Oct 27, 2015 - Bredesen Center for Interdisciplinary Research and Education, The University of Tennessee, ... Acetate was the direct substrate for exoelectrogenesis. ...... (29) Call, D.; Logan, B. E. Hydrogen production in a single chamber.
Article pubs.acs.org/est
Biotransformation of Furanic and Phenolic Compounds with Hydrogen Gas Production in a Microbial Electrolysis Cell Xiaofei Zeng,† Abhijeet P. Borole,‡,§ and Spyros G. Pavlostathis*,† †
School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0512, United States Biosciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States § Bredesen Center for Interdisciplinary Research and Education, The University of Tennessee, Knoxville, Tennessee 37996, United States ‡
S Supporting Information *
ABSTRACT: Furanic and phenolic compounds are problematic byproducts resulting from the breakdown of lignocellulosic biomass during biofuel production. The capacity of a microbial electrolysis cell (MEC) to produce hydrogen gas (H2) using a mixture of two furanic (furfural, FF; 5-hydroxymethyl furfural, HMF) and three phenolic (syringic acid, SA; vanillic acid, VA; and 4-hydroxybenzoic acid, HBA) compounds as the substrate in the bioanode was assessed. The rate and extent of biotransformation of the five compounds and efficiency of H2 production, as well as the structure of the anode microbial community, were investigated. The five compounds were completely transformed within 7-day batch runs and their biotransformation rate increased with increasing initial concentration. At an initial concentration of 1200 mg/L (8.7 mM) of the mixture of the five compounds, their biotransformation rate ranged from 0.85 to 2.34 mM/d. The anode Coulombic efficiency was 44−69%, which is comparable to that of wastewaterfed MECs. The H2 yield varied from 0.26 to 0.42 g H2−COD/g COD removed in the anode, and the bioanode volumenormalized H2 production rate was 0.07−0.1 L/L-d. The biotransformation of the five compounds took place via fermentation followed by exoelectrogenesis. The major identified fermentation products that did not transform further were catechol and phenol. Acetate was the direct substrate for exoelectrogenesis. Current and H2 production were inhibited at an initial substrate concentration of 1200 mg/L, resulting in acetate accumulation at a much higher level than that measured in other batch runs conducted with a lower initial concentration of the five compounds. The anode microbial community consisted of exoelectrogens, putative degraders of the five compounds, and syntrophic partners of exoelectrogens. The MEC H2 production demonstrated in this study is an alternative to the currently used process of reforming natural gas to supply H2 needed to upgrade bio-oils to stable hydrocarbon fuels.
■
INTRODUCTION Lignocellulosic biomass is a promising feedstock for the production of biofuels using thermal, chemical, or biological processes. Breakdown of lignocellulosic biomass, regardless of the process used, typically results in furanic and phenolic byproducts, which are inhibitory, problematic compounds. For instance, furanic and phenolic compounds in lignocellulosic hydrolysates are highly inhibitory to H2- and ethanol-producing fermentative microorganisms, which has been a persistent challenge in biofuel production through fermentation processes.1,2 In pyrolysis, a thermal process reforming biomass to bio-oil, the presence of polar and oxygen-rich compounds (e.g., furan aldehydes and phenolic acids) makes the bio-oil acidic and unstable, requiring H2 in the downstream hydrogenation process to upgrade the bio-oil to a stable fuel.3 To overcome the negative effect of furanic and phenolic compounds, physical and/or chemical removal, as well as microbial and enzymatic processes, have been explored.4 However, these methods have © 2015 American Chemical Society
the trade-offs of lowering biofuel yield and do not lessen the challenge of downstream wastewater treatment.4 Thus, a process to directly utilize furanic and phenolic compounds to produce biofuels would be an improvement over the existing methods. Microbial electrolysis cell (MEC) technology is a bioelectrochemical process, which produces H2. Exoelectrogenic bacteria in the anode oxidize organic substrates by transferring electrons to the electrode and then to the cathode. With a voltage input (>0.3 V), protons transferred from the anode to the cathode via a cation exchange membrane are reduced to H2.5 Unlike other H2-producing bioprocesses, the MEC produces H2 through an abiotic half-reaction in the cathode, Received: Revised: Accepted: Published: 13667
May 8, 2015 August 18, 2015 October 27, 2015 October 27, 2015 DOI: 10.1021/acs.est.5b02313 Environ. Sci. Technol. 2015, 49, 13667−13675
Article
Environmental Science & Technology
V (Table S2) compared to the E0′ value of −0.414 V for proton reduction to H2. Inoculum. The bioanode inoculum was a piece of biofilmattached carbon felt from a MFC anode developed at the Oak Ridge National Laboratory (Oak Ridge, TN), which over time had been fed with a mixture of furanic/phenolic compounds, corn stover hydrolysates, and switchgrass pyrolysis aqueous stream.9,16,17 The original inoculum of this MFC was a sample collected from a municipal anaerobic digester. The bioanode inoculum was further enriched in the present study in a different MFC anode, fed with the mixture of the two furanic and three phenolic compounds, prior to being transferred to a MEC anode. MFC Setup and Operation. An air-cathode MFC was set up to enrich the inoculum. The anode electrode was porous carbon felt (5 stripes, 1 cm × 1 cm × 10 cm each; Alfa Aesar) tied to a stainless steel rod. The anode chamber was a modified square glass bottle with an open channel on one side. The empty bed volume was 250 mL, and the liquid volume was 200 mL due to electrode displacement. The cathode was a membrane−electrode assembly with a surface area of 5.7 cm2 purchased from Fuel Cells Etc (College Station, TX), which was made of a cation exchange membrane and carbon cloth containing 0.5 mg/cm2 Pt. The cathode was clamped to the side channel extended from the anode chamber and exposed to air on one side. A piece of biofilm-attached carbon felt (approximately 1 cm × 1 cm × 3 cm) was placed in the center of the anode carbon felt electrode. The anode medium consisted of the following (in g/L): NH4Cl, 0.31; KCl, 0.13; NaH2PO4·H2O, 2.45; and Na2HPO4, 4.58; along with trace metals and vitamins.18 The pH of the medium was 7.0. The anolyte was deoxygenated by bubbling N2 through the liquid phase prior to use, and was continuously mixed magnetically. The MFC was maintained at room temperature (20−22 °C). A mixture of the five compounds at equal electron equivalents (each at 62.5 mg COD/L) and a total concentration of 200 mg/L (312 mg COD/L) was fed to the MFC anode once a week (7-day fedbatch). During the first ten feeding cycles (∼70 days), glucose (200 mg/L) was fed along with the five compounds to enhance microbial growth. A variable resistor was placed between the anode and cathode, and its resistance was gradually reduced from 500 to 250, and then to 100 Ω, in order to promote the growth of exoelectrogenic bacteria. The voltage was recorded by a potentiostat (Interface 1000, Gamry Instruments, Warminster, PA). Culture enrichment lasted for 6 months. The MFC activity stabilized with a mean maximum current of 1.25 mA, soluble COD (sCOD) removal of 50−60%, and Coulombic efficiency (CE) of 40−60%, measured over the last 20 feeding cycles. MEC Setup and Operation. An H-type MEC was developed with two square glass bottles separated by a cation exchange membrane (Nafion 117, 5.7 cm 2 ; Dupont, Wilmington, DE) and maintained at room temperature (20− 22 °C). Both chambers and the anode electrode had the same configuration as the above-described MFC anode. The cathode electrode was a carbon cloth containing 0.5 mg/cm2 of Pt (5 cm × 6 cm; Fuel Cell Etc). A gas collection buret using displacement of an acid brine solution (10% NaCl w/v, 2% H2SO4 v/v) was connected to each chamber headspace for gas volume measurement (Figure S1). The inoculation procedure and anolyte of the MEC were the same as for the above-described MFC. The MEC catholyte was
