基于二维相关近红外光谱技术测定辅料十二烷基硫酸钠

Measure Excipients Sodium Lauryl Sulfate according to the Two-Dimensional Correlation Near-infrared Light Spectrum Skill

为了测定缬沙坦和辅料十二烷基硫酸钠共存样品中辅料的含量，配制成含有十二烷基硫酸钠含量为0．30％～5．88％的缬沙坦样品共35个。以十二烷基硫酸钠浓度为外扰，分别研究了十二烷基硫酸钠和缬沙坦共存样品的二维相关近红外光谱特征，分别选择随浓度变化大的区间4119-4216cm-1、4235～4273cm-1、4312-4351cm-1、4484-4676cm-1、5732～5828cm-1、8132～8323cm-1为光谱的建模区间，利用偏最小二乘方法建立定量预测模型。结果表明：模型的交叉验证决定系数为0．8670，交叉验证均方根误差（RMSECV）为0．804，预测均方根误差（RMSEP）为0．775，对未知样品集预测结果的平均相对误差为0.25％，这表明二维相关近红外光谱技术选择的波段建立的定量模型具有较好的预测效果。

Objective： In order to measure the materials content in concomitant samples of valsartan and excipients sodium lauryl sulfate, 1 configured 35 valsartan samples with 0.30 %-5.88 % sodium lauryl sulfates. Methods： sodium dodecyl sulfate concentration outside interference, I studied the two-dimensional correlation near-infrared light spectrum characteristics of coexistence samples of sodium lauryl sulfate and valsartan, 1 selected the spectrum modeling interval which changes greatly with concentration, that is, 4119-4216 cm-1 , 4235-4273 cm-1, 4 312-4351 cm-1, 4484-4676 cm-1, 5732-5828 cm-1, 8132-8323 cm-1, then I established quantitative forecasting model by using partial least squares method. Result： The result showed that cross-validation determination coefficient of the model was 0.8670, the root mean square error of cross-validation （RMSECV） was 0.804, the root mean square error of prediction （RMSEP） was 0,775, and the average relative error of predicted result and unknown samples was 0.25 %. Conclusion： This shows that the predictive effect of quantitative model is good, which builds according to the waveband of the two-dimensional correlation near-infrared light spectrum skill. Keywords： two-dimensional correlation spectroscopy; near-infrared spectroscopy; sodium lauryl sulfate; pharmaceutical excipients.