image
Home Journals Progress in Chemistry
Progress in Chemistry

Abbreviation (ISO4): Prog Chem      Editor in chief: Jincai ZHAO

About  /  Aim & scope  /  Editorial board  /  Indexed  /  Contact  / 
Online first  /  Current issue  /  All issues  /  Top access  /  RSS

Progress in Chemistry >

2024 , Vol. 36 >Issue 2: 256 - 270

DOI: https://doi.org/10.7536/PC230615

18

Directed Preparation of 1,3-Propanediol From Glycerol Via Chemoselective Hydrogenolysis Over Bimetallic Catalyst: Active Sites, Structure-Functional Relationship and Mechanism

  • Man Yang , 1, * ,
  • Yuxiang Jiao 1 ,
  • Yujing Ren , 2, *
Expand
  • 1 School of Materials Science and Engineering, Xi’an University of Technology, Xi’an 710048, China
  • 2 Interdisciplinary Research Center of Biology & Catalysis, School of Life Sciences, Northwestern Polytechnical University, Xi’an 710072, China
* Corresponding author e-mail: (Man Yang);
(Yujing Ren).

Received date: 2023-06-19

  Revised date: 2023-11-11

  Online published: 2024-01-08

Supported by

National Key R&D Program of China(2023YFA1506603)

National Natural Science Foundation of China(22002118)

National Natural Science Foundation of China(22208262)

Postdoctoral Research Foundation of China(2020M683528)

Postdoctoral Research Foundation of China(2020TQ0245)

Natural Science Foundation of Shaanxi Provincial Department of Education(21JP086)

Talent Fund of Association for Science and Technology in Shaanxi, China(20230625)

Fold

Abstract

1,3-propanediol is one of the most important monomers in the polyester industry. Catalytic conversion of glycerol to 1,3-propanediol has important application value. In this article, we reviewed the research progress of bimetallic catalysts for the hydrogenolysis of glycerol to 1,3-propanediol, especially emphasizing Pt-W catalytic systems with high catalytic efficiency and great industrial application prospects. By reviewing the interaction between W species, with different microstructures and chemical environments, and Pt metal, as well as the structure-performance relationship between Pt-W dual sites and glycerol hydrogenolysis, the influence of in-situ generated Brønsted acid active species on catalytic activity, selectivity, and stability was summarized, the source of in-situ generated Brønsted acid and catalytic mechanism was discussed, and finally, the development of bimetallic catalysts for selective hydrogenolysis of glycerol to 1,3-propanediol was prospected.

Contents

1 Introduction

2 Catalyst system for selective hydrogenation of glycerol to 1,3-Propandiol

2.1 Tungsten-based catalyst

2.2 Rhenium-based catalyst

2.3 Other catalysts

3 Mechanism of selective hydrogenolysis of glycerol to 1, 3-propanediol

3.1 Dehydration-hydrogenation mechanism

3.2 Etherification-hydrogenation mechanism

3.3 Dehydrogenation-dehydration-hydrogenation mechanism

3.4 Chelation-hydrogenation mechanism

3.5 Mechanism of direct hydrogenolysis

4 Conclusion and outlook

Key words： glycerol; chemoselective hydrogenolysis; 1,3-propanediol; bimetallic catalyst; catalytic mechanism

Cite this article

Man Yang , Yuxiang Jiao , Yujing Ren . Directed Preparation of 1,3-Propanediol From Glycerol Via Chemoselective Hydrogenolysis Over Bimetallic Catalyst: Active Sites, Structure-Functional Relationship and Mechanism[J]. Progress in Chemistry, 2024 , 36(2) : 256 -270 . DOI: 10.7536/PC230615

1 Introduction

With the rapid development of the biodiesel industry, there is a large surplus of glycerol as a by-product of the industry (about 1 kg of glycerol is produced for every 10 kg of biodiesel produced). In addition to biodiesel by-product glycerol, the soap production industry also produces a large amount of by-product glycerol[1]. In addition, starch raw materials (grain, corn, sweet potato, etc.) Can also produce glycerol by microbial fermentation. In recent years, the use of biomass as a raw material to prepare polyols by hydrocracking has gradually emerged, which provides another potential path for the sustainable production of glycerol. Although glycerol has a wide range of applications in the cosmetics, food and pharmaceutical industries, its main use is still limited to the field of fine chemicals, and the existing market can not completely consume the surplus glycerol. Therefore, the conversion of excess glycerol into chemicals with higher added value has become an important topic in today's society[2].
Traditional glycerol conversion methods include biological conversion and chemical conversion. A large number of valuable fine chemicals and liquid fuel additives can be prepared by oxidation, hydrogenolysis, dehydration, pyrolysis gasification, transesterification, esterification, etherification, polymerization and carbonylation[3]. Among the many ways of glycerol conversion, 1,3-propanediol, the hydrogenolysis product, can be used not only as a synthetic raw material for antifreeze, plasticizer, detergent, preservative and emulsifier, but also in food, cosmetics and pharmaceutical industries. Its main use is the reaction of terephthalic acid with 1,3-propanediol to produce a new polyester PTT. PTT overcomes the rigidity of polyethylene terephthalate (PET) and the flexibility of polybutylene terephthalate (PBT), and combines the advantages of the softness of nylon, the bulkiness of acrylic and the stain resistance of polyester, which has attracted the attention of the fiber material industry and has been listed as one of the new chemical fiber materials in the 21st century. Based on the broad application prospect of 1,3-propanediol, the market demand for 1,3-propanediol is increasing. It is reported that the domestic demand for 1,3-propanediol in the downstream consumption market will be 42,800 tons in 2020. According to the current market price of 1,3-propanediol in circulation, the market size is about 1.3 billion yuan. With the continuous expansion of the downstream consumption market of 1,3-propanediol in China, it is estimated that the demand for 1,3-propanediol will reach about 42,800 tons per year in the next five years[4]. In addition, according to the forecast, by 2023, the market price of glycerol in China will be about 3,500 yuan/ton (99.5%), while the price of 1,3-propanediol (99.9%) will be about 29,000 yuan/ton, which is about 10 times that. However, the current commercial source of 1,3-propanediol is not obtained by glycerol hydrogenolysis, but by acrolein hydration hydrogenation and oxirane carbonylation. Although the direct hydrogenolysis of glycerol faces technical problems such as high pressure and low selectivity, the other two methods have problems such as non-renewable raw materials, large equipment investment, serious pollution in the production process and high production cost, which lead to the failure of large-scale industrial production of 1,3-propanediol[4]. On the other hand, 1,3-propanediol can also be obtained by fermentation of glycerol by microorganisms and disproportionation along the reduction pathway[5]. However, because microorganisms can only grow at a low concentration of glycerol, even though the biological method of 1,3-propanediol has the advantages of strong specificity and mild conditions, the problems of low productivity and long cycle still make it difficult to produce 1,3-propanediol on a large scale.
To sum up, the development of new chemical processes for the hydrogenation of glycerol to 1,3-propanediol is the future direction of large-scale production of 1,3-propanediol, which has important research significance and broad application prospects.

2 Development of Catalysts for Selective Hydrogenolysis of Glycerol to 1,3-Propanediol and Study on the Relationship between Active Sites and Structure

Glycerol hydrogenolysis involves a complex reaction network as shown in Fig. 1[6]. Theoretical studies show that the dehydration activation energy of the glycerol terminal hydroxyl group and the secondary hydroxyl group are very close (70.9 kcal/mol vs. 73.2 kcal/mol), and their proton affinities are almost the same (∼ 195 kcal/mole)[7]. In addition, the steric hindrance of hydrogenolysis of the secondary hydroxyl group is much greater than that of the terminal hydroxyl group. These unfavorable thermodynamic factors form a significant challenge for the selective preparation of 1,3-propanediol. Therefore, there is a need to develop efficient catalysts that make the formation of 1,3-propanediol kinetically more favorable. At present, the catalysts used in glycerol hydrogenolysis to 1,3-propanediol with high yield mainly include Pt-W and Ir-Re systems, as well as a small amount of other bifunctional catalytic systems. In this paper, the development of Pt-W catalyst system is mainly discussed, and its glycerol selective hydrogenolysis performance is reviewed, and the further understanding of the relationship between active site and structure-activity is focused on with the continuous development and in-depth study of the catalyst. The Pt-W catalytic system and Ir-Re system were further compared, and their characteristics and advantages and disadvantages were discussed.
图1 甘油氢解反应路径图[6]

Fig. 1 Diagram of the reaction path of glycerol hydrogenolysis[6]

2.1 Tungsten based catalyst

2.1.1 Tungstic acid

In the early studies, homogeneous catalysts were mainly used for the hydrogenolysis of glycerol to obtain 1,3-propanediol. In 1987, Celanese Company used Rh(CO)2(acac)2 as catalyst, added H2WO4, in 1-methyl-2-pyrrolidone solvent, under the condition of （CO/H2=1/2） of synthesis gas, the reaction temperature was 473 K, and the yield of 1,3-propanediol was 17%[8]. When H2SO4 was used instead of H2WO4, the desired product was not formed, indicating that H2WO4 does not play the role of acid alone in this reaction. At the same time, the amount of H2WO4 has a great influence on the reaction activity, and doubling the amount of H2WO4 will lead to a decrease in activity to 1/3 of the original. Later, tungstic acid was introduced to heterogeneous catalytic systems. In 2004, Chaminand et al. Added tungstic acid to Rh/C catalyst and reacted with sulfolane as solvent at 457 K and 80 bar hydrogen pressure for 168 H to obtain the target product 1,3-propanediol[9]. The reaction conversion was 32% and the selectivity was 12%. In 2012, Dam et al. Examined the effect of the addition of tungstic acid on the conversion of glycerol and the selectivity of 1,3-propanediol in four commercial catalyst systems, （Pt/Al2O3, Pt/SiO2, Pd/Al2O3 and Pd/SiO2）[10]. Compared with the addition of hydrochloric acid, the addition of tungstic acid significantly improved the selectivity of 1,3-propanediol. These experimental results show that the addition of tungstic acid is beneficial to the stabilization of the intermediate product (glycerol tungstate), thus improving the yield of the target product 1,3-propanediol.

2.1.2 Tungsten-containing heteropoly acid

Li Yongwang's research group of Shanxi Institute of Coal Chemistry, Chinese Academy of Sciences, studied the effect of tungsten-containing heteropoly acid on the hydrogenolysis performance of glycerol by loading silicotungstic acid on the surface of metal oxides. In 2009, Li Yongwang et al. Used silicon oxide as a carrier to prepare Cu/H4SiW12O40/SiO2 catalyst and used it in glycerol hydrogenolysis reaction[11]. The results showed that a conversion of 83. 4% and a selectivity of 32. 1% were obtained under the reaction conditions of 483 K and 0. 54 MPa. The research group further replaced Cu with noble metal Pt, and the selectivity of 1,3-propanediol could still be maintained at 31.4% when the glycerol conversion was 81.2%[12]. When the support is replaced by zirconia, although the main product is propanol, the catalyst has excellent stability due to the strong interaction between silicotungstic acid and the support. In the team's study, they verified that Brønsted acid (B acid) has a very important role in the hydrogenolysis of glycerol to 1,3-propanediol.
In 2013, Zhu Shanhui and others of the research group modified the catalyst by adding alkali metals (such as Li, K, Rb and Cs) to regulate B acid[13]. The results showed that the selectivity of 1,3-propanediol was greatly improved after the catalyst was further modified by alkali metal Li. Under the reaction conditions of 453 K and 5 MPa, the conversion of glycerol was 43. 5%, the selectivity of 1,3-propanediol was 53. 6%, and the reaction had excellent stability. Similarly, Chen Changlin's group used Pt-H3PW12O40/ZrO2 catalyst to carry out hydrogenolysis experiments on 60% glycerol aqueous solution[14]. Under the reaction conditions of 403 K, 4 MPa and LHSV of 0.25 h−1, the glycerol conversion was 53. 4% and the selectivity to 1,3-propanediol was 44. 5%, and the catalyst did not lose its activity after 100 H. The results showed that tungsten-containing heteropolyacid had more contact opportunities with the substrate than tungstic acid, and could cooperate with the metal component to catalyze the hydrogenolysis of glycerol. In the above reaction system, the yield of 1,3-propanediol was greatly improved when tungstic acid was replaced by tungsten-containing heteropoly acid.
The researchers preliminarily revealed that the active site of selective hydrogenolysis of glycerol mainly depends on the metal-acid bifunctional synergy between the well-contacted hydrogenated noble metal and acidic W species. Among them, the W species not only provides active B acid, but also plays a role in forming glycerol tungstate with glycerol, thus promoting the adsorption of glycerol and the stabilization of intermediate species. In addition, the strong interaction between the active site and the support has a significant effect on the stability of the catalyst.

2.1.3 Supported tungsten oxide

Due to the poor thermal stability of tungsten-containing heteropolyacid in liquid phase, the use of tungsten oxide （WO3） with better thermal stability to replace tungsten-containing heteropolyacid has become the main research direction of Pt-W catalytic system in recent years. In 2008, Kurosaka et Al. Used 1,3-dimethyl-2-imidazolidinone (DMI) as solvent to investigate the effects of different supports (Al-MCM-41, SiO2-Al2O3, Al2O3, anatase TiO2, H-Y zeolite and ZrO2）) on the performance of Pt-W bifunctional catalytic system.The results showed that the ZrO2 support could well disperse the active species of WO3, and the Pt/WO3/ZrO2 catalyst system had the best catalytic activity compared with other noble metals[15]. Under the reaction conditions of 443 K and 8 MPa, the yield of 1,3-propanediol can reach 24%. The authors believe that the interaction between Pt and W and the adsorption and activation of glycerol molecules by WO3 species lead to the high activity of the Pt/WO3/ZrO2 catalytic system. After that, Ding Yunjie's group investigated the activity of Pt-WO3/ZrO2 in different solvents: the conversion of glycerol and the selectivity of 1,3-propanediol were improved when DMI solvent was replaced by DMI-water mixed solvent[16]. The authors believe that the addition of protic solvent water can promote the formation of 1,3-propanediol. Li Debao's research team of Shanxi Institute of Coal Chemistry prepared ZrO2 carriers with different surface hydroxyl groups[17]. It is found that the surface hydroxyl groups can form Zr-O-W bonds with tungsten oxide, thus promoting the generation of active B acid sites. Zongbaoning's team found that Pt/WOx/ZrO2（ tetrahedron) was far more active and selective than Pt/WOx/ZrO2（ monoclinic) in glycerol hydrogenolysis[18]. This is because the ZrO2（ tetrahedron) has more B acid sites and is more favorable for the dispersion of noble metal Pt. At the same time, Chen Changlin's group used ZrO2 as the carrier to adjust the loading of metal and tungsten oxide, and found that when the content of W reached 10%, the WO3 was monolayer distribution, the B acid sites were the most, and the activity was the highest[19]. Under the reaction conditions of 403 K and 4 MPa, the selectivity of 1,3-propanediol could reach 45. 6% when the conversion of glycerol hydrogenolysis reached 70. 2% in a fixed-bed reactor. Li Yongwang's group achieved 52% 1,3-propanediol selectivity over SiO2 modified Pt/WOx/ZrO2 catalyst[20]. Pt and Polytungstate are considered to be the optimal active sites. The addition of silicon oxide is beneficial to the dispersion of Pt and W species, and makes the W species change from crystalline WO3 to polymeric W species. However, an excess of SiO2 would transform the polymeric W species into monodisperse W. Ma et al. Added appropriate Mn promoter to the Pt/WOx/ZrO2 catalyst to control the polymerization degree of WOx[21]. The results show that the addition of Mn leads to the formation of two-dimensional polymeric WOx, which interacts with Pt nanoparticles to promote the formation of B acid, thus accelerating the formation rate of 1,3-propanediol. Through these studies, we have a further understanding of the effect of WO3 dispersion morphology on glycerol hydrogenolysis to 1,3-propanediol. In the supported Pt-W catalyst system, the dispersion state of Pt particles and tungsten oxide plays a key role in the selective hydrogenolysis of glycerol. The improvement of the dispersion of Pt and tungsten oxide is helpful to improve the selective hydrogenolysis performance of glycerol, and the interaction between Pt-W or the interaction between Pt and tungsten oxide and the third component can effectively adjust the dispersion of Pt and tungstic oxide.
In addition to ZrO2, other oxide supports (including TiO2, SiO2, and Al2O3）) were also used to regulate Pt and WOx dispersion. Chu et al. Introduced tungsten oxide into the synthesis of Pt/TiO2 catalyst to introduce B acid sites, and the prepared Pt-W/TiO2 catalyst obtained 51.5% yield of 1,3-propanediol and 24.9% conversion of glycerol[22]. Zong Baoning et al. Found that the crystal structure of TiO2 has an important influence on hydrogen spillover and the interaction between Pt-Ti and W-Ti[23]. Xia Qineng et al. Modified sulfate on the surface of TiO2 to increase B acid sites for the system, and the glycerol conversion of Pt/W-S/Ti catalyst was 22. 5 times higher than that of Pt/W/Ti catalyst[24]. Furthermore, they explored the effects of TiO2 crystal form and tungsten oxide loading on reaction performance, and found that anatase TiO2 was significantly better than rutile TiO2, with excellent stability. The optimum tungsten oxide loading is 5 wt%, which is better than 2 wt% and 10 wt%. At this loading, the tungsten oxide has good dispersion and appropriate W domain, the glycerol conversion can reach 100%, and the yield of 1,3-propanediol is 41%[25]. Gong Leifeng et al of Ding Yunjie's research group prepared a series of Pt/WO3/TiO2/SiO2 catalysts[26]. Through X-ray diffraction (XRD) and transmission electron microscopy (TEM) characterization, the authors found that the introduction of TiO2 played a role in promoting the high dispersion of Pt. Tomishige et al. Used tungsten oxide to modify the Rh/SiO2, and the results showed that the addition of tungsten had little effect on the conversion of glycerol, but the selectivity of 1,3-propanediol was significantly improved[27]. Priya et al. Replaced the support with mesoporous silica SBA-15, and obtained 86% glycerol conversion and 42% 1,3-propanediol selectivity using 10 wt% glycerol aqueous solution as raw material under atmospheric pressure hydrogen. The authors considered that the introduction of tungsten oxide provided B acid sites favorable for the formation of 1,3-propanediol[28]. Darbha et al. Used SBA-15 as a support, which was consistent with that on ZrO2 and TiO2 supports, and also found that with the loading of tungsten oxide gradually increasing from 0.25 wt% to 15 wt%, the dispersion of Pt, the amount of B acid, and the yield of 1,3-propanediol all showed volcano-type changes, confirming that oligomeric tungsten oxide has appropriate amount of B acid and highly dispersed Pt species, which can promote the formation of 1,3-propanediol[29]. Subsequently, β molecular sieve and SAPO-34 molecular sieve were also studied and reported[30⇓~32].
In 2007, Suzuki et al. Used Al2O3 as the carrier of Pt-WOx system, and they impregnated ammonium paratungstate on commercial Pt/Al2O3 to prepare Pt(5 wt%)-W(5 wt%)/Al2O3 catalyst[33]. The catalyst performs excellently in liquid-phase glycerol hydrogenolysis, and the selectivity of 1,3-propanediol can reach 67% at 23% glycerol conversion. In 2013, Kaneda et al. Used boehmite, which is rich in surface hydroxyl groups, as a carrier to prepare a Pt/WOx/AlOOH catalyst[34]. The catalyst is the best catalyst for the hydrogenolysis of glycerol known at present, the conversion rate reaches 100%, and the selectivity of 1,3-propanediol reaches 66%. When the support is replaced by γ-Al2O3 with the same specific surface area or AlOOH with lower specific surface area, such a good effect can not be obtained, which is considered to be related to the number of Al-OH bonds. Similarly, Liu Lei's group also proposed in the Pt/WAlSi catalytic system that the abundant hydroxyl groups on the surface of AlSi can effectively promote the selective hydrogenolysis of glycerol[35]. In addition, researchers have tried to construct macroporous Al2O3 to promote the mass transfer of reactants in order to improve the catalytic performance[36]. On the basis of the above research, Li Yongwang's group prepared a series of Pt/yWOx/Al2O3 catalysts by sequential impregnation method. With the increase of y value (W loading), the morphology of tungsten species changed from single tungsten to polymeric tungsten species, and finally to crystalline WO3[37]. At the same time, the content of B acid also increased. Due to the abundant B acid sites, the strong electronic interaction of Pt and W species, and H overflow, the catalyst with 10 wt% tungsten oxide has the highest activity; The conversion of glycerol was 64.2% and the selectivity of 1,3-propanediol was 66.1%. Garc García-Fern Fernández et al. Prepared a series of Pt/WOx/Al2O3 catalysts with different Pt and WOx contents by sequential impregnation method, and found that the activity of the catalyst increased with the increase of Pt content[38]; With the increase of the content of WOx, the conversion of glycerol and the yield of 1,2-propanediol decreased gradually, but the yield of 1,3-propanediol increased first and then decreased. The authors related the selectivity of 1,3-propanediol to the state of W species, and found that the selectivity of 1,3-propanediol is highest when the content of polymeric W species reaches a maximum （WO3 before the crystal phase appears), because this polymeric W species can delocalize the electrons needed to form B acid. The authors believe that the higher content of Pt is easier to approach the WOx, and the strong interaction between Pt and tungsten oxide is also the main reason for obtaining high activity and selectivity. Wang Aiqin's team at Dalian Institute of Chemical Physics came to a similar conclusion in the study of selective hydrogenolysis of glycerol catalyzed by Pt/WOx/Al2O3, but the difference is that the advantage of two-dimensional polymeric tungsten oxide is that it can better disperse metal Pt.The interaction between Pt and tungsten oxide is regulated to promote the generation of more hydrogen overflow, and the in situ B acid generated by the dissociation and overflow of hydrogen to the surface of tungsten oxide is the active species of the reaction[39]. Two-dimensional polymeric tungsten oxide is more likely to react with overflowing hydrogen to form H+.
So far, the mechanism of glycerol hydrogenolysis to 1,3-propanediol catalyzed by Pt-W system has been further understood. It is believed that the dispersion of tungsten oxide determines the size of tungsten domains. With the increase of W density, the W species gradually changed from single W to polymeric W, and finally formed the WO3 crystal phase. At the same time, with the increase of W density, the content of B acid also showed a trend of first increasing and then decreasing. Most people believe that the in situ B acid produced by hydrogen overflow is the key factor to promote the production of 1,3-propanediol, and the different forms of W species (tungsten domain) have different electronic effects on the overflow H, thus affecting the production of B acid. When the W species is polymeric, the content of B acid is the highest, and the selectivity of 1,3-propanediol is the highest.
With the further study, the conclusion that the dispersion state of tungsten oxide is the key factor affecting the formation of Bronsted acid and catalytic performance has been gradually broken. Ma Xinbin's team recently found that in the WPt/SiO2 catalyst system, the WOx of dimerization can provide the most B acid sites[40]. Zong Baoning's team added a very small amount of tungsten oxide to the SBA-15 support to prepare Pt/W-SBA-15 (W/Si = 1/640) catalyst, in which WOx was mainly dispersed in the form of single W（WO4 tetrahedron)[41]. A high glycerol conversion of 86.8% and a 1,3-propanediol selectivity of 70.8% were obtained with this catalyst. Similarly, Zong Baoning's team prepared W-doped mesoporous silicon foam, promoted the mass transfer of reactant glycerol molecules in the pore channel by using the 3D large pore size of mesoporous silicon foam, and confirmed that the metal-oxide interface between Pt nanoparticles loaded on it and the surrounding monodisperse WO4 tetrahedron was the most active site, and the final yield of 1,3-propanediol could reach 63%[42]. Liu Lei's team constructed monodisperse WO4 species on the surface of Al2O3. With the addition of Pt nanoparticles, the catalyst achieved a high glycerol conversion of 51.7% and a 1,3-propanediol selectivity of 45.7% at a glycerol concentration of 30 wt%[43]. The yield of 1,3-propanediol decreased with the increase of W density in the above catalyst systems. These latest results indicate that the dispersion state of tungsten oxide is not the key factor determining the catalytic performance. However, it is undeniable that the structure and properties of tungsten oxide have a crucial impact on the reaction performance. Therefore, it is necessary to further design and characterize the catalysts to determine the effect of the intrinsic active sites of tungsten oxide on the performance of glycerol selective hydrogenolysis, so as to provide guidance for the further design of glycerol selective hydrogenolysis catalysts.

2.1.4 WOx as carrier

In the Pt/WOx system, tungsten oxide as a carrier can make the metal and tungsten oxide fully contact to play the maximum activity. More importantly, using tungsten oxide directly as the support can avoid the influence of its dispersion on the structural properties of the catalyst, which is conducive to in-depth exploration of the active center of tungsten oxide and its catalytic mechanism. Zhang Tao and Wang Aiqin of Dalian Institute of Chemical Physics have developed Pt/mesoporous TiW composite and Pt/mesoporous WOx catalyst successively, and achieved good performance.It was found that the mesoporous tungsten oxide had a relatively large specific surface area and more oxygen vacancies, which made Pt well dispersed, resulting in a conversion rate of 18% and a selectivity of 39. 3% for 1,3-propanediol, further broadening the scope of W research[44,45]. In 2016, the research group used WCl6 as a precursor to prepare WOx with large specific surface area and high defects by hydrothermal method, and used it as a carrier to prepare Pt/WOx catalyst for glycerol hydrogenolysis reaction. Under the mild reaction conditions of 413 K and 1 MPa, the conversion rate of 5% glycerol aqueous solution can reach 37.4%, and the selectivity of 1,3-propanediol is 35.1%[46]. The authors speculated that the heterolytic cleavage of H2 occurred at the interface between Pt and WOx to form H+/H− during the reaction, and this similar hindered Lewis acid-base pair site on the surface of the catalyst was the key to improve the selectivity of 1,3-propanediol. The team further introduced Au and Nb promoters into the reaction system, and by characterizing the structural properties of the catalyst before and after the introduction of the promoters, it was found that the interaction between Pt and W changed the exposed area of Pt.So that the hydrogen overflow effect is affected, and the surface structure of the tungsten oxide is changed, for example, the hindered Lewis acid-base pair sites on the surface of the tungsten oxide are increased, and the yield of the 1,3-propanediol is effectively improved[47⇓⇓~50]. Liu et al. Prepared a series of WOx supports with different oxygen vacancy contents by adjusting the amount of sodium oleate added during the hydrothermal synthesis of WOx[51]. The results show that the existence of oxygen vacancy is beneficial to the dispersion of Pt, enhances the interaction between Pt and WOx, and increases the Bronsted acid in the catalytic system. At the same time, the mechanism study shows that the oxygen vacancy can effectively adsorb the terminal hydroxyl group of glycerol and provide in situ B acid to break the secondary C — O bond. This study is a major step forward in revealing the true active site of tungsten oxide. In this paper, tungsten oxide was directly used as a carrier to support Pt nanoparticles, and a small amount of alumina was added as an additive, and it was confirmed that the addition of alumina had no effect on the dispersion and valence of Pt nanoparticles[52]. Therefore, by comparing the changes of the structure and properties of tungsten oxide before and after the addition of alumina, the real active sites of tungsten oxide were explored. The results showed that the introduction of alumina promoter increased the unsaturated coordination of tungsten oxide, which was positively correlated with the yield of 1,3-propanediol, indicating that the real active sites of tungsten oxide originated from its unsaturated coordination structure. The yield of 1,3-propanediol was increased by 2 times by introducing alumina promoter.

2.1.5 Catalytic stability

The Pt-W catalyst system has attracted much attention from the industry because of its excellent catalytic activity and regioselectivity. In this case, it is very important to study the deactivation mechanism of Pt-W catalyst system and design practical catalysts with long life. Recently, Wang Aiqin's team studied the stability of Pt/WOx/Al2O3 catalyst in 50% glycerol aqueous solution at a liquid volumetric hourly space velocity of 1 h−1, a gas volumetric hourly space velocity of 1000 h−1, a temperature of 453 K, and a hydrogen pressure of 5 MPa[53]. The results show that the deactivation process of the catalyst is a three-stage process. Among them, the catalytic performance decreased significantly in the first 100 H, and the glycerol conversion decreased by about 12%. Through further characterization of the catalyst after deactivation, it was found that the deactivation of the Pt/WOx/Al2O3 catalyst was mainly caused by Pt particle aggregation (dispersion decreased from 29. 1% to 10. 1%). The decrease in dispersion caused a decrease in the number of in situ B acid sites, which led to catalyst deactivation.

2.1.6 Brief summary

To sum up, the tungsten-based catalyst has excellent catalytic activity and good industrial application prospect. According to the research progress, the research of tungsten-based catalyst system is more and more concentrated, the support is mainly concentrated on zirconia, silica and alumina, and the active metal component is Pt. The other active component W comes from a variety of sources, including tungstic acid, phosphotungstic acid, silicotungsic acid, mesoporous tungsten oxide, composite mesoporous tungsten oxide, polytungstate, etc. Tungstic acid and heteropolyacid/heteropolyacid salt, supported Pt-W and Pt/bulk tungsten catalyst systems were discussed, and the key factor affecting the catalytic performance was revealed ：（1）WOx unsaturated coordination structure.Both experimental evidence and theoretical studies show that tungsten oxide has more unsaturated sites, which is easier to convert dissociated hydrogen into H+（B (acid) active species. (2) Pt dispersity The dissociation rate of ：H2 and the hydrogenation of unsaturated intermediate species are closely related to the Pt dispersity. The smaller the Pt particle, the higher the hydrogenation performance. (3) Pt-W interaction: The reaction occurs at the Pt-W interface and requires a tight fit between Pt and W. The strong interaction between Pt and W is closely related to the exposed area of Pt and hydrogen overflow, which affects the formation of in situ B acid. It can be seen that the active center site has been gradually recognized from the catalyst structure at this stage. However, in general, it is difficult to form an interface between small particles of Pt (even monoatomically dispersed Pt) and monodispersed W with highly unsaturated coordination. Therefore, the design and development of Pt-W interfacial catalysts with high efficiency will become an important research direction.
表1 钨基催化剂在甘油选择氢解制1,3-丙二醇反应中的应用

Table 1 Application of tungsten based catalyst in selective hydrogenolysis of glycerol to 1, 3-propanediol

Catalyst T（K） H2 pressure
（MPa）		 Solvent Glycerol phase Conversion（%） 1,3-Propanediol selectivity（%） ref
Rh(CO)2(acac)+H2WO4 473 32 (syngas) 1-methyl-2-pyrrolidinone liquid 48.0 44.0 8
Rh/C+H2WO4 453 8.0 Sulfone liquid 32.0 12.0 9
Pt/Al2O3 473 4.0 H2O liquid 49.0 28.0 10
Cu-HSiW/SiO2 483 0.54 - vapor 83.4 32.1 11
Pt-HSiW/SiO2 473 6 H2O liquid 81.2 31.4 12
Pt-LiSiW/ZrO2 453 5 H2O liquid 43.5 53.6 13
Pt-HPW/ZrO2 403 4 H2O liquid 53.4 44.5 14
Pt/WO3/ZrO2 443 8 1,3-dimethyl-2-imidazolidinone liquid 85.8 28.2 15
Pt/WO3/ZrO2 443 8 1,3-dimethyl-2-imidazolidinone- H2O liquid 31.6 34.9 16
Pt/W/SiZr 453 5 H2O liquid 90.1 44.5 17
Pt/WO3/ZrO2 413 8 H2O liquid 78.3 64.8 18
Pt/WO3/ZrO2 403 4 H2O liquid 70.2 45.6 19
Pt/WOx/ZrO2 453 5 - vapor 54.3 52 20
Pt/WOx/ZrO2 453 8 H2O liquid 56.2 50.6 21
Pt-W/TiO2 513 3 H2O liquid 24.9 51.5 22
Pt/TiO2+WOx/TiO2 423 4 H2O liquid 97.8 46.9 23
Pt/W-S/Ti 393 4 H2O liquid 100 36 24
Pt/W-S/Ti 393 4 H2O liquid 90 43 24
Pt/W/Ti 413 6 H2O liquid 70.7 58 25
Pt/WO3/TiO2/SiO2 453 5.5 H2O liquid 15.3 50.5 26
Pt/WO3/SBA-15 483 0.1 - vapor 86 42 28
Pt/W/SBA-15 463 8 H2O liquid 75.9 32.1 29
Pt-WOx/SAPO-34 483 6 H2O liquid 44.3 19.2 30
Pt-W/Al2O3 433 3 H2O liquid 23.0 67.0 33
Pt/WOx/AlOOH 453 5 H2O liquid 100 66 34
Pt/WAlSi 433 6 H2O liquid 48 56 34
Pt/WOx/Al2O3 433 5 H2O liquid 64.2 66.1 37
Pt/WOx/Al2O3 473 4.5 H2O liquid 53.1 51.9 38
Pt/WOx/Al2O3 453 5 H2O liquid 80.4 35.3 39
WPt/SiO2 453 8 H2O liquid 64.2 57.2 40
Pt/W-SBA-15 423 4 H2O liquid 86.8 70.8 41
Pt/W-MCFs 423 4 H2O liquid 97 65 42
Pt/WO4/Al2O3 453 5 H2O liquid 51.7 45.7 43
Pt/TiW 453 5.5 H2O liquid 24.2 33.5 44
Pt/m-WO3 453 5.5 H2O liquid 18.0 39.3 45
Pt/WOx 413 1 H2O liquid 37.4 35.1 46
AuPt/WOx 413 1 H2O liquid 81.4 51.6 47
Pt/Nb-WOx 433 5 H2O liquid 40.3 27.5 48
Pt/Au/WO3 428 5 H2O liquid 30.7 54.3 49
Pt/Al-WOx 433 3 H2O liquid 79.0 40.6 52

a Vapor: fixed-bed reactor, Liquid: reaction kettle reactor

2.2 Rhenium based catalyst

From 2010 to now, a series of M-ReOx/SiO2（M=Ir, Ru, Rh, Pd and Pt) and Ir-NOx/SiO2（Re, Cr, Mn, Mo, W and Ag) catalysts have been prepared by Tomishige's group in Japan[54]. Among them, Ir-ReOx/SiO2 catalyst showed excellent catalytic performance for glycerol hydrogenolysis to 1,3-propanediol. At the same time, the introduction of a small amount of H2SO4 into the reaction system can also effectively improve the reaction activity. Under the reaction conditions of hydrogen pressure of 8 MPa, temperature of 393 K, H+/Ir=1, and reaction time of 36 H, Ir-ReOx/SiO2（4 wt%Ir, Re/Ir = 1) showed the best catalytic performance, with glycerol conversion of 81% and 1,3-propanediol selectivity of 46%. Considering the corrosiveness of liquid sulfuric acid to equipment, the complexity of subsequent separation of products and the harm to the environment, the research group investigated more environmentally friendly solid acids such as molecular sieves, silicon aluminum oxides and ion exchange resin Amberlyst-70 to replace H2SO4[55]. Combined with the activity and stability results, H-ZSM-5 showed comparable promotion to H2SO4. The group then investigated the effect of metal promoters such as Rh, Ru, Pd, Ni, Co, Zn, Cu and Ag, and found that the introduction of 0. 9 wt% Ru promoter could promote the performance of the Ir-ReOx/SiO2 catalyst, and the conversion of 67% glycerol aqueous solution was 77. 9%, and the selectivity of 1,3-propanediol was 33. 9%[56]. Tomishige's group further studied the support of Ir-Re system catalyst, and the activity of Ir-Re catalyst supported by rutile TiO2 was much higher than that of SiO2 support reported before[57]. The mechanism study shows that the active site is still the Ir-Re interface, and the interaction between rutile TiO2 and Re can effectively adjust the coordination structure and valence of Re species, thus improving the catalytic performance. On this basis, Zhou Jinghong's team investigated the effect of Pt-Re particle size on the hydrogenolysis performance of glycerol, and the large particle size Pt-Re catalyst was beneficial to the C-O bond cleavage of secondary hydroxyl to produce 1,3-propanediol and n-propanol[58]. Subsequently, the team also investigated the effect of different supports (aluminum-containing and aluminum-free silicon materials) on the Ir-Re catalyst and the effect of calcination temperature on the catalytic performance of Ir-Re/KIT-6 (a mesoporous silicon). On the optimized catalyst, the selectivity of 1,3-propanediol was only 40%, and the conversion rate was 30%[59][60]. In addition, the research group also prepared a new Ir-Re alloy catalyst on KIT-6 support, with the assistance of Amberlyst-15 solid added acid at 393 K and 8 MPa hydrogen pressure, the conversion rate was 63% (20 wt% glycerol aqueous solution), and the formation rate of 1,3-propanediol was 25.6 mol1,3-PDmolIr−1·h−1, which was twice the catalytic activity of ordinary Ir-Re/KIT-6[61]. Seubsai et al. Directly used H-ZSM-5 as a carrier and observed the strong electronic interaction between Ir and Re, but the yield of 1,3-propanediol was only 2.8%[62]. He Dehua's group investigated the catalytic performance of Ru-Re catalyst system in the hydrogenolysis of glycerol to diol, but the yield of 1,3-propanediol was much lower than that of 1,2-propanediol[63,64]. Ding Yunjie's team designed a core-shell Ir-ReOx/SiO2 catalyst, which was concentrated on the outer surface to avoid further diffusion of glycerol into the inner shell for excessive hydrogenolysis, thus improving the selectivity of 1,3-propanediol, but the selectivity of 1,3-propanediol obtained on this catalyst was only 30%[65].
Table 2 compares some typical features of the two catalyst systems, Ir-Re and Pt-W. We can clearly see that the structures of the two are completely different. Ir-Re catalysts are characterized by loading three-dimensional ReOx clusters on the surface of metallic Ir nanoparticles, while Pt-W catalysts usually load metallic Pt nanoparticles on a two-dimensional WOx monolayer. In addition, external acid is often required to obtain excellent catalytic performance in rhenium-based catalyst systems. Compared with W, Re is expensive and easy to lose in aqueous phase reaction, so it is relatively difficult to regulate its activity. Therefore, the application prospect of rhenium-based catalyst system is poor, which leads to the delay of the research of Re-based catalyst in recent years.
表2 Pt-W和Ir-Re体系催化甘油选择氢解反应对比

Table 2 The comparison between Pt-W and Ir-Re catalysts in selective hydrogenolysis of glycerol

Ir-Re Pt-W
structure metal Ir load
3D ReOx		 2D WOx load metal Pt
Ir/Re or Pt/W atomic ratio ~1 1/4 ~ 1/2
reaction temperature 120 ℃ 160~180 ℃
H2 pressure 8 MPa 3~5 MPa
additive H+ none

2.3 Other catalyst

In addition to tungsten- and rhenium-based catalysts, there are some other catalysts that can promote the formation of 1,3-propanediol. In 2011, Oh et al. Synthesized sulfurized ZrO2 solid acid by sol-gel method using Zr(OBu)4 as precursor and n-propanol or sulfuric acid as solvent, and prepared 2%Pt-sulfated/ZrO2 catalyst[66]. Under the reaction conditions of 7. 3 MPa and 443 K, the conversion of glycerol catalyzed by 2%Pt-sulfated/ZrO2 was 66. 5% and 62. 9%, and the selectivity of 1, 3-propanediol was 83. 6% and 19. 6% in DMI and water, respectively, over the catalyst for 24 H. It is worth noting that 23.7% glycerol conversion and 32.1% 1,3-propanediol selectivity were still obtained when DMI was used as solvent without catalyst.
Priya et al. Used different supports and metal combinations to prepare catalysts to explore the performance of gas-phase glycerol hydrogenolysis reaction. In 2014, Priya et al. Used Al(NO)3·6H2O and (NH4)2HPO4 as precursors, hydrolyzed and calcined at 773 K to prepare AlPO4. H2PtCl6 was impregnated by impregnation method, and Pt/AlPO4 catalyst was obtained after calcination at 623 K[67]. The gas phase hydrogenolysis of glycerol over the catalyst was carried out at 0. 1 MPa, 533 K and water as solvent, and the glycerol conversion was 100% and the selectivity to 1,3-propanediol was 35. 4%. By comparing different carriers, the authors found that the acidity of the carrier and the dispersion of the metal were the main factors affecting the formation of 1,3-propanediol. In 2016, the authors selected mordenite （SiO2/Al2O3=20） as the support to prepare 2% Pt/HM catalyst by impregnation method, which was used for gas-phase glycerol hydrogenolysis reaction under the reaction conditions of 0.1 MPa and 498 K, with glycerol conversion of 94.9% and 1,3-propanediol selectivity of 48.6%[68]. In the same year, SBA-15 was loaded with 5 wt% Cu and applied to the gas-phase glycerol hydrogenolysis reaction. Under the reaction conditions of 0.1 MPa and 493 K, the glycerol conversion of 20% was 90.0%, and the selectivity of 1,3-propanediol was 5.0%[50].
In 2015, Vanama et al. Prepared a nanostructured MCM-41 support, which was loaded with noble metal Ru and applied to the gas-phase glycerol hydrogenolysis reaction. Under the reaction conditions of 0.1 MPa and 503 K, the glycerol conversion of 10% was 62.0%, and the selectivity of 1,3-propanediol was 20.0%[69]. Liu Qiying et al. Prepared Ir/Co nanorods with different morphologies by solvothermal method using H2IrCl6·6H2O and cobalt acetate tetrahydrate as precursors. The authors applied them to the hydrogenolysis reaction of glycerol. Under the reaction conditions of 3 MPa and 473 K, the conversion of 10% glycerol aqueous solution was 82.4%, and the selectivity of 1,3-propanediol was 23.1%[70].
In 2019, Cao Yong et al. Loaded monometallic Ir on H-ZSM-5 and prepared IrOx/H-ZSM-5 catalyst for selective hydrogenolysis of glycerol without Re and acid addition[71]. It was found that there was a positive linear correlation between the reactivity and the B acid sites, and the interface between Ir and H-ZSM-5 was the active site of B acid overflow. Finally, under the reaction conditions of 453 K and 8 MPa hydrogen pressure, the selectivity of 1,3-propanediol over the catalyst reached about 70%, and the conversion frequency (TOF) of glycerol was 4.5 h−1.
Although some non-Pt-W and Ir-Re catalysts have achieved satisfactory results in the selective hydrogenolysis of glycerol to 1,3-propanediol, the catalytic activity of most catalysts is far inferior to that of Pt-W and Ir-Re catalysts. Moreover, the preparation process is complicated and the cost is high, so there is little research in recent years.

3 Mechanism of Selective Hydrogenolysis of Glycerol to 1,3-Propanediol

The reaction mechanism of glycerol hydrogenolysis varies with different reaction systems, and there is no consensus on the reaction mechanism of glycerol hydrogenolysis. The study of the reaction mechanism is helpful for us to optimize the catalyst so as to obtain a higher yield of the target product. At present, five reaction mechanisms have been proposed: dehydration-hydrogenation mechanism, dehydrogenation-dehydration-hydrogenation mechanism, chelation-hydrogenation mechanism, etherification-hydrogenation mechanism and direct hydrogenolysis mechanism.

3.1 Dehydration and hydrogenation mechanism

The dehydration and hydrogenation mechanism is now recognized by people (see Figure 2 for the reaction formula). Glycerol is first dehydrated under the action of an acid catalyst to form an intermediate product, and then the intermediate product is rapidly hydrogenated to form a final product. According to this mechanism, the catalyst plays a dual role of metal and acid, with the acid site promoting the dehydration of glycerol and the metal site promoting the hydrogenation of intermediates.
图2 脱水-加氢机理[72]

Fig. 2 Dehydration-hydrogenation mechanism[72]

In terms of selectivity, it is widely believed that the presence of B acid is one of the reasons for the selective formation of 1,3-propanediol. In addition, many studies have correlated the content of B acid with the selectivity of 1,3-propanediol. Tomishige et al. In Japan added liquid sulfuric acid or solid acid H-ZSM-5 to Ir-Re catalyst, and Deng Chenghao et al. Introduced amberlyst-15 into Ir-Re catalyst, and the selectivity of 1,3-propanediol was significantly improved[54][55][58]. In addition, Li Yongwang et al. Found that the acid amount of the catalyst was related to the loading of W species through characterization, and the reaction activity was changed by adjusting the loading of W species[37]. Another strong evidence supporting this mechanism is that Priya et al. Used Pt-WO3/SBA-15 catalyst to detect hydroxyacetone in the catalytic hydrogenolysis of glycerol under atmospheric pressure of hydrogen, and its selectivity reached 17%[28].
In addition, hydroxyacetone is thermodynamically more stable than 3-hydroxypropionaldehyde, so it is thermodynamically preferred; The carbocation intermediate formed by the removal of the intermediate hydroxyl group from glycerol is more stable than the carbocation intermediate formed by the removal of the terminal hydroxyl group, so 3-hydroxypropionaldehyde is kinetically preferred. The competition between the two is also one of the reasons for determining the final product[44].

3.2 Etherification-hydrogenation mechanism

The etherification-hydrogenation mechanism is a supplement to the dehydration-hydrogenation mechanism (see Figure 3 for the reaction formula), in which glycerol is considered to dehydrate a molecule of water on the acid site and then hydrogenate to form propylene glycol. The difference is that the dehydration-hydrogenation mechanism holds that the enol and its keto (aldehyde) tautomer are formed after dehydration, while this mechanism holds that the dehydration is carried out within the molecule of glycerol, that is, the etherification of two adjacent hydroxyl groups of glycerol forms glycidol. The strong evidence of this mechanism is that Liu Haichao et al. And Li Yongwang et al. Detected glycidol as an intermediate product when using copper-based catalysts[73][74]. However, the biggest defect of this mechanism is that it can not judge the difficulty of producing 1,3-propanediol and 1,2-propanediol.
图3 醚化-加氢机理[74]

Fig. 3 Etherification-hydrogenation mechanism[74]

3.3 Mechanism of dehydrogenation-dehydration-hydrogenation

Glycerol hydrogenolysis under alkaline conditions is generally considered to follow the dehydrogenation-dehydration-hydrogenation mechanism (reaction formula is shown in Figure 4). Glycerol is first dehydrogenated to glyceraldehyde under the action of metal, and this process is reversible; Then glyceraldehyde is dehydrated to 2-hydroxyacrolein and hydrogenated to 1,2-propanediol (route a); Keto-enol tautomerism can also occur to produce methylglyoxal; Methylglyoxal can be hydrogenated to hydroxyacetone, and then hydrogenated to 1,2-propanediol (route B); Lactic acid can also be obtained by Cannizzaro reaction (route C); In addition, glyceraldehyde can undergo retro-aldol reaction to give hydroxyacetaldehyde and formaldehyde, which can be hydrogenated to give ethylene glycol and methanol (route d)[75]. There are studies supporting this mechanism by adding base (such as LiOH or NaOH) and catalyst activity correlation[76,77]. It is worth noting that the formation of 1,3-propanediol by this mechanism is difficult. Because it is difficult to obtain 3-hydroxyacrolein, the precursor of 1,3-alcohol, by dehydration of glyceraldehyde.
图4 脱氢-脱水-加氢机理[75]

Fig. 4 Dehydrogenation-dehydration-hydrogenation mechanism[75]

3.4 Chelation-hydrogenation mechanism

The reason for the chelation-hydrogenation engine was proposed by Chaminand et al., and the roadmap is shown in Figure 5[9]. They believe that the active metal can chelate with the two hydroxyl groups of glycerol to form a five-membered or six-membered ring, and then hydrogenolyze to obtain 1,2-propanediol and 1,3-propanediol. Because six-membered rings are more stable than five-membered rings (but the formation of six-membered rings is more hindered), they should be more inclined to form 1,3-propanediol, but their reaction products are 1.The predominance of 2-propanediol may be due to the fact that the ring formation step is not the key step of the reaction, or it may be due to the fact that the catalyst system used by the authors does not contain an active metal that can chelate the glycerol hydroxyl group. This mechanism does not give the scope of application for metal components, and the initial steps and specific routes of the reaction are not described in detail, so it is not highly recognized.
图5 螯合-加氢机理[9]

Fig. 5 Chelation-hydrogenation mechanism[9]

3.5 Direct hydrogenation mechanism

Among the glycerol hydrogenation reactions described above, the direct hydrogenation mechanism is the best speculation at present. The advantage of this mechanism is that it can reflect the difference of different acidic components. The essence lies in the good explanation of the reason why these two catalyst systems can selectively produce 1,3-propanediol: in the first step, the terminal hydroxyl group of glycerol is adsorbed on rhenium oxide or tungsten oxide, thus protecting the terminal hydroxyl group.
Around the reaction mechanism of Ir-Re/SiO2 system, Professor Tomishige of Japan has done a more in-depth study[27,54⇓~56,78]. Based on the reaction trend, kinetics, isotope labeling and other results of different substrates, he proposed a route (Fig. 6): the terminal hydroxyl group of glycerol first reacts with Re-OH to form 2,3-dihydroxyglyceride, then the H2 is activated at the interface between Ir and Re, and the resulting H− attacks the carbon on the secondary C-O bond of 2,3-dihydroxyglyceride, resulting in the cleavage of the C-O bond, dehydration to form 3-hydroxyglyceride, and finally hydrogenation to form 1,3-propanediol. The steric hindrance between the middle hydroxyl group of glycerol and rhenium oxide to form glyceride is greater than that between the terminal hydroxyl group, and the carbocation intermediate formed by attacking the middle hydroxyl group is more stable[79]. In addition, the six-membered ring formed by the activated H− on Ir attacking the middle hydroxyl group is more stable than the seven-membered ring formed by attacking the terminal hydroxyl group, so the terminal hydroxyl group of glycerol is protected to form 1,3-propanediol in the reaction[72].
图6 Ir-ReOx催化剂上直接氢解机理[72]

Fig. 6 Mechanism of direct hydrogenolysis on Ir-ReOx catalyst[72]

There are three indirect evidences to support this mechanism: first, when several different alcohols are used as substrates, the alcohols containing secondary hydroxyl groups have higher activity[72]; Secondly, when water was replaced by alcohols as solvent, the activity of the catalyst decreased a lot, which may be due to the competitive adsorption between alcohols and glycerol, thus occupying part of the active sites[27]; The third is to change the concentration of glycerol in the reaction solution, the amount of glycerol conversion is basically unchanged, which may be due to the fact that glycerol adsorption on the surface of rhenium oxide has been in a saturated state[72].
Similar to the Ir-Re/SiO2 system, Garc García-Fern Fernández et al also proposed a similar reaction mechanism for the Pt-WO3/Al2O3 system: tungsten oxide plays a role similar to rhenium oxide, and its surface hydroxyl can react with glycerol to form glyceride[38]. Several research teams have verified the adsorption of glycerol terminal hydroxyl groups and oxides by ATR infrared characterization[80,81].
Different from the Ir-Re system, in the Pt-W system, it is usually considered to be H+（B acid/protonic acid) to attack the oxygen on the secondary C-O bond. The roadmap is shown in Fig. 7. With the further study of the catalytic mechanism of the catalyst, the source of the H+ has been more deeply understood, and more evidence shows that the H+ in the reaction does not come from the intrinsic Bronsted acid sites on the surface of tungsten oxide. In the selective hydrogenolysis of glycerol to 1,3-propanediol catalyzed by Pt/WOx/ZrO2, Professor Ma Xinbin's research team found that the overflow of H2 was linearly related to the conversion rate of glycerol, indicating that H2 not only participated in the hydrogenation step, but also played an important role in the activation step of C — O bond[82]. Professor Wang Aiqin of Dalian Institute of Chemical Physics also found that H2 overflowed after dissociation on the metal surface.The dissociated H2 is heterolytically cleaved at the metal-oxide interface or the oxygen vacancy of the metal oxide during the overflow process to form a H+ species, which is positively correlated with the terminal diol product yield of C — O bond selective hydrogenolysis (Fig. 8)[39,46,47,83]. Similarly, Professor Zong Baoning and Professor Liu Lei have put forward similar conclusions in Pt/WOx/SBA-15 and Pt/WOx/Ti2O5 catalyst systems respectively[41][84].
图7 Pt-WOx催化剂上直接氢解机理[80]

Fig. 7 Mechanism of direct hydrogenolysis on Pt-WOx catalyst[80]

图8 Pt-WOx催化剂上原位B酸的测量及其与催化性能的关系[83]

Fig. 8 Measurement of in-situ B-acid on Pt-WOx catalyst and its relation to catalytic performance[83]

Focusing on this mechanism, the researchers further explored the special catalytic performance of Ir-Re and Pt-W combinations. Why is only the combination of Ir and Re, Pt and W the best for the selective hydrogenolysis of glycerol? Other combinations of precious metals and oxyphilic oxides are not satisfactory? This problem has been deeply puzzling researchers. In this context, our research team used tungsten oxide directly as a carrier to load Pt nanoparticles, added a small amount of alumina as an additive, and confirmed that the addition of alumina had no effect on the dispersion and valence of Pt nanoparticles[52]. Therefore, by comparing the changes of the structure and properties of tungsten oxide before and after the addition of alumina, the origin of the activity of tungsten oxide was explored. It is found that the addition of alumina promoter can effectively adjust the strong metal support interaction (SMSI) between noble metal and tungsten oxide, thus adjusting the exposure area of noble metal and regulating the hydrogen overflow. Oxygen in tungsten oxide is removed during hydrogen flooding, resulting in the formation of unsaturated defect sites. On the one hand, the defect sites on the surface of tungsten oxide contribute to the adsorption of glycerol terminal hydroxyl, on the other hand, they can further promote hydrogen overflow and activate it to produce H+（ Fig. 9). Therefore, the special combination of Pt and W can be attributed to the metal-support interaction of Pt and W matching. Avoid too strong interactions to coat the precious metal and too weak interactions, both of which affect hydrogen overflow.
图9 Pt-WOx催化剂上氧空位与1,3-丙二醇的关系及氧空位参与的直接氢解机理[52]

Fig. 9 Relationship between oxygen vacancies and 1,3-propanediol on Pt-WOx catalysts and the mechanism of direct hydrogenolysis involving oxygen vacancies[52]

The direct hydrogenolysis mechanism has a deeper understanding of the adsorption of glycerol and the stability of intermediate products, and each reaction step has been described in detail, and the role of tungsten and rhenium has also been explained. However, it is different from the H− induced direct hydrogenolysis mechanism of Ir-Re system and the tungsten oxide intrinsic B acid site induced direct hydrogenolysis mechanism. In the above catalytic mechanism, Pt is responsible for the dissociation of hydrogen, the dissociated H overflows at the Pt-W interface, and the H species overflowing to the surface of tungsten oxide will be converted into H+ active species at the unsaturated coordination of tungsten oxide. For the above three possible mechanisms, the development of in situ characterization techniques that can directly detect the intermediates is an important direction for the basic research of this reaction in the future.

4 Summary and Prospect

1,3-propanediol is an important synthetic monomer for many polymers in industry, especially for polyester PTT. In order to achieve sustainable development, it is of great significance to design a thermal catalytic system to use renewable glycerol as a raw material to replace the traditional fossil raw materials for the preparation of 1,3-propanediol. After nearly 30 years of development, the catalysts for glycerol hydrogenolysis to 1,3-propanediol have gradually changed from the original Ir-Re catalytic system to the more promising Pt-W catalytic system, and the reaction conditions have changed from organic solvents and high temperature and pressure (3. 1 MPa) to aqueous phase reaction, lower temperature and pressure (393 K). These changes make the reaction conditions more and more environmentally friendly and mild, and the activity and selectivity of 1,3-propanediol are gradually improved.
In this paper, the current representative catalysts for the conversion of glycerol to 1,3-propanediol are reviewed, and the catalytic performance of Pt-W catalytic system and the key factors affecting the catalytic activity and selectivity are emphatically sorted out. The results show that the excellent catalytic performance of the Pt-W catalyst system comes from the good match between Pt and W atoms, which is conducive to the effective transfer of hydrogen from the hydrogenation metal to the variable valence metal. In addition, the catalytic reaction mechanism dominated by Bronsted acid in Pt-W catalyst system was discussed in detail, and the source and action behavior of Bronsted acid were reviewed.
At present, the industrialization of the catalytic hydrogenolysis of glycerol to 1,3-propanediol needs to be further developed. The catalytic activity, selectivity and stability need to be further improved. It is hoped that this paper will provide useful information for the development of practical and efficient catalysts for the selective hydrogenolysis of glycerol to 1,3-propanediol.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Zhou C H, Zhao H, Tong D S, Wu L M, Yu W H. Catal. Rev., 2013, 55(4): 369.

[2]
Fan L M, Wang J H, Pei W. Zhejiang Chemical Industry, 2009, 40 (6): 22.

(樊利民, 王菊华, 裴文. 浙江化工, 2009, 40 (6): 22.)

[3]
Forsberg C W. Appl. Environ. Microbiol., 1987, 53(4): 639.

[4]
Li S, Li J. Fine and Specialty Chemicals, 2022, 30(3): 12.

(李烁, 李靖. 精细与专用化学品, 2022, 30(3): 12.)

[5]
Behr A, Eilting J, Irawadi K, Leschinski J, Lindner F. Green Chem., 2008, 10(1): 13.

[6]
Nakagawa Y, Tamura M, Tomishige K. J. Mater. Chem. A, 2014, 2(19): 6688.

[7]
Nimlos M R, Blanksby S J, Qian X H, Himmel M E, Johnson D K. J. Phys. Chem. A, 2006, 110(18): 6145.

[8]
Che T M. Production of propanediols: U.S. Patent, 1987, 4: 2.

[9]
Chaminand J, Djakovitch L A, Gallezot P, Marion P, Pinel C, Rosier C. Green Chem., 2004, 6(8): 359.

[10]
Ten Dam J, Djanashvili K, Kapteijn F, Hanefeld U. ChemCatChem, 2013, 5(2): 497.

[11]
Huang L, Zhu Y L, Zheng H Y, Ding G Q, Li Y W. Catal. Lett., 2009, 131(1): 312.

[12]
Zhu S H, Zhu Y L, Hao S L, Zheng H Y, Mo T, Li Y W. Green Chem., 2012, 14(9): 2607.

[13]
Zhu S H, Gao X Q, Zhu Y L, Zhu Y F, Xiang X M, Hu C X, Li Y W. Appl. Catal. B Environ., 2013, 140/141: 60.

[14]
Chen C L, Song M J, Qin L Z. J. Nanjing Univ. Technol. Nat. Sci. Ed., 2011, 33(1): 1.

陈长林, 宋敏洁, 秦丽珍. 南京工业大学学报(自然科学版), 2011, 33(1): 1.)

[15]
Kurosaka T, Maruyama H, Naribayashi I, Sasaki Y. Catal. Commun., 2008, 9(6): 1360.

[16]
Gong L F, Lv Y, Ding Y J, Lin R H, Li J W, Dong W D, Wang T, Chen W M. Chin. J. Catal., 2009, 30(12): 1189.

(龚磊峰, 吕元, 丁云杰, 林荣和, 李经伟, 董文达, 王涛, 陈维苗. 催化学报, 2009, 30(12): 1189.)

[17]
Zhao J X, Hou B, Guo H Q, Jia L T, Niu P Y, Chen C B, Xi H J, Li D B, Zhang J L. ChemCatChem, 2022, 14(14): e202200341.

[18]
Fan Y Q, Cheng S J, Wang H, Tian J, Xie S H, Pei Y, Qiao M H, Zong B N. Appl. Catal. B Environ., 2017, 217: 331.

[19]
Qin L Z, Song M J, Chen C L. Green Chem., 2010, 12(8): 1466.

[20]
Zhu S H, Gao X Q, Zhu Y L, Cui J L, Zheng H Y, Li Y W. Appl. Catal. B Environ., 2014, 158/159: 391.

[21]
Zhou W, Luo J, Wang Y, Liu J F, Zhao Y J, Wang S P, Ma X B. Appl. Catal. B Environ., 2019, 242: 410.

[22]
Luo Z C, Zhu Z G, Xiao R, Chu D W. Chem., 2023, 18(3): e202201046.

[23]
Zeng Y, Jiang L, Zhang X X, Xie S H, Pei Y, Zhou G B, Hua W M, Qiao M H, Li Z H, Zong B N. ACS Sustainable Chem. Eng., 2022, 10(29): 9532.

[24]
Zhou Z M, Jia H Y, Guo Y, Wang Y G, Liu X H, Xia Q N, Li X, Wang Y Q. ChemCatChem, 2021, 13(18): 3953.

[25]
Wang Y J, Zhou Z M, Wang C, Zhao L H, Xia Q N. Front. Chem., 2022, 10: 1004925.

[26]
Gong L F, Lu Y, Ding Y J, Lin R H, Li J W, Dong W D, Wang T, Chen W M. Appl. Catal. A Gen., 2010, 390(1/2): 119.

[27]
Shinmi Y, Koso S, Kubota T, Nakagawa Y, Tomishige K. Appl. Catal. B Environ., 2010, 94(3/4): 318.

[28]
Priya S S, Kumar V P, Kantam M L, Bhargava S K, Srikanth A, Chary K V R. Ind. Eng. Chem. Res., 2015, 54(37): 9104.

[29]
Bhowmik S, Enjamuri N, Sethia G, Akula V, Marimuthu B, Darbha S. Mol. Catal., 2022, 531: 112704.

[30]
Shi G J, Xu J Y, Song Z G, Cao Z, Jin K, Xu S H, Yan X T. Mol. Catal., 2018, 456: 22.

[31]
Miao G, Shi L, Zhou Z M, Zhu L J, Zhang Y F, Zhao X P, Luo H, Li S G, Kong L Z, Sun Y H. ACS Catal., 2020, 10(24): 15217.

[32]
Zhu L J, Sun M Y, Zhao X P, Zhang Y F, Luo H, Liu W, Miao G, Kong L Z. Sustain. Energy Fuels, 2021, 5(6): 1747.

[33]
Suzuki N, Takahashi M, Tamura M. JP Patent. 2012, 207326849A.

[34]
Arundhathi R, Mizugaki T, Mitsudome T, Jitsukawa K, Kaneda K. ChemSusChem, 2013, 6(8): 1345.

[35]
Feng S H, Zhao B B, Liu L, Dong J X. Ind. Eng. Chem. Res., 2017, 56(39): 11065.

[36]
Numpilai T, Cheng C K, Seubsai A, Faungnawakij K, Limtrakul J, Witoon T. Environ. Pollut., 2021, 272: 116029.

[37]
Zhu S H, Gao X Q, Zhu Y L, Li Y W. J. Mol. Catal. A Chem., 2015, 398: 391.

[38]
García-Fernández S, Gandarias I, Requies J, Güemez M B, Bennici S, Auroux A, Arias P L. J. Catal., 2015, 323: 65.

[39]
Lei N, Zhao X C, Hou B L, Yang M, Zhou M X, Liu F, Wang A Q, Zhang T. ChemCatChem, 2019, 11(16): 3903.

[40]
Zhou W, Li Y, Wang X F, Yao D W, Wang Y, Huang S Y, Li W, Zhao Y J, Wang S P, Ma X B. J. Catal., 2020, 388: 154.

[41]
Fan Y Q, Cheng S J, Wang H, Ye D H, Xie S H, Pei Y, Hu H R, Hua W M, Li Z H, Qiao M H, Zong B N. Green Chem., 2017, 19(9): 2174.

[42]
Cheng S J, Fan Y Q, Zhang X X, Zeng Y, Xie S H, Pei Y, Zeng G F, Qiao M H, Zong B N. Appl. Catal. B Environ., 2021, 297: 120428.

[43]
Zhao B B, Liang Y, Liu L, He Q, Dong J X. ChemCatChem, 2021, 13(16): 3695.

[44]
Zhang Y H, Zhao X C, Wang Y, Zhou L K, Zhang J Y, Wang J, Wang A Q, Zhang T. J. Mater. Chem. A, 2013, 1(11): 3724.

[45]
Liu L J, Zhang Y H, Wang A Q, Zhang T. Chin. J. Catal., 2012, 33(7/8): 1257.

[46]
Wang J, Zhao X C, Lei N, Li L, Zhang L L, Xu S T, Miao S, Pan X L, Wang A Q, Zhang T. ChemSusChem, 2016, 9(8): 784.

[47]
Zhao X C, Wang J, Yang M, Lei N, Li L, Hou B L, Miao S, Pan X L, Wang A Q, Zhang T. ChemSusChem, 2017, 10(5): 819.

[48]
Yang M, Zhao X C, Ren Y J, Wang J, Lei N, Wang A Q, Zhang T. Chin. J. Catal., 2018, 39(6): 1027.

[49]
Yang C J, Zhang F, Lei N, Yang M, Liu F, Miao Z L, Sun Y N, Zhao X C, Wang A Q. Chin. J. Catal., 2018, 39(8): 1366.

[50]
Harisekhar M, Kumar V P, Priya S S, Chary K V. J. Chem. Technol. Biotechnol., 2015, 90(10):1906.

[51]
Niu Y F, Zhao B B, Liang Y, Liu L, Dong J X. Ind. Eng. Chem. Res., 2020, 59(16): 7389.

[52]
Yang M, Wu K Y, Sun S D, Ren Y J. Appl. Catal. B Environ., 2022, 307: 121207.

[53]
Lei N, Miao Z L, Liu F, Wang H, Pan X L, Wang A Q, Zhang T. Chin. J. Catal., 2020, 41(8): 1261.

[54]
Nakagawa Y, Shinmi Y, Koso S, Tomishige K. J. Catal., 2010, 272(2): 191.

[55]
Nakagawa Y, Ning X H, Amada Y, Tomishige K. Appl. Catal. A Gen., 2012, 433/434: 128.

[56]
Tamura M, Amada Y, Liu S B, Yuan Z L, Nakagawa Y, Tomishige K. J. Mol. Catal. A Chem., 2014, 388/389: 177.

[57]
Liu L J, Asano T, Nakagawa Y, Tamura M, Okumura K, Tomishige K. ACS Catal., 2019, 9(12): 10913.

[58]
Deng C H, Duan X Z, Zhou J H, Chen D, Zhou X G, Yuan W K. Catal. Today, 2014, 234: 208.

[59]
Deng C H, Leng L, Duan X Z, Zhou J H, Zhou X G, Yuan W K. J. Mol. Catal. A Chem., 2015, 410: 81.

[60]
Deng C H, Leng L, Zhou J H, Zhou X G, Yuan W K. Chin. J. Catal., 2015, 36(10): 1750.

[61]
Deng C H, Duan X Z, Zhou J H, Zhou X G, Yuan W K, Scott S L. Catal. Sci. Technol., 2015, 5(3): 1540.

[62]
Chanklang S, Mondach W, Somchuea P, Witoon T, Chareonpanich M, Faungnawakij K, Seubsai A. Catal. Today, 2022, 397/399: 356.

[63]
Ma L, He D H. Top. Catal., 2009, 52(6): 834.

[64]
Ma L, He D H. Catal. Today, 2010, 149(1/2): 148.

[65]
Luo W T, Lyu Y, Gong L F, Du H, Wang T, Ding Y J. RSC Adv., 2016, 6(17): 13600.

[66]
Oh J, Dash S, Lee H. Green Chem., 2011, 13(8): 2004.

[67]
Priya S S, Kumar V P, Kantam M L, Bhargava S K, Chary K V R. Catal. Lett., 2014, 144(12): 2129.

[68]
Priya S S, Bhanuchander P, Kumar V P, Dumbre D K, Periasamy S R, Bhargava S K, Lakshmi Kantam M, Chary K V R. ACS Sustainable Chem. Eng., 2016, 4(3): 1212.

[69]
Vanama P K, Kumar A, Ginjupalli S R, Chary Komandur V R. Catal. Today, 2015, 250: 226.

[70]
Liu Q Y, Cao X F, Wang T J, Wang C G, Zhang Q, Ma L L. RSC Adv., 2015, 5(7): 4861.

[71]
Wan X Y, Zhang Q, Zhu M M, Zhao Y, Liu Y M, Zhou C M, Yang Y H, Cao Y. J. Catal., 2019, 375: 339.

[72]
Amada Y, Shinmi Y, Koso S, Kubota T, Nakagawa Y, Tomishige K. Appl. Catal. B Environ., 2011, 105(1/2): 117.

[73]
Wang S, Liu H C. Catal. Lett., 2007, 117(1): 62.

[74]
Huang L, Zhu Y L, Zheng H Y, Li Y W, Zeng Z Y. J. Chem. Technol. Biotechnol., 2008, 83(12): 1670.

[75]
Montassier C, Ménézo J C, Moukolo J, Naja J, Hoang L C, Barbier J, Boitiaux J P. J. Mol. Catal., 1991, 70(1): 65.

[76]
Falcone D D, Hack J H, Klyushin A Y, Knop-Gericke A, Schlögl R, Davis R J. ACS Catal., 2015, 5(10): 5679.

[77]
Yuan Z L, Wu P, Gao J, Lu X Y, Hou Z Y, Zheng X M. Catal. Lett., 2009, 130(1): 261.

[78]
Tomishige K, Nakagawa Y, Tamura M. Green Chem., 2017, 19(13): 2876.

[79]
Guan J, Chen X F, Peng G M, Wang X C, Cao Q, Lan Z G, Mu X D. Chin. J. Catal., 2013, 34(9): 1656.

[80]
García-Fernández S, Gandarias I, Requies J, Soulimani F, Arias P L, Weckhuysen B M. Appl. Catal. B Environ., 2017, 204: 260.

[81]
Copeland J R, Santillan I A, Schimming S M, Ewbank J L, Sievers C. J. Phys. Chem. C, 2013, 117(41): 21413.

[82]
Zhou W, Zhao Y J, Wang Y, Wang S P, Ma X B. ChemCatChem, 2016, 8(23): 3663.

[83]
Wang B, Liu F, Guan W X, Wang A Q, Zhang T. ACS Sustainable Chem. Eng., 2021, 9(16): 5705.

[84]
Zhao B B, Liang Y, Liu L, He Q, Dong J X. Green Chem., 2020, 22(23): 8254.

Options
Outlines

/

〈 〉