SPECTROSCOPY AND SPECTRAL ANALYSIS

News

Forgot your password?

Enter your e-mail address to receive an e-mail with your account information.

Journal Online

Archive

Links

	CAST

	CNCOS

	CNKI

	WanfangDATA

	CHEMSOC

	sinospectroscopy

	CPS

Quick Search

Adv Search

2025 Vol. 45, No. 09 Published: 2025-09-01

	2401	Application of Several New X-Ray Fluorescence Spectroscopy Techniques in Geological and Geochemical Analysis
		YUAN Jing, ZHANG Hua, SHI Lei, HUANG Hai-bo, TAN Gui-li, LIU Jian-kun, YU Jun-jie^*
		DOI: 10.3964/j.issn.1000-0593(2025)09-2401-09
		X-ray fluorescence spectroscopy (XRF) has been widely used in the field of geoscience due to its simple preparation, rapid analysis, environmental friendliness, high sensitivity, and non-destructive examination. Nowadays, Earth system science research has gradually become the focus of Earth science, encompassing global climate change, the interaction of the Earth's spheres, and the environmental changes caused by human activities. The boom in these areas raises new demands for elemental analysis, such as the spatial distribution and speciation of elements, real-time geochemical data in the field, accurate and rapid quantification of low atomic number elements, and the optimization of instrument resolution, sensitivity, detection limit, and so on. These demands also promote the emergence and progress of some new XRF techniques and their application in the field of geoscience. Synchrotron-based micro-X-ray fluorescence spectroscopy and X-ray absorption fine -structure spectroscopy permit in situ mapping of interest elements in samples at the sub-micrometer scale, owing to the superiority of the light source. This allows for the spatially resolved determination of element distribution and speciation, as well as the element oxidation state and coordination environment. The μ-XRF of a laboratory light source is time-saving and convenient compared to a synchrotron radiation device. Additionally, the analysis accuracy of low atomic number elements using laboratory μ-XRF has significantly improved through the upgrade of the X-ray optical tube, detector, and focusing optical system. PXRF enables the rapid analysis of field samples due to its miniaturization and simple sample processing. The XRF core scanner enables high-precision, continuous, and rapid scanning of large-batch cores, thereby overcoming the cyclical limitations of laboratory analysis. This paper reviews the application of several XRF techniques in geological and geochemical analysis over recent years, aiming to provide some ideas and inspiration for researchers in XRF studies and geoscience fields.
		2025 Vol. 45 (09): 2401-2409 [Abstract] ( 32 ) RICH HTML PDF (15413 KB) ( 41 )

	2410	Research Progress on Geographical Origin Discrimination of Traditional Chinese Medicines Based on Near-Infrared Spectroscopy
		ZHANG Xin-zhi¹, SAIMAITI Patiguli¹, ZHANG Lu-wen¹, YANG Ya-fei^2, BIAN Xi-hui^{1, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)09-2410-07
		The quality of traditional Chinese medicines (TCMs) significantly impacts their efficacy and medication safety, with the place of origin being one of the crucial factors. Traditional methods for identifying the origin of TCMs primarily include macroscopic identification and microscopic identification. The macroscopic identification methodsrely on manual observation and experience, which are prone to misjudgment. The microscopic identification methods require processes such as slicing and staining of the TCMs, which are cumbersome and not applicable to rare TCMs. In recent years, spectral analysis has attracted increasing attention in the identification of TCM origins due to its advantages of rapidness and non-destructiveness. Among these, near-infrared spectroscopy (NIR) is widely employed. Therefore, this review summarizes the research on the origin identification of TCMsusing NIR over the past 10 years. It conducts geographical zoning of TCMs and systematically summarizes the distribution of Dao-di herbs in China, as well as the research popularity of different TCMs for origin identification. Currently, research on the origin identification of TCMs mainly focuses on precious and easily counterfeited herbal medicines in terms of origin, such as Panax notoginseng, Panax ginseng, Gastrodia elata Blume, and Dendrobium officinale. In the application of NIR for origin identification of TCMs over the last decade, the combination of traditional NIR with chemometric methods has remained the mainstream approach. Although two-dimensional NIR correlation spectroscopy has been applied in various fields, including food, agriculture, and industry, its use for the origin identification of TCMs has been limited to the past two years, making it an emerging technology in this field. NIR hyperspectral imaging can simultaneously obtain spectral and image information of samples, thereby improving identification accuracy. The combination of NIR with ultraviolet-visible (UV-Vis) spectroscopy can obtain information on chemical bonds and conjugate relationships of compounds in samples. NIR combination with mid-infrared spectroscopy can provide more abundant information on molecular skeletons, and its combination with laser-induced breakdown spectroscopy (LIBS) can obtain information on molecular vibration and elemental composition. The multi-spectral hyphenated technology based on NIR achieves multi-dimensional information complementarity, effectively overcoming the limitations of single-spectral technology, and enhances the effectiveness of identifying the origin of TCMs. This review also summarizes the chemometric methods used in the origin identification of TCMs, including spectral pre-processing, variable selection, and chemical pattern recognition. Spectral pre-processing methods can be divided into smoothing, scattering correction, baseline correction, and scaling, which are often used to eliminate the influences of noise, baseline, and background. Variable selection methods remove redundant variables, further improving the accuracy of discrimination models. Deep learning algorithms are increasingly applied in the analysis of the origin identification of TCMs. This review provides a methodological framework for the rapid and accurate identification of the origin of TCMs.
		2025 Vol. 45 (09): 2410-2416 [Abstract] ( 27 ) RICH HTML PDF (4651 KB) ( 45 )

	2417	Research Progress in the Design of Carbazole-Based Organic Room Temperature Phosphorescence Materials
		WANG Xiao-ao¹, ZHAO Lu¹, BAI Yun-feng^{1, 2}, FENG Feng^1
		DOI: 10.3964/j.issn.1000-0593(2025)09-2417-11
		Organic room temperature phosphorescence (oRTP) materials have the advantages of good processability, excellent biocompatibility, low biotoxicity, and low cost, which have become the focus of functional materials research. Among them, carbazole-based oRTP materials have been rapidly developed -in terms of molecular structure design, performance regulation, and optimization, and have been more widely used in anti-counterfeiting, biooptical imaging, information encryption, and other applications. Carbazole is a nitrogen-containing heterocyclic compound with a rigid planar conjugated structure and excellent hole transport properties and thermal stability. In addition, carbazole's planarization structure is conducive to intermolecular stacking interaction, thus strengthening the intermolecular electron coupling effect and stabilizing the triplet excitons. In addition, carbazole, as an excellent chromophore, has been extensively modified to obtain better RTP properties. This paper describes the luminescence mechanism of oRTP materials. It summarizes the design strategies of carbazole-based oRTP materials with long phosphor lifetime, including: (1) Intermolecular electron coupling: face-to-face packing or H-aggregation can stabilize triplet excitons, which is generally conducive to efficient and long-lived RTP. (2) Spin-orbit coupling: the strong spin-orbit coupling of heavy atoms can efficiently promote intersystem crossing (ISC) of electrons from the first electron excited singlet state to the lowest excited triplet state (S₁-T₁) or from singlet state to triplet state (S_n-T_<i>n, n≥1) and induce phosphorescence emission. (3) Exciton separation systems: by constructing a distorted donor-acceptor system, the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) are separated, thereby reducing the energy gap (ΔE_ST) between singlet and triplet states and increasing the intersystem crossing rate (k_ISC).(4) Polymer encapsulation system: polymer encapsulation can build a dense and rigid external environment to effectively inhibit non-radiative transition and oxygen diffusion, significantly improve RTP efficiency, and extend RTP life. (5) Hydrogen bond system: the interaction of hydrogen bonds greatly weakens molecular vibration and non-radiation inactivation, improves its tolerance to temperature, and can shield water and oxygen. Then, the application of carbazole-based oRTP materials in anti-counterfeiting and other aspects is introduced. Finally, the challenges faced in this field and the research directions worthy of attention are discussed. In summary, challenges and opportunities coexist; it can be expected that -research on carbazole-based oRTP materials will become a new shining point in the field of organic luminescent materials. With continuous exploration and innovation, such materials will likely show a broader application prospect in the future.
		2025 Vol. 45 (09): 2417-2427 [Abstract] ( 26 ) RICH HTML PDF (28329 KB) ( 11 )

	2428	Reconstruction of Chinese Paintings Based on Hyperspectral Image Fusion Using Res-CAE Deep Learning
		ZHU Shi-hao¹, FENG Jie^1*, LI Xin-ting¹, SUN Li-cun¹, LIU Jie², YUAN Ping¹, YANG Ren-xiang¹, DENG Hong-yang¹
		DOI: 10.3964/j.issn.1000-0593(2025)09-2428-09
		Traditional color reproduction methods often suffer from complex preprocessing steps and reliance on subjective selection of spectral features. Moreover, the exclusive use of spectral reflectance data neglects spatial information, limiting reconstruction to isolated color points rather than full scenes. To overcome these limitations, this study proposes a deep learningbased method using a Residual-Convolutional Autoencoder (Res-CAE) to jointly extract and reconstruct spatial and spectral features from hyperspectral data cubes. The Res-CAE model was trained on the CAVE hyperspectral dataset and evaluated across five testing scenarios: a standard 24-color chart (X-Rite), a custom Chinese painting color chart, in-training and out-of-training random scenes, and a real Chinese painting scene captured under CIE standard observer conditions. Evaluation metrics included color difference (ΔE₀₀), root mean square error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM). Experimental results demonstrate that Res-CAE outperforms traditional methods, such as bilinear interpolation and principal component analysis (PCA), in both color fidelity and image quality. On the 24-color chart, the model achieved an average ΔE₀₀ of 0.694 5, RMSE of 0.009 2, PSNR of 35.92, and SSIM of 0.995 6. These results validate the effectiveness of Res-CAE in high-fidelity color reconstruction from hyperspectral data, offering practical value for digital preservation of traditional Chinese paintings.
		2025 Vol. 45 (09): 2428-2436 [Abstract] ( 27 ) RICH HTML PDF (27454 KB) ( 22 )

	2437	Research on Spectral Simulation Method of Space-Borne Limb Imaging Spectrometer
		ZHAO Min-jie¹, SI Fu-qi^1*, ZHOU Hai-jin¹, JIANG Yu¹, WANG Shi-mei¹, ZHAN Kai¹, YAN Ge²
		DOI: 10.3964/j.issn.1000-0593(2025)09-2437-08
		The spaceborne limb imaging spectrometer (LIS) operates in a sun-synchronous orbit and uses limb scanning to detect oxygen A-band airglow. Due to the strong absorption of oxygen, it is difficult to observe airglow emission from the ground, which limits the LIS's detection capability. Therefore, the spectral simulation method is discussed in this paper. Firstly, we obtain the LIS's working parameters, including a track altitude of 520 km, a scanning altitude of 10~100 km, and a scanning interval of 2 km. Secondly, based on laboratory calibration, spectral and radiation response parameters were obtained. The spectral calibration matched the lamp peaks and pixels. The polynomial fitting results show that the spectral range of LIS is 498.1~802.3 nm, with a spectral resolution of 1.38 nm obtained by Gaussian fitting. For radiometric calibration, the radiometric relative deviation at different pointing angles is ±0.5%. The least squares method was used to fit the radiance and response Digital Number (DN) values to obtain the radiation calibration coefficients for all integration times (ranging from 25 ms to 3 200 ms) with a radiation calibration uncertainty of 3.6%. Based on the airglow transmission length, the Mass Spectrometer and Incoherent Scatter(MSIS) model, and High-resolution Transmission molecular Absorption(HITRAN) database,the airglow transmittance is calculated. The results show that airglow transmittance is less affected by oxygen absorption above 80 km, with a transmittance of 0.9, and stronger oxygen absorption at 60 km, with a transmittance of less than 0.05. Based on the airglow observed by the Scanning Imaging Absorption spectrometer for Atmospheric CHartographY (SCIAMACHY), the onion-peeling method was used to obtain the volume emission rates, and then high-resolution airglow emissions were calculated, which vary at different target heights. Finally, the high-resolution airglow emission convolved with the response function, combined with the calibrated radiation response coefficient, yields the DN value of the airglow response of LIS. The results show that LIS can effectively detect airglow emission, and the spectral shape can characterize its temperature dependence. LIS has a good signal-to-noise ratio at the longest exposure time of 3.2 seconds. Through this simulation, the airglow detection capability and inversion algorithm of imaging spectrometers can be evaluated, providing scientific support for the exploration of the middle and upper atmosphere.
		2025 Vol. 45 (09): 2437-2444 [Abstract] ( 19 ) RICH HTML PDF (8949 KB) ( 7 )

