Raman spectrum identification based on the correlation score using the weighted segmental hit quality index.

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemometrics

Background:

• Conventional Raman spectra identification relies on the hit quality index (HQI), which is sensitive to intensity variations between instruments.
• Existing methods require time-consuming intensity calibration or standardization for each spectrometer, hindering rapid analysis.
• Instrumental differences, including laser wavelengths, complicate direct comparison of Raman spectra.

Purpose of the Study:

• To propose a novel, simplified scoring method for identifying Raman spectra acquired on different instruments.
• To enhance the accuracy and efficiency of spectral identification by overcoming intensity variation issues.
• To provide a robust alternative to conventional methods that require extensive instrument-specific calibration.

Main Methods:

• A new scoring method was developed, defined as the weighted sum of HQIs from spectrally segmented data using a windowing approach.
• Raman spectra of 10 different chemicals were recorded using three distinct instruments with varying laser wavelengths.
• The proposed method was evaluated on a dataset comprising 14,033 chemical spectra.

Main Results:

• The proposed scoring method successfully identified all 14,033 test chemicals without any errors.
• The method demonstrated robustness against variations in spectral intensity and instrumental parameters.
• Performance was validated across multiple instruments and laser wavelengths.

Conclusions:

• The novel weighted sum scoring method offers a significant improvement for Raman spectra identification across different instruments.
• This approach eliminates the need for laborious intensity calibration, saving time and resources.
• The proposed method is a promising and effective alternative to existing spectral identification techniques.