1 Introduction

Azo dyes are currently the most widely used dyes in the textile printing and dyeing industry. A considerable amount of highly colored wastewater containing azo dyes is generated during their production. At present, anaerobic digestion (AD)-aerobic processing remains an indispensable and crucial step in treating dye wastewater. Despite the low energy consumption and environmental friendliness of the anaerobic process, the presence of various inhibitory effects leads to a slow progress in anaerobic digestion (Dai et al. 2016). Biochar is prepared from discarded biomass or sludge under oxygen-isolated conditions at temperatures ranging from 300 to 900 ℃. In recent years, there have been numerous studies reporting the use of biochar to enhance anaerobic digestion for the removal of organic pollutants (Wang et al. 2020a, 2024). However, the physicochemical properties of traditional biochar limit its further application in the anaerobic process. Considering this, it is necessary to modify the surface of biochar to enhance its electron transfer capacity (Masebinu et al. 2019).

Redox mediators can enhance biochemical reactions by accelerating electron transfer from electron donors to organic compounds. It has been reported that quinone-based compounds often act as an electron shuttling agent and accelerate anaerobic biodegradation of pollutants (do Nascimento et al. 2021; Zhou et al. 2018). This process involves mechanisms that primarily involve the reduction of quinones via microbial respiration and the subsequent reduction of pollutants via the production of hydroquinone (Liu et al. 2018). However, continuous dosing of such substances in wastewater treatment plants is cost prohibitive. Therefore, immobilization of quinone-based redox media is an economically viable alternative. In our previously published report, prior research on enhancing the anaerobic decolorization of azo dyes using anthraquinone-loaded biochar (AQS-BC) was discussed (Wang et al. 2021a). In systems using carbonaceous materials as biocarriers, anaerobic sludge (AS) exists in two forms: flocculent sludge and biofilm grown on carrier surfaces. However, most current research on biochar as an additive in AD is limited to batch experiments of mixed systems, and thus lacks evidence regarding the roles played by biofilms.

Biofilm treatment of wastewater is a lucrative approach. Since solid–liquid separation can be achieved, biofilm formation can extend the residence time of biosolids and achieve good biomass retention (Wu et al. 2022). The formation of biofilms is generally considered to involve several processes—including microbial aggregation, surface attachment, and encapsulation (Cayetano et al. 2022). Fundamentally, it can be understood as the process whereby microorganisms come into contact with and are captured by the carrier material. This can be likened to the surface adsorption phenomenon in environmental chemistry. The concentration of solute in the liquid phase and the adsorption time are important factors affecting the adsorption capacity. Similarly, it can be considered that, in anaerobic systems, the sludge concentration in wastewater and the contact time between microorganisms and carriers affect biofilm growth. In addition, the organic load and pollutant concentration affect the development and reproduction of microorganisms. However, current research on the influence of these factors during the biofilm formation process remains scarce. More systematic and comprehensive studies are therefore warranted.

To the best of our knowledge, there is currently limited research on the treatment efficiency and stability of biofilms on carrier materials during long-term AD-based processes. The main purposes of this study were the following. (1) To separate biochar (including BC and AQS-BC) from flocculent sludge. Taking the biofilm system grown on the BC surface as a basis for comparison, this study focused on the performance of AQS-BC biofilm in the RR2 anaerobic decolorization process. (2) To study the effects of sludge concentration, contact time, carbon source concentration, and dye concentration on biofilm maturity. (3) To study the liquid- and gas-phase metabolites within the reaction system and to analyze the microbial growth statuses of the biofilms of the two materials as well as the characteristics of extracellular polymeric substance (EPS) in order to clarify the advantages of AQS-loading in the biofilm formation process. (4) To compare the levels of diversity and differences between bacterial microorganism surface compositions in BC and BC-AQS systems.

2 Materials and methods

2.1 Preparation of BC and AQS-BC

The procedure for preparing the biochar and AQS-loaded BC (AQS-BC) have been described in our previous report (Wang et al. 2021a). Briefly, the biochar was prepared by pyrolyzing sludge from a wastewater treatment plant at 500 °C for 2 h. Anthraquinone-2-sodium sulfonate (AQS) was loaded onto the surface of biochar as a quinone mediator. The biochar was surface-modified using Lucas reagent and the AQS was then loaded onto the surface of the biochar via the adsorption method.

2.2 Effect of sludge inoculation conditions on biofilm maturity

The batch experiments for the formation of AQS-BC biofilms were conducted in a series of 1 L anaerobic fermentation bottles. Four factors were selected as key influencing factors for the formation of biofilms: contact time between AS and the AQS-BC (2, 6, 12, 18, 24, and 36 h), inoculum sludge concentration (0.1, 0.5, 1, 2, 3, and 5 g VSS/L), glucose concentration (300, 500, 1000, 1500, 2000, and 3000 mg L–1), and RR2 concentration (100, 150, 200, 250, 300, and 400 mg L–1). After separation of AQS-BC and AS, decolorization experiment was conducted for 15 cycles to evaluate the formation of the biofilm. The RR2 and glucose concentrations of the simulated wastewater in the experiment were 100 mg L–1 and 1000 mg L–1, respectively. The pH was maintained at 6.8 ± 0.2. The simulated wastewater was blown off the dissolved oxygen with N2 for 10 min, and the bottles were put into a shaking table with a speed of 100 r min–1 and a temperature of 30 ℃. Each set of experiments was repeated three times. The achievement of a stable decolorization efficiency for RR2 was considered to indicate the successful initiation of biofilm formation.

