Cellulose dissolution and regeneration in ionic liquids: A computational perspective

Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Chemical Engineering Science journal homepage: www.elsevier.com/locate/ces

Cellulose dissolution and regeneration in ionic liquids: A computational perspective Krishna M. Gupta, Jianwen Jiang n Department of Chemical and Biomolecular Engineering, National University of Singapore, 117576, Singapore

H I G H L I G H T S







Computational studies for cellulose dissolution/regeneration in ILs are reviewed. Hydrogen-bonding plays a paramount role in both dissolution and regeneration. Cosolvents facilitate hydrogen-bonding of cellulose–IL and enhance dissolution. Anti-solvents destruct hydrogen-bonding of cellulose–IL and cause regeneration.

art ic l e i nf o

a b s t r a c t

Article history: Received 29 April 2014 Received in revised form 5 July 2014 Accepted 12 July 2014

To meet the increasing global energy demand and reduce the dependency on traditional fossil fuels, renewable biomass particularly cellulose has attracted considerable interest. Prior to processing and conversion into valuable products, cellulose needs to be pretreated (dissolved and then regenerated) via an environmentally benign route. Emerging as versatile solvents, ionic liquids (ILs) have been extensively examined for cellulose dissolution/regeneration. However, the underlying mechanisms of cellulose dissolution/regeneration in ILs remain elusive and the key governing factors are not fully understood at a microscopic level. This review summarizes the recent computational studies on cellulose pretreatment, including cellulose dissolution in neat ILs and IL/solvent mixtures, as well as cellulose regeneration by anti-solvents. Atom-resolution and time-resolved insights are provided to microscopically and fundamentally elucidate cellulose dissolution/regeneration, which are indispensable in the rational screening and design of new ILs for efficient cellulose pretreatment. Furthermore, the challenges for future computational exploration in this field are discussed. & 2014 Elsevier Ltd. All rights reserved.

Keywords: Solvent Anion Cation Hydrogen-bonding Interaction

1. Introduction 1.1. Biomass and cellulose Biomass is an abundant resource of renewable feedstock on the Earth (Graziani and Fornasiero, 2007). Substantial attention has been received towards the development of technically feasible methods to convert biomass into valuable products such as biofuels, chemicals and biomaterials (Muhammad et al., 2012). Particularly, the conversion of biomass to biofuels is appealing, which will reduce not only the dependence on fossil fuels, but also environmental pollution (Kunkes et al., 2008). The U.S. Department of Energy has set a target to achieve by 2025, i.e., nearly 30% transportation fuel is biofuels and 25% of organic compounds are renewable biochemicals (Ragauskas et al., 2006).

n

Corresponding author. E-mail address: [email protected] (J. Jiang).

Plants and plant-based (lignocellulosic) biomass contains three key components: cellulose, hemicellulose and lignin. The percentages of these components are approximately 50, 25 and 25 wt%, respectively; though they may vary depending on the source of finding and growth condition. Cellulose is a linear polymer of anhydroglucose units and exists in either crystalline or amorphous form; hemicellulose is an oligomer of glucose and xylose, composed of amorphous monosaccharide units; lignin is a cross-linked complex polymer mainly including syringyl and sinapyl units. The mechanical strength of plants is largely attributed to the lignin, which holds cell walls together and acts as a barrier preventing enzymatic attack to cellulose and hemicellulose (Petrus and Noordermeer, 2006). As the major component of biomass, cellulose has a global quantity of 700,000 billion tons. Nevertheless, only 0.1 billion tons of cellulose is currently being used for the production of paper, textiles, pharmaceutical compounds, etc. (Olivier-Bourbigou et al., 2010). Thus, a large amount of cellulose is still untapped. Cellulose is a polysaccharide composed of linear chains, from several hundred

http://dx.doi.org/10.1016/j.ces.2014.07.025 0009-2509/& 2014 Elsevier Ltd. All rights reserved.

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i

K.M. Gupta, J. Jiang / Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

2

to over ten thousand, linked by β (1-4) D-glucose units (i.e. glucosidic linkage). The hydroxyl groups along the chains are connected via hydrogen-bonds (H-bonds) in both parallel and anti-parallel fashion, as illustrated in Fig. 1. As a consequence of the H-bond network, cellulose possesses strong mechanical strength and cannot be easily dissolved in common solvents. For the widespread utilization of cellulose, the prime step in processing is cellulose dissolution. Traditionally, two types of solvents (derivatizing and nonderivatizing) are suggested to dissolve cellulose. The derivatizing solvents such as sodium hydroxide/carbon disulfide or sodium hydroxide/urea mixtures interact chemically with the hydroxyl groups of cellulose and form intermediates (Klemm et al., 2005; Pinkert et al., 2009). In contrast, the nonderivatzing solvents such as N-methylmorpholine-N-oxide monohydrate, N,Ndimethylacetamide/LiCl and dimethylsulfoxide (DMSO)/tetrabutylammonium fluoride trihydrate (TBAF) do not form intermediates (Hermanutz et al., 2008; Rosenau et al., 2001; Li and Zhao, 2007). Although these solvents are available and used by industry, they are not environmentally benign due to the lack of recyclability and the requirement of high temperature and pressure to operate (Zhu et al., 2006). Therefore, there is a critical need to develop alternative solvents to dissolve cellulose.

traditional volatile solvents and hence classified as “green” solvents for a broad spectrum of potential applications in both industrial-scale (Plechkova and Seddon, 2008) and laboratory scale (Olivier-Bourbigou et al., 2010). Among various applications schematically demonstrated in Fig. 3, ILs have been recommended for cellulose dissolution and regeneration. The first attempt using IL for cellulose dissolution was dated back to 1934 by Graenacher, who used N-ethylpyridinium chloride in the presence of N-containing bases (Graenacher, 1934). At that time, however, the practical importance of ILs was not realized. Only in 2002, Swatloski et al. found that 1-n-butyl-3methylimidazolium chloride [C4mim][Cl] could dissolve cellulose up to 25 wt% by microwave heating. They further reported that the dissolved cellulose could be readily regenerated by adding antisolvents such as water, ethanol, and acetone (Swatloski et al., 2002). Thereafter, 1-n-allyl-3-methylimidazolium chloride [Amim] Cl was tested for cellulose dissolution as well as regeneration (Zhang et al., 2005). Cellulose dissolution in six Cl  and [Ac]  -based ILs and its regeneration using water were investigated

1.2. Ionic liquids As a new class of solvents, ionic liquids (ILs) have attracted considerable interest. ILs are unique ionic materials with melting temperatures lower than 100 1C, substantially lower than normal salts (e.g. NaCl). If the melting temperatures are below room temperature, they are coined as room temperature ILs (RTILs). The most common cations in ILs are bulky, asymmetric and organic in nature, such as imidazolium, pyridinium, pyrrolidinium, ammonium, phosphonium, piperidinium, pyrazolium, thiazolium, and sulfonium. Anions may range from simple halides, inorganic ions to large organic ions. Fig. 2 depicts the chemical structures of typical cations and anions in ILs (Plechkova and Seddon, 2008). The salient characteristics distinguishing ILs from conventional solvents are the wide range of melting temperature ( 40 to 400 1C), high thermal stability (up to 400 1C), low vapor pressure, weakly coordinating properties, low flammability, high conductivity (both ionic and thermal), and broad electrochemical potential window (  4 to 4 V). Their physical and chemical properties can be tuned by the permutation of cations and anions, which is barely possible in conventional solvents (Freudenmann et al., 2011). Therefore, ILs have been considered as a good substitute for

Fig. 2. Typical cations and anions in ILs.

