Application of chemometric methods in Auger electron microanalysis of composites

Principal component analysis and cluster analysis have been used in AES investigations of fibre-matrix interactions in alumina fibre reinforced MgLi-alloys prepared by high-pressure infiltration. The samples have been fractured in ultra high vacuum to expose surfaces and interfaces without contamination. All main components exhibit Auger valence band transitions which change their shape with the chemical state. Chemometric methods have been utilized to identify characteristic peak shapes and to classify the investigated areas by composition. Fibre fracture surfaces are characterized by Al, magnesia and Li oxide formed by a redox reaction of alumina with Mg and Li penetrating along grain boundaries. For samples with high silica content a thin interfacial oxide layer on matrix grain boundaries and matrix-fibre interfaces has been found.     

Simplified, rapid and inexpensive extraction procedure for a high-performance liquid chromatographic method for determination of disopyramide and its main metabolite mono-N-dealkylated disopyramide in serum.

A simplified, rapid and inexpensive extraction procedure for the determination of the antiarrhythmic drug disopyramide and its main metabolite mono-N-desalkylated disopyramide in serum by high-performance liquid chromatography has been developed. The analysis uses ultraviolet detection at 254 nm, and a 5 micron reversed-phase column with a mobile phase of water-triethylamine-acetonitrile-PIC-B8 reagent. Serum extraction is performed with dichloromethane and 1 M sodium hydroxide. p-Chlorodisopyramide is used as internal standard. Recovery rates were 94.5% (S.D. 5.7%) for disopyramide, 96.8% (S.D. 2.2%) for mono-N-desalkylated disopyramide and 97.9% (S.D. 2.8%) for the internal standard.     

Resolution prediction and optimization of temperature programme in comprehensive two-dimensional gas chromatography

A model is developed for predicting the resolution of interested component pair and calculating the optimum temperature programming condition in the comprehensive two-dimensional gas chromatography (GC x GC). Based on at least three isothermal runs, retention times and the peak widths at half-height on both dimensions are predicted for any kind of linear temperature-programmed run on the first dimension and isothermal runs on the second dimension. The calculation of the optimum temperature programming condition is based on the prediction of the resolution of "difficult-to-separate components" in a given mixture. The resolution of all the neighboring peaks on the first dimension is obtained by the predicted retention time and peak width on the first dimension, the resolution on the second dimension is calculated only for the adjacent components with un-enough resolution on the first dimension and eluted within a same modulation period on the second dimension. The optimum temperature programming condition is acquired when the resolutions of all components of interest by GC x GC separation meet the analytical requirement and the analysis time is the shortest. The validity of the model has been proven by using it to predict and optimize GC x GC temperature programming condition of an alkylpyridine mixture.     

Investigation of a capillary electrophoretic approach for direct quantification of apolipoprotein A-I in serum

In the present study a rapid, reproducible and robust capillary electrophoresis (CE) procedure for the quantification of apolipoprotein A-I (Apo A-I) in serum without pretreatment has been developed (total run time, 11 min). The coefficients of variation (CV; n = 10) for the relative peak area are 1.8% at a concentration of 145 mg/dL and 1.6% at 196 mg/dL; and for the inter-assay 8.9% at 161 mg/dL (10 consecutive days), i.e., similar to the CVs of a high-throughput immunonephelometric routine assay. The CV for the migration time is 0.4% (n = 20). The robustness of the CE approach was tested in patient samples with hemolysis, hyperbilirubinemia and hyperlipidemia. A comparison of 99 Apo A-I serum values with results of a fixed-time immunonephelometric routine assay showed a positive constant bias of 60% (mean) for the immunonephelometric values, no deviation from linearity, but significant deviations in several samples. Investigations on interferences in the CE analyses gave no evidence that CE failed. Our study shows that CE is amenable to a fast analysis and a reproducible and reliable quantification of Apo A-I level in sera of various clinical samples.     

Application of probabilistic neural network in the clinical diagnosis of cancers based on clinical chemistry data

As a recently developed and powerful classification tool, probabilistic neural network was used to distinguish cancer patients from healthy persons according to the levels of nucleosides in human urine. Two datasets (containing 32 and 50 patterns, respectively) were investigated and the total consistency rate obtained was 100% for dataset 1 and 94% for dataset 2. To evaluate the performance of probabilistic neural network, linear discriminant analysis and learning vector quantization network were also applied to the classification problem. The results showed that the predictive ability of the probabilistic neural network is stronger than the others in this study. Moreover, the recognition rate for dataset 2 can achieve to 100% if combining these three methods together, which indicated the promising potential of clinical diagnosis by combining different methods.     

新型微填充柱－毛细管柱色谱操作条件的优化及灰色关联分析

通过正文设计实验方法优化了新型微填充柱。毛细管柱色谱操作条件，并利用灰色关联分析方法考察了操作条件与柱效的相关程度，对实验能起一定的指导作用。