2.3 The decolorization effect of flocculent sludge and biofilm

In our previously published report, both AQS-BC and BC were used in 15 cycles of cyclic decolorization experiments. The AS and biofilm on the carrier after the cyclic experiments was analyzed in this study. An air bag filled with nitrogen was connected to the anaerobic fermentation bottle to isolate the air, and deoxygenated deionized water was injected into the bottle to a final volume of 900 mL. After thoroughly mixing the solution, the sludge was immediately withdrawn from the bottle using a peristaltic pump to achieve separation of the AS and carrier. Following sedimentation of the AS flocs, the supernatant was decanted and the biomass of the sludge and biofilm was measured. Experiments were conducted by adding simulated dye wastewater at the same level of organic loading to investigate anaerobic decolorization efficiency of the separated AS and biofilm.

2.4 Analytical method

COD was determined using the rapid digestion spectrophotometric method (DRB200, Hach, USA)c. RR2 concentration was measured using an ultraviolet spectrophotometer (UV721-100, Jinghua Instrument, Shanghai) at a wavelength of 540 nm. The production of volatile fatty acids (acetic, propionic, and butyric acids) during the anaerobic biological treatment was detected using a gas chromatograph (GC-7890A, Agilent, USA) equipped with a flame ionization detector (FID). The injection port temperature was set to 250 ℃, and helium was used as the carrier gas. The biogas generated in the anaerobic fermentation bottle was collected using gas bags. This gas was extracted from the bags using a 10 mL syringe and rapidly injected into the injection port of the chromatograph. The composition of the produced gas (CH4 and H2) was measured via gas chromatography (GC Trace 1300, Thermo-Fisher Scientific, USA). To evaluate the electron transfer efficiency of the biofilm system, cyclic voltammetry (CV) analysis was conducted using an electrochemical workstation (CHI-760E, Chenhua, China). Iodonitrotetrazolium chloride (INT) was used as the electron acceptor to assess the electron transfer system (ETS) activity of the biofilms.

EPS was extracted using a heating method, and the specific process can be found in the Supplementary Information, Method 1. The polysaccharide (Comas et al. 2008) and protein (PN) in the EPS were determined via the sulfuric acid-anthrone and Lowry methods, respectively, with crystalline bovine serum albumin and glucose serving as standards. The composition and distribution of soluble organic compounds in the EPS were analyzed using a three-dimensional fluorescence spectrophotometer. The excitation spectrum was scanned from 200 to 400 nm, and the emission spectrum was scanned from 200 to 500 nm. The scanning step size was 5 nm and the scanning speed was 2000 nm min–1. Spatial distribution of biofilm EPS and biofilm thickness were observed using a confocal laser scanning microscope (CLSM, Leica TCS SP8, Germany), and three-dimensional image analysis was performed using ZEISS ZEN 2012 software. The pretreatment and staining methods for the biofilm samples can be found in Method S2 and Method S3 in Supplementary Information. Scanning electron microscopy (SEM, Zeiss Ultra60, Germany) was used to observe the morphological characteristics of the biofilms. All samples were fixed and pre-processed prior to observation.

The biomass of the biofilm was determined using a weight-based method. Briefly, appropriate samples of biochar with biofilm were collected, mixed with deionized water, and vigorously shaken to detach the biofilm. The mixed liquids were then poured into crucibles and dried at 105 °C for 24 h. The crucibles were then calcined at 600 °C for 2 h in a muffle furnace. The obtained VSS was recorded as W2 (mg VSS). The mass of the biochar after drying was recorded as W1 (g). The biofilm biomass per mass unit of biochar (M) was calculated using the following formula:

$$M = \frac{M_2 }{{M_1 }} \times 100\%$$

The unit of M was mg VSS/g biochar.

2.5 Microbial community analysis

High-throughput sequencing technology was used to characterize the microbial communities in the biofilms. After the biofilm on the biochar surface was shed, the sludge-water mixture was centrifuged using a refrigerated centrifuge at 4000 rpm. After centrifugation, the supernatant was discarded to obtain microbial samples of the biofilm. The DNA of the sludge microbial samples was extracted using a DNA extraction Fast DNA Spin Kit (MP Biomedicals, Solon, OH, USA). Universal primers 27F (5'-AGAGTTTGATCCTGGCTCAG-3') and 533R (5'-TTACCGCGGCTGCTGG CAC-3') were used to amplify the V3-V4 hypervariable region of the 16 S rRNA gene. Polymerase chain reaction (PCR) amplification was performed using a PCR gene amplifier (GeneAmp® 9700, ABI, USA). PCR amplification conditions are described in the Supplementary Information, Method S4. After completing the detection and quantification of the PCR products, Majorbio Bio-pharm Technology Co., Ltd. (Shanghai, China) performed the sequencing using the Illumina MiSeq PE300 high-throughput sequencing platform.

3 Result and discussion

3.1 Characteristics of BC and AQS-BC

The loading amount of AQS on the surface of BC was 148.09 mg g–1. The FTIR spectra of BC and AQS-BC are shown in Fig. 1a. The stretching vibration peak within the range of 1650–1630 cm−1 represented carbonyl groups, and the absorption intensity of this peak in the AQS-BC spectrum was higher than that of BC, indicating that AQS was successfully loaded onto the surface of BC. The BET results showed that the specific surface area of AQS-BC increased compared to BC. The larger specific surface area of AQS-BC facilitates the immobilization and growth of microorganisms. And the specific surface area of all materials slightly decreased after adsorption of RR2 in Table S1. In addition, in our past study, it was found that the electron accepting capacities (EAC) of AQS-BC and BC were 1.326 mmol e–g−1 and 0.608 mmol e–g−1, respectively, and the electron donating capacity (EDC) was 0.788 mmol e–g−1 and 0.329 mmo e–g−1, respectively. This indicated that the loading of AQS improved the electron gain and loss ability of biochar, which helps to accelerate electron transfer (Wang et al. 2021a).

