Insights into the Acute Stress of Glutaraldehyde Disinfectant on Short-Term Wet Anaerobic Digestion System of Pig Manure: Dose Response, Performance Variation, and Microbial Community Structure

Abstract

The outbreak of epidemics such as African swine fever has intensified the use of disinfectants in pig farms, resulting in an increasing residual concentration of disinfectants in environmental media; however, the high-frequency excessive use of disinfectants that damage pig farm manure anaerobic fermentation systems and their mechanisms has not attracted enough attention. Especially, the complex effects of residual disinfectants on anaerobic fermentation systems for pig manure remain poorly understood, thus impeding the application of disinfectants in practical anaerobic fermentation systems. Herein, we explored the effects of glutaraldehyde disinfectant on methane production, effluent physicochemical indices, and microbial communities in a fully automated methanogenic potential test system (AMPTSII). The results show that adding glutaraldehyde led to remarkable alterations in methane production, chemical oxygen demand (COD), volatile solids (VS), and polysaccharide and phosphorus concentrations. During the anaerobic process, the production of methane displayed a notable decrease of 5.0–98% in all glutaraldehyde treatments, and the trend was especially apparent for treatments containing high levels of glutaraldehyde. Comparisons of the effluent quality showed that in the presence of 0.002–0.04% glutaraldehyde, the COD and total phosphorus (TP) increased by 12–310% and 15–27%, respectively. Moreover, the addition of 0.01–0.08% glutaraldehyde decreased the ammonium (NH₄⁺-N) concentration and VS degradation rate by 7.7–15% and 4.9–26.2%. Furthermore, microbiological analysis showed that the glutaraldehyde treatments had adverse effects on the microbial community. Notably, certain functional bacteria were restrained, as highlighted by the decreases in relative abundance and microbial diversity by 1.3–17% and 0.06–21%, respectively. This study provides a theoretical basis for the rational use of disinfectants in anaerobic fermentation systems.

1. Introduction

In recent years, China’s pig farming industry has undergone intensive development to ensure domestic food security [1], which has led to the production of nearly 900 million tons of pig manure annually [2,3]. In the pig farms investigated, 52.89% of them cleaned pig manure by the dry cleaning method; the proportion of soaked manure farms was 29.75%. The proportion of cleaning pig manure and urine with a scraper was 8.26%. At present, most large-scale pig farms discharge sewage directly into farmland and river channels after precipitation and filtration, accounting for nearly 55%. After sewage treatment, only 18.26% is used for aquaculture; this is the random discharge of pig manure waste resources, which contains valuable organic matter and nutrients and also causes serious environmental pollution, such as heavy metal, ammonium (NH₄⁺-N), and chemical oxygen demand (COD) pollution [4]. Therefore, identifying effective methods for treating pig farm waste and promoting greening practices is necessary for ensuring sustainability in the pig farming industry [2,5,6].

Anaerobic fermentation has been identified as a potential green technology for the treatment of manure waste and agricultural residues and the production of biogas via a series of anaerobic bioreactions under completely enclosed anaerobic conditions [7,8]. However, poor methane yield and system instability represent ubiquitous problems during anaerobic fermentation, thus challenging the large-scale application of this technology [9,10]. Inhibitors of fermentation represent the major factor leading to issues with anaerobic reaction units owing to their high concentrations in fermentation systems [9]. For example, inputs of excessive ammonia, sulfide, metals, and organics break the balance between acid-forming microorganisms and methane-forming microorganisms, leading to chaos in the steady-state rate of methane yield and organic acid accumulation [11,12,13,14].

Disinfectants have evolved into a typical inhibitor of anaerobic fermentation methanogenesis. Recently, disinfectants have been widely used to block disease infection sources, as observed during the pandemics of African pig fever (ASF) and Coronavirus disease 2019 (COVID-19) [15,16,17]. In particular, external disinfectants are used on pig farms to eliminate the diffusion of viruses and pathogens during operational processes. Current disinfectants commonly used on pig farms include acids, bases, aldehydes, phenols, and quaternary ammonium salts [18,19,20]. However, in the sterilization process of pig farms, the residual disinfectant will enter the subsequent wastewater treatment system along with pig manure and urine and daily pig farm cleaning, thus adversely affecting the non-aerobic system and posing a threat to the surrounding environment [17,21]. Disinfectants were applied to perform AD in some studies; for example, Jiang et al. found that the residual disinfectants in piggery slurry inhibit the activity of microorganisms in AD, resulting in a decrease in biogas production, and even leading to the failure of anaerobic systems [22]. Wang et al. indicated that triclocarban (TCC) of 600 mg/kgTSS had an adverse impact on the methane production of AD from activated sludge, which was decreased by 13.8% than the control [23].

The broad-spectrum antibacterial properties of disinfectants enable them to inactivate pathogenic microorganisms while inhibiting or killing other non-pathogenic microorganisms, which adversely affects the main microbial function of sewage treatment systems, resulting in poor effluent quality and non-compliance with discharge standards [24].