	2445	Study on the Characteristics of Femtosecond Laser-Induced Cyano Chemiluminescence Under Low-Pressure Environment
		SHAN Yuan¹, YAN Hao², LIU Xu-hui², HAN Lei¹, LU Zheng¹, LIU Zi-han¹, LI Bo¹, GAO Qiang^1*
		DOI: 10.3964/j.issn.1000-0593(2025)09-2445-07
		Femtosecond laser molecular tagging velocimetry is currently the most mainstream non-intrusive velocimetric technique and is widely used in velocity field measurement in atmospheric pressure environments, especially having great advantages in supersonic and hypersonic flow field measurement. Currently, femtosecond laser molecular tagging velocimetry methods primarily include femtosecond laser electron excitation tagging velocimetry for tagging nitrogen molecules and femtosecond laser-induced cyano chemiluminescence velocimetry (FLICC) for tagging methane/nitrogen. Compared to femtosecond laser electron excitation tagging velocimetry, FLICC technology offers significant advantages in signal strength and duration, and its velocity measurement range and applicable scenarios have also been significantly expanded. This technology can provide technical support for flow velocity measurement in environments such as high-speed wind tunnels and aerospace propulsion systems. However, high-speed flow fields are often accompanied by low-pressure environments, such as those found in low-pressure wind tunnels on the ground, near-Earth, or in deep space. Therefore, it is of great significance to study flow field measurement technology in a low-pressure environment. Currently, the applicability of FLICC technology for velocity measurement in low-pressure environments remains undetermined. Since FLICC technology obtains velocity information by capturing the displacement of tagged luminescent molecules over a specific period, the intensity and duration of the tagged molecule's luminescence directly determine the feasibility of its application. In low-pressure environments, due to the decrease in particle number density, the interaction between the femtosecond laser and molecules, as well as the energy transfer between particles, is affected, thereby altering the intensity and lifetime of luminescence. The main focus of this study is to investigate the luminescence characteristics of FLICC under atmospheric to low-pressure environments. In the experiment, the pressure in the low-pressure chamber can be adjusted between 10 Pa and 0.1 MPa, and the chamber is filled with a 1% concentration of CH₄/N₂ mixture. A femtosecondlaser is incident into the low-pressure chamber and interacts with the mixed gas, inducing a chemical reaction to generate high-energy CN. Fluorescence is emitted through the transition of CN (B-X), and the spectrum is imaged using an ICCD camera and spectrometer. By capturing spectral information at different delays, the intensity and duration of various luminescence spectral lines of CN under different pressures are obtained, and a curve of spectral intensity versus pressure is established. The results show that as the pressure decreases, the luminescence intensity of CN gradually decreases. At the pressure of 10 Pa, it still maintains a good signal intensity. Through the delayed imaging of an ICCD camera, its fluorescence lifetime is about 5 μs, which meets the requirements of FLICC velocimetry technology. This research lays the foundation for the application of FLICC technology in low-pressure environments.
		2025 Vol. 45 (09): 2445-2451 [Abstract] ( 21 ) RICH HTML PDF (6612 KB) ( 7 )

	2452	Detection of Dissolved CO₂ in Water Based on Near-Infrared TDLAS and Degassed Membrane
		WEI Zhi-han, ZHANG Jun-sen, GU Xuan, XU Pan, HE Bo-yang, YE Biao, WEI Xin-xin, MENG Xin, XUE Guo-gang^, WANG Jing-jing^
		DOI: 10.3964/j.issn.1000-0593(2025)09-2452-07
		Inland waters, as a key component of the carbon cycle, the monitoring of dissolved gases within them is of paramount importance for evaluating carbon sources/sinks and climate change. Existing in-situ dissolved gas detection technologies for water primarily focus on dissolved gas detection in seawater, characterized by low detection limits, high technical complexity, elevated costs, large instrument size, and limited applicability in inland waters. In light of the high solubility of CO₂ in inland waters, the complex and variable environment, and the objective of cost reduction, this study constructed a measurement system for detecting the solubility of CO₂ in water by employing low-cost domestic butterfly-shaped near-infrared (1 603 nm) lasers, compact optical multi-pass cells, and degassing membranes. This system comprises a CO₂ gas detection system based on tunable diode laser absorption spectroscopy (TDLAS) and a water-gas separation device utilizing membrane degassing technology. The entire system can be completely integrated within two 15-inch instrument cases, achieving a high level of miniaturization. By measuring a set of CO₂ standard gases of varying concentrations, the calculation formula and performance of the TDLAS gas detection system for CO₂ gas phase concentration were determined. The lower limit of CO₂ measurement by the system is 3.91×10^-4 (integration time 5 s). An innovative and straightforward method for preparing standard aqueous solutions of CO₂, based on a gas detection system and Teflon gas bags, was proposed. The relationship between the liquid-phase CO₂ solubility and the gas-phase CO₂ concentration after degassing was thoroughly explored. It was determined that, under constant degassing process and parameters, a linear relationship exists between the solubility of CO₂ in water and the concentration of the evolved gas, and an inversion formula for the liquid-phase CO₂ solubility was established through linear fitting. The noise situation and detection sensitivity of the system were deeply investigated using methods such as Allan variance. The system's sensitivity for detecting the solubility of CO₂ in water reaches 1.15 μmol·L^-1 (integration time: 100 s), with a detection lower limit of approximately 3.22 μmol·L^-1. Finally, the solubility of CO₂ in various water samples was measured. The results indicate that this system can effectively detect the solubility of CO₂ in water samples such as tap water, pure water, river water, and fish tank water, and the measured values are largely consistent with the research and analysis results, demonstrating the outstanding performance and reliability of the device in practical application scenarios. This study offers an important technical reference for the development of small and low-cost in-situ dissolved gas detection equipment for water. It holds significant significance in water ecological environment monitoring and greenhouse gas management.
		2025 Vol. 45 (09): 2452-2458 [Abstract] ( 28 ) RICH HTML PDF (17002 KB) ( 13 )

	2459	Research on the Recognition of Mixed Mid-Infrared Spectra of Hazardous Chemicals Based on an Improved High-Order Residual Network
		FAN Bing-rui^{1, 2}, ZHAI Ai-ping¹, WANG Dong¹, LIANG Ting³, ZHANG Gen-wei^2, CAO Shu-ya^2
		DOI: 10.3964/j.issn.1000-0593(2025)09-2459-08
		Amid the rapid development of the chemical industry, the large-scale production and use of hazardous chemicals have brought significant environmental risks and potential threats to human health. Consequently, there is an urgent need to develop efficient monitoring and identification methods to address these challenges. Mid-infrared (MIR) spectroscopy, characterized by its unique molecular “fingerprint” recognition capability and high sensitivity, has been widely applied in the identification and analysis of hazardous chemicals. Meanwhile, recent breakthroughs in deep learning, particularly in feature extraction and object detection, have provided novel solutions for addressing complex data analysis problems. To meet the need for precise identification of hazardous chemicals under complex background conditions, this study proposes a novel model that combines Depthwise Separable Convolution (DSC) and a High-Order Residual Network (HORN), referred to as DSC-HORN. This model is designed to enhance further the efficiency of feature extraction and the accuracy of recognition for complex MIR spectral data. By leveraging depthwise separable convolution, the DSC-HORN model decomposes standard convolutions into depthwise and pointwise convolutions, resulting in a reduction of approximately 65.89% in parameter count and computational complexity. This optimization increases the operational speed of the DSC-HORN model by approximately four times compared to traditional one-dimensional convolutional neural networks (1D-CNNs), while significantly reducing memory consumption and enhancing its adaptability in resource-constrained environments. For the experiments, this study collected MIR spectral data of a 10% DMMP solution mixed with 11 types of background materials. After data preprocessing and removal of outlier samples, a total of 528 valid spectral datasets were obtained for model construction. To ensure balanced distribution of different labeled samples across subsets and to accurately reflect the overall data characteristics, the dataset was divided into training, validation, and testing sets using a stratified random sampling method, with a 6∶1∶3 ratio for training, validation, and testing, respectively. The experimental results demonstrated that the DSC-HORN model exhibits significant advantages in recognition efficiency and accuracy compared to existing models. Compared to traditional models such as K-Nearest Neighbors(KNN), Random Forest(RF), Robust Mode Regression Sparse Non-negative Matrix Factorization(MR-NMF), and Backpropagation Neural Networks(BP), as well as deep learning models like1D-Convolutional Neural Network(1D-CNN), the DSC-HORN model achieved a recognition accuracy of 96.84%. This represents improvements of 5.7%, 13.93%, 5.07%, 10.76%, and 6.33%, respectively, over the models above. These performance enhancements can be attributed to the parameter optimization achieved through depthwise separable convolution and the efficient feature extraction enabled by high-order residual connections. The results further confirm that the DSC-HORN model is not only an efficient and accurate MIR hazardous chemical recognition model but also provides a novel technical pathway for real-time monitoring and precise identification of hazardous chemicals. Additionally, the MIR recognition technology, based on the improved high-order residual network, has great potential to promote further the widespread application of MIR spectroscopy in complex scenarios.
		2025 Vol. 45 (09): 2459-2466 [Abstract] ( 19 ) RICH HTML PDF (7128 KB) ( 8 )

	2467	Spectroscopic Characterization of Turquoise From Nanhuatang, Shiyan, Hubei
		WANG Yi¹, YANG Ming-xing^{1, 2*}, DUN Jin-han¹, JIANG Yan¹, LIU Ling¹, YUAN Ye^{1, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)09-2467-08
		The EYS(E-Hu-Shaan) turquoise mining area is a significant turquoise-producing region in China, with the Nanhuatang turquoise deposit located in the Yunyang District of Shiyan City, Hubei Province. This study systematically investigates Nanhuatang turquoise samples using conventional gemmological testing, infrared spectroscopic analysis, Raman spectroscopy, laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS), and UV-visible spectroscopic analysis. The aim is to elucidate the mineralogical and spectroscopic characteristics of these samples and provide data support for provenance tracing.Results indicate that the chemical composition of Nanhuatang turquoise is complex, with significant enrichment in Na, Mg, K, Ca, Sc, Ti, V, Cr, Zn, Mo, and Ba. Notably, the concentrations of Ca, Cr, K, Mo, and Ba are considerably higher than those observed in other turquoise deposits. Rare earth element (REE) analysis reveals a left-dipping chondrite-normalized distribution pattern, characterized by low total REE content (ΣREE), relative depletion of light REEs, and relative enrichment of heavy REEs, with significant fractionation between the two groups. Raman spectral analysis showed that the main absorption peaks were located at 3 470 cm^-1 (hydroxyl group stretching vibration) and 1 040 cm^-1 (phosphate symmetry stretching vibration). UV-Vis absorption spectroscopy demonstrates that the coloration of turquoise is closely related to its chemical composition.The d—d electron jump of Cu²⁺ ions (670~690 nm absorption band) is a key factor in the blue colouration of turquoise, while the electron jump of Fe³⁺ ions is located at 428 nm.Based on the above analysis, the paper reveals the geological features of the Nanhuatang turquoise source area, providing data support for conducting turquoise provenance studies.
		2025 Vol. 45 (09): 2467-2474 [Abstract] ( 19 ) RICH HTML PDF (22269 KB) ( 20 )

