Appl Biochem Biotechnol (2015) 175:2258–2265 DOI 10.1007/s12010-014-1424-y
The Influence of the Buffering Capacity on the Production of Organic Acids and Alcohols from Wastewater in Anaerobic Reactor A. J. Silva & E. Pozzi & E. Foresti & M. Zaiat
Received: 23 May 2014 / Accepted: 18 November 2014 / Published online: 6 December 2014 # Springer Science+Business Media New York 2014
Abstract Some bacteria common in anaerobic digestion process can ferment a broad variety of organic compounds to organic acids, alcohols, and hydrogen, which can be used as biofuels. Researches are necessary to control the microbial interactions in favor of the alcohol production, as intermediary products of the anaerobic digestion of organic compounds. This paper reports on the effect of buffering capacity on the production of organic acids and alcohols from wastewater by a natural mixed bacterial culture. The hypothesis tested was that the increase of the buffering capacity by supplementation of sodium bicarbonate in the influent results in benefits for alcohol production by anaerobic fermentation of wastewater. When the influent was not supplemented with sodium bicarbonate, the chemical oxygen demand (COD)-ethanol and COD-methanol detected in the effluent corresponded to 22.5 and 12.7 % of the CODsucrose consumed. Otherwise, when the reactor was fed with influent containing 0.5 g/L of sodium bicarbonate, the COD-ethanol and COD-methanol were effluents that corresponded to 39.2 and 29.6 % of the COD-sucrose consumed. Therefore, the alcohol production by supplementation of the influent with sodium bicarbonate was 33.6 % higher than the fermentation of the influent without sodium bicarbonate. Keywords Anaerobic fermentation . Biofuel . Wastewater . Butanol . Ethanol
Introduction Bacteria from the Clostridium genus are common in anaerobic digestion processes of organic compounds. These microorganisms can ferment a broad variety of organic substrates and A. J. Silva (*) School of Agricultural Engineering, University of Campinas (UNICAMP), Campinas, Brazil e-mail: [email protected] E. Pozzi : E. Foresti : M. Zaiat Research Center, Development and Innovation in Environmental Engineering—CPDI-EA—São Carlos School of Engineering—EESC, University of São Paulo—USP—Environmental Engineering, Bloco 4-F. Av. João Dagnone, 1100 - Santa Angelina, São Carlos, SP 13563-120, Brazil M. Zaiat e-mail: [email protected]
Appl Biochem Biotechnol (2015) 175:2258–2265
2259
organic acids, such as pentose, hexose, monosaccharides, and polysaccharides, that are present in several industrial wastewaters, including those generated by the sugar cane industry, dairy industry, brewery plants, paper mills, poultry slaughterhouses, and sewage. Besides organic acids, the metabolic versatility and availability of Clostridium in water and soil has been the subject of researches with the purpose of producing biofuels, mainly hydrogen, ethanol, and butanol. Clostridium acetobutylicum, Clostridium beijerinckii, Clostridium saccharobutylicum, and Clostridium saccharoperbutylacetonicum produce butanol and ethanol through a two-phase anaerobic fermentation, which involves the production and assimilation of organic acids, mainly acetic and butyric acids, in addition to H2 and CO2. This process is called acetonebutanol-ethanol (ABE) fermentation and has prospered during the 1960s; the subsequent increase in the costs of feedstock and advances in petrochemical processes demotivated ABE fermentation [1]. Currently, there is a renewed interest in ABE fermentation by using agricultural waste as feedstock. However, this process is only possible by applying an inline-solvent recovery system due to product inhibition. C. acetobutylicum DSM 1731 produced up to 20 g/L of the ABE solvent on a medium containing potato powder 14 % (w/v) without solvent recovery, in fed-batch fermentation. When an inline perstraction system for solvent recovery was used, the ABE solvent production increased from 0.13 to 0.23 g/g of dry potato powder [2]. Alternatives for solvent recovered from ABE process are liquid phase adsorption and gas phase desorption [3], pervaporation [4], and vacuum process [5]. A three-phase system for the anaerobic digestion process driven toward acid fermentation and solvent production followed by a methane production phase could be applied to integrate the alcohol production and the complete anaerobic digestion process of wastewaters. The medium pH is very important for the production of organic acids and alcohols. The optimum pH for butyric acid production is 6.5 and for butanol production is 4.8 [1]. He et al. [6] reported that the most effective pH to obtain a high quantity of butyric acid is that at which only undissociated acid is present. Hartmanis and Gatenbeck [7] observed more butyric fermentation when the medium pH was maintained at 6.0 by adding potassium hydroxide, and the end products of this fermentation were predominantly butyric and acetic acids. Ezeji et al. [8] observed a sudden cessation of ABE production due to “acid crash,” which occasionally occurs in pH-uncontrolled batch fermentation. The increase of the buffering capacity of the medium results, in turn, in increased growth and carbohydrate utilization rates [1]. Many challenges need to be addressed before the optimization of the ABE fermentation from wastewaters in order to reduce the toxicity due to solvents, such as butanol. Moreover, large-scale butanol production plants use a specific microorganism such as C. acetobutylicum, and wastewater treatment plants operate with a natural microbial consortium involving interaction between interspecies. Researches are necessary to control the interactions in favor of the production of butanol and ethanol as intermediary products from the anaerobic digestion of organic compounds. In this context, this study evaluated the effect of the influent supplementation with sodium bicarbonate, aiming at increasing the buffering capacity of the reactor in order to benefit the alcohol production through a natural mixed bacterial culture in the anaerobic fermentation process.
