New QSPRs for Liquid Heat Capacity.

Area of Science:

• Physical Chemistry
• Computational Chemistry

Background:

• Quantitative Structure-Property Relationships (QSPRs) are established predictive models in chemistry and engineering.
• Existing QSPRs for liquid heat capacity are often limited to a single temperature (298.15 K) and struggle with oxygen-containing compounds.

Purpose of the Study:

• To develop novel QSPR models for predicting liquid heat capacity at various temperatures.
• To address limitations of existing QSPR models, particularly concerning temperature dependence and specific chemical functionalities.

Main Methods:

• Utilized data from the DIPPR database.
• Employed a novel search method for selecting molecular descriptors.
• Developed QSPR models for liquid heat capacity prediction.

Main Results:

• Successfully developed QSPR models for liquid heat capacity applicable across a range of temperatures.
• The novel descriptor selection method showed improvement over existing QSPR approaches.
• The models did not resolve prediction challenges associated with oxygen-containing chemical species.

Conclusions:

• QSPR models can effectively predict liquid heat capacity at various temperatures.
• The developed QSPR approach offers advancements in predicting thermophysical properties.
• Further research is needed to address the prediction of liquid heat capacity for oxygen-containing compounds.