- Email: [email protected]
Enhancing anaerobic digestion of complex organic waste with carbon-based conductive materials
Accepted Manuscript Enhancing anaerobic digestion of complex organic waste with carbon-based conductive materials Yan Dang, Dawn E. Holmes, Zhiqiang Zhao, Trevor L. Woodard, Yaobin Zhang, Dezhi Sun, Li-Ying Wang, Kelly P. Nevin, Derek R. Lovley PII: DOI: Reference:
S0960-8524(16)31252-4 http://dx.doi.org/10.1016/j.biortech.2016.08.114 BITE 17017
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
25 July 2016 29 August 2016 30 August 2016
Please cite this article as: Dang, Y., Holmes, D.E., Zhao, Z., Woodard, T.L., Zhang, Y., Sun, D., Wang, L-Y., Nevin, K.P., Lovley, D.R., Enhancing anaerobic digestion of complex organic waste with carbon-based conductive materials, Bioresource Technology (2016), doi: http://dx.doi.org/10.1016/j.biortech.2016.08.114
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Enhancing anaerobic digestion of complex organic waste with carbon-based conductive materials Yan Dang1,2*, Dawn E Holmes1,3, Zhiqiang Zhao1,4, Trevor L Woodard1, Yaobin Zhang1,4, Dezhi Sun1,2, Li-Ying Wang1, Kelly P Nevin1, and Derek R Lovley1 1
Department of Microbiology, University of Massachusetts Amherst, Morril IV N Science
Center, Amherst, MA 01003 2
Beijing Key Laboratory for Source Control Technology of Water Pollution, College of
Environmental Science & Engineering, Beijing Forestry University, 35 Tsinghua East Road, Beijing 100083, China 3
Department of Physical and Biological Sciences, Western New England University, 1215
Wilbraham Rd, Springfield, MA, 01119 4
Key Laboratory of Industrial Ecology and Environmental Engineering (Dalian University
of Technology), Ministry of Education, School of Environmental Science and Technology, Dalian University of Technology, Dalian 116024, China Corresponding Author *Yan Dang. Department of Microbiology, University of Massachusetts 9 Amherst, 639 North Pleasant Street, Amherst, MA 01003, USA.
1
Email: [email protected]; [email protected]
Abstract. The aim of this work was to study the methanogenic metabolism of dog food, a food waste surrogate, in laboratory-scale reactors with different carbon-based conductive materials. Carbon cloth, carbon felt, and granular activated carbon all permitted higher organic loading rates and promoted faster recovery of soured reactors than the control reactors. Microbial community analysis revealed that specific and substantial enrichments of Sporanaerobacter and Methanosarcina were present on the carbon cloth surface. These results, and the known ability of Sporanaerobacter species to transfer electrons to elemental sulfur, suggest that Sporanaerobacter species can participate in direct interspecies electron transfer with Methanosarcina species when carbon cloth is available as an electron transfer mediator.
Keywords: Anaerobic digestion; organic wastes; carbon-based conductive material; direct interspecies electron transfer; methanogenesis
1. Introduction Conversion of complex organic wastes to methane is an attractive treatment method but has the limitations of long start-up times, slow processing of wastes, and susceptibility to disruptions by organic overloading (Chen et al., 2008; Li et al., 2011; Mata-Alvarez et al., 2000; McCarty, 1964). Studies with simple substrates such as ethanol, propionate, and
2
butyrate have suggested that promoting direct interspecies electron transfer (DIET) with conductive materials may accelerate and stabilize the conversion of these substrates to methane (Cruz Viggi et al., 2014; Lee et al., 2016; Li et al., 2015; Liu et al., 2012; Zhao et al., 2015). However, the potential role of DIET in the conversion of more complex wastes to methane has not been intensively investigated. The concept of DIET was first developed with co-cultures of the Fe(III)-respiring bacteria Geobacter metallireducens and Geobacter sulfurreducens (Summers et al., 2010), but some methanogens can also participate in DIET. For example, ethanol metabolism to methane via DIET was documented in mixed microbial communities associated with anaerobic digesters (Morita et al., 2011; Rotaru et al., 2014b) as well as in defined cocultures with Geobacter metallireducens and either Methanosaeta harundinacea (Rotaru et al., 2014b) or Methanosarcina barkeri (Rotaru et al., 2014a). Studies also showed that addition of conductive carbon materials such as granular activated carbon (GAC) (Liu et al., 2012; Rotaru et al., 2014a), biochar (Chen et al., 2014b), or carbon cloth (Chen et al., 2014a) could accelerate ethanol metabolism to methane via DIET in co-cultures with G. metallireducens and M. barkeri. Propanol and butanol metabolism via DIET was also documented in co-cultures of G. metallireducens and either M. harundinacea or M. bakeri (Wang et al., 2016). However, these substrates were only partially oxidized to propionate and butyrate, even when conductive materials were included. In other studies, it was suggested that the conductive mineral magnetite, which is known to promote DIET (Kato et al., 2012b; Liu et al., 2015), could stimulate the complete oxidation of propionate (Cruz Viggi et al., 2014) and butyrate
3
(Li et al., 2015) via DIET with a methanogen, but the microorganisms involved were not identified and DIET was only inferred not directly demonstrated. These results indicate that it might be difficult for propionate and butyrate, important intermediates in anaerobic digestion, to be completely metabolized via DIET under methanogenic conditions (Wang et al., 2016). Acetate metabolism via DIET has not yet been demonstrated in methanogenic culture studies. However, co-cultures of G. metallireducens and G. sulfurreducens could completely oxidize acetate when fumarate was provided as the sole terminal electron acceptor (Wang et al., 2016). In addition, when rice paddy microcosms or anaerobic digesters were supplemented with conductive materials, acetate oxidation by Geobacter species and methane production rates were significantly stimulated (Kato et al., 2012a; Lee et al., 2016), suggesting that acetate was being metabolized via DIET in these environments. While studies have shown that the incorporation of conductive carbon materials into continuous flow reactors can enhance the conversion of ethanol (Zhao et al., 2015) and acetate (Lee et al., 2016) to methane and permit higher loading rates without digester failure (Xu et al., 2015; Zhao et al., 2015), there has been less investigation into the possibility that these conductive materials may also help stabilize methanogenic degradation of more complex organic materials. The purpose of the studies reported here was to investigate whether carbon-based conductive materials could enhance the conversion of complex wastes to methane in
4
anaerobic digesters. Therefore, laboratory-scale reactors supplemented with four different carbon-based materials (carbon cloth, carbon felt, GAC, and graphite) were fed with increasing concentrations of complex waste and methane conversion rates and volatile fatty acid (VFA) concentrations were monitored throughout the experiment. In addition, microbial communities associated with various reactor conditions were compared.
