On March 30, 2023, the U.S. Environmental Protection Agency’s (EPA) New Chemicals Program (NCP) held the second of a two-part webinar series on EPA’s new standardized process to assess risk and apply mitigation measures, as appropriate, for mixed metal oxides (MMO), including new and modified cathode active materials (CAM). Under the Toxic Substances Control Act (TSCA), EPA reviews new chemical substances before they enter the marketplace to ensure they do not present unreasonable risk to human health or the environment. In October 2022, EPA announced an “innovative effort” to help make its review of new MMOs, including new and modified CAMs, more efficient. More information on the initiative is available in our October 11, 2022, memorandum.

As described in EPA’s slides, the webinar focused on “one of the most time consuming parts of the new chemical review process for MMOs CAMs,” the human health risk assessment. The risk assessment includes the following steps: understanding the chemical; understanding exposure; understanding hazard; and determining risk. NCP follows EPA’s Guidance for Risk Assessment. For risks to humans, EPA looks at risks to workers, general population, and consumers; risk through dermal, oral (drinking water and fish ingestion), and inhalation routes; and whether the chemical substance is a carcinogen.

EPA noted the following key factors in human health risk assessment:

• Determination of bioavailability;

• Use of a component-based approach for MMOs;

• Sources of hazard information;

• Derivation of points of departure (POD); and

• Use of a risk calculator.

EPA states that this general approach applies to CAMs that are MMOs typically reviewed by EPA as respirable, poorly soluble particulates according to the TSCA New Chemicals Category Document. Using an industry list of example metals likely to be used, EPA developed a scientifically robust approach to assess risks for MMOs used as CAMs before receiving the new chemical submissions. EPA noted that since it developed this approach based on its experience with previously submitted CAMs and the expectation that the new chemical submission will be doped with a narrow subset of metals, the first step is to compare the new chemical substance with past submissions to determine if the submission is appropriately similar based on the metal composition and the physicochemical properties.

Under its approach, EPA assumes that the individual metal components are 100 percent bioavailable. According to EPA, this is supported by confidential business information (CBI) data and the published literature. The approach is component-based. EPA did not consider an additive approach appropriate because it would compound conservative assumptions. EPA stated that nearly all individual components are well-studied. EPA also has CBI data for several MMOs. To derive the PODs, EPA developed a “straight forward” decision tree based on its experience with previously submitted CAMs.

Regarding the exposure routes of interest, the approach assumes that the individual metal components are completely bioavailable following either oral or inhalation exposures. EPA expects dermal absorption to be negligible. EPA noted that analogue data for MMOs that contain cobalt or nickel are positive for skin sensitization in submitted studies, however, and EPA will evaluate skin sensitization for MMOs that contain these metals.

During the webinar, EPA demonstrated a risk calculator tool that it developed. Once the engineering and exposure tabs are completed by the appropriate teams, the human health assessor enters information regarding the composition. Once completed, the risk calculator tool will generate the conditions necessary for the consent order. The slides include example risk statements for workers and for general population drinking water. Once the tool is available for stakeholders to use, the submitter will be able to see immediately where mitigation will be required.

According to EPA, the following testing information would be helpful for human health risk assessments:

• Biosolubility (all submissions);

• Specific target organ toxicity (all submissions);

• Reproductive toxicity (only for MMOs that contain cobalt, nickel, boron, or lithium);

• Genetic toxicity (only for MMOs that contain cobalt or nickel);

• Skin sensitization (only for MMOs that contain cobalt or nickel); and

• Pulmonary effects (all submissions).

EPA will use the standard approach unless data are submitted to fill the data gaps for metal-specific endpoints. Both the decision tree framework and risk calculator are tools that submitters can use to understand better the potential risk of their CAMs/modified CAMs and what information is needed before they submit their premanufacture notice (PMN). EPA will update its approach if it receives new information from submitters or if the science evolves.

The webinar ended with a question and answer period. Asked if the risk calculator is publicly available, EPA responded that the tools are currently waiting for clearances. Once cleared, they will be publicly available. Once the tools are available, if there is enough interest or if there seems to be a need, EPA will offer training. At this time, EPA is not planning to do so, however.

EPA stated that submitters can submit their own dopant bioavailability data. EPA currently does not have much data and is mainly using public data. If the submitter provides specific bioavailability data for each component, EPA will use that data. According to EPA, it has not received such data to date. The risk calculator could be used for other metal-based components aside from MMOs, with caution. According to EPA, the calculator does work for other CAMs that it has received that are not currently on the TSCA Inventory. EPA has not tested this approach for generic metal oxides, however. EPA recommends that submitters arrange a prenotice meeting. Identifying the specific bioavailability data that EPA wants to see is best answered in a one-on-one prenotice meeting. EPA acknowledged that it does not have much experience with bioavailability data and will need to think through that question. EPA stated that the data it wants would probably be similar to what it sees in the literature, with similar conditions.

A stakeholder asked EPA if it was true that it has no CAM POD data for nanosized new chemical substances. EPA responded that it has not seen good repeated dose data on nano MMOs. If the data are out there, EPA asked that someone let it know. According to EPA, there seems to be concern that nanosized MMOs have the potential for other effects because of migration to other tissues.

Asked if the assessment tool will shorten the timeframe for review, EPA responded that it has not yet used the tool in an assessment. The idea is that the hard decisions have already been made, so the human health assessment should not take very long. In response to a question about whether there are efforts to incorporate new approach methodologies (NAM) in these risk assessments, EPA stated that it is very interested in applying NAMs where it can. According to EPA, the most important thing is determining bioavailability. The metals are fairly well-studied, and EPA knows what happens if metals are bioavailable.

In response to a question about whether EPA will reject a PMN if no exposure or use information is provided, particularly since it already has exposure and use parameters, EPA stated that submitters are not required to provide that information and that EPA will not reject the PMN as incomplete. EPA will use conservative assumptions, however. Under this approach, the default assumptions and scenario documents that EPA uses have not changed. EPA will do the same sort of exposure and release assessments, and it will use data provided by the submitter. According to EPA, the uncertainty is largely on the engineering/exposure side. With MMOs, one of the most difficult things is how to deal with worker inhalation, such as what nominal protection factor (NPF) do workers need to wear and what are the dust controls. Any data that submitters can provide that will help avoid the use of defaults, which are conservative, will help EPA.

Commentary

We applaud EPA’s continued application of its CAM category approach and its ongoing outreach on the category’s application. This category reflects EPA’s interest in using categories.

For decades, EPA built its categories based on its extensive experience with groups of chemicals and applied those approaches to review expeditiously PMNs within the categories. Contrary to some mischaracterizations of the use of categories pre-Lautenberg, the category approach reflects and optimizes efficiently extensive EPA experience with similar substances. EPA’s stepped up reliance on this approach is commendable. It will help EPA to deploy efficiently its limited resources without compromising risk to health or the environment and enable a more efficient path for CAMs to proceed to market with minimal delay. This is essential in the development of new battery technologies.

EPA’s webinar reiterated its view of CAMs and provided some additional details. It reflects our experience with this class of PMNs. Potential submitters may well evaluate bioavailability of metals, but absent high-tier data (e.g., a 90-day inhalation study) that demonstrate a new CAM is substantially less hazardous than CAMs for which EPA already has similar studies, EPA will issue orders with very low exposure limits and strict release controls. CAMs that do not contain manganese, cobalt, and/or nickel might especially benefit from data generation because the profile of such substances may depart substantially from data supporting existing CAM substances. Submitters should consider carefully how its product is similar to or departs from other CAMs to develop a strategy to best inform EPA about potential hazards and the necessary control measures that will be imposed.