Improving the Kinetics of H₂-Fueled Biological Methanation with Quinone-Based Redox Mediators

Abstract

The biomethanation process involves the conversion of CO₂ into a valuable energy carrier (i.e., methane) by methanogenic archaea. Since it can be operated at mild conditions, it is more sustainable than traditional chemical approaches. Nevertheless, the efficacy of biomethanation is limited by the low kinetics of the microbiological reaction and the poor solubility of H₂ in water. Herein, the effect of soluble (i.e., AQDS) and insoluble (i.e., biochar) quinone-based redox mediators on the kinetics of H₂-fueled biological methanation in bench-scale microcosms was investigated. Microcosms were set up in 120 mL serum bottles and were initially inoculated with a methanogenic sludge deriving from a lab-scale anaerobic digester treating food waste. As a result, the kinetics of H₂ consumption and CH₄ generation were greatly increased (p < 0.05) in presence of AQDS as compared to the control, accounting for up to +160% and +125% in the last experimental cycle, respectively. These findings could be explained by a two-step mechanism, whereby microbes used H₂ to quickly reduce AQDS into the highly soluble AH₂QDS, which in turn served as a more efficient electron donor for methanogenesis. In contrast, the used biochar had apparently an adverse effect on the biomethanation process.

1. Introduction

The release of carbon dioxide into the atmosphere is one of the main causes of global warming and ocean acidification [1]. These phenomena are, in turn, (co)responsible for several negative side-effects on the environment, such as the rise in sea levels, the reduction in ice caps and glaciers and the worsening of extreme events (i.e., heat waves, wildfire, floods, droughts and cyclones), thus threatening the whole biosphere [2]. It is well-known that human activities such as fossil fuels combustion and deforestation are the main contributors of CO₂ emissions [3], and, unfortunately, the trend of atmospheric CO₂ emissions is on the rise [4]. The Paris agreement adopted at the UN Climate Change Conference (COP 21) and adopted by 196 nations aims to take action against global warming and to limit the temperature increase to 1.5 °C above pre-industrial levels [5]. Mitigation strategies, such as carbon capture and storage (CCS) and carbon capture and utilization (CCU), are considered to be valid approaches to reduce the impact of CO₂ emissions [6]. For example, the conversion of CO₂ into methane via hydrogenation (i.e., methanation) couples the removal of CO₂ with the production of a valuable energy carrier [7], according to the following equation:

4H₂ + CO₂ → CH₄ + 2H₂O

(1)

Methanation is also utilized to upgrade the biogas produced during the anaerobic digestion process, as CO₂ is the main impurity present in the raw biogas, up to 45% by volume [8]. The molecular hydrogen necessary to carry out methanation is usually provided by means of water electrolysis, following the power-to-gas approach [9]. In this way, the surplus of energy generated by nonprogrammable renewable energy sources such as wind and photovoltaic can be stored in form of CH₄. It is worth noticing that methane is a more versatile chemical feedstock than H₂, since it allows easier and safer transportation and storage, has fewer barriers for commercial implementation and can count on the well-established natural gas industry and networks [10]. Furthermore, the energy density of CH₄ per volume (35.88 MJ m⁻³) is significantly higher than that of H₂ (10.79 MJ m⁻³) [11].

Traditionally, methanation from H₂ and CO₂ is carried out with a chemical process known as Sabatier reaction, which is carried out at high-temperature (e.g., 300–550 °C), pressure (e.g., 1–100 bar) and in presence of a metal catalyst (e.g., Ni, Fe, Ru) [12,13]. At present, chemical methanation is at a high technological readiness level (TRL) and currently is deployed at a pilot scale at different sites [14]. Conversely, low-temperature methanation can be achieved using biological processes, where CO₂ to CH₄ conversion is catalyzed by methanogenic archaea, in the so-called biomethanation [15]. The high energy consumption of the Sabatier process (26–35 kWh m⁻³ CH₄) [16], makes it less economically and environmentally sustainable than biomethanation.

Besides mild reaction conditions, biomethanation has several other advantages over catalytic methanation: (i), irreversibility of the biochemical reactions, (ii) less byproducts, (iii) insensitivity to the impurities and (iv) does not need a H₂:CO₂ ratio of 4 or higher in the feedstock gas mixture [17]. Furthermore, the self-replicating nature of the biological catalyst makes it less susceptible to irreversible inhibition by sulfur compounds and ammonia, allowing for a more robust process [18,19].

