (p < 0.05) (Fig. 3C). Although the relative Geobacter spp. activity increased in the AC-MECs during this period (9 ± 4% to 18 ± 9%, p = 0.27), it did not recover to previous levels (Fig. 3B), and current production did not increase (Fig. 5A). Inversely, after the substrate switch, the relative Geobacter spp. activity of the WW-MEC biofilm increased from 39 ± 10% to 62 ± 15% (p < 0.05) (Fig. 3B). The WW-MEC inverse Simpson diversity also decreased substantially from 6.73 ± 1.24 to 2.64 ± 0.96 (p < 0.05) (Fig. 3C). Non-metric multidimensional scaling (NMDS) revealed that the microbial communities of each MEC converged closer to one another after the substrate switch (Fig. 3D). These results indicate that, regardless of the substrate originally provided, acetate strongly selects for high Geobacter spp. activity, and complex substrates support a more diverse community.^31,32

For biosensing applications, the substrate switch suggested that EABs would perform best when grown on substrate that resembles the intended application, as changes in these conditions will shift the microbial community activity profile. Acetate-grown EABs, however, appear to adapt well to changes in substrate, even at high loading rates. It is important to note that both MECs appeared overloaded under high-strength influent and may have responded differently at low-strength influent where the observed relationship between current production and acetate concentration is stronger (Fig. 6B). Further work at longer operational periods should be done to assess the adaptability of MEC EABs to changes in substrate during continuous flow operation.

3.4. Both batch and continuous flow operation provide insight into MEC biosensing capacity

The high-resolution sampling series taken during batches 9–11 (Fig. 2D) shows two distinct ranges of response to acetate concentration (Fig. 6A). In both the AC- and WW-MECs, a positive linear relationship was observed between acetate concentration and MEC current, up to a limit where the signal plateaus. This pattern corroborates that the MEC current production increases linearly with increasing acetate concentration until the EAB metabolic saturation point is reached.^{18,24,49,52,53} Past the point of saturation, it appears that higher acetate concentrations may have an inhibitory effect on current density, however this is likely a product of latency in biofilm reactivation after substrate depletion of the previous batch. While a similar pattern is exhibited by both the AC-MECs and WW-MECs, the saturation point, and thus the linear range of acetate detection, was much lower for the WW-MECs at around 40 mg Ac⁻ L⁻¹, in comparison to the AC-MECs limitation at around 100 mg Ac⁻ L⁻¹, both with strong correlation (r² = 0.80 and 0.93, respectively; Table 2). During batch mode, peak current production followed by a steep drop is shown to occur within a consistent acetate concentration, especially for AC-MECs (Fig. 2D). During batch 9, 10, and 11, current production sharply declined at acetate concentrations between 151–218 mg L⁻¹. The lower end of this range coincides with the limit at which current production plateaus with respect to acetate concentration in batch mode (Fig. 6A), suggesting that the EAB acetate saturation concentration could be determined during batch mode where current production drops off.

As the MECs followed typical Monod growth kinetics, during continuous flow it is expected that current production aligns with MEC acetate concentration, regardless of influent strength. Continuous operation supported this hypothesis, as the AC-MECs gave a similar linear regression equation and equivalent fit to batch mode under the same influent strength (r² = 0.93; Table 2). Subsequent stepwise increases in influent strength decreased the slope of the linear regression equations, indicating that the EABs were less sensitive to acetate concentration (Fig. 6B). With r² = 0.84 and 0.68 for medium and high-strength influent, respectively, AC-MECs exhibited lower biosensing performance at higher influent concentrations of acetate. Decreased slopes and fit of the linear regression equations could indicate that drastic changes in acetate concentrations could result in biosensing latency and require recalibration of the biosensor. It has been previously reported that anodic EABs do not have a single response curve to acetate concentrations, but rather show shifts in their concentration-dependent behavior as they adapt to changes in acetate concentrations.⁵⁴ While the linear regression fit improved for WW-MECs with increasing influent strength (r² = 0.72 and 0.93 for medium and high-strength influent, respectively), the small range of acetate and current measurements and the similarity among the best fit lines suggested limited ability to assess the WW-MEC's biosensing capacity within the operational mode conducted in this study (Fig. S10†).

