Applied Spectroscopy

A single-frame approach to chemical imaging with high spectroscopic resolution is described that makes use of a second-generation dimension-reduction fiber-optic array. Laser-induced plume images are focused onto a 17 × 32 rectangular array of square close-packed 25 μm cross-sectional f/2 optical fibers that are drawn into a 544 × 1 distal array with serpentine ordering. The 544 × 1 side of the array is imaged with an f/2 spectrograph equipped with a holographic grating and a gated intensified charge-coupled device (ICCD) camera for spectral analysis. Software is used to extract the spatial/spectral information contained in the ICCD images and de-convolute them into wavelength-specific univariate reconstructed images or position-specific spectra that span an 86 nm wavelength space. Temporal resolution is provided by imaging sequential laser plumes with varying time delays after each laser pulse on the gated intensifier.

The physical release of single granules from individual rat peritoneal mast cells (RPMCs) was monitored by video recording of the degranulating cells by a high-resolution charge-coupled device (CCD) microscope system. The bright granular core disappears from the image as the vesicular content is dissolved on contact with the extra-cellular fluid. After a fixed time delay, the exocytotic product, serotonin, was detected by capillary electrophoresis coupled with laser-induced native fluorescence (LINF-CE). The timing of the two events is mostly correlated, which supports a fast release mechanism of the granular products.

A computational design tool has been developed and applied to a new fluorescence filtering scheme for the CH A²Δ_↔X²π transition, which uses diagonal excitation and observation to provide the highest possible excitation and emission strengths in systems with high quenching. A holographic notch filter is proposed to reject elastic scatter while a 10 nm bandpass filter isolates CH fluorescence from other emissions. The computer simulation of the fluorescence process was employed to determine the combination of filters and excitation wavelengths that provided the highest fluorescence signal. Optimum filter configurations were determined for a spectrum of pressures, temperatures, and quenching environments. The analysis shows that this technique is best suited to high-pressure environments where saturation is not possible, and the signal strength of this diagonal excitation/observation technique in such environments is at least an order of magnitude greater than that for previous off-diagonal techniques.

Solid-matrix fluorescence (SMF) and solid-matrix phosphorescence (SMP) have been used in conjunction with solid-phase microextraction to characterize mixtures of polycyclic aromatic hydrocarbons (PAHs) isolated from water. Whatman 1PS paper was used to extract the PAH from water, and then the isolated PAHs were directly identified on the paper by obtaining SMF and SMP spectra. The SMF and SMP properties of 10 PAH were obtained, and the PAHs in a two-component mixture, a three-component mixture, and a four-component mixture were easily identified by a combination of SMF and SMP. No external heavy atom was needed to acquire the SMP data. Benzo[e]pyrene gave a limit of detection of 6.2 pg/mL with SMP, and with SMF benzo[a]pyrene gave a limit of detection of 19 pg/mL.

NMR diffusion and relaxation measurements were applied to study the ibuprofen–protein interaction by changing the drug concentration in solutions containing 0.1 M human serum albumin (HSA). On the basis of the assumption that the drug molecules undergo fast exchange between the binding and free states, the association constant, K_a, for drug–HSA complexes is found to be about 60, and the number of binding sites, n, is about 50. The small association constant and the big binding site number suggest weak interactions between ibuprofen and HAS.

The interactions of methylene blue (MB, a cationic redox indicator and biological stain) and sodium dodecyl sulfate (SDS, a micelle-forming, anionic surfactant) in aqueous solution have been examined by using Rayleigh scattering, UV-visible absorption, and fluorescence spectroscopy. At SDS concentrations significantly below the critical micelle concentration (cmc), MB forms noncovalent dimers and aggregates with SDS that scatter light but do not fluoresce. For solutions containing 1 μM MB and < 3–5 mM SDS, shifts in the absorption spectrum characteristic of the formation of MB H-aggregates are noted. There appears to be little effect on the fluorescence emission spectrum, indicating that these MB aggregates do not fluoresce appreciably. At and above the known SDS cmc, MB is observed to interact with the micelles. The MB excited-state fluorescence lifetime (380 ps) remains constant until SDS micelles form, then increases to 615 ps. The MB rotational reorientation time similarly increases from 105 to 500 ps between 6 and 8 mM SDS. This finding suggests that the MB is encountering, on average, a microenvironment in the SDS micelles that is 5-fold more viscous than liquid water or the molar volume of the MB/SDS species that is reorienting is 5-fold larger than MB in water.

