Responses of biomass briquetting and pelleting to water-involved pretreatments and subsequent enzymatic hydrolysis

Bioresource Technology 151 (2014) 54–62

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Responses of biomass briquetting and pelleting to water-involved pretreatments and subsequent enzymatic hydrolysis Yang Li a,b, Xiaotong Li b, Fei Shen a,b,⇑, Zhanghong Wang b, Gang Yang b, Lili Lin b, Yanzong Zhang b, Yongmei Zeng c, Shihuai Deng a,b,⇑ a b c

Institute of Ecological and Environmental Sciences, Sichuan Agricultural University-Chengdu Campus, Chengdu, Sichuan 611130, PR China Rural Environment Protection Engineering & Technology Center of Sichuan Province, Sichuan Agricultural University-Chengdu Campus, Chengdu, Sichuan 611130, PR China Department of Information Consultation and Integration, Library of Sichuan Agricultural University-Chengdu Campus, Chengdu, Sichuan 611130, PR China

h i g h l i g h t s  Water-involved pretreatments and hydrolysis of briquettes & pellets were investigated.  Briquetting and pelletting improved diluted-NaOH pretreatment and enzymatic hydrolysis.  Pelletting made hydrothermal pretreatment easier resulting in better sugar yield.  Briquetting increased difficulty of hydrothermal pretreatment and sugar conversion.  Auto-swelling may facilitate water-involved pretreatments of the densified biomass.

a r t i c l e

i n f o

Article history: Received 7 August 2013 Received in revised form 8 October 2013 Accepted 14 October 2013 Available online 22 October 2013 Keywords: Biomass densification Auto-swelling Diluted-NaOH pretreatment Hydrothermal pretreatment Enzymatic hydrolysis

a b s t r a c t Although lignocellulosic biomass has been extensively regarded as the most important resource for bioethanol, the wide application was seriously restricted by the high transportation cost of biomass. Currently, biomass densification is regarded as an acceptable solution to this issue. Herein, briquettes, pellets and their corresponding undensified biomass were pretreated by diluted-NaOH and hydrothermal method to investigate the responses of biomass densification to these typical water-involved pretreatments and subsequent enzymatic hydrolysis. The densified biomass auto-swelling was initially investigated before pretreatment. Results indicated pellets could be totally auto-swollen in an hour, while it took about 24 h for briquettes. When diluted-NaOH pretreatment was performed, biomass briquetting and pelleting improved sugar conversion rate by 20.1% and 5.5% comparing with their corresponding undensified biomass. Pelleting improved sugar conversion rate by 7.0% after hydrothermal pretreatment comparing with the undensified biomass. However, briquetting disturbed hydrothermal pretreatment resulting in the decrease of sugar conversion rate by 15.0%. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Besides the serious energy security concerns, the increasing issues of CO2 emissions and pollutions from fossil energy consumption has strengthened the interests in the alternative, and non-petroleum-based sources energy (Alvira et al., 2010). Lignocellulose as the most abundant renewable biomass has a yearly supply of approximately 200 billion metric tons worldwide (Percival Zhang et al., 2006; Ragauskas et al., 2006). In this context, lignocellulosic biomass has been regarded as one of the most promising renewable sources for refining bioenergy or biofuel. However, the ⇑ Corresponding authors at: 211 Huimin Road, Wenjiang District, Chengdu, Sichuan 611130, PR China. Tel./fax: +86 28 86291390. E-mail addresses: fi[email protected] (F. Shen), [email protected] (S. Deng). 0960-8524/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2013.10.044

low density of lignocellulosic feedstocks and the high costs of the biomass handling, transportation, and storage seriously limit the wide application for biofuel production. One of ways to reduce the handling, transportation and storage costs is densification of the biomass raw materials into briquettes or pellets with the high densities (Kaliyan and Morey, 2010). The biomass densification technology was initially developed for the solid fuel production, and the densification systems for biomass have been adapted from other highly efficient processing industries like feed, food and pharmacy including pellet mill, briquette press, screw extruder, tabletizer, and agglomerator. Among these, the pellet mill and briquette press are the most common systems used for solid fuel production (Tumuluru et al., 2011). The typical products from these two densification systems are pellets and briquettes. Generally, the pelleting can convert finely ground lignocellulosic biomass into dense, free-flowing durable pellets,

