Effect of steam explosion pretreatment on the anaerobic digestion of rice straw

Abstract

Rice straw pretreated under various conditions of temperature, ranging from 200 to 220 °C, and time, from 60 to 240 s, was used as the substrate in an anaerobic biogas recovery process. Steam explosion-pretreated rice straw displayed a significant improvement in physicochemical properties compared to untreated rice straw. The biogas production rate increased in all pretreated rice straw systems with shortened start-up periods, and the highest biogas production rate reached 328.7 mL g⁻¹ TS (total solid content) under steam explosion pretreatment conditions of 200 °C/120 s, corresponding to a 51% increase. In addition, upon pretreatment at 200 °C/120 s, the degradation rates of cellulose and hemicellulose in the system reached 53.46% and 49.54%, which were 13.72% and 16.79% higher than in the control, respectively. PCR-DGGE analysis showed that the distributions of specific species of bacteria and archaea varied among different samples. There were clear differences in the bacterial population between pretreated and untreated groups during the set-up period and early stage. During the mid and final stages, pretreated systems had more diverse communities in the digester than the untreated system. Steam explosion pretreatment of the rice straw also led to the earlier presence of cellulolytic bacteria. Furthermore, the species of cellulolytic bacteria in pretreated systems were Clostridium sp. while those in the untreated reactor were Pseudomonas sp. The succession of archaea in the microbial community at different stages in the pretreated and untreated systems was not as obvious as that of bacteria. The results indicate that the physicochemical properties of rice straw were altered by the steam explosion pretreatment, which led to more efficient biogas production due to changes in the bacterial and archaeal species in the pretreated system.

---

1 Introduction

More than 700 million tons of crop straws are generated in China every year, and they constitute one of the most abundant crop residues.¹ The amount of rice straw generated in China is between 180 and 280 million tons annually, more than half of which remains unused.² Unsuitable disposal of rice straw not only wastes resources but also leads to environmental problems. Anaerobic digestion (AD) has been widely used as an efficient means of waste treatment and biogas harvesting throughout the world. This process demonstrates a wide range of beneficial effects on the environment, energy and human health, especially in rural areas of developing countries.³ Currently, many research studies are devoted to investigating the potential of using agricultural wastes to produce biogas.^4–7

However, the main components of rice straw are cellulose, hemicellulose and lignin, which together form the lignocellulose complex, and hydrolysis is the rate-limiting step in AD of rice straw.^8,9 Various methods of pretreating rice straw have been tried in efforts to enhance AD performance, including physical, chemical and biological pretreatment methods.^3,8,9 Steam explosion is one of the most effective techniques used to open up the lignocellulose complex, which involves high temperature heating combined with a rapid pressure drop that physically disrupts the lignocellulosic structures in the biomass fibers.^8,10 While many studies report steam explosion pretreatment of lignin-rich materials as a method for improving enzymatic hydrolysis and bioethanol production,¹⁰ there is little information on steam explosion as a pretreatment method of rice straw for biogas production. Chemical reactions take place during this pretreatment process due to the high temperatures. Polysaccharide compounds such as hemicellulose are hydrolyzed, leading to more rapid degradation rates during the AD process. Meanwhile, the structure of the rice straw could also be disrupted, and this would make it easier for the anaerobic bacteria to digest it. These advantages would lead to better performance of digesters with lower hydraulic retention times, thus reducing the energy required for fermenter homogenization. However, the effects of different steam explosion pretreatment conditions on the physical and chemical properties and biological methane potential are not currently clearly known; in particular, the relationship between the biogas potential and microbial diversity under different steam explosion conditions has not been investigated.

Therefore, to elucidate the mechanism of steam explosion pretreatment and its impact on the rice straw AD process, this study aimed to investigate the effects of different steam explosion pretreatment conditions on the rice straw structure, its chemical components and the performance of AD under batch conditions. Microbial community structures during the AD process were also determined using the denaturing gradient gel electrophoresis (DGGE) technique to reveal the correlations among microbial communities, fermentation performance and the pretreatment method of steam explosion in this AD system.

2 Materials and methods

2.1 Rice straw and inoculum

Air-dried rice straw with a moisture content of 8–10% was collected from a rice field in Siyang (Jiangsu province) and ground to a particle size of approximately 30 mm. Inoculums used in this study were from a biogas slurry, which was collected from a 300 m³ biogas plant in the campus of Nanjing Tech University (Nanjing, Jiangsu province, China) and maintained at 4 °C until use, and raw materials used in this biogas plant was pure rice straw. Table 1 shows selected physicochemical properties of the rice straw and inoculums.

