(356g) Hydrodeoxygenation of Guaiacol with Rh Based and CoMo and NiMo Catalysts
AIChE Annual Meeting
Tuesday, October 18, 2011 - 5:15pm to 5:35pm
Lignin is the most abundant nature-made polymer in the plant kingdom. Depolymerization of lignin can produce considerable amount of guaiacyl species such as guaiacol (GUA), vanillin, and euginol. This leads these species, especially GUA, to be a common lignin representative. An efficient upgrading process of GUA is deemed to be a potential route for lignin upgrading.
Hydrodeoxygenation (HDO) is a process by which hydrogen acts as the reductant for deoxygenation, usually operated in a batch or a continuous system with catalysts. Under H2-pressured hydrothermal conditions, oxygen can be removed in the form of water. Therefore, the upgraded products contain less oxygen, rendering them with a higher heating value and chemically more stable than their unrefined state.
This study investigates the GUA HDO chemistry. Two catalysts, Rh based catalysts (Rh, PtRh, and PdRh on ZrO2) and sulfided CoMo- and NiMo/Al2O3 were used. All of them were synthesized through the incipient wetness approach. Brunauer, Emmet, and Teller (BET) method, powder X-ray diffraction (XRD), temperature programmed reduction (TPR) and desorption (TPD) were used for catalyst characterization. Elemental analysis (EA) using combustion method was applied for sulfur estimation. Bimetallic PtRh and PdRh were found to be more reducible than monometallic Rh/ZO2. Properties of sulfided CoMo- and NiMo/Al2O3 were very similar to conventional hydrotreating catalysts.
An in-house designed, batch-type high-throughput system was employed for catalytic testing. About 0.23 g of GUA and 7.63 g of tetradecane were employed to set the reactant-to-solvent ratio close to 3 wt%. All trials were operated under a hydrogen pressure of 50 bar. Two sets of comparative tests were performed: 1) at 400 oC for 20, 40, and 60 minutes, and 2) at 300, 350, and 400 oC for 60 minutes. The experiments in this study identified eleven major species. Rh-based catalysts produced 2-methoxycyclohexanol, 2-methoxycyclohexanone, 1-methoxycyclohexane, cyclohexanol, and cyclohexanone. Sulfided CoMo and NiMo produced methoxybenzene, methylphenol, phenol, benzene, and cyclohexene. Both systems yielded cyclohexane.
Under the same reaction condition, Rh based catalysts were more active than sulfided CoMo- and NiMo/Al2O3. Among the Rh-based catalysts, the monometallic Rh catalyst generated the highest yield of cyclohexane (~42%) in the final stage. Moreover, the mono-oxygen species had the highest yield. The same trend appeared by illustrating carbon yields as a function of reaction time. This indicates that the monometallic Rh catalyst possessed the highest HDO reactivity of Rh-based catalysts. There is no correlation between activity and reducibility of Rh-based catalysts. The product distribution of GUA HDO by sulfided CoMo- and NiMo catalysts were similar. The two-oxygen containing species generated by the noble metal catalysts were absent. Moreover, significant amount of coke (~20%) were identified.
Plausible GUA HDO mechanisms by these two types of catalysts were thereby proposed. The first step of Rh-based catalysts was hydrogenation of GUA’s benzene ring, followed by demethoxylation and dehydroxylation. As for the sulfided CoMo and NiMo catalysts, GUA HDO began with demethylation, demethoxylation, and deoxygenation, followed by benzene ring saturation. This likely explains the difference in product distributions of these catalysts.