Gas Substrate Effects on Hydrogenotrophic Biomethanation in Flocculent and Granular Sludge Systems

Abstract

The biotechnological conversion of CO₂ to biomethane represents an energy-efficient, environmentally friendly, and sustainable approach within the waste-to-energy cycle. This process, in which CO₂ and H₂ are converted to biomethane in anaerobic bioreactors, is referred to as hydrogenotrophic biomethane production. While several studies have investigated hydrogenotrophic biomethane production, there is a lack of research comparing flocculent and granular sludge inoculum in continuously operated systems fed with a gas substrate. Both granular and flocculent sludge possess distinct advantages: granular sludge offers higher density, stronger microbial cohesion, and superior settling performance, whereas flocculent sludge provides faster substrate accessibility and more rapid initial microbial activity. In this study, two UASB (Upflow Anaerobic Sludge Blanket) reactors operated under mesophilic conditions were continuously fed with synthetic off-gas composed of pure H₂ and CO₂ in a 4:1 ratio and were compared in terms of microbial community shifts and their effects on hydrogenotrophic biomethane production. Biomethane production reached 75 ± 2% in the granular sludge reactor, significantly higher than the 64 ± 1.3% obtained with flocculent sludge. Although hydrogen consumption did not differ significantly, the granular sludge reactor exhibited higher CO₂ removal efficiency. Microbial analyses further revealed that granular sludge was more effective in supporting methanogenic archaea under conditions of gas substrate feeding. These findings offer advantageous suggestions for improving biogas production, enhancing waste gas management, and advancing sustainable energy generation.

1. Introduction

Biogas is recognized as a promising alternative to fossil fuels due to its clean and sustainable nature. However, its direct use as a fuel is limited by its high carbon dioxide (CO₂) content. To enhance its fuel quality, biogas must be upgraded to biomethane by removing CO₂, thereby increasing its calorific value and economic potential [1]. Several biogas upgrading technologies utilize biological approaches, such as microalgae for CO₂ absorption or hydrogenotrophic methanogenic archaea, which convert CO₂ and H₂ into CH₄ through in situ or ex situ pathways [2]. Microbial-based technologies present a cost-effective and environmentally friendly method for biogas upgrading [3]. Among these, the bioconversion of CO₂ into biomethane via the addition of hydrogen is regarded as a particularly promising biotechnology [4]. Hydrogenotrophic methanogens metabolically convert CO₂ and H₂ into CH₄ [5].

Biogas upgrading reactors typically consist of complex anaerobic microbial consortia [5]. The structure of the microbial community within the reactor significantly affects the performance of anaerobic systems. For example, free-living planktonic cells of smaller sizes can act as precursors to early-stage membrane biofouling, primarily through adhesion to the membrane surface [6]. Additionally, operating and environmental parameters such as nutrient availability, temperature, pH, and the presence of inhibitors play a critical role in shaping microbial communities and, consequently, influence methanogenesis efficiency [7,8]. Beyond microbial composition, the physical structure of the inoculum—whether flocculent or granular—is believed to impact methanogenic activity significantly [9]. The physical configuration of the anaerobic inoculum is particularly important during the rate-limiting step of methanogenesis, as it determines the efficiency of the overall anaerobic digestion process. Therefore, understanding the influence of granular floc structure on both methanogenesis and methanogen community diversity is essential [10].

Over three decades ago, it was noted that activated sludge flocs could aggregate into structures with diameters of 1–3 mm, known as granular sludge [11]. This type of sludge consists of densely packed microbial biomass in compact granules formed without the need for support material. Compared to flocculent biomass, granular sludge offers several advantages, including greater density, stronger microbial cohesion, and significantly higher settling rates—at least three times faster than flocs. These superior settling characteristics enable higher biomass retention, increased reactor efficiency, and enhanced contaminant removal [12]. Furthermore, granular sludge displays longer biomass retention times, greater tolerance to variations in contaminant loading, and enhanced resistance to environmental fluctuations such as pH and temperature. These properties make granular sludge a cost-effective and space-saving alternative to traditional flocculated sludge systems in wastewater treatment applications [13].

Previous studies have primarily examined these inoculum types in liquid-substrate systems or under batch conditions with organic feedstocks, resulting in a gap in understanding their performance in continuously operated, gas substrate-fed systems, particularly in relation to biomethane production. These studies suggest that the structure of the inoculum can affect reactor performance, depending on the substrate type used. Li et al. emphasized the differences between flocculent and granular sludge in systems utilizing completely autotrophic nitrogen removal over nitrite (CANON). They analyzed nitrogen removal performance, microbial community composition, and functional bacterial activity in both hybrid (floc-granule) and purely granular systems [14]. Similarly, Baltyn et al. performed a comprehensive comparison of the physical, structural, microbiological, and dewatering characteristics of flocculent and granular activated sludge sourced from the same wastewater [15]. Qiu et al. explored the removal and fate of sulfamethoxazole (SMX) in sulfate-reducing upflow sludge blanket (SRUSB) reactors inoculated with either flocculent or granular sulfate-reducing bacteria (SRB), highlighting the differences between the two inoculum types based on their physical, biological, and operational characteristics [16].