Table 1 Physicochemical characteristics of rice straw and inoculum^a

Substrate	TS (%)	VS (%)	C/N	Lignin (% TS)	Cellulose (% TS)	Hemicellulose (% TS)	Furfuraldehyde	5-Hydroxymethyl-2-furfural
a The results are shown as means ± deviation (n = 3). TS, total solid; VS, volatile solid.
Rice straw	89.49 ± 0.001	77.70 ± 0.015	70.88	22.14 ± 0.053	43.06 ± 0.020	21.17 ± 0.013	—	—
Inoculum	0.39 ± 0.001	0.21 ± 0.028	25	—	—	—	—	—

---

2.2 Steam explosion pretreatment

Steam explosion was chosen for the pretreatment of rice straw in this study because of its potential to disrupt the crystallinity of the cellulose, delignify the lignocellulosic materials and easily hydrolyze the hemicelluloses. However, some evidence has indicated that inhibitors such as furfural and hydroxymethylfurfural may also be formed during this process depending on the intensity of the pretreatment.^10,11 Therefore, the main factors (temperature and residence time) that affect steam pretreatment outcomes were investigated in this study. Rice straw was steam exploded at temperatures ranging from 200 to 220 °C at intervals of 10 °C. At each temperature, samples were kept in the reactor for 60, 90, 120 and 240 s, resulting in a total of 12 temperature/time combinations. After the retention time had elapsed, the pressure was reduced abruptly to atmospheric pressure.

The pretreated rice straw was then transferred into a flash tank and dried to constant weight at a temperature of 40 °C, then stored at 4 °C for later analysis and AD.

2.3 AD of the rice straw

Based on analysis of the degree of crystallinity, structural components and inhibitor content, various samples of pretreated rice straw (200 °C/60 s, 200 °C/90 s, 200 °C/120 s, 210 °C/60 s and 220 °C/90 s) were selected for the AD tests. Batch experiments were carried out in 500 mL anaerobic flasks with a working volume of 400 mL and final total solids content of 2%. The fermentation slurry was a mixture of 8.18 g dry rice straw, 140 mL inoculum and 260 mL tap water. The carbon-to-nitrogen (C : N) ratio was adjusted to 25 : 1 by adding (NH₄)₂SO₄. A control treatment using untreated rice straw was also set up. All experiments were conducted in triplicate. Oxygen in the anaerobic flasks was removed by sparging with nitrogen gas for 5 min, and then the flasks were capped with rubber stoppers and placed in an incubator (SPX-25013-D, Ningbo, China) at 38 ± 1 °C for 21 days. Gas sampling bags were used to collect the gas produced in each reactor. During the AD process, each anaerobic flask was shaken manually six times per day for 5 min. During incubation, samples for chemical analysis were withdrawn at different digestion times. Straw samples were washed three times in sterile sodium phosphate buffer (100 mM PBS buffer, pH 7.0). The slurry was carefully pipetted up and pelleted by centrifugation at 13 000 × g for 5 min. All processed samples were stored at −20 °C prior to DNA extraction.

2.4 Analytical methods

Substrates, inoculums and digestates were analyzed for all experiments. The pH, total solids (TS), total Kjeldahl nitrogen and volatile solids (VS) were measured according to standard methods.¹² Cellulose, hemicellulose and lignin contents were determined according to the methods of Van Soest and Wine.¹³ The crystal structure of the rice straw was examined by X-ray powder diffraction (XRD) (Rigaku Rotaflex D/max, Japan). To analyses the contents of the inhibitors furfural and hydroxymethylfurfural, the rice straw was mixed with Milli-Q water using a solid-to-water ratio of 1 : 50 (w/v, dry weight basis), and the mixture was shaken in a reciprocal shaker at 200 rpm for 60 min at 4 °C. The resulting suspension was centrifuged at 4000 × g for 20 min, and the supernatant was filtered through a 0.45 μm sterile membrane (GN-6 Metrice, Gelman Sciences, Ann Arbor, MI, USA). The concentrations of the furfural and hydroxymethylfurfural were determined using an Agilent 1260 liquid chromatograph equipped with an ultraviolet detector and a Bio-RaD HPX-87H (300 × 7.8 mm) column. The mobile phase was 5 mmol L⁻¹ H₂SO₄ with a flow rate of 0.8 mL min⁻¹. The sample injection volume was 20 μL, and the temperature of the column was maintained at 50 °C. Daily biogas production was recorded using the water displacement method. The corresponding cumulative biogas volume was also calculated. Methane content in the biogas produced was analyzed by gas chromatography (Modle 7890A, Agilent, USA). The degradation rate of the rice straw after fermentation was calculated by the weight decrease divided by the initial weight before digestion.