Fig.1
figure 1

(a) Fourier transform infrared (FT-IR) spectra of the AQS-BC and BC, (b) efficiency of RR2 decolorization by different systems, and multicycle experiment of RR2 decolorization of (c) anaerobic sludge and (d) biofilm after separation

Full size image

3.2 Decolorization efficiency analysis of RR2

3.2.1 Efficiency of RR2 decolorization by different systems

After separating AS with AQS-BC and BC, the anaerobic decolorization experiment of RR2 continued, and the results are shown in Fig. 1b. The decolorization rates of RR2 in the AQS-BC (biofilm) + RR2 and BC (biofilm) + RR2 were 92.36% and 73.95%, respectively, after 48 h. In comparison, the decolorization rates of RR2 in BC + RR2 and AQS-BC + RR2 were only 18.95% and 18.15%, respectively. This showed that the adsorption of the material only played a small role in the decolorization of RR2, and the main reason for the decolorization of RR2 was the role of biofilm.

3.2.2 Efficiency of RR2 decolorization by AS and biofilm

Over 15 cycles, RR2 decolorization efficiency by AS separated from the vector showed a gradual downward trend, as shown in Fig. 1c. The maximum decolorization efficiency of the AS separated from AQS-BC within 48 h was 55.75%, and that of the sludge separated from BC was 52.52%. Both maximum decolorization efficiency occurred within the first three cycles. Beginning on the 4th cycle, the RR2 decolorization efficiency of the two AS groups showed a steep downward trend. At the end of the 15th cycle, the RR2 decolorization efficiencies of the AS after separation from AQS-BC and BC within 48 h were only 10.64 and 9.58%, respectively, which were 80.93% and 81.76% lower than their respective maximum efficiencies. This indicates that after separated with AQS-BC and BC, the anaerobic decolorization ability of AS on RR2 recovered to the same level it had when no redox mediator was added. The impact of AQS-BC and BC on AS was therefore determined to be time-efficient, and the presence of mediator materials was beneficial to anaerobic decolorization. The relatively high RR2 decolorization efficiency observed in the early cycles suggested that the microorganisms were not highly sensitive to the presence or absence of mediator materials. AS was still able to degrade RR2, possibly because the addition of the redox mediators stimulated microbial aggregation and enhanced their adaptability to external environmental changes. In the following cycles, however, the original biochemical characteristics of the microorganisms changed with the degradation of the dyes, and the AS lost the "protective" effect of the carrier, gradually returning to the same state as the inoculated sludge without the carrier. This resulted in a decrease in RR2 decolorization efficiency (Li et al. 2022b).

The RR2 decolorization efficiencies of the AQS-BC and BC biofilms separated from AS are shown in Fig. 1d. The average RR2 decolorization efficiencies of the AQS-BC within 24 and 48 h were 84.91% and 94.06%, respectively, while those of the BC were 77.30% and 89.46%. The AQS-BC and BC biofilms continued to exhibit high decolorization efficiencies for RR2, with almost no lag phase. Moreover, the biofilm grown on the surface of the AQS-BC exhibited a higher RR2 decolorization efficiency than BC. This may be attributed to the fact that the AQS-BC had more quinone functional groups on its surface, which facilitated more participation of RR2 in biochemical reactions. Compared to AS, the biofilm formed on the carrier surface may have been able to come into closer contact with the functional groups on the carrier surface. This type of contact is advantageous for the formation of microbial communities related to RR2 decolorization on the surface, facilitating more efficient and rapid substance transport that increases decolorization efficiency (Ren et al. 2022). The biofilm formed on the AQS-BC carrier can directly adsorb the aniline substances formed after decolorization and improve its degradation efficiency under anaerobic conditions (Van der Zee and Cervantes 2009). This was confirmed by the average COD removal of 93.00% and 89.35% for AQS-BC and BC, respectively, within 48 h. Therefore, the formation of the biofilm is also beneficial for the degradation of benzidine-like substances in the biofilm, thereby mitigating their toxic effects on the microorganisms. This is one of the reasons why the biofilm was able to efficiently degrade RR2.

3.3 Study on anaerobic treatment performance of the biofilm

3.3.1 Effect on the composition and content of VFAs

The production and composition of VFAs produced by the AQS-BC and BC biofilm systems over one reaction cycle are shown in Fig. 2a. The four bar charts at each time point, from left to right, represent AQS-BC biofilm system with RR2, AQS-BC biofilm system without RR2, BC biofilm system with RR2 and BC biofilm system without RR2, respectively. Determination of the VFAs showed that acetic, propionic, and butyric acids were the main fermentation products. The content of all these three VFAs showed an initial increase followed by a decrease trend, with the levels of propionic and butyric acid staying essentially identical in all four experimental groups. Acetic acid was the dominant VFA, and its content varied more significantly between the AQS-BC and BC systems. In the two control groups without RR2, the acetic acid content showed an initial increase followed by a gradual decrease. By contrast, in the two experimental groups with RR2, the acetic acid accumulated, indicating that AQS impacted its consumption, which is consistent with previous study (Wang et al. 2018). The VFAs generated in the original anaerobic fermentation system were further metabolized by microorganisms to produce CH4 and CO2. The significant decrease in acetic acid content at 48 h indicated enhanced microbial activity that used acetic acid as a substrate. Considering that the decolorization efficiency had already reached ~ 90%, AQS may have promoted acetic acid to become the preferred electron donor for the reduction of RR2, thereby increasing the efficiency of the decolorization reaction. RR2 is an organic compound with oxidative properties that can inhibit the metabolic activity of acetolactic methanogens (Castañon et al. 2019). In that case, once the RR2 was degraded, its inhibitory effect on acetolactic methanogens likely also decreased, thus promoting the consumption of VFAs.

