TY - JOUR
T1 - Quantitative determination of CL-20 polymorphs by mid-infrared diffuse reflectance spectroscopy
AU - Pan, Qing
AU - Su, Peng Fei
AU - Gao, Hong Xu
AU - Wen, Xiao Yan
AU - Pang, Si Ping
AU - Sun, Cheng Hui
AU - Zhang, Gao
N1 - Publisher Copyright:
© 2016, Editorial Board of Chinese Chinese Journal of Energetic Materials. All right reserved.
PY - 2016/5/25
Y1 - 2016/5/25
N2 - The quantitative calibration models of three polymorphs α-CL-20, γ-CL-20, ε-CL-20 in ε-CL-20 product were established by mid-infrared diffuse reflectance spectroscopy combined with partial least squares (PLS) method of chemometrics, and the cross-validation and external validation to the models were performed. Results show that three models have high accuracy and precision, the determination coefficients of cross-validation (R2) are 0.9973, 0.9714, 0.9848, respectively, and the root mean square error of cross-validation (RMSECV) are 0.222%, 0.650%, 0.670%, respectively. In the external validation, the root mean square error of prediction (RMSEP) are 0.422%, 0.813%, 1.02%, respectively, and the paired t-test results show that there is no significant difference between predicted values and actual ones. The quantitative analysis method of CL-20 polymorphs based on the above models can be applied to the product quality test of ε-CL-20, and the studies on the polymorph stability of ε-CL-20 in technical process and long-time storage of explosives and propellants.
AB - The quantitative calibration models of three polymorphs α-CL-20, γ-CL-20, ε-CL-20 in ε-CL-20 product were established by mid-infrared diffuse reflectance spectroscopy combined with partial least squares (PLS) method of chemometrics, and the cross-validation and external validation to the models were performed. Results show that three models have high accuracy and precision, the determination coefficients of cross-validation (R2) are 0.9973, 0.9714, 0.9848, respectively, and the root mean square error of cross-validation (RMSECV) are 0.222%, 0.650%, 0.670%, respectively. In the external validation, the root mean square error of prediction (RMSEP) are 0.422%, 0.813%, 1.02%, respectively, and the paired t-test results show that there is no significant difference between predicted values and actual ones. The quantitative analysis method of CL-20 polymorphs based on the above models can be applied to the product quality test of ε-CL-20, and the studies on the polymorph stability of ε-CL-20 in technical process and long-time storage of explosives and propellants.
KW - CL-20
KW - Mid-infrared diffuse reflectance spectroscopy
KW - Partial least squares (PLS) method
KW - Quantitative analysis of polymorph
UR - http://www.scopus.com/inward/record.url?scp=84973889501&partnerID=8YFLogxK
U2 - 10.11943/j.issn.1006-9941.2016.05.013
DO - 10.11943/j.issn.1006-9941.2016.05.013
M3 - Article
AN - SCOPUS:84973889501
SN - 1006-9941
VL - 24
SP - 503
EP - 508
JO - Hanneng Cailiao/Chinese Journal of Energetic Materials
JF - Hanneng Cailiao/Chinese Journal of Energetic Materials
IS - 5
ER -