Fundamental studies of biomass fast pyrolysis for the direct production of molecules in the fuel range

Piotr Gawecki, Purdue University

Abstract

Worldwide depletion of oil reserves has long been a concern, and has sparked the drive to find alternative sources of energy, particularly for the transportation sector. The United States is the world's largest per capita consumer of energy, and petroleum accounts for more than 1/3 of all energy usage in the US; over 90% of transportation fuel is derived from petroleum. Thus, addressing the issue of declining resources is of paramount importance in this country. Conversion of biomass to fuels and value-added chemical has long been viewed as a viable solution to offset the need on imported oil, particularly for the light-duty vehicle fleet. Of the possible routes to accomplish this goal, we believe the thermochemical approach of pyrolysis to be best suited for the task. Pyrolysis is a promising approach for making fuels and value added chemicals from alternative sources, such as biomass, because it keeps most of the bonds intact, occurs at 400-600°C, and requires relatively simple reactors. However, the typical pyrolysis product, a viscous liquid referred to as bio-oil, has many undesirable qualities. Its high oxygen and water content give it a heating value that is half that of gasoline, it contains thousands of compounds that degrade over time, and it typically has a pH below 3. Storing such a mixture for an extended period of time allows the components to react with each other, and thus the bio-oil properties vary with time and the temperature at which it is stored. To upgrade the low energy density bio-oil to a fuel, it is necessary to remove as many oxygen atoms as possible while retaining the carbon atoms. Our group has proposed an integrated process to carry out pyrolysis followed by a catalytic cascade for the direct production of liquid fuels. We started off by developing a continuous solids feeder capable of operating at pressures up to 50 bar; this was incorporated into an optically-accessible continuous pyrolysis reactor operating at atmospheric pressure, with the aim of determining optimal flow characteristics inside the reactor. Using this setup, we were able to obtain a bio-oil, which was found to contain thousands of compounds, and was difficult to analyze using inhouse techniques. With this information, we turned our focus to a micro-scale batch pyrolyzer to determine the optimum operating conditions for creating a simpler product distribution. Using this apparatus, we discovered that for 50 μm cellulose particles, a heating rate of 1,000°C s-1 eliminates char formation when the sample is well dispersed, which is information that has since been adapted into the newest iteration of a continuous reactor in our laboratory. Interested in being able to monitor the pyrolysis products formed in real time, we developed a novel analytical method that employs atmospheric pressure chemical ionization and a linear quadrupole ion trap mass spectrometer to gently ionize all the fragments formed during cellulose pyrolysis. We discovered that the primary products of cellulose are a simple mixture of products closely related to the monomeric building block of cellulose, with the most abundant product a monomer with mass 162 Da. By modifying the setup slightly, we discovered that we are able to systematically manipulate the product distribution, which results in changing the monomer:dimer:trimer:tetramer ratio, by varying the surrounding gas temperature and residence time in the heated zone. These results have been verified using two continuous reactors, in which we were able to produce a simple bio-oil. Utilizing the versatility of the mass spectrometer, we were able to determine the presence of multiple isomers at 162 m/z, contrary to the generally-accepted view in literature of only one compound, levoglucosan, being formed during pyrolysis. Work is underway to determine the exact structures, as well as mechanisms to obtain those structures, and involves a joint computational and experimental approach. We have begun extending the knowledge gained for cellulose pyrolysis to real biomass systems, including novel plant mutants differing in the types and amounts of lignin, cellulose, and hemicellulose generated by our collaborators, in an effort to develop an optimized fuel crop.

Degree

Ph.D.

Advisors

Agrawal, Purdue University.

Subject Area

Alternative Energy|Chemical engineering|Environmental science

COinS