2.5 Denaturing gradient gel electrophoresis analysis

2.5.1 Genomic DNA extraction, PCR amplification of 16S rRNA gene fragments. To analyze the structure of the microbial communities in the rice straw anaerobic digester, microbial DNA was extracted using DNA Ultra CleanTM Soil DNA Isolation Kit (Mo BIO Laboratories, Inc. Carlsbad, CA, USA). A nested Polymerase Chain Reaction (PCR) technique was used to increase the sensitivity of PCR. The bacterial and archaeal 16S rRNA gene were amplified by the primers (357F-GC, 517R) and (PARCH340F-GC, PARCH519R),^14,15 respectively, and the PCR products, about 200-bp fragments of 16S rDNA genes, were used for denaturing gradient gel electrophoresis (DGGE) analysis.

2.5.2 Denaturing gradient gel electrophoresis analysis. DGGE was performed using the CBS Scientific DGGE-2401 system (C.B.S. Scientific Company, Inc., Del Mar, CA, USA). The products were separated on polyacrylamide gels (8%, w/v) using a denaturing gradient from 30% to 60% (for bacteria and archaea) where 100% denaturant contained 7 mol L⁻¹ urea and 40% (v/v) formamide. Electrophoresis was run for 12 h at 100 V in 1× TAE (40 mmol L⁻¹ Tris base, 40 mmol L⁻¹ glacial acid acetic, 1 mmol L⁻¹ EDTA) at a constant temperature of 60 °C. After electrophoresis, the gels were then stained for 20 min in 150 mL 1× TAE buffer containing ethidium bromide. Twenty four dominant DGGE bands, including bacterial and archaea, were excised from the gel using a sterile scalpel, and the total DNAs were eluted in 50 μL of sterilized ultrapure water at 4 °C overnight. The supernatants (1 μL) were used as the template in the reamplification PCR reaction mixture with the primer sets without GC clamps. The PCR products were purified with a PCR purification kit (Tiangen, Beijing, China) and sent to a commercial sequencing company (Genescript, Nanjing, China). The 16S rRNA sequences were compared with those in the database of the National Center for Biotechnology Information (NCBI) using the BLASTn program,¹⁶ a phylogenetic tree was constructed using the neighbor-joining method by MEGA software (Version 6.1 Beta), and 1000 bootstraps were used to assign confidence levels to the nodes in the tree.¹⁷

2.6 Statistical analysis

The data presented in the Results section are the mean and standard deviation of triplicate samples collected and analyzed. All of the figures presented include the standard deviations of the data and were drawn with Origin 8.5 software. Both cluster analyses and the sequencing analysis were performed in the experiment. For the cluster analyses, the results of DGGE profiles of bacteria and archaea were analyzed by Gelcompar II (Applied Maths, Sint-Martens-Latem, Belgium, USA) software.

The sequences obtained from the experiment were compared with the 16S rRNA gene sequences available in the GenBank database (http://blast.ncbi.nlm.nih.gov/) by BLAST search method, and the closely-related sequences downloaded for later analysis.

3 Results and discussion

3.1 Effect of steam explosion pretreatment on the physical properties of rice straw

XRD measurements of the straw under different steam explosion conditions are presented in Fig. 1. As can be seen, the diffraction trough around 2θ = 18° reflects an amorphous region, and this position became clear after steam explosion. The diffraction peak around 2θ = 21.93° is the maximum peak of cellulose. Compared with the control XRD in Fig. 1, the diffraction peak at 2θ = 21.9° became more acute with increased temperature and retention time. This phenomenon suggests that the crystallinity and components of the straw changed after steam explosion.

Fig. 1 X-ray diffraction patterns of rice straw under steam exploded by different temperature and retained time.

