One-pot pretreatment, saccharification and ethanol fermentation of lignocellulose based on acid–base mixture pretreatment

Abstract

Currently, for the production of cellulosic ethanol, multi-step unit operations, including pretreatment, solid/liquid (S/L) separation, solids washing, liquid detoxification, neutralization, enzymatic hydrolysis and fermentation, are the commonly required steps responsible for elevating the capital and operating costs. To simplify these steps, consolidated bioprocessing (CBP), focusing on the multi-functional microbial strains, was proposed. However, this process has not been commercialized yet. In this study, using an acid–base mixture as a pretreatment catalyst, pretreatment, saccharification and fermentation were performed in one pot without S/L separation, neutralization and detoxification. From the one-pot process based on the acid–base mixture pretreatment (190 °C, 2 min and 0.15 (w/v) acid–base mixture) and 15 FPU of cellulase per g glucan and Sacchromyces cerevisiae, 70.7% of the theoretical maximum ethanol yield (based on the initial amount of glucan in the untreated rice straw) was obtained. This was comparable to the estimated ethanol yield of 72.9%, assuming a 90% glucan recovery yield after pretreatment × a 90% glucose yield from saccharification × a 90% ethanol yield from ethanol fermentation performed in three separate pots. These results suggest that the entire slurry processing of lignocellulose in one pot could be an attractive way to achieve economic sustainability in the production of fuel from lignocellulose.

---

1 Introduction

The importance of biofuel production from lignocellulose is gaining prominence because of the associated environmental benefits and the concern for high dependence on petroleum.¹ The high recalcitrance of lignocellulose needs to be alleviated by using appropriate physicochemical pretreatment to increase the enzymatic digestibility of cellulose present in lignocellulose.^2,3 Because the physicochemical pretreatment processes are usually performed at extreme pH and/or high temperatures, the generation of sugar degradation products such as 2-furaldehyde (furfural) and 5-hydroxymethyl-2-furaldehyde (HMF) is inevitable.^4–6 These pretreatment byproducts inhibit microorganisms involved in biofuel synthesis (e.g., ethanol fermentation by yeast).⁷

The process of biofuel production from lignocellulose, which is represented by cellulosic ethanol, involves pretreatment, solid/liquid (S/L) separation, solids washing, liquid detoxification, enzymatic hydrolysis and ethanol fermentation (Fig. 1A). In the process of cellulosic ethanol production, the unit operations for the detoxification or removal of inhibitors, such as S/L separation, solids washing and liquid detoxification (e.g. overliming followed by acidification, chromatographic separation of sugar, etc.), cause a significant increase in the operating costs.^8–10 To avoid these steps, the development of either inhibitor-tolerant yeast or a novel pretreatment process that does not produce inhibitors is required.

Fig. 1 Schematic diagrams showing the pretreatment, saccharification and fermentation processes for cellulosic ethanol production. (A) Conventional multi-unit configuration, (B) whole slurry fermentation configuration,^14,15 (C) one-pot pretreatment and saccharification configuration¹⁶ and (D) the novel “one-pot pretreatment, saccharification and fermentation processes”.

In an attempt to reduce the production cost of cellulosic ethanol, the method of consolidated bioprocessing (CBP) was suggested. This process combines cellulase production, cellulose hydrolysis and ethanol fermentation into a single step, using a genetically engineered microorganism simultaneously capable of producing cellulase and fermenting ethanol.¹¹ However, the potential of CBP has not yet been realized.¹² Recently, to simplify the process, “whole slurry fermentation” involving the fermentation of the entire pretreated lignocellulose slurry without S/L separation was demonstrated (Fig. 1B).^13–15 Even in this process, a detoxification step such as activated carbon treatment¹⁴ or pH adjustment¹⁵ is required to remove inhibitors generated during the acid pretreatment. In the same context, a “one-pot pretreatment and saccharification process”, which combines ionic liquid (IL) pretreatment and saccharification (using an IL-tolerant enzyme cocktail) into a single-unit, was presented (Fig. 1C); however, IL needs to be separated and recovered from the hydrolysate prior to fermentation. Furthermore, simultaneous saccharification and fermentation (SSF) is not possible in this process configuration.¹⁶

In this study, we developed an integrated pretreatment, saccharification and fermentation process in a single reactor using an acid–base mixture as the pretreatment catalyst. This process does not require S/L separation, neutralization and detoxification, either after pretreatment or before SSF (Fig. 1D). This simplified process may highly impact the lignocellulose-based biofuels and biorefinery industries owing to a substantial reduction in the operating costs.

2 Experimental

2.1 Lignocellulose and compositional analyses

Rice straw used in this study was harvested in Yeonggwang, Korea in 2011. Rice straw was washed with tap water, air-dried and milled using a cutting mill (MF 10, IKA; Staufen, Germany). Rice straw was then sieved to generate particle sizes of 90–1000 μm. Carbohydrates and acid-insoluble lignin in rice straw were analyzed following the Laboratory Analytical Procedure (LAP) of the National Renewable Energy Laboratory (NREL; Golden, CO).¹⁷ The sugars, pretreatment by-products and fermentation products in the liquid fraction, as well as total solids and ash contents, were also measured following the LAP of NREL.^18–20

