Analysis of powders.
TEM analysis. TEM allows for direct viewing of solid electron-contrast primary nanoparticles or their agglomerates in consumer products. Representative TEM photomicrographs of tested powders are presented in , and the summary of the TEM image analysis results is presented in . We found electron-contrast particles in all of the tested powders. The electron beam did not appear to alter the structure of any of the particles observed [see Supplemental Material, p. 6
TEM photomicrographs of the tested cosmetic nanopowders (Nano) M (A–C), Nano D (D–F), and Nano K (G–I) and tested cosmetic regular powders (Reg) F (J–L), Reg G (M–O), and Reg E (P–R).
Characteristics of the tested cosmetic powder products obtained using different analysis methods.
All the primary particles (i.e., particles constituting the smallest dispersion level) in the sample of nanopowder M () were in the nanosize range. In fact, the largest observed particle was 45 nm in diameter. No free nanoparticles or individual agglomerates were observed—the level of agglomeration was very high, because all the nanoparticles in the samples of this product were continuously interconnected on the TEM grids. The sample of nanopowder D () contained no electron-contrast particles in the nanosize range. The only particles observed were > 5 μm (5,000 nm) in diameter and were not agglomerated. Nanopowder K () contained a wide size range of highly agglomerated particles, with most primary particles being in the nanosize range. Close examination of photomicrographs for regular powder F () showed nanosize particles in contact with larger particles. In the photomicrographs of regular powder G (), most of the surface of the TEM grid was covered with particles > 5 μm in diameter, with only a few separate nanoparticles. Regular powder E () contained a large number of nanoparticles that were agglomerated and attached to larger particles.
Based on the composition of the powders as provided by the manufacturers [see Supplemental Material, Table 1
)], we expect that the observed electron-contrast particles, including nanoparticles, contained silica (in all products with a possible exception of nanopowder K, for which information on composition was incomplete), talc (nanopowder D and regular powder G), mica (nanopowder D and regular powder F), aluminum hydroxide (nanopowder D), titanium dioxide and zinc oxide (nanopowder K), or kaolin and iron oxides (regular powders F and G).
Overall, based on TEM, we observed the highest abundance of nanoparticles in nanopowders M and K and in regular powder E.
LDS analysis. The summarized results of the LDS analysis are listed in . The instrument detected particles of 100 nm in nanopowders M and K and in regular powders F, G, and E. Size distributions of particles in these five powders were similar in shape, and all had mode diameters of 0.33 μm. Nanopowder D had a mode diameter of 0.66 μm ().
Size distributions of cosmetic powders by number as measured by the Mastersizer 2000. The data represent averages of three repeats.
Because the lower size limit of the LDS instrument used was 100 nm, particles with smaller diameter would not have been observed. However, the size distributions of these powders suggest that particles with diameters < 100 nm were likely present as well (). This assumption is supported by the fact that TEM also registered particles < 100 nm in these five products.
Notably, neither TEM nor LDS indicated nanoparticles in nanopowder D, which is marketed as nanotechnology based. Conversely, the same analysis techniques detected a high number of nanoparticles in regular powder E, which is not marketed as nanotechnology based. These findings suggest that information provided regarding the presence or absence of nanomaterials in consumer products may not always be confirmed by experimental techniques.
Analysis of airborne particles released during powder application. The size distributions and concentrations of aerosol particles released during the simulated application of cosmetic powders based on SMPS and APS are presented in . The mode diameters of the released particle size distributions that were sampled through the mannequin’s nostrils are provided in .
Figure 4 Size distributions of airborne cosmetic powders by number during their application to human mannequin face. The data represent averages of three repeats. (A) Electric mobility diameter measured by the SMPS: 14.1–723 nm measurement size range. (more ...)
A detailed description of the particle size distributions is provided in Supplemental Material
[see Supplemental Results, “Airborne Particle Measurement Results” (http://dx.doi.org/10.1289/ehp.1104350
)]. In brief, for particles < 25 nm in diameter, which are characterized by higher alveolar deposition efficiency compared with larger particles (International Commission on Radiological Protection 1994
), more variance in particle concentration was observed for nanopowders M and D and regular powders F and G () than for the rest of the products. The SMPS system is very sensitive to fluctuating particle concentrations. Therefore, we concluded that for these four cosmetic powders (M, D, F, and G), airborne nanoparticle concentration in the region < 25 nm in diameter was unstable over the course of cosmetic powder application. In general, peak nanoparticle number concentrations for particles < 25 nm in diameter were comparable to the highest concentrations observed for particles that were 25–723 nm in diameter.
Concentrations of nanoparticles between 25 and 100 nm in diameter differed among products (). It is notable that the highest total particle counts were measured during the application of regular powder E, which is not marketed as a nanotechnology-based product by its manufacturer. Nevertheless, the spherical shape of the silica particles observed in this cosmetic powder using TEM () suggests that they may have been engineered, which, if true, would make this product de facto nanotechnology based.
Airborne concentrations of particles between 100 nm and 20 μm in diameter () varied substantially among the different cosmetic powders. Particles across this entire range were measured during the application of both nanotechnology-based powders and regular powders, without obvious differences in the distributions between the nanopowders and regular powders. The products with the highest and lowest airborne concentrations varied within different particle size modes: fine (0.1–1 μm), accumulation (1–2.5 μm), coarse (2.5–10 μm), and supercoarse (> 10 μm), as defined by Lioy et al. (2006)
(). Notably, for particle diameters > approximately 1.5 μm, regular power E had the highest concentrations and nanopowder M had the lowest concentrations [for additional details, see Supplemental Material, pp. 6–7
It is important to note, however, that application of all nanopowders resulted in the release of particles as large as 20 μm (), and judging from the size distribution, even larger particles may have been released. As shown by the electron microscopy (), the nanoparticles were agglomerated in the cosmetic powders, which suggests that nanomaterial may have been present in all airborne particle size fractions generated in the personal breathing cloud by cosmetic powder application.
SMPS () and APS () measurements in the overlapping size range (500–700 nm or 0.5–0.7
μm) do not always agree. A discussion of potential causes for such differences is provided elsewhere (Nazarenko et al. 2011
Implications for exposure assessment and health risks.
Although deposition in the alveolar region of the lung is the highest for nanoparticles and agglomerates of nanoparticles < 100 nm in diameter, particles larger than approximately 0.3 μm (300 nm) in diameter can efficiently deposit in the non–gas-exchange region of the lung, with particles > 10 μm in diameter (supercoarse particles) depositing primarily in the head airways (Hinds 1999
). Therefore, inhalation of aerosol particles containing nanomaterials in both the 1–100 nm and 100 nm to 20 μm diameter size ranges, and possibly larger, and their potential deposition in all regions of the respiratory system, should be considered.
Our TEM data showed a predominance of agglomerated nanoparticles in nanopowders M and K, and a high number of agglomerated nanoparticles that were in contact with the surface of larger particles in regular powder E. Based on the TEM and aerosol measurement data, we expect that most of the airborne nanomaterial from cosmetic powders, especially by mass [see Supplemental Material, Figure 1
)], will be in agglomerated form in particle size fractions > 100 nm, which are usually not the focus of most toxicology studies involving nanomaterials. A similar phenomenon could be predicted for many other nanotechnology-based consumer products that release nanoparticles as agglomerates and/or as composites with larger particles.
Most toxicological studies of potential health effects of inhaled nanoparticles, including studies of murine models, have used aerosols in which individual nanoparticles or nanosize agglomerates are a dominant fraction. For example, Geiser et al. (2005)
administered a pure conditioned titanium dioxide aerosol with a 22-nm count median diameter into the rat respiratory system through an endotracheal tube. Sayes et al. (2010)
used a freshly generated silica aerosol with 37- and 83-nm mode diameter aerosols for nose inhalation exposure of rats. Based on size, such particles would primarily deposit in deep regions of the respiratory system.
By contrast, in consumer products such as cosmetic powders, our findings suggest that primary particles would likely coagulate among themselves and with other ingredients present in the product before its application. As a result, aerosols produced when cosmetic powders are used may be dominated by much larger particles, including agglomerates ≥ 10 μm in diameter. Consequently, application of cosmetic powders may result in inhaled nanomaterial deposition not only in the gas-exchange region of the lung (alveoli) but also in the non–gas-exchange regions (tracheobronchial and head airways). For example, nanoparticles ≤ 100 nm may form agglomerates > 10 μm (supercoarse-size particles) that deposit much higher up in the respiratory system than do nonagglomerated nanoparticles, that is, in the head airways rather than the alveolar and tracheobronchial regions (International Commission on Radiological Protection 1994
), resulting in completely different health effects. Use of pure nanomaterials, as in the experimental studies cited above, would lead to a much higher nanomaterial deposition in the deeper regions of the respiratory system than could be expected based on product exposure simulation. As a result, such studies would have a diminished capacity to predict human health effects due to exposure to actual nanotechnology-based products.
At the same time, depending on the breathing rate, the laryngeal jet may break up inhaled loose agglomerates as small as 1 μm in diameter (Li et al. 1996
) into smaller aggregates or individual particles that could deposit throughout the entire respiratory system. Therefore, quantitative nanoparticle exposure studies should take into account the polydisperse nature of aerosol produced during the use of nanotechnology-based consumer products and examine not only the exposure to and deposition of unbound nanoparticles, but also the fate, transport, and deposition of nanoparticle agglomerates in all regions of the respiratory system, including smaller aggregates or particles that may result from the breakup of larger nanoparticle agglomerates.
Our findings on potential nanomaterial inhalation exposure due to the use of actual consumer products emphasize that properties and effects of the pure nanomaterial ingredients cannot be used to predict actual consumer exposures and resulting health effects. Therefore, experimental techniques for toxicity studies of de facto nanotechnology-based consumer products must be developed. Results of such studies will provide guidance for the developing market of nanotechnology-based consumer products and help clarify the need and feasibility of its regulation.
We performed our measurements indoors at comfortable relative humidity levels: 40–50%. Application of the powders at different humidity conditions, especially at very low or very high levels, could possibly affect the extent of powder agglomeration and thus its deposition in the respiratory system. The effect of relative humidity and other environmental conditions on the extent of exposures should be addressed in future studies.