化学计量学在毛细管电泳中的应用

综述了最优化方法，多元校正方法、人工神经网络、曲线分辨、信号处理等化学计量学方法在毛细管电泳分离分析中的应用，展望了化 社量学在毛细管电泳中的应用前景，引用参考文献55篇。     

谱峰漂移校正技术结合三维平行因子分析方法用于分辨中药重叠高效液相色谱信号

谱峰漂移校正技术结合平行因子分析方法被应用于中药色谱重叠峰信号分辨. 谱峰漂移校正技术的使用, 使得数据的三线性得以加强, 更适合于平行因子分析方法的分辨. 本文通过对中药延胡索色谱数据分辨, 并与经典多元曲线分辨方法对比来验证谱峰漂移校正技术结合平行因子分析方法的正确性.     

多元非线性荧光校正的人工神经网络方法

由实验测定得知罗丹明Ｂ，丁基罗丹明Ｂ，曙红Ｂ组成的三组分混和荧光分析体系存在严重的荧光熄灭现象，其混和物的荧光光谱呈非线性，ＰＬＳ难以校正。本文成功地将ＢＰ－ＡＮＮ应用于此多元非线性荧光校正问题，完成了三组分的同时测定，得到了满意的结果。     

分析化学计量学的若干新进展

对分析化学计量学的若干新进展进行了讨论, 侧重从作者实验室所进行的研究工作的视角论述了高维数据的分辨、用于化学信号处理的形态学理论和方法、解决复杂化学问题的多元校正和化学模式识别以及联用色谱仪二维数据的分辨新方法的发展.     

基于遗传算法优化的约束背景双线性分解算法用于改进高效液相色谱灰色分析体系的校正结果

该文采用约束背景双线性分解算法(CBBL)对以高效液相色谱(HPLC)方法分离分析的灰色分析体系进行了多元校正研究。针对采用包括CBBL在内的矩阵校正方法处理HPLC灰色分析体系的固有缺陷,即在相关组分的色谱保留时间重现性较低的情形下多元校正的结果不理想,对CBBL方法进行了改进,即将待测组分的浓度与组分的色谱保留时间同时作为优化的参量引入CBBL,并采用遗传算法(GA)优化CBBL,对于模拟的组分保留时间飘移严重的HPLC灰色分析体系及保留时间重现性不佳的多种酚类化合物组成的实际HPLC灰色分析体系进行了多元校正分析,成功克服了经典CBBL的固有缺陷,取得了较理想的多元校正结果。另外,该研究所建议的方法的校正结果也显著优于传统的残差双线性分解法(RBL)以及秩消失因子分析法(RAFA)。     

维生素B的同时测定─—神经网络与紫外光谱法

研究了反向传播神经网络（ＢＰＮＮ）及改进算法，经比较发现ＢＰＮＮ在某些方面优于滤波分辨及多元校正。将其应用于多元分析及光谱定量，可不经分离同时测定Ｂ族维生素，操作简便，结果良好。从而开发了一种多组分同时光谱测定新方法。     

维生素B的同时测定——神经网络与紫外光谱法

研究了反向传播神经网络（ＢＰＮＮ）及改进算法，经比较发现ＢＰＮＮ在某些方面优于滤波分辨及多元校正，将其应用于多元分析及光谱定量，可不经分离同时测定Ｂ族维生素，操作简便，结果良好，从而开发一种多组分同时光谱测定新方法。     

基于量子波包方法的态-态分辨反应散射动力学计算

本文回顾了最近十几年利用量子波包方法研究气相分子反应散射动力学的工作进展,特别是在态-态分辨水平上的工作进展。比较详细地讨论了目前存在的利用量子波包方法计算态-态微分截面的几种方法。目前态-态分辨的波包动力学计算可以精确地预测三原子和四原子分子反应散射的各种信息,文章最后对几个典型的利用波包方法在态-态分辨水平上研究过的三原子和四原子反应散射体系做了讨论。     

化学计量学在我国的发展

评述了化学计量学在我国近年来的发展,包括多元校正与分辨、稳健方法、化学模式识别、化学定量构效关系、分子模拟与优化、化学专家系统与库主要分支,其中特别对近年来得到飞速发展的新技术如高维数据解析方法,人工神经网络、小波变换给出了较详细的介绍。     