2. Materials and Methods 2.1 Reactor design and incubation The eighteen reactors for this study were built from glass jars with a working volume of 2 liters and sealed with rubber stoppers. Each reactor had an influent/effluent port, a gas sampling port and a gas outlet port connected to a 10-liter gas collection bag. All reactors were incubated at 37 °C in the dark. The seed for the reactors was sludge obtained from the anaerobic digester of a municipal wastewater treatment plant in Pittsfield, Massachusetts. The total suspended sludge (TSS) of the seed sludge was about 13,000 mg/l and the ratio of volatile suspended sludge (VSS) to TSS was 0.72. It was stored under anaerobic conditions at 4 °C prior to use. To initiate the reactors they were each inoculated with 500 ml of seed sludge and 400 ml of deionized water. Five different materials (with similar conductive material geometric surface areas of ~ 1000 cm2) were added to triplicate reactors as follows: (1)
5
three pieces of carbon cloth (Zoroflex; buyactivatedcharcoal.com; 8×20×0.1 cm3; volume of 48 cm3 and geometric surface area of 977 cm2); (2) three pieces of carbon felt (AlfaAesar; MA; 8×17.5×0.6 cm3; volume of 252 cm3 and geometric surface area of 932 cm2); (3) 100 g of granular activated carbon (Sigma; Missouri; 8-20 mesh; volume of ca. 120 cm3 and geometric surface area of ca. 900 cm2); (4) 132 graphite rods (Sigma–Aldrich; Missouri; 13mm ×12 mm, volume of ca. 200 cm3 and surface area of ca.1000 cm2 ); or (5) three pieces of polyester cloth (Heavy Duty Fabric and CORDURA; http://heavydutyfabric.com; 8×20×0.1 cm3, volume of 48 cm3 and surface area of 977 cm2). One set of triplicate reactors received no materials. The initial composition of the artificial complex waste amendment was comprised of (per liter): commercial dog food (The Honest Kitchen; CA; total chemical oxygen demand (COD) of 1,336 ± 61 mg/g), 12.9 g; urea, 230 mg; KH2PO4, 110 mg; K2HPO4, 170 mg; Na2SO4, 50 mg; MgCl26H2O,100 mg; NiCl26H2O, 5 mg; CoCl26H2O, 6 mg and 2 ml each of previously described (Morita et al., 2011; Zhao et al., 2015) trace element and vitamin solutions. This dog food feedstock was blended for 5 min with a blender before being fed to the reactors. As the organic loading rate (OLR) of the reactors was increased, the dog food concentration in the artificial waste was increased stepwise from the initial concentration of 12.9 g/l to 25.8 g/l, 51.6 g/l, 64.5 g/l and 77.4 g/l. The pH of the artificial waste was adjusted to 7.0 with NaOH. Each reactor was fed with 100 ml of the artificial waste once a day to maintain a hydraulic retention time (HRT) of 10 days. The reactors were then shaken at 100 rpm for
6
21 h. Shaking was then stopped for 3 hours to allow sedimentation and 100 ml of effluent was discarded. This cycle was repeated for 10 days for each OLR. 2.2 Analytical methods TSS and VSS were analyzed in accordance with Standard Methods for the Examination of Water and Wastewater (APHA, 2012). COD was determined with Hach’s method 8000 (Hach DR/890 colorimeter procedures manual) (Morita et al., 2011). VFAs were measured with a high-performance liquid chromatograph (Bio-Rad; Hercules; California) (Nevin et al., 2008). pH was monitored with a pH analyzer (UB-10; Denver Instrument; Denver). The gas volume in the 10 L-gas-sampling bag was measured every 24 h with a digital mass flow meter (FMA4000; Omega; USA). The composition of biogas was analyzed with a gas chromatograph with a flame ionization detector (GC-8A; Shimadzu; Japan). Methane conversion rates were evaluated based on the established stoichiometric value of 15.6 mmol (CH4)/g (COD degraded) equal to 100 % efficiency (Timur & Özturk, 1999). 2.3 DNA extraction and barcoded amplicon sequencing Sludge samples (1.0 ml) and cloth material (1×1 cm2) were collected at the end of operation when the reactors were still in stable condition. The cloths were washed with phosphate-buffered saline (PBS; 0.13 M NaCl and 10 mM Na2HPO4 at pH 7.2) to remove any residual suspended sludge. All samples were immediately frozen in liquid nitrogen and
7
stored at -80°C. For DNA extraction, frozen samples were ground into a fine powder with a mortar and pestle before treatment with the BIO101 FastDNA soil kit (MP Biomedicals, Ohio) according to the manufacturer’s instructions. DNA was quantified with a NanoDrop spectrophotometer (Thermo, Fisher Scientific, Delaware). Extracted DNA was analyzed at the MR DNA sequencing facility (Shallowater, Texas). The primers 27F and 349F were used to sequence bacterial and archaeal 16S rRNA genes on a 454 platform (Roche). Sequence data was deposited in the Genbank database and assigned accession no. KT936838-KT937136.
3. Results and Discussion 3.1 Performance of the reactors with and without conductive carbon materials In the initial start up period, as the OLR was increased from 1.6 kg COD/(m3 day) to 3.3 and then to 6.7 kgCOD/(m3 day), all of the reactors performed in a similar manner (ttest, P < 0.05) with stable operation, high methane conversion rates (Figure 1A and 1B), stable pH (Figure 1C), and low concentrations of VFAs (Figure 2). Methane conversion rates from all six reactor systems exceeded 86% at an OLR of 6.7 kg COD/(m3 day), and the reactors supplemented with carbon cloth, GAC, and carbon felt only produced slightly more methane (109.6±3.0, 106.6±9.4 and 110.8±5.8 mmol/d; n=3) than graphite, polyester cloth or non-amended control reactors (101.2±3.2, 100.0±4.7 and 100.1±3.4 mmol/d; n=3).
8
However, when the OLR was increased to 8.5 kg COD/(m3 day), the performance of the reactors without conductive materials (i.e. non-amended controls or polyester cloth) or graphite rods declined (Figure 1). Methane production rates dropped to 43% or lower, VFAs accumulated to nearly 100 mM, and pHs sank to < 6.0 (Figure 1C). In contrast, reactors containing carbon cloth, GAC, or carbon felt continued to run stably and efficiently; total VFA concentrations were only 18 mM, 42 mM and 45 mM, respectively (Figure 2), and methane conversion rates were >85% (Figure 1B). A further increase of the OLR to 10.3 kg COD/(m3 day) immediately disrupted the methane conversion in the reactors with carbon felt, and the performance of the reactors with carbon cloth or GAC declined soon thereafter (Figure 1). The concentrations of total VFAs in the carbon cloth and carbon felt reactors reached ca. 90 mM (Figure 2) and methane production rates dropped significantly for all conditions (Figure 1). The reactors with GAC performed best at this OLR, continuing to convert 66 % of added organic load to methane, with total VFAs of ca. 60 mM and a pH of 7.1. 3.2 Reactor recovery In order to determine whether the failed non-amended control, graphite-amended, or polyester cloth-amended reactors could recover if the OLR was decreased, the OLR was reduced to 3.3 kg COD/(m3 day) on day 35 of operation (Figure 3). Methane production in polyester cloth-amended reactors began to slowly improve, but reactors with graphite rods or the non-amended control reactors were unable to recover (Figure 3). Therefore, after 13 days the OLR was further reduced to 0.85 kg COD/(m3 day). All three types of reactors
9
eventually recovered at this loading rate, substantially reducing VFA levels and steadily producing methane. The reactors with polyester cloth recovered faster than those with graphite rods, and the reactors with no materials were the slowest to recover (Figure 3). On day 45, the load to the reactors with GAC, carbon cloth, and carbon felt was also decreased to 3.3 kg COD/(m3 day). This load was further reduced to 0.85 kg COD/(m3 day) 13 days later, in order to make conditions for all 6 reactor-types comparable (Figure 4). All of these amended reactors recovered within 25 days, significantly faster than the graphite, polyester cloth or non-amended control reactors (Figure 4). 3.3 Microbial community analysis Reactor microbial community structure was also evaluated to gain insight into microbial factors that might lead to improved reactor performance. Although the reactors with GAC performed best, during operation GAC disintegrated into tiny carbon particles suspended in the sludge, making it difficult to differentiate between microorganisms attached to the GAC surface and those suspended in the sludge. Therefore, we focused on the microbial community attached to the carbon cloth, which also greatly improved reactor performance, and compared this community to those attached to non-conductive polyester cloth and in bulk sludge from the polyester cloth amended and non-amended control reactors. Bacterial and methanogenic sequences enriched on the surface of the carbon cloth differed significantly from the other samples (Figure 5). For example, bacteria in the genus Sporanaerobacter were much more abundant on the surface of the carbon cloth than on the
10
