This case study served as a range-finding survey of airborne nanomaterials emitted during common tasks performed in a laboratory that investigates environmental risks of engineered nanomaterials. In addition, this study allowed the NIOSH NTRC field team to test analytical equipment and methodologies under various laboratory conditions to evaluate potential occupational exposure to engineered nanomaterials. Specifically, this research effort combined semiquantitative airborne particle number concentrations with qualitative TEM imaging to provide a weight-of-evidence evaluation of whether engineered nanomaterials were released during laboratory tasks. To our knowledge, this is the first study to suggest that engineered nanoparticles may be released from aqueous suspensions during sonication. Results of this study imply that the commonly held belief that engineered nanomaterials in suspension during sonication pose low risk of inhalation exposure may need some reconsideration. This is especially true with regard to results from a recent international survey of nanomaterial firms and laboratories; in that study, Conti et al. (2008)
found that many workers in the field think nanomaterials pose no risk.
After accounting for background particle counts, we detected increased particle number concentrations during the handling of dry CNMs and also during the sonication of CNM suspensions (). An interesting observation during the present study was the differential behavior between hydrophobic and hydrophilic CNMs with regard to different laboratory processes. During material handling and weighing, we observed higher airborne particle number concentrations of the hydrophobic CNMs (C60 and raw MWCNT) compared with hydrophilic CNMs. Lower particle number concentrations of aerosolized CNMs at 300 and 500 nm were noted during sonication, yet cumulative particle number concentrations in the 10–1,000 nm size range were elevated compared with the handling process. This finding was more pronounced when raw MWCNTs were sonicated in moderately hard reconstituted water containing 100 mg/L NOM, suggesting that sonication of CNM suspensions may increase the number of smaller-sized CNM agglomerates (i.e., < 300 nm)—as would be expected with sonication—that were not detected by the HHPC-6 particle counter. An opposite pattern was observed when the hydrophilic CNMs (MWCNT-OH and CB) were compared. Very low particle number concentrations were detected during handling of hydrophilic CNMs, yet sonicating these hydrophilic CNMs, whether in a DI water suspension or a moderately hard reconstituted water suspension with 100 mg/L NOM, resulted in dramatically higher airborne particle number concentrations. From this finding, along with visual evidence provided by TEM examination of the air-sampling filters, we hypothesize that CNM agglomerates are being emitted to the laboratory atmosphere in water droplets. These data demonstrate that care should be exercised when handling dry hydrophobic CNMs and also when sonicating wet CNMs in suspension. A similar pattern of emissions and potential exposure was observed by Methner et al. (2007)
during a study of nanomaterial polymer laboratory workers.
In the present study, all filter-based air samples collected during weighing and transfer processes, with the exception of raw MWCNT, showed the presence of the engineered nanomaterial handled. Likewise, all samples collected during sonication, regardless of the nanomaterial in suspension, showed visual evidence of the presence of the engineered nanomaterial when analyzed by TEM. The majority of the images presented in indicate that single spheres or nanotubes are more the exception than the rule; most particles showed clear evidence of agglomeration. However, this may be due to the current methodology that uses 0.8-μm filter membranes, which may allow small, individual CNMs to pass through and thus be unavailable for analysis. The images shown in clearly provide strong visual evidence that emissions from specific tasks and processes can occur. No evidence of engineered nanomaterial was present on the background air filter sample collected.
Our data indicate that although suspensions may minimize aerosolization of CNMs relative to their dry form, sonication of such suspensions outside protective enclosures can result in aerosolization and thus potential exposure to nanosized particulates (). If sonication occurs outside an enclosure, as often occurs in laboratory settings, the proximity of the researcher’s breathing zone may result in inhalation of CNM particulates in water droplets and/or mists. Similarly, airborne water droplets can be generated by standard aquaria that use air stones or other air supplies to aerate test waters during long-term aquatic toxicology studies. The airborne CNMs in water droplets have the potential to cause pulmonary effects similar to those described for particulate matter, single-walled carbon nanotubes (SWCNT), and MWCNTs (Laks et al. 2008
; Lam et al. 2006
; Ma-Hock et al. 2009
; Mitchell et al. 2007
; Shvedova et al. 2008
). The mass balance of CNMs collected during the laboratory processes were not determined, so the mass of airborne CNMs is unknown, making it difficult to compare with CNM inhalation toxicity studies or to occupational exposure limits for carbon-based materials such as respirable graphite or particulate matter. However, Conti et al. (2008)
found that organizations that used nanomaterials in suspensions or embedded in matrices were less likely to make recommendations for respiratory protection. With sonication being a critical component of nanomaterial synthesis and deagglomeration, this survey result suggests that inhalation exposure may be an overlooked safety component during this commonly used laboratory process.
Graphical representation of potential exposure to engineered CNMs in the laboratory through inhalation and dermal contact.
Despite being housed in an enclosure during this experimental process evaluation, the sonicator has the potential to emit engineered nanomaterials when the enclosure door is opened after the sonication process is complete. If this occurs, airborne CNMs may be inhaled by laboratory workers. In addition, if the sealing gasket around the perimeter of the enclosure door is damaged or otherwise breached, release of aerosolized droplets to the laboratory atmosphere may result. Furthermore, these airborne CNM-containing water droplets have the potential to deposit on other surfaces within the sonication cabinet and in the laboratory. Once dried, CNM may become resuspended if disturbed and potentially result in exposure via inhalation. Finally, these nanomaterials may be available for dermal deposition if laboratory workers unknowingly contact contaminated surfaces with unprotected skin (e.g., hands, forearms). Currently, there are no occupational exposure limits specific to engineered nanomaterials (Methner 2008
); however, basic precautionary procedures and control equipment can dramatically reduce airborne releases of nanomaterials (NIOSH 2009
). Therefore, environmental scientists should implement a general or nano-specific environmental, health, and safety program at their organizations (Conti et al. 2008
), use personal protective equipment, and develop standard operational procedures to minimize potential hazards when working with engineered nanomaterials in environmentally relevant laboratory systems.
Although this preliminary research has generated some interesting and relevant findings, specific uncertainties associated with experimental design and implementation need to be addressed. First, this single-case study was designed to determine the relative magnitude of airborne nanomaterial emissions associated with tasks and materials used in environmental laboratory experiments. We used only a single data point for each of the tasks and materials during this first assessment. Thus, the data presented here are not statistically based. These data should be viewed as an indicator of the need for additional studies that focus on a robust statistically based experimental design, experimental variables, specific engineered nanomaterials, and sample collection. Second, the data interpretation can be confounding because of the two different particle counters used to measure airborne nano-sized particles. The two particle counters use different counting principles, counting efficiencies, and size ranges, so the data are not directly convertible to identical units. Our intentions were to show the size ranges and relative number concentrations on a task- specific basis. This way, a reader can examine the data separately according to task and determine which task emitted nanomaterials. Thus, these data should be interpreted as relative indicators of CNM release, especially since the data were adjusted by subtracting background particle number concentrations. Furthermore, because the direct-reading, real-time instrumentations are not material specific (e.g., MWCNTs or CB only) and cannot identify the chemical composition of the particles detected (e.g., MWCNT vs. background particulate matter or water droplets), we cannot definitively conclude that increases in particle number concentrations for a specific operation are due to a release of particulate material from that process. However, because the particle number concentrations in the lower size ranges were higher than background and the results of the TEM analyses yielded visual evidence of the engineered nanomaterials, we can conclude that a release occurred and that the potential for exposure exists.