Wang Hui, Barona David, Oladepo Sulayman, Williams Lisa, Hoe Susan, Lechuga-Ballesteros David, Vehring Reinhard
Department of Mechanical Engineering, University of Alberta, Edmonton, AB T6G 2G8, Canada.
Department of Mechanical Engineering, University of Alberta, Edmonton, AB T6G 2G8, Canada; Department of Chemistry, King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia.
J Pharm Biomed Anal. 2017 Jul 15;141:180-191. doi: 10.1016/j.jpba.2017.04.003. Epub 2017 Apr 23.
A new macro-Raman system equipped with a motorized translational sample stage and low-frequency shift capabilities was developed for bulk composition and homogeneity analysis of multi-component pharmaceutical powders. Different sampling methods including single spot and scanning measurement were compared. It was found that increasing sample volumes significantly improved the precision of quantitative composition analysis, especially for poorly mixed powders. The multi-pass cavity of the macro-Raman system increased effective sample volumes by 20 times from the sample volume defined by the collection optics, i.e., from 0.02μL to about 0.4μL. A stochastic model simulating the random sampling process of polydisperse microparticles was used to predict the sampling errors for a specific sample volume. Comparison of fluticasone propionate mass fractions of the commercial products Flixotide 250 and Seretide 500 simulated for different sampling volumes with experimentally measured compositions verified that the effective sample volume of a single point macro-Raman measurement in the multi-pass cavity of this instrument was between 0.3μL and 0.5μL. The macro-Raman system was also successfully used for blend uniformity analysis. It was concluded that demixing occurred in the binary mixture of l-leucine and d-mannitol from the observation that the sampling errors indicated by the standard deviations of measured leucine mass fractions increased during mixing, and the standard deviation values were all larger than the theoretical lower limit determined by the simulation. Since sample volume was shown to have a significant impact on measured homogeneity characteristics, it was concluded that powder homogeneity analysis results, i.e., the mean of individual test results and absolute and relative standard deviations, must be presented together with the effective sample volumes of the applied testing techniques for any measurement of powder homogeneity to be fully meaningful.
开发了一种新的宏观拉曼系统，该系统配备了电动平移样品台和低频位移功能，用于多组分药用粉末的整体成分和均匀性分析。比较了包括单点和扫描测量在内的不同采样方法。结果发现，增加样品体积可显著提高定量成分分析的精度，尤其是对于混合不佳的粉末。宏观拉曼系统的多程腔将有效样品体积从收集光学器件定义的样品体积增加了20倍，即从0.02μL增加到约0.4μL。使用模拟多分散微粒随机采样过程的随机模型来预测特定样品体积的采样误差。将不同采样体积模拟的商业产品辅舒酮250和舒利迭500的丙酸氟替卡松质量分数与实验测量的成分进行比较，验证了该仪器多程腔中单点宏观拉曼测量的有效样品体积在0.3μL至0.5μL之间。宏观拉曼系统也成功用于混合均匀性分析。从混合过程中亮氨酸质量分数测量标准差所表明采样误差增加且标准差均大于模拟确定的理论下限这一观察结果得出结论，在l-亮氨酸和d-甘露醇的二元混合物中发生了分层。由于样品体积对测量的均匀性特征有显著影响，因此得出结论，对于任何粉末均匀性测量，粉末均匀性分析结果，即各个测试结果的平均值以及绝对和相对标准差，必须与所应用测试技术的有效样品体积一起呈现，才有充分意义。