surface of the polyester, accounting for 34 % of the 16S rRNA gene sequences recovered (Figure 5A). In contrast, the proportion of Sporanaerobacter sequences attached to the surface of the polyester cloth and in the bulk sludge of the control and the polyester cloth reactors was < 1 % and Sporanaerobacter only accounted for 2 % of the sequences in the bulk sludge of the carbon cloth reactors (Figure 5A). These results suggest that the carbon cloth enhanced the growth of Sporanaerobacter. The best characterized Sporanaerobacter, S. acetigenes metabolizes fermentable substrates with the reduction of elemental sulfur (Hernandez-Eugenio et al., 2002). Although its ability to transfer electrons to conductive materials has not yet been evaluated, the capability of elemental sulfur reduction is very commonly linked to the ability to transfer electrons to other extracellular electron acceptors such as Fe(III) oxides or electrodes (Lovley et al., 2004). Therefore, it seems likely that the Sporanaerobacter species that were highly enriched on the carbon cloth were using the carbon cloth as an electrical connection to electron-accepting methanogens. Consistent with this concept was the finding that the methanogenic community on the carbon cloth was very distinct from any of the other environments sampled (Figure 5B). Methanosarcina species were only predominant on the carbon cloth surface, accounting for 68 % of the methanogen sequences recovered (Figure 5B). In all of the other environments Methanosaeta species were the predominant methanogens, even in the bulk sludge of the carbon cloth reactors. Studies have shown that both Methanosaeta (Rotaru et al., 2014b) and Methanosarcina (Rotaru et al., 2014a) species are capable of DIET when the cells are in direct contact with their partners through
11
biological electrical connections. However, DIET stimulated by conductive materials such as GAC (Liu et al., 2012), carbon cloth (Chen et al., 2014a), or biochar (Chen et al., 2014b) has only been documented in co-culture studies done with Geobacter metallireducens and Methanosarcina barkeri. Attempts to stimulate DIET with conductive carbon-based materials in co-cultures in which Methanosaeta haurundinacea replaced M. barkeri have been unsuccessful (see supplementary material, Figure S1). Thus, the enrichment of Methanosarcina on the cloth is consistent with a source of electrons being delivered by bacteria with the carbon cloth as a conduit. S. acetigenes metabolizes glucose solely to acetate and carbon dioxide when sulfur is available as an electron acceptor, but produces hydrogen when electron acceptors are not provided (Hernandez-Eugenio et al., 2002). In the polyester cloth and control reactors, methanogens known to specialize in the utilization of hydrogen as an electron donor (Methanomicrobiales and Methanobacteriales) were abundant in the bulk sludge, but not in the carbon cloth reactors. This finding is also consistent with our hypothesis that Sporanaerobacter can take advantage of the carbon cloth-mediated electrical conduit to Methanosarcina species to metabolize fermentable substrates with electron transfer to Methanosarcina rather than using a less energetically favorable fermentative metabolism to generate hydrogen. The proposed conversion of fermentable substrates primarily to acetate and carbon dioxide with electron transfer to Methanosarcina would be expected to result in the production of less propionate, butyrate and hydrogen, and limit the need for syntrophic metabolism of the VFAs. This is an important consideration because the accumulation of
12
propionate appeared to be associated with the inhibition of methane production at higher OLR (Figure 2). These results are consistent with previous studies that have shown that propionate accumulates in destabilized digester systems (Inanc et al., 1996; Li et al., 2012; McCarty & Smith, 1986; Ye et al., 2011). Carbon cloth was highly effective in maintaining low propionate concentrations at all but the highest organic loading rates (Figure 2). Methanosarcina may also benefit from accepting electrons from the carbon cloth, because the conversion of acetate to methane yields little energy and Methanosarcina typically grow slowly on acetate. Electrons obtained via DIET might enhance their metabolism and even increase their ability to produce methane by acetate decarboxylation. Improved growth and metabolism of Methanosarcina might account for lower acetate concentrations in reactors amended with carbon cloth compared to reactors without conductive materials. There was also an enrichment of Enterococcus on the carbon cloth, which comprised 10 % of the 16S rRNA gene sequences recovered (Figure 5A). In contrast, Enterococcus sequences accounted for less than 1 % of the sequences in the samples from the polyester and control reactors and only 1.5 % of the sequences in the bulk sludge of the carbon cloth reactor (Figure 5A). Enterococcus gallinarum, which metabolizes a diversity of fermentable substrates, is capable of extracellular electron transfer to Fe(III) and is electrochemically active (Kim et al., 2005). Thus, the Enterococcus species enriched on the carbon cloth may also have been contributing to the metabolism of fermentable substrates coupled with electron transfer to Methanosarcina via carbon cloth-mediated DIET.
13
There was no apparent enrichment on the carbon cloth of microorganisms that might be involved in syntrophic metabolism of short-chain fatty acids or ethanol via DIET. This included a lack of Geobacter species. These are the first results that provide evidence that DIET plays a role in stimulating and stabilizing anaerobic digestion of complex organic material by altering the metabolism of fermentable substrates. Isolation and characterization of Sporanaerobacter species enriched on the carbon cloth will be required to confirm that these organisms can participate in DIET mediated by carbon cloth, however their high abundance on the carbon cloth associated with the specific enrichment of Methanosarcina suggests that these organisms utilize a significantly different type of metabolism in the presence of carbon cloth than they do when grown with non-conductive polyester cloth. The results further demonstrate the potential benefits of conductive carbon materials in anaerobic digestion. Permanent incorporation of carbon cloth or similar conductive materials into single chamber anaerobic digesters is expected to be feasible and preferable to the use of particulate conductive materials that need to be continually replenished (Zhao et al., 2015). Graphite rods represent another conductive material that could serve as a permanent component of digesters. Graphite serves as an excellent surface for the growth of current-producing biofilms in microbial fuel cells (Franks & Nevin, 2010; Nevin et al., 2008), but we found that it was ineffective in promoting or stabilizing methanogenesis in the reactors. Additional studies are required to better understand this result.
14
It is clear from the results presented here, as well as many previous studies (Chen et al., 2014a; Chen et al., 2014b; Cruz Viggi et al., 2014; Kato et al., 2012a; Kato et al., 2012b; Lee et al., 2016; Li et al., 2015; Liu et al., 2012; Liu et al., 2015; Xu et al., 2015; Zhao et al., 2015), that conductive materials can enhance methane production from a diversity of organic substrates. However, the microbiology of this process is still poorly understood. Geobacter species are the only organisms that have been shown to donate electrons to electron accepting partners via DIET in culture studies. Yet in the studies reported here, as well as other recent studies (Baek et al., 2015; Beckmann et al., 2016; Luo et al.; Xu et al., 2015; Yamada et al., 2015; Zhao et al., 2016) Geobacter species were not enriched with conductive material amendments, suggesting that other microorganisms may participate in DIET. Further elucidation of which microbes can participate in DIET in methanogenic environments will be important for informed design of improved methanogenic digesters based on the DIET concept. 4. Conclusions These results demonstrate that some carbon conductive materials such as carbon cloth and GAC can stimulate methane production from complex organic material, permit higher OLRs, and promote faster recovery of soured reactors. Analysis of microbial community composition suggested a new concept for DIET. It appeared that conductive carbon cloth specifically enriched for Sporanaerobacter and Enterococcus species that could metabolize fermentable substrates and transfer those electrons to Methanosarcina species. This is a surprising finding because previous studies had only suggested the
15
possibility of accelerating the syntrophic metabolism of fermentation products when promoting DIET.
Acknowledgment This research was supported by the Environmental Research & Education Foundation. The first author Yan Dang acknowledges support from the China Scholarship Council (CSC) program (2014).
References 1. APHA. 2012. Standard methods for the examination of water and wastewater. American Public Health Association Washington, DC. 2. Baek, G., Kim, J., Cho, K., Bae, H., Lee, C. 2015. The biostimulation of anaerobic digestion with (semi) conductive ferric oxides: their potential for enhanced biomethanation. Applied microbiology and biotechnology, 99(23), 10355-10366. 3. Beckmann, S., Welte, C., Li, X., Oo, Y.M., Kroeninger, L., Heo, Y., Zhang, M., Ribeiro, D., Lee, M., Bhadbhade, M. 2016. Novel phenazine crystals enable direct electron transfer to methanogens in anaerobic digestion by redox potential modulation. Energy & Environmental Science. 4. Chen, S., Rotaru, A.-E., Liu, F., Philips, J., Woodard, T.L., Nevin, K.P., Lovley, D.R. 2014a. Carbon cloth stimulates direct interspecies electron transfer in syntrophic cocultures. Bioresource technology, 173, 82-86. 5. Chen, S., Rotaru, A.-E., Shrestha, P.M., Malvankar, N.S., Liu, F., Fan, W., Nevin, K.P., Lovley, D.R. 2014b. Promoting interspecies electron transfer with biochar. Scientific reports, 4. 6. Chen, Y., Cheng, J.J., Creamer, K.S. 2008. Inhibition of anaerobic digestion process: a review. Bioresource technology, 99(10), 4044-4064. 7. Cruz Viggi, C., Rossetti, S., Fazi, S., Paiano, P., Majone, M., Aulenta, F. 2014. Magnetite particles triggering a faster and more robust syntrophic pathway of methanogenic propionate degradation. Environmental science & technology, 48(13), 7536-7543. 8. Franks, A.E., Nevin, K.P. 2010. Microbial fuel cells, a current review. Energies, 3(5), 899-919. 9. Hernandez-Eugenio, G., Fardeau, M.-L., Cayol, J.-L., Patel, B.K., Thomas, P., Macarie, H., Garcia, J.-L., Ollivier, B. 2002. Sporanaerobacter acetigenes gen. nov., sp. nov., a novel acetogenic, facultatively sulfur-reducing bacterium. International journal of