Fig. 2
figure 2

(a) VFAs concentration, (b) cumulative methane production and (c) cumulative hydrogen production of the biofilm systems in one reaction cycle at a steady state

Full size image

3.3.2 Effect on biogas production

The cumulative production of CH4 and H2 during the decolorization process is shown in Fig. 2b, c. In the presence of RR2, the cumulative methane production of AQS-BC was 18.57 mL at 48 h, which was slightly higher than that of BC (17.31 mL). When RR2 was not added to the influent, there was no significant difference in methane production between AQS-BC and BC, with values of 19.46 mL and 19.68 mL, respectively. This indicates that in dye wastewater, AQS enhanced the methanogenic capability of the biofilm to a certain extent, whereas it does not demonstrate an advantage in wastewater without added dyes. Furthermore, in the systems containing RR2, methane production was lower than in those without RR2. Moreover, even as the decolorization efficiency stabilized, methane production continued to increase. This suggests that the presence of azo dyes inhibits methane production in anaerobic environments. There is likely competition for carbon sources between methanogens and bacteria involved in RR2 decolorization. The loading of AQS may enhance the competitiveness of RR2 decolorizing bacteria for these carbon sources. In the anaerobic decolorization process, the chromophoric group of RR2 was cleaved at its azo bond (–N=N–) by incoming electrons that may have come from VFAs (such as acetic, propionic, or butyric acids) and H2 generated during anaerobic fermentation (Li et al. 2014; Oliveira et al. 2023). Based on the cumulative hydrogen production, it was observed that the addition of RR2 resulted in lower hydrogen production for both the AQS-BC and BC groups compared to the two control ones. This suggests that the presence of RR2 led to a decrease in hydrogen production, possibly because hydrogen was preferentially used as the electron donor to reduce RR2, rather than the VFAs. In terms of H2, there may exist a competitive relationship between this process and methanogenesis performed by hydrogenotrophic methanogens (Dai et al. 2018). The cumulative H2 production in the AQS-BC system was higher than in the BC system, which may have shifted the reaction toward H2 as the electron donor used for the reduction of RR2.

3.4 The mechanism of enhanced anaerobic decolorization by the AQS-BC biofilm

3.4.1 Biochemical properties of biofilms

The thickness and biomass of the biofilms formed on the surfaces of the AQS-BC and BC are shown in Fig. 3a. After 15 cycles of the circulation experiment, the biomass of the biofilms formed on the surfaces of the two carriers was 137.45 mg VSS/g and 130.75 mg VSS/g, respectively. It was therefore evident that more microorganisms adhered to the surface of the AQS-BC. The increase in biomass likely contributed to the higher anaerobic decolorization efficiency observed in the system. The presence of some suspended biomass during the reaction indicated that the formation of the biofilm was a dynamic process involving both attachment and detachment (Sauer et al. 2022). This phenomenon could be a key factor that contributed to the long-term stable operation of the reaction system. The thickness of the biofilms formed on the surfaces of the AQS-BC and BC was 675.45 μm and 620.45 μm, respectively. The former had a higher biofilm thickness, which was consistent with the trend observed for biomass changes. In general, the thickness of the biofilm is associated with mass transfer efficiency in the biofilm, and a greater biofilm thickness often implies lower mass transfer efficiency. However, the higher decolorization and COD removal efficiency in the AQS-BC biofilm indicated that mass transfer was not significantly affected by the thickness of the biofilm. According to previous studies, the addition of redox mediators in the biodegradation of azo dyes can enhance the mass transfer efficiency in the EPS of microbial aggregates, thereby improving the decolorization efficiency (Dai et al. 2016). This was mainly due to the fact that that AQS can effectively transfer reducing equivalents from EPS to RR2 by changing its own redox state, which allowed RR2 to accelerate its reduction.

Fig. 3
figure 3

a Biomass and thickness of two biofilms. b EPS content of two biofilms. c ETS activity and (d) CV curves of different systems. 3D-EEM profiles of EPS in (e) AQS-BC biofilm and (f) BC biofilm

