Lab-scale fast-hydropyrolysis and vapor-phase catalytic hydrodeoxygenation for producing liquid fuel range hydrocarbons from intact biomass

Vinod Kumar Venkatakrishnan, Purdue University

Abstract

Liquid transportation fuels are primarily produced from petroleum-based non-renewable carbon sources. Sustainably available lignocellulosic biomass, as a renewable form of atmospheric carbon, could be utilized to produce hydrocarbon-based fuels with high energy density. One of the process options for this conversion is the H2Bioil process, where biomass is rapidly heated in a hydrogen environment to produce fast-hydropyrolysis vapors that are catalytically upgraded in downstream hydrodeoxygenation (HDO) to produce hydrocarbons. This process has been modeled to have high carbon and energy efficiencies of ~70% and ~75%, respectively. This dissertation presents the results of a lab-scale experimental proof-of-concept for the H2Bioil process for converting intact biomass to liquid fuel range hydrocarbons. Based on various prototype designs for high pressure (up to 68 bar) fast-pyrolysis in an inert environment, a cyclone-type fast-hydropyrolysis reactor system along with downstream vapor-phase catalytic HDO reactor was designed and constructed. A liquid chromatography-mass spectrometry based analytical technique was developed for quantitative compositional analysis of the cellulose pyrolysis liquid products. Levoglucosan and its isomers, cellobiosan, water and light oxygenates like formic acid, glycolaldehyde and hydroxyacetone are the major products of cellulose fast-pyrolysis. Increasing pyrolysis temperature in the range of 480 °C to 580 °C was found to increase the formation of light oxygenates, due to the increase in thermal cracking, and to decrease carbon recovery in the liquid. Comparison of cellulose fast-pyrolysis and fast-hydropyrolysis experiments showed that H2 does not play an important role in deoxygenation even up to 50 bar H 2 partial pressures in the absence of a downstream HDO catalyst. Candidate catalyst screening and previous work from our research group revealed that adding an oxophillic promoter, such as Mo, along with the hydrogenation function of Pt, could increase C-O bond scission. Hence, a 5wt%Pt-2.5wt%Mo catalyst supported on multi-walled carbon nanotubes (MWCNT) was tested for HDO of fast-hydropyrolysis vapors from cellulose, as a model biomass feedstock, and poplar, as a real biomass feedstock. The total C1-C8+ hydrocarbon yield (as % carbon of feed) with cellulose was ~73%, the liquid fuel range (C4+) hydrocarbon yield was ~55%, with a major fraction as C6 hydrocarbons from the HDO of levoglucosan and its isomers. The total C1-C8+ hydrocarbon yield (as % carbon of feed) with poplar was ~54%, and the liquid fuel range hydrocarbon yield (C4+) was ~32%, with a major fraction as C8+ hydrocarbons from the HDO of lignin fragments. Increasing the HDO temperature from 300 °C to 350 °C increased the C-C bond scission and led to higher yields of CO and lower yields of C4+ hydrocarbons. Independent control of fast-hydropyrolysis and HDO temperatures in the H2Bioil process helps in improving the overall C4+ hydrocarbon yields. For improving the overall carbon efficiency from the experimental proof-of-concept of the H2Bioil process, synergistic process integrations, involving gasification, combustion and reforming, have been suggested within the group for utilizing carbon from CO, char and C 1-C3 hydrocarbons to increase the yield of liquid fuel range (C4+) hydrocarbons.

Degree

Ph.D.

Advisors

Agrawal, Purdue University.

Subject Area

Chemical engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS