Catalytic hydrodeoxygenation of pyrolysis bio-oil to jet fuel: A review

Abstract

Bio-oil from biomass pyrolysis cannot directly substitute traditional fuel due to compositional deficiencies. Catalytic hydrodeoxygenation (HDO) is the critical and efficient step to upgrade crude bio-oil to high-quality bio-jet fuel by lowering the oxygen content and increasing the heating value. However, the hydrocracking reaction tends to reduce the liquid yield and increase the gas yield, causing carbon loss and producing hydrocarbons with a short carbon-chain. To obtain high-yield bio-jet fuel, the elucidation of the conversion process of biomass catalytic HDO is important in providing guidance for metal catalyst design and optimization of reaction conditions. Considering the complexity of crude bio-oil, this review aimed to investigate the catalytic HDO pathways with model compounds that present typical bio-oil components. First, it provided a comprehensive summary of the impact of physical and electronic structures of both noble and non-noble metals that include monometallic and bimetallic supported catalysts on regulating the conversion pathways and resulting product selectivity. The subsequent first principle calculations further corroborated reaction pathways of model compounds in atom-level on different catalyst surfaces with the experiments above and illustrated the favored C–O/C=O scission orders thermodynamically and kinetically. Then, it discussed hydrogenation effects of different H-donors (such as hydrogen and methane) and catalysts deactivation for economical and industrial consideration. Based on the descriptions above and recent researches, it also elaborated on catalytic HDO of biomass and bio-oil with multi-functional catalysts. Finally, it presented the challenges and future prospective of biomass catalytic HDO.

Graphical abstract

Keywords

biomass pyrolysis oil / bio-jet fuel / catalytic hydrodeoxygenation (HDO) / metal catalyst / reaction pathways

1 Introduction

In recent years, there has been a gradual increase in global consumption of fossil energy and carbon emission. In 2019, the global energy consumption amounted to 418 exajoules (EJ). Among the various energy sources, coal constituted 9.5%, while natural gas and oil accounted for 16.4% and 40.4%, respectively. Based on the projected policy scenarios, it is anticipated that the energy consumption will escalate to 516 EJ by 2040 [1]. According to the projections by the International Energy Agency (IEA), there will be an increasing supply for oil as the only fossil energy source from 2030 to 2050, as stated in the STEPS report. Furthermore, it is estimated that by 2050, 55% of the global oil consumption will be attributed to petrochemicals [2]. Nevertheless, it is important to note that developing regions generally possess lower capacities for emissions reduction compared to their developed counterparts. This disparity poses a significant challenge in achieving the long-term goal of limiting global warming to the 1.5 °C target set by the Intergovernmental Panel on Climate Change (IPCC) [3–5]. The utilization of fossil energy is expected to persistently rise in order to facilitate the advancement of developing economies, particularly in Asia [6,7]. Therefore, the adoption and exploitation of renewable energy sources have emerged as the predominant approach to substitute fossil fuels [8].

As a form of sustainable energy, the utilization of biomass as a primary resource is a net-zero-emission process. This characteristic promotes a transition toward low-carbon energy and plays a significant role in achieving the objectives outlined in the Paris Climate Agreement [9–12]. Currently, the utilization of hydrogen and electricity is prevalent. However, their application requires the reconstruction of existing facilities. Furthermore, replacing the energy supply for aviation, shipping, long-distance highway, and other transportation sectors with electricity or hydrogen remains challenging [13]. It is worth noting that the global aviation industry is experiencing a rapid growth. It is estimated by the International Air Transport Association (IATA) forecast [14] that by 2050 more than 10 billion passengers will travel approximately 22 trillion km by air each year, and without any further advancements in technology, fuels, or operations, this activity is projected to generate nearly 200 billion tons of CO₂ emission. Unlike electrified decarbonization options for road transport, which face limitations such as battery energy density and aviation safety, the path of the air transport industry to reducing emissions will have to rely on sustainable aviation fuels. Among renewable energy sources, biomass is the only one that can be converted into carbon-containing liquid fuels mainly through thermochemical conversion, which includes processes such as pyrolysis and hydrothermal treatment. Biomass energy is a plentiful resource derived from various sources, such as agricultural and forestry crops and residues [15], organic-rich wastewater, and animal excreta, among others. Various methods are employed to utilize biomass, including direct combustion [16,17], gasification [18,19], liquefaction [20–22], and emerging conversions such as the photocatalytic approach [23].

Fast pyrolysis is a widespread process to produce bio-oil from lignocellulosic biomass. Crude bio-oil is a viscous and stinky mixture with an opaque appearance mainly consisting of C3–C5 oxygenates derived from cellulose and hemicellulose such as acids, ketones, furans, alcohols, aldehydes, and various phenols derived from lignin. High viscosity, corrosion, instability and merely half higher heating value (HHV) of traditional fuel caused by high oxygen and water content make it incompatible with current transportation engineering [24]. Besides, it is important to note that the hydrocarbons comprising jet fuel typically consist of alkanes and aromatic hydrocarbons with a carbon number ranging from 8 to 16 while the carbon number of molecules in bio-oil is far fewer [25]. Therefore, both physical and chemical properties of crude bio-oil indicate that it cannot directly substitute jet fuel.

Catalytic hydrodeoxygenation (HDO) is an effective and promising process to upgrade crude bio-oil to closely resemble jet fuel. It can take place in the same reactor as hydropyrolysis or can be conducted sequentially. The incorporation of hydrogen and appropriate catalysts provides adequate hydrogen radicals that compensate for the deficiency of low H/C (hydrogen to carbon ratio) in biomass, thus inhibiting the formation of large oxygen-containing molecules and polycyclic aromatics, improving the C–C coupling, and finally enhancing the production of C8–C16 hydrocarbons. HDO can be categorized into direct deoxygenation (DDO) and hydrogenation (HYD). With different metal catalysts, hydrocracking (HCR), demethylation (DME), and demethoxylation (DMO), decarbonylation and decarboxylation (DCO) reactions also occur simultaneously with HDO. The influencing factors of HDO have been examined [13,26]. Gea et al. [27] provided a comprehensive analysis of the influence of working conditions, such as catalyst acidity, pressure, temperature, types of solvents, and residence time, on the HDO process. The catalysts commonly employed for the process of HDO encompass a range of materials such as metal oxides (e.g., alumina, zirconia, and titanium oxide), noble metals (e.g., Pt, Pd, and Ru), and non-noble metals (e.g., Ni, Mo, Fe, Cu, and Co) supported on acidic carriers (e.g., ZSM-5, HY, and MCM-41) [28]. Although catalytic HDO could optimize the bio-oil composition, the HCR reaction tends to reduce the yield of the liquid phase and increase the gas production. To deal with this deficiency and transform crude bio-oil to bio-jet fuel ideally, the illustration of reaction pathways is important to understand and further guide the designation of catalysts and selection of reaction conditions. Either the catalytic HDO on pyrolysis vapor or condensed bio-oil is complex because of the various components. Thus, representative model compounds are chosen to demonstrate the fundamental reaction mechanism of the typical oxygen functionalities in bio-oil. The primary model compounds for cellulose and hemicellulose derived oxygenates are some acids, alcohols, ketones (e.g. acetone, cyclopentanone, cyclohexanone, and 2-pentanone), furan and its derivatives (furfural, 5-methel furfural, and 5-hydroxymethylfurfural). Lignin derived phenols mixture and the main model compounds include guaiacol (GUA), eugenol, anisole, o-cresol, catechol and vanillin (VAN) [29–31].

There have been reviews on the catalytic HYD of biomass. However, a comprehensive summary of catalytic HDO pathways combining both experimental investigations and theoretical calculations is currently absent. As shown in Fig.1, this review aims to investigate the reaction pathways of biomass-derived oxygenates hydrodeoxygenation over multifunctional metal catalysts for aliphatic and aromatic hydrocarbons. It presents the impact of commonly used noble and non-noble metal-loaded mono and bimetallic catalysts on the depolymerization of phenols, furfural, and other biomass compounds. The catalysts are supported on materials such as zeolites, activated carbon, and metal oxides (Section 2). In addition, it combines the calculations based on density functional theory to infer the conversion mechanism of the model compounds by examining the adsorption configurations on various catalyst surfaces, the strength of adsorption energies, the dissociation energies of C–O and C=O bonds, and the reaction energy (Section 3). Moreover, it provides a summary of the role of various hydrogen sources such as hydrogen, methane, and alcohols, and their role in increasing the H/C and calorific value of the resulting products (Section 4) and evaluates catalysts deactivation (Section 5). Furthermore, it reviews catalytic HDO on actual biomass and condensed bio-oil with different metal catalysts (Section 6). It aims to present guidelines for bio-oil upgradation to bio-jet fuel and further catalytic HDO industrialization.

2 Catalytic HDO mechanism based on metal catalysts

Catalysts play a crucial role in regulating reaction pathways and product selectivity through the interaction between reactants and the surfaces of the catalysts. The catalytic reaction is significantly influenced by various factors such as the physical properties of the catalysts such as particle size, the surface morphology, the pore structure, and the degree of dispersion of the metal particles on the surface. Chemical properties mainly relate to acidity of supports such as metal oxides and solid acids that exhibit acid active sites which are distributed within the internal pores and external surfaces. These acid active sites can be categorized into Lewis acid sites and Brønsted acid ones [11,32,33]. The adsorption of oxygenates on the aforementioned active sites leads to various reactions, including cleavage, decarbonylation, decarboxylation, isomerization, and aromatization. The appropriate pore sizes can enhance the selectivity of desired hydrocarbons. Metal-loaded bifunctional catalysts exhibit a synergistic effect between the metal sites and acidic sites on the support. The incorporation of metal in the catalyst can effectively reduce the energy barrier for hydrogen dissociation and generate a significant amount of hydrogen radicals for the HDO process. On the other hand, the support exhibits a strong interaction with the metal particles, preventing their aggregation during the reaction and subsequent deactivation. Bimetallic catalysts have a synergistic effect between the two metals, where the presence of an auxiliary metal enhances the catalytic activity, selectivity, and deactivation resistance of the catalysts.

2.1 Single metal modified catalysts

2.1.1 Noble metal catalysts

Noble metals have exhibited exceptional HYD capabilities in biomass HDO. As summarized in Tab.1, despite their higher cost, noble metals such as Pt, Pd, and Ru are highly reusable and possess the ability to catalyze the effective removal of oxygen from the reactants. Metal oxides (e.g., ZrO₂, Al₂O₃, and TiO₂) and solid acids (e.g., SBA-15 and HZSM) are commonly utilized as support. Pd-based catalysts tend to improve the C–C coupling, while Pt- and Ru-based ones are more effective in cleaving C–C and C–O. The interaction between the metal and the support, known as the strong metal-support interaction (SMSI), plays a crucial role in achieving efficient HDO.

Noble metal Pt is widely utilized to remove the oxygen group of biomass derived oxygenates. VAN, a compound derived from the pyrolysis of lignin, possesses three distinct functional groups: aldehyde, methoxy, and hydroxyl. It is frequently employed as a moderator in studies investigating the HDO pathway. A synergistic interaction occurs between the reduction of surface-loaded Pt to the 0-valent state, and the presence of oxygen vacancies in the WO_3–x supports Pt/WO_3–x catalysts [34]. The VAN HDO pathway that includes initial cleavage of the C=O bond results in the deoxygenation of VAN to produce creosol. Additionally, under conditions of low hydrogen content, VAN tends to undergo decarbonylation, leading to the formation of GUA. The main mechanism for the production of methylcyclohexane (MCH) involves aryl ring HYD and C–O bond breakage. The presence of a partially hydrogenated aryl ring reduces the energy barrier for subsequent C–O bond breaking. As a result, the aryl ring of creosol undergoes preferential aryl ring HYD to generate COL (COL), which is then deoxygenated to obtain MCH. The WO_3–x material exhibits oxophilicity and its tunable oxygen vacancies contribute to stabilizing the metal particles that are loaded onto it, as well as facilitating the adsorption of oxygenated compounds. On the other hand, Pt primarily facilitates the dissociation and adsorption of hydrogen, thereby generating hydrogen radicals that are essential for HDO reaction.

Different supports loaded with the same metal yield vary HDO effects. Pd, being one of the extensively utilized noble metals, was employed as a catalyst on various supports including SiO₂, CeO₂, ZrO₂, TiO₂, and Nb₂O₅ for GUA HDO [36]. The selectivity of aromatic and alkane products was influenced by the dispersion of metal particles and the density of acid sites. The conversion pathways of GUA are illustrated in Fig.2(a). GUA undergoes sequential decreases in bond energies of C_Ar–OH, C_Ar–OCH₃, and C_ArO–CH₃, leading to its initial DME and subsequent production of catechol during the early stage of HDO. The hydrotreatments of Pd/SiO₂ and Pd/CeO₂ exhibit a greater significance, resulting in the production of cyclohexanone through HYD of intermediates. Subsequently, cyclohexanes obtained through further HDO. ZrO₂, TiO₂, and Nb₂O₅, on the other hand, exhibit a high affinity for the carbonyl group and convert it into a hydroxyl group because of oxophilic nature. The C–O bond energy can be reduced to various extents by utilizing specific metals and supports. In particular, the combination of Ru and CeO₂ can promote the cleavage of the C_Ar–OCH₃ bond [38]. For phenolics derived from lignin, HDO primarily involves the HYD of aryl rings and the breaking of C–O bonds to achieve saturated aryl rings. Additionally, further ring-opening reactions can lead to long-chain alkanes.

Oxygenated compounds, such as furfural and cyclopentanone, which are derived from the cellulose and hemicellulose, have the ability to produce bicyclic and tricyclic alkanes through the process of HDO using metal-acid catalysts. Recently, bifunctional nanoparticle catalysts encapsulating noble metal particles enclosed within zeolite were developed. By strategically positioning the active sites and controlling the diffusion of reactants and intermediates on the catalyst surface, researchers were able to convert cyclopentanone to the desired pathway, as shown in Fig.2(b) [43,44]. The Pt@H-BEA catalyst, consisting of Pt encapsulated within zeolite H-BEA, demonstrated the ability to catalyze the synthesis of C10 cycloalkanes (specifically bicyclopentane and decahydronaphthalene) from cyclopentanone through aldol condensation, HDO, and ring rearrangement reactions with an overall yield of 78% [43]. Encapsulation of Pt within the zeolite serves to prevent the HYD of cyclopentanone by contacting the metal sites first. Additionally, the presence of micropores in the zeolite enables the control over the selectivity of the resulting products. If only Pt is loaded onto the surface of BEA or HZSM-5, the predominant product formed is cyclopentane. Thus, the pore size of the zeolite skeleton encapsulating the Pt nanoparticles is decisive for the product distribution, while the acid-metal interface exhibits a synergistic catalytic effect. Pd/C and a range of acid catalysts could transfer 5-methylfuran (5-MF) or 5-hydroxymethylfurfural (5-HMF) into bicycloalkanes [39]. This transformation occurs through a three-step process, which involves a ring-arrangement reaction to form an intermediate known as 1,2,4-benzenetriol, followed by a C–C oxidative coupling reaction of the intermediate to form a dimer, and finally, the HDO of the dimer to produce alkanes.