2260
Appl Biochem Biotechnol (2015) 175:2258–2265
Material and Methods Experimental Apparatus The experimental apparatus is represented by the diagram in Fig. 1. A single continuous flow packed-bed acrylic reactor, with an overall volume of 3.17 L, was used in this experiment. The reactor had three sections: bottom (0.36 L), for influent distribution; middle (2.45 L), filled with polyethylene pellets used as support material; and top (0.36 L), to separate the liquid and gaseous phases. The three sections of the reactor were separated by stainless steel screens with 3-mm openings. The bed porosity was 0.48, thus resulting in a working volume of 1.81 L in the mid-section occupied by the support material. The biogas production was measured using a volumetric flow meter equipped with a pulse counter (Ritter®). The experiment took place inside a controlled temperature chamber at 25 °C (77 °F). Substrate The sucrose-based synthetic wastewater was prepared containing sucrose (1781.24 mg/L), urea (40.0 mg/L), nickel sulfate (1.00 mg/L), ferrous sulfate (5.00 mg/L), ferric chloride (0.50 mg/ L), calcium chloride (4.12 mg/L), cobalt chloride (0.08 mg/L), selenium oxide (0.072 mg/L), monobasic potassium phosphate (10.72 mg/L), dibasic potassium phosphate (2.60 mg/L), and dibasic sodium phosphate (5.52 mg/L). The substrate was stored in a reservoir maintained in a refrigerator with temperature set at about 4 °C (39.2 °F) in order to reduce the fermentation effects from outside the reactor. The pH, chemical oxygen demand (COD), and sucrose concentration resulted in 7.1±0.1, 2011±248, and 1395±133 mg/L, respectively. Inoculum The inoculation was obtained by an auto-fermentation strategy according to Leite et al. [9]. The substrate was maintained for 3 days in contact with the external environment, which
Fig. 1 Diagram of the experimental apparatus
Appl Biochem Biotechnol (2015) 175:2258–2265
2261
permitted the growth of microorganisms present in the atmosphere. Afterwards, the fermented substrate was recycled in the reactor in a closed circuit for 48 h. Reactor Operation The substrate was pumped from the reservoir into the reactor by a Hidrotec® diaphragm pump with flow rate set at 915 ± 0.085 L/h, resulting in an average hydraulic detention time of 2 h. This experiment was carried out in two steps, the first without the addition of sodium bicarbonate, while for the second, the influent received 0.5 g/L of sodium bicarbonate to increase the buffering capacity, with the purpose of attaining a better performance in the conversion of sucrose to VFA and alcohols. Chemical Analysis The sucrose concentrations, volatile fatty acids, alcohols, and pH were analyzed in both the influent and effluent of the reactors at least twice a week. For all analyses, except pH, samples were filtered in a 1.2-μm membrane. The concentrations of sucrose, volatile fatty acids, and alcohols were determined concomitantly by high performance liquid chromatography (HPLC) with UV/DAD and RID detectors [10]. Quantitative and Qualitative Analysis of the Biomass After operating the reactor, at each step, samples from the support material were collected to determine the biomass concentration, which was measured as total volatile suspended solids, and for molecular characterization. Three samples of the support material with attached biomass were taken from the mid-region, one from the center, one close to the top, and the other close to the bottom region. For biomass detachment from the support material, the samples were placed into a 500-mL Duran’s flask containing 100 mL of distilled water, manually stirred for 20 min. The analyses of total and volatile suspended solids were carried out according to Standard Methods for Examination of Water and Wastewater [11]. Part of the detached biomass was subjected to the molecular characterization to evaluate the phylogenetic similarity between the microorganisms in the three regions