Search

In order for catalytic biomass pyrolysis to be economically sustainable, low cost and highly efficient catalysts are needed. In a previous work, Magnesium oxide (MgO) basic catalysts produced from natural magnesite (MgCO3) were found to be highly efficient alternatives to zeolitic catalysts which apart from their relatively high cost, they are easily deactivated due to biomass alkali deposition. In order to validate these findings in pilot scale, these natural MgO catalysts were investigated in a circulating fluidized bed pilot scale unit and were compared to a commercially available ZSM-5 catalyst. This is to the best of our knowledge the first attempt to evaluate a basic catalyst in biomass fast pyrolysis in a pilot unit employing commercially relevant process technology. The basic sites of the MgO catalysts enhanced ketonization and aldol condensation reactions, as it was verified by 2DGC-TOFMS analyses of the produced bio-oils. Deoxygenation was achieved mainly via formation of CO2, while H2O yield was substantially reduced in comparison to the ZSM-5 catalyst. As a result, bio-oils richer in hydrogen were obtained by the use of MgOs. However, the MgO catalysts led to a significant increase of the catalytically produced coke compared to ZSM-5. The effect of MgO properties, such as surface area and basicity on product yields and bio-oil composition was elaborated. In contrast to the acidic ZSM-5, no alkali metals were found to deposit on MgO, indicating different deactivation mechanisms between acidic and basic catalysts in biomass fast pyrolysis.

The “ one-pot ” hydrolyltic hydrogenation of cellulose towards C2-C6 sugar alcohols has been recognized as one of the most promising biomass valorization routes for the production of high added-value chemicals. In this work, we studied the performance of Ru and Pt catalysts supported on micro/mesoporous activated carbon, in the hydrolytic hydrogenation of microcrystalline and ball-milled cellulose, in neat water, at 180 °C and at relatively low hydrogen pressure of 2 MPa. The impact of metal loading (1–5 wt.%), metal reduction method (H 2 at 350 °C or NaBH 4 ) and acidification (sulfonation) of the AC support on cellulose conversion and selectivity to the various products were systematically addressed. It was shown that Pt is significantly more selective towards hexitols (sorbitol and mannitol) compared to Ru, in glucose-rich reaction media, such as those offered by the easily hydrolyzed ball-milled cellulose. For example, the 5wt%Pt/AC-SO 3 H catalyst afforded hexitols yield of 69.5 wt.% (at 96.1% conversion) compared to 10.9 wt.% (at 95.2% wt.% conversion) obtained by the corresponding Ru catalyst, the latter being also selective towards glycerol and propane-1,2-diol (propylene glycol). A relatively moderate metal loading, such as in 3 wt.%Ru/AC-SO 3 H, was more favorable for hexitols production (44.5 wt.% yield, at 94.8 wt.% conversion) with Ru catalysts. These results were also verified by glucose hydrogenation experiments that were conducted at the same experimental conditions. Both Pt and Ru exhibited relatively high glucose hydrogenation activity towards hexitols, versus retro-aldol reactions that lead directly to smaller C2–C4 compounds, while the difference in the final product yields between the two metals was attributed to the higher hexitols hydrogenolysis (C C cleavage) reactivity of Ru. HRTEM data showed the formation of metallic crystalline Pt and Ru nanoparticles (≤4 nm, depending on loading) as well as of amorphous oxygen-rich M(O) x δ+ phases, which were also confirmed by the XPS data. The presence of these phases which may be a source of acidity, as well as the bacisity of the parent AC used in this study, were mainly responsible for inducing isomerization, retro-aldol and dehydration reactions leading eventually to increased glycerol and propylene glycol selectivity, as was observed for both low-metal catalysts, i.e. 1 wt.% Pt or Ru/AC.

The thermal and catalytic fast pyrolysis of biomass aims at the production of pyrolysis oil (bio-oil), which can be utilized as a source of chemicals or as a bio-crude for the production of hydrocarbon fuels. We investigated low-cost, naturally derived basic MgO materials as catalysts for the catalytic fast pyrolysis of lignocellulosic biomass as alternatives to classical acidic zeolite catalysts. The MgO catalysts were produced from natural magnesite mineral without any significant treatment besides calcination, crushing and sieving. Their structure, composition, porosity, morphology and surface properties were thoroughly examined by XRD, XRF, N 2 porosimetry, SEM, TEM, TPD-CO 2 and TPD-NH 3 . The physicochemical characteristics of the MgO catalysts depended mainly on the different production conditions (duration and temperature of calcination). Despite their negligible acidity, the MgO catalysts effectively reduced the oxygen content of the produced bio-oil and exhibited similar or even better performance compared to that of an industrial ZSM-5 catalyst formulation ( i.e. non-catalytic pyrolysis: 38.9 wt.% organic bio-oil with 38.7 wt.% O 2 ; ZSM-5 based catalyst: 20.7 wt.% organic bio-oil with 30.9 wt.% O 2 ; selected natural MgO catalysts: 25.7 wt.% organic bio-oil with 31.0 wt.% O 2 or 21.1 wt.% organic bio-oil with 28.4 wt.% O 2 ). The basic sites of the MgO catalysts favored reduction of acids and deoxygenation via ketonization and aldol condensation reactions, as indicated by the product distribution and the composition of the bio-oil. Oxygen was removed mainly via the preferred pathway of CO 2 formation, compared to CO and water as in the case of ZSM-5 zeolite. On the other hand, reaction coke slightly increased over the MgO catalysts as compared to ZSM-5; however, the MgO formed coke was oxidized/burnt at significantly lower temperatures compared to that of ZSM-5, thus enabling MgO regeneration by relatively mild calcination in air. A systematic correlation of product yields and oxygen content of bio-oil with the physicochemical properties of the MgO catalysts has been established.

This study investigates the effect of active metal and the nature and pre-treatment of the support in a series of K-promoted NiMo catalysts supported on activated carbon (AC) for the hydrogenation reaction of CO. The Ni Mo synergistic effect was found to be essential for the activation of CO. The use of activated carbon as support led to a three-fold increase in the selectivity toward higher alcohols, compared to the unsupported material, probably due to a reduction in the crystallinity of the active NiMoO4 phase and its partial transformation from alpha to beta form. Pre-treatment of the activated carbon prior to its use as support with HNO3 aqueous solution led to significant increase in CO conversion, related to the lower amount of ash and the much higher acidity of the acid treated sample. The use of a higher surface area acid-treated support also increases activity, with the K NiMo/AC2a catalyst exhibiting a 45% higher conversion at 280 °C compared to K NiMo/AC1a. The improvement in activity can be ascribed to better dispersion of the active phase on the higher surface area AC support and therefore higher exposure of active Ni O Mo sites. Overall, the K NiMo/AC2a catalyst exhibited the optimum performance, with high CO conversion (25% at 280 °C) and high production of oxygenates, with a space time yield to oxygenates of 141.5 mg/gcatalyst/h at 280 °C. A clear tendency of increasing higher alcohol production with decreasing acidity was evidenced for all K-promoted bimetallic NiMo catalysts.

The main objective of the present work was the study of different ZSM-5 catalytic formulations for the in situ upgrading of biomass pyrolysis vapors. An equilibrium, commercial diluted ZSM-5 catalyst was used as the base case, in comparison with a series of nickel (Ni) and cobalt (Co) modified variants at varying metal loading (1–10 wt.%). The product yields and the composition of the produced bio-oil were significantly affected by the use of all ZSM-5 catalytic materials, compared to the non-catalytic flash pyrolysis, producing less bio-oil but of better quality. Incorporation of transition metals (Ni or Co) in the commercial equilibrium/diluted ZSM-5 catalyst had an additional effect on the performance of the parent ZSM-5 catalyst, with respect to product yields and bio-oil composition, with the NiO modified catalysts being more reactive towards decreasing the organic phase and increasing the gaseous products, compared to the Co 3 O 4 supported catalysts. However, all the metal-modified catalysts exhibited limited reactivity towards water production, while simultaneously enhancing the production of aromatics and phenols. An interesting observation was the in situ reduction of the supported metal oxides during the pyrolysis reaction that eventually led to the formation of metallic Ni and Co species on the catalysts after reaction, which was verified by detailed XRD and HRTEM analysis of the used catalysts. The Co 3 O 4 supported ZSM-5 catalysts exhibited also a promising performance in lowering the oxygen content of the organic phase of bio-oil.

