ABOUT CCEI
RESEARCH
PEOPLE & PARTNERS
NEWS
RESOURCES
CONTACT US

ABOUT CCEI
RESEARCH
PEOPLE & PARTNERS
NEWS
RESOURCES
CONTACT US

Zhexi Lin

CCEI| People| Zhexi Lin

lin,-z

Zhexi Lin

E-mail: zl2453@columbia.edu
Institution:
Graduation Year:
Advisor(s): Jingguang Chen
Research Interests:

Email:

Phone:

CCEI Search

Member Portal

Login
Log Out
Forgot Password

AN ENERGY FRONTIER
RESEARCH CENTER

Funded by the Department of Energy

©2021 Catalysis Center for Energy Innovation. All Rights Reserved. | Legal Notices | Accessibility | Comments

Adhesives

Research goal: Synthesize performance-advantaged pressure-sensitive adhesives from leftover lignin to replace materials typically made from petroleum.

Approach: Develop efficient catalysts for depolymerization of lignin (main component in pulp and paper waste, bioethanol refineries, and future biorefineries) into functionalized aromatic monomers at high yield. Add designer molecules to each aromatic molecule, attach the units together, and finally, cast them onto sheets.

Learn more about adhesives

Waste-derived adhesives exhibit excellent adhesion to stainless steel without the addition of any tackifier or plasticizer, typically used in the manufacturing of adhesives from petroleum. We synthesize adhesives that are just as competitive as commercial tapes. Applications range from packaging and plastic wraps to labels and sticky notes.

Research highlights

We have introduced a strategy toward the development of high-performance, pressure-sensitive adhesives (PSAs) from raw biomass. By depolymerizing poplar wood with minimal purification, we have produced substituted aromatic compounds such as 4-propylsyringol and 4-propylguaiacol. Their efficient functionalization with either acrylate or methacrylate groups yield monomers that could easily be polymerized by a scalable, reversible addition-fragmentation chain-transfer process into polymeric materials with high glass-transition temperatures and robust thermal stabilities, especially relative to other potentially bio-based alternatives.

Lubricants

Research goal: This thrust focuses on innovative technologies that target new performance-advantaged bio-products. We develop sustainable, durable and high-performance bio-lubricants from inexpensive and sustainable renewables, including non-food biomass and fat, natural oils or waste cooking oils.

Approach: Develop various selective C-C coupling reaction chemistries via a slate of catalysts to build the carbon chain length starting from furans and long-chain functionalized molecules, such as fatty acids or their derivatives. Control with precision the number of carbons, the degree of branching and the stereospecificity of the resulting molecules. Scale up production. Evaluate typical performance metrics, such as viscosity index, pour point, volatility, and stability, of virgin (without additives) bio-lubricants relative to those of commercial, high-performance, additives-containing lubricants.

Learn more about lubricants

Lubricants represent a ~$61 billion global market at an annual consumption of 40 million tons. They have numerous applications in industrial machineries, automotive engines, agricultural equipment, aviation machinery, hydraulic fluids, hydropower, and transformers. Petroleum-based lubricants consist of C20-C50 molecules and are formulated with additives. The high volatility of low-carbon fractions causes thickening over time, requiring frequent lubricant replacements. High-performance applications under extreme conditions (e.g., in gas turbines and nuclear reactors) demand rigid specifications (e.g., low pour point, high thermal stability, low volatility, and high viscosity index). Synthetic, branched poly-alpha-olefin lubricants produced by olefin oligomerization meet some specifications, but selectively tuning their size and architecture is challenging. Use of less expensive, sustainable, and versatile carbon feedstocks to produce lubricants with controlled size and molecular architecture, and a narrow molecular weight distribution can improve lubricant properties and avoid complex formulations and high cost.