Y. Li et al. / Bioresource Technology 151 (2014) 54–62

which typically presents with the uniform cylindrical shapes with 6–25 mm in diameter and the length varies depending on diameter (length <5  diameter) (García-Maraver et al., 2011). The diameter of 6 and 8 mm pellets are the most common products with the unit density of 1125–1190 kg m3 (bulk density of 650–750 kg m3) (Finney et al., 2009; Tumuluru et al., 2010a,b). Unlike the pelleting systems, briquetting can handle the biomass with larger-sized particles and wider moisture contents. The briquettes could be produced from various briquetting systems, such as piston press, tabletizer, cuber, roller press, and the produced briquettes have typical shapes of cylinder and cuboid (Tumuluru et al., 2011). The sizes and shapes of briquettes are depended on the employed briquetting system. Generally, the cylindrical briquettes are 40–80 mm in diameter and 40–150 mm in length with unit density in the range of 800–1000 kg m3 (Song et al., 2010). The cubic briquettes are from 12.7  12.7 mm to 38.1  38.1 mm in cross section, and from 25.4 to 101.6 mm in length with the unit density >1000 kgm3 (Kaliyan and Vance Morey, 2009; Tumuluru et al., 2010b). Although different products could be obtained from different densification systems, the biomass densification is unexceptionally achieved by forcing the biomass particles together to create inter-particle bonding, which makes well-defined shapes and sizes. It is also well known that the mechanism of binding during densification can be from the formation of solid bridges. These solid bridges are developed by chemical reactions and sintering solidification, hardening of the melted substances, or crystallization of the dissolved materials (Kaliyan and Morey, 2010; Sastry and Fuerstenau, 1973). Therefore, the densified biomass will undoubtedly show different behaviors in their application comparing with the undensified biomass (Alevanau et al., 2011; Erlich et al., 2006). Recently, refining bioethanol from lignocellulosic biomass has been highlighted in many countries. For example, lignocellulosic ethanol has been proposed as one of the priority projects in ‘‘the Twelfth Five-year-plan’’ for energy development in China. US government also targeted 16 billion gallons of cellulosic ethanol by 2020. These means much more lignocellulosic biomass should be collected and transported to the potential biomass-to-ethanol plants. As a key issue, how to reduce the costs of collection and transportation has to be considered seriously. As mentioned above, the transportation costs could be apparently reduced by the biomass densification technology. However, as the most common products from densification, the possibility of biochemical conversion of briquettes and pellets has not been investigated thoroughly before the densification can be employed in the bioethanol production system. Moreover, the actions of high temperature and pressure are involved in the biomass densification, which may potentially affect the key downstream processes such as pretreatment and subsequent enzymatic hydrolysis (Theerarattananoon et al., 2012). Based on recent reference, it could be observed that pellets or briquettes can be auto-swollen and tend to be fragment, resulted in the loss of structural integrity and shape in high humidity conditions (Tooyserkani et al., 2012). Although this is one of the main challenges for the transports and storage of the commercial densified biomass, it can potentially decrease the energy consumption for re-crushing the densified biomass for pretreatment and hydrolysis because it can soak water and auto-swell resulting in fragment. Thus, two water-involved and cost-effective methods including hydrothermal pretreatment and alkalinity pretreatment were employed, respectively, to evaluate their responses to briquetting and pelleting in comparison of the densified biomass. Meanwhile, the subsequent enzymatic hydrolysis after the pretreatment also was performed to assess the fermentable sugar conversion from the densified and undensified biomass.

55

2. Methods 2.1. Raw materials The employed briquettes and pellets, and their corresponding raw materials in this work were collected from Heilongjiang Province in the North East of China. The raw materials for briquettes mainly consisted of corn leaves and corn stalks. The raw materials were pressed with a ring die briquette machine (cuber) (9SYH1200, WanGuo Bioenergy Technology Co., Ltd., Xuzhou, Jiangsu Province, China) with the working pressure of approximately 2600 kg cm2 and capability of approximately 1000–1600 kg h1. The diameter of the die ring was 960 mm with die size of 30  30 mm in cross section. The raw materials moisture and size should be controlled in the range of 15–30% (wet basis) and less than 50 mm before they were briquetted. The produced briquettes presented as cubic shape of approximately 30  30 mm in cross section, and 50–150 mm in length with the unit density of 1484 kg m3. The corn stalks were main raw materials for the pellets, which were produced by a ring die pelletizer with double rollers (FTHBCX350, Futen New-energy Technology Co., Ltd, Sanmenxia, Henan Province, China) with the capability of approximately 700–1200 kg h1. The diameter of the die ring was 350 mm with die size of 10.0 mm. The raw materials should be grounded with size of 10–30 mm and the moisture was required in the range of 15–30% for pelleting. The produced pellets presented to be a cylindrical shape with 10.0 mm in diameter and 15–20 mm in length, and the unit density was determined as 1176 kg m3. The raw materials for the briquettes and pellets were also collected for pretreatment and hydrolysis as the comparison from the grinding stream of the corresponding densification system. The determined moisture content of the briquettes and pellets and their corresponding raw materials were in the range of 8.4–10.27% (dry basis). 2.2. Pretreatment The hydrothermal pretreatment was performed in a high pressure reactor with working volume of 1.5 L and heating rate of 10 °C min1. 30.0 g densified/undensified biomass substrates were immersed in 450 mL distilled water for 24 h before pretreatment. The mixed slurry was input in the reactor undergoing approximately 20 min for increasing the pretreatment temperature to 200 °C. The pretreatment were maintained at 200 °C for 15 min with agitation rate of 150 rev.min1. Afterwards, the reactor was rapidly cooled down by inside stainless steel coil with tap water. No additional chemical catalysts were supplemented for the hydrothermal pretreatment. Similarly, 30.0 g densified/undensified biomass substrates were immersed in 450 mL 0.25 mol L1 NaOH aqueous solution for 24 h in 2.0 L glass bottles before the alkalinity pretreatment. The bottles were afterwards screw-sealed and input into an autoclave for pretreatment at 121 °C for 30 min. All the pretreated slurry were vacuumed and washed with tap water for 8–10 times, and the residual solids were stored in 4 °C for analysis and hydrolysis. 2.3. Enzymatic hydrolysis A commercial cellulase from Trichodema reesei ATCC26924 (Sigma, Missouri, USA) supplemented with cellobiase (Sigma, Missouri, USA) was used for the enzymatic hydrolysis. The employed cellulase activity was 700 FPU g1 and the cellobiase activity of cellobiase was 250 IU g1. The pretreated substrates were enzymatically hydrolyzed in 0.05 molL1 acetate buffer solution (pH = 5.0) with the substrate consistency of 2% (dry basis). The enzyme loading

56

Y. Li et al. / Bioresource Technology 151 (2014) 54–62

of cellulase and cellobiase were 20 FPU (g glucan)1 and 40 IU(g glucan)1. Before hydrolysis, 0.1 mL tetracycline (250 mg L1) was supplemented in hydrolysis to inhibit the microorganisms that may potentially consume the produced sugar. All runs of hydrolysis were performed in duplicate in the 150 mL screw-flasks with working volume of 20 mL. The flasks were incubated in a 50 °C incubator with 200 rev.min1 for 72 h. During the hydrolysis process, 800 lL samples were taken periodically, and boiled for 10 min to deactivate the enzymes and centrifuged at 15,493g, 4 °C for 10 min. The supernatants were collected and stored in a 20 °C freezer for further analysis.

where, PXylan is the xylan content of the pretreated biomass (%); OXylan is the xylan content of the original biomass (%).

2.4. Analysis

3. Results and discussion

The auto-swelling degree of pellets or briquettes was estimated according to the soaked water in a specified duration. The detailed protocols for the determination were as follows. 20.0 g briquettes or pellets were input into 400.0 mL distilled water or 0.25 mol L1 NaOH aqueous solution for auto-swelling. The auto-swollen biomass was filtered with a 40-meshes sieve for 1.0 h so that the unabsorbed water or NaOH aqueous solution can be separated for collection as completely as possible. Afterwards, the volume of the unabsorbed water or NaOH aqueous solution was measured. The auto-swelling degree (mL g1) of densified biomass was determined using the ratio of the soaked water or NaOH aqueous solution volume and the initial biomass weight (dry basis). The process was totally same for determining the degree at different autoswelling time points. All the determined results of auto-swelling degree were calculated by 3 repetitions at each time point. The Klason insoluble lignin content in the unpretreated and pretreated biomass were analyzed using the Tappi-T-22 om-88 as described in reference (Bura et al., 2002). The hydrolysate from the Klason insoluble lignin analysis was retained and analyzed for soluble lignin by an UV spectrophotometer at 205 nm. The main carbohydrate including glucose and xylose, in the hydrolysate from the Klason insoluble lignin analysis, were determined by a highperformance liquid chromatography (HPLC) (Flexar, PerkinElmer, Inc., Waltham, MA, USA) with a refractive index detector (at 50 °C).The sugars were separated by a sugar column (SH1011, Shodex, New York, USA) using 0.05 mol L1 H2SO4 as the mobile phase with the flow rate of 0.8 mL min1. The injection volume for the HPLC analysis was 100 lL and the column temperature was maintained at 60 °C for the sugar separation. The lactose (0.5 g L1, Sigma) was used as an internal standard. The obtained glucose and xylose content were used for calculating glucan and xylan content in the substrates. The lignin content and carbohydrate content in the biomass were determined with 3 repetitions, and the obtained result was the means of these repetitions. According to the mass of the recovered solid after pretreatments, and the determined lignin, xylan and glucan content in the biomass, the solid recovery, lignin removal, xylan recovery, and glucan recovery were defined in the following equations.

3.1. Chemical composition of pellets and briquettes

Solid recovery ð%Þ ¼

Wp  100 Wo

ð1Þ

where, Wp is the weight of pretreated biomass (dry basis) (g); Wo is the weight of the original biomass used for pretreatment (dry basis) (g).

  W p PLignin  100 Lignin removal ð%Þ ¼ 1   W o OLignin

ð2Þ

where, PLignin is the lignin content of the pretreated biomass (%); OLignin is the lignin content of the original biomass (%).