A series of C 60 catalysts supported on MCM-41 was prepared in the context of developing a highly active heterogeneous catalytic system for the singlet oxygen-mediated oxidation of organic substrates. Successive incipient wetness impregnation was used for preparing 3, 6, 9 and 12 wt.% C 60 onto MCM-41 surface. N 2 adsorption, FT-IR, XRD, HR-TEM, DRS and TGA, have been used to characterize the photocatalysts. It was found that C 60 is highly dispersed (especially in the 3 and 6 wt.% C 60 catalysts) within the mesopores or on the external surface of MCM-41, as clusters or crystallites/aggregates of very small size and presumably in the form of atomic layers. The mesoporous structure and characteristics (surface area, mesopore size and volume) of MCM-41 remain almost intact upon deposition of C 60 on its surface. The photocatalytic activity was assessed in the photocatalytic oxidation of olefins via the singlet oxygen ene reaction by studying the photooxygenation of 2-methyl-2-heptene as a probe reaction. The catalytic tests were carried out at 0–5 °C in CH 3 CN under oxygen atmosphere and using a 300 W xenon lamp as the light source. To investigate further the potential applications of our catalysts we examined various types of photooxidations including the [4 + 2] or [2 + 2] cycloaddition, the heteroatom oxidation with sulfur or phosphorus containing compounds and the oxidation of phenols. The heterogeneous photocatalysts obtained were proved to be very active exhibiting higher substrate conversion, TON and TOF values, compared to those of the unsupported (solid) C 60 and the corresponding photocatalysts supported on silica or γ-alumina. This was mainly attributed to the better dispersion of C 60 on the MCM-41. The conversion achieved over the catalysts increases with the amount of the supported C 60 up to a value of 6 wt.% and then it decreases whereas the TON and TOF values decrease monotonically as the amount of the supported C 60 increases following the increase in the size of the supported C 60 crystallites/aggregates. Shadow effects were not observed under our experimental conditions.

In the present work we studied the N2O decomposition activity of nano-Ru/-Al2O3 catalysts prepared by in situ reduction with ethylene glycol (EG), in comparison to catalysts prepared via typical incipient wetness procedure. The metal content, particle size, dispersion and reduction characteristics of the supported Ru catalysts were studied by means of ICP-AES, XRD, HRTEM, TPR-H2 and TPD-H2. The EG reduction method favored the formation of almost spherical metallic Ru nanoparticles with the size of 1–3 nm (HRTEM) and dispersion of 70–35% (TPD-H2), at Ru loading levels of 0.4–1 wt.%. The impregnatedcalcined catalysts exhibited larger Ru nanoparticles (10–80 nm, TEM) with very low degree of dispersion (∼10%). The activity of the EG-prepared catalysts for the decomposition of N2O in oxygen rich feeds and in the absence or presence of NO, H2O, SO2, CO and CO2, was significantly higher compared to that of the impregnated-calcined samples and comparable with that of the most active catalytic systems reported in the literature for the treatment of combustion processes off-gases. The superior activity of the former catalysts, especially of the sample with low Ru content (0.4 wt.% Ru, 1–2 nm particle size), was attributed both to the high dispersion of Ru (70% dispersion), i.e., the high surface area of Ru metal available for reaction, and to the higher intrinsic activity and lower apparent activation energy of the very small supported Ru nanoparticles prepared by the EG reduction method. The EG-prepared nano-Ru/-Al2O3 catalysts could retain their high N2O decomposition activity (higher than 90%) for long time-on-stream by applying intermediate, short term regenerations in reducing atmosphere at moderate temperatures (i.e., 500

The effect of the pretreatment (inert, oxidative, and reducing) of Ru/γ-Al 2 O 3 catalyst on its activity and stability in the decomposition of N 2 O in the absence or presence of O 2 , SO 2 , H 2 O and NO X was studied in the present work. Decomposition of pure N 2 O was slightly enhanced by the H 2 -pretreated catalyst (metallic Ru) compared to the O 2 - or He-pretreated ones, owing to a cyclic oxidation–reduction pathway of metallic Ru. The observed decrease of activity by O 2 or H 2 O addition was reversible compared to SO 2 which caused a strong, irreversible deactivation of the catalyst, irrespective of the type of pretreatment. This was attributed to the formation of stable sulphates, mainly those on RuO 2 surface, which could only be removed by regeneration under reducing (H 2 in He) atmosphere at temperatures of ca. 500 °C. Oxidative or inert regeneration required very high temperatures (i.e. >700 °C) in order to decompose these sulphates. A method of retaining N 2 O conversion activity very high (≥98%) for long reaction times is suggested and is based on frequent and short-time (ca. 10 min) regenerations of the catalyst under reducing atmosphere (ca. 5% H 2 in He). The effect of co-feeding various reducing agents, such as CO or C 3 H 6 , on the N 2 O conversion activity in the presence of O 2 , SO 2 , H 2 O and NO X is negligible, mainly because they are oxidized at relatively low temperatures in the O 2 -rich feeds used in this study.