of the reactor. The genomic DNA of the microorganisms was phenol-chloroform extracted according to Griffiths et al. [12] and used for amplification of a 16S rRNA fragment by polymerase chain reaction (PCR), employing a specific primer for the bacteria domains. The primer used was 968FGC (5′-AACGCGAAGAACCTTAC-3′) and 1392R (5′ACGGGCGGTGTGTAC-3′) according to Nielsen et al. [13]. The amplified DNA fragments were separated by denaturing gradient gel electrophoresis (DGGE), according to Muyzer et al. [14]. Electrophoresis gel was made with acrylamide/bis 40 % and the denaturant gradient of DNA (urea/formamide) 45–65 % for the domain bacteria. The electrophoresis was conducted at 65 °C (149 °F), 75 V for 16 h. After being stained with ethidium bromide for 20 min, the gel was transferred to an Eagle Eye TM III UV chamber (Stratagene) coupled to a computer with Eagle light UV software for visualization of the bands. The dendrogram was built using the Bionumeric software version 2.5, and Pearson’s correlation coefficient was used to analyze the phylogenetic similarity of the microorganisms.
Appl Biochem Biotechnol (2015) 175:2258–2265
2262
Statistical Analysis Data were statistically treated by hypothesis testing with significance level α equal to 0.05 using the BioEstat 5.0 software. Distribution frequency data were analyzed using the D’Agostino [15] and Lilliesfors [16] methods. Hypothesis t test to the parametric data and the median test [17] to the non-parametric data were applied.
Results and Discussion Initially, the similarities observed for the pH values in the influent and effluent streams indicated that the acid fermentation was poor, for both feeding conditions, i.e., influent with and without sodium bicarbonate. Only after 16 days of operation, it was verified that the effluent pH values were much lower (p
The Influence of the Buffering Capacity on the Production of Organic Acids and Alcohols from Wastewater in Anaerobic Reactor A. J. Silva & E. Pozzi & E. Foresti & M. Zaiat
Received: 23 May 2014 / Accepted: 18 November 2014 / Published online: 6 December 2014 # Springer Science+Business Media New York 2014
Abstract Some bacteria common in anaerobic digestion process can ferment a broad variety of organic compounds to organic acids, alcohols, and hydrogen, which can be used as biofuels. Researches are necessary to control the microbial interactions in favor of the alcohol production, as intermediary products of the anaerobic digestion of organic compounds. This paper reports on the effect of buffering capacity on the production of organic acids and alcohols from wastewater by a natural mixed bacterial culture. The hypothesis tested was that the increase of the buffering capacity by supplementation of sodium bicarbonate in the influent results in benefits for alcohol production by anaerobic fermentation of wastewater. When the influent was not supplemented with sodium bicarbonate, the chemical oxygen demand (COD)-ethanol and COD-methanol detected in the effluent corresponded to 22.5 and 12.7 % of the CODsucrose consumed. Otherwise, when the reactor was fed with influent containing 0.5 g/L of sodium bicarbonate, the COD-ethanol and COD-methanol were effluents that corresponded to 39.2 and 29.6 % of the COD-sucrose consumed. Therefore, the alcohol production by supplementation of the influent with sodium bicarbonate was 33.6 % higher than the fermentation of the influent without sodium bicarbonate. Keywords Anaerobic fermentation . Biofuel . Wastewater . Butanol . Ethanol
Introduction Bacteria from the Clostridium genus are common in anaerobic digestion processes of organic compounds. These microorganisms can ferment a broad variety of organic substrates and A. J. Silva (*) School of Agricultural Engineering, University of Campinas (UNICAMP), Campinas, Brazil e-mail: [email protected] E. Pozzi : E. Foresti : M. Zaiat Research Center, Development and Innovation in Environmental Engineering—CPDI-EA—São Carlos School of Engineering—EESC, University of São Paulo—USP—Environmental Engineering, Bloco 4-F. Av. João Dagnone, 1100 - Santa Angelina, São Carlos, SP 13563-120, Brazil M. Zaiat e-mail: [email protected]
