Gaining quantitative fidelity from Raman spectra in regimesof large and varying fluorescence

While Raman spectroscopy offers notable experimental advantages as a probe of complex mixtures, its application in practice often confronts samples that present an overwhelming fluorescence background. Here, we explore the efficacy of two particular Raman spectrometric strategies for quantitative analysis under conditions of high fluorescence interference. Calling upon very large datasets, we compare conventional Raman spectroscopy and Shifted Excitation Raman Difference Spectroscopy (SERDS) in an effort to determine which approach best overcomes obstacles presented by fluorescence under various experimental conditions. SERDS subtracts Raman spectra acquired at slightly different excitation wavelengths, which ideally removes an invariant fluorescence background. However, a question remains as to whether this difference waveform or the original spectrum, including the fluorescence, determines the sample composition with better accuracy. Calling upon stochastic simulations, we have constructed a 12- million spectrum database referring to binary mixtures of benzophenone and alanine in a fluorescent matrix representative of various experimental scenarios. We have found that multivariate regression models for the sample composition drawing upon conventional Raman libraries often achieve comparable or better prediction accuracy than SERDS for most fluorescence scenarios. The visually enhanced SERDS spectra do not necessarily translate to more accurate quantitative results. However, SERDS does outperform conventional Raman in scenarios involving highly variable, uncorrelated fluorescence backgrounds, effectively minimizing those challenging interferences. Both methods significantly benefit from preprocessing techniques such as Asymmetric Least Squares (ALS) and Discrete Wavelet Transforms (DWT), which enhance predictive accuracy by reducing baseline noise. This study emphasizes the importance of selecting the appropriate Raman analysis strategy based on specific fluorescence conditions, challenges assumptions about the superiority of visually distinct SERDS spectra, and provides new insights into leveraging Raman spectroscopy in real-world, fluorescence-rich environments.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.