2.1.2 Non-noble metal catalysts

Compared to noble metals, non-noble metals offer a more cost-effective alternative. In terms of physical properties, the interaction between the metal precursor solution and the support during the preparation process, working conditions, and the regeneration conditions, significantly impacts the dispersion and positioning of metal particles on the support. Chemical properties including quantity and strength of acidic sites on the metal and the support, as well as the interaction between the electronic structure of the metal when combined with different supports, also contribute to varying catalytic effects [45,46]. As listed in Tab.2, the non-noble metals that have been extensively studied include Co, Cu, Fe, and Ni. These metals are commonly supported by metal oxides or zeolites such as Al₂O₃, ZSM-5, TiO₂, ZrO₂, CeO₂, and SBA-15. Ni and Co metals usually promote HYD while other metals like Mo tend to enhance deoxygenation.

For instance, in the hydropyrolysis of pine pyrolysis gases, catalysts of Fe-loaded coke (Fe/CC-s and Fe/CC-n, where s and n represent the metal precursors iron sulfate and iron nitrate, respectively) were employed. It was observed that Fe/CC-n promoted the methanation, resulting in gas-phase products of 29.9%. On the other hand, the activity of Fe/CC-s was comparatively moderate, displaying a higher selectivity toward bio-oils, and aromatic hydrocarbons reached 78.55%–93.15% [56]. HDO of lignin was conducted using molecular sieves HBeta and metal oxides ZrO₂ as the supports (Fe/HBeta and Fe/ZrO₂, respectively) at 350 °C and 1 atm. Fe/ZrO₂ exhibited a superior HDO performance due to its mesoporous structure and moderate acid strength, resulting in an improved resistance to carbon accumulation [50]. The Fe/ZSM-5 catalyst was employed to exploit the synergistic effect between metal and solid acid for HDO of biomass residue and enabled the production of bio-oils with an HHV of 31.74 MJ/kg and a selectivity of aliphatic compounds of up to 72.96% [57]. Co and Fe are both Group VIII metals and exhibit similar chemical properties. However, in comparison to Fe, Co demonstrates a more pronounced HYD effect. The Co-Al₂O₃@USY catalyst incorporated highly dispersed Co nanoparticles along with an abundance of acidic sites on the surface of the USY. This unique combination enables the catalyst to facilitate various conversion pathways of GUA, as illustrated in Fig.3. The conversion of GUA to cyclohexane was reported to reach a maximum of 93.6% [58].

Among non-noble metal catalysts, Ni-based metals demonstrate a higher level of activity in HDO reactions, leading to an enhanced selectivity of monocyclic aromatic hydrocarbons. It promotes increased saturation of hydrocarbon products while reducing the formation of solid byproducts [59]. Ni loaded catalysts with various supports including SBA-15, Al-SBA-15, γ-Al₂O₃, microporous carbon, TiO₂, and CeO₂, have been shown to possess the ability to achieve high yields of cyclohexane [60]. Among these catalysts, the highest benzene selectivity (64%) was observed with Ni/C. Mo₂C loaded single-atom Ni catalysts possess a remarkable activity for completely transforming lignin-derived compounds into biofuel molecules. During the conversion of dihydroeugenol, Ni single-atom metal sites optimize the adsorption configuration of reactant from vertical to horizontal, which lowers the energy barrier and increases the selectivity of hydrocarbons [61]. In addition to single-atom catalysts, Ni-based nanoparticles catalysts also have a prime importance. Hu et al. [62], utilizing Ni/ZSM-5 nanoparticles, effectively converted a wide range of intermediates, including ketones, ethers and phenols into (cyclo) alkanes. The physical distribution of Ni sites exerts a significant influence on the selectivity of HDO products of anisole. Ni/BEA prepared by organic solvent has more metal sites located in mesopores, resulting in increasing metal sites accessibility and more high-T desorbed H species, which led to a higher (cyclo)alkanes selectivity, while other two prepared by impregnation and deposition-precipitation methods have more electron transfer from Ni to support because of the stronger metal-support interaction, resulting in a poor HYD activity and a more aromatics selectivity [63].

Supports, such as metal oxides and zeolite molecular sieves, possess unique pore structures, structural defects, and acid active sites. These characteristics enable them to adsorb hydroxyl, methoxy, and other oxygenated functional groups on oxygenated compounds. Additionally, both noble metals and transition metals can effectively facilitate the activation of hydrogen molecules, thereby generating a significant number of hydrogen free radicals for the hydrogen deoxygenation reaction. Furthermore, these metals promote the activation and cleavage of C–O and C=O bonds. Synergistic effects arising from the interaction between metal active sites and acid active sites can further increase the catalytic selectivity, catalytic activity, and stability of the catalyst.

2.2 Bimetallic catalysts

Compared with monometallic catalysts, the distribution of metal sites and the electronic structure of bimetallic catalysts have significant effects on the catalytic effect and reaction pathways. Structures such as core-shell, homogeneous alloy, single atom and metal-organic framework are widely investigated in HDO [61,64–66]. For example, the insertion of the metals Pt and Nb into the TiO₂ matrix increases the acid sites of Lewis and Brønsted acids, respectively, while the presence of Nb leads to more oxidized Pt species, producing more stable metal centers, which realized the promotion of C–O bond scission during acetone HDO [67].

2.2.1 Noble metal-based catalysts

The inclusion of a second metal can lower the expenses of noble metal and enhance the catalysts activity, which optimizes the dispersion of metal particles on the support surface and facilitates the reduction of second metal, as indicated in Tab.3. The hydrogen spillover from noble metals to other enhances HDO and the yield of hydrocarbons increases remarkably compared with monometallic catalysts. Noble metals combined with Ni enhance H₂ activation and produce more saturated aliphatics and those combined with other oxophilic metals produce more hydrocarbons.

The Ru/HY catalysts enhanced HDO conversion of lignin and the derived compounds with the introduction of metal (Fe, Ni, Cu, and Zn), mainly due to the increased acid sites, superior distribution of metal species and limited aggregation, and strengthened adsorption of oxygen groups. The second metal tuned the hydrogenolysis activity of Ru. The total mass of hydrocarbons reached 26–32 wt.% with RuCu/HY [31]. The Ru/Zn@Beta catalyst, which combined metal sites and acid active sites, was utilized to depolymerize Kraft lignin, resulting in a liquid phase product with an HHV of 30.39 MJ/kg [68]. The depolymerization pathway is illustrated in Fig.4(a), where Ru primarily activated hydrogen to produce a hydroxyl radical, while Zn adsorbed and weakened the C–O bond. The subsequent transfer of H to the vicinity of the adsorbed substrate facilitated the HDO process. The bimetallic PdRu/HAP catalyst was employed to carry out the HDO of C15 oxygenated compounds derived from lignocellulose. Under appropriate operating conditions, the yield of alkane products ranging from C8 to C15 was 84.24%. The incorporation of Pd into the catalyst system supplied additional electrons to the metal sites of Ru, further enhancing the HDO catalytic activity [69]. The potential reaction pathways are shown in Fig.4(c).

Noble metal Pd combined with Mo catalysts improved the conversion of anisole from 17% to 90% and the selectivity of HDO. Introduction of Mo modulated the crystalline, electronic state, redox properties, and adsorption behavior of Pd metal sites and thus altered the reaction pathway of anisole from HYD to C–O fracture (Fig.4(b)) [79]. The catalyst Pd–Zr/HZSM-5 exhibited an alloy structure that promoted the conversion of Pd–O–Zr solid solution to PdO–ZrO₂ metal oxide structure. The resulting oxygen vacancies and the surface charge of the bimetallic alloy particles contributed to the improved catalytic activity [80]. While combined with HYD metals, HYD is improved. The presence of Ru was found to promote the hydrogenolysis reaction. The combination of Ru and Pt significantly facilitated the conversion of the lignin-derived bio-oil compound 4-propylphenol, resulting in a maximum yield of propylcyclohexane of 100% [81]. Mortensen et al. [82] conducted a study on the preparation of hierarchical zeolites as supports to enhance the distribution of Pt–Ni bimetallic catalysts. The researchers aimed to encapsulate the metal within the internal pores of the supports to promote hydrogen activation. Anisole was successfully converted with Pt–Ni through DMO, dehydration, and further HYD. Additionally, a pathway involving, dehydration, deoxidation, and HYD was observed during the conversion process. The conversion rate reached 100%, resulting in a yield of cyclohexane in the product of 78.5 wt.% [84].

Bimetallic catalysts, which combine hydrotreating metals with oxophilic metals, exhibit a superior catalytic activity and selectivity for HDO compared to monometallic catalysts. Oxophilic metals, including Fe, Mo, Co, and W, tend to form metal-oxygen bonds when bonded with oxygen. On the other hand, noble metals such as Pd, Pt, and Ru activate hydrogen at their sites, resulting in the generation of hydrogen radicals. The hydrogen radicals are then transferred to the oxophilic metal sites due to the hydrogen spillover effect (HSPE). This process promotes the breaking of the C–O bond, which has a positive effect on the HDO reaction. Meanwhile, the electron gain/loss characteristics of bimetallic catalysts alter the electronic configurations of the catalysts. The electric effect additionally enhances the noble metal activity.

2.2.2 Non-noble bimetallic catalysts

Non-noble metal bimetallic catalysts exhibit a greater catalytic activity, selectivity, and stability for HDO process of biomass and its depolymerized oxygenates, compared to monometallic catalysts. Tab.4 demonstrates this trend, and the most commonly studied bimetallic catalysts include Ni–Mo, Ni–Fe, Cu–Ni, Ni–Co, Co–Fe, and others. Ni and Co are usually used as primary metals for their HYD effect and supplemented by the oxophilic metals such as Mo and Fe.

Biomass can be converted into liquid alkanes through hydropyrolysis by NiMo/HZSM-5 [83], Ni primarily facilitates HCR, and hydrogen saturation, promoting the production of CH₄, CO, and CO₂. However, excessive cracking occurs when only Ni is present. The addition of Mo enables methane oligomerization, thereby increasing the yield of liquid-phase products. Li et al. [84] investigated the catalytic effect of different NiMo bimetallic ratios on lignin hydropyrolysis. They found that Ni and Mo have an obvious synergistic effect, promoting phenolic methoxide removal and increasing hydrocarbon carbon yield by up to 25.82 C%. In contrast, Ni alone only facilitated phenolic hydroxyl removal and benzene ring hydrogen saturation reaction. Stummann et al. [85] performed biomass hydropyrolysis in a fluidized bed and discovered that the catalytic activity, carbon build-up resistance, and stability of CoMo/MgAl₂O₄ and NiMo/zeolite-Al₂O₃ hydropyrolysis were enhanced by loading them with bimetallics, as compared to the supports. The catalytic activities of bimetallic FeMo and NiMo were similar, and these bimetallics could promote C–C and C–O bond cleavage, enhance DCO, aromatization, and Diels-Alder reactions more effectively than Mo/ZSM-5 [86]. The transformation process is illustrated in Fig.5(a).

The addition of auxiliary metals alters the electronic structure of the catalyst, adjusts the original acidic strength, and modifies the number of oxygen vacancies in the support. This can effectively improve the HDO process. The catalysts Ni/ZrO₂, NiFe/ZrO₂, and NiCo/ZrO₂ [97] were prepared by loading Fe and Co onto ZrO₂. The reaction process of cresol on the surface of the catalysts is illustrated in Fig.5(b). The conversion of cresol was increased to 98% by using MeO–O_v–ZrO³⁺ on the ZrO₂ surface. This catalyst was able to adsorb oxygenated functional groups and activate the C–O bond. Hydrogen contacted with the Ni metal sites dissociated into hydroxyl radicals, which promoted aryl-ring HYD and C–O bond breaking. The introduction of reactive metals increased the oxygen vacancy concentration of the support and facilitated the adsorption of cresol. Bimetallic nickel-based catalysts (NiM@C, M=Co, Mo, and La) [98] showed a better conversion of COL, the target product, into lignin-derived dimers and monophenols compared to the conversion of COL using the single-metal Ni/C catalyst. The addition of a second metal that interacted synergistically with Ni enhanced the Brønsted acidity of the catalyst and promoted the DMO and HYD of phenol and GUA. The presence of two methoxyl groups on eugenol complicated the HDO process due to the site-blocking effect. The bimetallic catalysts were effective in promoting the cleavage of different C–O bonds, including β–O–4, α–O–4, and 4–O–5, which resulted in an enhanced HYD of the aryl ring and increased production of cycloalkanes. The NiMo/γ-Al₂O₃ in the conversion process of three-component model compounds derived from biomass was presented in Ref. [88]. The transformation pathways are shown in Fig.5(c), and furfural, 5-MFF, and 5-HMF are representative derivatives of cellulose and hemicellulose with various branched functional groups. GUA and VAN were selected as representatives of lignin, while benzaldehyde (BA) was chosen to further investigate the role of aldehyde functional groups. Model compounds were initially hydrocracked to generate small molecule oxygenated compounds such as furans, acids, and ketones in the HDO process, and then aromatic hydrocarbons were generated through HCR, HDO, isomerization, HYD, and alkylation in HDO. Additionally, some of the long-chain aliphatic compounds in the products were formed through hydrotreating, ring-opening, and dehydrating reactions, with the carbonyl group serving as the guiding moiety. Cellulose and hemicellulose were initially broken down into furan and ketone compounds. HDO primarily produced C1-7 compounds, including small molecule alkanes. The five-carbon ring could generate aromatic hydrocarbon products through the Diels-Alder reaction. Furan products underwent intermolecular coupling reactions facilitated by the carbonyl group, ultimately resulting in the formation of C8+ long-chain aliphatic hydrocarbons. Lignin mainly cleaved to obtain phenolic products that contain methoxy, phenolic hydroxyl, and propyl chains. Alkylation and isomerization in the protonation section were the main processes that generate monocyclic and polycyclic aromatic hydrocarbons.

The interaction between the support and the metal in a bimetallic catalyst is stronger compared to a single metal catalyst. The addition of the active metal increases oxygen vacancies, acidic active centers, and metal active centers on the support surface. This leads to changes in the adsorption configurations of oxygenated compounds and facilitates the breaking of the C–O bond, promoting HYD and hydrogen deoxygenation. Additionally, the introduction of the second metal reduces the size of the metal particles on the support surface, improves the uniformity in the distribution of metal active sites, and enhances the catalytic activity and stability of the catalyst.

3 Theoretical computation of reaction pathways investigation

It is challenging to experimentally clarify the specific reaction process and trend of oxygenated compounds on the surface of the catalyst. First-principle calculations based on the density functional theory can facilitate the investigations of catalytic pyrolysis reactions in biomass by calculating the energy changes at the atomic level. These calculations explored the adsorption properties of oxygenated functional groups on the active site, the energy barriers for breaking C–O bonds, and the changes in the electronic structure of the catalysts. This section examines the changes in reaction energy during the pyrolysis of lignin-derived phenolics, cellulosic and hemicellulose-derived furans, and other oxygen-containing compounds in the presence of metal catalysts.

3.1 Reaction paths for lignin-derived oxygenates

Theoretical studies on the catalytic conversion of lignin pyrolysis products primarily focus on common phenolic compounds like cresol, GUA, and anisole. Theoretical computational studies on lignin-derived phenolic HDO are summarized in Tab.5. The modal reaction path is affected by a variety of factors such as catalyst type, and structure. Noble metals facilitate ring saturation and non-noble metals promote C–O bond cleavage. Miao et al. [88] conducted quantum chemical calculations using GUA, VAN, and BA as model compounds for lignin depolymerization to investigate the ease of bond breaking of methoxy, hydroxy, and carbonyl groups during the HDO reaction over NiMo/γ-Al₂O₃ catalysts. The ease of bond breaking during the HDO reaction can be determined by the BO and f(0) values (x = maximum, n = minimum), which represent the order of bond breaking and free radical reaction sites. Fig.6(a) shows that the order of BO_x for the three is VAN > BA > GUA. Therefore, VAN is the first to undergo HDO reaction, followed by BA and GUA. Additionally, based on the f(0) values, aldehydes are most susceptible to free radical attack, leading to coupling and alkylation reactions that increase the carbon chain length. Methoxy and hydroxy groups, on the other hand, are less active [99].

