Jason A. Guicheteau

Bacillus Spore Classification via Surface-Enhanced Raman Spectroscopy and Principal Component Analysis

Surface-enhanced Raman spectroscopy (SERS) can provide rapid fingerprinting of biomaterial in a nondestructive manner. The adsorption of colloidal silver to biological material suppresses native biofluorescence while providing electromagnetic surface enhancement of the normal Raman signal. This work validates the applicability of qualitative SER spectroscopy for analysis of bacterial species by utilizing principal component analysis (PCA) to show discrimination of biological threat simulants, based upon multivariate statistical confidence limits bounding known data clusters. Gram-positive Bacillus spores (Bacillus atrophaeus, Bacillus anthracis, and Bacillus thuringiensis) are investigated along with the Gram-negative bacterium Pantoea agglomerans.

Semi-Automated Detection of Trace Explosives in Fingerprints on Strongly Interfering Surfaces with Raman Chemical Imaging

We have previously demonstrated the use of wide-field Raman chemical imaging (RCI) to detect and identify the presence of trace explosives in contaminated fingerprints. In this current work we demonstrate the detection of trace explosives in contaminated fingerprints on strongly Raman scattering surfaces such as plastics and painted metals using an automated background subtraction routine. We demonstrate the use of partial least squares subtraction to minimize the interfering surface spectral signatures, allowing the detection and identification of explosive materials in the corrected Raman images. The resulting analyses are then visually superimposed on the corresponding bright field images to physically locate traces of explosives. Additionally, we attempt to address the question of whether a complete RCI of a fingerprint is required for trace explosive detection or whether a simple non-imaging Raman spectrum is sufficient. This investigation further demonstrates the ability to nondestructively identify explosives on fingerprints present on commonly found surfaces such that the fingerprint remains intact for further biometric analysis.

Raman Chemical Imaging of Explosive-Contaminated Fingerprints

Raman chemical imaging (RCI) has been used to detect and identify explosives in contaminated fingerprints. Bright-field imaging is used to identify regions of interest within a fingerprint, which can then be examined to determine their chemical composition using RCI and fluorescence imaging. Results are presented where explosives in contaminated fingerprints are identified and their spatial distributions are obtained. Identification of explosives is obtained using Pearsons cosine cross-correlation technique using the characteristic region (500–1850 cm−1) of the spectrum. This study shows the ability to identify explosives nondestructively so that the fingerprint remains intact for further biometric analysis. Prospects for forensic examination of contaminated fingerprints are discussed.

Characterization of polymorphic states in energetic samples of 1,3,5-trinitro-1,3,5-triazine (RDX) fabricated using drop-on-demand inkjet technology.

The United States Army and the first responder community are evaluating optical detection systems for the trace detection of hazardous energetic materials. Fielded detection systems must be evaluated with the appropriate material concentrations to accurately identify the residue in theater. Trace levels of energetic materials have been observed in mutable polymorphic phases and, therefore, the systems being evaluated must be able to detect and accurately identify variant sample phases observed in spectral data. In this work, we report on the novel application of drop-on-demand technology for the fabrication of standardized trace 1,3,5-trinitro-1,3,5-triazine (RDX) samples. The drop-on-demand sample fabrication technique is compared both visually and spectrally to the more commonly used drop-and-dry technique. As the drop-on-demand technique allows for the fabrication of trace level hazard materials, concerted efforts focused on characterization of the polymorphic phase changes observed with low concentrations of RDX commonly used in drop-on-demand processing. This information is important when evaluating optical detection technologies using samples prepared with a drop-on-demand inkjet system, as the technology may be “trained” to detect the common bulk α phase of the explosive based on its spectral features but fall short in positively detecting a trace quantity of RDX (β-phase). We report the polymorphic shifts observed between α- and β-phases of this energetic material and discuss the conditions leading to the favoring of one phase over the other.

Bioaerosol Analysis with Raman Chemical Imaging Microspectroscopy

Raman chemical imaging microspectroscopy is evaluated as a technology for waterborne pathogen and bioaerosol detection. Raman imaging produces a three-dimensional data cube consisting of a Raman spectrum at every pixel in a microscope field of view. Binary and ternary mixtures including combinations of polystyrene beads, gram-positive Bacillus anthracis, B. thuringiensis, and B. atrophaeus spores, and B. cereus vegetative cells were investigated by Raman imaging for differentiation and characterization purposes. Bacillus spore aerosol sizes were varied to provide visual proof for corroboration of spectral assignments. Conventional applications of Raman imaging consist of differentiating relatively broad areas of a sample in a microscope field of view. The spectral angle mapping data analysis algorithm was used to compare a library spectrum with experimental spectra from pixels in the microscope field of view. This direct one-to-one matching is straightforward, does not require a training set, is independent of absolute spectral intensity, and only requires univariate statistics. Raman imaging is expanded in its capabilities to differentiate and distinguish between discrete 1-6 microm size bacterial species in single particles, clusters of mixed species, and bioaerosols with interference background particles.