The fit of the continuous mode linear regression may be explained when comparing the acetate concentrations measured during continuous mode to the acetate saturation range observed in batch mode (151–218 mg L⁻¹). When fed low-strength acetate during continuous mode, the MEC acetate concentrations (25–125 mg L⁻¹) were consistently below the batch mode acetate saturation range, and the linear relationship between acetate and current was strong and comparable to the linear range found in batch mode (r² = 0.93; Table 2). When fed medium- and high-strength acetate, the MEC acetate concentrations progressively extended beyond this saturation range (60–185 mg L⁻¹, and 198–250 mg L⁻¹, respectively). The linear relationship between acetate and current also progressively weakened (r² = 0.84 and 0.68, respectively). Decreasing linear fit associated with acetate concentrations increasing past the batch mode saturation range further suggests that batch mode operation may be useful in assessing the substrate saturation concentration of an MEC, but more investigation with higher resolution data is needed. Overall, strong linear relationships are observed between current production and low acetate concentrations. These MECs may still serve as an early warning system for higher acetate concentrations during system disruptions; however, the linear range of detection with strong fit must be extended if an MEC of this type is to be useful for higher strength quantification applications (e.g., monitoring anaerobic digesters).⁵⁵ Other studies commonly report regression fits of r² > 0.9 in coulometric and amperometric BESs suggesting this to be an acceptable threshold, however other statistical parameters such as standard error or 95% confidence intervals are important in assessing confidence in quantification.^{25–28,30,32,34,39} Further work must be done to optimize the MEC design, such as optimizing HRTs and the ratio of electrode surface area to MEC volume to increase the MEC metabolic range.²⁴

3.5. EAB communities are heavily influenced by substrate

At days 10, 37, 92, and 113, biofilm samples were taken from each MEC for microbial community analysis via 16S rRNA sequencing. These samples correspond to the initial inoculated community, the end of batch-fed mode, the end of normal continuous mode, and the end of the experiment after the substrate switch, respectively. DNA and RNA-sequencing was used to determine relative abundance and relative activity of the EAB. We emphasize relative activity data more heavily here as a more accurate measure of active microbial populations given that microbial growth rate and yield in such systems is low.^56,57 Upon initial inoculation (day 10), AC-MECs were heavily dominated by Geobacter spp., with a relative abundance and relative activity of 67 ± 6% and 91 ± 4% on average, respectively. Geobacter spp. also represented a significant portion of the WW-MEC communities upon inoculation, but at significantly lower relative abundance and relative activity, at 21 ± 5% and 25 ± 16% on average, respectively (p < 0.01).

Overall, it is clear that the AC influent was highly selective towards Geobacter spp., with the exception of the high-strength AC influent (Fig. 3B). After continuous mode with the high-strength AC influent, the relative activity of Geobacter spp. declined to only 9 ± 4% from 71 ± 12% (p < 0.01) at the end of batch mode (Fig. 3B). Interestingly, Methanothrix spp. replaced Geobacter spp. as the dominant population with a relative activity of 72 ± 5%, previously having negligible activity in the AC-MECs (Fig. 3B). It has been shown that high acetate could reduce the relative abundance of Geobacter spp.⁵⁸ Influent acetate in this study reached up to 30 mM, potentially accounting for the drop in relative activity of Geobacter spp. to 9%. Methanothrix spp. are known to have a high affinity for acetate but low growth rates and are usually outcompeted by the only other acetoclastic methanogens, Methanosarcina spp., at high acetate concentrations.^59–61 In this study, however, Methanosarcina spp. were negligible in the inoculum, potentially allowing Methanothrix spp. to be the sole acetoclastic methanogen and thus dominate the EAB at high acetate concentrations when Geobacter spp. were inhibited. Although methane was not measured, the combination of high Methanothrix spp. activity and high acetate consumption (88%) suggests acetate consumption via acetoclastic methanogenesis. Methanothrix spp. are also known to receive electrons via direct interspecies electron transfer (DIET) from Geobacter spp.^62,63 Therefore, it is also possible that electrons produced by Geobacter spp. were diverted from the electrode and instead used by Methanothrix spp. to reduce carbon dioxide to methane. The transition from Geobacter spp. to Methanothrix spp. in the activity profile likely explains the low current production despite high MEC acetate concentrations during high-strength continuous mode (Fig. 4B). Further work involving headspace methane measurements is needed to understand the role of methanogens in MECs such that design and operation can be optimized to reduce Geobacter spp. competition for acetate and ensure electron flow to the electrode.

