Background

Lignocellulosic biomass represents the abundant and renewable feedstock for the production of bio-based chemicals and fuels under the concept of sustainable biorefinery [1]. Poplar is proposed as a very promising biomass for biorefinery, owning to its relatively high content of carbohydrate and rapid growth rate. The bio-based ethanol production process from lignocellulosic biomass like poplar comprises pretreatment, enzymatic hydrolysis, fermentation and distillation, in which the pretreatment plays a key role in improving the ease of subsequent enzymatic hydrolysis and promoting the overall biomass utilization [45].

Results showed that acid-catalyzed BDO organosolv pretreatment was more effective in improving biomass saccharification (Fig. 1) and producing antioxidant lignin (Table 4) when comparing with alkali-catalyzed BDO pretreatment. Then, mass balance analysis based on acid-catalyzed BDO pretreatment was proposed (Fig. 6). As illustrated, after HCl-BDO pretreatment with HCl loading of 40 mM, the pretreatment hydrolysate containing C5 sugars and lignin was separated from pretreated solid, followed by precipitation. As a result, 19.42 g organosolv lignin was recovered from 100 g raw poplar biomass, which could be used as antioxidant (Table 3). The liquid stream containing 1.52 g cellulose-derived sugars and 8.17 g hemicellulose-derived sugars could be upgraded to furanic platform chemicals [17], like furfural. After acid hydrolysis (1.2 wt% sulphuric acid, 180 °C, 1 h) of the liquid stream, 3.58 g/L furfural could be produced from 11.68 g xylose (detail not shown). BDO solvent was proposed to be well separated from water using Mitsubishi SP70, a kind of macroporous adsorption resin (Mitsubishi Chemical Corporation, Japan) [18]. After separation, BDO could be reused in the pretreatment of lignocellulosic biomass [19], and water could be reused to wash the pretreated solid, reducing the possible inhibition of enzymatic hydrolysis and subsequent fermentation by BDO and degradation products from carbohydrate and lignin. Besides, 39.03 g glucan and 5.51 g xylan in the raw biomass could be converted to fermentable sugars after pretreatment and subsequent enzymatic hydrolysis. The enzymatic hydrolysate was concentrated to different initial glucose concentrations to estimate the fermentability of the enzymatic hydrolysate (Fig. 6). It was shown that 142.51 g/L glucose could be completely consumed at 24 h fermentation. And 68.12 g/L ethanol was produced, with ethanol yield of 93.73%, which verified the fermentability of enzymatic hydrolysate of HCl-BDO pretreated solid. Results suggested that acid-catalyzed BDO pretreatment, which maximized total sugar yield while enabling efficient production of antioxidant lignin from the poplar wood sawdust, had the potential to be a promising pretreatment approach of lignocellulosic biomass, since it not only diversified the bio-based products from biomass, but also encouraged the utilization of lignin as part of sustainable lignocellulosic biorefinery.

Fig. 6
figure 6

Mass balance of biorefinery based on acid-catalyzed BDO organosolv pretreatment to coproduce fermentable sugars and antioxidant lignin (a),  fermentation at different initial glucose concentrations (b), concentrations of enzymatic hydrolysis products (c)

Full size image

Conclusions

Results showed that BDO organosolv pretreatment was more effective in removing lignin from lignocellulosic biomass as compared to ethanol organosolv pretreatment. Besides, acid-catalyzed BDO pretreatment had better performance in reducing recalcitrance of poplar to achieve reasonable biomass saccharification through enzymatic hydrolysis than alkali-catalyzed BDO pretreatment, which was accompanied with greater cellulose accessibility resulted from higher degree of delignification and hemicellulose solubilization, as well as the more increase in fiber swelling and cellulose microfibril. The maximum sugar yield of 79.41% from original biomass was achieved after HCl-BDO pretreatment (170 °C, 1 h, with 40 mM HCl loading). Moreover, acid BDO pretreatment resulted in the formation of phenolic hydroxyl groups in lignin, which increased radical scavenging capacity of BDO organosolv lignin as a natural antioxidant.