Nevertheless, main limitation factors of the biomethanation process include the relatively low intrinsic kinetics of the microbiological reaction and the sluggish H₂ mass transfer mainly due to its low solubility in water (i.e., 23 times lower than that of CO₂) [20]. As a matter of fact, these combined factors ultimately require substantially larger reactor dimensions with respect to the catalytic methanation process [17,21,22].

In recent years, different strategies have been proposed to address some of these limitations, such as operating biomethanation reactors at high pressures (e.g., 10 bar) to increase the equilibrium concentration of dissolved H₂ and thus the concentration driving force for H₂ gas–liquid mass transfer or at high temperatures (e.g., 55 °C) to increase the kinetics of microbial reaction. For both strategies, however, the advantage of reduced capital expenses triggered by the smaller needed biomethanation reactor has to be balanced against the increase in operational expenses (OPEX) for the energy needed for operation at higher pressures or temperatures [23].

Soluble redox mediators such as anthraquinone-2,6-disulfonate (AQDS), neutral red and cysteine have also been being studied for the improvement of CH₄ generation in anaerobic digestion processes [24,25,26,27]. These electron shuttles can promote direct interspecies electron transfer (DIET) mechanisms [28], thus increasing the kinetics of biomethanogenesis. In particular, AQDS is a humic substance analogue that is widely used as model electron shuttle to stimulate microbial metabolism [29,30,31].

Analogously, biochar has also been studied as an amendment for increasing CH₄ content and reducing the residual CO₂ in anaerobic digesters [32]. Biochar is a carbonaceous material produced by pyrolysis of lignocellulosic biomass, such as agricultural residues [33]. This material can indeed enhance methanogenesis in several different ways: (i) providing a large surface area for microbial colonization, (ii) enriching the microbial communities, (iii) acting as a stimulator for buffering capacity and (iv) increasing the DIET between fermentative bacteria and methanogens thanks to its conductivity and to the redox properties of its surface-bound quinone moieties [33,34,35].

In this work, the impact of two different quinone-based redox mediators, one soluble (i.e., AQDS) and one insoluble (i.e., biochar), on the kinetics of H₂-fueled biological methanation in bench-scale microcosms was studied. The performances were evaluated in terms of hydrogen consumption, methane and VFAs generation with respect to unamended controls. The running hypothesis of this study is that mediators, upon reduction, could serve as direct electron donors for methanogenesis and sustain methane production at higher rates relative to H₂ alone, due to their higher solubility and/or availability in the liquid phase. Overall, the results of this study have relevance in the development of biogas upgrading technologies and power-to-gas processes to store renewable electric energy in form of methane.

2. Results and Discussion

2.1. Impact of Quinone-Based Mediators on H₂ Consumption and CH₄ Production

The trends of H₂ consumption and CH₄ and VFAs generation, in all the microcosms, over three successive feeding cycles is depicted in Figure 1. It is apparent that H₂ consumption was strictly linked with the production of CH₄. VFAs production was also present, but in much lower extent, accounting for less than 10% of H₂ utilization in all microcosms.

During the first feeding cycle, the average value of H₂ utilization rate in the microcosms supplemented with AQDS was 21.9 ± 2.3 mgCOD L⁻¹ d⁻¹ (Figure 2A), hence nearly twice as much the value observed in the unamended controls (10.4 ± 1.2 mgCOD L⁻¹ d⁻¹). By contrast, the rate of H₂ utilization in the biochar-supplemented microcosms was substantially lower (6.3 ± 2.4 mgCOD L⁻¹ d⁻¹) than in the unamended controls, suggesting a possible inhibitory effect exerted by the conductive material. A slightly different picture was noticed during the second feeding cycle, whereby H₂ utilization in the control microcosms and in those supplemented with AQDS (67.7 ± 4.4 mgCOD L⁻¹ d⁻¹ vs. 56.8 ± 1.3 mgCOD L⁻¹ d⁻¹) was statistically indistinguishable (p = 0.14). However, this apparent discrepancy is consistent with the fact that, although at the start of the cycle all bottles received the same dose of H₂, the resulting initial nominal concentration in the control microcosms (i.e., 278 mgCOD L⁻¹ d⁻¹) was almost 60% higher than the one measured in the AQDS microcosms (i.e., 180 mgCOD L⁻¹ d⁻¹), since the unamended controls had only partially consumed the H₂ supplied during the previous cycle. Accordingly, the positive effect of the quinone mediator on the kinetics of H₂ utilization in the AQDS-amended microcosms was masked by the higher H₂ concentration in the unamended controls. Moreover, in this second cycle H₂ utilization in the biochar-amended microcosms was lower (34.7 ± 0.2 mgCOD L⁻¹ d⁻¹) and strongly delayed compared to the other treatments.