Xylan recovery ð%Þ ¼

W p PXylan   100 W o OXylan

ð3Þ

Glucan recovery ð%Þ ¼

W p PGlucan   100 W o OGlucan

ð4Þ

where, PGlucan is the glucan content of the pretreated biomass (%); OGlucan is the glucan content of the original biomass (%). In addition, the content of the derived glucose in the hydrolysate by enzymatic hydrolysis was also determined by the HPLC with same method and conditions.

The chemical composition of biomass including cellulose, hemicellulose, lignin, protein, starch, and ash, can potentially affect the densification process. For example, the lignin in the biomass will be generally softened during the densification. The softened lignin could improve the binding of the biomass particles during the formation of pellets or briquettes (Tumuluru et al., 2011). Additionally, the moisture in the raw materials can form steam under high pressure, which hydrolyzes the hemicellulose and lignin into lower molecular carbohydrates, lignin products, sugar polymers, and other derivatives (Grover and Mishra, 1996). Consequently, the densification process will definitely change the chemical composition densified biomass and their subsequent conversion. The main composition of the briquettes, pellets and their corresponding undensified biomass were listed in Table 1. As mentioned above, the raw materials for briquettes and the pellets mainly included corn leaves and corn stalks. The main composition of the unbriquetted and unpelleted biomass were comparable based on the reported results of the corn leaves and stalks in reference (Su et al., 2006). As shown in Table 1, the decrease of xylan content (representing hemicellulose) and extractives could be observed after briquetting and pelleting. Similar results have been reported that the mannan (mainly representing hemicellulose of softwood) in pellets of Doulas fir was decreased obviously comparing with the unpelleted chips (Kumar et al., 2012). The hemicellulose was much more sensitive to thermo-shock than the other compositions such as cellulose and lignin, and the degradation of hemicellulose will potentially happen at high temperature and pressure during the process of briquetting and pelleting (Yang et al., 2007). Moreover, moisture in the biomass also could accelerate hemicellulose degradation due to steam formation at high pressure and temperature in densification process (Grover and Mishra, 1996). The decrease of acid-insoluble lignin content could be observed after the raw substrates were densified into briquettes and pellets. On the contrary, the acid-soluble lignin content was increased somewhat after densification because the ether bonds in lignin could be partially cleaved at high temperature produced from densification (Fang et al., 2012), and the intermediate lignin degradation compounds presented to be more soluble. The lignin content (the sum of acid-insoluble lignin and acid-soluble lignin) was decreased from 23.1% to 21.1% after briquetting, and no significant change can be observed in the pelleting process. These were probably attributed to the higher severity (temperature, pressure) in briquetting process comparing with the pelleting process, which consequently caused the potential lignin degradation more seriously. Moreover, the decrease of lignin also could be observed after densification of Doulas fir, wheat straw, corn stover, big bluestem and sorghum stalk (Kumar et al., 2012; Theerarattananoon et al., 2012). As another main composition of the lignocellulosic biomass, glucan content (representing cellulose) in the briquettes was a

57

Y. Li et al. / Bioresource Technology 151 (2014) 54–62 Table 1 The main composition of the briquettes, pellets and their undensified biomass. Composition (%) Glucan Xylan Acid-insoluble lignin Acid-soluble lignin Ethanol extractive Ash a

Unbriquetted biomass a

34.5 ± 0.1 21.4 ± 1.9 17.7 ± 0.1 5.4 ± 0.0 2.2 ± 0.4 5.1 ± 0.3

Briquettes

Unpelleted biomass

Pellets

37.2 ± 1.1 19.1 ± 1.8 13.2 ± 0.1 6.9 ± 0.1 1.8 ± 0.7 12.0 ± 1.1

35.2 ± 0.7 17.8 ± 0.3 17.1 ± 0.6 4.1 ± 0.8 6.5 ± 0.4 3.6 ± 0.1

33.9 ± 1.5 15.6 ± 0.8 15.0 ± 2.0 6.6 ± 0.4 1.7 ± 0.8 5.2 ± 0.0

The values in the table was listed in the form of ‘‘Mean ± Standard deviation’’.

Fig. 1. The auto-swelling behaviors of briquettes and pellets (s Pellets in NaOH solution; h Pellets in water; d Briquettes in NaOH solution; j Briquettes in water).

