- Email: [email protected]
Impact of biomass feedstock variability on acid-catalyzed alcoholysis performance
Fuel Processing Technology 180 (2018) 14–22
Contents lists available at ScienceDirect
Fuel Processing Technology journal homepage: www.elsevier.com/locate/fuproc
Research article
Impact of biomass feedstock variability on acid-catalyzed alcoholysis performance Tingting Zhao, Yuxuan Zhang, Guanglu Zhao, Xueli Chen, Lujia Han, Weihua Xiao
T
⁎
Laboratory of Biomass and Bioprocessing Engineering, College of Engineering, China Agricultural University, Beijing 100083, PR China
A R T I C LE I N FO
A B S T R A C T
Keywords: Feedstock types Alcoholysis performance Product distribution Lignin content Cellulose crystallinity index
Alcoholysis was an efficient way for liquefying lignocellulosic biomass into valuable platform chemicals. In this study, the impact of feedstock types on the acid-catalyzed alcoholysis performance was systematically investigated by comparing the alcoholysis ratio, product distribution and production of main products. Component analysis showed that three major chemical components could be efficiently degraded, further, hemicellulose and lignin were prone to be decomposed than cellulose. The time dependent alcoholysis ratio evolution was highly related to the feedstock types with the variation in lignin content and cellulose crystallinity index (CrI). The alcoholysis product distributions also varied significantly with different feedstocks according to the GC–MS analysis results. Ethyl levulinate (EL), furfural and ethyl p-hydroxybenzoate were identified as the major components in the products. The high mass yield 12.4% of EL was obtained from corn stover, and it was highly correlated to the cellulose content and CrI. The maximum mass yield of furfural was produced from wheat straw while a considerable amount of ethyl p-hydroxybenzoate was typically produced from poplar wood. With the comprehensive understanding of the impact of feedstock variability on the alcoholysis performance, our study could give an empirical evidence for product-oriented feedstock selection in biofuels and chemicals production.
1. Introduction Owing to diminishing petroleum reserves and environmental problems, such as global warming and air pollution, it is imperative to utilize biomass resources to produce fuels and chemicals [1–3]. Recently, liquefaction with alcoholic solvent (also called alcoholysis) has attracted considerable attention. As one of the major thermochemical processes, alcoholysis has many advantages, such as the low reaction temperature (< 400 °C) in the comparison with gasification and pyrolysis (400–1000 °C) [4], short reaction time (a few seconds or minutes) when compared with bio-chemical process (several days or weeks) [5]. More importantly, the alcoholysis can simultaneously facilitate the partial decomposition of lignin as well as the cellulose and hemicellulose through the solvolysis rather than remain it in residues [6]. Previous researches have demonstrated that abundant value-added platform chemicals, including glycosides, furan chemicals and levulinate esters, can be produced through the process [4,5,7–10]. However, due to the structural and chemical complexity of raw materials, selection of suitable biomass for biochemicals production is exceedingly necessary to be taken into consideration [5,11,12]. In recent decades, numerous researchers have explored the performance of different biomass, including agricultural residues, woody ⁎
materials and wastes from pulp industry, etc. [11–18]. Grisel found that a considerable amount of methyl glucosides and furfural could be generated after methanolysis of wheat straw at 175 °C with 40 mM H2SO4, the main by-products were levulinic acid and its alkyl ester [11]. Feng et al. used bamboo as an alcoholysis substrate and subcritical methanol as the solvent, 30.75 wt% of methyl levulinate was obtained with 2.5 wt% H2SO4 at 200 °C, meanwhile, glycosides and phenols (lignin derivatives) were also detected in a large amount [14]. Le Van Mao et al. have compared the alcoholysis products of paper pulp, pine wood and switch grass through two-step process, it was found that the production mixtures including ethyl levulinate, ethyl formate, ethyl acetate, 2-furfural and other compounds, were chemically identical but varied significantly on the yield. Besides, this study demonstrated that ethyl levulinate production depended closely on the cellulose content while 2-furfural derived from hemicellulose properly relied on the hemicellulose content of raw materials [16]. Recently, our group found that 34.9 mol% of methyl levulinate could be obtained as the main product when using corn stover as the substrate, and other major co-products were identified as levoglucosenone (LGO), furfural, and methyl 4-hydroxycinnamate [18]. The alcoholysis product distribution and target chemicals production have been generally explored in above researches. However, these results could not be compared due
Corresponding author. E-mail addresses: [email protected] (Y. Zhang), [email protected] (X. Chen), [email protected] (L. Han), [email protected] (W. Xiao).
https://doi.org/10.1016/j.fuproc.2018.08.003 Received 27 April 2018; Received in revised form 24 July 2018; Accepted 2 August 2018 0378-3820/ © 2018 Published by Elsevier B.V.
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
to the variation on reaction conditions and the definition of desired product. The study must be carried out at an identical condition to understand the effect of biomass feedstock variability on the alcoholysis performance. In this study, several common biomass resources, including herbaceous biomass (corn stover, wheat straw, rice straw, rape straw) and woody biomass (poplar wood) were chosen as the experimental substrate. The alcoholysis was conducted with ethanol as the liquefying agent at 180 °C as the reaction time ranging from 0 to 30 min. Microwave heating as a novel technology was applied in this study due to its versatile advantages, such as facilitating heating rates, saving processing times and reducing energy consumption, which cannot be observed in conventional heating methods [18–20]. The alcoholysis ratio, the alcoholysis product distribution, and the production of main products were systematically compared. In addition, the effects of innate variability on alcoholysis performance were discussed here. We hope that this study could profoundly give an empirical support for product-oriented feedstock selection in biochemicals production.
CrI (%) =
(I002 − Iam) × 100% I002
(1)
where I002 is the maximum intensity of 0 0 2 lattice diffraction of cellulose at 2θ = 22.5° and Iam is the peak intensities at around 2θ = 18° representing the amorphous part. 2.3. Microwave-assisted alcoholysis of biomass feedstocks Alcoholysis of each biomass feedstock was carried out in a Milestone microwave apparatus equipped with an internal temperature sensor and a magnetic stirrer. Sulfuric acid was employed as the acid catalyst. A mixture containing 5 g of biomass, 25 g of ethanol, and 0.75 g of sulfuric acid (3 wt%, based on ethanol amount) was introduced into a sealed Teflon reaction vessel (100 mL). The sample was heated to 180 °C for less than 90s under the initial microwave power of 600 W. Then the power was automatically adjusted to keep the temperature constant for a specific time (0–30 min). Subsequently, the vessel was cooled to room temperature using the air-cooling device of the microwave apparatus. All experiments were conducted in quadruplicate. Prior to analysis, 2 g of the resultant mixture was diluted with methanol in a 50-mL volumetric flask and then filtered through a membrane filter (0.22 μm).
2. Materials and methods 2.1. Materials and reagents Corn stover was obtained from the Shang Zhuang experiment station of China Agricultural University. Wheat straw and rape straw were collected from Shandong province. Rice straw and poplar wood were acquired from Hunan province and Hebei province respectively. All the lignocellulosic feedstocks were dried at 105 °C, and then milled and screened using a Taiwan RT-34 hammer mill with a 40-mesh screen. EL (99%), triethyl citrate (99%), LGO (98%), pyrimidine (> 98%), ethyl mercaptan (EtSH, > 98%), and BF3 (> 98%) were purchased from TCI. Glucose (99.5%), xylose (99%), furfural (98%), and N,O-bis (trimethylsilyl)acetamide (BSO, ≥95%) were supplied by Sigma Aldrich. All other chemicals (analytical reagents) were purchased from Beijing Chemical Plant and used as received.
2.4. Alcoholysis ratio The alcoholysis ratio, which indicates the extent of alcoholysis, was determined according to a reported method [24]. The residue was separated from the alcoholysis products using filter paper, washed thoroughly with 80% 1,4-dioxane, and then dried for 24 h at 105 °C. The alcoholysis ratio was calculated as follows:
Weight of residue (g) ⎞ Alcoholysis Ratio (%) = ⎜⎛1 − ⎟ × 100% Weight of total charged biomass (g) ⎠ ⎝ (2)
2.2. Chemical composition and structural property analysis
2.5. Chemical composition of residues
The chemical composition (cellulose, hemicellulose, lignin, and ash) of each biomass feedstock was determined using NREL/TP-510-42,618 [21]. The lignin monomer composition was analyzed through thioacidolysis according to Hoffmann et al. [22]. The corresponding gas chromatography–mass spectrometry (GC–MS) spectra are shown in Fig. S1 and the thioacidolysis monomer yields are presented in Table 1. Xray diffraction measurements were performed (2θ = 5°–40° with a step interval of 0.2°) with a Beijing Purkinje General diffractometer using Cu-Kα radiation at 40 kV and 30 mA. The cellulose crystallinity index could be calculated by the following equation according to an empirical method reported by Segal et al. [23]:
The chemical composition (cellulose, hemicellulose, and lignin) of each residue was determined according to NREL/TP-510-42,618 (Sluiter A et al., 2008) [21]. 2.6. Alcoholysis products analysis of biomass feedstocks 2.6.1. Qualitative analysis GC–MS (Agilent 7890A/5975C equipped with a DB-5MS capillary column (30 m × 0.25 mm × 0.25 μm)) was used to identify the alcoholysis products. The temperatures of the injection port, ion source, and quadrupole were 280, 230, and 150 °C, respectively. The oven was
Table 1 Chemical compositions and structural properties of various feedstocks. Lignocellulosic feedstocks
Corn stover
Cellulose (%) 32.8 ± 0.8 Hemicellulose (%) 18.4 ± 0.3 Lignin (%)a 14.8 ± 0.5 Ash (%) 5.6 ± 0.1 CrI (%) 41.2 ± 1.2 Thioacidolysis yields of H, G, and S monomers (μmol/g)b H 29.9 ± 2.0 G 38.7 ± 2.5 S 47.7 ± 2.2 Total content 116.3 ± 6.7 Molar ratio 0.77/1/1.23 a b
Wheat straw
Rice straw
Rape straw
Poplar wood
31.2 ± 0.5 17.2 ± 0.2 18.2 ± 0.3 8.5 ± 0.0 43.4 ± 1.1
37.1 15.6 19.9 10.1 43.7
26.4 ± 0.1 9.6 ± 0.1 20.1 ± 0.7 7.8 ± 0.3 41.7 ± 1.0
37.1 ± 0.7 10.5 ± 0.3 26.2 ± 0.5 8.6 ± 0.0 52.9 ± 1.4
29.6 ± 1.9 53.1 ± 1.9 56.5 ± 2.0 139.2 ± 5.8 0.56/1/1.07
31.5 ± 2.1 57.5 ± 3.2 53.9 ± 3.1 142.9 ± 8.4 0.55/1/0.94
27.5 ± 1.7 83.2 ± 3.8 101.9 ± 2.5 212.6 ± 8.0 0.33/1/1.23
40.5 ± 2.3 110.3 ± 3.3 148.4 ± 3.8 299.2 ± 9.4 0.37/1/1.35
The lignin content is the sum of acid-soluble and acid-insoluble lignin. The yields are relative to the mass of charged dry biomass. 15
± ± ± ± ±
0.1 0.0 0.9 0.1 1.1
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
programmed to hold an initial temperature of 50 °C for 3 min, ramp at 6 °C min−1 to 180 °C, and then ramp at 10 °C min−1 to the final temperature of 280 °C, which was held for 3 min. The carrier gas was helium and the flow rate was 1.0 mL min−1. The split ratio was 10:1 with an injection volume of 1.0 μL. Mass spectra were recorded from m/z 30 to 600 at a rate of 1.5 scans/s using an electron impact ionization energy of 70 eV. 2.6.2. Quantitative analysis The alcoholysis products were quantified using gas chromatography (GC, Shimadzu GC-2014C equipped with a flame ionization detector and a DB-5 capillary column). The temperature program was the same as that for GC–MS and the carrier gas was nitrogen. Cellulose, hemicellulose, and lignin conversion was determined from the mass change of the corresponding composition in the charged feedstock (Eq. (3)). The mole yield of main products was calculated according to the Eq. (4), and the “Theoretical molar yield of product” was calculated according to the Eq. (5). The mass yield of product based on the weight of charged lignocellulosic feedstock were given in Eq. (6). Fig. 1. Alcoholysis ratio of various feedstocks.
