Effect of Alkali and Alkaline Earth Metal Chlorides on Cellobiose

May 1, 2015 - ... Engineering, Curtin University, GPO Box U1987, Perth, Western Australia 6845, Australia ... Citation data is made available by parti...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/IECR

Effect of Alkali and Alkaline Earth Metal Chlorides on Cellobiose Decomposition in Hot-Compressed Water Yun Yu, Zainun Mohd Shafie, and Hongwei Wu* School of Chemical and Petroleum Engineering, Curtin University, GPO Box U1987, Perth, Western Australia 6845, Australia ABSTRACT: This paper reports a systematic study on the catalytic effect of alkali and alkaline earth metal (AAEM) chlorides on cellobiose decomposition in hot-compressed water (HCW) at 200−275 °C. The AAEM chlorides catalyze the cellobiose decomposition in HCW in the order of MgCl2 > CaCl2 > KCl > NaCl. The presence of AAEM chlorides not only increases the reaction rate but also alters the selectivities of primary reactions of cellobiose decomposition. The isomerization reactions to cellobiulose and glucosyl-mannose are strongly promoted by these cations due to their interactions with cellobiose. The hydrolysis reaction to glucose is also promoted as the hydrolysis of hydrated metal complexes generates H3O+. However, the promotion effect on hydrolysis reaction is much weaker, resulting in reduced glucose selectivity in AAEM chloride solutions. Depending on the AAEM species, the secondary decomposition reactions of those primary products are selectively catalyzed in AAEM chloride solutions, thus greatly influencing the product distribution of cellobiose decomposition in HCW.

1. INTRODUCTION Biomass hydrothermal conversion in hot-compressed water (HCW) is an important process for producing renewable biofuels and platform biochemicals,1 including polyols, furans, and organic acids.2 As a result of complex reactions,3 biomass conversion produces a liquid mixture of various compounds. Recent efforts were mainly on developing new processes for increasing the yield and selectivity of target compounds from monomer sugars (i.e., glucose) with the aid of catalysts4,5 and solvents.6,7 Model compounds including cellulose,8−16 cellobiose,17−23 and glucose24−27 are often deployed for exploring the fundamental reaction mechanism of biomass conversion in HCW. For example, our recent studies14 reported that the primary decomposition of cellulose in HCW produces glucose monomer and oligomers with various degrees of depolymerization (DPs). The further decomposition reactions of glucose oligomers in HCW are poorly documented in previous literature. A recent work proposed that the decomposition of glucose oligomers proceeds via isomerization of glucose at the reducing end into fructose, followed by the formation of fructose via hydrolysis.28 Using cellobiose as a model compound, our recent studies22,23 provided new insights on the decomposition of glucose oligomers in HCW. The dominant primary reactions are isomerization reactions to produce cellobiulose (glucosyl-fructose, GF) and glucosylmannose (GM) during cellobiose decomposition in HCW at 200−275 °C.23 Hydrolysis reaction to produce glucose contributes little to cellobiose primary decomposition but is promoted at increased temperatures and initial concentrations.23 The contribution of the retro-aldol condensation reaction to produce GE (glucosyl-erythrose) and glycolaldehyde is even smaller at 200−275 °C.23 Biomass contains abundant species of alkali and alkaline earth metals (AAEM), such as Na, K, Mg, and Ca. For example, the AAEM contents (on a dry weight basis) in mallee biomass can be as high as ∼0.6% for Na, ∼0.4% for K, ∼0.2% for Mg, and ∼2.7% for Ca.29 Previous studies30−34 have shown that © XXXX American Chemical Society

these inorganic species can influence biomass or sugar conversion in HCW. However, little is known on the effect of these inorganic species on cellobiose decomposition. Therefore, the purpose of this work is to investigate the effect of various AAEM chlorides on the mechanism and the reaction pathways of cellobiose decomposition in HCW.

2. EXPERIMENTAL SECTION 2.1. Materials and Reactor System. Standards and chemicals used in the experiments were purchased from Sigma-Aldrich, while GF and GM were synthesized by LC Scientific Inc. (Canada). A cellobiose solution with a concentration of 3 g L−1 (∼8.8 mM) was prepared. Then appropriate amounts of AAEM chlorides (NaCl, KCl, MgCl2, and CaCl2) were mixed with the cellobiose solution to prepare a set of solutions at a salt concentration of ∼88 mM (equivalent to a salt-to-cellobiose molar ratio of 10). Before experiments, the prepared solution was degassed to remove dissolved air using helium. The cellobiose decomposition experiments were carried out at 200−275 °C and 10 MPa using a continuous stainless tube reactor detailed elsewhere.22,23,27 The prepared solution was mixed with another stream of preheated water at a flow rate ratio of 1:2 (solution-to-water), making the final concentration of cellobiose and salt ∼2.9 and ∼29 mM, respectively. The residence time of reactant solution was adjusted by changing the reactor length and flow rate of the mixed stream. The temperature of effluent was rapidly cooled to room temperature in an ice water bath. 2.2. Sample Analysis. The liquid products were analyzed using higher performance anion exchange chromatography (Dionex ICS-5000) with pulsed amperometric detection and mass spectrometry (HPAEC-PAD-MS). The procedure for the Received: March 16, 2015 Revised: April 30, 2015 Accepted: May 1, 2015

A

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

Figure 1. Correlations between −ln[C(t)/C(0)] and residence time at various temperatures during cellobiose decomposition in HCW with the addition of various salts at a salt-to-cellobiose molar ratio of 10: (a) 200, (b) 225, (c) 250, and (d) 275 °C.