Intra-chain H-bond

Inter-chain H-bond

Plant Biomass

Cellulose Fig. 1. Cellulose network in plant biomass.

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i

K.M. Gupta, J. Jiang / Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Liquid Crystals

focused on cellulose regeneration from cellulose/IL mixtures. Finally, the summary and outlook are outlined in Section 4.

2. Cellulose dissolution

Applications

Applications

Applications

Separation

3

Solvents

Applications

Lubricants Fig. 3. Potential applications of ILs.

(Zhao et al., 2009). Considering viscosity and melting point, ILs with stronger basicity anions such as formate, acetate, or phosphate were suggested as possible surrogate to dissolve cellulose under mild condition (Cao et al., 2009). Among a series of anions, [Ac]  -based ILs were found to be better candidates than other counterparts (Hermanutz et al., 2008; Fukaya et al., 2008; Cruz et al., 2012). On the other hand, the effect of cation on cellulose dissolution was also reported. Liebert observed that ILs based on imidazolium, pyridinium and ammonium cations are potential candidates; however, phosphonium and sulfonium-based are not (Liebert, 2010). By adding cosolvents (DMSO, DMF and DMA) into [C4mim][Ac], Xu et al. found that cellulose solubility was enhanced preferentially due to cation solvation (Xu et al., 2013). As pointed out in several reviews for extensive experimental studies (Pinkert et al., 2009; Li et al., 2011; Tadesse and Luque, 2011; Wang et al., 2012; Brandt et al., 2013), cellulose dissolution/ regeneration in ILs are affected by many factors, e.g., the type of anion and cation, the basicity and H-bonding capability of anion, the position and length of side chain in cation. Nevertheless, the mechanisms of cellulose dissolution/regeneration are not fully elucidated.

1.3. Scope of the review Computational studies have been conducted towards in-depth understanding of cellulose dissolution/regeneration in ILs. At a molecular level, computations not only provide microscopic insights from bottom-up, but also secure quantitative interpretation of experimental observations. As such, structure-function relationships can be developed on the basis of molecular description, and used to screen and design new materials of interest. In this review, we aim to summarize recent computational studies, as tabulated in Table 1, for cellulose dissolution/regeneration in ILs. Though not so extensive compared with experimental studies, it is important to acquire the current status of computational endeavors in this field. Consequently, one can understand the progress achieved, as well as the challenges in future studies. Following this section, the computational studies for cellulose dissolution are presented in Section 2. Subsequently, Section 3 is

As mentioned above, experimental studies have revealed that many complex factors come into play in cellulose dissolution. There were even conflicting observations, for example, cellulose solubility in ILs was found to increase with the capability of H-bonding (Vitz et al., 2009; Zhao et al., 2008); however, a reversed trend was observed (Zhao et al., 2008). While some experiments suggested that cellulose dissolution is mainly governed by the interactions with anions (Remsing et al., 2006, 2008), others revealed that cations play a major role (Remsing et al., 2010; Zhang et al., 2005). Recently, Lu et al. prepared 13 ILs with various cations but identical anion [Ac]  . From systematic measurements, they concluded that cation does affect cellulose dissolution (Lu et al., 2014). Thus, the fundamental mechanism of cellulose dissolution remains elusive. Towards this end, a number of computational studies have been conducted for cellulose dissolution in ILs and IL/solvent mixtures. 2.1. ILs Derecskei and Derecskei-Kovacs reported a simulation study to calculate the solubility parameters of cellulose oligomers and its various derivatives (methyl cellulose, hydroxypropyl cellulose and carboxymethyl cellulose), as well as those of N,N0 -dialkyl-imidazolium ILs with Cl  and [CF3COO]  anions. The parameters were found to change linearly with the degree of polymerization. Due to the small number of ILs examined and the exclusion of thermal effect, they were unable to describe cellulose dissolution in ILs simply by traditional solubility parameter approach (Derecskei and Derecskei-Kovacs, 2006). Subsequent computational studies were largely focused on structural and energetic properties. For example, Youngs et al. carried out molecular dynamics (MD) simulations to investigate cellulose solvation in 1,3-dimethylimidazolium chloride [C1mim] Cl. The cellulose was mimicked by an isolated glucose. They found that Cl  anions form H-bonds with glucose, while cations interact weakly with glucose. Moreover, the most likely coordination environment for glucose was suggested to be three –OH groups H-bonded with three Cl  anions, and another two –OH groups bonded with a fourth Cl  anion (Youngs et al., 2006). In a later study, they examined glucose solvation in [C1min]Cl with different concentrations of glucose. Fig. 4 shows the three-dimensional probability distributions of anions and cations around an isolated glucose molecule. Apparently, the primary solvation shell around glucose is predominantly H-bonded with Cl  anions, although weak H-bonding is also present between the oxygen atom of glucose and the acidic hydrogen of imidazolium ring. At both low and high concentrations, the coordination number of Cl  anions around glucose was found to be 4. Relative to the cation, the position of glucose was observed to be either above or below the plane of imidazolium ring. The dominant contribution to glucose– [C1min]Cl interaction is the H-bonding (electrostatic) between – OH groups and Cl  anions, whereas a small van der Waals interaction between glucose and cation also contributes. Thus, they concluded that the cation can be tuned to adjust glucose/ cellulose dissolution in ILs (Youngs et al., 2007). Similarly, MD simulations were performed by Moyna and coworkers for the dissolution of 5 and 10 wt% cellobiose in [C4mim]Cl. Static and thermodynamic properties were collectively calculated. The structural features and H-bonding patterns between cellobiose and [C4mim]Cl were found to be consistent

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i

K.M. Gupta, J. Jiang / Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

4

Table 1 Computational studies for cellulose dissolution/regeneration. Cellulose dissolution Cellulose model

IL or IL/Solvent

Method

Force field/basis set

Ref.

Glucose derivatives Glucose Glucose Cellobiose Cellobiose 5, 6, 10, 20-mer Oligomers Glucose 2, 4, 6-mer Oligomers Dimethoxy-glucose 10-mer Oligomer 10-mer Oligomer Iβ Crystal Microfibril Microfibril Microfibril Small bundle Cellotriose Glucose 3  3 Structure 10-mer Oligomer Iβ crystal Glucose

[Cnmim]Cl and [Cnmim][CF3COO] [C1mim]Cl [C2mim][Ac] [C4mim]Cl [C4mim]Cl [C2mim][Ac] [C1mim]Cl [C4mim]Cl [C2mim][Ac] [Cnmim]Cl (n¼ 2–10) and [C4mpy]Cl [C2mim] þ various anions [C4mim][PF6] and [C4mim][Ac] [C4mim]Cl [C2mim][Ac] [C4mim]Cl [C4mim]Cl, [C2mim][Ac] and [C1mim][DMP] 42000 ILs 320 ILs 750 ILs [C4mim][Ac]/DMSO, DMF, CH3OH and H2O [Cnmim]Cl/DMSO and H2O (n¼ 1, 4 and 8) [C4mim][Ac]/DMSO and H2O

MD MD MD DFT MD MD DFT-D MD/DFT DFT MD MD MD MD MD MD MD COSMO-RS COSMO-RS COSMO-RS MD/DFT MD MD

COMPASS OPLS-AA OPLS-AA 6-31G(d) OPLS GLYCAM 6-311þ þG(2d,2p) GLYCAM/311þG(d,p) 6-31þ G(d) GLYCAM GLYCAM AMBER GLYCAM06 GLYCAM06 CHARMM OPLS

GLYCAM/6-311þ þ G(d,p) OPLS OPLS-AA

Derecskei and Derecskei-Kovacs (2006) Youngs et al. (2006, 2007) Youngs et al. (2011) Novoselov et al. (2007) Liu et al. (2007) (Liu et al., 2010) Janesko (2011) Xu et al. (2012) Ding et al. (2012) Zhao et al. (2012) Zhao et al. (2013a) Gupta et al. (2011) Mostofian et al. (2011, 2014) Liu et al. (2012) Gross et al. (2011, 2012), Cho et al. (2011) Rabideau et al. (2013) Kahlen et al. (2010) Casas et al. (2012) Casas et al. (2013) Zhao et al. (2013b) Huo et al. (2013) Andanson et al. (2014)

Cellulose model

IL/Anti-solvent

Method

Force field/basis set

Ref.

20-mer Oligomer Dimethoxy-glucose 10-mer Oligomer 10-mer Oligomer

[C2mim][Ac]/water [C2mim][Ac]/water [C4mim][Ac]/water [C4mim][Ac]/water, ethanol and acetone

MD DFT MD MD/MP2

GLYCAM 6-31þ G(d) AMBER AMBER/6-311þ þ G(d,p)

Liu et al. (2011) Ding et al. (2012) Gupta et al. (2013a) Gupta et al. (2013b)

Cellulose regeneration

Fig. 4. Three-dimensional probability distributions of anions (red/solid surface) and cations (blue/wireframe surface) around an isolated glucose molecule. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Reproduced with permission: Copyright 2007, American Chemical Society.

with NMR relaxation data, indicating the –OH groups interact with Cl  anions in an 1:1 ratio (Liu et al., 2007). Singh and coworkers examined the behavior of cellulose represented by glucose oligomers in [C2mim][Ac]. From structural analysis, the –OH groups of polysaccharides were found to form strong H-bonds with [Ac]  anions, while some cations were in close contact with the polysaccharides via hydrophobic interaction. The polysaccharides–IL interaction was stronger than that with either water or methanol. Furthermore, the conformational flexibility of (1,4)-glycosidic linkage in the IL was increased as compared with aqueous environment (Liu et al., 2010). Zhang and coworkers systematically investigated the effects of cationic structure on cellulose dissolution. [Cnmim]Cl (n¼2–10) and [C4mpy]Cl were considered by varying the heterocyclic structure

and the alkyl chain length. [C4mpy]Cl exhibits a better dissolution capability than [C4mim]Cl. For [Cnmim]Cl, the shorter the alkyl chain, the higher is the solubility. It was revealed the inclusion of electron-withdrawing groups in cation enhances cellulose solubility. As demonstrated in Fig. 5, the allyl group of 1-allyl-3-methyl imidazolium chloride [Amim]Cl can form H-bonds with the –OH groups of cellulose, leading to a higher solubility. Based on such observation from MD simulation, they suggested some promising groups such as –OH, –C  C–, –C¼ C– and –R–O–R0 for cation inclusion (Zhao et al., 2012). Further, they reported the effects of anionic structure for ILs having the same cation [C2mim] þ but various anions. The interactions between cellulose and anions were found to decrease as Cl  4[Ac]  4[(CH3O)–PO2]  4[SCN]  4 [PF6]  . Combining the effects of electronegativity of H-bond

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i

K.M. Gupta, J. Jiang / Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

5

NO. of H-bond

Fig. 5. H-bonds between the –OH group of cellulose and the allyl group of [Amin]Cl. O: red, N: blue, C: gray; H: cyan; Cl: green. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Reproduced with permission: Copyright 2012, John Wiley and Sons.

80

80

80

70

70

70

60

60

60

50

50

50

40

40

30

30

20

Bulk Surface

10 0

0

1

2

3

Time (ns)

4

5

40 Bulk Surface

20

20

10

10

0

0

1

2

3

Time (ns)

4

Bulk Surface

30

5

0

0

1

2

3

4

5

Time (ns)

Fig. 6. Inter-chain H-bonds at the bulk and surface layers of cellulose in cellulose/solvent systems (a) [C4mim][PF6], (b) [C4mim][Ac] and (c) water. Reproduced with permission: Copyright 2011, Elsevier.

