Influence of spectral resolution, spectral range and signal-to-noise ratio of Fourier transform infra-red spectra on identification of high explosive substances

In the contemporary spectroscopy there is a trend to record spectra with the highest possible spectral resolution. This is clearly justified if the spectral features in the spectrum are very narrow (for example infra-red spectra of gas samples). However there is a plethora of samples (in the liquid...

Full description

Saved in:
Bibliographic Details
Published in:Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy Molecular and biomolecular spectroscopy, 2018-01, Vol.188, p.106-112
Main Authors: Banas, Krzysztof, Banas, Agnieszka M., Heussler, Sascha P., Breese, Mark B.H.
Format: Article
Language:English
Subjects:
FTIR spectroscopy
R environment
Resolution
Synchrotron radiation
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the contemporary spectroscopy there is a trend to record spectra with the highest possible spectral resolution. This is clearly justified if the spectral features in the spectrum are very narrow (for example infra-red spectra of gas samples). However there is a plethora of samples (in the liquid and especially in the solid form) where there is a natural spectral peak broadening due to collisions and proximity predominately. Additionally there is a number of portable devices (spectrometers) with inherently restricted spectral resolution, spectral range or both, which are extremely useful in some field applications (archaeology, agriculture, food industry, cultural heritage, forensic science). In this paper the investigation of the influence of spectral resolution, spectral range and signal-to-noise ratio on the identification of high explosive substances by applying multivariate statistical methods on the Fourier transform infra-red spectral data sets is studied. All mathematical procedures on spectral data for dimension reduction, clustering and validation were implemented within R open source environment. Diagram showing workflow: from collecting FTIR spectra for various resolutions through pre-processing, dimension reduction to clustering and visualizing the results. [Display omitted] •Spectral data processing and analysis based on open source R environment•Investigation of the effect of spectral resolution and range on identification•Cluster quality assessment by stability and compactness measures
ISSN:1386-1425
DOI:10.1016/j.saa.2017.06.048