Appl Biochem Biotechnol (2015) 175:2258–2265
2259
organic acids, such as pentose, hexose, monosaccharides, and polysaccharides, that are present in several industrial wastewaters, including those generated by the sugar cane industry, dairy industry, brewery plants, paper mills, poultry slaughterhouses, and sewage. Besides organic acids, the metabolic versatility and availability of Clostridium in water and soil has been the subject of researches with the purpose of producing biofuels, mainly hydrogen, ethanol, and butanol. Clostridium acetobutylicum, Clostridium beijerinckii, Clostridium saccharobutylicum, and Clostridium saccharoperbutylacetonicum produce butanol and ethanol through a two-phase anaerobic fermentation, which involves the production and assimilation of organic acids, mainly acetic and butyric acids, in addition to H2 and CO2. This process is called acetonebutanol-ethanol (ABE) fermentation and has prospered during the 1960s; the subsequent increase in the costs of feedstock and advances in petrochemical processes demotivated ABE fermentation [1]. Currently, there is a renewed interest in ABE fermentation by using agricultural waste as feedstock. However, this process is only possible by applying an inline-solvent recovery system due to product inhibition. C. acetobutylicum DSM 1731 produced up to 20 g/L of the ABE solvent on a medium containing potato powder 14 % (w/v) without solvent recovery, in fed-batch fermentation. When an inline perstraction system for solvent recovery was used, the ABE solvent production increased from 0.13 to 0.23 g/g of dry potato powder [2]. Alternatives for solvent recovered from ABE process are liquid phase adsorption and gas phase desorption [3], pervaporation [4], and vacuum process [5]. A three-phase system for the anaerobic digestion process driven toward acid fermentation and solvent production followed by a methane production phase could be applied to integrate the alcohol production and the complete anaerobic digestion process of wastewaters. The medium pH is very important for the production of organic acids and alcohols. The optimum pH for butyric acid production is 6.5 and for butanol production is 4.8 [1]. He et al. [6] reported that the most effective pH to obtain a high quantity of butyric acid is that at which only undissociated acid is present. Hartmanis and Gatenbeck [7] observed more butyric fermentation when the medium pH was maintained at 6.0 by adding potassium hydroxide, and the end products of this fermentation were predominantly butyric and acetic acids. Ezeji et al. [8] observed a sudden cessation of ABE production due to “acid crash,” which occasionally occurs in pH-uncontrolled batch fermentation. The increase of the buffering capacity of the medium results, in turn, in increased growth and carbohydrate utilization rates [1]. Many challenges need to be addressed before the optimization of the ABE fermentation from wastewaters in order to reduce the toxicity due to solvents, such as butanol. Moreover, large-scale butanol production plants use a specific microorganism such as C. acetobutylicum, and wastewater treatment plants operate with a natural microbial consortium involving interaction between interspecies. Researches are necessary to control the interactions in favor of the production of butanol and ethanol as intermediary products from the anaerobic digestion of organic compounds. In this context, this study evaluated the effect of the influent supplementation with sodium bicarbonate, aiming at increasing the buffering capacity of the reactor in order to benefit the alcohol production through a natural mixed bacterial culture in the anaerobic fermentation process.