16
systematic and evolutionary microbiology, 52(4), 1217-1223. 10. Inanc, B., Matsui, S., Ide, S. 1996. Propionic acid accumulation and controlling factors in anaerobic treatment of carbohydrate: effects of H 2 and pH. Water Science and Technology, 34(5), 317-325. 11. Kato, S., Hashimoto, K., Watanabe, K. 2012a. Methanogenesis facilitated by electric syntrophy via (semi) conductive iron‐oxide minerals. Environmental microbiology, 14(7), 1646-1654. 12. Kato, S., Hashimoto, K., Watanabe, K. 2012b. Microbial interspecies electron transfer via electric currents through conductive minerals. Proceedings of the National Academy of Sciences, 109(25), 10042-10046. 13. Kim, G., Hyun, M., Chang, I., Kim, H., Park, H., Kim, B., Kim, S., Wimpenny, J., Weightman, A.J. 2005. Dissimilatory Fe (III) reduction by an electrochemically active lactic acid bacterium phylogenetically related to Enterococcus gallinarum isolated from submerged soil. Journal of applied microbiology, 99(4), 978-987. 14. Lee, J.-Y., Lee, S.-H., Park, H.-D. 2016. Enrichment of Specific Electro-Active Microorganisms and Enhancement of Methane Production by Adding Granular Activated Carbon in Anaerobic Reactors. Bioresource Technology. 15. Li, H., Chang, J., Liu, P., Fu, L., Ding, D., Lu, Y. 2015. Direct interspecies electron transfer accelerates syntrophic oxidation of butyrate in paddy soil enrichments. Environmental microbiology, 17(5), 1533-1547. 16. Li, J., Ban, Q., Zhang, L., Jha, A.K. 2012. Syntrophic Propionate Degradation in Anaerobic Digestion: A Review. International Journal of Agriculture & Biology, 14(5). 17. Li, Y., Park, S.Y., Zhu, J. 2011. Solid-state anaerobic digestion for methane production from organic waste. Renewable and Sustainable Energy Reviews, 15(1), 821-826. 18. Liu, F., Rotaru, A.-E., Shrestha, P.M., Malvankar, N.S., Nevin, K.P., Lovley, D.R. 2012. Promoting direct interspecies electron transfer with activated carbon. Energy & Environmental Science, 5(10), 8982-8989. 19. Liu, F., Rotaru, A.E., Shrestha, P.M., Malvankar, N.S., Nevin, K.P., Lovley, D.R. 2015. Magnetite compensates for the lack of a pilin‐associated c‐type cytochrome in extracellular electron exchange. Environmental microbiology, 17(3), 648-655. 20. Lovley, D.R., Holmes, D.E., Nevin, K.P. 2004. Dissimilatory fe (iii) and mn (iv) reduction. Advances in microbial physiology, 49, 219-286. 21. Luo, C., Lü, F., Shao, L., He, P. 2015. Application of eco-compatible biochar in anaerobic digestion to relieve acid stress and promote the selective colonization of functional microbes. Water research, 68, 710-718. 22. Mata-Alvarez, J., Mace, S., Llabres, P. 2000. Anaerobic digestion of organic solid wastes. An overview of research achievements and perspectives. Bioresource technology, 74(1), 3-16. 23. McCarty, P.L. 1964. Anaerobic waste treatment fundamentals. Public works, 95(9), 107-112. 24. McCarty, P.L., Smith, D.P. 1986. Anaerobic wastewater treatment. Environmental Science & Technology, 20(12), 1200-1206. 25. Morita, M., Malvankar, N.S., Franks, A.E., Summers, Z.M., Giloteaux, L., Rotaru, A.E., Rotaru, C., Lovley, D.R. 2011. Potential for direct interspecies electron transfer in
17
methanogenic wastewater digester aggregates. MBio, 2(4), e00159-11. 26. Nevin, K.P., Richter, H., Covalla, S., Johnson, J., Woodard, T., Orloff, A., Jia, H., Zhang, M., Lovley, D. 2008. Power output and columbic efficiencies from biofilms of Geobacter sulfurreducens comparable to mixed community microbial fuel cells. Environmental microbiology, 10(10), 2505-2514. 27. Rotaru, A.-E., Shrestha, P.M., Liu, F., Markovaite, B., Chen, S., Nevin, K.P., Lovley, D.R. 2014a. Direct interspecies electron transfer between Geobacter metallireducens and Methanosarcina barkeri. Applied and environmental microbiology, 80(15), 4599-4605. 28. Rotaru, A.-E., Shrestha, P.M., Liu, F., Shrestha, M., Shrestha, D., Embree, M., Zengler, K., Wardman, C., Nevin, K.P., Lovley, D.R. 2014b. A new model for electron flow during anaerobic digestion: direct interspecies electron transfer to Methanosaeta for the reduction of carbon dioxide to methane. Energy & Environmental Science, 7(1), 408-415. 29. Summers, Z.M., Fogarty, H.E., Leang, C., Franks, A.E., Malvankar, N.S., Lovley, D.R. 2010. Direct exchange of electrons within aggregates of an evolved syntrophic coculture of anaerobic bacteria. Science, 330(6009), 1413-1415. 30. Timur, H., Özturk, I. 1999. Anaerobic sequencing batch reactor treatment of landfill leachate. Water Research, 33(15), 3225-3230. 31. Wang, L.-Y., Nevin, K.P., Woodard, T.L., Mu, B.-Z., Lovley, D.R. 2016. Expanding the Diet for DIET: Electron Donors Supporting Direct Interspecies Electron Transfer (DIET) in Defined Co-Cultures. Frontiers in microbiology, 7. 32. Xu, S., He, C., Luo, L., Lü, F., He, P., Cui, L. 2015. Comparing activated carbon of different particle sizes on enhancing methane generation in upflow anaerobic digester. Bioresource technology, 196, 606-612. 33. Yamada, C., Kato, S., Ueno, Y., Ishii, M., Igarashi, Y. 2015. Conductive iron oxides accelerate thermophilic methanogenesis from acetate and propionate. Journal of bioscience and bioengineering, 119(6), 678-682. 34. Ye, J., Mu, Y., Cheng, X., Sun, D. 2011. Treatment of fresh leachate with high-strength organics and calcium from municipal solid waste incineration plant using UASB reactor. Bioresource technology, 102(9), 5498-5503. 35. Zhao, Z., Zhang, Y., Quan, X., Zhao, H. 2016. Evaluation on direct interspecies electron transfer in anaerobic sludge digestion of microbial electrolysis cell. Bioresource technology, 200, 235-244. 36. Zhao, Z., Zhang, Y., Woodard, T., Nevin, K., Lovley, D. 2015. Enhancing syntrophic metabolism in up-flow anaerobic sludge blanket reactors with conductive carbon materials. Bioresource technology, 191, 140-145.
18
Figure captions Figure 1. Methane production rate (A), average methane conversion rate for each OLR (B) and pH (C) of the six groups of reactors. The average methane conversion rates were evaluated by the total methane production and total organic loading in influent for each phase of loading. Error bars represent standard deviations of triplicate reactors. Figure 2. VFAs in the reactors after each cycle of operation with the increase of OLR. Error bars represent standard deviations of triplicate reactors. Figure 3. Recovery of reactors with polyester cloth, graphite rods, and non-amended control reactor. Arrows at the top of the figure represent the number of days it took for the reactors to recover. Error bars represent standard deviations of triplicate reactors. Figure 4. Recovery of reactors with carbon cloth, GAC, and carbon felt. Arrows at the top of the figure represent the number of days it took for the reactors to recover. Error bars represent standard deviations of triplicate reactors. Figure 5. Bacterial community (A), and archaeal community (B) associated with five different samples (surface of carbon cloth, sludge from carbon cloth amended reactor, surface of polyester cloth, sludge from polyester amended reactor, and sludge from nonamended control reactor). Sequences that accounted for less than 1% of the population were grouped into the category “Others”.
19
(A)
(B)
(C)
Figure 1. Methane production rate (A), average methane conversion rate for each OLR (B) and pH (C) of the six groups of reactors. The average methane conversion rates were evaluated by the total methane production and total organic loading in influent for each phase of loading. Error bars represent standard deviations of triplicate reactors.
Figure 2. VFAs in the reactors after each cycle of operation with the increase of OLR. Error bars represent standard deviations of triplicate reactors.
Figure 3. Recovery of reactors with polyester cloth, graphite rods, and non-amended control reactor. Arrows at the top of the figure represent the number of days it took for the reactors to recover. Error bars represent standard deviations of triplicate reactors.
Figure 4. Recovery of reactor groups with carbon cloth, GAC, and carbon felt. Arrows at the top of the figure represent the number of days it took for the reactors to recover. Error bars represent standard deviations of triplicate reactors.
(A)
(B)
Figure 5. Bacterial community (A), and archaeal community (B) associated with five different samples (surface of carbon cloth, sludge from carbon cloth amended reactor, surface of polyester cloth, sludge from polyester
amended reactor, and sludge from non-amended control reactor). Sequences that accounted for less than 1% of the population were grouped into the category “Others”.
