Applied Spectroscopy

An alternative approach to conduct attenuated total internal reflection infrared microspectroscopy is described with the use of cartridge-based hemispherical internal reflection elements. The study demonstrates that the devices can be employed on any infrared microscope having reflectance capabilities. A comparison shows that the method provides the same signal-to-noise ratio in comparison to transmission studies for equal sample sizes. In addition, a 4× decrease in the sampled areas inherent with the method was verified for samples 60 μm in diameter and larger. Examples are presented that demonstrate the method's capability of studying small isolated samples without the use of a contaminating mounting media. Examples are also presented that demonstrate the potential to study samples without the effects of diffraction.

An experimental investigation into the use of surface-modified sol-gel-coated Si attenuated total reflectance (ATR) mid-infrared sensors was completed to determine the detection limits for trace amounts of polar organic molecules in aqueous solutions. The surfaces of porous sol-gel films were modified to be mostly hydrophobic in order to concentrate the polar organic analytes in the film while largely excluding the water solvent. This modification of the Si ATR sensor coupled with multivariate partial least-squares calibration methods allowed approximately three orders of magnitude improvement in detection limits for acetone when compared to detection limits from univariate analysis of spectra obtained on uncoated Si ATR sensors. The comparable improvement for isopropanol in aqueous solution was over two orders of magnitude. Studies performed on mixtures of acetone and isopropanol in water confirmed that quantitative analyses could be performed on these mixture systems with cross-validated standard errors of prediction (CVSEP) of 0.5 and 8.7 ppm, respectively. Using three times the CVSEP as an estimate of the detection limit translates to detection limits of 1.5 and 26 ppm for acetone and isopropanol, respectively.

Partial least-squares (PLS) and principal component regression (PCR) methods applied to spectral data can generally provide excellent quantitative analysis precision, but extraction of qualitative spectral interpretation from the models can be more difficult. For example, we have achieved sensitivity in the parts-per-million range for polar organic compounds in aqueous solutions using infrared (IR) spectroscopy and modified sol-gel-coated ATR sensors. The interpretation of PLS or PCR loading vectors obtained from calibrations involving orthogonally designed solutions of acetone and isopropanol in water yields a misleading understanding of the mechanism for the IR detection of isopropanol on these sensors. Examination of the loading vectors from PLS or PCR or the first weight-loading vector from the PLS model would suggest that the spectral calibration is based largely on the interaction of the isopropanol with the surface modifier of the sol-gel coating. However, a classical least-squares (CLS) analysis of the data shows clearly that this interaction is not a significant source of the spectral calibration, but rather the calibration is primarily due to the spectroscopic signal of the isopropanol analyte. In this case, the misleading qualitative interpretation of the PLS and PCR models is the result of the spectral variation being dominated by the effects of spectrometer drift. CLS can overcome this problem if a parameter is included in the CLS calibration that adequately represents the drift. In the example presented here, time of spectral data collection is an appropriate drift-related parameter that can be added to the CLS calibration concentration model in order to provide the qualitative information needed to correctly interpret the spectral data. Other methods to include the effects of spectrometer drift in the CLS model are also presented.

Infrared absorption spectra have become widely used for in situ determination of concentrations in complex mixtures. However, experimental accuracy is often limited by lack of knowledge of the baselines. The mathematical treatment described below allows the spectra to be superimposed on any background with a shape as complex as a parabolic one. The concentrations are thus accurately determined whatever the baseline of the mixture spectrum and even those of the calibration spectra. A mathematical proof of this property is given within the framework of the presently used method, and tests of the efficiency of the method are presented.