2.2 Thermochemical pretreatment of lignocellulose

Ground dry rice straw (2 g) was soaked in 20 mL catalyst solutions comprising various mixing ratios and concentrations of acid–base mixtures in 100 mL vessels (SK-12 type; Milestone; Shelton, CT) equipped with a thermocouple. Pretreatment was performed by digesting the biomass and catalyst solution mixture in the vessels whilst ramping to 190 °C for 3 min and holding at 190 °C for 2 min, using a microwave digester (ETHOS EZ; Milestone). To prepare samples for the analyses of biomass compositions and enzymatic digestibilities of pretreated biomass, the insoluble solids were separated from the pretreated slurry by washing with 1 L of distilled water and filtering through a filter cloth (pore size of 22–25 μm; Calbiochem, La Jolla, CA) until the filtrate reached a pH ranging from 6 to 7. Some of the washed insoluble solids were transferred to aluminum dishes and placed in a vacuum drying oven at 45 °C for three days for analysing the composition of the solids. The other insoluble solids were stored at −20 °C for further experiments such as enzymatic hydrolysis and SSF. For the whole slurry fermentation of the pretreated biomass, the solid and liquid fractions from the pretreated slurry were directly subjected to SSF without an S/L separation. To quantify the sugar monomers, including glucose, xylose, galactose, arabinose and mannose, and to quantify ethanol and other byproducts, including furfural, HMF, acetic acid, formic acid and levulinic acid, Aminex HPX-87P (Bio-Rad; Hercules, CA) and Aminex HPX-87H (Bio-Rad) columns were used, respectively, for high pressure liquid chromatography (HPLC; Agilent Technologies, Waldbronn, Germany) as previously described.^14,15

2.3 Enzymatic hydrolysis of lignocellulose

To evaluate the effectiveness of the pretreatment, the untreated rice straw or the pretreated and washed rice straw was enzymatically hydrolyzed using 15 FPU g⁻¹ glucan derived from a commercial cellulase (Accellerase 1000; Genencor, Rochester, NY), following the LAP of NREL.²¹ In brief, a lignocellulosic biomass with 1% (w/v) of the final glucan concentration was added to 10 mL of 0.05 M sodium citrate buffer (pH 4.8) at 50 °C. The enzymatic digestibility was expressed as a percentage of the theoretical maximum glucose produced from the total amount of input glucan. An Aminex HPX-87P HPLC column was used to measure the amount of glucose produced from the enzymatic hydrolysis.

2.4 Simultaneous saccharification and fermentation

SSF was carried out to produce ethanol from untreated or pretreated rice straw following the LAP of NREL with a slight modification.²² Similar to the protocol followed for glucan and biomass loadings, a final glucan concentration of 3% (w/v) for the untreated or pretreated and washed rice straw, and a final biomass concentration of 6% (w/v), which was based on the inital untreated rice straw, for the whole slurry fermentation, were used in the SSF. After autoclaving the SSF media (1% yeast extract, 2% peptone and 0.05 M citrate buffer at pH 4.8) at 121 °C for 20 min, 15 or 60 FPU of cellulase (Accellerase 1000) g⁻¹ of glucan and 5% (v/v) of Saccharomyces cerevisiae D₅A (ATCC 200062)—grown in YPD media containing 1% (w/v) yeast extract, 2% (w/v) peptone and 2% (w/v) glucose—were added. Particularly for the whole slurry fermentation of the acid–base pre-treated rice straw, the SSF media components were added into the pretreated slurry after the acid–base pretreatment without any further operation. The SSF was conducted in a flask with a needle-pierced silicone stopper to vent the CO₂ produced during fermentation, in a shaking incubator for 60 h and maintained at 38 °C and 170 rpm. Ethanol yields were determined as a percentage of the theoretical maximum based on the glucan contained in the rice straw before the pretreatment.

2.5 Evaluation of cellulose accessibility to the enzyme

To analyze the cellulose accessibility of pretreated rice straw to cellulase, the binding capacity of the rice straw to a typical carbohydrate-binding module (CBM) of cellulase was quantified using a Type A surface binding CBM from Clostridium thermocellum (CtCBD3).^23–26 Recombinant CtCBD3 was prepared as previously described.^25,26 Bovine serum albumin (BSA; Sigma-Aldrich, St. Louis, MO) was used as a control protein in these binding experiments. For the binding analysis, 5 mg of the substrate (untreated or pretreated rice straw) was incubated with an excess amount (0.4 nM) of BSA or CtCBD3 in 500 μL of 0.05 M potassium phosphate buffer (pH 7). After 3.5 h, the binding mixture was separated into unbound protein in the supernatant, and bound protein in the pellet by centrifugation at 25 000 × g for 5 min. The amount of unbound protein was measured using the Bradford method.²⁷ The amount of bound protein was determined by subtracting the amount of unbound from the total protein initially added to the binding mixture.

3 Results and discussion

3.1 Effects of acid–base concentrations and mixture ratios on pretreatment

Previous studies have suggested pretreating lignocellulose by using sequential application of acid and base^28–30 or by using salt as a catalyst.^31–33 However, to the best of our knowledge, using a mixture of acid and base together has not been attempted for pretreating lignocellulose. When different mixing ratios of acid and base (i.e. HCl and NaOH of 1 : 4 to 4 : 1) were tested, the acidic region of the acid–base mixture gave the highest glucose yield (Fig. 2A). Other ratios of acid–base mixtures representing neutral or alkaline pH gave glucose yields lower than 15 g per 100 g of lignocellulose. The acid–base molar ratio of 4 : 1 was selected for all further pretreatment experiments.

