- Email: [email protected]
Corn stover for biogas production: Effect of steam explosion pretreatment on the gas yields and on the biodegradation kinetics of the primary structural compounds
Accepted Manuscript Corn stover for biogas production: effect of steam explosion pretreatment on the gas yields and on the biodegradation kinetics of the primary structural compounds Javier Lizasoain, Adrian Trulea, Johannes Gittinger, Iris Kral, Gerhard Piringer, Andreas Schedl, Paal J. Nielsen, Antje Potthast, Andreas Gronauer, Alexander Bauer PII: DOI: Reference:
S0960-8524(17)31347-0 http://dx.doi.org/10.1016/j.biortech.2017.08.042 BITE 18651
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
6 June 2017 7 August 2017 8 August 2017
Please cite this article as: Lizasoain, J., Trulea, A., Gittinger, J., Kral, I., Piringer, G., Schedl, A., Nielsen, P.J., Potthast, A., Gronauer, A., Bauer, A., Corn stover for biogas production: effect of steam explosion pretreatment on the gas yields and on the biodegradation kinetics of the primary structural compounds, Bioresource Technology (2017), doi: http://dx.doi.org/10.1016/j.biortech.2017.08.042
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.
Corn stover for biogas production: effect of steam explosion pretreatment on the gas yields and on the biodegradation kinetics of the primary structural compounds Javier Lizasoaina,b, Adrian Truleab, Johannes Gittingerb, Iris Krala,b, Gerhard Piringera,b, Andreas Schedld, Paal J. Nielsenc, Antje Potthast d, Andreas Gronauerb, Alexander Bauera,b,* a
AlpS-GmbH, Centre for Climate Change Adaptation Technologies, Grabenweg 68, A-6010 Innsbruck, Austria
b
University of Natural Resources and Life Sciences, Vienna, Department of Sustainable Agricultural Systems, Institute of Agricultural Engineering, Konrad-Lorenz-Strasse 24, A-3430 Tulln, Austria c
Cambi AS, Skysstasjon 11A, 1383 Asker, Norway
d
University of Natural Resources and Life Sciences, Vienna, Department of Chemistry, Institute of Chemistry of Renewable Resources, Konrad-Lorenz-Strasse 24, A-3430 Tulln, Austria * Corresponding author at: University of Natural Resources and Life Sciences, Vienna, Department of Sustainable Agricultural Systems, Institute of Agricultural Engineering, Konrad-Lorenz-Strasse 24, A-3430 Tulln, Austria. E-mail address: [email protected]
Abstract This study evaluated the effect of steam explosion on the chemical composition and biomethane potential of corn stover using temperatures ranging between 140 and 220 ºC and pretreatment times ranging between 2 and 15 minutes. Biodegradation kinetics during the anaerobic digestion of untreated and corn stover, pretreated at two different intensities, 140 °C for 5 minutes and 180 °C for 5 minutes, were studied in tandem. Results showed that pretreatment at 160 °C for 2 min improved the methane yield by 22%. Harsher pretreatment conditions led to lower hemicellulose contents and methane yields, as well as higher lignin contents, which may be due to the formation of pseudolignin. The biodegradation kinetics trial demonstrated that steam explosion enhances the degradation of structural carbohydrates and lignin.