Conversion (%) Weight of chemical composition in residual (g) ⎞⎟ = ⎜⎛1 − Weight of chemical composition in charged feedstock (g) ⎠ ⎝ × 100%
Mole of product (mol) ⎞ Mole Yield (mol%) = ⎜⎛ ⎟ × 100% Theoretical yield of product (mol) ⎠ ⎝
15 min, the alcoholysis ratio of poplar wood increased from 59.27% to 80.54%. However, corn stover still exhibited the highest alcoholysis ratio (87.68%). The maximum alcoholysis ratios of the feedstocks followed the order: corn stover > wheat straw ≈ rape straw > rice straw ≈ poplar wood. Notably, the alcoholysis ratio of rape straw reached a maximum at a reaction time of 25 min, which was much later than that of other feedstocks. In previous studies, several solvents were employed in the biomass liquefaction, such as methanol, glycerol, acetone, ethylene glycol and toluene [4,5,25–27]. The alcoholysis ratio of herbaceous feedstocks in this work were impressive when compared to these researches [4,5,27]. The maximum alcoholysis ratio of 80.54% of poplar wood was also comparable to the liquefaction result of 83.54% in 1-octanol [3].
(3)
(4)
Theoretical yield of product (mol) Cellulose (hemicellulose) content of feedstock charged (g) ⎞ = ⎛⎜ ⎟ ⎝ Molecular weight of anhydroglucose or anhydro xylose (g/mol) ⎠ × 100% (5)
Weight of product (g) ⎞ Mass Yield (%) = ⎜⎛ ⎟ × 100% Weight of feedstock charged (g) ⎠ ⎝
(6)
3.3. Chemical composition of alcoholysis residues Variations in the contents of the three major components in the feedstock residues are shown in Fig. 2. Hemicellulose mainly degraded within 5 min owing to its amorphous structure and low degree of polymerization (Fig. 2A) [28]. The lignin content also decreased simultaneously and only 5% of lignin remained in the poplar wood residue in the end. (Fig. 2B). A slight increase of lignin in corn stover residue at 5 min was probably because the degradation rate of hemicellulose was faster than that of lignin and cellulose. Liquefaction of large amounts of hemicellulose and lignin during the first 5 min augmented the cellulose content (Fig. 2C). After 5 min, the cellulose content decreased while the lignin content generally increased, indicating that the degradation rate of cellulose was much higher than the remaining lignin in the subsequent alcoholysis process. Meanwhile, it was found that cellulose in herbaceous biomass was more readily degraded than that in poplar wood. Following alcoholysis, about 5% of cellulose remained in the corn stover residue, while 18.73% remained in the poplar wood residue.
3. Results and discussion 3.1. Properties of various feedstocks The chemical compositions and structural properties of the five lignocellulosic biomass feedstocks are presented in Table 1. The cellulose content ranged from 26.4% (rape straw) to 37.1% (poplar wood), the hemicellulose content varied from 9.6% (rape straw) to 18.4% (corn stover), and the lignin content varied from 14.8% (corn stover) to 26.2% (poplar wood). As an amorphous heteropolymer, lignin consists of three monomers: p-hydroxylphenyl (H), guaiacyl (G), and syringyl (S) units. The contents of these three monomers were determined by thioacidolysis (Table 1). Poplar wood lignin had the highest total monomer content, with more G and S units than H units, which is consistent with a previous study [22]. Ash, which is not readily liquefied under acidic conditions, was found to have contents of 5.6% (corn stover) to 10.1% (rice straw). Finally, as a structural property, the CrI of each lignocellulosic biomass feedstock was determined, ranging from 41.2% (corn stover) to 52.9% (poplar wood).
3.4. Alcoholysis product distribution of various feedstocks
3.2. Alcoholysis ratio of various feedstocks
The alcoholysis products of various feedstocks were analyzed by GC–MS to evaluate changes in product composition and further understand the reaction complexity of different biomass. Owing to the complicated components, many chemical compounds were generated, leading to enormous chromatographic peaks in the GC–MS spectra (Fig. S2). According to our previous work, alcoholysis products can be classified into four categories based on their origin: cellulose-derived, hemicellulose-derived, lignin-derived, and compounds derived from
Fig. 1 showed that the alcoholysis ratio of each feedstock increased sharply over the first 5 min, then increased much gradually until reaching a maximum value, and finally decreased due to the polycondensation. The alcoholysis process varied significantly between the five feedstocks at 5 min, with the highest alcoholysis ratio of 82.22% obtained for corn stover. By increasing the reaction time from 5 to 16
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Fig. 2. The variation in the contents of chemical compositions of feedstock residues after alcoholysis. (A) hemicellulose, (B) lignin, and (C) cellulose. Table 2 Alcoholysis product distribution from various feedstocksa. No.
1 2 3 4 5 6 7 8 9 10 11 12 a
Category
Cellulose-derived
Hemicellulose-derived
Lignin-derived
Other compounds
Compound name
Ethyl levulinate Levoglucosenone Anhydroglucose Furfural Triethyl citrate Diethyl maleate Ethyl p-hydroxybenzoate Ethyl p-hydroxycinnamate Ethyl ferulate 3,5-dimethoxy-4-hydroxybenzaldehyde Diethyl sulfate Ethyl hexadecanoate
Relative area (%) Corn stover
Wheat straw
Rice straw
Rape straw
Poplar wood
40.23 3.27 1.07 4.45 1.08 0.74 – 4.85 2.46 – 3.83 1.22
26.05 6.01 1.39 7.81 2.51 – – 2.88 4.92 – 5.7 1.82
25.67 8.62 2.97 6.34 2.01 0.82 – 4.62 5.37 – 3.37 2.02
16.57 7.45 2.04 6.08 13.45 5.18 – – – – 3.53 0.97
25.52 3.96 1.58 4.13 – – 18.26 – – 2.63 1.58 –
Alcoholysis products were detected by GC–MS at 30 min (Fig. S1).
alcoholysis, whereas the relative areas of LGO and anhydroglucose were much smaller. In fact, LGO is formed by the pyrolysis of cellulose at an elevated temperature, but once exposure to acid solvent as the reaction proceeds, cellulose is readily decomposed into oligosaccharides and glucose. The glucose can react with the solvent to form several other compounds, such as ethyl-D-glucopyranoside, 2-
other components (e.g., wax and lipid) [15]. Typical alcoholysis products are listed in Table 2 and the fundamental reaction pathway was proposed in Scheme S1. The same cellulose derivatives were observed from all five feedstocks, but their proportions varied. EL represented 16–40% of the total area of alcoholysis products, making it the main compound obtained by 17
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Hemicellulose was easily decomposed within 5 min with numerous derivatives generated by 10 min (Table 4). The selectivity of triethyl citrate was nearly constant during the entire alcoholysis process, whereas that of furfural decreased after 10 min, likely because furfural is readily catalyzed to an insoluble substance through self-condensation or reaction with another active substance under acid conditions [41,42]. The maximum furfural mass yield ranged from 3.5 wt% (rape straw) to 5.5 wt% (wheat straw). The mass yield of triethyl citrate obtained by alcoholysis of rape straw was around 2.9 wt%, which was > 4 times higher than those obtained from the other herbaceous feedstocks. > 80% of lignin was degraded after 5 min, and poplar lignin was more easily broken down than herbaceous biomass lignin (Fig. 3A). This difference might result from ready cleavage of β-O-4′ linkages, the most common lignin linkage, during ethanol organosolv treatment [6]. Typically, the proportion of the β-O-4′ linkages in poplar lignin is around 88%, whereas that in herbaceous biomass is only 72% [17,43]. Notably, significant decreases of lignin conversion were observed for the corn stover, wheat straw, and rice straw around 5 min later. In fact, the repolymerization of lignin or its derives through the condensation between the aromatic C6 or C5 and carbonium ion was greatly enhanced under severe reaction conditions [44,45]. In addition, the lignin fragments also reacted with active compounds, such as polysaccharides and furans, derived from carbohydrates forming insoluble polymers [46]. Furthermore, some of these active compounds could be also repolymerized into “pseudo-lignin” which was detected as the lignin in the residues [45]. Owing to the excellent performance of poplar lignin, the mass yield of ethyl p-hydroxybenzoate from poplar wood was further explored (Fig. 3B). The yield sharply increased over the first 5 min and then a nearly constant value of ~2.6 wt% was reached, indicating that ethyl phydroxybenzoate was generated simultaneously with lignin decomposition and it was stable in the reaction system throughout the alcoholysis process. In general, synthesis of p-hydroxybenzoate esters is performed by direct esterification of p-hydroxybenzoic acid, which is a derivative of petroleum. Based on our results, poplar as a candidate can be used for the production. Besides, the productivity may be improved by several technologies, for example, utilizing extract lignin, reducing reaction temperature, and purifying the products.
(dimethoxymethyl)-5-(methoxymethyl) furan, and EL [10,11,29]. Under acidic conditions, anhydroglucose could be also readily formed through the dehydration of glucose [35]. Compared with the other feedstocks, corn stover produced more EL, but more LGO and anhydroglucose were formed from rice straw and rape straw. As a hemicellulose derivative, furfural was obtained from all five feedstocks. Triethyl citrate was not obtained from poplar wood and diethyl maleate was not detected for wheat straw and poplar wood. However, the relative areas of triethyl citrate and diethyl maleate from rape straw were much higher than those from the other herbaceous biomass feedstocks. This result might be ascribed to considerable amounts of maleic acid and citric acid exuded from rape plants [30]. Though the formation mechanism is not clear, our group has shown that trimethyl citrate is detected during methanolysis of hemicellulose [15]. Considerable differences were found among the lignin derivatives. Ethyl p-hydroxycinnamate and ethyl ferulate were detected for corn stover, wheat straw, and rice straw. Abundant amounts of ferulic acid and p-coumaric acid are found in the cell walls of grass straws, which accounts for the corresponding esters [31–33]. In contrast, ethyl p-hydroxybenzoate was typically observed from poplar wood. There are two possible synthetic routes for ethyl p-hydroxybenzoate formation. In the first route, as p-hydroxybenzoate esters correspond to the free phenolic pendant units originally attached to the γ-positions of poplar lignin, ethyl p-hydroxybenzoate could be obtained by transesterification. In the second route, owing to the rich content of G and S units, p-hydroxyphenyl could be formed from G and S propane units through a cleavage reaction under acidic conditions in ethanol [34,35]. A minor amount of 3,5-dimethoxy-4-hydroxybenzaldehyde, which originates from S units, was also obtained from poplar wood, likely because the content of S units in poplar lignin was much higher than that in the other feedstocks. Thus, the lignin derivatives can be regarded as characteristic compounds of different feedstocks. 3.5. Production of main alcoholysis products of various feedstocks To gain further insight into the alcoholysis progress, variations in the main products as a function of reaction time were analyzed. Table 3 shows the conversion of cellulose component in different feedstocks, mole yields and mass yields of the main derivatives. The conversion of cellulose at 5 min ranged from 42.7% to 73.6% and decreased in the following order: corn stover > wheat straw > rice straw > rape straw > poplar wood. These findings suggest that the depolymerization of cellulose from herbaceous biomass is easier than that from woody biomass. The selectivity of EL increased while that of LGO decreased as the reaction time increased from 5 to 30 min. The sum of the selectivity of these two compounds was always below 50%. We conjecture that most cellulose was converted to ethyl glucoside, an intermediate in the alcoholysis of cellulose to EL, but owing to a high activation energy, subsequent conversion was limited [29,36,37]. Previous studies also indicated that the mole yields of EL ranged from 37 mol% to 58 mol% when using different biomass feedstocks [16,18,38–40]. And the high yields above 50% were generally under optimal conditions (High liquid-solid ratio, high temperature, etc) [38,39]. Compared to the previous studies, 42.6 mol% yield of EL from corn stover in this work was quite considerable. To effectively evaluate EL production of the various feedstocks, we prolonged the reaction time (Fig. S3). A maximum EL mass yield of 12.4 wt% was obtained from corn stover at 30 min, when the reaction had reached equilibrium. Further, a large amount of humins was generated at equilibrium. Overall, the maximum EL mass yield of the five feedstocks decreased in the following order: corn stover (12.4 wt%) > rice straw (11.1 wt%) ≈ wheat straw (11.0 wt %) > poplar wood (8.5 wt%) > rape straw (7.2 wt%). Thus, corn stover should be considered a priority material for commercial EL production.
