微型色谱柱制备及其GC谱线分离性能的优化

Fabrication of Micro Chromatographic Columns and Optimization of Their GC Spectral Line Separation Performances

基于激光刻蚀加工技术设计并制备了新型蛇形沟道布局微型色谱柱芯片,采用硅材料衬底刻蚀加工,设计并制备了不同沟道布局的微型色谱柱,并对混合组分进行分离分析。选取成本较低、加工周期较短的紫外激光刻蚀进行微型色谱柱制备,并针对实验得到的色谱图存在的重叠峰问题,利用连续小波变换提取色谱重叠峰的特征点,构建反映重叠峰前肩峰、后肩峰、峰高的特征变量,将峰高、峰宽和峰位作为模型输入,子峰面积作为模型输出,提出基于反向传播(BP)神经网络的CW_TBP网络构建其映射关系,算法能够明确识别出7个重叠峰子峰,最终误差在0.05以内,拟合度在0.97以上,表明该方法可以对重叠峰进行高精度的识别。

Base on laser ablation process technology, a new type of serpentine channel layout micro chromatographic column chip was designed and fabricated. Using the silicon material substrate for etching process, the micro chromatographic columns with different channel layouts were designed and fabricated to separate and analyze the mixed components. The ultraviolet laser etching with low cost and short processing cycle was selected to prepare the micro chromatographic columns. For the problem of overlapping peaks in the chromatogram obtained from the experiment, the continuous wavelet transform was used to extract the characteristic points of the chromatographic overlapping peaks. Characteristic variables reflecting the front shoulder, back shoulder, and peak height of overlapping peaks were constructed. With the peak height, peak width and peak position as model inputs and the sub-peak area as the model output, the CWT_BP network based on the back propagation(BP) neural network was proposed to construct its mapping relationship. The algorithm can clearly identify 7 overlapping peaks with a final error within 0.05 and a fitting degree is above 0.97, proving that the overlapping peaks are identified with high precision by the proposed method.