High levels of disinfectants have a strong inhibitory effect on the methanation process, leading to significant reductions in the methane yield of 69.98% and 100% in the presence of 1.6 mL/L and 2.4 mL/L phenol disinfectant, respectively [25]. Disinfectants also contribute to the accumulation of volatile fatty acids and sludge solubilization [26,27]. These different effects have been attributed to the susceptibility of microbial activity to disinfectants. In particular, the growth of Methanosaeta and Methanobacterium is inhibited by the addition of high levels of disinfectants. Furthermore, disinfectants such as chlorine have a negative impact on nitrogen/COD removal depending on the disinfectant dose, which is attributed to the reduced activity of functional microorganisms [28]. Previous studies in different environments have also revealed that microorganisms can generate drug resistance under the selective pressure of disinfectants, resulting in the overgrowth of resistant microorganisms [28,29,30]. Disinfectant resistance genes, such as qacA, qacB, qacC, merb, and bcrABC, have been identified with the addition of quaternary ammonium salts and chlorine [29,31]. Furthermore, the enrichment of pathogenic dominant bacteria inhibits the growth of methanogenic and functional bacteria, such as ammonia-oxidizing bacteria, thereby hindering methane production and water purification efficiency.

Glutaraldehyde is an efficient and cost-effective disinfectant that can change the secondary structure of proteins, and it is widely used in the farming industry for disinfection [32]. However, the effects of glutaraldehyde on anaerobic fermentation, including its effects on effluent quality and microbial communities, remain unclear. Therefore, the effects of glutaraldehyde on methane production, microbial communities, and physicochemical parameters must be systematically and completely investigated. Herein, we report the multiscale effects of glutaraldehyde on anaerobic fermentation, including methane production, COD, NH₄⁺-N, total phosphorus (TP), TS, VS, polysaccharides, microbial relative abundance, microbial diversity, and genomic expression. This study provides a new reference for guiding the rational use of disinfectants on pig farms during epidemics.

2. Materials and Methods

2.1. Materials and Equipment

Anaerobically digested sludge with high activity was obtained from an anaerobic fermentation tank, and fresh pig manure was obtained from the normal excreta of fattening pigs at a large-scale pig farm in Nanchang, China. Glutaraldehyde solution (Shanghai Aladdin Bio-chemical Technology Co. Ltd., Shanghai, China) used as disinfectant in this experiment. A Multi-Parameter Water Quality Tester (5B-6C (V12), Lanzhou Lianhua Science and Technology Co. Ltd., Lanzhou, China) for testing COD, NH₄⁺-N, TP and TN. The pH was determined using a multi-combination meter (DZS-706, INESA Scientific Instrument Co. Ltd., Shanghai, China). A fully automatic methane potential testing system (AMPTS II, model XT5018GP) was purchased from BPC Instruments AB (Lund, Sweden). Basic physicochemical properties of raw materials, such as TS, VS, and pH values are shown in

Table 1. Basic physicochemical properties of the raw materials.

Materials	TS/%	VS/%	pH
Pig manure	27.57 ± 0.23	20.91 ± 0.19	7.32
sludge	2.51 ± 0.18	1.52 ± 0.16	7.65

.

2.2. Establishment and Operation of the Stimulated Anaerobic Fermentation Reactors

The fermentation test was performed using a fully Automated Methane Potential Testing System (AMPTS II, Bioprocess Control Co. Ltd., Lund, Sweden, Figure 1). Six groups of methanogenesis experiments by anaerobic fermentation were conducted in 500 mL glass containers containing 285 g of digested sludge. All experiments were performed in triplicate. The content of pig manure was set to 20 g/L (volatile solids (VS) ratio 2:1 mixed). A certain amount of ultrapure water was added to achieve a total solid content (TS) of 2% for the inoculated sludge (domesticated anaerobic sludge, 0.0152 g VS/mL) and pig manure (fresh pig manure without disinfectant, 0.2091 g VS/mL), and the mass concentration of the glutaraldehyde disinfectant was set to five gradients: 0% (CK group), 0.002% (C1 group), 0.01% (C2 group), 0.04% (C3 group), and 0.08% (C4 group). The digested sludge and pig manure were accurately weighed and then diluted to 500 mL with deionized water. Varying doses of glutaraldehyde were then added to the containers to obtain the appropriate concentrations. All glass reactors were incubated in a fully automated methane potential testing system at 37 °C for 15 days. During the incubation process, all reactors were agitated and stopped every 10 min. As a comparison, a blank incubation experiment (BK group) with the addition of digested sludge and deionized water was conducted under the same conditions.

Mixed solutions (10 mL aliquots) were withdrawn from each reactor on the first and last days of anaerobic fermentation. All suspensions were centrifuged at 4 °C and 8000 rpm for 5 min, and the supernatant was stored at 4 °C for the determination of physicochemical indices. The corresponding precipitates were then refrigerated at −80 °C to assess the microbial diversity.

2.3. Determination of Methane and Physicochemical Indices

Methane produced during anaerobic fermentation was collected using the AMPTS II device, which was equipped with a temperature-controlled fermentation unit, a CO₂ absorber bottle containing NaOH solution and a litmus indicator, and an array of 15 microgas flow meters with a resolution of 10 mL. The volume of daily biogas produced was automatically collected in batches throughout the anaerobic fermentation cycle, and the data were automatically corrected using the built-in temperature and pressure sensors to obtain accurate gas flow results.

