Feedback from Dow

The Dow Chemical Company

Posted on: March 23, 2007

We would like to commend the Environmental Defense and DuPont Nano Partnership for the preparation of the Nano Risk Framework document which is a significant contribution towards ensuring the responsible development of materials at the nanoscale. As presented, the document represents a well-disciplined yet flexible and adaptable approach. Notably, the framework builds upon the existing and well-established knowledge of the risk assessment process which is currently practiced within many organizations.

We would like to communicate the following points for consideration by Environmental Defense and DuPont with respect to future versions of this framework:

1. Application of this framework to a given material or process is based solely on a compound being defined as a nanomaterial. In contrast, inclusion into the regulatory framework of REACH is based on production volumes. It is suggested that specific comment should be provided on how the proposed framework will align with REACH.
2. The framework advocates the use of base datasets for the characterization and assessment of nanoscale materials. Although base datasets are not part of TSCA, they are a part of SIDS and REACH initiatives. We support the collection of appropriate hazard identification information for all of its product and non-product substances, regardless of whether these substances are new or existing inventory materials. However, data collection for hazard identification should follow a tiered approach, taking into consideration the volume, end use and potential exposure scenarios. Professional judgments should form the basis for decisions regarding what constitutes an adequate amount of hazard and exposure data that is necessary to evaluate the diverse spectrum of industrial compounds in commerce. We advocate that not all substances will require the same amount of hazard and exposure information. Each material will need to be evaluated on a case-by-case basis, taking into consideration the specific information about the material including its physical properties, end uses and potential for human and environmental exposures. Therefore, we propose that generation of complete base-datasets may not be required for certain nanoscale materials.
3. The assessment of Health Hazard as outlined in Box 3 (Page 35) includes the completion of an in vivo short term toxicity test; however, within this section, and throughout the document, there is no mention of alternatives to animal testing. It is felt that there should be some focus and consideration of these approaches. In this regard, a tiered approach to establishing a base-set of data should be considered (see above) and should involve alternatives to and/or optimization of animal use where appropriate.
4. Where possible, the conduct of the proposed base set experiments should be specifically aligned with the available OECD guidelines for the testing of chemicals. Furthermore, in cases where there is a suspected need to deviate from these guidelines, a complete reasoning for the change should be provided. One such example would include a more comprehensive rationale for the 90 day observation period in the 28 day in vivo study. In addition, the value of a single-dose instillation study is questionable when compared to the proposed alternatives of a repeated oral or inhalation exposure study.
5. The framework supports the use of “reasonable worst case” values and/or assumptions for the evaluation of risk in the absence of sufficient hazard or exposure data. It should be noted that in cases where this approach is applied and the results show no potential for an adverse effect, collecting more exposure data or further refinements to the characterization should not be necessary. Furthermore, while it is realized that it may be important to use conservative assumptions, there needs to be recognition that these scenarios may overestimate the risk. We believe when there is uncertainty in estimates of risk, presentation of single estimates of risk is misleading and provides a false sense of precision. Presenting a range of plausible risk estimates, along with a central estimate, conveys a more objective characterization of the magnitude of the risks. A “central estimate” of risk is the mean or average of the distribution; or a number which contains multiple estimates of risk based on different assumptions, weighted by their relative plausibility; or any estimate judged to be most representative of the distribution. The central estimate should neither understate nor overstate the risk, but rather, should provide the risk manager and the public with the expected risk. Influential risk assessments should characterize uncertainty by highlighting central estimates as well high-end and low-end estimates of risk. The practice of highlighting only high-end or only low-end estimates of risk is discouraged.
6. The framework document is written as if there is one party who is responsible for completing the entire nanomaterial evaluation, that party being the company producing the nanomaterial, or nanomaterial containing product. As there are a number of requests for down-stream information to which one company is not going to be privy, it is probably not realistic to put all the responsibility onto the one party. Having said that, and if the ED-DuPont partnership agrees, how will this be addressed within the workflow of the framework?
7. A minor comment involves the use of the term “lifecyle” in step 2 of the process which is currently termed “Profile Lifecycle(s)”. Although it is recognized that the term lifecycle is used here to highlight the need to comprehensively profile a nanomaterial’s properties, hazards and exposure throughout its lifecycle, the use of this term does not align with what is commonly referred to as “life cycle analysis”. It is recommended to replace “lifecycle(s)” in the naming of step 2 with a phrase such as “Evaluation of EH&S properties throughout the lifecycle”. This change should increase clarity and avoid confusion for readers.
8. Another minor comment involves the importance of including insurance underwriters as a key stakeholder in Step 5. Commercial insurance underwriters lack understanding of "Nanotech" products and are hesitant to provide product liability insurance without a thorough explanation of the product safety profile and applications. Recognizing this lack of understanding could have an impact on a company deciding to go forward with a new product offering. It is suggested that wording be expanded in Step 5, action element 3. "Discuss and consider business, legal and stakeholder issues". The proposed change to paragraph 3 would read: "Particularly important is attention to public, worker and insurance underwriter perspectives."

Once again, the Environmental Defense and DuPont Nano Partnership is to be commended for the preparation of this document. Furthermore, the opportunity to provide comments is greatly appreciated. Please do not hesitate to contact us should you require additional information/clarification on any of the provided comments.

Darrell Boverhof and Martin Debney
The Dow Chemical Company