Update on evaluation of enhanced darkfield microscopy and hyperspectral mapping for analysis of airborne nanoparticulate collected on filter-based media

N.M. Neu-Baker, A. Eastlake, S.A. Brenner
State University of New York Polytechnic Institute, College of Nanoscale Science,
United States

Keywords: enhanced darkfield microscopy, hyperspectral imaging, nanomaterials, filter-based samples, exposure assessment


Current best-known methods for engineered nanomaterial (ENM) exposure assessment in occupational environments include the capture of airborne ENMs onto filter media.1–3 The current standard method for the analysis of filter media is direct visualization via transmission electron microscopy (TEM) for particle sizing, count, and morphology, often coupled with compositional analysis, typically by energy-dispersive spectroscopy (EDS). This method is low-throughput, expensive, and time and resource-intensive. Enhanced darkfield microscopy (EDFM) with hyperspectral imaging (HSI) and mapping analysis is being evaluated as a high-throughput screening technique to rapidly identify ENMs of interest on filter media samples. We are exploring use of EDFM-HSI for the detection, characterization, and quantification of ENMs captured on filters with the intent to develop a standardized method and analytical protocols. The CytoViva EDFM-HSI system combines optical darkfield imaging and spectrophotometry to visualize and identify ENMs by capturing a spectrum from 400 – 1000 nm at each pixel in a hyperspectral image.4–6 Reference spectral libraries (RSLs) are created by collecting spectra from pixels corresponding to known materials for comparison to unknown samples. The spectral angle mapper (SAM) classifier algorithm can be used to identify the material of interest in images by matching the pixel spectrum with each spectrum in a RSL. Based on SAM results, an estimation of ENM concentration and size may be obtained. Preliminary work using multiwall carbon nanotubes (Figure 1) and silica ENMs (Figure 2) have provided promising results. Based on experience with carbon nanotubes, establishment of a limit of detection for EDFM-HSI for each material is critical. To accomplish this process, mixed cellulose ester (MCE) filters were exposed to an established concentration of the selected ENM using an acoustic generator. To produce statistically significant results, a given number of replicates was obtained for each concentration. The concentration range was chosen to encompass a sufficiently low concentration and span the recommended exposure limit (REL) for the ENM. All filters were then evaluated using EDFM-HSI. Filter samples were created using the same method for ultrafine titanium dioxide.7 Previous results indicate that these materials can be easily visualized by EDFM and, moreover, can be mapped using hyperspectral data. By using a given ENM concentration, filter replications, and using a material with an established REL, we have obtained data to determine a limit of detection for the EDFM-HSI method specific to titanium dioxide. Future steps include establishing limit of detection for additional ENMs (such as nanocellulose). There is need for additional evaluation tools when performing comprehensive exposure assessments of facilities that are handling emergent materials, such as ENMs, especially those that emphasize high-throughput, timely analysis, and are less expensive than the status quo. Analysis of the EDFM-HSI data obtained in this study will provide much needed insight into the potential future use of this method to determine the count, identification, and concentration of ENMs on field-collected filter-based samples. Future directions include expanding the EDFM-HSI protocol to additional filter media (i.e., polycarbonate) and to other ENMs, including mixed material exposures from field sampling.