Recent Advances and Remaining Challenges for the Spectroscopic Detection of Explosive Threats

In 2010, the U.S. Army initiated a program through the Edgewood Chemical Biological Center to identify viable spectroscopic signatures of explosives and initiate environmental persistence, fate, and transport studies for trace residues. These studies were ultimately designed to integrate these signatures into algorithms and experimentally evaluate sensor performance for explosives and precursor materials in existing chemical point and standoff detection systems. Accurate and validated optical cross sections and signatures are critical in benchmarking spectroscopic-based sensors. This program has provided important information for the scientists and engineers currently developing trace-detection solutions to the homemade explosive problem. With this information, the sensitivity of spectroscopic methods for explosives detection can now be quantitatively evaluated before the sensor is deployed and tested.

Ultraviolet resonance Raman spectroscopy of explosives in solution and the solid state.

Resonance Raman cross sections of common explosives have been measured by use of excitation wavelengths in the deep-UV from 229 to 262 nm. These measurements were performed both in solution and in the native solid state for comparison. While measurements of UV Raman cross sections in solution with an internal standard are straightforward and commonly found in the literature, measurements on the solid phase are rare. This is due to the difficulty in preparing a solid sample in which the molecules of the internal standard and absorbing analyte/explosive experience the same laser intensity. This requires producing solid samples that are mixtures of strongly absorbing explosives and an internal standard transparent at the UV wavelengths used. For the solid-state measurements, it is necessary to use nanostructured mixtures of the explosive and the internal standard in order to avoid this bias due to the strong UV absorption of the explosive. In this study we used a facile spray-drying technique where the analyte of interest was codeposited with the nonresonant standard onto an aluminum-coated microscope slide. The generated resonance enhancement profiles and quantitative UV-vis absorption spectra were then used to plot the relative Raman return as a function of excitation wavelength and particle size.

Assessing Metal Nanofabricated Substrates for Surface-Enhanced Raman Scattering (SERS) Activity and Reproducibility

Surface-enhanced Raman spectroscopy (SERS) has been shown to be an effective technique for increasing the detection sensitivity in chemical and biological applications. SERS has a distinct advantage over normal Raman spectroscopy, with enhancements typically greater than 104 over the normal Raman signal; however, this advantage in sensitivity comes with a caveat: controlling the spectroscopic reproducibility and enhancement activity of metal nanostructured substrates can be difficult. We present a survey and subsequent data analysis performed on several nanostructured substrates designed for SERS, including silver and gold colloids, silver nanorods, gold nanoshells, and commercially manufactured gold nanostructures.

Toward understanding the influence of intermolecular interactions and molecular orientation on the chemical enhancement of SERS.

Implementation of SERS as an analytical technique is limited because the factors that govern the enhancement of individual vibrational modes are not well understood. Although the chemical effect only accounts for up to two orders of magnitude enhancement, it can still have a significant impact on the consistency of chemical spectral signatures. We report on a combined theoretical and experimental study on the benzenethiol on silver and 4-mercaptophenol on silver systems. The primary and unique finding was that for the benzenethiol on silver system the inclusion of interaction between multiple benzenethiol analyte molecules was essential to account for the relative enhancements observed experimentally. An examination of the molecular orbitals showed sharing of electron density across the entire model of multiple benzenethiol molecules mediated by the metal atoms. The addition of multiple 4-mercaptophenol molecules to the theoretical model had little effect on the predicted spectra, and we attribute this to the fact that a much larger model is necessary to replicate the networks of hydrogen bonds. Molecular orientation was also found to affect the predicted spectra, and it was found that an upright position improved agreement between theoretical and experimental spectra. An analysis of the vibrational frequency shifts between the normal Raman spectrum of the neat compound and the SERS spectrum also suggests that both benzenethiol and 4-mercaptophenol are in an upright position.

Critical role of adsorption equilibria on the determination of surface-enhanced Raman enhancement.

Surface-enhanced Raman spectroscopy (SERS) is a useful technique for probing analyte-noble metal interactions and determining thermodynamic properties such as their surface reaction equilibrium constants and binding energies. In this study, we measure the binding equilibrium constants and Gibbs free energy of binding for a series of nitrogen-containing aromatic molecules adsorbed on Klarite substrates. A dual Langmuir dependence of the SERS intensity on concentration was observed for the six species studied, indicating the presence of at least two different binding energies. We relate the measured binding energies to the previously described SERS enhancement value (SEV) and show that the SEV is proportional to the traditional SERS enhancement factor G, with a constant of proportionality that is critically dependent on the adsorption equilibrium constant determined from the dual Langmuir isotherm. We believe the approach described is generally applicable to many SERS substrates, both as a prescriptive approach to determining their relative performance and as a probe of the substrates affinity for a target adsorbate.