An accurate correlation for estimation the pH of CO۲-saturated solutions: Implications for CO۲ sequestration

CO۲ sequestration, particularly through ocean injection, is a key method for reducing atmospheric CO۲ due to the oceans' capacity. However, this raises concerns about water pH changes. This study addresses gaps in previous research by employing a large, diverse dataset for accurate pH prediction under various conditions. Data on temperature, pressure, salinity, and pH of NaCl solutions saturated with CO۲ were collected. Pressure showed the highest correlation with pH, while temperature had the lowest. The data was divided into training and testing sets to develop a Gene Expression Programming (GEP) model. Error metrics for training and testing sets showed R² values of ۰.۹۰۳۷ and ۰.۹۰۳۵, and RMSE values of ۰.۱۱۷۲ and ۰.۱۱۸۲. Residual error plots indicated good performance across pressure and salinity ranges. Sensitivity analysis confirmed pressure had the highest influence on pH output. The model provides valuable insights for predicting pH changes in CO۲-saturated NaCl solutions, aiding CO۲ sequestration efforts and environmental studies.

Keywords:

pH estimation , CO۲ sequestration , Ocean , Machine learning