Geobacter spp. played an important role in the WW-MEC EAB communities as well, with an average relative activity between 23–39% (Fig. 3B), but the WW influent selected for a much more diverse community with an average inverse Simpson diversity between 6.36–9.87 (Fig. 3C). Major community members found in the WW-MEC EAB, such as Acetobacteroides spp., Macellibacteroides spp., and Dysgonomonas spp., are known fermentative organisms and were likely responsible for fermenting complex biopolymers into acetate, hydrogen, and carbon dioxide that would subsequently support Geobacter spp. and methanogens like Methanothrix spp. and Methanospirillum spp.^64–71

Although the AC-MEC and WW-MEC EABs were quite distinct from one another, when their substrate compositions were switched, their microbial communities tended to converge. WW-MECs fed high-strength AC showed an average Geobacter spp. relative activity of 62 ± 15% (Fig. 6B) and their inverse Simpson diversity decreased from 6.73 ± 1.24 to 2.64 ± 0.96 (p < 0.05) (Fig. 3C). Although the AC-MEC Geobacter spp. relative activity was already low after being fed high-strength AC, when switched to high-strength WW, the EAB inverse Simpson diversity increased from 1.94 ± 0.20 to 6.96 ± 2.44 (p < 0.05) (Fig. 3C). Additionally, NMDS showed decreased distance of the two MEC communities after the substrate switch, further showing convergence (Fig. 3D).

4. Conclusion

This study addressed the need for investigation into the performance of real-time, low-cost SCMEC biosensors in representative, continuous-flow wastewater conditions comparing simple and complex substrates. It was demonstrated that, while an MEC EAB community is not dependent solely on the initial growth substrate, EAB community activity, and thus electroactive performance, is heavily dependent on the substrate provided. The MEC group being fed complex substrate consistently had a more diverse EABs compared to the MEC group being fed solely acetate. While acetate-grown EABs were primarily dominated by Geobacter spp. and far less diverse than those grown on complex substrate, they adapted well to consume complex substrate, whereas the complex substrate-grown EABs did not adapt well to consume high acetate concentrations. It should be noted that the effect of the substrate on the biofilm community has implications for biosensor studies using acetate as the sole BOD source; acetate selects more highly for Geobacter spp. than complex substrate and thus should not be the sole carbon source. A mixture of representative substrates should instead be used when investigating BESs for total BOD quantification.

At every biofilm sampling timepoint, the MEC group producing higher current consistently had higher relative activity of Geobacter spp. regardless of substrate composition. This positive relationship between Geobacter spp. activity and current production might be relevant in assessing the fitness of an MEC EAB for acetate biosensing. While continuous mode operation should be prioritized to enable real-time monitoring, batch mode operation could be useful in further assessing the acetate biosensing capacity of MECs. Overall, the AC-MECs exhibited strong biosensing performance at low acetate concentrations, with a linear fit of r² = 0.93 up to 100 mg L⁻¹ acetate, whereas the WW-MECs only had a linear fit up to 40 mg L⁻¹ with a fit of r² = 0.80. Future studies should further investigate the optimization of reactor design and operation to increase the linear range of acetate detection while minimizing cost and thus make MECs suitable for real-life applications, such as anaerobic digestion monitoring.

Data availability

The authors confirm that the data that support the findings of this study are available within the article and its ESI.† Raw data that support the findings of this study are available from the corresponding author upon reasonable request.

