Applied Spectroscopy最新文献

This work implements a mid-level data fusion methodology on spectral data from handheld X-ray fluorescence and laser-induced breakdown spectroscopy analyzers to quantify plutonium surrogate (CeO${}_2$) contamination in soil samples for the first time. Spectral data from each analyzer were used independently to train supervised machine learning regressions to predict Ce concentration. Fused features from both data sets were then used to train the same models, comparing prediction performance by evaluating model precision and sensitivity. Fusing principal component scores from the two sensors yielded an order of magnitude improvement in precision and sensitivity of predictions made with an artificial neural network, compared to predictions made by models trained on independent sensor data. Lastly, a boosted ensemble trained on the fused spectral features yielded an ideal predictor with root-mean-squared error on the order of 10^-6 and calculated limit of detection order 10^-5 wt$\%$.

Volatile organic compounds (VOCs) are an ever-growing hazard for health and environment due to their increased emissions and accumulation in the air. Quantum cascade laser-based infrared (QCL-IR) sensors hold significant promise for gas monitoring, thanks to their compact, rugged design, high laser intensity, and high molecule-specific detection capabilities within the mid-infrared spectrum's fingerprint region. In this work, tunable external cavity QCLs were complemented by an innovative germanium-on-silicon integrated optics waveguide sensing platform with integrated microlenses for efficient backside optical interfacing for the tunable laser spectrometer. The waveguide chip was coated with a mesoporous silica coating, thereby increasing the signal by adsorptive enhancement of VOCs while at the same time limiting water vapor interferences. Different least square fitting methods were explored to deconvolute the resulting spectra, showing subparts-per-million by volume (sub-ppmv) limits of detection and enrichment factors of up to 22 000 while keeping the footprint of the setup small (29 × 23 × 11 cm³). Finally, a use-case simulation for the continuous detection of VOCs in a process analytical technology environment confirmed the high potential of the technique for the monitoring of contaminants. By successfully demonstrating the use of photonic waveguides for the monitoring of VOCs, this work offers a promising avenue for the further development of fully integrated sensors on a chip.

Environmental fluorescence measurements sometimes use water Raman scattering as an internal standard to compensate for path length, lensing effects, and turbidity. Fluorescent dissolved organic matter (FDOM) in water may interfere strongly with the measurement of this reference. However, fluorescence in fluid solution is largely unpolarized, while the OH stretching Raman band of water is always strongly polarized. Using an environmental sample from Lake Wateree in South Carolina, USA, we demonstrate that judicious use of this polarization allows for a significant level of improvement in the contrast or visibility of the water Raman band relative to FDOM.

Combining near-infrared (NIR) and mid-infrared (MIR) spectroscopy to cover both the fundamental and overtone combination molecular vibrational resonances allows more robust analytical methods to be used, such as two-dimensional correlation spectroscopy. However, due to the strong differences in molar absorption coefficients and transparency of the optical material, it is inherently difficult to perform NIR and MIR spectroscopy on aqueous samples using a single instrument. Combining spectra from different instruments and sample presentations can result in unwanted spectral variations, which can influence the prediction models and mitigate the advantages of the combination approaches. In this work, a more consistent instrument response is achieved by combining a single supercontinuum (SC) laser spanning from 1000 to 4000 nm as the light source, with an attenuated total reflection crystal and a transmission cuvette in a single-path configuration. Using this approach, NIR-MIR correlation spectroscopy is demonstrated using a set of 22 aqueous samples with varying concentrations of ethanol, sucrose, and ʟ-proline.

Leptospirosis is an acute bacterial febrile disease affecting humans and animals in many tropical and subtropical countries. This work presents an optimization of surface-enhanced Raman spectroscopy (SERS) substrates to probe vibrational spectroscopic detail from Leptospira deoxyribonucleic acid (DNA). The pathogenic gene of LipL32 was used as a biomarker. The SERS substrates were based on a photonic crystal (PC) structure embedded with bi-metallic gold and silver nanoparticles (PC@AuAg NPs). The localized plasmonic resonance of AuAg NPs was coupled to the Raman modes of the target through SERS interaction. Prior to detection, the AuAg NPs were functionalized with chemical linkers to facilitate specific conjugation between metallic surfaces and DNA biomolecules. The immobilization and hybridization of probe DNA to their complementary target DNA (cDNA) created duplex formation for detection. The configuration was also tested with non-complementary DNA to verify detection specificity. Prominent SERS peaks were recorded, and the characteristic intensity decreased after cDNA hybridization due to less base interaction after complementary pairing. Distinct SERS behavior from the negative control test was also observed in non-complementary interaction. The configuration is highly attractive and can be potentially extended for sensitive and label-free detection of leptospiral DNA, paving the way for alternative diagnosis of leptospirosis.

Fourier transform spectrometers typically use a presumed monochromatic reference source to track and correct errors in optical path difference changes. This paper will conduct a theoretical analysis to show that non-monochromatic light sources with symmetric spectral profiles can also be used as reference sources without adding errors. An experiment was carried out using a symmetric broadband superluminescent diode (SLED) as reference light to measure the spectrum of some other SLED light sources to experimentally demonstrate this finding.

Attenuated total reflection (ATR) spectroscopy in infrared is a standard tool used in most analytical labs, as it allows a rapid chemical analysis with virtually no sample preparation. However, when the sample contains materials with a high refractive index, special care must be taken as the resulting data may be severely biased. This article reports a theoretical approach to correcting distorted ATR spectra. Starting from Snell's law, Lorenz model and Fresnel's equations are combined to obtain the complex relationship between optical constants. With calculating the real and imaginary parts, that is, $n\left(\nu \right)$ and $k\left(\nu \right)$, respectively, of the complex refractive index from the absorption spectrum, a model for mixtures comprising of a liquid and a solid is established. The effects of distortion and potential misinterpretation of the data are discussed. Proof-of-concept experiments with mixtures of carbonaceous materials and toluene confirm the theoretically predicted observations.

Raman spectroscopy is widely used for material detection due to its specificity, but its application to spectral recognition often faces limitations due to insufficient training data, unlike fields such as image recognition. Traditional machine learning or basic neural networks are commonly used, but they have limited ability to achieve high precision. We have proposed a novel approach that combines the Triplet network (TN) and K-nearest neighbor (KNN) techniques to address this issue. TN maps the Raman spectral sequences to a 128-dimensional Euclidean space to extract features, enabling the features in the new space to more accurately represent the similarities or differences between spectra, and then utilizes the KNN algorithm to perform classification tasks in this feature space. Our method exhibits superior performance in recognizing unknown Raman spectra with minimal training samples per class. We employed a handheld Raman spectrometer with an excitation wavelength of 785 nm to collect the Raman spectra of 36 samples, including 28 safe materials and eight hazardous materials. Using only one spectrum as a support set for each category, the hazardous samples were successfully distinguished from the safe samples with an accuracy of 99.6%. Additionally, our model offers adaptability without requiring exhaustive retraining when adding new prediction classes. In situations with high background fluorescence, the TN performs better in measuring the distance between spectra of the same class than traditional distance measurement methods.

In this paper, a new model is presented for estimation of the blood glucose level from the measured near-infrared absorbance. The model has been developed in such a way that the regression coefficients of this linear relation have been approximated by considering only the molar absorptivity of the glucose and the obtained coefficients have been utilized to estimate the blood glucose levels from the measured absorbances. The estimation of the blood glucose concentrations by this blind approach exhibited an acceptable accuracy in comparison to the more accurate principal components regression method. The blood sample absorbances have been measured using a Fourier transform infrared device while the blood glucose levels have been determined by a commercial finger-prick glucometer device.

Respirable dust mass is a prevalent occupational health hazard to the mining workforce. Mineral matrices observed in the mine environment are complex, time varying, and heterogeneous. This poses a challenge to assessing dust exposure using Fourier transform infrared (FT-IR) spectrometry as calibrations for constituent dust species (e.g., crystalline silica) have historically been trained using homogeneous standards or simple mixtures therein. Investigations have considered direct-on-filter analysis, which collects FT-IR spectra directly from sampling filters for calibration, as an alternative. Direct-on-filter analysis using a partial least squares (PLS) method has gained particular interest recently due to the potential to rapidly quantify multiple species from a single filter at the mine site. By design, heterogeneity, and its presumed impact on method accuracy, cannot be addressed in the laboratory when using a direct-on-filter approach motivating the need for more advanced calibration approaches. When heterogeneity is present, mixture of experts (MoE) finite mixture models offer a promising and novel alternative to PLS direct-on-filter analysis as MoE incorporates cluster discovery, regression, and outlier identification into model fitting. Three MoE models of increasing complexity were tasked with determining respirable dust mass in 243 field samples from thirteen active coal, limestone, sandstone, and silver mines. All MoE models, including those using only "expert" spectroscopic predictors or a combination of expert and categorical "gate" variables (e.g., mine type), significantly outperform PLS in terms of accuracy (α = 0.05). Decomposing bias by mine type shows that accuracy generally improves across all types considered when MoE models are not overfitted. The MoE method's effectiveness was linked to its ability to endogenously classify outliers as well as possibly to the use of an additional cluster model for mass predictions. Overall, MoE methods appear as a capable and novel tool to addressing problems of heterogeneity for direct-on-filter quantitative analysis.