Applied Spectroscopy

Hyperspectral imaging was originally developed for remote sensing and astronomical applications, but adaptations of this technology have been of great benefit to the preservation of cultural heritage. Developments in noninvasive analytical techniques have advanced the preservation of cultural heritage materials by enabling the identification and analysis of a range of materials, utilizing their unique spectral response to nondestructively determine chemical composition, and determining states of deterioration and change due to environmental conditions. When used as a tool for noninvasive characterization of cultural heritage, these spectral imaging systems allow the collection of chemical identification information about materials without sampling, which is a critical factor for cultural heritage materials. The United States Library of Congress has been developing the application of hyperspectral imaging to the preservation and analysis of cultural heritage materials as a powerful noncontact technique. It allows noninvasive characterization of materials, by identifying and characterizing colorants, inks, and substrates with narrow-band illumination to protect the object while also monitoring deterioration or changes due to exhibit and other environmental conditions. Contiguous illumination from the ultraviolet, visible, and infrared spectral regions allows the capture of lost, obscured, and deteriorated information. The resulting image cube allows greater capabilities for mapping and coordinating a range of complementary chemical and spectral analyses. The capabilities of this technique are illustrated by a review of results from analysis of the Waldseemüller World Map, the L'Enfant plan for Washington, D.C., and the first draft of the U.S. Declaration of Independence.

A multimodal methodology for spectral imaging of cells is presented. The spectral imaging setup uses a transmission diffraction grating on a light microscope to concurrently record spectral images of cells and cellular organelles by fluorescence, darkfield, brightfield, and differential interference contrast (DIC) spectral microscopy. Initially, the setup was applied for fluorescence spectral imaging of yeast and mammalian cells labeled with multiple fluorophores. Fluorescence signals originating from fluorescently labeled biomolecules in cells were collected through triple or single filter cubes, separated by the grating, and imaged using a charge-coupled device (CCD) camera. Cellular components such as nuclei, cytoskeleton, and mitochondria were spatially separated by the fluorescence spectra of the fluorophores present in them, providing detailed multi-colored spectral images of cells. Additionally, the grating-based spectral microscope enabled measurement of scattering and absorption spectra of unlabeled cells and stained tissue sections using darkfield and brightfield or DIC spectral microscopy, respectively. The presented spectral imaging methodology provides a readily affordable approach for multimodal spectral characterization of biological cells and other specimens.

Human PC-3 prostate cancer cells were incubated in the presence of two cardenolides, i.e., ouabain and 19-hydroxy-2“-oxovoruscharin. Their effects were monitored by infrared spectroscopy of the cells after different exposure times to the cardenolides. Analysis of changes in absorbance intensities indicated that, for both compounds, the absorbance at one wavenumber with a minor contribution of a second wavenumber is sufficient to build a linear model accurate enough to assign more than 97% of the spectra to their correct time slot. Student t-tests and two-dimensional correlation analysis (2D-COS) indicated that both drugs have very similar effects on PC-3 cells. However, asynchronous 2D maps revealed significant differences and allowed the sequence of the spectral changes to be determined: 1395 → 1695 cm⁻¹ for ouabain, and 1400 → 1655 → 1100 → 1250 → 1020 cm⁻¹” for 19-hydroxy-2“-oxovoruscharin. 2D correlation map subtraction allowed the identification of very specific differences in the impact of both compounds on PC-3 cells, in particular the ability of 19-hydroxy-2”-oxovoruscharin to affect nucleic acid of PC-3 cells.

The molecular composition of the organic and inorganic matrices of bone undergoes alterations during maturation. The aim of this study was to compare Fourier transform infrared (FT-IR) and near-infrared (NIR) Raman microspectroscopy techniques for characterization of the composition of growing and developing bone from young to skeletally mature rabbits. Moreover, the specificity and differences of the techniques for determining bone composition were clarified. The humeri of female New Zealand White rabbits, with age range from young to skeletally mature animals (four age groups, n = 7 per group), were studied. Spectral peak areas, intensities, and ratios related to organic and inorganic matrices of bone were analyzed and compared between the age groups and between FT-IR and Raman microspectroscopic techniques. Specifically, the degree of mineralization, type-B carbonate substitution, crystallinity of hydroxyapatite (HA), mineral content, and collagen maturity were examined. Significant changes during maturation were observed in various compositional parameters with one or both techniques. Overall, the compositional parameters calculated from the Raman spectra correlated with analogous parameters calculated from the IR spectra. Collagen cross-linking (XLR), as determined through peak fitting and directly from the IR spectra, were highly correlated. The mineral/matrix ratio in the Raman spectra was evaluated with multiple different peaks representing the organic matrix. The results showed high correlation with each other. After comparison with the bone mineral density (BMD) values from micro-computed tomography (micro-CT) imaging measurements and crystal size from XRD measurements, it is suggested that Raman microspectroscopy is more sensitive than FT-IR microspectroscopy for the inorganic matrix of the bone. In the literature, similar spectroscopic parameters obtained with FT-IR and NIR Raman microspectroscopic techniques are often compared. According to the present results, however, caution is required when performing this kind of comparison.