At the start of the third cycle, all the microcosms were supplied with the same initial H₂ concentration, and the rate of H₂ consumption in the AQDS-amended microcosms (54.2 ± 5.3 mgCOD L⁻¹ d⁻¹) returned to be substantially (p < 0.05) higher than in the control microcosms (20.9 ± 0.1 mgCOD L⁻¹ d⁻¹), as expected. Besides that, this third feeding cycle further confirmed the apparent inhibitory effect of biochar on H₂ utilization, which proceeded at lower rate (3.3 ± 1.2 mgCOD L⁻¹ d⁻¹) as compared to the unamended controls.

The low performance of the biochar amended microcosms seems to be in apparent contrast with many findings reported in the literature, whereby biochar is used to enhance methanogenesis [33]. In the majority of these studies, however, biochar was applied during the anaerobic digestion of (waste) organic compounds, while here the only carbon source for methanogens was CO₂. Hence, this behaviour could possibly be ascribed to the adsorption properties of biochar itself, whose capability of capturing CO₂ may have resulted in a reduced substrate availability for the methanogens, and in turn in a limited methanogenic activity [36,37,38].

As previously noted, H₂ consumption was closely linked to CH₄ production in all treatments, and also acetate production to a lesser extent. Figure 2B shows the average rate of CH₄ production in the different microcosms over the three feeding cycles, which nicely mirrored the values and trends reported for H₂ consumption. Concerning VFAs, a slightly higher production (Figure 1) was obtained in the presence of biochar (up to nearly 25 mgCOD L⁻¹, at the end of the second feeding cycle), but the relative contribution to H₂ utilization remained, however, substantially lower than 10%, in all treatments.

Finally,

Table 1. Mass balance for all treatments, relative to the whole experimental period.

	Consumed H2/mgCOD L−1	Produced CH4+ VFAs/mgCOD L−1	Recovery/%
Biochar	513 ± 36	516 ± 2	101
AQDS	610 ± 53	717± 68	117
Control	639 ± 14	725 ± 16	113

presents an overall COD balance for the above reported experiments relative to the whole experimental period. Notably, the conversion of the consumed H₂ into metabolic products (i.e., methane and VFAs) ranged between 101–117% in all treatments, thus confirming that all key metabolic processes were properly identified and quantified.

2.2. Deciphering the Mechanisms of AQDS-Enhanced Methanogenesis

Results of the above-described microcosm incubations highlighted the intriguing capacity of AQDS, a soluble quinone-based redox mediator, to enhance the kinetics of H₂ bioconversion into CH₄. This finding could be explained by assuming the following two-step reaction mechanism: firstly, AQDS is rapidly reduced by H₂ into AH₂QDS, this latter then served as a direct electron donor for CH₄ generation (Figure 3), according to the following reaction stoichiometry:

Step 1: 4H₂ + 4AQDS → 4AH₂QDS

(2)

Step 2: 4AH₂QDS + CO₂ → CH₄ + 4AQDS + 2H₂O

(3)

