阿立哌唑粉末近红外光谱定量模型的建立与应用  

Establishment and application of near-infrared spectroscopy quantitative model of aripiprazole content

作　　者：王薇青 张钰婷 陆峰 于治国[1] WANG Weiqing;ZHANG Yuting;LU Feng;YU Zhiguo(School of Pharmacy,Shenyang Pharmaceutical University,Shenyang 110016,China;College of Pharmacy,the Second Military Medical University,Shanghai 200433,China)

机构地区：[1]沈阳药科大学药学院,辽宁沈阳110016 [2]海军军医大学药学院,上海200433

出　　处：《沈阳药科大学学报》2023年第1期36-41,49,共7页Journal of Shenyang Pharmaceutical University

基　　金：国家重点研发计划“国家质量基础的共性技术研究与应用”(2017YFF0210103);上海市产业转型升级发展专项(JJ-YJCX-01-17-2237)。

摘　　要：目的建立阿立哌唑粉末近红外光谱定量模型,探讨近红外光谱技术监测阿立哌唑原辅料混合终点的同时预测粉末中阿立哌唑含量的可行性。方法采集含量范围在84.60%~112.8%内的阿立哌唑粉末样品的近红外光谱,使用高效液相色谱法测定样品的实际含量,选择多元散射校正结合Savitzky-Golay卷积平滑的预处理方法,建模波段为5550.00~7400.00 cm^(-1),使用偏最小二乘法建立含量定量模型。使用建立的模型分别预测离线样品和经过光谱校正的在线样品,并用HPLC法测定的含量进行验证。结果模型校正均方根误差为2.94,校正相关系数为0.9552。离线样品的预测均方根误差为4.82,在线样品的预测均方根误差为3.34,离线样品和在线样品的预测结果与HPLC法测定的结果一致(P>0.05)。结论建立的基于近红外光谱技术的阿立哌唑粉末定量模型在监测阿立哌唑原辅料混合终点的同时可在线准确预测粉末中阿立哌唑的含量。Objective To establish a near-infrared spectroscopy quantitative model of aripiprazole content,and to explore the feasibility of monitoring the mixed endpoint of aripiprazole and predicting the content of aripiprazole by near-infrared spectroscopy.Methods The near infrared spectra of aripiprazole samples with content range of 84.60%-112.80% were collected,and the actual contents of the samples were determined by HPLC.The pretreatment method of multiplicative signal correction(MSC)+Savitzky-Golay filter was selected.The modeling region was 5550.00-7400.00 cm^(-1),and the quantitative content model was established by partial least squares method.This model was used to predict the off-line samples and the in-line samples with spectral correction,respectively,and was verified with the contents determined by HPLC.Results The root mean square error of calibration was 2.94,and the calibration correlation coefficient was 0.9552.The root mean square error of prediction of the off-line samples was 4.82,and that of the in-line samples was 3.34.The prediction results of the off-line and in-line samples were consistent with those of HPLC(P>0.05).Conclusion Near-infrared spectroscopy quantitative model of aripiprazole content can be used to predict in-line samples,and near-infrared spectroscopy can be used to predict aripiprazole content while monitoring the mixed endpoint.

关 键 词：阿立哌唑 近红外光谱技术 含量 在线 

分 类 号：R917[医药卫生—药物分析学]