Research highlights

We have developed novel synthesis strategies to produce three broad classes of bio-lubricants from inexpensive and sustainable non-food biomass and natural oils or waste cooking oils. Yields ranging from 80 to over 90% have been achieved. These molecules contain furan rings, or fully saturated furan rings, or fully deoxygenated branched alkanes. Importantly, we can control with precision the number of carbons, the molecular structure, the degree of aromaticity and the oxygen level. Our carbon-carbon coupling synthesis strategies include: (1) acylation of furan; (2) hydroxyalkylation/alkylation of 2-alkylfurans with aldehydes; (3) conjugate addition-hydroxyalkylation/alkylation (CA-HAA) of 2-alkylfurans with enals; and (4) ketonic decarboxylation of fatty acids followed by aldol condensation. The derived furan-based lubricants can be further upgraded via hydrogenation or hydrodeoxygenation. New, inexpensive metal oxide on metal hydrodeoxygenation catalysts have been developed. The properties of these bio-lubricants are on par with or better than those of commercial mineral oils and synthetic lubricants.

We have introduced a novel, two-step strategy to synthesize benzene and branched cyclic alkane lubricant base-oils from lignin-derived guaiacol and lauryl aldehyde (produced from natural oils or biomass). The reaction pathway involves carbon–carbon coupling through Brønsted acid-catalyzed hydroxyalkylation/alkylation (HAA), followed by hydrodeoxygenation (HDO). We optimized the reaction conditions to achieve a maximum guaiacol conversion of 90%, with a product consisting of 70% benzene lubricant, 22% aldol condensation side products, and 8% unreacted guaiacols. Subsequent HDO of the products over Ir-ReOx/SiO₂ produced a lubricant-ranged mixture of cyclic and branched alkanes (C₂₄), at 82% yield, and small fractions of dodecyl cyclohexane and C-C cracked alkanes with carbon numbers between C₁₅ and C₁₀. Quantification of viscosity properties (kinematic viscosity, viscosity index, and Noack volatility) indicated that the synthesized biobased lubricant base-oil was comparable to commercial petroleum-derived poly α-olefin Group III, IV, and refrigerant base oils. This approach provides a unique pathway for upgrading lignin-derived monomers into replacements of petroleum-derived base oils. We are expanding to additional lignin-based paths to lubricants and replace fatty acids with sugar-derived molecules.

Dienes

Research goal: Synthesize dienes, such as butadiene and isoprene, from sugar-derived oxacyclopentanes (tetrahydrofurans).

Objectives: Develop selective catalysts for the ring opening and dehydration of tetrahydrofurans. To that end, we are working in two directions.

First, we are developing and extending a new class of catalysts, dubbed P-zeosils, consisting of phosphoric acid supported on inert all-silica zeolites. The P-zeosils were discovered by CCEI researchers and after extensive characterization and computer simulations were found to behave like Brønsted acids but with properties quite distinct from typical aluminosilicate zeolites. For one, P-zeosils present a distribution of dynamically changing active sites whose nature and activity can be modulated by varying the temperature, pressure and moisture content. Using advanced characterization techniques (e.g., solid-state NMR spectroscopy) we are uncovering their structural secrets and developing principles that will allow us to extend P-zeosils into a broader class of materials consisting of dynamically confined inorganic acids, for example, using metal-organic frameworks as supports.

Second, we are venturing into the Lewis acid catalyst space and metal oxides, in particular. After extensive testing, CCEI researchers have discovered that zirconia (ZrO₂) consistently exhibited the highest selectivity for 1,3-butadiene (~90%) when reacted with tetrahydrofuran.

Learn more about dienes

Dienes are important, high-volume chemicals used as feedstock for polymers. They can also be converted to larger alkanes for diesel, surfactants and lubricants via alkylation, acylation and hydroformylation. Diene monomers are a worldwide feedstock integral to growing economies. The shale-gas revolution and concomitant reduction in mid-range hydrocarbons, market pressure on butadiene, isoprene (C5) and piperylenes (linear C5), the growing rubbers and elastomers market and consumer interest in renewables make dienes attractive targets. Butadiene production reached 15 million tons in 2018, with applications in tough styrene-butadiene rubber, hard acrylonitrile-butadiene-styrene, and stretchy nitrile-butadiene rubber. The butadiene growth rate in the next decade is projected to be 3%, while isoprene, a key monomer in car tire manufacturing, is similarly growing, especially in Asia-Pacific. We are seeking transformative routes to renewable dienes to cover market demands.