The finding that H₂ utilization in the presence of AQDS proceeded at higher rates relative to the unamended controls clearly indicates that mediator reduction is a faster reaction compared to CO₂ reduction. To ascertain whether the H₂-driven reduction of AQDS into AH₂QDS was a biotic or abiotic process, ad hoc cyclic voltammetry (CV) experiments were carried, as depicted in Figure 4. To start with, a CV conducted in an anaerobic mineral medium supplemented with 0.1 mM AQDS was recorded, which revealed the characteristic redox signature of the mediator, with a couple of clearly identifiable anodic and cathodic peaks with formal potential E_f −0.30 V vs. Ag/AgCl. To verify whether H₂ could abiotically reduce AQDS, an excess of H₂ (60 mL) was added to the headspace of the bioelectrochemical cell, which was kept vigorously stirred at 500 rpm for 24 h to favour H₂ dissolution into the liquid phase, and possibly its reaction with AQDS. Thereafter, the CV was repeated to ascertain the possible accumulation of AH₂QDS which, in principle, would result in an increased intensity of the anodic oxidation peak in the CV. Interestingly, the CVs of AQDS recorded both in the absence and in presence of H₂ were almost indistinguishable, thus indicating that H₂ could not abiotically reduce AQDS under the tested conditions. Based on this result, it could be then hypothesized that the reduction of AQDS into AH₂QDS during the microcosm experiments was biologically catalyzed by microorganisms occurring in the anaerobic sludge used as inoculum. Indeed, the capability of AQDS to serve as a respiratory electron acceptor by anaerobic microorganisms which use H₂ as electron donor has been widely documented in the literature [39]. Finally, a positive control experiment was carried out to confirm that the employed electrochemical technique was sensitive enough to detect changes in the concentration of AH₂QDS. In brief, the working electrode was kept polarized (for 24 h) at −0.7 V vs. Ag/AgCl to electrochemically reduce the mediator in solution. Notably, the CV recorded at the end of the chronoamperometry (Figure 4) clearly revealed a substantial increase in the intensity of the anodic current, thus confirming the validity of the adopted experimental approach.

Based on these experiments, the enhancement in the H₂ utilization rate could be attributed to the presence within the microbial culture of microorganisms capable to “respire” AQDS and rapidly convert it into AH₂QDS using H₂ as electron donor. On the other hand, the observed corresponding enhancement of the rate of CH₄ production warrants further considerations. One possibility is that, since AH₂QDS is more soluble than H₂ and was supplied at a relatively high concentration (i.e., 0.5 g L⁻¹), the reduced form of the mediator pooled up in the bulk liquid, ultimately driving methane production at higher rates compared to H₂. Another intriguing possibility is that, while H₂ could only serve as an electron donor in the metabolism of hydrogenophilic methanogens, AH₂QDS could fuel methane production by both hydrogenophilic and acetoclastic methanogens, thereby increasing the overall concentration of active biocatalysts participating in the methanation process.

Regardless the underlying mechanisms ultimately responsible for the observed enhancement of methane production, it is worth noting that the herein reported results are fully in agreement with the few previous literature studies investigating the impact of AQDS on methane production that documented its stimulatory effect (Figure 5).

As an example, Xu and colleagues [24] observed a 10-fold enhancement of methane production from acetate (10 mmol L⁻¹) as compared to an unamended control in presence of AQDS (initial concentration 5 mmol L⁻¹). A five-fold enhancement was instead observed when the mediator was provided at a lower concentration of 0.5 mmol L⁻¹. Microbiological analyses indicated that methane production was most likely catalyzed by Methanosarcina species, microorganisms capable of producing methane from either acetate, H₂ and CO₂ as well as other simple substrates such as methanol. In a more recent study, 0.1 mmol L⁻¹ AQDS enhanced by nearly 30% the production of methane from acetate by Methanosaeta [25]. Using paddy soil as inoculum and propionate as electron donor, Zhuang and colleagues [27] reported a 15% increase in the rate of methane production in the presence of AQDS. The hydrogenophilic methanogen Methanobacterium was the microorganisms most likely responsible for methane production.

Taken as a whole, the above-reported studies seem to confirm that the capability to use AH₂QDS as electron donor is widely distributed among methanogenic organisms.

3. Materials and Methods

3.1. Chemicals

All chemicals used for the experiments were of analytical grade and were purchased from Merck KGaA (Darmstadt, Germany). De-ionized water (Millipore, Darmstadt, Germany) was used to prepare the microbial medium and all other solutions.