Liquid-substrate systems are generally crafted for waste management and energy recovery, while gas-substrate systems concentrate on biogas purification and the production of high-quality fuels [17]. Zakoura et al. evaluated the performance of three mesophilic UASB bioreactors treating olive mill wastewater (OMW), comparing the impacts of flocculent and granular anaerobic inoculum under various organic loading rates [18]. In another study, the performance of anaerobic reactors seeded with flocculent and granular sludge was directly compared under different organic loading rates, revealing greater stability and methane production in flocculent-based systems, despite the absence of granule formation [19]. A comparative study assessed flocculent and granular anaerobic sludges for their capacity to degrade long-chain fatty acids (LCFAs), demonstrating that both biomass types displayed distinct metabolic capabilities and structural behaviors during LCFA conversion [20].

While anaerobic digestion with liquid substrates mainly targets organic matter degradation and biogas production, gas substrate-based systems (CO₂ and H₂) aim to enhance biogas quality by generating high-purity biomethane. Hydrogenotrophic biomethane production from gaseous substrates can be categorized into in situ and ex situ approaches. In in situ systems, H₂ is introduced alongside organic feedstock, whereas in ex situ systems, only CO₂ and H₂ are supplied to produce biomethane [21,22].

Despite the growing interest in ex-situ biogas upgrading, research on how gas substrate feeding in such systems influences flocculent versus granular inoculum, their effects on biomethane production, and the subsequent changes in microbial community composition remains limited. Pan et al. evaluated batch reactors supplied with formate, hydrogen, and acetate, utilizing both intact and crushed granules [9]. Wu et al. investigated how variations in granule size affect biomethane production under different chemical oxygen demand (COD) levels, providing further insight into the significance of sludge structure [23]. Pan et al. explored the performance of various substrates (e.g., organic waste or mixtures) in batch anaerobic systems, emphasizing the effects of substrate variability on methane production and system stability; however, their study was confined to batch mode and did not address the dynamics of continuous-flow systems. Wu et al. assessed the impact of anaerobic granule sizes (small, medium, and large) on biogas production at a microscale, concentrating on convective diffusion and mass transfer mechanisms. They demonstrated that larger granules enhanced mass transfer due to an increased permeable area; however, their analysis was granule-centric and did not consider floc structures or gas substrate-fed continuous systems.

In this study, we investigated biomethane production and microbial population dynamics under hydrogenotrophic conditions with continuous gas substrate feeding, using both flocculent and granular sludges. Immobilized UASB reactors were operated under mesophilic conditions and continuously fed at a fixed gas flow rate, employing two distinct inoculum configurations. Inoculum samples were collected before and after reactor operation for microbial diversity analysis. Significant differences in microbial community structure were noted between the two sludge types. This study addresses a critical gap by comparing floc and granule inoculum in a gas substrate-fed (H₂/CO₂) continuous-flow system.

2. Materials and Methods

2.1. Inoculum

Anaerobic and mesophilic flocculent microbial inoculum was obtained from a biogas plant. The inoculum was sieved through a 2 mm mesh to remove large particles prior to processing. Anaerobic and mesophilic granular inoculum was collected from a wastewater treatment plant that processes beverage waste. Samples of both sludge types were collected before reactor operation and stored at −20 °C for subsequent microbial diversity analysis. The total solids (TS), volatile solids (VS), and pH of the flocculent inoculum were determined as 83 mg/L, 38 mg/L, and 8.2, respectively. The TS, VS, and pH of the granular inoculum were determined as 89 mg/L, 41 mg/L, and 8.1, respectively. No preconditioning was applied to the inoculum.

2.2. Reactor Operation

The UASB bioreactor was utilized for both floc and granular inoculum. The total volume of the reactor was 4.5 L, with a wet volume of 3.1 L. For both experiments, the bioreactor was filled with MBBR (Moving Bed Biofilm Reactor) for immobilization purposes. A gas mixture of H₂ and CO₂ was prepared in a pre-gas chamber before being introduced into the reactor at a ratio of 4 mol H₂ to 1 mol CO₂. The volumetric gas supply of H₂ and CO₂ was regulated by two mass flow controllers (ALLICAT), achieving a total gas supply rate of 6 m³ gas/m³ reactor/day (Figure 1).

