Skip booklet index and go to page content

Risk Assessment/Management

Comparative Analysis of Current Risk Assessment Approaches for Nanomaterials

In order to develop an understanding of the methodologies and tools used by the Canada/US Programs to support regulatory decisions, a comparative analysis of current risk assessment and risk management approaches to nanomaterials in Canada (under CEPA 1999) and in the US (under TSCA) was conducted using a case-study approach. Industry stakeholders were asked to nominate potential nanomaterials for the case-study; nominated nanomaterials were subsequently ranked against relevant criteria such as commercialization and market availability. After a thorough analysis, Multiwalled Carbon Nanotubes (MWCNTs) were selected.  To facilitate the case-study, guidance for the conduct of the comparative analysis was drafted (Appendix D), and information exchanged between the Canada/US Programs.

From that comparative analysis, it was concluded that the Canada/US Programs have similar conservative approaches to the risk assessment and risk management of nanomaterials.  Additionally, identified risks, including areas of uncertainty, are mitigated through processes specifically designed to reduce exposures and allow further assessment for potential scenarios of direct human exposure and/or increased environmental release.  The Canada/US Programs agree that sufficient scientific knowledge does not yet exist on the behaviour, toxicity, mode of action and environmental fate of nanomaterials to allow for the development of a risk assessment methodology that would benefit from the use of analogue/read-across information.  As scientific knowledge evolves, understanding of these relationships will allow risk assessment to evolve from a case-by-case to a more generalized paradigm. 

Material Selection

The process for selecting a nanomaterial for the case study included a request for nominations from stakeholders.  A letter was sent to stakeholders in both Canada and the US on October 9, 2012, inviting nomination of candidate substances that could be considered for the case study (Appendix B). Four nanomaterials were nominated: Multiwalled Carbon Nanotubes (MWCNTs), Nano Crystalline Cellulose (NCC), Nano Silver (nAg) and Nano Titanium Dioxide (nTiO2).

The following criteria (See Appendix C for detailed criteria) were considered in the selection of an appropriate nanomaterial for the purposes of conducting a comparative analysis of current risk assessment methods and risk management approaches[1]:

  • commercialization and market availability;
  • substance composition/ production;
  • information availability; and,
  • relevance to RCC.

Based on the responses/information that was provided from stakeholders, MWCNTs were selected for the first nanomaterial case study.  Table 1 briefly summarizes the information received from stakeholders and the rationale for selecting MWCNTs.

Table 1: Summary ranking of nanomaterials nominations

Commercial ApplicationsResins, coatings and compositesPaints, coatings, composites, oil drillingTextilesNone provided
Regulatory status in Canada and USNotified in BOTH countriesNotified in Canada onlyNOT notified to either ProgramNOT notified to either Program
Type of available informationPhysical-chemical, volumes, toxicity, leachability, etc.Physical-chemical, volumes, toxicity, industrial releasePhysical-chemicalNone provided
Relevance Rank to RCC1234

NOTE: Future work may include conducting additional case studies from the other nominated nanomaterials with packages of greater or lesser information.

Comparison of Risk Assessment Methods

Though statutory differences surrounding pre-market notification requirements, information/data packages, assessment timelines and target populations of interest do exist, the overarching principles for the assessment and management of nanomaterials are consistent between the Canada/US Programs.  Because very few nanomaterial-specific evaluation methodologies, tools or models exist in Canada or the US, existing methodologies for the assessment of chemical substances are being used in both countries, where appropriate.  Numerous similarities in the Canadian and US chemical assessment protocols were revealed during the comparative analysis exercise. (Detailed comparison available in Appendix E.) The following key observations, risk assessment approaches and methods were noted during the discussions.

Identity/Physical-chemical Properties

Identity and chemical structure of traditional chemicals are usually confirmed through spectroscopic (e.g., infrared), resonance (e.g., nuclear magnetic resonance) and/or mass spectrometric (e.g., ionization mass spectrometry) techniques.  Additional particle-based techniques such as microscopic (e.g., scanning electron microscopy or transmission electron microscopy), light scattering (e.g., dynamic light scattering), and electronic are needed in order to adequately identify and characterize a nanomaterial.   However, there are currently very few standardized methodologies for these techniques, and conducting these measurements in complex environments presents many challenges.

The physical-chemical properties relevant to risk assessments of traditional chemicals -  including end-points such as melting-point, boiling point, vapour pressure, and water solubility -  are of limited value for nanomaterials.   End-points such as particle size distribution, surface charge, aggregation state, and specific surface area are more relevant in determining the toxicity, environmental fate, and behaviour of nanomaterials.  