Author contributions

CRediT: Connor Sauceda conceptualization, data curation, formal analysis, investigation, methodology, validation, visualization, writing – original draft, writing – review & editing; Adam L. Smith conceptualization, funding acquisition, project administration, supervision, validation, writing – review & editing.

Conflicts of interest

We do not have any conflicts of interest to declare.

Acknowledgements

Connor Sauceda was partially supported by a Provost Fellowship from the University of Southern California and a Graduate Research Fellowship from the National Science Foundation (Grant No. DGE-1842487). We would like to thank Dr. Haluk Beyenal and Dr. Jerome Babauta for their invaluable insight into bioelectrochemical systems.

References

1. I. A. Cruz, L. R. S. Andrade, R. N. Bharagava, A. K. Nadda, M. Bilal and R. T. Figueiredo, et al. An overview of process monitoring for anaerobic digestion, Biosyst. Eng., 2021, 207, 106–119 CrossRef CAS .
2. W. Zhang, N. B. Tooker and A. V. Mueller, Enabling wastewater treatment process automation: leveraging innovations in real-time sensing, data analysis, and online controls, Environ. Sci.:Water Res. Technol., 2020, 6(11), 2973–2992 RSC .
3. F. Ejeian, P. Etedali, H. A. Mansouri-Tehrani, A. Soozanipour, Z. X. Low and M. Asadnia, et al. Biosensors for wastewater monitoring: A review, Biosens. Bioelectron., 2018, 118, 66–79 CrossRef CAS .
4. W. Bourgeois, J. E. Burgess and R. M. Stuetz, On-line monitoring of wastewater quality: a review, J. Chem. Technol. Biotechnol., 2001, 76(4), 337–348 CrossRef CAS .
5. D. Dubber and N. F. Gray, Replacement of chemical oxygen demand (COD) with total organic carbon (TOC) for monitoring wastewater treatment performance to minimize disposal of toxic analytical waste, J. Environ. Sci. Health, Part A: Toxic/Hazard. Subst. Environ. Eng., 2010, 45(12), 1595–1600 CrossRef CAS .
6. S. Jouanneau, L. Recoules, M. J. Durand, A. Boukabache, V. Picot and Y. Primault, et al. Methods for assessing biochemical oxygen demand (BOD): A review, Water Res., 2014, 49(1), 62–82 CrossRef CAS PubMed .
7. V. A. Arlyapov, Y. V. Plekhanova, O. A. Kamanina, H. Nakamura and A. N. Reshetilov, Microbial Biosensors for Rapid Determination of Biochemical Oxygen Demand: Approaches, Tendencies and Development Prospects, Biosensors, 2022, 12(10), 842 CrossRef CAS PubMed .
8. USEPA, Method 410.4, Revision 2.0: The Determination of Chemical Oxygen Demand by Semi-Automated Colorimetry, 1993 Search PubMed.
9. Y. Cui, B. Lai and X. Tang, Microbial Fuel Cell-Based Biosensors, Biosensors, 2019, 9(3), 92 CrossRef CAS .
10. A. Adekunle, V. Raghavan and B. Tartakovsky, A comparison of microbial fuel cell and microbial electrolysis cell biosensors for real-time environmental monitoring, Bioelectrochemistry, 2019, 126, 105–112 CrossRef CAS PubMed .
11. Y. Zhou, S. Zheng and W. Qin, Electrochemical biochemical oxygen demand biosensors and their applications in aquatic environmental monitoring, Sens. Bio-Sens. Res., 2024, 44, 100642 CrossRef .
12. California State Water Resources Control Board, SBDDW-23-001 Direct Potable Reuse Regulations, Jun, 2024 Search PubMed.
13. T. Zheng, J. Li, Y. Ji, W. Zhang, Y. Fang and F. Xin, et al. Progress and Prospects of Bioelectrochemical Systems: Electron Transfer and Its Applications in the Microbial Metabolism, Front. Bioeng. Biotechnol., 2020, 8, 495918 Search PubMed .
