Insights into Hydrothermal Decomposition of Cellobiose in Gamma

Jun 21, 2017 - The results clearly demonstrate that, in comparison to water only, GVL/water solvent is a promising reaction medium for glucose recover...
0 downloads 14 Views 1MB Size
Article pubs.acs.org/IECR

Insights into Hydrothermal Decomposition of Cellobiose in GammaValerolactone/Water Mixtures Bing Song, Yun Yu,* and Hongwei Wu* Department of Chemical Engineering, Curtin University, GPO Box U1987, Perth Western Australia 6845, Australia ABSTRACT: This study provides new insights into the solvent effect of gamma-valerolactone (GVL) during hydrothermal decomposition of cellobiose at 150−200 °C in GVL/water mixtures. It was found that GVL addition (even at a small concentration of 0.03%) strongly suppresses the isomerization reaction, which is the dominant primary reaction of cellobiose decomposition in water. In GVL/water mixtures, the selectivity of primary hydrolysis reaction rapidly increases from ∼10% to ∼80% when the GVL concentration increases from 0 to 10%. Due to the enhanced hydrolysis reaction in GVL/water mixtures, this study achieves a maximal glucose yield of ∼66%, which is the highest glucose yield reported so far from cellobiose decomposition under acid-free conditions. The results clearly demonstrate that, in comparison to water only, GVL/water solvent is a promising reaction medium for glucose recovery under acid-free conditions, obviously due to the significantly enhanced hydrolysis reaction and suppressed isomerization reaction in GVL/water mixtures.

1. INTRODUCTION Lignocellulosic biomass has been widely considered to produce renewable biofuels and biochemicals via various biorefinery processes.1 Because of the recalcitrance of lignocellulosic biomass, the biorefinery processes generally suffer from either slow conversion or low selectivity, making the process economically unfeasible.2 Various catalysts and solvents have been shown to improve the conversion or selectivity,3,4 but at the expense of high process costs. To make biorefinery sustainable, it is of critical importance to develop efficient processes to convert biomass into high-value biofuels or biochemicals without significantly increasing the process costs. Sugar monomers (i.e., glucose and xylose) are important intermediate products for the subsequent production of biofuels or key platform biochemicals.5,6 However, the current sugar production process is expensive, requiring either enzymes or strong acids as catalysts.7 Recently, gamma-valerolactone (GVL) has been reported as a green solvent for biomass fractionation, sugar production, and the production of biochemicals from sugar compounds.8−12 It was also found that high yields of monosaccharides can be achieved from lignocellulosic biomass via hydrothermal processing in GVL/ water mixture at low temperatures (150−210 °C) under acidic conditions, with a production cost economically competitive with the traditional sugar production processes.13 GVL itself can also be produced from biomass-derived sugar monomers by catalytic conversion,14 making the process based on GVL/water system more sustainable. While GVL has been widely used for biomass hydrothermal processing,9,14,15 the underlying reaction mechanism during biomass hydrothermal processing in GVL/water system remains unclear. Under acidic conditions, GVL can increase © XXXX American Chemical Society

the hydrolysis reaction rate and reduce the activation energy for biomass hydrolysis.9 However, little has been reported on the effect of GVL during biomass hydrothermal processing under acid-free conditions. In this regard, this study employs cellobiose as a model compound of biomass to investigate the fundamental reaction mechanisms of cellobiose decomposition in GVL/water mixtures under acid-free conditions.

2. EXPERIMENTAL SECTION 2.1. Materials and Experimental Procedures. D(+)-cellobiose (≥99.0%) and D-(+)-glucose (≥99.9%) together with other standards and reagents used in this study were obtained from Sigma-Aldrich, except cellobiulose (glucosylfructose, GF), which was synthesized by LC Scientific Inc. (Canada). In this study, the cellobiose decomposition experiments were performed at 150−200 °C in water and GVL/water mixtures using a batch reactor, which was made from 316 stainless steel tubing and fittings from Swagelok, similar to those used elsewhere.16,17 Cellobiose solutions were prepared at a concentration of 1 g/L in water and various GVL/water mixtures (at a wide range of GVL concentration of 0.03−75% on a volume basis). In each experiment, ∼10 mL of cellobiose solution was loaded into the reactor, and the loaded solution was calculated to occupy 95% of the reactor volume under reaction conditions. After purging with helium to remove the air in the reactor, the Received: Revised: Accepted: Published: A

May 16, 2017 June 20, 2017 June 20, 2017 June 21, 2017 DOI: 10.1021/acs.iecr.7b02012 Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX

Article

Industrial & Engineering Chemistry Research

Figure 1. Conversion of cellobiose and yields of glucose and glucosyl-fructose (GF) at 150−200 °C in water and GVL/water mixtures: (a) water; (b) 0.03% GVL; (c) 0.3% GVL; (d) 1% GVL; (e) 5% GVL; (f) 10% GVL; (g) 25% GVL; (h) 50% GVL; (i) 75% GVL.

reactor was placed vertically in a fluidized sand bath (model: Techne SBL-2) and preheated to desired reaction temperatures in 3 min. After holding at the reaction temperature for a desired reaction time, the reactor was lifted out of the sand bath and placed in an ice water bath for rapid cooling to room temperature. After the experiment, the liquid sample was collected and analyzed by high performance anion exchange chromatography with pulsed amperometric detection (HPAEC-PAD) via a Thermo Fisher ICS-5000 system equipped with a CarboPac PA20 column, following a method detailed elsewhere.18 It should be noted that all the samples were extensively diluted (i.e., by 200−1000 times depending on GVL concentration) to satisfy the requirements for sample analysis using HPAEC-PAD19 and eliminate the interference of GVL during HPAEC-PAD analysis. As a result, the method is not able to detect some reaction products with low concentrations (e.g., retro-aldol reaction products) in the sample because such extensive dilutions substantially reduce the concentrations of those products. 2.2. Data Processing. The conversion of cellobiose (X), the yield (Yi), and the selectivity (Si) of a compound i at a residence time of t can be calculated on a carbon basis, using the following equations: X=

C(0) − C(t ) C(0)

(1)

Yi =

Ciai C(0)a

(2)

Si =

Yi ciai = X [C(0) − C(t )]a

(3)

where Ci is the concentration of a compound i (mg L−1) in the product; C(0) is the concentration of cellobiose (mg L−1) in the reactant; C(t) is the concentration of cellobiose (mg L−1) in the product collected at a reaction time t; and ai and a are the carbon contents (wt %) of the compound i and cellobiose, respectively. Assuming the cellobiose decomposition follows first-order kinetics, the reaction rate k (min−1) of cellobiose decomposition can be determined by eq 4:

−ln

C(t ) = kt C(0)

(4)

where t (min) is the reaction time.

3. RESULTS AND DISCUSSION 3.1. Conversion of Cellobiose and Yields of Primary Products during Cellobiose Decomposition in Water and GVL/Water Mixtures. At temperatures below 300 °C, cellobiose decomposition in hot-compressed water proceeds via four primary reactions, including isomerization to produce cellobiulose (glucosyl-fructose, GF) at a selectivity of 63−81% and glucosyl-mannose (GM) at a selectivity of 8−12%, the hydrolysis reaction to produce glucose at a selectivity of 6− 27%, and the retro-aldol reaction to glucosyl-erythrose (GE) and glycolaldehyde at a selectivity of