Table 1. Reaction Rate Constants and Kinetics Parameters of Cellobiose Decomposition under Various Reaction Conditions reaction rate constant (s−1) control (water only) NaCl (29 mM) KCl (29 mM) CaCl2 (29 mM) MgCl2 (29 mM)

kinetic parameters

200 °C

225 °C

250 °C

275 °C

activation energy (kJ mol−1)

0.0025 0.0027 0.0030 0.0037 0.0049

0.0069 0.0085 0.0112 0.0139 0.0184

0.0263 0.0283 0.0328 0.0522 0.0593

0.0936 0.1013 0.1071 0.1167 0.1510

104.9 103.9 101.7 101.0 99.0

pre-exponential factor (−) 8.33 7.35 5.10 5.46 4.29

× × × × ×

108 108 108 108 108

constant k (s−1) of cellobiose decomposition can be determined by −ln[C(t)/C(0)] = kt, where C(0) and C(t) are the concentrations of cellobiose (mg L−1) in the reactant and the product collected at a residence time t. The residence time t can be calculated using t = ρV/F, where V is the volume of the reactor (m3), ρ is the water density (kg m−3) under the reaction conditions, and F is the total mass flow rate (kg s−1) of solution. On the basis of the cellobiose concentration in the liquid product at various residence times, the relationship between −ln[C(t)/C(0)] and residence time t at various temperatures can be plotted in Figure 1. A linear relationship between −ln[C(t)/C(0)] and residence time t is observed for all reaction conditions. On the basis of the slope of the linear curve, the reaction rate constants of cellobiose decomposition under various reaction conditions are determined and presented in Table 1. The addition of AAEM chlorides indeed accelerates cellobiose decomposition in HCW. At a given temperature and salt concentration, the reaction rate of cellobiose decomposition follows the order MgCl2 > CaCl2 > KCl > NaCl. Compared to the reaction rate constant of control experiment (water only), the reaction rate constant of cellobiose decomposition is almost doubled in the MgCl2 solution at all temperatures. On the basis of the reaction rate constants at various temperatures, the activation energy and pre-exponential factor of cellobiose decomposition in various AAEM chloride solutions can be also estimated and are listed in

HPAEC-PAD-MS analysis was detailed elsewhere. 22 A CarboPac PA20 column was used to separate the compounds in the liquid products. Experiments were at least repeated, and the standard errors were ∼2−7%. Total carbon contents of selected samples at high temperatures were also determined by a total organic carbon (TOC) analyzer (Shimadzu TOC-VCPH), confirming the negligible gas or solid residue formed from cellobiose decomposition under the reaction conditions in this study. The pH value of liquid sample after reaction was also measured at room temperature by an acid−base titrator (MEP Oil Titrino plus 848) to confirm the formation of organic acids. The cellobiose conversion (X) on a carbon basis can be calculated using the equation X = [C(0) − C(t)]/C(0), where C(0) and C(t) are the concentrations of cellobiose (mg L−1) in the reactant and the product collected at a residence time t. The yield (Yi) and selectivity (Si) of compound i at a residence time of t can be determined on a carbon basis using the equations Yi = (ci × ai)/[C(0) × a] and Si = Yi /X, where ci is the concentration of compound i (mg L−1) in the product and ai and a are the carbon contents (wt %) of compound i and cellobiose, respectively.

3. RESULTS AND DISCUSSION 3.1. Effect of AAEM Chlorides on the Kinetics of Cellobiose Decomposition in HCW. If the cellobiose decomposition follows first-order kinetics, the reaction rate B

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

Figure 2. Effect of AAEM chlorides on the yields of cellobiose and its primary products from cellobiose decomposition at various temperatures, expressed on a carbon basis: (a) cellobiose at 200 °C; (b) cellobiose at 225 °C; (c) cellobiose at 250 °C; (d) cellobiose at 275 °C; (e) GF at 200 °C; (f) GF at 225 °C; (g) GF at 250 °C; (h) GF at 275 °C; (i) GM at 200 °C; (j) GM at 225 °C; (k) GM at 250 °C; (l) GM at 275 °C; (m) glucose at 200 °C; (n) glucose at 225 °C; (o) glucose at 250 °C; (p) glucose at 275 °C.

However, at high temperatures (>250 °C), the primary products can further decompose to other secondary products, depending on the catalytic effect of AAEM chlorides on the secondary reactions. Therefore, the yields of GF, GM, and glucose are determined by the trade-off between their formation from cellobiose decomposition and their decomposition to other products. A careful comparison for the cellobiose conversion and the yields of primary products at the same residence time suggests that the decomposition of GF and GM commences at a cellobiose conversion of ∼60%, while glucose starts to decompose at a higher cellobiose conversion of ∼80%. Therefore, the addition of AAEM chlorides promotes both the formation and the secondary decomposition of primary products from cellobiose decomposition. The AAEM chlorides exhibit a different catalytic effect on the secondary decomposition of GF, GM, and glucose. For GF and GM, the secondary decomposition is promoted in all AAEM chloride solutions, following the order MgCl2 > CaCl2 > KCl > NaCl. For example, at 250 °C. the yields of GF and GM in the MgCl2 solution are initially highest but become lowest when the residence time increases to ∼56 s. When the temperature increases to 275 °C, the yields of GF and GM in the MgCl2 solution are lowest even at a short residence time of ∼6 s. Since MgCl2 has the strongest effect on the formation and decomposition of GF and GM, the maximal yields of GF and GM are obtained in the MgCl2 solution, being ∼32% (at 250 °C and a residence time of ∼5 s) and ∼3% (at 250 °C and a

