On October 18, 2022, the U.S. Environmental Protection Agency’s (EPA) New Chemicals Program held a webinar on EPA’s process for assessing the potential risks of new chemicals under Section 5 of the Toxic Substances Control Act (TSCA) and the types of data EPA considers in this assessment. Specifically, the webinar covered examples of data (quantitative and qualitative) that are not likely to be accepted for engineering assessment, considerations EPA makes when evaluating data, and clarifications of common misconceptions in EPA’s new chemical assessments. The webinar was the second in a series of webinars intended to increase the efficiency and transparency of EPA’s new chemical determinations. EPA will hold a third webinar to communicate additional information intended to help submitters of new chemicals supplement complete initial review submissions. EPA will post information on the third webinar as it becomes available. In July 2022, EPA hosted the first webinar, analyzing common issues that cause EPA to have to rework risk assessments. More information on the first webinar is available in our July 28, 2022, memorandum. The slides for the second webinar are available online.

Ritesh Tiwari, EPA, provided an overview of the engineering assessment that EPA prepares for new chemical substances (NCS). The scope of the assessment covers the complete lifecycle of the NCS, including manufacturing (including importing), processing, and industrial and commercial use (until the NCS is no longer available for release and exposure). EPA’s initial review engineering report (IRER) estimates environmental releases (to air, water, incineration, and/or landfill) and occupational exposures (worker exposure via inhalation and dermal routes). Tiwari emphasized that new chemical assessments intended to be screening-level assessments represent a high-end exposure scenario (90th percentile or above of the expected distribution).

For most release sources and activities, EPA has standard models that it uses. Tiwari stated that it might surprise stakeholders that EPA prefers to use information from the submitter or claims based on physical-chemical properties that are found to be acceptable after EPA’s evaluation. According to Tiwari, submitted information may fall into the following categories:

• Quantitative Information: This type of information is numerical in nature. Examples include quantity of chemical released from a source activity or the duration and frequency of release and exposure; and

• Qualitative Information: This type of information is descriptive in nature. Examples include general statements regarding release/exposure potential of NCS, description of engineering control, and media of release.

Tiwari stated that EPA wants to draw attention to some gross inadequacies that it sees. These types of statements are unlikely to be accepted for the engineering assessment (emphasis in EPA’s slides):

• Information less conservative than standard EPA models, Generic Scenarios (GS), or Organization for Economic Cooperation and Development (OECD) Emission Scenario Documents (ESD) without substantiation or supporting documents;

• Claims for a site not controlled by the submitter without substantiation and supporting documentation from the third party;

• Claims for a large number of unknown customer sites not under submitter control;

• A single data point provided, or multiple data points provided without supporting details (e.g., sampling/test method, equipment description, worker activity description); and

• Claims based on engineering control to be installed in the future.

According to Tiwari, when evaluating quantitative data, EPA considers whether data are reliable, representative, and accessible and clear or variable and uncertain. When evaluating qualitative data, EPA is unlikely to accept claims that are not substantiated with supporting evidence. Tiwari stated that examples of helpful supporting documents include:

• For a site controlled by the submitter: Visuals, site operation documents, and claims based on the NCS physical-chemical properties with supporting test data detail; and

• For a site NOT controlled by the submitter: Third-party visuals, operation documents, customer notification, and claims based on physical-chemical properties with supporting test data details.

Tiwari noted that it is unlikely that EPA will accept claims or statements on waste disposal methods at customer facilities if the submitter does not provide supporting documents such as how it will notify customers about how to manage wastes containing the NCS.

If a submitter does not provide relevant information, or does not include adequate substantiation, EPA will apply conservative assumptions and estimate releases and exposures using Chemical Screening Tool for Exposures and Environmental Releases (ChemSTEER) models, OECD ESDs, and/or GS. According to EPA, GS/ESD provide conservative, screening-level estimates of environmental releases and worker exposures for specific industry sectors or exposure scenarios. Tiwari noted that some estimates may result in release/exposure amounts that are likely to be higher, or at least higher than average, than amounts that actually occur in real world practice. EPA also searches prior new chemical submissions (i.e., premanufacture notices (PMN)/low volume exemptions (LVE)) and may use relevant information from similar past assessments.

Tiwari clarified the following common misconceptions:

• EPA’s assessment covers the entire lifecycle of the NCS, beyond manufacturing and import;

• Engineering assessment focuses on the NCS and not on the bulk material (if the NCS is formulated with other substances);

• EPA considers process activities conducted at elevated temperatures that may lead to releases and exposures not otherwise anticipated;

• EPA considers whether solid materials that are not typically airborne as manufactured would present inhalation exposure potential due to attrition (generation of smaller particles) during transportation; and

• When supporting information is provided, EPA evaluates the information for acceptability before applying it to the assessment.

During the engineering pre-screen process, EPA reviews submissions to ensure that they are complete. If information is not available, the submitter should indicate that it is not known or reasonably ascertainable. During pre-screening, EPA does not confirm that supporting information is provided for every statement made in the submission. EPA conducts a more detailed review of the information when performing its assessment.

The webinar included several case studies from past TSCA Section 5 submissions. The case studies discuss how EPA evaluates submitted information and determines whether it is acceptable for the engineering assessment. According to EPA, worker inhalation exposure from particulates is a frequent area of rework. EPA selected several case studies to cover situations where submitter claims were either accepted or not accepted and provided rationales for each type of determination. EPA expects manufacturing, processing, and use operations involving handling, transferring, unloading, or loading the NCS in solid forms to present potential exposure for workers to total and respirable particles. In the absence of specific and substantiated information from the submitter, EPA will assess inhalation exposure to total and respirable particulates using either the applicable ESD or the Occupational Safety and Health Administration (OSHA) Particulates Not Otherwise Regulated (PNOR) Total and Respirable Dust, Permissible Exposure Limit (PEL) Model.

The webinar concluded with questions and answers. Ariel Hou, EPA, answered most of the questions. She began by addressing several questions from the July 2022 webinar that were not answered due to time constraints. The first question was whether EPA still does a pre-notice consultation and if it can be used to address potential data gaps or misunderstandings before EPA begins its assessment. Hou stated that EPA still does pre-notice consultations, but she noted that the meeting is not intended to obtain EPA’s opinion on the likelihood of the submission being approved. The next question asked for more information about the modeling assumptions and default values that EPA uses. The Office of Pollution Prevention and Toxics’ (OPPT) predictive tools and models are published at https://www.epa.gov/tsca-screening-tools/using-predictive-methods-assess-exposure-and-fate-under-tsca#fate. Specifically, default model input values for models such as ChemSTEER can be found in the User Guide, available at https://www.epa.gov/tsca-screening-tools/chemsteer-quick-start-guide-and-user-guide-tsca-predictive-screening-tool. In response to a question concerning whether EPA plans on updating the 2018 Points to Consider document or issue similar new guidance, Hou stated that EPA does not currently plan to update the Points to Consider document. EPA welcomes any feedback, however, and will consider it during the next update.

The questions from the October 18, 2022, webinar included a suggestion that since worker exposure falls under OSHA, EPA should work with OSHA on exposure. According to Hou, EPA coordinates with both OSHA and the National Institute for Occupational Safety and Health (NIOSH). Another question asked about evaluating hand-to-mouth incidental exposures during the worker exposure assessment. Hou responded that worker exposure assessments for new chemicals under TSCA Section 5 do not assess the hand-to-mouth pathway and that at this time, it is not assessed on a routine basis. Asked what EPA considers to be a large number of customer sites not under submitter control, Hou stated that it varies and could be tens or hundreds of use sites. A webinar attendee asked if EPA would provide training on the tools that it uses so that submitters might have a better idea of how their NCSs would be assessed. According to Hou, training was offered in the past through the Sustainable Futures Program. EPA has heard that there is a lot of interest and is considering it internally. EPA hopes to resume training when it has sufficient resources to do so.

A webinar attendee asked what EPA meant by visuals or pictures. Hou responded that if the question refers to the slides on the types of substantiation, she has seen diagrams to be helpful in terms of how the chemical is used, as well as photographs of the processing area showing where the engineering control is installed in relation to worker exposure. Tiwari stated that if a submitter makes a statement about closed transfer, then including a hand drawing is not a good idea, but photographs showing how the NCS is transferred within the controlled environment can help. Asked if the New Chemicals Division (NCD) will provide specific feedback when it determines that data are not acceptable, Hou responded that she would like to move toward that. If a submitter is told that its data are not acceptable but no reason is provided, Hou suggested reaching out to the program manager to discuss that specific PMN.

Commentary

This webinar represents EPA’s continued efforts to improve the quality of new chemical submissions and increase the transparency of EPA’s review of submissions. The webinar’s inclusion of specific examples from submissions of both what EPA found acceptable and unacceptable was helpful. In one example, EPA compared particle size data provided in two cases. In one case, the submitter provided only a barely legible graph; EPA understandably did not rely on that data. In the other case, EPA stated that the submitter provided full details from ten samples evaluated at two different labs. It is not clear where the line is between one unsupported data point and 20 well-supported data points. What is clear is that when providing measured data, EPA needs more details about how the data were gathered and how representative the data are. Such information is common in toxicity studies, less so for industrial hygiene or exposure studies.

EPA stated and we have found that photographs of equipment can help EPA better understand the conditions of use at facilities and are very helpful in supporting representations of handling substances in closed systems. This is especially true when substances are handled in automated systems, including being handled in systems that are designed to protect the product as well as the worker. Although EPA did not so state, in our experience, videos can also be helpful. In either case, photos and videos provide insight into the specific processes and conditions of use. Submitters should keep in mind that EPA is not expert in the specific processes at the submitter’s site, so spending some additional effort to educate EPA will very likely improve the outcome of a submission. Neither the photos nor the videos need to be high production value. Most mobile phones provide sufficient quality for these needs. As with all aspects of a submission, companies should review carefully any information submitted to EPA to ensure that confidential business information is protected.

Another common area of weakness in submissions is waste handling procedures. EPA stated, and our experience confirms, that it is insufficient to simply state that waste is sent offsite for proper disposal. EPA needs more details about the waste handling procedures and how those procedures address EPA’s concerns for potential releases and exposures.

Although it should not be news to submitters nearly six and a half years after TSCA reform, it is clear from EPA’s presentations and our experience that EPA needs much more detail to document better the conditions of use. It is still unclear to what extent all of this information will do to address EPA’s view of what conditions of use are reasonably foreseen and therefore reduce the chance that EPA issues an order, but this level of detail can at least improve the chances of EPA granting an LVE.