little higher than the unbriquetted biomass. Unlike briquettes, the content in pellets presented to be a little lower comparing with the unpelleted biomass. In addition, ash content was an important item in the standards of densification in many countries or areas. In this work, the ash content in the employed densified biomass was both higher than their corresponding undensified biomass. Especially, the ash in the briquettes was 12.0%, which was obviously higher than the unbriquetted biomass (5.1%). Furthermore, it was also obviously higher than the pellets (3.6%) and the unpelleted biomass (5.2%). Generally, the increase of ash content in the densified biomass may partially attribute to the losses of other composition in the biomass after densification. For example, xylan, lignin and extractives content showed more or less decrease after densification in this work (see Table 1). In addition, based on the communication with the supplier of briquettes, bentonite is generally supplemented as a binder (approximately 2.5%–6.0%) during the biomass briquetting with cuber system in Heilongjiang Province. Moreover, cubing operators often find it necessary to add a binder to increase cubic briquettes durability, and typical binders used are bentonite, hydrated lime, starch, lingo-sulfonates, agro colloids, and other commercial binders (Sokhansanj and Turhollow, 2004). However, the binders are seldom required in the pellet mill system (Tumuluru et al., 2011). This may be another reason for especially higher content in the employed briquettes in this work. 3.2. Comparison of the auto-swelling degree of pellets and briquettes The extremely compact bonding in the densified substrates with high density and hardness will greatly reduce their sensitivity

to the pretreatment, and they thereby should be crushed prior to pretreatment to increase their surface. The auto-swelling of the densified biomass, including the pellets and the briquettes, has ever been observed in the water-involved solutions. Therefore, the water-involved pretreatments (hydrothermal pretreatment and diluted-NaOH pretreatment) were employed in this work. In this part, the densified biomass (pellets and briquettes) were immersed in the water or the NaOH solution before the pretreatment so that it could be swollen well to weaken the potentially negative influences on the subsequent pretreatment. Their auto-swelling behaviors were presented in Fig. 1. Overall, it took about 24 h to achieve the maximum water or NaOH solution absorption for the briquettes, which could be regarded as the complete auto-swelling state. However, it took about 1.0 h to achieve the complete auto-swelling for the pellets in water or the NaOH solution, which was significantly shorter than that of the briquettes. As mentioned above, the compression severity for briquettes was generally higher than that of pellets, which could enhance the solid bridge or the inter-particle attraction forces (Kaliyan and Morey, 2010). In addition, the binder of bentonite was supplemented for briquetting, which also could slacken the auto-swelling rate for briquettes. By contrast, the auto-swelling rate of briquettes or pellets in NaOH solution appeared to be relatively easier than that their auto-swelling in water according to Fig 1. It may be due to that part of lignin degradation or dissolution in alkalinity conditions, which could potentially improve the water absorption during the auto-swelling process (Alvira et al., 2010). Moreover, the lignin in the biomass was partially decomposed and released from the cell

58

Y. Li et al. / Bioresource Technology 151 (2014) 54–62

Table 2 The main composition of the briquettes, pellets and their undensified biomass after pretreatment. Pretreatment type

a

Main composition(%)

Unbriquetted biomass a

Briquettes

Unpelleted biomass

Pellets

Diluted-NaOH pretreatment

Glucan Xylan Lignin

66.4 ± 2.1 18.7 ± 0.2 7.3 ± 1.7

55.4 ± 3.7 16.6 ± 0.3 6.3 ± 0.5

58.6 ± 2.7 6.1 ± 1.3 25.7 ± 1.0

50.6 ± 1.0 4.9 ± 2.8 24.1 ± 0.1

Hydrothermal pretreatment

Glucan Xylan Lignin

58.6 ± 2.2 5.4 ± 0.4 29.0 ± 0.0

52.0 ± 2.5 6.0 ± 1.0 23.9 ± 1.0

56.6 ± 3.6 5.6 ± 0.8 31.8 ± 0.3

46.0 ± 0.6 4.0 ± 0.2 34.3 ± 1.4

The values in the table was listed in the form of ‘‘Mean ± Standard deviation’’.

Fig. 2. The solid recovery, lignin removal, glucan and xylan recovery after (a) diluted-NaOH pretreatment and (b) hydrothermal pretreatment.

walls in process of densification. Thus, it was easier accessible for NaOH solution, which could correspondingly accelerate the lignin degradation and dissolution and consequently improved the autoswelling.

3.3. Responses of densification to the pretreatment After the diluted-NaOH and hydrothermal pretreatment, the main composition of the briquettes, pellets and their undensified

Y. Li et al. / Bioresource Technology 151 (2014) 54–62

59

Fig. 3. The sugar conversion rate during enzymatic hydrolysis of alkalinity-pretreated biomass (j Briquettes; d Unbriquetted biomass; h Pellets; s Unpelleted biomass).

