Impact of Alkaline Pretreatment Condition on Enzymatic Hydrolysis of Sugarcane Bagasse and Pretreatment Cost

A combined severity factor (RCSF) which is usually used to evaluate the effectiveness of hydrothermal pretreatment at above 100 °C had been developed to assess the influence of temperature, time, and alkali loading on pretreatment and enzymatic hydrolysis of lignocellulose. It is not suitable for evaluating alkaline pretreatment effectiveness at lower than 100 °C. According to the reported deducing process, this study modified the expression of RCSF=logCOH−n×t×eTr−Tb/14.75\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ {R}_{\mathrm{CSF}}=\log \left[{C}_{OH^{-}}^n\times t\times {e}^{\left({T}_r-{T}_b\right)/14.75}\right] $$\end{document} as RCSF=logCOH−×t×e−13700/Tr+273+36.2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ {R}_{\mathrm{CSF}}=\log \left\{{C}_{OH^{-}}\times t\times {e}^{\left[-13700/\left( Tr+273\right)+36.2\right]}\right\} $$\end{document} which is easier and more reasonable to assess the effectiveness of alkaline pretreatment. It showed that RCSF exhibited linear trend with lignin removal, and quadratic curve relation with enzymatic hydrolysis efficiency (EHE) at the same temperature. The EHE of alkali-treated SCB could attain the maximum value at lower RCSF, which indicated that it was not necessary to continuously enhance strength of alkaline pretreatment for improving EHE. Within a certain temperature range, the alkali loading was more important than temperature and time to influence pretreatment effectiveness and EHE. Furthermore, the contribution of temperature, time, and alkali loading to pretreatment cost which was seldom concerned was investigated in this work. The alkali loading contributed more than 70% to the pretreatment cost. This study laid the foundation of further optimizing alkaline pretreatment to reduce cost for its practical application.


Introduction
Biorefinery of lignocellulosic biomass to biofuels and bioproducts has gained more and more attention due to its sustainability and low carbon discharge [1,2]. Lignocellulose is mainly composed of cellulose, hemicellulose, and lignin which entangle with each other to endow the lignocellulose with recalcitrance towards biodegradation [3,4]. An ideal pretreatment is an indispensable step to break down the compact structure of lignocellulose for the following steps of enzymatic saccharification and microbial conversion [3,5].
Various physical, chemical, and biological tools have been adopted to pretreat lignocellulose like microwave, ultrasound, inorganic and organic solvents, microorganism, and so on. Pretreatment technologies including liquid hot water, diluted acid solution, steam explosion, and diluted alkaline solution have been employed for pilot-scale biorefinery plant of lignocellulose. Different from other three pretreatment technologies which are conducted under the condition of above 100°C and saturated steam pressure to mainly degrade hemicellulose [6], the alkaline pretreatment is more promising for industrial application due to its moderate condition which can be realized at lower temperatures than 100°C without being pressurized and high lignin removal efficiency [7][8][9]. NaOH is a common alkaline reagent to be used for pretreating lignocellulose due to its lower cost than potassium hydroxide (KOH) and higher solubility than calcium hydroxide (Ca(OH) 2 ). However, the alkali-treated lignocellulose is often washed for several times to remove inhibitors and alkalis for guaranteeing high bioconversion efficiency, which would consume plenty of water and generate a large quantity of wastewater. The technologies including wastewater recycling [10] and direct bioconversion without washing step [11,12] had been developed to overcome the obvious shortcomings of alkaline pretreatment process. Additionally, the attempt of recycling liquid waste from alkaline pretreatment as fertilizer [13] or something else would make alkaline pretreatment more feasible for actual application. Furthermore, to find out the main factors influencing pretreatment cost is another meaningful work for selecting proper and economical alkaline pretreatment condition for practical utilization. Based on the recycling technology of liquid wastes [13], the alkaline pretreatment cost investigated in this work is highly dependent on energy consumption (relating to temperature and time), water consumption, and alkali loading. Seldom studies reported the factors influencing alkaline pretreatment cost with experimental tests.
As is known, the effectiveness of alkaline pretreatment is chiefly related to temperature, time, and alkali loading. The combined severity factor (R CSF ) which is modified from hydrothermal pretreatment [14,15] had been used to describe the combined effects of temperature, time, and alkali loading on alkaline pretreatment and enzymatic hydrolysis [16,17].
where R CSF is the combined severity factor, C is the concentration of hydroxyl ion (mol/L), t is the reaction time (min), T r is the temperature maintained during the reaction time (°C), T b is the base temperature (°C) which is acknowledged as 100°C, and n is an arbitrary constant.  (1) is deduced under the assumption that the reaction obeys to the first-order kinetics and the rate constant has relation to temperature via Arrhenius equation [14,15,18]. Therefore, in this study, the expression of (T r − 100)/14.75 is replaced by −13700/(T r + 273)+36.2 for alkaline pretreatment conducted below 100°C. The impact of alkaline pretreatment condition on lignin removal and enzymatic hydrolysis was switched to discuss the correlations between R CSF and lignin removal, and R CSF and enzymatic hydrolysis. The contribution of pretreatment condition on pretreatment cost was investigated in this study. The quota of temperature and time to pretreatment cost was reflected as energy consumption. At the similar enzymatic hydrolysis efficiency, the alkali loading shared more cost.