14. J. T. Babauta and H. Beyenal, Introduction to Electrochemically Active Biofilms, in Biofilms in Bioelectrochemical Systems: From Laboratory Practice to Data Interpretation, 2015, pp. 1–36 Search PubMed .
15. Y. Fan, E. Sharbrough and H. Liu, Quantification of the internal resistance distribution of microbial fuel cells, Environ. Sci. Technol., 2008, 42(21), 8101–8107 CrossRef CAS PubMed .
16. N. J. Koffi and S. Okabe, High voltage generation from wastewater by microbial fuel cells equipped with a newly designed low voltage booster multiplier (LVBM), Sci. Rep., 2020, 10(1), 1–9 CrossRef .
17. T. Chen, H. Liu and J. Li, Research on minimizing the MFC internal resistance via a shared electrode MFC-MEC coupling system, Biochem. Eng. J., 2024, 203, 109195 CrossRef CAS .
18. X. Jin, X. Li, N. Zhao, I. Angelidaki and Y. Zhang, Bio-electrolytic sensor for rapid monitoring of volatile fatty acids in anaerobic digestion process, Water Res., 2017, 111, 74–80 CrossRef CAS .
19. T. Hua, S. Li, F. Li, Q. Zhou and B. S. Ondon, Microbial electrolysis cell as an emerging versatile technology: a review on its potential application, advance and challenge, J. Chem. Technol. Biotechnol., 2019, 94(6), 1697–1711 CrossRef CAS .
20. P. Clauwaert and W. Verstraete, Methanogenesis in membraneless microbial electrolysis cells, Appl. Microbiol. Biotechnol., 2009, 82(5), 829–836 CrossRef CAS PubMed .
21. Y. Koul, V. Devda, S. Varjani, W. Guo, H. H. Ngo and M. J. Taherzadeh, et al. Microbial electrolysis: a promising approach for treatment and resource recovery from industrial wastewater, Bioengineered, 2022, 13(4), 8115 CrossRef CAS .
22. Y. Wang, X. Liu, M. Wang, P. Zhang, Y. Zong and Q. Zhang, A single-chamber microbial fuel cell for rapid determination of biochemical oxygen demand using low-cost activated carbon as cathode catalyst, Environ. Technol., 2018, 39(24), 3228–3237 CrossRef CAS PubMed .
23. T. W. von Zuben, A. G. Salles and J. A. Bonacin, Low-cost open-source potentiostats: A comprehensive review of DIY solutions and fundamental concepts of electronics and its integration with electrochemistry, Electrochim. Acta, 2024, 498, 144619 CrossRef CAS .
24. J. Kretzschmar, L. F. M. Rosa, J. Zosel, M. Mertig, J. Liebetrau and F. Harnisch, A Microbial Biosensor Platform for Inline Quantification of Acetate in Anaerobic Digestion: Potential and Challenges, Chem. Eng. Technol., 2016, 39(4), 637–642 CrossRef CAS .
25. G. A. Huerta-Miranda, L. M. Rodríguez-Torres, A. L. Martínez-García, D. A. Mazón-Montijo, A. Hernández-Eligio and M. Miranda-Hernández, et al. Geobacter sulfurreducens electroactive biofilms on Fe2O3/FTO support-electrodes for developing a sodium acetate electrochemical biosensor, Biosens. Bioelectron.:X, 2023, 14, 100370 CAS .
26. M. Di Lorenzo, T. P. Curtis, I. M. Head and K. Scott, A single-chamber microbial fuel cell as a biosensor for wastewaters, Water Res., 2009, 43(13), 3145–3154 CrossRef CAS PubMed .
27. S. Wang, S. Tian, P. Zhang, J. Ye, X. Tao and F. Li, et al. Enhancement of biological oxygen demand detection with a microbial fuel cell using potassium permanganate as cathodic electron acceptor, J. Environ. Manage., 2019, 252, 109682 CrossRef CAS PubMed .
28. O. Modin and B. M. Wilén, A novel bioelectrochemical BOD sensor operating with voltage input, Water Res., 2012, 46(18), 6113–6120 CrossRef CAS .