The crystallinity of the rice straw was analyzed by XRD, and the results are shown in Table 2. The crystallinity of steam exploded rice straw was higher than that of untreated rice straw (control). The average crystallinities of treated rice straw were 39%, 49% and 51% at temperatures of 200 °C, 210 °C and 220 °C, respectively, which were 1%, 11% and 14% higher than the control. In the same treatment time, the crystallinity of treated rice straw gradually increased with increasing temperature. Meanwhile, the crystallinity of treated rice straw also increased with extension of the treatment time at the same temperature. This phenomenon suggests that the cellulose structure of the straw was rearranged, and part of the amorphous zone was transformed to a crystalline region at high temperature and pressure. Similar results were observed by Zhao et al.¹⁸

Table 2 Crystallinity of rice straw under different steam exploded condition^a

Dwell time/s	Crystallinity (%)
	200 °C	210 °C	220 °C
a The crystallinity of the control rice straw is 38%.
60	0.31 ± 0.01	0.41 ± 0.02	0.47 ± 0.03
90	0.34 ± 0.04	0.48 ± 0.04	0.47 ± 0.02
120	0.45 ± 0.01	0.49 ± 0.01	0.52 ± 0.01
240	0.46 ± 0.02	0.57 ± 0.03	0.58 ± 0.04

---

3.2 Variation in straw components under different steam explosion condition

The major constituents of the rice straw are cellulose, hemicellulose and lignin polymer. Of these, cellulose is a polymer material made of glucose moieties connected by β-1,4-glycosidic bonds that is interlaced with the hemicellulose and lignin.⁹ Table 3 shows the variations in cellulose, hemicellulose and lignin polymers under different steam explosion conditions. The proportion of cellulose substantially increased under conditions of 200 °C/90 s, 210 °C/120 s, 200 °C/240 s and 210 °C/240 s, with the highest content of cellulose occurring at 210 °C/240 s, which was 1.12 times that of the control. The changes in cellulose content may be due to the following two factors: firstly, cellulose, hemicellulose and lignin in the straw were degraded inevitably under high temperature and pressure conditions; secondly, the relative content of cellulose increased at high temperature and pressure because hemicellulose and lignin were degraded more than cellulose under these conditions. Compared to cellulose, the content of hemicellulose was generally reduced in the treated rice straw. Over the same treatment time, the content of hemicellulose in the treated rice straw gradually decreased as the steam explosion temperature increased, as it did with increased retention time at a given steam explosion temperature. The proportion of hemicellulose reached 67.89% under the conditions of 220 °C/240 s. Variations in the lignin content displayed the same trends as the hemicellulose content.

Table 3 Componential changes of rice straw before and after steam explosion pretreatment

Different treatments	Cellulose (%)	Hemicellulose (%)	Lignin (%)
200 °C, 60 s	41.42 ± 0.11	19.24 ± 0.12	20.57 ± 0.11
200 °C, 90 s	43.89 ± 0.23	20.62 ± 0.11	21.41 ± 0.12
200 °C, 120 s	42.25 ± 0.09	23.12 ± 0.06	21.13 ± 0.07
200 °C, 240 s	47.49 ± 0.06	15.79 ± 0.05	19.68 ± 0.11
210 °C, 60 s	42.41 ± 0.11	19.46 ± 0.11	21.34 ± 0.08
210 °C, 90 s	40.59 ± 0.12	18.26 ± 0.06	21.31 ± 0.10
210 °C, 120 s	46.74 ± 0.05	17.58 ± 0.05	21.18 ± 0.13
210 °C, 240 s	48.19 ± 0.06	13.27 ± 0.14	18.73 ± 0.07
220 °C, 60 s	40.21 ± 0.02	17.42 ± 0.05	21.52 ± 0.09
220 °C, 90 s	40.57 ± 0.10	13.91 ± 0.12	22.15 ± 0.10
220 °C, 120 s	40.86 ± 0.12	10.47 ± 0.07	22.85 ± 0.11
220 °C, 240 s	38.12 ± 0.06	6.78 ± 0.09	23.73 ± 0.06
Control	43.05 ± 0.04	21.16 ± 0.08	22.16 ± 0.12