The quasi-primary kinetic equation was used to analyze the kinetic characteristics of biochemical methane production during the entire anaerobic digestion process [7,33]:

Y(t) = a∗(1 − exp(−b∗t))

where Y(t) = Cumulative biochemical methane production (mLCH₄⋅gVS⁻¹) at time t (day).

The physicochemical indices of anaerobic fermentation, including COD, total nitrogen (TN), TP, ammonium (NH₄⁺-N), pH, and polysaccharides were measured by UV–vis spectroscopy (SP-1920, Shanghai Spectral Instrument Co., Shanghai, China) according to standard methods (HJ828-2017 [Ministry of Ecology and Environment of the People’s Republic of China: Beijing, China, 2017], HJ636-2012 [Ministry of Ecology and Environment of the People’s Republic of China: Beijing, China, 2012], GB11893-89 [State Administration for Market Regulation: Beijing, China, 1989], HJ535-2009 [Ministry of Ecology and Environment of the People’s Republic of China: Beijing, China, 2009], GB6920-86 [Ministry of Ecology and Environment of the People’s Republic of China: Beijing, China, 1986], and GB/T35818-2018 [State Administration for Market Regulation: Beijing, China, 2018]). TS and VS were measured by electronic balance (BSA124S-CW, Sartorius, Germany) according to the standard method (GB11901-89 [State Administration for Market Regulation: Beijing, China, 1989]).

2.4. Microbial Community Analysis

The microbial community of the anaerobic sludge in the six fermentation systems was evaluated at the beginning and end of the reaction. DNA was extracted using an E.Z.N.A. Soil DNA Spin gene Kit (Omega Bio-tek, Norcross, GA, USA). The extracted DNA products were run on a 1.0% agarose gel, and the quantity and purity were determined using a NanoDrop ND-2000 UV–VIS spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). 16S rRNA (V3–V4 regions) was used as the bacteria-specific fragment with primers 338F and 806R [34]. Sequencing libraries were generated using a NEXTFLEX Rapid^® DNA-Seq Kit (Bluescape Scientific Co. Ltd., Beijing, China). The library was sequenced on an Illumina MiSeq PE30 (Illumina, Inc., California, USA) platform. Quantitative Insights into Microbial Ecology (QIIME) was used to quantify and analyze the raw sequences, which were clustered into operational taxonomic units (OTUs) at the 97% similarity level.

Alpha diversity indices, including Shannon, Simpson, Chao1, Ace, and the observed species, were calculated based on the OTUs. The Ace and Chao indices were used to estimate the microbial species abundance, and the Shannon and Simpson indices were used to reflect the microbial diversity [35,36].

It was decided to assess the disturbance of bacterial communities in anaerobic systems by means of a violin diagram. The violin plot is an evolution of the box plot that reports the density distribution of observed values for a univariate distribution in the form of a kernel density plot. A map of the kernel density of the data is reported symmetrically by both sides of the distribution, which explains the typical shape of the graph, hence, the name [37].

The Kyoto Encyclopedia of Genes and Genomes (KEGG) orthology results predicted by PICRUSt2 were compared with each immediate homologous gene in the KEGG database to obtain the genes related to the predicted functions of the samples during the anaerobic fermentation experiments.

2.5. Data Analysis

Graphic plotting and statistical analysis to determine significant differences were performed using Origin 8.5 (OriginLab Corporation, Northampton, MA, USA) and GraphPad Prism 9 software (GraphPad Software Inc., San Diego, CA, USA). Microbial data were analyzed using the Majorbio Cloud platform (Shanghai Majorbio Technology Co., Ltd., Shanghai, China) [38].

For the difference analysis, we used the Kruskal–Wallis H test, which is a method to generalize the Wilcox rank sum test of two independent samples to the non-parametric test of multiple independent samples (greater than or equal to 3). The analysis allows for the analysis of significant differences in species across multiple groups of samples. In the P-value test, the Fdr method is used: First, the m P-values are sorted in order from small to large, starting from p₍₁₎ to p₍₂₎, p₍₃₎... Then, compare them one by one until you find the largest p_(i) satisfaction:

$${\mathrm{p}}_{(i)}\le {\displaystyle \frac{i}{m}}\alpha$$

Once found, reject all previous null hypotheses H(i), i = 1,2,3... i.