Gas feeding was implemented from the bottom of the reactor. To measure the flow of gas in the headspace, a gas flow meter (ALICAT, Whisper, Marana, AZ, USA) was connected to the outlet of the reactor. A hot water circulator maintained the operating temperature at 37 °C. Since the reactor lacked internal temperature and pH probes, direct measurements of internal temperature were not possible. pH levels were monitored weekly with pH indicator paper, while external temperature was recorded daily using a thermometer. Throughout the operation, the working pH of both reactors consistently ranged from 8.0 to 8.5. During the entire reaction period, no mixing, liquid recirculation, or additional liquid feeding occurred.

Both experiments were sustained until a pseudo steady-state condition was reached. Coefficient of variation (COV) less than 20% was considered for the pseudo steady state condition for each performance parameter. COV is the percentage of the standard deviation of methane percentages in the head gas relative to the average methane percentages as explained in detail by Daglioglu et al. [24]. The reactors were operated for a total of 40 days. Steady-state conditions were achieved on day 24 in the flocculent sludge reactor (FIR) and on day 22 in the granular sludge reactor (GIR).

2.3. Calculations

The biomethane production per reactor volume, along with hydrogen and carbon dioxide consumption, was calculated according to [25]. Statistical analyses were conducted using SPSS software (version 18.0) to identify statistically significant differences in the percentage of biomethane present in the effluent, as well as in biomethane formation, hydrogen consumption, and carbon dioxide consumption. Tukey’s post-hoc test was employed for multiple comparisons to pinpoint the specific differences.

2.4. Analytical Methods

The biomethane content of the headspace gas was determined by injecting a 5 mL gas sample into a gas chromatograph (GC) (6890N Agilent, Santa Clara, CA, USA) that was equipped with a thermal conductivity detector and a Hayesep D 80/100 packed column. The temperatures for the injector, detector, and column were maintained at 120, 140, and 35 °C, respectively. Argon (Linde, Izmir, Türkiye) served as the carrier gas at a flow rate of 20 mL/min. A calibration gas mixture containing high-purity hydrogen, methane, and carbon dioxide (30% H₂, 30% CH₄, and 30% CO₂) was utilized for the GC calibration.

2.5. Microbial Community Analysis

Samples were collected uniformly, with two taken prior to reactor operation and two taken after its completion for analysis. To address variations in sequencing depth across samples (FIR start: 79.027; FIR finish: 80.224; GIR start: 66.708; GIR finish: 71.594), raw read counts were normalized using the Counts Per Million (CPM) method, and sequencing depth ratios between samples were calculated for sequencing statistics. The Shannon and Simpson indices were applied for taxonomic analysis.

2.5.1. DNA Isolation

Genomic DNA isolation from samples was obtained from the “Quick-DNA TM Fungal/Bacterial Miniprep Kit, Cat. No.: D6005.” The amount and purity of the isolated DNA were determined fluorometrically by Qubit.

2.5.2. Amplification of the 16S rRNA V3-V4 Region

The V3-V4 regions of the 16S rRNA gene used for species identification were amplified using the universal 341F-805R primer sequences on a SimpliAmp Thermal Cycler (ThermoFisher Scientific, Waltham, MA, USA). The V3-V4 regions of the 16S rRNA gene used for archaeal identification were amplified using the universal ARC787F-ARC1559R primer sequences on a SimpliAmp Thermal Cycler. The primer sequences are listed below.

• 341F: CCTACGGGNGGCWGCAG
• 805R: GACTACHVGGGTATCTAATCC
• ARC787F: ATTAGATACCCSBGTAGTCC
• ARC1559R: GCCATGCACCWCCTCT

PCR conditions are given below.

• 95 °C 10 min—initial denaturation (HS enzyme used)
• 35 cycles:

  —

  95 °C for 30 s—denaturation

  —

  53–48 °C for 30 s—annealing (touchdown PCR)

  —

  72 °C for 15 s—extension

• The temperature was reduced to 4 °C and PCR was completed.

2.5.3. Library Preparation and Sequencing

Illumina’s “Nextera XT DNA Library Prep Kit, Cat. No.: FC-131-1096” (San Diego, CA, USA) was used for library preparation of 16S rRNA V3-V4 and ARC787F-ARC1559R amplicon products. For indexing, “TG Nextera XT Index Kit v2 Set A (96 Indices, 384 Samples), Cat. No.: TG-131-2001” was used. PCR purification processes were performed with “AMPure XP beads” from Beckman Coulter (San Jose, CA, USA). Sequencing was performed with Illumina’s Miseq platform as paired-end (PE) 2 × 150 base reads. A minimum of ≥50,000 readings were carried out per sample.