Full size image

EPS is considered to be a substance secreted by microorganisms in response to external environmental pressures, forming a protective layer to create a stable living environment for the microorganisms (Yu et al. 2023). PN and PS are the main components of biofilm EPS, accounting for approximately 70 ~ 80% of the total mass. In order to further elucidate the effect of AQS loading on biofilm EPS, the contents of PN and PS in EPS were quantitatively analyzed. As is shown in Fig. 3b, the PS contents in the EPS of the AQS-BC and BC biofilms were 12.65 mg g–1 VSS and 9.83 mg g–1 VSS, respectively, and their respective PN contents were 33.83 mg g–1 VSS and 23.47 mg g–1 VSS. The PN and PS contents in the EPS of the AQS-BC biofilm were 28.7% and 44.1% higher, respectively, than those in the BC biofilm. This provides evidence that the loading of AQS promotes the secretion of EPS, with a greater effect on PN than PS. The increase of PN in EPS can enhance microbial aggregation, which is mainly due to the special spatial structure of spiral folding of PN itself. And PS has a high content of dextran and shows a negative charge. When the content of PS increases, the repulsive force between microbial flocs increases, leading to sludge loosening, which reduces the efficiency of the microbial for pollutant treatment. However, the biofilm didn’t show any signs of structural damage, which could be attributed to the more significant role of PN in maintaining the structural stability of the biofilm. PS, by contrast, was thought to play a more significant role in increasing the thickness of the biofilm. As has been mentioned in previous reports, PS in EPS mainly exhibits hydrophilic characteristics, while PN exhibits hydrophobic ones (Wang et al. 2020b). An increase in the PN/PS ratio can therefore promote microbial cell aggregation. The PN/PS ratio in the EPS of the AQS-BC and BC was 2.68 and 2.39, respectively. This also indicated that the biofilm on the AQS-BC surface was more stable structurally.

The 3D-EEM spectra of EPS of the AQS-BC and BC biofilms are shown in Fig. 3e, f. Generally, 3D-EEM spectra can be divided into five regions. The characteristic peaks in the spectra of both biofilms were found to be distributed in region I (Em: 250–400 nm/ Ex: 200–380 nm) and II (Em: 250–400 nm/ Ex: 380–500 nm), representing soluble microbial by-products (SMPs) and humic-like substances, respectively (Li et al. 2022a). Compared to the AQS-BC spectra, the BC spectra showed only one characteristic peak in region I, indicating the presence of humic-like substances. Humic acid is a redox mediator, and the oxygen-containing functional groups in its structure can also accelerate electron transfer. Humic acid-like substances may also originate from the decomposition of cells or biomacromolecules. The increase in SMP implies an increase in metabolic activity (Barker and Stuckey 1999). Taken together with the results of a previous analysis of biofilm biomass, the higher biomass observed in the AQS-BC biofilm may also explain the higher levels of SMP. It also suggests that the microorganisms in the AQS-BC biofilm had higher metabolic activity.

3.4.2 Electrochemical properties of biofilms

ETS activity is an effective indicator reflecting the electron transfer activity within microbial cells and the overall biological activity of sludge. The ETS activity of the biofilm is shown in Fig. 3c, and the ETS activity of AQS-BC and BC biofilms were 125.2% and 88.4% higher, respectively, than that observed in the flocculent sludge. This showed that loading AQS on BC could improve the ETS activity of the biofilm formed on its surface to a certain extent. This could be because the biochar is involved in extracellular electron transfer. In particular, AQS (as an electron acceptor) participates in extracellular respiration, also known as humus respiration. As a result, ETS activity is enhanced.

CV curves of AQS-BC and BC surface biofilms are shown in Fig. 3d. Compared to BC, AQS-BC showed a more obvious reduction peak. The integral area of the closed loop reflects the charge–discharge performance of the system. Through calculation, it was found that the charge–discharge performance of the AQS-BC biofilm system improved by 5.20% compared to BC, demonstrating better redox activity in the AQS-BC biofilm system (Xu et al. 2023). The BC loaded with AQS enhanced electron transfer in the anaerobic biofilm system.

3.4.3 Surface structural characteristics of biofilms

In order to understand the influence of AQS loading on the morphology of biofilms formed on biochar surfaces, SEM was used to observe both biofilms (Fig. 4a–d). At 1500× magnification, it was observed that many microorganisms and bacterial jelly clusters were attached to the surfaces of both carriers, clustering together to form the main surface structure of the biofilm. This indicated that both materials can serve as effective microbial carriers to promote microbial enrichment. The irregular and complex morphology of the biofilm also indicated the heterogeneity of the biofilm structure. At 5000× , it became clear that many Streptococci and short rod-shaped bacteria were attached to the AQS-BC surface. Among them, the Streptococci were more dispersed, while the rod-shaped bacteria were present in aggregated states. There were fewer Streptococci on the BC surface, and the rod-shaped bacteria were more dispersed. A small number of cocci were also observed. It was hypothesized that AQS promoted more EPS secretion by the microorganisms and also helped increase the diversity of the microbial communities in the biofilm system, which is consistent with previous research (Lu et al. 2021). The presence of numerous Streptococci, rod-shaped bacteria, and filamentous bacteria on the AQS-BC surface was likely related to the synergistic degradation of RR2. This analysis provided a microscopic exploration of the morphological characteristics of the key microorganisms involved in more efficient anaerobic decolorization of RR2. Compared to BC, the microorganisms on the AQS-BC surface exhibited more spatially intricate structures. The presence of these biofilm structures implied that the microorganisms had better cohesion and stability, which likely enhanced their probability of contact with nutrients in wastewater. This is also naturally more conducive to the contact between pollutants and redox mediators on the surface of carriers throughout the biofilm, improving the efficiency of their participation in biochemical reactions.