At this point, the correction of FDR is complete. Alternatively, leaving α unchanged, correct the p-value to mp_(i)/i, which is also called the Q value: q-value (i) = m × p_(i)/i < α, the confidence interval is 95%

3. Results

3.1. Effects of Glutaraldehyde on Effluent Properties in the Anaerobic Fermentation System

Comparisons of COD concentrations in different fermentation systems showed that glutaraldehyde led to abnormal changes in COD removal during anaerobic fermentation depending on glutaraldehyde dose (Figure 2a). For example, 0.002 g/L glutaraldehyde reduced the COD content in the solutions by 16% at the pre-reaction period compared with the CK group, whereas greater amounts of glutaraldehyde led to increases in the COD content. The high-level glutaraldehyde treatments (0.01–0.08%) displayed increasing COD values of 670–1003 mg/L in the initial reaction phase, with values that were 0.045 to 0.64-fold higher than that of CK. This might be attributed to the increase in COD-degrading microorganisms with low-dose glutaraldehyde treatments and the decrease in these microorganisms with high-dose glutaraldehyde treatments in the early reaction stage. At the post-reaction period, the concentration of COD increased rapidly from 507 mg/L to 2075 mg/L with increasing glutaraldehyde doses from 0 to 0.04%, and it subsequently decreased to 1282 mg/L with a further increase in glutaraldehyde dose to 0.08%. The post-reaction results showed that the COD concentrations of CK decreased by 24%, respectively, compared with that of the pre-reaction period, indicating a normal anaerobic process. This demonstrated that the inoculated anaerobic sludge maintained stable activity in the anaerobic sludge in the absence of glutaraldehyde. However, COD showed obvious increases in the presence of glutaraldehyde, except in the 0.01% treatment, further revealing the significant inhibitory effects of increased exposure to high-dose glutaraldehyde. These results suggest that the glutaraldehyde treatments seriously inhibited the normal operation of the anaerobic fermentation system based on the glutaraldehyde dose and reaction time.

The concentrations of TN and NH₄⁺-N increased gradually with reaction time during normal anaerobic fermentation (Figure 2c,d). In the pre-reaction period, the concentrations of NH₄⁺-N for the C1 and C2 groups decreased by 9.8% and 7.3%, respectively, in the C3 group, there was a notable increase in the value, amounting to 157.67 mg/L (p < 0.05). This difference may be attributed to the opposite response mechanisms of ammonification microorganisms to treatments with low and high doses of disinfectants. The inhibition of NH₄⁺-N production was highlighted by the post-reaction results. Increasing the glutaraldehyde dose to 0.01–0.08% resulted in a 14–71% decrease in NH₄⁺-N concentration at the post-reaction period. However, the effects of the glutaraldehyde treatments on TN release were not significant (p > 0.05).

The addition of glutaraldehyde led to notable differences in residual phosphorus in the pre-reaction and post-reaction periods (Figure 2b). The aqueous concentrations of phosphorus ranged from 18 to 36 mg/L with the addition of glutaraldehyde in the initial reaction period, and these values were 35–51% lower than that of the control treatment. This indicated that the decomposition of pig manure decreased with the addition of glutaraldehyde. However, the concentration of phosphorus increased at the post-reaction period by 15–27% after treatment with 0.002–0.04% glutaraldehyde, suggesting that the addition of glutaraldehyde resulted in an excess release of phosphorus or a decrease in phosphorus-degrading capability. This finding was supported by the evident increase in phosphorus levels by 71–150% in the post-reaction period after the addition of glutaraldehyde relative to the pre-reaction period.

An abnormal release of polysaccharides was observed in the different fermentation systems with glutaraldehyde addition, and it was associated with variations in COD removal. Polysaccharide concentrations of the fermentation liquor in the C1 and C2 groups decreased by 31% and 11%, respectively, and subsequently increased by 69–104% with further increases in the glutaraldehyde dose to 0.04–0.08% (Figure 2e). Furthermore, the post-reaction results showed that the addition of glutaraldehyde resulted in a notable increase in polysaccharides in the suspensions. In particular, the aqueous concentration of polysaccharide ranged from 39 to 71 mg/L for the groups treated with 0.01–0.08% glutaraldehyde, and these values were 1.0- to 2.7-fold higher than those without glutaraldehyde addition.

The effects of different glutaraldehyde treatments on the aqueous pH during the anaerobic process are shown in Figure 2f. The differences between the C1 and C2 groups were statistically insignificant (p > 0.05). However, the pH values in the reaction systems treated with 0.04% and 0.08% glutaraldehyde dropped slightly in the post-reaction period from 7.1 to 6.2 and 6.9 (p < 0.05), respectively. This increased acidity of the aqueous solution was mostly related to the hydrolysis of glutaraldehyde and the accumulation of volatile fatty acids (VFAs) [39,40].

VS and TS are the key indicators in anaerobic fermentation systems. VS represents the content of available organic matter in the substrate, and its degradation rate demonstrates the degree of microbial transformation and utilization of the substrate in an anaerobic fermentation system. As shown in Figure 2g,h, during the post-reaction period, the contents of VS and TS in each group increased linearly with the increase in glutaraldehyde dosage, in which the degradation rates of VS were 11.5% and 5.4% (p < 0.05), and the degradation rates of TS were 5.65% and 3.25% (p < 0.001) in the C3 and C4 groups, respectively.

