Prediction of nematic transition temperatures of quinoline-containing liquid crystalline compounds from molecular structure

ABSTRACT

A quantitative structure–property relationship study was performed on 51 quinoline-containing liquid crystalline compounds to predict their nematic–isotropic transition temperatures using chemical structure-based descriptors calculated by CODESSA software. The descriptors were also analysed by using principal component analysis. Essentials accounting for a reliable model were all considered carefully during model construction and assessment process. Four variables were selected out by best multilinear regression and were used as inputs to perform the multiple linear regressions. The obtained model was validated both internally and externally. For the test set, it gave a predictive correlation coefficient (R) of 0.9914, root mean square error of 10.5253 and absolute average relative deviation of 5.2213%, respectively. The applicability domain was checked by the leverage approach to verify prediction reliability. In addition, model fit and regression diagnostics were done through the influence plot which reflects the hat values, studentised residuals and Cook’s distance statistics of each sample. The relationships between the descriptors and the nematic–isotropic transition temperature of compounds were discussed. The odd-even effect in the transition temperatures of mesogens in the same homologous series was also discussed.

Graphical Abstract

KEYWORDS:

• Nematic–isotropic transition temperature
• quinoline
• CODESSA software
• best multilinear regression
• influence plot

Acknowledgements

The authors thank the R Development Core Team for affording the free R software. This work was supported by the Young Scholars Science Foundation of Lanzhou Jiaotong University: [Grant Number 2011009].

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplemental meterial

Supplemental data for this article can be accessed here.

Additional information

Funding

This work was supported by the Young Scholars Science Foundation of Lanzhou Jiaotong University: [Grant Number 2011009].