Human Health Risk Assessment for Biodiesel Production, Distribution and Use in Canada

Health Canada assessed the potential human health implications of the widespread use of biodiesel in Canada, considering the production, distribution, storage and use stages in the lifecycle of biodiesel fuel. The general approach employed is as comprehensive as the available information allows and is comparative in nature, i.e., the impacts of biodiesel blends are compared to those of conventional ultra low sulphur diesel (ULSD) and presented as relative risks and benefits. The primary consideration of this analysis is the potential impact of biodiesel use on mobile sector emissions and atmospheric concentrations of air pollutants.

The Government of Canada put in place a 2% renewable content requirement in diesel and heating oil on July 1^st, 2011, as outlined in the Regulations Amending the Renewable Fuels Regulations (P.C. 2011-795 June 29, 2011) and published in the Canada Gazette, Part II, on July 20, 2011. ^{Footnote 1} The Regulation does not specify the use of biodiesel fuel in distillate or heating fuels. Rather, any liquid fuel meeting the definition of renewable fuel as per the Regulation, produced from one or more of the designated feedstocks, and complying with the maximum content specified may be acceptable.^{Footnote 2}

Biodiesel is a mixture of fatty acid alkyl esters produced from vegetable oils and animal fats via transesterification with an alcohol (generally methanol). The combinations of fatty acids in fats and oils can vary substantially depending on the source material and influence the resulting biodiesel properties. Biodiesel is usually blended with ULSD and the resulting blends are denoted by BX, where X indicates the percent of biodiesel in a blend, on a volume basis (e.g., B5 is a blend comprised of 5% by volume biodiesel and 95% by volume ULSD). Biodiesel blends up to B20 can generally be used in most compression ignition engines without any modifications.

Biodiesel production facilities rely on technologies and processes that vary according to the type of feedstock and the level of integration or complexity of a facility. Biodiesel production activities can lead to a variety of emissions or releases to water, air, and soil. Risks and hazards associated with biodiesel facilities are common to other industrial sectors (e.g., combustion emissions, fugitive emissions, and spills) and can be limited by mitigation strategies, both behavioural and technological. The most common air emissions from biodiesel production plants are methanol (during transesterification), hexane (during oil extraction), and criteria air contaminants (CACs) (e.g., particulate matter (PM) emissions from fuel-powered generators). Total amounts emitted are expected to be relatively low and to meet regulatory requirements, based on information from the National Pollutant Release Inventory and environmental assessments of various Canadian biodiesel production facilities. Ambient air concentrations near facilities are expected to meet air quality guidelines. Emissions of heavy metals and air toxics are generally expected to be minimal, as specific activities that would lead to significant emissions of these pollutants have not been identified.

No extensive database or tool has been specifically developed to predict the environmental fate and transport of biodiesel releases, and empirical values for numerous physical and chemical properties of biodiesel fuel components are not available. Health Canada conducted screening level environmental fate and transport modelling of different biodiesel fuel spill scenarios to identify key potential impacts. The modelling results of neat ULSD, neat biodiesel, and biodiesel blends show that biodiesel fuel components are projected to travel less than ULSD fuel components, as expected based on biodiesel's physical and chemical characteristics, notably the greater biodegradation rate of biodiesel fuel components compared to diesel fuel fractions. Notwithstanding the modelling uncertainties (i.e., assumptions and input data), the limited mobility of biodiesel fuel components can be considered a benefit since the soil and groundwater contamination is expected to be relatively contained and, consequently, have less impact on environmental and human health than petroleum fuels following releases.

With regards to the use of biodiesel fuels in on-road heavy-duty diesel vehicles (HDDVs) and its impact on exhaust emissions compared to ULSD, the following trends are noted:

• considerable reductions in PM, CO, hydrocarbon, volatile organic compound, and polycyclic aromatic hydrocarbon (PAH) emissions;
• no net impact or a slight increase in NO_X emissions; and
• no significant impact on the efficiency of after-treatment devices.

For the current assessment, the impacts of biodiesel use on Canadian fleet-wide mobile source emissions were estimated with the MOBILE6.2C model in collaboration with Environment Canada. It was assumed that biodiesel affects emissions of on-road HDDVs only. Table ES.1 shows the percent change in fleet-average HDDV emissions estimated for B5 and B20 in comparison to ULSD, for 2006, 2010, and 2020.

Biodiesel is projected to have less impact on HDDV exhaust emissions in 2020 due to the turn-over of the Canadian HDDV fleet, as 2010 and beyond model-year HDDVs are equipped with new engine technologies and exhaust emission controls in order to meet more stringent emission standards.

Air quality modelling was undertaken to investigate the impact of biodiesel blends on air pollution in Canada. Specifically, MOBILE6.2C results of Canadian mobile source emissions for the basecase and biodiesel scenarios were used as input to the air quality modelling.^{Footnote 3} Photochemical modelling for the current project was conducted with A Unified Regional Air quality Modelling System (AURAMS) in collaboration with Environment Canada (see Table ES.2 for scenarios). National scenarios of biodiesel use were modelled on a 22.5-km grid covering the whole country. In addition, a 2-week high pollution episode was modelled at high resolution (3-km grid) over the Montréal region.

It is predicted that the national use of B5 and B20 under 2006 conditions would result in small (less than 1%) but non-negligible changes in air quality compared to ULSD use. In general, PM_2.5 and O₃ concentrations decrease in urban areas and increase in surrounding areas. CO concentrations are expected to decrease in all regions. For the 2020 projections, changes in predicted air quality are very small (less than 0.5%) and often close to model detection limits. Ozone and PM_2.5 concentrations are generally reduced in urban centres, but increase slightly in surrounding areas. CO concentrations are reduced in most areas. The smaller air quality impacts of biodiesel use in 2020 are due to the significant reductions in basecase fleet emissions in 2020 compared to 2006, as a result of the introduction of cleaner vehicles.

