Hyperspectral microscopy is an advanced visualization technique that combines hyperspectral imaging with state-of-the-art optics and computer software to enable rapid identification of nanomaterials. Since hyperspectral datacubes are large, their acquisition is complicated and time-consuming.

Despite the efficiency of spectral scanning in acquiring hyperspectral datacubes, this technique cannot be extended to large numbers of spectral bands because of their low light levels due to narrowband filters and mechanical difficulties while using large filter wheels.

However, applying a digital micromirror device (DMD) can circumvent the above drawbacks during spectral multiplexing. Utilizing a single DMD avoids the need for large filter wheels by promoting arbitrary spectral programming.

In an article published in The Journal of Physical Chemistry C, a brightfield DMD-based multiplexing microscope was employed to investigate the two-dimensional (2D) nanomaterials. Furthermore, the effectiveness of the DMD-based microscopy was demonstrated by measuring the thickness of few-layer graphene and molybdenum sulfide (MoS₂) from their corresponding contrast spectra, which were later compared to their theoretical curves for validation.

Hyperspectral Microscopy to Characterize 2D Materials

Atomically thin semiconducting 2D materials are extensively applied in nanophotonics, and the outstanding optical properties of these 2D materials play a critical role in many applications. Hence, accurate characterization of these 2D materials is critical to employ them in device structures to pattern the necessary electrical contacts.

Hyperspectral microscopy is a spectral imaging modality that can obtain a sample’s full spectroscopic information and render it in image form, and is one technique that is being developed and explored to address current analytical challenges for nanoscale 2D materials.

Hyperspectral microscopy involves the functional combination of a traditional high-resolution microscope and spectrometer. The motivation behind developing this technique for biomedical applications comes from an interest in the biological sample’s emission or reflectance spectrum, which contains important structural, biochemical, or physiological information.

The unique optical properties of 2D materials are largely dependent on the number of atomic layers. Hyperspectral imaging microscopy shows a large potential for rapid and accurate thickness mapping.

Hyperspectral Microscopy of 2D Materials

In the present study, the DMD was employed to encode the illumination source’s spectral content and overcome the mechanical difficulties of hyperspectral microscopy in terms of imaging with a filter wheel. This method promoted Hadamard multiplexing in the spectral content of the sample, improving the light throughput without affecting the signal-to-noise ratio.

Although using DMD as a programmable spectral filter was previously reported, this was the first work that applied it to hyperspectral microscopy of nanomaterials. The proposed multiplexing microscope was composed of illumination and an imager. While the illumination side was employed with a hyperspectral projector, the imager consisted of a reflective brightfield microscope.

Moreover, the microscope’s entrance had a biconvex lens that focused the incident light to the back focal plane of the objective to realize Koehler’s illumination. On the other hand, the objective lens focused the illumination that was spectrally programmed down to the sample and collected the light reflected.

The bandwidth and spectral resolution of the microscope were measured using tantalum sulfide (TaS₂) since it is highly reflective across the visible region. The two hyperspectral images obtained revealed that the topographical features in transmission mode were more than in reflection mode.

Measuring the exciton peaks in MoS₂ and comparing them to the theoretical result computed using Fresnel’s equations showed good agreement with the theoretical spectra for monolayer and bilayer MoS₂.

Furthermore, the image of graphite nanosheets at the camera and the reconstructed hyperspectral image showed regions with multiple spatially separated flakes. The reconstructed image helped to optically determine the thickness of the flakes at different parts of the nanosheet.

Conclusion

In summary, diffraction-limited, fast, large-field-of-view hyperspectral microscopy was demonstrated to contrast spectroscopy. The proposed system could be applied for characterizing novel devices and thin film heterostructures. Additional modifications to the hyperspectral microscope can enable different experiments.

For example, the sample, transmission, and reflection hyperspectral imaging can be concurrently achieved with a long working distance objective. Hyperspectral imaging of TaS₂ with three regions of differing thickness revealed that the topographical features in transmission mode were more than in reflection mode.

On the other hand, for the samples that evolve over time, performing hyperspectral video microscopy allowed sampling of both spectral and temporal dimensions. Moreover, single-pixel imaging could be naturally incorporated into the system by utilizing DMD and a single detector instead of a camera.

This enabled hyperspectral microscopy in the infrared, which otherwise becomes expensive for cameras. The spatial, temporal, and spectral information was captured on a single detector followed by reconstruction using compressive sensing recovery algorithms.