Materials
Sugarcane bagasse (SCB) used in this experiment was collected from Guangxi Zhuang Autonomous Region (China). After being washed, it was dried to a constant weight in an oven at 105°C and then screened with 40-60 mesh sifter. After componential detection with the laboratory analytical procedure (NREL/TP-510-42618) [19], the cellulose, hemicellulose, and lignin contents of SCB were 43.98 ± 0.57%, 23.62 ± 0.36%, and 23.61 ± 3.84%. The sodium hydroxide (NaOH) was bought from Shanghai Macklin Biochemical Co., Ltd. (China). The cellulase was purchased from Imperial Jade Bio-Technology Co., Ltd. (China). The filter paper activity of the cellulase was 197.5 FPU/g of powder, which was tested by Ghose's method [20].

Alkali Pretreatment
The alkaline pretreatment was performed as the following procedure to evaluate the influence of temperature, time, and NaOH loading on pretreatment effectiveness and cost. SCB was respectively added in 150-mL Erlenmeyer flasks containing 1%, 2%, 3%, and 4% (w/v) NaOH solution to form solid-to-liquid ratio of 1:10. The flasks were placed in a water bath shaker with reciprocating rotation speed of 150 rpm at 60°C, 70°C, and 80°C for 1, 2, 3, and 4 h, respectively. The electrical consumption of each pretreatment was recorded by metering socket. The treated SCB was washed with water to neutral pH and then oven-dried at 50°C to a constant weight.

Enzymatic Hydrolysis
The NaOH-treated SCB was added into 25-mL Erlenmeyer flask containing 5 mL, 0.05 M acetate buffer (pH 4.8), to achieve solid-to-liquid ratio of 1:10. The cellulase was loaded with 20 FPU/g cellulose. In order to prevent water loss, the flasks were sealed with glass plugs and Parafilm in the hydrolysis process. The enzymatic hydrolysis was conducted at 50°C, 150 rpm for 72 h. Meanwhile, the electricity consumption during hydrolysis was also recorded by metering socket. And the enzymatic hydrolysis efficiency (EHE) was calculated by the followed formula: where c glucose and c xylose (mg/mL) are the concentrations of glucose and xylose in the enzymatic hydrolysate; v (mL) is the volume of enzymatic hydrolysate; 0.9 and 0.88 are the dehydration coefficients of the synthesis of glucose to glucan and xylose to xylan, respectively; and m glucan and m xylan (mg) are the mass of glucan and xylan in NaOH-treated SCB, respectively.

Calculation of the Combined Severity Factor
The combined severity factor (CSF) which describes the combined effect of temperature, time, and alkaline concentration was analogized from the equation describing the combined effect of temperature, time, and acid concentration for hydrothermal pretreatment [18,21]. It was modified as Eq. (1) to reflect the strength of alkaline pretreatment. The arbitrary constant n in Eq. (1) is used to make the value of R CSF positive. Since R CSF is employed to exhibit the strength of pretreatment, it is not necessary to be positive. Based on the above description, it is finally modified as the following Eq. (3) with n = 1.
where R CSF , t, and T r are the same as Eq. (1), and C OH − is the concentration of hydroxyl ion (mol/L).

Results and Discussion
Correlation of the Combined Severity Factor with NaOH Pretreatment SCB was treated at 60, 70, and 80°C with different NaOH concentrations for different times, respectively. The compositional changes of SCB and R CSF values after NaOH pretreatment were shown in Table 1. Under the condition of same temperature and NaOH loading, the relative contents of glucan (namely, cellulose) and lignin tended to slightly increase and decrease, respectively. The relative xylan (namely, hemicellulose) content changed little and then decreased slightly with NaOH loading surpassed 2% (w/v). Under the condition of same temperature and time, the relative glucan content obviously showed increasing tendency, while the relative contents of xylan and lignin got the reverse trend with NaOH dosage increasing. Under the condition of same time and NaOH loading, the temperature could weakly influence the amount changes of glucan and xylan but exhibit obvious impact on lignin removal. Generally, lignin removal is used as an indicator for evaluating the effectiveness of NaOH pretreatment under different conditions due to the chief action of alkaline pretreatment on lignin degradation [7,22]. The numeric size of R CSF value reflects the pretreatment strength which is resulted by temperature, time, and alkali concentration during pretreatment. The same R CSF values appeared for NaOH pretreatment under the same temperature with different NaOH concentrations for different times (Table 1), which is in line with the previous reports [16,17]. The relationship between R CSF and lignin removal was shown in Fig. 1. The maximum lignin removal of 77.94% reached at R CSF = 1.122 (80°C, 3% NaOH for 4 h) has little difference with that of 77.40% obtained at 180°C, 10% NaOH/ethanol for 30 min [23]. R CSF tended to be linear correlation with the lignin removal, which is corresponding to the previous report [17]. The NaOH concentration was more important for lignin removal than pretreatment time at the same R CSF such as sequential increment of lignin removal at 60°C, 1% NaOH for 4 h, 60°C, 2% NaOH for 2 h, and 60°C, 4% NaOH for 1 h (Table 1 and Fig. 1). The close R CSF values obtained under different temperatures; different NaOH loadings for different times showed that NaOH loading was at the first place to influence lignin removal, followed by temperature (Table 1 and Fig. 1). Additionally, the lignin removal has little difference when the pretreatment was conducted at 80°C with over 2% NaOH for more than 2 h. It indicated that lignin removal would not be improved further along with the intensification of alkaline pretreatment when it attained a limiting value.

