Nanoscale materials, including nanoparticles (NPs) and ultra-thin films (UTF), are finding increased use in many fields, such as biomedical, optics, energy conversion, agricultural sciences, among others. In the case of nanoparticles, there are techniques available to characterize different properties, but few can yield information on their elemental composition, dimensions, and distribution. Techniques of choice, e.g. electron microscopy, require exceedingly long measurement time. Thus, there is a need to develop techniques with high sample throughput. We recently demonstrated the sizing, elemental analysis, and distribution characterization of NPs within a few seconds using GDOES-EM with traditional hyperspectral imaging (HSI) [1]. However, sputter sampling in GDOES consumes the sample, such that traditional HSI techniques can lead to loss of information/resolution due to sequential scanning in at least one dimension, which is compounded for multi-elemental analysis and can be particularly limiting for nanoscale materials characterization. To overcome this limitation, we developed the concept of glow-discharge optical emission coded aperture spectral imaging elemental mapping (GOCAM), which incorporates compressive coded aperture spectral imaging [2]. Here, the convolution of one spatial and the spectral dimensions enables to “fold” the 3D HSI data cube into two dimensions, which allows to obtain the full data cube simultaneously with a single snapshot. Later, the data is deconvoluted in software using compressed sensing (CS) convex iterative algorithms, optimized for best spatial and spectral fidelity, to recover the full field-of-view (FOV) from the under-sampled images. In the first part of this talk, we will present on the experimental realization of GOCAM for NP characterization in a fraction of a second.

In the case of UTF, GDOES is a sought-after technique because it can provide fast depth resolution on the nanometer scale at the near surface. However, GDOES operating conditions must be set to yield best crater shape, as opposed to best limits of detection (LODs). Furthermore, best GDOES LODs are reported for bulk analysis where one can collect data for long enough time to obtain the best statistics. Nevertheless, the resolution in depth relies on how fast one can measure the highly transient signals, which limits sampling statistics and in turn also compromises LODs. Moreover, slit-based λ selection devices in commercial GDOES instruments favour high λ resolution (very thin slits) at the expense of light throughput. We propose overcoming these limitations with a technique called glow discharge optical emission coded aperture spectroscopy (GOCAS). Here, both high λ resolution and high light throughput are achieved with a spectrograph featuring a coded aperture with multiple thin entrance slits, an array detector to collect the convoluted spectra, as well as CS algorithms (optimized only for best spectral fidelity) to recover the original spectra from the whole under-sampled FOV. In the second part of this talk, we will present on GOCAS quantitative figures of merit and improved LODs at the fastest acquisition times compared to the traditional single-slit approach.

References
[1] K. Finch, A. Hernandez, G. Gamez; Anal. Chem., 2023, 95, 4, 2269–2277.
[2] H. Agrawaal, G. Gamez; Anal. Chim. Acta, 2024, 1321, 343001.