	2475	A Study of Gas Absorption Spectral Inversion Method Based on K-MLE With Wide Concentration Range
		YU Qin-yi^{1, 2}, FAN Bo-qiang¹, YOU Kun¹, HE Ying¹, ZHANG Shi-qi^{1, 2}, CAI Ao-xue^{1, 2}, HAN Ling-ran^{1, 2}, ZHANG Yu-jun^1*
		DOI: 10.3964/j.issn.1000-0593(2025)09-2475-09
		The gas absorption spectroscopy technique, based on the Lambert-Beer Law, is extensively employed for the detection of various pollutant gases. However, challenges arise due to the limitations of devices, the saturation absorption of gases, and other interfering factors. When measuring gases at high concentrations, these issues create a nonlinear relationship between absorbance and concentration, resulting in significant deviations in gas detection across a wide concentration range.In this paper, we leverage the linear signal correlation between the measured spectra and reference spectra to reflect the relationship between the concentrations of the two. We propose employing Maximum Likelihood Estimation (MLE) to determine the linear correlation coefficient K between spectra signals, facilitating the inversion of gas concentrationsover a wide concentration range.To determine the linear correlation coefficients across variable spectral intervals in the gas absorption band, we implement a moving internal algorithm with adjustable parameters. The interval-wise statistical inference is carried out on the set of correlation coefficients, and optimal parameter estimators, K-values, are derived by constructing a piecewise likelihood function. This approach effectively mitigates fluctuations in measurement results caused by signal perturbation.Ultimately, gas concentration inversion across a wide concentration range is achieved through a combination of concentration threshold and compensation equation. Using sulfur dioxide (SO₂) gas concentration detection as a case study, we process thirty sets of spectral datasets acquired under each predetermined concentration condition. The accuracy of the concentration inversion methodologywas quantified through the relative error of the mean concentrations calculated from thirty replicate measurements. At the same time,its stability wasevaluatedusing the standard deviation of the concentration values. Under identical low-concentration conditions, the broadband method, narrowband method, and K-MLE algorithm demonstrated mean retrieved concentration errors of 6.14%, 22.13%, and 0.11% respectively. Under the premise of ensuring the stability of the results, the K-MLE algorithm achieved superior measurement accuracy. Furthermore, across a wide concentration range of 0~508 μL·L^-1, the maximum mean error for concentration inversion of the K-MLE method is only 0.98%, with a maximum standard deviation of 1.25, outperforming the other two conventional methods. The results demonstrate that the K-MLE method offers an effective approach for nonlinear offset compensation, enabling accurate and stable gas concentration inversion across a wide range of concentrations.
		2025 Vol. 45 (09): 2475-2483 [Abstract] ( 26 ) RICH HTML PDF (6515 KB) ( 3 )

	2484	A Waste Plastic Classification Method Based on Laser Spectral Fusion Detection Technology
		FANG Jia-xuan, DONG Xi-wen, XU Zi-rui, QU Dong-ming, YANG Guang^, SUN Hui-hui^
		DOI: 10.3964/j.issn.1000-0593(2025)09-2484-07
		Plastic is a commonly used polymer in daily life. With the increasing amount of waste plastic, the resulting environmental pollution has become more severe, making the classification and recycling of waste plastic an urgent issue. Different types of plastics require different recycling methods, so researching plastic classification methods is of great significance. Laser-Induced Breakdown Spectroscopy (LIBS) is an elemental analysis technique based on atomic emission spectroscopy, offering advantages such as rapid analysis, no sample pretreatment required, and in-situ analysis, which provides convenience for plastic classification. Raman Spectroscopy (RS) is a molecular structure characterization technique based on Raman scattering theory, which offers advantages such as simultaneous multi-element analysis, low sample quantity requirements, and minimal sample damage, also facilitating plastic classification. This paper will utilize LIBS and RS technologies to collect spectral information from both atomic and molecular perspectives of plastics, and then merge the two types of spectral information to obtain a fused spectrum. By using LIBS spectra, RS spectra, and fused spectra in conjunction with the Random Forest machine learning algorithm (RF) to build models for plastic classification and identification, a comparison of the classification accuracy of the three models reveals that the fused spectrum can improve classification accuracy. During the model-building process, with the same number of test sets, the number of training sets affects both the model construction time and classification accuracy. Experiments were conducted on the accuracy and model construction time for different ratios of test sets to training sets, concluding that a ratio of 1∶3 is the most suitable, achieving an accuracy of 96%. In addition to the impact of the training set, the preprocessing methods of spectral data also affect the classification accuracy of the plastic fusion spectrum. The experiment-employed a sparsity-based baseline estimation denoising method to process the fusion spectral data and rebuild the model, thereby increasing the classification accuracy of plastics to 100%. The experimental results indicate that when the ratio of the test set to the training set is 1∶3, the fused spectral data has a significant advantage in classification accuracy compared to single spectral data. The classification accuracy of the preprocessed fused spectral data can be improved to 100%.
		2025 Vol. 45 (09): 2484-2490 [Abstract] ( 26 ) RICH HTML PDF (7704 KB) ( 16 )

	2491	Origin Discrimination and Soluble Protein Content Prediction of Dried Daylily Based on Near Infrared Spectroscopy
		ZHANG Xue-li^{1, 2}, YANG Hao^{1, 2}, LI Chen-fei^{1, 2}, SUN Yi-le^{1, 2}, LIU Zong-lin^{1, 2}, ZHENG De-cong^{1, 2}, SONG Hai-yan^{1, 2*}
		DOI: 10.3964/j.issn.1000-0593(2025)09-2491-05
		Daylily is rich in nutrients and has high edible, medicinal, and economic value. It has many producing areas in China. The origin discrimination and soluble protein content prediction of daylilies are of great significance to the quality management of daylilies, the establishment of an agricultural product brand, and the development of the local economy. Because fresh daylily contains a variety of alkaloids, it is not suitable to eat in large quantities. Therefore, most of the daylilies on the market are dried daylilies. In this paper, the origin discrimination model and soluble protein content prediction model for dried daylily were established based on near-infrared spectroscopy. To address the issues of low discrimination accuracy and inaccurate content prediction in the original algorithm, the model was enhanced, resulting in a significant improvement in accuracy through the combination of various preprocessing methods and characteristic wavelength screening algorithms. In this study, Partial Least Squares Discriminant Analysis (PLS-DA), Random Forest (RF), and Support Vector Machine (SVM) were combined with Multiplicative Scatter Correction (MSC), Standard Normal Variate (SNV) and Savitzky-Golay smoothing (SG) respectively to establish the origin discrimination models of dried daylily and compare the model discrimination results. The experimental results show that PLS-DA combined with MSC has the best effect on origin discrimination, with an accuracy of 93.33%. The precision and recall of the three origins are all above 85%, with an average precision of 91.9% and an average recall of 91.9%. It demonstrates that the model exhibits good accuracy and stability, and can effectively distinguish the origin of dried daylilies. At the same time, Partial Least Squares Regression (PLSR) was combined with a variety of preprocessing methods and three characteristic wavelength screening algorithms: Unobserved Variable Elimination (UVE), Competitive Adaptive Reweighted Sampling (CARS) and Successive Projections Algorithm (SPA), respectively, to establish the prediction models of soluble protein content of dried daylily and compare the prediction results. The results show that the model established by PLSR, combined with SG and CARS, has the best predictive effect. The determination coefficient R² reached 0.981 5, and the Root Mean Square Error of Prediction (RMSEP) was 0.021 4 g·kg^-1. Compared with the original PLSR, the R² increased by 0.12, and the RMSEP decreased by 0.033 1 g·kg^-1. This prediction model can well predict the soluble protein content of dried daylily.
		2025 Vol. 45 (09): 2491-2495 [Abstract] ( 22 ) RICH HTML PDF (3846 KB) ( 19 )

	2496	Identification of Sandstone Lithology Based on Hyperspectral Combined With Support Vector Machine Algorithm
		JIAO Long¹, LI Ying¹, TONG Rong-chao², WANG Cai-ling³
		DOI: 10.3964/j.issn.1000-0593(2025)09-2496-06
		Accurate identification of sandstone lithology is crucial in resource exploration, geological engineering, and building material research. Hyperspectral is an emerging analytical method with the advantages of spectral integration, a large amount of information, fast analysis speed, and non-destructive testing. It overcomes the problems of long time and complex procedures of traditional analytical methods. The support vector machine (SVM) method has strong learning and generalization ability, and is a fast and accurate analysis method. Therefore, hyperspectral analysis combined with support vector machine modeling established the identification method of different lithological sandstones. Four types of sandstone samples with different lithologies were collected, and their hyperspectral data were collected. Hyperspectral data were preprocessed by standard normal variable transformation (SNV), multiple scattering correction (MSC), and Savitzky-Golay smoothing method (SG), respectively. After that, partial least squares discriminant analysis (PLS-DA) and SVM methods were used to establish classification models. In the SVM model, the Gaussian kernel function (RBF kernel) is selected to establish the SVM classification model. The grid search method optimizes the penalty parameter C and the kernel function gamma parameter in the SVM. The value of C is determined to be (0.1, 1, 10, 100), and the radial basis function gamma parameter is (0.01, 0.1, 1, 10). 16 parameter combinations are formed, and the classification models are established respectively. The five-fold cross-validation classification accuracy of the best PLSDA and SVM models reaches 93.20% and 96.40%. The best MSC-PLSDA and SNV-SVM models established can accurately identify the test set samples. The classification accuracy of the test set of the MSC-PLSDA model reaches 89.00%, and the corresponding F1 values reach more than 80%. The classification accuracy of the test set of the MSC-SVM model reaches 96%, and the corresponding F1 value of the model reaches more than 90%. Among them, the recognition accuracy of argillaceous and fine-grained sandstone is the highest, and the F1 value reaches 100%. The results show that hyperspectral technology combined with the support vector machine method is a reliable method for sandstone lithology identification and analysis, and the spectral preprocessing method has a significant impact on the identification accuracy, which provides a new idea for sandstone lithology identification and analysis systems.
		2025 Vol. 45 (09): 2496-2501 [Abstract] ( 28 ) RICH HTML PDF (8758 KB) ( 18 )

	2502	Raman Spectral Sample Data Enhancement Method Based on Voigt Function
		BU Zi-chuan, LIU Ji-hong^*, REN Kai-li, LIU Chi, ZHANG Jia-geng, YAN Xue-wen
		DOI: 10.3964/j.issn.1000-0593(2025)09-2502-09
		Raman spectroscopy technology plays an important role in modern analytical chemistry and physical chemistry, particularly in providing important information about the structure and properties of substances. In recent years, the development of deep learning and artificial intelligence technologies has provided new directions for Raman spectroscopy analysis, excelling in spectral classification and identification. However, the performance of deep learning models is highly dependent on the scale and quality of the data. The acquisition of Raman spectroscopy data is time-consuming and of a single type, which cannot meet the model's training needs. Therefore, data enhancement techniques should be employed to augment the training data. Due to the problems above, a Raman spectrum sample data enhancement method based on the Voigt function is proposed. By fitting the spectral peaks in the spectrum with a Voigt function, each fitting peak is randomly shifted left and right, and the amplitude is changed within the specified range. Finally, all the fitting peaks are linearly superimposed to achieve the purpose of enhancing the data set. In this paper, this method is compared with three commonly used Raman spectroscopy data enhancement methods (noise addition method, offset method, and left-right translation method) under a set of Raman spectroscopy data with small sample sizes and multiple categories. Two evaluation indexes evaluate the quality of the enhanced spectra generated by different methods: structural similarity (SSIM) and PCA total interpretation variance ratio, and the models were trained and tested using three classification models: k-nearest neighbors (KNN), support vector machine (SVM), and one-dimensional convolutional neural network (1D-CNN) to evaluate the classification performance and generalization ability of the models. The results show that the Raman-enhanced spectrum obtained by the Voigt function method performs well in both evaluation indices. In the classification model trained on the enhanced data set, the accuracy of the verification set and the original data set in the three classification models namely, the noise addition method, the offset method, and the Voigt function method is 100%. In contrast, the classification model of the left and right translation method performs poorly, with accuracies of 99.80% and 96.35% (KNN), 98.75% and 100% (SVM), and 94.89% and 98.54% (1D-CNN), respectively. In simulated generated abnormal data, models trained with data enhanced by common methods performed poorly for certain types of abnormal data, while models trained with data enhanced by the Voigt function method performed excellently in various types of abnormal data. In summary, the Raman spectroscopy sample data augmentation method based on the Voigt function can effectively increase the diversity of the augmented samples, and the trained models exhibit good generalization ability and robustness, making them suitable for scenarios that require high abnormal data processing capabilities and high generalization ability. This method has certain application value in the field of Raman spectroscopy analysis technology.
		2025 Vol. 45 (09): 2502-2510 [Abstract] ( 30 ) RICH HTML PDF (8102 KB) ( 12 )

