Exposure to Nanoparticles - Environmental and Human Health Impacts of Nanotechnology - page 343

Filtration theory is well understood and has been extensively described by several
authors (e.g. Brown, 1993). As an aerosol penetrates through a fi lter, the trajectories
of the particles deviate from the streamline due to various well understood mecha-
nisms. As a result, particles may collide with the fi lter elements (fi bres) and become
deposited on them. The mechanisms include diffusion, interception, initial impac-
tion and gravitational settling. Electrostatic forces can also play a role in some fi lter
types.
For particles less than 100 nm, Brownian diffusion is the dominant mechanism
(Lee and Mukund, 2001). Filtration effi ciency due to Brownian diffusion increases
as particle size decreases. Brownian diffusion is caused by collisions between par-
ticles and the air molecules to create random paths that the particles follow. The
random motion increases the probability of a particle contacting one of the fi lter
elements. Once the particle is collected onto a surface it will adhere to it due to the
Van der Waals' forces. Therefore, fi lters are likely to be good collectors of
nanoparticles.
Current methods for certifi cation of HEPA fi lters and for respirator fi lters
do not routinely require testing at particle sizes in the nanometre size range.
Internationally recognised standards for HEPA fi lters (DOE, 1998) require that
the fi lter is challenged with an aerosol with a mass median diameter of 300 nm
and that the particle collection effi ciency is greater than 99.97%. Three hundred
nanometres is considered to be a much more penetrating aerosol for these
fi lters than nanometre size particles due to the decrease of Brownian diffusion
at this particle size. Similarly, European Standards for respirator fi lter cartridges
(CEN, 2000) and for fi ltering face pieces (CEN, 2001) require that these
systems are tested against sodium chloride aerosols with a mass median diameter
of 300 nm. Again, this is based on an expectation that this would be the most pen-
etrating size.
However, some authors have suggested that penetration of nanometre particles
through wire screens (fi lters) can deviate from the classical penetration models if
the effect of thermal rebound is signifi cant. Wang (1996) used a modifi ed penetra-
tion model including the effect of thermal rebound to compare with the experi-
mental results of Ichitsubo et al. (1996). He found good agreement using model
parameters derived from literature. Both experiment and theory suggested signifi -
cant thermal rebound and increased penetration for particles smaller than 2 nm.
Similar results have been reported by Otani (2002), who found that the particle
rebound may increase the penetration for platinum nanoparticles through circular
diffusion tubes when the particles are smaller than 2 nm.

8.4.2.3 Use of Personal Protective Equipment

Use of PPE such as respirators and air-fed devices may be used (as a fi nal option)
as a method of control for any airborne hazard. All of these devices depend
on fi ltration as a means of cleaning the air prior to it being breathed by the worker.
The discussion relating to fi ltration applies equally here. It should be expected
that, for all but the smallest nanoparticles ( < 2 nm), the fi ltration effi ciency will be

high.