3.6. Effect of feedstock properties on alcoholysis performance 3.6.1. Effect of feedstock properties on alcoholysis ratio As stated before, the alcoholysis ratio within 5 min varied with different feedstocks. This result was probably attributed to the variation on the intrinsic properties of biomass. As shown in Fig. 4, neither the cellulose nor the hemicellulose contents was correlated with the alcoholysis ratio at 5 min. In contrast, there was a strong negative correlation (R2 = 0.9573) between the lignin content and the alcoholysis ratio. Owing to its highly cross-linked structure, lignin is regarded as the most recalcitrant biopolymer in plant cell walls [5,43,47,48]. Thus, the presence of lignin protects cellulose, hemicellulose, and other components from acid attack [18,47]. Zhang et al. suggested that the crystalline part of biomass consisting of strong inter- and intrachain hydrogen bonds hinders enzyme accessibility [49]. Therefore, the crystallinity might also influence the alcoholysis process. As shown in Fig. 5A, there was a weak negative correlation between CrI and the alcoholysis ratio at 5 min. However, there was a strong negative correlation between cellulose conversion (5 min) and CrI, except for rape straw (Fig. 5B). These results indicate that CrI affected the initial alcoholysis process to some extent through the degradation of cellulose. Considerable deviation was observed for the cellulose conversion of rape straw, likely because its lignin and other chemical compositions (lipid, protein, and soluble carbohydrates, etc.) decomposed prior than cellulose. A weak negative correlation was found between CrI and the maximum alcoholysis ratio (Fig. 5C). However, as hemicellulose and most 18
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Table 3 Conversion of cellulose component, mole yields and mass yields of main derivatives from various feedstocks. Time (min)
Conversion (%)
Mole yield (mol%) Ethyl levulinate
Corn stover
Wheat straw
Rice straw
Rape straw
Poplar wood
5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30
73.6 89.3 94.6 94.6 96.6 96.8 63.3 83.2 91.3 91.8 95.0 97.1 57.2 77.4 85.0 90.6 90.5 93.6 44.2 73.7 80.4 85.8 91.8 89.7 42.7 67.4 82.1 85.6 82.0 89.1
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
a
1.7 1.1b 0.2c 0.2c 2.0c 0.4c 0.2a 3.6b 0.2c 3.3cd 1.2cd 1.4d 0.8a 1.5b 3.4c 0.9d 0.9d 1.1d 4.1a 1.2b 0.5c 2.5cd 0.8d 3.5d 4.3a 2.0b 1.5c 0.5c 4.0c 1.5c
a
16.0 ± 0.1 24.7 ± 2.4b 32.9 ± 0.8c 35.7 ± 2.5cd 39.8 ± 1.9de 44.6 ± 2.3e 10.7 ± 1.3a 18.9 ± 0.5b 28.7 ± 1.5c 26.5 ± 1.8c 33.7 ± 0.9d 38.1 ± 0.8e 9.5 ± 0.9a 17.7 ± 0.5b 21.8 ± 1.4c 25.9 ± 0.8d 29.7 ± 0.8e 33.4 ± 0.5f 8.3 ± 0.17a 13.9 ± 0.18b 17.9 ± 1.2c 18.9 ± 0.8c 24.9 ± 0.8d 26.5 ± 1.5d 8.2 ± 1.0a 15.1 ± 2.1b 19.3 ± 1.2c 23.3 ± 0.7d 24.9 ± 2.0d 25.7 ± 0.7d
Mass Yield (wt%) Levoglucosenone 6.9 7.0 5.9 4.8 4.8 4.1 7.2 7.2 6.6 6.8 5.5 6.0 6.1 6.5 6.3 6.0 5.1 4.5 8.3 8.5 8.5 7.5 6.7 6.7 5.4 6.4 6.9 7.7 6.3 5.8
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
a
0.0 0.0b 0.4c 0.4d 0.2d 0.3d 0.1a 0.1a 0.6ab 0.1ab 0.5c 0.1bc 0.6bc 0.1c 0.6bc 0.3bc 0.1ab 0.8a 0.1ab 0.3a 0.3a 0.6ab 0.9b 1.2b 0.5a 0.1bc 0.2c 0.4c 0.1bc 0.2ab
Ethyl levulinate a
4.7 ± 0.0 7.2 ± 0.7b 9.6 ± 0.2c 10.4 ± 0.7cd 11.6 ± 0.5de 12.4 ± 0.7e 3.0 ± 0.4a 5.2 ± 0.1b 8.0 ± 0.4c 8.6 ± 0.5c 9.4 ± 0.3d 10.6 ± 0.2e 3.1 ± 0.3a 5.8 ± 0.2b 7.19 ± 0.5c 8.55 ± 0.3d 9.8 ± 0.3e 11.0 ± 0.2f 2.0 ± 0.0a 3.3 ± 0.0b 4.2 ± 0.3c 4.4 ± 0.2c 5.3 ± 0.4d 6.2 ± 0.4e 2.7 ± 0.3a 5.0 ± 0.7b 6.4 ± 0.4c 7.7 ± 0.2d 8.2 ± 0.7d 8.5 ± 0.2d
Levoglucosenone 1.8 1.8 1.5 1.2 1.2 1.1 1.7 1.7 1.6 1.6 1.3 1.4 1.8 1.9 1.8 1.7 1.5 1.3 1.7 1.8 1.8 1.5 1.4 1.4 1.6 1.8 2.0 1.9 1.8 1.7
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.0a 0.0a 0.1b 0.1c 0.1c 0.1c 0.0a 0.0a 0.2ab 0.0ab 0.1c 0.0bc 0.2ab 0.0b 0.2ab 0.1ab 0.0ac 0.2c 0.0ab 0.1a 0.1a 0.1ab 0.2b 0.2b 0.1a 0.0bc 0.1c 0.10c 0.0bc 0.1ab
Different superscript letters in the same column indicate significant differences (p < 0.05). Table 4 Conversion of hemicellulose component, mole yields and mass yields of main derivatives from various feedstocks. Time (min)
Conversion (%)
Mole yield (mol%) Furfural
Corn stover
Wheat straw
Rice straw
Rape straw
Poplar wood
5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30
98.7 99.2 99.4 99.3 99.2 99.1 98.9 100.0 100.0 100.0 100.0 100.0 97.7 100.0 100.0 100.0 100.0 100.0 97.3 100.0 100.0 100.0 100.0 100.0 94.1 100.0 100.0 100.0 100.0 100.0
28.4 30.7 27.7 24.4 23.8 22.1 44.1 39.7 38.6 35.6 32.2 32.1 36.7 38.0 33.8 35.8 33.3 27.5 58.3 60.7 56.8 53.6 55.9 54.5 41.9 45.1 44.3 43.0 42.7 39.8
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
1.2a 0.5b 0.3a 1.5c 0.2cd 0.8d 1.8a 1.6b 0.1bd 0.6bd 1.3cd 0.8d 0.3a 1.0ab 2.6ab 2.3ab 0.7a 1.7c 1.2a 0.5ab 0.6abc 0.1c 3.5bc 1.4c 1.6ab 1.0d 0.7c 0.4bc 0.4bc 0.3a
nd: not detected. Different superscript letters in the same column indicate significant differences (p < 0.05). 19
Mass yield (%) Triethyl citrate
Furfural
2.1 ± 0.2a 2.2 ± 0.2a 2.2 ± 0.1a 2.1 ± 0.2a 2.2 ± 0.1a 2.0 ± 0.1a 2.4 ± 0.1a 2.4 ± 0.1a 2.5 ± 0.1a 2.4 ± 0.0a 2.4 ± 0.1a 2.3 ± 0.1a 2.8 ± 0.2a 3.0 ± 0.0abc 3.0 ± 0.0abc 3.2 ± 0.0bc 3.2 ± 0.8c 3.0 ± 0.0ab 14.4 ± 0.0ab 14.7 ± 0.3ab 14.2 ± 1.3ab 12.4 ± 1.2a 15.3 ± 1.5b 14.8 ± 0.5ab nd nd nd nd nd nd
4.1 4.3 4.0 3.8 3.9 3.8 5.5 5.0 4.8 4.5 4.0 4.0 4.0 4.3 3.8 4.1 3.8 3.1 3.2 3.5 3.4 3.3 3.3 3.1 3.8 4.1 3.7 3.3 3.2 3.0
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.1ab 0.0a 0.0abc 0.0c 0.2bc 0.1bc 0.5a 0.2ab 0.0bc 0.1bcd 0.2d 0.1d 0.0ab 0.1a 0.3ab 0.3ab 0.1b 0.2c 0.1ad 0.0b 0.0bc 0.0bcd 0.0cd 0.0a 0.2a 0.1b 0.0a 0.2c 0.0cd 0.1d
Triethyl citrate 0.8 0.9 0.9 0.8 0.8 0.8 0.9 0.9 0.9 0.9 0.9 0.8 0.9 1.0 1.0 1.0 1.1 1.0 2.9 2.9 2.8 2.5 3.1 3.0 nd nd nd nd nd nd
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.1a 0.1a 0.0a 0.0a 0.0a 0.0a 0.0a 0.0a 0.1a 0.0a 0.1a 0.0a 0.1a 0.0abc 0.0abc 0.0bc 0.0c 0.0ab 0.0ab 0.1ab 0.3ab 0.2a 0.3b 0.1ab
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Fig. 3. Lignin conversion as a function of reaction time (A) and ethyl p-hydroxybenzoate mass yield of poplar wood (B).
Changes in the cellulose content always corresponded to similar change in the EL mass yield, indicating that the cellulose content likely played a more important role in EL formation. However, the cellulose content does not predict the EL mass yield perfectly. For instance, poplar wood, which had a high cellulose content, yielded a relatively low amount of EL after alcoholysis. As discussed before, the high cellulose crystallinity of poplar wood likely suppressed the degradation of cellulose; thus, in addition to the cellulose content, cellulose crystallinity also affected EL production to some extent. Based on this result, biomass feedstocks with low cellulose crystallinity and high cellulose can be chosen to produce the EL in the industrial production. There was a weak positive correlation between the furfural mass yield and the hemicellulose content (Fig. 6B). While corn stover had the highest hemicellulose content, wheat straw exhibited the highest furfural mass yield, likely because furfural was readily converted into humins through polycondensation reactions at high temperature.
4. Conclusion Fig. 4. Correlation between the contents of different components and the alcoholysis ratio at 5 min.
The impact of biomass feedstock variability (corn stover, wheat straw, rice straw, rape straw and poplar wood) on alcoholysis performance was investigated. The maximum alcoholysis ratios of various feedstocks descended in the following order: corn stover > wheat straw ≈ rape straw > rice straw ≈ poplar wood. Most of the hemicellulose and over 80% of lignin in different feedstocks decomposed within 5 min, while cellulose gradually degraded throughout the process. Lignin content and cellulose crystallite hindered the alcoholysis successively. Considering the production distribution of different biomass, the cellulose derivatives were chemically identical but varied in the proportions. Hemicellulose and lignin derivatives of corn stover, wheat straw and rice straw were similar. Rape straw could obtain more triethyl citrate and diethyl maleate from hemicellulose while poplar lignin produced ethyl p-hydroxybenzoate as a characteristic compound. As the main product, the maximum EL mass yield of the five feedstocks decreased in the following order: corn stover > rice straw > wheat straw > poplar wood > rape straw. Wheat straw could produce more furfural than other feedstocks while poplar wood with a considerable amount of ethyl p-hydroxybenzoate made it a potential material for aromatic compounds production. Comprehensive investigation on the impact of feedstock variability on the acid-catalyzed alcoholysis performance could profoundly understand the alcoholysis mechanism and give some practical supports on selecting appropriate biomass for different industrial production in future.
lignin were decomposed after 5 min, degradation of any remaining cellulose was the dominant process during subsequent alcoholysis. A strong correlation was found between the incremental alcoholysis ratio and CrI (Fig. 5D), indicating that cellulose crystallinity played an important role in the subsequent alcoholysis process. In addition, samples with identical lignin contents and CrI (wheat straw and rice straw) showed different alcoholysis ratios, which may be explained by the difference in the ash content of these samples. In summary, the alcoholysis process was mainly affected by lignin content and CrI successively. Due to the rapid decomposition of lignin, reducing the crystallinity of single feedstock by some methods (ball milling, freeze-dried) can be effectively facilitate the process.