3.2. Microcosms Setup

The microcosms were prepared using 120 mL serum bottles sealed with Teflon-faced butyl rubber stoppers. The bottles were filled with 60 mL of mineral medium, prepared as follows (g/L): NaHCO₃ 4.2; NH₄Cl 0.5; MgCl₂•6H₂O 0.1; CaCl₂•2H₂O 0.05; K₂HPO₄ 8; KH₂PO₄ 8. To the mineral medium, 10mL L⁻¹ of trace metal solution and 1ml L⁻¹ of vitamin solution were added. The trace metal solution contained (g L⁻¹): Nitrilotriacetic acid 4.5; FeSO₄•7H₂O 0.556; MnSO₄•H₂O 0.086; CoCl₂•6H₂O 0.17; ZnSO₄•7H₂O 0.21; H₃BO₃ 0.019; NiCl₂ 0.02; Na₂MoO₄ 0.01. The vitamin solution was composed of (g L⁻¹): Biotin (B7) 0.02; Folic acid (B9) 0.02; Pyridoxine (B6) 0.1; Thiamine (B1) 0.05; Riboflavin (B2) 0.05; Nicotinic acid (B3) 0.05; Pantothenic acid (B5) 0.05; Cyanocobalamin (B12) 0.002; 4-aminobenzoic acid (B10) 0.05. The final pH of the medium was 7. The microcosms were amended by adding the selected redox mediator in each bottle, as shown in Figure 6. In this way, three different conditions were tested: in the first type 5 g L⁻¹ of AQDS (Merck KGaA, Darmstadt, Germany) were added, the second type was amended with 25 g L⁻¹ of biochar powder, and finally no mediator was supplemented in the control microcosms.

The biochar powder used for this study was obtained by gasifying pine wood at approximately 850 °C. The elemental composition of the material is as follows: 78%C; 4.18%Ca; 1.48% K; 0.67%Si; 0.64%Mn and 0.46%Fe. Furthermore, this biochar consisted in a micro-mesoporous material, having a specific surface area of 343 ± 2 m² g⁻¹, an electrical conductivity >2 S m⁻¹, and a pore volume value of 0.383 cm³ g⁻¹. The micro- and mesopores are continuously distributed in a 20–1000 Å range, and appear most abundant in the 20–100 Å range. No pre-treatments were conducted on the biochar prior to its use in the experiments. Additional information on the material can be found elsewhere [40].

After that, the microcosms were inoculated with 6 mL of anaerobic sludge (~25 gVSS/L) collected from a lab-scale anaerobic digester treating food waste. All microcosms were purged with a N₂/CO₂ (70/30 v/v) gas mixture to remove O₂ and establish anaerobic conditions. Subsequently, 15 mL of headspace were replaced with pure H₂, resulting in a H₂/CO₂ ratio of 25/22% (v/v). During the whole study, the microcosms were constantly stirred by means of a magnetic stirrer and kept at room temperature (24 ± 3 °C).

The experiment lasted for 25 days during which three feeding cycles were performed. At the beginning of each cycle, H₂ was provided in the headspace of the microcosms. At the end of cycles 2 the headspace was purged with N₂/CO₂ (70/30 v/v) to restore the CO₂ level and remove the produced CH₄.

3.3. Analytical Methods

The liquid phase of the microcosms was constantly analyzed during the experiments to monitor the production of VFAs. Liquid samples (1 µL) were filtered (0.45 μm, nylon, Sartorius, IT), acidified with 100 µL of oxalic acid 3M and injected with a syringe into a gas-chromatograph (GC, Perkin–Elmer 8500, Waltham, MA, USA) equipped with a flame ionization detector (FID), mounting a glass-packed column (Carbopack B-DA80/120 4% CW20M; ID 2 mm; length 2 m; Supelco) and operated as follows: gas carrier N₂; flux 40 mL min⁻¹; column temperature 175 °C; injector temperature 200 °C; detector temperature 200 °C.

In order to monitor H₂ consumption and CH₄ generation, 50 μL of gaseous samples were collected with a gastight syringe (Hamilton, Reno, NV, USA) from the headspace of the bottles and injected into a gas-chromatograph (GC, Perkin–Elmer Auto System, Waltham, MA, USA) equipped with a thermal conductivity detector (TCD), mounting a steel packed column (60/80 Carbopack B/1% SP-1000; ID 2mm; length 2 m; Supelco) and operated as follows: gas carrier N₂; flux 20 mL min⁻¹; column temperature 150 °C; injector temperature 200 °C; detector temperature 200 °C. Gas-phase concentrations were converted into nominal concentrations (i.e., total mass divided by the liquid phase in the bottle concentrations) using tabulated Henry’s Law constants [41].

3.4. Cyclic Voltammograms

To study the redox activity of AQDS, a two-chamber electrochemical cell was set up, with a three-electrode configuration. In brief, a glassy carbon rod (length: 7.5 cm, ø: 0.5 cm, HTW GmbH, Thierhaupten, Germany) served as the working electrode, a graphite rod (purity: 99.995%, length: 7.5 cm, ø: 0.6 cm; Sigma-Aldrich, Milan, Italy) was the counter electrode, and an Ag/AgCl electrode was used as reference. Both chambers were filled with phosphate buffer 0.5 M, pH 7 and were kept separated by Nafion^® membrane (Dupont Co., Wilmington, DE, USA), which was pretreated as follows: the membrane is initially placed in a boiling solution of H₂O₂ 3% for 2 h, then it is soaked in boiling demineralized water for 2 h, after that it is put in a boiling solution of H₂SO₄ for 2 h, and finally it is soaked again in boiling demineralized water for 2 h [42]. AQDS was added to the chamber of the working electrode in a concentration of 100 µM. With this setup, cyclic voltammetries were performed at a scan rate of 20 mV s⁻¹ by means of an IVIUMnSTAT potentiostat (IVIUM Technologies, The Netherlands). First of all, the voltammograms were recorded in absence of externally supplied H₂. Then, 60 mL of H₂ were injected in the headspace of the half-cell where the working electrode was present, and the voltammetry was repeated. Finally, the working electrode was polarized at −0.7 V vs. Ag/AgCl for three days in order to reduce all AQDS present in solution, and the cyclic voltammetry was repeated once more.

3.5. Calculations

The concentrations of H₂, CH₄ and VFAs were converted into mgCOD L⁻¹ using the following conversion factors (mg mgCOD⁻¹): acetic acid 1.066; propionic acid 1.51; butyric acid 1.81; H₂ 8; CH₄ 4.

The average rates of H₂ consumption and CH₄ production were calculated for each cycle as the slope of the linear regression line obtained from the concentration vs. time plot. The regression line was calculated using a function of the program Microsoft Excel^® (Microsoft, Redmond, WA, USA).

4. Conclusions

This study shows that soluble quinone-based redox molecules such as AQDS can substantially improve the kinetics of H₂-fueled biological methanation. Indeed, the kinetics of H₂ consumption and CH₄ generation were greatly increased in presence of AQDS as compared to the controls. As a matter of fact, the hydrogen consumption rates in AQDS-amended microcosms were increased by 110 and 160% as compared to the unamended controls during the first and the third cycle, respectively, while methane generation rates were 181 and 125% higher. These results could be explained by assuming a two-step reaction mechanism, where AQDS is rapidly reduced into AH₂QDS by H₂, and then this latter is used as direct electron donor for methanogenesis. Since we demonstrated with a series of cyclic voltammetries in presence ad absence of H₂ that AQDS could not be reduced abiotically, it seems likely that the increase in the H₂ consumption rate could be due to the presence of microorganisms capable to “respire” AQDS using H₂ as electron donor, according to a process widely reported in the literature [39]. We hypothesized two possible mechanisms that explain the increased rate of methanogenesis: (i) AH₂QDS pooled up in the bulk liquid, thus increasing the availability of electron donors for the methanogens as compared to the less soluble H₂, or (ii) unlike H₂, AH₂QDS was metabolized by both hydrogenophilic and acetoclastic methanogens, thus increasing the overall concentration of active biocatalysts participating in the methanation process.

It is worth noticing that the results obtained in this work are fully in agreement with the few literature studies that investigated the stimulatory effect of AQDS on methane production.

On the contrary, biochar had an adverse effect on methane production, which seems in contrast with the many pieces of literature evidence that report its positive effect on methanogenesis. However, in most studies biochar was applied to the anaerobic digestion of organic compounds, while in the present study CO₂ was the only carbon source. Therefore, this phenomenon is probably due to the absorption capability of biochar: the capture of CO₂ may have resulted in a reduced substrate availability for the methanogens, and thus in a reduced methanogenic activity.

These findings pave the way for innovative applications of AQDS-assisted methanogenesis, such as in the field of CH₄ generation in microbial electrolysis cells (MECs). Furthermore, the presented results have relevance in the development of biogas upgrading technologies and power-to-gas processes, that allow the storage of renewable electric energy in form of methane. Clearly, further work is needed to define the optimal mediator supplementation strategies as well as to examine its long-term chemical and biochemical activity within anaerobic bioreactors.