Biomedical Engineering Reference
In-Depth Information
and to the uncertainty associated with the conventional products. Several
methods are available to quantify uncertainty.
Sensitivity studies
can be car-
ried out by, for example, looking at variation in impacts when varying each
input variable between its lower and upper confidence limits. This enables
the practitioner to identify the most influential parameters for the consid-
ered nanoproducts and to help focus data search on the most relevant data.
Uncertainty propagation
can be carried out using more or less transparent
and numerically intensive methods. Hong et al.
10
propose a parsimonious
Taylor Series expansion approach based on lognormal distributions that
apply well to the type of calculations used to assess nanoparticle impacts,
with zero-bounded multiplicative factors. It enables the analyst to easily
determine the contribution of every parameter to the overall uncertainty.
Monte-Carlo approaches can handle a wider range of distributions
204,205
but
has often been limited to uncertainty propagation in the sole inventory,
thus the need to have uncertainty in impact assessment explicitly defined
by LCIA developers, users applying them to nanoproducts, and to have
this LCIA uncertainty contribution integrated in the mostly used LCA
software.
Interpreting uncertainty in the nano-LCA context and for sound deci-
sion making:
As emphasized by Rosenbaum et al.,
7
uncertainty on chemi-
cal characterization factors (their impact per kilogram emitted) is relatively
high, within a factor of 100-1000 for human health and 10-100 for freshwa-
ter ecotoxicity. Such a precision of 2 to 3 orders of magnitude is, however,
significantly lower than the roughly 10 to 12 orders of magnitude variation
between the impacts of the different chemicals.
The inventories and impact characterization factors calculated for nano-
products and nanoemissions must therefore be used in a way that reflects
the large variation of 10 orders of magnitude between chemical character-
ization factors as well as the 3 orders of magnitude uncertainty on the indi-
vidual factors. This means that contributions of 1%, 5%, or 90% to the total
human toxicity score are essentially equal but significantly larger than those
of a conventional chemical contributing to less than one per thousand or less
than one per million of the total score. In practice, this means that for LCA
practitioners, these toxicity factors are very useful to identify the 10 most
important toxics pertinent for their nano-applications. The life cycle toxicity
scores thus enable the identification of all chemicals contributing more than,
for example, one thousandth to the total score. In most applications, this will
allow the practitioner to identify whether direct nanoparticle emissions may
be relevant compared with the 10 chemicals to look at in priority, and per-
haps, more importantly, to disregard 400 other substances whose impacts
are not significant for the considered application. For the nanosilver T-shirt,
it will be interesting to identify the order of magnitude of impacts related
to nanosilver emissions during washing compared with detergent impacts,
especially if the number of washing cycle per functional unit is reduced due
to the antibacterial effect of nanosilver.