	2511	Composition and Color Characteristics of Ming Dynasty White Glazed Tiles of Yaoli Kiln in Jingdezhen
		YU Yong-bin¹, CHEN Tian-min^2*, QIU Xiao-xin³, WU Jun-ming³
		DOI: 10.3964/j.issn.1000-0593(2025)09-2511-06
		In this study，energy dispersive X-ray fluorescence spectroscopy and a colorimeter were used to test and analyze the composition and chroma of the body glaze of the Ming Dynasty white glaze ceramic tile samples from Yaoli kiln in Jingdezhen，and to explore their composition，color, and raw material formula. The results show that the overall characteristics of the body composition of the white glazed ceramic tiles in the kiln are similar，with the typical characteristics of the Southern “high silicon and low aluminum” ceramic body. The average content of Al₂O₃ is about 21.4%，the average content of SiO₂ is 71.2%，the average content of Fe₂O₃ is 1.2%，and the total content of flux oxides K₂O，Na₂O，CaO, and MgO is 5.1%，which is close to the chemical composition of Jingdezhen white porcelain body in the early Ming Dynasty. Kaolin is introduced into the body of the Yaoli kiln ceramic tile，and the “binary formula” of porcelain stone and kaolin is used to make the quality of the ceramic tile body better. The main cosolvent in the glaze of white glazed tiles unearthed from the Yaoli kiln in Jingdezhen is likely CaO, with an average content of 10.29%. The contents of K₂O and Na₂O are relatively low, with an average of 3.22% and 0.52%, respectively. The CaO of Yaoli ceramic tile glaze should come from glaze ash，and glaze ash or plant ash should be introduced into the glaze formula. There are many kinds of glaze ash and glaze fruit，resulting in a large fluctuation of Fe₂O₃ content in porcelain glaze. There are 9 pieces of calcium alkali glaze and 11 pieces of calcium glaze，and the glaze formula is unstable. Chromaticity analysis shows that the main wavelength of Jingdezhen white glaze ceramic tile is approximately 515 nm, which falls within the green band of visible light waves, closely resembling the green glaze of white glaze ceramic tiles in the Ming Dynasty. The iron content ( average value is about 0.48%，0.61%，0.64%，0.99% ) in the glaze formula of the four regional samples gradually increased from yellow to light and green to green，which is consistent with the glaze color of the Ming Dynasty green and white glaze ceramic tiles of Yaoli Kiln in Jingdezhen.
		2025 Vol. 45 (09): 2511-2516 [Abstract] ( 24 ) RICH HTML PDF (6399 KB) ( 9 )

	2517	Online Monitoring for Surface Integrity of γ-TiAl in Laser Shock Peening Based on Plasma Spectroscopy
		SHI Guang-yuan¹, WANG Yuan-bin^{1, 2, 3*}, WANG Ying-hao¹, SHAN Meng-jie¹, DING Lei-yi¹, CUI Min-chao^{1, 2, 3}, LUO Ming^{1, 2, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)09-2517-09
		This study focuses on the plasma spectroscopy phenomena during the Laser Shock Peening (LSP) process, aiming to explore the correlation between spectral characteristics and surface integrity. It proposes an online monitoring method based on plasma spectroscopy to enable intelligent monitoring of the LSP process. First, the parameters of the plasma spectroscopy acquisition system are introduced, followed by a detailed analysis of the plasma characteristics during LSP. Using the Boltzmann two-line method with N Ⅱ 500.515 nm and N Ⅱ 399.5 nm spectral lines, the apparent plasma temperature is estimated to range between 16 000 and 22 000 K. Additionally, the electron density, calculated using the Hα 656.27 nm spectral line, is approximatelyin the range of 2.287 to 3.612×10¹⁶ cm^-3. As the laser power density increases, both plasma temperature and electron density exhibit an overall increasing trend. However, fluctuations are observed due to the inherent spatiotemporal instability of the LSP process.Subsequently, the study investigates the strengthening effects of LSP on the surface integrity of γ-TiAl alloy. Experimental findings show that as the laser power density increases, the surface residual stress rises from an initial value of -61.498 to -444.224 MPa, while the surface Vickers hardness improves from 317.8 to 385.5 HV. Based on these experimental data, polynomial fitting models are developed to predict surface residual stress and Vickers hardness, using the intensity ratio of the Hα line and the N Ⅱ 500. 515 nm line as independent variables. Both models achieve determination coefficients (R²) exceeding 90%, providing a reliable foundation for quantitative predictions of surface integrity. To enable efficient online monitoring, the study proposes an end-to-end monitoring method based on a CNN-Transformer deep learning architecture. By processing plasma spectroscopy data, the model performs monitoring of the LSP process. Experimental results demonstrate a classification accuracy of 99.3%, highlighting the approach's efficiency and reliability for online monitoring. In conclusion, by integrating physical modeling and deep learning techniques, this study establishes the correlation between plasma spectroscopy and surface integrity during the LSP process, providing an innovative and reliable approach for intelligent online monitoring of LSP.
		2025 Vol. 45 (09): 2517-2525 [Abstract] ( 22 ) RICH HTML PDF (9193 KB) ( 18 )

	2526	Dual-Mode Fluorescence Enhancement of Skin-Print Using NaYF₄:Yb, Er/CDs Nanocomposites
		LI Wei-lin¹, WANG Meng^2*, XU Ying¹, HUANG Yi-xiang³, LIU Jian-yu¹
		DOI: 10.3964/j.issn.1000-0593(2025)09-2526-09
		Skin-print development, as one of the essential technologies in the field of forensic science, is the basic premise for skin-print analysis and identification. Based on the research achievements of latent finger-print development using fluorescent nanomaterials, this paper puts forward a method of skin-print development followed by a strategy of dual-mode fluorescent enhancement using the nanocomposite (NCs) composed of rare earth doped up-conversion fluorescent nanomaterials (UC NMs) and carbon dots (CDs), to improve the quality of skin-print development. NaYF₄:Yb, Er UC NMs and CDs down-conversion (DC) fluorescent NMs were respectively synthesized by thermal decomposition method and solvothermal method, then NaYF₄:Yb, Er/CDs NCs with both UC and DC fluorescent properties were thus formed by loading the CDs onto the surface of NaYF₄:Yb, Er NMs. Through optimization, the doping ratio of the CDs suspension to NaYF₄:Yb, Er powder was determined to be 0.017 5 μL·mg^-1. The as-prepared NCs were quasi-spherical in shape, with an average diameter of 22.2 nm, and were modified with carboxyl groups on their surface. The characteristic peaks assigned to NaYF₄:Yb, Er, and CDs appeared in both the Raman spectrum and the X-ray diffraction spectrum of NaYF₄:Yb, Er/CDs NCs. The NCs exhibited an ultraviolet (UV) absorption peak at 359 nm and displayed a DC fluorescent emission at 468 nm upon excitation with 365 nm UV light. Additionally, the NCs exhibited a near-infrared (NIR) absorption peak at 977 nm, which could produce three UC fluorescent emissions at 520, 540, and 654 nm upon excitation with 980 nm NIR light. In addition, it was confirmed that both the CDs suspension and the NaYF₄:Yb, Er/CDs powder exhibit wavelength-dependent excitation properties. The as-prepared NCs were subsequently used for powder-dusting development, followed by dual-mode fluorescence enhancement of latent skin prints, including fingerprints, palm prints, footprints, lip prints, and pinna prints. Experimental results showed that, excited with 365~415 nm light under DC fluorescence enhancement mode, the skin-print could emit seven colors of fluorescence, including red, orange, yellow, yellow-green, green, blue-green, and blue; excited with 980 nm light under UC fluorescence enhancement mode, the skin-print could emit green fluorescence. After powder-dusting development and fluorescence enhancement, a strong contrast between the developing signal and background noise could be achieved, allowing for well-defined outlines, coherent ridges, and clear minutiae to be obtained. Additionally, a strong contrast between papillary ridges and furrows could be observed. The skin-print development and enhancement using NaYF₄:Yb, Er/CDs NCs proposed in this study was proved to be suitable for dealing with the skin-prints on smooth objects with high sensitivity, contrast, and selectivity.
		2025 Vol. 45 (09): 2526-2534 [Abstract] ( 16 ) RICH HTML PDF (72000 KB) ( 8 )

	2535	Optimization of Data Quality Objectives Under Control of Near Infrared System for Diesel Cetane Number
		LI Ying¹, PAN Zhi-qiang², WANG Dou-wen^3*
		DOI: 10.3964/j.issn.1000-0593(2025)09-2535-08
		The DQO modeling for the NIR-CN system has not yet seen mature research results, in which comparison with data from the Waukesha CN engine was conducted to alleviate the composite matrix effect of matching error and the uncontrollable β risk at a single level. Compared to the power function, quadratic linear combination, and slope radian angles, the DR bias correction model's maximum likelihood estimate is found to be the best-fitting line. The DR modeling, which was derived from the variable iteration of two-dimensional data weighted by the CSS, was established under the premise of the i.i.d. assumption. Consequently, the correlation coefficient cannot be regarded as the criterion for judging the model, but rather through the lack of fit and the AD goodness-of-fit test. This test can effectively mitigate the cumulative impact and interaction interference of ACF, and maximize compensation for the limitations of subjective trend analysis in control charts. The research results indicate that the DR model aligns more closely with the DQO robustness and the actual situation of the NIR system. The NIR system was previously limited to a single CN level discussion; now it has been expanded to study variations at multiple levels. Different decisions lead to different uncertainties, and wrong decisions will incur additional cost losses. Under the uniform principle of risk and cost, the variation of CN over time will inevitably lead to the choice of CSS. As long as the DQO set by the NIR system is subsequently met, the risk arising from the use conditions can be controlled within the demonstrated CSS range.
		2025 Vol. 45 (09): 2535-2542 [Abstract] ( 23 ) RICH HTML PDF (2753 KB) ( 8 )

	2543	Study on Infrared Absorption Spectra of Ethanol in Small Slits
		LIU Yun, LI Chun, GUO Da-bo, LEI Wen^, YUAN Guang^
		DOI: 10.3964/j.issn.1000-0593(2025)09-2543-07
		Infrared (IR) spectroscopy is a tool routinely used to analyze substances. It is based on the resonant absorption of infrared radiation by molecular vibrations to obtain information about the structure and chemical composition of a sample. IR enhancement is a crucial topic in IR research, and increasing the IR absorption range is a common approach. However, IR enhancement under small cavity conditions has unique application requirements. The enhancement of infrared radiation capacity within the small cavity suggests the potential for increased infrared absorption. In this study, the infrared spectral absorption properties of ethanol molecules in slit cavities are investigated. A wedge-shaped slit structure was developed, creating slit regions of varying thicknesses between the double-polished single-crystal silicon wafer and zinc selenide plate, with slit thicknesses less than 1 μm, along with an external temperature control device. The temperature dependence of the infrared absorption of ethanol in slits of different thicknesses was investigated. The infrared absorption spectrum of ethanol exhibits prominent absorption peaks in the wavenumber ranges of 3 690~2 650 and 1 150~1 010 cm^-1. The variations in the positions of the main absorption peaks and the non-absorption peaks of ethanol with temperature were compared. It was found that the absorbance at the non-absorption peaks increases with the intensity of blackbody radiation. In contrast, the intensity at the main absorption peaks of ethanol decreases relative to the intensity of blackbody radiation. This reflects that the radiative capacity of ethanol molecules in the slit decreases with temperature. For different Slit thicknesses, the absorbance of ethanol molecules at the same temperature varies. The maximum absorbance for different vibrational modes corresponds to different slit thicknesses. The slit thickness that shows the maximum absorbance at wavenumbers of 3 335, 2 975, 2 930, and 2 881 cm^-1 is 0.73 μm. For the wavenumbers of 1 090 and 1 050 cm^-1, the slit thickness that exhibits the best enhancement in absorbance is greater than or equal to 0.64 μm. These results show that the slit cavity can enhance the IR absorption of ethanol molecules; however, different vibrational modes of ethanol molecules correspond to different optimal slit thicknesses. The absorption coefficients of the characteristic absorption peaks of ethanol decrease with increasing temperature.
		2025 Vol. 45 (09): 2543-2549 [Abstract] ( 24 ) RICH HTML PDF (5214 KB) ( 6 )