Fig. 4
figure 4

SEM image of biofilm: (a) 1500× amplification of AQS-BC; (b) 5000× amplification of AQS-BC; (c) 1500 times amplification of BC; (d) 5000× amplification of BC. CLSM images of (e) AQS-BC biofilm and (f) BC biofilm

Full size image

CLSM can be used to visually characterize the composition and distribution of cells and EPS in biofilms (Wagner et al. 2009). The CLSM spectra of the AQS-BC and BC surface biofilms are shown in Fig. 4e, f. α-d-glucopyranose, β- d-glucopyranos, protein and total cells correspond to the red, green, pink and purple areas, respectively. The two biofilm structures exhibited non-uniform and irregular states, with the distribution of polysaccharides and proteins intertwining around the cells. The aforementioned state could also be confirmed through SEM image, indicating that EPS played the role of a "bridge" and binder in the formation of biofilm (Wang et al. 2020c). Specifically, the structure of the biofilm on the AQS-BC surface was tighter, and the fluorescence intensity of β- d-glucopyranose, the main component of EPS polysaccharides, was higher. Proteins were widely distributed in the biofilms, indicating their important role. The high total cell fluorescence intensity in both biofilms indicated that both AQS-BC and BC can serve as good microbial carriers that help maintain cell viability. However, the loading of AQS onto BC can stimulate microorganisms to secrete more proteins and polysaccharides. This forms the basis for maintaining the higher stability and metabolic activity of AQS-BC-based biofilms. These results are consistent with the results of polysaccharide and protein content observed in a previous study, which was concluded that AQS effectively altered the microstructure and physicochemical properties of the sludge, contributing to maintaining a higher RR2 degradation efficiency.

3.4.4 Comparative analysis of microbial diversity

The α-diversity index of microorganisms can reflect the diversity of a microbial community. As is shown in Table S2, the sequencing depth (coverage) for this analysis was consistently > 0.999, indicating that the majority of microorganisms in the samples were detected. The Shannon diversity indexes for the AQS-BC and BC microbial communities were 5.214 and 5.127, respectively, while their respective Simpson indexes were 0.011 and 0.012. The Shannon index primarily measures the diversity and richness of species in an ecosystem, with a higher value indicating a more diverse ecosystem. By contrast, the Simpson index focuses more on the relative abundance of dominant species. A lower Simpson index suggests lower dominance, with more species contributing to the microbial community. The ACE indexes of the AQS-BC and BC biofilms were 754.6 and 664.7, respectively, while their Chao1 indexes were 751.2 and 663.8, respectively. Both the ACE and Chao1 indexes aim to estimate the number of unobserved species, with larger values indicating a greater number of sample species. Comparing the α-diversity indexes of the microorganisms in the two biofilms, it could be observed that the loading of AQS effectively increased the species richness and biological diversity of the microbial community. Higher microbial population richness can effectively protect microorganisms from adverse effects caused by environmental changes and functional decline.

The similarity of microbial communities in the biofilms of AQS-BC and BC was analyzed by β-diversity comparison. The direction of the sparse curve flattens with the increase of the number of sequences in Fig. 5a, indicating that the sequencing data effectively reflect the information about most microorganisms in the biofilm microbial community. The differences of microbial communities in biofilms are shown in Fig. 5b. The number of ASVs common to the two biofilms was 501, accounting for 35.33% of all ASVs (1418), which indicated that biochar as a carrier had a consistent effect on the formation of microbial communities. The number of unique ASVs in the biofilm communities on the AQS-BC and BC surfaces was 524 and 393, respectively, indicating that there was a significant difference between the two microbial communities. The difference was mainly caused by AQS loading, which is also consistent with the results of the α-diversity analysis. It is worth noting that the number of unique ASVs in AQS-BC was greater than that in BC, suggesting that AQS loading had a greater impact on the diversity of the biofilm communities.

Fig. 5
figure 5

Similarity analysis of the microbial communities in the biofilms (a) rarefaction curves and (b) Venn diagram, relative abundance of bacterial communities at the (c) phylum and (d) genus levels in two biofilms

Full size image

3.4.5 Comparative analysis of microbial community structure

In order to further analyze the differences in microbial community structure composition within the AQS-BC and BC biofilms, a structural analysis of microbial communities at the phylum and genus levels was conducted, and the results are presented in Fig. 5c. The main microorganisms found in both biofilms included Bacteroidota, Proteobacteria, Patescibacteria, Firmicutes, Spirochaetota, Chloroflexi, Synergistota, Actinobacteriota and Desulfobacterota. Bacteroidota represented the microbial phylum with the highest relative abundance in both biofilms, accounting for 26.22% and 26.72% in the AQS-BC and BC samples, respectively. The relative abundance of Proteobacteria, Firmicutes, Actinobacteriota, and Desulfobacterota in the two biofilm communities was 20.07% and 13.58%, 15.48% and 13.45%, 3.08% and 2.25%, 3.19% and 1.98%, respectively. The abundance of these phyla in the AQS-BC biofilm was higher than that in BC, indicating that AQS primarily affected these four phyla. The Proteobacteria phylum can degrade a variety of organic pollutants and has been widely reported to become the dominant bacterial phylum in anaerobic digestion reactors (Liu et al. 2024). This suggests that the loading of AQS has an impact on the microbial community structure. Microorganisms belonging to the Firmicutes phylum exhibit a strong capacity to promote the biological transformation of easily reducible pollutants under anaerobic conditions (Shao et al. 2023). The Actinobacteriota phylum is often detected in extremely polluted environments, aquatic habitats, and complex wastewater treatment systems, indicating its potentially significant role in the degradation of azo dyes under anaerobic conditions. The increase in the relative abundance of these functional microorganisms suggested that a large number of strains are synergistically involved in the RR2 degradation mechanism. It also helps to explain why the decolorization efficiency of the AQS-BC biofilm was higher than that of the BC.

