Chemocatalytic Conversion of Cellulose into Key Platform Chemicals

Tianjin University of Science and Technology, Tianjin 300457, China Jiangsu Provincial Key Laboratory of Pulp and Paper Science and Technology, Nanjing Forestry University, Nanjing 210037, China State Key Laboratory of Tree Genetics and Breeding, Northeast Forestry University, Harbin 150040, China State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou 510640, China Davidson School of Chemical Engineering, Purdue University, 480 Stadium Mall Drive, West Lafayette, IN 47907, USA


Introduction
With the increasing global demand for renewable and valuable chemicals and fuels, social reliance on biomass materials and sustainable technologies is imminent [1][2][3]. These technologies are booming and will be able to effectively use renewable resources to decrease their dependence on nonrenewable resources in order to promote social progress and development [4][5][6][7]. Currently, a promising strategy for alleviating the depletion of energy is to convert lignocellulosic biomass into high-value chemicals and fuels, which has become the current research hotspots . As an environmentally friendly, renewable, sustainable, inexpensive, and nonfood biomass resource, lignocelluloses have gained widespread interest for the production of valuable fuels and chemicals with a variety of designed processing technologies [8,9,[35][36][37][38][39][40][41][42][43][44]. Lignocelluloses are very abundant in the world; it is estimated that the annual global production is about 1.1 × 10 11 metric tons [45]. Thus, researchers have explored the use of lignocellulose as relatively low-cost carbon resource to produce key platform chemicals and fuels. It is well known that lignocellulose as an important biomass resource consists mainly of cellulose, hemicellulose, and lignin [46]. Cellulose accounts for the largest proportion (30-55% wt%) of lignocellulose compared to lignin (25-30% wt%) and hemicellulose (25-30% wt%) [47][48][49][50][51][52][53][54][55][56][57], and it is a linear polymer consisted of numerous glucose units covalently linked by β-1,4-glycosidic bonds and closely bound by intramolecular and intermolecular hydrogen-bond networks (Scheme 1) [58][59][60][61][62][63][64][65][66]. Most cellulose is locked in the form of lignocellulose in nature; hemicellulose and lignin need be removed by extensive cleaning and purification processes. For example, Sasaki et al. reported that semibatch hydrothermal treatment could remove hemicellulose and lignin and recover cellulose from sugarcane bagasse biomass without any catalyst or organic solvent [67]. An average of about 700,000 billion metric tons of cellulose is synthesized annually through photosynthesis of plants using solar energy in the world, but only 0.1 billion metric tons of cellulose is being used as feedstock in paper, food, pharmaceuticals, and textiles [68][69][70][71][72][73]. Therefore, a large amount of underutilized cellulose is wasted every year, and new technologies need to be developed to make better use of cellulose resources. In order to reduce excessive dependence on fossil fuels, it is very important to convert cellulosic biomass into renewable and valuable chemicals and fuels. As the most abundant biorenewable materials on earth, the conversion of cellulose to renewable chemicals and fuels to satisfy the growing global energy demand will attract widespread attention of academia and industrial fields [18,. However, apart from papermaking and materials industry, a low reactivity of cellulose has seriously hindered its application in the chemical industry. Recently, researchers have designed and screened various heterogeneous or homogeneous catalysts under a variety of reaction conditions; heterogeneous or homogeneous catalysis for the transformation of cellulose into key platform molecules is expected to overcome this problem [98][99][100][101][102].
In this review, we focus on the recent advances in homogeneous or heterogeneous catalysis for direct conversion of cellulose into key platform chemicals, particularly glucose, polyols, and furans, except for the production of biofuels. The thermochemical such as pyrolysis and gasification, or enzymatic process, is beyond the scope of this review. We review the recent process of direct chemocatalytic conversion of cellulose to key platform chemicals (Scheme 2), which is helpful for researchers to build a deeper understanding of existing chemical processes on the value-added utilization of cellulose and rationally design a more efficient chemical catalytic conversion system for cellulose.

Pretreatment Technologies for Decreasing Rigid Structure of Cellulose
Due to tight van der Waals interactions and intramolecular and intermolecular hydrogen-bonding networks, cellulose is chemically stable, structurally rigid, and insoluble in water or common organic solvents [103][104][105][106][107][108][109][110][111]. The recalcitrance of cellulose is a bottleneck that affects its sustainability and cost-effective utilization for the conversion of cellulosic biomass into key platform chemicals.  [72,[112][113][114][115]. Among these pretreatment techniques for dissolving the robust structure of cellulose, milling methods as typical mechanical techniques have been widely used, such as ball-milling [116][117][118]. This can disrupt the crystal structure of native cellulose by cleaving hydrogen bonds in cellulose during processing and reduce the degree of polymerization and decrease the particle size and crystallinity of cellulose, thereby increasing the glucose yield [53,[119][120][121][122][123][124][125][126][127][128][129][130][131][132][133][134][135][136]. Kobayashi et al. found that the glucose yield was 88% obtained by mixed ball-milling (60 rpm for 2 days) of catalyst (K26, the purified carbon) and microcrystalline cellulose (MCC), which was much higher than that obtained by single ball-milling (30.0%, 60 rpm for 2 days), indicating that mixed ballmilling was more efficient than single ball-milling for the pretreatment of MCC under the same conditions (180°C and 20 min) ( Table 1, entry 1) [136]. This is due to the fact that mixed ball-milling improves the contact between MCC and catalyst and also promotes the collision of catalyst and MCC at the solid-solid interface. Suzuki et al. reported that jet-milling treatment only reduced particle size without decreasing the crystallinity index, whereas ball-milling and rob-milling processes converted the crystalline of cellulose into amorphous within one hour [127,137]. Kafy  In addition, chemical methods are also used to pretreat cellulose and its derivatives. Deng et al. reported that after commercial cellulose was treated by different concentrations of phosphoric acid (H 3 PO 4 ), its crystallinity decreased from 85% to 79-35%, and the DP of cellulose reduced from 221 to 209-106, and the decreases in DP and crystallinity of cellulose raised the conversion of treated cellulose to sorbitol over the Ru/CNT catalyst (Table 1, entry 4) [140]. The authors demonstrated that the concentration of H 3 PO 4 , treatment temperature, and treatment time had important effects on the crystallinity and DP of cellulose. Cellulose is hardly soluble in water, and it can only be dissolved at relatively high temperatures [141,142]. In order to improve the utilization of cellulose, it is particularly important to explore solvents of cellulose to deal with it as soluble substrates, which could increase the reactivity of cellulose and enhance the conversion of cellulose into high-value chemicals [143]. Burchard et al. demonstrated that a complex was formed between Schweizer's reagent and cellulose, and the reagent dissociated the hydrogen bonding of cellulose and dissolved cellulose [144]. The researchers reported that the mixtures of N,N-dimethylacetamide (DMAc) and lithium chloride (LiCl) could dissolve cellulose owing to cleaving hydrogen bonds of cellulose with the chloride ion [145][146][147]. Recently, ionic liquids (ILs) as cellulose-dissolved green solvents have attracted extensive attention [72,[148][149][150][151][152][153]; this is because ILs have many excellent properties as compared to other traditional solvents, such as controllable adjustment of physicochemical properties, low vapor pressure, wide liquid range, high chemical and thermal stability, high solvation ability, and wide electrochemical window [154][155][156]. Swatloski and coworkers demonstrated that ILs 1-butyl-3-methylimidazolium chloride ([BMIM]Cl) could dissolve cellulose without derivatization [157]. The dissolution mechanism of cellulose in ILs is largely believed to be formation of hydrogen bonds between hydroxyl groups of cellulose and anions of ILs [158]. Li and Zhao demonstrated that the glucose yield reached 43% at 100°C for 540 min in the homogeneous hydrolysis of cellulose using IL 1-butyl-3-methylimidazolium chloride ([C 4 mim]Cl) as a solvent to completely dissolve the cellulose and dilute sulfuric acid (H 2 SO 4 , mass ratio of acid/cellulose = 0 11) as a catalyst without pretreatment (Table 1, entry 5) [159]. The glucose yield was higher than just water as a solvent (dilute H 2 SO 4 as a catalyst, mass ratio of acid/cellulose = 0 92) or concentrated sulfuric acid (65 wt% and higher) as a catalyst. Binder and Raines observed that the yield of cellulose hydrolysis to glucose was increased to nearly 90% in 120-240 min by adding water gradually to IL-HCl solution [ILs 1-ethyl-3-methylimidazolium chloride ([EMIM]Cl) as a solvent and HCl (20 wt%, mass ratio of HCl/cellulose) as a catalyst] (Table 1, entry 6) [51]. Li and coworkers showed that ILs with unsaturated heterocyclic cations generally dissolved cellulose better [160]. Li et al. reviewed in detail various structures and properties of ILs that dissolved cellulose in recent years [161]. With the rapid development of science and technology, it is believed that more efficient pretreatment technologies will be applied for the pretreatment and conversion of cellulose.

Direct Catalytic Conversion of Cellulose into Key Platform Molecules
3.1. Glucose. It is well known that efficient hydrolysis of cellulose as a renewable bulk and nonfood carbon source to glucose is an important challenge for the use of realization of biorefinery, which is the starting point of the overall catalytic conversion chain (Scheme 2) [162][163][164][165][166][167][168].
Glucose is a versatile platform chemical to value-added products such as 5-hydroxymethylfurfural (5-HMF), bioethanol, 3 International Journal of Polymer Science and biodegradable plastics [102]. Early, researchers used mineral acids such as phosphoric acid (H 3 PO 4 ), sulfuric acid (H 2 SO 4 ), hydrochloric acid (HCl), perchloric acid (HClO 4 ), hydrofluoric acid (HF), and nitric acid (HNO 3 ) as representative homogeneous catalysts for the hydrolysis of cellulose into glucose [162,[169][170][171][172][173]. These mineral acids are cheap and widely used in the early hydrolysis of cellulose to glucose. However, they have some insurmountable bottlenecks owing to their difficulties in poor recyclability, wastewater treatment, reactor corrosion, and other issues [98,[174][175][176]. Therefore, there is an urgent need to develop a green and sustainable hydrolysis process of cellulose into glucose.
Recently, solid catalysts, such as carbonaceous acids, metal oxides, supported metals, H-form zeolites, acid resins, heteropoly acids, magnetic acids, and functionalized silicas, have attracted much attention as promising and appealing catalysts for the hydrolysis of cellulose into glucose due to their excellent properties, such as easy separation, recoverability and reusability from product solutions, and adjustable functional structure of catalysts [92,98,131,163,[217][218][219][220][221][222][223][224][225][226][227][228]. Su et al. reported that the pretreatment of cellulose and cow dung-based carbonaceous catalyst (CD-C) via mixed ballmilling effectively enhanced the yield of glucose with 59.3% compared with unmilled catalysts (3.6% glucose yield). Moreover, the addition of trace hydrochloric acid (0.015 wt% HCl) could significantly increase the glucose yield to 74% under the same reaction conditions. The excellent performance of the hydrolysis of cellulose to glucose was attributed to the synergistic effect of the mixed ball-milling pretreatment and the addition of dilute HCl ( Table 2, entry 3) [220]. Yang and Pan demonstrated an important improvement in hydrolysis of cellulose to glucose, in which a bifunctional mesoporous polymeric catalyst bearing boronic acid as cellulose-binding groups and sulfonic acid as cellulose-hydrolytic groups was prepared, resulting in an excellent hydrolysis performance (Table 2, entry 4, Scheme 4) [229]. The increase in hydrolysis performance was mainly due to the synergistic effects of more cellulose-attracting groups (boronic acid) and cellulosehydrolyzing groups (sulfonic acid) and larger surface area of the prepared polymeric solid acids. Zhang et al. reported an effective carbonaceous solid acid catalyst (10-SGOC) for the hydrolysis of cellulose to glucose, which was prepared by hydrothermal carbonization of cellulose using graphene oxide as a structure-directing agent and subsequent concentrated H 2 SO 4 as a sulfonating agent [230]. The yield and selectivity of glucose reached 17.76% and 94.22% at 160°C for 5 h over 10-SGOC catalyst, respectively (Table 2, entry 5). 10-SGOC exhibited excellent glucose yield and selectivity in the case of low mass ratio of catalyst/cellulose compared with previously reported sulfonated solid acid catalysts [130,217,231,232]. This indicated that the combination of functional groups such as -OH, -COOH, and -SO 3 H and layered structure with good hydrophilicity promoted a synergistic effect for the 10-SGOC catalyst, thereby better facilitating the contact of the catalyst with cellulose and effectively diffusing glucose into water. Li [235]. This might be due to a decrease in Lewis acidity of the modified HY catalyst and a decrease in contact probability of Lewis acid sites with sugar caused by the outer silicalite-1 layer.

Polyols.
The conversion of resource-abundant, renewable, and nonedible cellulosic biomass to high-value chemicals is beneficial to the development of a sustainable society. In particular, the direct catalytic conversion of cellulose to polyols is an important part of biorefinery, which has attracted extensive interest to meet the world's energy needs [236][237][238][239][240][241][242]. In this section, we focus on the recent developments in the chemical catalytic conversion of cellulosic biomass into ethylene glycol (EG) and sorbitol, which are widely used in the food industry, pharmaceutical production, cosmetics, and so on.

Ethylene Glycol (EG).
With the increase of the market demand of renewable products from biomass, it is highly desirable to design a series of effective catalysts which can control the selectivity of polyols [169,. The direct conversion of cellulose to value-added polyols in biorefinery is an attractive and promising approach. Although significant advances have been made in the transformation of cellulose to polyols, the production of quite valuable EG is still challenging [88,[264][265][266][267][268][269][270][271][272].
Hamdy et al. reported a novel 3Al-15W-3Ni catalyst prepared by one-pot hydrothermal sol-gel process for selective hydrogenolysis of cellulose into EG. The catalyst showed the multiple active sites and exhibited a high cellulose conversion (100%) and an excellent yield of EG with 76 [277]. The addition of Ru species significantly increased the active site of W 5+ , thereby promoting the retro-aldol condensation reaction of glucose and enhancing the etching of H + in aqueous solution for the hydrogenolysis of cellulose to EG, thus obtaining a high yield of EG from cellulose.

Sorbitol.
With the gradual depletion of nonrenewable resources, such as fossil fuels and coal, cellulosic biomass, as a large-scale renewable and sustainable carbon source on earth, has attracted wide attention for its catalytic conversion to value-added chemicals, such as the hydrolytic hydrogenation of cellulose to produce sorbitol [140,[278][279][280][281][282][283][284][285][286][287]. Zhang [288]. This indicated that the modification of Ni species improved the catalytic reaction activity, and a synergistic effect might be promoted between active Ni/Cu and Cu sites.
Xi and coworkers reported a novel mesoporous niobium phosphate-supported bifunctional Ru catalyst (5%Ru/NbOPO 4 -pH2), which exhibited excellent performance for the selective transformation of cellulose to sorbitol [289]. When the mixture of cellulose and 5%Ru/NbOPO 4 -pH2 was ball-milled for 10 h and the hydrolytic hydrogenation reaction was conducted in water, the sorbitol yield was

5-Hydroxymethylfurfural (5-HMF). 5-HMF is a promis-
ing and versatile biomass-based platform molecule from biorefinery carbohydrates, which can be used to produce various chemicals and liquid fuels currently derived from nonrenewable fossil resources [293][294][295][296]. It is worth noting that the production of 5-HMF from renewable cellulose has become an integral part of biorefinery and attracted extensive attention in recent years . Li et al. prepared a sulfonated poly(phenylene sulfide) (SPPS) catalyst with strong Brønsted acid sites and a sulfonation degree of 21.8 mol%; when it was used for the direct conversion of cellulose to 5-HMF in IL 1-methyl-3-ethyl imidazolium bromide ([EMIM]Br) solvent, the yield of 5-HMF with 68.2% was obtained at 180°C for 4 h ( Table 5, entry 1) [333]. DFT calculations indicated that the −SO 3 H groups in SPPS play an important role in catalytic conversion, and it acts as a proton donor for Brønsted acids and acts as a proton acceptor as a conjugate base. In addition, the anions and cations of ILs together with SPPS-SO 3 H are conducive to stabilizing the transition states and reaction intermediates, which also leads to the easy conversion of cellulose to 5-HMF. Zhang [337] yield of 49.3% was obtained at 150°C for 30 min under the conditions of 7 mol% of CrCl 3 and 5 mol% of tetramethylammonium perrhenate. The effective catalytic activity of CrCl 3 /[R 4 N]ReO 4 for the production of 5-HMF from cellulose was attributed to the bifunctional catalytic system of the catalyst, which was better than the single CrCl 3 catalyst (  [336]. What is noteworthy is that the Nb/C-50 catalyst was agglomerated particles and contained suitable Brønsted/Lewis acid sites and weak to medium acid strength, which favored the conversion of cellulose to 5-HMF. Zhao et al. reported an interesting bifunctional catalytic system using AlCl 3 as the Lewis acidic catalyst and H 3 PO 4 as Brønsted acidic catalyst, and when the AlCl 3 -H 3 PO 4 catalyst was employed for the transformation of cellulose to 5-HMF, the highest yield of 5-HMF with 49.42% was achieved in a single-phase reaction system of 1,2-dimethoxyethane (DMOE) and water at 180°C for 120 min under the reaction conditions of mole ratio of 1: 0.8 of AlCl 3 / H 3 PO 4 and volumetric ratio of 7: 1 of DMOE/H 2 O (Table 5, entry 5) [337].
Shen and coworkers reported that a high catalytic activity for the conversion of cellulose to LA with a yield of 39.4% was achieved in the presence of IL 1-(4-sulfonic acid) butyl-3-methylimidazolium hydrogen sulfate ([BSMim]HSO 4 ) with addition of water at 120°C for 120 min (   (Table 7, entry 1) [358]. The bifunctional catalyst exhibited an excellent lactic acid yield because the combination of Al(III) and Sn(II) cations promoted the catalytic conversion of cellulose to lactic acid and limited the side reaction.
A series of erbium ion-exchanged montmorillonite K10 catalysts were prepared by Wang et al. and used as solid acid catalysts for the conversion of cellulose to lactic acid. Interestingly, Er/K10(S)-3 exhibited the highest yield of lactic acid with 67.6% in aqueous solution at 240°C for 30 min under 2.0 MPa of N 2 pressure (Table 7, entry 2) [359]. Following this work, Wang et al. reported a series of Er/β-zeolite catalysts prepared using erbium species grafted on β-zeolites, which were used for one-pot hydrothermal transformation of cellulose to lactic acid [360]. 57.9% lactic acid yield was achieved over the 12.4 wt% Er/deAlb-2 catalyst (synthesized by erbium species grafted on de-aluminated β-zeolite) with a Si/Al ratio of 159 for the conversion of cellulose into lactic acid in H 2 O at 240°C for 30 min under 2.0 MPa of N 2 pressure (Table 7, entry 3). The excellent activity of Er/deAlb-2 might be due to its higher erbium content, larger average pore diameter, and external surface area, which promoted contact between active erbium ions and reactant molecules.
Wattanapaphawong et al. studied the effect of a large number of transition metal oxides on the catalytic activity of cellulose conversion to lactic acid. They found that ZRO-7 exhibited the highest yield of lactic acid from cellulose with 21.2% in H 2 O at 199.85°C for 6 h compared with transition metal oxide catalysts, such as Al 2 O 3 , V 2 O 5 , CeO 2 , Y 2 O 3 , Ga 2 O 3 , and MgO (Table 7, entry 4) [361]. The remarkable catalytic activity of ZRO-7 was due to the large amount of base and acid sites on the ZRO-7 catalyst. Following this work, the 10%ZrO 2 -Al 2 O 3 catalyst synthesized by Wattanapaphawong et al. was used for the production of lactic acid from cellulose, leading to 25.3% lactic acid yield at 199.85°C for 6 h in aqueous solution (Table 7, entry 5), which was superior to ZRO-7 [362]. Compared to pure ZrO 2 , the ZrO 2 -Al 2 O 3 catalysts contained more Lewis acid sites and fewer base sites. The researchers suggested that the Lewis acid sites on ZrO 2 -Al 2 O 3 play a more important role than do basic sites for the production of lactic acid from cellulose.

Conclusions
Direct chemocatalytic conversion of lignocellulosic biomass to renewable and value-added chemicals has attracted worldwide attention in order to build up sustainable societies. Cellulose is the most abundant, renewable, and nonedible biomass resource in lignocellulosic biomass. However, the recalcitrance and low reactivity of cellulose have hampered its sustainability and cost-effective utilization for chemical industry except for papermarking. It has been expected that homogeneous or heterogeneous catalysts for the direct chemocatalytic conversion of cellulosic biomass into high-value platform chemicals can overcome this problem due to the fact that various types of pretreatment as well as homogeneous or heterogeneous catalysts can be designed and applied under a wide range of reaction conditions. In this review, we have summarized the recent pretreatment methods for decreasing the rigid structure of cellulose and catalytic conversion of cellulose into key platform chemicals, such as glucose, ethylene glycol, sorbitol, 5-hydroxymethylfurfural, levulinic acid, and lactic acid, via a variety of homogeneous or heterogeneous catalysts. This work will be helpful for researchers to build a deeper understanding of existing chemical processes on the value-added utilization of cellulosic biomass to high-value chemicals and rationally design a more efficient chemical catalytic conversion system for cellulosic biomass.

Conflicts of Interest
The authors declare no conflict of interest.