Scholarlink

<p>Novel analytical methodologies integrating UV spectrophotometric techniques with chemometric models were developed for the simultaneous determination of cinnarizine (CIN), domperidone (DOM), and benzophenone (BNZ), a carcinogenic degradation product of CIN, without prior separation. Despite their clinical significance, no existing methods have been reported for their simultaneous quantification. This approach aligns with green and white analytical chemistry principles, offering an environmentally sustainable solution. Predictive models were constructed using Classical Least Squares (CLS), Partial Least Squares (PLS), and Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS). A key methodological advancement was the application of D-optimal design via MATLAB’s Candexch algorithm, generating a strategically balanced validation dataset of 13 mixtures and overcoming limits of conventional data splitting in chemometric modelling. The methods demonstrated robust linearity across concentration ranges of 4–20&#xa0;µg/mL for CIN, 3–15&#xa0;µg/mL for DOM, and 1–5&#xa0;µg/mL for BNZ. Calibration performance was excellent, with root mean square errors of calibration (RMSEC) values of 0.036–0.062 for CLS, 0.013–0.024 for PLS, and 0.009–0.012 for MCR-ALS. Notably, MCR-ALS exhibited the smallest and most consistent RMSEC range, demonstrating superior accuracy and stability compared to other chemometric models. Validation studies confirmed excellent method performance with recovery rates between 98 and 102%. The root mean square errors of prediction (RMSEP) for the validation set were 0.042–0.201 for CLS, 0.035–0.187 for PLS, and 0.022–0.154 for MCR-ALS, with MCR-ALS consistently exhibiting exceptional predictive capability. The environmental sustainability credentials of the methodology were comprehensively evaluated using nine distinct evaluation tools: NEMI, Complex GAPI, AGREE, BAGI, RGB12, SDAGI, Carbon Footprint Analysis, GSST, and NQS. These rigorous assessments confirmed the method's exceptional environmental compatibility while maintaining analytical excellence, positioning this approach as an ideal sustainable alternative for pharmaceutical quality control applications aligned with UN Sustainable Development Goals.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

MATLAB-candexch algorithm-enhanced UV spectrophotometric-chemometric models for green, blue, and white determination of cinnarizine, domperidone, and carcinogenic impurity in pharmaceuticals: NQS assessment and UN-SDGs integration

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月12日

查看全文
  • Hussein N. Ghanem,
  • Asmaa A. El-Zaher,
  • Enas A. Taha,
  • Ahmed Emad F. Abbas,
  • Sally T. Mahmoud

摘要

Novel analytical methodologies integrating UV spectrophotometric techniques with chemometric models were developed for the simultaneous determination of cinnarizine (CIN), domperidone (DOM), and benzophenone (BNZ), a carcinogenic degradation product of CIN, without prior separation. Despite their clinical significance, no existing methods have been reported for their simultaneous quantification. This approach aligns with green and white analytical chemistry principles, offering an environmentally sustainable solution. Predictive models were constructed using Classical Least Squares (CLS), Partial Least Squares (PLS), and Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS). A key methodological advancement was the application of D-optimal design via MATLAB’s Candexch algorithm, generating a strategically balanced validation dataset of 13 mixtures and overcoming limits of conventional data splitting in chemometric modelling. The methods demonstrated robust linearity across concentration ranges of 4–20 µg/mL for CIN, 3–15 µg/mL for DOM, and 1–5 µg/mL for BNZ. Calibration performance was excellent, with root mean square errors of calibration (RMSEC) values of 0.036–0.062 for CLS, 0.013–0.024 for PLS, and 0.009–0.012 for MCR-ALS. Notably, MCR-ALS exhibited the smallest and most consistent RMSEC range, demonstrating superior accuracy and stability compared to other chemometric models. Validation studies confirmed excellent method performance with recovery rates between 98 and 102%. The root mean square errors of prediction (RMSEP) for the validation set were 0.042–0.201 for CLS, 0.035–0.187 for PLS, and 0.022–0.154 for MCR-ALS, with MCR-ALS consistently exhibiting exceptional predictive capability. The environmental sustainability credentials of the methodology were comprehensively evaluated using nine distinct evaluation tools: NEMI, Complex GAPI, AGREE, BAGI, RGB12, SDAGI, Carbon Footprint Analysis, GSST, and NQS. These rigorous assessments confirmed the method's exceptional environmental compatibility while maintaining analytical excellence, positioning this approach as an ideal sustainable alternative for pharmaceutical quality control applications aligned with UN Sustainable Development Goals.