Formation of in situ sulfurous acid from the decomposition of butadiene sulfone in water at temperatures above 90 °C and 6 h. Sulfurous acid acts as a Brønsted acid catalyst and butadiene sulfone as an organic solvent in the deconstruction of lignocellulosic biomass. (Plant cell wall depiction by U.S. DOE Genomic Science Program modi fi ed from a previous study. 18 ) 

Context in source publication

Context 1

... depletion of non-renewable sources of energy and rising oil prices have prompted a surge in research to develop alternatives to fuels and chemicals from renewable materials. Biofuels and chemicals production from edible biomass, such as corn or sugarcane, is a mature and reliable technology. However, these feedstocks are not a long-term energy solution because suitable cropland is not readily available and may compete with food production leading to higher food prices. 1 On the other hand, non-food biomass sources ( i.e. lignocellulose) are renewable and sustainable, do not interfere with the food chain and can grow on unfertile lands. 1 Miscanthus × giganteus , a C 4 perennial grass, is an energy crop with low nutrient and water requirements, and more productive in biomass per acre than other energy crops, such as switchgrass or poplar. 2,3 Researchers are attempting to break the complex and pro- tective chemical structure of lignocellulosic biomass in order to isolate its fundamental sugar units which can be trans- formed into biofuels and chemicals. Cellulose and hemicellulose are the main sugar polymers in biomass and, together with lignin, represent the key constituents of plant cell walls. The function of cellulose (glucan) (30 – 50%), which consists of glucose monomers connected by β (1 – 4) hydrogen bond linkages, is to provide structure to the plant. Hemicellulose (20 – 30%) is an amorphous and complex mixture of oligosaccharides intertwined among the crystalline cellulose fibers and linked to cellulose via hydrogen bonds. The sugar monomers of hemicellulose are the pentoses xylose and arabinose and the hexoses glucose, galactose and mannose; and with xylose being the predominant monomer in Miscanthus , 4 we mostly refer to its hemicellulosic component as xylan. Lignin (10 – 30%), the third component of importance in biomass, is a non-sugar polyphenolic polymer consisting primarily of p -cou- maryl, coniferyl and sinapyl alcohol monomers chiefly linked via ether linkages. 5,6 Lignin covalently links to hemicellulose mostly through ester linkages 7 and protects both cellulose and hemicellulose from enzymatic attack. 8 “ Pretreatment ” is the term used for the physical, biological, chemical or physicochemical process 8 of deconstructing the cell wall matrix to remove or alter the lignin and hemicellulose struc- tures, 9,10 and to preserve the cellulose fraction. 11 It is the step before hydrolysis of the sugar biopolymers into simple monosac- charides, via enzymes or organic/inorganic chemicals. Physical pretreatments aim at size reduction and mechanical decrystalliza- tion of the plant cell wall constituents ( e.g. , milling), whereas biological pretreatments are based on natural wood-attacking microorganisms that grow on the biomass and have the ability to degrade lignin. 8 Chemical pretreatments are generally more e ff ective in removing a greater amount of lignin and/or hemicellulose, 12 and in exposing or solubilizing cellulose ( e.g. , alkali, dilute and concentrated Brønsted acids, peroxides, organic solvents, ionic liquids). 13 – 16 Physicochemical pretreatments generally combine physical and chemical methods by treating the biomass with a chemical under high pressure, followed by rapid decompression ( e.g. , ammonia fiber expansion (AFEX), CO 2 explosion). 17 All these methods present advantages and disadvan- tages, and the reader is referred to comprehensive reviews on the matter. 14 – 18 However, a common theme is that pretreatment remains the most expensive process step in the production of biofuels and chemicals from lignocellulosic biomass. This paper explores the potential of switchable butadiene sulfone and water (BSW) as a feasible chemical pretreatment of lignocellulosic biomass to remove hemicellulose and lignin and expose cellulose for processing of these biopolymers to value-added products. This solvent combination was tested by M. Kassner for the pretreatment of corn stover. The author showed SEM micrographs as evidence of destruction of biomass; however, no chemical analysis was performed due to separation di ffi culties. 19 To the best of our knowledge, there have been no other studies exploring the possibilities of BSW as a lignocellulose pretreatment option. Butadiene sulfone (BS) or sulfolene (C 4 H 6 O 2 S) is a β , γ -unsaturated cyclic sulfone (m.p. 65 °C) 20 with the unique ability to “ switch ” from solvent to gaseous 1,3-butadiene and sulfur dioxide in a reversible reaction 21,22 that favors BS below 100 °C and the gases above 100 °C. 23 In the presence of water, sulfur dioxide forms in situ sulfurous acid 23 which can act as a catalyst in deconstructing lignocellulosic biomass, 19 and as such, it can break ester bonds connecting hemicellulose and lignin, with the delocalization of the amorphous structure of hemicellulose and subsequent acid hydrolysis. 