Dissertations / Theses on the topic 'Catalytic hydrodeoxygenation'

The objective of this project is to investigate hydrodeoxygenation (HDO), a crucial step in the treatment of bio-oil, on transition metal phosphide catalysts. The study focuses on reactions of simple oxygenated compounds present in bio-oil -- ethanol and 2-methyltetrahydrofuran (2-MTHF). The findings from this project provide fundamental knowledge towards the hydrodeoxygenation of more complex bio-oil compounds. Ultimately, the knowledge contributes to the design of optimum catalysts for upgrading bio-oil. A series of transition metal phosphides was prepared and tested; however, the focus was on Ni2P/SiO2. Characterization techniques such as X-ray diffraction (XRD), temperature-programmed reduction and desorption (TPR and TPD), X-ray photoelectron spectroscopy (XPS), and chemisorption were used. In situ Fourier transform infrared (FTIR) spectroscopy was employed to monitor the surface of Ni2P during various experiments such as: CO and pyridine adsorption and transient state of ethanol and 2-MTHF reactions. The use of these techniques allowed for a better understanding of the role of the catalyst during deoxygenation.<br>Ph. D.

Biomass is the only renewable resource that can be used to produce liquid fuels. The pyrolysis of biomass produces a crude bio-oil liquid. This study investigated the catalytic hydrotreatment of bio-oil to produce advanced liquid biofuels. The study focused on the understanding of reaction mechanisms for the formation of light fuel species and solid coke. The results will be useful for the development of low-emission biofuel technologies using non-food lignocellulosic biomass resources.

The present Ph.D. thesis provides some examples of innovative 2nd generation catalytic processes for the conversion of renewable raw materials into green value-added chemicals. In particular, D-glucose and some its derivatives, all ideally representing waste materials of dedicated biomasses, agricultural residues, or solid organic urban waste exploitation in the biorefinery plant, were converted into useful chemical building-blocks. After a brief introduction to the topic and the description of the experimental method, each chapter of the work is based on one or more scientific articles either published or submitted. Among the possible catalytic reactions, the hydrogenation, oxidation and hydrodeoxygenation were investigated: for this purpose, several novel catalysts were synthetized, tested and characterized. The catalysts were made from precursor solutions with the incipient wet impregnation method or with the solution combustion synthesis, depending by the catalyst type. The conversion of glucose and some glucose-derivatives was typically performed under gentle operative conditions and in aqueous mean. Pt-based catalysts were tested for glucose conversion to adipic acid in a two-step process carried out in water solution without any pH control by investigating the effect of a series of supporting materials (active carbon, alumina, silica and ceria). The process consisted in the D-saccharic acid (SacA) production by D-glucose catalytic wet air oxidation followed by a hydrodeoxygenation treatment of SacA to adipic acid (AA) with the same catalyst. The main limit of using Pt for D-glucose oxidation is represented by the catalyst inhibition operated by the first product of the reaction, the gluconic acid (GluA), which prevents the consecutive reaction of SacA formation, but a deeper investigation of the reaction scheme allowed us to assess that over Pt/alumina the consecutive oxidation of gluconic acid to SacA is slightly favored under uncontrolled pH too.We have demonstrated that a 5.2 wt.