472304 Selective Hydrogenation of Bio-Oil Model Compounds over Molybdenum Carbide Supported Catalysts

Monday, November 14, 2016: 12:45 PM
Franciscan B (Hilton San Francisco Union Square)
Sarah W. Paleg1, Joshua Schaidle2 and Levi T. Thompson1, (1)Department of Chemical Engineering, University of Michigan, Ann Arbor, MI, (2)National Bioenergy Center, National Renewable Energy Laboratory, Golden, MI

The increasing complexity and cost of drilling for oil, the need to decrease global net CO2 emissions, and the desire to continue using liquid transportation fuels have motivated research to produce liquid fuels from biomass. Fast pyrolysis is a thermochemical conversion process in which biomass is reacted at high temperatures (near 500°C) at atmospheric pressure for very short residence times (on the order of seconds) in the absence of added molecular oxygen. Bio-oil from conventional fast pyrolysis is corrosive, has poor heating value, is thermally unstable, cannot be blended with conventional fuels without the addition of surfactant, and can become very viscous with aging[i]. Selective hydrogenation can improve the properties of bio-oil by increasing its thermostability and energy density, and potentially reducing acidity and therefore improving the ability to blend it with conventional fuels.

1H and 13C NMR analyses of pyrolysis bio-oils have shown that typical functional groups include aldehydes, ketones, hetero-aromatics, aliphatics, methoxys, alcohols, alkenes, dibenzenes, esters, benzyls, esters, and carboxylic acids[ii]. Considering the molecular character of pyrolysis bio-oil, crotonaldehyde was used as the model compound for work described in this paper as it is a simple aldehyde and can be used to probe a variety of different reaction pathways. The catalysts are based on Mo carbide (Mo2C). Molybdenum carbide is active for a variety of hydrogenation reactions including the hydrogenation of CO to hydrocarbons (Fischer-Tropsch reaction)[iii] and naphthalene to tetralin[iv]. In addition, this material can possess, depending on preparation and treatment conditions, multiple types of surface sites (e.g. acidic, basic and metallic sites)[v],[vi],[vii] and is stable under conditions of fast pyrolysis5. Transition metal and alkali promotion of Mo2C has been shown to cause shifts in the selectivity for a number of reactions, including shifting the selectivity for CO hydrogenation (Fischer-Tropsch reaction) to alcohols and heavier hydrocarbons, depending on the promoter[viii],[ix]. The goal of this work was to understand how metal promotion of Mo2C-based catalysts could be used to shift selectivity of crotonaldehyde hydrogenation.

The Mo2C catalysts were synthesized by a temperature-programmed reaction method as described previously[x]. Metal salt was added to the native (unpassivated) Mo2C catalyst by incipient wetness in an oxygen-free inert atmosphere. Then, the material was dried at 110°C for 2 hrs in H2, reduced in H2 at 450°C for 4 hrs, quenched to room temperature, and passivated in 1% O2/He for at least 6 hrs. All of the catalysts were characterized by x-ray powder diffraction, N2 physisorption, and temperature programmed desorption of NH3 and CO2. The crotonaldehyde hydrogenation experiments were performed in a flow reactor system equipped with a bubbler to introduce vapors of crotonaldehyde, and the reactor effluent was analyzed with a gas chromatograph equipped with flame ionization and thermal conductivity detectors. The catalyst was loaded into a quartz u-tube reactor and supported on quartz wool. Prior to the rate and selectivity measurements, the catalysts were reduced in 15% CH4/H2 for 4 h at 590 °C. All reactions were performed between 275-350°C in atmospheric pressure. The reaction mixture consisted of 0.7 mol% crotonaldehyde, 4.2 mol% H2, and 95.1% N2. The H2/acetic acid molar ratio (6) was chosen to achieve approximately 2 times the stoichiometric amount of H2required for complete saturation and deoxygenation of crotonaldehyde to butane.

The unpromoted Mo2C favored formation of butyraldehyde and butene (ca. 40% each). Since butene is believed to be a secondary hydrogenation-dehydration product of crotyl alcohol[xi], there is not a clear preference for a specific pathway on the unpromoted catalyst. Selectivities to butyraldehyde, butene, and butadiene decreased with increasing potassium coverage, while selectivity to furan increased with potassium coverage. At coverages equivalent to 50% of a monolayer (ML), the furan selectivity was ca. 80%, while the selectivities to the other three products were each less than 10%. There was no significant further increase in the selectivity upon increasing the potassium loading, suggesting that the effect of potassium promotion was maximized at ~50% ML.

Alkali promotion is known to affect the acid-base properties of a catalyst, and acid-base properties are also known to affect selectivity6,[xii]. Acid and base site densities of the catalyst surfaces were quantified via temperature-programmed desorption of NH3 and CO2, respectively. While acid site densities decreased with increasing potassium loading, base site density increased with potassium promotion. The highest base site density (~4.3x1014 molecules/cm2) was for the K/Mo2C catalyst with 1.1ML-equivalent coverage of potassium; the highest acid site density was on the bulk Mo2C and was 1.5x1014 molecules/cm2. There was a strong correlation between the furan production rate and the base site densities, implicating base sites the location of the rate-determining step. Furan appears to be produced from crotyl alcohol intermediate, indicating that potassium promotion enhanced the hydrogenation of crotonaldehyde to crotyl alcohol.

The model compound results presented in this paper demonstrate the ability to tune the product selectivity of molybdenum carbide catalysts by adding potassium. Tuning product selectivity may prove invaluable in upgrading and stabilizing bio-oils. Performing simultaneous carbon-carbon coupling and hydrogenation-dehydration reactions could ultimately lead to an upgraded bio-oil with lesser oxygen content, greater hydrogen content and a higher average carbon number.

[i] (a) Czernik, S.; Bridgwater, A.V.; Energy & Fuels, 2004, 18, 590-598; (b) Zhang, Q.; Chang, J.; Tiejun, W.; Yung, X. Energy Conversion and Management, 2007, 48, 87-92.

[ii] Mullen, C.A.; Strahan, G.D.; Boateng, A.A. Energy & Fuels, 2009, 23, 2707-2718.

[iii] Schaidle, J.A.; Thompson, L.T. Journal of Catalysis, 2015, 329, 325-334.

[iv] Pang, M.; Liu, C.; Xia, W.; Muhler, M.; Liang, C. Green Chemistry, 2012, 14, 1272-1276.

[v] Bej, S. K.; Bennett, C. A.; Thompson, L. T., Applied Catalysis A: General, 2003, 250, 197-208.

[vi] Bej, S. K.; Thompson, L. T., Applied Catalysis A: General, 2004, 264, 141-150.

[vii] Sullivan, M. M.; Held, J. T.; Bhan, A., Journal of Catalysis, 2015, 326, 82-91.

[viii] Park, K.Y.; Seo, W.K.; Lee, J.S. Catalysis Letters, 1991, 11, 349-356.

[ix] Griboval-Constant, A.; Giraudon, J.-M.; Leclerq, G.; Leclercq, L. Applied Catalysis A: General, 2004, 260, 34-45.

[x] J.A. Schaidle, N.M. Schweitzer, O.T. Ajenifujah, L.T. Thompson, J. Catalysis 289 (2012) 210-217.

[xi] Zanella, R.; Louis, C.; Giorgio, S.; Touroude, R. Journal of Catalysis, 2004, 223, 328-339.

[xii] Kotarba, A.; Adamski, G.; Piskorz, W.; Sojka, Z.; Sayag, C.; Djega-Mariadassou, G. D. Journal of Physical Chemistry B, 2004, 108, 2885-2892.

Extended Abstract: File Not Uploaded