Highlights
• • • • •
Conductive materials stimulated methanogenesis in reactors treating food wastes Conductive materials promoted faster recovery of soured reactors Fermentative Sporanaerobacter and Enterococcus were enriched on carbon cloth Methanosarcina was also specifically enriched on carbon cloth The results suggest fermenter-methanogen DIET aided metabolism of complex waste
20
21
S0960-8524(16)31252-4 http://dx.doi.org/10.1016/j.biortech.2016.08.114 BITE 17017
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
25 July 2016 29 August 2016 30 August 2016
Please cite this article as: Dang, Y., Holmes, D.E., Zhao, Z., Woodard, T.L., Zhang, Y., Sun, D., Wang, L-Y., Nevin, K.P., Lovley, D.R., Enhancing anaerobic digestion of complex organic waste with carbon-based conductive materials, Bioresource Technology (2016), doi: http://dx.doi.org/10.1016/j.biortech.2016.08.114
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Enhancing anaerobic digestion of complex organic waste with carbon-based conductive materials Yan Dang1,2*, Dawn E Holmes1,3, Zhiqiang Zhao1,4, Trevor L Woodard1, Yaobin Zhang1,4, Dezhi Sun1,2, Li-Ying Wang1, Kelly P Nevin1, and Derek R Lovley1 1
Department of Microbiology, University of Massachusetts Amherst, Morril IV N Science
Center, Amherst, MA 01003 2
Beijing Key Laboratory for Source Control Technology of Water Pollution, College of
Environmental Science & Engineering, Beijing Forestry University, 35 Tsinghua East Road, Beijing 100083, China 3
Department of Physical and Biological Sciences, Western New England University, 1215
Wilbraham Rd, Springfield, MA, 01119 4
Key Laboratory of Industrial Ecology and Environmental Engineering (Dalian University
of Technology), Ministry of Education, School of Environmental Science and Technology, Dalian University of Technology, Dalian 116024, China Corresponding Author *Yan Dang. Department of Microbiology, University of Massachusetts 9 Amherst, 639 North Pleasant Street, Amherst, MA 01003, USA.
1
Email: [email protected]; [email protected]
Abstract. The aim of this work was to study the methanogenic metabolism of dog food, a food waste surrogate, in laboratory-scale reactors with different carbon-based conductive materials. Carbon cloth, carbon felt, and granular activated carbon all permitted higher organic loading rates and promoted faster recovery of soured reactors than the control reactors. Microbial community analysis revealed that specific and substantial enrichments of Sporanaerobacter and Methanosarcina were present on the carbon cloth surface. These results, and the known ability of Sporanaerobacter species to transfer electrons to elemental sulfur, suggest that Sporanaerobacter species can participate in direct interspecies electron transfer with Methanosarcina species when carbon cloth is available as an electron transfer mediator.
Keywords: Anaerobic digestion; organic wastes; carbon-based conductive material; direct interspecies electron transfer; methanogenesis
1. Introduction Conversion of complex organic wastes to methane is an attractive treatment method but has the limitations of long start-up times, slow processing of wastes, and susceptibility to disruptions by organic overloading (Chen et al., 2008; Li et al., 2011; Mata-Alvarez et al., 2000; McCarty, 1964). Studies with simple substrates such as ethanol, propionate, and
2
butyrate have suggested that promoting direct interspecies electron transfer (DIET) with conductive materials may accelerate and stabilize the conversion of these substrates to methane (Cruz Viggi et al., 2014; Lee et al., 2016; Li et al., 2015; Liu et al., 2012; Zhao et al., 2015). However, the potential role of DIET in the conversion of more complex wastes to methane has not been intensively investigated. The concept of DIET was first developed with co-cultures of the Fe(III)-respiring bacteria Geobacter metallireducens and Geobacter sulfurreducens (Summers et al., 2010), but some methanogens can also participate in DIET. For example, ethanol metabolism to methane via DIET was documented in mixed microbial communities associated with anaerobic digesters (Morita et al., 2011; Rotaru et al., 2014b) as well as in defined cocultures with Geobacter metallireducens and either Methanosaeta harundinacea (Rotaru et al., 2014b) or Methanosarcina barkeri (Rotaru et al., 2014a). Studies also showed that addition of conductive carbon materials such as granular activated carbon (GAC) (Liu et al., 2012; Rotaru et al., 2014a), biochar (Chen et al., 2014b), or carbon cloth (Chen et al., 2014a) could accelerate ethanol metabolism to methane via DIET in co-cultures with G. metallireducens and M. barkeri. Propanol and butanol metabolism via DIET was also documented in co-cultures of G. metallireducens and either M. harundinacea or M. bakeri (Wang et al., 2016). However, these substrates were only partially oxidized to propionate and butyrate, even when conductive materials were included. In other studies, it was suggested that the conductive mineral magnetite, which is known to promote DIET (Kato et al., 2012b; Liu et al., 2015), could stimulate the complete oxidation of propionate (Cruz Viggi et al., 2014) and butyrate
3
(Li et al., 2015) via DIET with a methanogen, but the microorganisms involved were not identified and DIET was only inferred not directly demonstrated. These results indicate that it might be difficult for propionate and butyrate, important intermediates in anaerobic digestion, to be completely metabolized via DIET under methanogenic conditions (Wang et al., 2016). Acetate metabolism via DIET has not yet been demonstrated in methanogenic culture studies. However, co-cultures of G. metallireducens and G. sulfurreducens could completely oxidize acetate when fumarate was provided as the sole terminal electron acceptor (Wang et al., 2016). In addition, when rice paddy microcosms or anaerobic digesters were supplemented with conductive materials, acetate oxidation by Geobacter species and methane production rates were significantly stimulated (Kato et al., 2012a; Lee et al., 2016), suggesting that acetate was being metabolized via DIET in these environments. While studies have shown that the incorporation of conductive carbon materials into continuous flow reactors can enhance the conversion of ethanol (Zhao et al., 2015) and acetate (Lee et al., 2016) to methane and permit higher loading rates without digester failure (Xu et al., 2015; Zhao et al., 2015), there has been less investigation into the possibility that these conductive materials may also help stabilize methanogenic degradation of more complex organic materials. The purpose of the studies reported here was to investigate whether carbon-based conductive materials could enhance the conversion of complex wastes to methane in
4
anaerobic digesters. Therefore, laboratory-scale reactors supplemented with four different carbon-based materials (carbon cloth, carbon felt, GAC, and graphite) were fed with increasing concentrations of complex waste and methane conversion rates and volatile fatty acid (VFA) concentrations were monitored throughout the experiment. In addition, microbial communities associated with various reactor conditions were compared.