Research highlights

CCEI researchers discovered that phosphoric acid supported on otherwise inert all-silica zeolites [P-zeosils: e.g., P-Beta, P-MFI and P-SPP (self-pillared pentasil)] with Si/P>10 enables unprecedented (97%) p-xylene yields by selectively catalyzing dehydration over undesired alkylation and oligomerization reactions. Soon thereafter, we explored P-zeosils to produce butadiene and isoprene via ring opening and dehydration of tetrahydofuran (THF) and methyl-tetrahydrofuran (MTHF). We showed that the selectivity to dienes varies dramatically across catalysts. P-zeosils exhibit significantly higher yield to butadiene (75% at 90% THF conversion), isoprene and pentadiene (35 and 15%, respectively, at 70% 3-MTHF conversion).

We have undertaken detailed first-principles mechanistic and kinetic studies of the dehydra-decyclization of tetrahydrofuran (THF) to butadiene over H-ZSM5, P-ZSM5 and P-BEA in order to understand the origins of the selectivity of P-zeosils. We have mapped out comprehensive reaction networks that include pathways to butadiene and the major side-reaction pathway (retro Prins-condensation) to formaldehyde and propene. On H-ZSM5, the selectivity for butadiene is ca. 50%. Similar studies for P-BEA and P-ZSM5 for P-BEA predict selectivity to butadiene as high as 93%. Micro-kinetic analysis has shown that ring opening of THF is both rate-limiting and selectivity-controlling for butadiene formation.

We have characterized the state and evolution of phosphorus in P-zeosils, before, during, and after reaction, as well as elucidated the effect of the zeolite framework structure (SPP vs. BEA) on the distribution and hydrolytic stability of the phosphorus sites in P-zeosils using NMR spectroscopy. Solid-state NMR with magic-angle spinning (MAS) in combination with Dynamic Nuclear Polarization (DNP)-induced signal enhancement was used to acquire site-specific structural and dynamic information about the P-site structures, their distribution, solvent accessibility and hydrolytic stability, as well as to study the impact of the zeosil framework on the distribution and stability of the acidic P-sites. We further employed operando MAS-NMR spectroscopy to study catalytic reactions in P-zeosils. We identified the P-sites in P-SPP with a low P-loading (Si/P = 27) using SS ³¹P NMR and frequency-selective detection. The spectra revealed a previously unappreciated diversity of P-sites, including evidence for surface-bound oligomers. In dry P-SPP, essentially all P-sites are anchored to the solid phase, including mono- and dinuclear sites containing the [Si−O−P−O−P−O−Si] motif. These sites evolve rapidly when exposed to humidity, even at room temperature, in accordance with the ab initio calculations performed by CCEI researchers. The partially hydrolyzed sites have a wide range of acidities, as inferred from calculated LUMO energies. Initial cleavage of some P−O−Si linkages results in an evolving mixture of surface-bound, mono- and oligo-nuclear P-sites with increased acidity. Subsequent P−O−P cleavage leads to a decrease in acidity as the P-sites are eventually converted to H₃PO₄ (Q⁰).

Surfactants

Research goal: This thrust focuses on innovative technologies that target new performance-advantaged bio-products. We develop sustainable, durable and high-performance surfactants from inexpensive and sustainable renewables.

Approach: Develop carbon-carbon (C-C) coupling strategies on the furan molecule to build chains of varying size and branching to produce aliphatic chains that can be utilized as hydrophobic tails after partial hydrodeoxygenation. Sulfonate the resulting molecule to build a hydrophilic head. Scale up production. Evaluate performance of detergents against that of commercial products. One C-C coupling chemistry entails the acylation of furan. The indirect route, acylation by carboxylic acid anhydrides, is process intensive and carbon inefficient: stoichiometrically, half of the anhydride backbone ends up as by-product. On the other hand, directly acylating with the carboxylic acid itself is not wasteful as water is the only by-product.

Learn more about surfactants

Surfactant molecules are a core class of chemicals broadly relevant to industry, personal care products and environmental applications, comprising a $40 billion/year market worldwide that is expanding faster (>5%) than global trends in energy and GDP. Significant industries include cleansers, detergents and soap, while other formulated products include paints, coatings and drilling fluids. Surfactants are applied throughout the agricultural sector in herbicides and insecticides (crop protection) of more than $3 billion annually. Most detergents are washed down the drain, while almost all agricultural surfactants are found in the food supply. This has led to publicly announced programs to produce renewable surfactants with traits not currently available.

Research highlights

We have developed strategies to synthesize surfactants from bio-renewable feedstocks as they have potential environmental benefits in addition to unique physicochemical properties that could improve performance and expand applicability. We have shown that sugar-derived furans can be linked with triglyceride-derived fatty acid chains via Friedel−Crafts acylation within single layer (SPP) zeolite catalysts to yield oleo-furan surfactants with superior properties compared to standard detergents. These alkylfuran surfactants independently suppress the effects of hard water while permitting broad size tunability, structure and function. These molecules can be optimized for superior capability of forming micelles and of solubilizing in water. Insights into the fundamentals of acylation chemistry have been obtained. Improved catalysts and processes are being developed.

Lignocellulose Hydrolysis

Research goal: Develop energy efficient, sustainable processes for the selective deconstruction of lignocellulose to high yields of monosaccharides and intact lignin.

Approach: Develop molten salt hydrate catalytic systems for the efficient dissolution, exfoliation, and deconstruction of biomass. Convert the resulting cellulose and hemi-cellulose into high yield sugars. Develop efficient separation strategies of the sugars.

Learn more about lignocellulose hydrolysis

The complex structure of lignocellulose and its strong hydrogen bonds make its deconstruction a great challenge for the second-generation cellulosic biorefineries. Deconstruction of biomass has been one of the biggest hurdles toward commercialization. Enzymatic processes can be used but they are inherently slow. Harsh conditions, employing acids, alkalis, ammonia/CO₂ or liquid hot water, ionic liquids (IL), and ball milling, are energy intensive. Developing energy efficient, sustainable processes for the selective hydrolysis of lignocellulose into cellulose and hemi-cellulose, and their conversion to monosaccharides, and intact lignin under mild conditions is essential.

Research highlights

We have developed a one-pot saccharification strategy that uses inorganic salt solutions and less water and combines the lignocellulose pretreatment and saccharification steps under mild conditions (85 C, 1 h). At sufficiently low water content, the solution is called molten salt hydrate. This process intensification also uses less energy and can produce up to 95% theoretical glucose and xylose, from multiple lignocellulose sources. This process produces sugars at faster rates owing to exfoliation of the biopolymers, the greater accessibility of the porous biomass fibers by the salt solution, and the very effective proton-catalyzed de-etherification of the C-O-C bonds. Subsequent biphasic integration of the saccharification with dehydration enables reactive extraction of products such as 5-hydroxymethylfurfural (HMF) and furfural from the homogeneous hydrolysates. The cooperativity of Lewis and Brønsted acidity and high viscosity of the salt solution result in higher HMF and furfural yields (>70 mol%) compared with non-salt systems. Techno-economic analysis of the process has shown significant cost advantages. In a parallel effort, we have developed carbon adsorbents for separation of soluble sugars from the salt solution. A biphasic, single pot, two-step reactor was introduced to directly convert the lignocellulose to furans, without the need for separation of sugars. More recently, we have provided fundamental insights into the origins of super-acidity of these systems and rationalized why certain salts synergize with certain inorganic acids (when added in very low concentration) to create such effective dissolution and catalyst media. Further, we developed a NMR method to quantify the acidity in such complex media.

Aromatics

Research goal: Selective upgrade of furans to aromatics for renewable plastics.

Approach: Develop strategies for high selectivity Diels-Alder (DA) reaction of furans and of substituted furans with dienophiles via cycloaddition and dehydration of the resulting cycloadducts.

Learn more about aromatics

Aromatics constitute a large market, with BTX (benzene-toluene-paraxylene) being one of the highest-volume and considerably energy-intensive chemicals made from crude oil. As steam cracker operators are capitalizing on the shale-gas revolution and inexpensive natural gas liquids are pushing naphtha feedstocks aside, the development of alternative routes for the production of aromatics from renewables, such as biomass, is receiving considerable attention. Of the possible routes garnering interest, CCEI scientists have proposed the dehydrative aromatization of the Diels-Alder product between bio-based furans with an appropriate dienophile; a tandem scheme where the cycloadduct obtained from the Diels-Alder is catalytically dehydrated to an aromatic. The strategy was showcased by the formation of p-xylene from 2,5-dimethylfuran and ethylene over the Brønsted-acidic catalysts H-Y and H-BEA. The same strategy has been used to convert methylfuran and ethylene to toluene; ethylene and methyl 5-(methoxymethyl)-furoate to methyl 4-(methoxymethyl)-benzoate; ethylene and the dimethyl ester of 2,5-furandicarboxylic acid to dimethyl terephthalate; furan and maleic anhydride to phthalic anhydride; and furan and methyl acrylate to methylbenzoate. and furan and ethylene to benzene.

Research highlights

In order to curtail undesired side reactions promoted by Brønsted acids, CCEI researchers turned their attention to Lewis-acidic zeotypes, specifically isomorphically substituted zeotypes with Lewis-acidic centers, such as Sn-BEA, Zr-BEA, Ti-BEA, Hf-BEA and Zn-BEA. This multi-institution effort led to the synthesis of numerous aromatics: methyl 4-(methoxymethyl)-benzoate from ethylene and methyl 5-(methoxymethyl)-furoate; dimethyl terephthalate from ethylene and the dimethyl ester of 2,5-furandicarboxylic acid; as well as p-xylene. These efforts led to the remarkable finding that Lewis-acidic zeotypes are as effective as the Brønsted-acidic H-BEA in influencing the rate of the tandem scheme, and in particular the dehydration of oxanorbornene derivatives.

The fleeting existence of the Diels-Alder cycloadduct does not allow one to decouple the two steps of the tandem scheme and thus elucidate whether the zeotypic Brønsted and Lewis acids have any effect on the rate of the Diels-Alder step itself. With furans being among the less reactive dienes, accelerating the cycloaddition reaction will be a pivotal step in the success of producing aromatics by dehydrative aromatization of the Diels-Alder product of biomass-derived furans and appropriate dienophiles. Calculations have asserted that when ethylene is the dienophile, neither Brønsted nor Lewis acid zeotypes can influence the rate of the Diels-Alder step, because the furanic molecule binds more strongly to the active site of the catalysts.

CCEI catalysis experts have achieved to synthesize methylbenzoate by Diels-Alder aromatization of furan and electrophilically activated methylacrylate over the zeotypic Lewis acids Sn-BEA, Zr-BEA, and Hf-BEA. With the aid of computational chemistry, they were able to show that these Lewis acids are capable of accelerating the Diels-Alder reaction heterogeneously.

In order to curtail by-product formation, CCEI researchers have also been exploring phosphorous-containing siliceous zeolites. They were able to synthesize zeolites with BEA (P-BEA) and MFI topology (P-SPP). These catalysts are highly selective and stable for this reaction, reaching an unprecedented p-xylene yield of 97%. The superior properties of phosphorous-containing siliceous zeolite catalysts are attributed to the nature of the acid sites which exclusively catalyze the dehydration of cycloadduct intermediate, and remarkably minimize the alkylation and oligomerization reactions.

Hydrodeoxygenation
(HDO)

Research goal: Selectively upgrade furans to fuel-grade compounds and to (partially) deoxygenated chemicals.

Approach: Develop design principles of efficient hydrodeoxygenation catalysts and for ring opening of furanics and hydrodeoxygenation by combining multiscale modeling and experiments. Synthesize multisite/multifunctional materials with nearly atomic scale control to carry out these processes selectively. Perform kinetic and spectroscopic studies for upgrading of furan-based compounds.

Learn more about HDO

The HDO Thrust has aimed to develop science and catalysts for the upgrade of furan-based compounds.

Furans, obtained from dehydration of sugars or pyrolysis of lignocellulose and oxygenated fuel precursors obtained from chain growth chemistry of furans are important intermediate platforms of lignocellulosic biomass.

However, owing to their oxygen-rich content, their transformation to fuels and chemicals requires selective oxygen removal. Upgrade of furans requires transformative technology to improve product yields and introduce new chemicals and to reduce catalyst cost.

Research highlights

We discovered that moderately reducible metal oxides, such as RuOx, can contain multifunctional sites that can selectively hydrogenate carbonyl groups, compared to the ring, via the MPV mechanism over Lewis acidic sites, and then hydrodeoxygenate the side OH groups on vacancies. The latter behave like redox centers, where a single concerted mechanism occurs, entailing radical formation and ring conjugation to facilitate the C-OH bond scission. While RuOx is an effective and selective catalyst, it is being reduced during catalysis. In order to improve catalyst stability, this concept was extended further. Computational studies allowed for the first time to establish catalyst design principles, which can ultimately lead to the discovery of the next-generation of C-O activation catalysts. First, we have explored the activity and stability tradeoff of single metal oxides and demonstrated a volcano curve where the activity/stability is correlated with the bulk Gibbs free energy as well as surface properties of the oxides. RuOx and IrOx are the best single metal oxide catalysts. Second, we used colloidal synthesis and atomic layer deposition (ALD) of bimetallics to create metal core/oxide monolayer thick shell, e.g., cobalt-oxide-overcoated Pt nanoparticles and Cu/Ni catalysts. These catalysts activate hydrogen and importantly do not get reduced; they can excitingly provide nearly quantitative yield in converting HMF to dimethyl furan (DMF).

In a parallel effort, we developed efficient and highly selective catalysts for (1) ring-opening catalytic pathways to convert biomass-derived furans and tetrahydrofurans into adipic acid, an essential monomer in 6,6-nylon; (2) deoxydehydration of vicinal diols to olefins, a class of compounds with a multitude of uses. A first generation of catalysts was obtained by adopting transition metals and structural motifs from homogeneous catalysts, e.g., oxide-supported oxo-rhenium; (3) non-precious metal oxide catalysts for the ring-opening of furanics. We have demonstrated that the use of reduced Cu-Co-Al mixed metal oxides achieves the formation of 1,5-pentanediol with upward yields of 45% and 1,2-pentanediol with upward yields of 17%. This is the highest reported yield for 1,5-pentanediol when compared with other catalysts with a similar mechanism; (4) production of mono and di-ethers of furans suitable for detergents and fuels via the MVP reaction over Lewis acid centers, e.g., Sn-containing zeolites, and etherification over acid sites.

Furans

Research goal: Produce furans or other value-added chemicals by efficient transformation of sugars derived from cellulose and hemi-cellulose.

Approach: Develop sugar transformations in Lewis acid zeolites and over homogeneous Lewis acid catalysts. Develop tandem reaction schemes, design principles, associated Lewis and Bronsted acid zeolite and other bifunctional catalysts for these transformations. Elucidate mechanisms and kinetics.

Learn more about furans

Furans are potentially the most versatile platform molecules for converting lignocellulose into various bio-products. CCEI has demonstrated a number of new bio-products derived from furans. Yet, furans production has been hindered by low yields due to various byproducts, such as humins. The furans thrust aimed at paradigm changing approaches for the selective transformations of sugars to furans and their further conversion to value-added chemicals and fuels.

Research highlights

Inspired by an earlier demonstration of intermolecular Meerwein-Ponndorf-Verley (MPV) carbonyl reduction and Oppenauer alcohol oxidation (MPVO) in Sn-BEA, we demonstrated sugar isomerization by an analogous intramolecular MPVO cycle. We proposed that the Sn⁴⁺ centers catalyze the intramolecular H-shift to fructose similarly to the reaction catalyzed by D-xylose isomerase metalloenzymes. Using ¹¹⁹Sn MAS NMR spectroscopy and isotopic labelling (²H, ¹³C), we fully characterized the Sn-BEA active site and showed the Sn centers in their “open” configuration ((-SiO)₃-Sn-OH) can be distinguished from those in their “closed” configuration [(-SiO)₄-Sn]. Using DFT calculations, we elucidated the glucose-fructose isomerization mechanism and showed that the “open” framework stannanol (Sn-OH) sites are catalytically more active than the “closed” ones and suggested that group-cooperativity—often encountered in enzymatic catalysis—determines selectivity to fructose, via 1,2-hydride shift, or to mannose (epimerization), via 1,2-carbon shift.

We contributed to the selective formation of 5-hydroxymethylfurfural (HMF) from fructose, with emphasis on understanding why certain organic solvents (e.g., DMSO, neat or as co-solvent) accelerate fructose dehydration and minimize HMF loss to levulinic and formic acids and humins. Using quantum mechanics/molecular mechanics molecular dynamics simulations we showed that rate-limiting elementary steps of the dehydration are profoundly influenced by solvent dynamics and re-organization. We developed the COSMO-SAC modeling platform to predict solvents for reactive extraction to increase the yield of furans during dehydration. We introduced reactive adsorption and suitable adsorbents as an alternative to reactive extraction and showed that this can be a competitive process at lower temperatures. Finally, we developed the first kinetic models for dehydration of fructose to HMF and byproducts and for glucose isomerization to fructose and used these to optimize operating conditions.

We employed speciation modeling, ab initio molecular dynamics and a suit of experimental methods to elucidate the mechanism of HMF formation in the glucose-fructose and xylose-xylulose isomerization using metal salts as homogenous Lewis acid catalysts. We showed that the homogeneous aldol-ketose isomerization mechanism is similar to that over Sn-BEA. Hydrolysis of the salt produces Brønsted acidity that drives fructose dehydration to HMF and further to formic and levulinic acids.

We developed methodologies that enable the synthesis of hierarchical zeolites. We introduced two representative hierarchical architectures (self-pillared MFI and 3-dimensionally-ordered-mesoporous-imprinted (3DOm-i) zeolites) with nanometer-sized microporous domains accessible through larger pores, ensuring coexistence of micropore activity and selectivity with fast mass transport. These hierarchical zeolites can be synthesized in pure silica or aluminosilicate form and now with metals (e.g., Sn) incorporated in the frameworks.

We achieved moderate-temperature (around 100 °C) retro-aldol reactions of various hexoses in aqueous and alcoholic media with catalysts traditionally known for their capacity to catalyze 1,2-intramolecular carbon shift reactions of aldoses, for example, various molybdenum oxide and molybdate species and alkali-exchanged stannosilicate molecular sieves, paving the road to alkyl lactates from renewables.

Fuels

Research goal: Synthesis of renewable jet fuels from non-food biomass.

Approach: Develop various selective C-C coupling reaction chemistries via a slate of catalysts to build the carbon chain length starting from furans. Scale up production.

Research highlights

We developed a low-temperature, selective and energy efficient strategy for renewable jet fuels from non-food biomass. Low-carbon furans (e.g., methylfuran) and carbonyl compounds are first coupled to high-carbon (C12-C15) furylmethanes with up to 95% yield. Furylmethanes with branched-chain backbones then undergo HDO to jet fuel-range alkanes at high yields (99%). We have discovered that an inexpensive catalyst, improved graphene oxide (IGO), is super-active for C-C coupling of methylfuran and carbonyl compounds at 60 °C, under neat conditions. These energy- and cost-efficient conditions enable the C-C coupling via a hydroxyalkylation/alkylation (HAA) reaction in a scalable process. After C-C coupling, the furylmethanes undergo hydroxydehydrogenation over Pd/CHf(OTf)4 and Ir ReOx/SiO2 catalysts. The acid sites catalyze ring opening of saturated or unsaturated furylmethanes, enabling lower reaction temperatures. Inexpensive metal/metal oxide catalysts are being exploited and the fundamentals of catalyst active sites, mechanisms, and kinetics in hydrodeoxygenation are being developed.