摘要
应用二阶导数红外光谱对羊毛纤维、蚕丝纤维及其混纺纤维进行鉴别,并建立羊毛/蚕丝混纺纤维含量的二阶导数红外光谱定量分析模型。实验结果表明:应用二阶导数谱图中的1265±1、996±1、(974±1) cm-1一组吸收峰可以对蚕丝纤维进行定性。应用二阶导数红外谱图并结合偏最小二乘法(PLS ),选取835.00~495.00 cm-1作为建模时的峰范围建立定量模型。该定量方法数据重复性较好,标准偏差≤1.51;定量模型的回归系数及定标均方根误差分别为R:0.99943和RM S EC :1.16;与化学溶解法相比,两者的定量结果与实际含量的绝对偏差相差不大,但红外光谱法的回收率稍低,在87.76%~107.82%之间。
This paper applies second derivative infrared spectroscopy to distinguish wool ,silk fibers and wool/silk blend fibers and establishes second derivative infrared spectroscopy quantitative analysis model of wool/silk blend fiber ratio .The experimental results show that a group of absorption peak in sec-ond derivative infrared spectroscopy (1 265 ± 1 ,(996 ± 1)cm -1 and (974 ± 1)cm -1 ) can be used to deter-mine the nature of silk fiber .The quantitative model is established through selecting 835 .00~495 .00 cm -1 as the peak scope in combination of second derivative infrared spectroscopy and partial least squares (PLS) .This quantitative method owns good data repeatability with the standard deviation ≤1 .51 .The re-gression coefficient and root-mean-square error of the model are R :0 .999 43 and RMSEC :1 .16 respective-ly .Compared with chemical dissolution method ,the absolute deviation of the quantitative results and the actual content is not large , but the recovery rate of infrared spectroscopy is slightly lower , between 87.76% and 107 .82% .
出处
《浙江理工大学学报(自然科学版)》
2014年第2期107-111,共5页
Journal of Zhejiang Sci-Tech University(Natural Sciences)
关键词
羊毛
蚕丝
二阶导数红外光谱
定量
纤维鉴别
混纺纤维
wool
silk
second derivative infrared spectroscopy
fiber identification
blend fiber