Near-infrared single-beam spectra are used to build partial least-squares (PLS) calibration models for the determination of glucose in biological matrices. Two different data sets of the same sample constituents are used in this investigation. The glucose samples consist of an aqueous matrix of varied concentrations of bovine serum albumin (BSA) and triacetin. The BSA and triacetin are models for blood proteins and triglycerides, respectively. Due to the effects of intensity variation in the single-beam spectra, calibration models obtained with unprocessed spectra are not as good as those computed with the corresponding spectra in absorbance units. When this intensity variation is reduced through the use of multiplicative signal correction (MSC), a spectral normalization method, or a logarithmic transform, the resulting models are as good as or better than those obtained in the analysis of absorbance spectra. An attempt is made to model the nonlinear relationship between single-beam spectral intensities and glucose concentrations by use of step-wise quadratic PLS (QPLS) models. The QPLS models are found to perform better than linear PLS models in some cases (e.g., with MSC-corrected single-beam spectra). The effect of digital filtering on the calibration models computed with single-beam spectra is also studied. The results obtained with and without filtering are found to be similar in terms of model performance, but the models based on filtered single-beam intensities require fewer latent variables and perform more consistently as a group. A final test is performed to compare the robustness of calibration models computed with single-beam spectra to those based on absorbance spectra. When applied to spectra that lie outside the time span of the calibration data, the models based on single-beam spectra are still competitive with those computed with absorbance spectra.

A series of solid sodium (hydroxy)aluminates has been prepared from highly caustic aqueous sodium aluminate solutions. The coordination geometry of the aluminum in these compounds was established by ²⁷Al magic angle spinning nuclear magnetic resonance (²⁷Al-MAS-NMR) spectroscopy and was used to identify the Raman and IR vibration frequencies characteristic of tetrahedral (Raman: 440 cm⁻¹; IR: 823 cm⁻¹ and 540 cm⁻¹) and octahedral (Raman: 490–500 cm⁻¹; IR: 728 cm⁻¹) aluminum sites. The vibrational spectra of these solids differ markedly from those observed for concentrated aqueous solutions, and it appears that solid-state vibrational spectra cannot be used to predict the structure of aluminate species existing in solutions.

The infrared absorption, infrared emission, and Fourier transform (FT)-Raman spectra of a series of gibbsites well defined by X-ray diffraction have been obtained. Hydroxyl stretching frequencies were found ∼ at 3670, 3620, 3524, 3452, 3395, 3375, and ∼ 3300 cm⁻¹. Hydroxyl deformation vibrations were observed at 1059, 1023, 969, 938, and 915 cm⁻¹. Hydroxyl stretching bands were observed in the Raman spectra at 3524, 3436, and 3365 cm⁻¹ and correspond well with the three infrared bands. These bands are both Raman and infrared active. The bands at ∼ 3670, 3620, and 3395 cm⁻¹ are infrared active only. The hydroxyl stretching frequencies show a pronounced blue shift, while the hydroxyl deformation modes show a pronounced red shift. Infrared absorption bands were observed at 3413, 3283, and 3096 cm⁻¹ for the hydroxyl stretching frequencies and at 1024, 969, and 914 cm⁻¹ for the hydroxyl deformation frequencies. Low-frequency infrared absorption vibrations were found at ∼ 860, 838, 800, 747, 666, ∼ 625, 585, 560, 522, 452, and 423 cm⁻¹ and infrared emission bands at 834, 778, 728, 652, 640, 609, 580, 512, and 489 cm⁻¹. The infrared emission low-frequency bands moved to higher frequencies upon thermal treatment. The dehydroxylation of gibbsite was followed by the combination of infrared emission spectroscopy and differential thermal analysis over the 200 to 750 °C temperature ranges. Dehydroxylation is followed by the loss of intensity of the hydroxyl stretching frequencies observed at 3620 and 3351 cm⁻¹ and by the loss of intensity of the hydroxyl deformation modes at 1024 cm⁻¹. Dehydroxylation starts at 220 °C and is complete by 350 °C. Some variation in the gibbsite endotherms was found between the synthetic and natural gibbsite dehydroxylations. Spectral changes in the low-frequency bands confirm that dehydroxylation commenced at 220 °C.

We report what is believed to be the first detection of chiral recognition in the inclusion complexes of cyclodextrins by Raman spectroscopy. The spectra of inclusion complexes of N-benzoyl-d-leucine or N-benzoyl-l-leucine with β-cyclodextrin were recorded in the 1500–1800 and 3000–3150 cm⁻¹ regions. These chiral molecules were chosen to provide group frequencies (NH₂⁺ and benzoyl C=O) near the chiral center of the guest as well as COO⁻ and phenyl CH for probing the guest molecule without spectral interference from the vibrational bands of the cyclodextrin host. Frequency shifts and intensity decreases were observed for some Raman bands of N-benzoyl-leucines upon inclusion in the cyclodextrin cavity. In the Raman spectra of the inclusion complex, the frequency shifts were greater for N-benzoyl-d-leucine than for N-benzoyl-l-leucine.

Luminescence and Raman measurements of metal (Er³⁺, Tb³⁺, Sm³⁺, Eu³⁺, UO₂²⁺, and Na⁺) nitrates from multiple spectral windows are combined and analyzed to determine whether this multispectral approach provides any benefit over the best single-measurement, single-window technique by itself. The individual measurements we combined were obtained under the following conditions (1) excited at 488 nm and detected from 500 to 730 nm, (2) excited at 514.5 nm and detected from 530 to 700 nm, and (3) excited at 785 nm and detected from 800 nm to 920 nm. The first two conditions were excellent for producing fluorescence, while the latter was best for Raman spectroscopy because it avoided most fluorescence. Single-window measurements, particularly the near-infrared Raman spectroscopy of the nitrate counterions, were sufficiently similar to make visual identification of the metal salts from the spectra impossible. The combined data set provided easy visual differentiation between the metal salts. However, we found that hierarchical cluster analysis of the single-window measurements could also be used to identify the metal nitrates even when only the Raman spectral windows of the nitrates were studied. To select windows for a multispectral approach rationally, we define an efficacy parameter, π_T, for each possible combination of measurement windows, plus a formula for cost estimation for the multispectral system. This parameter is evaluated for the data presented in the paper, with the conclusion that the least costly acceptable combination consists of only two windows with the same excitation source. In cases such as these, pooled multispectral analysis is proposed as an excellent means of selecting the appropriate technique and wavelength range for an optimal single measurement set, but subsets of the pooled data may function equally well or even better for specific monitoring tasks, especially when cost analysis is necessary. Means of optimizing costs are described briefly, as are means of performing the efficacy calculation when analyses are not of equal importance.

Autofluorescence spectra, autofluorescence images, and near-infrared (NIR) reflectance spectra were tested to determine intramuscular fat and connective tissue in meat slices (longissimus dorsi) from 45 Norwegian Red cattle. Excitation wavelength 332 nm was used to generate fluorescence that was spectroscopically measured and imaged over a total area of 33 and 21 cm², respectively, for each sample. Five types of image features were extracted: the angle measurement technique, the autocovariance spectrum, measures derived from the gray-level co-occurrence matrix, singular value decomposition, and percentage of bright areas in the images. The fat content was 0.5–10.5% (w/wet w) and connective tissue 0.42–0.92% (w/wet w). Partial least-squares regression resulted in the following correlations (R) and root mean square errors of cross-validation (RMSECV) for fat: autofluorescence spectroscopy (0.84, 1.10%), NIR (0.87, 1.01%), and the best image feature (0.78, 1.30%). The best result was obtained by combining autofluorescence spectra with image features (0.92, 0.80%). Connective tissue was poorly predicted (R = 0.41) because of narrow range. The study demonstrates that the combination of spectral data and structural information from images can result in better and more robust prediction models.

The effects of different matrices on the analyte vertical emission profile in the inductively coupled plasma were evaluated with use of a monodisperse droplet sample introduction system in combination with a pneumatic nebulizer. The influence of matrix in the gas phase was separated from influences in the desolvation and volatilization steps. The results suggest that vertical shifts in the analyte emission profiles occur primarily during droplet desolvation and vaporization. The vertical shifts can be correlated to the size of the desolvated particles; larger particles produce emission profiles closer to the load coil. Possible effects of radiative heat transfer were also studied. Strongly absorbing species can increase the evaporation rate of the solvent and produce vertical shifts toward the load coil.

Monodisperse droplet sample introduction was used to study the ionization process in the inductively coupled plasma (ICP). The influence of the matrix on the analyte ion number density was evaluated by using laser-excited atomic fluorescence spectroscopy (LEAFS). The results suggest that the presence of 0.1 M NaCl enhances the ionization of the analyte species. With the addition of the NaCl, the spatially integrated Ba(II) signal decreases, apparently due to enhanced production of Ba²⁺. Added NaCl accelerates the appearance of Sc⁺, but does not affect the spatially integrated signal once ionization is complete. The possible effects on the fluorescence signal of quenching caused by the presence of the NaCl were evaluated by comparison of time-resolved and time-integrated detection. Quenching makes a small contribution to the fluorescence intensity change produced by the matrix but does not account for large loss in signal from singly ionized Ba.

A standard cross-flow nebulizer and commercially available microconcentric nebulizer (MCN-100) have been compared for capillary zone electrophoresis (CZE) inductively coupled plasma mass spectrometry (ICP-MS) measurements. Metallothionein samples were separated and detected to characterize the performance of the two nebulizers for chemical speciation measurements. The MCN-100 offered improved sensitivity and lower detection limits compared to the cross-flow nebulizer, but provided slightly poorer resolution. The detection limit for ¹¹⁴Cd in metallothionein solutions was 90 ng/g with the cross-flow nebulizer and 10 ng/g with the MCN-100 for ∼ 4 nL injections. These values correspond to absolute detection limits of 360 fg Cd in the injected sample with the cross-flow nebulizer and 40 fg Cd for the MCN-100. Quantitation of Cd in metallothioneins (rabbit liver and horse kidney) with the use of a well-characterized rabbit liver metallothionein sample as the calibration standard is reported. Measured Cd concentrations agreed with results obtained by both graphite furnace atomic absorption spectroscopy (AAS) and solution nebulization ICP-MS.

The chemical state of analyte species collected on a water-cooled silica tube during atom-trapping atomic absorption spectrometric determination is investigated with the use of X-ray photoelectron spectroscopy (XPS) for Bi and Mn. Analysis of the Bi 4f_7/2 peak reveals that the chemical state of Bi is +3 during initial trapping (before the atomization stage), but an additional 0-valence state of Bi is also observed after the atomization stage. With the use of the measured Mn 2p_3/2 binding energy together with the observed 3s multiplet splitting, the chemical state of Mn is determined as +2 in all stages. Together with our previous determination of 0 valence for Au, it is now postulated that the stability of certain valence states of the three elements (Au, Bi, and Mn) on the silica matrix can be correlated to their electrochemical reduction potentials.

A Raman microprobe employing a card-based spectrograph and a HeNe excitation source was constructed and evaluated. Particular emphasis was placed on the low-cost room-temperature charge-coupled device (CCD) employed in the instrument. This detector exhibited a relatively high fixed pattern noise [198 e⁻/pixel/s, root mean squared (rms)] and read noise (35 e⁻, rms) compared to conventional charge-coupled detectors employed for Raman spectroscopy. However, by compensating for the fixed pattern noise and signal averaging, we obtained Raman spectra of a variety of common materials with reasonable signal-to-noise ratios at room temperature. Spectra having good signal-to-noise characteristics were collected on polystyrene spheres as small as 5 μm in diameter and a nylon fiber 23 μm in diameter. Although the sensitivity of the detector is limited for weak scatterers, the instrument can be used in specific industrial applications and for educational purposes. Suggestions are made for improving the system.