College of Chemistry & Chemical Engineering, Lanzhou University, Lanzhou 730000, PR China.
College of Chemistry & Chemical Engineering, Lanzhou University, Lanzhou 730000, PR China.
Spectrochim Acta A Mol Biomol Spectrosc. 2022 Oct 15;279:121413. doi: 10.1016/j.saa.2022.121413. Epub 2022 May 27.
As one of the important factors in chemical production, catalyst content directly affects the process of reaction and the quality of products. The quantitative analysis of trace catalyst in homogeneous reaction system is still faced with great challenges. In this work, a simple and effective approach to the rapid determination of trace homogeneous catalyst (THC) was proposed based on UV-vis spectrophotometry. Wavelet transform and Tchebichef curve moment methods were combined with gray wolf algorithm to extract the feature information from the original UV-vis spectra of samples. Then the partial least-squares model was established. The predictive correlation coefficient (R) was 0.9842, and the limit of quantification was 0.07 ‰. The intra-day and inter-day precision were 3.97 % and 4.36 %, respectively. The spiked recoveries of three different concentrations in actual samples were between 97.6 and 101.9 %. The results indicated that the obtained model was satisfactory and could be used in practical measurement. Compared with the conventional modeling methods, the proposed approach was more accurate and reliable, which provided a feasible new pathway for enterprise product quality control.
作为化学生产的重要因素之一，催化剂含量直接影响反应过程和产品质量。均相反应体系中痕量催化剂的定量分析仍然面临巨大挑战。在这项工作中，提出了一种基于紫外-可见分光光度法的快速测定痕量均相催化剂（THC）的简单有效方法。小波变换和 Chebichef 曲线矩法与灰狼算法相结合，从样品的原始紫外-可见光谱中提取特征信息。然后建立偏最小二乘模型。预测相关系数（R）为 0.9842，定量限为 0.07‰。日内和日间精密度分别为 3.97%和 4.36%。在实际样品中三个不同浓度的加标回收率在 97.6%至 101.9%之间。结果表明，所得到的模型是令人满意的，可以用于实际测量。与传统的建模方法相比，所提出的方法更加准确可靠，为企业产品质量控制提供了一种可行的新途径。