Short-term high resolution modelling of the Montréal urban area revealed similarly small changes in air quality. High-resolution modelling provided enhanced spatial resolution of air quality impacts, bringing to light different air quality phenomena caused by smaller scale meteorological regimes and a more detailed distribution of mobile emission sources, such as the impacts of major bridges and highways.

Health Canada's Air Quality Benefits Assessment Tool (AQBAT) was used to quantify Canadian morbidity and mortality risks/benefits from changes in CAC concentrations associated with the use of B5 or B20 compared to ULSD in the on-road HDDV fleet, in either 2006 or 2020. In 2006, annual B5 or summertime B20 use are associated with a reduction of about five to seven premature mortalities as well as minimal reductions in hospital admissions, emergency room visits and other morbidity outcomes, due primarily to minor reductions in PM_2.5 and O₃ levels. The health benefits associated with biodiesel use are expected to be reduced by 2020 due to the incorporation of new emission control technologies in the HDDV fleet.

Qualitative consideration of the available mobile source air toxics emissions data indicate that minimal reductions are expected for air concentrations of benzene, 1,3-butadiene, acetaldehyde, formaldehyde, acrolein and PAHs in association with the use of biodiesel, which may translate into very minor reductions in human exposure to these pollutants, particularly near roads that are heavily used by HDDVs. However, the emissions benefits and any associated reductions in human exposures are expected to diminish by 2020.

A toxicological review of biodiesel exhaust was conducted with two objectives: to determine if biodiesel exhaust has a similar, reduced or greater impact than diesel exhaust in terms of specific health effects; and to attribute any difference in the magnitude of effects observed (between biodiesel and diesel exhaust) to a change in the level of a specific physicochemical parameter(s) in the exhaust.

A review of several studies determined that biodiesel exhaust is unlikely to exceed diesel exhaust in terms of respiratory effects. Only two studies were reviewed that examined cardiovascular effects of biodiesel exhaust. Based on this limited data set, it was not possible to draw any conclusions as to how biodiesel and diesel exhaust compare with respect to cardiovascular effects.

A review of outcomes relevant to the initiation of carcinogenesis indicated that biodiesel and diesel exhaust are similar in terms of clastogenicity, biodiesel exhaust has a similar or lower effect on biochemical events (reactive oxygen species, inflammation) associated with genetic instability, and biodiesel is equal to or exceeds diesel with respect to cytotoxicity. The majority of studies investigating mutagenicity demonstrated that PM extract from biodiesel exhaust is potentially less mutagenic than diesel exhaust PM extract.

Only one inhalation study considered reproductive and developmental effects, neurological effects, and systemic effects resulting from exposure to biodiesel exhaust. Given that this study did not include a diesel treatment, it was not possible to draw any comparison between biodiesel and diesel exhaust. Dermal exposure to biodiesel was also considered because of potential exposure during refuelling. However, skin irritation, a potential outcome of this type of exposure, was not considered in the study reviewed.

No information was available for immunological effects resulting from exposure to biodiesel exhaust.

Regarding the second objective, it was determined that toxicological studies investigating respiratory, cardiovascular, and outcomes associated with initiation of carcinogenesis increasingly reflect efforts to ascribe differences in biological responses between biodiesel and diesel exhaust to differences in physicochemical characteristics between the two fuels. However, for most studies, differences in individual pollutant levels between biodiesel and diesel exhaust have not been specifically linked to changes in a given biological response.

A review was conducted to examine the risk that inhalation exposure of the Bovine Spongiform Encephalopathy (BSE) infectious agent may occur in the general population as a result of the combustion of biodiesel made from Specified Risk Material (SRM) derived tallow. The risk was considered negligible provided that SRM and tallow destined for biodiesel production are processed to achieve a tallow purity standard of not more than 0.15% insoluble impurity content, as per Canadian Food Inspection Agency directives. In a second scenario, in which the insoluble content of the SRM-derived tallow exceeds 0.15% and contains BSE agents, it is expected that biodiesel manufacturing and combustion processes would contribute to a reduction in the risk of inhalation exposure to BSE agents.

The potential for allergic reactions in the general population following inhalation exposure to exhaust from soy-based biodiesel was investigated due to the fact that soy is one the main foods eliciting allergic reactions. It was concluded that denaturation and hydrolysis of proteins during biodiesel production as well as purification processes are likely to reduce the allergenicity of biodiesel. In the event that allergenic proteins survive the latter processes, it is highly probable they would be destroyed during the combustion process given that temperatures in diesel engines are significantly higher than those which cause significant alterations in protein structure, thus eliminating the potential for allergic reactions.

A review of the major fuel additive categories that are likely to be used in biodiesel fuels in Canada was carried out. The review included key background and toxicity information for different types of additives as well as specific products. There is a relatively high level of uncertainty associated with additives due to the fact that it is difficult to predict which products will be used on a consistent basis in biodiesel blends and because there is relatively limited toxicological and exposure information available for these compounds.

Overall Conclusion

Although the scenarios examined in this assessment do not replicate specific existing Canadian biodiesel use policies, they were selected in order to provide an overall picture of potential health impacts of biodiesel use in Canada. Overall, the use of B5 or B20 nationally is expected to result in very minimal air quality and health benefits/risks, and these are likely to diminish over time. Although substantial modelling and data limitations remain, the currently available evidence suggests that the incremental health impacts associated with the widespread use of low level biodiesel blends in Canada as compared to the use of ULSD are expected to be minimal.

To obtain an electronic copy of the Human Health Risk Assessment for Biodiesel Production, Distribution and Use in Canada, please contact hc.air.sc@canada.ca. The report is also available for download.