3.6.2. Effect of feedstock properties on main products mass yields The alcoholysis product composition essentially depended on the chemical composition of the original feedstocks, especially their characteristic components (Table 1). Thus, the effect of the feedstock properties on the mass yields of the main derivatives was analyzed. The relative importance of cellulose content and cellulose crystallinity on EL mass yield is shown in Fig. 6A for the various feedstocks. Similar trends were observed for the cellulose content and EL mass yield, whereas the trend for cellulose crystallinity values was different. 20
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Fig. 5. Correlation between CrI and alcoholysis ratio (A), cellulose conversion (B) at the reaction time of 5 min; Correlation between CrI and the maximum alcoholysis ratio (C), the increment of alcoholysis ratio (D).
Acknowledgements
2016YFE0112800).
The authors are extremely grateful to the financial support by the National Natural Science Foundation of China (No. 31671572) and the National Key Research and Development Program of China (No.
Funding This work was financially supported by the National Natural Science
Fig. 6. Effect of cellulose content and CrI on EL mass yield at 30 min (A) and correlation between hemicellulose content and furfural mass yield at 10 min (B). 21
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Foundation of China (No. 31671572) and the National Key Research and Development Program of China (No. 2016YFE0112800).
42618, (2008). [22] L. Hoffmann, S. Besseau, P. Geoffroy, C. Ritzenthaler, D. Meyer, C. Lapierre, B. Pollet, M. Legrand, Silencing of hydroxycinnamoyl-coenzyme a shikimate/quinate hydroxycinnamoyltransferase affects phenylpropanoid biosynthesis, Plant Cell 16 (2004) 1446–1465. [23] L. Segal, J.J. Creely, A.E. Martin Jr., C.M. Conrad, An empirical method for estimating the degree of crystallinity of native cellulose using the X-ray diffractometer, Text. Res. J. 29 (1959) 786–794. [24] T. Yamada, H. Ono, Rapid liquefaction of lignocellulosic waste by using ethylene carbonate, Bioresour. Technol. 70 (1999) 61–67. [25] Y. Wang, J. Wu, Y. Wan, H. Lei, F. Yu, P. Chen, X. Lin, Y. Liu, R. Ruan, Liquefaction of corn stover using industrial biodiesel glycerol, Int. J. Agric. Biol. Eng. 2 (2009) 32–40. [26] Q. Zhai, F. Li, F. Wang, J. Xu, J. Jiang, Z. Cai, Liquefaction of poplar biomass for value-added platform chemicals, Cellulose 25 (2018) 4663–4675. [27] L. Cao, C. Zhang, S. Hao, G. Luo, S. Zhang, Effect of glycerol as co-solvent on yields of bio-oil from rice straw through hydrothermal liquefaction, Bioresour. Technol. 220 (2016) 471–478. [28] Y. Lu, N.S. Mosier, Kinetic modeling analysis of maleic acid-catalyzed hemicellulose hydrolysis in corn stover, Biotechnol. Bioeng. 101 (2008) 1170–1181. [29] X. Hu, C.Z. Li, Levulinic esters from the acid-catalysed reactions of sugars and alcohols as part of a bio-refinery, Green Chem. 13 (2011) 1676–1679. [30] E. Hoffland, Quantitative evaluation of the role of organic acid exudation in the mobilization of rock phosphate by rape, Plant Soil 140 (1992) 279–289. [31] C. Crestini, D.S. Argyropoulos, Structural analysis of wheat straw lignin by quantitative 31P and 2D NMR spectroscopy. The occurrence of ester bonds and α-O-4 substructures, J. Agric. Food Chem. 45 (1997) 1212–1219. [32] G. Jacquet, B. Pollet, C. Lapierre, New ether-linked ferulic acid-coniferyl alcohol dimers identified in grass straws, J. Agric. Food Chem. 43 (1995) 2746–2751. [33] J. Ralph, R.D. HatTield, S. Quideau, R.F. Helm, J.H. Grabber, H.G. Jug, Pathway of p-coumaric acid incorporation into maize lignin as revealed by NMR, J. Am. Chem. Soc. 116 (1994) 9448–9456. [34] J. Ralph, T. Akiyama, H.D. Coleman, S.D. Mansfield, Effects on lignin structure of coumarate 3-hydroxylase downregulation in poplar, Bioenergy Res. 5 (2012) 1009–1019. [35] K. Sarkanen, C. Schuerch, Lignin structure. XI. A quantitative study of the alcoholysis of lignin, J. Am. Chem. Soc. 79 (1957) 4203–4209. [36] W. Deng, M. Liu, Q. Zhang, Y. Wang, Direct transformation of cellulose into methyl and ethyl glucosides in methanol and ethanol media catalyzed by heteropolyacids, Catal. Today 164 (2011) 461–466. [37] F. Yu, R. Zhong, H. Chong, M. Smet, W. Dehaen, B.F. Sels, Fast catalytic conversion of recalcitrant cellulose into alkyl levulinates and levulinic acid in the presence of soluble and recoverable sulfonated hyperbranched poly(arylene oxindole)s, Green Chem. 19 (2017) 153–163. [38] C. Chang, G. Xu, X. Jiang, Production of ethyl levulinate by direct conversion of wheat straw in ethanol media, Bioresour. Technol. 121 (2012) 93–99. [39] Y. Zhang, X. Wang, T. Hou, H. Liu, L. Han, W. Xiao, Efficient microwave-assisted production of biofuel ethyl levulinate from corn stover in ethanol medium, J. Energy Chem. 27 (2018) 890–897. [40] C. Chang, G. Xu, W. Zhu, J. Bai, S. Fang, One-pot production of a liquid biofuel candidate—Ethyl levulinate from glucose and furfural residues using a combination of extremely low sulfuric acid and zeolite USY, Fuel 140 (2015) 365–370. [41] A.S. Mamman, J. Lee, Y. Kim, I.T. Hwang, N. Park, Y.K. Hwang, J. Chang, J. Hwang, Furfural: Hemicellulose/xylose- derived biochemical, Biofuels Bioprod. Biorefin. 2 (2008) 438–454. [42] R. Weingarten, G.A. Tompsett, W.C. Conner Jr., G.W. Huber, Design of solid acid catalysts for aqueous-phase dehydration of carbohydrates: the role of Lewis and Brønsted acid sites, J. Catal. 279 (2011) 174–182. [43] A. Barakat, F. Monlau, A. Solhy, H. Carrere, Mechanical dissociation and fragmentation of lignocellulosic biomass: effect of initial moisture, biochemical and structural proprieties on energy requirement, Appl. Energy 142 (2015) 240–246. [44] R. El Hage, N. Brosse, P. Sannigrahi, A. Ragauskas, Effects of process severity on the chemical structure of Miscanthus ethanol organosolv lignin, Polym. Degrad. Stab. 95 (2010) 997–1003. [45] R. El Hage, L. Chrusciel, L. Desharnais, N. Brosse, Effect of autohydrolysis of Miscanthus x giganteus on lignin structure and organosolv delignification, Bioresour. Technol. 101 (2010) 9321–9329. [46] X. Ouyang, X. Huang, B.M.S. Hendriks, M.D. Boot, E.J.M. Hensen, Coupling organosolv fractionation and reductive depolymerization of woody biomass in a twostep catalytic process, Green Chem. 20 (2018) 2308–2319. [47] G. Cheng, P. Varanasi, C. Li, H. Liu, Y.B. Melnichenko, B.A. Simmons, M.S. Kent, S. Singh, Transition of cellulose crystalline structure and surface morphology of biomass as a function of ionic liquid pretreatment and its relation to enzymatic hydrolysis, Biomacromolecules 12 (2011) 933–941. [48] A. Demirbas, Effect of lignin content on aqueous liquefaction products of biomass, Energy Convers. Manag. 41 (2000) 1601–1607. [49] Y.P. Zhang, L.R. Lynd, Toward an Aggregated Understanding of Enzymatic Hydrolysis of Cellulose: Noncomplexed Cellulase Systems, 88 (2004), pp. 797–824.
Notes The authors declare no competing financial interest. Abbreviations EL CrI LGO BSO
ethyl levulinate cellulose crystallinity index levoglucosenone N,O-bis(trimethylsilyl)acetamide
Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.fuproc.2018.08.003. References [1] G.W. Huber, S. Iborra, A. Corma, Synthesis of transportation fuels from biomass: chemistry, catalysts, and engineering, Chem. Rev. 106 (2006) 4044–4098. [2] E.L. Kunkes, D.A. Simonetti, R.M. West, J.C. Serrano-Ruiz, C.A. Gärtner, J.A. Dumesic, Catalytic conversion of biomass to monofunctional hydrocarbons and targeted liquid-fuel classes, Science 322 (2008) 417–421. [3] X.W. Zou, T.F. Qin, Y. Wang, L.H. Huang, Mechanisms and product specialties of the alcoholysis processes of poplar components, Energy Fuel 25 (2011) 3786–3792. [4] Z. Liu, F. Zhang, Effects of various solvents on the liquefaction of biomass to produce fuels and chemical feedstocks, Energy Convers. Manag. 49 (2008) 3498–3504. [5] H.J. Huang, X.Z. Yuan, Recent progress in the direct liquefaction of typical biomass, Prog. Energy Combust. Sci. 49 (2015) 59–80. [6] B.B. Hallac, Y. Pu, A.J. Ragauskas, Chemical transformations of Buddleja davidii lignin during ethanol organosolv pretreatment, Biomass Bioenergy 24 (2010) 2723–2732. [7] S. Branda, J. Kim, Liquefaction of major lignocellulosic biomass constituents in supercritical ethanol, Energy 80 (2015) 64–74. [8] S. Kang, X. Li, J. Fan, J. Chang, Hydrothermal conversion of lignin: a review, Renew. Sust. Energ. Rev. 27 (2013) 546–558. [9] H. Pan, Synthesis of polymers from organic solvent liquefied biomass: a review, Renew. Sust. Energ. Rev. 15 (2011) 3454–3463. [10] R. Van Putten, J.C. van der Waal, E. de Jong, C.B. Rasrendra, H.J. Heeres, J.G. de Vries, Hydroxymethylfurfural, a versatile platform chemical made from renewable resources, Chem. Rev. 113 (2013) 1499–1597. [11] R.J.H. Grisel, J.C. van der Waal, E. de Jong, W.J.J. Huijgen, Acid catalysed alcoholysis of wheat straw: Towards second generation furan-derivatives, Catal. Today 223 (2014) 3–10. [12] L. Peng, L. Lin, H. Li, Q. Yang, Conversion of carbohydrates biomass into levulinate esters using heterogeneous catalysts, Appl. Energy 88 (2011) 4590–4596. [13] P. Duan, B. Jin, Y. Xu, Y. Yang, X. Bai, F. Wang, L. Zhang, J. Miao, Thermo-chemical conversion of Chlorella pyrenoidosa to liquid biofuels, Bioresour. Technol. 133 (2013) 197–205. [14] J. Feng, J. Jiang, J. Xu, Z. Yang, K. Wang, Q. Guan, S. Chen, Preparation of methyl levulinate from fractionation of direct liquefied bamboo biomass, Appl. Energy 154 (2015) 520–527. [15] H. Huang, S. Ramaswamy, W. Al-Dajani, U. Tschirner, R. Cairncross, Effect of biomass species and plant size on cellulosic ethanol: a comparative process and economic analysis, Biomass Bioenergy 33 (2009) 234–246. [16] R. Le Van Mao, Q. Zhao, G. Dima, D. Petraccone, New process for the acid-catalyzed conversion of cellulosic biomass (AC3B) into alkyl levulinates and other esters using a unique one-pot system of reaction and product extraction, Catal. Lett. 141 (2011) 271–276. [17] X. Meng, Y. Pu, C.G. Yoo, M. Li, G. Bali, An in-depth understanding of biomass recalcitrance using natural poplar variants as the feedstock, ChemSusChem 10 (2017) 139–150. [18] W. Xiao, X. Chen, Y. Zhang, T. Qu, L. Han, Product analysis for microwave-assisted methanolysis of lignocellulose, Energy Fuel 30 (2016) 8246–8251. [19] C.O. Kappe, D. Dallinger, Controlled microwave heating in modern organic synthesis: highlights from the 2004-2008 literature, Mol. Divers. 13 (2009) 71–193. [20] E.S. Kim, S. Liu, M.M. Abu-Omar, N.S. Mosier, Selective conversion of biomass hemicellulose to furfural using maleic acid with microwave heating, Energy Fuel 26 (2012) 1298–1304. [21] A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter, D. Templeton, D. Crocker, Determination of Structural Carbohydrates and Lignin in Biomass, NREL/TP-510-
22
Contents lists available at ScienceDirect
Fuel Processing Technology journal homepage: www.elsevier.com/locate/fuproc
Research article
Impact of biomass feedstock variability on acid-catalyzed alcoholysis performance Tingting Zhao, Yuxuan Zhang, Guanglu Zhao, Xueli Chen, Lujia Han, Weihua Xiao
T
⁎
Laboratory of Biomass and Bioprocessing Engineering, College of Engineering, China Agricultural University, Beijing 100083, PR China
A R T I C LE I N FO
A B S T R A C T
Keywords: Feedstock types Alcoholysis performance Product distribution Lignin content Cellulose crystallinity index
Alcoholysis was an efficient way for liquefying lignocellulosic biomass into valuable platform chemicals. In this study, the impact of feedstock types on the acid-catalyzed alcoholysis performance was systematically investigated by comparing the alcoholysis ratio, product distribution and production of main products. Component analysis showed that three major chemical components could be efficiently degraded, further, hemicellulose and lignin were prone to be decomposed than cellulose. The time dependent alcoholysis ratio evolution was highly related to the feedstock types with the variation in lignin content and cellulose crystallinity index (CrI). The alcoholysis product distributions also varied significantly with different feedstocks according to the GC–MS analysis results. Ethyl levulinate (EL), furfural and ethyl p-hydroxybenzoate were identified as the major components in the products. The high mass yield 12.4% of EL was obtained from corn stover, and it was highly correlated to the cellulose content and CrI. The maximum mass yield of furfural was produced from wheat straw while a considerable amount of ethyl p-hydroxybenzoate was typically produced from poplar wood. With the comprehensive understanding of the impact of feedstock variability on the alcoholysis performance, our study could give an empirical evidence for product-oriented feedstock selection in biofuels and chemicals production.