Fig. 2 Effects of (A) different molar ratios of acid (HCl)–base (NaOH) mixtures (at the final total concentration of 0.05 M) and (B) total concentrations of the acid–base mixtures (with an acid : base molar ratio of 4 : 1) and their effect on the enzymatic digestibility of the pretreated and washed rice straw, on the basis of the total dry wt of the untreated rice straw (input). Pretreatment was conducted using various concentrations of acid–base mixture at 190 °C and with a solids loading of 10% (w/v), a 3 min ramping to 190 °C and a 2 min holding at 190 °C in a microwave digester. Enzymatic hydrolysis was conducted using 15 FPU of cellulase (Accellerase 1000) g⁻¹ glucan at 50 °C (pH 4.8) and at 200 rpm for 50 h.

Based on the selected optimal acid–base mixture ratio, as shown in Fig. 2A, various total concentrations of the acid–base mixture (0.01–1 M) were tested during the pretreatment of rice straw at 190 °C (Fig. 2B). In the case of pretreatment without any acid–base mixture, the enzymatic digestibility was higher than that of the untreated rice straw, probably due to the slight removal of lignin by the acidic water at high temperatures.³⁴ As the total concentrations of acid–base mixtures increased from 0 M to 0.05 M, the glucose yield significantly increased (from 9.6 g to 27.2 g per 100 g of lignocellulose). However, a further increase in the concentration of the acid–base mixture did not result into any significant increase in the enzymatic digestibility. For example, when the total concentrations of the acid–base mixtures increased to 0.5 M or 1 M, the produced glucose yield significantly decreased and the recovery yield of the insoluble solids was reduced to less than 40%. Therefore, a 0.05 M acid–base mixture (0.04 M of HCl and 0.01 M of NaOH) was selected as the optimum concentration for the pretreatment of rice straw, for which the enzymatic digestibility was 75.9% of the theoretical maximum glucose, obtained using pretreated and washed rice straw. Although the optimum molar concentration of the acid–base mixture used in this study, which is equivalent to 0.15% (w/v) of HCl and 0.04% (w/v) of NaOH, is much lower than that of the acid (0.5–5% [w/v])^35,36 or base (0.5–3% [w/v] for NaOH (ref. 37) and 2–14% [w/w] for NH₃)^38,39 catalysts in other pretreatment studies, the resulting enzymatic digestibility was comparable to that obtained from the conventionally pretreated rice straw, which also uses an acid or base, showing a glucose yield of ∼70%.^12,39,40

3.2 Effects of acid, base and salt on pretreatment

To investigate the mechanism of pretreatment using the acid–base mixture, several control experiments were performed using combinations of catalysts such as acid (HCl), base (NaOH), salt (NaCl) and acid with salt (Fig. 3). In these experiments, the amount of salt (i.e. 0.01 M NaOH) added to the acid was determined using the molar ratio of the acid–base mixture (i.e., 0.04 M HCl and 0.01 M NaOH) to simulate the formation of salt in the acid–base mixture. Among the different combinations of the acid, base and salt used, the acid–base mixture gave the highest enzymatic digestibility to the pretreated and washed rice straw (Fig. 3). The HCl pretreatment resulted in an enzymatic digestibility that was 68.8% of the theoretical maximum glucose yield, but using NaOH or NaCl resulted in 32.0% and 33.8% enzymatic digestibilities, respectively. In particular, the combination of HCl and NaCl, which was designed to simulate the acid–base mixture involving the acid–base reaction, did not result into an enzymatic digestibility as high as that found with the acid–base mixture. Therefore, the possible effect of the formation of NaCl and the remaining HCl from the acid–base mixture in the rice straw pretreatment was not simulated by the mixture of NaCl and HCl.

Fig. 3 Enzymatic digestibilities (% theoretical maximum glucose from the remaining glucan in the pretreated and washed rice straw) of the rice straw samples pretreated with various catalysts: 0.04 M HCl; 0.01 M NaOH; an acid–base mixture of 0.04 M HCl and 0.01 M NaOH; 0.01 M NaCl; 0.01 M NaCl; and 0.03 M HCl. Enzymatic hydrolysis was conducted using 15 FPU of cellulase (Accellerase 1000) g⁻¹ glucan at 50 °C (pH 4.8) and at 200 rpm for 50 h.

The compositions of the rice straw pretreated using the acid–base mixture, HCl, NaOH or NaCl were analyzed (Table 1). In the untreated rice straw, the total amount of carbohydrate and lignin was 70.9%, and this value was comparable to the amount derived from rice straw in other studies.^39,40 The recovery yields of the insoluble solids following NaOH or NaCl pretreatment were much higher than those following the acid–base mixture treatment, or the HCl pretreatment. This can be related to the lower enzymatic digestibility following NaOH or NaCl pretreatment as compared to the acid–base mixture or HCl pretreatment. In particular, xylan was substantially reduced after pretreatment when using HCl or the acid–base mixture. Glucan was also reduced, but to a lesser degree. These results are consistent with the typical characteristics of acid pretreatment.^14,41 Specifically, in the acid–base mixture pretreatment, the recovery yields of glucan, xylan and lignin in the insoluble solids were 90.1%, 37.2% and 60.6%, respectively. Accordingly, a higher amount of xylose was recovered in the liquid fraction of the pretreated rice straw when using the acid–base mixture or HCl, compared to NaOH or NaCl. When using the acid–base mixture, the lignin removal was comparable to that when using NaOH; however, the generation of acetic acid was lower than that when using HCl. Moreover, when using the acid–base mixture, furfural and HMF productions were lower than when using HCl. All of these results indicate that the acid–base mixture pretreatment has the characteristics of both acid and alkali pretreatments probably due to the actions on biomass by H⁺ from acid and OH⁻ from base before being associated into water. The acid–base mixture pretreatment generated less inhibitors than the acid pretreatment (Table 1). Taken together, the acid–base mixture pretreatment shows positive aspects of the acid and alkali pretreatments, including solubilizing hemicellulose and removing lignin, respectively.