Table 1. Without salt addition, the cellobiose decomposition has an apparent activation energy of ∼104.9 kJ mol−1. The addition of salt leads to a small reduction in the apparent activation energy of cellobiose decomposition in HCW, i.e., to ∼103.9 kJ mol−1 for NaCl, ∼101.7 kJ mol−1 for KCl, ∼99.0 kJ mol−1 for MgCl2, and ∼101.0 kJ mol−1 for CaCl2. 3.2. Effect of AAEM Chlorides on the Primary Products of Cellobiose Decomposition in HCW. The above results clearly demonstrate that the cellobiose decomposition reaction is accelerated in AAEM chloride solutions, especially for the MgCl2 solution. However, it is not clear which primary reaction of cellobiose decomposition is affected by AAEM chlorides. Our recent study22 found that cellobiose decomposes via a series of main primary reactions, i.e., isomerization reactions to form GF and GM, hydrolysis reaction to form glucose, and retro-aldol reaction to form GE and glycolaldehyde. On the basis of the concentrations in the liquid products, the yields of cellobiose and its main primary decomposition products GF, GM, and glucose were calculated on a carbon basis. Figure 2 compares the effect of AAEM chlorides on the yields of cellobiose and its main primary decomposition products at 200−275 °C and a salt concentration of ∼29 mM. The yields of three main primary products are all enhanced in AAEM chloride solutions. Therefore, three main primary reactions of cellobiose decomposition are all promoted by AAEM chlorides, and such promotion effect also follows the similar order of MgCl2 > CaCl2 > KCl > NaCl. C

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

Figure 3. Effect of AAEM chlorides on the selectivities of primary products from cellobiose decomposition at various temperatures, expressed on a carbon basis: (a) GF at 200 °C; (b) GF at 225 °C; (c) GF at 250 °C; (d) GF at 275 °C; (e) GM at 200 °C; (f) GM at 225 °C; (g) GM at 250 °C; (h) GM at 275 °C; (i) glucose at 200 °C; (j) glucose at 225 °C; (k) glucose at 250 °C; (l) glucose at 275 °C.

Figure 4. Effect of AAEM chlorides on the pH value of liquid product (measured at room temperature after reaction) at various cellobiose conversions. (a) 200, (b) 225, (c) 250, and (d) 275 °C.

residence time of ∼10 s), respectively. For glucose, only the alkaline earth metal chlorides exhibit a strong catalytic effect on its decomposition, while the effect of alkali metal chlorides is weak. Therefore, the maximal glucose yields in the alkali metal chloride solutions are quite similar to that in water. It is also noteworthy that the maximal glucose yield is obtained at a longer residence time than those for GF and GM. This is because glucose is both a primary product (from cellobiose

hydrolysis) and a secondary product (from the hydrolysis of GF and GM) during cellobiose decomposition in HCW. The enhancements to the three major primary reactions in AAEM chloride solutions are different as reflected by the selectivity of each primary reaction. Figure 3 shows that the addition of AAEM chlorides increases the selectivities of isomerization reactions to produce GF and GM but decreases the selectivity of the hydrolysis reaction to produce glucose. Therefore, the AAEM chlorides are more significant in D

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

Figure 5. Effect of AAEM chlorides on the relative selectivity of GE from cellobiose decomposition at various temperatures, expressed on a basis of peak area: (a) 200 °C, (b) 225 °C, (c) 250 °C, and (d) 275 °C.

Figure 6. Effect of AAEM chlorides on the yields and selectivities of fructose and mannose from cellobiose decomposition at various temperatures, expressed on a carbon basis: (a) yield of fructose at 200 °C; (b) yield of fructose at 225 °C; (c) yield of fructose at 250 °C; (d) yield of fructose at 275 °C; (e) yield of mannose at 200 °C; (f) yield of mannose at 225 °C; (g) yield of mannose at 250 °C; (h) yield of mannose at 275 °C; (i) selectivity of fructose at 200 °C; (j) selectivity of fructose at 225 °C; (k) selectivity of fructose at 250 °C; (l) selectivity of fructose at 275 °C; (m) selectivity of mannose at 200 °C; (n) selectivity of mannose at 225 °C; (o) selectivity of mannose at 250 °C; (p) selectivity of mannose at 275 °C.

promoting isomerization reactions, particularly the alkaline earth metal chlorides, because the selectivities of isomerization reactions are higher in the alkaline earth metal chloride solutions. Compared to GF and GM, which show decreasing selectivities with increasing cellobiose conversion, the glucose selectivity initially increases with increasing cellobiose conversion, reaches a maximum of ∼20−33% (depending on AAEM species and temperature), and then starts to decrease at a late stage of cellobiose decomposition (i.e., >80% conversion). The increase of glucose selectivity can be due to the fact that glucose is also a secondary product from the hydrolysis of GF and GM. Another reason is that cellobiose decomposition also results in the formation of organic acids,22

which may catalyze the hydrolysis reactions of cellobiose or its primary products to produce glucose. However, glucose is further decomposed to other products, leading to the reduction in glucose selectivity at increased cellobiose conversions. Figure 3 also demonstrates that AAEM chlorides not only influence the initial selectivity of glucose but also affect the evolution of glucose selectivity during cellobiose conversion. In an alkali metal (K or Na) chloride solution, the glucose selectivity is initially lower but increases to similar levels to those in water. The increased selectivity of glucose is more likely due to the formation of organic acids as the pH values (measured at room temperature after reaction) of liquid products rapidly reduces from ∼7 to ∼5 (see Figure 4) in the E