1. Introduction Owing to diminishing petroleum reserves and environmental problems, such as global warming and air pollution, it is imperative to utilize biomass resources to produce fuels and chemicals [1–3]. Recently, liquefaction with alcoholic solvent (also called alcoholysis) has attracted considerable attention. As one of the major thermochemical processes, alcoholysis has many advantages, such as the low reaction temperature (< 400 °C) in the comparison with gasification and pyrolysis (400–1000 °C) [4], short reaction time (a few seconds or minutes) when compared with bio-chemical process (several days or weeks) [5]. More importantly, the alcoholysis can simultaneously facilitate the partial decomposition of lignin as well as the cellulose and hemicellulose through the solvolysis rather than remain it in residues [6]. Previous researches have demonstrated that abundant value-added platform chemicals, including glycosides, furan chemicals and levulinate esters, can be produced through the process [4,5,7–10]. However, due to the structural and chemical complexity of raw materials, selection of suitable biomass for biochemicals production is exceedingly necessary to be taken into consideration [5,11,12]. In recent decades, numerous researchers have explored the performance of different biomass, including agricultural residues, woody ⁎
materials and wastes from pulp industry, etc. [11–18]. Grisel found that a considerable amount of methyl glucosides and furfural could be generated after methanolysis of wheat straw at 175 °C with 40 mM H2SO4, the main by-products were levulinic acid and its alkyl ester [11]. Feng et al. used bamboo as an alcoholysis substrate and subcritical methanol as the solvent, 30.75 wt% of methyl levulinate was obtained with 2.5 wt% H2SO4 at 200 °C, meanwhile, glycosides and phenols (lignin derivatives) were also detected in a large amount [14]. Le Van Mao et al. have compared the alcoholysis products of paper pulp, pine wood and switch grass through two-step process, it was found that the production mixtures including ethyl levulinate, ethyl formate, ethyl acetate, 2-furfural and other compounds, were chemically identical but varied significantly on the yield. Besides, this study demonstrated that ethyl levulinate production depended closely on the cellulose content while 2-furfural derived from hemicellulose properly relied on the hemicellulose content of raw materials [16]. Recently, our group found that 34.9 mol% of methyl levulinate could be obtained as the main product when using corn stover as the substrate, and other major co-products were identified as levoglucosenone (LGO), furfural, and methyl 4-hydroxycinnamate [18]. The alcoholysis product distribution and target chemicals production have been generally explored in above researches. However, these results could not be compared due
Corresponding author. E-mail addresses: [email protected] (Y. Zhang), [email protected] (X. Chen), [email protected] (L. Han), [email protected] (W. Xiao).
https://doi.org/10.1016/j.fuproc.2018.08.003 Received 27 April 2018; Received in revised form 24 July 2018; Accepted 2 August 2018 0378-3820/ © 2018 Published by Elsevier B.V.
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
to the variation on reaction conditions and the definition of desired product. The study must be carried out at an identical condition to understand the effect of biomass feedstock variability on the alcoholysis performance. In this study, several common biomass resources, including herbaceous biomass (corn stover, wheat straw, rice straw, rape straw) and woody biomass (poplar wood) were chosen as the experimental substrate. The alcoholysis was conducted with ethanol as the liquefying agent at 180 °C as the reaction time ranging from 0 to 30 min. Microwave heating as a novel technology was applied in this study due to its versatile advantages, such as facilitating heating rates, saving processing times and reducing energy consumption, which cannot be observed in conventional heating methods [18–20]. The alcoholysis ratio, the alcoholysis product distribution, and the production of main products were systematically compared. In addition, the effects of innate variability on alcoholysis performance were discussed here. We hope that this study could profoundly give an empirical support for product-oriented feedstock selection in biochemicals production.
CrI (%) =
(I002 − Iam) × 100% I002
(1)
where I002 is the maximum intensity of 0 0 2 lattice diffraction of cellulose at 2θ = 22.5° and Iam is the peak intensities at around 2θ = 18° representing the amorphous part. 2.3. Microwave-assisted alcoholysis of biomass feedstocks Alcoholysis of each biomass feedstock was carried out in a Milestone microwave apparatus equipped with an internal temperature sensor and a magnetic stirrer. Sulfuric acid was employed as the acid catalyst. A mixture containing 5 g of biomass, 25 g of ethanol, and 0.75 g of sulfuric acid (3 wt%, based on ethanol amount) was introduced into a sealed Teflon reaction vessel (100 mL). The sample was heated to 180 °C for less than 90s under the initial microwave power of 600 W. Then the power was automatically adjusted to keep the temperature constant for a specific time (0–30 min). Subsequently, the vessel was cooled to room temperature using the air-cooling device of the microwave apparatus. All experiments were conducted in quadruplicate. Prior to analysis, 2 g of the resultant mixture was diluted with methanol in a 50-mL volumetric flask and then filtered through a membrane filter (0.22 μm).
2. Materials and methods 2.1. Materials and reagents Corn stover was obtained from the Shang Zhuang experiment station of China Agricultural University. Wheat straw and rape straw were collected from Shandong province. Rice straw and poplar wood were acquired from Hunan province and Hebei province respectively. All the lignocellulosic feedstocks were dried at 105 °C, and then milled and screened using a Taiwan RT-34 hammer mill with a 40-mesh screen. EL (99%), triethyl citrate (99%), LGO (98%), pyrimidine (> 98%), ethyl mercaptan (EtSH, > 98%), and BF3 (> 98%) were purchased from TCI. Glucose (99.5%), xylose (99%), furfural (98%), and N,O-bis (trimethylsilyl)acetamide (BSO, ≥95%) were supplied by Sigma Aldrich. All other chemicals (analytical reagents) were purchased from Beijing Chemical Plant and used as received.
2.4. Alcoholysis ratio The alcoholysis ratio, which indicates the extent of alcoholysis, was determined according to a reported method [24]. The residue was separated from the alcoholysis products using filter paper, washed thoroughly with 80% 1,4-dioxane, and then dried for 24 h at 105 °C. The alcoholysis ratio was calculated as follows:
Weight of residue (g) ⎞ Alcoholysis Ratio (%) = ⎜⎛1 − ⎟ × 100% Weight of total charged biomass (g) ⎠ ⎝ (2)
2.2. Chemical composition and structural property analysis
2.5. Chemical composition of residues
The chemical composition (cellulose, hemicellulose, lignin, and ash) of each biomass feedstock was determined using NREL/TP-510-42,618 [21]. The lignin monomer composition was analyzed through thioacidolysis according to Hoffmann et al. [22]. The corresponding gas chromatography–mass spectrometry (GC–MS) spectra are shown in Fig. S1 and the thioacidolysis monomer yields are presented in Table 1. Xray diffraction measurements were performed (2θ = 5°–40° with a step interval of 0.2°) with a Beijing Purkinje General diffractometer using Cu-Kα radiation at 40 kV and 30 mA. The cellulose crystallinity index could be calculated by the following equation according to an empirical method reported by Segal et al. [23]:
The chemical composition (cellulose, hemicellulose, and lignin) of each residue was determined according to NREL/TP-510-42,618 (Sluiter A et al., 2008) [21]. 2.6. Alcoholysis products analysis of biomass feedstocks 2.6.1. Qualitative analysis GC–MS (Agilent 7890A/5975C equipped with a DB-5MS capillary column (30 m × 0.25 mm × 0.25 μm)) was used to identify the alcoholysis products. The temperatures of the injection port, ion source, and quadrupole were 280, 230, and 150 °C, respectively. The oven was
Table 1 Chemical compositions and structural properties of various feedstocks. Lignocellulosic feedstocks
Corn stover
Cellulose (%) 32.8 ± 0.8 Hemicellulose (%) 18.4 ± 0.3 Lignin (%)a 14.8 ± 0.5 Ash (%) 5.6 ± 0.1 CrI (%) 41.2 ± 1.2 Thioacidolysis yields of H, G, and S monomers (μmol/g)b H 29.9 ± 2.0 G 38.7 ± 2.5 S 47.7 ± 2.2 Total content 116.3 ± 6.7 Molar ratio 0.77/1/1.23 a b
Wheat straw
Rice straw
Rape straw
Poplar wood
31.2 ± 0.5 17.2 ± 0.2 18.2 ± 0.3 8.5 ± 0.0 43.4 ± 1.1
37.1 15.6 19.9 10.1 43.7
26.4 ± 0.1 9.6 ± 0.1 20.1 ± 0.7 7.8 ± 0.3 41.7 ± 1.0
37.1 ± 0.7 10.5 ± 0.3 26.2 ± 0.5 8.6 ± 0.0 52.9 ± 1.4
29.6 ± 1.9 53.1 ± 1.9 56.5 ± 2.0 139.2 ± 5.8 0.56/1/1.07
31.5 ± 2.1 57.5 ± 3.2 53.9 ± 3.1 142.9 ± 8.4 0.55/1/0.94
27.5 ± 1.7 83.2 ± 3.8 101.9 ± 2.5 212.6 ± 8.0 0.33/1/1.23
40.5 ± 2.3 110.3 ± 3.3 148.4 ± 3.8 299.2 ± 9.4 0.37/1/1.35
The lignin content is the sum of acid-soluble and acid-insoluble lignin. The yields are relative to the mass of charged dry biomass. 15
± ± ± ± ±
0.1 0.0 0.9 0.1 1.1
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
programmed to hold an initial temperature of 50 °C for 3 min, ramp at 6 °C min−1 to 180 °C, and then ramp at 10 °C min−1 to the final temperature of 280 °C, which was held for 3 min. The carrier gas was helium and the flow rate was 1.0 mL min−1. The split ratio was 10:1 with an injection volume of 1.0 μL. Mass spectra were recorded from m/z 30 to 600 at a rate of 1.5 scans/s using an electron impact ionization energy of 70 eV. 2.6.2. Quantitative analysis The alcoholysis products were quantified using gas chromatography (GC, Shimadzu GC-2014C equipped with a flame ionization detector and a DB-5 capillary column). The temperature program was the same as that for GC–MS and the carrier gas was nitrogen. Cellulose, hemicellulose, and lignin conversion was determined from the mass change of the corresponding composition in the charged feedstock (Eq. (3)). The mole yield of main products was calculated according to the Eq. (4), and the “Theoretical molar yield of product” was calculated according to the Eq. (5). The mass yield of product based on the weight of charged lignocellulosic feedstock were given in Eq. (6). Fig. 1. Alcoholysis ratio of various feedstocks.