29. Y. Yi, T. Zhao, B. Xie, Y. Zang and H. Liu, Dual detection of biochemical oxygen demand and nitrate in water based on bidirectional Shewanella loihica electron transfer, Bioresour. Technol., 2020, 309, 123402 CrossRef CAS PubMed .
30. Z. Wang, C. Liao, Z. Zhong, S. Liu, M. Li and X. Wang, Design, optimization and application of a highly sensitive microbial electrolytic cell-based BOD biosensor, Environ. Res., 2023, 216, 114533 CrossRef CAS PubMed .
31. K. J. Chae, M. J. Choi, J. W. Lee, K. Y. Kim and I. S. Kim, Effect of different substrates on the performance, bacterial diversity, and bacterial viability in microbial fuel cells, Bioresour. Technol., 2009, 100(14), 3518–3525 CrossRef CAS PubMed .
32. M. W. A. Spurr, E. H. Yu, K. Scott and I. M. Head, Extending the dynamic range of biochemical oxygen demand sensing with multi-stage microbial fuel cells, Environ. Sci.:Water Res. Technol., 2018, 4(12), 2029–2040 RSC .
33. Y. Jiang, P. Liang, P. Liu, B. Miao, Y. Bian and H. Zhang, et al. Enhancement of the sensitivity of a microbial fuel cell sensor by transient-state operation, Environ. Sci.:Water Res. Technol., 2017, 3(3), 472–479 RSC .
34. N. Xiao, R. Wu, J. J. Huang and P. R. Selvaganapathy, Development of a xurographically fabricated miniaturized low-cost, high-performance microbial fuel cell and its application for sensing biological oxygen demand, Sens. Actuators, B, 2020, 304, 127432 CrossRef CAS .
35. A. S. Commault, G. Lear, S. Bouvier, L. Feiler, J. Karacs and R. J. Weld, Geobacter-dominated biofilms used as amperometric BOD sensors, Biochem. Eng. J., 2016, 109, 88–95 CrossRef CAS .
36. Y. Ruiz, J. A. Baeza and A. Guisasola, Microbial electrolysis cell performance using non-buffered and low conductivity wastewaters, Chem. Eng. J., 2016, 289, 341–348 CrossRef CAS .
37. F. Guo, J. T. Babauta and H. Beyenal, The effect of additional salinity on performance of a phosphate buffer saline buffered three-electrode bioelectrochemical system inoculated with wastewater, Bioresour. Technol., 2021, 320, 124291 CrossRef CAS .
38. Z. Wang, C. Liao, Z. Zhong, S. Liu, M. Li and X. Wang, Design, optimization and application of a highly sensitive microbial electrolytic cell-based BOD biosensor, Environ. Res., 2023, 216, 114533 CrossRef CAS .
39. G. M. Tardy, B. Lóránt, M. Gyalai-Korpos, V. Bakos, D. Simpson and I. Goryanin, Microbial fuel cell biosensor for the determination of biochemical oxygen demand of wastewater samples containing readily and slowly biodegradable organics, Biotechnol. Lett., 2021, 43(2), 445–454 CrossRef CAS PubMed .
40. C. Gimkiewicz and F. Harnisch, Waste Water Derived Electroactive Microbial Biofilms: Growth, Maintenance, and Basic Characterization, J. Visualized Exp., 2013,(123), e55617 Search PubMed .
41. S. Aiyuk and W. Verstraete, Sedimentological evolution in an UASB treating SYNTHES, a new representative synthetic sewage, at low loading rates, Bioresour. Technol., 2004, 93(3), 269–278 CrossRef CAS PubMed .
42. J. G. Caporaso, C. L. Lauber, W. A. Walters, D. Berg-Lyons, C. A. Lozupone and P. J. Turnbaugh, et al. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample, Proc. Natl. Acad. Sci. U. S. A., 2011, 108(SUPPL. 1), 4516–4522 CrossRef CAS .