biomass were listed in Table 2. The lignin content presented an obvious decrease when the alkalinity pretreatment was performed on the briquettes and the unbriquetted biomass (see Tables 1 and 2). Unlike the briquettes, a little bit of increase of lignin content in the pellets and the unpelleted biomass after the alkalinity pretreatment can be observed (see Table 2). The decrease of xylan content can be observed in the pellets, the briquettes and their undensified biomass after NaOH pretreatment, respectively. When the hydrothermal pretreatment was performed on the pellets, briquettes and their corresponding undensified biomass, the xylan content was typically decreased in all pretreated biomass. Consequently, the lignin and glucan content were increased after hydrothermal pretreatment. Generally, the lower solid recovery after alkalinity pretreatment implies the more lignin is removed or the more insoluble carbohydrates are degraded into soluble products. According to Fig. 2a, the solid recovery was 37.1% and 40.7% for the briquettes and the

unbriquetted biomass, respectively, when the alkalinity pretreatment was performed, and it was 37.7% and 42.1% for the pellets and the unpelleted biomass, respectively. Furthermore, the lignin removal was well responded to the solid recovery after the alkalinity pretreatment. Concretely, the lignin removal of the briquettes and the pellets were both higher than their corresponding undensified biomass. There were 87.1% and 85.9% lignin removal for the briquettes and the unbriquetted biomass, respectively, and 54.0% and 43.2% for the pellets and the unpelleted biomass. On the contrary, the 61.3% and 61.6% glucan recovery of the briquettes and the pellets after pretreatment were significantly lower than that of the unbriquetted biomass (90.0%) and unpelleted biomass (78.3%). Similarly, xylan recovery of the briquettes and the pellets after pretreatment were 35.6% and 12.8%, which presented to be lower than their corresponding undensified biomass (39.0% for unbriquetted biomass and 16.1% for the unpelleted biomass) (see Fig. 2a). Based on these results, it could be primarily judged that

60

Y. Li et al. / Bioresource Technology 151 (2014) 54–62

Fig. 4. The sugar conversion rate during enzymatic hydrolysis of hydrothermal-pretreated biomass. (j Briquettes; d Unbriquetted biomass; h Pellets; s Unpelleted biomass).

the briquetting and pelleting positively responded to the NaOH pretreatment comparing with their undensified biomass. When the hydrothermal pretreatment was performed for the briquettes, the pellets and their corresponding undensified biomass, the solid recovery of the briquettes and the unbriquetted biomass was 62.2% and 44.0%, respectively. However, it was 48.6% and 53.7%, respectively, for the pellets and the unpelleted biomass (see Fig. 2b). The solid recovery results meant that the employed pretreatment severity on the unbriquetted biomass and pellets was more excessive comparing with the briquettes and unpelleted biomass. Additionally, it was reported that the hydrothermal pretreatment was typically characterized with the hemicellulose removal (Petersen et al., 2009). Thus, the higher xylan recovery based on the pretreated biomass meant the lower efficiency of hydrothermal pretreatment. According to Fig. 2b, the xylan recovery was 21.7% and 12.2% for the briquettes and the unbriquetted biomass, and 13.5% and 18.8% for the pellets and

the unpelleted biomass. Obviously, the pelleting responded to the hydrothermal pretreatment better in contrast to the unpelleted biomass, and the briquetting appeared to decrease the pretreatment efficiency. Moreover, the lower glucan recovery after the hydrothermal pretreatment indicated the more glucan may be degraded or cleaved, which also meant the pretreatment severity was more excessive on the substrates. The glucan recovery of the pellets after the hydrothermal pretreatment was 72.1%, which was significantly lower than the unpelleted biomass (96.3%). However, 96.5% glucan can be recovered in the briquettes after the pretreatment comparing with the unbriquetted biomass (82.1%). These results again proved that the biomass pelleting positively responded to the hydrothermal pretreatment. However, the pretreatment difficulty may be potentially enhanced by the biomass briquetting. Besides, there were about 34.8% and 38.6% lignin removal after the hydrothermal pretreatment for the briquettes and the unbriquetted biomass, respectively. Unlike the briquettes, the lignin