3.2. Effects of Glutaraldehyde on Methane Production in Anaerobic Fermentation System

The addition of glutaraldehyde resulted in an apparent decline in methane yield throughout the anaerobic fermentation process. In particular, the cumulative methane production in the anaerobic reactors was negatively correlated with the glutaraldehyde dose (r = −0.92, p = 0.027) (Figure 3). The daily gas production in the Blank group initially increased and subsequently decreased, with a cumulative methane production of 85.10 ± 3.99 mL. This indicates that the domesticated anaerobic sludge exhibited high activity. After the addition of pig manure as a substrate for anaerobic digestion, methane production increased to 128 mL/d. Strikingly, the daily methane production in the anaerobic reactor increased continuously for the C1 group and plateaued at 131 mL within the first three days of the reaction. Methane production in the control treatment ceased on the tenth day, and the accumulated amount reached 569 mL. Correspondingly, methane production in the system treated with 0.002% glutaraldehyde ceased on the 12th day, and the cumulative methane production reached 599 mL, showing a 5.3% enhancement relative to the control system. However, for the anaerobic reactors treated with 0.01–0.08% glutaraldehyde, the peak methane production values and cumulative methane amounts were only 12–47 mL/d and 11–514 mL, respectively, which were 63–81% and 9.7–98% lower than those of the control treatment, respectively. This suggests that glutaraldehyde inhibits methane production during anaerobic fermentation in a dose-dependent manner. This was further verified using a reaction kinetics equation.

The data on reaction time and CH₄ production rate were subjected to analysis, which revealed that the entire anaerobic digestion gas production process, with the exception of group C4 (CH₄ production for only 1 d), is consistent with the quasi-primary kinetic equation. The following characteristics were identified as being typical of this process the BK, CK, C1, and C2 groups were G(t) = 84.123∗ (1 − exp)^{(−0.466 t)}, G(t) = 650 ∗ (1 − exp^(−0.245t)), G(t) = 644.554∗ (1 − exp^(−0.266t)), and G(t) = 160.78∗ (1 − exp)^(−0.065t), respectively, and the correlation coefficients (R²) were 0.99108, 0.98576, 0.9853, and 0.9136, respectively. Equilibrium methane production slightly increased from 640 mL to 644 mL with increasing glutaraldehyde levels from 0 to 0.002%, although it subsequently dropped drastically to 34 mL with a further increase in glutaraldehyde level to 0.04%.

3.3. Effect of Glutaraldehyde on Microbial Richness and Diversity

A total of 1,636,945 high-quality sequences were obtained via high-throughput sequencing of 36 samples, and 37,904 to 55,302 sequences were obtained per sample, with an average of 45,470 sequences. All samples contained 51 phyla, 99 classes, 228 orders, 394 families, 773 genera, 1371 species, and 2636 OTUs.

Microbial community analysis demonstrated that glutaraldehyde significantly affected microbial richness and diversity. In the initial reaction phase, 0.002% glutaraldehyde led to 5.3% and 6.7% increases in the Ace and Chao indices, respectively (Figure 4a,b). This suggests that the low-dose glutaraldehyde treatment slightly enhanced the microbial richness and diversity, which is conducive to the smooth operation of anaerobic fermentation. However, the Ace and Chao indices of anaerobic sludge in the presence of 0.01–0.08% glutaraldehyde ranged from 374.5 to 498.1, which was lower than the values for the control treatment (~510). This suggests that the high-dose glutaraldehyde treatment impeded microbial richness and diversity, which is unfavorable for methane production during anaerobic fermentation. This negative impact was further demonstrated by the variations in the microbial community during the post-reaction period. The Ace and Chao indices of bacteria on day 15 significantly differed among the samples treated with 0.04–0.08% glutaraldehyde and the control (p < 0.05). Most glutaraldehyde treatments led to values of 366.5–447 for the Ace and Chao indices, which were low compared with that of the control group (453) after fermentation. Regarding species diversity (Figure 4c,d), the Shannon index of the anaerobic systems treated by glutaraldehyde was respectively 2.0–2.6 and 2.7–3.4 lower in the pre- and post-reaction periods compared with the control system. Correspondingly, the Simpson index of the anaerobic sludge in the presence of glutaraldehyde was 0.081–0.19, showing an evident increase relative to the control treatment (0.07). These results suggest that glutaraldehyde adversely affects microbial diversity during anaerobic fermentation. Notably, the Ace and Chao indices in the system treated with 0.01% glutaraldehyde reached high values of 536 and 556, respectively, which might be attributable to the growth-promoting effect of low-dose glutaraldehyde.

As can be seen in the plot of intergroup differences in colony disorder indices (Figure 4e,f), the colony became disordered with the addition of glutaraldehyde. The degree of disturbance was positively correlated with glutaraldehyde concentration, ranging from −0.44 (CK) to 1.15 (C4). However, 0.01% glutaraldehyde treatment led to colony disorder indices declining from 0.75 at the beginning to 0.42 at the end of the reaction. It suggests that the microbial community in the 0.01% glutaraldehyde treatment not only developed a certain degree of resistance to the disinfectant as the reaction progressed, but also adapted to the environment and carried out a certain degree of vital activities.