Correlation of the Combined Severity Factor with Enzymatic Hydrolysis
SCB treated under different conditions was enzymatically hydrolyzed at 50°C for 72 h with addition of 20 FPU cellulase/g cellulose. From Fig. 2, the maximum EHE of SCB treated at 70°C , 4% NaOH for 4 h, is 73.41% which is almost the same as those of SCB treated at 140~160°C , 5% NaOH mixed with 60% ethanol for 30 min [24]. EHEs of SCB treated at same temperature for the same time were improved obviously with NaOH concentration increasing from 1 to 2%, while had little difference when NaOH concentration was in the range from 2 to 4%. Meanwhile, EHEs of SCB treated with over 2% NaOH for the same time were changed little with pretreatment temperature increasing from 60 to 80°C. When NaOH loading achieved to 3%, the EHEs of NaOH-treated SCB had little differences no matter how different the pretreatment time and temperature were. It could be also found that EHEs of SCB treated with 2% NaOH at 80°C for 2 h, 3 h, and 4 h had the maximum values around 70%, as well as that of SCB treated with 3% NaOH at 80°C for 1 h (Fig. 2). As the pretreatment time prolonged for over 2 h and NaOH loading increased more than 2%, EHEs of SCB treated at different temperatures were close to the maximum value. It meant that the optimum enzymatic hydrolysis of SCB treated with NaOH solution could be attained at lower energy consumption which is related to pretreatment temperature and time. The quadratic correlation between R CSF and EHE also showed that EHE had maximum value and would not be improved with R CSF Fig. 1 Correlation between the combined severity factor (R CSF ) and lignin removal; Exp is short for experimental data, Fitted is short for fitted data, and green line fits the whole data obtained under all conditions increasing (Fig. 3). The EHE closed to the maximum value could be obtained at 60°C, 3% NaOH for 3 h, whose R CSF was only higher than three conditions at 70°C, while lower than all conditions at 80°C. It hinted that a milder condition of alkaline pretreatment would realize maximum EHE of alkali-treated lignocellulose to reduce pretreatment cost.

Contribution of Temperature, Time, and Alkali on Pretreatment Cost
Under the same solid concentration and stirring speed, energy consumption and alkali loading are two main factors influencing pretreatment cost. Based on the previously constructed process that direct enzymatic hydrolysis of alkali-treated SCB without washing step [12], the three parameters-temperature, time, and alkali loading-are involved in the combined severity factor to constitute the elements of alkaline pretreatment cost. The pretreatment conditions which made EHE of treated SCB 5% less than the maximum EHE of SCB treated with 4% NaOH at 70°C for 4 h were selected to compare the cost. Through monitoring electricity, the costs of pretreating 1.6 kg SCB with solid-liquid ratio of 1:10 were estimated ( Table 2). It could be found that the lowest pretreatment cost was attained under the condition of 2% NaOH at 70°C for 3 h. Pretreatment conducted with more alkali loading at higher temperature for longer time got the higher cost, while the EHEs of treated SCB had little difference. The alkali consumption contributed a large portion of cost which ranged from 71.40 to 93.34%. The pretreatment cost under the condition of 3% NaOH at 60°C for 3 h is

Conclusion
The combined severity factor (R CSF ) inclines to present linear correlation with lignin removal and quadratic correlation with enzymatic hydrolysis efficiency. The alkali loading plays a more important role than temperature and time for lignin removal with the same and similar R CSF . The optimum enzymatic hydrolysis efficiency could be obtained for milder alkali-treated sugarcane bagasse. The alkali loading has a larger contribution on pretreatment cost than energy consumption. The temperature acts a more obvious role than time for pretreatment cost under the same alkali loading.  Availability of Data and Material The data and material were obtained from experimental tests in duplicate and described clearly in the text.

Declarations
Ethical Approval Not applicable

Consent to Publish Not applicable
Conflict of Interest The authors declare that they have no conflict of interest.