2.5.4. Bioinformatics Analysis of Raw Data

Raw data reads (FASTQ) were QC checked, trimmed, and sorted into the OTU classes with the Kraken Metagenomics system. The Kraken application assigns taxonomic tags to short DNA sequences with high precision and speed [26].

3. Results and Discussion

3.1. Biomethane Production

Biomethane yield, head gas composition, and the consumption of hydrogen and CO₂ were studied in continuously gas-fed reactors inoculated with two structurally distinct types of sludge. In the reactor inoculated with granular sludge, the biomethane content of the head gas was measured at 75 ± 2%, while the reactor with flocculent sludge recorded a biomethane content of 64 ± 1.3% (Figure 2). In the flocculent inoculum reactor (FIR), hydrogen consumption decreased rapidly after day 15 and reached zero by day 20. The biomethane content in the headspace gas was observed to stabilize by day 19. Conversely, in the granular inoculum reactor (GIR), hydrogen consumption remained high until day 15. Although a significant amount had been consumed by this point, approximately 2% hydrogen was still detected in the head gas. Biomethane production per unit reactor volume was calculated to be 2.77 ± 0.08 m³ CH₄/m³ reactor/day in the GIR and 2.48 ± 0.09 m³ CH₄/m³ reactor/day in the FIR (Figure 3).

Hydrogen consumption during the steady-state phase of the reactors is illustrated in Figure 4a. The flocculent inoculum reactor (FIR) achieved 100% hydrogen consumption, while the granular inoculum reactor (GIR) reached slightly lower levels at 98%. A limited number of studies focusing on gas substrates have explored the relationship between structure, hydrogen and CO₂ consumption, and methane production. Pan et al. utilized the Gompertz model to highlight that the H₂/CO₂ methanogenesis rate was greater in crushed granules, which was attributed to the increased availability of hydrogen-consuming methanogens, such as Methanothermobacter thermautotrophicus. Additionally, the lag phase was longer in crushed granules (0.29 days compared to 0.02 days), indicating that the granule structure enhances H₂ accessibility [9].

In their study on granule size, Wu et al. emphasized the significance of granule dimensions, attributing the superior efficiency of larger granules to their biofilm-like porous structure. The study noted that these granules provide an extensive gas–biomass contact area, facilitating more effective dissolution of H₂ and improved delivery to microorganisms, thereby offering a mass transfer advantage. According to the modified Fick’s diffusion law, larger granules with shorter channel lengths (5 × 10⁴ m/g) and greater pore volumes (0.14 mL/g) promote higher substrate diffusion rates and enhance microbial bioactivity [23].

Hydrogen consumption has been recognized as a limiting factor in systems designed for hydrogenotrophic biomethane production [1]. To address this limitation, various studies have examined the impacts of operational parameters, including gas recirculation, mixing intensity, and reactor configuration [27]. For instance, Luo and Angelidaki [28] reported an increase in methane content from 57% to 75% by improving the mixing rate, while Wahid and Horn [29] noted a rise from 77% to 80% through gas recirculation. Additional factors, such as hollow fiber membranes, temperature, and pressure, have also been found to affect biomethane content [30,31,32]. In comparison to this study, the FIR exhibited a lower methane content of 64%, whereas the GIR showed methane levels of 75%, which are consistent with those reported in the literature.

In TBF (Three-Phase Bed) reactors, the gas–liquid interfacial area formed on the immobilized bed is maximized, enabling the injected gases to be distributed more efficiently and uniformly. Furthermore, as biofilms rapidly consume available nutrients, high concentration gradients of gases develop across the three-phase system. According to Henry’s law, this phenomenon promotes the transport of gases into the biofilm. Consequently, H₂ mass transfer is passively enhanced without the need for liquid mixing, diffusion devices, or gas recirculation. An additional advantage of this approach is that biomethanation occurs at atmospheric pressure, thereby reducing technical requirements and economic costs [33,34].

In the present study, both bioreactors demonstrated high hydrogen consumption, attributed to the use of immobilized materials. These materials likely enhance hydrogen uptake by lowering gas velocity within the reactor and creating a stable microenvironment that supports microbial activity [2,3]. However, methane yields in the headspace gas—particularly in the FIR—were lower than those reported in other studies. Logroño et al. [8] indicate that the origin and microbial diversity of the inoculum are crucial factors in the biomethanation process, especially in systems that depend on complex microbial consortia.