Notable changes in the composition of the microbial flora and corresponding dominant flora were observed in the different experimental groups based on microbial community analysis. At the phylum level, the bacteria with high abundance mainly belonged to Firmicutes, Bacteroidota, Proteobacteria, and Chloroflexi, among which Firmicutes was the first dominant phylum during the anaerobic fermentation process (Figure 5a). The relative abundance of Firmicutes increased logarithmically as the dose of glutaraldehyde increased at the pre-reaction period (R² = 0.93), accounting for 75–96% of the total bacteria. However, the relative abundances of Bacteroidota, Proteobacteria, and Chloroflexi decreased dramatically as the dose of glutaraldehyde increased. Among them, the most significant downward trend was observed for Chloroflexi, with a typical decrease in relative abundance from 5.3% in the control treatment to 0.25% in the 0.04% glutaraldehyde treatment at the pre-reaction period of anaerobic fermentation (Figure 5d). Compared with the pre-reaction period, the relative abundance of Firmicutes in the post-reaction period showed a decreasing trend, although it still ranked the highest among all microflora. In contrast, the relative abundances of Proteobacteria, Bacteroidota, and Chloroflexi increased during the post-reaction period, with Bacteroidota showing an obvious increase from 1.4% to 29% in the reactor with 0.04% glutaraldehyde. The relative abundances of other key microflora, such as Thermotogae and Patescibacteria, increased to varying degrees with prolonged anaerobic digestion time. However, this increasing trend was not observed at high doses of glutaraldehyde (0.04% and 0.08%), with the relative abundance decreasing to zero at the post-reaction period. This suggests that functional microorganisms disappeared as a result of the stress caused by high doses of glutaraldehyde.

Significant alterations in the composition of microorganisms were also observed in the lower-level classifications. At the genus level (Figure 5b), the dominant microbial flora were Clostridium_sensu_stricto_1, Trichococcus, Terrisporobacter, Turicibacter, and Romboutsias. The relative abundance of Clostridium_sensu_stricto_1 increased logarithmically with increasing glutaraldehyde doses at the beginning of the reaction (R² = 0.85), accounting for 10–37% of the total bacteria. The relative abundance of Trichococcus was enhanced slightly from 42% to 52% with increasing glutaraldehyde levels from 0 to 0.002%, followed by a sharp decline to 4.5% with a further increase in glutaraldehyde levels to 0.08%. Compared with the pre-reaction period, the relative abundance of the dominant microbial flora in the post-reaction period decreased by 9–18% with 0.04–0.08% glutaraldehyde, suggesting that high doses of glutaraldehyde resulted in the attenuation of dominant microbial flora. Notably, some rare bacterial genera showed significant increases in abundance following the addition of glutaraldehyde. Typically, Bacteroides shifted to the second most abundant species after treatment with 0.04% glutaraldehyde, with a relative abundance of 25% at the post-reaction period (Figure 5f). Bacteroides ferment a wide range of sugars to produce acids and gases that are the main contributors to VFAs in anaerobic digestion systems [41], a phenomenon that inevitably leads to changes in pH, as evidenced by the results in Figure 2f. Syntrophomonas, a well-known genus of hydrogen-acetic acid-producing bacteria [42], underwent a significant reduction in abundance with the addition of glutaraldehyde, with relative abundance decreased by 77.4–99.9% with the addition of 0.01–0.08% glutaraldehyde (Figure 5e). This also elucidates the reduction in methane gas production.

The effects of the microbial community on the main physicochemical indices were assessed through a correlation analysis based on the relative microbial abundance at the genus level (Figure 6). The top three dominant genera in terms of abundance in the samples were analyzed for redundancy with respect to environmental factors and sample groups (Figure 6a), including Clostridium_sensu_stricto_1, Trichococcus, and Terrisporobacter, in which Trichococcus showed a highly significant negative correlation (r = −0.63, p < 0.001) with COD, while the remaining genera showed the opposite trend, with Clostridium_sensu_stricto_1 and Terrisporobacter having the greatest effect (r = 0.38 and r = 0.40, p < 0.05). For the physicochemical indicators, half of the colonies showed a significant positive correlation with the indicators (r = 0.44–0.75, p < 0.05), while the remaining colonies were highly significantly negatively correlated with COD conversion (r = −0.90–0.63, p < 0.001). The colonies that had highly significantly different correlations in terms of TN and NH₄⁺-N transformation showed similar effects. With regard to TP, the number of significant positive or negative correlations exhibited was comparable. It is evident that the area in question boasts a variety of distinctive flora; for example, Turicibacter did not significantly affect the COD, NH₄⁺-N, TN, and TP physicochemical indices and Trichococcus was only highly significantly negatively correlated with COD (r = −0.49, p < 0.05).

Functional changes in the microbial community under exposure to different glutaraldehyde concentrations were further determined (Figure 7). The results showed that the genes related to transporter proteins and signaling (ABC series) had the highest percentage of information, with a total percentage of more than 60%, indicating that the cells maintain the basic physiological characteristics [43]. Among the 50 identified gene fragments, a number require further investigation. For example, K02760 (celA, chbB) and K02761 (celB, chbC) are gene function fragments corresponding to carbohydrate (starch and sucrose) metabolism. In the pre-reaction period, as the glutaraldehyde concentration increased, these gene fragments were significantly inhibited in the C2, C3, and C4 groups compared to the CK group (p < 0.05). This further explains the trend of rising polysaccharides in the pre-reaction period in Figure 2e. Moreover, we observed that in the post-reaction period, a significant difference was observed in the C2 group compared to the remaining groups (p < 0.05), and this phenomenon corresponded to the abnormal changes in polysaccharide concentration in the C2 group in Figure 2e. In addition, K03657 (uvrD, pcrA) and K03406 (mcp) are functional fragments related to cell repair and cell chemotaxis, and they were significantly changed in the C2, C3, and C4 groups compared with the CK group (p < 0.05). This change was attributed to the migration of bacteria in response to changes in the glutaraldehyde concentration gradient, which was directional. In combination with the changes in the drug resistance genome in K01448 (amiABC), it could be predicted that the addition of glutaraldehyde resulted in a certain degree of cellular damage. Furthermore, the higher the concentration, the greater the degree of damage, and the greater the changes in the expression of cellular resistance genes [44].