Y. Li et al. / Bioresource Technology 151 (2014) 54–62

removal of pellets was 9.9%, which was higher than that of the unpelleted biomass (8.1%). The lignin removal results also implied that the hydrothermal pretreatment efficiency can potentially improved by the biomass pelleting. 3.4. Responses of densification to the subsequent enzymatic hydrolysis The pretreated substrates were enzymatically hydrolyzed to evaluate the responses of densification to sugar production. The sugar conversion rates of the diluted-NaOH pretreated substrates were presented in Fig. 3. 96.8% and 80.6% sugar conversion rate of the pretreated briquettes and unbriquetted biomass could be obtained after 72 h hydrolysis (Fig. 3a). Based on the analysis of variance (ANOVA), the sugar conversion rate of the briquettes was significantly higher than that of the unbriquetted biomass (p < 0.05). The sugar conversion rate of the pretreated pellets was 74.5% (see Fig. 3b). Although it was higher than that of the unpelleted biomass, the difference was not significant (p > 0.05) according to the ANOVA results. Based on these results, it was found that the enzymatic hydrolysis efficiency positively correlated to the alkalinity pretreatment. As a result, biomass briquetted and pelleting could improve the alkalinity pretreatment and correspondingly presented to be more beneficial to sugar release in enzymatic hydrolysis process. The high sugar conversion of the densified biomass was obtained, which was mainly attribute to much more hemicellulose removal in the densified biomass by the alkalinity pretreatment (see Fig. 2a). In addition, the thermal softening or cleavage of lignin that occurred during the densification also likely contributed to make the lignin removal much easier in the alkalinity pretreatment, which also potentially improved the sugar conversion from the densified biomass by enzymatic hydrolysis (Rijal et al., 2012). Moreover, similar results for the pellets of switch grass, corn stover, wheat straw, big bluestem and sorghum stalk were also observed in references using acid pretreatment (Rijal et al., 2012; Theerarattananoon et al., 2012). When the hydrothermal pretreated biomass was hydrolyzed, the sugar conversion rates varied in the range of 49.7–58.5% for the briquettes, pellets and their corresponding undensified biomass according to Fig. 4. The sugar conversion rate (at 72 h) of the pretreated unbriquetted biomass was increased by17.6% (see Fig. 4a), which was significantly higher than that of the briquettes (p < 0.05). Accroding to Fig. 4b, there was 7.0% increase of sugar conversion rate as the pretreated pellets were enzymatically hydrolyzed after 72 h comparing with the unpelleted biomass, but the difference was not significant (p > 0.05). According to the reference, the corn stover were hydrothermally pretreated at 200 °C for 15 min, and approximately 45% sugar conversion rate was obtained from the pretreated substrates with the cellulase loading of 30 FPU(g glucan)1 (Mosier et al., 2005). The obtained results in current work were comparable with the sugar conversion rate in the mentioned reference. Combining the results in Fig. 2b, the xylan removal of the pellets was higher than the unpelleted biomass, which meant the hydrothermal pretreatment on the pellets should be better. Consequently, the sugar conversion rate of the pretreated pellets (53.8%) was reasonably better than the unpelleted biomass (50.3%). It was also no surprising that the sugar conversion rate of pretreated briquettes was 49.7%, which was significantly lower than that of the unbriquetted biomass (58.5%). As a result, the biomass pelleting positively responded to hydrothermal pretreatment and enzymatic hydrolysis comparing with the unpelleted biomass. However, the responses of biomass briquetting to hydrothermal pretreatment and enzymatic hydrolysis were negative comparing with the unbriquetted bioamss. Overall, the hydrolysis of the hydrothermal pretreated biomass was not better than that of the alkalinity pretreatment. This mainly

61

because lignin content in the recovered biomass after the hydrothermal pretreatment was relatively high varying from 24% to 34% (see Table 2). The enzymes in the hydrolysis could be absorbed by the excessive residual lignin in the substrates, which may seriously limit the sugar release from the biomass. (Tu et al., 2009). In addition, the responses of the densified and undensified biomass to water-involved pretreatments and the subsequent enzymatic hydrolysis were mainly compared in this work. However, the responses of different types of densification to a specified pretreatment and the corresponding enzymatic hydrolysis may be potentially affected by main factors, such as the type of raw materials, the biomass particle sizes for densification, and the type of biomass densification technologies. Thus, further investigations should be performed on these issues.

4. Conclusions According to the analysis above, it could be found that the biomass briquetting and pelleting both presented positive responses to the alkalinity pretreatment and the subsequent enzymatic hydrolysis comparing with the undensified biomass. However, the biomass briquetting showed negative responses to hydrothermal pretreatment in comparison with the unbriquetted biomass, and resulted in lower sugar conversion rate during subsequent enzymatic hydrolysis. Unlike the briquettes, the pelleting did not negatively affect the hydrothermal pretreatment and enzymatic hydrolysis comparing with the unpelleted biomass. In addition, the auto-swelling for densification biomass can be employed to facilitate the water-involved pretreatments and the subsequent sugar conversion.

Acknowledgements We give our appreciations to the National Natural Science Foundation of China (No. 21306120) and the Department of Science and Technology of Sichuan Province (No. 2013SZ0090) for the funding supports to this work.

