Page 131 - Advances in bioenergy (2016)

remains a problem, as up to approximately 30% of the carbon initially present in the feedstock
has been found to form coke. The catalyst must be regenerated in order to be reused and unless
the coke can be burned off in a way that produces useful heat, it represents carbon that is
wasted. The liquid products of this type of catalytic pyrolysis separate into a carbon-rich
phase (which usually contains about 10% oxygen) and an aqueous phase. A significant amount
of carbon is usually still dissolved in the aqueous phase, and may be lost. These considerations
reduce the economic viability of this approach.

It is possible to upgrade condensed liquid pyrolysis oil, eliminating oxygen from the molecular
structure of the species in the oil. 11-13 It may also be possible to produce a significant yield of

liquid containing far less oxygen. In practice, catalytic hydrodeoxygenation is required,
utilizing pressurized hydrogen. Through hydrodeoxygenation, oxygen can be removed from the
oil, which reduces the ability of compounds in the oil to react with one another and creates oil
that is more similar to that derived from traditional petroleum feedstocks. However, this
approach is also problematic. Work carried out with solid catalysts, gaseous hydrogen, and
pyrolysis oil in pressurized reactors has demonstrated that partial pressures of hydrogen up to
13
2500 psig (170 barg) are required for the desired reactions to proceed. Undesirable
rearrangements and polymerizations of molecules in the oil occur simultaneously with the
desired deoxygenation reactions, and coke formation may rapidly deactivate the catalyst. The
presence of liquid water in pyrolysis oil further complicates this approach, as catalysts and

their supports must be capable of tolerating exposure to water.
14
Recently, a major developer of petroleum-related technologies worked to evaluate the
potential for utilizing biomass-derived feedstocks in petroleum refineries. Two classes of
materials were examined: vegetable oils and pyrolysis oils derived from forest waste. The
former can be readily hydrotreated to form ‘green diesel’ (which consists of hydrocarbons
without the oxygenates found in biodiesel). Green diesel was found to be economically
attractive only if derived from a low-cost feedstock, such as brown grease, or if vegetable oil
feedstocks were subsidized. This approach has been implemented on a commercial scale, but
the oils required are not available from lignocellulosic feedstocks. The authors reported that
economically attractive hydrocarbon fuels can be generated from the pyrolysis of a
lignocellulosic feedstock only if a significant number of practical and economic hurdles are
overcome, particularly the above-discussed ones.