The top 30 genera identified by relative abundance are shown in Fig. 5d. The top 10 were norank_f__norank_o__Saccharimonadales, norank_f__Bacteroidetes_vadinHA17, Lentimicrobium, Dechloromonas, norank_f__Synergistaceae, norank_f__Lentimicrobiace, norank_f__norank_o__norank_c__SJA-28, Desulfovibrio, Brachymonas, and norank_f__norank_o__PeM15. The genus norank_f__norank_o__Saccharimonadales belongs to the phylum Patescibacteria, and is an important genus in the anaerobic microbial process for COD removal and pollutant degradation. The norank_f__Lentimicrobiaceae and Lentimicrobium genera are often detected in anaerobic reactors used for the industrial-scale treatment of organic wastewater containing high-strength starch (Clagnan et al. 2023). The genus norank_f_Bacteroidetes_vadinHA17 can degrade glucose into acetate, propionate, and H2/CO2 during anaerobic fermentation processes (Wang et al. 2021b). Brachymonas in the biofilm can stimulate the secretion of EPS, maintaining the stability of the entire biofilm structure (Jia et al. 2021). The higher relative abundance of Brachymonas found in the AQS-BC is consistent with the results of the EPS analysis. Desulfovibrio belongs to a type of sulfate-reducing bacteria. Studies have shown that some microorganisms of the Desulfovibrio genus can enhance the degradation of aromatic compounds (Qian et al. 2021). Dechloromonas is a genus of bacteria that has demonstrated quinone-reducing ability and is often found when redox mediators participate in anaerobic digestion processes. The norank_f_Synergistaceae genus can improve anaerobic digestion performance and produce VFAs (Meng et al. 2022). It represents an important bacterial genus in the hydrolysis acidification process and acetyltrophication pathway. Acidovorax demonstrated potential in the degradation of aromatic compounds and denitrification, which partly explains the lower concentration of aniline in the AQS-BC biofilm system. Brachymonas plays a crucial role in the degradation of phenolic substances, and phenol is a degradation product of aniline, which is derived from the degradation of Reactive Red 2. Therefore, an increase in the content of Brachymonas is beneficial for the further metabolism of aniline.

3.5 Effect of inoculation conditions on biofilm formation

3.5.1 Contact time

The formation of biofilm mainly relies on the growth of microorganisms on the surface of the carrier, rather than solely on the attachment of suspended microorganisms. Therefore, it is important to use foreign microorganisms to inoculate the biofilm system and thus shorten its cultivation cycle. The RR2 decolorization efficiency of biofilms under different contact times is shown in Fig. 6a. In the six experimental groups, the decolorization efficiency of the biofilm system within 48 h continued to increase within 15 cycles. The decolorization efficiency of the biofilm increased gradually at 2, 6, and 12 h of contact time, showing a relatively slower rate of increase. These three groups required a longer time to reach a relatively stable decolorization efficiency. On the 30th day, the decolorization efficiencies reached 92.43%, 89.76%, and 91.29%, respectively. By contrast, when the contact times were 18 h, 24 h, and 36 h, the decolorization efficiencies of the biofilm increased faster. By the 8th cycle (i.e., day 16), the decolorization efficiency of these groups exceeded 80%, while the other three groups remained below 60%. The decolorization efficiency of these three groups tended to be stable on days 22, 20, and 22, respectively, indicating that the biofilm had gradually entered a mature state. It is worth mentioning that when the contact time was 36 h, the growth trend and final stability value of biofilm decolorization efficiency did not seem to show obvious differences compared to that at 24 h, implying that an excessively long contact time does not accelerate the biofilm maturation process. These results indicate that the contact time has a significant impact on the formation of the biofilm, and an appropriate contact time can effectively shorten the time required for a biofilm to reach its stable state.

Fig. 6
figure 6

Effect of (a) contact time, (b) sludge concentration, (c) glucose concentration and (d) RR2 concentration on the formation process of biofilm

Full size image

3.5.2 Sludge concentration

The flocculent AS was diluted to different sludge concentrations, and the effect of sludge concentration on biofilm formation was investigated for a contact time of 24 h. The results are shown in Fig. 6b. The RR2 decolorization efficiency was significantly lower when the inoculum sludge concentrations were 0.1 and 0.5 g VSS/L, indicating that lower inoculum sludge concentrations delayed the formation of biofilms to some degree. When the inoculum sludge concentration exceeded 1 g VSS/L, the RR2 decolorization efficiency stabilized at > 90% by the 10th cycle (i.e., day 20), indicating that the biofilm had matured and could perform anaerobic decolorization stably and efficiently. The excessively high inoculum sludge concentration did not significantly shorten the formation time of the biofilm. This may be because the initial microbial attachment to the carrier surface was saturated, and the microbial attachment did not increase with the increase in the inoculum sludge concentration in the system. Therefore, under this experimental condition, the appropriate sludge concentration was 1 g VSS/L.