acceptor, alkyl chain length and electron-withdrawing group, they concluded that anions should possess high electron density, short alkyl chain and avoid electron-withdrawing group for efficient cellulose dissolution (Zhao et al., 2013a). In the above simulation studies discussed, cellulose was simply represented as glucose or a single oligomer. However, natural cellulose exists in crystalline form (sometimes amorphous) and thus more accurate description is desired. In this regard, Gupta et al. used Iβ crystalline structure of cellulose to investigate its dissolution in [C4mim][PF6] and [C4mim][Ac], as well as water. The predominant H-bonds within cellulose crystal (intra-chain O2H2∙∙∙O6 and O3H3∙∙∙O5 and the inter-chain O6H6∙∙∙O3) were identified by simulation. To define H-bonds, two geometrical criteria were adopted: (a) the distance between a donor and an acceptor o 0.35 nm and (b) the hydrogen-donor–acceptor angle o301. Upon contact with the two ILs or water, the number of the H-bonds at the bulk layer of cellulose remains nearly a constant. As shown in Fig. 6; however, there is a decrease in the number of inter-chain H-bonds at the surface layer of cellulose. This implies the breaking of H-bonds at the surface, as attributed to the

orientational change of  OH groups exerted by cellulose/solvent interaction. The simulation study reveals that solvation leads to the breaking of H-bonds at the cellulose surface and suggests that inter-chain H-bonding is critical to govern cellulose dissolution (Gupta et al., 2011). Mostofian et al. simulated the solvation of cellulose microfibril in [C4mim]Cl. For a 36-chain fibril in Iβ crystalline form, they found that the hydroxymethyl groups of cellulose surface are more exposed to IL as compared with water, and the hydrophilic surface is the major interaction site between cellulose and IL. The exposed surface of cellulose (glucose units) is surrounded by cations in a preferred parallel orientation (Mostofian et al., 2011). They further investigated the role of cations and anions at the initial stage of cellulose dissolution in [C4mim]Cl. The dominant polar interactions between Cl  anions and the surface –OH groups of cellulose disrupt the H-bonds in glucan chains. On the other hand, the cations reside on hydrophobic surface by non-polar interactions, which compensates the interactions between stacked cellulose layers and thus stabilizing detached glucan chains (Mostofian et al., 2014). In a similar manner, Singh and coworkers examined the

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i

K.M. Gupta, J. Jiang / Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

6

dissolution of microcrystalline cellulose Iβ in [C2mim][Ac]. The microfibril was mimicked by 3 layers of 3 cellulose chains in each layer, arranged in Iβ structure. Both cation and anion were observed to penetrate into cellulose crystal; particularly, the anion forms strong H-bonds with cellulose. The hydroxymethyl groups of cellulose dissolved in the IL tend to adopt a gauche-trans conformation, in contract to the preferential conformation in air or water (Liu et al., 2012). In addition to H-bonds analysis for cellulose dissolution, thermodynamic driving force for cellulose dissolution in ILs was explored by Chu and coworkers. By considering two extreme states of cellulose (a crystalline microfibril and dissociated glucan chains) in [Bmim]Cl, they revealed important molecular features in determining solubility: (1) the perturbation of solvent structure by dissolved glucan chains and (2) the interactions of both cation and anion with the moieties of glucan residues forming intersheet contacts. It was proposed that the indicative signature of an effective solvent for cellulose dissolution is the ability to interact with glucan chains along axial direction and disrupt the intersheet contacts of cellulose (Gross et al., 2011). Subsequently, they carried out simulations to unravel the mechanism of cellulose deconstruction by peeling off an 11-residues glucan chain from a cellulose

Fig. 7. Potentials of mean force (kcal/mol) for peeling off the corner glucan chain from cellulose microfibril in [Bmim]Cl and water. Reproduced with permission: Copyright 2011, American Chemical Society.

2 ns

microfibril. In [Bmim]Cl, the anions strongly interact with hydroxyl groups, whereas the cations couple to the side chains and linker oxygens in peeled-off state. Conversely, water couples to the hydroxyl and side-chain groups more strongly than the sugar rings and linker oxygens. As indicated by Fig. 7, the potential of mean force for peeling-off the glucan chain in [Bmim]Cl is  2 kcal/mol per glucose residue, more favorable than in water (Cho et al., 2011). They also examined the entropic contribution to cellulose dissolution via MD simulations and the two-phase thermodynamic model. In both [Bmim]Cl and water, the entropy associated with the solvent degree of freedom is reduced upon cellulose dissolution. Moreover, the solvent entropy reduction in [Bmim][Cl] is significantly smaller than in water. Both the total entropy change and the solvent–glucan interaction in [Bmim]Cl contribute favorably to cellulose dissolution (Gross et al., 2012). In a recent all-atom MD simulation study, Rabideau et al. explored the mechanism for the breakup of small cellulose bundles with Iα and Iβ structures in [C4mim]Cl, [C2mim][Ac] and [C1mim][DMP]. In all the three ILs, substantial breakup of the bundles was observed. As illustrated in Fig. 8, the anions are strongly bound to the –OH groups of the bundle and form negatively charged complex, which weakens the intrastrand H-bonds and reduces the strand rigidity. Meanwhile, the cations are associated with the complex, push the individual chains apart, and initiate the breakup. Among the three ILs, [C2mim][Ac] was found to have the potential to peel-off individual strands from the main bundle. Detailed analysis showed that the intrasheet H-bond breakage is important to peel-off individual chains despite the interstrand H-bonds breakage. Finally, they concluded that cations might play a significant role in the initial breakup of cellulose bundle during dissolution (Rabideau et al., 2013). Different from molecular simulation method applied in the above discussed studies, a predictive theory namely COSMO-RS (Klamt, 2005) was implemented to examine cellulose dissolution in ILs. Kahlen et al. described the combinatorial contributions to chemical potential by COSMOS-RS and estimated cellulose solubilities (modeled as cellotriose) in over 2000 ILs. They observed that anions rather than cations play a dominant role in cellulose dissolution. Qualitatively good agreement was found between the theoretical results and scarce experimental data. On this basis, some new ILs were suggested as potential candidates for cellulose dissolution (Kahlen et al., 2010). Similarly, Casas et al. used COSMOS-RS to analyze cellulose solubilities in 320 ILs from a

50 ns

55 ns

Fig. 8. Anions (top) and cations (bottom) near cellulose Iα bundle in [C4mim]Cl after 2 ns (a,d), 50 ns (b,e), and 55 ns (c,f). Reproduced with permission: Copyright 2013, American Chemical Society.

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i

K.M. Gupta, J. Jiang / Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

combination of 20 cations and 16 anions. Based on the intermolecular interactions contributing to mixing properties, the excess enthalpies of cellulose/IL mixtures were evaluated to assess the affinities of cellulose for different ILs. The dissolution is governed by the H-bonding between anions and cellulose, while the electrostatic and van der Waals interactions are insignificant. The most promising anions were identified to be acetate, formate and chloride (Casas et al., 2012). They carried forward COSMO-RS to predict cellulose solubilities in a wide collection of 750 ILs by combining 25 cations and 30 anions. The cellulose was represented by a 3  3 structure consisting of nine (M1–M9) monomers, all linked to the surroundings by H-bonds. The excess enthalpy and activity coefficient were used as reference parameters to characterize solubility. For good solubility, the excess enthalpy is generally negative (exothermic) and the activity coefficient is o1. Furthermore, a criterion for cellulose dissolution in ILs was proposed, i.e., cellulose is soluble if the activity coefficient o0.85 or vice versa. On this basis, the soluble and insoluble regions in 750 ILs were determined. From COSMO-RS analysis, the ILs favorable for dissolution readily act as H-bond acceptors; while those with large anions and dispersed charges tend to be poor solvents (Casas et al., 2013).

7

Fig. 9. Cosolvent effect on cellulose dissolution in [C4mim][Ac]. Reproduced with permission: Copyright 2013, American Chemical Society.

2.2. IL/solvent mixtures Xu et al. experimentally reported that cellulose dissolution in [C4mim][Ac] could be enhanced by adding aprotic polar solvents. The effects of molar ratio of solvent to IL, anionic structure and the nature of solvent (DMSO, DMF and DMA) were examined in detail. The enhancement was attributed to the increased number of free [Ac]  anions, as a result of the preferential solvation of [C4mim] þ by aprotic polar solvents (Xu et al., 2013). From molecular simulations and density-functional theory (DFT) calculations, Zhang and coworkers quantitatively examined the effects of four typical solvents (DMSO, DMF, CH3OH and H2O) on cellulose dissolution in [C4mim][Ac]. They found that cellulose dissolution in IL/solvent mixtures is governed by the electrostatic interaction and H-bonding between cellulose and [Ac]  anions, which may be substantially influenced by the addition of solvent. Dissolution is improved by adding the aprotic solvents (DMSO and DMF), which solvate the cations and anions thus breaking apart the cation–anion pairs. As a consequence, more unpaired [Ac]  anions participate in the H-bonding with cellulose and lead to a high solubility. Such a mechanism of cosolvent effect is schematically demonstrated in Fig. 9, and consistent with the experimental observation mentioned above. In contrast, the protic solvents (CH3OH and H2O) strongly solvate [Ac]  anions, inhibit the H-bonding between cellulose and [Ac]  , thus reducing cellulose solubility (Zhao et al., 2013b). Separately, Wang and coworkers presented a molecular view for the interfacial behavior between cellulose Iβ and ILs (EmimCl, BmimCl and OmimCl) upon adding solvents (DMSO and water). As indicated by the density profiles and pair energy distributions (PEDs), Cl  anions interact more strongly with the cellulose surface than cations, and form multiple H-bond patterns with the –OH groups of cellulose. In the presence of DMSO or water, the H-bond patterns and PEDs are noticeably altered. As shown in Fig. 10, the PED between Cl  and cellulose surface in neat Bmim/Cl exhibits a single band at approximately 37 kcal/mol. Upon adding 20 wt% of DMSO, the PED splits into two distinguishable bands with a major band at  40 kcal/mol and a minor band at 20 kcal/mol. Thus, the interaction between Cl  and cellulose is enhanced and leads to a higher solubility of cellulose. Upon adding 33% of water, in contrast, the major band of the PED shifts from 37 to  33 kcal/mol. This implies that the interaction between [Cl]  and cellulose is reduced, thus water is not a cosolvent for

Fig. 10. Pair energy distributions between Cl  and cellulose surface in neat BmimCl, BmimCl/DMSO and BmimCl/water mixtures. Reproduced with permission: Copyright 2013, American Chemical Society.

cellulose dissolution. They suggested that a combined analysis of both PEDs and H-bond patterns would provide insightful information for the selection and design of solvents for cellulose dissolution (Huo et al., 2013). Combining polarized-light optical microscopy and MD simulation, Andanson et al. attempted to understand the role of DMSO in the dissolution of microcrystalline cellulose in [C4mim][Ac]. The addition of DMSO was experimentally found to reduce the viscosity of IL and increase the ionic conductivity, without affecting the cation–anion dissociation. This observation was confirmed by simulation, showing that the H-bond network between glucose and [Ac]  anions is not significantly altered in the presence of 0.5 mol fraction of DMSO. Furthermore, there is no specific interaction between DMSO and glucose. Therefore, the improved cellulose dissolution by DMSO was attributed to the facilitated mass transport by reducing viscosity, while the cation–anion and IL–cellulose interactions are not substantially affected. As a comparison, water disrupts IL–cellulose interactions and thus cannot act as a cosolvent (Andanson et al., 2014).

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i

K.M. Gupta, J. Jiang / Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

8

3. Cellulose regeneration After dissolution in an IL, cellulose is required to be regenerated from cellulose/IL mixture for further processing; meanwhile, the IL can be recycled and reused. Towards this end, a handful of experimental studies have been reported. For example, Swatloski et al. used water, ethanol and acetone as anti-solvents to regenerate cellulose from [C4mim]Cl (Swatloski et al., 2002). By using water, Zhang et al. regenerated cellulose and recycled [Amim]Cl (Zhang et al., 2005). For fast enzymatic hydrolysis of cellulose, Zhao et al. conducted a systematic study on cellulose regeneration in several Cl  and [Ac]  -based ILs, and found that regenerated cellulose are less crystalline (Zhao et al., 2009). Recently, Hauru et al. examined cellulose regeneration by water from its mixtures with ILs including [C2mim][Ac], [TMGH][EtCO2], and [TMGH][Ac] (Hauru et al., 2012). These experiments reveal the easy regeneration of cellulose, but the mechanism of cellulose regeneration from ILs is not completely understood. Currently, there are only few computational studies reported to unravel the role of anti-solvents in cellulose regeneration. Ding et al. applied DFT calculations and experimental techniques to examine cellulose regeneration from [C2mim][Ac]. The H-bonds formed between the –OH groups of cellulose and [Ac]  anions were found to diminish by adding H2O. The regenerated cellulose was converted to structure II from structure I in original sample. They suggested that H2O is more preferentially to form H-bonds with [Ac]  ; as a consequence, the H-bonds between cellulose are connected again leading to precipitation (Ding et al., 2012). From MD simulations for cellulose in [C2mim][Ac]/H2O mixtures, Liu et al. demonstrated that the structure of [C2mim][Ac] is altered by H2O, which in turn disrupts the interaction between cellulose and [C2mim][Ac]. Specifically, the number of H-bonds between cellulose and H2O increases upon H2O diffusing within cellulose; meanwhile, the number of H-bonds between cellulose and anion decreases. Thus, the H-bonding network of cellulose-anion-H2O is formed as illustrated in Fig. 11. This is a key intermediate step, which displaces cation from the solvation shell of cellulose and leads to cellulose precipitation (Liu et al., 2011). In the above two computational studies, cellulose was mimicked as dimethoxy-glucose or a single oligomer. To realistically examine cellulose regeneration at a finite concentration, more accurate description is desired. Gupta et al. provided a molecular insight

into cellulose regeneration from cellulose/[C4mim][Ac] by using H2O as anti-solvent. The concentration of cellulose examined was 11.5 wt%, corresponding to experimental measured solubility at 40 1C. With increasing H2O percentage from 0 to 80 wt%, they found that the cellulose–[Ac]  interaction strength drops. The addition of H2O leads to the destruction of cellulose–[Ac]  H-bonds, and the subsequent formation of cellulose–cellulose H-bonds and [Ac]  –H2O H-bonds. On this basis, they proposed a mechanism as demonstrated in Fig. 12. The regenerated cellulose is amorphous, as evidenced by the significantly different torsional angle distributions of hydroxymethyl groups from crystalline structure. In addition, cellulose regeneration was found to be prompted at a higher temperature, due to the enhanced cellulose–cellulose interaction (Gupta et al., 2013a). Combining MD simulations and ab-inito calculations, Gupta et al. further explored the role of other anti-solvents (ethanol and acetone) for cellulose regeneration from cellulose/[C4mim][Ac]. As revealed by the structural analysis, cellulose–[C4mim][Ac] possess the strongest interactions in acetone, but the weakest in water. Nevertheless, cellulose–cellulose interact the most strongly in water, followed by ethanol and acetone. In water, the number of cellulose–[Ac]  H-bonds is the smallest and the number of cellulose–cellulose H-bonds is the largest. As confirmed by the strongest binding energy between [Ac]  and water, shown in Fig. 13, this study reveals that water is the most effective among the three solvents to break cellulose–[Ac]  H-bonds and lead to cellulose regeneration (Gupta et al., 2013b).

4. Summary and outlook This review presents an overview of recent computational studies on cellulose dissolution/regeneration in ILs. From a microscopic level, clearer and deeper understanding has been progressively achieved. For both dissolution and regeneration, H-bonding plays a paramount role. Dissolution in ILs is initiated by the disruption of H-bonds in cellulose, driven by the formation of Hbonds between cellulose and anions, as well as the hydrophobic interactions with cations. Upon adding cosolvents, cation–anion pairs are partially dissociated and more anions interact with cellulose, leading to enhanced dissolution. For regeneration, antisolvents possess strong affinity for ILs, destruct the H-bonds

H2O

Fig. 11. A key intermediate step of cellulose regeneration from [C2mim][Ac] by adding H2O. Reproduced with permission: Copyright 2011, American Chemical Society.

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i

K.M. Gupta, J. Jiang / Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

[Ac]−

9

[Ac]−

H2O

Fig. 12. A mechanism of cellulose regeneration by adding H2O in cellulose/[C4mim][Ac]. Reproduced with permission: Copyright 2013, Royal Society of Chemistry.

B.E. = −75.58

B.E. = −73.97

B.E.= −55.92



Fig. 13. Binding energies (kJ/mol) of [Ac] with (a) water, (b) ethanol and (c) acetone. Reproduced with permission: Copyright 2013, Royal Society of Chemistry.

between cellulose and ILs; consequently, cellulose–cellulose reform H-bonds and precipitate. To design new ILs for efficient cellulose dissolution, several factors should be taken into account, such as anions with high ability to form H-bonds, short alkychains on both cations and anions, non electron-withdrawing groups in anions, as well as electron-withdrawing groups in cations. For cellulose regeneration, anti-solvents are expected to interact with ILs more strongly than with cellulose. Obviously, computations have become an important tool to provide microscopic insights into cellulose dissolution/regeneration in ILs. Nevertheless, a few challenges exist for future computational exploration in this field. (1) While quantum chemical calculations are applicable to the basic units of cellulose (e.g. glucose and its derivatives), classical molecular simulations are mostly conducted for small cellulose oligomers or finite fibrils. Currently, it is formidable to calculate large units using high-level quantum chemistry methods or simulate long polysaccharide chains. With the growing of computational power, however, this hurdle will be overcome progressively. (2) The accuracy of molecular simulations exclusively relies on the force field used to represent the system of interest. Although the classical force fields (e.g. AMBER and OPLS) have been adopted primarily for imidazolium-based ILs, they cannot well represent the thermodynamic and structural properties of certain cations and anions. Therefore, transferable and reliable force fields are highly desired for a wide spectrum of ILs and further for cellulose/IL systems. (3) Most simulation studies are performed within a few nanoseconds. Such a time scale is not sufficient to observe the whole process of cellulose dissolution and regeneration. An alternative is to implement coarse-grained simulation method. A prerequisite

for doing this is, nevertheless, the availability of appropriate force fields at the coarse-grained level. (4) As discussed above, cellulose dissolution can be enhanced by cosolvents; on the other hand, cellulose is regenerated by anti-solvents. Currently, only a handful of solvents have been examined and there is a need to test the role of other solvents. Considering the vast number of possible solvents, the testing is a time-consuming task. To a certain extent, these challenges should be addressed in the future computational studies towards more accurate and reliable microscopic description of cellulose dissolution/regeneration in ILs.

Acknowledgments The authors are grateful to the National University of Singapore and the Ministry of Education of Singapore for financial support. References Andanson, J.-M., Bordes, E., Devemy, J., Leroux, F., Padua, A.A.H., Gomes, M.F.C., 2014. Understanding the role of cosolvents in the dissolution of cellulose in ionic liquids. Green Chem. 16, 2528–2538. Brandt, A., Grasvik, J., Hallett, J.P., Welton, T., 2013. Deconstruction of lignocellulosic biomass with ionic liquids. Green Chem. 15, 550–583. Cao, Y., Wu, J., Zhang, J., Li, H.Q., Zhang, Y., He, J.S., 2009. Room temperature ionic liquids: a new and versatile platform for cellulose processing and derivatization. Chem. Eng. J. 147, 13–21. Casas, A., Palomar, J., Alonso, M.V., Oliet, M., Omar, S., Rodriguez, F., 2012. Comparison of lignin and cellulose solubilities in ionic liquids by COSMO-RS analysis and experimental validation. Ind. Crops Prod. 37, 155–163. Casas, A., Omar, S., Palomar, J., Oliet, M., Alonso, M.V., Rodriguez, F., 2013. Relation between differential solubility of cellulose and lignin in ionic liquids and activity coefficients. RSC Adv. 3, 3453–3460.