43. P. D. Schloss, S. L. Westcott, T. Ryabin, J. R. Hall, M. Hartmann and E. B. Hollister, et al. Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities, Appl. Environ. Microbiol., 2009, 75(23), 7537–7541 CrossRef CAS .
44. C. Quast, E. Pruesse, P. Yilmaz, J. Gerken, T. Schweer and P. Yarza, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools, Nucleic Acids Res., 2013, 41(D1), D590–D596 CrossRef CAS PubMed .
45. R Core Team, R Foundation for Statistical Computing, R: A language and environment for statistical computing, 2024, Available from: https://www.R-project.org/ Search PubMed.
46. A. M. Speers and G. Reguera, Electron donors supporting growth and electroactivity of geobacter sulfurreducens anode biofilms, Appl. Environ. Microbiol., 2012, 78(2), 437–444 CrossRef CAS PubMed .
47. S. Freguia, E. H. Teh, N. Boon, K. M. Leung, J. Keller and K. Rabaey, Microbial fuel cells operating on mixed fatty acids, Bioresour. Technol., 2010, 101(4), 1233–1238 CrossRef CAS .
48. J. Kretzschmar, C. Koch, J. Liebetrau, M. Mertig and F. Harnisch, Electroactive biofilms as sensor for volatile fatty acids: Cross sensitivity, response dynamics, latency and stability, Sens. Actuators, B, 2017, 241, 466–472 CrossRef CAS .
49. E. Atci, J. T. Babauta, S. T. Sultana and H. Beyenal, Microbiosensor for the detection of acetate in electrode-respiring biofilms, Biosens. Bioelectron., 2016, 81, 517–523 CrossRef CAS .
50. A. Esteve-Núñez, M. Rothermich, M. Sharma and D. Lovley, Growth of Geobacter sulfurreducens under nutrient-limiting conditions in continuous culture, Environ. Microbiol., 2005, 7(5), 641–648 CrossRef PubMed .
51. E. L. King, K. Tuncay, P. Ortoleva and C. Meile, In Silico Geobacter sulfurreducens Metabolism and Its Representation in Reactive Transport Models, Appl. Environ. Microbiol., 2008, 75(1), 83 CrossRef PubMed .
52. H. Sun, Y. Zhang, S. Wu, R. Dong and I. Angelidaki, Innovative operation of microbial fuel cell-based biosensor for selective monitoring of acetate during anaerobic digestion, Sci. Total Environ., 2019, 655, 1439–1447 CrossRef CAS .
53. J. M. Tront, J. D. Fortner, M. Plötze, J. B. Hughes and A. M. Puzrin, Microbial fuel cell biosensor for in situ assessment of microbial activity, Biosens. Bioelectron., 2008, 24(4), 586–590 CrossRef CAS .
54. F. Kubannek, J. Block, B. Munirathinam and R. Krull, Reaction kinetics of anodic biofilms under changing substrate concentrations: Uncovering shifts in Nernst-Monod curves via substrate pulses, Eng. Life Sci., 2022, 22(3–4), 152–164 CrossRef CAS .
55. A. Prévoteau and K. Rabaey, Electroactive Biofilms for Sensing: Reflections and Perspectives, ACS Sens., 2017, 2(8), 1072–1085 CrossRef PubMed .
56. D. R. Bond and D. R. Lovley, Electricity production by Geobacter sulfurreducens attached to electrodes, Appl. Environ. Microbiol., 2003, 69(3), 1548–1555 CrossRef CAS .
57. J. De Vrieze, L. Regueiro, R. Props, R. Vilchez-Vargas, R. Jáuregui and D. H. Pieper, et al. Presence does not imply activity: DNA and RNA patterns differ in response to salt perturbation in anaerobic digestion, Biotechnol. Biofuels, 2016, 9(1), 244 CrossRef .
58. T. Li, Q. Zhou, L. Zhou, Y. Yan, C. Liao and L. Wan, et al. Acetate limitation selects Geobacter from mixed inoculum and reduces polysaccharide in electroactive biofilm, Water Res., 2020, 177, 115776 CrossRef CAS .