The extent of leaching of indicator dyes of varying molecular mass, following encapsulation into hydrated sol-gel monoliths prepared from tetramethylorthosilicate (TMOS), was investigated. Dye modification of a water-soluble phthalocyanine with poly(ethylene glycol) side chains (2000 MW) reduced leaching relative to the unmodified dye, but significant losses were still observed. Monoliths doped with fluorescein modified with dextrans of higher molecular weight were investigated in an attempt to determine the minimum size necessary for dye retention. In comparison to monoliths doped with unmodified fluorescein, attachment to a dextran carrier significantly decreased the extent of dye leaching. However, even a 70 000 MW carrier was not sufficient to prevent some loss of dye. In contrast, no leaching of myoglobin (18 000 MW) occurred, which shows that when the entrapped molecule has a organized tertiary structure, diffusion within the pores is eliminated. Malachite green and its dextran conjugate were used to examine the effect of electrostatic interactions on leaching. Minimal loss of both the dye and the conjugate was observed, which indicates that when a favorable interaction between the dye and sol-gel matrix exists, attachment to a macromolecular carrier may not be necessary to prevent leaching.

Two systems have been compared for the on-line production of infrared spectra of compounds separated by high-performance liquid chromatography (HPLC). System 1 had a thermospray interface to evaporate the mobile-phase solvents and deposit the solutes onto a moving stainless steel belt, for direct analysis by reflection–absorption IR spectrometry. In system 2, the column effluent was split 1:6 and pumped through an ultrasonic nebulizer. The spray was desolvated at reduced pressure in a heated transfer tube and the solutes deposited onto a ZnSe window for direct analysis by transmission IR spectrometry. When system 1 was used for the analysis of reactive dyes, eluent containing ammonium acetate caused a large background spectrum, as a film of unknown composition was formed on the surface of the stainless steel belt. An alternative procedure, without ammonium acetate, was developed with a mixed-mode column containing C18 and cation exchange particles, but variable retention times were obtained. System 2 could be used satisfactorily with eluents containing ammonium acetate, and it produced dye spectra of better resolution than those obtained with system 1. When the dyes were examined under a microscope, the material on the ZnSe window was more uniform and more evenly distributed on a narrower track than were the deposits on the stainless steel belt. Consequently, the peak shape of the IR chromatograms was better with system 2, and a lower detection limit was achieved for reactive blue 74 (9 ng). When mixtures of the pesticides metolcarb and carbofuran were analyzed, a thermospray temperature of 196 °C was required to desolvate the effluent from the HPLC column in system 1. At this temperature, thermal degradation of the pesticides occurred, and no spectrum was recorded from the belt surface. However, with system 2, deposition of the pesticides was achieved at 90 °C, without degradation, and IR spectra/chromatograms were obtained. Overall, the study showed that the infrared chromatograph (IRC) was the more efficient and flexible interface for HPLC Fourier transform infrared (FT-IR) spectrometry.

Temperature-programmed diffuse reflectance infrared Fourier transform spectroscopy (TPDRIFTS) has been used to characterize a reacted selective catalytic reduction (SCR) vanadia/titania catalyst. At room temperature the reacted catalyst presents ammonium ions adsorbed on the surface. On heating under N₂ flow, a surface redox reaction takes place; ammonium ions are oxidized to bridged nitrosyls species, while V⁵⁺ = O groups are reduced to V³⁺. Reduction of V⁵⁺ species to V³⁺ ions occurs at lower temperatures than those observed in conventional TPR processes under H₂ atmospheres. At higher temperatures, bridged nitrosyls species become monodentate, first in an angular configuration and then in a linear one. Previous data on the SCR reaction let us propose that they are desorbed from the surface as gaseous N₂O.

The diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) spectrum of the peeled wood of kenaf (Hibiscus cannabinus L.), cotton (Gossypium barbadense), and pine (Pinus brutia) revealed three broad absorptions in the region of 1800–1540 cm⁻¹. The deconvolution of spectra in this region revealed the existence of 10 major peaks contributing to that spectrum in the aforementioned region. The absorption peaks at 1710 and 1608 cm⁻¹ have been assigned to carboxylic (–COOH) and carboxylate (–COO⁻) groups, respectively. Since their relative size is pH dependent, the ratio of their area was used as a pH indicator. A linear relationship was then established between that ratio and the amount of acid need to bring the pH of the wood to 3, as it is measured by the pH determination of a water extract of the wood. This is the pH at which the wood is processed during manufacturing of chipboards.

We have used a miniaturized Wollaston wire resistive thermometer as a probe to record infrared absorption spectra of polymeric samples by detecting photothermally induced temperature fluctuations at the sample surface. This method opens the way to absorption Fourier transform infrared spectroscopy/microscopy with a spatial resolution that is no longer diffraction limited, but is determined instead by the size of the contact between probe and sample. At present, this is on the order of a few hundred nanometers. The thermal probe, of a type used in scanning thermal microscopy and microthermal analysis, allows us to detect the photothermal response of a specimen exposed to the beam of a Fourier transform infrared spectrometer and heated thereby. The signal from this probe measures the resulting temperature fluctuations, and thus provides an interferogram which replaces the interferogram normally obtained by means of direct detection of the IR transmitted by a sample.

Experimental improvements have been made in the UV-visible absorption spectroscopy technique applied to propellant flame diagnostics. The two-dimensional feature of an intensified charge-coupled device (CCD) detector was used to simultaneously record multiple, spatially distinct absorption spectra over a region of 0.35 cm. Temporal resolution has been increased to 1 ms by pulsing a simmering xenon arc lamp. The resulting increase in the light intensity by 30 to 70 times over the nonpulsed output provides the necessary light flux to achieve single-pulse, multiple absorption spectra. Species with low concentrations can be measured with the inclusion of multiple-pass optics to increase the effective pathlength through the combustion region. Due to broadband UV-visible absorption observed in propellant flame spectra, typically only 20% of the incident light is transmitted. However, inclusion of a neutral-density filter during the measurement of the incident intensity I₀ allows the dynamic range of the detector to be effectively increased by a factor of 5. With these improvements, temperature and OH and NO concentration maps, with 1 ms temporal resolution, have been determined with two different propellant flames.

We have completed an experimental study to investigate the use of infrared emission spectroscopy (IRES) for the quantitative analysis of borophosphosilicate glass (BPSG) thin films on silicon monitor wafers. Experimental parameters investigated included temperatures within the range used in the microelectronics industry to produce these films so that the potential for using the IRES technique for real-time monitoring of the film deposition process could be evaluated. The film properties that were investigated included boron content, phosphorus content, film thickness, and film temperature. The studies were conducted over two temperature ranges, 125 to 225 °C and 300 to 400 °C. The latter temperature range includes realistic processing temperatures for the chemical vapor deposition (CVD) of the BPSG films. Partial least-squares (PLS) multivariate calibration methods were applied to spectral and film property calibration data. The cross-validated standard errors of prediction (CVSEP) from the PLS analysis of the IRES spectra of 21 calibration samples each measured at six temperatures in the 300 to 400 °C range were found to be 0.09 wt % for B, 0.08 wt % for P, 3.6 nm for film thickness, and 1.9 °C for temperature. Upon lowering the spectral resolution from 4 to 32 cm⁻¹ and decreasing the number of spectral scans from 128 to 1, we were able to determine that all the film properties could be measured in less than one second to the precision required for the manufacture and quality control of integrated circuits. Thus, real-time in situ monitoring of BPSG thin films formed by CVD deposition on Si monitor wafers is possible with the methods reported here.

Dehydroxylation of diaspore has been followed by a combination of differential thermal analysis (DTA) and infrared emission spectroscopy (IES). The DTA endotherm of diaspore is not simple, and two inflections are observed at 505 and 531 °C. Dehydroxylation shows that diaspore loses 15% of its weight upon dehydroxylation. The Mn-substituted diaspore shows major endotherms at 620 and 649 °C with a broad endotherm at 539 °C. Infrared absorption bands of diaspore were observed at 3365, 3284, and 3095 cm⁻¹ in the hydroxyl stretching region and at 911, 755, 706, 668, 649, 570, and 544 cm⁻¹ in the low-frequency region. Infrared emission bands were observed at 3239 and 2958 cm⁻¹ in the hydroxyl stretching region. Low-frequency infrared emission bands in agreement with the absorption bands were observed at around 916, 752, 706, 668, 649, 570, and 544 cm⁻¹. The variation of intensity of the hydroxyl stretching frequencies with temperature follows a pattern similar to that for the DTA curves.

The objective of this study was to determine the potential of using mid- or near-infrared diffuse reflectance spectra to construct food ingredient spectral libraries for product identification and checking. Samples (106) consisting of buttermilk, dehydrated onion, cheese and milk-egg powders, wheat flours, and two powdered seasonings were scanned “as is” (not diluted with KBr using diffuse reflectance) at 4 and 16 cm⁻¹ resolution in the mid-infrared on a Digilab FTS-60 and a Perkin-Elmer Model 2000 and in the near-infrared on an FTS-60 (4 and 16 cm⁻¹ resolution) and NIRSystems Model 6500 scanning monochromator (10 nm bandwidth). A custom-made sample transport device was used on the FTS-60, a rotating sample cup on the NIRSystems 6500, and a stationary cell for the Perkin-Elmer 2000. Every third sample of each group was used as a test sample and searched against a library containing the remaining samples. Results showed that only full spectrum based searches with the use of Euclidian distance or correlation (with or without a first derivative) were useful. All unknowns were correctly classified by using near-infrared spectra generated on either the scanning monochromator or the FTS-60 (4 cm⁻¹ resolution) or by using any of the mid-infrared spectra. Results demonstrated that near- or mid-infrared spectral libraries of powdered food ingredients can be used for product identification and checking.

Silica optical fibers have been modified to improve their sensitivity as evanescent wave sensors for liquids. A section of fiber is stripped of its cladding and bent or coiled, which significantly increases the evanescent wave interaction with the surrounding environment. The effect of sensor shape, solute concentration, refractive index, and temperature on the sensor response is discussed.

A portable fiber-optic Raman system comprising a diode laser, echelle spectrograph, charge-coupled device (CCD) detector, and filtered fiber-optic probe has been used to analyze chemical agents and other toxic chemicals in sealed glass containers. These containers include ampoules and bottles that are contents of chemical agent identification sets (CAIS) developed for use in training military personnel in chemical agent identification, safe handling, and decontamination. Real-time nonintrusive analysis of these sets is required so that the items containing chemical agents can be identified for proper disposal. This paper details the laboratory measurement of Raman spectra of CAIS chemicals, the determination of detection limits for mustard gas and lewisite in chloroform, and the analysis of CAIS items in the field.

A procedure has been developed to permit recalculation of a Raman spectrum that has been taken on a source spectrometer to allow its quantitative use with data taken from a target system. The procedure is based on accurate characterization of both the source and target spectral abscissas by means of a neon spectrum that is recorded at the time the Raman measurements are made. Corrections are made for wavelength-dependent behavior of both systems, for differences in dispersion for the two systems, and for variation in the number, spacing, and size of the detector elements. An example is presented in which the fingerprint and CH regions of the Raman spectrum of cyclohexane are transferred from higher to lower resolution context.

Vapor-deposited Ag and Au films exhibit high surface-enhanced Raman scattering (SERS) sensitivity. With thin metal films, SERS spectra can be obtained by using a “backside” configuration in which the laser light is focused through the substrate onto the metal film/liquid (or gas) interface. Use of the “backside” configuration is particularly advantageous for obtaining SERS spectra of aqueous samples. Because excitation occurs only at the metal/film/liquid interface, both attenuation of the signal by water and fluorescence interference are minimized. However, noble metals, such as Ag and Au, do not adhere well to glass substrates, and significant exposure to water damages the films. In this paper, three methods of adhering Au to glass and their effect on the SERS response are examined. The methods include the use of a buffer metal between the SERS-active metal film and the glass substrate; the use of (3-mercaptopropyl)trimethoxysilane (MCTMS); and the use of organometallic paints. Of the methods examined, the best film adherence and SERS response was obtained by using MCTMS to bind the metal film to a chemically etched glass slide.

Normal, polarized, and cross-polarized Raman spectra of a single crystal of cafarsite, Ca₈(Ti, Fe, Mn)_6–7(AsO₃)₁₂·4H₂O, from Binntal (Switzerland) are reported. Comparison with known metal oxides was used to achieve assignments of the observed bands. Differences between the polarized and cross-polarized spectra helped to identify the symmetric and asymmetric modes. The bands at 258, 319, 721, and 763 cm⁻¹ are assigned to the ν₄(E), ν₂(A₁), ν₃(E), and ν₁ arsenite modes after comparison to AsF₃. The bands at 204, 437, and 620 cm⁻¹ are assigned to the asymmetric Fe(2)–O modes, while the bands at 232, 288, 464, and 506 cm⁻¹ are the symmetric ones. The two bands at 837 and 880 cm⁻¹ are ascribed to the symmetric and asymmetric Fe(1)–O modes, respectively. The Ti–O modes are assigned to the bands at 204 and 648 cm⁻¹ (asymmetric) and to 319, 390, and 802 cm⁻¹ (symmetric). Bands at 288 and 587 cm⁻¹ represent the symmetric and asymmetric modes, respectively, of the Ca–O bonds. On the basis of the Raman spectra, no distinction can be made between the two crystallographic Ca sites. Also no difference can be made between the two crystallographic Mn sites in cafarsite. Mn–O bands are observed at 335, 437 (asymmetric), 464, 506, and 601 cm⁻¹ (symmetric).