Ian R. Lewis

Interpretation of Raman Spectra of Nitro-Containing Explosive Materials. Part I: Group Frequency and Structural Class Membership

Fourier transform (FT)-Raman spectroscopy has been used to obtain high-quality spectra of 32 explosive materials. The majority of the spectra of these explosives have not previously been reported. Twenty-eight of the explosives have been categorized into three classes (nitrates esters, nitro-aromatics, and nitramines) based on their chemical structure, the position of the antisymmetric and symmetric stretching vibrations of the nitro group, and the shapes of the band envelopes. The spectra of exceptional explosives are discussed in terms of their unique structures or compositions.

Raman Spectrometry with Fiber-Optic Sampling

Raman spectrometry has historically been limited to the study of pure, nonfluorescent samples. During the period 1966-1986, several workers demonstrated that the use of deep-red or near-infrared (NIR) lasers would allow the observation of Raman spectra without exciting fluorescence. The ability to obtain Raman spectra in the deep red or NIR could not be generally realized in reasonable measurement times, primarily because of the insensitivity of the available detectors (photomultiplier tubes or diode arrays). In addition, the krypton or solid-state ion-pumped lasers required for these measurements were very expensive.

Raman spectroscopic studies of explosive materials: towards a fieldable explosives detector

Abstract Raman spectroscopy, with red (632.8 nm) and near-infrared (785 and 1064 nm) excitation, has been used to obtain high quality spectra of neat explosives. Samples with dimensions from a minimum size of 10 μm have been analyzed utilizing a Raman microprobe fitted with a charge-coupled device (CCD) array detector. Little sample fluorescence is observed for 23 of the 32 high explosives using 632.8 nm excitation and all of the samples can be measured with a 1064 nm Nd:YAG laser. 785 nm radiation affords an excellent compromise between sensitivity and fluorescence suppression. Problems of instrumentation and sample handling have been investigated. Spectra have been obtained for explosives, both neat and in plastic and glass containers. The feasibility of sampling explosives through colored glass, which is highly fluorescent in the visible, is also demonstrated.

Raman spectrometry and neural networks for the classification of wood types—1

Abstract In this work the coupling of near infrared (NIR) Fourier-transform (FT) Raman spectroscopy and neural computing for spectral feature extraction and classification of woods is reported. A NIR FT-Raman spectrometer operating at 1064 nm was used for all measurements; particular attention was paid to the effects of sample fluorescence and heating. It was demonstrated that fluorescence rejection is accomplished only for the lighter colored woods and that fluorescence was found to be severe for 10 of the 71 woods studied in this work even using excitation at 1064 nm. It was further found that hardwoods were no more or less susceptible to sample heating than softwoods. Feed-forward neural networks were used to extract the principal features of wood spectra at resolutions of 4, 8 and 16 cm −1 and to classify spectra as either temperate hardwoods or temperate softwoods. Neural networks were constructed using zero and two processing elements in the hidden layer. It was shown that neural networks with two hidden processing elements perform near optimally, since each hidden layer processing element may function as either a hardwood or softwood feature detector. This work represents the first time that FT-Raman spectroscopy and neural network technology have been coupled for spectral feature extraction and classification.

Comparison of sampling techniques for in-line monitoring using Raman spectroscopy.

Raman spectroscopy is currently of interest as a process monitoring tool for pharmaceutical unit operations. In this study, the performance characteristics of Raman spectrometers with different sampling optics have been investigated in the context of process monitoring, with emphasis being placed on assessing homogeneity in powder blends and following changes in solid-state form during wet granulation. A novel large spot non-contact Raman sampling device was compared with a traditional small spot size non-contact sampling device and an immersion probe. The large spot non-contact optics provided significant advantages over the standard systems both as a result of the enhanced sampling volume and because of the greater robustness of the system to fluctuations in the sampling distance during the wet granulation process.

VIBRATIONAL SPECTROSCOPIC STUDIES OF ASBESTOS AND COMPARISON OF SUITABILITY FOR REMOTE ANALYSIS

Abstract Raman, mid-infrared (MIR) and near-infrared (NIR) spectroscopic methods have been applied to the analysis of fibrous (asbestos) and non-fibrous forms of serpentine and amphibole minerals. This paper describes the first application of NIR diffuse reflectance spectroscopy to the study of asbestos minerals. In addition, a more generally applicable Raman spectroscopic study, in terms of multiple excitation wavelengths and asbestos materials, than has previously been made is reported. Improvements in spectral quality (especially in the OH stretching region) over previously published MIR spectroscopic data are also reported. The results of this work indicate the NIR diffuse reflectance spectroscopy appears to be the preferred method for the remote analysis of asbestos due to the relatively simple spectra in the wavenumber range 7400-6900 cm −1 , the high signal-to-noise ratio and spectral contrast of the spectra, the capability of using silica fiber-optics cables, and the time required to perform the analysis relative to the other vibrational spectroscopic techniques.

Interpretation of Raman Spectra of Nitro-Containing Explosive Materials. Part II: The Implementation of Neural, Fuzzy, and Statistical Models for Unsupervised Pattern Recognition:

The implementation of neural, fuzzy, and statistical models for the unsupervised pattern recognition and clustering of Fourier transform (FT)-Raman spectra of explosive materials is reported. In this work a statistical pattern recognition technique based on the concept of nearest-neighbors classification is described. Also the first application of both fuzzy clustering and a fuzzified Kohonen clustering network for the analysis of vibrational spectra is presented. Fuzzified Kohonen networks were found to perform as well as or better than the traditional fuzzy clustering technique. The unsupervised pattern recognition techniques, without the need for a priori structural information, yielded results which were comparable with those obtained by using a combination of a priori structural information and manual group-frequency analysis. This work demonstrates, via the use of a nitro-containing explosive data set, the utility of unsupervised pattern recognition techniques for the clustering, novelty detection, prototyping, and feature mapping of Raman spectra. The results of this work are directly applicable to the characterization of Raman spectra of explosives recorded with fiber-optic sampling.

Anti-Stokes Raman Spectrometry with 1064-nm Excitation: An Effective Instrumental Approach for Field Detection of Explosives

Anti-Stokes Raman spectra of 28 explosive materials were obtained with 1064-nm excitation using fiber-optic sampling and a dispersive spectrograph equipped with a charge-coupled device (CCD) array detector. By using a silicon CCD detector, anti-Stokes features could clearly be observed for the majority of samples from −250 to −1650 cm−1. Using the fiber-optic probe, spectra were routinely obtained from samples positioned up to twelve meters from the spectrograph within 240 s. The utility of an anti-Stokes correction routine is demonstrated, which routine allowed anti-Stokes spectra measured with 1064-nm excitation to be successfully searched and identified against libraries of Stokes spectra obtained using a Fourier transform (FT) Raman system equipped with a 1064-nm Nd: YAG laser.

Measurement of Spatial Resolution and Sensitivity in Transmission and Backscattering Raman Spectroscopy of Opaque Samples: Impact on Pharmaceutical Quality Control and Raman Tomography:

A practical methodology is described that allows measurement of spatial resolution and sensitivity of Raman spectroscopy in backscatter and transmission modes under conditions where photon migration dominates, i.e., with turbid or opaque samples. For the first time under such conditions the width and intensity of the point spread function (PSF) has been accurately measured as a function of sample thickness and depth below the surface. In transmission mode, the lateral resolution for objects in the bulk degraded linearly with sample thickness, but the resolution was much better for objects near either surface, being determined by the diameter of the probe beam and collection aperture irrespective of sample thickness. In other words, buried objects appear to be larger than ones near either surface. The absolute transmitted signal decreased significantly with sample thickness, but objects in the bulk yielded higher signals than those at either surface. In transmission, materials are sampled preferentially in the bulk, which has ramifications for quantitative analysis. In backscattering mode, objects near the probed surface were detected much more effectively than in the bulk, and the resolution worsened linearly with depth below the surface. These results are highly relevant in circumstances in which one is trying to detect or image buried objects in opaque media, for example Raman tomography of biological tissues or compositional and structural analysis of pharmaceutical tablets. Finally, the observations were in good agreement with Monte Carlo simulations and, provided one is in the diffusion regime, were insensitive to the choice of transport length, which shows that a simple model can be used to predict instrument performance for a given excitation and collection geometry.

Raman spectrometry and neural networks for the classification of wood types. 2. Kohonen self-organizing maps

Abstract One- and two-dimensional Kohonen self-organizing maps (SOMs) were successfully used for the unsupervised differentiation of the Fourier transform Raman spectra of hardwoods from softwoods. The SOMs were also applied to differentiate temperate woods from tropical woods, and results showed that the two types of woods could only be partly differentiated. A semi-quantitative method that is based on the Euclidean distances of the weight matrix has been developed to assist the automatic clustering of the neurons in a two-dimensional SOM.