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

Figure 7. Effect of AAEM chlorides on the yields and selectivities of 5-HMF and cellobiosan from cellobiose decomposition at various temperatures, expressed on a carbon basis: (a) yield of 5-HMF at 200 °C; (b) yield of 5-HMF at 225 °C; (c) yield of 5-HMF at 250 °C; (d) yield of 5-HMF at 275 °C; (e) yield of cellobiosan at 200 °C; (f) yield of cellobiosan at 225 °C; (g) yield of cellobiosan at 250 °C; (h) yield of cellobiosan at 275 °C; (i) selectivity of 5-HMF at 200 °C; (j) selectivity of 5-HMF at 225 °C; (k) selectivity of 5-HMF at 250 °C; (l) selectivity of 5-HMF at 275 °C; (m) selectivity of cellobiosan at 200 °C; (n) selectivity of cellobiosan at 225 °C; (o) selectivity of cellobiosan at 250 °C; (p) selectivity of cellobiosan at 275 °C.

chlorides also promote the decomposition of GE, resulting in the rapid decrease in the GE selectivity as cellobiose conversion increases. 3.3. Effect of AAEM Chlorides on the Secondary Products of Cellobiose Decomposition in HCW. Further investigation was also taken to examine the yields and selectivities of the major products from the secondary decomposition of the primary products from cellobiose decomposition in HCW. Figure 6 illustrates the effect of AAEM chlorides on the yields and selectivities of fructose and mannose, which are the hydrolysis products of GF and GM. The AAEM chlorides increase the formation of fructose and mannose at low temperatures (i.e., 250 °C), the fructose and mannose are easily degraded, particularly in the alkaline earth metal chloride solutions. However, it is interesting to see that the selectivities of fructose and mannose as a function of cellobiose conversion are not significantly affected by the addition of AAEM chlorides at all temperatures. The aforementioned results (see Figure 3) have shown that the selectivities of GF and GM are increased in all AAEM chloride solutions so that the selectivities of fructose and mannose as the main hydrolysis products for GF and GM are expected to increase. It is also noted that the selectivities of fructose and mannose are much lower than those of glucose. For example, the maximal selectivity of fructose in the NaCl

alkali metal chloride solutions even at the early stage of cellobiose conversion (i.e., ∼10−20%). The formed organic acids can further catalyze the hydrolysis reaction, leading to the increase in glucose selectivity in the alkali metal chloride solutions. The similar glucose selectivity also indicates the little catalytic effect of alkali metal chlorides on glucose decomposition. However, in an alkaline earth metal (Mg or Ca) chloride solutions, the glucose selectivity is much lower than those in water or the alkali metal chloride solutions. Although the addition of the alkaline earth metal chlorides also promotes the organic acid formation (see the pH value in Figure 3), the lower glucose selectivity suggests that the catalytic effect of the alkaline earth metal chlorides on glucose decomposition is much stronger, especially in the case of MgCl2. Our previous work22 also identified that retro-aldol condensation reaction to form GE and glycolaldehyde is also a primary reaction of cellobiose decomposition, but its contribution is small compared to other primary reactions. Although the yield of GE is not quantified in this study due to the unavailability of the standard, the relative selectivity of GE was analyzed according to the method developed elsewhere.22 The results in Figure 5 demonstrate that the retro-aldol condensation reaction to form GE and glycolaldehyde is also promoted in AAEM chloride solutions, following the similar order MgCl2 > CaCl2 > KCl > NaCl. However, the AAEM F

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

Figure 8. Effect of AAEM chlorides on the yields and selectivities of glycolaldehyde and erythrose from cellobiose decomposition at various temperatures, expressed on a carbon basis: (a) yield of glycolaldehyde at 200 °C; (b) yield of glycolaldehyde at 225 °C; (c) yield of glycolaldehyde at 250 °C; (d) yield of glycolaldehyde at 275 °C; (e) yield of erythrose at 200 °C; (f) yield of erythrose at 225 °C; (g) yield of erythrose at 250 °C; (h) yield of erythrose at 275 °C; (i) selectivity of glycolaldehyde at 200 °C; (j) selectivity of glycolaldehyde at 225 °C; (k) selectivity of glycolaldehyde at 250 °C; (l) selectivity of glycolaldehyde at 275 °C; (m) selectivity of erythrose at 200 °C; (n) selectivity of erythrose at 225 °C; (o) selectivity of erythrose at 250 °C; (p) selectivity of erythrose at 275 °C.

solution is ∼5% at 250 °C, compared to that of ∼30% for glucose under the same condition. It is known that fructose is more prone to decomposition than glucose.24 Therefore, the results suggest that both formation and decomposition of fructose and mannose are promoted at similar levels, leading to the observed insignificant effect of AAEM chlorides on the selectivities of fructose and mannose. It also appears that the alkaline earth metal chlorides, which have a more significant influence on fructose formation, also have a more significant influence on fructose decomposition. Figure 7 shows the effect of AAEM chlorides on the yields and selectivies of 5-HMF and cellobiosan that are the dehydration products. The yield and selectivity of 5-HMF at 200 °C are low but increase with temperature. At temperatures > 225 °C, the 5-HMF yield is substantially increased in AAEM chloride solutions. The maximal 5-HMF yield is ∼24% at 275 °C in the CaCl2 solution, indicating that 5-HMF is a major secondary product from cellobiose decomposition. The data also show that the addition of alkaline earth metal chlorides has a more significant effect on the 5-HMF formation but also strongly influences the degradation of 5-HMF. Compared to that for the alkaline earth metal chlorides, the addition of the alkali metal chlorides only promotes the 5-HMF formation. Thus, the addition of the alkali metal chlorides results in an

increase in the 5-HMF selectivity, whereas the addition of the alkaline earth metal chlorides leads to a reduction in the 5HMF selectivity. However, the yield of cellobiosan is relatively low ( CaCl2 > KCl > NaCl. For erythrose, it has a maximal yield of ∼2.6% at G

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

Figure 9. Main reaction network of cellobiose decomposition in HCW, as catalyzed by Lewis acids (i.e., AAEM cations) and Brønsted acid (i.e., H+).