Keywords: biomass, agricultural residues, maize straw, biofuel, anaerobic digestion
1
1
Introduction
Plant-based resources, such as agricultural residues, can help reduce the use of fossilbased raw materials. Lignocellulosic biomass is the most abundant source of unutilized biomass and its use avoids the competition between food, feed and energy production (Lin and Tanaka, 2006). The use of this type of alternative biomass for the production of gaseous and liquid biofuels has met with growing interest in the last decade (Ho et al., 2014). Agricultural residues can provide large amounts of biomass in a sustainable manner for biogas generation. However, the composition and structure of the lignocellulosic complex limits and often restricts biodegradability during anaerobic digestion (AD), reducing methane production and thus the efficiency of the whole process (Hendriks and Zeeman, 2009; Risberg et al., 2013). In order to weaken and break the links in the lignocellulosic complex and enhance microorganism accessibility to the carbohydrate polymers, a pretreatment step is necessary (Alvira et al., 2010). Steam explosion is currently one of the most attractive and investigated pretreatment methods for both biogas and ethanol production from lignocellulosic biomass. It involves heating up the biomass to a maximum of 240 ºC under high pressure (up to 34 bar) for a few minutes. Then, the pressure is released abruptly, causing an explosive decompression of the lignocellulosic biomass. Steam explosion has been proven to be effective for both biogas and ethanol production using different input materials such as wood (Horn et al., 2011a), grasses (Lizasoain et al., 2016), agricultural residues (Bauer et al., 2009), bioethanol by-products (De Paoli et al., 2011) and municipal waste (Li et al., 2007). The effectiveness of pretreatment on lignocellulosic biomass is generally assessed by measuring the methane yields during its subsequent anaerobic digestion. While it is true 2
that this approach provides important information about the effect of the pretreatment on the final biogas yields, no actual knowledge of the improved biomass degradability can be gained. To learn more about how pretreatment methods improve AD, it is necessary to better understand the biodegradation kinetics of lignocellulosic biomass during the digestion process. Techniques that evaluate the degradation of biomass, such as the rumen derived anaerobic digestion process (RUDAD) and the rumen simulation technique (RUSITEC) have been applied in the field of animal husbandry (Bayané and Guiot, 2011; Machmüller et al., 1998). However, techniques to measure the biodegradation kinetics of biomass during AD are very scarce and only few approaches have been documented, such as the studies performed by Stopp et al. (2009) and Theuretzbacher et al. (2015). The latter is considered a suitable model to investigate the degradation of lignocellulosic biomass during AD and will serve as a basis for this study. The literature on steam exploded corn stover for biogas purposes is very scarce despite corn stover being a widespread source of biomass with an enormous potential for biogas production and steam explosion a technology that is rapidly expanding. Li at al., (2015) studied the effects of steam explosion pretreatment, potassium hydroxide and the sequent pretreatment of potassium hydroxide and steam explosion on the biomethane potential of corn stover. While steam explosion increased the biomethane potential yields by 55% in comparison to the untreated sample, an increase of 80% was obtained after the sequent pretreatment of potassium hydroxide and steam explosion. Siddhu et al. (2016) evaluated the effect of thermal potassium hydroxide, steam explosion and the co-pretreatment of thermal potassium hydroxide and steam explosion on methane yields. They reported maximum improvements in the methane yields in comparison to
3
the untreated biomass of 56%, 40% and 88%, respectively. In addition, Ji et al. (2016) studied the effect of pretreatment with calcium hydroxide, steam explosion and the copretreatment of calcium hydroxide and steam explosion in corn stover. The maximum biomethane yields for each type of pretreatment increased 53%, 34% and 62%, respectively over the yields obtained for untreated corn stover. Despite the positive effect of such co-pretreatments on laboratory-scale trials, the utilization of catalysts in full-scale plants has numerous drawbacks as they are generally expensive, corrode the machinery, modify the pH of the reactor content and have negative environmental repercussions. Therefore, optimizing pretreatment systems that only require the use of heat and water, as steam explosion does, is of great interest for the development of sustainable biofuels. Despite many authors having studied the effects of different pretreatments on the chemical characteristics of biomass or on biomethane yields, there is still a lack of knowledge about the degradation process of biomass during AD. Thus, the aim of this study was to determine the effect of steam explosion on the biomass characterization and methane yields of corn stover without the addition of catalysts and to determine the biodegradation kinetics of the main structural compounds of untreated and pretreated corn stover during the AD process. 2
Material and methods 2.1
Raw material and steam explosion pretreatment
The corn stover used for this study was grown and harvested under standard conditions in Ardud (Romania) in November 2013. The stover was dried on the field, compressed in a round bale and sent to the University of Natural Resources and Life Sciences,
4
Vienna, where it was chopped to a final length of less than 10 cm. The stover was then stored under dry conditions at 4 ºC until pretreatment. The stover was pretreated in March 2014 at the Norwegian University of Life Sciences (NMBU) in Ås (Norway) with a steam explosion unit (Cambi, Asker, Norway), whose performance is described by Horn et al. (2011b). The unit consisted of a 25 KW electric-heated steam boiler (Parat, Flekkefjord, Norway) that could achieve a maximum pressure of 34 bar. An automatic air-actuated valve released the steam into a 20 L pressure vessel that contained the biomass to be pretreated. This valve ensured keeping the pressure in the vessel at the target value. Air was removed from the vessel before every pretreatment was carried out by direct injection of steam until saturation was reached. When each pretreatment was finished, another air-actuated valve was responsible for a rapid pressure reduction in the vessel. This pressure difference transported the pretreated stover to a removable bucket, from which the samples were taken for further tests. The pretreatment step was carried out with temperatures ranging from 140 to 220 °C, using intervals of 20 °C (Table 1). Residence times in the pressure vessel were 2, 5, 10 and 15 minutes. The amount of sample used in all pretreatments was 500 grams. After every pretreatment, the unit was cleaned by running three steam-only pretreatments and flushing the flash tank and bucket with clean tap water. All steam-exploded samples were subsequently vacuum stored at 4 °C until physico-chemical analysis and biogas tests were carried out.