---

3.3 Effects of different steam explosion pretreatments on biogas production during the AD process

The cumulative biogas production from the pretreated rice straw system during AD process, as shown in Fig. 2 was higher than that of the untreated rice straw system. The highest cumulative biogas production reached 2689 mL for the 200 °C/120 s pretreated system, which was 74.80% higher than the control system. The biogas production rate of the pretreated rice straw was obviously higher than that of untreated rice straw, and the highest rate of 328.7 mL g⁻¹ TS was reached after pretreatment at 200 °C/120 s, which was 51% higher than for the untreated rice straw system. Cumulative biogas production in the 200 °C/90 s and 200 °C/60 s pretreated systems was also higher by 49.58% and 45.66%, respectively, than in the control system. The average methane contents in the biogas were also higher in the pretreated rice straw systems: 52%, 57%, 60%, 54%, 52% and 50% in the 200 °C/60 s, 200 °C/90 s, 200 °C/120 s, 210 °C/60 s, 220 °C/90 s and control systems, respectively. Meanwhile, the starting time of the AD was also shortened in the pretreated systems. The pretreated systems began to produce biogas during the first day of operation, whereas biogas production began in the untreated system 4 days later. Furfural, hydroxymethylfurfural and soluble phenolic compounds that are inhibitory to microbial growth were generated in the process at higher pretreatment severities (high temperature or long time).^19,20 In this study, the inhibitors (furfuraldehyde and 5-hydroxymethyl-2-furfural) were increased with the increase of temperature and time (data not shown), so the optimal biogas production not obtained in the highest temperature and longest time.

Fig. 2 Cumulative biogas production changes along with fermentation period.

3.4 Chemical composition and degradation rate of the rice straw after the AD process

After AD, cellulose, hemicellulose and other substances in the rice straw were reduced by microbial degradation and utilization. Fig. 3 shows the variations in cellulose, hemicellulose and lignin polymer contents in different steam explosion systems before and after AD. Among these, the highest cellulose and hemicellulose degradation rates were found in the 200 °C/120 s pretreated system, in which the cellulose and hemicellulose degradation rates reached 53.46% and 49.54%, which were 13.72% and 16.79%, respectively, higher than the control system. Since rice straw lignin cannot be directly used by microbial organisms, the proportion of lignin in the digested sample had increased. Table 4 shows that the rice straw degradation rate in the pretreated system after AD was higher than that in the untreated system. At a temperature of 200 °C, the rice straw degradation rate increased with increments in the duration of high pressure. The highest rice straw degradation rate reached 66.53% for the 200 °C/120 s system, which was 11.12% higher than that in the control system. These results suggest that the degradation rate of rice straw after AD increased with increases in the steam explosion temperature and time.

Fig. 3 Changes of chemical composition of rice straw before and after fermentation (1-before anaerobic fermentation (200 °C, 60 s); 2-after anaerobic fermentation (200 °C, 60 s); 3-before anaerobic fermentation (200 °C, 90 s); 4-after anaerobic fermentation (200 °C, 90 s); 5-before anaerobic fermentation (200 °C, 120 s); 6-after anaerobic fermentation (200 °C, 120 s); 7-before anaerobic fermentation (210 °C, 60 s); 8-after anaerobic fermentation (210 °C, 60 s); 9-before anaerobic fermentation (220 °C, 90 s); 10-after anaerobic fermentation (220 °C, 90 s); 11-before anaerobic fermentation (control); 12-after anaerobic fermentation (control)).

Table 4 Degradation rate of rice straw after AD process^a

	TS = 2%
	Dry weight before AD (g)	Dry weight after AD (g)	Biodegradation rate (%)
a Rice straw was first washed three times using the distilled water, then dried to the constant weight under 105 °C.
200 °C, 60 s	8.18	3.60	55.99
200 °C, 90 s	8.18	3.16	61.34
200 °C, 120 s	8.18	2.74	66.53
210 °C, 60 s	8.18	3.14	61.64
220 °C, 90 s	8.18	2.75	66.38
Control	8.18	3.65	55.41

---

3.5 Cluster analysis of the microbial community based on the 16S rDNA gene

The dynamic changes in the microbial community structure (bacterial and archaeal communities) during the AD process were detected by PCR-DGGE analysis. The results are shown in Fig. 4a and b. Each band on the DGGE profile represents a specific species in the microbial community while the intensity of a band indicates the relative abundance of the corresponding microbial species. The bacterial community in all sampled stages mainly consisted of the phylotypes Firmicutes, Spirochaetes, Planctomycetes, Bacteroidetes and Proteobacteria (Fig. 4a and 5). Nearly half of the bands resolved in the DGGE belonged to Bacteroidetes, while bands 6, 10, 12, 15 and 16 belonged to Firmicutes, bands 7, 8 and 17 to Spirochaetes, band 3 to Planctomycetes and band 14 to Proteobacteria. The archaeal community in all sampled stages mainly consisted of the phylotypes Thermoplasmata, Methanomicrobia and Crenarchaeota (Fig. 4b and 6). Most of the bands excised from the archaeal gel were affiliated to Methanomicrobia while bands 5 and 7 belonged to Crenarchaeota and band 10 to Thermoplasmata. Generally, the species of bacteria and archaea in the pretreated and untreated systems were similar; however, the distribution of specific species varied among samples. Similarities among samples collected at different periods were analyzed by Pearson correlation as evidenced in Fig. 4a and b. The composition and distribution of bacteria and archaea were regrouped by Pearson correlation analysis. Analysis revealed that there were some differences in both bacterial and archaeal community structures among the samples collected from different periods of the AD process in the pretreated and untreated systems.

Fig. 4 DGGE cluster analysis of PCR-amplified 16S rDNA fragments from the bacteria (a) and archaea (b) community during AD process. (A, B, C, D sample was the set-up period, early stage, medium term and final period in the pretreated system; E, F, G, H sample was the set-up period, early stage, medium term and final period in the untreated system).

Fig. 5 Neighbor-joining tree presenting the bacterial phylogenetic affiliation to the DGGE band sequences.

Fig. 6 Neighbor-joining tree presenting the archaeal phylogenetic affiliation to the DGGE band sequences.

3.6 Sequencing analysis of the microbial community by DGGE

To gain more information about the dominant microbial community in the AD process, the major bands of bacteria and archaea in the PCR-DGGE profile were sequenced and identified via high similarities to those in the GenBank database, and the results are shown in Fig. 5 and 6. At the beginning of the batch process (A, B and E, F, Fig. 4a), the bacterial community consisted of 1 Planctomycetes (band 3), 3 Spirochaetes (bands 7, 8 and 17), 4 Bacteroidetes (bands 5, 9, 11 and 13) and 4 Firmicutes (bands 10, 12, 15 and 16) species. The bacterial populations were clearly different in pretreated and untreated groups during the set-up period and early stage. Phylotypes that had high similarity to uncultured Bacteroidetes (bands 11 and 13), uncultured Spirochaetes (band 7) and Sporosarcina psychrophila (band 16) were only present in the untreated groups while Clostridium sp. (band 15) was only found in the pretreated groups.

It has been reported that some genera in Clostridium sp. are responsible for hydrolyzing cellulolytic substrates, as they produce several putative enzymes that hydrolyze the β-1,3 and β-1,4-glycosidic bonds of mixed-linkage polysaccharides, but also glucans or lichenin.²¹ Clostridium sp. are capable of hydrolyzing cellulose and various carbohydrates to acetate, butyrate and hydrogen²² and their prevalence in stable H₂-producing systems has already been documented.²³ In contrast, uncultured Bacteroidetes (bands 11 and 13), uncultured Spirochaetes (band 7) and Sporosarcina psychrophila (band 16) are mainly present in sludge-derived digesters.²¹ Hence, it may be speculated that pretreatment enhances the hydrolysis of cellulolytic substances while unpretreated systems perform more or less like an ordinary digester.

A remarkable change in the structure and composition of the bacterial community occurred at the midpoint of this experiment, as evidenced by the low Pearson correlation index among the DGGE banding patterns (Fig. 4a, lane C and G). Changes in microbial diversity between two different periods of a batch study are common; however, the dynamics of different treatments at the same stage should be attributable to the pretreatment. Compared with untreated lane G, pretreatment of the substrates (lane C) increased the diversities of the bacterial community. In particular, uncultured Bacteroidetes (band 5 and 12), uncultured bacterium (band 6), uncultured Spirochaetes (band 7), uncultured Firmicutes (band 10) and Clostridium sp. (band 15) appeared only after pretreatment, whereas uncultured Bacteroidetes (bands 11) were present in the untreated group. The presence of more diverse microbes in samples with pretreatment should be due to an improvement in the biodegradability of the substrate caused by steam explosion. A similar observation indicated that steam explosion could increase the area available to the degrading microbes and also partially hydrolyze some components, particularly hemicellulose.²⁴ In particular, the appearance of uncultured Firmicutes (band 10) and Clostridium sp. (band 15) in pretreated samples confirmed the superiority of steam explosion in increasing the biodegradability of cellulolytic substrates.²¹