% Pt on γ-alumina sample, the catalysts presenting the larger amount of strong Brønsted acid sites, was the best material for obtaining SacA (with a molar yield of about 13.5%); afterward, by performing the halogen-promoted hydrodeoxygenation of the resulting solution, the SacA was quantitatively converted into AA (and thus the overall adipic acid molar yield from glucose was about 13.0%). Effectively, The efficient conversion of common biomass derivatives, as D-glucose, into value-added chemicals has received a great deal of attention in the last few years. Several heterogeneous catalytic systems, characterized by noble metals, have already been investigated for the Catalytic Wet Air Oxidation (CWAO) of derived biomass. Nevertheless, the redox effect of such catalysts on biobased compounds has not been described in detail. In the present thesis, some perovskite type oxides (Fe, Co, Mn) that present high redox properties and stability under hydrothermal conditions have been tested to establish their ability to convert D-glucose into C6 aldaric acid, lactic acid (LacA) and levulinic acid (LevA). The influence of the reaction temperature, and the affinity of the catalysts to hydrogen and oxygen on the distribution of the liquid products have been investigated. In the best conditions, 50.3 mol.% and 69.5 mol.% of lactic and levulinic acid have been obtained by employing LaCoO3 and LaMnO3, respectively. Apart from the oxidative effect, which has led to several oxidation products, a high reductive effect of the catalysts has enabled the conversion of some key intermediates, such as pyruvic acid (PyrA) and hydroxymethylfurfural (HMF), into the desired products. LaMnO3, which has resulted to be the most oxidizable/reducible catalyst over a low temperature range, has shown the best performance of the studied perovskite type oxides; it has been found to promote the conversion of hydroxymethylfurfural to levulinic acid and to give the highest overall molar yield. Moreover, performing catalysts have been synthetized through incipient wet impregnation and tested for cis,cis-muconic acid (ccMA) hydrogenation to adipic acid. Before the hydrogenation, the investigation on the solubility of ccMA dissolution in different polar solvents has been carried out by characterizing and modelling the dissolution as a function of temperature. Water, ethanol, 2-propanol and acetic acid have been investigated as solvents in the range temperatures from a 298.15 to 348.15 K. Owing to the absence of ccMA solubility data, the reliability of the adopted experimental set-up was validated comparing published and experimental solubility data of a similar compound, that is, AA. From the results, it has emerged that the employed system is appropriate for the determination of molar fractions of an organic compound dissolved in polar solvents. The molar fraction and temperature were correlated using the Apelblat equation model, which is applied for the mathematic fitting of experimental data. A total relative average deviation of 3.54% was obtained for the experimental results and the solubility data obtained with the model, thus attesting the adequacy for this study. The use of Apelblat equation also allowed to determine the apparent molar enthalpy and molar entropy of dissolution. The dissolution of ccMA in water, ethanol, 2-propanol and acetic acid, over temperatures ranging from 298.15 to 348.15 K, has been shown to be endothermic. The activity of Pt-based catalysts has been compared with a Ni-based catalyst at a gentle condition. A supported 14.2 wt.% Ni on γ-alumina converted 100% of muconic acid, yielding 99.4 mol.% of AA. Finally, the oxidative cracking of 5-keto-L-aldonic acids to tartaric acid (TarA) was successfully performed at room temperature and atmospheric pressure in a carbonate buffer (pH = 10.34), by employing various V-based catalysts. The performance of the novel heterogeneous V-based catalysts was compared with the one of a conventional homogeneous system. The effect of the catalysts was obvious and 2%VOx/ZrO2 was found to be the best catalyst for the 5-keto-aldonic acids conversion to tartaric acid. The tartaric acid selectivity was equal to 74.5% and 44.3% starting from the 5-keto-D-gluconic acid (5kGl) and 5-keto-L-galactonic acid (5kGa), respectively. The best performances in terms of tartaric acid selectivity were obtained at the beginning of reaction, and about one fourth of the carbon moles were converted into tartaric acid after 24 h of reaction. The substrate was entirely converted after 24 h indicating that several by-products were also produced during the reaction. So, an overall reaction pathway was supposed and the effect of the vanadium structure to the catalytic activity was hypothesized. Moreover, the reaction mechanism of the 5-keto-aldonic acids conversion to tartaric acid promoted by the anchoring VOx-support bond was described.

