Risk Assessment of Manufactured Nanomaterials - Environmental and Human Health Impacts of Nanotechnology - page 397

assessment should continue to be used (Crane and Handy, 2007; Crane et al. , 2008 ),
but additional measurements may need to be adopted to account for novel proper-
ties as they emerge.
The exposure assessment is perhaps the most problematic, mainly because of a
lack of data on measured levels of nanomaterials in the environment, although this
is less of a problem in the workplace where the source of the material is easily
identifi ed. Environmental fate and behaviour studies are needed for nanomaterials
in the environment so that exposure point concentrations can be estimated. There
is a massive background of natural nanomaterials already present in the environ-
ment and manufactured nanomaterials would account for a trivial proportion of
the total (Handy et al. , 2008b). There are also technical issues to overcome, such
as measurement methods for nanomaterials in complex environmental matrices,
but this is not fundamentally different to any other new substance that would simi-
larly require a measurement technique to be developed and approved.
The risk analysis step should also consider the physical form of the product and
what it is used for (Hansen et al. , 2008). Much of the current hazard data on nano-
materials relates to nanoparticles in a free rather than fi xed (or embedded) form,
and it is important to provide this context at the outset (Ozin and Arsenault, 2006).
There is a concern that the risk will change with the product life cycle. For example,
a product coated with carbon nanotubes in a resin matrix will presumably eventu-
ally degrade to release individual particles. The context of the risk would depend
on how the product wears and where it is eventually disposed; a carbon nanotube
coated tennis racket would presumably be disposed of in landfi ll, but there are
uncertainties about the long term erosion of nanomaterials in landfi ll. This risk
might be offset by the increased strength and durability that might be imparted by
making the product from nanomaterials leading to a longer product life. Similar
arguments apply to the deterioration of nanoproducts in any occupational setting
and the risk of long term, low level exposure may then be an issue. Clearly, the
complexity of behaviour and toxic effects of nanoparticles generate uncertainties
which have signifi cant implications for the risk analysis. Nonetheless, the uncertain-
ties are regarded as manageable through collecting data from new experiments to
reduce the level of uncertainty of the hazard and exposure issues.
A further need is to consider the actual fate of nanomaterials in environmental
matrices. Many nanomaterials are likely to form aggregates in a wide range of
natural waters and the question arises as to whether materials should be dispersed
in laboratory tests or allowed to aggregate (Crane and Handy, 2007; Crane et al. ,
2008). It might be technically feasible to develop a lowest observed effect concentra-
tion (LOEC) or no observed adverse effect level (NOAEL) using one or more
standard test methods, but these may have little relevance for actual exposures in
complex natural systems. This is not a new problem and the inevitable difference
between standardised laboratory tests and the real environment is just another
source of uncertainty. However, there is no reason to believe this aspect is any better
or worse than that of other new chemicals. High uncertainty in hazard assessment
might mean the requirement for higher assessment factors when calculating, for
example, Predicted No Effect Concentrations (PNEC) from LOEC data. In addi-
tion, the tendency of nanomaterials to aggregate as mass concentration increases