Power Generation and Microbial Communities in Microbial Fuel Cell Powered by Tobacco Wastewater

Abstract

The tobacco production process generates a substantial amount of wastewater characterized by high organics and low biodegradability, which poses a significant risk of severe environmental pollution. In order to explore a clean and low-cost technology for tobacco wastewater treatment, this study constructed two-chamber MFCs and investigated the performance of tobacco wastewater treatment and electricity generation capacity at room temperature. The incorporation of carbon sources (e.g., glucose, acetate, propionate, and butyrate) in wastewater could enhance the removal of COD, total nitrogen and ammonia nitrogen in wastewater. After three cycles, the maximum COD removal rate reached 75.97 ± 1.49%, while the maximum total nitrogen removal and ammonia nitrogen removal rates were 46.95 ± 1.77% and 48.31 ± 1.16%, respectively. Meanwhile, the maximum voltage output of 0.67 V was observed, and the maximum power density was 717.04 mW/m². The microbial community analysis revealed that Trichococcus and Acinetobacter were present in high abundance in MFCs, which may play a significant role in electricity generation and wastewater treatment. These results demonstrate that MFC is applicable for tobacco wastewater treatment, providing both theoretical foundation and technical references for the large-scale practical application of MFC technology in tobacco wastewater treatment.

1. Introduction

Tobacco sheets are produced by reprocessing and combining tobacco materials such as tobacco stems, tobacco dust, and low-grade tobacco leaves, which are generated during re-baking or cigarette manufacturing [1]. The use of tobacco sheets in the production of tobacco products is widespread due to their ability to enhance the utilization of tobacco, thereby reducing costs and minimizing the release of nicotine and tar [2]. The primary methods of producing tobacco sheets include roller process, slurry process, and papermaking-reconstituted process. The latter method yields a product with advantageous physical properties, including low density, air permeability, and good combustibility [3]. These attributes make it a preferred choice among cigarette production enterprises and have led to its extensive use in the manufacture of tobacco products, such as traditional cigarettes and novel heat-not-born cigarettes. However, compared to other production methods, the papermaking-reconstituted tobacco sheets production process generates a considerable amount of recalcitrant and highly toxic wastewater that is and significantly detrimental to the environment [4]. Despite the investigation of electrocoagulation [5], coagulation-air floatation [6], and Fenton processes [7], these methods continue to present challenges, including high energy consumption and cost. For example, when using traditional electrochemical technologies or Fenton process for wastewater treatment, some problems such as electrode corrosion [8], secondary iron pollution, and stringent pH requirements may limit their feasibility for large-scale of high-concentration wastewater treatment [9]. Biochemical treatment methods remain the primary approach to tobacco wastewater treatment. However, traditional anaerobic digestion still requires a significant amount of heat input to maintain microbial activity, which undoubtedly increases energy consumption costs and leads to the waste of energy from organic matters in the wastewater [10]. Therefore, it is imperative to develop a new energy-saving, clean, and environmentally friendly tobacco wastewater treatment technology.

Microbial fuel cell (MFC) is a bio-electrochemical technology that can concurrently treat wastewater and recover electrical energy from many kinds of wastewater. Its operational principle involves leveraging microbial metabolic activities to directly convert organic pollutants in wastewater into electrical energy, aligning with the contemporary concept of “waste to wealth” [11,12]. This technology has emerged as a prominent research focus in wastewater treatment in recent years [13]. To date, numerous reports have demonstrated the efficiency of MFCs in treating a range of wastewater, including domestic wastewater, kitchen wastewater, petroleum wastewater, and so on [14,15,16]. These findings substantiate the potential of MFCs as a versatile wastewater treatment technology with promising prospects. Furthermore, MFC possesses the advantages of cleanliness, efficiency, and compatibility with traditional biological wastewater treatment, thereby reducing energy consumption during the wastewater treatment process [17]. However, no research reports on the application of MFC technology in tobacco wastewater treatment have been retrieved in the current literature. The feasibility of using MFC for treating tobacco wastewater remains to be verified.