2. Materials and Methods 2.1 Reactor design and incubation The eighteen reactors for this study were built from glass jars with a working volume of 2 liters and sealed with rubber stoppers. Each reactor had an influent/effluent port, a gas sampling port and a gas outlet port connected to a 10-liter gas collection bag. All reactors were incubated at 37 °C in the dark. The seed for the reactors was sludge obtained from the anaerobic digester of a municipal wastewater treatment plant in Pittsfield, Massachusetts. The total suspended sludge (TSS) of the seed sludge was about 13,000 mg/l and the ratio of volatile suspended sludge (VSS) to TSS was 0.72. It was stored under anaerobic conditions at 4 °C prior to use. To initiate the reactors they were each inoculated with 500 ml of seed sludge and 400 ml of deionized water. Five different materials (with similar conductive material geometric surface areas of ~ 1000 cm2) were added to triplicate reactors as follows: (1)
5
three pieces of carbon cloth (Zoroflex; buyactivatedcharcoal.com; 8×20×0.1 cm3; volume of 48 cm3 and geometric surface area of 977 cm2); (2) three pieces of carbon felt (AlfaAesar; MA; 8×17.5×0.6 cm3; volume of 252 cm3 and geometric surface area of 932 cm2); (3) 100 g of granular activated carbon (Sigma; Missouri; 8-20 mesh; volume of ca. 120 cm3 and geometric surface area of ca. 900 cm2); (4) 132 graphite rods (Sigma–Aldrich; Missouri; 13mm ×12 mm, volume of ca. 200 cm3 and surface area of ca.1000 cm2 ); or (5) three pieces of polyester cloth (Heavy Duty Fabric and CORDURA; http://heavydutyfabric.com; 8×20×0.1 cm3, volume of 48 cm3 and surface area of 977 cm2). One set of triplicate reactors received no materials. The initial composition of the artificial complex waste amendment was comprised of (per liter): commercial dog food (The Honest Kitchen; CA; total chemical oxygen demand (COD) of 1,336 ± 61 mg/g), 12.9 g; urea, 230 mg; KH2PO4, 110 mg; K2HPO4, 170 mg; Na2SO4, 50 mg; MgCl26H2O,100 mg; NiCl26H2O, 5 mg; CoCl26H2O, 6 mg and 2 ml each of previously described (Morita et al., 2011; Zhao et al., 2015) trace element and vitamin solutions. This dog food feedstock was blended for 5 min with a blender before being fed to the reactors. As the organic loading rate (OLR) of the reactors was increased, the dog food concentration in the artificial waste was increased stepwise from the initial concentration of 12.9 g/l to 25.8 g/l, 51.6 g/l, 64.5 g/l and 77.4 g/l. The pH of the artificial waste was adjusted to 7.0 with NaOH. Each reactor was fed with 100 ml of the artificial waste once a day to maintain a hydraulic retention time (HRT) of 10 days. The reactors were then shaken at 100 rpm for
6
21 h. Shaking was then stopped for 3 hours to allow sedimentation and 100 ml of effluent was discarded. This cycle was repeated for 10 days for each OLR. 2.2 Analytical methods TSS and VSS were analyzed in accordance with Standard Methods for the Examination of Water and Wastewater (APHA, 2012). COD was determined with Hach’s method 8000 (Hach DR/890 colorimeter procedures manual) (Morita et al., 2011). VFAs were measured with a high-performance liquid chromatograph (Bio-Rad; Hercules; California) (Nevin et al., 2008). pH was monitored with a pH analyzer (UB-10; Denver Instrument; Denver). The gas volume in the 10 L-gas-sampling bag was measured every 24 h with a digital mass flow meter (FMA4000; Omega; USA). The composition of biogas was analyzed with a gas chromatograph with a flame ionization detector (GC-8A; Shimadzu; Japan). Methane conversion rates were evaluated based on the established stoichiometric value of 15.6 mmol (CH4)/g (COD degraded) equal to 100 % efficiency (Timur & Özturk, 1999). 2.3 DNA extraction and barcoded amplicon sequencing Sludge samples (1.0 ml) and cloth material (1×1 cm2) were collected at the end of operation when the reactors were still in stable condition. The cloths were washed with phosphate-buffered saline (PBS; 0.13 M NaCl and 10 mM Na2HPO4 at pH 7.2) to remove any residual suspended sludge. All samples were immediately frozen in liquid nitrogen and
7
stored at -80°C. For DNA extraction, frozen samples were ground into a fine powder with a mortar and pestle before treatment with the BIO101 FastDNA soil kit (MP Biomedicals, Ohio) according to the manufacturer’s instructions. DNA was quantified with a NanoDrop spectrophotometer (Thermo, Fisher Scientific, Delaware). Extracted DNA was analyzed at the MR DNA sequencing facility (Shallowater, Texas). The primers 27F and 349F were used to sequence bacterial and archaeal 16S rRNA genes on a 454 platform (Roche). Sequence data was deposited in the Genbank database and assigned accession no. KT936838-KT937136.
3. Results and Discussion 3.1 Performance of the reactors with and without conductive carbon materials In the initial start up period, as the OLR was increased from 1.6 kg COD/(m3 day) to 3.3 and then to 6.7 kgCOD/(m3 day), all of the reactors performed in a similar manner (ttest, P < 0.05) with stable operation, high methane conversion rates (Figure 1A and 1B), stable pH (Figure 1C), and low concentrations of VFAs (Figure 2). Methane conversion rates from all six reactor systems exceeded 86% at an OLR of 6.7 kg COD/(m3 day), and the reactors supplemented with carbon cloth, GAC, and carbon felt only produced slightly more methane (109.6±3.0, 106.6±9.4 and 110.8±5.8 mmol/d; n=3) than graphite, polyester cloth or non-amended control reactors (101.2±3.2, 100.0±4.7 and 100.1±3.4 mmol/d; n=3).
8
However, when the OLR was increased to 8.5 kg COD/(m3 day), the performance of the reactors without conductive materials (i.e. non-amended controls or polyester cloth) or graphite rods declined (Figure 1). Methane production rates dropped to 43% or lower, VFAs accumulated to nearly 100 mM, and pHs sank to < 6.0 (Figure 1C). In contrast, reactors containing carbon cloth, GAC, or carbon felt continued to run stably and efficiently; total VFA concentrations were only 18 mM, 42 mM and 45 mM, respectively (Figure 2), and methane conversion rates were >85% (Figure 1B). A further increase of the OLR to 10.3 kg COD/(m3 day) immediately disrupted the methane conversion in the reactors with carbon felt, and the performance of the reactors with carbon cloth or GAC declined soon thereafter (Figure 1). The concentrations of total VFAs in the carbon cloth and carbon felt reactors reached ca. 90 mM (Figure 2) and methane production rates dropped significantly for all conditions (Figure 1). The reactors with GAC performed best at this OLR, continuing to convert 66 % of added organic load to methane, with total VFAs of ca. 60 mM and a pH of 7.1. 3.2 Reactor recovery In order to determine whether the failed non-amended control, graphite-amended, or polyester cloth-amended reactors could recover if the OLR was decreased, the OLR was reduced to 3.3 kg COD/(m3 day) on day 35 of operation (Figure 3). Methane production in polyester cloth-amended reactors began to slowly improve, but reactors with graphite rods or the non-amended control reactors were unable to recover (Figure 3). Therefore, after 13 days the OLR was further reduced to 0.85 kg COD/(m3 day). All three types of reactors
9
eventually recovered at this loading rate, substantially reducing VFA levels and steadily producing methane. The reactors with polyester cloth recovered faster than those with graphite rods, and the reactors with no materials were the slowest to recover (Figure 3). On day 45, the load to the reactors with GAC, carbon cloth, and carbon felt was also decreased to 3.3 kg COD/(m3 day). This load was further reduced to 0.85 kg COD/(m3 day) 13 days later, in order to make conditions for all 6 reactor-types comparable (Figure 4). All of these amended reactors recovered within 25 days, significantly faster than the graphite, polyester cloth or non-amended control reactors (Figure 4). 3.3 Microbial community analysis Reactor microbial community structure was also evaluated to gain insight into microbial factors that might lead to improved reactor performance. Although the reactors with GAC performed best, during operation GAC disintegrated into tiny carbon particles suspended in the sludge, making it difficult to differentiate between microorganisms attached to the GAC surface and those suspended in the sludge. Therefore, we focused on the microbial community attached to the carbon cloth, which also greatly improved reactor performance, and compared this community to those attached to non-conductive polyester cloth and in bulk sludge from the polyester cloth amended and non-amended control reactors. Bacterial and methanogenic sequences enriched on the surface of the carbon cloth differed significantly from the other samples (Figure 5). For example, bacteria in the genus Sporanaerobacter were much more abundant on the surface of the carbon cloth than on the
10