2260
Appl Biochem Biotechnol (2015) 175:2258–2265
Material and Methods Experimental Apparatus The experimental apparatus is represented by the diagram in Fig. 1. A single continuous flow packed-bed acrylic reactor, with an overall volume of 3.17 L, was used in this experiment. The reactor had three sections: bottom (0.36 L), for influent distribution; middle (2.45 L), filled with polyethylene pellets used as support material; and top (0.36 L), to separate the liquid and gaseous phases. The three sections of the reactor were separated by stainless steel screens with 3-mm openings. The bed porosity was 0.48, thus resulting in a working volume of 1.81 L in the mid-section occupied by the support material. The biogas production was measured using a volumetric flow meter equipped with a pulse counter (Ritter®). The experiment took place inside a controlled temperature chamber at 25 °C (77 °F). Substrate The sucrose-based synthetic wastewater was prepared containing sucrose (1781.24 mg/L), urea (40.0 mg/L), nickel sulfate (1.00 mg/L), ferrous sulfate (5.00 mg/L), ferric chloride (0.50 mg/ L), calcium chloride (4.12 mg/L), cobalt chloride (0.08 mg/L), selenium oxide (0.072 mg/L), monobasic potassium phosphate (10.72 mg/L), dibasic potassium phosphate (2.60 mg/L), and dibasic sodium phosphate (5.52 mg/L). The substrate was stored in a reservoir maintained in a refrigerator with temperature set at about 4 °C (39.2 °F) in order to reduce the fermentation effects from outside the reactor. The pH, chemical oxygen demand (COD), and sucrose concentration resulted in 7.1±0.1, 2011±248, and 1395±133 mg/L, respectively. Inoculum The inoculation was obtained by an auto-fermentation strategy according to Leite et al. [9]. The substrate was maintained for 3 days in contact with the external environment, which
Fig. 1 Diagram of the experimental apparatus
Appl Biochem Biotechnol (2015) 175:2258–2265
2261
permitted the growth of microorganisms present in the atmosphere. Afterwards, the fermented substrate was recycled in the reactor in a closed circuit for 48 h. Reactor Operation The substrate was pumped from the reservoir into the reactor by a Hidrotec® diaphragm pump with flow rate set at 915 ± 0.085 L/h, resulting in an average hydraulic detention time of 2 h. This experiment was carried out in two steps, the first without the addition of sodium bicarbonate, while for the second, the influent received 0.5 g/L of sodium bicarbonate to increase the buffering capacity, with the purpose of attaining a better performance in the conversion of sucrose to VFA and alcohols. Chemical Analysis The sucrose concentrations, volatile fatty acids, alcohols, and pH were analyzed in both the influent and effluent of the reactors at least twice a week. For all analyses, except pH, samples were filtered in a 1.2-μm membrane. The concentrations of sucrose, volatile fatty acids, and alcohols were determined concomitantly by high performance liquid chromatography (HPLC) with UV/DAD and RID detectors [10]. Quantitative and Qualitative Analysis of the Biomass After operating the reactor, at each step, samples from the support material were collected to determine the biomass concentration, which was measured as total volatile suspended solids, and for molecular characterization. Three samples of the support material with attached biomass were taken from the mid-region, one from the center, one close to the top, and the other close to the bottom region. For biomass detachment from the support material, the samples were placed into a 500-mL Duran’s flask containing 100 mL of distilled water, manually stirred for 20 min. The analyses of total and volatile suspended solids were carried out according to Standard Methods for Examination of Water and Wastewater [11]. Part of the detached biomass was subjected to the molecular characterization to evaluate the phylogenetic similarity between the microorganisms in the three regions of the reactor. The genomic DNA of the microorganisms was phenol-chloroform extracted according to Griffiths et al. [12] and used for amplification of a 16S rRNA fragment by polymerase chain reaction (PCR), employing a specific primer for the bacteria domains. The primer used was 968FGC (5′-AACGCGAAGAACCTTAC-3′) and 1392R (5′ACGGGCGGTGTGTAC-3′) according to Nielsen et al. [13]. The amplified DNA fragments were separated by denaturing gradient gel electrophoresis (DGGE), according to Muyzer et al. [14]. Electrophoresis gel was made with acrylamide/bis 40 % and the denaturant gradient of DNA (urea/formamide) 45–65 % for the domain bacteria. The electrophoresis was conducted at 65 °C (149 °F), 75 V for 16 h. After being stained with ethidium bromide for 20 min, the gel was transferred to an Eagle Eye TM III UV chamber (Stratagene) coupled to a computer with Eagle light UV software for visualization of the bands. The dendrogram was built using the Bionumeric software version 2.5, and Pearson’s correlation coefficient was used to analyze the phylogenetic similarity of the microorganisms.
Appl Biochem Biotechnol (2015) 175:2258–2265
2262
Statistical Analysis Data were statistically treated by hypothesis testing with significance level α equal to 0.05 using the BioEstat 5.0 software. Distribution frequency data were analyzed using the D’Agostino [15] and Lilliesfors [16] methods. Hypothesis t test to the parametric data and the median test [17] to the non-parametric data were applied.
Results and Discussion Initially, the similarities observed for the pH values in the influent and effluent streams indicated that the acid fermentation was poor, for both feeding conditions, i.e., influent with and without sodium bicarbonate. Only after 16 days of operation, it was verified that the effluent pH values were much lower (p