	2550	Determination and Analysis of Rb/Sr Ratio in Yixing Purple Sand Using Energy Dispersive X-Ray Fluorescence Spectrometry
		XU Wei-xuan¹, CHEN Wen-bin^2*, SHAN Ai-xian³
		DOI: 10.3964/j.issn.1000-0593(2025)09-2550-07
		The natural isotope⁸⁷Rb, a trace element rubidium in geology, is radioactive and can be used to determine geological age. The ratio of rubidium to strontium (Rb/Sr) and the strontium isotope ratio (⁸⁷Sr/⁸⁶Sr) are commonly used scientific methods for studying the provenance characteristics of ancient ceramic raw materials. ⁸⁷Sr is primarily formed through the radioactive decay of ⁸⁷Rb. The rate of rubidium decaying into strontium is related to the chemical weathering history and geological age. Since the Ming Dynasty, Huanglong Mountain in northwest Dingshu Town, Yixing, has served as the primary source of purple clay raw materials due to its abundant mineral deposits. This experiment aims to perform a non-destructive determination of Rb, Sr, and the Rb/Sr ratio using energy-dispersive X-ray fluorescence spectrometry on a select number of samples from historical Yixing purple sand works of different periods. The Rb/Sr ratio in the ore materials is related to the depth of the ore layer, influenced by factors such as chemical weathering history and leaching of surface water and rainwater. The deeper the geological layer corresponding to the raw materials, the lower the Rb/Sr ratio; shallow surface ore materials experience greater leaching of Sr, resulting in a higher Rb/Sr ratio.The experimental results show that during the Ming and Qing dynasties, when traditional hand tools were used for mining, the Rb/Sr ratio of the samples was relatively close. From the late Qing dynasty to the 1960s, due to modern mining operations (projects of Wells No. 1 to No. 3) based on the foundations of the Ming and Qing quarries, the Rb/Sr ratio of the samples was slightly lower than during the Ming and Qing periods, though the difference was not significant. From the 1970s to the 1980s, the primary source of ore materials was mechanized deep-well mining at Well No. 4, which had a digging depth of -80 meters, significantly exceeding the previous depths of -20 to -50 meters, resulting in a notable decrease in the Rb/Sr ratio of the samples. In recent decades, most of the ore materials from Huanglong Mountain have been sourced from surface layers or manually excavated materials, resulting in Rb/Sr ratios in the samples that are generally higher than those of historical samples from earlier periods. According to the experimental data, the Rb/Sr ratios of purple clay samples from different periods are related to the depth of the ore layers. They are consistent with the documented depths of mined layers corresponding to each era. Quantitative analysis performed using EDXRF, facilitated by calibration curves, endows this method with high operability. The method is simple to operate and provides rapid analysis, facilitating a quick comparison between a large number of reference data and test samples. This has certain application prospects for tracing the provenance of Yixing Huanglongshan purple sand ore materials and identifying the age of their products.
		2025 Vol. 45 (09): 2550-2556 [Abstract] ( 28 ) RICH HTML PDF (9315 KB) ( 16 )

	2557	Study on Identification Methods of Pericarpium Citri Reticulatae Based on Time-Gated Raman Spectroscopy and Support Vector Machine
		FENG Hao-heng^{1, 2, 3, 4}, LI Ke-qing^{5, 6}, NIU Yuan-yuan^{2, 4}, SUN Qing-di^{2, 4}, XU Jin-chang^{2, 4}, FANG Guang-you^{1, 2, 3, 4}, CHEN Jian^{5, 6}, HU Min^{7, 8, 9}, WANG Zhen-you^{1, 2, 3, 4}
		DOI: 10.3964/j.issn.1000-0593(2025)09-2557-06
		The efficacy and price of Pericarpium Citri Reticulatae (PCR) are significantly influenced by its origin and harvest year, necessitating a rapidand effective identification method.Traditional chemical analysis techniques, though accurate, are complex, expensive, and time-consuming. Raman spectroscopy, with its high specificity and non-destructive detection capabilities, is a promising method for rapid detection. However, the strong fluorescence background of PCR limits the application of conventional Raman spectroscopy. To address this issue, this study combines time-gated Raman (TG-Raman) spectroscopy with support vector machine (SVM) classification models, proposing an efficient and non-destructive method for identifying PCR. The study selected six groups of PCR samples from different origins and ages in Xinhui District, Jiangmen City, Guangdong Province, as well as one crafted PCR sample, and compared the 532 nm, 1 064 nm continuous wave, and 532 nm TG-Raman spectra. The experimental results demonstrate that TG-Raman spectroscopy effectively eliminates fluorescence interference, thereby significantly enhancing the signal-to-noise ratio of the Raman signals and facilitating the extraction of more characteristic Raman peaks for chemical components.The key Raman peaks of PCR were observed at 856, 1 084, 1 112, 1 264, 1 300, 1 340, 1 456, 1 607, and 2 935 cm^-1. Spectral analysis revealed that the main chemical components of PCR include pectin, cellulose, fatty acids, and flavonoids. Among these, the flavonoid characteristic peak at 1 607 cm^-1 exhibited significant intensity variation, making it a key marker for distinguishing PCR from different origins and ages. Based on the extracted spectral features, the study constructed SVM classification models using various kernel functions, optimizing model parameters, and found that the radial basis function (RBF) kernel performed best. After training and testing on different PCR samples, the highest classification accuracy reached 96.43%, fully demonstrating the excellent classification performance of the combination of TG-Raman spectroscopy and SVM. The study indicates that this method is efficient and non-destructive, enabling accurate identification of PCR samples in a short time, with broad application potential in origin tracing and age identification. In conclusion, this study -presents a novel non-destructive detection technique for PCR and other medicinal materials, offering significant advantages in addressing the time-consuming and high-cost issues associated with traditional chemical analysis methods. This technology offers robust technical support for the quality control, authenticity verification, and traceability of medicinal materials, with wide-ranging application prospects.
		2025 Vol. 45 (09): 2557-2562 [Abstract] ( 23 ) RICH HTML PDF (5239 KB) ( 8 )

	2563	Study on Remote Analysis Method of Insulator Contamination Grades Based on Laser-Induced Breakdown Spectroscopy
		GUAN Zi-ran, HU Cong, SHI Qiao^*, WU Hui-feng, HE Wen-feng
		DOI: 10.3964/j.issn.1000-0593(2025)09-2563-06
		Insulators are critical components in transmission lines, playing a vital role in supporting and insulating, especially in ultra-high-voltage (UHV) transmission lines. However, the accumulation of industrial dust and other pollutants on the surface of insulators leads to a decline in insulating performance, which can trigger contamination flashover and cause significant damage to the transmission system. Therefore, monitoring the contamination level of transmission line insulators is a key factor in ensuring the safe and reliable operation of the power grid. Laser-induced breakdown spectroscopy (LIBS) is an in situ, rapid elemental analysis technique that offers the advantage of on-site, non-destructive analysis without requiring sample preparation. The remote sensing capability is one of the distinctive strengths of LIBS. In this study, artificial contamination was used as the target for analysis. A novel remote LIBS analyzer was used to conduct remote analysis of the elemental composition and contamination levels of glass insulator surfaces. A novel in-situ, rapid analytical method for determining the contamination level of insulators using remote LIBS was established. In the experimental process, under working conditions of a 2-meter testing distance, a laser energy output of 50 mJ, a laser frequency of 20 Hz, an integration time of 2.0 seconds, and a delay time of 2.0 μs, effective qualitative determination of elements such as Mg, Si, Al, Ca, and Na in artificial contamination was achieved. The quantitative analysis revealed a good linear relationship between the Na intensity in the contamination and the equivalent salt density. This indicates that the remote LIBS analyzer has a strong spectral response to Na in the contamination. Using the characteristic spectral lines of soluble salts, such as Mg and Na, obtained from the LIBS spectra, principal component analysis (PCA) and K-nearest neighbors (KNN) algorithms were employed to cluster and distinguish contamination levels effectively. The KNN classification model achieved an accuracy of 94.4%, a precision of 93.7%, and a recall of 96.4%, demonstrating its high effectiveness in identifying contamination levels. This study demonstrates that remote LIBS can achieve in-situ multi-element analysis of contamination on insulator surfaces. Combined with machine learning, it enables direct recognition of contamination levels. This provides methodological support for the further development of LIBS technology in power industry applications and the development of targeted analysis equipment for future in-situ applications.
		2025 Vol. 45 (09): 2563-2568 [Abstract] ( 25 ) RICH HTML PDF (7906 KB) ( 10 )

	2569	Estimation of Soil Organic Matter Content in Coal Mining Tensile Fracture Area Based on Spectral Index
		GUO Hui^{1, 2}, HAN Zi-wei^{1, 2*}, WU Dou-qing^{1, 2}
		DOI: 10.3964/j.issn.1000-0593(2025)09-2569-09
		The large-scale and high-intensity mining of coal leads to the formation of cracks in the ground, which destroy the structure of the soil and affect the soil quality. To quickly estimate the content of soil organic matter (SOM) in the coal stretching fissure area, soil samples were collected from the fissure area at Zhuzhuang coal mine, Huaibei City, China. The spectrum and SOM content of the soil samples were then determined. Inverse log reflectance (LR), first order differential reflectance (FD), and continuum removal(CR) were performed on the original spectrum. Then, the difference index, ratio index, and normalized difference index of any two band combinations were calculated. The Pearson correlation coefficient (PCC) was combined with the maximum relevance minimum redundancy (mRMR) algorithm to extract one-dimensional spectral bands and two-dimensional spectral indices, respectively. In the end, the two algorithms, PLSR and eXtreme Gradient Boosting (XGBoost), were used to construct a SOM content estimation model for the mining fissure area. The accuracy of the model was then tested and evaluated. The results show that: (1) The high intensity mining of coal leads to the formation of cracks in the ground, which destroys the structure of the soil and affects the soil quality. It also accelerates the loss of SOM and fine soil particles in the soil, and the coefficient of variation of SOM in the study area reaches 61.32%; (2)Regardless of one-dimensional spectral band or spectral index, the model based on FD spectrum has the highest accuracy; (2) The correlation between the two-dimensional spectral index and SOM content is significantly better than that of the one-dimensional spectral band, and the prediction model based on the spectral index has higher prediction accuracy; (3) Compared with the one-dimensional spectral band, the difference index (DI), ratio index (RI), and normalized difference index (NDI) have a stronger correlation with SOM, and the FD-DI index has the highest correlation, with a correlation coefficient of 0.88; (4) The accuracy of XGBoost based model is better than PLSR model, among which FD-NDI-XGBoost model has the highest accuracy, and its R², RMSE and RPD are 0.83, 0.49 mg·kg^-1 and 2.44, respectively. The experimental results can provide a technical reference for the hyperspectral estimation of SOM content in the coal mining tensile fracture area.
		2025 Vol. 45 (09): 2569-2577 [Abstract] ( 26 ) RICH HTML PDF (16109 KB) ( 10 )

	2578	Rapid Estimation of Cobalt Content in Lateritic Cobalt Ores: a Quantitative Inversion Study of VNIR-SWIR Spectra
		MEI Jia-cheng^{1, 2}, WANG Xue^{1, 2}, ZHANG Hong-rui^3, LIU Lei^{1, 2}
		DOI: 10.3964/j.issn.1000-0593(2025)09-2578-07
		Cobalt is a global strategic mineral resource. Laterite-type cobalt deposits are large and shallow, and are important targets for cobalt exploration. The traditional cobalt exploration process often relies on indoor testing and analysis to determine the level of cobalt content in field outcrops and the degree of its mineralization. The cobalt elemental content testing method often involves a complex sample preparation process. It relies on large-scale high-precision instrumentation, which makes it difficult to meet the demand for rapid testing and exploration. Visible-near infrared and short-wave infrared (VNIR-SWIR) spectroscopy offers the advantages of portability, high efficiency, and non-destructiveness to the samples. It demonstrates excellent applicability to field sample testing scenarios. The measured spectral analysis of the samples shows that 620~810 nm reflects the absorption of cobalt ions and iron ions, 810~1 200 nm reflects the absorption characteristics of Fe²⁺, 1 350～1 450 and 1 850～2 040 nm reflect the absorption characteristics of —OH and H₂O, and 2 140～2 260 and 2 260～2 360 nm reflect the absorption characteristics of Al—OH and Mg—OH, respectively; accordingly, the sensitive wavelength range and the characteristic absorption characteristics of the samples are selected. Accordingly, the range of sensitive bands, the characteristic absorption peak parameters and the ratio of sensitive bands were selected as the spectral combination parameters, and the XGBoost (Extreme Gradient Boosting) regression algorithm was applied to establish the cobalt content quantitative inversion model; based on which, parameter optimization was carried out to obtain the optimal cobalt content quantitative analysis model, and the validation set had the values of R² 0.95, the RMSE was 89.19, the RPD was 4.35, and the model inversion accuracy was high. The histogram of feature importance shows that the sensitive band of cobalt element ranges from 620~810 nm, and the accuracy of the model is significantly improved after increasing the weights of absorption features of minerals closely related to cobalt content (chlorite, serpentine). The above results-demonstrate that the cobalt content of lateritic cobalt ore samples can be accurately estimated based on VNIR-SWIR spectra. The model, which incorporates combined spectral parameters, has the capability of rapidly determining cobalt content in field outcrops, offering significant application value for lateritic cobalt ore exploration.
		2025 Vol. 45 (09): 2578-2584 [Abstract] ( 20 ) RICH HTML PDF (9325 KB) ( 6 )