Table 1 Composition of the rice straw pretreated with different catalysts^ab

	Untreated	Acid–base mixture (0.04 M HCl–0.01 M NaOH)	HCl (0.04 M)	NaOH (0.01 M)	NaCl (0.01 M)
a Pretreatment conditions were 190 °C, 3 min ramping with 2 min holding time and 10% (w/v) solids loading.b Experimental data are expressed as means ± standard deviations.c Includes xylose, galactose and arabinose in the liquid fractions.d NA: not applicable.
Component from insoluble solids (g per 100 g dry rice straw before pretreatment)
Insoluble solids recovery yield	NA^d	55.6 ± 2.5	55.9 ± 2.6	75.5 ± 0.8	85.0 ± 3.9
Glucan	35.8 ± 1.5	32.3 ± 0.6	30.9 ± 0.3	33.7 ± 0.0	33.1 ± 0.4
Xylan	10.5 ± 1.4	3.9 ± 0.5	2.9 ± 0.2	10.7 ± 0.2	10.5 ± 0.4
Galactan	3.3 ± 0.3	1.5 ± 1.4	2.2 ± 0.0	2.9 ± 0.0	3.4 ± 0.0
Arabinan	3.1 ± 0.5	1.8 ± 0.0	1.7 ± 0.0	2.3 ± 0.0	2.6 ± 0.0
Lignin	18.2 ± 1.3	11.0 ± 0.3	10.7 ± 0.0	10.8 ± 0.3	12.2 ± 0.2
Component from dissolved solids (g per 100 g rice straw before pretreatment)
Glucose	NA^d	4.9 ± 0.2	6.1 ± 0.0	0.5 ± 0.0	1.5 ± 0.0
Hemicellulosic monomer^c	NA^d	10.2 ± 0.1	14.9 ± 0.1	0.5 ± 0.0	1.3 ± 0.1
Acetic acid	NA^d	1.5 ± 0.0	2.0 ± 0.1	1.6 ± 0.0	0.3 ± 0.1
HMF	NA^d	0.4 ± 0.0	0.5 ± 0.0	0.0 ± 0.0	0.0 ± 0.0
Furfural	NA^d	0.6 ± 0.0	0.7 ± 0.0	0.0 ± 0.0	0.0 ± 0.0

---

3.3 Correlation of xylan and lignin removal with the enzymatic digestibility of acid–base mixture pretreated rice straw

The amounts of major biomass components of pretreated rice straw, such as glucan, xylan and lignin, were correlated with the enzymatic digestibilities of the acid–base mixture pretreated rice straw (Fig. 4). The removal of glucan did not show a high correlation coefficient with an increase in enzymatic digestibility. However, the losses of both xylan and lignin after the acid–base mixture pretreatment were highly correlated with an increase in enzymatic digestibility. These results imply that the removal of hemicellulose (e.g. xylan) and lignin may have contributed to the increased enzymatic digestibility by the pretreatment using the acid–base mixture. It is known that removal of hemicellulose contributes to the increase of enzymatic digestibility in acid pretreatment⁴¹ and that the removal of lignin contributes to the increase of enzymatic digestibility in alkali pretreatment.^38,42 The acid–base mixture pretreatment used in this study, resulted in the significant removal of both hemicellulose and lignin, thereby increasing the enzymatic digestibility of the pretreated biomass.

Fig. 4 Correlation of the biomass components of rice straw pretreated using the acid–base mixtures under various conditions, such as pHs and total concentrations of the acid–base mixtures, with the corresponding enzymatic digestibilities. (A) Glucan, (B) xylan and (C) lignin.

3.4 Evaluation of cellulose accessibility to the enzyme using a CBM

The cellulose accessibility of the acid–base mixture pretreated rice straw to the enzyme was tested by protein binding analysis using CtCBD3, a Type A CBM that is the typical for cellulose surface binding in cellulase.²⁴ Untreated and pretreated rice straw samples did not show significant differences in non-specific protein binding, with BSA as a control (Fig. 5). When CtCBD3 was used, the pretreated rice straw exhibited approximately two-fold higher binding capacity than the untreated rice straw. These results suggest that the acid–base mixture pretreatment significantly improved cellulose accessibility to the CBM, which may have contributed to the increased enzymatic digestibility of the acid–base mixture pretreated rice straw as shown in Fig. 3. In a previous study using alkali pretreatment of lignocellulose with more than 40% lignin removal,⁴² CtCBD3 binding was not higher than in the untreated lignocellulose. This was probably due to the redistribution or condensation of lignin after alkali pretreatment.^42,43 However, despite a similar degree of lignin removal in this study, removal of both lignin and hemicellulose may have transformed the pretreated rice straw into a structure that was more accessible to the CBM.^26,44 Moreover, transformation of lignin structures by acid in the acid–base mixture may reduce the unproductive binding of CBM to lignin.⁴⁵ This unique feature of the acid–base mixture pretreatment in the removal and modification of both hemicellulose and lignin, improves the accessibility of cellulose (in lignocellulose) to the enzyme.