Another significant aspect of this study was the absence of preconditioning on the inocula. In previous studies that employed preconditioning, inocula were exclusively fed H₂/CO₂ for a defined period, thereby steering the microbial community toward the desired metabolic reactions [8]. In this work, hydrogen consumption was noted in both reactors; however, methane yield was higher in the GIR. This difference was interpreted as stemming from the GIRs more protected structure and its enhanced diffusion rate [9].

In contrast to hydrogen consumption, CO₂ consumption in the granular bioreactor (GIR) was found to be twice as high as in the flocculent inoculum reactor (FIR). This finding suggests that the hydrogen supplied to the GIR was more effectively utilized by hydrogenotrophic microorganisms, resulting in increased biomethane production and a higher biomethane concentration in the headspace compared to the FIR. Statistically significant differences were observed between the GIR and FIR regarding the headspace biomethane percentage, H₂ and CO₂ consumption, and volumetric biomethane production, as analyzed using Tukey HSD post-hoc tests in SPSS software (p = 0.000 < 0.05).

Conversely, the FIR achieved complete hydrogen consumption but exhibited higher CO₂ emissions and lower biomethane yields, indicating that hydrogen was likely utilized by competing microbial populations. A more robust hydrogenotrophic archaeal community was detected in the GIR by the end of the experiment, supported by the detailed microbial community analysis presented in Section 3.2. Additionally, Xu et al. [35] reported that anaerobic granules are more suitable for hydrogenotrophic biomethane production and possess a notably high CO₂ sequestration capacity under hydrogen-fed conditions [36].

3.2. Microbial Community

Microbial community shifts in the FIR and GIR reactors were analyzed before and after the operation (

Table 1. Microbial community shifts in the FIR.

Start-Up of FIR			Finish of FlR
Taxa	Count	%	Taxa	Count	%
Planococcus sp. MB-3u-03	9189	30.53	Planococcus sp. MB-3u-03	6137	22.86
Methanosaeta harundinacea	5534	18.39	Methanosaeta harundinacea	5249	19.55
Methanobrevibacter sp. AbM4	3209	10.66	Methanobrevibacter sp. AbM4	2352	8.76
Methanothrix soehngenii	1922	6.39	Methanothrix soehngenii	1523	5.67
Neisseria gonorrhoeae	1148	3.81	Escherichia coli	1395	5.2
Escherichia coli	1056	3.51	Staphylococcus aureus	1271	4.73
Staphylococcus cohnii	942	3.13	Methanobrevibacter smithii	1117	4.16
Staphylococcus aureus	818	2.72	Neisseria gonorrhoeae	798	2.97
Methanocorpusculum labreanum	715	2.38	Staphylococcus cohnii	773	2.88
Methanobrevibacter smithii	639	2.12	Methanosarcina barkeri	488	1.82

and

Table 2. Microbial community shifts in the GIR.

Start-Up of GIR			Finish of GlR
Taxa	Count	%	Taxa	Count	%
Staphylococcus aureus	33,605	62.98	Methanothrix soehngenii	6337	19.7
Methanothrix soehngenii	4253	7.97	Methanospirillum hungatei	5596	17.4
Methanospirillum hungatei	2146	4.02	Planococcus sp. MB-3u-03	3706	11.52
Methanoculleus bourgensis	2111	3.96	Methanosaeta harundinacea	2808	8.73
Methanosaeta harundinacea	1864	3.49	Staphylococcus aureus	2337	7.27
Methanosarcina thermophila	1285	2.41	Staphylococcus cohnii	1543	4.8
Methanofollis liminatans	1025	1.92	Methanosphaerula palustris	1285	3.99
Methanosphaerula palustris	894	1.68	Methanobrevibacter smithii	943	2.93
Methanobrevibacter smithii	836	1.57	Neisseria gonorrhoeae	715	2.22
Methanolacinia petrolearia	553	1.04	Escherichia coli	657	2.04

). Microbial abundances were normalized using the CPM method for Group 1 and Group 2 to account for differences in sequencing depth (FIR start/FIR end = 0.9851, GIR start/GIR end = 0.9317), and depth ratios were calculated. Tukey’s HSD test for taxonomic diversity showed an increase between FIR Start and FIR End (Shannon: −0.397, p = 0.015; Simpson: −0.0313, p = 0.0157) and between GIR Start and GIR End (Shannon: −1.2020, p < 0.001; Simpson: −0.3001, p < 0.001).

In the FIR, the bacterial population decreased from 76% to 64%, while the archaeal population increased from 24% to 36%. In contrast, the changes in microbial composition were less pronounced in the GIR. The bacterial population in the GIR decreased slightly from 59% to 55%, while the archaeal population increased from 41% to 45% after the operation.