Similar to the mid-stage, the bacterial community in the samples treated with steam explosion was more diverse than in the untreated samples during the final stage. Bands of typical microbes in anaerobic digesters such as uncultured Spirochaetes (band 7) and uncultured Bacteroidetes (bands 11) disappeared, whereas the putative cellulolytic bacteria Pseudomonas sp. (band 14) were present in the group without pretreatment, indicating a delayed process without pretreatment.

As evidenced by the previous description, at each stage of this experiment diversities in the bacterial communities of the groups exposed to steam explosion were different from those of the untreated group, although they shared some bands. During the mid and final stages, pretreated groups displayed communities in the digester that were more diverse than those in the untreated reactor. The major difference was in the cellulolytic bacteria. After treatment with steam explosion, the cellulolytic bacteria were present earlier than in the untreated group. Furthermore, the main species of cellulolytic bacteria in the pretreatment groups were Clostridium sp. while they were Pseudomonas sp. in the untreated reactor.

Generally, species succession within the archaea community among the different stages of AD was not as obvious as that of bacteria. At the beginning of the batch process (A, B and E, F, Fig. 4b), the archaea community consisted of 4 Methanomicrobia (bands 1, 7, 9 and 10), 4 Crenarchaeota (bands 3, 4, 5 and 6) and 1 Thermoplasmata (band 8). Methanomicrobia was the major group for CH₄ production, and Methanospirillum was responsible for hydrogenotrophic methanogenesis²⁵ while Methanosarcinales is known to utilize both acetate and hydrogen.²⁶ The only difference between the pretreated groups and the untreated ones was the disappearance of bands 4 and 8 in lane A, which correspond to uncultured Crenarchaeota and uncultured Thermoplasmata. Crenarchaeota is reported to co-exist with Thermoplasmata and methanogens but their exact function during the generation of CH₄ is not clear.²⁷ Thermoplasmata is one of the methylotrophic methanogens, the presence of which in lane E should be due to the relatively lower abundances of other groups. Thermoplasmata is one of the methylotrophic methanogens, the presence of which in lane E should be due to the relatively lower abundances of other groups.²⁸ Similarly, in the mid-stage, bands 5 and 8 were not present in the groups treated with steam explosion whereas they were found in untreated lane G. An almost identical profile of the bands in lanes D and H was found in the final stage.

4 Conclusion

Steam explosion serves as a useful pretreatment method for promoting the generation of biogas during AD of rice straw as well as improving the biodegradability of cellulose and hemicellulose. Anaerobic biogas recovery of up to 328.7 mL g⁻¹ TS under the conditions of 200 °C/120 s in the pretreated system was achieved, corresponding to a 51% increase compared to the control. Meanwhile, the highest degradation rates of cellulose and hemicellulose in the system (200 °C/120 s) reached 53.46% and 49.54%, respectively, corresponding to 13.72% and 16.79% increases. The obvious differences in microbial profiles between the pretreated system and the control were another indication of the efficiency of the pretreatment method for rice straw. The results indicated that steam explosion is a viable pretreatment method for improving biogas production efficiency from rice straw. Further energy consumption reduction of steam explosion pretreatment and its application in the large-scale biogas plant were still needed in the future studies.

Acknowledgements

This work was supported by the National Basic Research Program of China (2013CB733500), the National Natural Science Foundation of China (21307058), the National Key Technology Support Program of China (2014BAC33B00), the Jiangsu Province Science Foundation for Youths (BK20130931), and the Key Science and Technology Project of Jiangsu Province (BE2016389).