275 °C in the NaCl solution. The addition of alkali metal chlorides increases its yield but with insignificant affect on its selectivity. The addition of alkaline earth metal chlorides initially increases the erythrose yield but reduces its yield at increased conversions. Obviously, the alkaline earth metal chlorides have a more significant catalytic effect on erythrose decomposition. 3.4. Discussion on the Catalytic Mechanism of AAEM Chlorides on Cellobiose Decomposition. The results of this study clearly show that the addition of AAEM chlorides enhances the cellobiose decomposition in HCW, depending on the AAEM species. At the same salt concentration, the effect of AAEM chlorides follows the order MgCl2 > CaCl2 > KCl > NaCl. It is found that the addition of AAEM chlorides greatly affects both the primary and the secondary reactions of cellobiose decomposition in HCW. Although the effect of AAEM chlorides on cellobiose decomposition is complex, there may be several reasons resulting in the observed effect. First, the addition of AAEM chlorides leads to changes in water properties, especially dielectric constant and ion product.35,36 It is known that for a reaction with a transition state more polar than the reactant an increase in the solvent polarity can increase the reaction rate.37 The effect of salt on hydrothermal decomposition may be related to an increase in the solvent polarity.34,37 However, there are insufficient data and knowledge on the dielectric constant of various salt solutions under subcritical and supercritical conditions. Our study employed a very low salt concentration (0.029 M) so that the change in dielectric constant due to salt addition is expected to be small.36 Therefore, the significant increase in the reaction rate upon the addition of salts (especially for the alkaline earth metal chlorides) in the solutions cannot be explained by the increase in the solvent polarity alone. Ion product of HCW is the other important property that can be influenced by salt addition. A higher ion product can provide better catalysis for sugar decomposition in HCW.38,39 It was reported that salt addition can increase the ion product of water at temperatures up to 100 °C.35 Unfortunately, there are limited data on the ion product of various salt solutions under subcritical and supercritical conditions. Consequently, in order to understand the effect of salts on sugar decomposition, further research is required to investigate how these HCW properties are influenced by salt addition. Second, it is known that sugar forms complexes with cations in aqueous solution at room temperature, particularly with the alkaline earth metal cations.40−43 At increased temperatures, such interactions between sugar and cations could strongly affect the cellobiose decomposition reaction.43 The complexation between cations and cellobiose creates positively charged

molecules that could promote various cellobiose decomposition reactions, depending on the coordination of cations. For example, Lewis acids (i.e., CrCl3) were recently reported to catalyze the isomerization reaction of aldoses to ketoses (i.e., glucose to fructose,5 xylose to xylulose44) via the coordination of Cr3+, mostly like in the form of [Cr(OH)]2+.5 It is known that AAEM cations are weak Lewis acids and hence can catalyze the isomerization reaction via the coordination of AAEM cations to form complexes with cellobiose. The Lewis acidity of the alkaline earth metal cations is stronger than that of the alkali metal cations, leading to a more significant effect on the formation of GF and GM in the MgCl2 and CaCl2 solutions. It is noteworthy that the formation of glucose is slightly enhanced during cellobiose decomposition in AAEM chloride solutions. This is probably because cations also form hydrated complexes (i.e., [M(H2O)n]z+) in aqueous solution. Such hydrated complexes undergo hydrolysis reactions to release H3O+ via the following reaction [M(H 2O)n ]z + + H 2O ↔ [M(H 2O)n − 1OH](z − 1) + + H3O+ (1)

Therefore, hydrated metal complexes act as Brønsted acids, which can catalyze the hydrolysis reaction. The dissociation constant (pKa) of the cation is related to the charge and radius of the cation. The pKa values for monovalent cations (i.e., 14.2 for Na+ and 14.5 for K+) are higher than those of divalent cations (i.e., 11.4 for Mg2+ and 12.8 for Ca2+).45 At high temperatures, more H+ can be generated in the MgCl2 and CaCl2 solutions. For example, the H+ concentration in the MgCl2 solution (29 mM) at 250 °C is calculated to be ∼6.6 × 10−5 M based on the literature data,46,47 while the H+ concentration in water at 250 °C is only ∼2.4 × 10−6 M.48 Therefore, the actual H+ concentration in the MgCl2 solution (29 mM) at reaction temperature increases by more than one order. This explains the more significant effect of MgCl2 and CaCl2 on glucose formation via hydrolysis reaction. Although both isomerization and hydrolysis reactions are promoted, such promotion effect is more significant for isomerization reactions, leading to the increased selectivities of isomerization reactions. It is also noted that the alkaline earth metal cations have a more significant effect to promote the isomerization reactions; thus, more substantial increases in the selectivities of isomerization reactions occur in the alkaline earth metal chloride solutions. A recent work5 indicated that the dehydration of fructose to 5-HMF and the decomposition of 5-HMF to levulinic acid and formic acid are also catalyzed by Brønsted acids. Therefore, the stronger Brønsted acidity derived from the alkaline earth metal cations not only catalyzes the hydrolysis reactions (i.e., GF to glucose and fructose, GM to glucose and mannose) but also H

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

in the selectivities of isomerization reactions. Compared to those in the NaCl and KCl solutions, the selectivities of isomerization reactions in the MgCl2 and CaCl2 solutions are higher. The secondary decomposition reactions of those primary products are also selectively catalyzed depending on the AAEM species. The study highlights the important role of AAEM cations (i.e., as Lewis acids and Brønsted acid) in catalyzing the reaction network of cellobiose decomposition in HCW.

favors the dehydration reactions (fructose to 5-HMF) and the decomposition of 5-HMF to levulinic acid and formic acid. This in turn explains the lower selectivities of 5-HMF in the alkaline earth metal chloride solutions. Compared to that of the alkaline earth metal cations, the ability of alkali metal cations to generate H3O+ is weak. Therefore, the degradation of 5-HMF is not favored in the alkali metal chloride solutions, resulting in higher 5-HMF selectivities. Third, the anion Cl− may also have some interactions with sugar during cellobiose decomposition. It was reported that Cl− enhances furfural formation from xylose in dilute aqueous acidic solution, due to the promotion of 1,2-enediol formation from xylose.49 A 13C NMR study indicated that Cl− anions donated electrons to C1, C3, and C5 carbons of glucose in saturated solution of MgCl2,50 thus affecting the glucose decomposition reaction. This suggests that the anion Cl− also plays a role in cellobiose decomposition in HCW, i.e., promoting isomerization reaction. However, under the reaction condition in this study, the role of Cl− is expected to be much smaller than that of cation, since the reaction rate of cellobiose in the MgCl2 solution is much faster than that in the CaCl2 solution at the same salt concentration. Figure 9 summarizes the effect of AAEM chlorides on the main reaction network of cellobiose decomposition in HCW. Cellobiose decomposition in HCW mainly proceeds via the isomerization reactions to form GF and GM, which are catalyzed by AAEM cations as Lewis acids, due to the coordination of AAEM cations to form complexes. The stronger Lewis acidity of the alkaline earth metal cations is largely responsible for the higher selectivities of cellobiose isomerization reactions in the alkaline earth metal chloride solutions. In contrast, the Brønsted acid has little effect on the isomerization reactions during cellobiose decomposition in HCW, as revealed in our recent work.51 Further hydrolysis of GF and GM requires the action of H+ to produce glucose, fructose, and mannose. Once glucose is formed, it is further catalyzed by the alkaline earth metal cations to fructose or mannose via isomerization reactions. The catalytic effect of the alkali metal cations on glucose isomerization is weak, probably due to the weak Lewis acidity of the alkali metal cations to form complexes with glucose. Fructose is rapidly dehydrated to 5HMF and then to levulinic acid. Both reactions are catalyzed by H+. The hydrolysis of hydrated metal complexes of the alkaline earth metal cations generates more H+, thus promoting the transformation of fructose to levulinic acid via 5-HMF. It should be mentioned that the retro-aldol condensation to form GE is also promoted in the AAEM chloride solutions. Further hydrolysis of GE leads to formation of glucose and erythrose. Erythrose is then decomposed into glycolaldehyde, which is most likely catalyzed by the AAEM cations, eventually resulting in the increase in the selectivity of glycolaldehyde.



AUTHOR INFORMATION

Corresponding Author

*Fax: +61-8-92662681. E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was initially supported by the Centre for Research into Energy for Sustainable Transport (CREST) through the Western Australian Government Centre of Excellence Program. The authors are grateful to the partial financial support received from the Australian Research Council’s Discovery Projects Program. Z.M.S. also acknowledges the Malaysian Agricultural Research and Development Institute (MARDI) for the support of her Ph.D. scholarship.



REFERENCES

(1) Peterson, A. A.; Vogel, F.; Lachance, R. P.; Fröling, M.; Antal, M. J., J.; Tester, J. W. Thermochemical biofuel production in hydrothermal media: A review of sub- and supercritical water technologies. Energy Environ. Sci. 2008, 1, 32−65. (2) Bozell, J. J.; Petersen, G. R. Technology development for the production of biobased products from biorefinery carbohydratesthe US Department of Energy’s “Top 10” revisited. Green Chem. 2010, 12, 539−554. (3) Yu, Y.; Lou, X.; Wu, H. Some Recent Advances in Hydrolysis of Biomass in Hot-Compressed Water and Its Comparisons with Other Hydrolysis Methods. Energy Fuels 2008, 22, 46−60. (4) Holm, M. S.; Saravanamurugan, S.; Taaming, E. Conversion of Sugars to Lactic Acid Derivatives Using Heterogeneous Zeotype Catalysts. Science 2010, 328, 602−605. (5) Choudhary, V.; Mushrif, S. H.; Ho, C.; Anderko, A.; Nikolakis, V.; Marinkovic, N. S.; Frenkel, A. I.; Sander, S. I.; Vlachos, D. G. Insights into the Interplay of Lewis and Brønsted Acid Catalysts in Glucose and Fructose Conversion to 5-(Hydroxymethyl)furfural and Levulinic Acid in Aqueous Media. J. Am. Chem. Soc. 2013, 135, 3997− 4006. (6) Zhao, H.; Holladay, J. E.; Brown, H.; Zhang, Z. C. Metal Chlorides in Ionic Liquid Solvents Convert Sugars to 5-Hydroxymethylfurfural. Science 2007, 316, 1597−1600. (7) Román-Leshkov, Y.; Chheda, J. N.; Dumesic, J. A. Phase Modifiers Promote Efficient Production of Hydroxymethylfurfural from Fructose. Science 2006, 312, 1933−1937. (8) Zhang, Y.; Ikeda, A.; Hori, N.; Takemura, A.; Ono, H.; Yamada, T. Characterization of liquefied product from cellulose with phenol in the presence of sulfuric acid. Bioresour. Technol. 2006, 97, 313−321. (9) Tolonen, L. K.; Juvonen, M.; Niemelä, K.; Mikkelson, A.; Tenkanen, M.; Sixta, H. Supercritical water treatment for cellooligosaccharide production from microcrystalline cellulose. Carbohydr. Res. 2015, 401, 16−23. (10) Adschiri, T.; Hirose, S.; Malaluan, R.; Arai, K. Noncatalytic Conversion of Cellulose in Supercritical and Subcritical Water. J. Chem. Eng. Jpn. 1993, 26, 676−680. (11) Sakaki, T.; Shibata, M.; Miki, T.; Hirosue, H.; Hayashi, N. Decomposition of Cellulose in Near-Critical Water and Fermentability of the Products. Energy Fuels 1996, 10, 684−688.