Materials and methods

Materials

Poplar sawdust, with a moisture content of 7.33 ± 0.21%, was collected from Xuzhou, Jiangsu Province, China. Chemicals including ethanol, 1,4-butanediol (1,4-BDO, AR, > 99% purity), hydrochloric acid (HCl, 95–98 wt%) and sodium hydroxide (NaOH, AR, > 96% purity) were obtained from Sinopharm Chemical Reagent Co., Ltd. Commercial enzyme blend CTec2 (SAE0020, filter paper activity of 185 FPU/g, protein content of 233 mg/g, endoglucanse, exoglucanase, β-glucosidase, xylanase activity was 2297.8, 114.9, 3451.6 and 8902.9 U/g, respectively) was obtained from Sigma–Aldrich.

Acid- and alkali-catalyzed BDO pretreatments

Acid-catalyzed ethanol organosolv pretreatment (HCl-ethanol) was performed as follows: 100 g dry mass poplar was soaked in an aqueous solution (ethanol–water ratio of 65:35, v/v) containing 30, 40 and 50 mM HCl at a solid-to-liquid ratio of 1:7 (g: mL). The mixture was pretreated at 170 °C for 60 min.

Acid-catalyzed BDO organosolv pretreatment (HCl-BDO) was performed as follows: 100 g dry mass of poplar was soaked in an aqueous solution (BDO-water ratio of 65:35, v/v) containing 10, 20, 30, 40 and 50 mM HCl at a solid-to-liquid ratio of 1:7 (g: mL). The mixture was pretreated at 170 °C for 60 min [13]. The oil bath (GSC-30L, Yushen Instruments Company, China) was heated to 170 °C during 30 min and maintained at the temperature for 60 min. Alkali-catalyzed BDO pretreatment (NaOH-BDO) was carried out under the same conditions but using an aqueous solution containing 250, 300, 350, 400 and 450 mM NaOH. After pretreatment, the reactor was cooled in tap water. Solid fraction was separated from pretreatment liquor through vacuum filtration, then washed by 2100 mL water. The washed solid, as water-insoluble fraction (WIF), was kept at 4 °C for further use. For HCl-BDO pretreatment, the pretreatment liquor and washing water were combined, as water-soluble fraction (WSF), for lignin recovery by precipitation and determination of the sugar concentration. For NaOH-BDO pretreatment, H2SO4 was added to pretreatment liquor and washing water to lower the pH to around 2.0 for lignin precipitation. After precipitation, the solid was separated by centrifugation, then rinsed by hot water and freeze-dried for 72 h to recover lignin.

$$\text{Lignin yield} (\%)=\frac{\mathrm{Lignin\,recovered\,through\,precipitation }\,(\mathrm{g})}{\mathrm{Lignin\,in\,raw\,biomass }\,(\mathrm{g})}$$
(1)

Enzymatic hydrolysis of the pretreated substrates

Enzymatic hydrolysis was carried out on the BDO pretreated and washed solid at 50 °C, pH 4.8, 180 rpm for 72 h, in air shaker. In enzymatic hydrolysis, cellulose loading was 2% (w/v), and enzyme loading was 20 FPU cellulase/g cellulose. After 72 h enzymatic hydrolysis, flasks were taken out of the air shaker. Samples were taken from enzymatic hydrolysate. Enzymes were inactivated by heating to 100 °C for 5 min and subsequently stored at − 4 °C until sugar analysis was performed by HPLC. All experiments were performed in duplicate.

$$\text{Glucan hydrolysis yield} (\%)=\frac{\mathrm{Glucose\,in\,enzymatic\,hydrolysate }\,\left(\mathrm{g}\right)\times 0.9}{\mathrm{Glucan\,loading\,for\,enzymatic\,hydrolysis }\,(\mathrm{g})}\times 100$$
(2)
$$\text{Xylan hydrolysis yield} (\%)=\frac{\mathrm{Xylose\,in\,enzymatic\,hydrolysate }\,\left(\mathrm{g}\right)\times 0.88}{\mathrm{Xylan\,loading\,for\,enzymatic\,hydrolysis }\,\left(\mathrm{g}\right)} \times 100$$
(3)
$$\text{Total sugar yield} (\%)=\frac{\mathrm{Sugars\,released\,in\,pretreatment\,and\,enzymatic\,hydrolysis }\,(\mathrm{g})}{\mathrm{Cellulose\,and\,hemicellulose\,in\,raw\,biomass }\,(\mathrm{g})} \times 100$$
(4)