Fig. 5 Comparison of the protein binding capacities of the untreated rice straw and the acid–base mixture pretreated rice straw. The rice straw was incubated with BSA or CtCBD3 in 50 mM potassium phosphate buffer (pH 7) at 4 °C for 3.5 h.

3.5 Saccharification and ethanol fermentation of the acid–base mixture pretreated rice straw

Washed solids of rice straw pretreated at the optimal conditions (0.05 M acid–base mixture composed of 0.04 M HCl and 0.01 M NaOH, 3 min ramping to 190 °C and 2 min holding at 190 °C) were hydrolyzed with 15 FPU Accellerase 1000 g⁻¹ glucan (Fig. 6A). The glucose yield from the untreated rice straw (control) was only 15.6% of the theoretical maximum glucose at 72 h. The final glucose yield at 72 h from the acid–base mixture pretreated and washed rice straw was 75.2%; however, more than 70% of the final glucose yield was achieved after 6 h. This relatively fast hydrolysis is uncommon in pretreated lignocellulose since previous studies have reported that approximately 40–65% of glucose yields are obtained after 6–12 h.^14,38,39 Depending on the process economics, it would be preferable to compromise the hydrolysis time rather than obtaining a lower maximum glucose yield. The fast reactivity of the acid–base pretreated rice straw (Fig. 6A) can be attributed to the high accessibility of cellulose to the enzyme (Fig. 5).

Fig. 6 (A) Saccharification of the washed solids of the acid–base mixture pretreated rice straw using 15 FPU of cellulase (Accellerase 1000) g⁻¹ glucan at 50 °C (pH 4.8) and at 200 rpm. (B) Simultaneous saccharification and ethanol fermentation (SSF) of the whole slurry or the washed solids of the acid–base mixture pretreated rice straw. For the SSF, 15 FPU Accellerase 1000 g⁻¹ glucan and Saccharomyces cerevisiae D₅A were added and cultivation was performed at 38 °C (pH 4.8) and at 170 rpm for 60 h.

To test the applicability of the acid–base mixture pretreatment in “the one-pot pretreatment, saccharification and fermentation”, the rice straw was pretreated with the acid–base mixture and then directly proceeded to the SSF by the addition of 15 FPU cellulase g⁻¹ glucan with buffer (0.05 M) and S. cerevisiae D₅A. The neutralization, conditioning, and detoxification steps were not performed (Fig. 6B). In order to utilize the entire liquid faction or the whole slurry of pretreated biomass, conditioning or neutralization of the liquid fraction is required after the acid pre-treatment of lignocellulose.^14,46 Even after acid neutralization, the formation of salts has been shown to inhibit microbial cell growth.^14,15 However, in this study, neutralization and conditioning were not necessary. It is possible that this was due to the concentration of acid–base mixture being less than 10 times lower than that found in the conventional dilute-acid pretreatments, or because of the partial neutralization in the acid–base mixture. Furthermore, the inhibitory effect of salts could have been negligible because of the lower amounts of acid and base used in the pretreatment during this study. Although the ethanol yield was determined based on the amount of glucan before pretreatment, (in which all the losses of glucan during pretreatment were accounted for), the ethanol yield from the whole slurry fermentation after 60 h reached 70.7% with 15 FPU of the enzyme. This ethanol yield was comparable to the ethanol yield (72.9%) estimated based on the assumption of a 90% glucan recovery yield after pretreatment × a 90% glucose yield from saccharification × a 90% ethanol yield from ethanol fermentation. When the washed solids of the pretreated rice straw were used alone in the SSF, the maximum yield of ethanol was only 49.7% with 15 FPU of the enzyme (after 48 h), based on the initial glucan before pretreatment. The ∼30% lower ethanol yield when using the washed solids could only be compared with the fermentation of the whole slurry, because the liquid fraction of the pretreated slurry was discarded. Similarly, in a previous study using dilute-acid pretreated and washed corn stover, an ethanol yield of only 41.9% was obtained, based on both glucan and xylan amounts before pretreatment because of the sugar loss during the S/L separation and solids washing.⁴⁷ Therefore, whole slurry fermentation has the advantage of using all the available glucan and glucose both in the solid and liquid phases of the pretreated slurry.^14,15,48 Moreover, the acid–base mixture pretreatment of this study has the advantage of both enhancing the final ethanol yield, and eliminating the conditioning and neutralization steps.

3.6 Generalization of acid–base mixture pretreatment

To validate the pretreatment effectiveness of the acid–base mixture as a pretreatment catalyst, different combinations of acid–base mixtures, which included H₂SO₄, KOH and NH₃, other than the HCl–NaOH mixture were tested. In these experiments, pretreatment was performed using the acid–base molar ratios of 4 : 1, 1 : 1 and 1 : 4 in the acid–base mixture, but other pretreatment conditions were fixed at the optimal pretreatment conditions selected for the HCl–NaOH mixture (ESI Fig. 1†). Of all the acid–base combinations, the acid-rich acid–base mixture (i.e. the molar ratio of 4 : 1 = acid : base) resulted into a significantly higher enzymatic digestibility of the pretreated and washed rice straw than equal or base-rich acid–base mixtures. For example, when H₂SO₄–KOH (ESI Fig. 1B†) was used, the highest enzymatic digestibility of 73.9%, was obtained. H₂SO₄–NaOH (ESI Fig. 1A†) and H₂SO–NH₃ (ESI Fig. 1C†) mixtures resulted into ∼65% enzymatic digestibility. HCl–KOH (ESI Fig. 1D†) and HCl–NH₃ (ESI Fig. 1E†) mixtures resulted into approximately 50–60% enzymatic digestibilities. In general, during lignocellulose pretreatment, removal of hemicellulose by the acid gives more positive effects on enhancing the enzymatic digestibility than the removal of lignin by an alkali.⁴⁹ These results indicate that an acid-rich acid–base mixture could be used as an effective catalyst after optimizing the total concentration and other pretreatment conditions.