This work describes the behavior of 1,6-diphenyl-1,3,5-hexatriene (DPH) in ethanol/water mixtures. The dependence of DPH photophysical properties (absorption and fluorescence emission) on the water percentage in ethanol indicates that DPH undergoes self-aggregation processes in solvent conditions above a critical water content. Evidence such as an additional absorption band, Beer's law deviation, kinetic behavior, and other experimental results obtained from temperature variation and surfactant addition demonstrated the presence of several types of DPH aggregates. Resonance light scattering measurements proved that the aggregate grew in water-rich media by a self-catalyzed process.

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.

We very recently reported a new spectroscopic application for expanding the versatility of surface Raman called “shell-isolated nanoparticle-enhanced Raman spectroscopy” or “SHINERS”. The most important and most difficult part of the SHINERS experiment is the effective transfer of the strong electromagnetic field from a gold core through the isolating silica or alumina shell to the probed surface. For this it is essential that the chemically inert dielectric shell be ultrathin (2–5 nm) yet pinhole-free. Herein we describe experimental and theoretical aspects of our SHINERS method in more detail. We provide a protocol for the synthesis and characterization of optimized shell-isolated nanoparticles (SHINs), and we examine the advantages of SHINERS nanoparticles over bare gold nanoparticles. We also present high-quality Raman spectra obtained from gold and platinum single-crystal surfaces in an electrochemical environment by our SHINERS technique. SHINERS is a simple and cost-effective approach that expands the flexibility of surface-enhanced Raman scattering (SERS) for an unprecedented diversity of applications in materials and surface sciences.

This paper presents a Fourier transform infrared (FT-IR) spectrometer calibration procedure based on an unusual source made from a spectrally selective surface. An alternative solution to the usual calibrators has been developed to cope with the tight mass budget of an instrument devoted to Mars surface exploration. The designed system has proved effective, in terms of achievable radiometric accuracy, despite the drawbacks due to the significant reflectivity of the sources. The proposed procedure is a standard “two-source” approach in which both cold and hot sources are thermally controlled surfaces, similar to an optical solar reflector, associated to a filament lamp. Such a system allows the required signal to be achieved in the 2–25 μm instrument wavelength range. Source optimization was performed using, as a cost function, the computed radiometric uncertainty, while the required absolute accuracy of the instrument was imposed as the optimization constraint.

Back-surface mirrors are needed as reference materials for vibrationally resonant sum-frequency generation (VR-SFG) probing of liquid-solid interfaces. Conventional noble metal mirrors are not suitable for back-surface applications due to the presence of a metal adhesion layer (chromium or titanium) between the window substrate and the reflective metal surface. Using vapor deposited 3-mercaptopropyltrimethoxysilane (MPTMS) as a bi-functional adhesion promoter, gold mirrors were fabricated on fused silica substrates. These mirrors exhibit excellent gold adhesion as determined by the Scotch® tape test. They also produce minimal spectroscopic interference in the C–H stretching region (2800–3000 cm⁻¹), as characterized by VR-SFG. These mirrors are thus robust and can be used as back-surface mirrors for a variety of applications, including reference mirrors for VR-SFG.

Laser-induced breakdown spectroscopy (LIBS) using conditional data analysis was applied to aqueous suspensions of palladium particles in the reformate water of palladium-based proton exchange membrane fuel cells. A significant amount of palladium was found in the water, indicating degradation of the fuel-cell cathode catalytic layers. The palladium particle-size detection limit was found to be about 400 nm. Calibration procedures to quantify the palladium concentration are discussed.

Controlling the composition of mixtures is critical for quality control in a wide variety of applications. There is a need for rapid, on-site measurements to optimize processes in real time. Ultrasound easily penetrates opaque samples and containers, yet currently provides minimal chemical information. We have developed a general approach to determine the volume fraction of a liquid in mixtures with multiple components. Ultrasound waves propagating through a medium undergo distortion processes that are characteristic of the chemical bonding composition. The distortion of the waveform can be measured in the ultrasound frequency profile. An ultrasound pulse-through configuration with matching 5 MHz transducers was used to analyze mixtures of water, methanol, and ethanol. Multilinear regression analysis was used to determine the volume fraction of all components in a series of mixtures. Using this technique, volume fractions were determined simultaneously with correlation coefficients (r²) greater than 0.98 in two-component mixtures. Determination of volume fractions in three-component mixtures ranging from 65–100% water also showed correlation coefficients of 0.91 for methanol and 0.94 for ethanol. This technique is attractive for process monitoring due to the short measurement time and the simple methodology that excludes sample pretreatment.

A wide variety of digital filters exist for processing one-dimensional (1D) signals; however, the application of some filters results in pronounced systematic distortions in band shapes and band intensities. In the present contribution, filtering is achieved by optimization in which a general objective function is constructed that possesses a number of desirable qualities, such as (1) smoothness of the resulting spectrum as well as (2) statistical constraints on the residual. Since the residual is explicitly used in the optimization, one can control systematic distortions and therefore avoid over-filtering. In tests using a variety of synthetic as well as real 1D spectroscopic data, the filter adequately preserves both band shapes and band intensities. In addition, the filter appears to accommodate homoscedastic, heteroscedastic, and frequency-dependent noise. Examples of its application and usefulness to powder X-ray diffraction (PXRD), Raman, and Fourier transform infrared (FT-IR) emission data are provided. Tests with synthetic data indicate that considerable noise reduction can be achieved in many applications. Finally, an iterative form of the filter is presented. This iterative form further minimizes distortions in band shapes and band intensities when very high levels of denoising are desired. The present filtering approach is an alternative to existing filters, particular when the quality of the residual is important to the user.

A variety of data smoothing techniques exist to address the issue of noise in spectroscopic data. The vast majority, however, require parameter specification by a knowledgeable user, which is typically accomplished by trial and error. In most situations, optimized parameters represent a compromise between noise reduction and signal preservation. In this work, we demonstrate a nonparametric regression approach to spectral smoothing using a spatially adaptive penalized least squares (SAPLS) approach. An iterative optimization procedure is employed that permits gradual flexibility in the smooth fit when statistically significant trends based on multiscale statistics assuming white Gaussian noise are detected. With an estimate of the noise level in the spectrum the procedure is fully automatic with a specified confidence level for the statistics. Potential application to the heteroscedastic noise case is also demonstrated. Performance was assessed in simulations conducted on several synthetic spectra using traditional error measures as well as comparisons of local extrema in the resulting smoothed signals to those in the true spectra. For the simulated spectra, a best case comparison with the Savitzky–Golay smoothing via an exhaustive parameter search was performed while the SAPLS method was assessed for automated application. The application to several dissimilar experimentally obtained Raman spectra is also presented.

A measurement method and apparatus was developed to measure continuously toxic metal compounds in industrial water samples. The method was demonstrated by using copper as a sample metal. Water was injected into the sample line and subsequently into a nitrogen plasma jet, in which the samples comprising the metal compound dissolved in water were decomposed. The transmitted monochromatic light was detected and the absorbance caused by copper atoms was measured. The absorbance and metal concentration were used to calculate sensitivity and detection limits for the studied metal. The sensitivity, limit of detection, and quantification for copper were 0.45 ± 0.02, 0.25 ± 0.01, and 0.85 ± 0.04 ppm, respectively.

A quantitative point measurement of total sodium ([Na]_total) and potassium ([K]_total) in the plume of a burning particle of Australian Loy Yang brown coal (23 ± 3 mg) and of pine wood pellets (63 ± 3 mg) was performed using laser-induced breakdown spectroscopy (LIBS) in a laminar premixed methane flame at equivalence ratios (Φ) of 1.149 and 1.336. Calibration was performed using atomic sodium or potassium generated by evaporation of droplets of sodium sulfite (Na₂SO₃) or potassium sulfate (K₂SO₄) solutions seeded into the flame. The calibration compensated for the absorption by atomic alkalis in the seeded flame, which is significant at high concentrations of solution. This allowed quantitative measurements of sodium (Na) and potassium (K) released into the flame during the three phases of combustion, namely devolatilization, char, and ash cooking. The [Na]_total in the plume released from the combustion of pine wood pellets during the devolatilization was found to reach up to 13 ppm. The maximum concentration of total sodium ([Na]^max_total) and potassium ([K]^max_total) released during the char phase of burning coal particles for Φ = 1.149 was found to be 9.27 and 5.90 ppm, respectively. The [Na]^max_total and [K]^max_total released during the char phase of burning wood particles for Φ = 1.149 was found to be 15.1 and 45.3 ppm, respectively. For the case of Φ = 1.336, the [Na]^max_total and [K]^max_total were found to be 13.9 and 6.67 ppm during the char phase from burning coal particles, respectively, and 21.1 and 39.7 ppm, respectively, from burning wood particles. The concentration of alkali species was higher during the ash phase. The limit of detection (LOD) of sodium and potassium with LIBS in the present arrangement was estimated to be 29 and 72 ppb, respectively.

A goethite-based catalyst was evaluated using in-situ X-ray absorption near-edge structure (XANES) spectroscopy during methane oxidation under increasing reaction temperature. Determination of rank by median absolute deviation (DRMAD), evolving factor analysis (EFA), and multivariate curve resolution (MCR) were used to detect the species present in the catalyst during the reaction and determine their concentration profiles and their pure spectra. The reactants and reaction products were monitored on-line by mass spectrometer. The goethite-based catalyst was active for methane oxidation, with CO₂ and H₂O as the main products. DRMAD and EFA were useful to determine the number of chemical species present in the catalyst structure during reactions. The catalyst presented phase transition during the reaction from goethite to maghemite according to XANES spectra determined by MCR. On the other hand, it was verified that the catalyst presented phase transition from goethite to wüstite in the process in the absence of the oxidant (O₂).