There is a growing need to develop the processes to produce renewable fuels and chemicals due to the economical, political, and environmental concerns associated with the fossil fuels. One of the most promising methods for a small scale conversion of biomass into liquid fuels is fast pyrolysis. The liquid product obtained from the fast pyrolysis of biomass is called pyrolysis oil or bio-oil. It is a complex mixture of more than 300 compounds resulting from the depolymerization of biomass building blocks, cellulose; hemi-cellulose; and lignin. Bio-oils have low heating value, high moisture content, are acidic, contain solid char particles, are incompatible with existing petroleum based fuels, are thermally unstable, and degrade with time. They cannot be used directly in a diesel or a gasoline internal combustion engine. One of the challenges with the bio-oil is that it is unstable and can phase separate when stored for long. Its viscosity and molecular weight increases with time. It is important to identify the factors responsible for the bio-oil instability and to stabilize the bio-oil. The stability analysis of the bio-oil showed that the high molecular weight lignin oligomers in the bio-oil are mainly responsible for the instability of bio-oil. The viscosity increase in the bio-oil was due to two reasons: increase in the average molecular weight and increase in the concentration of high molecular weight oligomers. Char can be removed from the bio-oil by microfiltration using ceramic membranes with pore sizes less than 1 µm. Removal of char does not affect the bio-oil stability but is desired as char can cause difficulty in further processing of the bio-oil. Nanofiltration and low temperature hydrogenation were found to be the promising techniques to stabilize the bio-oil. Bio-oil must be catalytically converted into fuels and chemicals if it is to be used as a feedstock to make renewable fuels and chemicals. The water soluble fraction of bio-oil (WSBO) was found to contain C2 to C6 oxygenated hydrocarbons with various functionalities. In this study we showed that both hydrogen and alkanes can be produced with high yields from WSBO using aqueous phase processing. Hydrogen was produced by aqueous phase reforming over Pt/Al2O3 catalyst. Alkanes were produced by hydrodeoxygenation over Pt/SiO2-Al2O3. Both of these processes were preceded by a low temperature hydrogenation step over Ru/C catalyst. This step was critical to achieve high yields of hydrogen and alkanes. WSBO was also converted to gasoline-range alcohols and C2 to C6 diols with up to 46% carbon yield by a two-stage hydrogenation process over Ru/C catalyst (125 °C) followed by over Pt/C (250 °C) catalyst. Temperature and pressure can be used to tune the product selectivity. The hydroprocessing of bio-oil was followed by zeolite upgrading to produce C6 to C8 aromatic hydrocarbons and C2 to C4 olefins. Up to 70% carbon yield to aromatics and olefins was achieved from the hydrogenated aqueous fraction of bio-oil. The hydroprocessing steps prior to the zeolite upgrading increases the thermal stability of bio-oil as well as the intrinsic hydrogen content. Increasing the thermal stability of bio-oil results in reduced coke yields in zeolite upgrading, whereas, increasing the intrinsic hydrogen content results in more oxygen being removed from bio-oil as H2O than CO and CO2. This results in higher carbon yields to aromatic hydrocarbon and olefins. Integrating hydroprocessing with zeolite upgrading produces a narrow product spectrum and reduces the hydrogen requirement of the process as compared to processes solely based on hydrotreating. Increasing the yield of petrochemical products from biomass therefore requires hydrogen, thus cost of hydrogen dictates the maximum economic potential of the process.

<div>Biomass-derived fuels have long been considered as a possible replacement for traditional liquid fuels derived from petroleum. However, biomass as a feedstock requires significant refinement prior to application as a liquid fuel. The H2Bioil process has previously been proposed in which biomass is pyrolyzed and the resulting vapors are passed over a catalyst bed for upgrading to hydrocarbon products in a hydrogen environment [1]. A PtMo catalyst has been developed for the complete hydrodeoxygenation (HDO) of biomass pyrolysis vapors to hydrocarbons [2]. However, the product hydrocarbons contain a large fraction of molecules smaller than C4 which would not be suitable as liquid fuels. In fast hydropyrolysis of poplar followed by hydrodeoxygenation over a PtMo/MWCNT catalyst at 25 bar H2 and 300oC, only 32.1% of carbon is captured in C4 – C8 products; 21.7% of carbon is captured in C1 – C3 hydrocarbons [2]. Here, approaches are examined to increase selectivity of H2Bioil to desired products. Aldol condensation catalysts could be used prior to the HDO catalyst in order to increase the carbon number of products. These products would then be hydrodeoxygenated to hydrocarbons of greater average carbon number than with an HDO catalyst alone. Application of a 2% Cu/TiO2 catalyst to a classic aldehyde model compound, butanal, shows high selectivity towards aldol condensation products at low H2 pressures. In more complex systems which more closely resemble biomass pyrolysis vapors, this catalyst also shows significant yields to aldol condensation products, but substantial carbon losses presumed to be due to coke formation are observed. Both glycolaldehyde, a significant product of biomass pyrolysis, and cellulose, a component polymer of biomass, have been pyrolyzed and passed through aldol condensation followed by hydrodeoxygenation in a pulsed fixed-bed microreactor. Glycolaldehyde aldol condensation resulted in the formation of products in the C2-C¬9 range, while the major aldol condensation products observed from cellulose were C7 and C8 products. Carbon losses in