Based on the lack of appropriate standardized methodologies needed to properly identify nanomaterials, the Canada/US Programs compared what kind of information they request from notifiers to identify similarities and differences in the two countries’ approaches.  Key findings are as follows:

  • Standard physical-chemical properties provided with new substances submissions do not usually contain adequate information for nanomaterial characterization or relevant physical-chemical properties such as particle size distribution, shape, and degree of agglomeration/aggregation.  Both countries feel this information is crucial to understanding the behaviour and toxicity of the nanomaterial.
  • Both jurisdictions request similar data sets to better characterize the substance (e.g., length, diameter, size, SSA, agglomeration/aggregation, purity, etc.).
  • For the most part, the OECD Test Guidelines for Chemicals are considered to be adequate for the testing of nanomaterials by both jurisdictions with some modifications.  However, some standard physical-chemical test guidelines for chemicals are not considered appropriate for assessing nanomaterials (e.g., OECD water solubility test guideline 105).
  • Both jurisdictions agree that no models are available to validate or predict physical-chemical data on nanomaterials. Appropriateness/relevance of data provided is assessed by professional judgment.

Environmental Fate and Release

Quantifying the releases from industrial activities of nanomaterials is not currently possible. In traditional chemical risk assessments, the releases of a substance can be appropriately quantified using mass balance models and traditional physical-chemical end-points. Emission scenario documents[2] from the OECD provide a solid basis for calculating the release from many different chemical uses; as well, models based on physical-chemical parameters, such as vapour pressure and degradation, can be used to predict behaviour in, for example, sewage treatment facilities to help calculate predicted environmental concentrations and the environmental fate of a substance.

There are no emission scenario documents specific to releases of nanomaterials. Release scenario documents relevant to particles do exist; however their applicability to nanomaterials is still uncertain. As can be seen in the high-level findings below, the Canada/US Programs rely on these particle-based release documents, or use conservative release scenarios, to calculate predicted environmental concentrations.

Most of the endpoints used for traditional chemicals cannot be used for nanomaterials, and no suitable models exist to adequately quantify or predict their environmental fate; additional properties such as aggregation state, surface charge, and surface area are needed to determine their environmental fate from a qualitative perspective. For example, carbon nanotubes may undergo settling through hetero- and homo- aggregation in rivers, embedding in sediments; they may stay in the water column through stabilization with natural organic matter; and they may be mobile or immobile in the soil, depending on their surface charge.

Key findings on environmental fate and release are as follows:

  • Due to the uncertainties associated with predicting the partitioning behaviour of nanomaterials, the Canada/US Programs assume that nanomaterials will likely be present in all compartments: water, soil, sediment, and air.
  • Both jurisdictions believe that the potential for long-range transport of nanomaterials is low; however, the potential for vectorization (i.e., nanoparticles binding to other particulates and undergoing transport) is uncertain and considered possible. 
  • Both jurisdictions believe that octanol-water partitioning coefficient (Log KOW) is not a useful predictor for bioconcentration/bioaccumulation for nanomaterials since their behaviour at the octanol-water interface is governed by different mechanisms than traditional chemicals (Hou et al., 2013)[3].
  • Overall, Canadian and US assumptions and approaches to environmental fate and release are similar: both countries consider industrial releases to waters through wastewater treatment processes to be a major route of release (similar to traditional chemicals)[4], releases to air are considered where appropriate, and land-spray applications are considered in cases where the substance partitions to biosolids. However, in the absence of experimental or predictive data, conservative values may be used for estimating exposures (e.g., 0% removal at a sewage treatment plant).
  • Environmental releases consider two worst-case scenarios: 100% release to the aquatic compartment from wastewater treatment (i.e., no removal from treatment) and 100% release to biosolids (i.e., 100% removal from treatment). These two extremes allow the Canada/US Programs to develop conservative scenarios to calculate predicted environmental concentrations.

Exposure Assessment

In a traditional pre-market chemical risk assessment, exposure is estimated following accepted paradigms and exposure scenarios based on known applications and the use-patterns identified. However, in the case of nanomaterials, it is currently not possible to accurately quantify exposure because nanomaterials can undergo various transformations during their lifecycle, including aggregation/agglomeration, changes in shape, and being surface-coated by different substances. In the absence of relevant information, the traditional chemical risk assessment framework is used to inform potential exposure scenarios, and conservative qualitative estimates of exposures are developed.  Exposure assessments of nanomaterials are generally conservative, assuming worst-case scenarios.  It is expected that as more models emerge and understanding increases, better identification and quantification of relevant exposures for nanomaterials will be developed.

  • Both Canada and the US agree it is impossible to quantify exposure with much certainty due to the novel nature of nanomaterials, uncertainty in release/fate (see 4.2), and the uncertainty of applying existing models.
  • The applicability of some data to ‘real world’ exposure is considered questionable; for example,  how does the surface of a nanomaterial change when it is released from an end-use coating, and is the information on the pure nanomaterial sufficient to address that exposure?

Hazard Assessment

Traditional chemical hazard assessments in the Canada/US Programs rely on test data for the notified substance, analogue/read-across data and in-silico modelling (e.g., QSAR). Hazard assessments of nanomaterials generally lack substance-specific test data, have limited access to adequate analogue data and no access to reliable in-silico models.  Key findings pertaining to the hazard assessment of nanomaterials are as follows:

  • For the most part, the EPA and OECD Test Guidelines for Chemicals are considered to be adequate for the testing of nanomaterials by both jurisdictions, albeit with some modifications and additional physical-chemical characterisation data.  However, some standard toxicity test guidelines for chemicals are not appropriate for assessing nanomaterials (e.g., bacterial reverse mutation test for genotoxicity).
  • Physical-chemical characterization of the test material is critical for adequately interpreting study results, assessing the toxicity of nanomaterials, and for comparison to other nanomaterials.
  • Data on the bulk-sized substances and similar nano-scale materials are used as qualitative indicators in hazard identification but not in point-of-departure quantification of hazard.
  • The suitability of analogs for read-across is determined based on degree of similarity between relevant physical-chemical properties of the new substance and those of the proposed analogs (e.g., size, shape, number of walls, and surface chemistry).
  • There is agreement in the interpretation of toxicity test results and critical endpoints by regulators in both jurisdictions.
  • No software models are used in Canada or the US for prediction of human health hazard or ecotoxicity.

Overarching Similarities and Differences in Risk Assessment Approaches

Overarching similarities in Canada/US approaches to nanomaterial risk assessment are as follows:

  • Both jurisdictions apply conservative assumptions about human health hazard and ecotoxicity potential when there is little test data available suitable for hazard or risk assessment of nanomaterials.
  • The Canada/US Programs rely heavily on published peer-reviewed scientific literature to fill data gaps.
  • Better characterization of nanomaterials  is considered essential by both Programs for assessing the applicability/reliability of any supplied test data for conducting quantitative risk assessments, and for applying methods of comparison (such as analogue/read-across) when possible.
  • Risk management recommendations reflect both the degree of risk and the level of uncertainty (e.g., the applicability of the available data and the lack of substance-specific information).

Differences in assessment approaches and in methods used were most often the result of differences in regulatory mandates or data availability.  The following identifies differences in observations, risk assessment approaches and methods between the Programs:

  • In general, Canada receives more test data than the US at the time of notification, due to the test data requirements prescribed under the New Substances Notification regulations (NSNR) in Canada; no similar requirements are prescribed in the US under TSCA.
  • The occupational workplace is a focus of attention in US risk assessments, while in Canada, general population exposure and lifecycle considerations are the focus in risk assessments of nanomaterials under CEPA 1999.
  • Consequently, Pre-Manufacture Notices (PMNs) in the US usually contain detailed information on manufacturing processes and industrial releases not usually received with submissions to Canada.
  • Canadian exposure assessments usually include consideration of potential use scenarios in addition to those identified in the submission, while in the US assessments generally only consider applications reported/mentioned in the submission.
  • The US receives a greater number of nanomaterial submissions and consequently has greater access to information on potential analogs to support the use of analogue/read-across information in quantitative risk assessment and decisions-making. As a result, the EPA uses analog data for conducting quantitative risk assessment of some PMN nanomaterials
  • Most of these differences are due to differences in regulations, rather than how information is reviewed and/or evaluated in the two countries. 

Comparison of Risk Management Approaches Used

Canada and the US have similar approaches for the management of nanomaterials. Both countries try to manage exposure in order to manage risk. This can be done through the application of significant new use type provisions for those notified substances of concern, or, those that may impact on health or the environment. Both countries also work closely with industry in order to best mitigate potential risks to the environment or human health.

Identified risks or areas of uncertainty are mitigated through an approach designed to reduce exposures and allow further assessment of potential scenarios of increased environmental release and/or direct human exposure. 

Key Conclusions

The US and Canada have similar approaches to risk assessment and risk management of nanomaterials based on case-by-case assessments and utilizing a conservative screening approach.  Identified risks or areas of uncertainty are mitigated by methodologies designed to reduce exposures and to allow further assessment for potential scenarios of increased environmental release and/or direct human exposure. 

Both jurisdictions agree that sufficiently comprehensive scientific knowledge on the behaviour, toxicity, mode of action and environmental fate of nanomaterials does not yet exist to allow for routinely conducting quantitative risk assessment, but as scientific knowledge evolves, understanding of these relationships will allow increasingly more data to be incorporated into assessments. 

Differences in exposure assessment scope were discussed during the comparative analysis exercise.  These include a focus on occupational exposure in US assessments as compared to a focus on general population exposure in Canadian assessments, differences in the initial data sets submitted to both jurisdictions and differences in the extent to which assessments would cover potential uses and life cycle considerations.  These differences stem from the respective regulatory frameworks rather than differences in approaches as to how data is evaluated. 

Based on the analysis and discussions that occurred during the case study/comparative analysis exercises, it became clear that the development of common approaches for the systematic assessment of physical-chemical properties, environmental fate, exposure and hazard of nanomaterials would contribute to increase consistency, transparency and predictability between the Canada/US Programs.

[1] Note: Nomination/identification of a nanomaterial for the purposes of this comparative analysis does not constitute a finding that the nanomaterial presents a risk to human health or the environment

[3] Hou, W.; Westerhoff, P.; Posner, J.Env. Sci. Processes & Impacts, 2013, 15, 103-122

[4] Gottschalk, F.; Sonderer, T.; Scholz, R.; Nowack, B. ES&T, 2009, 43, 9216-9222.