In this study, we used wastewater from the anaerobic sewage tank of papermaking-reconstituted tobacco sheets as the inocula in two-chamber MFCs. Firstly, different external resistors were used to investigate the impact of external resistance on the enrichment of bioanodes. After all MFCs were successfully started up, the effects of different external carbon sources on the wastewater treatment and electricity generation performance of MFCs were examined. The electrical energy generation and electrochemical characteristics of MFCs were measured, and the tobacco wastewater treatment efficiencies were determined. According to previous studies, the functional microorganisms in the bioanode of MFC are basically bacteria [18]. The microbial community structures differences involved in electrical energy generation at different stages and with different extra carbon sources in actual tobacco wastewater were evaluated by 16S RNA gene pyro sequencing. These findings further revealed that the microbial community in these MFCs was markedly distinct from those observed in other similar systems.

2. Materials and Methods

2.1. Construction of MFCs

To study the bioanode under stable conditions and prevent the infiltration of oxygen, five groups of identical two-chamber MFCs were constructed [19], utilizing carbon brushes (5 × 6 cm²) as anodes and graphite plates (5.2 × 5.2 cm²) as cathodes (Figure 1). The main body of MFC was fabricated using transparent colorless polymethyl methacrylate (PMMA) and assembled in the laboratory. The anode and cathode compartment volumes were 250 mL and 150 mL, respectively. A cation-exchange membrane (CEM; CTCM-1, 9 × 12 cm², China) was installed between the bioanode and cathode to prevent oxygen diffusion and mitigate chemical interference from catholyte constituents. A titanium wire with a diameter of 0.1 cm was threaded through the graphite plate cathode. The two electrodes were spaced 10 cm apart and connected via conductive clips and an external resistor. The experiment was conducted in two stages: Stage I: bioanode enrichment stage; Stage II: tobacco wastewater treatment stage. In the enrichment stage (Stage I), the cathodic solution was 50 mM potassium ferricyanide, simultaneous enrichment was conducted with external resistances of 47 Ω (R1), 100 Ω (R2), 200 Ω (R3), 510 Ω (R4), and 1000 Ω (R5). The inocula of anode was tobacco wastewater from the anaerobic treatment tank of reconstituted tobacco sheet in the papermaking process. Once a significant decrease in voltages in all five MFCs was observed, the solutions in anode and cathode chambers were renewed. When the maximum voltages of MFCs were similar over two cycles, this indicates that MFCs have been successfully started up and entered the tobacco wastewater treatment stage (Stage II). During Stage II, in order to investigate and compare the effects of common commercial extra carbon sources on electrical energy generation and tobacco wastewater treatment efficiency, all external resistances were replaced with 1000 Ω, the anode solution was replaced with real tobacco wastewater supplemented with glucose (R1), acetate (R2), propionate (R3), and butyrate (R4) at a concentration of 0.5 g L⁻¹, and tobacco wastewater with no substrate supplementation (R5) [20,21]. There were a total of three cycles in Stage II, and based on previous research observations, each cycle lasted for nine days. After replacing the solutions in the anode and cathode chambers, all MFCs were aerated with N₂ for 15 min to maintain an anaerobic environment. All reactors were operated at room temperature (23 ± 2 °C).

2.2. Analytical Method

The output voltage on the external resistor was recorded at five-minute intervals by a battery monitoring system (Xinwei, Shenzhen, China) and subsequently processed by the data acquisition software (BTS Client 8.0.0.471). Polarization curves and power density curves were obtained by varying the external resistance through a variable resistance box (ZX21, Dongmao, Shanghai, China). During the polarization process, the external resistance (ranging from 1500 Ω to 10 Ω) was first adjusted and stabilized for 20 min [22]. Subsequently, the output voltage was recorded after an additional 15 min stabilization period using a digital multimeter (UNI-T UT33D, Dongwan, China). The current and power density were calculated according to Ohm’s law.

Cyclic voltammetry (CV) analysis was conducted using an electrochemical workstation (CHI 660E, Chenhua, Shanghai, China) with an anodic biofilm serving as the working electrode, a platinum electrode as the counter electrode, and an Ag/AgCl electrode as the reference electrode. The scanning range was set to −0.8 to 0.8 V, with a scanning speed of 10 mV/s. A multiparameter water quality meter (GLKRUI, GL-200, Shandong, China) was used to determine the chemical oxygen demand (COD) via potassium dichromate oxidation, as well as the concentration of ammonia nitrogen (NH₄⁺-N) and total nitrogen (TN) using Nessler’s reagent spectrophotometric method and Alkalinepotassium persulfate digestion UV spectrophotometric method, respectively. Using a syringe, samples of the anode solution were collected and analyzed on days 0, 1, 2, 3, 6, and 9 of each cycle, respectively. Prior to further analysis, the sample was filtered through a 0.45 μm filter membrane.