glycolaldehyde aldol condensation were reduced under operation at increased hydrogen partial pressures, supporting the hypothesis that increasing selectivity to hydrogenation products can reduce coke formation from primary aldol condensation products. </div><div>The use of feeds which have undergone genetic modification and/or pretreatment by other catalytic processes may also lead to improvements in overall product selectivity. The influence of genetic modifications to poplar lignin on the pyrolysis plus HDO process are investigated, and it is found that these materials have no effect on the final product distribution. The product distribution from a poplar sample which has had lignin catalytically removed is also examined, with the conclusion that the product distribution strongly resembles that of cellulose, however the lignin-removed sample shows high selectivity towards char which is not seen from cellulose. </div><div><br></div>

Catalytic hydrogenation is one of the most important classes of reactions in industrial chemistry. Generally, these reactions are limited by the equilibrium thermodynamics and, especially when using H2 as a hydrogenation agent, the use of catalysts and heterogeneous catalysts is of paramount importance. Generally, hydrogenations are conducted in the gas phase or in the liquid phase, depending on the substrate to be hydrogenated. In this thesis, the focus was on liquid phase hydrogenations. Specifically, the thesis was divided into four main chapters: 1) a general introduction to hydrogenation reactions, reactors, thermodynamics, kinetics, catalysis and safety; 2) the direct synthesis of H2O2 by using membrane reactors; 3) the hydrodeoxygenation/hydrocracking of microalgae oils by using Ni/BEA (hierarchical) zeolites; 4) The hydrodeoxygenation of furfural to 2-methylfuran. All these products are of relevant industrial interest. Regarding the first research topic, about the direct synthesis of H2O2, three families of catalytic membranes have been prepared by dispersing Pd on asymmetric alumina membranes. By using 1) reduction with hydrazine in an ultrasonic bath, 2) impregnation-decomposition and 3) sol-immobilisation techniques, Pd nanoparticles (NPs) with different particle size distributions were prepared. The prepared catalytic membranes were tested for the direct synthesis of H2O2 in a membrane reactor operating in semi-batch in the presence of H2SO4 and, eventually, KBr as promoters. All the catalytic membranes were tested in the fully reduced state and/or after pre-oxidative treatments (calcination). The catalytic membranes have been characterised before, and after testing by using transmission electron microscopy (TEM), Temperature-programmed reduction (TPR) and diffuse reflectance infrared Fourier transform spectroscopy (DRIFT). Sol-immobilisation and impregnation-decomposition have been identified as promising preparation techniques. Compared to other reported procedures, catalytic membranes prepared by using the sol-immobilisation technique have shown the greatest selectivities and productivities in their reduced form. This was related to promotion effects active in the presence of polyvinyl alcohol, the capping agent used for obtaining colloidal Pd NPs in the case of the sol-immobilisation. However, during the tests, by using this technique, extensive deactivation phenomena were observed. In line with the extensive literature about the direct synthesis of H2O2, the calcination pretreatment led to improved selectivity and productivity. However, the results have shown a marked dependence on the Pd NPs size. The experimental results have been analysed by using several kinetic models in order to understand the origin of the deactivation phenomena observed for SI membranes, obtaining insights on the promotion effect of PVA and analysing the nature of the active sites for pre-oxidised catalytic membranes. For studying the HDO/HC of microalgae oils to Green Jet Fuel, four catalysts were prepared by using two commercial Bea zeolites supplied by Zeolyst (CP811E-75, CP814E) and a homemade SBA-15. The deposition of 8%wt of the Ni active metal was performed by incipient wetness. To minimise diffusional problems inside the pore of zeolites, the desilication treatment was performed on the commercial BETA zeolite CP811E-75 determining an increase of both, surface area and volume in the mesopore range. The prepared Ni-based catalysts were tested for methyl palmitate conversion in a batch autoclave reactor. Analysis of the results revealed the role of the surface acidity of the catalysts for the hydrogenolysis of methyl palmitate to palmitic acid, further evidencing the role of the greater available surface area and mesoporosity by the desilicated Ni/CP811E-75 in directing the selectivity toward the production of the C12 fraction. The third topic, the hydrodeoxygenation of furfural to 2-methylfuran was developed in collaboration with Avantium, The Netherlands. The hydrogenation of furfural is a very versatile reaction which can be used for obtaining a variety of high-value chemicals and potential biofuels. Between these potential biofuels, 2-methylfuran has been recently identified as high-value fuel/octane booster and has triggered an increasing the interest by the scientific community toward this reaction. The hydrogenation of furfural has been studied by many research groups, producing a large number of publications. However, the variation in the conditions used for this synthesis is huge, and a comparison with commercial catalysts is missing. Furthermore, the effect of the solvent has been only scarcely analysed. In this study, a large number of commercially available catalysts, comprising Cu, Ni, Pd, Pt, Ru, Rh dispersed on common supports or used as Raney catalysts were analysed using a high-throughput approach. The effect of the solvent and temperature was further analysed and discussed.