surface of the polyester, accounting for 34 % of the 16S rRNA gene sequences recovered (Figure 5A). In contrast, the proportion of Sporanaerobacter sequences attached to the surface of the polyester cloth and in the bulk sludge of the control and the polyester cloth reactors was < 1 % and Sporanaerobacter only accounted for 2 % of the sequences in the bulk sludge of the carbon cloth reactors (Figure 5A). These results suggest that the carbon cloth enhanced the growth of Sporanaerobacter. The best characterized Sporanaerobacter, S. acetigenes metabolizes fermentable substrates with the reduction of elemental sulfur (Hernandez-Eugenio et al., 2002). Although its ability to transfer electrons to conductive materials has not yet been evaluated, the capability of elemental sulfur reduction is very commonly linked to the ability to transfer electrons to other extracellular electron acceptors such as Fe(III) oxides or electrodes (Lovley et al., 2004). Therefore, it seems likely that the Sporanaerobacter species that were highly enriched on the carbon cloth were using the carbon cloth as an electrical connection to electron-accepting methanogens. Consistent with this concept was the finding that the methanogenic community on the carbon cloth was very distinct from any of the other environments sampled (Figure 5B). Methanosarcina species were only predominant on the carbon cloth surface, accounting for 68 % of the methanogen sequences recovered (Figure 5B). In all of the other environments Methanosaeta species were the predominant methanogens, even in the bulk sludge of the carbon cloth reactors. Studies have shown that both Methanosaeta (Rotaru et al., 2014b) and Methanosarcina (Rotaru et al., 2014a) species are capable of DIET when the cells are in direct contact with their partners through
11
biological electrical connections. However, DIET stimulated by conductive materials such as GAC (Liu et al., 2012), carbon cloth (Chen et al., 2014a), or biochar (Chen et al., 2014b) has only been documented in co-culture studies done with Geobacter metallireducens and Methanosarcina barkeri. Attempts to stimulate DIET with conductive carbon-based materials in co-cultures in which Methanosaeta haurundinacea replaced M. barkeri have been unsuccessful (see supplementary material, Figure S1). Thus, the enrichment of Methanosarcina on the cloth is consistent with a source of electrons being delivered by bacteria with the carbon cloth as a conduit. S. acetigenes metabolizes glucose solely to acetate and carbon dioxide when sulfur is available as an electron acceptor, but produces hydrogen when electron acceptors are not provided (Hernandez-Eugenio et al., 2002). In the polyester cloth and control reactors, methanogens known to specialize in the utilization of hydrogen as an electron donor (Methanomicrobiales and Methanobacteriales) were abundant in the bulk sludge, but not in the carbon cloth reactors. This finding is also consistent with our hypothesis that Sporanaerobacter can take advantage of the carbon cloth-mediated electrical conduit to Methanosarcina species to metabolize fermentable substrates with electron transfer to Methanosarcina rather than using a less energetically favorable fermentative metabolism to generate hydrogen. The proposed conversion of fermentable substrates primarily to acetate and carbon dioxide with electron transfer to Methanosarcina would be expected to result in the production of less propionate, butyrate and hydrogen, and limit the need for syntrophic metabolism of the VFAs. This is an important consideration because the accumulation of
12
propionate appeared to be associated with the inhibition of methane production at higher OLR (Figure 2). These results are consistent with previous studies that have shown that propionate accumulates in destabilized digester systems (Inanc et al., 1996; Li et al., 2012; McCarty & Smith, 1986; Ye et al., 2011). Carbon cloth was highly effective in maintaining low propionate concentrations at all but the highest organic loading rates (Figure 2). Methanosarcina may also benefit from accepting electrons from the carbon cloth, because the conversion of acetate to methane yields little energy and Methanosarcina typically grow slowly on acetate. Electrons obtained via DIET might enhance their metabolism and even increase their ability to produce methane by acetate decarboxylation. Improved growth and metabolism of Methanosarcina might account for lower acetate concentrations in reactors amended with carbon cloth compared to reactors without conductive materials. There was also an enrichment of Enterococcus on the carbon cloth, which comprised 10 % of the 16S rRNA gene sequences recovered (Figure 5A). In contrast, Enterococcus sequences accounted for less than 1 % of the sequences in the samples from the polyester and control reactors and only 1.5 % of the sequences in the bulk sludge of the carbon cloth reactor (Figure 5A). Enterococcus gallinarum, which metabolizes a diversity of fermentable substrates, is capable of extracellular electron transfer to Fe(III) and is electrochemically active (Kim et al., 2005). Thus, the Enterococcus species enriched on the carbon cloth may also have been contributing to the metabolism of fermentable substrates coupled with electron transfer to Methanosarcina via carbon cloth-mediated DIET.
13
There was no apparent enrichment on the carbon cloth of microorganisms that might be involved in syntrophic metabolism of short-chain fatty acids or ethanol via DIET. This included a lack of Geobacter species. These are the first results that provide evidence that DIET plays a role in stimulating and stabilizing anaerobic digestion of complex organic material by altering the metabolism of fermentable substrates. Isolation and characterization of Sporanaerobacter species enriched on the carbon cloth will be required to confirm that these organisms can participate in DIET mediated by carbon cloth, however their high abundance on the carbon cloth associated with the specific enrichment of Methanosarcina suggests that these organisms utilize a significantly different type of metabolism in the presence of carbon cloth than they do when grown with non-conductive polyester cloth. The results further demonstrate the potential benefits of conductive carbon materials in anaerobic digestion. Permanent incorporation of carbon cloth or similar conductive materials into single chamber anaerobic digesters is expected to be feasible and preferable to the use of particulate conductive materials that need to be continually replenished (Zhao et al., 2015). Graphite rods represent another conductive material that could serve as a permanent component of digesters. Graphite serves as an excellent surface for the growth of current-producing biofilms in microbial fuel cells (Franks & Nevin, 2010; Nevin et al., 2008), but we found that it was ineffective in promoting or stabilizing methanogenesis in the reactors. Additional studies are required to better understand this result.
14
It is clear from the results presented here, as well as many previous studies (Chen et al., 2014a; Chen et al., 2014b; Cruz Viggi et al., 2014; Kato et al., 2012a; Kato et al., 2012b; Lee et al., 2016; Li et al., 2015; Liu et al., 2012; Liu et al., 2015; Xu et al., 2015; Zhao et al., 2015), that conductive materials can enhance methane production from a diversity of organic substrates. However, the microbiology of this process is still poorly understood. Geobacter species are the only organisms that have been shown to donate electrons to electron accepting partners via DIET in culture studies. Yet in the studies reported here, as well as other recent studies (Baek et al., 2015; Beckmann et al., 2016; Luo et al.; Xu et al., 2015; Yamada et al., 2015; Zhao et al., 2016) Geobacter species were not enriched with conductive material amendments, suggesting that other microorganisms may participate in DIET. Further elucidation of which microbes can participate in DIET in methanogenic environments will be important for informed design of improved methanogenic digesters based on the DIET concept. 4. Conclusions These results demonstrate that some carbon conductive materials such as carbon cloth and GAC can stimulate methane production from complex organic material, permit higher OLRs, and promote faster recovery of soured reactors. Analysis of microbial community composition suggested a new concept for DIET. It appeared that conductive carbon cloth specifically enriched for Sporanaerobacter and Enterococcus species that could metabolize fermentable substrates and transfer those electrons to Methanosarcina species. This is a surprising finding because previous studies had only suggested the
15
possibility of accelerating the syntrophic metabolism of fermentation products when promoting DIET.
Acknowledgment This research was supported by the Environmental Research & Education Foundation. The first author Yan Dang acknowledges support from the China Scholarship Council (CSC) program (2014).
References 1. APHA. 2012. Standard methods for the examination of water and wastewater. American Public Health Association Washington, DC. 2. Baek, G., Kim, J., Cho, K., Bae, H., Lee, C. 2015. The biostimulation of anaerobic digestion with (semi) conductive ferric oxides: their potential for enhanced biomethanation. Applied microbiology and biotechnology, 99(23), 10355-10366. 3. Beckmann, S., Welte, C., Li, X., Oo, Y.M., Kroeninger, L., Heo, Y., Zhang, M., Ribeiro, D., Lee, M., Bhadbhade, M. 2016. Novel phenazine crystals enable direct electron transfer to methanogens in anaerobic digestion by redox potential modulation. Energy & Environmental Science. 4. Chen, S., Rotaru, A.-E., Liu, F., Philips, J., Woodard, T.L., Nevin, K.P., Lovley, D.R. 2014a. Carbon cloth stimulates direct interspecies electron transfer in syntrophic cocultures. Bioresource technology, 173, 82-86. 5. Chen, S., Rotaru, A.-E., Shrestha, P.M., Malvankar, N.S., Liu, F., Fan, W., Nevin, K.P., Lovley, D.R. 2014b. Promoting interspecies electron transfer with biochar. Scientific reports, 4. 6. Chen, Y., Cheng, J.J., Creamer, K.S. 2008. Inhibition of anaerobic digestion process: a review. Bioresource technology, 99(10), 4044-4064. 7. Cruz Viggi, C., Rossetti, S., Fazi, S., Paiano, P., Majone, M., Aulenta, F. 2014. Magnetite particles triggering a faster and more robust syntrophic pathway of methanogenic propionate degradation. Environmental science & technology, 48(13), 7536-7543. 8. Franks, A.E., Nevin, K.P. 2010. Microbial fuel cells, a current review. Energies, 3(5), 899-919. 9. Hernandez-Eugenio, G., Fardeau, M.-L., Cayol, J.-L., Patel, B.K., Thomas, P., Macarie, H., Garcia, J.-L., Ollivier, B. 2002. Sporanaerobacter acetigenes gen. nov., sp. nov., a novel acetogenic, facultatively sulfur-reducing bacterium. International journal of