Coulombic efficiency refers to the ratio of the number of electrons utilized by microorganisms for electricity generation to the number of electrons theoretically released from the oxidation of substrates, and it can be used to evaluate the energy conversion efficiency of MFCs. Based on the voltage output and COD removal, the coulombic efficiency of MFC was calculated according to the following equation:

$$CE={\displaystyle \frac{M{\int }_0^{T}Idt}{nVF({COD}_0-{COD}_T)}}$$

where M represents the molecular weight of oxygen (32 g/mol), I is the current at the moment of T, T is the cycle time (s), n is the number of electrons exchanged per mole of oxygen (4), V is the volume of the reactor (m³), and F is Faraday’s constant (96,458 C/mol). COD₀ and COD_T denote the initial and final COD values (mg/L), respectively.

2.3. Microbial Community Analysis

To further explore the functional microorganisms in the bioanode, bacterial diversity analyses were conducted when the MFCs were running stably during Stage I and Stage II, respectively. Each sample of bristle material from the bioanode was collected using sterile scissors. The genomic DNA was extracted using the Power Soil DNA Isolation Kit (Mo-Bio, Carlsbad, CA, USA). The V3–V4 region of the 16S rRNA gene was amplified using primers 338F (5′-ACTCCTACGGGGAGGCAGCAG-3′) and 806R (5′-GGACTACHVGGGGTWTCTAAT-3′). The resulting amplicons were sequenced using the Illumina MiSeq PE 300 sequencing platform (Majorbio Biopharm Technology Co., Shanghai, China). A total of 175,497,322 and 150,714,632 bases, along with 418,682 and 357,335 sequences, were obtained in Stage I and Stage II, respectively. The sequences were clustered into operational taxonomic units (OTUs) based on a 97% similarity threshold. Representative sequences from each OTU were selected for classification.

3. Result and Discussion

3.1. Power Generation and Organic Removal

The electrical energy generation characteristics of MFCs were investigated after approximately three weeks (four cycles) after bioanode enrichment. Figure 2a shows that stable output voltages were obtained from MFCs with varying external resistances. R1–R4 obtained low maximum voltage values during the first two cycles, while R5 maintained comparable maximum voltage values from the second to fourth cycle. Prior research has indicated that external resistance may influence the thickness of the biofilm. The reduction in external resistance increased the discrepancy between the anodic potential and the electron acceptor redox potential, leading to a shift in energy from microbial growth to extracellular polymer formation, which in turn prolonged the acclimation period [23]. In the final two cycles, the maximum output voltages of MFCs were maintained at approximately 0.05 V (R1), 0.06 V (R2), 0.20 V (R3), 0.38 V (R4), and 0.56 V (R5), indicating that the MFCs had been successfully activated.

It is well known that the extra carbon sources can serve as additional electron donors to affect wastewater treatment efficiency and power generation capacity [24,25]. In order to further explore the impact of carbon sources on tobacco wastewater treatment in MFCs, a range of carbon sources (glucose (R1), acetate (R2), propionate (R3), and butyrate (R4)) were added as substrates into the tobacco wastewater in Stage II. In Stage II, all MFCs initially exhibited a low maximum voltage of approximately 0.61 V in the initial cycle. In subsequent cycles, the maximum voltage of R1–R4 increased to approximately 0.67 V, while that of R5 remained at approximately 0.62 V.

The polarization and power density curves were obtained by adjusting the external resistance, and the results demonstrated that the open-circuit voltages of R1–R5 were 0.72 V (R1), 0.73 V (R2), 0.71 V (R3), 0.71 V (R4), and 0.68 V (R5), respectively. The maximum power densities were 717.04 mW/m² (R1), 693.60 mW/m² (R2), 532.00 mW/m² (R3), 515.29 mW/m² (R4), and 477.90 mW/m² (R5), respectively (Figure 2b), which are similar to or higher than those obtained in MFCs treating domestic wastewater (422 mW/m²) [26], organic acid fermentation wastewater (543.75 mW/m²) [27] and paper recycling wastewater (501 mW/m²) [28]. As the external resistance was adjusted from 1500 Ω to 100 Ω, the voltage variation tended to decrease linearly, which is a typical ohmic polarization stage. When the external resistance was decreased from 100 Ω to 50 Ω, the mass transfer resistance became the primary resistance to electron transfer, resulting in a significant decline in power density.