	2585	Near-Infrared Modeling for Total Flavonoid and Protein Contents in Buckwheat Leaves Based on CARS Feature Extraction
		ZHU Li-wei, DU Qian-xi, TANG Guo-hong, LI Hong-you, ZHANG Xiao-na, CHEN Qing-fu, SHI Tao-xiong^*
		DOI: 10.3964/j.issn.1000-0593(2025)09-2585-05
		To meet the requirements of buckwheat quality determination and breeding work, the Competitive Adaptive Re-Weighted Sampling (CARS) algorithm was used in this study to extract the characteristic spectrum and combined with the quantitative partial least squares method to rapidly determine the total flavonoid and protein content in buckwheat leaves. First, the Kennard-Stone (KS) algorithm was used to split the training and test sets. The training set's average, maximum, and minimum total flavonoid contents were 55.8, 92.5 and 28.1 mg·g^-1, respectively. The test set's average, maximum, and minimum total flavonoid contents were 71.0, 99.8 and 31.5 mg·g^-1, respectively. The training set's average, maximum, and minimum protein contents were 169.6, 331.0 and 121.2 mg·g^-1, respectively. The samples' average, maximum, and minimum protein contents in the test set are 158.2, 183.0 and 129.1 mg·g^-1, respectively. Then use Normalization, Normalization + Multiplicative Scatter Correction (MSC), Normalization + Standard Normal Variate Transform (SNV), Normalization + First Derivative, Normalization + Second Order Derivative, Normalization + Savitzky-Golay Smoothing Filter (SG) to preprocess the spectrum in the wavelength range from 4 000 to 12 000 cm^-1, then use CARS algorithm to extract the characteristic spectrum, and finally use the partial least squares method to build prediction models. Through a comprehensive analysis of the coefficient of determination of the training model (R_c), the coefficient of determination of the test model (R_p), the root mean square error of cross-validation (RMSECV), the root mean square error of the test model (RMSEP) and the residual predictive deviation (RPD), we obtain the best model for the prediction of total flavonoid and protein in buckwheat. Three available prediction models for total flavonoidswere constructed. The best prediction model used 46 characteristic wavenumber points out of 1102 wavenumber points. The preprocessing method used was normalization + first derivative. The model's R_c, R_p, RMSECV, RMSEP, and RPD were 0.997, 0.933, 0.170, 0.829 and 2.893, respectively. Four available protein prediction models were created, the best of which used 42 characteristic wavenumber points, and the preprocessing method used was normalization + second derivative. The model'sR_c, R_p, RMSECV, RMSEP, and RPD are 0.998, 0.965, 0.202, 0.353 and 3.849, respectively. The results show that the application of the KS algorithm and CARS algorithm in building the near-infrared spectroscopy model requires fewer samples to build a reliable prediction model, enables the rapid determination of total flavonoids and protein of buckwheat leaves, and provides powerful tools for buckwheat breeding.
		2025 Vol. 45 (09): 2585-2589 [Abstract] ( 23 ) RICH HTML PDF (1502 KB) ( 14 )

	2590	Hyperspectral Estimation of Selenium Content in Selenium-Rich Tea Based on Feature Selection and Machine Learning
		WEN Zhu¹, GUO Song¹, SHU Tian¹, ZHAO Long-cai^{2, 3}
		DOI: 10.3964/j.issn.1000-0593(2025)09-2590-07
		Selenium is one of the important nutrient indices in selenium-rich tea, and its content determines the economic and nutritional value of selenium-rich tea. Hyperspectral remote sensing inversion technology has the characteristics of non-destructive, real-time, and rapid monitoring. This study utilizes the selenium content in selenium-rich tea from the Nangong River tea garden in Kaiyang County, Guizhou Province, and corresponding canopy non-imaging hyperspectral data as source data. The Savitzky-Golay second-order smoothing filter was used to preprocess the primary spectrum, and the potential of the primary spectral data was explored through first-order derivative transformation and continuum removal transformation. The independent variables for the modeling were obtained using a band elimination combination and various feature selection algorithms. Multiple inversion models of selenium content in tea were constructed using different algorithms. The results showed that: (1) the combination of spectral transformation and spectral index could enhance the ability of retrieving selenium content from the primary spectrum. (2) SPA was better than UVE overall; Continuum removal spectrum was superior to the primary spectrum and the first derivative spectrum. (3) The accuracy of the multi-factor model was better than that of the factor model, and the performance of ELMR in the multi-factor model was the best. Among all the models, the SPA-ELMR model under the continuum removal spectrum had the highest accuracy. The coefficient of determination (R²) and normalized root mean square error (nRMSE) of this model were 0.689 and 18.869%, respectively, and the corresponding verification R² and nRMSE were 0.627 and 20.429%, respectively. In this study, the response relationship between selenium content in tea and spectral reflectance at specific growth stages was discussed. A single-factor inversion model and a multi-factor inversion model with appropriate accuracy were constructed, providing a theoretical basis for the rapid and non-destructive monitoring of selenium content in tea. Also, they provided some technical support for the digital construction of tea gardens.
		2025 Vol. 45 (09): 2590-2596 [Abstract] ( 23 ) RICH HTML PDF (17187 KB) ( 17 )

	2597	Digital Non-Destructive Testing of Internal Color of Potatoes and Rapid Identification of Yellow and White Core Potatoes
		WANG Wei, NIE Sen^*, LI Yong-yu, PENG Yan-kun, MA Shao-jin, ZHANG Yue-xiang
		DOI: 10.3964/j.issn.1000-0593(2025)09-2597-09
		Color is an important quality indicator for agricultural products. The internal color of potatoes directly affects the sensory quality of their processed products. The rapid and non-destructive digital detection of the internal color of potatoes, as well as the quick classification of yellow-core and white-core potatoes, is of significant importance for advancing China's staple potato industry and the rapid rise of the prepared food industry. This study aims to achieve rapid, non-destructive digital characterization of the internal color of potatoes. It utilizes two self-developed, portable, multi-quality non-destructive detection devices based on a mini-spectrometer and a discrete spectral sensor: one is a laboratory-based, portable multi-quality non-destructive detection device for potatoes, and the other is a handheld, multi-quality non-destructive detection device for potatoes. A total of 209 potato samples from 26 different varieties, grown in various regions, were selected. Continuous spectral and discrete spectral data were collected, and the internal color parameters L^，a^ and b^* of the potato samples were measured using a colorimeter. First, based on the L^，a^ and b^* values measured by the colorimeter, an SVM classification model was established to distinguish between yellow-core and white-core potatoes. The threshold plane for distinguishing between yellow- and white-core potatoes was determined, which will serve as the basis for the next step in the non-destructive and rapid identification of yellow- and white-core potatoes. Secondly, based on the continuous spectral data collected by the portable continuous spectral device from the 209 potato samples, SNV preprocessing combined with the Random Frog Jump (RF) algorithm was used to select 200 characteristic wavelengths to establish a PLSR (Partial Least Squares Regression) prediction model for the L^，a^ and b^* parameters of potatoes. The root mean square errors (RMSE) for the validation set of L^，a^ and b^* were 1.278 8, 0.081 6, and 1.407 1, respectively. Similarly, for the discrete spectral data collected by the handheld spectral device, after SNV preprocessing, PLSR prediction models for the L^，a^ and b^* parameters of potatoes were also established. The RMSE for the validation set of L^，a^ and b^* were 1.278 8, 0.081 6, and 1.407 1, respectively. The results showed that the prediction models for the internal color parameters of potatoes established with both devices can meet the demand for rapid, non-destructive, digital detection of potato internal color in the field. Finally, 52 potato samples from 26 varieties, which were not involved in model training, were selected for external validation of the L^，a^ and b^* predicted values using both devices. For the portable continuous spectral device, the maximum absolute residuals for L^，a^ and b^* were 2.617 3, 0.141 3, and 2.779 1, respectively, and the mean residuals were 0.857 7, 0.049 0, and 0.697 2, respectively. For the handheld discrete spectral device, the maximum absolute residuals for L^，a^ and b^* were 3.262 8, 0.203 4, and 3.519 5, respectively, and the mean residuals were 1.093 0, 0.066 7, and 1.268 8, respectively. Based on the L^，a^ and b^* predicted values from both devices, rapid non-destructive identification of yellow and white-core potatoes was performed using the previously established classification threshold plane. The classification accuracy for yellow and white-core potatoes was 92.31% for the portable continuous spectral device and 86.54% for the handheld discrete spectral device. This technology enables the rapid, non-destructive, and real-time digital detection of potato internal color, as well as the quick classification of yellow- and white-core potatoes in the field. It provides technical support for the entire potato industry chain, including planting, processing, and sales.
		2025 Vol. 45 (09): 2597-2605 [Abstract] ( 23 ) RICH HTML PDF (11566 KB) ( 12 )

	2606	Role of Oxidant and Fluoride in Stone Coal-Type Vanadium Ore Leaching Process Based on SEM and Element Mapping Analysis
		LIU Ming-bao^{1, 2}, ZHANG Tian^{1, 2}, LI Feng^{1, 2}, LIU Heng^{1, 2}, YIN Wan-zhong³
		DOI: 10.3964/j.issn.1000-0593(2025)09-2606-08
		Vanadium-bearing mica in stone coal vanadium ore serves as one of the important vanadium source minerals. While the presence of oxidizing agents and fluorides significantly enhances vanadium leaching efficiency, the underlying mechanisms remain incompletely understood. This study focuses on the ShanYang stone coal vanadium deposit in Shangluo, Shaanxi Province. Using a thermostatic waterbath-sulfuric acid leaching technique，the effects of oxidizing agents and fluoride systems, including binary fluorides of GroupⅠ, Ⅱ, Ⅲ metals, hydrofluorides (K⁺/Na⁺/NH⁺₄), and hexafluoro-silicates/aluminates/phosphates, on vanadium leaching are systematically investigated. Through SEM and element mapping analysis, the primary pathways by which oxidants and fluorides promote vanadium leaching are clarified for the first time. Results indicate that the “pore corrosion” and “crevice corrosion” induced by the oxidant and fluorides are the main ways for efficient vanadium leaching. The effects of Group Ⅰ, Ⅱ metallic fluorides have been found to follow the order: CsF>KF>NaF>LiF, BaF₂>SrF₂>CaF₂>MgF₂，aligning with the lattice energies of these fluorides and the standard Gibbs free energy changes for HF production via reactions of these fluorides with leaching agent(sulfuric acid). In hexafluorosilicate/aluminate/phosphate systems, Na₂SiF₆ predominantly induces “pore corrosion”, while (NH₄)₂SiF₆ generates extensive “crevice corrosion”traces. Faster HF mass transfer in fissure corrosion explains their divergent leaching efficiencies at varying dosages. Na₃AlF₆ forms colloidal aggregates that hinder mass transfer, resulting in lower leaching efficiency compared to (NH4)₃AlF₆, as confirmed by Al mapping in the residues. PF^-₆ exhibits weaker dissociation than SiF^2-₆/AlF^3-₆ due to stronger P—F bonds (vs. Si—F/Al—F), making NH₄PF₆ less effective. Hydrofluorides (KHF₂>NH₄HF₂>NaHF₂) achieve superior leaching via “pore corrosion”, with KHF₂ achieving 96.0% vanadium recovery-the highest among all fluorides. SEM confirms the presence of abundant corrosion pores in KHF₂ residues. This work provides theoretical and technical foundations for optimizing vanadium extraction from stone coal.
		2025 Vol. 45 (09): 2606-2613 [Abstract] ( 24 ) RICH HTML PDF (23199 KB) ( 6 )

	2614	Design of a Lightweight and Compact Self-Aligned PG Imaging Spectrometer
		WANG Shi-qi, LIU Zi-rui, LI Qin-hao, ZHANG Xue-min^*
		DOI: 10.3964/j.issn.1000-0593(2025)09-2614-06
		Imaging spectrometers are capable of acquiring information in both the spatial and spectral dimensions of a target, and have been widely used in agriculture, industrial production, military and other fields. To comply with the development trend of unmanned aerial vehicle (UAV) lightweight development trend on the imaging spectrometer load miniaturization, lightweight requirements, and at the same time to ensure the high imaging quality, high spectral resolution of the performance needs, a new type of lightweight and small self-aligned PG imaging spectrometer optical system design is proposed. The thesis combines a Dyson-type imaging spectrometer and prism-grating-prism (PGP) features through an in-depth study of existing imaging spectrometer types. The introduction of the image-space telecentric lens at the same time to assume the function ofcollimating lens and imaging lens, greatly reducing the size and quality of the optical system, effectively reducing the cost of processing, assembly and adjustment; the use of prisms and plane reflective grating combination constitutes the PG dispersion module, instead of concave grating, and at the same time retains the ability to correct the spectral bending of the PGP. The designed imaging spectrometer features a slit length of 10 mm, a working wavelength range of 450 to 900 nm, a relative aperture of 1∶3, and a full spectral resolution of better than 0.8 nm when the slit width is 10 μm. The envelope volume of the optical system is only 123 mm×44 mm×44 mm, and the MTF of the full field of view exceeds 0.5 at 100 lp·mm^-1 , approaching the diffraction limit. The RMS of each wavelength point array is smaller than that of the Airy spot indicating good imaging quality of the system. Sodium dual line separation can be effectively achieved under laboratory conditions.
		2025 Vol. 45 (09): 2614-2619 [Abstract] ( 21 ) RICH HTML PDF (13782 KB) ( 20 )

	2620	Influence of Powder Particle Size on X-Ray Photoelectron Spectroscopy
		ZHU Xiao-dong, WANG Jie-ru
		DOI: 10.3964/j.issn.1000-0593(2025)09-2620-05
		X-ray Photoelectron Spectroscopy (XPS) is a powerful surface analysis technique that can analyze all elements except hydrogen and helium. It can analyze the surface chemical states of almost all solid and ionic liquid materials, achieving qualitative or semi-quantitative analysis of the material surface. Combined with ion etching or angle-resolved analysis, it can also -facilitate in-depth analysis and interface analysis of materials, as well as the analysis of electrical properties, such as the valence band. It is widely applied in semiconductor materials, polymer materials, catalytic materials, metallurgy, corrosion, and other fields. Solid powder material is the most common test object in XPS measurements, and its common preparation method is to sprinkle or press the powder into the tape (tape method). Generally speaking, under the same test conditions, the XPS signal of coarse solid samples is weaker, while the influence mechanism of powder particle size on XPS signals remains unclear. This study focuses on the interaction mechanism between particle size and substrate effects on XPS signals for powder samples prepared by the adhesive method, systematically investigating the impact of particle size distribution on signal intensity, peak characteristics, and elemental quantitative analysis. Using alumina (Al₂O₃) powder as a model system, XPS analysis [wide scan and high-resolution scans of Al(2p), Si(2p), C(1s), and O(1s)] on gradient-sized samples (4.5～500 μm)，SEM morphological characterization, and the variation of the contents of elements have been studied. The results reveal a non-monotonic relationship between particle size and testing signals. With the increase of Al₂O₃ particle size (greater than 150 μm), the peak intensity of Al(2p) gradually decreases, and the half-height width gradually increases. This correlates with increased surface roughness, which affects XPS signals. When the particle size of Al₂O₃ powder is less than 13 μm, the peak intensity of Al(2p) shows a decreasing trend, while the Si(2p) signal from the substrate tape increases. It demonstrates unavoidable substrate interference. This study provides a scientific basis and theoretical guidance for optimizing XPS testing methods of solid powder samples.
		2025 Vol. 45 (09): 2620-2624 [Abstract] ( 20 ) RICH HTML PDF (7944 KB) ( 8 )

	2625	Hyperspectral Image Visualization and Recognition of Bloodstains and Their Analogues Based on Attention Mechanism
		HU Wei-cheng, LI Yun-peng^*, WANG Hong-wei, DAI Xue-jing, WANG Hua-peng
		DOI: 10.3964/j.issn.1000-0593(2025)09-2625-07
		Bloodstains are key litigation evidence for exposing and confirming crimes. Still, due to the complexity of the crime scene environment and the diversity of objects, it is challenging to quickly and accurately detect bloodstains and identify their analogues during scene investigation. Given this, this paper proposes Hybrid-PSA, a hyperspectral visual classification model for bloodstain identification based on the attention mechanism, which enables the visual classification and identification of bloodstains and their analogues, such as blood, ketchup, artificial blood, and acrylic paint. The Hybrid-PSA model was designed hierarchically, with a core containing a set of three-level 3D convolutional layers, a polarized self-attention module (PSA), and a 2D convolutional layer. The PSA module is embedded through residual linkage, which employs a dual-branching structure to achieve feature optimization; the channel branch retains 1/2 spectral band and compresses the spatial dimensions to 1×1 to focus on correlation features between spectral bands; the spatial branch maintains the original resolution and compresses the number of channels to 1 to model spatial features. This dual-polarization mechanism strikes a balance between spectral integrity and spatial information modeling, enabling the model to accurately focus on key feature regions while simultaneously improving its feature capture capability and computational efficiency with only a small increase in parameters. To validate the performance of the Hybrid-PSA model, ablation experiments are performed on the 3D convolution module and the attention module using a limited number of training samples, experimentally comparing them with 3DCNN and Hybrid-SN on the publicly available bloodstain hyperspectral dataset, Hyper Blood. The experimental results show that Hybrid-PSA improves the model accuracy from 96% to 99.08% with only a 0.02% increase in the number of parameters, resulting in a 3.08% improvement in accuracy. In terms of visual recognition, 3D-CNN and Hybrid-SN-exhibit high misidentification rates and struggle to classify blood and blood-like traces accurately on red and black backgrounds. The null-spectrum fusion strategy and residual linking of Hybrid-PSA enable the polarized self-attention mechanism to better adapt to the trace target shape and accurately capture the trace boundaries in each iteration, thereby avoiding misclassification of traces at the boundaries due to similarity in color. Avoid misclassification at the boundary due to color similarity, and the visualization classification effect is significantly better than the other two models. The bloodstain classification model based on the attention mechanism proposed in this paper is characterized by high classification accuracy, an excellent visualization effect, and strong generalization ability, and is able to quickly, accurately, and non-destructively discover and identify bloodstains in complex investigation environments.
		2025 Vol. 45 (09): 2625-2631 [Abstract] ( 22 ) RICH HTML PDF (10521 KB) ( 7 )

	2632	Cross-Modal Dual-Channel Camouflaged Object Detection Method for Visible-Spectrum Image
		CHENG Yu-hu, WU Shi-jia, WANG Hao-yu, WANG Xue-song^*
		DOI: 10.3964/j.issn.1000-0593(2025)09-2632-10
		The camouflaged object detection (COD) task for visible-spectrum images aims to utilize visible-spectrum information to detect camouflaged objects that are visually consistent with their surrounding environment. This visual consistency poses challenges such as difficulty in distinguishing object boundaries and learning discriminative features, which limit the effectiveness of existing object detection methods for COD. A Cross-modal Dynamic Collaborative Dual-channel Network (CDCDN) is proposed to explore the potential of global-local multi-level visual perception and visual-language models in COD. First, to address the challenge of distinguishing object boundaries, a dynamic, collaborative, dual-channel module is designed. Through the dual channels, the detection process is decoupled into global information localizationand local feature refinement, enabling object detection and optimization from a multi-level visual perspective. A dynamic information collaboration and fusion mechanism is established, through which global and local information are mutually complemented and corrected by global gating constraints and local perception correction. The spatial capture capability of the model is enhanced in scenarios with blurred object boundaries. To address the difficulty in learning discriminative features, a cross-modal scene-object matching module is designed. By incorporating a pre-trained VLM, this module establishes cross-modal interactions between the visual and language modalities, thereby enhancing the distinction between objects and backgrounds in the feature space and improving the model's semantic discrimination in scenes with limited discriminative features. CDCDN is evaluated on the MHCD2022 and COD10K datasets using the mAP@0.5, mAP@0.5∶0.95, and mAP@0.75 metrics. CDCDN achieves scores of 67.6%, 42.6%, 48.4% on the MHCD2022 dataset, and 67.9%, 40.6%, 41.0% on the COD10K dataset, respectively. Compared to five mainstream object detection methods, including Faster R-CNN, DETR, Lite-DETR, YOLOv5, and YOLOv10, CDCDN achieves the best detection accuracy across all three metrics.Visualization of detection results in four common camouflaged scenes -barren land, grassland, woodland, and snowfield -demonstrates the adaptability of CDCDN to various scenes. In an ablation study, the key components of CDCDN are incrementally removed to systematically evaluate their contributions, with results showing that each component significantly enhances the model's detection performance. Comprehensive experimental results indicate that CDCDN can accurately detect camouflaged objects with high visual consistency to their surroundings, providing a novel solution for COD.
		2025 Vol. 45 (09): 2632-2641 [Abstract] ( 19 ) RICH HTML PDF (30305 KB) ( 2 )

	2642	Broadband Dielectric Characterization of Fe₂O₃ and Fe₃O₄ Using Terahertz Time Domain Spectroscopy
		ZHAI Min, XIAO Bin, PAN Hao-yue, HE Wen-long
		DOI: 10.3964/j.issn.1000-0593(2025)09-2642-06
		Surface quality is an important indicator for evaluating the quality of hot-rolled coil steel, which directly affects the service life and performance stability of hot-rolled strip steel products. Dense mill scale defects form on the surface of the steel substrate during the production process of hot-rolled strip steel, and their composition is mainly composed of iron oxides. To ensure the quality of subsequent coating and plating processes, understanding the thickness information of mill scale on the surface of hot-rolled strip steel can effectively reduce the risk of insufficient or excessive pickling. Because terahertz waves are located between infrared and microwaves in the electromagnetic spectrum, and reflect with minimal attenuation on the surface of polar materials such as metals, terahertz time-of-flight tomography (TOFT) meets the technical requirements of non-destructive characterization of micron-level covering layers of steel-based materials. Owing to the difficulty of non-destructive mechanical stripping, the optical properties of the mill scale on the surface of hot-rolled steel strip in the terahertz frequency band are preliminarily estimated based on existing values in the published literature, rather than values obtained through experiments, as a result of which, unexpected errors between the scale thickness calculated based on terahertz results and the nominal value. To accurately determine the thickness distribution of iron scale, hematite (Fe₂O₃) and magnetite (Fe₃O₄) powders were thoroughly mixed with polyethylene (PE) and pressed into pellets. Transmission experiments were conducted on the Fe₂O₃/PE and Fe₃O₄/PE pellets using terahertz time-domain spectroscopy (THz-TDS), and the optical parameters of Fe₂O₃ and Fe₃O₄ inclusions at 10% mass fraction in the frequency range [350 GHz~2.5 THz] were then accurately calculated using the Maxwell-Garnett effective medium theory and Vegard's law. The refractive indexn, absorption coefficientα, and conductivity σ of Fe₂O₃ and Fe₃O₄ at 1 THz are 3.96 and 5.18, 10.13 and 25.58 cm^-1, and 1.2 and 4.97 S·m^-1, respectively. The method outlined here will also be of interest to a range of powder materials that may need to evaluate their optical properties in the THz band. The research results have laid a theoretical foundation for the future implementation of terahertz-based technology in online monitoring of the quality of hot-rolled steel products within the complex steel production environment, providing a non-contact and non-destructive method. This has important engineering practical significance for promoting the widespread application of terahertz technology.
		2025 Vol. 45 (09): 2642-2647 [Abstract] ( 20 ) RICH HTML PDF (3868 KB) ( 8 )