59. A. Conklin, H. D. Stensel and J. Ferguson, Growth Kinetics and Competition Between Methanosarcina and Methanosaeta in Mesophilic Anaerobic Digestion, Water Environ. Res., 2006, 78(5), 486–496 CrossRef CAS PubMed .
60. M. B. Kurade, S. Saha, E. S. Salama, S. M. Patil, S. P. Govindwar and B. H. Jeon, Acetoclastic methanogenesis led by Methanosarcina in anaerobic co-digestion of fats, oil and grease for enhanced production of methane, Bioresour. Technol., 2019, 272, 351–359 CrossRef CAS PubMed .
61. M. S. M. Jetten, A. J. M. Stams and A. J. B. Zehnder, Methanogenesis from acetate: a comparison of the acetate metabolism in Methanothrix soehngenii and Methanosarcina spp, FEMS Microbiol. Rev., 1992, 8(3–4), 181–197 CrossRef .
62. J. Zhou, J. A. Smith, M. Li and D. E. Holmes, Methane production by Methanothrix thermoacetophila via direct interspecies electron transfer with Geobacter metallireducens, MBio, 2023, 14(4), e0036023 CrossRef PubMed .
63. D. E. Holmes, P. M. Shrestha, D. J. F. Walker, Y. Dang, K. P. Nevin and T. L. Woodard, et al. Metatranscriptomic Evidence for Direct Interspecies Electron Transfer between Geobacter and Methanothrix Species in Methanogenic Rice Paddy Soils, Appl. Environ. Microbiol., 2017, 83(9), e00223-17 CrossRef PubMed .
64. X. L. Su, Q. Tian, J. Zhang, X. Z. Yuan, X. S. Shi and R. B. Guo, et al. Acetobacteroides hydrogenigenes gen. nov., sp. nov., an anaerobic hydrogen-producing bacterium in the family Rikenellaceae isolated from a reed swamp, Int. J. Syst. Evol. Microbiol., 2014, 64(Pt 9), 2986–2991 CrossRef CAS PubMed .
65. S. F. Fu, J. N. Ding, Y. Zhang, Y. F. Li, R. Zhu and X. Z. Yuan, et al. Exposure to polystyrene nanoplastic leads to inhibition of anaerobic digestion system, Sci. Total Environ., 2018, 625, 64–70 CrossRef CAS PubMed .
66. L. Jabari, H. Gannoun, J. L. Cayol, A. Hedi, M. Sakamoto and E. Falsen, et al. Macellibacteroides fermentans gen. nov., sp. nov., a member of the family Porphyromonadaceae isolated from an upflow anaerobic filter treating abattoir wastewaters, Int. J. Syst. Evol. Microbiol., 2012, 62(Pt 10), 2522–2527 CrossRef CAS PubMed .
67. Z. W. He, C. C. Tang, W. Z. Liu, Y. X. Ren, Z. C. Guo and A. J. Zhou, et al. Enhanced short-chain fatty acids production from waste activated sludge with alkaline followed by potassium ferrate treatment, Bioresour. Technol., 2019, 289, 121642 CrossRef CAS PubMed .
68. B. Zhang, X. Xu and L. Zhu, Structure and function of the microbial consortia of activated sludge in typical municipal wastewater treatment plants in winter, Sci. Rep., 2017, 7(1), 1–11 CrossRef PubMed .
69. T. Murakami, T. Segawa, N. Takeuchi, G. B. Sepúlveda, P. Labarca and S. Kohshima, et al. Metagenomic analyses highlight the symbiotic association between the glacier stonefly Andiperla willinki and its bacterial gut community, Environ. Microbiol., 2018, 20(11), 4170–4183 CrossRef CAS PubMed .
70. I. Owusu-Agyeman, E. Plaza and Z. Cetecioglu, Long-term alkaline volatile fatty acids production from waste streams: Impact of pH and dominance of Dysgonomonadaceae, Bioresour. Technol., 2022, 346, 126621 CrossRef CAS PubMed .
71. B. Shamurad, P. Sallis, E. Petropoulos, S. Tabraiz, C. Ospina and P. Leary, et al. Stable biogas production from single-stage anaerobic digestion of food waste, Appl. Energy, 2020, 263, 114609 CrossRef CAS .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ew00233h

This journal is © The Royal Society of Chemistry 2025