The COD value of the original tobacco wastewater was 1645.33 ± 104.36 mg L⁻¹ (R5). The addition of external carbon sources resulted in higher COD removal rate, although higher initial COD values were obtained on the addition of external carbon sources (R1–R4) (Table S3). After discharge, the COD removal efficiencies were 69.66 ± 2.56% (R1), 74.58 ± 0.81% (R2), 73.51 ± 0.85% (R3), 75.97 ± 1.49% (R4), and 56.72 ± 1.97% (R5), respectively. In the initial cycle, the initial COD values were slightly higher than those observed in the subsequent two cycles. This discrepancy may be attributed to the presence of residual organic matter in Stage I. Additionally, the lower voltage output suggested that the microbial community may have been subjected to external environmental influences, with exoelectrogens utilizing the organic matter for growth rather than electrical energy generation. With regard to nitrogen removal, the TN removal efficiencies were 34.88 ± 6.93% (R1), 42.86 ± 4.92% (R2), 45.52 ± 1.69% (R3), 46.95 ± 1.77% (R4), and 37.96 ± 1.18% (R5). The NH₄⁺-N removal efficiencies were 37.65 ± 1.55% (R1), 42.63 ± 1.88% (R2), 46.47 ± 0.40% (R3), 48.31 ± 1.16% (R4), and 37.23 ± 0.81% (R5), respectively. It is well known that glucose and acetate are the most readily utilized substrates by microorganisms for denitrification and power generation. However, in this study, the MFCs fed with propionate (R3) and butyrate (R4) exhibited higher denitrification rates and comparable voltage outputs relative to the MFCs fed with glucose (R1) and acetate (R2). This observation may be attributed to the enrichment of the associated functional microorganisms [20]. Nevertheless, the nitrogen removal efficiency of all MFCs in this study was not entirely satisfactory. This could be attributed to the competitive relationship between electricity-producing microbial communities and functional microorganisms in laboratory-scale experiments. It is hoped that future research will focus on the coupling of anaerobic digestion and MFC technology to adjust the dynamic balance between the bioanode and the wastewater microbial community, thereby enhancing nitrogen removal and electricity generation efficiency.

It is also noteworthy that in Stage II, the voltage output of all the MFCs exhibited a bimodal distribution. It was demonstrated that the bimodal nature was associated with the degradation of organic matter [27]. At the outset of each cycle, the voltage output values exhibited a rapid decline, while the rate of decline slowed and a second peak was observed following the second day. In terms of COD dynamics, all MFCs exhibited a rapid decline in COD values during the initial two days, followed by a deceleration in the decreasing trend after that period (Figure 2d). This pattern aligns with the observed voltage output changes. These results demonstrated that the supplementary carbon sources were effectively utilized by the microorganisms. The microorganisms initially utilized readily available organic matter (e.g., glucose or volatile fatty acids (VFAs)) present in the wastewater, and subsequently consumed more complex organic matter (e.g., cellulose) for power generation. For R1 and R5, there was a rapid decrease in pH from 0 to 2 d, followed by a slower decline after 2 d, and a rebound subsequently. In contrast, for R2–R4 with added organic acids, pH was increased from 0 d to 1 d and then exhibited a similar trend (Figure 2c), indicating that glucose and organic matter in tobacco wastewater may be initially degraded into organic acids, which were then utilized for power generation.

Based on the voltage output and COD degradation values, the coulomb efficiencies of the MFCs in the Stage II were 7.75 ± 1.74% (R1), 6.12 ± 2.05% (R2), 6.10 ± 1.22% (R3), 5.34 ± 1.16% (R4), and 11.16 ± 3.08% (R5), respectively. Overall, the addition of carbon sources did not enhance the coulombic efficiencies of the MFCs (R1–R4) in comparison with R5. This indicated that not all electrons released from the oxidation of the carbon source were transferred to the anode, leading to a decrease in coulombic efficiency. The remaining electrons may have been utilized by microorganisms for their own biomass synthesis or for functions such as denitrification. However, it did result in an improvement in electrical energy generation and COD removal efficiencies.