References Alevanau, A., Ahmed, I., Gupta, A.K., Yang, W., Blasiak, W., 2011. Parameters of high temperature steam gasification of original and pulverised wood pellets. Fuel Process. Technol. 92, 2068–2074. Alvira, P., Tomás-Pejó, E., Ballesteros, M., Negro, M., 2010. Pretreatment technologies for an efficient bioethanol production process based on enzymatic hydrolysis: a review. Bioresour. Technol. 101, 4851–4861. Bura, R., Mansfield, S.D., Saddler, J.N., Bothast, R.J., 2002. SO2-catalyzed steam explosion of corn fiber for ethanol production. Appl. Biochem. Biotechnol. 98, 59–72. Erlich, C., Björnbom, E., Bolado, D., Giner, M., Fransson, T.H., 2006. Pyrolysis and gasification of pellets from sugar cane bagasse and wood. Fuel 85, 1535–1540. Fang, C., Mariotti, N., Cloutier, A., Koubaa, A., Blanchet, P., 2012. Densification of wood veneers by compression combined with heat and steam. Eur. J. Wood Wood Prod. 70, 155–163. Finney, K.N., Sharifi, V.N., Swithenbank, J., 2009. Fuel pelletization with a binder: part i identification of a suitable binder for spent mushroom compost coal tailing pellets. Energy Fuels 23, 3195–3202. García-Maraver, A., Popov, V., Zamorano, M., 2011. A review of European standards for pellet quality. Renew. Energy 36, 3537–3540. Grover, P., Mishra, S., 1996. Biomass briquetting: technology and practices. In: Regional Wood Energy Development Program in Asia, Tech. Report GCP/RAS/ 154/NET. Food and Agricultural Organization of the United Nations, Bangkok, Thailand. Kaliyan, N., Vance Morey, R., 2009. Factors affecting strength and durability of densified biomass products. Biomass Bioenergy 33, 337–359. Kaliyan, N., Morey, R.V., 2010. Natural binders and solid bridge type binding mechanisms in briquettes and pellets made from corn stover and switchgrass. Bioresour. Technol. 101, 1082–1090. Kumar, L., Tooyserkani, Z., Sokhansanj, S., Saddler, J., 2012. Does densification influence the steam pretreatment and enzymatic hydrolysis of softwoods to sugars? Bioresour. Technol. 121, 190–198.

62

Y. Li et al. / Bioresource Technology 151 (2014) 54–62

Mosier, N., Hendrickson, R., Ho, N., Sedlak, M., Ladisch, M.R., 2005. Optimization of pH controlled liquid hot water pretreatment of corn stover. Bioresour. Technol. 96, 1986–1993. Percival Zhang, Y.H., Himmel, M.E., Mielenz, J.R., 2006. Outlook for cellulase improvement: screening and selection strategies. Biotechnol. Adv. 24, 452–481. Petersen, M.Ø., Larsen, J., Thomsen, M.H., 2009. Optimization of hydrothermal pretreatment of wheat straw for production of bioethanol at low water consumption without addition of chemicals. Biomass Bioenergy 33, 834–840. Ragauskas, A.J., Williams, C.K., Davison, B.H., Britovsek, G., Cairney, J., Eckert, C.A., Frederick, W.J., Hallett, J.P., Leak, D.J., Liotta, C.L., 2006. The path forward for biofuels and biomaterials. Science 311, 484–489. Rijal, B., Igathinathane, C., Karki, B., Yu, M., Pryor, S.W., 2012. Combined effect of pelleting and pretreatment on enzymatic hydrolysis of switchgrass. Bioresour. Technol. 116, 36–41. Sastry, K., Fuerstenau, D., 1973. Mechanisms of agglomerate growth in green pelletization. Powder Technol. 7, 97–105. Sokhansanj, S., Turhollow, A., 2004. Biomass densification-cubing operations and costs for corn stover. Appl. Eng. Agric. 20, 495–502. Song, Y., Tumuluru, J., Iroba, K., Tabil, L., Xin, M., Meda, V., 2010. Material and operating variables affecting the physical quality of biomass briquettes. In: Proceedings of the XVIIthWorld Congress of the International Commission of Agricultural and Biosystems Engineering (CIGR), June 13–17, Québec, Canada.

Su, D., Sun, J., Liu, P., LÜ, Y., 2006. Effects of different pretreatment modes on the enzymatic digestibility of corn leaf and corn stalk. Chin. J. Chem. Eng. 14, 796–801. Theerarattananoon, K., Xu, F., Wilson, J., Staggenborg, S., Mckinney, L., Vadlani, P., Pei, Z., Wang, D., 2012. Effects of the pelleting conditions on chemical composition and sugar yield of corn stover, big bluestem, wheat straw, and sorghum stalk pellets. Bioprocess Biosys. Eng. 35, 615–623. Tooyserkani, Z., Kumar, L., Sokhansanj, S., Saddler, J., Bi, X.T., Jim Lim, C., Lau, A., Melin, S., 2012. SO2-catalyzed steam pretreatment enhances the strength and stability of softwood pellets. Bioresour. Technol. 130, 59–68. Tu, M., Pan, X., Saddler, J.N., 2009. Adsorption of cellulase on cellulolytic enzyme lignin from lodgepole pine. J. Agric. Food Chem. 57, 7771–7778. Tumuluru, J., Sokhansanj, S., Lim, C.J., Bi, X., Lau, A., Melin, S., Oveisi, E., Sowlati, T., 2010. Quality of wood pellets produced in British Columbia for export. Appl. Eng. Agric. 26, 1013–1020. Tumuluru, J.S., Wright, C.T., Hess, J.R., Kenney, K.L., 2011. A review of biomass densification systems to develop uniform feedstock commodities for bioenergy application. Biofuels Bioprod. Biorefining 5, 683–707. Tumuluru, J.S., Wright, C.T., Kenny, K.L., Hess, J.R. 2010b. A review on biomass densification technologies for energy application. Available at: