Process for encapsulation of zeolite encapsulated transition metal particles

Encapsulating Ni nanoparticles in MFI zeolites through a post-synthetic treatment enhances ethylene glycol yield and selectivity by addressing the limitations of existing catalysts, achieving high yields and improved catalyst stability for biomass conversion.

WO2024243137A1PCT designated stage expired Publication Date: 2024-11-28SURENDHRAN ROSHAAN +1

Patent Information

Application Number
PCT/US2024/030223
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-19
Filing Date
2024-05-20
Publication Date
2024-11-28

AI Technical Summary

Technical Problem

Current methods for converting biomass to ethylene glycol face challenges such as low selectivity and yield due to the indiscriminate reduction of sugars by Ni-based catalysts, equilibrium limitations in retro-aldol reactions, and instability of first-row transition metals in alkaline synthesis conditions, leading to poor selectivity and efficiency in ethylene glycol production.

Method used

A post-synthetic treatment for encapsulating Ni nanoparticles inside MFI zeolites, using a high-silica MFI zeolite impregnated with Ni and reducing it, followed by leaching with a bulky acid to remove external Ni particles, resulting in size-selective hydrogenation catalysts that improve ethylene glycol yield by minimizing sugar alcohol formation.

Benefits of technology

This approach achieves a 63% yield of ethylene glycol from glucose and extends to polysaccharides with minimal production of hexitols, demonstrating improved selectivity and reusability of the catalysts, and is scalable and cost-effective.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2024030223_28112024_PF_FP_ABST
    Figure US2024030223_28112024_PF_FP_ABST
Patent Text Reader

Abstract

A method of forming a catalyst material by selective encapsulation of one or more transition metals within a porous substrate by introducing a transition metal onto a microporous and / or mesoporous support material by impregnation or ion exchange to deposit the transition metal as metal or metal oxide nanoparticles on the support material and treating the material obtained an organic to substantially leach any transition metal deposited outside the pores of the support material. Also, a catalyst composition, produced by the method, and a method for producing short-chain polyols from biomass-derived mono- and / or polysaccharides using the catalyst composition.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] PROCESS FOR ENCAPSULATION OF ZEOLITE ENCAPSULATED TRANSITION METAL PARTICLES

[0002] Background of the Invention

[0003] Increasing interest in the production of commercial chemicals from renewable sources has motivated significant research in the field of biomass conversion to ethylene glycol (EG).

[0004] Ethylene glycol (EG) is a high-volume chemical used primarily in the synthesis of polymers, as a solvent, and as a heat-transfer fluid. The prevalent industrial process for ethy lene gly col production involves the epoxidation of ethylene, followed by the hy dration of the ethylene oxide intermediate, which requires multiple reactors and separation units [1], A majority of the global ethylene glycol supply is produced through this method utilizing ethylene sourced predominantly from non-renewable natural gas, naphtha, and coal [1], The direct conversion of cellulose or glucose to ethylene glycol has been the subject of research in recent years [2- 4] .

[0005] The direct conversion of cellulosic biomass to glycols is believed to have been first reported in 2008 using a Ni-W2C catalyst [5], Since then, a number have studies have explored the impact of reaction conditions, the choice of retro-aldol and hydrogenation catalysts, supports, and catalyst composition on ethylene glycol yields [2-4], The W catalyst is responsible for scission of glucose to glycol aldehyde (GA) via the retro-aldol reaction. However, the retro-aldol reaction is equilibrium limited and favors glucose over glycolaldehyde and erythrose [6], This equilibrium limitation can be overcome by utilizing a one-pot tandem strategy that irreversibly transforms the glycolaldehyde to ethylene glycol via hydrogenation. Such systems that combined a W-based retro-aldol catalyst and a Ni or Ru based hydrogenation catalyst show 70% yields of ethylene glycol from cellulose [2-4],

[0006] However, typical hydrogenation catalysts like Ni / SiO2 are not size-selective and catalyze the reduction of both glucose and glycolaldehyde. In general, Ni-based catalysts exhibit indiscriminate reduction of sugars, resulting in a mixture of C2-C6 polyols [7-9], Thus, the selectivity towards ethylene glycol is sensitive to Ni-W compositions that optimize retro-aldol and hydrogenation rates [9], At high Ni:W ratios, with greater exposed Ni surface area, the selectivity to ethylene glycol is drastically reduced and more sorbitol is produced, while at low Ni:W ratios, the coverage ofNi by W contributes to improved ethylene glycol selectivity [9,10], Consequently, the ethylene glycol selectivity for solid mixtures of Ni and W catalysts are worse than for the composite Ni-W catalysts

[0011] , Additionally, under reaction conditions, dissolution of the active W species as soluble tungstic acid exposes more Ni surface sites, shifting the effective Ni:W ratio to higher values. This is anticipated to lower ethylene glycol selectivity and increase sorbitol yield when the catalyst is reused [4,12],

[0007] While direct conversion of cellulose is a valuable strategy, the challenges associated with processing raw biomass have necessitated cellulose and hemicellulose hydrolysis to glucose and xylose prior to upgrading of the sugars to valuable chemicals [13.14], However, the Ni-W systems developed for cellulose conversion exhibit lower selectivity with glucose [9,11], This is primarily due to the semi-batch nature of cellulose conversion, which involves hydrolysis followed by retro-aldol and hydrogenation, resulting in a low steady-state concentration of glucose, consequently reducing the extent of side reactions such as glucose hydrogenation to sorbitol, isomerization, and dehydration to 5- hydroxy methyl furfural and other byproducts. Additionally, low glucose concentrations and high temperatures favor more glycolaldehyde production in the equilibrium limited retro- aldol reaction. For these reasons, glucose conversion to ethylene glycol over Ni-W has only been able to achieve < 30% yield in batch studies [9,11], Semi-batch systems can reach close to 60% ethylene glycol yield with Ni-W catalysts

[0012] . or 75% yield with Ru-W catalysts

[0015] , but still requires harsh conditions (> 220 °C and 50 barg H2).

[0008] Zeolite-encapsulated metal catalysts have been a growing field of interest with several strategies for the synthesis of such materials [18-21], Methods for selective encapsulation of metals in zeolites can be classified into 3 groups - direct hydrothermal synthesis, inter-zeolite transformation, and recrystallization of zeolite or silica particles [18— 22], While these methods are more commonly used for encapsulation of precious metals (Pt, Ru, Rh etc), their application to 1st row transition metals is more limited with a few notable examples. This can largely be attributed to the lower stability of 1st row transition metals under alkaline synthesis conditions. Iglesia and co-workers have shown encapsulation of Ni inside ZSM-5 by direct hydrothermal synthesis with chelating ligands [23,24], and through inter-zeolite transformation

[0025] , Alternatively, recrystallization of metal impregnated silica or zeolite precursors can make hollow-zeolite particles with encaged metal particles [10,22,26,27], In general, these methods have all encapsulated Ni inside aluminosilicate zeolites, but the Bronsted acidity in aluminosilicate zeolites can catalyze dehydration reactions of sugars, resulting in poor selecti \ i ty

[0009] Prior work on glucose isomerization over Sn-zeolites has shown the lower activity over medium-pore Sn-MFI in comparison to large-pore materials like Sn-BEA

[0016] , This has been attributed to the greater transport resistance for glucose in medium pore zeolites

[0001] , Additionally, this property of MFI zeolites to exclude hexoses has been leveraged for retro-aldol and 1,2-hydride shift reaction for the tandem production of alkyd lactates from fructose

[0017] ,

[0010] While major advancements in this field over the last 15 years have enabled the development of new processes for the conversion of lignocellulosic biomass, it is imperative to improve the process selectivity and yield to enable commercial implementation.

[0011] Summary of the Invention

[0012] To this end, the inventors developed a simple, scalable, post-synthetic treatment for the encapsulation of Ni nanoparticles inside MFI zeolites (Ni@MFI). To improve the hydrogenation selectivity7in these systems, active Ni nanoparticles are encapsulated inside MFI zeolites. By utilizing a zeolite-encapsulated Ni catalyst, glucose hydrogenation can be minimized allowing for improved ethylene glycol selectivity.

[0013] The size-selective sieving of glucose (Glu) from glycolaldehyde (GA) by the MFI framework enables significant improvement in ethylene glycol yield by minimizing sugar alcohol formation (sorbitol and mannitol). This is particularly relevant for direct conversion of glucose to ethylene glycol which can achieve 63% yield with <2% sugar alcohols in batch operation at approximately^ 50 °C lower reaction temperature (170 °C vs 230 °C). The use of size-selective hydrogenation catalyst enabled a 50 °C reduction in reaction temperature and the highest reported yields of ethylene glycol from glucose in batch operation. As a consequence of the size-selectivity, the Ni@MFI catalyst loading is agnostic to the system performance for Ni:W > 0.2. This is in stark contrast with previously reported Ni-W composite catalysts or unencapsulated Ni / SiCh, where ethylene glycol selectivity is strongly dependent on the catalyst composition. This improvement in sizeselectivity can also be extended to polysaccharides (starch and cellulose).

[0014] To this end, a method for encapsulation of Ni inside zeolite particles by a simple post-synthetic treatment has been developed. Commercially available high-silica MFI zeolite is first impregnated with Ni and reduced. The resulting material has Ni particles located on the external surface as well as inside the pore network. Subsequently, the Ni particles on the external surface are removed by leaching with a bulky acid that is excluded from the internal pore network of the zeolite. The leached catalysts which exhibit size- selective hydrogenation are used in the tandem system to improve ethylene glycol selectivity by minimizing hexose hydrogenation.

[0015] By encapsulating the active Ni nanoparticles inside zeolitic frameworks, size- selective hydrogenation of glycolaldehyde by limiting the transport of glucose to the active site is achieve. MFI framework type materials, with largest pore diameter of 5.5 A, can exclude glucose due to its larger size and provide the necessary selectivity for glycolaldehyde reduction to ethylene glycol. These size-selective hydrogenation catalysts are combined with W-based retro-aldol catalysts to enable the direct conversion of mono- and polysaccharides to ethylene glycol in high yields, with minimal production of hexitols (sorbitol and mannitol). The reusability of these catalysts is also demonstrated.

[0016] The invention enables the preparation of such materials by introducing metal particles exclusively inside the porous zeolite structure. As a post-synthetic strategy, the methodology works both for zeolites synthesized in the laboratory or purchased from a zeolite vendor. The process is to first indiscriminately introduce metal nanoparticles to the zeolite material, with some metal ending up on the inside of the framework and some on the outside. Subsequently, sterically bulky acids are used to dissolve all the metal particles on the external surface of the zeolite crystal, while retaining all the metal particles inside the pores. With this methodology, zeolite catalysts containing first-row transition metals such as manganese, iron, cobalt, nickel, copper, and zinc can be prepared.

[0017] Optionally, galvanic displacement allows the composition of the encapsulated metal nanoparticle to be expanded to include precious metals such as ruthenium, rhodium, palladium, platinum, silver, iridium, and gold. Galvanic displacement leverages the difference in reduction potentials between first-row transition metals and precious metals to replace the former with the latter. As a result, zeolite-encapsulated metal particles containing any of the listed metals can be made as mono or bimetallic particles. Since this method does not use high temperatures and complex chemicals, it can be a reliable and scalable method to make such materials.

[0018] The inventive methodology for encapsulation uses an acid leaching and galvanic displacement methodology to create the material, in distinction to current technologies, which construct / reconstruct the silicate or aluminosilicate zeolite around a metal particle. The inventive method has the advantage of being more broadly applicable to a range of zeolitic structures. It can be easily implemented with commercially available precursors, making it a cheaper alternative to exiting methods. The materials are reproducible and scalable. It enables the successful encapsulation of first row transition metals, which is challenging for existing methods due to the instability of the metal salts under alkaline conditions. The optional galvanic displacement method allows the inclusion of other catalytically relevant metals. The resulting product can be bimetallic or monometallic, which is of interest for a wide variety of applications.

[0019] According to a first aspect, a method of forming a catalyst material by selective encapsulation of one or more transition metals within a porous substrate is provided, the method comprising the steps of: a. introducing a transition metal onto a microporous and / or mesoporous support material by impregnation or ion exchange of a salt or organometallic compound of the transition metal; b. drying the support material onto which the transition metal has been introduced in step a; c. thermally treating the dried support matenal obtained in step b at a temperature of 250 C to 700 C in the presence of an inert gas, oxygen, or hydrogen to at least partially agglomerate the transition metal as metal or metal oxide nanoparticles deposited on the support material; and d. treating the thermally treated support material obtained in step c with an aqueous solution of an organic acid having a smallest molecular dimension that is larger than a largest dimension of a pore opening of the microporous and / or mesoporous support material to substantially leach any transition metal deposited on the support material as metal or metal oxide outside the pores of the support material. According to another aspect, a method according to the first aspect is provided, wherein the transition metal comprises manganese, iron, cobalt, nickel, copper, and / or zinc, preferably nickel.

[0020] According to another aspect, a method according to the first or second aspect is provided, wherein the support material comprises a microporous and / or mesoporous aluminosilicate, aluminophosphate. and / or silicoaluminophosphate zeolite and / or zeotype material or a microporous and / or mesoporous carbon material, preferably a microporous aluminosilicate zeolite and / or zeotype material, more preferably an MFI zeolite, most preferably a high-silica MFI zeolite.

[0021] According to another aspect, a method according to any of the previous aspects is provided, wherein step d is performed with an oxidizing agent, preferably oxygen and / or hydrogen peroxide.

[0022] According to another aspect, a method according to any of the previous aspects is provided, wherein the organic acid comprises acetic and / or trimesic acid and the support material comprises an MFI zeolite.

[0023] According to another aspect, a method according to any of the previous aspects is provided, further comprising the steps of: e. reducing the catalyst material obtained in step l.d of the first aspect under hydrogen or in the presence of a chemical reducing agent; and f. contacting the reduced catalyst composition obtained in step e of the present aspect under an inert and / or hydrogen atmosphere with a solution of a noble metal salt in which the noble metal is present in an oxidized state, wherein the noble metal is introduced into the pores of the support material.

[0024] According to another aspect, a method according to any of the previous aspects is provided, wherein the noble metal comprises one or more of platinum, palladium, ruthenium, silver, gold, rhodium, and iridium, preferably platinum.

[0025] According to another aspect, a method according to any of the previous aspects is provided, wherein the noble metal is encapsulated as a monometallic or bimetallic particle within the porous substrate. According to another aspect, a method according to any of the previous aspects is provided, wherein the noble metal salt solution comprises water or a polar organic solvent, preferably water and / or an alcohol.

[0026] According to another aspect, a catalyst composition is provided, produced by the method of any of the previous aspects.

[0027] According to another aspect, a method for producing short-chain polyols from biomass-derived mono- and / or polysaccharides using the catalyst composition of the previous aspect is provided, comprising the steps of: a. reacting an aqueous solution comprising biomass-derived mono- and / or polysaccharides comprising one or more of cellulose, hemicellulose, glucose, fructose, xylose, glycolaldehyde, starch, and cellulose with hydrogen in the presence of the catalyst composition and optionally one or more retro- aldol catalysts at a temperature of 50 C to 250 C, preferably 100 C to 200 C, more preferably 125 C to 200 C, most preferably 150 C to 175 C, a pressure of 5 to 100 barg, preferably 5 to 75 barg. more preferably 10 to 50 barg. most preferably 20 to 40 barg. and a pH of 2 to 6, preferably 4 to 5, for a time of 0.5 to 24 hours, preferably 1 to 10 hours, and at a catalyst loading of 0. 1 wt.% to 10 wt.%, based on the solution weight; and b. recovering one or more of the short-chain polyols from the aqueous solution.

[0028] According to another aspect, a catalyst composition is provided, comprising a microporous and / or mesoporous high-silica-content MFI zeolite and / or zeotype support material encapsulating nanoparticles comprising nickel and / or nickel oxide.

[0029] According to another aspect, a catalyst composition according to the previous aspect is provided, additionally comprising one or more noble metals encapsulated in the support material.

[0030] According to another aspect, a catalyst composition according to either of the previous two aspects is provided, wherein the one or more noble metals comprise one or more of platinum, palladium, ruthenium, silver, gold, rhodium, and iridium. According to another aspect, a catalyst composition according to any of the previous three aspects is provided, wherein the support material has an external surface that is substantially free of the nickel and / or nickel oxide and / or the one or more noble metals.

[0031] Brief Description of the Drawing Figures

[0032] FIG. 1 plots change in Ni content for 4% Ni / MFI leached with different acids;

[0033] FIG. 2 plots GA reduction over Ni / MFI leached with acetic, citric, or trimesic acid;

[0034] FIG. 3 plots glucose reduction over Ni / MFI leached with acetic, citric, or trimesic acid;

[0035] FIG. 4 plots XRD and N2 Physisorption for parent MFI;

[0036] FIG. 5 plots TPR of 5% Ni / SiCh;

[0037] FIG. 6 plots TPR of (I) Ni@MFI after leaching, (II) Ni / MFI and (III) Ni / SiCh after impregnation and reduction at 350 °C;

[0038] FIG. 7 plots TPR of (I) Ni / MFI after leaching, (II) Ni / MFI and (III) Ni / SiCh after impregnation and reduction at 350 °C;

[0039] FIG. 8 plots TPR of fresh and used Ni@MFI;

[0040] FIG. 9 plots glucose and glycolaldehyde yield over Ni catalysts at different temmperatures;

[0041] FIG. 10 compares glucose and glycolaldehyde hydrogenation for Ni / MFI and Ni@MFI 600 °C;

[0042] FIG. 11 is a reaction schematic showing the similarity between epimerization and retro-aldol reaction of Mo and W catalysts;

[0043] FIG. 12 plots C-shift and retro-aldol activity for (NH4)eMo7O24 and Na<5[H2Wi204o] catalysts; FIG. 13 is two tables showing C-Shift reactions for different Mo and W oxides as catalysts;

[0044] FIGS. 14A-14E plot GCMS analysis for byproducts of reactions of glycolaldehyse amd glucose under slow hydrogenation regimes;

[0045] FIG. 15 compares tandem reaction of glucose over unencapsulated and encapsulated Ni catalysts at 140 °C;

[0046] FIG. 16 shows the impact of Ni@MFI loading on product distribution for glucose conversion to ethylene glycol and the effect of Ni:W mass ratio on the ethylene glycol yield for Ni / SiCh and Ni@MFI;

[0047] FIGS. 17A and 17B show the effect of catalyst loading on product distribution for glucose conversion to ethylene glycol;

[0048] FIG. 18 shows the impact of SMT loading at fixed Ni@MFI loading, impact of Ni@MFI loading on the product distribution for glucose conversion to ethylene glycol at higher W loading, and impact of W loading on the EG yield for tandem system with Ni@MFI;

[0049] FIGS. 19A and 19B show the effect of W loading and differences for Ni / SiO2 and Ni@MFI;

[0050] FIG. 20 shows the effect of reaction temperature on product distribution of glucose conversion to ethylene glycol;

[0051] FIGS. 21A and 21B show the effect pf glucose concentration on product distribution;

[0052] FIG. 22 shows reuse experiments for glucose and starch tandem reactions;

[0053] FIG. 23 shows the impact of reaction time on the product distribution of tandem reaction with starch; FIG. 24 shows the impact of impact of temperature and solution pH on tandem reaction of starch;

[0054] FIG. 25 is an XRD of fresh Microcrystalline Cellulose (MCC) and after ball milling for different times;

[0055] FIGS. 26A and 26B show the impact of ball milling on tandem reactions of cellulose;

[0056] FIG. 27 shows the difference in the product distribution of the tandem system with SMT paired with Ni / SiO-2, Ni / ZSM-5 or Ni'^ZSM-5;

[0057] FIG. 28 is a table of BET surface are and C-values for different Sil-1 and MFI materials;

[0058] FIG. 29 shows BET transform and linear fit for MFI, Sil-1, Ni / MFI, Ni / Sil-1, Ni@MFI and Ni@Sil-l;

[0059] FIG. 30 shows fitted regions for t-plot for MFI. Sil-1, Ni / MFI, Ni / Sil-1, Ni@MFI and Ni@Sil-l;

[0060] FIG. 31 describes calibration of the TCD for temperature programmed reductions (TPR);

[0061] FIG. 32 describes calibrations for x-ray fluorescence spectroscopy (XRF);

[0062] FIG. 33 shows retention times and calibration statistics for different analytes quantified herein;

[0063] FIG. 34 describes calculations for the degree of encapsulation and effectiveness factor;

[0064] FIG. 35 shows the synthetic protocol for encapsulation of Ni particles inside MFI zeolite;

[0065] FIG. 36 shows the impact of different leaching conditions for 4% Ni@Sil- 1 on sorbitol yields; FIG. 37 shows conversion, yield, and <D for glucose and glycolaldehyde hydrogenation overNi@Sil-l and Ni@MFI materials;

[0066] FIG. 38 shows glucose and glycolaldehyde hydrogenation of parent 4% Ni / Sil-1 and leached Ni@Sil-l;

[0067] FIG. 49 shows Ch:Ni for synthesis at different scales;

[0068] FIG. 40 shows change in Ni content for 4% Ni / MFI leached with different acids;

[0069] FIG. 41 shows surface areas and micropore volume for Sil-1 and MFI materials:

[0070] FIG. 42 shows glucose and glycolaldehyde reduction over 4% Ni / MFI leached with acetic, citric or trimesic acid;

[0071] FIG. 43 shows X-ray diffractograms for MFI and Sil-1 materials;

[0072] FIG. 44 shows a TPR of impregnated 5% Ni / SiO2;

[0073] FIG. 45 shows TPRs for Ni / MFI, Ni / SiCh, and Ni@MFI materials;

[0074] FIG. 46 shows glucose and glycolaldehyde hydrogenation for Ni@MFI reduced at 600 °C;

[0075] FIG. 47 shows a TPR of Ni@MFI after use in reaction under 170 °C;

[0076] FIG. 48 shows quantitative TPR results for Ni materials;

[0077] FIG. 49 shows glucose and glycolaldehyde hydrogenation over Ni catalysts at different temperatures;

[0078] FIG. 50 shows the degree of encapsulation for Ni@MFI and Ni@Sil-l materials;

[0079] FIG. 51 shows effectiveness factors for MFI and Sil-1 materials;

[0080] FIG. 52 shows a synthetic method for galvanic displacement deposition of Pt2+on Ni. @ Sil- 1 ; FIG. 53 describes the change in cell potential for galvanic displacement deposition ofPt2+;

[0081] FIG. 54 shows the impact of galvanic displacement deposition of Pt2+onto Ni@Sil- 1 on glucose and glycolaldehyde hydrogenation;

[0082] FIGS. 55A and 55B describe calibration curves and statistics for glucose reactions;

[0083] FIG. 56 shows C -shift and retro-aldol activity for ammonium heptamolybdate and sodium metatungstate catalysts;

[0084] FIG. 57 shows C-shift and retro-aldol activity for different Mo and W oxides with 55 mM glucose at 140 °C;

[0085] FIG. 58 shows the effect of temperature on C-shift and retro-aldol reactions;

[0086] FIG. 59 shows post-reaction GCMS analysis for identification of products without catalyst;

[0087] FIG. 60 shows post-reaction GCMS analysis for identification of products with Ni@MFI;

[0088] FIG. 61 shows post-reaction GCMS analysis for identification of products with SMT catalyst;

[0089] FIG. 62 shows post-reaction GCMS analysis for identification of products with both SMT and Ni@MFI catalysts;

[0090] FIG. 63 shows post-reaction GCMS analysis for identification of products with both SMT and Ni@MFI;

[0091] FIG. 64 shows the impact of reducing pretreatment with Ni / SiO2 on the C-shift and retro-aldol activity of Mo and W oxide catalysts;

[0092] FIG. 65 shows interactions between MoOs and Ni catalysts;

[0093] FIG. 66 shows a reaction network for the conversion of glucose to polyols; FIG. 67 shows a comparison of tandem reaction of glucose over 2 loadings of unencapsulated and encapsulated Ni catalysts;

[0094] FIG. 68 shows the impact of W loading on the ethylene glycol yield for tandem system with Ni@MFI and Ni / SiCh;

[0095] FIG. 69 shows the impact of Ni r / MFI and Ni / SiCh loading on product distribution for glucose conversion to ethylene glycol;

[0096] FIG. 70 shows reuse experiments for tandem glucose or starch conversion to ethylene glycol over Ni r / MFI catalyst;

[0097] FIG. 71 shows the impact of Ni / SiO2 loading on the product distribution for glucose conversion to ethylene glycol at higher W loading;

[0098] FIG. 72 shows the impact of SMT loading at fixed Ni@MFI loading and the impact of Ni@MFI loading on the product distribution for glucose conversion to ethylene glycol at higher SMT loading;

[0099] FIG. 73 shows the impact of reaction temperature on product distribution of glucose conversion to ethylene glycol and the impact of product distribution at 100% conversion for different temperatures achieved by changing reaction time;

[0100] FIG. 74 shows a time profile for tandem reaction of glucose at 140 °C and 160 °C;

[0101] FIG. 75 shows the impact of glucose concentration on product distribution and the impact of Glucose:W ratio with fixed Ni:W ratio at higher glucose concentrations;

[0102] FIG. 76 shows equilibrium yield of glycolaldehyde at different initial concentration glucose;

[0103] FIG. 77 shows the impact of reaction time on the product distribution of tandem reaction with starch;

[0104] FIG. 78 shows reuse experiments for tandem glucose or starch conversion to ethylene glycol over Ni <7 FI catalyst; FIG. 79 is an XRD of fresh microciystalline cellulose (MCC) and after ball milling for different times; and

[0105] FIG. 80 shows the impact of ball milling on cellulose conversion to ethylene glycol and the impact of solution pH on BMC conversion to ethylene glycol.

[0106] Detailed Description of the Invention

[0107] Definitions

[0108] “First row transition metals" refers to the following elements: Manganese (Mn), Iron (Fe), Cobalt (Co), Nickel (Ni), Copper (Cu), and Zinc (Zn).

[0109] “Microporous materials” refers to aluminosilicate zeolite and zeotype materials, preferably of “high silica” composition. This can also refer to aluminophosphate (A1PO) and silicoaluminophosphate (SAPO) materials of similar structures. This can also refer to microporous carbon materials.

[0110] “Mesoporous materials” refers to zeolite and zeotype materials that possess both micropores and mesopores (often referred to as hierarchical zeolites).

[0111] “Method for selective encapsulation” refers to a selective acid leaching of transition metal nanoparticles from the external surfaces of transition metal impregnated microporous or mesoporous materials.

[0112] Method of Forming, a Catalyst Material

[0113] A method forming a catalyst material by selective encapsulation of one or more transition metals within a porous substrate is provided. The method comprises the steps of: a. introducing a transition metal onto a microporous and / or mesoporous support material by impregnation or ion exchange of a salt or organometallic compound of the transition metal; b. drying the support material onto which the transition metal has been introduced in step a; c. thermally treating the dried support material obtained in step b at a temperature of 250 C to 700 C in the presence of an inert gas, oxygen, or hydrogen to at least partially agglomerate the transition metal as metal or metal oxide nanoparticles deposited on the support material; and d. treating the thermally treated support material obtained in step c with an aqueous solution of an organic acid having a smallest molecular dimension that is larger than a largest dimension of a pore opening of the microporous and / or mesoporous support material to substantially leach any transition metal deposited on the support material as metal or metal oxide outside the pores of the support material.

[0114] The transition metal, which catalyzes hydrogenation, preferably comprises one or more first-row transition metals selected from manganese, iron, cobalt, nickel, copper, and / or zinc, preferably nickel. Second- and third-row transition metals are less preferred for the method mentioned above as they are challenging to re-oxidize and dissolve. However, they may optionally also be incorporated into the catalyst material, as will be described further below.

[0115] The support material can be any microporous or mesoporous material capable of supporting catalytic transition metal species, including microporous aluminosilicate zeolite and zeotype materials, preferably of "‘high silica’7composition. Zeolites are a class of microporous alumino-silicate materials with important applications for catalysis, ion exchange, gas separation, and adsorption. Also known as molecular sieves, these materials have sub-nanometer pore mouths capable of separating molecules by size. Based on these structural characteristics, zeolitic materials are broadly classified as small pore, medium pore, or large pore. Other suitable microporous materials include aluminophosphate (A1PO) and silicoaluminophosphate (SAPO) materials of similar structures, as well as microporous carbon materials. Also suitable are mesoporous zeolite and zeotype materials that possess both micropores and mesopores (often referred to as hierarchical zeolites). The support material is preferably a microporous aluminosilicate zeolite and / or zeotype material, preferably a zeolite of MFI, CHA, LTA, BEA, FAU, and / or RHO type, more preferably an MFI type zeolite, most preferably a high-silica MFI-type zeolite.

[0116] Zeolite encapsulated transition materials are made by using either a commercially available or laboratory-synthesized high silica zeolite as the support. The calcined zeolite support is impregnated with a first-row transition metal salt and then subjected to reducing treatment to create metal particles. At this stage, the material has both particles encapsulated in the zeolite and non-encapsulated particles on the external surface of the zeolite crystal. Exclusive encapsulation is achieved by treating the impregnated material wi th multiple leaching treatments involving a sterically bulky organic acid (one that is too large to enter the pores of the zeolite) and an optional oxidant (O2 or H2O2) to facilitate the leaching process.

[0117] The transition metal is introduced onto the microporous and / or mesoporous support material by impregnation or ion exchange of a salt or organometallic compound of the transition metal from an aqueous solution, as is well-known in the art. After drying, the thermal treatment forms nanoparticles of the transition metal or an oxide thereof on the internal and external surfaces of the support material. Thermally treated support material thus obtained is further treated with an aqueous solution of an organic acid having a smallest molecular dimension that is larger than a largest dimension of a pore opening of the microporous and / or mesoporous support material to substantially leach any transition metal deposited on the support material as metal or metal oxide outside the pores of the support material. The organic acid solution may include an oxidizing agent, preferably oxygen and / or hydrogen peroxide, to aid with the leaching process.

[0118] Any organic acid is capable of use in the leaching process, provided, of course, that it is capable of being excluded from the interior of the support material pores by its molecular size relative to the pore size of the support material. Suitable acids may be selected from sulfonic and / or carboxylic acids, for example, acetic acid, citric acid, benzoic acid, isophthalic acid, trimesic acid, and polystyrene sulfonic acid. The main criterion for the choice of acid is the molecular size; the choice of acid depends on the support material pore size. For example, trimesic acid is suitable for MFI; a larger acid would be necessary for other zeolites with larger pores. Pore sizes of typical zeolites: CHA - 3.7 A; LTA - 4.1 A; MFI - 5.5 A; BEA - 6.7 A; FAU - 7.4 A. In a particularly preferred embodiment, the support material comprises an MFI-type zeolite and the organic acid comprises acetic and / or trimesic acid.

[0119] In a further embodiment, the catalyst material encapsulating the transition metal obtained as described above can be further modified to encapsulate one or more additional catalytic metals within the porous support material. Second and third row catalytically active transition metals, including noble metals, are not suitable for the encapsulating method just described as they are challenging to re-oxidize and dissolve in the leaching step. However, once the first transition metal is encapsulated according to the method previously described, one or more additional catalytic metals can be introduced into and encapsulated within the support material by a galvanic or redox process. Preferred among the catalytically active second- or third-row transition metals platinum, palladium, ruthenium, silver, gold, rhodium, and iridium, most preferably platinum.

[0120] Due to the difference in reduction potential between first row transition metals and second- or third-row transition metals, the initially encapsulated first-row transition metal can be partially or even entirety replaced within the support material by galvanic displacement from a solution containing one or more second- or third-row transition metals or metal ions. The one or more additional metals or salts thereof are dissolved in an aqueous or polar organic solvent, preferably water and / or an alcohol, In an alternate embodiment, the catalyst material encapsulating the first transition metal obtained as described above is reduced with H2 and the one or more additional metals are deposited directly onto the initially encapsulated transition metal by liquid-phase reductive deposition to produce an encapsulated bi- or multi-metallic catalyst material.

[0121] Method of Producing Short-Chain Polyols

[0122] A method for producing short-chain polyols from biomass-derived mono- and / or polysaccharides is provided, using the catalyst composition formed by selective encapsulation as just described. While the present disclosure in one aspect of the invention pertains to the production of ethylene glycol, the catalyst materials produced according to the invention are useful for producing mixtures of two- and three-carbon polyols, including ethylene glycol, propylene glycol, and glycerol, for example, from biomass feedstocks. Depending on the biomass feedstock, the yields of these polyols using the same catalyst system will vary, but the main products are ethylene and propylene glycols.

[0123] Biomass generally describes plant-derived carbonaceous matter including agricultural products and byproducts, hard and softwood from timber industries and forestland, paper, cardboard, certain textile wastes and food wastes. The precise mixture of polyols / glycols will depend on the starting sugars in the biomass-derived feedstock. Such feedstocks can include sugars like glucose, fructose, or xylose or polysaccharides like cellulose or hemicellulose. Any of the aldoses and ketoses are useful; fructose and xylose can produce more propylene glycol based on the same reaction pathway. Based on natural abundance and commercial availability the most likely feedstocks will be glucose, fructose or xylose, along with the corresponding polysaccharides starch, cellulose and hemicellulose. The reaction system can be extended to starch and microcrystalline cellulose as substrates, with modifications to reaction conditions for enhanced polysaccharide hydrolysis.

[0124] The method of producing short-chain polyols uses a tandem reaction scheme, wherein the feedstock sugars are fragmented by retro-aldol reaction, and the smaller glycoaldehyde fragments are hydrogenated to yield the desired short-chain polyols, such as ethylene glycol. This method requires the use of two catalysts: a retro-aldol catalyst and the selectively encapsulated mono- or multi-metallic catalyst prepared according to the present invention as previously described. Molybdenum and tungsten oxides have shown uti 1 i ty as retro-aldol catalysts, but due to the deactivation of MoOx in the presence of Ni and H2 under reaction conditions, soluble tungsten oxides are preferred as the retro-aldol catalyst. Tungsten is believed to be the most suited catalyst for retro-aldol catalysis under the conditions for practicing the method, but any combination of retro-aldol catalyst and inventively encapsulated catalyst could be used, with a readily available tungsten catalyst (oxide / carbide / salt) being most preferred. The desired size-selectivity is retained in the tandem system with sodium metatungstate as the retro-aldol catalyst and an inventively encapsulated hydrogenation catalyst. The method of producing short-chain polyols includes the step of reacting an aqueous solution comprising biomass-derived mono- and / or polysaccharides comprising one or more of hemicellulose, glucose, fructose, xylose, glycolaldehyde, starch, and cellulose with hydrogen in the presence of the catalyst composition and optionally one or more retro-aldol catalysts. The concentration of biomass-derived sugar or sugars in the feedstock solution can range from 0.1 wt.% to 10 wt.%, preferably from 0.1 wt.% to 5 wt. %, more preferably from 0.1 wt.% to 4 wt. %, more preferably from 0. 1 wt.% to 3 wt. %, more preferably from 0. 1 wt.% to 2 wt. %. and most preferably from 0. 1 wt.% to 1 wt. %, based on the weight of the solution. The process is carried out at a temperature of 50 C to 250 C, preferably 100 C to 200 C, more preferably 125 C to 200 C, most preferably 150 C to 175 C, at a pressure of 5 to 100 barg, preferably 5 to 75 barg, more preferably 10 to 50 barg, most preferably 20 to 40 barg, at a pH of 2 to 6, preferably 4 to 5, for a reaction time of 0.5 to 24 hours, preferably 1 to 10 hours, and at a catalyst loading of 0.1 wt.% to 10 wt.%, based on the solution weight. The one or more of the short-chain polyols can be recovered from the aqueous solution.

[0125] Catalyst Composition

[0126] A catalyst composition, produced by the method as previously described, is also provided. Preferably, the selectively encapsulated catalyst material comprises a microporous and / or mesoporous high-silica-content MFI zeolite and / or zeotype support material encapsulating nanoparticles comprising nickel and / or nickel oxide. High silica content means preferably as silica content of at least 90 wt.% or higher, including 100 wt.%. The selectively encapsulated catalyst material may also contain one or more additional noble metals encapsulated in the support material. The noble metals are preferably chosen from one or more of platinum, palladium, ruthenium, silver, gold, rhodium, and iridium, most preferably platinum. The support material of the selectively encapsulated catalyst composition has an external surface that is substantially free of nickel and / or nickel oxide and / or the one or more noble metals that are encapsulated within the support material pores.

[0127] Within this specification embodiments have been described in a way which enables a clear and concise specification to be written, but it is intended and will be appreciated that embodiments may be variously combined or separated without departing from the invention. For example, it will be appreciated that all preferred features described herein are applicable to all aspects of the invention described herein.

[0128] In some embodiments, the invention herein can be construed as excluding any element or process step that does not materially affect the basic and novel characteristics of the compositions and methods described herein. Additionally, in some embodiments, the invention can be construed as excluding any element or process step not specified herein.

[0129] The foregoing description of various forms of the invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Numerous modifications or variations are possible in light of the above teachings. The forms discussed were chosen and described to provide the best illustration of the principles of the invention and its practical application to thereby enable one of ordinary7skill in the art to utilize the invention in various forms and with various modifications as are suited to the particular use contemplated. All such modifications and variations are within the scope of the invention as determined by the appended claims when interpreted in accordance wi th the breadth to which they are fairly, legally, and equitably entitled.

[0130] The following non-limiting examples further sen e to illustrate various aspects and embodiments of the invention.

[0131] EXAMPLES

[0132] Example 1

[0133] The impact of reaction parameters such as catalyst loadings, temperature, pH, and substrate concentration were studied and optimal conditions for high selectivity was identified for both glucose and starch.

[0134] Methods

[0135] Materials:

[0136] Nickel (II) nitrate hexahydrate (Alfa Aesar, 19.8% Ni min), trimesic acid (Fisher Scientific, 98%), citric acid (Alfa Aesar, 99+%), acetic acid (Fisher Scientific, glacial), sulfuric acid (Fisher Scientific, concentrated), H2O2 solution (Fisher Scientific, 30%), fumed silica (Cab-O-Sil M5), high silica MFI zeolite (Zeolyst, ZD97037, Lot #: 1822-15), D-glucose (Fisher Scientific, 38.97% C), starch (Fisher scientific, 38.25% C), microcrystalline cellulose (Alfa Aesar, 40.21% C), D-fructose (Fisher Scientific), D- mannose (Alfa Aesar, 99%), glycol aldehyde dimer (Sigma Aldrich), erythrose (Sigma Aldrich, > 75%), D-sorbitol (Acros Organics, 97%), D-mannitol (TCI Chemicals, 99%), meso-erythritol (Alfa Aesar, 99%), ethylene glycol (Fisher Scientific). 1,2-propylene glycol (Fisher Scientific), 1,2-butanediol (TCI Chemical, 98%). hydroxy acetone (Alfa Aesar, 95%), l-hydroxy-2-butanone (Ambeed, 96%), sodium metatungstate monohydrate (Alfa Aesar, > 87% WO3), ammonium metatungstate hydrate (Sigma Aldrich, > 85% WO3), ammonium molybdate hydrate (Alfa Aesar, 99.999% metal basis), tungsten (VI) oxide (Alfa Aesar. 99.998% metal basis), and molybdenum (VI) oxide (Alfa Aesar, 99.5%) were purchased. Di. Water was obtained from an in-house Bamstead Mega-Pure Still MP-1 water purification system with conductivity71 pS / cm (1 MQ cm). CHN content for glucose, starch, and cellulose was measured using Elementar CHN analyzer (Std. Error < ±0.5%).

[0137] Synthesis of Ni(afMFI:

[0138] The parent zeolite was first calcined at 550 °C for 8 h in 100 ml / min air and subsequently impregnated with 2.5 ml / g zeolite Ni(NO3)2-6H2O solution and dried at 50 °C for 12 h. This material was reduced at 350 °C for 2 h in 100 ml / min 10% H2:90% Ar. The reduced material was denoted as Ni / MFI. Selective Ni leaching was performed in either 50 ml centrifuge tubes (for Ni / MFI batches < 0.5 g) or in 500 ml PP bottles (for batches 1 g < Ni / MFI < 4g). The Ni / MFI catalyst, 2 g / g catalyst trimesic acid (0. 16 M), 60 ml di H2O / g catalyst, and 0.4 ml / g catalyst H2O2 solution was added to the vessel, sealed, and heated at 85 °C for 16 h. This was repeated twice to make the leached Ni@MFI. The resulting Ni@MFI was collected by centrifugation and washed first with ethanol and subsequently with water before being dried at 80 °C.

[0139] Synthesis of Ni / SiO2:

[0140] Fumed silica was impregnated with a solution Ni(NO3)2- 6H2O, and dried at 80 °C overnight. The dried powder was reduced at 350 °C for 2 h in 100 ml / min 10% H2:90% Ar. Catalyst Characterization:

[0141] X-ray diffractograms were collected on Bruker D8 XRD. N2 Physisorption was performed with a Micromeritics 3Flex Surface Characterization instrument. Micropore volume was calculated using t-plot method. Temperature programmed reduction (TPR) was collected on Micromeritics AutoChem II. Typically, 100 - 200 mg of catalyst was packed into a flow-thru U-tube and pre-treated before analysis in 30 ml / min Ar at 300 °C for 1 h. TPR was performed with 30 ml / min 10% H2:Ar and 5 or 10 °C / min ramp rate. Optionally, 10% O2:He or Ar can also be used to treat the catalysts. Ni content for catalysts was measured with a Rigaku SuperMini 200 WDXRF.

[0142] Reactions:

[0143] All hydrogenation reactions were performed in a Parr reactor equipped with an internal thermocouple, Teflon liner, and stir bar. In general, the reactor was charged with 5 - 15 ml of aqueous solution of glycolaldehyde, glucose, starch, or cellulose, the retro-aldol catalyst, and / or the hydrogenation catalyst. The reactor was then sealed, purged 5 times with H2, and pressurized with H2 to 30 barg at room temperature (20 °C). The reactor was then heated up to reactor temperature. Heat-up time was 10 min. Upon completion, the reactor was cooled to room temperature, vented, and 10 pL / ml-reaction of a sucrose solution (0.3 M) was added as an external standard. The reaction solution was then filtered and analyzed by an Agilent 1260 Infinity II HPLC equipped with Hi-Plex Ca column (7.7 x 300 mm, 8 pm), and RID and operated with 0.6 ml / min di. H2O as mobile phase. All analytes were identified and quantified by HPLC (calibrations and quantification statistics in Tables 1 and 2). Byproducts contributing to the C-gap in reactions were identified using Agilent 7890B GCMS equipped with an HP-Innowax (30 m, 0.25 pm film) column. The experimental error for conversion and yields for all analytes was ± 2%.

[0144] Table 1: Mean, standard deviation, and standard error in the mean for an example glucose reaction. curves for all analytes. I - Sucrose (external standard).

[0145] Results and Discussion

[0146] Encapsulation process and catalyst characterization:

[0147] The general procedure for the synthesis of zeolite encapsulated Ni catalyst (Ni@MFI) is shown in Scheme 1. Acid dissolution of metallic Ni and NiO particles has been extensively studied, primarily the use of strong mineral acids, but also known to occur with weak organic acids. By leveraging the size-exclusion property of zeolites, a bulky acid can be used to selectively dissolve Ni particles from the external surface of a 4% Ni / MFI material. Acetic acid was used as the “small'’ acid while citric and trimesic acids were tested as bulky acids to be excluded by the zeolite.

[0148] Trimesic acid was also considered since it is structurally akin to 1,3,5-triisoporpyl benzene (kinetic diameter = 0.84 nm) which has been shown to experience diffusion limitations over MFI zeolites (ZSM-5). The loss of Ni from the parent Ni / MFI due to different acids can be seen in FIG. 1. A sharp decrease in the Ni content can be observed after the first leaching. For successive leaching, the extent of Ni dissolution seems to be significantly lowered. After 3 leaching runs, the Ni content for citric and trimesic acid leached materials remains largely unchanged w hereas that of the acetic acid leached material exhibits an observable loss in Ni content. This can be further corroborated by the glucose and glycolaldehyde hydrogenation activity as seen in FIGS. 2 and 3. All materials show low activity for glucose reduction to sorbitol since the external Ni particles should be dissolved by all acids. However, the trend in glycolaldehyde reduction is acetic acid<citric acid < trimesic acid. This is expected as acetic acid should not be excluded from the zeolite and be able to dissolve Ni from the internal pore network resulting in fewer active sites for the reaction. The sudden decrease in the rate of Ni dissolution, particularly for acetic acid, was suspected to be due to a limitation of Ch transport. To test this H2O2 w as added to the 4th leaching run but no change in the extent of leaching was observed. Increasing the amount of acid to 0.6 M did not drastically change the extent of Ni leaching either (5th leaching). However, a slight change was observed in the glycolaldehyde reduction activity with the citric acid leached material exhibiting lower activity than the trimesic acid leached material (FIGS 2, 3). Ultimately since trimesic acid showed the highest glycolaldehyde reduction activity, it was used as the standard acid for preparation of Ni@MFI. Acid leaching of Ni / SiCh materials exhibit total dissolution of Ni with no observable hydrogenation activity.

[0149] To ensure a standard protocol, all subsequent tests were performed with 4% Ni / MFI leached with trimesic acid and H2O2. The typical loadings and textural properties for all catalysts are summarized in Table 3. . . Micropore volume BET Surface Area Ni Content

[0150] Material .3, .2, ,

[0151] (cm / g) (m / g) (mass %)

[0152] MFI 0 107 313

[0153] Ni / MFI 0.109 319 4.0

[0154] Ni@MFI 0.111 327 1.1

[0155] Ni / SiO2- - 5.7

[0156] Table 3: Textural properties and Ni loadings for all catalysts

[0157] Ex-situ characterization of the leached material shows no major difference from the parent zeolite. XRD shows characteristic MFI reflections for all materials and no distinguishable loss in crystallinity after leaching (FIG. 4). N2 physisorption (FIG. 4) shows the micro and mesoporosity of the parent material is retained with no major change in the micropore volume (Table 1). This shows that the zeolite structure is retained and is not degraded by the leaching treatments. TPR of impregnated Ni / SiCh shows a reduction peak at 366 °C which is consistent with the reduction of bulk NiO to Ni particles (FIG. 5). Based on this the reduction of the impregnated material was performed at 350 °C for 2 h. After bulk reduction to make Ni / MFI, the material is cooled to room temperature and stored open to air which results in surface and sub-surface oxidation of the Ni particles. TPR of the reduced 4% Ni / MFI and 5.7% Ni / SiCh shows 2 distinct peaks (FIG. 6) - 1. A low temperature peak (< 150 °C) which is commonly seen for surface oxides on metallic Ni particles; 2 - A broad high temperature peak (> 400 °C) which is attributed to “strongly interacting” Ni particles or small NiO particles. For the leached Ni@MFI material, there is no low temperature peak, but the high temperature feature was still observed (FIG. 6). Following TPR, the sample was cooled to room temperature under Ar and reoxidized at 30 °C in 10% Ch:He stream for 1 h. The sample was then purged and another TPR was acquired (TPR2; FIG. 7). This TPR exhibits similar characteristics with a low temperature peak at 100 °C for Ni / SiCh and Ni / MFI, and 2 additional peaks at 180 °C and 240 °C for Ni / MFI and Ni@MFI. However. Ni@MFI reduced at 600 °C prior to reaction shows total loss in size selectivity7. This could be because reduction at 600 °C results in sintering of the encapsulated particles and manifests as a loss in size selectivity (FIG. 8). In contrast, TPR of spent Ni@MFI catalyst, after reaction at 30 barg H2 and 170 °C, exhibits drastically different peaks in the same range as the fresh Ni@MFI but with different relative intensities (TPR 3, FIG. 9). Based on these observations, it is suspected that some of the encapsulated Ni in Ni@MFI is reduced under reaction conditions and is responsible for the size selectivity observed for Glu and GA hydrogenation.

[0158] The size-selectivity of leached Ni@MFI is evident in glucose and glycolaldehyde reduction (FIG. 9) which shows high ethylene glycol yield and low sorbitol yields with increasing conversion. In contrast, Ni / SiCh and Ni / MFI both show significant glucose reduction. At elevated temperature, glucose hydrogenation is still observed over the Ni@MFI material. The degree of Ni encapsulation was measured by comparison of rate for glycolaldehyde and glucose hydrogenation at 140 °C and calculated to be 0.93 (FIG. 10). The high DOE for Ni@MFI indicates that the size-selectivity of the material is from the exclusion of larger substrates by the zeolite framework.

[0159] Retro-aldol reaction:

[0160] The ability of Mo- and W-based oxides to make strong complexes with sugars and catalyze epimerization reactions by rearrangement of the C backbone has been well documented. The transition state for epimerization and retro-aldol is expected to be similar and involves the Ca-C bond scission of the sugar backbone (FIG. 11). The epimerization and retro-aldol activity of Mo and W polyoxometalate clusters is seen in FIGS. 12 and 13. Mo based catalysts exhibit a higher activity for epimerization and can rapidly establish the epimer equilibrium when compared to W. Interestingly, the yield of glycolaldehyde is similar for both W and Mo catalysts. The soluble tungstate salts exhibit higher activity than WO3 (FIG. 13), which can be attributed to the low solubility of WO3 under reaction conditions. This is consistent with literature reports for W catalyzed glucose retro-aldol reactions. Additionally, the activity7of the tungstate salts is independent of the corresponding cations (Na+ vs NH4+; FIG. 13).

[0161] Examples of Mo retro-aldol reactions of monosaccharides have been reported for the conversion of glucose to glycolic acid and fructose to alkyl lactates. In contrast, glucose or cellulose conversion to ethylene glycol with W is only seen at higher temperatures and Mo materials have shown minimal activity in this system even at higher temperatures. Additionally due to the reactivity of sugars under these conditions the sugars are also irreversibly consumed by side reactions which convolutes measurement of the equilibrium at high temperatures (FIG. 13). When MoOs or soluble molybdates are first pre-treated with Ni / SiCh under reaction conditions in the absence of the sugar, the epimerization and retro-aldol activity is completely shut down (FIG. 13). This can be explained by the deactivation of Mo by reduction in the presence of a Ni catalyst. However, W does not exhibit the same behavior and instead retains retro-aldol activity (FIG. 13). For this reason, W based materials were chosen as the retro-aldol catalyst. To simplify the system and better study the impact of the TWO catalysts in the tandem system, SMT was used as the retro- aldol catalyst for tandem studies.

[0162] Tandem reaction - Glucose: Scheme 2: Reaction network for the conversion of glucose to polyols

[0163] Scheme 2 shows the reaction network for the direct conversion of glucose to ethylene glycol. Glucose can be converted to glycolaldehyde and ery throse by a retro-aldol catalyst. Erythrose can undergo further retro-aldol reaction to produce 2 equivalents of glycolaldehyde. The glycolaldehyde can then be hydrogenated to ethylene glycol over aNi catalyst. The retro-aldol reaction is equilibrium limited and strongly favors glucose over glycolaldehyde. A one-pot tandem strategy comprising of a retro-aldol and size-selective hydrogenation catalyst can leverage the hydrogenation of glycolaldehyde to overcome the retro-aldol equilibrium limitation. In addition, glucose isomerization to fructose can also be observed. The fructose retro-aldol and hydrogenation pathway lead to the formation of C3 polyols like propylene glycol. The C4 sugars also undergo dehydration and hydrogenation reactions to make 1,2-butanediol. For clarity, the products of other sugar reactions such as epimenzation and isomerization are omitted from the scheme but mannose, and threose are also observed. In general, sugar solutions can rapidly equilibrate to a complex mixture of isomers and epimers in the presence of the necessary catalysts. This mixture can subsequently yield a wide range of Ci -C6 degradation products including furfurals, carboxylic acids, and alcohols (FIGS. 14A-14E). Hydrogenation provides a pathway to thermodynamically stable polyols enabling efficient conversion of sugars to desired products over other degradation products. The product distribution is thus sensitive to the hydrogenation rate and can be described by 2 scenarios:

[0164] 1) Hydrogenation is rate-limiting - Under this condition, sugar intermediates can equilibrate resulting in poor selectivity due to competing degradation reactions.

[0165] 2) Hydrogenation rate is comparable to retro-aldol rate - Under this condition, high selectivity to polyols can be achieved due to rapid hydrogenation of the sugar intermediates. By preventing the accumulation of these intermediates, the selectivity to degradation products is minimized.

[0166] FIG. 15 shows the tandem reaction with glucose using SMT as the retro-aldol catalyst and 3 different hydrogenation catalysts - Ni / SiCh, Ni / MFI, and Ni@MFI. High loadings of these catalysts are used to ensure fast hydrogenation and high selectivity to the polyols. The catalysts with unencapsulated Ni yield significant quantities of hexitols while Ni@MFI yields primarily ethylene glycol. The lower glucose conversion and absence of hexitols at lower temperature (140 °C) over Ni@MFI is consistent with the exclusion of glucose by the MFI framework. The size-selectivity of Ni@MFI is retained at higher temperatures (170 °C) and conversions. The retro-aldol reaction is endothermic and increasing reaction temperature favors higher yields of glycolaldehyde. This can be seen clearly with the higher yield of EG with the unencapsulated Ni catalysts at 170 °C. The clear distinction in the selectivity between Ni / SiCh or Ni / MFI, and Ni@MFI provides strong evidence for the need for size-selectivity in the tandem system.

[0167] The need for fast hydrogenation is evident in FIG. 16 which shows the effect of Ni@MFI and Ni / SiCh loading on the tandem system. By increasing the amount of Ni@MFI, we can increase the net rate of hydrogenation for a fixed retro-aldol rate. At low loadings, less than 50% total selectivity to polyols is observed due to the formation of byproducts via degradation reactions. As the catalyst amount is increased, the selectivity towards ethylene glycol increases reaching a plateau around 65%. At around 40 mg Ni@MFI / g reaction, the hydrogenation rate is the dominant pathway for C-flux with a majority going towards ethylene glycol and a small fraction going towards propylene glycol and 1,2-butanediol. Increasing beyond this does not improve the yield of ethylene glycol since the selectivity is now decided by the rate of glycolaldehyde hydrogenation vs C3 and C4 dehydration / hydrogenation reactions. Since both these reactions involve hydrogenation, the selectivity is no longer affected by changes in hydrogenation rate.

[0168] The balance in retro-aldol and hydrogenation rates can be achieved by maintaining the catalyst ratio at an optimal value. A clear trend is observed for ethylene glycol yield as a function of Ni:W ratio. For Ni:W < 0.2, ethylene glycol yield decreases with Ni:W ratio. This is due to slow hydrogenation that results in a larger selectivity towards degradation products. At Ni:W > 0.2, hydrogenation is the dominant pathway for sugars which consequently leads to ethylene glycol yields around 65%. The ethylene glycol yield is agnostic to Ni:W ratio above the threshold value which is expected since the reaction network is no longer rate limited by hydrogenation and so independent of the amount of hydrogenation catalyst. In contrast, with unencapsulated Ni / SiCh, a single optimal value corresponding to a local maximum in EG yield is observed. A similar trend is observed with previously reported Ni-W catalyst for cellulose conversion. Deviating from this Ni:W ratio in either direction results in a loss in selectivity (FIGS. 17A-17B) due to competing glucose hydrogenation (at Ni:W > 0.2) or C3 and C4 dehydration / hydrogenation (at Ni:W < 0.04). The threshold Ni:W ratio is different for different SMT loading as seen in FIG. 18 for 10 mg SMT / g reaction vs 3 mg SMT / g reaction. This is because the primary- criteria to ensure high selectivity- to ethylene glycol is maintaining fast hydrogenation to rapidly consume the glycolaldehyde produced by the retro-aldol reaction to minimize the degradation of accumulated glycolaldehyde and erythrose (FIGS. 14A-14E). Operating at a fixed Ni@MFI loading at increasing SMT loadings results in a loss in EG selectivity7(FIG. 18). Thus, when retro-aldol rates are increased by increasing the SMT loading, the hydrogenation rate must also be increased resulting in a different threshold Ni:W ratio. However, as long as the Ni:W ratio at any SMT loading is above the threshold value, the EG yield is invariant to changes in catalyst loading and results in stable performance. Changing the SMT loading for the unencapsulated Ni / SiO2 catalyst results in a narrower optimal window with steeper decline due to over hydrogenation as seen in FIG. 19.

[0169] The impact of reaction temperature can be seen in FIG. 20. While ethylene glycol is produced at 140 °C, this is accompanied by mannose, and fructose which arise from the epimerization and isomerization of the sugar. The hydrogenation rate can be increased byincreasing temperature which results in the highest yield of 63% at 170 °C. At temperatures greater than 170 °C more propylene glycol and 1,2-butanediol are produced resulting in lower ethylene glycol selectivity7. Under a fixed set of reaction conditions and catalyst loadings, lower glucose concentrations result in higher ethylene glycol yields (FIG. 21 A). At higher glucose concentrations, a large C-gap is observed due to the formation of degradation products which implies that the rate of retro-aldol and hydrogenation are not sufficient to compete with that of homogenous side reactions. Operating the tandem system for 29 mM glucose concentration with the same Glu:W and W:Ni ratios as that of the optimal conditions for 12 mM glucose offers slight improvement to the overall ethylene glycol yield (47% from 39%) but is still worse than the performance at lower glucose concentrations (FIG. 2 IB). This emphasizes the impact of low glucose concentrations on minimizing the extent homogenous side reactions and improving the C-flux towards ethylene glycol. Additionally, the Ni@MFI catalyst is stable under reaction conditions and can be reused up to 3 times without loss in activity (FIG. 22).

[0170] Tandem reaction - Starch and Cellulose:

[0171] The optimal conditions identified for the tandem reaction with glucose were then applied to polysaccharides such as starch and cellulose. In general, for polysaccharide substrates, hydrolysis is rate limiting which often necessitates the use of higher temperatures and longer reaction times than glucose systems. The difference in reaction rates is evident in FIG. 23 which shows the slower production of EG from starch when compared to glucose under the same conditions. By extending the reaction time to 16 h we see ethylene glycol yields comparable to when glucose is used as the substrate. The polysaccharide serves as a stable reservoir of glucose resulting in a low glucose concentration at any given time. This is analogous to a semi-batch system and is expected to improve the selectivity' towards ethylene glycol. Hydrolysis is acid-catalyzed, and the reaction rate can be increased significantly by decreasing the solution pH as shown in FIG. 24. Consequently. 58% ethylene glycol yield is achieved at pH 2.8 after 4 h at 170 °C. By increasing the reaction temperature, the ethylene glycol yield can be further increased to 76%. In general, acidic conditions have faster rates of hydrolysis and ethylene glycol production, and this effect is more clearly observed at lower temperatures which are also expected to be at lower conversions. The Ni@MFI catalyst maintains a high selectivity to ethylene glycol with minimal sorbitol production as observed for glucose which clearly7demonstrates the benefit of size selectivity in this system. It should be noted that the solution pH after reaction is in the range of 4.3 - 5.0, regardless of the ini tai pH. The disproportionation of sodium metatungstate into a broad distribution of multinuclear tungsten species has been reported previously and could act as a buffer resulting in this change in solution pH after reaction. A particular concern is the dissolution of Ni in acidic solutions under reaction conditions. While some dissolution is expected, no major change in the product distribution is observed after 3 catalyst recycles (FIG. 22). The observed stability could be due to the net buffering of the solution pH by SMT and the reducing reaction conditions (30 bar H2) that can minimize Ni dissolution. The structure of the polysaccharide can influence the rate of hydrolysis with starch being easier to hydrolyze than cellulose. The highly ordered structure of cellulose chains results in strong inter-chain hydrogen bonds forming crystalline domains evident in the XRD of microcrystalline cellulose (MCC) (FIG. 25). Consequently, more forcing conditions are necessary to overcome the stabilizing interactions of cellulose and facilitate hydrolysis. This is particularly evident in the slower rate of ethylene glycol production in the tandem system (FIG. 26A). Amorphization of MCC by ball milling can partially disrupt the strong interactions and facilitate faster hydrolysis of cellulose chains. The amorphization can be tracked by the loss in XRD intensity (FIG. 25) and ball milled cellulose (BMC) shows significantly faster rates for ethylene glycol production (FIG. 26B). Like starch, the addition of acid and higher reaction temperatures can improve the performance of the tandem system resulting in ethylene glycol yields of 50%. Attempting to optimize the ethylene glycol yield through changes in reaction parameters (temperature, time, and pH) was unsuccessful and ethylene glycol yield did not exceed 55%. Ni@MFI still exhibits lower sorbitol production than the unencapsulated Ni / SiChfor cellulose, but the lower ethylene glycol yield requires further investigation to optimize the reaction conditions. Alternatively, the development of pre-processing methods for raw cellulosic biomass can be beneficial by providing more easily processable polysaccharide structures or monosaccharides as the feedstock for reaction.

[0172] Conclusion

[0173] In this work, we have developed a simple, post-synthetic method for preparing zeolite encapsulated Ni particles and applied these materials for direct conversion of biomass to ethylene glycol. By preventing glucose access to active Ni hydrogenation catalyst, the selectivity to sorbitol and mannitol is significantly reduced. This enables high ethylene glycol yields from glucose at significantly lower temperature in batch operation than previous reports. Optimizing reaction conditions, we were able to achieve 63% ethylene glycol yield with glucose and 75% ethylene glycol yield with starch. The size- selective hydrogenation catalyst enables significant changes to the optimal reaction window, particularly for glucose, by lowering reaction temperature, decoupling retro-aldol and hydrogenation activity' from the catalyst composition, and expanding the range of optimal catalyst loading. Example 2

[0174] Materials and Methods

[0175] Supported nickel catalysts were synthesized by a two-step process using commercially available fumed silica or high-silica ZSM-5 (SiO2:A12O3 = 1000) and Ni(NO3)2.6H2O. First the calcined support was impregnated with a solution of Ni(NO3)2 in water, dried at 50 °C for 12 h, and reduced in a tube furnace at 350 °C for 2 h under a 100 ml / min 10:90 H2:Ar stream, yielding Ni / SiO2 or Ni / ZSM-5. The encapsulated Ni@ZSM-5 was prepared from Ni / ZSM-5 using a proprietary post-synthetic method. Sodium metatungstate (SMT) was used as the retro-aldol catalyst. Catalytic tests were performed in a high-pressure Parr reactor that was loaded with a solution of glucose or glycolaldehyde, the retro-aldol and hydrogenation catalysts, and pressurized with hydrogen before being heated to the reaction temperature. Quantitative analysis of the reaction solutions was performed using Agilent 1260 Infinity II HPLC with a 300 mm Hi-Plex Ca column and a refractive index detector (RID).

[0176] Results and Discussion

[0177] The Ni / ZSM-5 prepared by impregnation and reduction contains metallic Ni nanoparticles located both on the external surface of the zeolite crystals and encapsulated within the pore network. We have developed a simple post-synthetic treatment of Ni / ZSM- 5 that selectively dissolves Ni nanoparticles on the external surface of the zeolite support, resulting in a material that is more size-selective. Indeed, this behavior is observed over Ni@ZSM-5 both in independent tests of glucose and glycolaldehyde hydrogenation, and in tandem reactions for the direct conversion of glucose to EG. FIG. 27 shows the difference in the product distribution of the tandem system with SMT paired with Ni / SiO2, Ni / ZSM-5 or Ni@ZSM-5. The catalysts with unencapsulated Ni yield significant quantities of hexitols while Ni@ZSM-5 yields primarily EG. The lower glucose conversion and absence of hexitols at lower temperature (140 °C) over Ni@ZSM-5 is consistent with the exclusion of glucose by the MFI framework (Figure la). The size selectivity7is retained by Ni@ZSM-5 at higher temperature (170 °C) and conversion (Figure lb).

[0178] Example 3 Materials:

[0179] Nickel (II) nitrate hexahydrate (Alfa Aesar. 19.8% Ni min), trimesic acid (Fisher Scientific, 98%), citric acid (Alfa Aesar, 99+%), acetic acid (Fisher Scientific, glacial), sulfuric acid (Fisher Scientific, concentrated), H2O2 solution (Fisher Scientific, 30%), fumed silica (Cab-O-Sil M5), high silica MFI zeolite (Zeolyst, ZD97037, Lot #: 1822-15), D-glucose (Fisher Scientific, 38.97% C), D-fructose (Fisher Scientific), D-mannose (Alfa Aesar, 99%), glycolaldehyde dimer (Sigma Aldrich), erythrose (Sigma Aldrich, > 75%), D- sorbitol (Acros Organics, 97%), D-mannitol (TCI Chemicals, 99%), mesoerythritol (Alfa Aesar, 99%), ethylene glycol (Fisher Scientific), 1 ,2-propylene glycol (Fisher Scientific), 1,2-butanediol (TCI Chemical, 98%), hydroxy acetone (Alfa Aesar, 95%), l-hydroxy-2- butanone (Ambeed. 96%), tetraethylorthosilicate (TEOS, Sigma Aldrich, > 99% ), tetra-n- propyl ammonium hydroxide (TPAOH, Sigma Aldrich, 40% in H2O) were purchased and used as is. Di. water was obtained from an in-house Bamstead Mega-Pure Still MP-1 water purification system with conductivity 1 pS / cm (1 MQ cm). CEIN content for glucose was measured using Elementar CEIN analyzer (Std. Error < ±0.5%).

[0180] Synthesis of Nii@,MFI:

[0181] The commercial zeolite, denoted MFI, was first calcined at 550 °C for 8 h in 100 ml / min air and subsequently impregnated with 2.5 ml / g zeolite Ni(NCh)2-6H2O solution and dried at 50 °C for 12 h. This material was reduced at 350 °C for 2 h in 100 ml / min 10% H2:90% Ar. The reduced material was denoted as Ni / MFI. Selective Ni leaching was performed in either 50 ml centrifuge tubes (for Ni / MFI batches < 0.5 g) or in 500 ml PP bottles (for batches 1 g < Ni / MFI < 4g). The Ni / MFI catalyst, 2 g / g catalyst trimesic acid (0.16 M), 60 ml di H2O / g catalyst, and 0.4 ml / g catalyst 30 wt.% H2O2 solution was added to the vessel, sealed, and heated at 85 °C for 16 h. This was repeated twice to make the leached Ni@MFI. The resulting Ni@MFI was collected by centrifugation and washed first with ethanol and subsequently with water before being dried at 80 °C.

[0182] Synthesis of Silicalite-1:

[0183] Silicalite- 1 (Sil-1) was prepared following the procedure reported in Li, S.; Tuel, A.;

[0184] Laprune, D.; Meunier, F.; Farrusseng, D. Chemistry of Materials 2015, 27, 276-282. In a typical synthesis, 28.33 g TEOS was taken in a PFA beaker. A I M solution of TPAOH solution was prepared and added dropwise in 10 ml portions to TEOS with stirring. The mixture was stirred for 10 mins after each addition. After the final addition the mixture was stirred for 1 h to ensure a homogenous solution. The mixture was left to stir open to the atmosphere for several hours, over which 31.28 g di. water was added in intervals, till all the TEOS was hydrolyzed and the expected ethanol mass loss was measured. The final gel composition was 1 SiO2: 0.4 TPAOH: 35 H2O. Typically, the gel was allowed to dry to a mass approximately 1-3 g less than the expected mass for the gel to ensure total ethanol evaporation. Di. water was then added to bring the gel to the expected mass. The gel was transferred to PTFE liners, assembled in synthesis autoclaves, and crystallized at 170 °C for 5 days. The Sil-1 was collected by centrifugation and washed with di. water till the pH of the rinsate was neutral. The recovered solids were dried at 80 °C overnight and calcined at 550 °C for 8 h to remove the SDA. The yield after 5 days was -90%.

[0185] Synthesis ofNi@Sil-l:

[0186] The impregnation and leaching for Ni@Sil-l was performed similar to Ni@MFI. All leaching for Ni / Sil-1 materials were performed at 0.5 g scale in centrifuge tubes. For developing the synthetic protocol, reduction temperature, acid leaching conditions were varied as discussed in next section.

[0187] Synthesis of Ni / SiQ2:

[0188] Fumed silica was impregnated with a solution Ni(NO3)2-6H2O, and dried at 80 °C overnight. The dried powder was reduced at 350 °C for 2 h in 100 ml / min 10% H2:90% Ar.

[0189] Catalyst Characterization:

[0190] X-ray diflfractograms were collected on Bruker D8 XRD. The 20 range was 5° - 50° with a step size of 0.02° and dwell time of 0.5 s. N2 Physisorption was performed with a Micromeritics 3Flex Surface Characterization instrument. In a ty pical experiment, 0. 1 - 0.2 g of the sample was loaded into a sample tube and degassed at 250 °C under vacuum overnight. The degassed sample was then cooled to room temperature and backfilled with N2 (Grade 5.0). N2 adsorption isotherms were obtained at liquid N2 temperatures with 10s equilibration time for each point. Micropore volume was calculated using t-plot method. BET surface area was calculated using a linear region for 0.01< P / PO < 0.05. BET surface area values were obtained from the linear region of the BET transform as shown in FIG. 29. The linear region was adjusted from the typical region of 0.05 < P / PO < 0.3 to 0.005 < P / PO < 0.05 to satisfy the conditions for reliable BET measurements. The surface areas, C- values, R2 for each fit are shown in FIG. 28 for different P / PO ranges. Additional details regarding the analysis are provided in FIGS. 28-30.

[0191] Temperature programmed reduction (TPR) data was collected on Micromeritics AutoChem II. Typically, 100 - 200 mg of catalyst was packed into a flow-thru U-tube and pre-treated in 30 ml / min Ar at 300 °C for 1 h. TPR was performed with 30 ml / min 10% H2:Ar and 5 or 10 °C / min ramp rate. Optionally. 10% O2:He was used to treat the catalysts to simulate reoxidation under ambient conditions. Calibrations and calculations for TPR results are provided in FIG. 31. Ni content for catalysts was measured with a Rigaku SuperMini 200 WDXRF. Typically, 1.5 - 2 g of catalyst pow der was placed in a 30 mm sample cup with a polypropylene film. XRF spectra were measured based on an empirical quantitative method built with Ni / SiO2 standards. Standards were run before each set of samples to adjust for X-ray drift across different days. Calibrations for quantification are provided in FIG. 32.

[0192] Reactions:

[0193] All hydrogenation reactions w ere performed in a Parr reactor equipped with an internal thermocouple, Teflon liner, and stir bar. In general, the reactor was charged with 5 - 15 ml of aqueous solution of 0. 1 - 1 wt.% glycolaldehyde or glucose, and the hydrogenation catalyst. The reactor was then sealed, purged 5 times with H2, and pressurized with H2 to 30 barg at room temperature (20 °C). The reactor w as then heated up to reaction temperature. Heat-up time was 10 min. Upon completion, the reactor was cooled to room temperature, vented, and 10 pL / ml-reaction of a sucrose solution (0.3 M) was added as an external standard. The reaction solution was then filtered and analyzed by an Agilent 1260 Infinity II HPLC equipped with Hi-Plex Ca column (7.7 x 300 mm, 8 pm), Refractive Index Detector (RID), and operated with 0.6 ml / min di. H2O as mobile phase. All analytes were identified and quantified by HPLC (calibrations and quantification statistics in FIG. 33). Volatile byproducts contributing to the C-gap in reactions were identified using Agilent 7890B GCMS equipped with an HP-Innowax (30 m, 0.25 pm film) column. The experimental error for conversion and yields for all analytes was ± 2%. Calculations for Degree of Encapsulation (DOE) and effectiveness factor (r|) are shown in FIG. 34 (Equations A.3 - A.7).

[0194] Synthetic Strategy’ for Encapsulation

[0195] Contrary' to previously reported strategies, which focus on constructing zeolite frameworks around metal particles, the inventive method focuses on the removal of Ni from an existing zeolite framework. Since this method does not involve zeolite synthesis, any high silica zeolite material can be used as the starting material. The general procedure for the synthesis of zeolite encapsulated Ni catalyst is shown in FIG. 35. Ni particles are supported onto this material through impregnation and oxidation / reduction. The resulting material has Ni particles located on the external surface as well as inside the pore network. Acid dissolution of metallic Ni and NiO particles has been extensively studied, primarily with strong mineral acids, but is also known to occur with weak organic acids. By leveraging the size-exclusion property of zeolites, a bulky organic acid can be used to selectively dissolve Ni particles from the external surface of the material. The choice of acid is investigated, and the conditions for successful leaching are explored. The initial method development was performed on hydrothermally synthesized silicalite-1. This method was then applied to commercial high silica MFI zeolite from Zeolyst. This was also tested with other small-pore aluminosilicate zeolites such as LTA, and CHA.

[0196] Effect of synthesis parameters

[0197] Preliminary testing for acid dissolution of Ni from Ni / SiCh was performed with acetic acid, benzoic acid, isophthalic acid, and trimesic acid. In all cases, Ni dissolution was possible in water at 80 - 90 °C despite the low aqueous solubility of the benzoic acid derivatives. In organic solvents (acetonitrile and DMSO) capable of dissolving all the acids, only acetic acid showed Ni dissolution. Based on this acid leaching using benzoic acids derivatives in water was attempted on a 4% Ni / Sil-1. Of these candidates, trimesic acid was the most promising showing size-selectivity after repeated leaching runs. Trimesic acid is structurally akin to 1,3,5-triisopropyl benzene (kinetic diameter ~ 0.84 nm) which has been shown to experience diffusion limitations over MFI zeolites (ZSM-5). To further investigate the effect of synthesis conditions, repeated leaching and glucose hydrogenation tests were performed (FIG. 36). Glucose hydrogenation is used as the primary method for characterizing these materials since size-selectivity in hydrogenation is the desired property. Ex-situ techniques cannot easily distinguish betw een unencapsulated and encapsulated particles and cannot be used as the criteria for optimizing synthesis conditions. The changes to glucose hydrogenation activity for leached materials are quantified through a ratio of mass normalized sorbitol yields between the leached and parent materials as shown in Equation 2. 1. A decrease in <D is expected upon leaching with encapsulated materials showing <D ~ 0. Glycolaldehyde hydrogenation for materials with 0 = 0 confirms the presence of encapsulated Ni particles (FIGS. 37, 38). = Yield of Sorbitol per g leached catalyst Yield of Sorbitol per g parent catalyst (2.1)

[0198] The distribution of Ni particles through the zeolite is governed by the impregnation and reduction conditions. Drying conditions after impregnation can have significant impact of resulting particle size sue to hydrolysis of Ni salts and the formation of larger Ni(0H)2 particles. This phenomenon can be qualitatively observed by the change in color of the impregnated zeolite from green to yellow after drying at 80 °C for 12 h. For this reason, the impregnated zeolite was dried at 50 °C for 12 h which still resulted in a green powder. The reduction temperature can impact the resulting Ni distribution as sintering of Ni particles is know n to occur at temperatures greater than 450 °C. Lowering the reduction temperature from 450 °C to 350 °C resulted in faster leaching and, is consistent with more agglomeration of Ni particles on the external surface at higher temperatures resulting in a smaller fraction of encapsulated particles. Regardless of reduction temperature, multiple leaching runs (~ 5) are necessary to create a size-selective material. However, the addition of H2O2 was seen to accelerate the rate of acid leaching and consistently resulted in a size- selective material after 2 treatments (FIG. 36). The addition of small amounts of H2O2 has been shown to accelerate the rate of Ni dissolution using mineral acids. The dissolution of metallic Ni is a function of acid strength and oxidation potential and can be viewed as a coupled redox reaction occurring at a shorted electrochemical cell. Ni oxidation to NiO or Ni(0H)2 is the anodic reaction and O2 reduction is the cathodic reaction. The acidic environment enables the dissolution of NiO or Ni(OH)2. The thermodynamic driving force for this reaction can be estimated through the standard cell potential E0, for this redox pair to be 1.49 V vs S.H.E. All Sil-1 leaching was performed in a 50 ml centrifuge tube as listed in FIG. 39. Consequently, the addition of H2O2 can increase the concentration of strong oxidant in solution and enable faster leaching. The exact mechanism of H2O2 was not investigated, but its clear benefit to synthesis is noted and was used as part of the standard protocol.

[0199] While this method was able to produce Ni@Sil- 1 with the required selectivity, the overall activity w as very low7and not suitable for the tandem conversion of glucose to ethylene glycol (FIG. 38). Coincidentally, the use of high silica MFI zeolite from Zeolyst resulted in higher activity after leaching.

[0200] The commercial MFI material, available in large quantities, allowed for easy scale- up of syntheses to 4 g per vessel, from the previous Sil-1 batches of 0.5 g per vessel (FIG. 39). This enables more consistent production of materials, and elemental characterization of the Ni content using XRF. Successful scale-up was achieved through the use of a 500 ml polypropylene bottle as the synthesis vessel and a rotation oven for stirring / agitation.

[0201] To ensure similar leaching behavior with MFI as Sil-1, a series of acid leaching was performed with different acids without added H2O2. Acetic acid was used as the “small” acid while citric and trimesic acids were tested as bulky acids to be excluded by the MFI material. Citric acid has been used previously for a similar application, but it is unclear whether it is entirety excluded by the zeolite. Polystyrene sulfonic acid has also been used to dissolve Ni from the external surface of zeolites, but the large size of these polymer structure (~ 80 kDa), prevents total dissolution from external sites, particularly for materials with mesoporosity. The loss of Ni from the parent Ni / MFI due to different acids can be seen in FIG. 38. A sharp decrease in the Ni content can be observed after the first leaching. For successive leaching, the extent of Ni dissolution seems to be significantly lowered. After 3 leaching runs, the Ni content for citric and trimesic acid leached materials remains largely unchanged whereas that of the acetic acid leached material exhibits an observable loss in Ni content. This can be further corroborated by the glucose and glycolaldehyde hydrogenation activity as seen in FIG. 40. All materials show low activity for glucose reduction to sorbitol since the external Ni particles should be dissolved by all acids. However, the trend in glycol aldehyde reduction is acetic acid < citric acid < trimesic acid. This is expected as acetic acid should not be excluded from the zeolite and be able to dissolve Ni from the internal pore network resulting in fewer active sites for the reaction. The sudden decrease in the rate of Ni dissolution, particularly for acetic acid, was suspected to be due to a limitation of O2 transport. To test this H2O2 was added to the 4thleaching run but no change in the extent of leaching was observed. Increasing the amount of acid to 0.6 M did not drastically change the extent of Ni leaching based on XRF (5th leaching). However, a clear decrease in the glycolaldehyde hydrogenation was observed for all materials. Notably, citric acid showed the largest decrease in reduction activity, even more than acetic acid. This could be explained by considering the pKa and solubility for each of these acids (refer FIG. 33). Triprotic acids such as citric and trimesic acids have greater buffering capacity’ than acetic acid at the same concentrations. The lower solubility of trimesic acid in water further limits the achievable pH for this system in comparison to citric acid, particularly at high concentration (0.6 M). The loss in hydrogenation activity implies that all of the acids tested are not perfectly excluded and the access to encapsulated Ni is limited by the diffusivity of each acid. Ultimately, since trimesic acid showed the highest glycolaldehyde reduction activity; it was used as the standard acid for preparation of Ni@MFI. Acid leaching of Ni / SiCh exhibits total dissolution ofNi with no observable hydrogenation activity. Based on these results, it can be concluded that excessive leaching will result in dissolution of encapsulated Ni particles and trimesic acid’s superior leaching performance is due to a balance in the solubility, pH, and zeolite pore diffusivity.

[0202] A distinct difference between the leaching of Ni / MFI and Ni / Sil-1 was noted, with Ni / MFI leaching seeming to occur much faster than that for Ni / Sil-1 under analogous conditions (trimesic acid, no H2O2). This is clearly evident when comparing 0 values for different leached materials as shown in FIG. 37 . In all cases leaching with Ni / MFI is faster that Ni / Sil-1 and results in a size-selective material. Ni / MFI materials show higher amounts of sorbitol than Ni / Sil-1 as evidenced by the difference in the <D values. This is likely due to differences in pore-structure and transport properties of the support. The addition of H2O2 which w as crucial for Ni / Sil-1 leaching showed minimal effect in Ni / MFI leaching. The zeolite itself is not expected to change the rate of Ni dissolution. The main difference is that the Ni / MFI leaching was performed at a 2 g scale in a polypropylene bottle with rotation while the Ni / Sil-1 leaching was performed at a 0.5 g scale in a centrifuge tube with magnetic stirring. The 2 parameters that have changed in the scale up are the O2:Ni stoichiometry and the mode of mixing. In the large-scale leaching, the polypropylene bottle is attached to a rotation rack set to ~ 40 rpm, which results in improved contact between the gas and liquid. Operating under similar O2:Ni ratio for Ni / MFI as the Ni / Sil-1 systems still show faster leaching for Ni / MFI. Furthermore, the leaching in the Ni / MFI system seems to be invariant to small changes in the O2:Ni ratio. These results suggest that the dissolution of Ni is limited by O2 transport and can be improved through greater agitation to increase the gas-liquid interfacial area. If this agitation is not possible due to mode of mixing, a liquid oxidant can be added to overcome the O2 limitation. To ensure a standard protocol, all subsequent tests were performed with 4% Ni / MFI leached with trimesic acid and H2O2. under rotation for increased mixing.

[0203] Catalyst Characterization

[0204] Physical Characterization

[0205] The typical loadings and textural properties for all catalysts are summarized in FIG. 41. Ex-situ characterization of the leached material shows no major difference from the parent zeolite. XRD shows characteristic MFI reflections for all materials and no distinguishable loss in crystallinity after leaching (FIG. 42). N2 physisorption (FIG. 43) shows the micro and mesoporosity of the parent material is retained with no major change in the micropore volume (FIG. 41). This shows that the zeolite structure is retained and is not degraded by the leaching treatments. A clear difference between the N2 isotherms for Sil-1 and the commercial MFI is noted. In particular, the MFI material shows hysteresis at P / PO > 0.4 which is characteristic of mesopores. The Sil-1 does not show this behavior which is expected of a purely microporous material. There also exists a well-defined low- pressure hysteresis at 0. 1 < P / P0 < 0.3 for MFI, which is below the N2 stability limit for capillary condensation. This behavior has been reported previously for MFI materials and is hypothesized to be due to a phase transition for N2 adsorbate. It should also be noted that while the Sil-1 material does not show low-pressure hysteresis, there exists a second knee at ~ 0.3 P / PO similar to the MFI material, possibly indicative of a similar phenomenon at a lesser extent.

[0206] Site Characterization by Temperature Programmed Reduction

[0207] TPR of impregnated Ni / SiCh shows a reduction peak at 366 °C which is consistent with the reduction of bulk NiO to Ni particles (FIG. 44). After bulk reduction at 350 °C to make Ni / MFI, the material is cooled to room temperature and stored open to air which results in surface and sub-surface oxidation of the Ni particles. TPR of the reduced 4% Ni / MFI and 5.7% Ni / SiO2 shows 2 distinct peaks (FIG. 45 (A)) - 1. A low' temperature peak (< 150 °C) which is commonly seen for surface oxides on metallic Ni particles; 2 - A broad high temperature peak (> 400 °C) which is attributed to '‘strongly interacting’’ Ni particles or small NiO particles. For the leached Ni@MFI material, there is no low temperature peak, but the high temperature feature was still observed (FIG. 45 (A)). Following TPR. the sample was cooled to room temperature under Ar and reoxidized at 30 °C in 10% O2:He stream for 1 h. The sample was then purged and another TPR was acquired (FIG. 45 (B)). This TPR exhibits similar characteristics with a low' temperature peak at 100 °C for Ni / SiO2 and Ni / MFI, and 2 additional peaks at 180 °C and 240 °C for Ni / MFI and Ni@MFI. However, Ni@MFI reduced at 600 °C prior to reaction results in sintering of the encapsulated particles which manifests as a loss in size selectivity (FIG. 46). In contrast, TPR of spent Ni@MFI catalyst, after reaction at 30 barg H and 170 °C, exhibits peaks in the same range as the fresh Ni@MFI but with different relative intensities (FIG. 47). Quantitative analysis of TPR results is provided in FIG. 48. In all cases high temperature reduction of Ni catalysts in 10% H2:Ar results in a decreased H2:Ni ratio indicating less H2 consumption than the untreated material. This behavior is consistent with sintering of Ni particles effectively reducing the particle dispersion and surface area prone to oxidation. In contrast, high pressure reduced Ni@MFI (30 barg H2, 170 °C) shows similar H2:Ni ratio as the untreated material and continues to maintain size-selectivity. Based on these TPR results, it can be concluded that acid leaching of Ni impregnated MFI zeolite can result in encapsulation of NiO particles, which upon reduction under reaction conditions generate the active Ni particles for size-selective hydrogenation.

[0208] Size-Selectivity for Sugar Hydrogenation The size-selectivity of leached Ni@MFI is evident in glucose and glycolaldehyde reduction (FIG. 49) which shows high ethylene glycol yield and low sorbitol yields with increasing conversion. In contrast, Ni / SiCh and Ni / MFI both show significant glucose reduction. At elevated temperature, glucose hydrogenation is still observed over the Ni@MFI material. The size-selectivity' for Ni@MFI can be quantified through a degree of encapsulation calculated through Equation A.6 (FIG. 34). The degree of Ni encapsulation was measured by comparison of rate for glycolaldehyde and glucose hydrogenation at 140 °C and calculated to be 0.88 for Ni@MFI and 0.97 for Ni@Sil-l (FIG. 50). The high DOE for these materials indicates that the size-selectivity of the material is from the exclusion of larger substrates by the zeolite framework. As mentioned previously, the hydrogenation activity of leached Ni@Sil-l and Ni@MFI varies significantly (FIG. 47). Sil-1 shows better size-selectivity, as evidenced by the higher DOE, but the overall activity for glycolaldehyde reduction is much lower. An estimate of the effectiveness factor for glucose and glycolaldehyde hydrogenation over the leached catalysts is provided with respect to the activity' of unencapsulated Ni / SiO2 (FIG. 51). The unencapsulated Ni / SiO2 is assumed to be the true rate in the absence of internal transport limitations since the Ni particles are supported on a non-porous fumed silica. The glucose effectiveness factor for leached Ni@MFI and Ni@Sil-l is significantly smaller than that of the parent material as expected for the exclusion of glucose by the zeolite framework. Comparing the effectiveness factor for Ni / MFI (0.53) and Ni / Sil-1 (0.82) shows that glucose hydrogenation is more restricted on MFI materials over Sil-1 materials. This could be due to the mesoporosity of Ni / MFI materials resulting in diffusional constraints not present for Ni / Sil-1 . Nonetheless, the leaching treatment of both these materials removed Ni from both the external surface and mesopores resulting in a significant reduction in the effectiveness factor.

[0209] Expanding the Scope for Encapsulation

[0210] Galvanic Displacement Deposition for Introducing Precious Metals

[0211] Precious metal like Pt, Pd and Ru are known to be excellent hydrogenation catalysts with higher mass activity than Ni. The encapsulation of these metals could then result in a material with improved hydrogenation rates. To achieve this, a galvanic displacement deposition strategy was devised using Ni^Sil-l as a template (FIG. 52). The Ni particles can act as electrodes in a shorted electrochemical cell undergoing simultaneous Ni oxidation / dissolution and Pt reduction. The reaction is thermodynamically driven by the differences in the reduction potentials of the 2 metals (EONi = -0.257; EOPt = 0.73) and can be described through the Nemst equation A.8 (FIG. 53). The general procedure involves the reduction of Ni under Eh to first create metallic Ni particles. This is then exposed to a Pt2+solution under inert atmosphere at room temperature. The resulting material is then washed and tested for hydrogenation activity. The addition of Pt significantly increases the glycolaldehyde hydrogenation rate as shown in FIG. 54. This method can be adapted for the preparation of bimetallic catalysts on non-conductive supports and extend traditional electrocatalyst synthesis methods to thermal catalysts.

[0212] Encapsulation with other 1st row transition metals and zeolite frameworks

[0213] A defining property7of this synthetic method is the ability7to dissolve metal particles. This is a common property of first row transition metals. Thus, this method could be implemented with other first row transition metals such as Mn, Fe, Co, Cu and Zn. There exists a large body of w ork on the corrosion of these metals which can be used as a basis to identify the best conditions (pH, temperature, additives) for successful encapsulation of these metals in zeolites. Since this method focuses on the removal of metal, it is more amenable to use with other ready-to-use zeolite supports. A main criterion is the stability of the zeolite under the acid leaching conditions. This restricts the use of this method to ”high- silica” zeolites and zeotype with Si: Al > 20-30. When subjected to these sy thesis conditions, traditional LTA (Si:Al ~5) undergoes dealumination and the framework collapses. In contrast ZSM-5 (Si: Al 30) and CHA (Si: Al ~20) retain the zeolite structure. In general, this method is applicable to any acid stable microporous material with the appropriate acid to ensure removal of metal from only the external surface.

[0214] Conclusion

[0215] This example outlines the development of a simple and scalable method for synthesis of zeolite encapsulated Ni catalysts for size-selective hydrogenation applications. This method can be used with a high silica zeolite as the support and does not require hydrothermal conditions. By using a bulky organic acid that exhibits strong transport limitations for the zeolite, Ni particles on the external surface can be selectively removed. Acid leaching of Ni was found to be rate limited by O2 transport, which can be mitigated at larger scales with better agitation to improve gas liquid surface area. Alternatively, H2O2 can be added as a strong liquid oxidant for faster oxidation in systems with inadequate mixing. Physical characterization of synthesized catalyst shows no change in the zeolite structure after leaching treatments. Reduction characteristics of MFI materials, reveals that NiO particles are encapsulated by the zeolite and are reduced in-situ under high H2 pressure to Ni particles exhibiting size-selectivity. The size-selectivity of leached Ni@MFI and Ni@Sil- 1 catalysts was characterized using glucose and glycolaldehyde hydrogenation reactions. These materials exhibit a high degree of encapsulation of 0.88 and 0.97 for Ni@MFI and Ni@Sil- 1 respectively. The size-selectivity is shown to be due to transport limitations for glucose over MFI and Sil-1 materials through comparisons of effectiveness factor.

[0216] Example 4

[0217] Materials:

[0218] D-glucose (Fisher Scientific, 38.97% C), D-fructose (Fisher Scientific), D-mannose (Alfa Aesar. 99%), glycolaldehyde dimer (Sigma Aldrich), erythrose (Sigma Aldrich, > 75%), Nickel (II) nitrate hexahydrate (Alfa Aesar, 19.8% Ni min), fumed silica (Cab-O-Sil M5), sodium metatungstate monohydrate (Alfa Aesar, > 87% WO3), ammonium metatungstate hydrate (Sigma Aldrich, > 85% WO3), ammonium molybdate hydrate (Alfa Aesar. 99.999% metal basis), tungsten (VI) oxide (Alfa Aesar, 99.998% metal basis), and molybdenum (VI) oxide (Alfa Aesar, 99.5%) were purchased and used as is. Di water was obtained from an in-house Bamstead Mega-Pure Still MP-1 water purification system with conductivity 1 pS / cm (1 MQ cm).

[0219] C-shift and Retro- Aldol Reactions:

[0220] C-Shift and retro-aldol reactions under H2 atmospheres were performed with a Parr reactor equipped with an internal thermocouple, Teflon liner, and stir bar. In general, the reactor was charged with 5 - 15 ml of aqueous solution of glycol aldehyde, or glucose, the retro-aldol catalyst, and / or the hydrogenation catalyst. The reactor was then sealed, purged 5 times with H2, and pressurized with H2 to 30 barg at room temperature (20 °C). The reactor was then heated up to reactor temperature. Heatup time was 10 min. Upon completion, the reactor was cooled to room temperature, vented, and 10 pL / ml -reaction of a sucrose solution (0.3 M) was added as an external standard. The reaction solution was then filtered and analyzed by an Agilent 1260 Infinity II HPLC equipped with Hi-Plex Ca column (7.7 x 300 mm, 8 pm), and RID and operated with 0.6 ml / min di. H2O as mobile phase. All analytes were identified and quantified by HPLC (calibrations and quantification statistics in FIGS. 55A and 55B). Byproducts contributing to the C-gap in reactions were identified using Agilent 7890B GCMS equipped with an HP-Innowax (30 m, 0.25 pm film) column. The experimental error for conversion and yields for all analytes was ± 2%. Reactions under N2 atmospheres at pressures < 5 barg was performed in a glass tube reactor (Ace Glass. 8648-23) equipped with a gas addition port, pressure gauge, rupture disk (200 psig), and a pressure relief valve (150 psig). In a typical experiment, an aqueous solution of glycolaldehyde, or glucose, the retro-aldol catalyst, and / or the hydrogenation catalyst was added to the glass reactor, sealed, purged 5 times with N2, and pressurized with N2 to the operating pressure < 5 barg at room temperature. The reactor was then heated up to reactor temperature in a silicon oil bath. Post-reaction analysis was the same as that for reaction in the Parr reactor with sucrose as the standard. Low temperature C-shift reactions under air at temperatures < 100 °C were performed in PP centrifuge tubes following a similar procedure as mentioned above.

[0221] Synthesis of LiNbMoQe:

[0222] LiNbMoOo was synthesized using the procedure reported in Takagaki, A.; Furusato, S.; Kikuchi. R.; Oyama, S. T. ChemSusChem 2015, 8, 3769-3772. In a typical synthesis. Li2CCh, Nb20s, and MoOs were mixed as a 1 : 1 : 1 solid mixture in a crucible. This mixture was heated in 200 ml / min air at 580 °C for 24 h with an intermediate mixing / grinding after 12 h.

[0223] Synthesis of Mo3Nb2Oi4:

[0224] MosNb2Oi4 was prepared by through a procedure adapted from Afanasiev. P. The Journal of Physical Chemistry B 2005. 109, 18293-18300. MoO3 and Nb20s was mixed in a 3:2 ratio as a solid or through an alkaline solution (pH 10). The resulting solid mixture was heated in air at 700 °C for 12 - 72 h. The formation of Mo3Nb2Oi4 was followed by XRD. Both methods resulted in the formation of the green Mo3Nb2Oi4 phase.

[0225] Synthesis of MoNbVTeO (Ml):

[0226] Ml was synthesized following the procedure reported in Annamalai. L.; Liu, Y: Ezenwa, S.; Dang, Y; Suib, S. L.; Deshlahra, P. ACS Catalysis 2018, 8, 7051-7067.

[0227] Synthesis of Zr(MoC>4):

[0228] Zr(MoO4) was prepared following a procedure from Briand, L. E.; Hirt, A. M.; Wachs, I. E. Journal of Catalysis 2001, 202, 268-278. 0.57 g ZrCh (0.5 Zr:Mo) was dissolved in 7.6 g H2O and 0. 76 g citric acid (0.8 CA: Mo) was added. This solution was stirred till a homogenous solution was obtained. 0.86 g AMo was dissolved in 7.6 g water. The AMo solution was added to the Zr solution and dried in a steam batch till a glass texture is observed. This material was dried further in a vacuum oven at 70 °C overnight. The dried material was ground and calcined at 700 °C for 24 h.

[0229] Synthesis of MoOs / TiCh:

[0230] MoCL / TiCh was prepared by impregnation. MoOs loading was targeted based on the surface area of the TiCh support (225 m2 / g). 0.43 g AMo was dissolved in 10 ml H2O and added to 1 g of TiCh support. The mixture was heated at 70 °C till all the water was evaporated. The solid was then dried at 80 °C in an oven and then calcined in air at 500 °C for 4 h.

[0231] Mo vs W Oxides

[0232] Mo based catalysts exhibit a higher mass activity' for epimerization and can rapidly establish the epimer equilibrium when compared to W (FIG. 56). Table 3.2 (FIG.57) lists C-shift and retro-aldol activities for Mo and W catalysts under representative reaction conditions (140 °C, H2). It should be noted that the measured yields of mannose and glycolaldehyde, while assumed to be at equilibrium, may not be representative of the equilibrium at reaction temperature but rather at room temperature after cooling the reactor. MoOs shows similar C-shift activity7but slightly higher retro-aldol activity' when compared to ammonium heptamolybdate ((NH4)6Mo?O24, AMo). At 140 °C, all Mo catalysts approached C-shift equilibrium during the heat-up time of 10 minutes. In contrast, sodium metatungstate (NE [H2W12O40], SMT), shows a slow increase in the yield of C-shift and retro-aldol products (FIG. 56), indicative of slower kinetics. The activity of the tungstate salts is independent of the cation identity as shown in Table 3.2 (FIG. 57). Unlike the analogous Mo catalysts, the soluble metatungstate salts exhibit higher activity than WO3 (FIG. 57). which can be attributed to the low solubility of WO3 under reaction conditions. Increasing reaction temperature and time results in increased degradation of sugars and larger C-gap for both Mo and W catalysts (FIG. 58). This is a combination of uncatalyzed dehydration reactions and molybdate or tungstate-catalyzed disproportionation reactions (FIGS. 59-63).

[0233] Reactions were performed at 170 °C wi th high concentration of glycolaldehyde to identify any products that are made. Control reactions for glycolaldehyde without catalyst show minimal degradation products. However, with SMT catalyst significant C-gap and number of new species are seen. Glycolaldehyde over Ni@MFI hydrogneation catalyst only shows ethylene glycol and 1 -hydroxy -2-butanone arising from hydrogenation of C2 and C4 sugars. Tandem reactions with glycolaldehyde and glucose at low loadings of Ni@MFI, i.e smaller Ni:W, show the largest C-gap. Analysis of these products show species similar to that from glycolaldehyde reaction over SMT. Reactions with glucose do not result in as many volatile products and humin formation is observed. HPLC results show the presence of 5-hydroxymethylfurfural (5-HMF). For all W catalysts, increasing catalyst loading increases the yield of C-shift and retro-aldol products. Nonetheless, Mo catalysts outperform the corresponding W counterparts under all conditions without hydrogenation catalyst. Despite this clear difference in performance, and applications in multiple biomass reaction systems, Mo-based retro-aldol catalysts have not been used in hydrogenation systems.

[0234] The application of a tandem retro-aldol hydrogenation system requires that all components be stable in the reaction environment without detrimental interactions with other catalysts or substrates. When MoOs or soluble molybdates are first pre-treated with Ni / SiCh under reaction conditions in the absence of the sugar, the epimerization and retro- aldol activity is completely shut down (FIG. 64). This effect is only observed with metallic Ni particles under N2 or H2 atmospheres, which implies the deactivation of Mo by reduction in the presence of aNi catalyst. W oxides do not exhibit the same behavior and retain retro- aldol activity even at higher temperatures (FIG. 64). For this reason, W based materials were chosen as the retro-aldol catalyst. To simplify the system and better study the impact of the 2 catalysts in the tandem system, SMT was used as the retro-aldol catalyst for tandem studies.

[0235] Conclusion

[0236] Based on the well-studied C-shift reaction, Mo and W oxides were chosen as potential retro-aldol catalysts. While Mo oxides are more active than W oxides, the incompatibility with hydrogenation conditions makes them unsuitable for biomass conversion to ethylene glycol. For this reason, W oxides, particularly soluble SMT, w as chosen as the retro-aldol catalysts.

[0237] Example 5

[0238] Materials:

[0239] Sulfuric acid (Fisher Scientific, concentrated), D-glucose (Fisher Scientific, 38.97% C), starch (Fisher scientific, 38.25% C), microcrystalline cellulose (MCC, Alfa Aesar, 40.21% C), D-fructose (Fisher Scientific), D-mannose (AlfaAesar, 99%), glycolaldehyde dimer (Sigma Aldrich), erythrose (Sigma Aldrich, > 75%). D-sorbitol (Acros Organics, 97%), D-mannitol (TCI Chemicals, 99%), meso-erythritol (Alfa Aesar, 99%), ethylene glycol (Fisher Scientific), 1,2-propylene glycol (Fisher Scientific), 1,2-butanediol (TCI Chemical, 98%), hydroxy acetone (Alfa Aesar, 95%), 1 -hydroxy -2-butanone (Ambeed, 96%), sodium metatungstate monohydrate (AlfaAesar, 87% WO3) were purchased and used as is. Ni / SiO2, Ni / MFI and Ni@MFI catalysts were prepared as described in chapter 2. Di. Water was obtained from an in-house Bamstead Mega-Pure Still MP-1 water purification system with conductivity 1 pS / cm (1 MQ cm). CHN content for glucose was measured using Elementar CHN analyzer (Std. Error < ±0.5%).

[0240] Ball-milling: Ball-milling of microcrystalline cellulose was performed with [instrument], using zirconia beads (size and number). Milling was performed in 5-minute intervals and the amorphization was followed with XRD. Final material after 120 minutes was denoted BMC and used as is for reactions.

[0241] Reactions:

[0242] All reactions were performed in a Parr reactor equipped with an internal thermocouple, Teflon liner, and stir bar. In general, the reactor was charged with 5 - 15 ml of aqueous solution of glycolaldehyde, glucose, starch, or cellulose, the retro-aldol catalyst, and / or the hydrogenation catalyst. The reactor was then sealed, purged 5 times with H2, and pressurized with H2 to 30 barg at room temperature (20 °C). The reactor was then heated up to reactor temperature. Heat-up time was 10 min. Upon completion, the reactor was cooled to room temperature, vented, and 10 pL / ml -reacti on of a sucrose solution (0.3 M) was added as an external standard. The reaction solution was then filtered and analyzed by an Agilent 1260 Infinity II HPLC equipped with Hi-Plex Ca column (7.7 x 300 mm, 8 pm), and RID and operated with 0.6 ml / min di. H2O as mobile phase. All analytes were identified and quantified by HPLC (calibrations and quantification statistics in FIGS. 55A and 55B). Byproducts contributing to the C-gap in reactions were identified using Agilent 7890B GCMS equipped with an HP-Innowax (30 m. 0.25 pm film) column. The experimental error for conversion and yields for all analytes was ± 2%.

[0243] Reaction Network and Kinetic Considerations

[0244] FIG. 66 shows the reaction network for the direct conversion of glucose to ethylene glycol. Glucose can be converted to glycolaldehyde and erythrose by a retro aldol catalyst. Ery throse can undergo further retro-aldol reaction to produce 2 equivalents of glycolaldehyde. The glycolaldehyde can then be hydrogenated to ethylene glycol over aNi catalyst. A one-pot tandem strategy' comprising of a retro-aldol and size-selective hydrogenation catalyst can leverage the hydrogenation of glycol aldehyde to overcome the retro-aldol equilibrium limitation. In addition to C-shift and retro aldol products, glucose isomerization to fructose can also be observed. The fructose retro-aldol and hydrogenation pathway lead to the formation of Cs polyols like propylene glycol. The C4 sugars also undergo dehydration and hydrogenation reactions to make 1,2-butanediol. For clarity, the products of other sugar reactions such as epimerization and isomerization are omitted from the scheme but mannose, and threose are also observed. Sugar solutions can rapidly equilibrate to a complex mixture of isomers and epimers in the presence of the necessary catalysts. This mixture can subsequently yield a wide range of Ci -Ce degradation products including furfurals, carboxylic acids, and alcohols occurring via uncatalyzed and W- catalyzed homogenous reactions. Hydrogenation provides a pathway to thermodynamically stable polyols enabling efficient conversion of sugars to desired products over other degradation products. The general system involves a pair of rapidly equilibrating species leading to different products through irreversible secondary7reactions.

[0245] Glucose Conversion to Ethylene Glycol: Effect of Reaction Parameters

[0246] Comparison of Encapsulated and Unencapsulated Catalysts

[0247] FIG. 67 shows the tandem reaction with glucose using SMT as the retro-aldol catalyst and 3 different hydrogenation catalysts - Ni / SiCh, Ni / MFI, and Ni@MFI. The catalysts with unencapsulated Ni yield significant quantities of hexitols while Ni@MFI yields primarily ethylene glycol in all reactions. The ethylene glycol selectivity over unencapsulated Ni / SiCh and Ni / MFI is strongly dependent on the catalyst loading even at low reaction temperatures around 140 °C. In contrast, the lower glucose conversion and absence of hexitols at lower temperature (140 °C) over Ni@MFI is consistent with the exclusion of glucose by the MFI framework. Glucose conversion is dependent on unencapsulated catalyst loading but invariant to Ni tyMFI loading at 140 °C, which suggests different rate controlling regimes between encapsulated and unencapsulated catalysts. This is further confirmed by the ethylene glycol selectivity which decreases with increased catalyst loading for unencapsulated catalyst but increases with increased Ni@MFI loading. Increasing the reaction temperature to 170 °C results in 100% conversion for all catalysts. The size-selectivity of Ni@MFI is retained at higher temperatures (170 °C) and conversions, resulting in higher ethylene glycol yields. All catalysts at both loadings can provide improvements to ethylene glycol selectivity at 170 °C in comparison to that at 140 °C. This is due to the shift in the endothermic retro-aldol equilibrium towards increased glycolaldehyde production resulting in higher yields of ethylene glycol. However, the selectivity trends at 170 °C with changing loadings are similar to those at 140 °C for unencapsulated and encapsulated catalysts.

[0248] Impact of Hydrogenation Catalyst Loading

[0249] The impact of hydrogenation catalyst loading on product selectivity is related to the relative rates of hydrogenation and retro-aldol as described in the previous section. The need for fast hydrogenation is evident in FIG. 68 which shows the effect of Ni@MFI and Ni / SiO2 loading on the tandem system. Ethylene glycol selectivity over Ni / SiCh is strongly dependent on catalyst composition, i.e., Ni:W ratio, with a single optimal value corresponding to a local maximum in ethylene glycol yield. The impact of increasing Ni / SiCh loading on the product distribution (FIG. 69) can be explained by a change in the selectivity7. Across all loadings, the selectivity of total hydrogenation to degradation products (i.e. carbon gap in 4.4) is constant around 70%. This implies that hydrogenation rates are faster than / comparable to degradation rates. At low loadings (Ni:W ~ 0.03), the hydrogenation product distribution includes ethylene glycol, propylene glycol, butanediol, and sorbitol, where hydrogenation is slower than sugar equilibrium reactions resulting in hydrogenation products from all sugars. Increasing Ni loading (Ni:W ~ 0.05 - 0. 12) shifts the system to where hydrogenation product distribution is mainly between ethylene glycol, butanediol, and sorbitol. The lack of propylene glycol products implies that hydrogenation is faster than isomerization but slower than retro-aldol and so C-flux towards fructose and propylene glycol is reduced. Further increasing the loading (Ni:W > 0.2) results in only ethylene glycol, and sorbitol, indicative of where the selectivity7is governed by glucose hydrogenation and glucose retro-aldol. Consequently, the selectivity to ethylene glycol decreases and more sorbitol is produced. A similar analysis can be performed for the impact of Ni@MFI loading on product distribution. It should be noted that Ni@MFI imposes an additional restriction on hydrogenation due to its size-selectivity7. For this reason, hydrogenation selectivity is mainly distributed between C2 - C4 products. At low loadings (Ni:W < 0.2), less than 50% total selectivity to polyols is observed due to the formation of byproducts via degradation reactions. This is characteristic of where slow hydrogenation results in rapid equilibration and degradation of sugars. As the catalyst amount is increased (Ni:W > 0.2), the ethylene glycol yield increases reaching a plateau around 65%. At around 40 mg Ni@MFI / g reaction, the hydrogenation rate is the dominant pathway for C-flux with a majority going towards ethylene glycol and a small fraction going towards propylene glycol, and butanediol. Increasing beyond this does not improve the yield of ethylene glycol since the hydrogenation selectivity is now decided by the rate of glycolaldehyde hydrogenation vs C3 and C4 dehydration / hydrogenation reactions. The presence of propylene glycol indicates that hydrogenation is not faster than isomerization, even at the highest loading tested. This difference in the loading behavior between Ni / SiCh and Ni@MFI can be explained by the factors governing hydrogenation kinetics in these systems. Hydrogenation over Ni@MFI is controlled by the diffusion of glycolaldehyde, which means that simply increasing the catalyst loading above a certain threshold will not increase the reaction rate. This is because the diffusion length is an intrinsic property7for this material. Further increases in the reaction rate can be achieved by reducing the transport limitation and increasing the effectiveness factor by engineering the smaller zeolite cry stal sizes, hierarchical structures, and / or improving the intrinsic hydrogenation rate by incorporating precious metals such as Pt or Ru. In addition to the size-selectivity the encapsulated Ni@MFI also affords lower operation temperatures and independence of ethylene glycol selectivity from catalyst composition. This can provide significant benefit when designing a continuous process by expanding the operational window for both reaction temperature and catalyst loading without compromising on selectivity. The Ni@MFI catalyst retains the activity and selectivity7for up to 3 reuses operating at optimal conditions as shown in FIG. 70.

[0250] Impact of Retro-Aldol C atalyst Loading

[0251] Changing the retro-aldol catalyst (SMT) loading has minimal impact on the trends in hydrogenation selectivity for Ni / SiCh is consistent with the observation from the previous section (FIG. 70). Increasing the hydrogenation catalyst loading at the higher SMT loading results in the same trend in hydrogenation product distribution (FIG. 71). The overall selectivity between hydrogenation and degradation is largely unaffected since hydrogenation is still much faster. The impact of SMT loading on Ni@MFI systems is more complex due to effects of SMT on the retro-aldol and degradation rates. Control reactions of glycolaldehyde with SMT only show an increased production of Ci - Ce disproportionation products in comparison to glycolaldehyde stock solutions under the same conditions (FIGS. 59-63). Increasing SMT loading at a fixed Ni@MFI loading results in a slight decrease in the ethylene glycol selectivity7(FIG. 72). Under these conditions, the retro-aldol reaction is expected to reach equilibrium, and so this decrease in ethylene glycol selectivity is due to increase in the W-catalyzed degradation reactions. The primary criteria to ensure high selectivity to ethylene glycol is maintaining fast hydrogenation to rapidly consume the glycolaldehyde produced by the retro-aldol reaction to minimize the degradation of equilibrated sugars. Thus, when increasing the SMT loading, the hydrogenation rate (i.e. Ni@MFI loading) must also be increased to maintain consistent operation (FIG. 72). This results in a different threshold Ni:W ratio for each SMT loading. However, as long as the Ni:W ratio at any SMT loading is above the threshold value, the ethylene glycol yield is invariant to changes in catalyst loading and results in stable performance. Thus, these results imply that the overall ethylene glycol selectivity can be maintained as long as the relative rates are balanced, and the system is operated above the corresponding threshold Ni:W ratio.

[0252] Impact of Temperature

[0253] The impact of reaction temperature can be seen in FIG. 73. While ethylene glycol can be produced at 140 °C, this is accompanied by mannose, and fructose due to the epimerization and isomerization of glucose. The hydrogenation rate can be increased by increasing temperature which results in the highest yield of 63% at 170 °C. At temperatures greater than 170 °C more propylene glycol and 1,2-butanediol are produced resulting in lower ethylene glycol selectivity. The time profile at lower temperatures is shown in FIG. 74, indicating that the size-selectivity is maintained throughout the reaction even in the presence of unreacted glucose. The overall polyol yields and selectivities at 100% conversion was found to be similar across temperatures, even as low as 140 °C (FIG. 73). Thus, the use of a size-selective hydrogenation catalyst allows for lowering the operation temperature by 50 - 100 °C relative to other reported systems without compromising reaction selectivity.

[0254] Impact of Glucose Concentration

[0255] Under a fixed set of reaction conditions and catalyst loadings, lower glucose concentrations result in higher ethylene glycol yields (FIG. 75). This is a consequence of the impact of glucose concentration on the retro-aldol equilibrium, which favors the production of glycolaldehyde at lower concentration of sugar (FIG. 76). At higher glucose concentrations, a large C-gap is observed due to the formation of degradation products which implies that the rate of retro-aldol and hydrogenation are not sufficient to compete with that of side reactions. Operating the tandem system for 29 mM glucose concentration with the same Glu:W and W:Ni ratios as that of the optimal conditions for 12 mM glucose offers slight improvement to the overall ethylene glycol yield (47% from 39%) but is still worse than the performance at lower glucose concentrations (Figure 4.8). This emphasizes the impact of low glucose concentrations on minimizing the extent homogenous side reactions and improving the C-flux towards ethylene glycol.

[0256] Extension of Tandem Reaction to Polysaccharides

[0257] Reactions with Starch

[0258] The optimal conditions identified for the tandem reaction with glucose were then applied to polysaccharides such as starch and cellulose. In general, for polysaccharide substrates, hydrolysis is rate limiting which often necessitates the use of higher temperatures and longer reaction times when compared to glucose systems. The difference in reaction rates is evident in FIG. 77 which shows the slower production of ethylene glycol from starch under the same conditions as glucose reactions. Extending the reaction time to 16 h, results in ethylene glycol yields comparable to using glucose as the substrate. The polysaccharide serves as a stable reservoir of glucose resulting in a low glucose concentration at any given time. This is analogous to a semi-batch system and is expected to improve the selectivity towards ethylene glycol. Hydrolysis is acid-catalyzed, and the reaction rate can be increased significantly by decreasing the solution pH as shown in 4.10. Consequently, 58% ethylene glycol yield is achieved at an initial pH of 2.8 after 4 h at 170 °C. By increasing the reaction temperature, the ethylene glycol yield can be further increased to 76%. In general, acidic conditions are expected to have faster rates of hydrolysis and improved ethylene glycol production, but this effect is more clearly observed at lower temperatures, and consequently lower conversions. The Ni@MFI catalyst maintains a high selectivity to ethylene glycol with minimal sorbitol production as observed for glucose which clearly demonstrates the benefit of size selectivity in this system. It should be noted that the solution pH after reaction is in the range of 4.3 - 5.0, regardless of the inital pH. The disproportionation of sodium metatungstate into a broad distribution of multinuclear tungsten species acts as a buffer resisting change in solution pH during reaction. A particular concern is the dissolution of Ni in acidic solutions under reaction conditions. While some dissolution is expected, no major change in the product distribution is observed after 3 catalyst recycles (FIG. 78). The observed stability could be due to the net buffering of the solution pH by SMT and the reducing reaction conditions (30 bar H2) that can minimize Ni dissolution.

[0259] Reactions with Cellulose

[0260] The structure of the polysaccharide can influence the rate of hydrolysis with starch being easier to hydrolyze than cellulose. The highly ordered structure of cellulose chains results in strong inter-chain hydrogen bonds forming cry stalline domains evident in the XRD of microcrystalline cellulose (MCC) (FIG. 79). Consequently, more forcing conditions are necessary to overcome the stabilizing interactions of cellulose and facilitate hydrolysis. This is particularly evident in the slower rate of ethylene glycol production in the tandem system (FIG. 80). Amorphization of MCC by ball milling can partially disrupt the strong interactions and facilitate faster hydrolysis of cellulose chains. The amorphization can be tracked by the loss in XRD intensity (FIG. 79) and ball milled cellulose (BMC) shows significantly faster rates for ethylene glycol production (FIG. 80).

[0261] Like starch, the addition of acid and higher reaction temperatures can improve the performance of the tandem system resulting in ethylene glycol yields of 50% (FIG. 80). Attempting to optimize the ethylene glycol yield through changes in reaction parameters (temperature, time, and pH) was unsuccessful with a maximum ethylene glycol yield of 55%. Ni@MFI still exhibited lower sorbitol production than the unencapsulated Ni / SiCh for cellulose.

[0262] Conclusions

[0263] This example describes the performance of the size-selective Ni@MFI catalysts for the tandem conversion of mono- and polysaccharides to ethylene glycol. The size selectivity of Ni@MFI affords several advantages for glucose conversion such as minimal sorbitol production, ethylene glycol selectivity independent from catalyst composition, and lowered reaction temperature. The ideal operating conditions can be identified based on an analysis of relative rates and can be tuned by adjusting the Ni:W ratio to be greater than a threshold value. The combination of these factors enables the highest reported glycol (C2 and C3) yields of 75% from a batch glucose system at 140 - 170 °C. The selectivity of the catalyst was also successfully demonstrated with polysaccharides.

[0264] REFERENCES

[0265] [1] Yue, FL; Zhao, Y.; Ma, X.; Gong, J. Ethylene Glycol: Properties, Synthesis, and Applications. Chem Soc Rev 2012, 41 (11), 4218-4244. https: / / doi.org / 10.1039 / c2cFIG. 26359a.

[0266] [2] Baniamerian, H.; Hoj, M.; Beier, M. J.; Jensen, A. D. Catalytic Conversion of Sugars and Polysaccharides to Glycols: A Review. Appl Catal B 2023. 330, 122650. https: / / doi.Org / 10.1016 / J.APCATB.2023.122650.

[0267] [3] Zheng, M.; Pang, J.; Sun, R.; Wang, A.; Zhang, T. Selectivity' Control for Cellulose to Diols: Dancing on Eggs. ACS Catal 2017, 7 (3), 1939-1954. https: / / doi.org / 10.1021 / ACSCATAL.6B03469 / ASSET / IMAGES / MEDIUM / CS-2016- 03469W_0008.GIF.

[0268] [4] Wang, A.; Zhang, T. One-Pot Conversion of Cellulose to Ethylene Glycol with Multifunctional Tungsten-Based Catalysts. Acc Chem Res 2013. https: / / doi.org / 10.1021 / ar3002156.

[0269] [5] Ji, N. ; Zhang, T. ; Zheng, M. ; Wang. A. ; Wang, H. ; Wang. X. ; Chen, J. G. Direct Catalytic Conversion of Cellulose into Ethylene Glycol Using Nickel-Promoted Tungsten Carbide Catalysts. Angewandte Chemie 2008, 120 (44), 8638-8641. https: / / doi.org / 10.1002 / ange.200803233.

[0270] [6] Zhang, J.; Hou, B.; Wang, A.; Li, Z.; Wang, H.; Zhang, T. Kinetic Study of the Competitive Hydrogenation of Glycolaldehyde and Glucose on Ru / C with or without AMT. Aiche Journal 2015. https: / / doi.org / 10.1002 / aic.14639.

[0271] [7] Cao. Y; Wang, J.; Kang, M.; Zhu, Y. Efficient Synthesis of Ethylene Glycol from Cellulose over Ni-WO3 / SBA-15 Catalysts. J Mol Catal A Chem 2014, 381, 46-53. https : / / doi . org / 10. 1016 / j . molcata.2013.10.002.

[0272] [8] Ribeiro, L. S.; Pires, A. L. F.; Orfao, J. J. de M.; Pereira, M. F. R. Paving the Way towards an Eco- and Budget-Friendly One-Pot Catalytic Conversion of Cellulose and Lignocellulosic Residues into Ethylene Glycol over Ni-W / CNT Catalysts. Renew Energy 2022, 200, 1008-1022. https: / / doi.Org / 10.1016 / j.renene.2022.10.026. [9] Cao, Y.; Wang, J.; Kang, M.; Zhu, Y. Catalytic Conversion of Glucose and Cellobiose to Ethylene Glycol over Ni-WO3 / SBA-15 Catalysts. RSCAdv, 2015, 5, 90904- 12 https: / / doi.org / 10. 1039 / c5ral5400f.

[0273]

[0010] Li, S.; Tuel, A.; Laprune, D.; Meunier, F.; Farrusseng, D. Transition-Metal Nanoparticles in Hollow Zeolite Single Crystals as Bifunctional and Size-Selective Hydrogenation Catalysts. Chemistry of Materials 2015, 27 (1), 276-282. https : / / doi. org / 10.1021 / cm503921 f.

[0274]

[0011] Cao, Y.; Wang, J.; Kang, M.; Zhu, Y. Catalytic Conversion of Glucose and Cellobiose into Ethylene Glycol over Various Tungsten-Based Catalysts. Journal of Fuel Chemistry and Technology! 2016. https: / / doi.org / 10.1016 / sl872-5813(16)30038-x.

[0275]

[0012] Ooms, R.; Dusselier, M.t Geboers, J. A.; Op De Beeck, B.; Verhaeven, R.;

[0276] Gobechiya, E.; Martens, J. A.; Redl, A.; Seis, B. F. Conversion of Sugars to Ethylene Glycol with Nickel Tungsten Carbide in a Fed-Batch Reactor: High Productivity and Reaction Network Elucidation. Green Chemistry 2014, 16 (2), 695-707. https: / / doi.org / 10.1039 / C3GC41431K.

[0277]

[0013] Artz, J.; Palkovits, R. Cellulose-Based Platform Chemical: The Path to Application. Curr Opin Green Sustain Chem 2018, 14, 14-18. https: / / doi.Org / 10.1016 / J.COGSC.2018.05.005.

[0278]

[0014] Guleria, A.; Kumari, G.; Saravanamurugan, S. Cellulose Valorization to Potential Platform Chemicals. Biomass, Biofuels, Biochemicals: Recent Advances in Development of Platform Chemicals 2020, 433-457. https: / / doi.org / 10.1016 / B978-0-444-64307-0.00017-2.

[0279]

[0015] Zhao, G.; Zheng, M.; Sun, R.; Tai, Z.; Pang, J.; Wang, A.; Wang, X.; Zhang, T. Ethylene Glycol Production from Glucose over W-Ru Catalysts: Maximizing Yield by Kinetic Modeling and Simulation. AIChE Journal 2017, 63 (6), 2072-2080. https: / / doi.org / 10.1002 / aic.15589.

[0280]

[0016] Lew, C. M.; Rajabbeigi, N.; Tsapatsis, M. Tin-Containing Zeolite for the Isomerization of Cellulosic Sugars. Microporous and Mesoporous Materials 2012, 153, 55- 58. https: / / doi.Org / 10.1016 / J.MICROMESO.2011.12.020.

[0281]

[0017] Orazov, M.; Davis, M. E. Tandem Catalysis for the Production of Alkyl Lactates from Ketohexoses at Moderate Temperatures. Proc Natl Acad Sci U S A 2015. https: / / doi.org / 10.1073 / pnas.1516466112.

[0282]

[0018] Farrusseng, D.; Tuel, A. Perspectives on Zeolite-Encapsulated Metal Nanoparticles and Their Applications in Catalysis. New Journal of Chemistry 2016, 40 (5), 3933-3949. https: / / doi.org / 10.1039 / C5NJ02608C.

[0283]

[0019] Wang, H.; Wang, L.; Xiao, F. S. Metal@zeolite Hybrid Materials for Catalysis. ACS Cent Sci 2020, 6 (10), 1685-1697. https: / / doi.Org / 10.1021 / ACSCENTSCI.0C01130 / ASSET / IMAGES / LARGE / OC0C01130_0 006. JPEG.

[0020] Gao, C.; Lyu, F.; Yin, Y. Encapsulated Metal Nanoparticles for Catalysis. Chemical Reviews. American Chemical Society 2020. https: / / doi.org / 10.1021 / acs.chemrev.0c00237.

[0284]

[0021] Dai, C.; Du, K.; Song, C.; Guo, X. Recent Progress in Synthesis and Application of Zeolite-Encapsulated Metal Catalysts. In Advances in Catalysis; Academic Press, 2020; Vol. 67, pp 91-133. https: / / doi.org / 10.1016 / bs.acat.2020.10.001.

[0285]

[0022] Kwok, K. M.; Ong, S. W. D.; Chen, L.; Zeng, FI. C. Transformation of Stober Silica Spheres to Hollow Hierarchical Single-Crystal ZSM-5 Zeolites with Encapsulated Metal Nanocatalysts for Selective Catalysis. ACS Appl Mater Interfaces 2019, 11 (16), 14774- 14785. https: / / doi.org / 10.1021 / acsami.9b00630.

[0286]

[0023] Choi, M.; Wu, Z.; Iglesia, E. Mercaptosilane- Assisted Synthesis of Metal Clusters within Zeolites and Catalytic Consequences of Encapsulation. J Am Chem Soc 2010, 132 (26), 9129-9137. https: / / doi.org / 10.1021 / jal02778e.

[0287]

[0024] Otto, T; Zones, S. I.; Iglesia, E. Synthetic Strategies for the Encapsulation of Nanoparticles of Ni, Co, and Fe Oxides within Crystalline Microporous Aluminosilicates. Microporous and Mesoporous Materials 2018, 270, 10-23. https: / / doi.Org / 10.1016 / j.micromeso.2018.04.045.

[0288]

[0025] Goel, S.; Zones, S. I.; Iglesia, E. Encapsulation of Metal Clusters within MFI via Interzeolite Transformations and Direct Hydrothermal Syntheses and Catalytic Consequences of Their Confinement. J Am Chem Soc 2014, 136 (43), 15280-15290. https: / / doi.org / 10.1021 / ja507956m.

[0289]

[0026] Pagis, C.; Morgado Prates, A. R.; Farrusseng, D.; Bats. N.; Tuel, A. Hollow Zeolite Structures: An Overview of Synthesis Methods. Chemistry of Materials 2016, 28 (15), 5205-5223. https: / / doi.org / 10.1021 / ACS.CHEMMATER.6B02172 / ASSET / IMAGES / LARGE / CM- 2016-02172Y_0020. JPEG.

[0290]

[0027] Li, S.; Boucheron, T; Tuel, A.; Farrusseng, D.; Meunier, F. Size-Selective Hydrogenation at the Subnanometer Scale over Platinum Nanoparticles Encapsulated in Silicalite-1 Single Crystal Hollow Shells. Chemical Communications 2014, 50 (15), 1824. https: / / doi.org / 10.1039 / c3cc48648f.

Claims

What is claimed:

1. A method of forming a catalyst material by selective encapsulation of one or more transition metals within a porous substrate, comprising the steps of: a. introducing a transition metal onto a microporous and / or mesoporous support material by impregnation or ion exchange of a salt or organometallic compound of the transition metal; b. drying the support material onto which the transition metal has been introduced in step a; c. thermally treating the dried support material obtained in step b at a temperature of 250 C to 700 C in the presence of an inert gas, oxygen, or hydrogen to at least partially agglomerate the transition metal as metal or metal oxide nanoparticles deposited on the support material; and d. treating the thermally treated support material obtained in step c with an aqueous solution of an organic acid having a smallest molecular dimension that is larger than a largest dimension of a pore opening of the microporous and / or mesoporous support material to substantially leach any transition metal deposited on the support material as metal or metal oxide outside the pores of the support material.

2. The method of claim 1, wherein the transition metal comprises manganese, iron, cobalt, nickel, copper, and / or zinc, preferably nickel.

3. The method of claim 1 or claim 2, wherein the support material comprises a microporous and / or mesoporous aluminosilicate, aluminophosphate, and / or silicoaluminophosphate zeolite and / or zeotype material or a microporous and / or mesoporous carbon material, preferably a microporous aluminosilicate zeolite and / or zeotype material, more preferably an MFI zeolite, most preferably a high-silica MFI zeolite.

4. The method of any of claims 1 to 3, wherein step d is performed with an oxidizing agent, preferably oxygen and / or hydrogen peroxide.

5. The method of any of claims 1 to 4, wherein the organic acid comprises acetic and / or trimesic acid and the support material comprises an MFI zeolite.

6. The method of any of claims 1 to 5, further comprising the steps of: e. reducing the catalyst material obtained in step I d under hydrogen or in the presence of a chemical reducing agent; and f. contacting the reduced catalyst composition obtained in step e under an inert and / or hydrogen atmosphere with a solution of a noble metal salt in which the noble metal is present in an oxidized state, wherein the noble metal is introduced into the pores of the support material.

7. The method of claim 6, wherein the noble metal comprises one or more of platinum, palladium, ruthenium, silver, gold, rhodium, and iridium, preferably platinum.

8. The method of claim 6 or claim 7, wherein the noble metal is encapsulated as a monometallic or bimetallic particle within the porous substrate.

9. The method according to any of claims 6 to 8, wherein the noble metal salt solution comprises water or a polar organic solvent, preferably water and / or an alcohol.

10. A catalyst composition, produced by the method of any of claims 1 to 9.

11. A method for producing short-chain polyols from biomass-derived mono- and / or polysaccharides using the catalyst composition of claim 10, comprising the steps of: a. reacting an aqueous solution comprising biomass-derived mono- and / or polysaccharides comprising one or more of hemicellulose, glucose, fructose, xylose, glycolaldehyde, starch, and cellulose with hydrogen in the presence of the catalyst composition and optionally one or more retro-aldol catalysts at a temperature of 50 C to 250 C, preferably 100 C to 200 C, more preferably 125 C to 200 C, most preferably 150 C to 175 C, a pressure of 5 to 100 barg, preferably 5 to 75 barg, more preferably 10 to 50 barg, most preferably 20 to 40 barg, and a pH of 2 to 6, preferably 4 to 5, for a time of 0.5 to 24 hours, preferably 1 to 10 hours, and at a catalyst loading of 0.1 wt.% to 10 wt.%, based on the solution weight; andb. recovering one or more of the short-chain polyols from the aqueous solution.

12. A catalyst composition comprising a microporous and / or mesoporous high-silica- content MFI zeolite and / or zeotype support material encapsulating nanoparticles comprising nickel and / or nickel oxide.

13. The catalyst composition of claim 12, additionally comprising one or more noble metals encapsulated in the support material.

14. The catalyst composition of claim 12 or claim 13, wherein the one or more noble metals comprise one or more of platinum, palladium, ruthenium, silver, gold, rhodium, and iridium.

15. The catalyst composition of any of claims 12 to 14, wherein the support material has an external surface that is substantially free of the nickel and / or nickel oxide and / or the one or more noble metals.

Citation Information

Patent Citations

  • Method of producing zeolite encapsulated nanoparticles

    US20160129429A1

  • Methods for selective leaching and extraction of precious metals in organic solvents

    US20180112289A1

  • Methods and Compositions for Isolation of Copper Group Metals

    US20200277683A1

  • Catalyst for the dehydroaromatisation of methane and mixtures containing methane

    US8742189B2

Cited By

  • Zn-Pd bimetal hierarchical pore ZSM-5 molecular sieve and preparation method and application thereof

    CN120861136A

  • A Zn-Pd bimetallic hierarchical pore ZSM-5 molecular sieve, a preparation method and application thereof

    CN120861136B

  • Preparation method and application of catalyst for treating high-salt phenol-containing wastewater

    CN121402117A