3.2. Anode Biofilm Electrochemical Activity

The biocatalytic activity of anode biofilm in the stage of tobacco wastewater treatment was investigated by cyclic voltammetry. As shown in Figure 3, distinguishable oxidation peaks emerged at −0.05 V and +0.25 V potentials, while reduction peaks emerged at −0.30 V and +0.1 V potentials. This evidence substantiates the involvement of the redox mediator in the redox reaction and underscores its pivotal role in the anodic electron transfer. Compared to R5, the capacities of R1–R4 were markedly higher, indicating that the incorporation of a carbon source facilitated the growth of exoelectrogens and the electrochemical activity of the anode biofilm.

3.3. Microbial Community Analysis

In order to elucidate the dynamics shifts in the anode microbial community, the microbial community structures of MFCs between the two stages were studied. Figure 4a,c present the microbial community composition of bioanodes at the phylum level, the predominant bacterial groups were identified as Firmicutes and Proteobacteria, representing 67.81–81.87% and 69.29–94.50% of the microbial communities in the Stage I and Stage II, respectively. Previous research also demonstrated that Firmicutes and Proteobacteria are the predominant members of electrochemically active bacteria and play an important role in MFC for electrical energy generation [29]. Furthermore, Synergistota, Bacteroidota, and Chloroflexi were found to be differentially enriched in Stage I and Stage II, respectively. Synergistota plays a significant role in the degradation of amino acids and proteins, which is capable of fermenting glucose to acetate, suggesting that the enrichment of Synergistota may be associated with yeast and beef extract in the nutrient solution [30]. Bacteroidota and Chloroflexi have the capacity of degrading complex organics and thus enhancing wastewater treatment, as has been extensively reported in studies of MFC. The relative abundance of Bacteroidota and Chloroflexi decreased in Stage I, while increased in Stage II. There is currently no evidence to suggest a correlation between their ability to power generation.

Figure 4b,d shows the microbial community composition of MFCs at the genus level. Exiguobacterium dominated the original wastewater, with a relative abundance of 67.41%. It has been reported that Exiguobacterium is capable of degrading various organic compounds and possesses heterotrophic nitrification and denitrification abilities [31], indicating that it played a major role in the degradation and nitrogen removal processes in anaerobic biochemical treatment of wastewater. However, Exiguobacterium was not enriched in Stage I and Stage II. The anodic bacterial communities enriched by the different external resistance in Stage I were found to be generally similar and consisted mainly of Trichococcus, Acinetobacter, and Proteocatella. In Stage II, the bioanode communities were mainly composed of Trichococcus, Brevundimonas, and Acinetobacter. Trichococcus may play an important role in power generation and pollutants degradation. Brevundimonas has been reported to have capabilities of power production and denitrification [32]. As for Acinetobacter, it has been shown to be electrochemically active in the MFC, suggesting that Acinetobacter may be involved in the power generation [33,34]. In R1, Trichococcus had the highest abundance of 72.40%, compared with a lower abundance of 5.44% in R4, while the exoelectrogens Streptococcus [35] and Proteiniclasticum had abundances of 13.06% and 12.11%, respectively. Proteiniclasticum has the capacity to degrade proteins to acetate, which may play an important role in denitrification in the MFC [36]. In addition, Flavobacterium and Stenotrophomonas were enriched in R2 and R4, R2 and R3, respectively. It was reported that Flavobacterium and Stenotrophomonas have the ability of denitrification and complex organic matters degradation, and their increased abundance contributes to wastewater treatment [37,38,39]. Exoelectrogens are the core of MFC and determine the performance of MFC in terms of power generation, and microorganisms capable of generating electric currents have always been of interest to researchers [18]. To the best of our knowledge, the exoelectrogens found so far mainly belong to Proteobacteria. For example, in the vast majority of studies on wastewater treatment of MFC, the highest abundance in the bioanode is of the Geobacter or Desulfobulbaceae, which act as typical exoelectrogens and play a major role in MFCs for electrical energy generation [40]; however, their presence was not detected in this study.