Conversion (%) Weight of chemical composition in residual (g) ⎞⎟ = ⎜⎛1 − Weight of chemical composition in charged feedstock (g) ⎠ ⎝ × 100%
Mole of product (mol) ⎞ Mole Yield (mol%) = ⎜⎛ ⎟ × 100% Theoretical yield of product (mol) ⎠ ⎝
15 min, the alcoholysis ratio of poplar wood increased from 59.27% to 80.54%. However, corn stover still exhibited the highest alcoholysis ratio (87.68%). The maximum alcoholysis ratios of the feedstocks followed the order: corn stover > wheat straw ≈ rape straw > rice straw ≈ poplar wood. Notably, the alcoholysis ratio of rape straw reached a maximum at a reaction time of 25 min, which was much later than that of other feedstocks. In previous studies, several solvents were employed in the biomass liquefaction, such as methanol, glycerol, acetone, ethylene glycol and toluene [4,5,25–27]. The alcoholysis ratio of herbaceous feedstocks in this work were impressive when compared to these researches [4,5,27]. The maximum alcoholysis ratio of 80.54% of poplar wood was also comparable to the liquefaction result of 83.54% in 1-octanol [3].
(3)
(4)
Theoretical yield of product (mol) Cellulose (hemicellulose) content of feedstock charged (g) ⎞ = ⎛⎜ ⎟ ⎝ Molecular weight of anhydroglucose or anhydro xylose (g/mol) ⎠ × 100% (5)
Weight of product (g) ⎞ Mass Yield (%) = ⎜⎛ ⎟ × 100% Weight of feedstock charged (g) ⎠ ⎝
(6)
3.3. Chemical composition of alcoholysis residues Variations in the contents of the three major components in the feedstock residues are shown in Fig. 2. Hemicellulose mainly degraded within 5 min owing to its amorphous structure and low degree of polymerization (Fig. 2A) [28]. The lignin content also decreased simultaneously and only 5% of lignin remained in the poplar wood residue in the end. (Fig. 2B). A slight increase of lignin in corn stover residue at 5 min was probably because the degradation rate of hemicellulose was faster than that of lignin and cellulose. Liquefaction of large amounts of hemicellulose and lignin during the first 5 min augmented the cellulose content (Fig. 2C). After 5 min, the cellulose content decreased while the lignin content generally increased, indicating that the degradation rate of cellulose was much higher than the remaining lignin in the subsequent alcoholysis process. Meanwhile, it was found that cellulose in herbaceous biomass was more readily degraded than that in poplar wood. Following alcoholysis, about 5% of cellulose remained in the corn stover residue, while 18.73% remained in the poplar wood residue.
3. Results and discussion 3.1. Properties of various feedstocks The chemical compositions and structural properties of the five lignocellulosic biomass feedstocks are presented in Table 1. The cellulose content ranged from 26.4% (rape straw) to 37.1% (poplar wood), the hemicellulose content varied from 9.6% (rape straw) to 18.4% (corn stover), and the lignin content varied from 14.8% (corn stover) to 26.2% (poplar wood). As an amorphous heteropolymer, lignin consists of three monomers: p-hydroxylphenyl (H), guaiacyl (G), and syringyl (S) units. The contents of these three monomers were determined by thioacidolysis (Table 1). Poplar wood lignin had the highest total monomer content, with more G and S units than H units, which is consistent with a previous study [22]. Ash, which is not readily liquefied under acidic conditions, was found to have contents of 5.6% (corn stover) to 10.1% (rice straw). Finally, as a structural property, the CrI of each lignocellulosic biomass feedstock was determined, ranging from 41.2% (corn stover) to 52.9% (poplar wood).
3.4. Alcoholysis product distribution of various feedstocks
3.2. Alcoholysis ratio of various feedstocks
The alcoholysis products of various feedstocks were analyzed by GC–MS to evaluate changes in product composition and further understand the reaction complexity of different biomass. Owing to the complicated components, many chemical compounds were generated, leading to enormous chromatographic peaks in the GC–MS spectra (Fig. S2). According to our previous work, alcoholysis products can be classified into four categories based on their origin: cellulose-derived, hemicellulose-derived, lignin-derived, and compounds derived from
Fig. 1 showed that the alcoholysis ratio of each feedstock increased sharply over the first 5 min, then increased much gradually until reaching a maximum value, and finally decreased due to the polycondensation. The alcoholysis process varied significantly between the five feedstocks at 5 min, with the highest alcoholysis ratio of 82.22% obtained for corn stover. By increasing the reaction time from 5 to 16
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Fig. 2. The variation in the contents of chemical compositions of feedstock residues after alcoholysis. (A) hemicellulose, (B) lignin, and (C) cellulose. Table 2 Alcoholysis product distribution from various feedstocksa. No.
1 2 3 4 5 6 7 8 9 10 11 12 a
Category
Cellulose-derived
Hemicellulose-derived
Lignin-derived
Other compounds
Compound name
Ethyl levulinate Levoglucosenone Anhydroglucose Furfural Triethyl citrate Diethyl maleate Ethyl p-hydroxybenzoate Ethyl p-hydroxycinnamate Ethyl ferulate 3,5-dimethoxy-4-hydroxybenzaldehyde Diethyl sulfate Ethyl hexadecanoate
Relative area (%) Corn stover
Wheat straw
Rice straw
Rape straw
Poplar wood
40.23 3.27 1.07 4.45 1.08 0.74 – 4.85 2.46 – 3.83 1.22
26.05 6.01 1.39 7.81 2.51 – – 2.88 4.92 – 5.7 1.82
25.67 8.62 2.97 6.34 2.01 0.82 – 4.62 5.37 – 3.37 2.02
16.57 7.45 2.04 6.08 13.45 5.18 – – – – 3.53 0.97
25.52 3.96 1.58 4.13 – – 18.26 – – 2.63 1.58 –
Alcoholysis products were detected by GC–MS at 30 min (Fig. S1).
alcoholysis, whereas the relative areas of LGO and anhydroglucose were much smaller. In fact, LGO is formed by the pyrolysis of cellulose at an elevated temperature, but once exposure to acid solvent as the reaction proceeds, cellulose is readily decomposed into oligosaccharides and glucose. The glucose can react with the solvent to form several other compounds, such as ethyl-D-glucopyranoside, 2-
other components (e.g., wax and lipid) [15]. Typical alcoholysis products are listed in Table 2 and the fundamental reaction pathway was proposed in Scheme S1. The same cellulose derivatives were observed from all five feedstocks, but their proportions varied. EL represented 16–40% of the total area of alcoholysis products, making it the main compound obtained by 17
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Hemicellulose was easily decomposed within 5 min with numerous derivatives generated by 10 min (Table 4). The selectivity of triethyl citrate was nearly constant during the entire alcoholysis process, whereas that of furfural decreased after 10 min, likely because furfural is readily catalyzed to an insoluble substance through self-condensation or reaction with another active substance under acid conditions [41,42]. The maximum furfural mass yield ranged from 3.5 wt% (rape straw) to 5.5 wt% (wheat straw). The mass yield of triethyl citrate obtained by alcoholysis of rape straw was around 2.9 wt%, which was > 4 times higher than those obtained from the other herbaceous feedstocks. > 80% of lignin was degraded after 5 min, and poplar lignin was more easily broken down than herbaceous biomass lignin (Fig. 3A). This difference might result from ready cleavage of β-O-4′ linkages, the most common lignin linkage, during ethanol organosolv treatment [6]. Typically, the proportion of the β-O-4′ linkages in poplar lignin is around 88%, whereas that in herbaceous biomass is only 72% [17,43]. Notably, significant decreases of lignin conversion were observed for the corn stover, wheat straw, and rice straw around 5 min later. In fact, the repolymerization of lignin or its derives through the condensation between the aromatic C6 or C5 and carbonium ion was greatly enhanced under severe reaction conditions [44,45]. In addition, the lignin fragments also reacted with active compounds, such as polysaccharides and furans, derived from carbohydrates forming insoluble polymers [46]. Furthermore, some of these active compounds could be also repolymerized into “pseudo-lignin” which was detected as the lignin in the residues [45]. Owing to the excellent performance of poplar lignin, the mass yield of ethyl p-hydroxybenzoate from poplar wood was further explored (Fig. 3B). The yield sharply increased over the first 5 min and then a nearly constant value of ~2.6 wt% was reached, indicating that ethyl phydroxybenzoate was generated simultaneously with lignin decomposition and it was stable in the reaction system throughout the alcoholysis process. In general, synthesis of p-hydroxybenzoate esters is performed by direct esterification of p-hydroxybenzoic acid, which is a derivative of petroleum. Based on our results, poplar as a candidate can be used for the production. Besides, the productivity may be improved by several technologies, for example, utilizing extract lignin, reducing reaction temperature, and purifying the products.
(dimethoxymethyl)-5-(methoxymethyl) furan, and EL [10,11,29]. Under acidic conditions, anhydroglucose could be also readily formed through the dehydration of glucose [35]. Compared with the other feedstocks, corn stover produced more EL, but more LGO and anhydroglucose were formed from rice straw and rape straw. As a hemicellulose derivative, furfural was obtained from all five feedstocks. Triethyl citrate was not obtained from poplar wood and diethyl maleate was not detected for wheat straw and poplar wood. However, the relative areas of triethyl citrate and diethyl maleate from rape straw were much higher than those from the other herbaceous biomass feedstocks. This result might be ascribed to considerable amounts of maleic acid and citric acid exuded from rape plants [30]. Though the formation mechanism is not clear, our group has shown that trimethyl citrate is detected during methanolysis of hemicellulose [15]. Considerable differences were found among the lignin derivatives. Ethyl p-hydroxycinnamate and ethyl ferulate were detected for corn stover, wheat straw, and rice straw. Abundant amounts of ferulic acid and p-coumaric acid are found in the cell walls of grass straws, which accounts for the corresponding esters [31–33]. In contrast, ethyl p-hydroxybenzoate was typically observed from poplar wood. There are two possible synthetic routes for ethyl p-hydroxybenzoate formation. In the first route, as p-hydroxybenzoate esters correspond to the free phenolic pendant units originally attached to the γ-positions of poplar lignin, ethyl p-hydroxybenzoate could be obtained by transesterification. In the second route, owing to the rich content of G and S units, p-hydroxyphenyl could be formed from G and S propane units through a cleavage reaction under acidic conditions in ethanol [34,35]. A minor amount of 3,5-dimethoxy-4-hydroxybenzaldehyde, which originates from S units, was also obtained from poplar wood, likely because the content of S units in poplar lignin was much higher than that in the other feedstocks. Thus, the lignin derivatives can be regarded as characteristic compounds of different feedstocks. 3.5. Production of main alcoholysis products of various feedstocks To gain further insight into the alcoholysis progress, variations in the main products as a function of reaction time were analyzed. Table 3 shows the conversion of cellulose component in different feedstocks, mole yields and mass yields of the main derivatives. The conversion of cellulose at 5 min ranged from 42.7% to 73.6% and decreased in the following order: corn stover > wheat straw > rice straw > rape straw > poplar wood. These findings suggest that the depolymerization of cellulose from herbaceous biomass is easier than that from woody biomass. The selectivity of EL increased while that of LGO decreased as the reaction time increased from 5 to 30 min. The sum of the selectivity of these two compounds was always below 50%. We conjecture that most cellulose was converted to ethyl glucoside, an intermediate in the alcoholysis of cellulose to EL, but owing to a high activation energy, subsequent conversion was limited [29,36,37]. Previous studies also indicated that the mole yields of EL ranged from 37 mol% to 58 mol% when using different biomass feedstocks [16,18,38–40]. And the high yields above 50% were generally under optimal conditions (High liquid-solid ratio, high temperature, etc) [38,39]. Compared to the previous studies, 42.6 mol% yield of EL from corn stover in this work was quite considerable. To effectively evaluate EL production of the various feedstocks, we prolonged the reaction time (Fig. S3). A maximum EL mass yield of 12.4 wt% was obtained from corn stover at 30 min, when the reaction had reached equilibrium. Further, a large amount of humins was generated at equilibrium. Overall, the maximum EL mass yield of the five feedstocks decreased in the following order: corn stover (12.4 wt%) > rice straw (11.1 wt%) ≈ wheat straw (11.0 wt %) > poplar wood (8.5 wt%) > rape straw (7.2 wt%). Thus, corn stover should be considered a priority material for commercial EL production.
