Standoff raman spectroscopy system for remote chemical detection

Evan Cull, Michael E. Gehm, B. D. Guenther, David J. BradyProceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE 2005 · DOI

TL;DR

This paper introduces a novel coded-aperture spectrometer design for Remote Raman Spectroscopy (RRS) that breaks the traditional trade-off between spectral resolution and light throughput. This allows for greater flexibility in controlling the size of the interrogation region on a target, making RRS systems more adaptable for real-world security and environmental monitoring applications.

What Problem Does This Paper Address?

Remote Raman Spectroscopy (RRS) is a powerful technique for identifying chemical compounds from a distance by analyzing the unique "fingerprint" of Raman-scattered light. However, traditional RRS systems face a fundamental challenge: a trade-off between spectral resolution and light throughput. To achieve high spectral resolution, a narrow slit is needed in the spectrometer, which limits the amount of light collected, especially from diffuse sources like Raman scattering. Conversely, a wider slit increases throughput but sacrifices spectral resolution. This limitation makes it difficult to optimize RRS systems for operational needs, such as scanning large areas or dealing with varying target conditions, without compromising performance.

Methods

The core innovation presented is the use of coded apertures in place of the traditional spectrometer slit. Instead of a single narrow slit, the authors employ a two-dimensional transmission pattern, specifically a Hadamard matrix mask. This mask has a pattern of opaque and transparent regions. The key idea is that the spectral resolution is determined by the size of the individual features in the mask, while the light throughput is determined by the total number of transparent features. This decouples the resolution-throughput trade-off.

The system works by projecting a laser onto a target, collecting the scattered Raman photons with a telescope, and then directing them through the coded aperture mask into the spectrometer. The detector records a transformed version of the spectrum. To recover the actual spectrum, a computational inversion algorithm, specifically a non-negative least squares inversion, is applied to the detector data. The paper also details the design of their RRS system, including a 532 nm laser and a Schmidt-Cassegrain telescope.

Datasets

Not discussed in the paper. The experiments used isopropanol as the target chemical, with varying concentrations diluted in water.

Key Insights and Ideas

  • Decoupling Resolution and Throughput: The most significant insight is that by using coded apertures, particularly Hadamard masks, the traditional optical compromise between spectral resolution and light throughput can be overcome. This is a fundamental shift in spectrometer design for RRS.
  • Operational Flexibility: This decoupling directly translates to operational flexibility. The size of the laser interrogation region on the target can now be chosen based on practical requirements (e.g., scanning speed, power density) rather than being dictated by the spectrometer's optical limitations.
  • Computational Spectrometry: The coded aperture approach necessitates a computational step to reconstruct the spectrum. This highlights the growing importance of algorithmic processing in optical sensing.
  • Potential for Improved Performance: The paper suggests that this approach can lead to RRS systems that are more adaptable and effective in real-world scenarios, where target conditions are often unpredictable.

Key Results

The authors demonstrated a proof-of-concept RRS system using coded apertures. They compared the performance of their system with different order Hadamard masks (N=12, 24, 40) against a traditional slit-based spectrometer.

  • Signal-to-Noise Ratio (SNR) Improvement: The coded aperture systems consistently outperformed the slit-based system in terms of SNR. The SNR ratio of the coded aperture to the corresponding slit was:
  • Order-40 mask: 1.9
    • Order-24 mask: 1.6
    • Order-12 mask: 1.1
  • Detection of Diluted Samples: The system was able to detect isopropanol at concentrations as low as 6.25%, although fluorescence from the sample container became a dominant feature at lower concentrations.

Limitations and Future Work

  • Fluorescence Interference: At low concentrations of the target chemical, broad fluorescence from the sample container became a significant issue, masking the Raman signal.
  • Pulsed Laser and Gated CCD: The authors suggest that future improvements could include using a pulsed laser to increase target illumination power and a gated CCD to reject background light, further enhancing performance.
  • Optimal Code Selection: Further research is needed to determine the optimal type of coded aperture (Hadamard or other) for RRS applications.
  • Mask Feature Size: While the mask feature size of 36 µm was used, the spectral resolution was practically limited to approximately 0.6 nm/pixel. Optimizing this feature size could further improve resolution.

Summary updated

Abstract

We have developed a class of aperture coding schemes for Remote Raman Spectrometers (RRS) that remove the traditional trade-off between throughput and spectral resolution. As a result, the size of the remote interrogation region can be driven by operational, rather than optical considerations. In this paper we present the design of our coded-aperture standoff spectroscopy system as well as experimental data collected while making remote measurements.
BibTeX
@inproceedings{cull2005standoff,
  author = {Evan Cull and Michael E. Gehm and B. D. Guenther and David J. Brady},
  title = {{Standoff raman spectroscopy system for remote chemical detection}},
  booktitle = {Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE},
  year = {2005},
  doi = {10.1117/12.626170},
}
Figure