Chemometric techniques such as partial least squares combined with discriminant analysis (PLS–DA) and
artificial neural networks (ANN) analysis were used to enhance the detection, discrimination and quantification
of chemical warfare agents simulants. Triethyl phosphate (TEP) mixed with commercial products
in their original containers was analyzed through the container walls using fiber-optic-coupled
Raman spectroscopy. Experiments were performed by employing a custom built optical fiber probe operating
at 488 nm. Detection was accomplished using mixtures of the contents of the commercial bottles
and water. The bottle materials included green plastic, green glass, clear plastic, clear glass, amber glass
and white plastic. To account for the low scattering-peak intensities of some bottle materials, integration
times were increased. Short integration times provided no information for amber glass and white plastic.
The limits of detection were on the order of 1–5%, depending on bottle materials and contents. Good discrimination
was achieved with PLS–DA when models were generated from a dataset originating from the
same type of bottle material. ANN performed better when large sets of data were used, discriminating
TEP from bottle materials and contents, as well as accurately classifying over 90% of the data.
Chemometric techniques such as partial least squares combined with discriminant analysis (PLS–DA) andartificial neural networks (ANN) analysis were used to enhance the detection, discrimination and quantificationof chemical warfare agents simulants. Triethyl phosphate (TEP) mixed with commercial productsin their original containers was analyzed through the container walls using fiber-optic-coupledRaman spectroscopy. Experiments were performed by employing a custom built optical fiber probe operatingat 488 nm. Detection was accomplished using mixtures of the contents of the commercial bottlesand water. The bottle materials included green plastic, green glass, clear plastic, clear glass, amber glassand white plastic. To account for the low scattering-peak intensities of some bottle materials, integrationtimes were increased. Short integration times provided no information for amber glass and white plastic.The limits of detection were on the order of 1–5%, depending on bottle materials and contents. Good discriminationwas achieved with PLS–DA when models were generated from a dataset originating from thesame type of bottle material. ANN performed better when large sets of data were used, discriminatingTEP from bottle materials and contents, as well as accurately classifying over 90% of the data.
การแปล กรุณารอสักครู่..
Chemometric techniques such as partial least squares combined with discriminant analysis (PLS–DA) and
artificial neural networks (ANN) analysis were used to enhance the detection, discrimination and quantification
of chemical warfare agents simulants. Triethyl phosphate (TEP) mixed with commercial products
in their original containers was analyzed through the container walls using fiber-optic-coupled
Raman spectroscopy. Experiments were performed by employing a custom built optical fiber probe operating
at 488 nm. Detection was accomplished using mixtures of the contents of the commercial bottles
and water. The bottle materials included green plastic, green glass, clear plastic, clear glass, amber glass
and white plastic. To account for the low scattering-peak intensities of some bottle materials, integration
times were increased. Short integration times provided no information for amber glass and white plastic.
The limits of detection were on the order of 1–5%, depending on bottle materials and contents. Good discrimination
was achieved with PLS–DA when models were generated from a dataset originating from the
same type of bottle material. ANN performed better when large sets of data were used, discriminating
TEP from bottle materials and contents, as well as accurately classifying over 90% of the data.
การแปล กรุณารอสักครู่..