Before the operation in the FIR, the microbial community was primarily composed of Euryarchaeota (45%), Firmicutes (31%), Proteobacteria (11%), and Actinobacteria (9%). After the operation, the proportion of Euryarchaeota increased to 60%, while Firmicutes decreased to 21%. In the GIR, the initial community was dominated by Firmicutes (52%), Euryarchaeota (44%), and Proteobacteria (2%). After the operation, the composition changed to Euryarchaeota (65%), Firmicutes (20%), and Actinobacteria (7%). In both reactors, there was a notable decline in Firmicutes, and Euryarchaeota emerged as the dominant group by the end of the process. It is evident that when environmental conditions—such as substrate feeding; temperature; or pH—are altered; microbial species that are best adapted to these changes gain a competitive advantage. Other species may persist by forming spores or metabolizing residual organic matter from the biomass, in addition to the dominant microorganisms. As a result, even heterotrophic strains can survive in hydrogenotrophic environments when carbon sources are limited to CO₂. Adapted microbial consortia derived from biogas sludge may still contain a diverse range of bacterial and archaeal species, depending on specific environmental conditions [37].

While Methanosarcinales was prevalent in both reactors, Methanobacteriales was more dominant in the FIR, whereas Methanomicrobiales was initially more abundant in the GIR. Following the reactor operations, all three methanogenic orders exhibited increased relative abundances. In the FIR, Methanosarcinales (30%) and Methanobacteriales (29%) were the dominant groups, while in the GIR, Methanomicrobiales (27%) and Methanosarcinales (25%) were predominant. The Methanosarcinaceae family is known for its metabolic versatility, capable of utilizing both acetate and hydrogen. Under elevated hydrogen concentrations, this family tends to shift toward hydrogen utilization [32].

In the FIR system, biomethane production was primarily facilitated by a stable presence of acetoclastic methanogens such as Methanosaeta harundinacea and Methanothrix soehngenii, along with hydrogenotrophic methanogens including Methanobrevibacter sp., Methanobrevibacter smithii, and Methanosarcina barkeri. In the GIR system, biomethane production was primarily facilitated by two functional groups of methanogenic archaea: acetoclastic methanogens such as Methanothrix soehngenii and Methanosaeta harundinacea, which convert acetate into methane, and hydrogenotrophic methanogens like Methanospirillum hungatei, Methanobrevibacter smithii, and Methanoculleus bourgensis, which utilize hydrogen and carbon dioxide.

Although Planococcus sp. MB-3u-03 was observed to be the most dominant species in the baseline and final samples, there was an 8% decrease in FIR species diversity. The Methanosaeta harundinacea, Methanobrevibacter sp. AbM4, and Methanothrix soehngenii species maintained their presence before and after the reactor. Methanobrevibacter sp. AbM4 was a member of Methanobacteriales [38]. In the context of this study, Methanobrevibacter sp. AbM4 likely played an important role in hydrogenotrophic methanogenesis, especially in the FIR system, where hydrogen is fully consumed but the overall biomethane yield remains lower compared to GIR. Members of the genus Methanobrevibacter are well adapted to high hydrogen partial pressure environments and are commonly found in anaerobic systems, including ruminant intestines and biogas reactors. Their ability to efficiently utilize H₂ and CO₂ to produce methane gives them a competitive advantage under these conditions. The dominance of Methanobrevibacter sp. AbM4 in the FIR may indicate that a hydrogenotrophic pathway operates efficiently but is limited by factors such as competition with homoacetogens or suboptimal syntrophic interactions, which may have limited overall methane productivity. Conversely, the structured environment of the GIR may have favored other hydrogenotrophic archaea with higher methane yields. The presence of Methanobrevibacter sp. AbM4 reflects the functional diversity and adaptability of the microbial community in response to sludge structure and reactor conditions [39].

Although the reactors were only fed hydrogen and carbon dioxide, the acetoclastic species Methanothrix soehngenii and Methanosaeta harundinacea remained among the dominant microorganisms. In a mixed culture, the competition among microorganisms for the same substrate can support this dominance. It can be argued that homoacetogens produce acetate when hydrogen is shared between themselves and hydrogenotrophic species [38]. Methanothrix soehngenii and Methanosaeta harundinacea are highly specialized and exhibit a strong affinity for acetate, allowing them to thrive in environments with low acetate concentrations where other methanogens may be outcompeted. Their dominance is frequently linked to stable and efficient methane production, particularly in systems with well-developed granular sludge or biofilms. M. soehngenii plays a crucial role in maintaining the structural integrity of microbial granules, which facilitates syntrophic interactions with fermentative and acetogenic bacteria. Although it does not utilize hydrogen, its presence enhances the function of hydrogenotrophic methanogens by ensuring the complete conversion of volatile fatty acids into biogas, ultimately improving overall process stability and methane yield in anaerobic digestion systems [7,40,41,42].