3.6. Effect of feedstock properties on alcoholysis performance 3.6.1. Effect of feedstock properties on alcoholysis ratio As stated before, the alcoholysis ratio within 5 min varied with different feedstocks. This result was probably attributed to the variation on the intrinsic properties of biomass. As shown in Fig. 4, neither the cellulose nor the hemicellulose contents was correlated with the alcoholysis ratio at 5 min. In contrast, there was a strong negative correlation (R2 = 0.9573) between the lignin content and the alcoholysis ratio. Owing to its highly cross-linked structure, lignin is regarded as the most recalcitrant biopolymer in plant cell walls [5,43,47,48]. Thus, the presence of lignin protects cellulose, hemicellulose, and other components from acid attack [18,47]. Zhang et al. suggested that the crystalline part of biomass consisting of strong inter- and intrachain hydrogen bonds hinders enzyme accessibility [49]. Therefore, the crystallinity might also influence the alcoholysis process. As shown in Fig. 5A, there was a weak negative correlation between CrI and the alcoholysis ratio at 5 min. However, there was a strong negative correlation between cellulose conversion (5 min) and CrI, except for rape straw (Fig. 5B). These results indicate that CrI affected the initial alcoholysis process to some extent through the degradation of cellulose. Considerable deviation was observed for the cellulose conversion of rape straw, likely because its lignin and other chemical compositions (lipid, protein, and soluble carbohydrates, etc.) decomposed prior than cellulose. A weak negative correlation was found between CrI and the maximum alcoholysis ratio (Fig. 5C). However, as hemicellulose and most 18
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Table 3 Conversion of cellulose component, mole yields and mass yields of main derivatives from various feedstocks. Time (min)
Conversion (%)
Mole yield (mol%) Ethyl levulinate
Corn stover
Wheat straw
Rice straw
Rape straw
Poplar wood
5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30
73.6 89.3 94.6 94.6 96.6 96.8 63.3 83.2 91.3 91.8 95.0 97.1 57.2 77.4 85.0 90.6 90.5 93.6 44.2 73.7 80.4 85.8 91.8 89.7 42.7 67.4 82.1 85.6 82.0 89.1
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
a
1.7 1.1b 0.2c 0.2c 2.0c 0.4c 0.2a 3.6b 0.2c 3.3cd 1.2cd 1.4d 0.8a 1.5b 3.4c 0.9d 0.9d 1.1d 4.1a 1.2b 0.5c 2.5cd 0.8d 3.5d 4.3a 2.0b 1.5c 0.5c 4.0c 1.5c
a
16.0 ± 0.1 24.7 ± 2.4b 32.9 ± 0.8c 35.7 ± 2.5cd 39.8 ± 1.9de 44.6 ± 2.3e 10.7 ± 1.3a 18.9 ± 0.5b 28.7 ± 1.5c 26.5 ± 1.8c 33.7 ± 0.9d 38.1 ± 0.8e 9.5 ± 0.9a 17.7 ± 0.5b 21.8 ± 1.4c 25.9 ± 0.8d 29.7 ± 0.8e 33.4 ± 0.5f 8.3 ± 0.17a 13.9 ± 0.18b 17.9 ± 1.2c 18.9 ± 0.8c 24.9 ± 0.8d 26.5 ± 1.5d 8.2 ± 1.0a 15.1 ± 2.1b 19.3 ± 1.2c 23.3 ± 0.7d 24.9 ± 2.0d 25.7 ± 0.7d
Mass Yield (wt%) Levoglucosenone 6.9 7.0 5.9 4.8 4.8 4.1 7.2 7.2 6.6 6.8 5.5 6.0 6.1 6.5 6.3 6.0 5.1 4.5 8.3 8.5 8.5 7.5 6.7 6.7 5.4 6.4 6.9 7.7 6.3 5.8
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
a
0.0 0.0b 0.4c 0.4d 0.2d 0.3d 0.1a 0.1a 0.6ab 0.1ab 0.5c 0.1bc 0.6bc 0.1c 0.6bc 0.3bc 0.1ab 0.8a 0.1ab 0.3a 0.3a 0.6ab 0.9b 1.2b 0.5a 0.1bc 0.2c 0.4c 0.1bc 0.2ab
Ethyl levulinate a
4.7 ± 0.0 7.2 ± 0.7b 9.6 ± 0.2c 10.4 ± 0.7cd 11.6 ± 0.5de 12.4 ± 0.7e 3.0 ± 0.4a 5.2 ± 0.1b 8.0 ± 0.4c 8.6 ± 0.5c 9.4 ± 0.3d 10.6 ± 0.2e 3.1 ± 0.3a 5.8 ± 0.2b 7.19 ± 0.5c 8.55 ± 0.3d 9.8 ± 0.3e 11.0 ± 0.2f 2.0 ± 0.0a 3.3 ± 0.0b 4.2 ± 0.3c 4.4 ± 0.2c 5.3 ± 0.4d 6.2 ± 0.4e 2.7 ± 0.3a 5.0 ± 0.7b 6.4 ± 0.4c 7.7 ± 0.2d 8.2 ± 0.7d 8.5 ± 0.2d
Levoglucosenone 1.8 1.8 1.5 1.2 1.2 1.1 1.7 1.7 1.6 1.6 1.3 1.4 1.8 1.9 1.8 1.7 1.5 1.3 1.7 1.8 1.8 1.5 1.4 1.4 1.6 1.8 2.0 1.9 1.8 1.7
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.0a 0.0a 0.1b 0.1c 0.1c 0.1c 0.0a 0.0a 0.2ab 0.0ab 0.1c 0.0bc 0.2ab 0.0b 0.2ab 0.1ab 0.0ac 0.2c 0.0ab 0.1a 0.1a 0.1ab 0.2b 0.2b 0.1a 0.0bc 0.1c 0.10c 0.0bc 0.1ab
Different superscript letters in the same column indicate significant differences (p < 0.05). Table 4 Conversion of hemicellulose component, mole yields and mass yields of main derivatives from various feedstocks. Time (min)
Conversion (%)
Mole yield (mol%) Furfural
Corn stover
Wheat straw
Rice straw
Rape straw
Poplar wood
5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30 5 10 15 20 25 30
98.7 99.2 99.4 99.3 99.2 99.1 98.9 100.0 100.0 100.0 100.0 100.0 97.7 100.0 100.0 100.0 100.0 100.0 97.3 100.0 100.0 100.0 100.0 100.0 94.1 100.0 100.0 100.0 100.0 100.0
28.4 30.7 27.7 24.4 23.8 22.1 44.1 39.7 38.6 35.6 32.2 32.1 36.7 38.0 33.8 35.8 33.3 27.5 58.3 60.7 56.8 53.6 55.9 54.5 41.9 45.1 44.3 43.0 42.7 39.8
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
1.2a 0.5b 0.3a 1.5c 0.2cd 0.8d 1.8a 1.6b 0.1bd 0.6bd 1.3cd 0.8d 0.3a 1.0ab 2.6ab 2.3ab 0.7a 1.7c 1.2a 0.5ab 0.6abc 0.1c 3.5bc 1.4c 1.6ab 1.0d 0.7c 0.4bc 0.4bc 0.3a
nd: not detected. Different superscript letters in the same column indicate significant differences (p < 0.05). 19
Mass yield (%) Triethyl citrate
Furfural
2.1 ± 0.2a 2.2 ± 0.2a 2.2 ± 0.1a 2.1 ± 0.2a 2.2 ± 0.1a 2.0 ± 0.1a 2.4 ± 0.1a 2.4 ± 0.1a 2.5 ± 0.1a 2.4 ± 0.0a 2.4 ± 0.1a 2.3 ± 0.1a 2.8 ± 0.2a 3.0 ± 0.0abc 3.0 ± 0.0abc 3.2 ± 0.0bc 3.2 ± 0.8c 3.0 ± 0.0ab 14.4 ± 0.0ab 14.7 ± 0.3ab 14.2 ± 1.3ab 12.4 ± 1.2a 15.3 ± 1.5b 14.8 ± 0.5ab nd nd nd nd nd nd
4.1 4.3 4.0 3.8 3.9 3.8 5.5 5.0 4.8 4.5 4.0 4.0 4.0 4.3 3.8 4.1 3.8 3.1 3.2 3.5 3.4 3.3 3.3 3.1 3.8 4.1 3.7 3.3 3.2 3.0
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.1ab 0.0a 0.0abc 0.0c 0.2bc 0.1bc 0.5a 0.2ab 0.0bc 0.1bcd 0.2d 0.1d 0.0ab 0.1a 0.3ab 0.3ab 0.1b 0.2c 0.1ad 0.0b 0.0bc 0.0bcd 0.0cd 0.0a 0.2a 0.1b 0.0a 0.2c 0.0cd 0.1d
Triethyl citrate 0.8 0.9 0.9 0.8 0.8 0.8 0.9 0.9 0.9 0.9 0.9 0.8 0.9 1.0 1.0 1.0 1.1 1.0 2.9 2.9 2.8 2.5 3.1 3.0 nd nd nd nd nd nd
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
0.1a 0.1a 0.0a 0.0a 0.0a 0.0a 0.0a 0.0a 0.1a 0.0a 0.1a 0.0a 0.1a 0.0abc 0.0abc 0.0bc 0.0c 0.0ab 0.0ab 0.1ab 0.3ab 0.2a 0.3b 0.1ab
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Fig. 3. Lignin conversion as a function of reaction time (A) and ethyl p-hydroxybenzoate mass yield of poplar wood (B).
Changes in the cellulose content always corresponded to similar change in the EL mass yield, indicating that the cellulose content likely played a more important role in EL formation. However, the cellulose content does not predict the EL mass yield perfectly. For instance, poplar wood, which had a high cellulose content, yielded a relatively low amount of EL after alcoholysis. As discussed before, the high cellulose crystallinity of poplar wood likely suppressed the degradation of cellulose; thus, in addition to the cellulose content, cellulose crystallinity also affected EL production to some extent. Based on this result, biomass feedstocks with low cellulose crystallinity and high cellulose can be chosen to produce the EL in the industrial production. There was a weak positive correlation between the furfural mass yield and the hemicellulose content (Fig. 6B). While corn stover had the highest hemicellulose content, wheat straw exhibited the highest furfural mass yield, likely because furfural was readily converted into humins through polycondensation reactions at high temperature.
4. Conclusion Fig. 4. Correlation between the contents of different components and the alcoholysis ratio at 5 min.
The impact of biomass feedstock variability (corn stover, wheat straw, rice straw, rape straw and poplar wood) on alcoholysis performance was investigated. The maximum alcoholysis ratios of various feedstocks descended in the following order: corn stover > wheat straw ≈ rape straw > rice straw ≈ poplar wood. Most of the hemicellulose and over 80% of lignin in different feedstocks decomposed within 5 min, while cellulose gradually degraded throughout the process. Lignin content and cellulose crystallite hindered the alcoholysis successively. Considering the production distribution of different biomass, the cellulose derivatives were chemically identical but varied in the proportions. Hemicellulose and lignin derivatives of corn stover, wheat straw and rice straw were similar. Rape straw could obtain more triethyl citrate and diethyl maleate from hemicellulose while poplar lignin produced ethyl p-hydroxybenzoate as a characteristic compound. As the main product, the maximum EL mass yield of the five feedstocks decreased in the following order: corn stover > rice straw > wheat straw > poplar wood > rape straw. Wheat straw could produce more furfural than other feedstocks while poplar wood with a considerable amount of ethyl p-hydroxybenzoate made it a potential material for aromatic compounds production. Comprehensive investigation on the impact of feedstock variability on the acid-catalyzed alcoholysis performance could profoundly understand the alcoholysis mechanism and give some practical supports on selecting appropriate biomass for different industrial production in future.
lignin were decomposed after 5 min, degradation of any remaining cellulose was the dominant process during subsequent alcoholysis. A strong correlation was found between the incremental alcoholysis ratio and CrI (Fig. 5D), indicating that cellulose crystallinity played an important role in the subsequent alcoholysis process. In addition, samples with identical lignin contents and CrI (wheat straw and rice straw) showed different alcoholysis ratios, which may be explained by the difference in the ash content of these samples. In summary, the alcoholysis process was mainly affected by lignin content and CrI successively. Due to the rapid decomposition of lignin, reducing the crystallinity of single feedstock by some methods (ball milling, freeze-dried) can be effectively facilitate the process.