16
systematic and evolutionary microbiology, 52(4), 1217-1223. 10. Inanc, B., Matsui, S., Ide, S. 1996. Propionic acid accumulation and controlling factors in anaerobic treatment of carbohydrate: effects of H 2 and pH. Water Science and Technology, 34(5), 317-325. 11. Kato, S., Hashimoto, K., Watanabe, K. 2012a. Methanogenesis facilitated by electric syntrophy via (semi) conductive iron‐oxide minerals. Environmental microbiology, 14(7), 1646-1654. 12. Kato, S., Hashimoto, K., Watanabe, K. 2012b. Microbial interspecies electron transfer via electric currents through conductive minerals. Proceedings of the National Academy of Sciences, 109(25), 10042-10046. 13. Kim, G., Hyun, M., Chang, I., Kim, H., Park, H., Kim, B., Kim, S., Wimpenny, J., Weightman, A.J. 2005. Dissimilatory Fe (III) reduction by an electrochemically active lactic acid bacterium phylogenetically related to Enterococcus gallinarum isolated from submerged soil. Journal of applied microbiology, 99(4), 978-987. 14. Lee, J.-Y., Lee, S.-H., Park, H.-D. 2016. Enrichment of Specific Electro-Active Microorganisms and Enhancement of Methane Production by Adding Granular Activated Carbon in Anaerobic Reactors. Bioresource Technology. 15. Li, H., Chang, J., Liu, P., Fu, L., Ding, D., Lu, Y. 2015. Direct interspecies electron transfer accelerates syntrophic oxidation of butyrate in paddy soil enrichments. Environmental microbiology, 17(5), 1533-1547. 16. Li, J., Ban, Q., Zhang, L., Jha, A.K. 2012. Syntrophic Propionate Degradation in Anaerobic Digestion: A Review. International Journal of Agriculture & Biology, 14(5). 17. Li, Y., Park, S.Y., Zhu, J. 2011. Solid-state anaerobic digestion for methane production from organic waste. Renewable and Sustainable Energy Reviews, 15(1), 821-826. 18. Liu, F., Rotaru, A.-E., Shrestha, P.M., Malvankar, N.S., Nevin, K.P., Lovley, D.R. 2012. Promoting direct interspecies electron transfer with activated carbon. Energy & Environmental Science, 5(10), 8982-8989. 19. Liu, F., Rotaru, A.E., Shrestha, P.M., Malvankar, N.S., Nevin, K.P., Lovley, D.R. 2015. Magnetite compensates for the lack of a pilin‐associated c‐type cytochrome in extracellular electron exchange. Environmental microbiology, 17(3), 648-655. 20. Lovley, D.R., Holmes, D.E., Nevin, K.P. 2004. Dissimilatory fe (iii) and mn (iv) reduction. Advances in microbial physiology, 49, 219-286. 21. Luo, C., Lü, F., Shao, L., He, P. 2015. Application of eco-compatible biochar in anaerobic digestion to relieve acid stress and promote the selective colonization of functional microbes. Water research, 68, 710-718. 22. Mata-Alvarez, J., Mace, S., Llabres, P. 2000. Anaerobic digestion of organic solid wastes. An overview of research achievements and perspectives. Bioresource technology, 74(1), 3-16. 23. McCarty, P.L. 1964. Anaerobic waste treatment fundamentals. Public works, 95(9), 107-112. 24. McCarty, P.L., Smith, D.P. 1986. Anaerobic wastewater treatment. Environmental Science & Technology, 20(12), 1200-1206. 25. Morita, M., Malvankar, N.S., Franks, A.E., Summers, Z.M., Giloteaux, L., Rotaru, A.E., Rotaru, C., Lovley, D.R. 2011. Potential for direct interspecies electron transfer in
17
methanogenic wastewater digester aggregates. MBio, 2(4), e00159-11. 26. Nevin, K.P., Richter, H., Covalla, S., Johnson, J., Woodard, T., Orloff, A., Jia, H., Zhang, M., Lovley, D. 2008. Power output and columbic efficiencies from biofilms of Geobacter sulfurreducens comparable to mixed community microbial fuel cells. Environmental microbiology, 10(10), 2505-2514. 27. Rotaru, A.-E., Shrestha, P.M., Liu, F., Markovaite, B., Chen, S., Nevin, K.P., Lovley, D.R. 2014a. Direct interspecies electron transfer between Geobacter metallireducens and Methanosarcina barkeri. Applied and environmental microbiology, 80(15), 4599-4605. 28. Rotaru, A.-E., Shrestha, P.M., Liu, F., Shrestha, M., Shrestha, D., Embree, M., Zengler, K., Wardman, C., Nevin, K.P., Lovley, D.R. 2014b. A new model for electron flow during anaerobic digestion: direct interspecies electron transfer to Methanosaeta for the reduction of carbon dioxide to methane. Energy & Environmental Science, 7(1), 408-415. 29. Summers, Z.M., Fogarty, H.E., Leang, C., Franks, A.E., Malvankar, N.S., Lovley, D.R. 2010. Direct exchange of electrons within aggregates of an evolved syntrophic coculture of anaerobic bacteria. Science, 330(6009), 1413-1415. 30. Timur, H., Özturk, I. 1999. Anaerobic sequencing batch reactor treatment of landfill leachate. Water Research, 33(15), 3225-3230. 31. Wang, L.-Y., Nevin, K.P., Woodard, T.L., Mu, B.-Z., Lovley, D.R. 2016. Expanding the Diet for DIET: Electron Donors Supporting Direct Interspecies Electron Transfer (DIET) in Defined Co-Cultures. Frontiers in microbiology, 7. 32. Xu, S., He, C., Luo, L., Lü, F., He, P., Cui, L. 2015. Comparing activated carbon of different particle sizes on enhancing methane generation in upflow anaerobic digester. Bioresource technology, 196, 606-612. 33. Yamada, C., Kato, S., Ueno, Y., Ishii, M., Igarashi, Y. 2015. Conductive iron oxides accelerate thermophilic methanogenesis from acetate and propionate. Journal of bioscience and bioengineering, 119(6), 678-682. 34. Ye, J., Mu, Y., Cheng, X., Sun, D. 2011. Treatment of fresh leachate with high-strength organics and calcium from municipal solid waste incineration plant using UASB reactor. Bioresource technology, 102(9), 5498-5503. 35. Zhao, Z., Zhang, Y., Quan, X., Zhao, H. 2016. Evaluation on direct interspecies electron transfer in anaerobic sludge digestion of microbial electrolysis cell. Bioresource technology, 200, 235-244. 36. Zhao, Z., Zhang, Y., Woodard, T., Nevin, K., Lovley, D. 2015. Enhancing syntrophic metabolism in up-flow anaerobic sludge blanket reactors with conductive carbon materials. Bioresource technology, 191, 140-145.
18
Figure captions Figure 1. Methane production rate (A), average methane conversion rate for each OLR (B) and pH (C) of the six groups of reactors. The average methane conversion rates were evaluated by the total methane production and total organic loading in influent for each phase of loading. Error bars represent standard deviations of triplicate reactors. Figure 2. VFAs in the reactors after each cycle of operation with the increase of OLR. Error bars represent standard deviations of triplicate reactors. Figure 3. Recovery of reactors with polyester cloth, graphite rods, and non-amended control reactor. Arrows at the top of the figure represent the number of days it took for the reactors to recover. Error bars represent standard deviations of triplicate reactors. Figure 4. Recovery of reactors with carbon cloth, GAC, and carbon felt. Arrows at the top of the figure represent the number of days it took for the reactors to recover. Error bars represent standard deviations of triplicate reactors. Figure 5. Bacterial community (A), and archaeal community (B) associated with five different samples (surface of carbon cloth, sludge from carbon cloth amended reactor, surface of polyester cloth, sludge from polyester amended reactor, and sludge from nonamended control reactor). Sequences that accounted for less than 1% of the population were grouped into the category “Others”.
19
(A)
(B)
(C)
Figure 1. Methane production rate (A), average methane conversion rate for each OLR (B) and pH (C) of the six groups of reactors. The average methane conversion rates were evaluated by the total methane production and total organic loading in influent for each phase of loading. Error bars represent standard deviations of triplicate reactors.
Figure 2. VFAs in the reactors after each cycle of operation with the increase of OLR. Error bars represent standard deviations of triplicate reactors.
Figure 3. Recovery of reactors with polyester cloth, graphite rods, and non-amended control reactor. Arrows at the top of the figure represent the number of days it took for the reactors to recover. Error bars represent standard deviations of triplicate reactors.
Figure 4. Recovery of reactor groups with carbon cloth, GAC, and carbon felt. Arrows at the top of the figure represent the number of days it took for the reactors to recover. Error bars represent standard deviations of triplicate reactors.
(A)
(B)
Figure 5. Bacterial community (A), and archaeal community (B) associated with five different samples (surface of carbon cloth, sludge from carbon cloth amended reactor, surface of polyester cloth, sludge from polyester
amended reactor, and sludge from non-amended control reactor). Sequences that accounted for less than 1% of the population were grouped into the category “Others”.
Highlights
• • • • •
Conductive materials stimulated methanogenesis in reactors treating food wastes Conductive materials promoted faster recovery of soured reactors Fermentative Sporanaerobacter and Enterococcus were enriched on carbon cloth Methanosarcina was also specifically enriched on carbon cloth The results suggest fermenter-methanogen DIET aided metabolism of complex waste
20
21