On the evaluation of the carbon dioxide solubility in polymers using gene expression programming

Area of Science:

• Polymer science
• Chemical engineering
• Computational chemistry

Background:

• Carbon dioxide (CO2) solubility in polymers is vital for material design and chemical processes.
• Accurate prediction models are needed for efficient engineering applications.

Purpose of the Study:

• To develop predictive models for CO2 solubility in polystyrene (PS), polybutylene succinate-co-adipate (PBSA), and polybutylene succinate (PBS) using gene expression programming (GEP).
• To analyze the influence of pressure and temperature on CO2 solubility.
• To validate the statistical robustness of the developed correlations.

Main Methods:

• Gene expression programming (GEP) was employed to generate correlations for CO2 solubility.
• Henry's law was used for trend analysis concerning pressure and temperature.
• The leverage approach was applied for outlier detection.

Main Results:

• GEP models achieved high prediction accuracy with average absolute relative errors of 9.71% (PS), 5.87% (PBSA), and 1.63% (PBS).
• Trend analysis confirmed that CO2 solubility increases with pressure and decreases with temperature.
• Outlier detection validated the statistical significance of the developed correlations.

Conclusions:

• GEP provides an efficient method for predicting CO2 solubility in the specified polymers.
• The developed correlations offer reliable predictions for engineering applications.
• The study confirms established thermodynamic trends for gas solubility in polymers.