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i

10

K.M. Gupta, J. Jiang / Chemical Engineering Science ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Cho, H.M., Gross, A.S., Chu, J.W., 2011. Dissecting force interactions in cellulose deconstruction reveals the required solvent versatility for overcoming biomass recalcitrance. J. Am. Chem. Soc. 133, 14033–14041. Cruz, H., Fanselow, M., Holbrey, J.D., Seddon, K.R., 2012. Determining relative rates of cellulose dissolution in ionic liquids through in situ viscosity measurement. Chem. Commun. 48, 5620–5622. Derecskei, B., Derecskei-Kovacs, A., 2006. Molecular dynamic studies of the compatibility of some cellulose derivatives with selected ionic liquids. Mol. Simul. 32, 109–115. Ding, Z.D., Chi, Z., Gu, W.X., Gu, S.M., Liu, J.H., Wang, H.J., 2012. Theoretical and experimental investigation on dissolution and regeneration of cellulose in ionic liquid. Carbohydr. Polym. 89, 7–16. Freudenmann, D., Wolf, S., Wolff, M., Feldmann, C., 2011. Ionic liquids: new perspectives for inorganic synthesis? Angew. Chem. Int. Ed. 50, 11050–11060. Fukaya, Y., Hayashi, K., Wada, M., Ohno, H., 2008. Cellulose dissolution with polar ionic liquids under mild conditions: required factors for anions. Green Chem. 10, 44–46. Graenacher, C., 1934. US Patent, 1943176. Graziani, M., Fornasiero, P., 2007. Renewable Resources and Renewable Energy: A Global Challenge. CRC Press/Taylor & Francis Group LLC, Boca Raton. Gross, A.S., Bell, A.T., Chu, J.-W., 2011. Thermodynamics of cellulose solvation in water and the ionic liquid 1-butyl-3-methylimidazolim chloride. J. Phys. Chem. B 115, 13433–13440. Gross, A.S., Bell, A.T., Chu, J.W., 2012. Entropy of cellulose dissolution in water and in the ionic liquid 1-butyl-3-methylimidazolim chloride. Phys. Chem. Chem. Phys. 14, 8425–8430. Gupta, K.M., Hu, Z., Jiang, J., 2011. Mechanistic understanding of interactions between cellulose and ionic liquids: a molecular simulation study. Polymer 52, 5904–5911. Gupta, K.M., Hu, Z., Jiang, J., 2013a. Molecular insight into cellulose regeneration from a cellulose/ionic liquid mixture: effects of water concentration and temperature. RSC Adv. 3, 4425–4433. Gupta, K.M., Hu, Z., Jiang, J., 2013b. Cellulose regeneration from a cellulose/ionic liquid mixture: the role of anti-solvents. RSC Adv. 3, 12794–12801. Hauru, L.K.J., Hummel, M., King, A.W.T., Kilpeläinen, I., Sixta, H., 2012. Role of solvent parameters in the regeneration of cellulose from ionic liquid solutions. Biomacromolecules 13, 2896–2905. Hermanutz, F., Gaehr, F., Uerdingen, E., Meister, F., Kosan, B., 2008. New developments in dissolving and processing of cellulose in ionic liquids. Macromol. Symp. 262, 23–27. Huo, F., Liu, Z., Wang, W., 2013. Cosolvent or antisolvent? a molecular view of the interface between ionic liquids and cellulose upon addition of another molecular solvent. J. Phys. Chem. B 117, 11780–11792. Janesko, B.G., 2011. Modeling interactions between lignocellulose and ionic liquids using DFT-D. Phys. Chem. Chem. Phys. 13, 11393–11401. Kahlen, J., Masuch, K., Leonhard, K., 2010. Modelling cellulose solubilities in ionic liquids using COSMO-RS. Green Chem. 12, 2172–2181. Klamt, A., 2005. COSMO-RS: From Quantum Chemistry to Fluid Phase Thermodynamics and Drug Design. Elsevier, Amsterdam. Klemm, D., Heublein, B., Fink, H.-P., Bohn, A., 2005. Cellulose: fascinating biopolymer and sustainable raw material. Angew. Chem. Int. Ed. 44, 3358–3393. Kunkes, E.L., Simonetti, D.A., West, R.M., Serrano-Ruiz, J.C., Gärtner, C.A., Dumesic, J. A., 2008. Catalytic conversion of biomass to monofunctional hydrocarbons and targeted liquid-fuel classes. Science 322, 417–421. Li, C., Zhao, Z.K., 2007. Efficient acid-catalyzed hydrolysis of cellulose in ionic liquid. Adv. Synth. Catal. 349, 1847–1850. Li, W.Y., Sun, N., Stoner, B., Jiang, X.Y., Lu, X.M., Rogers, R.D., 2011. Rapid dissolution of lignocellulosic biomass in ionic liquids using temperatures above the glass transition of lignin. Green Chem. 13, 2038–2047. Liebert, T., 2010. Cellulose solvents – remarkable history, bright future. ACS Symp. Ser. 1033, 3–54. Liu, H.B., Sale, K.L., Holmes, B.M., Simmons, B.A., Singh, S., 2010. Understanding the interactions of cellulose with ionic liquids: a molecular dynamics study. J. Phys. Chem. B 114, 4293–4301. Liu, H.B., Sale, K.L., Simmons, B.A., Singh, S., 2011. Molecular dynamics study of polysaccharides in binary solvent mixtures of an ionic liquid and water. J. Phys. Chem. B 115, 10251–10258. Liu, H.B., Cheng, G., Kent, M., Stavila, V., Simmons, B.A., Sale, K.L., Singh, S., 2012. Simulations reveal conformational changes of methylhydroxyl groups during dissolution of cellulose Iβ in ionic liquid 1-ethyl-3-methylimidazolium acetate. J. Phys. Chem. B 116, 8131–8138. Liu, Z.W., Remsing, R.C., Moore, P.B., Moyna, G., 2007. Molecular dynamics study of the mechanism of cellulose dissolution in the ionic liquid 1-n-butyl-3-methylimidazolium chloride. ACS Symp. Ser. 975, 335–350. Lu, B., Xu, A., Wang, J., 2014. Cation does matter: how cationic structure affects the dissolution of cellulose in ionic liquids. Green Chem. 16, 1326–1335. Mostofian, B., Smith, J., Cheng, X., 2011. The solvation structures of cellulose microfibrils in ionic liquids. Interdiscip. Sci. Comput. Life Sci. 3, 308–320. Mostofian, B., Smith, J., Cheng, X., 2014. Simulation of a cellulose fiber in ionic liquid suggests a synergistic approach to dissolution. Cellulose 21, 983–997.