3.7 Process overview

Based on the results of this study, 100 g dry wt of rice straw was pre-treated with the acid–base mixture and added to 0.05 M sodium citrate buffer, 15 FPU cellulase per g glucan and S. cerevisiae for the SSF in the same reactor that was used for the pretreatment (Fig. 7). From this one-pot pretreatment, saccharification and fermentation, 14.4 g ethanol was obtained, which is 70.7% of the theoretical maximum yield of ethanol from the initial glucan in the rice straw before pretreatment, which also accounts for the glucan loss throughout the process. When the insoluble solids obtained from washing the pretreated rice straw were used for the SSF, only 10.1 g of ethanol (49.7% of the theoretical maximum) is obtained, because the liquid phase of the pretreated rice straw containing the residual sugars and inhibitors is not used in the fermentation. Because the high titer of the ethanol production is also important to achieve economic sustainability in the production of fuels and commodity products from lignocellulose,⁴⁸ the proposed one-pot pretreatment, saccharification and fermentation should be investigated for high solid loadings of lignocellulose in the future. Furthermore, because of the large amount of xylose solubilised during the pretreatment in this study, if a yeast-engineered to ferment xylose is used in ethanol fermentation, the overall ethanol production could be significantly increased.

Fig. 7 Mass balances for the processing of rice straw to produce ethanol. The rice straw was pretreated with the HCl (0.04 M)–NaOH (0.01 M) mixture at 190 °C. The pretreated rice straw was hydrolyzed with 15 FPU of cellulase (Accellerase 1000) g⁻¹ glucan and fermented with S. cerevisiae D₅A at 38 °C for 60 h. The ethanol yield was determined based on the total glucan content in the untreated rice straw before pretreatment.

3.8 Comparison of estimated ethanol production costs

Although pretreatment, saccharification and fermentation in one pot may provide advantages in the process economics, many other factors, such as enzyme cost, reaction conditions used, saccharification and fermentation efficiencies, etc., are also important in determining the commercial success of fuel production using lignocellulosic biomass. Therefore, using the most updated NREL's cellulosic ethanol production cost estimation of different process schemes, the cost analysis of the one-pot process of the present study was implemented and compared.⁸ The four scenarios are mainly differentiated by the modes of operation, which include separate conditioning and separate fermentation (SCSF), separate conditioning and whole slurry fermentation (SCWF), whole slurry conditioning and fermentation (WCF) and acid–base mixture one-pot pretreatment, saccharification and fermentation (ABM one-pot; this study) (ESI Fig. 2†). SCSF and SCWF are conventional and already proven technology. WCF is one of the advanced technologies but is still semi-practical since either it was demonstrated at a low solids loading (6% w/v)¹⁴ or it is being developed at a high solids loading (20%, w/v) at NREL.⁸ Being experimentally by the enzymatic hydrolysis and fermenation at a 20% (w/v) solids loading.⁸ Unfortunately, the process models of SCSF, SCWF and WCF do not accurately represent any existing industrial processes because no commercial cellulosic ethanol plants exist to date. The most advanced model technology (i.e. WCF) and ABM one-pot share a lot of similarities such as lower loadings of pretreatment catalysts, neutralizing agents and enzymes, faster saccharification and fermentation and slightly lower ethanol yields due to the reduced enzyme loading.

On the basis of these scenarios, the estimated costs for the production of a gallon of ethanol were compared, considering the operating costs and cost of installed equipment (Fig. 8). The costs of ethanol for SCSF, SCWF, WCF and ABM one-pot were $6.47, $6.42, $5.95 and $5.07, respectively.⁸ A detailed contribution of each factor has been presented in ESI Table 1.† As compared to WCF, ABM one-pot showed lower ethanol production costs due to the lesser loading of pretreatment catalysts, no requirement of neutralizing agents and water washing, and lower non-enzyme conversion-related costs. In addition, less loading of the enzyme leads to a further decrease in the enzyme conversion-related cost. Because S/L separation, solids washing, conditioning and pH adjustment using neutralizing agents are not necessary in the ABM one-pot process, the corresponding equipment costs also get deducted. Although the ethanol yield of this study (70.7 gal/dry ton of rice straw), was lower than that of the NREL study (79 gal/dry ton of corn stover), the ABM one-pot process reduced other facility costs for boilers, storage spaces, etc. Taken together, by estimating the ethanol production costs, it became apparent that the ABM one-pot pretreatment, saccharification and fermentation scheme would be economically promising. However, in order to increase the ethanol yields and titres, more research on the high loadings of solids needs to be investigated.