The presence of Trichococcus with high abundance in both stages and all MFCs has been presented in this study, indicating its adaptation to tobacco wastewater and its important role in MFC, which piqued our interest. Trichococcus, which is a Gram-positive bacterium belonging to Firmicutes, is capable of fermenting a wide range of carbohydrates, including glucose [41], which explains the high abundance of Firmicutes in the MFC, and the addition of glucose a carbon source leading to a high enrichment of Trichococcus. Indeed, the presence of Trichococcus in microbial electrochemical systems is not uncommon, but it was usually present in lower abundance in association with other exoelectrogens (e.g., Geobacter), which were thought to ferment glucose for the production of acetate as well as denitrification [42,43]. Despite the lack of direct evidence for the ability of power generation in Trichococcus, however, previous studies have shown that Trichococcus possesses Fe (III) reducing ability and may engage in direct interspecies electron transfer with Methanosaeta, which is very similar to the characteristics of exoelectrogens [44,45]. In addition, Trichococcus possesses genes in extracellular electron transfer (EET) locus, including dmkA, fmnB, and ndh2 [46]. These results imply that Trichococcus has electroproduction capacity. R1 with highly enriched Trichococcus still exhibited excellent electrochemical performance. Regarding the EET ability of Trichococcus, further confirmation is needed, such as by performing pure culture experiments.

3.4. Bacterial Function Prediction

To further examine the distinctions in function between communities, metabolic pathways were predicted based on community composition using PICRUST2 and in combination with the Kyoto Encyclopedia of Genes and Genomes (KEGG) catalog. Metabolism was the dominant pathway at KEGG level 1, which accounted for 74.2–76.28%. The top five rankings at KEGG Level 2 were global and overview maps (38.36–39.23%), carbohydrate metabolism (9.14–11.13%), amino acid metabolism (6.40–7.88%), energy metabolism (3.84–4.11%), and membrane transport (3.37–5.09%), respectively (Figure 5a). As shown in Figure 5b, at KEGG Level 3, the relative abundance of nitrogen metabolism (ko00910) was higher in R2-R4, with R4 (0.31%) being the highest, followed by R3 (0.30%) and R2 (0.27%). These suggested that the addition of VFAs increased denitrogenation-related functional bacteria, indicating high denitrogenation efficiency. Compared to R5, R2-R4 fed with VFAs showed upregulation of fatty acid degradation (ko00071), glyoxylate and dicarboxylate metabolism (ko00630), and TCA cycle (ko00020). In the case of R1 fed with glucose, the TCA cycle was downregulated, while pathways associated with carbohydrate metabolism were upregulated, including glycolysis/gluconeogenesis (ko00010), the pentose phosphate pathway (ko00030), and amino sugar and nucleotide sugar metabolism (ko00520). These results indicated that glucose enhances the activity of fermentation bacteria and produces nicotinamide adenine dinucleotide (NADH) through fermentation. In contrast, VFAs enter the TCA cycle and the glyoxylate cycle primarily through fatty acid degradation, generating energy for pollutant removal and power generation [47].

4. Conclusions

This study successfully demonstrated the feasibility of MFC for tobacco wastewater treatment and promising potential for coupling with traditional biochemical treatment of tobacco wastewater. The addition of extra carbon sources could increase the voltage output, COD, and nitrogen removal efficiency, while it may have enriched other functional microorganisms; however, it did not enhance the coulombic efficiency. The maximum power density supported by glucose was 717.04 mW/m² in the tobacco wastewater treatment stage. This was higher than that of the MFCs with acetate (693.60 mW/m²), propionate (532.00 mW/m²), and butyrate (515.29 mW/m²), and without extra carbon sources (477.90 mW/m²) in tobacco wastewater. The highest TN and NH₄⁺-N removal efficiencies were 46.95 ± 1.77% and 48.31 ± 1.16%, respectively. The microbial community analysis indicated that anodic biofilms, predominantly Trichococcus and Acinetobacter, may play a pivotal role in power generation and pollutants removal in the MFC for treating tobacco wastewater. Trichococcus may serve as a novel exoelectrogen for power generation, demonstrating remarkable performance in the MFC at high abundance (72.40%). It is expected that further research will be conducted to verify the electricity-producing performance and study the functions of Trichococcus.