Muhammad, N., Man, Z., Khalil, M.A.B., 2012. Ionic liquid—a future solvent for the enhanced uses of wood biomass. Eur. J. Wood Prod. 70, 125–133. Novoselov, N.P., Sashina, E.S., Petrenko, V.E., Zaborsky, M., 2007. Study of dissolution of cellulose in ionic liquids by computer modeling. Fibre Chem. 39, 153–158. Olivier-Bourbigou, H., Magna, L., Morvan, D., 2010. Ionic liquids and catalysis: recent progress from knowledge to applications. Appl. Catal. 373, 1–56. Petrus, L., Noordermeer, M.A., 2006. Biomass to biofuels, a chemical perspective. Green Chem. 8, 861–867. Pinkert, A., Marsh, K.N., Pang, S.S., Staiger, M.P., 2009. Ionic liquids and their interaction with cellulose. Chem. Rev. 109, 6712–6728. Plechkova, N.V., Seddon, K.R., 2008. Applications of ionic liquids in the chemical industry. Chem. Soc. Rev. 37, 123–150. Rabideau, B.D., Agarwal, A., Ismail, A.E., 2013. Observed mechanism for the breakup of small bundles of cellulose Iα and Iβ in ionic liquids from molecular dynamics simulations. J. Phys. Chem. B 117, 3469–3479. Ragauskas, A.J., Williams, C.K., Davison, B.H., Britovsek, G., Cairney, J., Eckert, C.A., Frederick, W.J., Hallett, J.P., Leak, D.J., Liotta, C.L., 2006. The path forward for biofuels and biomaterials. Science 311, 484–489. Remsing, R.C., Swatloski, R.P., Rogers, R.D., Moyna, G., 2006. Mechanism of cellulose dissolution in the ionic liquid 1-n-butyl-3-methylimidazolium chloride: a 13C and 35/37Cl NMR relaxation study on model systems. Chem. Commun., 1271–1273. Remsing, R.C., Hernandez, G., Swatloski, R.P., Massefski, W.W., Rogers, R.D., Moyna, G., 2008. Solvation of carbohydrates in n,n'-dialkylimidazolium ionic liquids: a multinuclear nmr spectroscopy study. J. Phys. Chem. B 112, 11071–11078. Remsing, R.C., Petrik, I.D., Liu, Z., Moyna, G., 2010. Comment on “NMR spectroscopic studies of cellobiose solvation in emimac aimed to understand the dissolution mechanism of cellulose in ionic liquids”. Phys. Chem. Chem. Phys. 12, 14827–14828. Rosenau, T., Potthast, A., Sixta, H., Kosma, P., 2001. The chemistry of side reactions and byproduct formation in the system nmmo/cellulose (lyocell process). Prog. Polym. Sci. 26, 1763–1837. Swatloski, R.P., Spear, S.K., Holbrey, J.D., Rogers, R.D., 2002. Dissolution of cellose with ionic liquids. J. Am. Chem. Soc. 124, 4974–4975. Tadesse, H., Luque, R., 2011. Advances on biomass pretreatment using ionic liquids: an overview. Energy Environ. Sci. 4, 3913–3929. Vitz, J., Erdmenger, T., Haensch, C., Schubert, U.S., 2009. Extended dissolution studies of cellulose in imidazolium based ionic liquids. Green Chem. 11, 417–424. Wang, H., Gurau, G., Rogers, R.D., 2012. Ionic liquid processing of cellulose. Chem. Soc. Rev. 41, 1519–1537. Xu, A.R., Zhang, Y.J., Zhao, Y., Wang, J.J., 2013. Cellulose dissolution at ambient temperature: role of preferential solvation of cations of ionic liquids by a cosolvent. Carbohydr. Polym. 92, 540–544. Xu, H., Pan, W.X., Wang, R.X., Zhang, D.J., Liu, C.B., 2012. Understanding the mechanism of cellulose dissolution in 1-butyl-3-methylimidazolium chloride ionic liquid via quantum chemistry calculations and molecular dynamics simulations. J. Comput. Aided Mol. Des. 26, 329–337. Youngs, T.G.A., Holbrey, J.D., Deetlefs, M., Nieuwenhuyzen, M., Gomes, M.F.C., Hardacre, C., 2006. A molecular dynamics study of glucose solvation in the ionic liquid 1,3-dimethylimidazolium chloride. ChemPhysChem 7, 2279–2281. Youngs, T.G.A., Hardacre, C., Holbrey, J.D., 2007. Glucose solvation by the ionic liquid 1,3-dimethylimidazolium chloride: a simulation study. J. Phys. Chem. B 111, 13765–13774. Youngs, T.G.A., Holbrey, J.D., Mullan, C.L., Norman, S.E., Lagunas, M.C., D'Agostino, C., Mantle, M.D., Gladden, L.F., Bowron, D.T., Hardacre, C., 2011. Glucose solvation by the ionic liquid 1,3-dimethylimidazolium chloride: a simulation study. Chem. Sci. 2, 1594–1605. Zhang, H., Wu, J., Zhang, J., He, J., 2005. 1-allyl-3-methylimidazolium chloride room temperature ionic liquid: a new and powerful nonderivatizing solvent for cellulose. Macromolecules 38, 8272–8277. Zhao, H., Baker, G.A., Song, Z., Olubajo, O., Crittle, T., Peters, D., 2008. Designing enzyme-compatible ionic liquids that can dissolve carbohydrates. Green Chem. 10, 696–705. Zhao, H., Jones, C.I.L., Baker, G.A., Xia, S., Olubajo, O., Person, V.N., 2009. Regenerating cellulose from ionic liquids for an accelerated enzymatic hydrolysis. J. Biotechnol. 139, 47–54. Zhao, Y.L., Liu, X.M., Wang, J.J., Zhang, S.J., 2012. Effects of cationic structure on cellulose dissolution in ionic liquids: a molecular dynamics study. ChemPhysChem 13, 3126–3133. Zhao, Y.L., Liu, X.M., Wang, J.J., Zhang, S.J., 2013a. Effects of anionic structure on the dissolution of cellulose in ionic liquids revealed by molecular simulation. Carbohydr. Polym. 94, 723–730. Zhao, Y.L., Liu, X.M., Wang, J.J., Zhang, S.J., 2013b. Insight into the cosolvent effect of cellulose dissolution in imidazolium-based ionic liquid systems. J. Phys. Chem. B 117, 9042–9049. Zhu, S., Wu, Y., Chen, Q., Yu, Z., Wang, C., Jin, S., Ding, Y., Wu, G., 2006. Dissolution of cellulose with ionic liquids and its application: a mini-review. Green Chem. 8, 325–327.

Please cite this article as: Gupta, K.M., Jiang, J., Cellulose dissolution and regeneration in ionic liquids: A computational perspective. Chem. Eng. Sci. (2014), http://dx.doi.org/10.1016/j.ces.2014.07.025i