Fig. 8 Comparison of the estimated ethanol production costs for the four different process schemes, namely, separate conditioning and separate fermentation (SCSF), separate conditioning and whole slurry fermentation (SCWF), whole slurry conditioning and fermentation (WCF) and one-pot pretreatment, saccharification and fermentation using an acid–base mixture (ABM one-pot). The costs were estimated on the basis of the NREL Technical Report.⁸

4 Conclusions

The majority of current lignocellulosic pretreatments require S/L separation, washing or neutralization and detoxification, after pretreatment and before SSF or a separate hydrolysis and fermentation (SHF), thereby increasing the overall costs. Therefore, to establish a cost-effective cellulosic ethanol process, the post-pretreatment steps need to be minimized. The “whole slurry processing in one pot” described here may provide a solution. The pre-treatment should not use too much acid or base catalyst in order to avoid generating large amounts of inhibitors. Considering all the costs with regards to the post-pretreatment processing steps in the conventional chemical pretreatment, the one-pot process, integrating the pretreatment, saccharification and fermentation, (steps based on the acid–base pretreatment), shows apparent advantages in the process economics for producing a commodity product using lignocellulose. However, the exact mechanism of the acid–base pretreatment has not been revealed yet.

Acknowledgements

This work was supported by grants from the National Research Foundation of Korea (2013M1A2A2072597) and the Advanced Biomass R&D Center of Korea (2011-0031353), funded by the Korean Government (MSIP). Facility support at Korea University Food Safety Hall for the Institute of Biomedical Science and Food Safety is also acknowledged.

