Open Collections

Su, Jianping

University of British Columbia

Several oil refineries have repurposed their operations to “stand-alone” biorefineries that upgrade oleochemical feedstocks to produce lower carbon-intensive fuels. The refining of biogenic feedstocks produces a range of lower-carbon intensive fuels with the nature of the feedstock, the refinery configuration, the source of hydrogen, all contributing to the carbon intensity of the process and the final fuels. Policies have also encouraged some refineries to co-process biogenic feedstock rather than to repurpose. This has proven to be a less capital-intensive way of reducing the carbon intensity of the refinery and its products. Although, at this time, oleochemical feedstocks predominate as co-processing feedstocks, in the longer term, biocrudes are expected to predominate. The work primarily focused on co-processing of oleochemical feedstocks with the recognition that any results will be needed for future biocrude co-processing. Initial work assessed if a stable baseline could be established and whether a 6% co-processing ratio would allow any changes in the yields of the various process streams to be readily detected. However, the minimum changes detected indicated that the “signal” resulting from processing biogenic feedstocks was not strong enough to overcome the background “noise” of normal petroleum refining. Thus, to amplify detection, multiple linear regression models which made use of year-long commercial operational data, while normalizing critical changing variables, were developed. These models were successfully used to quantify the impact of co-processing lipid feedstocks on the yields of existing petroleum operations. However, it was apparent that the results from modelling were not as consistent as desired for all fractions. The difficulties in obtaining direct measurements highlighted the practical problems in sampling the various fuel streams. A combination of direct measurements and long-term process data provided the most practical solution, particularly for hard-to-measure fractions such as the “green coke” that was generated during co-processing of biogenic feedstocks. Although tracking the “green molecules” proved challenging, these methods will be needed to improve the quality of needed data (e.g., mass balance and hydrogen consumption) if we are to determine the carbon intensity of the various fuels produced as a result of co-processing and for the refinery to obtain “policy credits”.

Text

2022-03-02

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/80905

Forestry

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/