4. CONCLUSIONS This study investigates the effect of AAEM chlorides on cellobiose decomposition in HCW at 200−275 °C. The addition of AAEM chlorides enhances cellobiose conversion in the order MgCl2 > CaCl2 > KCl > NaCl, and the reaction rate constant of cellobiose decomposition is almost doubled in MgCl2 solution. As the primary reactions of cellobiose decomposition, both isomerization reactions and hydrolysis reaction are promoted in the AAEM chloride solutions. However, the promotion effect of AAEM chlorides on isomerization reactions is more significant, resulting in increases I

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research (12) Sasaki, M.; Kabyemela, B.; Malaluan, R.; Hirose, S.; Takeda, N.; Adschiri, T.; Arai, K. Cellulose hydrolysis in subcritical and supercritical water. J. Supercrit. Fluids 1998, 13, 261−268. (13) Yu, Y.; Wu, H. Characteristics and Precipitation of Glucose Oligomers in the Fresh Liquid Products Obtained from the Hydrolysis of Cellulose in Hot-Compressed Water. Ind. Eng. Chem. Res. 2009, 48, 10682−10690. (14) Yu, Y.; Wu, H. Understanding the Primary Liquid Products of Cellulose Hydrolysis in Hot-Compressed Water at Various Reaction Temperatures. Energy Fuels 2010, 24, 1963−1971. (15) Yu, Y.; Wu, H. Significant Differences in the Hydrolysis Behavior of Amorphouse and Crystalline Portions within Microcrystalline Cellulose in Hot-Compressed Water. Ind. Eng. Chem. Res. 2010, 49, 3902−3909. (16) Yu, Y.; Wu, H. Evolution of Primary Liquid Products and Evidence of in situ Structural Changes in Cellulose with Conversion during Hydrolysis in Hot-Compressed Water. Ind. Eng. Chem. Res. 2010, 49, 3919−3925. (17) Liang, X.; Montoya, A.; Haynes, B. S. Local Site Selectivity and Conformational Structures in the Glycosidic Bond Scission of Cellobiose. J. Phys. Chem. B 2011, 115, 10682−10691. (18) Bobleter, O.; Bonn, G. The Hydrothermolysis of Cellobiose and Its Reaction-Product D-Glucose. Carbohydr. Res. 1983, 124, 185−193. (19) Kabyemela, B. M.; Takigawa, M.; Adschiri, T.; Malaluan, R. M.; Arai, K. Mechanism and Kinetics of Cellobiose Decomposition in Suband Supercritical Water. Ind. Eng. Chem. Res. 1998, 37, 357−361. (20) Sasaki, M.; Furukawa, M.; Minami, K.; Adschiri, T.; Arai, K. Kinetics and Mechanism of Cellobiose Hydrolysis and Retro-Aldol Condensation in Subcritical and Supercritical Water. Ind. Eng. Chem. Res. 2002, 41, 6642−6649. (21) Park, J. H.; Park, S. D. Kinetics of Cellobiose Decomposition under Subcritical and Supercritical Water in Continuous Flow System. Korean J. Chem. Eng. 2002, 19, 960−966. (22) Yu, Y.; Mohd Shafie, Z.; Wu, H. Cellobiose Decomposition in Hot-Compressed Water: Importance of Isomerization Reactions. Ind. Eng. Chem. Res. 2013, 52, 17006−17014. (23) Mohd Shafie, Z.; Yu, Y.; Wu, H. Insights into the Primary Decomposition Mechanism of Cellobiose under Hydrothermal Conditions. Ind. Eng. Chem. Res. 2014, 53, 14607−14616. (24) Kabyemela, B. M.; Adschiri, T.; Malaluan, R. M.; Arai, K. Kinetics of Glucose Epimerization and Decomposition in Subcritical and Supercritical Water. Ind. Eng. Chem. Res. 1997, 36, 1552−1558. (25) Kabyemela, B. M.; Adachi, T.; Malaluan, R. M.; Arai, K. Glucose and Fructose Decomposition in Subcritical and Supercritical Water: Detailed Reaction Pathway, Mechanims, and Kinetics. Ind. Eng. Chem. Res. 1999, 38, 2888−2895. (26) Matsumura, Y.; Yanachi, S.; Yoshida, T. Glucose Decomposition Kinetics in Water at 25 MPa in the Temperature Range of 448−673 K. Ind. Eng. Chem. Res. 2006, 45, 1875−1879. (27) Yu, Y.; Wu, H. Kinetics and Mechanism of Glucose Decomposition in Hot-Compressed Water: Effect of Initial Glucose Concentration. Ind. Eng. Chem. Res. 2011, 50, 10500−10508. (28) Kimura, H.; Nakahara, N.; Matubayashi, N. Noncatalytic hydrothermal elimination of the terminal D-glucose unit from maltoand cello-oligosaccharides through transformation to D-Fructose. J. Phys. Chem. A 2012, 116, 10039−10049. (29) Kong, Z.; Liaw, S. B.; Gao, X.; Yu, Y.; Wu, H. Leaching characteristics of inherent inorganic nutrients in biochars from the slow and fast pyrolysis of mallee biomass. Fuel 2014, 128, 433−441. (30) Liu, L.; Sun, J.; Cai, C.; Wang, S.; Pei, H.; Zhang, J. Corn stover pretreatment by inorganic salts and its effects on hemicellulose and cellulose degradation. Bioresour. Technol. 2009, 100, 5865−5871. (31) Yu, Q.; Zhuang, X.; Yuan, Z.; Qi, W.; Wang, Q.; Tan, X. The effect of metal salts on the decomposition of sweet sorghum bagasse in flow-through liquid hot water. Bioresour. Technol. 2011, 102, 3445− 3450. (32) Kang, K. E.; Park, D.-H.; Jeong, G.-T. Effects of inorganic salts on pretreatment of Miscanthus straw. Bioresour. Technol. 2013, 132, 160−165.

(33) Liu, C.; Wyman, C. E. The enhancement of xylose monomer and xylotriose degradation by inorganic salts in aqueous solutions at 180 °C. Carbohyd. Res. 2006, 341, 2550−2556. (34) Wu, X.; Fu, J.; Lu, X. Hydrothermal decomposition of glucose and fructose with inorganic and organic salts. Bioresour. Technol. 2012, 119, 48−54. (35) Maeda, M.; Hisada, O.; Kinjo, Y.; Ito, K. Estimation of salt and temperature effects on ion product of water in aqueous solution. Bull. Chem. Soc. Jpn. 1987, 60, 3233−3239. (36) Valiskó, M.; Boda, D. The effect of concentration- and temperature-dependent dielectric constant on the activity coefficient of NaCl electrolyte solutions. J. Chem. Phys. 2014, 140, 234508. (37) Torry, L. A.; Kaminsky, R.; Klein, M. T.; Klotz, M. R. The Effect of Salts on Hydrolysis in Supercritical and Near-Critical Water: Reactivity and Availability. J. Supercrit. Fluids 1992, 5, 163−168. (38) Akiya, N.; Savage, P. E. Roles of Water for Chemical Reactions in High-Temperature Water. Chem. Rev. 2002, 102, 2725−2750. (39) Hunter, S. E.; Savage, P. E. Recent advances in acid-and basecatalyzed organic synthesis in high-temperature liquid water. Chem. Eng. Sci. 2004, 59, 4903−4909. (40) Angyal, S. J.; Davies, K. P. Complexing of Sugars with Metal Ions. J. Chem. Soc. D 1971, 500−501. (41) Angyal, S. J. Complex formation between sugars and metal ions. Pure Appl. Chem. 1973, 35, 131−146. (42) Gyurcsik, B.; Nagy, L. Carbohydrates as ligands: coordination equilibria and structure of the metal complexes. Coord. Chem. Rev. 2000, 203, 81−149. (43) Banipal, P. K.; Chahal nee Hundal, A. K.; Banipal, T. S. Effect of magnesium chloride (2:1 electrolyte) on the aqueous solution behavior of some saccharides over the temperature range of 288.15−318.15 K: a volumetric approach. Carbohydr. Res. 2010, 345, 2262−2271. (44) Choudhary, V.; Sander, S. I.; Vlachos, D. G. Conversion of Xylose to Furfural Using Lewis and Brønsted Acid Catalysts in Aqueous Media. ACS Catal. 2012, 2, 2022−2028. (45) Baes, C. F.; Mesmer, R. E. The Hydrolysis of Cations; Wiley: New York, 1976. (46) Frantz, J. D.; Marshall, W. L. Electrical Conductances and Ionization Constants of Calcium Chloride and Magnesium Chloride in Aqueous Solutions at Temperatures to 600 °C and Pressures to 400 bar. Am. J. Sci. 1982, 282, 1666−1693. (47) Palmer, D. A.; Wesolowski, D. J. Potentiometric Measurements of the First Hydrolysis Quotient of Magnesium(II) to 250 °C and 5 Molal Ionic Strength (NaCI). J. Solution Chem. 1997, 26, 217. (48) Bandura, A. V.; Lvov, S. N. The ionization constant of water over wide ranges of temperature and density. J. Phys. Chem. Ref. Data 2006, 35, 15−30. (49) Marcotullio, G.; De Jong, W. Chloride ions enhance furfural formation from D-xylose in dilute aqueous acidic solutions. Green Chem. 2010, 12, 1739−1746. (50) Tyrlik, S. K.; Szerszen, D.; Szymanski, S. The existence of the anions-guiding-rail in hexoses-C-13 NMR-study of Cl- donation on C1, C-3 and C-5 in hexoses with axial hydrogens connected to these carbons. New J. Chem. 1995, 19, 1019−1021. (51) Mohd Shafie, Z.; Yu, Y.; Wu, H. Effect of Initial pH on Hydrothermal Decomposition of Cellobiose under Weakly Acidic Conditions. Fuel 2015, DOI: 10.1016/j.fuel.2015.05.023.

J

DOI: 10.1021/acs.iecr.5b01007 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX