pKa prediction for acidic phosphorus-containing compounds using multiple linear regression with computational...

Area of Science:

• Computational Chemistry
• Physical Chemistry
• Organic Chemistry

Background:

• Acidic phosphorus-containing compounds are crucial as extractants in various chemical processes.
• Predicting the acidity (pKa) of these molecules is essential for optimizing their performance.
• Existing methods for pKa prediction may be limited in scope or accuracy.

Purpose of the Study:

• To establish a quantitative structure-property relationship (QSPR) for the pKa of acidic phosphorus-containing molecules.
• To develop a predictive model for pKa based on molecular structural parameters.
• To facilitate the rational design of novel phosphorus-based extractants with desired acidity.

Main Methods:

• Collected a dataset of 96 acidic phosphorus-containing molecules with pKa values ranging from 1.88 to 6.26.
• Utilized density functional theory (DFT) calculations to determine key structural parameters for each molecule.
• Employed multiple linear regression (MLR) analysis to correlate structural parameters with experimental pKa values, using training and test sets.

Main Results:

• A robust multiple linear regression model was developed, accurately describing the relationship between molecular structure and pKa.
• The model achieved high R-squared values of 0.974 for the training set and 0.966 for the test set, indicating excellent predictive performance.
• The derived regression equation quantitatively elucidates the influence of specific structural features on the acidity of these compounds.

Conclusions:

• The developed regression model provides a significant tool for predicting the pKa of new acidic phosphorus-containing molecules.
• This quantitative structure-property relationship is valuable for the targeted design and synthesis of improved phosphorus-based extractants.
• The findings contribute to a deeper understanding of the factors governing the acidity of organophosphorus compounds.