Catalysis is an important phenomenon in nature and is involved in all areas of the chemical industry. Homogeneous and heterogeneous catalysis are key techniques in chemical production. The development of efficient, stable, and environmentally friendly catalytic methods is receiving much attention because of the increasing global shortage of energy and serious environmental damage. A number of efforts have been made in the past for designing rationally catalysts to achieve desirable catalytic activities, selectivities, and stabilities for specific reactions [1]. The current energy crisis is affecting economies in every region, particularly in fossil-oil-consuming countries, because of rapidly decreasing available stocks, therefore biomass conversion to fuels and chemicals using various approaches, including thermal, biological, and chemocatalytic processes, is attracting great attention worldwide [2].

The catalytic conversion of biomass and derivatives to chemicals has been the subject of intense research efforts during the past decade. As the most accessible renewable form of carbon, lignocellulosic biomass (lignin, cellulose, and hemicellulose) is the most logical carbon-based feedstock for the manufacture of fuels, commodities, high-value-added chemicals, and new bio-based materials such as bioplastics [3]. Conversion can be achieved using various reactions, including hydrogenation, dehydrogenation, dehydroxylation/ hydrogenolysis, and oxidation; these reactions can be used in combination with acid/base-catalyzed reactions such as hydrolysis and dehydration to achieve one-pot multistep conversions (Fig. 1). Among various biomass conversions, large-scale production of biodiesel has been achieved using transesterification of triglycerides in vegetable oil and animal fats. The amount of glycerol generated as a by-product of biodiesel production is ca. 10 wt%. It is therefore necessary to find more effective approaches to using this oversupply of glycerol; this will be crucial for the biodiesel economy. Possible reactions for glycerol valorization include hydrodeoxygenation, hydrogenation, hydrogenolysis, dehydration, and isomerization [4]. As a polylol, glycerol is an important biochemical that can be used as a platform chemical for the production of high-added-value products [5]. This has led to investigation of the valorization of this highly versatile molecule, which is, because of its three hydroxyl groups, a potential starting material for numerous value-added products (Fig. 1). Reactions investigated include steam reforming, aqueous-phase reforming, hydrogenolysis, oxidation, dehydration, esterification, etherification, carboxylation, acetalization, and chlorination [6]. Various glycerol conversion techniques have been developed and glycerol has become a potential renewable resource for the production of various high-value-added chemicals. A representative chemocatalytic conversion of biomass to high-value-added products is summarized in Fig. 1.

Various types of chemicals can be obtained from renewable biomass as sustainable feedstocks using catalytic approaches. Numerous solid catalysts have been developed for the transformation of biomass feedstocks to high-value-added chemicals and fuels. Heterogeneous catalysts for biomass conversion are classified into several main groups, depending on their structures and substrate activation properties [3, 7]: (a) micro-and mesoporous materials, (b) metal oxides, (c) supported metal catalysts, (d) sulfonated polymers, and (e) metal phosphates. Because of their low water tolerances, limited numbers of surface active sites, and poor thermal stabilities during reactions, the catalytic applications of metal oxides, supported metal catalysts, and polymer catalysts are restricted. Porous metal phosphates are highly attractive for use as heterogeneous catalysts. Zirconium phosphates (ZrPs), in particular, have high thermal stabilities and water tolerances, and form sediments easily [8], and their medium to strong acidities make them compatible with polar media, including aqueous phases. These properties are advantageous in biomass conversions, which always involve a high temperature and aqueous media. Furthermore, the acidic sites in the ZrP skeletal structure are flexible and adjustable. It should be noted that because of the synergistic effect of combining metal centers and acidic sites in metal-modified ZrP catalysts, they have a broad range of potential applications in a large number of specific reactions [9-13].

ZrPs are a family of transition-metal phosphate materials. Various methods have been reported for the synthesis of amorphous ZrP, mesoporous ZrP, and crystalline α/γ/τ-ZrP, with structures and morphologies that can be varied by tuning the synthetic conditions [14]. The general formulas for α-, γ-, and τ-ZP are α-Zr(HPO₄)₂·H₂O, γ-Zr(PO₄)(H₂PO₄)·2H₂O, and τ-Zr(HPO₄)₂, respectively [15-17]. Different types of ZrP contain different phosphorus hydroxyl groups and numbers of water molecules. The crystalline ZrP formula indicates that the theoretical P/Zr ratio in crystalline ZrP is ca. 2, whereas the P/Zr ratio of amorphous ZrP ranges from 0 to 2 [18, 19]. Nanosized α-ZrP can be converted to τ-ZrP by drying a gel in an oven at 120 ℃ for 15 h, but this conversion is not observed for microcrystalline α-ZrP under the same conditions [20]. α-ZrP can be synthesized under more acidic conditions than γ-ZrP. The different α/γ/τ-ZrP phases can be clearly identified using X-ray diffraction (XRD); the XRD pattern shows ZrO₆ octahedra connected with phosphonate tetrahedra via oxygen bridging bonds [20, 21].

Amorphous ZrP can be conveniently prepared by precipitation. Dropwise addition of an aqueous solution of NH₄H₂PO₄ to an aqueous solution of ZrOCl₂·8H₂O at a specific P/Zr molar ratio and then stirring the mixture overnight at room temperature give amorphous ZrP [8]. This preparation method is simple and convenient. Mesoporous ZrP can be obtained by using a different zirconium complex precursor and phosphate source, and can be synthesized using the following methods. (1) A post-synthesis method using the surfactant hexadecyl trimethyl ammonium bromide as a structure-directing agent, followed by post-treating the surfactant-assisted zirconium oxide mesophase with phosphoric acid solution [19]. (2) Evaporation-induced self-assembly, using non-ionic Pluronic F127 or P123 as a structure-directing agent in an acidic non-aqueous medium [22, 23]. Mesoporous ZrP has useful properties such as a high surface area, narrow size distribution, specific particle morphology, and high acid strength. Large pores aid diffusion of large organic molecules, enhancing the catalytic ability. The high catalytic activity can be attributed to a large surface area and the presence of a large number of acidic sites on the material surface. The catalysts show negligible loss of catalytic activity in stability tests, suggesting that the mesoporous ZrP framework has high chemical stability and the catalyst can be recycled. The thermal stability of the acidic sites on the resulting metal phosphates enables their use under harsh conditions such as high temperature and in the presence of water-steam.

Generally, α-and γ-type layered ZrPs are isomorphous compounds. The layers in γ-ZrP are more rigid than those in the α-analog, giving stronger acidity and faster intercalation [24]; τ-ZrP has a three-dimensional structure [17]. Among these metal phosphates, α-ZrP has been most widely investigated and used in catalytic reactions. It is a typical layered compound with a well-defined structure and has both medium-strength acidic (-8.2 < H₀ ≤ -3.0) and weak acidic sites (-3.0 < H₀ ≤ +6.8). The acid density is about 0.93 mmol/g, determined using the Hammett indicator method [8], and the surface area is ca. 27.5 m²/g [15]. Because of its particular acidic and porous properties, α-ZrP can act as both a catalyst support and an efficient catalyst directly.

Crystalline α-ZrP can be obtained using various strategies, including refluxing, hydrothermal, and precipitation methods, and α-ZrP with a wide range of aspect ratios can be obtained [25]. Recently, a sustainable route that uses only zirconium oxychloride and concentrated phosphoric acid to form highly crystalline α-ZrP within hours was reported [26]. The morphology of α-ZrP can be changed from platelets to rod-shaped particles by fluoride addition, and the particle size can be changed from nanometers to microns by tuning the temperature and time.

The ZrP layered structure consists of zirconium ions in a semi-planar arrangement, located slightly above and below the mean plane. Each Zr⁴⁺ ion is connected through the oxygen atoms of phosphate groups above and below [27] (Fig. 2). These hydroxyl groups connected with phosphorus atoms are responsible for the Brønsted acidity and the zirconium atom centers provide the Lewis acidity. The Brønsted and Lewis acidities are controlled by changing the phosphorus and zirconium sources. α-ZrP and γ-ZrP have high cation-exchange capabilities of ca. 6.64 and 6.27 mmol H⁺/g respectively, because of their high densities of acidic sites and reactive surface hydroxyl groups. The interlayer distances in α-ZrP and γ-ZrP are 0.76 and 1.22 nm, respectively [24, 28, 29].

Various acid catalysts, e.g., homogeneous mineral acids, metal chlorides (including those of Cr, Al, Ge, and Zn), ionic liquids, and heterogeneous catalysts, including metal oxides, heteropolyacids, ion-exchange resins, zeolites, functionalized carbonaceous materials, mesoporous silica materials, and ammonium and hydrotalcite-based materials, have been used to transform biomass-derived platform molecules [30]. The ZrP acid strength has been investigated qualitatively using temperature-programmed NH₃ desorption. Desorption peaks were observed in the ranges 100-200, 200-400, and 400-600 ℃, attributed to NH₃ chemisorbed on weak, medium, and strong acidic sites, respectively [18, 23, 31]. ZrPs have therefore become increasingly popular as solid acid catalysts in the past decade. In this perspective, we focus on their recent applications in catalytic transformations of biomass-derived platform molecules via reactions such as dehydration, hydrogenation, hydrogenolysis, and oxidation, and also clarify the role of the surface active sites of ZrP catalysts.

Dehydration is an important path for the conversion of biomass to useful chemicals. Biomass dehydration involves the conversion of glucose, fructose, and sorbitol to 5-hydroxymethylfurfural (HMF). It has been reported that the Lewis acidic sites on heterogeneous catalysts promote glucose isomerization to fructose via an intermolecular hydride shift, and the Brønsted acidic sites promote further dehydration of fructose to HMF [32].

An amorphous ZrP layer deposited on the surface of open-cell Al foam showed high catalytic activity in fructose dehydration to HMF [33]. The Al foam was fully and uniformly covered by the ZrP layer, with good mechanical adherence to the support. Electrophoretic deposition has been used to form an amorphous ZrP layer on the surface of open-cell Al foam. The surface area of the ZrP layer on the foam surface was 198 m²/g, which is higher than that of the uncoated foam, 89 m²/g, as a result of stabilization of the particles on the foam surface, preventing agglomeration [33]. Compared with other reported catalysts, ZrP catalysts are more selective for HMF in fructose conversion because of their inherent acid strength. The HMF selectivity correlates with the contribution of Brønsted acidic sites, which give high selectivity; the Lewis acidic sites decrease the HMF selectivity because fructose condenses rapidly to give humins at Lewis acidic sites in the ZrP framework [34].

Investigations of the acidic properties of similar phosphates of Al, Ti, Nb, and Zr suggest that the amount of strong Brønsted acidic sites increases with decreasing electronegativity of the metal. The selectivity for HMF increases with increasing ratio of Brønsted to Lewis acidic sites, accompanied by a decrease in the amount of isolated Lewis acidic sites [35]. Nijhuis et al. [36] prepared a solid foam-structured acid catalyst by coating ZrP on Al foam. An excess of Lewis acidic sites on the catalyst led to unselective glucose transformation to humins. Modification of the catalyst by silylation or deeper treatment with phosphoric acid significantly increases the HMF selectivity because of synergism between protonated phosphate groups and nearby metal Lewis acidic sites on the silylated catalyst.

Huber et al. [37] investigated solid acidic metal-(IV) phosphates such as ZrPs and tin phosphate (SnP) as catalysts for producing levulinic acid from glucose. Lewis acidic sites mainly produce fructose via isomerization reactions and undesired degradation products (humins). In addition, HMF is produced on both Brønsted acidic sites (protonated phosphate groups) and Lewis acidic sites (metal sites), whereas levulinic acid is exclusively produced on Brønsted acidic sites in the ZrP skeleton. A ZrP catalyst is more favorable for levulinic acid production because of its inherently high surface area and enhanced Brønsted acidity. A study using ³¹P magic-angle spinning nuclear magnetic resonance spectroscopy showed that of four phosphorus coordination states, polyphosphate species were present in the highest amount. The higher amount of polyphosphate species could be the reason for the enhanced acidity. The length of the polyphosphate chains could also play a key role in determining the concentrations of acidic sites. Because glucose is abundant in nature, present in polysaccharide forms such as starch or in cellulose in biomass, levulinic acid production from glucose via a two-step acid-catalyzed route is desirable (Fig. 3).

The synthesis of HMF from fructose was achieved using amorphous ZrP as the catalyst in water under microwave heating. A high yield of HMF was obtained using a high temperature and short reaction time [38]. ZrP catalysts have different acidic properties depending on the total amount of acidic sites, the Brønsted/Lewis acidic site ratio, and the acid strength. ZrP catalysts are more selective for HMF than niobium phosphate (NbP). This can be explained by the predominance of medium-strength Brønsted acidic sites in the ZrP catalysts because stronger acidic sites promote the formation of undesired heavy compounds. Furthermore, ZrP catalysts show good hydrothermal stabilities during reactions and in recycling tests. Mesoporous ZrP obtained using a hydrothermal method, with a high surface area of 407 m²/g, gave excellent yields of HMF from fructose (80%), glucose (63%), and sucrose (61%) [39]. This catalyst has high acidity because of the incorporation of zirconium and phosphorus on the mesoporous surface, and a high external surface area, which exposes a larger number of acidic sites on the surface of the material. The catalyst showed a negligible decline in activity after 5 cycles, on the basis of the HMF yield.

A series of ordered mesoporous zirconium oxophosphates with different P/Zr molar ratios (M-ZrP-x) were prepared using one-pot evaporation-induced self-assembly [40]. The concentrations of Brønsted and Lewis acidic sites were adjusted by varying the P/Zr molar ratio; a high yield (72%) of HMF was obtained under relatively mild conditions. The catalytic activity was retained after 12 cycles; this is ascribed to an ordered mesoporous structure and high thermal stability. The mesostructure of ZrP (P/Zr = 0.75) promotes mass transfer, and more active centers are in contact with the substrate, which improves the catalytic activity. The introduction of chromium into mesoporous ZrP (ZrP-Cr) gives a heterogeneous catalyst with an increased number of active sites for the catalytic conversion of carbohydrates to HMF [9]. Recycling experiments indicated that the catalyst was stable at 100 ℃, and could be reused six times without significant loss of catalytic activity. Mesoporous ZrP obtained by a hydrothermal method using organic amines as templates has been investigated as a solid catalyst for the aqueous-phase dehydration of xylose to furfural [41]. Compared with TiO₂/carbon black, Zr (W, Al) mixed oxides, and H-MCM-22 catalysts, the ZrP was synthesized by hydrothermal method with amorphous structure after calcination treatment (ZrP-HT-Am-C) gave a higher conversion (96%) and yield (52%). This shows that the open structure with numerous active Brønsted and Lewis acidic sites in the ZrP skeleton is responsible for isomerization and dehydration, and gives a catalyst with excellent efficiency. The catalyst is easily regenerated by thermal treatment in air and shows stable activity.

Among various sorbitol-derived chemicals, isosorbide, a product of sorbitol dehydration, is one of the most important derivatives because of its rigid molecular structure and chiral centers; it has applications in areas such as polymers and medicine. The dehydration of sorbitol proceeds via two steps. In the first dehydration step, two types of monoanhydrosorbitol are formed. The first type (1, 4-and 3, 6-anhydrosorbitol) reacts sequentially to give isosorbide, and the second type (1, 5-and 2, 5-anhydrosorbitol) is stable toward further dehydration. A porous ZrP catalyst was obtained using a hydrothermal method and its use in the dehydration of sorbitol to isosorbide under water-free conditions was studied. Its catalytic activity in isosorbide production was better than those of Amberlyst 15, Nafion, and niobic acid catalysts [42], probably because of its high surface area of 148 m²/g, which exposed a large number of active acidic sites at the catalyst surface, and its high acidity arised from incorporation of zirconium and phosphorus on the pore surfaces [42]. The sorbitol conversion was still 100% up to the fifth run. The isosorbide selectivity declined only slightly up to the fifth run, indicating that the catalyst has good reusability. Hoelderich et al. [43] used a commercially available ZrP catalyst for sorbitol dehydration. This conversion involves water molecules and Brønsted acidic centers in the solid ZrP framework. Moderate and strong acidic sites are favorable for this process. The strong acidity of ZrP gradually increases with increasing calcination temperature, whereas the total acidity abruptly decreases, which corresponds to enhanced catalytic activity. The product distribution indicates a simple pathway for isosorbide formation, as shown in Fig. 4.

Hierarchically porous ZrP monoliths with high surface areas were used as heterogeneous acid catalysts for the dehydration of xylose to furfural [44]. The ZrP monolith calcined at 600 ℃ has a large proportion of medium-strong acidic sites and a small proportion of strong acidic sites. High-temperature calcination leads to a decrease in the number of Brønsted acidic sites and an increase in the number of Lewis acidic sites compared with the as-synthesized ZrP monolith. The high accessibility and availability of acidic sites result in good catalytic activity, and easy separation of the monolith and resistance to the liquid medium improve the recyclability.

2, 3-Butanediol (BDO) is an excellent feedstock for expanding the range of available bio-based chemicals. It can be dehydrated catalytically using a ZrP catalyst [45]. The dehydration of BDO provides an alternative green route to important chemicals such as methyl ethyl ketone and 1, 3-butadiene. It can be used as an antifreeze agent, solvent, and precursor for plastic manufacturing. Brønsted and Lewis acidic sites in the ZrP skeleton can both participate in BDO dehydration. The correlation between the catalytic activity in BDO dehydration and the acid strength distribution over phosphate catalysts (ZrP, NbP, aluminum phosphate (AlP), titanium phosphate (TiP), and boron phosphates) shows that acidic sites with medium strength are the most active in dehydration.

In recent years, the catalytic conversion of glycerol to acrolein via a double-dehydration reaction has become an important route for using glycerol resources. Acrolein is an important and versatile chemical intermediate and raw material for the production of acrylic acid esters, super-absorbent polymers, super-absorbents, detergents, dispersants, flocculants, and value-added derivatives [46]. Recently, Hou et al. [8] prepared three ZrPs using precipitation, hydrothermal synthesis, and impregnation methods. The preparation method crucially affects the texture and surface acidity of the ZrP. Amorphous ZrP obtained using a precipitation method has weak and medium-strength acidities, and affords highly selective catalysts for acrolein production (yield 62%-82%) by the gas-phase dehydration of glycerol; this is a better performance than those of ZrP catalysts prepared using other methods. The catalytic dehydration of glycerol can provide a cost-effective and sustainable alternative route to acrolein (Fig. 5). It has been reported that calcining the deactivated catalyst in air easily regenerates the deactivated catalyst, with its original activity. The surface acidic sites and thermal stability of ZrP play important roles in the activity, acrolein selectivity, and catalyst lifetime.

Transition metals and some main-group metals provide excellent metal centers for building metal phosphate skeletons because of their abilities to gain and lose electrons and/or their corresponding redox abilities. The H⁺ in P-OH can be exchanged with divalent, trivalent, tetravalent, and pentavalent metal phosphates, and their considerable insolubility and stability permit their use under harsh conditions. In a recent study, ZrP was used as an acidic support and various metal cations, including Al, Ti, and Nb, were introduced into the ZrP matrix to produce new catalytic sites. These materials have been used as heterogeneous acid catalysts for the acid dehydration of biomass-derived molecules to fine chemicals because of the presence of Lewis (metal sites) and Brønsted (protonated phosphate group) acidic sites [35].

A series of vanadium phosphorus oxide (VPO) catalysts supported on porous ZrP were prepared using a solid-solid wetting method, and the catalytic activities of these materials were examined. The active VPO species were well dispersed on the ZrP support because of the strong interactions between the active phase and the support. The active phase consisted mainly of vanadyl pyrophosphate, along with a small amount of vanadium orthophosphate [47]. The supported VPO catalysts gave better conversions and selectivities for acrolein than the pure VPO catalyst because of their appropriate acidic properties, higher surface area, i.e., 152 m²/g, and higher pore diameter, i.e., 7.4 nm. The acidity of the catalyst, especially the strong acidic sites, has a detrimental effect on acrolein formation during glycerol dehydration.

In summary, ZrP-derived catalysts are widely used in the preparation of fine chemicals via biomass dehydration because of their medium-strength acidic sites and the high thermal stability of the ZrP structure. The Brønsted and Lewis acidic sites can be adjusted for specific catalytic reactions. These results for catalytic dehydration over ZrP catalysts are summarized in Table 1.

A series of highly efficient ZrP catalysts were developed for the transfer hydrogenation of levulinate esters to γ-valerolactone using 2-propanol as the hydrogen donor [18]. ZrPs (P/Zr = 1) are generally hydrophilic because of their P-OH groups. The physicochemical properties, particularly the hydrophobicity, Lewis to Brønsted acidic sites ratio and Lewis acid strength were subtly tuned by adjusting the P/Zr molar proportion. Excellent transfer hydrogenation activity at 210 ℃ was achieved. The optimum catalyst has high thermal stability and recyclability for at least 10 reaction cycles. The hydrophobicity of the catalyst has a positive effect on transfer hydrogenation of the carbonyl group and the hydrophobicity of the alcohol is significantly enhanced by elongation of the carbon chain. The hydrophobicity and steric hindrance of levulinate esters significantly affect γ-valerolactone production. The hydrophobicity of the substrate increases with increasing length of the carbon chain of the alkyl group.

A facile and template-free route for synthesizing a series of heterogeneous nitrogen-containing alkyl triphosphonate-metal hybrids with enhanced Lewis acidic and basic sites was reported. Their catalytic activities in converting biomass-derived carbonyl compounds to the corresponding alcohols in 2-propanol were investigated [48]. A quantitative yield of furfuryl alcohol (FFA) was obtained from furfural over an organo-triphosphate-zirconium hybrid (ZrPN) under mild conditions. A mesoporous ZrPN nanohybrid with a high surface area and enhanced Lewis basic (alkyl phosphate and amino groups) sites adjacent to acidic (zirconium) sites, with an appropriate basic/acidic site ratio (1/0.7) in ZrPN, significantly improved the yield of FFA. The ZrPN was thermally stable up to 250 ℃ and the performance of ZrPN was almost constant (>98% selectivity) in six consecutive runs.

Mesoporous and orderly layered zirconium benzyl phosphonate has been synthesized for the first time via simple and template-free assembly of ortho-, meta-, or para-xylylene diphosphonates (o-, m-, or p-PhP) containing zirconium [49]. The Lewis acidic (Zr⁴⁺) and basic (PO₃^2-) centers in this robust heterogeneous catalyst acted synergistically in one-pot cascade transformations of m-PhPZ. The catalyst with a high Lewis acidic/basic site ratio (1/0.7) gave an excellent performance under mild conditions in the transfer hydrogenation of carbonyl compounds, including those in bioaldehydes and ketones, with near quantitative yields and no loss of Zr species during the reaction.

In addition to transfer hydrogenation, catalytic hydrogenation with molecular hydrogen has been widely performed over ZrP-derived catalysts. Heterogeneous catalysts with Pd supported on different acidic supports have been developed [10]. In comparison with various supports, the specific Brønsted acidity of a ZrP support effectively accelerated cleavage of the C-O bond in a furan ring. It has been proposed that this hydrogenolysis of HMF to 1, 6-hexanediol involves six major steps, i.e., steps 1-6 in Fig. 6. It is suggested that the surface acidity, which is important for ring cleavage, the presence of a transition metal, which is important for formic acid (FA) dissociation, and use of in situ-generated hydrogen for hydrogenation are important in governing 1, 6-hexanediol formation from HMF. The product was isolated using silica-gel column chromatography and the 1, 6-hexanediol yield was 39.6%. The catalyst was easily recovered by centrifugation, and recycled without any significant loss of activity even after 5 cycles.

Huber et al. [11] used modified aqueous-phase hydrodeoxygenation (APHDO) to selectively produce high-octane gasoline-range molecules from both C5 and C6 aqueous carbohydrate solutions over a stable, selective, and active Pt/ZrP catalyst. The APHDO of sugar alcohols involves several key reactions, particularly C-O bond cleavage at Brønsted acidic sites, which are present on ZrP catalysts. Gasoline-range compounds can be produced selectively from biomass by adjusting the concentration and strength of the acidic sites on bifunctional catalysts, and the reaction conditions. Pt/ZrP was stable with no significant deactivation after 200 h time-on-stream. Pt, Zr, or P were detected ( < 7 ppm) in the product stream, indicating that no leaching occurred.

Glycerol hydrogenolysis normally proceeds via a dehydration-hydrogenation route [50]. Initially, glycerol undergoes dehydration-hydrogenation at the acidic and metal sites on the catalyst to give 1, 2-propanediol and 1, 3-propanediol, respectively. The produced propanediols then undergo further dehydration-hydrogenation to give 1-propanol and 2-propanol, respectively (Fig 7). The synthesis of 1, 2-propanediol, 1, 3-propanediol, and propanols from biomass-derived glycerol via hydrogenolysis under ambient condition is an important industrial process because these products have a range of uses, including synthesis of polyester resins, detergents, flavoring agents, cosmetics, pharmaceuticals, and paints.

Metal phosphate-supported Pt catalysts (Pt/AlP, Pt/TiP, Pt/ZrP, and Pt/NbP) have been used for the direct vapor-phase hydrogenolysis of glycerol under mild conditions to produce biopropanols [12]. The excellent catalytic activity of a Pt-supported metal phosphate is attributed to the strong acidity of the catalyst and well dispersed Pt on the metal phosphate surface. The catalyst was stable for a long reaction time, namely 20 h, and reusable, with only a slight decline in the catalytic activity. The one-pot hydrogenolysis of biomass-derived glycerol to 1-propanol over sequential two-layer catalysts in a continuous-flow fixed-bed reactor has been investigated [31]. The ZrP layer with strong acidic sites, packed in the upper layer, dehydrated glycerol to acrolein; Ru catalysts supported on SiO₂, Al₂O₃, Nb₂O₅, Hβ, and HZSM-5 packed in the second layer gave sequential hydrogenation of acrolein to 1-propanol. The second-layer catalyst, with weak acidic sites (2% Ru/SiO₂), gave the highest selectivity for 1-propanol. Carbonaceous deposits were the main reason for deactivation and the deactivated catalysts were conveniently regenerated by calcination in air.

Combinations of ZrP materials with large surface areas and mesoporous structures and commercial Ru/C have been evaluated for the hydrolytic hydrogenation of cellulose to sorbitol/mannitol [51]. The enhanced yield is attributed to the Lewis and Brønsted acidic sites on the ZrP surface. These improve surface adsorption and activation of cellulose, which may accelerate the rate-determining step, i.e., cellulose hydrolysis. Compared with various other solid acid catalysts (γ-Al₂O₃, ZrO₂, SiO₂-ZrO₂, H-MOR, HZSM-5, Amberlyst 15, ZrW, AlW, and ZrS), ZrP improved cellulose adsorption and promoted its depolymerization to cellobiose and glucose. These were immediately hydrogenated to C6 alditols over Ru/C, resulting in a high yield (85.5%) of C6 alditols from cellulose. The weak adsorption of C6 alditols on ZrP inhibited the dehydration of C6 alditols to sorbitan. Furthermore, ZrP showed excellent hydrothermal stability and could be reused in several runs [52].

A mesoporous binary phosphate (ZrNbP) with a high surface area, i.e., 170 m²/g, and a large average pore size, i.e., 8 nm, was obtained using a sol-gel method. It was used as a support for Pt nanoparticles [53]. The Pt nanoparticles remained in a monodispersed state on the support. The hydrogenation of furfural and subsequent alcoholysis of FFA were performed over this catalyst using an integrated method. Among the metal-supported catalysts Ir/ZrNbP, Ru/ZrNbP, Rh/ZrNbP, Re/ZrNbP, and Pt/ZrO₂SO₄^2-, the Pt/ZrNbP catalyst gave the highest selectivity, namely 75.7%, for levulinate derivatives. The integrated process shown in Fig. 8 was developed for the direct conversion of furfural to levulinate ester. The catalytic activity and selectivity had not changed significantly after the third run.

Carbohydrates are one of the most important types of biomass feedstock. Catalytic dehydration of C6 carbohydrates generates HMF, which can be used as a precursor for the synthesis of valuable chemicals, high-energy-density fuels, and polymer monomers. Fig. 9 shows that HMF oxidation produces several important furan chemicals, including 2, 5-diformylfuran (DFF), 5-hydroxymethyl-2-furancarboxylic acid (HMFCA), 5-formyl-2-furancarboxylic acid (FFCA), and 2, 5-furandicarboxylic acid (FDCA). A Ru-based ZrP catalyst has been used in the oxidation of biomass-derived HMF under mild conditions. This catalyst was prepared by the exchange of Ru³⁺ with H⁺ in the ZrP structure [13]. The reaction solvent significantly affected the catalytic activity, and the best results were achieved using aromatic solvents. Three oxidation products were detected in the reaction system; DFF was the main oxidation product.

Mesoporous ZrP catalysts have also been used in biodiesel reactions for the effective conversion of long-chain fatty acids to their respective methyl esters [22]. The high acid density, large surface area, and large pore size of mesoporous ZrP favor mass diffusion of substrates, which improves the catalytic activity. These catalysts efficiently converted long-chain fatty acids to their corresponding methyl ester with methanol, and different alcohols to their corresponding esters with glacial acetic acid. This mesoporous solid acid catalyst was recycled 5 times with negligible loss of activity, suggesting high chemical stability of the mesoporous zirconium oxophosphate framework, and good reusability. In addition, Cu species can be incorporated into the α-ZrP layered structure through ion exchange to give a ZPCu catalyst [54]. This has been used as an efficient catalyst for the acetylation of a wide range of alcohols and phenols with acetic anhydride in good to excellent yields under solvent-free conditions, indicating good thermal stability and water resistance during the reaction.

ZrPs have numerous potential applications as supports and catalysts for liquid-phase and vapor-phase reactions. Many approaches to the design of efficient ZrP-based heterogeneous catalysts for various biomass transformations have been proposed. ZrP materials are promising alternative catalysts because of their properties such as high water tolerance, excellent structural stability, and tunable Brønsted and Lewis acidic sites; these all enhance the catalytic activity in acid-related biomass conversions. In particular, ZrP materials can be modified by introducing inorganic or organic compounds into the ZrP matrix, especially by intercalation with metal ions, metal complexes, small organic groups, and polymers, creating unexpected active sites for specific catalytic reactions.

In summary, the synthesis and applications of ZrP-based catalysts are evolving into an exciting research area. Although much effort has been made, further in-depth extensive studies are still required, e.g., to investigate improvements in the synthetic conditions, adjustment of the catalyst acidicity, and expansion of their use in biomass conversion. The development of more efficient ZrP-derived catalysts and more accurate structural design are required for clarification of the catalytic mechanism. These robust ZrP catalysts will play a crucial role in facilitating various versatile and individual organic reactions, especially the conversion of biomass-derived platform molecules. We believe that the structural design of ZrP catalysts will have a particularly large impact on improving their performances, promoting sustainable chemical production, and providing various opportunities and challenges in this field.

Acknowledgments

We gratefully acknowledge the financial support from the National Natural Science Foundation of China (21373082, 21773061), the Innovation Program of Shanghai Municipal Education Commission (15ZZ031), and the Fundamental Research Funds for the Central Universities.