Non-perturbative cathodoluminescence microscopy of beam-sensitive materials

Abstract

Cathodoluminescence microscopy is now a well-established and powerful tool for probing the photonic properties of nanoscale materials, but in many cases, nanophotonic materials are easily damaged by the electron-beam doses necessary to achieve reasonable cathodoluminescence signal-to-noise ratios. Two-dimensional materials have proven particularly susceptible to beam-induced modifications, yielding both obstacles to high spatial-resolution measurement and opportunities for beam-induced patterning of quantum photonic systems. Here pan-sharpening techniques are applied to cathodoluminescence microscopy in order to address these challenges and experimentally demonstrate the promise of pan-sharpening for minimally-perturbative high-spatial-resolution spectrum imaging of beam-sensitive materials.

Keywords: 2D materials; cathodoluminescence; color centers.

Figure 1:

Combined SEM-CL analysis of a prototypical hBN flake. (a) SEM image of an hBN flake with horizontal width of 10.5 μm. (b) Time-series CL spectra acquired at a single point on the hBN flake highlighting the beam-induced changes in the hBN CL spectrum as a function of increasing dose. (c) Spectral components and (d) intensity maps generated by non-negative matrix factorization of CL spectrum image acquired in conventional rastered CL spectrum imaging modality.

Figure 2:

Benchmark PS-CL results generated from a single CL hyperspectral image of an hBN flake with dimensions of 100 × 100 × 1024 (horizontal, vertical, wavelength) pixels with a horizontal field of view of 30 μm. A panchromatic image was generated from the spectrum image by summing along the wavelength axis while spectrum images with reduced spatial resolution were generated by binning spatial pixels together. The Brovey pan-sharpening algorithm was used to reconstruct a hyperspectral CL image from these datasets. NMF reconstructions of the pan-sharpened CL images are shown for data generated from the panchromatic image and (a) the complete 100 × 100 × 1024 spectrum image, (b) a 50 × 50 × 1024 spectrum image, (c) a 25 × 25 × 1024 spectrum image, (d) a 12 × 12 × 1024 spectrum image, (e) a 6 × 6 × 1024 spectrum image, and (f) a 1 × 1 × 1024 spectrum image.

Figure 3:

Calculated structural similarity index as a function of compression ratio for the PS-CL data shown in Figure 2. The compression ratio is calculated based on the compression of the hyperspectral data prior to pan sharpening.

Figure 4:

Three component NMF reconstructions of hBN PS-CL images generated from (a) an 1,168 × 1,034 pixel panchromatic image and a 39 × 35 pixel hyperspectral image and (b) an 1,168 × 1,034 panchromatic image and a 117 × 105 pixel hyperspectral image. The horizontal field of view is 10.5 μm.