混合线性分析法的原理及应用

对一种新近发展的多元校正方法－混合线性分析法（hybrid linear analysis,HLA)的原理进行了深入探讨和阐述，用该法进行了模拟体系和复方头孢氨苄制剂的分析，在主因子数确定，单位谱的获得，波长选择、适用范围等方面进行了讨论。结果表明，HLA法可给出优于偏量小二乘法（PLS）的结果，方法易行，结果准确。在复杂体系分析中具有较好的应用前景。     

Estimation of rate constants from growth and decay data

Many experiments in chemical kinetics are initiated by a fast pulse, such as electric discharge, shock wave, flash lamp, or laser. After this pulse one observes the production and subsequent decay of a reactive intermediate. One then postulates a mechanism and adjusts the associated rate constants so as to minimize the difference between the results of the experiment and the prediction of the mechanism. The parameters to be estimated are usually strongly correlated, so that it is not possible to determine them separately. These estimated parameters are of little value unless we can also estimate statistically valid confidence limits for them. The difficulties are discussed which frequently arise in estimating parameters and confidence limits for a kinetic mechanism which is widely used in interpreting laser excitation and fluorescence measurements, that is, first-order production and decay. These difficulties, and methods for dealing with them, are illustrated with realistic data. The estimation problem is particularly ill conditioned when the production and loss rates are nearly equal. In some experimental systems this can be avoided, but in others it is inevitable.     

Multivariate optimization and simultaneous determination of hydride and non-hydride-forming elements in samples of a wide pH range using dual-mode sample introduction with plasma techniques: application on leachates from cement mortar material

Analytical methods have been developed for the simultaneous determination of hydride-forming (As, Sb) and non-hydride-forming (Cr, Mo, V) elements in aqueous samples of a wide pH range (pH 3–13). The methods used dual-mode (DM) sample introduction with ICP-AES and ICP-MS instruments. The effect of selected experimental variables, i.e., sample pH and concentrations of HNO₃, thiourea, and NaBH₄, were studied in a multivariate way using face-centered central composite design (FC-CCD). Compromised optimum values of the experimental parameters were identified using a response optimizer. The statistically found optimum values were verified experimentally. The methods provided improved sensitivities for the hydride-forming elements compared with the respective conventional nebulization (Neb) systems by factors of 67 (As) and 64 (Sb) for ICP-AES and 36 (As) and 54 (Sb) for ICP-MS. Slight sensitivity improvements were also observed for the non-hydride-forming elements. The limits of detection (LOD) of As and Sb were lowered, respectively, to 0.8 and 0.9 μg L⁻¹ with the DM-ICP-AES system and to 0.01 and 0.02 μg L⁻¹ with the DM-ICP-MS system. The short-term stabilities of both methods were between 2.1 and 5.4%. The methods were applied for the analysis of leachates of a cement mortar material prepared in the pH range 3–13. The elemental concentration of the leachates determined by the two DM methods were statistically compared with the values obtained from Neb-ICP-MS analysis; the values showed good agreement at the 95% confidence level. Quantitative spike recoveries were obtained for the analytes from most of the leachates using both DM methods. Figure Schematic of the dual-mode sample introduction system used in combination with ICP-AES and ICP-MS for the simultaneous determination of hydride and non-hydride-forming elements     

Characterization and discrimination of pollen by Raman microscopy

The chemical characterization and discrimination of allergy-relevant pollen (common ragweed (Ambrosia artemisiifolia), white birch (Betula pendula), English oak (Quercus robur), and European linden (Tilia cordata)) has been studied by Raman microscopy. Spectra were obtained at different excitation wavelengths (514, 633, and 780 nm) and various methods were examined to minimize the strong fluorescence background. The use of a He–Ne laser (633 nm) for excitation yields high-quality single pollen Raman spectra, which contain multiple bands due to pollen components such as carotenoids, proteins, nucleic acids, carbohydrates, and lipids. Multivariate classification, i.e. principal component analysis (PCA) and hierarchical cluster analysis, demonstrated the validity of the approach for discrimination between different pollen species.     

Simultaneous flow injection analysis of cadmium and lead with differential pulse voltammetric detection

A flow injection (FI) method using multiple differential pulse voltammetric detection for the simultaneous determination of two metal ions was developed and applied to the resolution of Cd(II)-Pb(II) mixtures. The metals are detected by applying two sequential pulses to a three-electrode voltammetric system that uses a flow-through cell accommodating a static mercury-drop working electrode. The influence of the electrode area, flow-rate, pulse frequency, pulse width and sampling time was investigated. Under the experimental conditions used, the two metals were found to interfere with each other. The use of a neural network allows the simultaneous determination of both, in mixtures, with good accuracy. The proposed method is applicable to other complex systems involving different working electrodes and more than two electroactive species.     

Dissimilar chromatographic systems to indicate and identify antioxidants from Mallotus species

The genus of Mallotus contains several species commonly used as traditional medicines in oriental countries. A data set containing 39 Mallotus samples, differing in species, cultivation conditions, harvest season and/or part of the plant was used to develop fingerprints on two dissimilar chromatographic systems. An exploratory analysis with principal component analysis (PCA) was performed on both data sets individually. The results were also combined to obtain additional information on the unknown samples included in the data set. Furthermore, the antioxidant activity of the samples was measured and modelled as a function of the fingerprints using the orthogonal projections to latent structures (O-PLS) technique. The regression coefficients of the models were studied to indicate the peaks potentially responsible for the antioxidant activity. The indicated peaks were analyzed and identified by HPLC coupled to mass spectrometry (HPLC-MS). Because of the complexity of biological samples, it was aspired to separate co-eluting components based on the significant difference in chromatographic selectivity on the dissimilar systems and consequently obtain additional, complementary information on the contribution of the individual components to the antioxidant activity. The results illustrate the potential use of dissimilar chromatographic systems. Several initially co-eluting compounds could be separated on the dissimilar system. The corresponding regression coefficients provided complementary information on the potential antioxidant activity of the separated compounds.     

On the study of nonlinear dynamics of complex chemical reaction systems

When driven far from equilibrium,nonlinear chemical reactions often show a variety of self-organization behavior,including chemical oscillations,waves,chaos and patterns[1].Recently,the study of such nonlinear phenomena in‘complex’systems,such as the li…     

Droplet‐based microfluidics systems in biomedical applications

Microfluidics has made a very impressive progress in the past decades due to its unique and instinctive advantages. Droplet‐based microfluidic systems show excellent compatibility with many chemical and biological reagents and are capable of performing variety of operations that can implement microreactor, complex multiple core–shell structure, and many applications in biomedical research such as drug encapsulation, targeted drug delivery systems, and multifunctionalization on carriers. Droplet‐based systems have been directly used to synthesize particles and encapsulate many biological entities for biomedicine applications due to their powerful encapsulation capability and facile versatility. In this paper, we review its origin, deviation, and evolution to draw a clear future, especially for droplet‐based biomedical applications. This paper will focus on droplet generation, variations and complication as starter, and logistically lead to the numerous typical applications in biomedical research. Finally, we will summarize both its challenge and future prospects relevant to its droplet‐based biomedical applications.     

Reduction of interferences in graphite furnace atomic absorption spectrometry by multiple linear regression modelling

The multivariate effects of Na, K, Mg and Ca as nitrates on the electrothermal atomisation of manganese, cadmium and iron were studied by multiple linear regression modelling. Since the models proved to efficiently predict the effects of the considered matrix elements in a wide range of concentrations, they were applied to correct the interferences occurring in the determination of trace elements in seawater after pre-concentration of the analytes. In order to obtain a statistically significant number of samples, a large volume of the certified seawater reference materials CASS-3 and NASS-3 was treated with Chelex-100 resin; then, the chelating resin was separated from the solution, divided into several sub-samples, each of them was eluted with nitric acid and analysed by electrothermal atomic absorption spectrometry (for trace element determinations) and inductively coupled plasma optical emission spectrometry (for matrix element determinations). To minimise any other systematic error besides that due to matrix effects, accuracy of the pre-concentration step and contamination levels of the procedure were checked by inductively coupled plasma mass spectrometric measurements. Analytical results obtained by applying the multiple linear regression models were compared with those obtained with other calibration methods, such as external calibration using acid-based standards, external calibration using matrix-matched standards and the analyte addition technique. Empirical models proved to efficiently reduce interferences occurring in the analysis of real samples, allowing an improvement of accuracy better than for other calibration methods.