3.6.2. Effect of feedstock properties on main products mass yields The alcoholysis product composition essentially depended on the chemical composition of the original feedstocks, especially their characteristic components (Table 1). Thus, the effect of the feedstock properties on the mass yields of the main derivatives was analyzed. The relative importance of cellulose content and cellulose crystallinity on EL mass yield is shown in Fig. 6A for the various feedstocks. Similar trends were observed for the cellulose content and EL mass yield, whereas the trend for cellulose crystallinity values was different. 20
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Fig. 5. Correlation between CrI and alcoholysis ratio (A), cellulose conversion (B) at the reaction time of 5 min; Correlation between CrI and the maximum alcoholysis ratio (C), the increment of alcoholysis ratio (D).
Acknowledgements
2016YFE0112800).
The authors are extremely grateful to the financial support by the National Natural Science Foundation of China (No. 31671572) and the National Key Research and Development Program of China (No.
Funding This work was financially supported by the National Natural Science
Fig. 6. Effect of cellulose content and CrI on EL mass yield at 30 min (A) and correlation between hemicellulose content and furfural mass yield at 10 min (B). 21
Fuel Processing Technology 180 (2018) 14–22
T. Zhao et al.
Foundation of China (No. 31671572) and the National Key Research and Development Program of China (No. 2016YFE0112800).
42618, (2008). [22] L. Hoffmann, S. Besseau, P. Geoffroy, C. Ritzenthaler, D. Meyer, C. Lapierre, B. Pollet, M. Legrand, Silencing of hydroxycinnamoyl-coenzyme a shikimate/quinate hydroxycinnamoyltransferase affects phenylpropanoid biosynthesis, Plant Cell 16 (2004) 1446–1465. [23] L. Segal, J.J. Creely, A.E. Martin Jr., C.M. Conrad, An empirical method for estimating the degree of crystallinity of native cellulose using the X-ray diffractometer, Text. Res. J. 29 (1959) 786–794. [24] T. Yamada, H. Ono, Rapid liquefaction of lignocellulosic waste by using ethylene carbonate, Bioresour. Technol. 70 (1999) 61–67. [25] Y. Wang, J. Wu, Y. Wan, H. Lei, F. Yu, P. Chen, X. Lin, Y. Liu, R. Ruan, Liquefaction of corn stover using industrial biodiesel glycerol, Int. J. Agric. Biol. Eng. 2 (2009) 32–40. [26] Q. Zhai, F. Li, F. Wang, J. Xu, J. Jiang, Z. Cai, Liquefaction of poplar biomass for value-added platform chemicals, Cellulose 25 (2018) 4663–4675. [27] L. Cao, C. Zhang, S. Hao, G. Luo, S. Zhang, Effect of glycerol as co-solvent on yields of bio-oil from rice straw through hydrothermal liquefaction, Bioresour. Technol. 220 (2016) 471–478. [28] Y. Lu, N.S. Mosier, Kinetic modeling analysis of maleic acid-catalyzed hemicellulose hydrolysis in corn stover, Biotechnol. Bioeng. 101 (2008) 1170–1181. [29] X. Hu, C.Z. Li, Levulinic esters from the acid-catalysed reactions of sugars and alcohols as part of a bio-refinery, Green Chem. 13 (2011) 1676–1679. [30] E. Hoffland, Quantitative evaluation of the role of organic acid exudation in the mobilization of rock phosphate by rape, Plant Soil 140 (1992) 279–289. [31] C. Crestini, D.S. Argyropoulos, Structural analysis of wheat straw lignin by quantitative 31P and 2D NMR spectroscopy. The occurrence of ester bonds and α-O-4 substructures, J. Agric. Food Chem. 45 (1997) 1212–1219. [32] G. Jacquet, B. Pollet, C. Lapierre, New ether-linked ferulic acid-coniferyl alcohol dimers identified in grass straws, J. Agric. Food Chem. 43 (1995) 2746–2751. [33] J. Ralph, R.D. HatTield, S. Quideau, R.F. Helm, J.H. Grabber, H.G. Jug, Pathway of p-coumaric acid incorporation into maize lignin as revealed by NMR, J. Am. Chem. Soc. 116 (1994) 9448–9456. [34] J. Ralph, T. Akiyama, H.D. Coleman, S.D. Mansfield, Effects on lignin structure of coumarate 3-hydroxylase downregulation in poplar, Bioenergy Res. 5 (2012) 1009–1019. [35] K. Sarkanen, C. Schuerch, Lignin structure. XI. A quantitative study of the alcoholysis of lignin, J. Am. Chem. Soc. 79 (1957) 4203–4209. [36] W. Deng, M. Liu, Q. Zhang, Y. Wang, Direct transformation of cellulose into methyl and ethyl glucosides in methanol and ethanol media catalyzed by heteropolyacids, Catal. Today 164 (2011) 461–466. [37] F. Yu, R. Zhong, H. Chong, M. Smet, W. Dehaen, B.F. Sels, Fast catalytic conversion of recalcitrant cellulose into alkyl levulinates and levulinic acid in the presence of soluble and recoverable sulfonated hyperbranched poly(arylene oxindole)s, Green Chem. 19 (2017) 153–163. [38] C. Chang, G. Xu, X. Jiang, Production of ethyl levulinate by direct conversion of wheat straw in ethanol media, Bioresour. Technol. 121 (2012) 93–99. [39] Y. Zhang, X. Wang, T. Hou, H. Liu, L. Han, W. Xiao, Efficient microwave-assisted production of biofuel ethyl levulinate from corn stover in ethanol medium, J. Energy Chem. 27 (2018) 890–897. [40] C. Chang, G. Xu, W. Zhu, J. Bai, S. Fang, One-pot production of a liquid biofuel candidate—Ethyl levulinate from glucose and furfural residues using a combination of extremely low sulfuric acid and zeolite USY, Fuel 140 (2015) 365–370. [41] A.S. Mamman, J. Lee, Y. Kim, I.T. Hwang, N. Park, Y.K. Hwang, J. Chang, J. Hwang, Furfural: Hemicellulose/xylose- derived biochemical, Biofuels Bioprod. Biorefin. 2 (2008) 438–454. [42] R. Weingarten, G.A. Tompsett, W.C. Conner Jr., G.W. Huber, Design of solid acid catalysts for aqueous-phase dehydration of carbohydrates: the role of Lewis and Brønsted acid sites, J. Catal. 279 (2011) 174–182. [43] A. Barakat, F. Monlau, A. Solhy, H. Carrere, Mechanical dissociation and fragmentation of lignocellulosic biomass: effect of initial moisture, biochemical and structural proprieties on energy requirement, Appl. Energy 142 (2015) 240–246. [44] R. El Hage, N. Brosse, P. Sannigrahi, A. Ragauskas, Effects of process severity on the chemical structure of Miscanthus ethanol organosolv lignin, Polym. Degrad. Stab. 95 (2010) 997–1003. [45] R. El Hage, L. Chrusciel, L. Desharnais, N. Brosse, Effect of autohydrolysis of Miscanthus x giganteus on lignin structure and organosolv delignification, Bioresour. Technol. 101 (2010) 9321–9329. [46] X. Ouyang, X. Huang, B.M.S. Hendriks, M.D. Boot, E.J.M. Hensen, Coupling organosolv fractionation and reductive depolymerization of woody biomass in a twostep catalytic process, Green Chem. 20 (2018) 2308–2319. [47] G. Cheng, P. Varanasi, C. Li, H. Liu, Y.B. Melnichenko, B.A. Simmons, M.S. Kent, S. Singh, Transition of cellulose crystalline structure and surface morphology of biomass as a function of ionic liquid pretreatment and its relation to enzymatic hydrolysis, Biomacromolecules 12 (2011) 933–941. [48] A. Demirbas, Effect of lignin content on aqueous liquefaction products of biomass, Energy Convers. Manag. 41 (2000) 1601–1607. [49] Y.P. Zhang, L.R. Lynd, Toward an Aggregated Understanding of Enzymatic Hydrolysis of Cellulose: Noncomplexed Cellulase Systems, 88 (2004), pp. 797–824.
Notes The authors declare no competing financial interest. Abbreviations EL CrI LGO BSO
ethyl levulinate cellulose crystallinity index levoglucosenone N,O-bis(trimethylsilyl)acetamide
Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.fuproc.2018.08.003. References [1] G.W. Huber, S. Iborra, A. Corma, Synthesis of transportation fuels from biomass: chemistry, catalysts, and engineering, Chem. Rev. 106 (2006) 4044–4098. [2] E.L. Kunkes, D.A. Simonetti, R.M. West, J.C. Serrano-Ruiz, C.A. Gärtner, J.A. Dumesic, Catalytic conversion of biomass to monofunctional hydrocarbons and targeted liquid-fuel classes, Science 322 (2008) 417–421. [3] X.W. Zou, T.F. Qin, Y. Wang, L.H. Huang, Mechanisms and product specialties of the alcoholysis processes of poplar components, Energy Fuel 25 (2011) 3786–3792. [4] Z. Liu, F. Zhang, Effects of various solvents on the liquefaction of biomass to produce fuels and chemical feedstocks, Energy Convers. Manag. 49 (2008) 3498–3504. [5] H.J. Huang, X.Z. Yuan, Recent progress in the direct liquefaction of typical biomass, Prog. Energy Combust. Sci. 49 (2015) 59–80. [6] B.B. Hallac, Y. Pu, A.J. Ragauskas, Chemical transformations of Buddleja davidii lignin during ethanol organosolv pretreatment, Biomass Bioenergy 24 (2010) 2723–2732. [7] S. Branda, J. Kim, Liquefaction of major lignocellulosic biomass constituents in supercritical ethanol, Energy 80 (2015) 64–74. [8] S. Kang, X. Li, J. Fan, J. Chang, Hydrothermal conversion of lignin: a review, Renew. Sust. Energ. Rev. 27 (2013) 546–558. [9] H. Pan, Synthesis of polymers from organic solvent liquefied biomass: a review, Renew. Sust. Energ. Rev. 15 (2011) 3454–3463. [10] R. Van Putten, J.C. van der Waal, E. de Jong, C.B. Rasrendra, H.J. Heeres, J.G. de Vries, Hydroxymethylfurfural, a versatile platform chemical made from renewable resources, Chem. Rev. 113 (2013) 1499–1597. [11] R.J.H. Grisel, J.C. van der Waal, E. de Jong, W.J.J. Huijgen, Acid catalysed alcoholysis of wheat straw: Towards second generation furan-derivatives, Catal. Today 223 (2014) 3–10. [12] L. Peng, L. Lin, H. Li, Q. Yang, Conversion of carbohydrates biomass into levulinate esters using heterogeneous catalysts, Appl. Energy 88 (2011) 4590–4596. [13] P. Duan, B. Jin, Y. Xu, Y. Yang, X. Bai, F. Wang, L. Zhang, J. Miao, Thermo-chemical conversion of Chlorella pyrenoidosa to liquid biofuels, Bioresour. Technol. 133 (2013) 197–205. [14] J. Feng, J. Jiang, J. Xu, Z. Yang, K. Wang, Q. Guan, S. Chen, Preparation of methyl levulinate from fractionation of direct liquefied bamboo biomass, Appl. Energy 154 (2015) 520–527. [15] H. Huang, S. Ramaswamy, W. Al-Dajani, U. Tschirner, R. Cairncross, Effect of biomass species and plant size on cellulosic ethanol: a comparative process and economic analysis, Biomass Bioenergy 33 (2009) 234–246. [16] R. Le Van Mao, Q. Zhao, G. Dima, D. Petraccone, New process for the acid-catalyzed conversion of cellulosic biomass (AC3B) into alkyl levulinates and other esters using a unique one-pot system of reaction and product extraction, Catal. Lett. 141 (2011) 271–276. [17] X. Meng, Y. Pu, C.G. Yoo, M. Li, G. Bali, An in-depth understanding of biomass recalcitrance using natural poplar variants as the feedstock, ChemSusChem 10 (2017) 139–150. [18] W. Xiao, X. Chen, Y. Zhang, T. Qu, L. Han, Product analysis for microwave-assisted methanolysis of lignocellulose, Energy Fuel 30 (2016) 8246–8251. [19] C.O. Kappe, D. Dallinger, Controlled microwave heating in modern organic synthesis: highlights from the 2004-2008 literature, Mol. Divers. 13 (2009) 71–193. [20] E.S. Kim, S. Liu, M.M. Abu-Omar, N.S. Mosier, Selective conversion of biomass hemicellulose to furfural using maleic acid with microwave heating, Energy Fuel 26 (2012) 1298–1304. [21] A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter, D. Templeton, D. Crocker, Determination of Structural Carbohydrates and Lignin in Biomass, NREL/TP-510-
22