with the supply of electrons from the microbially assisted halfreaction in the anode. This feature of MEC eliminates the need for H2-producing bacteria, which are highly susceptible to furanic and phenolic compounds.1,6 Thus, the MEC is potentially able to convert these problematic compounds to H2, which in turn can be used in the hydrogenation process and thus minimize the external H2 supply currently generated by reforming natural gas. Given the advantage of the MEC technology, recent studies have investigated the potential of integrating MEC in biofuel production platforms. Lignocellulosic effluent, refinery wastewater, and switchgrass-derived bio-oil aqueous phase have been treated in MECs with simultaneous H2 production.7−9 However, the fate and effect of furanic and phenolic compounds, which are the problematic components of the above-mentioned feedstocks, remain unknown. This study investigated a MEC for H2 production from two furanic and three phenolic compounds. This is the first attempt to understand the fate of specific inhibitory compounds and their potential to produce H2 in a MEC. The most commonly used substrates in MECs are acetate and wastewater.10−12 However, acetate is a favorable organic substrate for exoelectrogens,13,14 whereas, furanic and phenolic compounds are less biodegradable and have not been reported as direct substrates for exoelectrogens. On the other hand, wastewater streams generally include easily biodegradable components, which may be the major contributors to H2 production. Two previous studies used furanic and phenolic compounds as the substrate in the anode of a microbial fuel cell (MFC), but had mixed results. Catal et al. reported that, with the exception of HMF, use of nine individual furanic and phenolic compounds (2-furaldehyde, syringaldehyde, vanillin, trans-cinnamic acid, trans-4-hydroxy-3-methoxy-cinnamic acid, 4-hydroxy-cinnamic acid, 3,5-dimethoxy-4-hydroxy-cinnamic acids, benzyl alcohol, and acetophenone) did not generate voltage.15 In contrast, Borole et al. demonstrated that an anode microbial consortium was able to convert furfural, HMF, 4-hydroxybezaldehyde, hydroxyacetophenone, and vanillic acid to electricity.16 Thus, the question remains whether a MEC anode microbial community can use furanic and phenolic compounds as the sole carbon and energy source to produce H2 in the cathode. The overall goal of this study is to produce H2 through the biotransformation of specific furanic and phenolic compounds using MEC technology. The specific objective of the work presented here was to enrich a microbial community, assess its capacity to biotransform two furanic and three phenolic compounds, and produce H2 in a batch-fed MEC. Biotransformation of the selected furanic and phenolic compounds, formation of metabolites, and production of current and H2, as well as the structure of the anode microbial community, were investigated.
■
MATERIALS AND METHODS Chemicals. Furfural (FF, 99%), 5-hydroxymethyl furfural (HMF, ≥ 99%), syringic acid (SA, ≥ 95%), and 4hydroxybenzoic acid (HBA, ≥ 99%) were purchased from Sigma-Aldrich (St. Louis, MO). Vanillic acid (VA, ≥ 99%) was purchased from Alfa Aesar (Ward Hill, MA). The five compounds are soluble in water (1.5−364 g/L at 25 °C), not volatile (Henry’s law constant =10−14 to 10−6 atm-m3/mol), and have a low hydrophobicity (log Kow = −0.09 to 1.58) (Supporting Information Table S1). The standard potential at pH 7.0 (E0′) of the five compounds is from −0.388 to −0.303 13668
DOI: 10.1021/acs.est.5b02313 Environ. Sci. Technol. 2015, 49, 13667−13675
Article
Environmental Science & Technology
wavelength of 210 nm, except that the eluent was 5 mM H2SO4 without any organic solvent. Metabolites of the five compounds were identified using an LC/MS/MS unit (Agilent 1260 Infinity LC system, 6410 Triple Quad MSD) equipped with a Kinetex biphenyl column (3 × 150 mm, 5 μm; Phenomenex, Torrance, CA). The eluent consisted of (A) 5 mM ammonium acetate with 0.5% acetic acid in acetonitrile (v/v) and (B) 5 mM ammonium acetate in 0.5% acetic acid (v/v) at a flow rate of 0.5 mL/min, using gradient elution as follows: eluent A was increased from 2% to 30% in 2.3 min and to 90% in 1.2 min, and then was maintained at 90% for 2.5 min. The MS/MS was operated in both positive and negative modes at 100 eV in an m/z range of 50−250. The product ions of the same molecular weight as hypothesized metabolites were fragmented, and the fragmentation patterns were compared with those of purchased pure chemicals. Soluble chemical oxygen demand (sCOD) and pH were measured following procedures outlined in Standard Methods.19 Total gas production was measured by the acid brine solution displacement in the burets, equilibrated to 1 atm. Headspace gas composition (i.e., H2, CO2, and CH4) was determined with a gas chromatography unit equipped with two columns and two thermal conductivity detectors.20 Calculations. Coulombic efficiency and H2 yield were calculated as previously described.21 Current density was normalized to either the empty bed volume of the anode chamber (250 mL) for comparison with single-chamber MECs, or the projected surface area of the Nafion membrane (5.7 cm2), assuming the membrane surface area was limiting, due to the narrow channel of the H-type reactor.22
a 100 mM phosphate buffer (pH 7.0), deoxygenated by bubbling N2 prior to use. Both the anolyte and catholyte were replaced at the beginning of each feeding cycle. During the startup, the anode chamber was amended with 200 mg/L of the five-compounds mixture (same composition as the MFC feed). After the startup, which lasted for 9 weeks, the total initial concentration of the substrate mixture was increased from 200 to 400, 800, and then to 1200 mg/L. A voltage of +0.6 V was set to the MEC anode relative to the cathode, and the current was recorded every 4 h by the potentiostat. The duration of each feeding cycle was 6−7 days until the current dropped below 0.2 mA. The anolyte and catholyte were continuously mixed magnetically, and the anode and cathode headspaces were initially filled with N2. Two controls were evaluated. Control 1 was used to investigate the stability of the five compounds in the presence of the porous carbon felt and anolyte. Four serum bottles, each containing 100 mL of anolyte and 200 mg/L compound mixture, were kept under N2 headspaces. Two of the bottles contained carbon felt with equivalent quantity (v/v) as in the MEC anode. The concentration of the five compounds was monitored for 7 days. Control 2, set up with a biomass-free anode electrode in the MEC, was used to evaluate the potential contribution of the applied voltage on current production and transformation of the five compounds in the absence of microbial activity. Microbial Community Analysis. Microbial community analysis of the MEC anode was performed after 9 weeks (9 feedings) from the startup. A piece of anode electrode with attached biofilm (approximately 1 cm × 1 cm × 3 cm) was washed several times with the anolyte and then cut into small pieces (1% abundance) was constructed by applying the neighbor-joining algorithm using the program MEGA 6.06. The tree topology was evaluated by bootstrap resampling analysis of 1000 data sets. The representative sequences of the abundant species have been deposited to GenBank, National Center for Biotechnology Information (NCBI; www.ncbi.nlm.nih.gov/) with sequence accession numbers from KT124613 to KT124626. Analytical Methods. The furanic and phenolic compounds were quantified using a high-performance liquid chromatography (HPLC) unit equipped with a UV−vis detector (Agilent 1100, Santa Clara, CA). A HPX-87H column (BioRad, Hercules, CA) was used with an eluent of 15% acetonitrile in 5 mM H2SO4 (v/v) at a flow rate of 0.6 mL/min.16 The wavelength of 280 and 210 nm was used for the furanic and phenolic compounds, respectively. Acetate and other volatile fatty acids were quantified by the same HPLC method at the
■
RESULTS AND DISCUSSION MEC Startup. During the MEC startup period, fed-batch addition of 200 mg/L compound mixture was conducted in repetitive 7-day feeding cycles. Stable maximum current and H2 production was observed by day 28 (Figure S2) and the startup period continued for another 35 days (5 feeding cycles) to confirm stable performance. During the MEC operation, the anolyte and catholyte pH was in the range of 6.7−7.0 and 7.0− 7.3, respectively. The maximum current density (Imax) was 0.16 ± 0.04 mA/cm2 or 3.6 ± 0.9 A/m3, cumulative H2 production was 19.3 ± 1.2 mL (20 °C, 1 atm), and Coulombic efficiency was 44 ± 12% over the last 5 feeding cycles. The five compounds were completely transformed, with sCOD removal of 57 ± 10% during each feeding cycle. An abiotic control assay (Control 2), conducted under the same MEC conditions, with the exception that the anode was not inoculated, resulted in negligible current (