References

1. V. Arantes and J. N. Saddler, Biotechnol. Biofuels, 2011, 4, 16.
2. J. A. Rollin, Z. Zhu, N. Sathitsuksanoh and Y.-H. P. Zhang, Biotechnol. Bioeng., 2011, 108, 22–30.
3. Q. Q. Wang, Z. He, Z. Zhu, Y.-H. P. Zhang, Y. Ni, X. L. Luo and J. Y. Zhu, Biotechnol. Bioeng., 2012, 109, 381–389.
4. J. R. M. Almeida, T. Modig, A. Petersson, B. Hähn-Hägerdal, G. Lidén and M. F. Gorwa-Grauslund, J. Chem. Technol. Biotechnol., 2007, 82, 340–349.
5. H. B. Klinke, A. B. Thomsen and B. K. Ahring, Appl. Microbiol. Biotechnol., 2004, 66, 10–26.
6. Z. L. Liu, Appl. Microbiol. Biotechnol., 2011, 90, 809–825.
7. M. Ask, M. Bettiga, V. Mapelli and L. Olsson, Biotechnol. Biofuels, 2013, 6, 22.
8. D. Humbird, R. Davis, L. Tao, C. Kinchin, D. Hsu, A. Aden, P. Schoen, J. Lukas, B. Olthof, M. Worley, D. Sexton and D. Dudgeon, Process Design and Economics for Biochemical Conversion of Lignocellulosic Biomass to Ethanol: Dilute-Acid Pretreatment and Enzymatic Hydrolysis of Corn Stover, National Renewable Energy Laboratory, NREL Technical Report, NREL/TP-5100–47764, Golden, CO, 2011.
9. J. Houghton, S. Weatherwax and J. Ferrell, Breaking the Biological Barriers to Cellulosic Ethanol: A Joint Research Agenda, DOE/SC-0095, 2005, ed. U.S. D.o. Energy.
10. L. J. Jönsson, B. Alriksson and N.-O. Nilvebrant, Biotechnol. Biofuels, 2013, 6, 16.
11. L. R. Lynd, C. E. Wyman and T. U. Gerngross, Biotechnol. Prog., 1999, 15, 777–793.
12. M. Jin, C. Gunawan, N. Uppugundla, V. Balan and B. E. Dale, Energy Environ. Sci., 2012, 5, 7168–7175.
13. A. Dutta, N. Dowe, K. N. Ibsen, D. J. Schell and A. Aden, Biotechnol. Prog., 2010, 26, 64–72.
14. Y. H. Jung, I. J. Kim, H. K. Kim and K. H. Kim, Bioresour. Technol., 2013, 132, 109–114.
15. Y. H. Jung, I. J. Kim, H. K. Kim and K. H. Kim, Bioprocess Biosyst. Eng., 2014, 37, 659–665.
16. J. Shi, J. M. Gladden, N. Sathitsuksanoh, P. Kambam, L. Sandoval, D. Mitra, S. Zhang, A. George, S. W. Singer, B. A. Simmons and S. Singh, Green Chem., 2013, 15, 2579–2589.
17. A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter, D. Templeton and D. Crocker, Laboratory Analytical Procedure: Determination of Structural Carbohydrates and Lignin in Biomass, National Renewable Energy Laboratory, NREL/TP-510-42618, Golden, CO, 2008.
18. A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter and D. Templeton, Laboratory Analytical Procedure: Determination of Sugars, Byproducts, and Degradation Products in Liquid Fraction Process Samples, National Renewable Energy Laboratory, NREL/TP-510-42623, Golden, CO, 2006.
19. A. Sluiter, B. Hames, D. Hyman, C. Payne, R. Ruiz, C. Scarlata, J. Sluiter, D. Templeton and J. Wolfe, Laboratory Analytical Procedure: Determination of Total Solids in Biomass and Total Dissolved Solids in Liquid Process Samples, National Renewable Energy Laboratory, NREL/TP-510-42621, Golden, CO, 2008.
20. A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter and D. Templeton, Laboratory Analytical Procedure: Determination of Ash in Biomass, National Renewable Energy Laboratory, NREL/TP-510-42622, Golden, CO, 2005.
21. M. Selig, N. Weiss and Y. Ji, Laboratory Analytical Procedure: Enzymatic Saccharification of Lignocellulosic Biomass, National Renewable Energy Laboratory, NREL/TP-510-42629, Golden, CO, 2008.
22. N. Dowe and J. McMillan, Laboratory Analytical Procedure: SSF Experimental Protocols. Lignocellulosic Biomass Hydrolysis and Fermentation, National Renewable Energy Laboratory, NREL/TP-510–42630, Golden, CO, 2001.
23. A. B. Boraston, D. N. Bolam, H. J. Gilbert and G. J. Davies, Biochem. J., 2004, 382, 769–781.
24. A. L. Creagh, E. Ong, E. Jervis, D. G. Kilburn and C. A. Haynes, Proc. Natl. Acad. Sci. U. S. A., 1996, 93, 12229–12234.
25. I. J. Kim, H.-J. Ko, T.-W. Kim, I.-G. Choi and K. H. Kim, Biotechnol. Bioeng., 2013, 110, 401–407.
26. I. J. Kim, H.-J. Ko, T.-W. Kim, K. H. Nam, I.-G. Choi and K. H. Kim, Appl. Microbiol. Biotechnol., 2013, 97, 5381–5388.
27. M. M. Bradford, Anal. Biochem., 1976, 72, 248–254.
28. J.-W. Kim, K. S. Kim, J.-S. Lee, S. M. Park, H.-Y. Cho, J. C. Park and J. S. Kim, Bioresour. Technol., 2011, 102, 8992–8999.
29. S. Kim, J. M. Park, J.-W. Seo and C. H. Kim, Bioresour. Technol., 2012, 109, 229–233.
30. S. Zhu, Y. Wu, Z. Yu, C. Wang, F. Yu, S. Jin, Y. Ding, R. Chi, J. Liao and Y. Zhang, Biosystems Eng, 2006, 93, 279–283.
31. I. Kim, M. S. U. Rehman, K. H. Kim and J.-I. Han, Bioresour. Technol., 2013, 135, 635–639.
32. C. Liu and C. E. Wyman, Carbohydr. Res., 2006, 341, 2550–2556.
33. R. Xing, A. V. Subrahmanyam, H. Olcay, W. Qi, G. P. van Walsum, H. Pendse and G. W. Huber, Green Chem., 2010, 12, 1933–1946.
34. G. Brodeur, E. Yau, K. Badal, J. Collier, K. B. Ramachandran and S. Ramakrishnan, Enzyme Res., 2011, 2011, 17.
35. V. B. Agbor, N. Cicek, R. Sparling, A. Berlin and D. B. Levin, Biotechnol. Adv., 2011, 29, 675–685.
36. P. Kumar, D. M. Barrett, M. J. Delwiche and P. Stroeve, Ind. Eng. Chem. Res., 2009, 48, 3713–3729.
37. Z. Wang, D. R. Keshwani, A. P. Redding and J. J. Cheng, Bioresour. Technol., 2010, 101, 3583–3585.
38. Y. H. Jung, I. J. Kim, J. J. Kim, K. K. Oh, J.-I. Han, I.-G. Choi and K. H. Kim, Bioresour. Technol., 2011, 102, 7307–7312.
39. J. K. Ko, J. S. Bak, M. W. Jung, H. J. Lee, I.-G. Choi, T. H. Kim and K. H. Kim, Bioresour. Technol., 2009, 100, 4374–4380.
40. T.-C. Hsu, G.-L. Guo, W.-H. Chen and W.-S. Hwang, Bioresour. Technol., 2010, 101, 4907–4913.
41. K. H. Kim, M. Tucker and Q. Nguyen, Bioresour. Technol., 2005, 96, 1249–1255.
42. Y. H. Jung, I. J. Kim, J.-I. Han, I.-G. Choi and K. H. Kim, Bioresour. Technol., 2011, 102, 9806–9809.
43. D. Gregg and J. N. Saddler, Appl. Biochem. Biotechnol., 1996, 57–58, 711–727.
44. S. Lv, Q. Yu, X. Zhuang, Z. Yuan, W. Wang, Q. Wang, W. Qi and X. Tan, BioEnergy Res., 2013, 6, 1128–1134.
45. H. Lou, J. Y. Zhu, T. Q. Lan, H. Lai and X. Qiu, ChemSusChem, 2013, 6, 919–927.
46. S. Larsson, E. Palmqvist, B. Hahn-Hägerdal, C. Tengborg, K. Stenberg, G. Zacchi and N.-O. Nilvebrant, Enzyme Microb. Technol., 1999, 24, 151–159.
47. N. Uppugundla, L. da Costa Sousa, S. P. S. Chundawat, X. Yu, B. Simmons, S. Singh, X. Gao, R. Kumar, C. E. Wyman, B. E. Dale and V. Balan, Biotechnol. Biofuels, 2014, 7, 72.
48. H. Zhou, J. Y. Zhu, X. Luo, S.-Y. Leu, X. Wu, R. Gleisner, B. S. Dien, R. E. Hector, D. Yang, X. Qiu, E. Horn and J. Negron, Ind. Eng. Chem. Res., 2013, 52, 16057–16065.
49. S.-Y. Leu and J. Y. Zhu, BioEnergy Res., 2013, 6, 405–415.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra10092a

---