3.5.3 Glucose concentration

Glucose, as an electron donor, is an essential component for the stable decolorization of RR2. As a nutrient, glucose can also play a crucial role in the formation of biofilms. The effect of different initial glucose concentrations on the RR2 decolorization efficiency during the biofilm formation process was investigated, and the results are shown in Fig. 6c. When the glucose concentrations were 300 mg L–1 and 500 mg L–1, the RR2 decolorization efficiency remained low at 30 days, reaching only 40.43% and 55.76%, respectively. According to the electron balance, the cleavage of 1 mol of azo bonds requires 4 mol of electrons (corresponding to 64 mg COD). Therefore, 300 mg L–1 of glucose can theoretically provide sufficient electrons to achieve the reductive decolorization of 100 mg L–1 RR2. Although the majority of heterotrophic microorganisms in nature can use glucose as carbon source for metabolic synthesis, the lower organic content in anaerobic environments is not conducive to the growth and reproduction of sludge microorganisms, thus resulting in a lower decolorization efficiency (Liang et al. 2022). When the glucose concentration increased to 1000 mg L–1, the decolorization ability of the biofilm significantly improved. After the 9th cycle, the decolorization efficiency exceeded 90%, indicating that an adequate supply of easily degradable carbon sources was necessary for the anaerobic decolorization process. This may be because there is a competitive relationship between the decolorization and methanogenesis processes in the anaerobic reaction system. Adequate glucose provides sufficient nutrients, which is beneficial for the proliferation and metabolic activities of various microorganisms, thus shortening the maturation time of the biofilm. Further increasing the glucose concentration resulted in no significant difference in the biofilm formation time. With further increases in glucose concentration, there was no significant difference in the formation time of the biofilm, which meant that the carbon source available for the organisms related to the decolorization reaction had reached saturation.

3.5.4 RR2 concentration

As pollutant, RR2 exerts a non-negligible impact on the activity of microorganisms. The initial RR2 concentration also played an important role in the formation of the biofilms in this study. The effects of initial RR2 concentration on biofilm formation are detailed in Fig. 6d. When the initial RR2 concentrations were 100 and 200 mg L–1, the respective biofilms already exhibited efficient decolorization capability by the 9th cycle. When the RR2 concentrations were 100 and 150 mg L–1, the respective biofilms achieved efficient decolorization by the 9th cycle. At 30 days, the decolorization efficiencies reached 92.5% and 92.04%, respectively. At concentrations of 200 and 250 mg L–1, although the final decolorization efficiency also exceeded 90%, the biofilm required more cycles (11th and 12th ones) to reach a stable state. When the initial concentrations of RR2 were 300 and 400 mg L–1, the RR2 decolorization efficiencies were only 41.23% and 55.73%, respectively. When the RR2 concentration was lower, the maturation time of the microorganisms in the biofilm was found to be shorter. However, the presence of RR2 was also able to stimulate the growth of the biofilm, promoting the secretion of EPS. When the RR2 concentration was relatively high, both RR2 and its degradation products inhibited the activity of the microorganisms to some degree, leading to a decrease in the ability of the biofilm to degrade RR2 (Wang et al. 2018). Additionally, owing to the limited availability of glucose as an electron donor, the proportion of reduced RR2 molecules decreased gradually as the initial RR2 concentration increased. This resulted in a decrease in decolorization efficiency. This part of the study also elucidated the concentration threshold of dye wastewater that was conducive to biofilm formation. For anaerobic biofilm systems being used to treat dye wastewater, controlling the initial dye concentration in the influent is important for promoting the maturation of the biofilm.

3.6 Roles of AQS-BC biofilm in enhanced decolorization

Compared to the original BC, AQS-BC has a higher content of quinone groups on its surface (mmol g–1), indicating higher redox activity and stronger electron acceptance and donation capabilities. After glucose, acting as an electron donor, is oxidized intracellularly, the generated electrons are transferred to the extracellular BC or AQS-BC. The reversible conversion of quinone groups between their oxidized and reduced states can significantly accelerate electron transfer. Compared to dispersed activated sludge treatment systems, the formation of biofilms can shorten the distance for electron transfer between microorganisms and carrier materials, thereby mediating direct interspecies electron transfer during syntrophic metabolic processes. A thicker and denser biofilm was observed on the surface of the AQS-BC samples. Moreover, this biofilm had higher levels of proteins and polysaccharides that were more evenly distributed, resulting in a denser and more stable biofilm structure. Additionally, the microbial activity of the AQS-BC biofilm was higher, and showed stronger redox capabilities with better oxidation–reduction activity. An analysis of the product transformation during decolorization revealed that, compared to BC, the AQS-BC biofilm more readily utilized acetic acid as an electron donor for anaerobic decolorization. After decolorization, acetic acid was further consumed for methanogenesis. Indeed, the presence of AQS did not seem to reduce the accumulation of VFAs, but rather alleviated the inhibition of methanogenesis by RR2. The 16S rRNA sequencing results in this study indicate that, as a carrier, AQS-BC could alter the microbial community structure of its biofilm. This in turn led to a higher relative abundance of various functional microorganisms in the AQS-BC biofilm, resulting in a microbial community structure with enhanced decolorization efficiency. The role of AQS-BC biofilm in enhancing the anaerobic decolorization process are shown in Fig. 7.

Fig. 7
figure 7

The role of AQS-BC biofilm in enhancing the anaerobic decolorization process (Created with BioRender.com)

Full size image

4 Conclusion

This study investigated the anaerobic RR2 decolorization efficiency of a biofilm formed on the surface of the anthraquinone-loaded carrier (AQS-BC) and the mechanism behind its enhanced RR2 decolorization ability. Despite the removal of flocculent sludge, the biofilm system still exhibited stable decolorization capability. Within 48 h, the average decolorization efficiencies of RR2 by the AQS-BC and BC biofilms were 94.06% and 89.46%, respectively. A series of experiments revealed that the optimal conditions for biofilm formation were a 24-h contact time, 1 g VSS/L of sludge inoculation, and a glucose concentration of 1000 mg L–1. The loading of AQS facilitated the degradation of organic matter, increasing the production of VFAs and methane in dye wastewater. Additionally, compared to BC, the AQS-BC biofilm exhibited greater thickness and biomass, a higher EPS content, and superior electrochemical properties. Lastly, AQS-BC effectively increased the abundance and biodiversity of specific functional microbial communities.