The transformation pathways of phenols on catalyst surfaces with varying degrees of HYD and oxophilicity differ. The HDO of phenols is significantly influenced by three commonly used metals: Pt, Ru, and Fe. The mechanism of methoxide removal by these metals was investigated taking anisole as an example. The activation energies of C_aryl–O, C_alkyl–O, and C–H bonding, as well as the reactive energies, were calculated for the surfaces of Pt(111), Ru(0001), and Fe(110) on the anisole surface of anisole [113]. Anisole (Ph–O–CH₃) is adsorbed on the surface of Pt (111). The activation energies required for direct DMO, DME, and removal of hydrogen on one methoxide were 250, 200, and 125 kJ/mol, respectively. The activation energies required for the removal of the second hydrogen after removal of one hydrogen on methoxide (Ph–O–CH₂) were lower than those for DMO and DME. After the two-step dehydrogenation (Ph–O–CH), the activation energy of C_alkyl–O was lower than that of C–H and Caryl–O. This was attributed to the fact that the removal of hydrogen on the methyl group in the previous stage promoted the dissociation of C_alkyl–O [114]. Consistent with the Pt surface reaction, the initial step of anisole deoxygenation on both Ru and Fe surfaces involves the dehydrogenation of methyl to form Ph–O–CH₂. However, the activation energy for further dehydrogenation of Ph–O–CH₂ to form Ph–O–CH on the Ru(0001) surface is 14 kJ/mol, which is similar to that of C_alkyl–O (24 kJ/mol). On the other hand, on the Fe(110) surface, the activation energy for further dehydrogenation of Ph–O–CH₂ (32 kJ/mol) is higher than that of C_alkyl–O (12 kJ/mol). The reason for this is that the oxophilicity of Ru and Fe is higher than that of Pt, and the former two are more prone to breaking the C–O bond. In particular, Fe exhibits the strongest oxophilicity, and the DME of anisole produces phenoxy intermediates. This intermediate can undergo either HYD or deoxygenation to yield phenol or phenyl, as illustrated in Fig.6(b).

The stronger HYD ability of Pt leads to the predominance of HYD, while on the highly oxophilic Fe surface, the cleavage of the C–O bond is the main reaction. This is consistent with experimental results showing that phenol and benzene are the main products in Pt- and Ru-catalyzed reactions, while Fe-catalyzed HDO of anisole only yields benzene. The affinity of the metal for oxygen significantly influences the conversion pathway of oxygenated compounds in the HDO process, thereby altering the composition of the reaction products. The strength of the C–M bond is also influenced by the oxopilicity. The team conducted calculations on the m-cresol HDO pathway on Pt(111) and Ru(0001) surfaces [115]. They discovered that the adsorption energies of both the reactant m-cresol and the product benzene were higher on the Ru surface compared to Pt. This indicates that the C–M bonding energy is higher on the Ru surface. On the Pt surface, the energy barrier for DDO of m-cresol was found to be as high as 242 kJ/mol. However, the energy barriers for the formation of keto tautomer intermediate, followed by hydrodehydration to produce toluene or HYD to obtain 3-methylcyclohexanone, were lower. M-cresol was dehydroxylated and hydrogenated on the Ru surface to produce toluene. The Ru surface facilitates the formation of C1–C5 small molecule hydrocarbon products through the C–C rupture of C₇H₇* intermediates. The reaction path is illustrated in Fig.7(a).

Based on Refs. [116,117], there are three potential pathways for GUA HDO as illustrated in Fig.7(b). In Path I, GUA undergoes dehydration to obtain 2-methoxy-cyclohexanol (MOCOL), which is then demethylated to obtain COL. Finally, COL is dehydrated to yield cyclohexane. In Path II, GUA undergoes DMO followed by HYD, saturation, and dehydroxylation to obtain CHA. The saturation and dehydroxylation reactions of the Path III pathway are the reverses of Path II, in that the bond energies of Caryl–OH (468 kJ/mol) and C_aryl–OCH₃ (422 kJ/mol) are higher than those of C_alkyl–OH (385 kJ/mol) and C_alkyl–OCH₃ (339 kJ/mol) for GUA [118]. The hydrogen saturation of the aryl ring is prioritized to facilitate the subsequent HDO process. In the three pathways, methoxy is removed before hydroxyl, and the DMO of MOCOL and the dehydroxylation reaction of COL are rate-limiting steps. The conversion pathways and product distributions of GUA can be altered by adjusting the oxophilicity of the catalyst. Additionally, the interaction between the metal sites of bimetallic catalysts can also impact the process. The oxygen vacancies in the bimetallic Ni–Mo catalyst can be adjusted by changing the Mo loading. The adsorption energies of GUA and its HDO intermediates, COL and MOCOL, were calculated on the surfaces of Ni (111), NiMo (111), and MoO_x-Ni (111) [93], as depicted in Fig.7(c). Calculations showed that the benzene ring of GUA adsorbed parallel to the metal surface, with the C–O bond lengthening. The adsorption energies on the three surfaces were 2.45, 2.68, and 2.77 eV, respectively. After hydrogen saturation, the adsorption energies of the benzene ring decreased. The intermediate product, MOCOL, had the longest bond lengths and strongest adsorption energies on the MoO_x-Ni (111) surfaces, with the oxygen vacancy introduced by MoO_x promoting the DMO reaction. The other product, COL, had the longest bond on the surface of NiMo (111). These findings, combined with catalyst characterization results, suggest that the Ni–Mo alloy facilitates hydroxyl removal.

Carbonyl groups, along with methoxy and hydroxyl groups, are commonly found oxygenated functional groups in phenolic branched chains. The addition of CoO_x to CoNi bimetallic catalysts creates oxygen vacancies, altering the electronic properties of the catalyst surface and influencing the VAN HDO reaction pathway [90]. The adsorption of VAN, MMP and 4-MP on Co (101), Ni (111), and CoNi-111-Co₆O₇-Ov surfaces was calculated, as shown in Fig.7(d). The phenolic hydroxyl group, methoxy group, and oxygen atom of the carbonyl group of VAN were found to be adsorbed on the Co surface. Oxygen and carbon atoms from the carbonyl groups of VAN adsorbed onto the Ni sites. The most stable adsorption occurred on the surface of CoNi-111-Co₆O₇-O_v (−3.87 eV), where the carbonyl carbons were positioned on the oxygen vacancies of CoO_x. This arrangement facilitated the activation of C=O and the DDO process, leading to the formation of MMP. MMP undergoes DMO on the Co surface and forms oxygen vacancies, resulting in the formation of 4-MP. On the other hand, adsorption on the Ni surface leads to HYD and saturation, resulting in the formation of 4-methyl-2-methoxy-cyclohexanol (MCYL). Fig.7(d) illustrates the HDO pathway of VAN on the surface of CoO_x-decorated CoNi catalysts. In this pathway, oxygen vacancies preferentially adsorb the carbonyl group and activate C=O, converting VAN to MMP intermediates through DDO. The Co metal sites on the catalyst surface subsequently promote the removal of methoxy, resulting in the formation of 4-MP. Finally, 4-MP undergoes HYD to produce MCYL.

Lignin-derived phenols undergo tautomerization on the surfaces of hydrogenated metals like Pt, Pd, and Ni, resulting in the formation of ketone intermediates. These intermediates can then be further converted into hydrocarbons through processes such as HYD and dehydroxylation. Additionally, this type of metal promotes the dissociation of hydrogen, leading to an increased concentration of hydrogen atoms on the reaction surface. This is beneficial for enhancing the HYD of oxygenated compounds. On the other hand, DDO can occur on the surfaces of oxophilic metals like Fe and Mo, breaking the C–O bond. This is attributed to the abundance of oxygen vacancies in their structure and the strong M–O interaction. The use of multifunctional catalysts loaded with two different metals can enhance the HDO of phenols through intermetallic synergistic effects. This allows for the hydrogen saturation of aryl rings, resulting in the production of saturated ketones, alcohols, and other intermediates. The order of the HDO reaction of phenols varies depending on the oxopilicity and acidity of the catalyst, particularly in the presence of oxygenated groups like methoxy and hydroxyl. In this case, the hydrogen on the methoxy group is preferentially removed, which simultaneously reduces the activation of the C_aryl–O bond, leading to the removal of the methoxy group. Subsequently, the hydroxyl group will be eliminated through dehydration, resulting in complete deoxygenation.

3.2 Reaction paths for cellulose- and hemicellulose-derived oxygenates

Typical model compounds for depolymerization of cellulose and hemicellulose include FF, MFF, and HMF. These compounds have five-membered rings, which allow for a more thorough investigation of the HDO pathway. Tab.6 presents theoretical computational studies on the HDO process of oxygenated compounds derived from cellulose and hemicellulose. The main rate-determining step is the HYD of carbonyl group. H is more adsorbed on noble metals and vacancies generated on oxophilic metals tend to adsorb oxygen-containing group. Miao et al. [88] analyzed the bond strength of each C–O bond in FF, HMF, and HMF molecules on the active sites of NiMo/γ-Al₂O₃ catalyst. They also examined the ease of reaction with free radicals, such as hydrogen. The results, shown in Fig.8(a), revealed that the C=O bond of the carbonyl group was the most difficult to break, while the oxygen atom was the most easily broken by the carbonyl group and the oxygen atom. The oxygen atom is highly susceptible to attack by free radicals. Among the three modal compounds, HMF is preferentially reacted due to its longer and more branched chain, which is structurally unstable. Additionally, HMF and MMF are more likely to form PAHs products through coupling reactions compared to FFA.

The potential pathways for furan HDO are determined through calculations of reactions involving C–O and C–C bond breaking, hydrogen saturation, and ring opening on the surfaces of various transition metals (Ni, Co, Rh, Ru, Pt, Pd, Fe, and Ir) [119]. The initial stage of furan conversion involves either ring opening or partial HYD. In the ring opening reaction, the breaking of the C–O bond, the energy barriers are highest on Ni and Co surfaces, while they are lowest on Ir and Fe surfaces. Partial HYD enhances ring-opening reactions on the surfaces of Ni, Fe, Rh, and Co, while the energy barriers for these reactions increase on the surfaces of the other four metals. The hydrogenated intermediates were either further hydrogenated to produce alcohols or directly deoxygenated to yield hydrocarbons. Additionally, the alcohols underwent dehydroxylation to generate hydrocarbon products. The oxophilicity of transition metals affects product selectivity. Ir, Pt, Pd, and Rh surfaces tend to produce more linear chain oxygenates, while Ni, Ru, and Fe surfaces have a higher selectivity for fully deoxygenated linear products (Fig.8(b)) [119]. The hydrodeoxygenation products of furfural on Cu, Ni, Pd, Pt, Re, Rh, and Ru are influenced by the properties of the metal [134]. Furfural adsorbs onto metal surfaces in a parallel manner. However, during the adsorption process, the aldehyde group moves away from Pd and Pt surfaces and becomes closer to other metals. The potential transformation pathways of furfural are illustrated in Fig.8(c). On the Pd surface, the initial cycloaddition of furfural to saturation or aldehyde group HYD does not occur in a significant sequential manner. However, on other metal surfaces, the selective HYD of the aldehyde group is followed by deoxygenation as the first step.

The interaction between metals can enhance the HDO reaction. When ReO_x is modified on the Ni surface, hydrogen dissociates on the Ni surface, and furfural binds to the ReO_x surface. Furfural alcohol is then hydrogenated by hydrogen atoms diffusing from the Ni site. The hydrogenated furfural alcohol has a lower adsorption energy on the ReO_x surface and undergoes a HYD reaction on the Ni surface to obtain THFA [130]. The HDO of furfural on Pd(111), Ni/Pd(111) and Ir/Pd(111) catalyst surfaces was compared. The path diagrams and reaction energy barriers are shown in Fig.8(d). It is observed that the activation energy is significantly reduced when using bimetallic catalysts compared to monometallic Pd. The rate-limiting step in the transformation process is the final HYD reaction, C₄H₃O* + H*→C₄H₄O [135]. The atomic arrangement position, in addition to the metal properties, also affects the adsorption conformation of furfural. Furfural adsorbs in a bridging-type manner on the surface of NiMo intermetallic compound catalysts, resulting in the highest 2-MF yield (99%) compared to Ni and NiMo alloy alone [136]. There are multiple pathways for converting furfural to furfuryl alcohol (FA) and 2-MF. On the CuNiCu(111) surface, the HDO conversion primarily occurs through the following pathway [137]. The HYD of furfrual (FF) to FA can result in either F–CH₂O or F–CHOH, with calculations indicating that the former is more kinetically favored. The intermediate product, F–CH₂O, undergoes HYD to form the alcohol F–CH₂OH, which then dehydroxylates and hydroxylates to produce 2-MF (F–CH₃).

5-HMF is a commonly studied depolymerizing oxygenated compound found in biomass. Its HDO pathway has been extensively researched. The potential conversion pathways of 5-HMF to 2,5-DMF on Pd, Cu, and PdCu bimetallic catalysts are shown in Fig.9(a) [138]. The adsorption energy of 2,5-DMF on the Pd(111) is highly favorable (−2.47 eV), suggesting that monometallic Pd catalysts may facilitate the formation of 2,5-dimethyltetrahydrofurfural as the final product. The highest selectivity of bimetallic PdCu for 2,5-DMF products is achieved through low reaction energy barriers, weak adsorption energy of products, and enhanced hydrogen dissociation. After calculating the energy barriers for hydrogenolysis, HYD, and other reactions, it is discovered that on the surfaces of Pd and PdCu catalysts, 5-HMF is first dehydroxylated to produce 5-MF. The aldehyde group of the 5-MF is then hydrogenated to form a hydroxyalkyl intermediate, which is ultimately dehydroxylated to yield 2,5-DMF. On the other hand, the aldehyde group is preferentially hydrogenated on the surface of the Cu catalyst, resulting in the formation of 2,5-bis(hydroxymethyl)furan, which is subsequently dehydroxylated to produce the 2,5-DMF product. When isopropanol is used as a hydrogen source, 5-HMF can undergo esterification by HDO with activated carbon supported Cu catalysts, resulting in the production of ester meso-inspection products. The potential conversion pathways are shown in Fig.9(b) [139].

The main depolymerization products of cellulose and hemicellulose are furan, FF, and 5-HMF. The conversion pathways of HDO are diverse, which can either remove oxygenated functional groups or achieve complete deoxygenation. At the early stage of the reaction, either ring opening or carbonyl HYD occurs. The latter leads to the formation of alcohol intermediates which then undergo dehydration to obtain the final product. Various calculations of bond-breaking energy for C–O and C=O at different molecular positions have shown that the reaction path is influenced by the type of catalyst. For instance, in the case of 5-HMF, metals such as Pd and Pt, which promote H₂ dissociation, tend to remove OH and form H₂O, followed by the formation of 5-MFF and 5-MFA. Other metals, such as Cu and Mo, initially promote the HYD of C=O due to the relatively high barrier for direct formation of H₂O. In the HDO process, the main reactions include cyclic HYD, C=O HYD, ring opening, DCO. Different types of metal catalysts have varying selectivity for the products. The refined products include furan, FA, 5-MF, pentanol, cyclopentanol, THFA, MTHF, and pentanes.

4 Effects of different H-donors on catalytic HDO

In the HDO process of biomass-derived oxygenates, the choice of external hydrogen sources has a significant impact on product selectivity. Factors to consider when selecting H-donors include their activation ability, economic feasibility, and the solubility of reactants in the liquid-phase system. The commonly used sources of hydrogen are pure hydrogen and small molecule alkane gases like methane, 2-propanol, and tetralin.

4.1 Hydrogen

H₂ is the most commonly used hydrogen donor. It provides hydrogen radicals to react with oxygenated compounds, leading to deoxygenation through dehydration. This method reduces carbon loss compared to inert atmospheres that rely on decarboxylation and decarbonylation. Hydrogen radicals can react with biomass pyrolysis products, such as acids and ketones, and inhibit polymerization, thereby reducing char yield [13,88]. As shown in Tab.7, hydrogen spillover happened between noble metals and non-noble metals or supports is the major contributor to the effectiveness of HDO. Among those researches, bimetallic catalysts demonstrate superior deoxygenation results than monometallic ones.

In the HDO of biomass without catalysts, hydrogen is used to generate free radicals that participate in the reaction. However, this process requires relatively harsh conditions and results in significant deoxygenation of crude bio-oil, but the yield of hydrocarbons is not high. The carbon yield of ketones obtained from cellulose hydropyrolysis reached 27.2%. At a pyrolysis temperature of 500 °C and a hydrogen pressure of 3 MPa, levoglucosan and acids were completely converted. Additionally, a small number of aromatics, aliphatic hydrocarbons, and phenols were detected. The addition of hydrogen facilitated the depolymerization and deoxygenation of cellulose macromolecules, resulting in an increase in the H/C of the purified product from 1.4 to 1.9. The depolymerized cellulose product was converted to levoglucosan through the addition of hydrogen. Levoglucosan can undergo DDO, cleavage, and ring-opening reactions to produce acids, furans, and C3,4 ketones. Acids can be converted to aldehydes through HDO, and furan products can be transformed into straight-chain and cyclic aliphatic hydrocarbons through HDO, isomerization, C–C coupling, and ring-opening reactions. Additionally, aromatics can be obtained through the Diels-Alder reaction with cellulose HDO. The potential reaction pathways are shown in Fig.10 [150]. After hydropyrolysis of lignin, there was a significant decrease in the yield of methoxyphenol, while there was an increase in the yields of aromatic compounds and methane. HYD and pressurization can facilitate the depolymerization of lignin and the removal of methoxy groups from the phenolic ring. During the process of HDO, the alkyl group obtained from HYD and C–C bond cleavage of the phenolic branched chain can replace the methoxy group, resulting in the formation of alkylphenol. The alkylphenol then undergoes further reactions, where the phenolic hydroxyl group is replaced by hydrogen, leading to the generation of aromatics. Under high-pressure conditions, the aromatic ring can undergo hydrotreating, saturation, and ring-opening reactions, resulting in the production of a small amount of long-chain aliphatic hydrocarbons [151].

Hydropyrolysis of biomass primarily produces oxygenated compounds, with minimal amounts of aromatic and aliphatic hydrocarbons. The addition of catalysts, particularly metal-based catalysts, has been found to significantly increase the yield of hydrocarbons, due to their HSPE [152]. The occurrence of HSPE requires two conditions: 1) metal particles that can dissociate the hydrogen source to obtain hydrogen atoms, and 2) supports with active sites for capturing hydrogen atoms and pore structures for facilitating their transfer. The metal species that exhibit significant hydrogen spillover phenomenon include noble metals like Ru, Pd, Pt, as well as common metals like Ni and Co.

The use of reducible materials (such as CeO₂, TiO₂, WO₃, and Fe₂O₃) as supports can lead to the HSPE, which promotes the HDO reaction. In this process, hydrogen molecules (H₂) dissociate into atoms and then diffuse as pairs of protons and electrons on the catalyst surface. In the case of Pt/CeO₂, the diffused electrons can reduce Ce⁴⁺ to Ce³⁺, while the protons are transferred from the Pt sites to the support surface and bind to the surface oxygen ions [153]. The interaction between the metal and the support can enhance the activity of the metal sites, promoting the HSPE. Additionally, this interaction partially reduces the support and enhances the activity of the catalyst in the HDO process [154]. For TiO₂, the support itself contains a higher number of oxygen vacancies. The hydrogen spillover process facilitates the continuous reduction of oxidized Ti and the regeneration of depleted oxygen vacancies. These vacancies serve not only for the adsorption of oxygenated compounds but also for promoting the diffusion of hydrogen atoms. Non-reducible supports such as SiO₂, Al₂O₃, and zeolites are not considered to have HSPE. It is commonly believed that the transfer of H on the surface of these supports occurs through ion exchange. However, this process is hindered by high H-D exchange barriers, making it difficult to directly prove the hydrogen spillover phenomenon on the surfaces of non-reducible supports at present [155]. For non-reducing supports, hydrogen spillover can be achieved through three methods. The first method involves high-temperature treatment to create defective spots on the support structure, which allows for the immobilization of hydrogen atoms [152]. The second method is that the addition of oxygenated functional groups can be used to capture free hydrogen atoms and facilitate their transfer to the surface of the support material [156]. The third method is that hydrogen transfer can occur through the acidic sites of supports like zeolites, Brønsted acid sites and Lewis acid sites favor mobile diffusion of protons and electrons. However, further investigation is needed to understand the mechanism of hydrogen spillover on non-reducing support surfaces [157].

The HSPE of supported metal catalysts is synergistically manifested through various active sites, including acidic sites, metal sites, oxygen vacancies, and metal-support interactions. These active sites play a significant role in enhancing the HDO reactions of biomass and its derived oxygenated compounds. Phenol undergoes deoxygenation on the Ru/TiO₂ surface, primarily through a process known as DDO, resulting in the production of benzene. During this reaction, hydrogen atoms dissociate from the Ru sites and migrate to the reduced titania sites on the TiO₂ surface. The deoxygenation of phenol occurs through the adsorption of the hydroxyl group of phenol [158]. VAN was converted to 92.5% using a Pt-WO_3–x catalyst. This high conversion rate can be mainly attributed to the synergistic effect of hydrogen spillover from Pt and the presence of oxygen vacancies in the WO_3–x support [34].

Hydrogen can significantly enhance the biomass liquefaction process, resulting in the production of bio-oil with increased H/C and calorific value. When no catalyst is present, the reaction requires a high pressure to achieve hydrogen dissociation, resulting in primarily mono-oxygen compounds. After the introduction of a metal catalyst, hydrogen atoms will dissociate at the metal sites on the catalyst surface and then transfer to the acid active sites on the support surface for the HDO reaction. The HSPE, which occurs in catalysts that combine reducible supports and noble metals, increases the number of active sites on the catalyst surface. This effect enhances the efficiency of hydrogen utilization and improves the activity, selectivity, and stability of the catalyst. Ultimately, it optimizes the distribution of oil-phase product components and increases the yield of aromatic and aliphatic hydrocarbons.

4.2 Methane

Hydrogen is the most commonly studied source of hydrogen, but it is expensive. Methane is a cheaper and more abundant alternative to hydrogen, and is readily available as a major component of natural gas. This makes it a viable source of hydrogen supply. Tab.8 lists some studies on the use of methane as a pyrolysis gas. Methane can be co-converted with biomass pyrolysis intermediates to obtain the desired hydrocarbon product under favorable operating conditions and metal catalysts. Compared with phenols, cellulose and hemicellulose derived compounds are more inclined to react with methane because the former is more susceptible to ring opening.

Methane can contribute protons and carbon moieties in HDO reactions. Its high H/C can effectively promote the HDO of intermediates and inhibit the formation of carbon-rich precursors. In addition to serving as a hydrogen supply, dehydrogenated methyls and even methylene groups from methane can generate aromatic hydrocarbons under suitable conditions and catalyst action through dehydroaromatization [167]. Methane can also be used as a carbon source in a series of reactions involving the depolymerization of oxygenated compounds (e.g., phenol, furan, and furfural) and their modal compounds HDO, which can enhance the carbon yield of hydrocarbon products. The primary challenge in utilizing methane as a hydrogen source is the activation of the C–H bond. This is due to the stable tetrahedral structure of the methane molecule and the high bonding energy of C–H (435 kJ/mol) [167,168]. Methane can be activated on the surfaces of various metal active sites, such as zeolites loaded with metals like Mn, Fe, Co, Ni, Cu, and Zn [169]. The catalysts that have been extensively studied are Mo and Zn-loaded HZSM-5 catalysts. In these catalysts, Mo is present as MoO₃ on the surface of the supports, while Zn can be transformed into [Zn⁺–O–Zn²⁺] or Zn²⁺. In the first two cases, methane undergoes homolytic activation ({\mathrm{C}\mathrm{H}}_3^{\delta +}-{\mathrm{H}}^{\delta -}) [170–172]. In the third active site, methane undergoes heterolytic activation ({\mathrm{C}\mathrm{H}}_3^{\delta -}-{\mathrm{H}}^{\delta +}) [173].

The mechanism of methane in the HDO reaction of biomass-derived oxygenated compounds is often studied using model compounds such as furans, alcohols, acids, and phenolics derived from cellulose, hemicellulose, and lignin. Methane can promote alkylation, isomerization, and aromatization of intermediates in the presence of catalysts. The Moogi team [159] discovered that the Ga/HZSM-5 catalyst and depolymerized methane atmosphere had the highest yield of BTEX at 9.58 wt.%. The presence of (GaO)⁺ active sites on the catalyst surface is crucial for activating methane and oxygenated compounds. The metal Ga promotes the conversion of methane to alkenes which then undergo arylation, oligomerization, and Diels-Alder reactions with oxygenated compounds like acids and ketones. This process produces aromatics. The team also studied the role of Zn- and Mo-loaded HZSM-5 [160] in a methane-rich environment. The Zn–Mo bimetal significantly increased the BTEX yield (19.93 wt.%) by promoting the activation of methane and hydrogen. Additionally, ZnMo increased the number of acid sites on the HZSM-5 surface, enhancing aromatization, HDO, and alkylation reactions. The synergistic interaction between ZnGa/ZSM-5 bimetallic catalysts and the acid modification of the support by the metal also promotes reactions such as aromatization and HDO [174]. During the lignin conversion process, 37.4 wt.% of aromatics were produced, with a BTEX selectivity reaching 62.2 wt.%, and NMR (nuclear magnetic resonance spectroscopy) tests indicated that methane can undergo alkylation to form aryl groups, as well as cyclization with alkyl groups to generate aromatics. The purpose of metal loading on zeolite surfaces is to modify the support acidity, which in turn affects the degree of methane activation. The acidity and type of acidic sites are important factors in this process. The role of framework and extra-framework aluminum sites in the structural framework of zeolites is crucial in the co-aromatization reactions of methane and hydrocarbons [175].

Acids and ketones are frequently found in biomass pyrolysis products and are responsible for the low heating value and corrosiveness of biomass pyrolysis oils. As a result, model compounds of acids and ketones are commonly utilized in studies on hydrocarbon conversion pathways. Fig.11(a) [176] illustrates the potential transformation pathways of acids and ketones catalyzed by zeolite in a methane environment. During the ring-forming process of acids and ketones, carbon-containing radicals like methyl ethyl groups are produced through decarbonylation or decarboxylation. These carbon-containing radicals then undergo aromatization or isomerization reactions, where methane is added to the benzene ring and the benzyl group. Other oxygenated compounds, like furfural, can undergo cleavage and decarbonylation reactions in the presence of a catalyst. This produces short-chain alkanes, which then react with methane to form monocyclic aromatic hydrocarbons through arylation and the Diels-Alder reaction. The reaction pathway is shown in Fig.11(b) [165]. The HDO of phenols mainly includes DDO, tautomerization, ring opening, and others to generate a variety of radicals. These radicals then participate in radical aromatization and Diels-Alder reactions, resulting in the formation of aromatic hydrocarbons. Additionally, methane can undergo alkylation reactions to directly adduct to the aromatic ring of phenols. Fig.11(c) [167] illustrates the possible pathways of these reactions.

Methane can increase the H/C of biomass-derived oxygenated compounds and inhibit the formation of carbon precursors like polycyclic aromatic hydrocarbons (PAHs). Additionally, metal-loaded zeolite catalysts, such as Ga and Zn, can reduce the activation energy for C–H bonds, promoting the conversion of acids, ketones, phenolics, and increasing the yield of hydrocarbons, particularly monocyclic aromatic hydrocarbons, in a cost-effective manner compared to hydrogen.

4.3 Other H-donors

Other hydrogen sources, such as alcohols [177–180], plastic wastes [181,182], acids [183], naphthalene [184] are also widely applied on the HDO of biomass-derived oxygenates. The aqueous-phase system offers a greater variety of hydrogen sources. Li et al. [178] investigated the HDO conversion of ethylphenol under mild conditions and at atmospheric pressure using methanol as a hydrogen supply source. They found that the hydrogen generated by using the aqueous-phase reforming of methanol and the activation of hydrogen molecules are different. The former has a faster transfer diffusion rate on the catalyst surface, resulting in a conversion of ethylphenol close to 50% with the use of noble metal catalysts. In subsequent mechanistic investigations, the team optimized the catalytic system by combining hydrogen and methanol. They discovered that the main pathway for HDO of DHE involved DMO followed by HYD of the aryl ring, resulting in the production of propylcyclohexane. The reduction mechanisms of different externally supplied hydrogen sources vary. When using only hydrogen or methanol as the hydrogen source, the resulting products contain a mixture of alcohols and phenolic compounds. However, when both hydrogen and methanol are added, fully deoxygenated hydrocarbon products are obtained. Methanol primarily promotes reciprocal isomerization while hydrogen mainly promotes hydrogen saturation of the aromatic ring.

Shafaghat et al. [185] chose primary alcohols (such as methanol, ethanol, 1-propanol, and 1-butanol) and secondary alcohols (such as 2-propanol and 2-butanol) as hydrogen sources for the HDO of GUA and pyrolysis oil. It was discovered that secondary alcohols are effective in promoting hydrogen transfer reactions, resulting in improved HDO. Under Pd/C catalysis, GUA conversion exceeded 98.5 wt.%, although only 2-propanol achieved complete deoxygenation. In the HDO of pyrolysis oil, 2-BuOH was identified as the most effective hydrogen source. In the HDO reaction of pyrolysis oil, the HHV of the refined oil increased from 13.21 to 25.86 MJ/kg when 2-BuOH was used as a hydrogen source. In the aqueous catalytic system of the noble metal Pd-Au BMNP, using formic acid as a hydrogen source [183], the HDO of VAN could be achieved at ambient temperature and pressure. The yield of MMP can reach up to 99%, and the catalyst shows almost no deactivation after five cycles of experiments. In addition to hydrogen in the gas-phase system, alcohols are also sources of hydrogen.

Hydrogen is the most widely researched source of hydrogen supply and has been shown to contribute significantly to biomass HDO, but hydrogen is expensive to prepare and store and requires a high-pressure condition. Compared with hydrogen, CH₄, as the major component of natural gas, is much cost-effective and has become an underutilized resource for HDO. He et al. [164] used Zn/ZSM-5 to upgrade pyrolysis bio-oil under methane and the highest oil yield reached 9.98 wt.%. Other H-donors used in solution are also quite cost competitive with hydrogen. However, the solvents are also involved in the separation of products after the reaction and increase process complexity.

5 Catalysts deactivation

Catalyst stability and reusability are important indicators of economic efficiency. The catalytic activity weakens as the reaction time increases, and the extent of this weakening depends on factors such as temperature and pressure conditions, catalyst metal and support type, and the type of feedstock [13,82,186,187]. As shown in Tab.9, the deactivation of catalysts is basically divided into coking, poisoning, metal sintering, and leaching [188,189]. Noble metals form less coke and present a better deactivation resistance than non-noble metals because of their ability to promote HYD. Bimetallic catalysts have improved metal stability compared to monometallic catalysts because of optimized distribution of metal sites. The spent catalysts are usually regenerated by burning under the temperature not higher than calcination. The detailed process contains washing with organic solvents, burning in air to remove the coke, and finally reducing in hydrogen-containing gas [58,68,83,177]. Coking occurs during the catalytic refining process when certain aromatic compounds polymerize or oxygenated compounds over-aromatized in the acid active site. This leads to the formation of polymers that cover the surface of the catalyst support, blocking the pores and rendering the internal active site of the catalyst inaccessible to reactants [177]. The extent of coking depends on the acidic strength of the catalyst, the structure of the pore channel of the support, and other properties. Catalyst poisoning deactivation in aqueous phase systems occurs when reactive substances with a strong adsorption energy, such as water and CO, occupy the active sites on the catalyst surface. This not only hinders the desired HDO reaction, but also damages the structure of the supports, such as poorly hydrothermally stabilized γ-Al₂O₃. In gas-phase systems, these substances include a small amount of ash in biomass feedstocks and alkali metals. Metal sintering and leaching occur when metal particles migrate on the support surface due to unstable loading, excessive temperature, and pressure during the reaction process. This leads to the gradual accumulation of large metal clusters or the gradual loss of metal particles. The formation of carbon deposits and metal agglomeration are the primary causes of deactivation in metal-support catalysts. For catalysts poisoning, introducing promoter elements that can selectively adsorb or react with poisons or developing selective adsorbents or scavengers that can selectively remove specific poisons from the reaction mixture could prevent them from affecting the active sites of the catalyst. In addition, utilizing high surface area supports, such as mesoporous materials or zeolites, could disperse active metal particles and inhibit sintering. As for metal leaching, applying coatings or modifying the catalyst surface to create a protective layer could reduce the leaching of active metal species, and choosing catalyst supports that enhance the stability of the active metal is also useful.

The lower H/C of biomass is the main reason for the over-aromatization of intermediates into carbonaceous precursors. The addition of an external hydrogen source, along with the hydrotreating capability of noble metals, greatly enhances the resistance to carbonaceous formation and decelerates the deactivation process. The Pt/NbOPO₄ catalyst demonstrated an outstanding resistance to carbon accumulation during the cyclic HDO process. After four cycles, the yields of hexanes and pentanes decreased from 41.3% and 4.6% to 37.5% and 4.0%, respectively. The catalytic activity was only slightly weakened, which could be attributed to two factors: the use of mild reaction conditions to prevent excessive accumulation of metal particles, and the selection of non-aqueous single-phase solvents to inhibit metal leaching and preserve catalyst structure [70]. Brønsted acid sites on the surface of supports, such as HZSM-5, promote the generation of carbon positive ions by providing protons. These protons indirectly contribute to the formation of carbon deposits. Additionally, the microporous structure of HZSM-5 is prone to deactivation. However, loading metals onto the catalyst can significantly enhance its resistance to carbon deposits. Pd–Zr/HZSM-5 maintained the selectivity of ethylbenzene, the target product, after six consecutive cycles of catalyzing the ketone HDO [80]. Encapsulating metal particles within the support pores can enhance catalyst stability. The selectivity of C10 cyclic hydrocarbons, which were converted from cyclopentanone using Pt@H-BEA nanoparticle catalysts, remained stable at approximately 78% after three cycles. Additionally, the loading of Pt nanoparticles remained almost unchanged [43].

Therefore, to improve the overall durability of catalysts, the working conditions of the HDO reaction should be optimized. This includes temperature and solvent selection to minimize metal leaching and sintering. Additionally, choosing a metal-support combination with strong interaction can stabilize the metal particles on the catalyst surface, preventing migration during the HDO process. Alternatively, special loading structures like metal-encapsulated catalysts can be prepared.

6 HDO of actual biomass and bio-oil

Up to the present, numerous efforts have been devoted to the HDO of real feedstocks, either on biomass or pyrolysis bio-oil [88,193,194]. Based on the discussion above, noble-metal catalysts demonstrate a better performance compared with transition metal in most cases. However, the latter is more commonly employed because it is more economical. In addition to catalyst selection, reaction conditions including temperature, pressure and residence time are also required to optimize for assistance. Miao et al. [88] conducted poplar hydropyrolysis with NiMo/Al₂O₃ in fluidized bed to increase selectivity of aviation fuel components. At an upgrading temperature of 300–350 °C, the HHV of the oil phase reached 46.2 MJ/kg with a mass yield of 6.4 wt.%. Furthermore, the main components were aliphatic compounds (72.6%) formed through HDO and C–C coupling reaction, among which 59.9% owned a carbon number of more than 8. Wu et al. [195] utilized carbon-based catalyst modified with NiMo and investigated the effects of catalyst preparation. Catalytic hydropyrolysis, at 500 °C, 1 atm H₂, using ball milling-melting method, showed the highest liquid yield of 35.0 wt.% and H/C of 1.46, and the O/C was reduced to 0.34. The synergistic effect of Ni⁰ and Mo²⁺/Mo^δ+ were responsible for hydrogen activation and oxygen elimination. With the same metals, different supports also have an impact on the conversion process and products distribution. Yan et al. [22] employed NiFe loaded Escott, BEA and Escott-BEA catalysts on biomass catalytic hydropyrolysis. The NiFe/Escott-BEA has the weakest metal-support interaction and appropriate acidity, which enhanced hydrogen adsorption, resulting in the highest organic phase yield. Incorporation of BEA increased aromatic hydrocarbon selectivity and prevented excessive HCR of the liquid phase, while BEA or Escott alone caused a higher C2–C5 gas yield or Ni particles aggregation. Apart from aromatic hydrocarbons, HDO to obtain oxygen-containing biofuels include ketones, alcohols and furans get attention in recent years. Wang et al. [196] utilized Pd/Al₂O₃ to upgrade cellulosic waste pyrolysis vapor. With the upgrading temperature of 400 °C and 0.3 MPa H₂, the highest yield of oxygen-containing biofuels reached (171.1 mg/g) and the conversion was 89.2%. Pd modified catalysts showed an excellent stability and the coke yield is only 2.1 wt.% after three cycles. Based on noble metal Ru, Liu et al. [197] produced extremely low oxygen-content aviation fuel (0.4 wt.%) on multi-step pilot operation units which included two stages of HDO upgradation. Biomass depolymerization took place before deoxygenation and solutions containing furfural and levulinic acid were transported into the following HYD and HDO units, while the upgradation conditions were 340 °C and 4 MPa H₂. The ultimate biofuel realized a selectivity of 50.4% of isoparaffins with the carbon number ranging from C8–C15, while other components were n-paraffins (12.2%), cycloparaffins (19.0%), aromatic hydrocarbons (11.3%), and alkenes (5.6%). Furthermore, the biofuel HHV was 45.5 MJ/kg with a mass yield of 10.6%.

The effective HDO conversion of lignin that is the largest and most inactive part of biomass is frequently explored. Chen et al. [192] combined noble metal Ru and transition metal Zn to convert lignin to aromatics. Attributed to the interaction of Ru and ZnO, the selectivity of desired aromatics reached 51.3% at 240 °C and 2 MPa H₂. Incorporation of ZnO tuned the acidity of Ru and changed its electronic properties for a better activity, thus improving C–O cleavage, and the oxophilic ZnO itself facilitated the adsorption of oxygen intermediates at the meantime. Ma et al. [198] chose Ru/Nb₂O₅ series catalysts to convert lignin, resulting in a 99.1% selectivity of C7–C9 hydrocarbons and an 88.1% selectivity of C7–C9 aromatics. The research demonstrated that the Lewis acid sites of Ru played an important role in deoxygenation, while the electronic density and particle also mattered.

As to biomass pyrolysis bio-oil, Li et al. [199] conducted catalytic hydroprocessing of crude bio-oil with biochar supported CoMo catalysts in a fixed-bed between 400 and 425 °C, at 10 MPa H₂. The oil HHV and yield reached 43.9 MJ/kg and 25.7%, with the desired C7–C11 hydrocarbons accounting for 84.9 mol.% and the main component being i-paraffins. Noble metal series catalysts including Pd, Pt, and Ru were utilized in two-stage HDO of pyrolysis bio-oil [200]. The final deoxygenation degree reached 97.6 wt.% and a bio-carbon of up to 53.4 wt.% was maintained in the biofuel, with the total oil yield of more than 90 wt.%. Ketonization, aldol-condensation and some hydrogen activation reactions in stage I, aldol-condensation in stage II, were the main reactions attributed to C6–C10 n-alkanes and cycloalkanes. (Fig.12). Under the condition of 350 °C and 60 bar, Khanh Tran et al. [190] obtained a 70.46 wt.% yield of liquid fuel from sawdust pyrolysis bio-oil with an HHV of 34.22 MJ/kg over activated carbon supported Co catalyst, among which the C8 hydrocarbons yield was 20.40 wt.%.

Based on optimized HDO conditions such as temperature and pressure, biomass pyrolysis vapor or bio-oil could be remarkably upgraded as the oxygenates are substantially removed and the HHV closely resembles the aviation fuel. Multifunctional metal catalysts lead to varied products distribution by regulating reaction paths, mainly including HCR, aldol-condensation, HYD and deoxygenation. Although noble metals are limited by its high cost, the high HYD efficiency and stability they demonstrated are useful guidelines to design other metal catalysts. Thus, non-noble metals combined with active supports need more attention for industrialization. Considering economy, sustainability and efficiency, direct conversion from biomass to bio-jet fuel through catalytic fast hydropyrolysis is more promising for further development compared with bio-oil upgradation.

7 Conclusions and outlook

7.1 Conclusions

The HDO of biomass to bio-jet fuels is a highly promising method within the realm of utilizing renewable and sustainable energy resources. In the realm of aviation, an industry that is challenging to transition toward electrification and HYD, bio-based fuels have emerged as a viable solution for mitigating greenhouse gas emissions. This paper presents a comprehensive overview of the reaction mechanism involved in the catalytic HDO technology, which has garnered significant interest in the realm of biomass liquefaction to produce liquid fuels. In this paper, the mechanism of HDO was illustrated from both experimental and computational perspectives, focusing on model compounds derived from lignocellulosic biomass, such as furfurals, furans, and phenolics. In the experiments involving model compounds, the presence of metal catalysts has been found to exert a substantial influence on the selectivity of the resulting products. This can be attributed to the distinct electronic structures exhibited by noble and non-noble metals, which in turn give rise to varying effects on reactions such as HYD and C–O breaking. In addition to the distribution pattern and size of the metal particles on the surface of the supports, the presence of structural defective sites and facial functional groups on the supports, as well as the interactions between the metal and the supports, also play a crucial role in determining the selectivity of both intermediate and final products. In addition to the extensively researched hydrogen, cost-effective alternatives such as methane and methanol have been found to exhibit a relatively efficient HDO of oxygenated compounds under mild reaction conditions. This paper provides a comprehensive overview of catalyst deactivation in the context of the catalytic HDO reaction. It explores the various factors that contribute to catalyst deactivation and examines the stability of metal catalysts commonly used in this reaction. Theoretical calculations, on the other hand, provide a microscopic explanation for the facilitation of hydrogen dissociation and the diffusion of hydrogen radicals from noble metallic sites to other active sites. This phenomenon is known as the HSPE. The calculations also explain the promotion of strong adsorption of oxygenated functional groups of model compounds and the breaking of C–O bonds at the oxygen vacancies on the reducible supports or a variety of non-noble metals. Additionally, the calculations summarize thermodynamically and kinetically favored pathways by comparing the activation barriers for C–O and C–C bond activation and the reaction energy. In conclusion, biomass HDO holds tremendous promise as a sustainable pathway for the production of sustainable aviation fuel. The advancements in understanding reaction pathways, the development of catalysts, and the combination of theoretical and experimental approaches have significantly contributed to the progress in this field.

7.2 Challenges and perspective

However, there are still numerous challenges that need to be addressed in the mechanistic investigation of biomass catalytic HDO and its subsequent implementation in industrial applications.

The enhancement of liquid yield and selectivity of target hydrocarbons derived from biomass catalytic HDO is necessary. Due to the HCR effect induced by hydrogen radicals, along with the reaction conditions of elevated temperature and pressure, the production of C1–C4 hydrocarbons in the gas products is enhanced. This growth in yield leads to a reduction in the production of organic-phase compounds and also results in a significant carbon loss. Additionally, there are notable variations in the distribution of carbon numbers among the hydrocarbons obtained from catalytic HDO when compared to the jet fuel fractions. The hydrocarbons obtained have a low carbon number and are predominantly composed of aromatic products. The distribution of carbon numbers for alkanes derived from cellulose and hemicellulose ranges from C4 to C6. Meanwhile, achieving the full saturation of the benzene ring and the ring-opening reaction of phenolics derived from lignin, as well as obtaining high selectivity of target alkanes in the final products, poses significant challenges. Therefore, it is imperative to conduct further investigations into the mechanism of alkylation reactions, carbon coupling, and other carbon chain growth reactions involving oxygenated compounds.

The elucidation of the catalyst deactivation mechanism necessitates further investigation. The deactivation of the catalyst can be reduced by utilizing hydrogen as an external source. However, due to the high cost of hydrogen, alternative options such as methane and alcohols can be employed as hydrogen sources for the reaction. Compared to hydrogen, these alternative sources possess a higher carbon content, thereby increasing the potential for catalyst deactivation. During the process, hydrogen atoms are dissociated, resulting in the formation of carbon moieties. These carbon moieties have a tendency to polymerize, leading to the formation of oligomers. Consequently, a carbonaceous deposit is formed on the active sites of the catalysts. This deposit hampers the performance of the catalysts and contributes to carbon loss. Therefore, further investigation is required to understand the deactivation mechanism in HDO reactions caused by hydrogen sources or other contributing factors.

The mechanism of catalytic HDO for real biomass remains uncertain. In addition to phenols and furfurals, the process of biomass depolymerization yields a significant quantity of small molecule oxygenated compounds, including acids, esters, and ketones. However, the reaction pathways of these compounds have not been thoroughly investigated. The depolymerization, adsorption on the catalyst surface, and the subsequent HYD, deoxygenation, isomerization, and other reactions of actual biomass exhibit a high level of complexity. There is still a significant disparity between the studies conducted on model compounds and real-life scenarios. To understand the biomass HDO process, it is imperative to develop more mechanistic approaches.

Scaling up bio-oil production to meet aviation fuel demands poses economic and logistical challenges. The cost of producing bio-oil at a scale competitive with traditional aviation fuels remains a significant hurdle. Meanwhile, meeting stringent aviation fuel standards and regulations requires rigorous testing and certification processes. To ensure bio-oil to meet these standards is essential for its acceptance and integration into existing aviation infrastructure. The life cycle analysis of bio-oil production, from feedstock cultivation to end-use, to ensure overall environmental benefits also require consideration.

Strategic integration of bio-oil processing units within existing oil refineries and aviation fuel production facilities can enhance synergies and reduce production costs which still needs collaborative efforts between governments, research institutions, and aviation industry stakeholders. Incentives and policies supporting sustainable aviation fuels can drive industry-wide transition and establish a framework for the widespread adoption of bio-oil-based aviation fuels.

Based on the mechanism work performed on biomass-derived oxygenates, it is crucial to employ mechanistic studies to guide the design and synthesis of catalysts that exhibit both a high HDO efficiency and resistance to deactivation. Furthermore, these studies can contribute to the overall improvement of the process, enabling the production of desired hydrocarbons from biomass. Further studies will be conducted to explore the underlying mechanisms involved in the formation of biofuels.

References

[1] InternationalEnergy Agency (IEA). Outlook—Key World Energy Statistics 2021—Analysis. 2023–11-13, available at the website of IEA

[2] InternationalEnergy Agency (IEA). Overview—World Energy Outlook 2021—Analysis. 2023–11-13, available at the website of IEA

[3] Alabi O, Abubakar A, Werkmeister A. . Keeping the lights on or off: Tracking the progress of access to electricity for sustainable development in Nigeria. GeoJournal, 2022, 88(2): 1535–1558

[4] Ganti G, Gidden M J, Smith C J. . Uncompensated claims to fair emission space risk putting Paris Agreement goals out of reach. Environmental Research Letters, 2023, 18(2): 024040

[5] IntergovernmentalPanel on Climate Change (IPCC). IPCC meets to approve the final component of the Sixth Assessment Report. 2023–11-13, available at the website of IPCC

[6] USEnergy Information Administration (EIA). International Energy Outlook 2023. 2023–11-13, available at the website of EIA

[7] Su C W, Pang L D, Qin M. . The spillover effects among fossil fuel, renewables and carbon markets: Evidence under the dual dilemma of climate change and energy crises. Energy, 2023, 274: 127304

[8] Ju Y, Liu R, Fu L. Engineering fronts in fields of Energy and Electrical Science and Technologies in the report of Engineering Fronts 2022. Frontiers in Energy, 2023, 17(1): 5–8

[9] Osman A I, Farghali M, Ihara I. . Materials, fuels, upgrading, economy, and life cycle assessment of the pyrolysis of algal and lignocellulosic biomass: A review. Environmental Chemistry Letters, 2023, 21(3): 1419–1476

[10] Mujtaba M, Fernandes Fraceto L, Fazeli M. . Lignocellulosic biomass from agricultural waste to the circular economy: A review with focus on biofuels, biocomposites and bioplastics. Journal of Cleaner Production, 2023, 402: 136815

[11] Bhoi P R, Ouedraogo A S, Soloiu V. . Recent advances on catalysts for improving hydrocarbon compounds in bio-oil of biomass catalytic pyrolysis. Renewable & Sustainable Energy Reviews, 2020, 121: 109676

[12] Popp J, Lakner Z, Harangi-Rákos M. . The effect of bioenergy expansion: Food, energy, and environment. Renewable & Sustainable Energy Reviews, 2014, 32: 559–578

[13] Dabros T M H, Stummann M Z, Høj M. . transportation fuels from biomass fast pyrolysis, catalytic hydrodeoxygenation, and catalytic fast hydropyrolysis. Progress in Energy and Combustion Science, 2018, 68: 268–309

[14] AviationBenefits Beyond Borders. Waypoint 2050. 2023–11-13, available at the website of Aviationbenefits

[15] Andrade M C, Gorgulho Silva C D O, de Souza Moreira L R. . Crop residues: Applications of lignocellulosic biomass in the context of a biorefinery. Frontiers in Energy, 2022, 16(2): 224–245

[16] Xu Y, Liu X, Qi J. . Compositional and structural study of ash deposits spatially distributed in superheaters of a large biomass-fired CFB boiler. Frontiers in Energy, 2021, 15(2): 449–459

[17] Pan P, Wu Y, Chen H. Performance evaluation of an improved biomass-fired cogeneration system simultaneously using extraction steam, cooling water, and feedwater for heating. Frontiers in Energy, 2022, 16(2): 321–335

[18] Guo S, Wei X, Che D. . Experimental study on influence of operating parameters on tar components from corn straw gasification in fluidized bed. Frontiers in Energy, 2021, 15(2): 374–383

[19] Khalifa O, Xu M, Zhang R. . Steam reforming of toluene as a tar model compound with modified nickel-based catalyst. Frontiers in Energy, 2022, 16(3): 492–501

[20] Li H, Hou C, Zhai Y. . Selective preparation for biofuels and high value chemicals based on biochar catalysts. Frontiers in Energy, 2023, 17(5): 635–653

[21] Yu H, Wu J, Wei W. . Synthesis of magnetic carbonaceous acid derived from waste garlic peel for biodiesel production via esterification. Frontiers in Energy, 2023, 17(1): 176–187

[22] Yan P, Nur Azreena I, Peng H. . Catalytic hydropyrolysis of biomass using natural zeolite-based catalysts. Chemical Engineering Journal, 2023, 476: 146630

[23] Liu X, Chen Z, Lu S. . Heterogeneous photocatalytic conversion of biomass to biofuels: A review. Chemical Engineering Journal, 2023, 476: 146794

[24] Bridgwater A. Fast pyrolysis processes for biomass. Renewable & Sustainable Energy Reviews, 2000, 4(1): 1–73

[25] Wang M, Dewil R, Maniatis K. . Biomass-derived aviation fuels: Challenges and perspective. Progress in Energy and Combustion Science, 2019, 74: 31–49

[26] Yang Y, Xu X, He H. . The catalytic hydrodeoxygenation of bio-oil for upgradation from lignocellulosic biomass. International Journal of Biological Macromolecules, 2023, 242: 124773

[27] Gea S, Hutapea Y A, Piliang A F R. . A comprehensive review of experimental parameters in bio-oil upgrading from pyrolysis of biomass to biofuel through catalytic hydrodeoxygenation. BioEnergy Research, 2023, 16(1): 325–347

[28] Ouedraogo A S, Bhoi P R. Recent progress of metals supported catalysts for hydrodeoxygenation of biomass derived pyrolysis oil. Journal of Cleaner Production, 2020, 253: 119957

[29] Figueirêdo M B, Jotic Z, Deuss P J. . Hydrotreatment of pyrolytic lignins to aromatics and phenolics using heterogeneous catalysts. Fuel Processing Technology, 2019, 189: 28–38

[30] Sun J, Shao S, Hu X. . Synthesis of oxygen-containing precursors of aviation fuel via carbonylation of the aqueous bio-oil fraction followed by C–C coupling. ACS Sustainable Chemistry & Engineering, 2022, 10(33): 11030–11040

[31] Wang H, Ruan H, Feng M. . One-pot process for hydrodeoxygenation of lignin to alkanes using Ru-based bimetallic and bifunctional catalysts supported on zeolite Y. ChemSusChem, 2017, 10(8): 1846–1856

[32] Xue S, Luo Z, Sun H. . Product regulation and catalyst deactivation during ex-situ catalytic fast pyrolysis of biomass over nickel-molybdenum bimetallic modified micro-mesoporous zeolites and clays. Bioresource Technology, 2022, 364: 128081

[33] Tian X, Wang Y, Zeng Z. . Research progress on the role of common metal catalysts in biomass pyrolysis: A state-of-the-art review. Green Chemistry, 2022, 24(10): 3922–3942

[34] Sun M, Zhang Y, Liu W. . Synergy of metallic Pt and oxygen vacancy sites in Pt–WO_3−x catalysts for efficiently promoting vanillin hydrodeoxygenation to methylcyclohexane. Green Chemistry, 2022, 24(24): 9489–9495

[35] Shivhare A, Hunns J A, Durndell L J. . Metal–acid synergy: Hydrodeoxygenation of anisole over Pt/Al-SBA-15. ChemSusChem, 2020, 13(18): 4775–4775

[36] Teles C A, De Souza P M, Rabelo-Neto R C. . Reaction pathways for the HDO of guaiacol over supported Pd catalysts: Effect of support type in the deoxygenation of hydroxyl and methoxy groups. Molecular Catalysis, 2022, 523: 111491

[37] Feng S, Liu X, Su Z. . Low temperature catalytic hydrodeoxygenation of lignin-derived phenols to COLs over the Ru/SBA-15 catalyst. RSC Advances, 2022, 12(15): 9352–9362

[38] Zhang K, Meng Q, Wu H. . Selective hydrodeoxygenation of aromatics to COLs over Ru single atoms supported on CeO₂. Journal of the American Chemical Society, 2022, 144(45): 20834–20846

[39] Cai T, Deng Q, Peng H. . Synthesis of renewable C–C cyclic compounds and high-density biofuels using 5-hydromethylfurfural as a reactant. Green Chemistry, 2020, 22(8): 2468–2473

[40] Yan P, Wang H, Liao Y. . Synthesis of renewable diesel and jet fuels from bio-based furanics via hydroxyalkylation/alkylation (HAA) over S O4 2−/TiO₂ and hydrodeoxygenation (HDO) reactions. Fuel, 2023, 342: 127685

[41] Gao Z, Zhou Z, Wang M. . Highly dispersed Pd anchored on heteropolyacid modified ZrO₂ for high efficient hydrodeoxygenation of lignin-derivatives. Fuel, 2023, 334: 126768

[42] Wang L, Yang Y, Shi Y. . Single-atom catalysts with metal-acid synergistic effect toward hydrodeoxygenation tandem reactions. Chem Catalysis, 2022, 3(1): 100483

[43] Cho H J, Kim D, Xu B. Pore size engineering enabled selectivity control in tandem catalytic upgrading of cyclopentanone on zeolite-encapsulated Pt nanoparticles. ACS Catalysis, 2020, 10(15): 8850–8859

[44] Deng Q, Peng H, Yang Z. . A one-pot synthesis of high-density biofuels through bifunctional mesoporous zeolite-encapsulated Pd catalysts. Applied Catalysis B: Environmental, 2023, 337: 122982

[45] Ly H V, Kim J, Hwang H T. . Catalytic hydrodeoxygenation of fast pyrolysis bio-oil from saccharina japonica alga for bio-oil upgrading. Catalysts, 2019, 9(12): 1043

[46] Song Q L, Zhao Y P, Wu F P. . Selective hydrogenolysis of lignin-derived aryl ethers over Co/C@N catalysts. Renewable Energy, 2020, 148: 729–738

[47] Kumar A, Biswas B, Saini K. . Py-GC/MS study of prot lignin with cobalt impregnated titania, ceria and zirconia catalysts. Renewable Energy, 2021, 172: 121–129

[48] Song W, Liu Y, Baráth E. . Synergistic effects of Ni and acid sites for hydrogenation and C–O bond cleavage of substituted phenols. Green Chemistry, 2015, 17(2): 1204–1218

[49] Saidi M, Moradi P. Catalytic hydrotreatment of lignin-derived pyrolysis bio-oils using Cu/γ-Al₂O₃ catalyst: Reaction network development and kinetic study of anisole upgrading. International Journal of Energy Research, 2021, 45(6): 8267–8284

[50] Sirous-Rezaei P, Park Y K. Catalytic hydropyrolysis of lignin: Suppression of coke formation in mild hydrodeoxygenation of lignin-derived phenolics. Chemical Engineering Journal, 2020, 386: 121348

[51] Lonchay W, Bagnato G, Sanna A. Highly selective hydropyrolysis of lignin waste to benzene, toluene and xylene in presence of zirconia supported iron catalyst. Bioresource Technology, 2022, 361: 127727

[52] Afreen G, Mittal D, Upadhyayula S. Biomass-derived phenolics conversion to C10–C13 range fuel precursors over metal ion-exchanged zeolites: Physicochemical characterization of catalysts and process parameter optimization. Renewable Energy, 2020, 149: 489–507

[53] Zhang J, Mao D, Zhang H. . Improving furfural hydrogenation selectivity by enhanced Ni-TiO₂ electronic interaction. Applied Catalysis A, General, 2023, 660: 119206

[54] Xue Y, Sharma A, Huo J. . Low-pressure two-stage catalytic hydropyrolysis of lignin and lignin-derived phenolic monomers using zeolite-based bifunctional catalysts. Journal of Analytical and Applied Pyrolysis, 2020, 146: 104779

[55] Yan P, Bryant G, Li M M J. . Shape selectivity of zeolite catalysts for the hydrodeoxygenation of biocrude oil and its model compounds. Microporous and Mesoporous Materials, 2020, 309: 110561

[56] Zheng N, Wang J. Distinctly different performances of two iron-doped charcoals in catalytic hydrocracking of pine wood hydropyrolysis vapor to methane or upgraded bio-oil. Energy & Fuels, 2020, 34(1): 546–556

[57] Guan H, Ding W, Liu S. . Catalytic hydrothermal liquefaction of Chinese herb residue for the production of high-quality bio-oil. International Journal of Hydrogen Energy, 2023, 48(30): 11205–11213

[58] Ji N, Cheng S, Jia Z. . Fabricating bifunctional Co−Al₂O₃ @USY catalyst via in-situ growth method for mild hydrodeoxygenation of lignin to naphthenes. ChemCatChem, 2022, 14(12): e202200274

[59] Gamliel D P, Bollas G M, Valla J A. Bifunctional Ni-ZSM-5 catalysts for the pyrolysis and hydropyrolysis of biomass. Energy Technology, 2017, 5(1): 172–182

[60] Yang Y, Ochoa-Hernández C, De La Peña O’Shea V A. . Effect of metal-support interaction on the selective hydrodeoxygenation of anisole to aromatics over Ni-based catalysts. Applied Catalysis B: Environmental, 2014, 145: 91–100

[61] Guo H, Zhao J, Chen Y. . Mechanistic insights into hydrodeoxygenation of lignin derivatives over Ni single atoms supported on Mo₂C. ACS Catalysis, 2024, 14(2): 703–717

[62] Hu Y, Li X, Liu M. . Ni-based nanoparticles catalyzed hydrodeoxygenation of ketones, ethers, and phenols to (cyclo) aliphatic compounds. ACS Sustainable Chemistry & Engineering, 2023, 11(42): 15302–15314

[63] Yan P, Tian X, Kennedy E M. . The role of Ni sites located in mesopores in the selectivity of anisole hydrodeoxygenation. Catalysis Science & Technology, 2022, 12(7): 2184–2196

[64] Wang W, Zhang H, Zhou F. . Al-doped core-shell-structured Ni@mesoporous silica for highly selective hydrodeoxygenation of lignin-derived aldehydes. ACS Applied Materials & Interfaces, 2023, 15(28): 33654–33664

[65] Chen C, Ji X, Xiong Y. . Ni/Ce co-doping metal–organic framework catalysts with oxygen vacancy for catalytic transfer hydrodeoxygenation of lignin derivatives vanillin. Chemical Engineering Journal, 2024, 481: 148555

[66] Gong X, Li N, Li Y. . The catalytic hydrogenation of furfural to 2-methylfuran over the Mg–Al oxides supported Co–Ni bimetallic catalysts. Molecular Catalysis, 2022, 531: 112651

[67] Strapasson G B, Sousa L S, Báfero G B. . Acidity modulation of Pt-supported catalyst enhances C–O bond cleavage over acetone hydrodeoxygenation. Applied Catalysis B: Environmental, 2023, 335: 122863

[68] Zhu L, Li W, Zhang H. . Bimetallic ruthenium- and zinc-doped beta zeolite for efficiently depolymerizing Kraft lignin. Fuel, 2023, 349: 128766

[69] Li X. Efficient hydrodeoxygenation of lignocellulose derivative oxygenates to aviation fuel range alkanes using Pd–Ru/hydroxyapatite catalysts. Fuel Processing Technology, 2022, 232: 107263

[70] Xia Q, Chen Z, Shao Y. . Direct hydrodeoxygenation of raw woody biomass into liquid alkanes. Nature Communications, 2016, 7(1): 11162

[71] Zhao M, Hu J, Wu S. . Hydrodeoxygenation of lignin-derived phenolics over facile prepared bimetallic RuCoN_x/NC. Fuel, 2022, 308: 121979

[72] Resende K A, Teles C A, Jacobs G. . Hydrodeoxygenation of phenol over zirconia supported Pd bimetallic catalysts. The effect of second metal on catalyst performance. Applied Catalysis B: Environmental, 2018, 232: 213–231

[73] Zhao Y P, Wu F P, Song Q L. . Hydrodeoxygenation of lignin model compounds to alkanes over Pd-Ni/HZSM-5 catalysts. Journal of the Energy Institute, 2020, 93(3): 899–910

[74] Shu R, Li R, Liu Y. . Enhanced adsorption properties of bimetallic RuCo catalyst for the hydrodeoxygenation of phenolic compounds and raw lignin-oil. Chemical Engineering Science, 2020, 227: 115920

[75] Liu T, Tian Z, Zhang W. . Selective hydrodeoxygenation of lignin-derived phenols to alkyl COLs over highly dispersed RuFe bimetallic catalysts. Fuel, 2023, 339: 126916

[76] Sunil More G, Rajendra Kanchan D, Banerjee A. . Selective catalytic hydrodeoxygenation of vanillin to 2-methoxy-4-methyl phenol and 4-methyl cyclohexanol over Pd/CuFe₂O₄ and PdNi/ CuFe₂O₄ catalysts. Chemical Engineering Journal, 2023, 462: 142110

[77] Li R, Qiu J, Chen H. . Hydrodeoxygenation of phenolic compounds and raw lignin-oil over bimetallic RuNi catalyst: An experimental and modeling study focusing on adsorption properties. Fuel, 2020, 281: 118758

[78] Zhang J, Sun K, Li D. . Pd-Ni bimetallic nanoparticles supported on active carbon as an efficient catalyst for hydrodeoxygenation of aldehydes. Applied Catalysis A, General, 2019, 569: 190–195

[79] Liu X. In-situ studies on the synergistic effect of Pd–Mo bimetallic catalyst for anisole hydrodeoxygenation. Molecular Catalysis, 2022, 230: 112591

[80] Ding W, Li H, Zong R. . Controlled hydrodeoxygenation of biobased ketones and aldehydes over an alloyed Pd–Zr catalyst under mild conditions. ACS Sustainable Chemistry & Engineering, 2021, 9(9): 3498–3508

[81] Salakhum S, Yutthalekha T, Shetsiri S. . Bifunctional and bimetallic Pt–Ru/HZSM-5 nanoparticles for the mild hydrodeoxygenation of lignin-derived 4-propylphenol. ACS Applied Nano Materials, 2019, 2(2): 1053–1062

[82] Mortensen P M, Gardini D, Damsgaard C D. . Deactivation of Ni-MoS₂ by bio-oil impurities during hydrodeoxygenation of phenol and octanol. Applied Catalysis A, General, 2016, 523: 159–170

[83] Chandler D S, Seufitelli G V S, Resende F L P. Catalytic route for the production of alkanes from hydropyrolysis of biomass. Energy & Fuels, 2020, 34(10): 12573–12585

[84] Li T, Su J, Wang H. . Catalytic hydropyrolysis of lignin using NiMo-doped catalysts: Catalyst evaluation and mechanism analysis. Applied Energy, 2022, 316: 119115

[85] Stummann M Z, Høj M, Davidsen B. . Effect of the catalyst in fluid bed catalytic hydropyrolysis. Catalysis Today, 2020, 355: 96–109

[86] Su J, Li T, Luo G. . Co-hydropyrolysis of pine and HDPE over bimetallic catalysts: Efficient BTEX production and process mechanism analysis. Fuel Processing Technology, 2023, 249: 107845

[87] He C, Ruan T, Ouyang X. . Selective hydrodeoxygenation of monophenolics from lignin bio-oil for preparing cyclohexanol and its derivatives over Ni–Co/Al₂O₃-MgO catalyst. Industrial Crops and Products, 2023, 202: 117045

[88] Miao F, Luo Z, Zhou Q. . Study on the reaction mechanism of C8+ aliphatic hydrocarbons obtained directly from biomass by hydropyrolysis vapor upgrading. Chemical Engineering Journal, 2023, 464: 142639

[89] Khromova S A, Smirnov A A, Bulavchenko O A. . Anisole hydrodeoxygenation over Ni–Cu bimetallic catalysts: The effect of Ni/Cu ratio on selectivity. Applied Catalysis A, General, 2014, 470: 261–270

[90] Liu M, Zhang J, Zheng L. . Significant promotion of surface oxygen vacancies on bimetallic CoNi nanocatalysts for hydrodeoxygenation of biomass-derived vanillin to produce methylCOL. ACS Sustainable Chemistry & Engineering, 2020, 8(15): 6075–6089

[91] Blanco E, Carrales-Alvarado D, Belen Dongil A. . Effect of the support functionalization of mono- and bimetallic Ni/Co supported on graphene in hydrodeoxygenation of guaiacol. Industrial & Engineering Chemistry Research, 2021, 60(51): 18870–18879

[92] Zhang Y, Wang W, Fan G. . Defect-decorated NiFe bimetallic nanocatalysts for the enhanced hydrodeoxygenation of guaiacol. ChemCatChem, 2022, 14(19): e202200585

[93] Zhang Y, Fan G, Yang L. . Cooperative effects between Ni–Mo alloy sites and defective structures over hierarchical Ni–Mo bimetallic catalysts enable the enhanced hydrodeoxygenation activity. ACS Sustainable Chemistry & Engineering, 2021, 9(34): 11604–11615

[94] Han Q, Rehman M U, Wang J. . The synergistic effect between Ni sites and Ni–Fe alloy sites on hydrodeoxygenation of lignin-derived phenols. Applied Catalysis B: Environmental, 2019, 253: 348–358

[95] Lei X, Du X, Xin H. . Chemical-switching strategy for the production of green biofuel on NiCo/MCM-41 catalysts by tuning atmosphere. Fuel, 2022, 315: 123118

[96] Jaswal A, Singh P P, Kar A K. . Production of 2-methyl furan, a promising 2nd generation biofuel, by the vapor phase hydrodeoxygenation of biomass-derived furfural over TiO₂ supported Cu Ni bimetallic catalysts. Fuel Processing Technology, 2023, 245: 107726

[97] Lu K L, Yin F, Wei X Y. . Promotional effect of metallic Co and Fe on Ni-based catalysts for p-cresol deoxygenation. Fuel, 2022, 321: 124033

[98] Zhou M H, Xue Y Q, Ge F. MOF-derived NiM@C catalysts (M = Co, Mo, La) for in-situ hydrogenation/hydrodeoxygenation of lignin-derived phenols to cycloalkanes/COL. Fuel, 2022, 329: 125446

[99] Zhou X, Rauchfuss T B. Production of hybrid diesel fuel precursors from carbohydrates and petrochemicals using formic acid as a reactive solvent. ChemSusChem, 2013, 6(2): 383–388

[100] LiL, NieX, ChenY, et al. Computational insights into the hydrodeoxygenation of phenolic compounds over Pt–Fe catalysts. Journal of Physical Chemistry, 2021, 125, 26: 14239–14252

[101] Liu X, An W, Wang Y. . Hydrodeoxygenation of guaiacol over bimetallic Fe-alloyed (Ni, Pt) surfaces: Reaction mechanism, transition-state scaling relations and descriptor for predicting C–O bond scission reactivity. Catalysis Science & Technology, 2018, 8(8): 2146–2158

[102] Konadu D, Kwawu C R, Tia R. . Mechanism of guaiacol hydrodeoxygenation on Cu(111): Insights from density functional theory studies. Catalysts, 2021, 11(4): 523

[103] Fraga G, Yin Y, Konarova M. . Hydrocarbon hydrogen carriers for catalytic transfer hydrogenation of guaiacol. International Journal of Hydrogen Energy, 2020, 45(51): 27381–27391

[104] Gao M, Tan H, Zhu P. . Why phenol is selectively hydrogenated to cyclohexanol on Ru(0001): An experimental and theoretical study. Applied Surface Science, 2021, 558: 149880

[105] Zhou J, An W, Wang Z. . Hydrodeoxygenation of phenol over Ni-based bimetallic single-atom surface alloys: Mechanism, kinetics and descriptor. Catalysis Science & Technology, 2019, 9(16): 4314–4326

[106] Nie X, Zhang Z, Wang H. . Effect of surface structure and Pd doping of Fe catalysts on the selective hydrodeoxygenation of phenol. Catalysis Today, 2021, 371: 189–203

[107] Wu B, Li L, Wang H. . Role of MoO_x/Ni(111) interfacial sites in direct deoxygenation of phenol toward benzene. Catalysis Science & Technology, 2023, 13(7): 2201–2211

[108] Tan H, Rong S, Zhao R. . Targeted conversion of model phenolics in pyrolysis bio-oils to arenes via hydrodeoxygenation over MoO_x/BaO@SBA-15 catalyst. Chemical Engineering Journal, 2022, 438: 135577

[109] Itthibenchapong V, Chakthranont P, Sattayanon C. . Understanding the promoter effect of bifunctional (Pt, Ni, Cu)-MoO_3–x/TiO₂ catalysts for the hydrodeoxygenation of p-cresol: A combined DFT and experimental study. Applied Surface Science, 2021, 547: 149170

[110] Khan T S, Singh D, Samal P P. . Mechanistic investigations on the catalytic transfer hydrogenation of lignin-derived monomers over Ru catalysts: Theoretical and kinetic studies. ACS Sustainable Chemistry & Engineering, 2021, 9(42): 14040–14050

[111] Yan P, Tian X, Kennedy E M. . Influence of promoters (Fe, Mo, W) on the structural and catalytic properties of Ni/BEA for guaiacol hydrodeoxygenation. ACS Sustainable Chemistry & Engineering, 2021, 9(46): 15673–15682

[112] Liu R, An W, Stepped M. Stepped M@Pt(211) (M = Co, Fe, Mo) single-atom alloys promote the deoxygenation of lignin-derived phenolics: Mechanism, kinetics, and descriptors. Catalysis Science & Technology, 2021, 11(21): 7047–7059

[113] Tan Q, Wang G, Long A. . Mechanistic analysis of the role of metal oxophilicity in the hydrodeoxygenation of anisole. Journal of Catalysis, 2017, 347: 102–115

[114] Chia M, Pagán-Torres Y J, Hibbitts D. . Selective hydrogenolysis of polyols and cyclic ethers over bifunctional surface sites on rhodium–rhenium catalysts. Journal of the American Chemical Society, 2011, 133(32): 12675–12689

[115] Tan Q, Wang G, Nie L. . Different product distributions and mechanistic aspects of the hydrodeoxygenation of m-cresol over platinum and ruthenium catalysts. ACS Catalysis, 2015, 5(11): 6271–6283

[116] Deepa A K, Dhepe P L. Function of metals and supports on the hydrodeoxygenation of phenolic compounds. ChemPlusChem, 2014, 79(11): 1573–1583

[117] Runnebaum R C, Nimmanwudipong T, Block D E. . Catalytic conversion of compounds representative of lignin-derived bio-oils: A reaction network for guaiacol, anisole, 4-methylanisole, and cyclohexanone conversion catalysed by Pt/γ-Al₂O₃. Catalysis Science & Technology, 2012, 2(1): 113–118

[118] Furimsky E. Catalytic hydrodeoxygenation. Applied Catalysis A, General, 2000, 199(2): 147–190

[119] Kanchan D R, Banerjee A. Linear scaling relationships for furan hydrodeoxygenation over transition metal and bimetallic surfaces. ChemSusChem, 2023, 16(18): e202300491

[120] Fang W, Liu S, Steffensen A K. On the role of Cu⁺ and CuNi alloy phases in mesoporous CuNi catalyst for furfural hydrogenation. ACS Catalysis, 2023, 13(13): 8437–8444

[121] Chen L, Shi Y, Chen C. . Precise control over local atomic structures in Ni–Mo bimetallic alloys for the hydrodeoxygenation reaction: A combination between density functional theory and microkinetic modeling. Journal of Physical Chemistry C, 2022, 126(9): 4319–4328

[122] Zhang Z, Zhang Z, Zhang X. . Single pot selective conversion of furfural into 2-methylfuran over a Co-CoO_x/AC bifunctional catalyst. Applied Surface Science, 2023, 612: 155871

[123] Hamid A H, Ali L, Shittu T. . Transformation of levoglucosan into liquid fuel via catalytic upgrading over Ni-CeO₂ catalysts. Molecular Catalysis, 2023, 547: 113382

[124] Asada D, Ikeda T, Muraoka K. . Density functional theory study of deoxydehydration reaction by TiO₂-supported monomeric and dimeric molybdenum oxide catalysts. Journal of Physical Chemistry C, 2022, 126(48): 20375–20387

[125] Banerjee A, Mushrif S H. Reaction pathways for the deoxygenation of biomass-pyrolysis-derived bio-oil on Ru: A DFT study using furfural as a model compound. ChemCatChem, 2017, 9(14): 2828–2838

[126] Wang X B, Xie Z Z, Guo L. . Mechanism of dibenzofuran hydrodeoxygenation on the surface of Pt(111): A DFT study. Catalysis Today, 2021, 364: 220–228

[127] Chen H, Liu J, Li W. . Mechanism insights into the decarbonylation of furfural to furan over Ni/MgO: A molecular simulation study. Energy & Fuels, 2023, 37(14): 10594–10602

[128] Zhang J, Jia Z, Yu S. . Regulating the Cu⁰–Cu⁺ ratio to enhance metal-support interaction for selective hydrogenation of furfural under mild conditions. Chemical Engineering Journal, 2023, 468: 143755

[129] Pino N, Sitthisa S, Tan Q. . Structure, activity, and selectivity of bimetallic Pd–Fe/SiO₂ and Pd–Fe/γ-Al₂O₃ catalysts for the conversion of furfural. Journal of Catalysis, 2017, 350: 30–40

[130] Lin W, Chen Y, Zhang Y. . Surface synergetic effects of Ni–ReO_x for promoting the mild hydrogenation of furfural to tetrahydrofurfuryl alcohol. ACS Catalysis, 2023, 13(17): 11256–11267

[131] Luo J, Cheng Y, Niu H. . Efficient Cu/FeO_x catalyst with developed structure for catalytic transfer hydrogenation of furfural. Journal of Catalysis, 2022, 413: 575–587

[132] Zhao H, Liao X, Cui H. . Efficient Cu–Co bimetallic catalysts for the selective hydrogenation of furfural to furfuryl alcohol. Fuel, 2023, 351: 128887

[133] Yuan E, Wang C, Wu C. . Constructing a Pd–Co interface to tailor a d-band center for highly efficient hydroconversion of furfural over cobalt oxide-supported Pd catalysts. ACS Applied Materials & Interfaces, 2023, 15(37): 43845–43858

[134] Šivec R, Huš M, Likozar B. . Furfural hydrogenation over Cu, Ni, Pd, Pt, Re, Rh and Ru catalysts: Ab initio modelling of adsorption, desorption and reaction micro-kinetics. Chemical Engineering Journal, 2022, 436: 135070

[135] Xue J, Wang Y, Meng Y. . Theoretical investigation of decarbonylation mechanism of furfural on Pd(111) and M/Pd(111) (M = Ru, Ni, Ir) surfaces. Molecular Catalysis, 2020, 493: 111054

[136] Liu W, Yang Y, Chen L. . Atomically-ordered active sites in NiMo intermetallic compound toward low-pressure hydrodeoxygenation of furfural. Applied Catalysis B: Environmental, 2021, 282: 119569

[137] Shi Y. Exploring the reaction mechanisms of furfural hydrodeoxygenation on a CuNiCu(111) bimetallic catalyst surface from computation. ACS Omega, 2020, 5(29): 18040–18049

[138] Gunawan R, Cahyadi H S, Insyani R. . Density functional theory investigation of the conversion of 5-(hydroxymethyl)furfural into 2,5-dimethylfuran over the Pd(111), Cu(111), and Cu₃Pd(111) surfaces. Journal of Physical Chemistry C, 2021, 125(19): 10295–10317

[139] Hsiao Y W, Zong X, Zhou J. . Selective hydrodeoxygenation of 5-hydroxymethylfurfural (HMF) to 2,5-dimethylfuran (DMF) over carbon supported copper catalysts using isopropyl alcohol as a hydrogen donor. Applied Catalysis B: Environmental, 2022, 317: 121790

[140] Zhang Y, Zhan S, Liu K. . Heterogeneous hydrogenation with hydrogen spillover enabled by nitrogen vacancies on boron nitride-supported Pd nanoparticles. Angewandte Chemie International Edition 62, 2023, 62(9): e202217191

[141] Kim H, Yang S, Lim Y H. . Enhancement in the metal efficiency of Ru/TiO₂ catalyst for guaiacol hydrogenation via hydrogen spillover in the liquid phase. Journal of Catalysis, 2022, 410: 93–102

[142] Redina E A, Vikanova K V, Kapustin G I. . Selective room-temperature hydrogenation of carbonyl compounds under atmospheric pressure over platinum nanoparticles supported on ceria-zirconia mixed oxide. European Journal of Organic Chemistry, 2019, 2019(26): 4159–4170

[143] Kasabe M M, Kotkar V R, Dongare M K. . Phenol hydrogenation to COL catalysed by palladium supported on CuO/CeO₂. Chemistry, an Asian Journal, 2023, 18(11): e202300119

[144] Yung M M, Foo G S, Sievers C. Role of Pt during hydrodeoxygenation of biomass pyrolysis vapors over Pt/HBEA. Catalysis Today, 2018, 302: 151–160

[145] Rong S, Tan H, Pang Z. . Synergetic effect between Pd clusters and oxygen vacancies in hierarchical Nb₂O₅ for lignin-derived phenol hydrodeoxygenation into benzene. Renewable Energy, 2022, 187: 271–281

[146] Rios-Escobedo R, Ortiz-Santos E, Colín-Luna J A. . Anisole hydrodeoxygenation: A comparative study of Ni/TiO₂-ZrO₂ and commercial TiO₂ supported Ni and NiRu catalysts. Topics in Catalysis, 2022, 65(13–16): 1448–1461

[147] Zanuttini M S, Gross M, Marchetti G. . Furfural hydrodeoxygenation on iron and platinum catalysts. Applied Catalysis A, General, 2019, 587: 117217

[148] Guo L, Tian Y, He X. . Hydrodeoxygenation of phenolics over uniformly dispersed Pt–Ni alloys supported by self-pillared ZSM-5 nanosheets. Fuel, 2022, 322: 124082

[149] Wu X, Liu C, Wang H. . Origin of strong metal-support interactions between Pt and anatase TiO₂ facets for hydrodeoxygenation of m-cresol on Pt/TiO₂ catalysts. Journal of Catalysis, 2023, 418: 203–215

[150] Li T, Miao K, Zhao Z. . Understanding cellulose pyrolysis under hydrogen atmosphere. Energy Conversion and Management, 2022, 254: 115195

[151] Wang H, Li T, Su J. . Noncatalytic hydropyrolysis of lignin in a high pressure micro-pyrolyzer. Fuel Processing Technology, 2022, 233: 107289

[152] Geng Y, Li H. Hydrogen spillover-enhanced heterogeneously catalyzed hydrodeoxygenation for biomass upgrading, ChemSusChem, 2022, 15(8): e202102495

[153] Alayoglu S, An K, Melaet G. . Pt-mediated reversible reduction and expansion of CeO₂ in Pt nanoparticle/mesoporous CeO₂ catalyst: In situ X-ray spectroscopy and diffraction studies under redox (H₂ and O₂) atmospheres. Journal of Physical Chemistry C, 2013, 117(50): 26608–26616

[154] Ahmed F, Alam M K, Muira R. . Adsorption and dissociation of molecular hydrogen on Pt/CeO₂ catalyst in the hydrogen spillover process: A quantum chemical molecular dynamics study. Applied Surface Science, 2010, 256(24): 7643–7652

[155] Messou D, Bernardin V, Meunier F. . Origin of the synergistic effect between TiO₂ crystalline phases in the Ni/TiO₂-catalyzed CO₂ methanation reaction. Journal of Catalysis, 2021, 398: 14–28

[156] Zhang S, Xia Z, Zhang M. . Boosting selective hydrogenation through hydrogen spillover on supported-metal catalysts at room temperature. Applied Catalysis B: Environmental, 2021, 297: 120418

[157] Shin H, Choi M, Kim H. A mechanistic model for hydrogen activation, spillover, and its chemical reaction in a zeolite-encapsulated Pt catalyst. Physical Chemistry Chemical Physics, 2016, 18(10): 7035–7041

[158] Newman C, Zhou X, Goundie B. . Effects of support identity and metal dispersion in supported ruthenium hydrodeoxygenation catalysts. Applied Catalysis A, General, 2014, 477: 64–74

[159] Moogi S, Jae J, Kannapu H P R. . Enhancement of aromatics from catalytic pyrolysis of yellow poplar: Role of hydrogen and methane decomposition. Bioresource Technology, 2020, 315: 123835

[160] Moogi S, Pyo S, Farooq A. . Biomass enhancement of bioaromatics production from food waste through catalytic pyrolysis over Zn and Mo-loaded HZSM-5 under an environment of decomposed methane. Chemical Engineering Journal, 2022, 446: 137215

[161] Austin D, Wang A, He P. . Catalytic valorization of biomass derived glycerol under methane: Effect of catalyst synthesis method. Fuel, 2018, 216: 218–226

[162] Peng H, Wang A, He P. . Solvent-free catalytic conversion of xylose with methane to aromatics over Zn–Cr modified zeolite catalyst. Fuel, 2019, 253: 988–996

[163] Cheng Y T, Huber G W. Production of targeted aromatics by using Diels–Alder classes of reactions with furans and olefins over ZSM-5. Green Chemistry, 2012, 14(11): 3114

[164] He P, Shan W, Xiao Y. . Performance of Zn/ZSM-5 for in situ catalytic upgrading of pyrolysis bio-oil by methane. Topics in Catalysis, 2016, 59(1): 86–93

[165] Wang A, Austin D, Qian H. . Catalytic valorization of furfural under methane environment. ACS Sustainable Chemistry & Engineering, 2018, 6(7): 8891–8903

[166] Tshikesho R S, Kumar A, Huhnke R L. . Catalytic co-pyrolysis of red cedar with methane to produce upgraded bio-oil. Bioresource Technology, 2019, 285: 121299

[167] Wang A, Austin D, Song H. Investigations of thermochemical upgrading of biomass and its model compounds: Opportunities for methane utilization. Fuel, 2019, 246: 443–453

[168] Tang P, Zhu Q, Wu Z. . Methane activation: The past and future. Energy & Environmental Science, 2014, 7(8): 2580–2591

[169] Doan H A, Wang X, Snurr R Q. Computational screening of supported metal oxide nanoclusters for methane activation: Insights into homolytic versus heterolytic C–H bond dissociation. Journal of Physical Chemistry Letters, 2023, 14(21): 5018–5024

[170] Xu Y, Liu S, Guo X. . Methane activation without using oxidants over Mo/HZSM-5 zeolite catalysts. Catalysis Letters, 1995, 30(1–4): 135–149

[171] Xu J, Zheng A, Wang X. . Room temperature activation of methane over Zn modified H-ZSM-5 zeolites: Insight from solid-state NMR and theoretical calculations. Chemical Science, 2012, 3(10): 2932

[172] Luzgin M V, Rogov V A, Arzumanov S S. . Understanding methane aromatization on a Zn-modified high-silica zeolite. Angewandte Chemie International Edition, 2008, 47(24): 4559–4562

[173] Gabrienko A A, Arzumanov S S, Luzgin M V. . Methane activation on Zn²⁺-exchanged ZSM-5 zeolites. The effect of molecular oxygen addition. Journal of Physical Chemistry C, 2015, 119(44): 24910–24918

[174] Wang A, Song H. Maximizing the production of aromatic hydrocarbons from lignin conversion by coupling methane activation. Bioresource Technology, 2018, 268: 505–513

[175] He P, Wang A, Meng S. . Impact of Al sites on the methane co-aromatization with alkanes over Zn/HZSM-5. Catalysis Today, 2019, 323: 94–104

[176] Wang A, Austin D, Karmakar A. . Methane upgrading of acetic acid as a model compound for a biomass-derived liquid over a modified zeolite catalyst. ACS Catalysis, 2017, 7(5): 3681–3692

[177] Du L, Luo Z, Wang K. . Catalytic co-conversion of poplar pyrolysis vapor and methanol for aromatics production via ex-situ configuration. Journal of Analytical and Applied Pyrolysis, 2022, 165: 105571

[178] Li S, Liu B, Truong J. . One-pot hydrodeoxygenation (HDO) of lignin monomers to C9 hydrocarbons co-catalysed by Ru/C and Nb₂O₅. Green Chemistry, 2020, 22(21): 7406–7416

[179] Lu Q, Yang X, Zhu X. Analysis on chemical and physical properties of bio-oil pyrolyzed from rice husk. Journal of Analytical and Applied Pyrolysis, 2008, 82(2): 191–198

[180] Pidtasang B, Sukkasi S, Pattiya A. Effect of in-situ addition of alcohol on yields and properties of bio-oil derived from fast pyrolysis of eucalyptus bark. Journal of Analytical and Applied Pyrolysis, 2016, 120: 82–93

[181] Kim Y M, Jae J, Kim B S. . Catalytic co-pyrolysis of torrefied yellow poplar and high-density polyethylene using microporous HZSM-5 and mesoporous Al-MCM-41 catalysts. Energy Conversion and Management, 2017, 149: 966–973

[182] Ahmed M H M, Batalha N, Mahmudul H M D. . A review on advanced catalytic co-pyrolysis of biomass and hydrogen-rich feedstock: Insights into synergistic effect, catalyst development and reaction mechanism. Bioresource Technology, 2020, 310: 123457

[183] Wu P, Zhao D, Lu G. . Supported Pd–Au bimetallic nanoparticles as an efficient catalyst for the hydrodeoxygenation of vanillin with formic acid at room temperature. Green Chemistry, 2022, 24(3): 1096–1102

[184] Alemán-Vázquez L O, Cano-Domínguez J L, García-Gutiérrez J L. Effect of tetralin, decalin and naphthalene as hydrogen donors in the upgrading of heavy oils. Procedia Engineering, 2012, 42: 532–539

[185] Shafaghat H, Lee I G, Jae J. . Pd/C catalyzed transfer hydrogenation of pyrolysis oil using 2-propanol as hydrogen source. Chemical Engineering Journal, 2019, 377: 119986

[186] Boscagli C, Raffelt K, Grunwaldt J D. Reactivity of platform molecules in pyrolysis oil and in water during hydrotreatment over nickel and ruthenium catalysts. Biomass and Bioenergy, 2017, 106: 63–73

[187] Boscagli C, Yang C, Welle A. . Effect of pyrolysis oil components on the activity and selectivity of nickel-based catalysts during hydrotreatment. Applied Catalysis A, General, 2017, 544: 161–172

[188] Ambursa M M, Juan J C, Yahaya Y. . A review on catalytic hydrodeoxygenation of lignin to transportation fuels by using nickel-based catalysts. Renewable & Sustainable Energy Reviews, 2021, 138: 110667

[189] Lan X, Hensen E J M, Weber T. Hydrodeoxygenation of guaiacol over Ni₂P/SiO₂–reaction mechanism and catalyst deactivation. Applied Catalysis A, General, 2018, 550: 57–66

[190] Khanh TranQVu Ly HAnh VoT, . Highly selective hydrodeoxygenation of wood pallet sawdust pyrolysis oil to methyl phenol derivatives using cobalt and iron on activated carbon supported catalysts. Energy Conversion and Management: X, 2022,14: () 10018410.1016/j.ecmx.2022.100184

[191] Tian Y, Guo L, Qiao C. . Dynamics-driven tailoring of sub-nanometric Pt–Ni bimetals confined in hierarchical zeolite for catalytic hydrodeoxygenation. Applied Catalysis B: Environmental, 2023, 336: 122945

[192] Chen S, Yan P, Yu X. . Conversion of lignin to high yields of aromatics over Ru–ZnO/SBA-15 bifunctional catalysts. Renewable Energy, 2023, 215: 118919

[193] Hita I, Cordero-Lanzac T, Kekäläinen T. . In-depth analysis of raw bio-oil and its hydrodeoxygenated products for a comprehensive catalyst performance evaluation. ACS Sustainable Chemistry & Engineering, 2020, 8(50): 18433–18445

[194] Cordero-Lanzac T, Palos R, Hita I. . Revealing the pathways of catalyst deactivation by coke during the hydrodeoxygenation of raw bio-oil. Applied Catalysis B: Environmental, 2018, 239: 513–524

[195] Wu T, Dang Q, Wu Y. . Catalytic hydropyrolysis of biomass over NiMo bimetallic carbon-based catalysts. Journal of Environmental Chemical Engineering, 2023, 11(3): 110024

[196] Wang J, Jiang J, Li D. . Integrated hydropyrolysis and vapor-phase hydrodeoxygenation process with Pd/Al₂O₃ for production of advanced oxygen-containing biofuels from cellulosic wastes. Fuel Processing Technology, 2024, 254: 107948

[197] Liu Y, Chen L, Chen Y. . Pilot study on production of aviation fuel from catalytic conversion of corn stover. Bioresource Technology, 2023, 372: 128653

[198] Ma D, Lu S, Liu X. . Depolymerization and hydrodeoxygenation of lignin to aromatic hydrocarbons with a Ru catalyst on a variety of Nb-based supports. Chinese Journal of Catalysis, 2019, 40(4): 609–617

[199] Li R, Zhao Z, Zhang B. . Catalytic hydroprocessing of white pine pyrolysis bio-oil over cobalt-molybdenum carbide in a continuous packed-bed reactor. BioEnergy Research, 2021, 14(2): 588–597

[200] Onwudili J A, Scaldaferri C A. Catalytic upgrading of intermediate pyrolysis bio-oil to hydrocarbon-rich liquid biofuel via a novel two-stage solvent-assisted process. Fuel, 2023, 352: 129015

RIGHTS & PERMISSIONS

Higher Education Press 2024