Antioxidant activity of lignin

Antioxidant activity of lignin was measured as radical scavenging activity using 2, 2-diphenyl-1-picrylhydrazyl (DPPH) method. The recovered lignin was dissolved in dioxane/water solution (9/1, v/v), with concentration ranging from 40 mg/L to 200 mg/L. Of the lignin solution, 0.1 mL was added to 3.9 mL of DPPH ethanolic solution (25 mg/L). The mixture was kept at 25 °C for 30 min. Absorbance of the solutions was measured at 517 nm. Inhibition percentage (IP) was calculated [46] and plotted as a function of lignin concentration, in which EC50 (lignin concentration needed to obtain 50% IP) was obtained. Radical scavenging index (RSI), as the inverse of EC50, was used to evaluate the antioxidant activity of lignin. Higher RSI indicated better antioxidant activity of lignin.

Fermentation

After enzymatic hydrolysis of HCl (40 Mm)-BDO pretreatment and enzymatic hydrolysis, the enzymatic hydrolysate was concentrated to achieve different initial glucose concentrations of 75 g/L (Low concentration, L), 110 g/L (Medium concentration, M) and 145 g/L (High concentration, H) for fermentation by Saccharomyces cerevisiae to produce bioethanol [47]. Glucose fermentation was performed at 30 °C, 100 rpm, pH 5.5 with cell density of OD600nm = 4, for 24 h, The fermentation experiments were done in duplicate.

$$\text{Ethanol yield} (\%)=\frac{\text{Weight}\, \text{of}\,\text{produced}\,\text{ethanol}}{0.51\times\text{Weight} \,\text{of}\, \text{glucose}\,\text{in}\, \text{fermentation}} \times 100$$
(5)

Analytical methods

Chemical components of biomass samples were analyzed by following the method developed by the US National Renewable Energy Laboratory [48]. The water-soluble fraction (WSF) was subjected to an acid hydrolysis (4% H2SO4, 121 °C for 1 h), and the sugars (glucose, xylose, arabinose) in the liquid fraction were determined for mass balance analysis. The sugars and ethanol concentration was determined using a high performance liquid chromatography (HPLC) system (Agilent 1100) with a refractive index (RI) detector. The separation was performed on Bio-rad Aminex HPX-87H column (300 × 7.8 mm) with 5 mM H2SO4 as the eluent at a flow rate of 0.6 mL/min.

Water retention value (WRV) measurement was performed according to TAPPI UM 256 for evaluation of biomass fiber swelling [49]. To assess cellulose accessibility, staining method by DR28 was carried out as described elsewhere [50]. The crystallinity of biomass samples was measured using an Ultim IV X-ray diffractometer (XRD) equipped with a Cu Kα radiation source (λ = 0.15406 nm), which was scanned at the range of 2θ = 5° − 50° with a rate of 5°/min. Crystallinity (CrI) was calculated as described before [51]. CrI/cellulose was defined as the ratio of the calculated CrI to the cellulose content of biomass. CrI might represent the total crystallinity in biomass rather than the cellulose itself, and the CrI/cellulose ratio was suggested as an appropriate mean to estimate true crystallinity in native cellulose [49].

Scanning electron microscope (SEM) was used to observe surface morphology of untreated and BDO pretreated samples with different catalyst loadings at magnification of 1 K. The chemical structure of recovered lignin from BDO pretreatment was determined by attenuated total reflection Fourier transform infrared spectra (ATR-FTIR, Spectrum Two, PerkinElmer, US). Spectra of each biomass sample ranged from 500 to 4000 cm−1 at a spectral resolution of 4 cm−1 with an average from 64 scans. Gel permeation chromatography (GPC, Waters 1525 system, US) equipped with Agilent PL-gel MIXED-C column and Waters 2414 refractive index (RI) detector was used to determine weight-average (Mw) and number-average (Mn) molecular weights of recovered lignin. Polydispersity index (PDI) was calculated as Mw/Mn. Tetrahydrofuran (THF) was used as the mobile phase at a flow rate of 1.0 mL/min. Polystyrene narrow standards were used as calibration standards [53].