Case Study Challenges in Human Health Hazard and Risk Assessment of Nanoscale Silver - Safety of Nanomaterials along Their Lifecycle - page 421

identification of the material usually includes information on the main constitu-
ents as well as on impurities and additives (Box 19.1). For nanosilver, there are a
significant number of potentially relevant impurities and additives, including citrate,
various amines (e.g., polyethylenimine), Tween 20, polyvinylpyrrolidone (PVP), cetyl
trimethylammonium bromide (CTAB), sodium dodecylsulfate (SDS), (poly)saccha-
rides, dextran, (poly)ethylene glycol, polyacrylic acid, functional thiols, inorganic
coatings, and others. Although impurities resulting from the production process may
in principle be removed from the nanomaterial, others have been purposely added
to, for example, prevent aggregation of the nanosilver particles. The minimum set
of information that should be provided for constituents, impurities, and additives
includes the chemical names, molecular and structural formulas, and content or typi-
cal concentration ranges. Unfortunately, such information is currently rarely pro-
vided on stabilizers and residual reducing agents in study reports published in the
open literature. It is widely accepted that some substances including many inorganic
minerals and also nanomaterials require additional identifiers such as crystallinity
to allow for unequivocal description. Currently, there is no full consensus on the
type of additional information that should be part of the substance identification
of a nanomaterial. In many cases the debate is whether a certain parameter such as
morphology, size, and size distribution should be regarded as a characterizer of the
substance rather than an identifier (JRC 2011). Significant research activity (compare
Section II of this topic, especially Chapter 9) is ongoing to elucidate those nanomate-
rial characteristics that have a major impact on the hazard potential and should thus
be relevant identifiers in the context of risk assessment.
For nanosilver, different particle morphologies can be obtained depending on the
production protocol: In addition to spheres that dominate in the literature, pyramids,
cubes, plates, and rods can be generated (Wiley et al. 2007). It has been demonstrated
that chemical reactivity differs between nanosilver shapes as morphology influences
accessibility of the different crystal faces of the silver (Xu et al. 2006). Therefore,
nanosilver shape may also influence toxicity and may deserve consideration as a sub-
stance identifier in risk assessment. In addition, the crystallite size may also vary for
particles of comparable dimensions. Depending on the number of nucleation sites,
monocrystalline, twinned, and multiple twinned crystalline particles have been
described for nanosilver (Wiley et al. 2007).
Nanosilver particles can be synthesized with narrow monodisperse but nonover-
lapping size distributions of, for example, 25 ± 3 and 70 ± 4 nm, as demonstrated by
Li et al. (2012a). For most other nanomaterials, reaction products contain particles
of a wider size range or show a polydisperse distribution resulting from aggregation
during synthesis. In in vitro experiments, an inverse correlation of nanosilver parti-
cle size and cytotoxicity has been shown; whereas, the situation in vivo including the
effect of particle size on toxicokinetics is less clear (Carlson et al. 2008; Lankveld
et al. 2010; Li et al. 2012b).
The OECD (2012) currently recommends considering at least chemical composi-
tion, size and size range, crystallinity, surface coatings, and morphology to identify
the nanoscaled constituents (see Box 19.2 as an example). However, the OECD doc-
ument also states that where further information on potentially relevant properties is
available, such as zeta potential, redox potential, specific surface area, and others,