Methanospirillum hungatei, a hydrogenotrophic species, increased in abundance from 4% at the beginning of the granular anaerobic reactor (GIR) operation to 17% by the end of the process. Its presence was more pronounced in the GIR, correlating with the higher biomethane content and superior hydrogen (H₂) consumption observed in this system. As a syntrophic partner, M. hungatei plays a crucial role in facilitating interspecies hydrogen transfer, which is essential for the degradation of fatty acids and other intermediates under low redox conditions. The structural advantages of granular sludges, such as increased surface area and microbial stratification, likely created a more favorable microenvironment for M. hungatei, enhancing its metabolic activity. In contrast, although hydrogen consumption was complete in the floccular reactor (FIR), the comparatively lower methane yield suggests that M. hungatei and similar hydrogenotrophic populations were either less dominant or less metabolically active, potentially due to competition from homoacetogens or limitations in floc structure. These findings align with previous investigations that emphasize the value of Methanospirillum spp. in optimizing methane production in hydrogen-rich anaerobic environments. Depending on the hydrogen feeding, the population of Methanobrevibacter smithii increased, and Methanosarcina barkeri was also present in the population as a hydrogenotrophic species.

It was found that the abundance of methanogenic archaea in GIR was more conspicuous, especially after the process. The results also indicated that there were more surviving pathogen species in the floc bioreactor. It was found that the granular structure was more protective against these pathogens, and the methane yield was correspondingly high. The presence of pathogenic microorganisms may pose a risk in terms of industrial/environmental implications. As discussed in Section 3.1, it is considered important to perform preconditioning of the inoculum before starting work to prevent such situations. Previous studies have shown that the granular structure better protects the microbial biota. Granules can improve microbial retention capacity [14]. In addition, the granule structure improves the stability and productivity of the natural ecosystem of the sludge and maintains the stability of the process [14]. The granules are well-suited for substrate transport and are highly bioactive due to their internal structure (large pore size, high porosity, and short diffusion distance), which can improve biogas production [23].

Although Planococcus sp. MB-3u-03 was not in the initial GIR, it was found to be one of the predominant species after the bioreactor. While an increase in the populations of Methanosaeta harundinacea, Methanosphaerula palustris, and Methanobrevibacter smithii was observed, it was observed that the bands of Methanosarcina thermophila, which is an acetoclastic species, Methanofollis liminatans, which oxidizes secondary alcohols, and Methanolacinia petrolearia disappeared. In line with the results of this study, Ahring et al. [42] revealed that H₂ inhibited acetate metabolism of Methanosarcina thermophila. Staphylococcus aureus population decreased significantly, and after the reactor, Staphylococcus cohnii was observed among the dominant species in the GIR. Another study showed that total biogas yields from chicken manure were significantly lowered by the addition of S. Aureus which is a pathogenic microorganism [43].

The observed persistence of pathogens such as Enterococcus spp. and Clostridium spp. in floc-based anaerobic systems poses significant environmental and public health concerns. Both Horan et al. [44] and Smith et al. [45] emphasize that mesophilic anaerobic digestion (MAD) may not reliably achieve complete pathogen inactivation, particularly under suboptimal operational conditions such as inadequate mixing, short-circuiting, or insufficient retention time. Horan et al. found that Campylobacter jejuni showed minimal die-off (~0.36 log₁₀) even under compliant MAD conditions, while Salmonella and E. coli required additional treatment to achieve ≥3-log reductions. Similarly, Smith et al. demonstrated that MAD does not directly rely on heat for pathogen removal but instead depends on microbial competition and substrate depletion, making it more sensitive to digester design and operation. Therefore, in systems like floc-based reactors where mixing and contact efficiency may be compromised, the survival of enteric pathogens could result in land and water contamination if digestates are reused without further treatment. The continued presence of pathogenic microorganisms in biosolids poses risks of soil contamination, groundwater pollution, and potential transmission to humans through food chains [44,45]. Post-treatment processes such as UV disinfection or chlorination should be implemented to enhance pathogen inactivation in floc-based systems, ensuring effective reduction of residual pathogens [46].

3.3. Sustainability of the System

The bioconversion of CO₂ into biomethane through biomethanation represents a promising strategy for biogas upgrading from a microbiological standpoint. In this process, microorganisms serve as biological catalysts, facilitating the reduction of carbon dioxide with hydrogen to produce methane [47]. In ex-situ biogas upgrading, CO₂ derived from sources such as biogas or syngas is combined with externally supplied H₂, which is introduced into the liquid phase of a reactor harboring hydrogenotrophic microorganisms [48].

The scale-up studies on industrial biogas upgrading reveal the fundamental aspects that must be addressed for the industrialization of this technology. The findings emphasize that scaling up biomethane production is not merely a matter of reactor sizing but requires a holistic optimization of multiple interrelated factors [49]. In pilot-scale applications, the comparison of parallel and series operations demonstrated that methane purity exceeding 97% can be achieved in series mode. Moreover, the system’s ability to rapidly recover its performance after interruptions in hydrogen supply highlights a significant advantage for integration with fluctuating renewable energy systems [50]. In the transition from laboratory to pilot scale, reactor engineering parameters such as hydrogen dissolution, gas residence time (GRT), and gas–liquid mass transfer are emphasized as critical factors in process design [51]. In this context, the present study demonstrated that high H₂ consumption can be achieved without the need for mixing, elevated pressure, or temperature adjustments, owing to the contribution of immobilized materials. These results support the industrial applicability of biomethane production and underscore the critical optimization domains for scale-up, including nutrient management, gas distribution, and dynamic operation [51].

From an environmental perspective, Life Cycle Assessment (LCA) has emerged as a valuable tool for evaluating the environmental impacts of biomethane production and identifying strategies to improve its sustainability. LCA studies have compared the environmental footprint of upgrading biogas to biomethane with that of using biogas directly [52].

One of the most critical challenges in biomethane production lies in sourcing hydrogen, which is predominantly generated via electrolysis—a process that requires substantial energy input to supply the hydrogen needed by methanogenic archaea. Liu et al. [53] evaluated the energy balance of H₂ electrolysis, reporting an energy demand of 622 MJ for hydrogen generation, compared to an energy content of only 14 MJ in the resulting methane, yielding a net energy balance of −608 MJ. Elyasi et al. [54] compared biological upgrading with water scrubbing and modeled ex-situ biological methanation in a trickle-bed filter reactor (TBFR) operating at 55 °C, producing an outlet gas containing 98.5% CH₄. In their scenario, hydrogen was produced in an electrolyzer powered by surplus wind energy. The results demonstrated that biological upgrading outperformed water scrubbing in three environmental impact categories—Human Health; Climate Change; and Resource Use.

Vo et al. [55] assessed three upgrading configurations: amine scrubbing, amine scrubbing followed by ex situ biological methanation, and ex situ biological methanation alone. The system boundaries encompassed silage production, transportation, biogas generation, and upgrading via amine scrubbing, electrolysis, and biological methanation. Across all configurations, electrolysis emerged as the dominant contributor to environmental impacts, accounting for 80% of GHG emissions, 86% of acidification, 70% of ozone depletion, and 85% of particulate matter formation, except in the case of eutrophication potential.

In a separate economic analysis, Vo et al. [56] estimated the minimum selling price (MSP) of renewable methane at €0.76/m³ for amine scrubbing, €1.50/m³ for amine scrubbing plus biological methanation, and €1.43/m³ for biological methanation alone, assuming an electricity price of €0.10/kWh for hydrogen production and a grass silage cost of €27/t. The study further highlighted that electricity price strongly influences the cost of renewable methane in the latter two scenarios and concluded that direct biogas injection into the methanation reactor is economically preferable to CO₂ capture from biogas prior to methanation.

Considering all the results of this study, it can be concluded that biological methanation systems are suitable for use with renewable energy applications within the Power-to-Gas framework.

4. Conclusions

This study examined the impact of gas substrate feeding on biomethane production and microbial dynamics in biomethane reactors using different microbial sludges. The results revealed that both granular and flocculate inoculum consumed hydrogen similarly. However, the granular sludge reactor produced biogas with higher methane content. Regarding microbial population dynamics, the granular sludge exhibited less microbial shift, likely due to the protective structure of the granules.

The results of this investigation can help engineers and operators of biogas plants create indicators to evaluate the potential of different substrates for the capture and conversion of CO₂ into biomethane, a renewable energy source that is presently undergoing pilot testing. Capturing CO₂ and transforming it into valuable products addresses a critical environmental challenge and could significantly contribute to climate change mitigation. For future industrial applications, conducting larger-scale and longer-term studies is recommended.

The findings indicate the potential for granular sludge reactors to enhance methane yield and improve microbial stability, which are critical for the efficient and sustainable operation of biomethanation systems. The greater resilience of granular sludge to microbial fluctuations indicates its suitability as a promising approach for industrial-scale applications. Additionally, optimizing gas substrate feeding strategies based on these insights could support the shift toward cleaner energy production and contribute to broader sustainability objectives in renewable bioenergy.