5
2.2
Substrate characterization
The analysis of structural carbohydrates was carried out to gain a deeper understanding of the effect of steam explosion on the chemical properties of corn stover. The composition of untreated and steam exploded biomass was determined by analyzing the following parameters: pH, dry matter, raw ash, volatile solids, cellulose, hemicellulose, acid insoluble lignin (AIL) and crude protein. The dry matter content of the corn stover was determined by drying the samples in an oven at 105 °C until constant weight was reached. The volatile solids (VS) were determined after dry oxidation in a muffle furnace as the difference between the dry matter and the raw ash content (Sluiter et al., 2008a). The cellulose, hemicellulose and lignin of native and pretreated samples were analyzed in duplicates using a two-step acid hydrolysis (Sluiter, 2008b). The dried biomass was milled until the entire sample passed through a 1 mm sieve. 300 mg ± 10 mg of the samples were weighted into pressure tubes. Then 3 ml of 72% sulfuric acid was added and the mixture was incubated at 30 ºC for 60 min. In the second step, 84 ml of deionized water was added to reduce the sulfuric acid to 4% and then the samples were incubated for one hour at 121 ºC in an autoclave. In addition to the samples, a set of sugar recovery standards (SRS) was prepared. The structural carbohydrates were analyzed from the hydrolysis liquor. To determine the acid insoluble lignin (AIL), the autoclaved hydrolysis solution was filtered using filtering crucibles and the remaining insoluble residue was washed, dried overnight at 105 ºC and weighed. Then, the ash content of the dry residue was determined in a muffle furnace at 500 ºC. Chromatographic analysis of the soluble sugars glucose, xylose, galactose, mannose and arabinose was carried out on an Agilent 1100 HPLC system (Agilent Technologies, 6
Waldbronn, Germany) set up with a 7.8 x 100 mm Rezex RFQ-Fast Fruit H+ column (Phenomenex, CA, USA) heated to 80 ºC. The mobile phase consisted of 5 mM sulfuric acid and the flow rate used was 0.4 ml per minute. The HPLC samples were prepared by 5-fold dilution of the previously filtered hydrolysis liquor with HPLC-grade water, followed by a second filtration using membrane filter Filtropur S with a pore size of 0.2 um (Sarstedt, Nümbrecht, Germany). The eluted monosaccharides were detected by recording the refractive index and the analytes were identified and quantified by running standards. Chromatograms were recorded, integrated and analyzed using the Chromeleon 6.8 chromatography software from Dionex (Sunnyvale, CA, USA). To determine the protein content, total nitrogen was determined in duplicates according to the Kjeldahl method using the SpeedDigester K-439 (Büchi, Flawil, Switzerland) and a distillation unit type B-324 (Büchi, Flawil, Switzerland). The protein content was determined from the total nitrogen by applying the conversion factor 6.25 (ConklinBrittain et al., 1999). In addition, the water soluble fraction (WSF) was determined. For this, around 50 grams of biomass was introduced in nylon filter bags. The bags were washed with deionized water until the runoff was clear (following the same procedure explained in section 2.4). The amount of biomass lost during this step is defined as the WSF. Harvesting and baling operations can add substantial amounts of soil-derived ash to the baled corn stover (Bonner et al., 2014). In order to avoid distorted results due to heterogeneous soil contamination, the physico-chemical parameters showed hereinafter are provided as a percentage of volatile solids (VS) instead of dry matter (DM).
7
2.3
Specific biogas and methane yield determination
AD trials were performed in triplicate to determine the specific biogas and methane yields of untreated and pretreated corn stover according to VDI 4630 (2006), using eudiometer batch fermenters with 250 ml capacity. The corn stover samples and the inoculum were weighed out in a ratio of 1:3 based on the VS content. Every fermenter was filled with 200 ml of inoculum consisting of a mixture of two inocula, which came from a previous BMP test using steam-exploded biomass, and from a biogas plant located in Margarethen am Moos (Austria) that was using lignocellulosic biomass and manure as input material. The mixing of the two inocula was carried out in a 1:1 volume ratio and then the mixed inoculum was sieved and diluted to 4% DM. Prior to the performance of the test, the inoculum was stored for 10 days at mesophilic conditions in order to reduce the residual gas production. No nutritional additives were used and microcrystalline cellulose was used as control. The fermenters were continuously stirred with a magnetic system and maintained in water baths at mesophilic conditions (37.5 °C) over the course of the 49 days that the process took place. The production of biogas and methane was monitored on a daily basis and all gas volumes are reported under conditions of 273.15 K and 101.33 kPa per kilogram of volatile solids (L kg-1 VS). The methane (CH4) and carbon dioxide (CO2) contents in the biogas were simultaneously analyzed with the portable gas analyzer “Dräger X-am 7000” (Dräger, Lübeck, Germany). The gas analyzer was calibrated weekly with a gas standard consisting of 33% CH4 and 33% CO2 (Messer, Gumpoldskirchen, Austria). In addition, the portable gas analyzer was used to determine the content of H2, O2 and H2S.
8
2.4
Biodegradation kinetics
The biodegradation kinetics of the main structural components of untreated and pretreated biomass (at 140 ºC and 180 ºC, both for 5 minutes) were investigated using a modification of the rumen simulation technique (RUSITEC) developed by Czerkawski and Breckenridge (1977). The selection of the two steam explosion conditions (140 ºC and 180 ºC for 5 minutes) was based on a pre-test, which revealed that low and mild intensity pretreatments provide higher gas yields. Based on the original method, nitrogen and ash-free nylon filter bags F57 (Ankom Technology, NY, USA) with 25 µm porosity were filled with the samples, sealed with a welding impulse sealer AIE-200 (American International Electric, CA, USA) and fed into three 20-L digesters using one digester per biomass type. Then, the digesters were filled with the same inoculum described in section 2.3 in order to reach an inoculumsubstrate ratio (ISR) of 2 based on DM content. The three digesters were located in independent water baths set at 37.5 °C and equipped with a water lock to allow the produced biogas to exit the digester. At the start of the experiment, and after every sample removal, the headspace of the digesters was flushed with nitrogen. For each variant, digesters were filled with five filter bags. The bags were withdrawn for chemical analysis after 3, 7, 10, 20, and 40 days after the beginning of the experiment. Initial concentrations of the main structural components were determined in a sixth filter bag, which was named “day-0-bag”. Before the chemical analyses were carried out, all bags, including the day-0-bag, were washed with deionized water until the runoff was clear and then dried in an oven at 45 °C. The chemical analyses performed in this trial were dry matter, raw ash, volatile solids,
9
cellulose, hemicellulose and AIL. The methodology used for these analyses is the same as described above in section 2.2. 2.5
Statistical analysis
The tables and figures provided in this study present mean values and standard deviations of the experiments carried out. The computer program SPSS version 21 (SPSS Inc, Chicago, USA) was used to statistically analyze all the data. Homogeneity of variances was preliminarily checked using Levene's test. Multiple means were compared with one-way ANOVA followed by Scheffé test for post hoc comparison. The level of significance was set at p < 0.05.
3
Results and discussion 3.1
Pretreatment effects on corn stover composition
Table 1 shows the influence of steam explosion pretreatment at various intensities on specific physicochemical parameters. As it can be seen, all pretreated samples reported lower pH values than the untreated control (pH 7.5), decreasing from 6.9, obtained at 160 °C for 2 minutes, to 3.9, obtained at 220 °C for 10 minutes. Steam explosion pretreatment causes acetyl groups in the hemicellulose fraction to hydrolyze, leading to a build-up of acetic and uronic acids, thus reducing the pH value in the pretreated samples (Laser et al., 2002). The released acids offer some beneficial effects for the subsequent AD process as they promote higher hydrolysis rates of the cell wall components (Boussaid et al., 2000). Reductions in the pH after intense steam explosion conditions have been observed in different types of biomass (Bauer et al., 2014; Edwards, 2005).