7 The involvement of BS in lignin solubilization, during or after xylan hydrolysis, is dis- cussed in this study (Fig. 1). The “ switchable ” capacity of BS also allows for recovery of the sulfur dioxide and butadiene decomposition gases for recombination into the solvent at potentially high yields. These gases are the raw materials for commercial production of BS and the current large scale availability for both production and recovery of BS 24 will allow for a potential transfer of these operations into a biorefinery. The facile and scalable recyclability of BS is an advantage over traditional acid/alkali pretreatments and novel ionic-liquid based pretreatments. The former would involve neutralization and further processing to remove contaminant salt by-products 23,25 and the latter require di ffi cult recycling processes not yet at large scale operations. 26 Another advantage of BSW pretreatment is the inexpensive commercial availability of BS, compared for instance, to ionic liquids, which are 5 – 20 times more expensive than conventional solvents at the laboratory scale 27 and require in many cases the addition of an acid-catalyst. 28 For example, from the Sigma-Aldrich catalog, 500 g of BS (98%) cost ∼ US$50 and 50 g of the e ff ective ionic liquid 1-ethyl-3-methylimidazolium acetate (97%) cost ∼ US$800. Pretreating lignocellulosic biomass with BSW is not without challenges. The decomposition gases tend to polymerize into mostly diene polymers, causing loss of the solvent and possible lower sugar yields during enzymatic processes due to the inhibition e ff ect of dienes. Also, carbohydrate degradation can occur at low water concentrations. Solutions for both challenges are addressed in this study. Our work represents the first quantitative analysis on the e ff ectiveness of BSW pretreatment of biomass ( Miscanthus ) through improved reaction conditions and good separation of the pretreated Miscanthus solids from the liquid. After pretreatment, the in situ production of sulfurous acid removed most of the xylan in Miscanthus and BS was involved in the solubilization of more than half of the lignin, with minor glucan degradation. We explored the decomposition of BS from the liquid after pretreatment for potential solvent reformation and for the isolation of lignin and hydrolyzed xylan. Butadiene sulfone (98%), Avicel® PH-101 (microcrystalline cellulose), xylan (from birchwood), lignin (alkali), cellobiose (99%), glucose (99%) and xylose (99%) were purchased from Sigma-Aldrich (St. Louis, MO) and used as received. Miscanthus × giganteus was provided by the Energy Biosciences Institute at the University of Illinois at Urbana-Champaign. Miscanthus was milled to pass a 1 mm screen using a knife mill (Retsch GmbH, model SM 2000, Haan, Germany). The particle size was standardized to a range of − 20/+80 mesh sieve (ASTM, U.S. Standard Sieve Series, Soiltest, Inc., Evan- ston, IL) using a Ro-Tap shaker (model D-4325, Dual Manufac- turing Co. Inc., Chicago, IL). Biomass was then packaged in Ziploc® bags and stored in a cold room at 0 °C until use. Fol- lowing the solids loading used by Kassner, 19 0.5 g of biomass, 10 g of butadiene sulfone, and 0 – 5 g of deionized water were added to a 140 mL glass pressure tube and capped with a PTFE plug (#8648-30 and #5845-47, respectively, Ace Glass, Vineland, NJ). The tube was preheated in an oil bath at 80 °C for 15 minutes and transferred to another oil bath on top of an Isotemp® stirring hotplate (Fisher Scientific, Pittsburgh, PA) that provided temperature and stirring control at 90 °C, 100 °C and 110 °C from 6 to 30 h at 600 rpm. After pretreatment, the solution was vacuum filtered using previously tared filtering crucibles (Coors #60531, Golden, CO). During filtration, a heat gun was used to keep the solution liquid 19 at a temperature of 70 – 75 °C. The filtrate was stored and all remaining solids in the tubes were transferred to the crucibles using at least 50 mL of hot deionized water, which also served as a rinsing medium. Crucibles and solids were dried for 12 h at 110 °C for gravimetric analysis and subsequent chemical composition analysis. The composition of glucan, xylan and lignin in raw Miscanthus was determined according to the NREL Laboratory Analytical Procedures (LAP) “ Determination of structural carbohydrates and lignin in biomass ” 29 (Table 1). As prerequisite for chemical composition of raw biomass and to avoid analysis interfer- ence, moisture content of biomass was less than 10% and non-structural elements were removed via a two-step extraction with water and ethanol (95% v/v) for 80 and 120 minutes, respectively, with an automatic Soxhlet apparatus (Soxtec TM System HT6 1043 extraction unit and 1046 service unit, FOSS, Hillerød, Denmark). The temperature of the Soxtec oil service unit was 150 °C for water extraction and 140 °C for ethanol extraction. The amount of extractable material was determined gravimetrically and reported as % extractives. 30 The extractive- free solids were dried for 24 h at 30 °C to a moisture content of 10% or less prior to ...