Research Doctorate - Doctor of Philosophy (PhD)<br>The utilisation of biomass has drawn widespread attention due to the fast depletion of fossil energy resources and environmental challenges. Biocrude oil, derived from pyrolysis or liquefaction of lignocellulosic biomass, contains a high level of oxygen leading to some detrimental properties (high viscosity, high corrosivity, low heating value and low thermal stability). Therefore, hydrodeoxygenation of the biocrude is necessary to remove the oxygen atoms and produce value-added fuels and chemicals. Hydrodeoxygenation (HDO) is a two-step process which involves the hydrogenation and deoxygenation. The hydrogenation commonly occurs in the presence of metal sites, and the acid sites play a crucial role in the deoxygenation. Therefore, the HDO catalysts are routinely composed of metals (such as Pt, Pd, Ru, and Ni) and supports (ZrO₂, Al₂O₃ and zeolites). Due to the complexity of biocrude oil, a model compound (guaiacol), which possesses two different oxygen-containing groups, was employed in this study. In the first step, the Ru/BEA and Ru/ZSM-5 with varied Si/Al ratios were studied to examine the influence of supports with varying acidity and pore size on HDO of biocrude oil and guaiacol in a batch-type reactor operated at 4.0 MPa hydrogen. It was observed that a decrease in the Si/Al ratio of the support generated an increase in the yield of cyclohexane and a decrease in the yield of 2-methoxycyclohexanol in HDO of guaiacol. Both Ru/BEA and Ru/ZSM-5, possessing low Si/Al ratios, displayed a high activity for HDO for guaiacol while only Ru/BEA catalyst exhibits a high activity for HDO of biocrude oil. Catalyst characterisation shows that the Ru/BEA catalyst, with a low Si/Al ratio, not only possesses strong B acid sites but also contains extensive mesoporosity. Notably, these mesopores appear to facilitate the hydrogenation, deoxygenation, and ring-opening of large oxygenated and condensed-ring hydrocarbons in biocrude oil which then leads to a high yield of cycloalkanes. As expected, the Ru/Al₂O₃ and Ru/SiO₂ catalysts exhibit a high hydrogenation activity but a low deoxygenation activity in the HDO of guaiacol and biocrude oil due to the absence of B acid sites. These results suggest that the larger pore support, with strong B acid sites, engendered the observed HDO activity. The reaction pathway for the main components of biocrude oil was proposed based on the observed reaction product distribution. Although Ru-based catalysts display a high HDO activity, the high cost could hinder their wide application in industry. Therefore, metallic Ni, which is low cost, non-sulfided, and has the advantage of high hydrogenation activity was employed as the main metal phase in this project. The influence of catalyst pore size and shape selectivity on the catalytic hydrodeoxygenation (HDO) of biocrude oil has been investigated by comparing the activity of nickel catalysts on the supports of different pore sizes towards model compounds of increasing dimension. Five model compounds (guaiacol, anisole, phenanthrene, pyrene, and 1,3,5-trimethoxybenzene), and five zeolite supports (small-pore ZSM-5, medium-pore MOR, large-pore Beta and Y, and mesoporous Al-MCM-41) have been investigated. Hydrodeoxygenation and hydrogenation activities were determined on the basis of the yield of the associated cycloalkanes. All catalysts show a high HDO activity for small molecules (anisole and guaiacol) while the catalysts with medium and large pore supports display high HDO activity for 1,3,5-trimethoxybenzene. Furthermore, the large pore catalysts (Ni/Y and Ni/Beta) exhibit high hydrogenation activity for phenanthrene, while only the extra-large pore size catalyst (Ni/Y) presents good hydrogenation and HDO activities for all model compounds. The mesoporous Ni/Al-MCM-41 catalyst shows low HDO and hydrogenation activities for large model compounds, which can be ascribed to its low metal dispersion and low concentration of acid sites. In addition, the Ni/Beta and Ni/ZSM-5 were also tested in HDO of biocrude oil. Ni/Beta catalyst displays a higher yield of cycloalkanes than that of Ni/ZSM-5, which confirms that the selection of catalyst support can have impact on the product distribution in HDO of biocrude. The BEA supported Ni catalyst, which possesses high concentration of acid sites and mesopores, displays a high HDO activity for guaiacol and biocrude oil, however, results from batch reactor