At the genus level, Venn diagrams were used to count the species of microorganisms co-occurring in the 0.01–0.08% glutaraldehyde treatment prior to the reaction, and the results are shown in Figure 7C. The top five dominant genera in terms of abundance are discussed separately: Marinospirillum has the highest relative abundance (43.26%), and relevant literature indicates that there is an obvious correlation between its abundance change and gas production, with higher abundance producing worse gas [45]. Acholeplasma accounts for 8.65% and has been reported to utilize oligopeptides and amino acids released from dead cells as an energy source [46]. Myroides and Aerococcus, on the other hand, are typical drug-resistant bacteria [47]. Noting that the K02760 (celA, chbB) and K02761 (celB, chbC) genes were more strongly expressed in the C2 group than in the other groups after the reaction, the exclusive strains were analyzed using Venn diagrams, and it was found that the dominant strains of Bifidobacterium, Sedimentobacterium, and Pseudoalteromonas were all global and an overview of secondary metabolic pathways, carbohydrate metabolism, amino acid metabolism, and energy metabolism key microorganisms in the mapping.

4. Discussion

Increasing doses of glutaraldehyde adversely affected the anaerobic fermentation process. Compared to other commonly used disinfectants at corresponding doses, Xu [25] found that cumulative methane production decreased by 35.72% and 96.86% when phenol concentration was increased from 0.004% to 0.02% and 0.04%, respectively, which was attributed to the fact that phenol mainly inhibited the activities of acetate kinase and coenzyme F₄₂₀ in the methanation stage. Similarly, Wang et al. [48] found that the maximum yield of methane production decreased from 157.5 mL/g VSS to 135.7 mL/g VSS when the dose of TCC exposure was increased from 0 to 0.6%, and it also pointed out that the presence of TCC promotes the release of organic matter from the sludge, which alters the functional group structure and surface morphology of the sludge, and furthermore, the presence of TCC leads to acidification and significantly inhibited methanogenesis. In contrast, the disinfection mechanism of glutaraldehyde mainly relies on the alkylation of two reactive aldehyde groups, which act directly or indirectly on different types of biological protein molecules, disrupting the secondary structure of proteins and thus reducing the carbon removal efficiency of the effluent [49]. As described in Section 3.1, both 0.04 and 0.08% glutaraldehyde treatments led to a significant increase in COD concentration throughout the anaerobic fermentation process, while 0.08% glutaraldehyde completely inhibited microbial activity. This phenomenon can be confirmed not only from the methane gas production data but also from the changes in VS. When 0.04–0.08% glutaraldehyde was present in the anaerobic reactor, the cumulative methane production was 94–98% lower than that of the control, while on the contrary, the cumulative methane and daily methane production of the anaerobic system treated with 0.002% glutaraldehyde was higher than that of the control, which was mainly due to the accelerated microbial energy metabolism, electron transfer, and immunity under the stimulation of 0.002% glutaraldehyde. Notably, the daily methane production in the anaerobic fermentation system treated with 0.01% glutaraldehyde started to rebound from the peak on the third day after a continuous decline for three days until it returned to the peak again on the ninth day (47 mL/d). This is because 0.01% glutaraldehyde did not inactivate all microorganisms in the anaerobic sludge, resulting in some microorganisms surviving and evolving into dominant flora, such as Trichococcus, which ferments glucose to produce acetic acid and carbon dioxide [49,50]. Whereas the anaerobic system, in the presence of 0.01% glutaraldehyde, showed an increase in the relative abundance (37%) of Trichococcus, which became the first dominant flora after the reaction, even though it was ranked fourth prior to the reaction. The degradation rate of VS reveals the extent of microbial conversion and utilization of substrates in anaerobic digestion systems; anaerobic reactors in the presence of 0.002% glutaraldehyde showed a 1.6% higher rate of degradation of VS, compared to the control treatment, in the anaerobic reactor. The anaerobic reactor at 0.002% glutaraldehyde had a 1.6% higher VS degradation rate than the control treatment, while the anaerobic reactor in the presence of 0.08% glutaraldehyde had a 24.5% lower VS degradation rate than the control treatment (Figure 2g).