10
As Table 1 shows, the DM content of the untreated sample was 86% FM. This value decreased strongly in all pretreated samples, ranging from 51.6%, after pretreatment at 180 °C for 2 min, to 20.3%, after pretreatment at 200 °C for 15 min. In general terms, DM content decreased under longer pretreatment times. This trend was caused by the utilization of direct steam injection to reach the high pressures and temperatures needed in the steam explosion reactor. Higher temperatures and longer pretreatment times required more steam to maintain the target temperature and therefore, more water was added to the samples. This observation is in line with results from previous studies (Bauer et al., 2009; Bauer et al., 2014; Theuretzbacher et al., 2015). <>
S0960-8524(17)31347-0 http://dx.doi.org/10.1016/j.biortech.2017.08.042 BITE 18651
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
6 June 2017 7 August 2017 8 August 2017
Please cite this article as: Lizasoain, J., Trulea, A., Gittinger, J., Kral, I., Piringer, G., Schedl, A., Nielsen, P.J., Potthast, A., Gronauer, A., Bauer, A., Corn stover for biogas production: effect of steam explosion pretreatment on the gas yields and on the biodegradation kinetics of the primary structural compounds, Bioresource Technology (2017), doi: http://dx.doi.org/10.1016/j.biortech.2017.08.042
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.
Corn stover for biogas production: effect of steam explosion pretreatment on the gas yields and on the biodegradation kinetics of the primary structural compounds Javier Lizasoaina,b, Adrian Truleab, Johannes Gittingerb, Iris Krala,b, Gerhard Piringera,b, Andreas Schedld, Paal J. Nielsenc, Antje Potthast d, Andreas Gronauerb, Alexander Bauera,b,* a
AlpS-GmbH, Centre for Climate Change Adaptation Technologies, Grabenweg 68, A-6010 Innsbruck, Austria
b
University of Natural Resources and Life Sciences, Vienna, Department of Sustainable Agricultural Systems, Institute of Agricultural Engineering, Konrad-Lorenz-Strasse 24, A-3430 Tulln, Austria c
Cambi AS, Skysstasjon 11A, 1383 Asker, Norway
d
University of Natural Resources and Life Sciences, Vienna, Department of Chemistry, Institute of Chemistry of Renewable Resources, Konrad-Lorenz-Strasse 24, A-3430 Tulln, Austria * Corresponding author at: University of Natural Resources and Life Sciences, Vienna, Department of Sustainable Agricultural Systems, Institute of Agricultural Engineering, Konrad-Lorenz-Strasse 24, A-3430 Tulln, Austria. E-mail address: [email protected]
Abstract This study evaluated the effect of steam explosion on the chemical composition and biomethane potential of corn stover using temperatures ranging between 140 and 220 ºC and pretreatment times ranging between 2 and 15 minutes. Biodegradation kinetics during the anaerobic digestion of untreated and corn stover, pretreated at two different intensities, 140 °C for 5 minutes and 180 °C for 5 minutes, were studied in tandem. Results showed that pretreatment at 160 °C for 2 min improved the methane yield by 22%. Harsher pretreatment conditions led to lower hemicellulose contents and methane yields, as well as higher lignin contents, which may be due to the formation of pseudolignin. The biodegradation kinetics trial demonstrated that steam explosion enhances the degradation of structural carbohydrates and lignin.
Keywords: biomass, agricultural residues, maize straw, biofuel, anaerobic digestion
1
1
Introduction
Plant-based resources, such as agricultural residues, can help reduce the use of fossilbased raw materials. Lignocellulosic biomass is the most abundant source of unutilized biomass and its use avoids the competition between food, feed and energy production (Lin and Tanaka, 2006). The use of this type of alternative biomass for the production of gaseous and liquid biofuels has met with growing interest in the last decade (Ho et al., 2014). Agricultural residues can provide large amounts of biomass in a sustainable manner for biogas generation. However, the composition and structure of the lignocellulosic complex limits and often restricts biodegradability during anaerobic digestion (AD), reducing methane production and thus the efficiency of the whole process (Hendriks and Zeeman, 2009; Risberg et al., 2013). In order to weaken and break the links in the lignocellulosic complex and enhance microorganism accessibility to the carbohydrate polymers, a pretreatment step is necessary (Alvira et al., 2010). Steam explosion is currently one of the most attractive and investigated pretreatment methods for both biogas and ethanol production from lignocellulosic biomass. It involves heating up the biomass to a maximum of 240 ºC under high pressure (up to 34 bar) for a few minutes. Then, the pressure is released abruptly, causing an explosive decompression of the lignocellulosic biomass. Steam explosion has been proven to be effective for both biogas and ethanol production using different input materials such as wood (Horn et al., 2011a), grasses (Lizasoain et al., 2016), agricultural residues (Bauer et al., 2009), bioethanol by-products (De Paoli et al., 2011) and municipal waste (Li et al., 2007). The effectiveness of pretreatment on lignocellulosic biomass is generally assessed by measuring the methane yields during its subsequent anaerobic digestion. While it is true 2
that this approach provides important information about the effect of the pretreatment on the final biogas yields, no actual knowledge of the improved biomass degradability can be gained. To learn more about how pretreatment methods improve AD, it is necessary to better understand the biodegradation kinetics of lignocellulosic biomass during the digestion process. Techniques that evaluate the degradation of biomass, such as the rumen derived anaerobic digestion process (RUDAD) and the rumen simulation technique (RUSITEC) have been applied in the field of animal husbandry (Bayané and Guiot, 2011; Machmüller et al., 1998). However, techniques to measure the biodegradation kinetics of biomass during AD are very scarce and only few approaches have been documented, such as the studies performed by Stopp et al. (2009) and Theuretzbacher et al. (2015). The latter is considered a suitable model to investigate the degradation of lignocellulosic biomass during AD and will serve as a basis for this study. The literature on steam exploded corn stover for biogas purposes is very scarce despite corn stover being a widespread source of biomass with an enormous potential for biogas production and steam explosion a technology that is rapidly expanding. Li at al., (2015) studied the effects of steam explosion pretreatment, potassium hydroxide and the sequent pretreatment of potassium hydroxide and steam explosion on the biomethane potential of corn stover. While steam explosion increased the biomethane potential yields by 55% in comparison to the untreated sample, an increase of 80% was obtained after the sequent pretreatment of potassium hydroxide and steam explosion. Siddhu et al. (2016) evaluated the effect of thermal potassium hydroxide, steam explosion and the co-pretreatment of thermal potassium hydroxide and steam explosion on methane yields. They reported maximum improvements in the methane yields in comparison to
3
the untreated biomass of 56%, 40% and 88%, respectively. In addition, Ji et al. (2016) studied the effect of pretreatment with calcium hydroxide, steam explosion and the copretreatment of calcium hydroxide and steam explosion in corn stover. The maximum biomethane yields for each type of pretreatment increased 53%, 34% and 62%, respectively over the yields obtained for untreated corn stover. Despite the positive effect of such co-pretreatments on laboratory-scale trials, the utilization of catalysts in full-scale plants has numerous drawbacks as they are generally expensive, corrode the machinery, modify the pH of the reactor content and have negative environmental repercussions. Therefore, optimizing pretreatment systems that only require the use of heat and water, as steam explosion does, is of great interest for the development of sustainable biofuels. Despite many authors having studied the effects of different pretreatments on the chemical characteristics of biomass or on biomethane yields, there is still a lack of knowledge about the degradation process of biomass during AD. Thus, the aim of this study was to determine the effect of steam explosion on the biomass characterization and methane yields of corn stover without the addition of catalysts and to determine the biodegradation kinetics of the main structural compounds of untreated and pretreated corn stover during the AD process. 2
Material and methods 2.1
Raw material and steam explosion pretreatment
The corn stover used for this study was grown and harvested under standard conditions in Ardud (Romania) in November 2013. The stover was dried on the field, compressed in a round bale and sent to the University of Natural Resources and Life Sciences,
4
Vienna, where it was chopped to a final length of less than 10 cm. The stover was then stored under dry conditions at 4 ºC until pretreatment. The stover was pretreated in March 2014 at the Norwegian University of Life Sciences (NMBU) in Ås (Norway) with a steam explosion unit (Cambi, Asker, Norway), whose performance is described by Horn et al. (2011b). The unit consisted of a 25 KW electric-heated steam boiler (Parat, Flekkefjord, Norway) that could achieve a maximum pressure of 34 bar. An automatic air-actuated valve released the steam into a 20 L pressure vessel that contained the biomass to be pretreated. This valve ensured keeping the pressure in the vessel at the target value. Air was removed from the vessel before every pretreatment was carried out by direct injection of steam until saturation was reached. When each pretreatment was finished, another air-actuated valve was responsible for a rapid pressure reduction in the vessel. This pressure difference transported the pretreated stover to a removable bucket, from which the samples were taken for further tests. The pretreatment step was carried out with temperatures ranging from 140 to 220 °C, using intervals of 20 °C (Table 1). Residence times in the pressure vessel were 2, 5, 10 and 15 minutes. The amount of sample used in all pretreatments was 500 grams. After every pretreatment, the unit was cleaned by running three steam-only pretreatments and flushing the flash tank and bucket with clean tap water. All steam-exploded samples were subsequently vacuum stored at 4 °C until physico-chemical analysis and biogas tests were carried out.
