Quantum cascade laser back‐reflection spectroscopy at grazing‐angle incidence using the fast Fourier transform as a data preprocessing algorithm

• Published in: Journal of chemometrics Vol. 33; no. 9
• Main Authors: Pacheco‐Londoño, Leonardo C., Galán‐Freyle, Nataly J., Figueroa‐Navedo, Amanda M, Infante‐Castillo, Ricardo, Ruiz‐Caballero, José L., Hernández‐Rivera, Samuel P.
• Format: Journal Article
• Language: English
• Published: Chichester Wiley Subscription Services, Inc 01-09-2019
• Subjects: Absorption; Algorithms; Analytical methods; Angle of reflection; Chemometrics; Contaminants; Detection limits; Explosives; Explosives detection; fast Fourier transform (FFT); Fast Fourier transformations; Fourier transforms; Grazing; grazing angle MIR laser spectroscopy; Infrared reflection; Infrared spectroscopy; irbesartan/RDX; partial least squares (PLS); Preprocessing; principal component analysis (PCA); Principal components analysis; Quantum cascade lasers; RDX; Reflectance; Routines; Spectra; Spectroscopy; Spectrum analysis; Substrates; Thin films
• ISSN: 0886-9383; 1099-128X
• DOI: 10.1002/cem.3167

A simple optical layout for a grazing‐angle probe (GAP) mount for coupling to a midinfrared (MIR) quantum cascade laser (QCL) spectrometer is described. This assembly enables reflectance measurements at high incident angles. In the case of optically thin films and deposits on MIR reflective substrates, a double‐pass effect occurs, which is accompanied by the absorption of deposited samples in a reflection‐absorption infrared spectroscopy modality. The optical system allows MIR light to pass through the sample twice.
Applications to cleaning validation and detection of traces of explosives using the QCL‐GAP is reported. Principal component analysis and partial least squares multivariate chemometrics methods were employed to analyze MIR spectra to evaluate an analytical methodology for confirming the presence of residues of pharmaceutically active ingredients (irbesartan) and of traces of explosives (cyclotrimethylenetrinitramine [RDX]) that have been deposited on metallic substrates. The performance of spectral preprocessing via fast Fourier transform (FFT) analysis was evaluated for the ability to extract more powerful and accurate information from the obtained reflectance spectra. According to the figures of merit of this new technique, FFT with chemometric routines can obtain sensitivity and specificity values of 1.000. The limits of detection that were obtained for irbesartan and RDX were 26 and 8 ng/cm2, respectively. The experimental results demonstrate that the proposed system, when used together with proper chemometrics routines, constitutes a powerful tool for the development of methodologies that have lower detection limits for a range of applications that involve detecting traces of analytes that reside on substrates as contaminants.
A grazing‐angle probe was coupledp to a quantum cascade laser (QCL). A reflection‐absorption‐infrared spectroscopy (RAIRS) effect was observed for optically thin samples. Principal component analysis (PCA), artial least squares (PLS), and Fast Fourier transform (FFT) methods were employed to evaluate an analytical methodology for confirming the presence of traces of irbesartan (for cleaning validation) and RDX (explosives detection) deposited on metallic substrates. Limits of detection obtained for irbesartan and RDX were 26 and 8 ng/cm2, respectively. The proposed system constitutes a powerful tool for the development of methodologies for trace detection.