cannot provide the information about turnover frequency of active sites. Therefore BEA zeolite, with different Si/Al ratios (12.5, 25, 175) and varying metal loadings (2.3 ~23.4 wt%), was studied in a flow reactor to examine the influence of catalyst acidity and the structure of Ni on the performance, with particular emphasis on the change in product selectivity. The ratio of cyclohexane formation rate to the concentration of acid sites in reduced catalysts was found to be roughly constant when HDO of guaiacol in a flow reactor over 15.7 wt% Ni/BEA catalysts with different acid site concentrations, demonstrating the deoxygenation activity increases with increasing number of acid sites. On the other hand, the materials with the majority of isolated Ni sites (prepared by ion exchange) showed no cyclohexane yield at 230 degree. However, at higher temperatures, the formation of cyclohexane was observed, as a result of a consecutive reaction where catechol was generated by acid sites and subsequently reduced to cyclohexane on isolated Ni sites. With respect to catalysts with higher Ni-loading, the selectivity towards cyclohexane increases with an increased Ni loading up to 15.7 wt%. This is attributed to the formation of larger Ni nanoparticles upon H₂ reduction. A higher concentration of nickel hydrides compared to isolated Ni sites was observed by H₂-TPD and H₂-FTIR. The nickel hydrides are believed to be crucial intermediates in the hydrogenation reaction. Based on the product distribution over catalysts containing mainly isolated Ni sites and the Ni nanoparticles, two different reaction pathways were proposed. Traditionally, catalysts with high metal loadings are indispensable for improved catalytic performance. Supported metal catalysts are typically prepared by incipient wetness impregnation, which inevitably gives rise to inhomogeneous metal distribution and leads to large metal particles formed via agglomeration. Therefore, highly dispersed Ni/BEA catalysts prepared via ion-exchange-deposition-precipitation (IDP) utilising careful pH control were conducted. In comparison, catalysts prepared by incipient wetness impregnation (IWI) and deposition-precipitation (DP) methods were also investigated and IDP catalysts were shown to have higher dispersion. A significantly increased selectivity toward hydrocarbons was observed over these carefully prepared catalysts. The presence of nickel hydrides was confirmed by H₂-TPD and H₂-FTIR. IDP catalyst exhibits a higher metal dispersion and higher concentration of nickel hydrides than impregnated and DP catalysts, while larger Ni nanoparticles formed in impregnated catalysts show a higher concentration of nickel hydrides per surface Ni. The guaiacol conversion was not significantly affected by the catalyst preparation method, while the product selectivity was altered. Higher cyclohexane formation rate was detected over IDP catalysts compared to DP and impregnated catalysts. Besides, cyclohexane formation rate presents a positive linear correlation with the concentration of nickel hydrides, suggesting nickel hydrides play a crucial role in the hydrodeoxygenation reaction. The factors affecting catalyst deactivation during HDO in a continuous-flow reactor were investigated in order to gain insight into the deactivation mechanism of BEA-supported Ni catalyst. Phenolic-OH group moieties accelerate catalyst deactivation through the production of condensed-ring compounds, which then leads to the blockage of pores. In contrast, the yield of cycloalkane did not change with time-on-stream when using toluene, cyclohexanol and anisole as feeds, suggesting aromatic-ring, alkyl-OH and aromatic-OCH₃ have negligible effect on catalyst deactivation. Operation at low weight hour space velocity (WHSV) values increase the yield of cycloalkane and reduce the rate of catalyst deactivation. High metal loadings can increase selectivity to cyclohexane, however, the production of condensed-ring products is also enhanced, which is likely to be the result of an increased concentration of surface cyclohexane carbocations. Furthermore, high metal loadings play a limited role in preventing the catalyst from deactivation. In addition, the activity and stability of catalysts were positively affected by an increase in reaction temperature (up to 230°C). Based on the product distribution observed, a coupling reaction pathway (leading to the formation of condensed-ring products and cycloalkanes) is proposed.