This post is the first of a number to come that will examine in some detail specific chemicals and chemical categories for which EPA has made questionable or flawed risk decisions under ChAMP. Many of these problems can be traced to EPA's near-exclusive reliance on incomplete or poor-quality data provided by manufacturers, or its need to resort to unsupported assumptions in the absence of sufficient data. For each posting, we'll summarize what is known about production and use of the chemical(s); describe EPA's hazard, exposure, risk and priority rankings; and then discuss why we question or disagree with EPA's decisions. First up: a category of three alkyl nitriles.
The alkyl nitriles category is comprised of three chemicals: propionitrile (CAS# 107-12-0), butyronitrile (CAS# 109-74-0), and isobutyronitrile (CAS# 78-82-0). Their annual production volumes range from 3 to 30 million pounds. EPA's 2006 Inventory Update Rule data list a single producer for each chemical; other companies may have claimed their identity to be confidential business information (CBI):
Aggregate 2005 volume (mil lbs)
(there may be others that have claimed their identity CBI)
10 to < 50
1 to < 10
Eastman Chemical Company
1 to < 10
Eastman Chemical Company
According to EPA based on data supplied by their manufacturers, these chemicals are used primarily as chemical intermediates in the manufacture of insecticides, other industrial chemicals and pharmaceuticals. However, other industrial uses as catalysts, dielectric fluids and solvents were also reported, as well as possible use of one of the chemicals as a gasoline additive. EPA states that no commercial or consumer uses of these chemicals have been reported by their manufacturers. Two of the chemicals, propionitrile and isobutyronitrile, are regulated as Extremely Hazardous Substances under the Clean Air Act.
Hazard rankings: Although these chemicals are not persistent or bioaccumulative, EPA found that they are toxic following repeated dosing and can cause developmental defects. As a result, EPA classified these chemicals as highly hazardous to human health. In contrast, EPA ranked these chemicals as low hazard to aquatic organisms, based on testing results in fish, invertebrates and aquatic plants submitted by the manufacturers.
Exposure rankings: EPA concludes that environmental exposures will be low even if there are releases, because the chemicals would rapidly break down. Information submitted by manufacturers under EPA's Inventory Update Rule (IUR) did not indicate any commercial or consumer uses, including in products intended for use by children. EPA relies on this information to conclude that exposures to consumers and children are low. Other evidence beyond the IUR suggesting uses in pharmaceuticals and gasoline is not considered by EPA in making these rankings. EPA ranks worker exposure potential as high, as these are volatile chemicals that could readily partition to air, if not contained.
Risk rankings: Despite the high hazard ranking, EPA's risk conclusions emphasize the anticipated low exposure to characterize human health risks for the general public, consumers and children as low. EPA ranks the potential risk to workers as high.
Prioritization ranking: Despite finding both high human health hazard and high potential exposures and therefore risks to workers, EPA considers the alkyl nitriles category to be of low priority. EPA bases this ranking primarily on the existence of a National Institute for Occupational Safety and Health (NIOSH) recommended exposure limit (REL), which EPA claims if implemented would manage risks to workers.
Why We Disagree:
1. EPA has no basis to assume that the mere existence of the NIOSH recommended exposure limit for one of these chemicals means that it is being used or complied with. The REL is not a regulatory standard and EPA provides no empirical evidence whatsoever to indicate the extent to which it is being met in workplaces handling these chemicals. EPA itself states the reason why RELs should not be considered in ranking exposure: "these limits are not enforceable" (emphasis added; see p. 16 of EPA's Methodology for Risk-Based Prioritization Under ChAMP). Yet it then proceeds to reduce the priority of a category of highly hazardous chemicals to low on that very basis!
Likewise, EPA invokes the mere existence of a NIOSH/International Program on Chemical Safety (IPCS) International Chemical Safety Card for a second of the three chemicals and a NIOSH recommendation for establishing a workplace exposure standard for nitriles – a recommendation issued in 1978 that has never been acted on by OSHA! – as justification its low-priority ranking.
2. While we agree that the human health hazard posed by alkyl nitriles is high, we question EPA's low ranking for eco-hazard. Readily available data not referenced by EPA, which we quickly found via a simple Google search, indicates that both NIOSH and the International Program on Chemical Safety (IPCS) indicates that at least two of these chemicals may be ecotoxic: one may be hazardous to birds (butyronitrile), and another to aquatic organisms (proprionitrile).
In contrast, EPA appears to have relied solely on data submitted by the manufacturers of these chemicals, despite the fact that IPCS is listed as one of the supplemental sources EPA claims to search for hazard data (see Appendix B of EPA's Methodology document). For at least one of these chemicals, proprionitrile, EPA clearly knew of the existence of the IPCS document – because it (selectively) cited its mere existence as a basis for its low-priority ranking (see point 1 above).
The extent of ecological hazard is an issue that needs to be resolved, not ignored as will result from EPA's low-priority ranking for these chemicals.
3. EPA gives too much credence to the unsubstantiated use and exposure information provided by manufacturers. For two of the three chemicals, EPA has publicly provided a description of the industrial processing information supplied by the manufacturer under the IUR; for the third, that information was provided to EPA but claimed to be confidential. By accepting the limited use information provided under the IUR at face value, EPA assumes that 100% of production goes into use as intermediates. Yet EPA cites evidence from other sources suggesting other uses of these chemicals.
4. EPA bases most of its low-exposure findings on the fact that alkyl nitriles are used primarily as intermediates to make other industrial chemicals as well as pesticides and pharmaceuticals. Yet EPA fails to provide any data or even to discuss the potential for the final products to contain unreacted starting chemical (termed "residual" though the amounts can be significant). If present, the final product could be a source of exposure to the starting chemical.
While industrial chemical products made using alkyl nitriles are regulated under TSCA, EPA points out that pharmaceutical and pesticide products are not. Regardless, all sources of exposure to TSCA-regulated chemicals used as intermediates – including via residuals present in products – need to be considered. There is no mention of the nature of the products produced or how they are used. Exposure potential to such residuals in pharmaceuticals is obvious. For pesticides, worker and environmental exposures can be expected, and if approved for residential use, consumers applying the pesticides and children playing nearby could be exposed as well.
Exposure to residuals of the highly hazardous alkyl nitriles in products may or may not be significant, but EPA's silence on this question effectively means it has simply ignored it and assumed that there will be no or low exposure via this route. And its final prioritization decision relegating this chemical to low priority effectively closes the book on such questions.
5. EPA uses its exceedingly weak exposure characterization to justify the decision to assign low human health risk rankings for the general public, consumers, and children. This could readily result in underestimating risk, rather than taking a health-protective approach, which is the appropriate outcome of a screening-level risk characterization.
As a consequence of the hasty risk decisions made by EPA, important questions regarding the actual risks posed by these chemicals will not be answered: Could environmental organisms really be harmed? Are there residual levels of these chemicals in consumer products? Are workers adequately protected?