5
2.2
Substrate characterization
The analysis of structural carbohydrates was carried out to gain a deeper understanding of the effect of steam explosion on the chemical properties of corn stover. The composition of untreated and steam exploded biomass was determined by analyzing the following parameters: pH, dry matter, raw ash, volatile solids, cellulose, hemicellulose, acid insoluble lignin (AIL) and crude protein. The dry matter content of the corn stover was determined by drying the samples in an oven at 105 °C until constant weight was reached. The volatile solids (VS) were determined after dry oxidation in a muffle furnace as the difference between the dry matter and the raw ash content (Sluiter et al., 2008a). The cellulose, hemicellulose and lignin of native and pretreated samples were analyzed in duplicates using a two-step acid hydrolysis (Sluiter, 2008b). The dried biomass was milled until the entire sample passed through a 1 mm sieve. 300 mg ± 10 mg of the samples were weighted into pressure tubes. Then 3 ml of 72% sulfuric acid was added and the mixture was incubated at 30 ºC for 60 min. In the second step, 84 ml of deionized water was added to reduce the sulfuric acid to 4% and then the samples were incubated for one hour at 121 ºC in an autoclave. In addition to the samples, a set of sugar recovery standards (SRS) was prepared. The structural carbohydrates were analyzed from the hydrolysis liquor. To determine the acid insoluble lignin (AIL), the autoclaved hydrolysis solution was filtered using filtering crucibles and the remaining insoluble residue was washed, dried overnight at 105 ºC and weighed. Then, the ash content of the dry residue was determined in a muffle furnace at 500 ºC. Chromatographic analysis of the soluble sugars glucose, xylose, galactose, mannose and arabinose was carried out on an Agilent 1100 HPLC system (Agilent Technologies, 6
Waldbronn, Germany) set up with a 7.8 x 100 mm Rezex RFQ-Fast Fruit H+ column (Phenomenex, CA, USA) heated to 80 ºC. The mobile phase consisted of 5 mM sulfuric acid and the flow rate used was 0.4 ml per minute. The HPLC samples were prepared by 5-fold dilution of the previously filtered hydrolysis liquor with HPLC-grade water, followed by a second filtration using membrane filter Filtropur S with a pore size of 0.2 um (Sarstedt, Nümbrecht, Germany). The eluted monosaccharides were detected by recording the refractive index and the analytes were identified and quantified by running standards. Chromatograms were recorded, integrated and analyzed using the Chromeleon 6.8 chromatography software from Dionex (Sunnyvale, CA, USA). To determine the protein content, total nitrogen was determined in duplicates according to the Kjeldahl method using the SpeedDigester K-439 (Büchi, Flawil, Switzerland) and a distillation unit type B-324 (Büchi, Flawil, Switzerland). The protein content was determined from the total nitrogen by applying the conversion factor 6.25 (ConklinBrittain et al., 1999). In addition, the water soluble fraction (WSF) was determined. For this, around 50 grams of biomass was introduced in nylon filter bags. The bags were washed with deionized water until the runoff was clear (following the same procedure explained in section 2.4). The amount of biomass lost during this step is defined as the WSF. Harvesting and baling operations can add substantial amounts of soil-derived ash to the baled corn stover (Bonner et al., 2014). In order to avoid distorted results due to heterogeneous soil contamination, the physico-chemical parameters showed hereinafter are provided as a percentage of volatile solids (VS) instead of dry matter (DM).
7
2.3
Specific biogas and methane yield determination
AD trials were performed in triplicate to determine the specific biogas and methane yields of untreated and pretreated corn stover according to VDI 4630 (2006), using eudiometer batch fermenters with 250 ml capacity. The corn stover samples and the inoculum were weighed out in a ratio of 1:3 based on the VS content. Every fermenter was filled with 200 ml of inoculum consisting of a mixture of two inocula, which came from a previous BMP test using steam-exploded biomass, and from a biogas plant located in Margarethen am Moos (Austria) that was using lignocellulosic biomass and manure as input material. The mixing of the two inocula was carried out in a 1:1 volume ratio and then the mixed inoculum was sieved and diluted to 4% DM. Prior to the performance of the test, the inoculum was stored for 10 days at mesophilic conditions in order to reduce the residual gas production. No nutritional additives were used and microcrystalline cellulose was used as control. The fermenters were continuously stirred with a magnetic system and maintained in water baths at mesophilic conditions (37.5 °C) over the course of the 49 days that the process took place. The production of biogas and methane was monitored on a daily basis and all gas volumes are reported under conditions of 273.15 K and 101.33 kPa per kilogram of volatile solids (L kg-1 VS). The methane (CH4) and carbon dioxide (CO2) contents in the biogas were simultaneously analyzed with the portable gas analyzer “Dräger X-am 7000” (Dräger, Lübeck, Germany). The gas analyzer was calibrated weekly with a gas standard consisting of 33% CH4 and 33% CO2 (Messer, Gumpoldskirchen, Austria). In addition, the portable gas analyzer was used to determine the content of H2, O2 and H2S.
8
2.4
Biodegradation kinetics
The biodegradation kinetics of the main structural components of untreated and pretreated biomass (at 140 ºC and 180 ºC, both for 5 minutes) were investigated using a modification of the rumen simulation technique (RUSITEC) developed by Czerkawski and Breckenridge (1977). The selection of the two steam explosion conditions (140 ºC and 180 ºC for 5 minutes) was based on a pre-test, which revealed that low and mild intensity pretreatments provide higher gas yields. Based on the original method, nitrogen and ash-free nylon filter bags F57 (Ankom Technology, NY, USA) with 25 µm porosity were filled with the samples, sealed with a welding impulse sealer AIE-200 (American International Electric, CA, USA) and fed into three 20-L digesters using one digester per biomass type. Then, the digesters were filled with the same inoculum described in section 2.3 in order to reach an inoculumsubstrate ratio (ISR) of 2 based on DM content. The three digesters were located in independent water baths set at 37.5 °C and equipped with a water lock to allow the produced biogas to exit the digester. At the start of the experiment, and after every sample removal, the headspace of the digesters was flushed with nitrogen. For each variant, digesters were filled with five filter bags. The bags were withdrawn for chemical analysis after 3, 7, 10, 20, and 40 days after the beginning of the experiment. Initial concentrations of the main structural components were determined in a sixth filter bag, which was named “day-0-bag”. Before the chemical analyses were carried out, all bags, including the day-0-bag, were washed with deionized water until the runoff was clear and then dried in an oven at 45 °C. The chemical analyses performed in this trial were dry matter, raw ash, volatile solids,
9
cellulose, hemicellulose and AIL. The methodology used for these analyses is the same as described above in section 2.2. 2.5
Statistical analysis
The tables and figures provided in this study present mean values and standard deviations of the experiments carried out. The computer program SPSS version 21 (SPSS Inc, Chicago, USA) was used to statistically analyze all the data. Homogeneity of variances was preliminarily checked using Levene's test. Multiple means were compared with one-way ANOVA followed by Scheffé test for post hoc comparison. The level of significance was set at p < 0.05.
3
Results and discussion 3.1
Pretreatment effects on corn stover composition
Table 1 shows the influence of steam explosion pretreatment at various intensities on specific physicochemical parameters. As it can be seen, all pretreated samples reported lower pH values than the untreated control (pH 7.5), decreasing from 6.9, obtained at 160 °C for 2 minutes, to 3.9, obtained at 220 °C for 10 minutes. Steam explosion pretreatment causes acetyl groups in the hemicellulose fraction to hydrolyze, leading to a build-up of acetic and uronic acids, thus reducing the pH value in the pretreated samples (Laser et al., 2002). The released acids offer some beneficial effects for the subsequent AD process as they promote higher hydrolysis rates of the cell wall components (Boussaid et al., 2000). Reductions in the pH after intense steam explosion conditions have been observed in different types of biomass (Bauer et al., 2014; Edwards, 2005).
10
As Table 1 shows, the DM content of the untreated sample was 86% FM. This value decreased strongly in all pretreated samples, ranging from 51.6%, after pretreatment at 180 °C for 2 min, to 20.3%, after pretreatment at 200 °C for 15 min. In general terms, DM content decreased under longer pretreatment times. This trend was caused by the utilization of direct steam injection to reach the high pressures and temperatures needed in the steam explosion reactor. Higher temperatures and longer pretreatment times required more steam to maintain the target temperature and therefore, more water was added to the samples. This observation is in line with results from previous studies (Bauer et al., 2009; Bauer et al., 2014; Theuretzbacher et al., 2015). <