References

1. Q. L. Wang, W. Li, X. Gao and S. J. Li, Bioresour. Technol., 2016, 201, 208–214.
2. S. J. Yang, H. P. He, S. L. Lu, D. Chen and J. X. Zhu, Atmos. Environ., 2008, 42, 1961–1969.
3. Z. Y. Yan, Z. L. Song, D. Li, Y. X. Yuan, X. F. Liu and T. Zheng, Bioresour. Technol., 2015, 177, 266–273.
4. M. Krishania, V. Kumar, V. K. Vijay and A. Malik, Fuel, 2013, 106, 1–9.
5. F. Theuretzbacher, J. Lizasoain, C. Lefever, M. K. Saylor, R. Enguidanos, N. Weran, A. Gronauer and A. Bauer, Bioresour. Technol., 2015, 179, 299–305.
6. S. F. Fu, F. Wang, X. Z. Yuan, Z. M. Yang, S. J. Luo, C. S. Wang and R. B. Guo, Bioresour. Technol., 2015, 175, 203–208.
7. J. Jimenez, Y. Guardia-Puebla, M. E. Cisneros-Ortiz, J. M. Morgan-Sagastume, G. Guerra and A. Noyol, Chem. Eng. J., 2015, 259, 703–714.
8. V. Vivekanand, P. Ryden, S. J. Horn, H. S. Tapp, N. Wellner, V. G. H. Eijsink and K. W. Waldron, Bioresour. Technol., 2012, 123, 608–615.
9. J. Singh, M. Suhag and A. Dhaka, Carbohydr. Polym., 2015, 117, 624–631.
10. M. M. Estevez, R. Linjordet and J. Morken, Bioresour. Technol., 2012, 104, 749–756.
11. A. Singh and N. R. Bishnoi, Ind. Crops Prod., 2013, 41, 221–226.
12. APHA, AWWA and WEF, Effect of steam explosion pretreatment on the anaerobic digestion of rice straw, American Public Health Association, Washington DC, 21st edn, 2005.
13. P. J. Van Soest and R. H. Wine, J. - Assoc. Off. Anal. Chem., 1967, 50, 50–55.
14. S. Wang, N. Nomura, T. Nakajima and H. Uchiyama, J. Biosci. Bioeng., 2012, 113, 624–630.
15. H. K. Sang and K. Nakasaki, J. Ind. Eng. Chem., 2015, 2, 443–450.
16. M. Khemkhao, S. Techkarnjanaruk and C. Phalakornkule, Bioresour. Technol., 2015, 177, 17–27.
17. X. Y. Yong, Y. Q. Cui, L. H. Chen, W. Ran, Q. R. Shen and X. M. Yang, Appl. Microbiol. Biotechnol., 2011, 92, 717–725.
18. H. B. Zhao, J. H. Kwak, Z. C. Zhang, H. M. Brown, B. W. Arey and J. E. Holladay, Carbohydr. Polym., 2007, 68, 235–241.
19. H. Jorgensen, J. B. Kristensen and C. Felby, Biofuels, Bioprod. Biorefin., 2007, 1, 119–134.
20. R. Martin-Sampedro, M. E. Eugenio, J. C. Garcia, F. Lopez, J. C. Villar and M. J. Diaz, Biomass Bioenergy, 2012, 42, 97–106.
21. S. Weiß, W. Somitsch, I. Klymiuk, S. Trajanoski and G. M. Guebitz, Bioresour. Technol., 2016, 207, 244–251.
22. I. Valdez-Vazquez and H. M. Poggi-Varaldo, Int. J. Hydrogen Energy, 2009, 34, 3639–3646.
23. G. Cai, B. Jin, P. Monis and C. Saint, Biotechnol. Bioeng., 2013, 110, 338–342.
24. F. Theuretzbacher, J. Lizasoain, C. Lefever, M. K. Saylor, R. Enguidanos, N. Weran, A. Gronauer and A. Bauer, Bioresour. Technol., 2015, 179, 299–305.
25. S. N. Parshina, A. V. Ermakova, M. Bomberg and E. N. Detkova, Int. J. Syst. Evol. Microbiol., 2014, 64, 180–186.
26. L. Sun, P. B. Pope, V. G. Eijsink and A. Schnürer, Microb. Biotechnol., 2015, 8, 815–827.
27. R. Chouari, S. Guermazi and A. Sghir, World J. Microbiol. Biotechnol., 2015, 31, 805–812.
28. M. Poulsen, C. Schwab, B. B. Jensen, R. M. Engberg, A. Spang, N. Canibe, O. Højberg, G. Milinovich, L. Fragner, C. Schleper, W. Weckwerth, P. Lund, A. Schramm and T. Urich, Nat. Commun., 2013, 4, 66–78.

---