	2648	A Diagnostic Method for Adzuki Bean Rust Based on an Improved E-DWT Algorithm and Deep Learning Model
		FU Qiang¹, GUAN Hai-ou^1*, LI Jia-qi²
		DOI: 10.3964/j.issn.1000-0593(2025)09-2648-10
		Adzuki bean rust is a common fungal disease that significantly reduces crop yield by infecting leaves and impairing photosynthesis. This paper proposes a novel diagnostic method for adzuki bean rust based on an improved Empirical Mode Decomposition-Wavelet Transform (E-DWT) algorithm and a deep learning model. Using “Baoqinghong” adzuki beans as the experimental material, 960 leaf spectral datasets were collected over 10 days using a handheld visible/near-infrared spectrometer, obtaining reflectance data in the wavelength range of 326~1 075 nm. First, the improved E-DWT algorithm was applied for spectral denoising. This algorithm combines Empirical Mode Decomposition (EMD) and wavelet threshold denoising technology to retain effective signal information while removing noise. The optimal wavelet basis function (sym5) and the number of decomposition layers (4 layers) were determined by comparing the RMSE and SNR indicators. To further reduce redundancy in high-dimensional data, the Successive Projections Algorithm (SPA) was employed to select 12 representative wavelengths from the initial 750 features, resulting in a 98.4% reduction in feature wavelength count. Next, the Gramian Angular Field (GAF) method was employed to convert the one-dimensional wavelength sequence into a two-dimensional spectral image, thereby enhancing correlations between different bands for subsequent model training. The designed deep learning model combines a Convolutional Neural Network (CNN) with a Convolutional Block Attention Module (CBAM). The CBAM module effectively discriminates the weights of different feature wavelengths and time nodes through channel and spatial attention mechanisms, enabling the model to focus on key features influencing adzuki bean rust identification. Experimental results show that the CBAM-CNN model achieved 99.31% accuracy in the training set, 98.33% accuracy in the test set, and 98.89% recall, significantly outperforming traditional CNN models. Compared to existing methods, the proposed model exhibits superior performance in terms of recognition accuracy, stability, and training convergence speed. Additionally, the model structure is more concise, which optimizes parameter adjustment and improves operability in practical applications. In conclusion, the proposed diagnostic method based on the improved E-DWT algorithm and CBAM-CNN model not only achieves efficient and precise disease detection but also provides a theoretical foundation and technical support for constructing data-driven crop disease diagnosis systems in the future.
		2025 Vol. 45 (09): 2648-2657 [Abstract] ( 26 ) RICH HTML PDF (16174 KB) ( 6 )

	2658	Exploration of Spectral Variation and Element Identification Under Environmental Geological Data Simulation of Heavy Metal Pollution
		HU Lin-zhen^{1, 3, 6}, XIA Tian⁴, ZHANG Chao^{1, 2, 3*}, YANG Ke-ming⁵, GAO Xue-zheng^{1, 3}, LI Xiao-lei^{1, 3}, WAN Ming-ming⁷
		DOI: 10.3964/j.issn.1000-0593(2025)09-2658-08
		In the interdisciplinary field of modern agriculture and environmental science, the study of changes in the spectral characteristics of crops contaminated with heavy metals is gradually becoming a hot topic. When crops are contaminated with heavy metals, their internal physiological structure and biochemical composition change, which is directly reflected in their spectral characteristics. The variation information generated by spectral changes becomes a crucial basis for monitoring heavy metal pollution. This study conducted pot experiments on maize plants contaminated with different concentrations of heavy metals, specifically copper and lead, in the laboratory. It measured the reflectance spectra of maize leaves under various concentration gradients of copper and lead pollution, as well as key data such as the copper and lead content in maize leaves. A comprehensive, detailed, and specialized dataset was constructed for maize plants contaminated with heavy metals copper and lead. And focusing on the spectrum of maize leaves- from a unique perspective in the frequency domain, we will conduct an in-depth exploration of its Full spectral range and sub-spectral range. By innovatively combining time-frequency analysis methods, a method called leaf-sensitive Spectral Interval Detection Method (SIDM) was proposed. Based on SIDM, spectral Variation Characteristic Parameters (SVCP) for leaf spectra were further proposed, which serve as “biomarkers” for crop contamination status and are of great significance for studying the intrinsic correlation between variation characteristic parameters and leaf heavy metal content. Meanwhile, compare it with conventional spectral indices to explore the spectral range sensitive to copper and lead pollution. On this basis, a leaf Spectral Transformation Method (STM) was ingeniously constructed by combining a nonlinear time-frequency distribution. Through experimental verification, STM can clearly distinguish different types of copper and lead pollution. SIDM has successfully enhanced and accurately extracted weak information on copper and lead pollution in leaves, making the originally weak and difficult-to-detect pollution signals visible. More importantly, a highly specific spectral range for copper and lead pollution has been identified, laying a solid foundation for the development of more accurate and efficient heavy metal pollution monitoring technologies in the future. STM has advantages in distinguishing spectral differences between samples with and without heavy metal pollution, and can intuitively categorize the element types of maize contaminated with copper and lead, effectively promoting the development of spectral technology for monitoring heavy metal pollution in crops.
		2025 Vol. 45 (09): 2658-2665 [Abstract] ( 24 ) RICH HTML PDF (16990 KB) ( 3 )

	2666	A Multi-Layer Attention Convolutional Neural Network Model for Fine Classification of Hyperspectral Images in Rare Earth Mining Areas
		FAN Xiao-yong¹, LI Heng-kai^1*, LIU Kun-ming², WANG Xiu-li³, YU Yang¹, LI Xiao-yu¹
		DOI: 10.3964/j.issn.1000-0593(2025)09-2666-10
		Ion-adsorption-type rare earth minerals are important strategic resources. Long-term extensive mining has led to severe surface damage in mining areas, posing significant challenges to the ecological environment. Accurate and detailed land use information is a critical foundation for ecological restoration and process monitoring in mining areas. Hyperspectral imagery is considered an effective means for large-scale monitoring of mining areas to obtain land use information. However, the complexity of the land cover and the information redundancy in hyperspectral images pose challenges for fine classification. This study proposes a fine classification method for rare earth mining areas based on object-oriented thinking and a multi-layer attention convolutional neural network (OCTC). First, a scale parameter estimation model was used to quantitatively analyze the optimal segmentation scale at multiple levels of the rare earth mining area images. Four types of image features—spectral, index, texture, and geometric—were extracted from the segmented images. Then, an optimal feature combination was obtained through distance separability analysis. Based on this, a multi-layer attention convolutional neural network model (OCTC) was used for classification. This model is an improved version of the 1D-CNN, integrating the Transformer and CBAM to enhance the model's feature extraction capabilities and overall classification accuracy. To verify the method's effectiveness, Zhuhai-1 hyperspectral remote sensing imagery was used as the data source, and the Jiangxi Gan'nan Lingbei rare earth mining area served as the study region. The proposed method was compared with KNN, RF, and 1D-CNN classification methods for accuracy analysis. The results demonstrate that the proposed method effectively mitigates salt-and-pepper noise, maintains good overall classification integrity, and achieves the highest classification accuracy. The overall accuracy reached 88.11%, representing an improvement of 1.22% to 8.84% compared to other classification methods, with the Kappa coefficient increasing by 0.015 9 to 0.109 0. This method can provide valuable reference and scientific insights for the fine classification of land use and production monitoring, as well as environmental protection management in rare earth mining areas.
		2025 Vol. 45 (09): 2666-2675 [Abstract] ( 24 ) RICH HTML PDF (23066 KB) ( 11 )

	2676	Research on Mural Line Drawing Enhancement Method Combining Wavelet Transform and Hyperspectral Imagery
		DUAN Lu-nan^{1, 2}, ZHANG Ai-wu^{1, 2*}, CHEN Yun-sheng^{1, 2}, GAO Feng³, GUO Ju-wen³
		DOI: 10.3964/j.issn.1000-0593(2025)09-2676-08
		Ancient murals have often suffered from blurring and loss of line structures over time, making their interpretation challenging. Hyperspectral imaging technology, capable of capturing subtle variations in material and energy, provides valuable information for enhancing these faint or missing details. Therefore, this paper proposes a method for enhancing mural line information that combines the wavelet transform with hyperspectral imaging data. Firstly, the dimensionality of the mural hyperspectral image was reduced by using the Minimum Noise Fraction (MNF) transform. The optimal MNF band image was selected using the maximum average gradient method. The MNF results were used to extract the pure end members, and the corresponding abundance maps were inverted by fully constrained least squares spectral unmixing. The line abundance map was combined with the optimal MNF band image through band operations to obtain a linear feature-enhanced image. Then, the true color image was synthesized using the inverse MNF transformation. Both the linear feature-enhanced image and the true color image were processed with Gaussian filtering to enhance detail. Haar wavelet decomposition was applied to both images. The corresponding high-frequency components were fused, while the low-frequency component from the true color image was retained to reconstruct the final color image with enhanced line features. The experimental validation on murals from Yiju Temple (Shanxi) shows that, in comparison with a PCA-based enhancement technique, the proposed approach achieves an increase of 0.083 7 in average gradient and 15.253 1 in edge intensity, indicating more effective enhancement of line features and offering valuable insights for mural preservation and restoration efforts.
		2025 Vol. 45 (09): 2676-2683 [Abstract] ( 27 ) RICH HTML PDF (69039 KB) ( 13 )

	2684	Preparation and Luminescence Performance of Li₂Mg₃TiO₆:Sm³⁺ Orange-Red Phosphor
		LI Peng-cheng¹, LI Zhao^2, WANG Wei-gang², ZHOU Jun^3
		DOI: 10.3964/j.issn.1000-0593(2025)09-2684-09
		A series of Li₂Mg₃TiO₆:xSm³⁺ (0.005≤x≤0.010) orange-red phosphors was prepared using the high-temperature solid-state method. The luminescent transition mechanism of the phosphor was analyzed, and the packaging of LED devices was conducted. The results indicate that the Li₂Mg₃TiO₆:Sm³⁺ prepared by the high-temperature solid-state method has a pure phase, uniform particle distribution, and an average particle size of 3 μm. The main excitation peak of Li₂Mg₃TiO₆:Sm³⁺ is 344 nm, and the main emission peak is 677 nm, with an optimal doping concentration of 5%. At 350 K, the relative luminescent intensity of the sample is 63.5%, and the thermal activation energy ΔE is 0.363 eV. The color coordinates of Li₂Mg₃TiO₆:0.05Sm³⁺ calculated by CIE are located in the red-light region (0.630 8, 0.358 4). The packaged LED device emits red light with good color rendering Li₂Mg₃TiO₆:Sm³⁺ red phosphor is expected to be used in white light LEDs.
		2025 Vol. 45 (09): 2684-2692 [Abstract] ( 23 ) RICH HTML PDF (27995 KB) ( 8 )

	2693	Methods to Improve Ultraviolet Aging Resistance for Glass Fiber Reinforced Polypropylene Composite Materials
		HAN Yu¹, XIAO Han^2*, ZHANG Zhi-cheng², WU Fu-mei², DONG Yang²
		DOI: 10.3964/j.issn.1000-0593(2025)09-2693-08
		This study investigated the mechanical, thermal, and morphological properties of glass fiber-reinforced polypropylene (GFPP) composite plates before and after ultraviolet (UV) aging, to evaluate the effects of UV-resistant additives and protective films on enhancing UV durability. The findings provide technical support for the application of composite materials in lightweight freight vehicle bodies. The results demonstrated that the tensile strength of unmodified samples decreased by 19% after 2 000 hours of UV exposure. In contrast, samples treated with anti-UV additives and coated with UV-resistant films exhibited only a 4% reduction in mechanical performance. Fourier transform infrared spectroscopy (FTIR) was employed to quantitatively analyze the evolution of carbonyl groups (C═O), revealing that synergistic UV protection effectively suppressed molecular chain scission. Specifically, unprotected samples showed a significant increase in the carbonyl absorption peak at 1 740 cm^-1 (a characteristic marker of polypropylene degradation), while protected samples exhibited minimal changes. Scanning electron microscopy (SEM) further confirmed severe surface pulverization and cracking in unprotected samples, whereas protected samples maintained structural integrity.
		2025 Vol. 45 (09): 2693-2700 [Abstract] ( 17 ) RICH HTML PDF (16955 KB) ( 5 )

　CopyRight © OFFICE OF SPECTROSCOPY AND SPECTRAL ANALYSIS
　Address: NO.76 Collage South Road of Beijing 　Postalcode:100081
　Tel. (010)62181070　E-mail:chngpxygpfx@vip.sina.com
　Support by Beijing Magtech Co.ltd　E-mail:support@magtech.com.cn