Notably, the addition of glutaraldehyde reduced the TP content at the beginning of the reaction, which was attributable to the negative effects of glutaraldehyde on TP determination and the slow decomposition of pig manure. Potassium persulfate has been shown to be reduced by reducing organic substances, such as glutaraldehyde, thus resulting in fewer phosphorous-containing substances being oxidized [51,52]. However, 0.08% glutaraldehyde can inhibit the activity of organic-degrading bacteria, resulting in a slow release of phosphorus from pig residues. However, excess phosphorus accumulation was found in the post-reaction period. The microbial cell structure can be destroyed and hydrolyzed after a long-term reaction with glutaraldehyde, leading to the release of large amounts of nitrogen and phosphorus into the solution. In addition, bacteria such as polysaccharomycetes (GAOs) and polyphosphorylated organisms (PAOs) die with the depletion of nutrients and are responsible for the abundant accumulation of polysaccharides and TP [53]. In addition, activated sludge is more likely to release phosphorus into the water under anaerobic conditions [54] because the redox potential at the water–soil interface is low under anaerobic conditions. However, aqueous Fe(III) is easily reduced to Fe(II), resulting in the gradual conversion of insoluble ferric compounds into soluble ferric salts, which allows a large number of phosphate ions (PO₄³⁻) and ferrous ions Fe(II) to enter the sludge solutions [55].

Combined with the microbial species diversity indices in Figure 4, the results show that on day 0 of the reaction, except for the 0.002% glutaraldehyde group, where no significant changes in the Ace and Chao indices occurred, the remaining experimental groups exhibited adverse effects on microbial diversity (Shannon and Simpson indices) and species abundance (Ace and Chao indices). As the fermentation time was prolonged, the Ace and Chao indices decreased in all experimental groups except in group 0.01% glutaraldehyde, the Shannon index increased in all experimental groups, and the Simpson index showed the opposite trend, suggesting that the end period of the experiment increased the species diversity of microbial flora but decreased species abundance compared to the initiation period. The reason for this may be that 0.08% glutaraldehyde had an inhibitory effect on microbial growth, thus leading to a decrease in the abundance and diversity of microbial communities at the beginning of anaerobic fermentation. However, 0.002% glutaraldehyde did not affect microbial growth and even played a certain role in promoting growth. For example, the Ace index of the 0.002% glutaraldehyde group increased to 536.4 compared with that of the CK group (509.2). Combined with MDI index analyses, 0.08% glutaraldehyde can lead to the disruption of microbial communities, including loss of diversity, disruption of abundance, and their colonization [56]. This phenomenon further supports the analysis of the initial changes in the physicochemical indices mentioned above. With prolonged fermentation time, the 0.01% glutaraldehyde group produced resistant flora with a significant degree of dominance, which led to an increase in the diversity of microflora [57,58]. This speculation can be supported based on the changes in daily methane production in Figure 3b and the change in the K03657 (uvrD, pcrA), K03406 (mcp), and K01448 (amiABC) genomes in Figure 7. The daily methane production of the 0.01% glutaraldehyde group was lower than that of the CK group in the first 7 days after the addition of glutaraldehyde, although the inhibitory effect of glutaraldehyde on the microorganisms was weakened after the 7th day and the daily methane production exceeded that of the CK group. As for the genomic response, the expression of the above-mentioned genes related to cell repair and drug resistance was significantly altered in the 0.01% glutaraldehyde group compared with the CK group (p < 0.05).

Meanwhile, as shown in Figure 5, all the test groups exhibited an increase in the abundance of Firmicutes as the disinfectant dose increased, whereas the abundance of Proteobacteria showed an overall decrease as the glutaraldehyde dose increased, in the pre-reaction period. This may be due to the hydrolysis of microorganisms in the pig manure or activated sludge by glutaraldehyde, which increased the concentration of available organic matter in the anaerobic fermentation system, thus providing sufficient nutrients for the growth of Firmicutes [42]. However, Proteobacteria, which includes parthenogenetic or specialized anaerobic or oligotrophic bacteria [59], showed decreased abundance.

5. Conclusions

Adding glutaraldehyde resulted in notable alterations in methane production and biological sewage treatment performance during the anaerobic fermentation process in a dose-dependent manner. In the state of 0.002% glutaraldehyde addition, the cumulative methane increased by 5.27%, whereas in the case of 0.01–0.08% glutaraldehyde addition, the cumulative methane decreased by 9.7–98%, which was attributed to the irreversible damage caused by glutaraldehyde to the methanogenic microorganisms in the anaerobic sludge. Correspondingly, COD and TP increased by 12–310% and 15–27%, respectively, which is attributable to the adverse effects of glutaraldehyde on organic-degrading bacteria. Microbial community analysis showed that the 0.08% glutaraldehyde treatments impeded microbial richness and diversity, although some bacteria exhibited resistance and growth upon stimulation with 0.01% glutaraldehyde. Glutaraldehyde also drove the enrichment of certain microorganisms such as Firmicutes, which showed strong adaptability under pressure from glutaraldehyde over a wide concentration range. The complex effects of residual disinfectants on anaerobic fermentation systems for pig manure remain poorly understood, which impedes the application of disinfectants in practical anaerobic fermentation systems. Our study fills this gap in the literature by providing insights for guiding the rational use of glutaraldehyde as a disinfectant in pig anaerobic fermentation systems during epidemics.