Direct absorption spectroscopy baseline fitting for blended absorption features

Quantification of Elevated Hydrogen Cyanide (HCN) Concentration Typical in a Residential Fire Environment Using Mid-IR Tunable Diode Laser

A versatile portable tunable diode laser based measurement system for measuring elevated concentrations of hydrogen cyanide (HCN) in a time-resolved manner is developed for application in the fire environment. The direct absorption tunable diode laser spectroscopy (DA-TDLAS) technique is employed using the R11 absorption line centered at 3345.3 cm^-1 (2989.27 nm) in the fundamental C-H stretching band (ν₁) of the HCN absorption spectrum. The measurement system is validated using calibration gas of known HCN concentration and the relative uncertainty in measurement of HCN concentration is 4.1% at 1500 ppm. HCN concentration is measured with a sampling frequency of 1 Hz, in gas sampled from 1.5 m, 0.9 m, and 0.3 m heights in the Fireground Exposure Simulator (FES) prop at the University of Illinois Fire Service Institute, Champaign, Illinois. The immediately dangerous to life and health (IDLH) concentration of 50 parts per million (ppm) is exceeded at all the three sampling heights. A maximum concentration of 295 ppm is measured at the 1.5 m height. The HCN measurement system, expanded to measure HCN simultaneously from two sampling locations, is then deployed in two full-scale experiments designed to simulate a realistic residential fire environment at the Delaware County Emergency Services Training Center, Sharon Hill, Pennsylvania.

Dual-logarithmic demodulation method application in a wide gas optical thickness range

Direct absorption spectroscopy (DAS) is an extremely practical and effective technology to detect gas concentration in site applications. Dual-beam subtraction is one of the most common demodulation methods in DAS, yet this method cannot solve the problem of absolute absorption curve nonlinearization in a wide optical thickness range. A real-time and practical dual-logarithmic demodulation method is proposed and proved to be robust when the optical thickness is much greater than linear region. Moreover, the error of optical thickness peak is only 1.18% between the dual-logarithmic demodulation system and simulation after correcting the dual-beam subtraction demodulation system under a 300 K, 1 atm, and 3 m absorption path. When the range of optical thickness peak of acetylene is from 0.0252 to 2.5335 at 1532.83 nm, the peak voltages always maintain satisfactory linearity (R-square=0.9989).

Gaussian process regression for direct laser absorption spectroscopy in complex combustion environments

Tunable diode laser absorption spectroscopy (TDLAS) has been proved to be a powerful diagnostic tool in combustion research. However, current methods for post-processing a large number of blended spectral lines are often inadequate both in terms of processing speed and accuracy. The present study verifies the application of Gaussian process regression (GPR) on processing direct absorption spectroscopy data in combustion environments to infer gas properties directly from the absorbance spectra. Parallelly-composed generic single-output GPR models and multi-output GPR models based on linear model of coregionalization (LMC) are trained using simulated spectral data at set test matrix to determine multiple unknown thermodynamic properties simultaneously from the absorbance spectra. The results indicate that compared to typical data processing methods by line profile fitting, the GPR models are proved to be feasible for accurate inference of multiple gas properties over a wide spectral range with a manifold of blended lines. While further validation and optimization work can be done, parallelly composed single-output GPR model demonstrates sufficient accuracy and efficiency for the demand of temperature and concentration inference.

Blind source separation of coexisting background in Raman spectra

Due to the Raman signal coexists with other scatter spectra which leads to the low ratio of the wanted signal and high background, the appropriate method should be applied to enhance this ratio. The nature of raw spectra is a multi-source system, so its determinacy must be ensured by multi-input. Besides, the faithfulness of output should be provided. Then, the huge fall within the frequencies of Raman and background almost satisfies separating demand for independent component analysis (ICA), and this analysis can give help to the achievement of the two type signals classing and estimate the optimal number of source and match ICA output signals to Raman or background. Thus, based on ICA and the mixing-entropy criteria, the background and Raman adapting calibration kit (BRACK) method is proposed, which is a kind of multiple raw spectral inputs and multiple output (MIMO) method. This method firstly divides the raw data into two parts of Raman and background by ICA, identifies Raman signal by entropy criterion, then restores the part of Raman signal. BRACK method obtains several advantages, for instance, well-adapted, no need for any additional option or extra-intervention, high fidelity, and no unwanted external information. In principle, the correction of background and Raman signals can be expected to be completed by BRACK method.

Terahertz fingerprint characterization of 2,4-dichlorophenoxyacetic acid and its enhanced detection in food matrices combined with spectral baseline correction

Rapid and accurate detection of pesticide residues in food matrices are of great significance to food safety. This study aimed to characterize the fingerprint peaks of 2,4-dichlorophenoxyacetic acid (2,4-D) and to enhance its detection accuracy in food matrices by using terahertz (THz) time-domain spectroscopy. Density functional theory was used to simulate molecular dynamics of 2,4-D peaks (1.35, 1.60, 2.37 and 3.00 THz). Four baseline correction methods, including asymmetric least squares smoothing (AsLS), adaptive iteratively reweighted penalized least squares (AirPLS), background correction (Backcor), baseline estimation and denoising with sparsity (BEADS) were compared and used to eliminate spectral baselines of Zizania latifolia (ZIZLA), rice and maize containing 2,4-D residues, from 0.1 to 4 THz. Based on the peak information of 1.35 THz, the detection limit and accuracy of 2,4-D residues in these food matrices were significantly improved after THz spectral baseline correction, providing a new feasibility for food safety and agricultural applications.

Laser Absorption Sensing Systems: Challenges, Modeling, and Design Optimization

Laser absorption spectroscopy (LAS) is a promising diagnostic method capable of providing high-bandwidth, species-specific sensing, and highly quantitative measurements. This review aims at providing general guidelines from the perspective of LAS sensor system design for realizing quantitative species diagnostics in combustion-related environments. A brief overview of representative detection limits and bandwidths achieved in different measurement scenarios is first provided to understand measurement needs and identify design targets. Different measurement schemes including direct absorption spectroscopy (DAS), wavelength modulation spectroscopy (WMS), and their variations are discussed and compared in terms of advantages and limitations. Based on the analysis of the major sources of noise including electronic, optical, and environmental noises, strategies of noise reduction and design optimization are categorized and compared. This addresses various means of laser control parameter optimization and data processing algorithms such as baseline extraction, in situ laser characterization, and wavelet analysis. There is still a large gap between the current sensor capabilities and the demands of combustion and engine diagnostic research. This calls for a profound understanding of the underlying fundamentals of a LAS sensing system in terms of optics, spectroscopy, and signal processing.