G4CEP: A G4 theory modification by including pseudopotential for molecules containing first-, second- and third-row...

Area of Science:

• Computational Chemistry
• Quantum Chemistry
• Theoretical Chemistry

Background:

• The G4 all-electron composite method is a benchmark for high-accuracy thermochemical calculations.
• Developing accurate pseudopotential methods is crucial for extending high-level computational chemistry to larger systems.

Purpose of the Study:

• To develop and benchmark the G4CEP composite method, an adaptation of the G4 method using compact effective pseudopotentials (CEP).
• To evaluate the accuracy of G4CEP for predicting thermochemical data, including enthalpies of formation and electron affinities.

Main Methods:

• The G4CEP method was derived from the G4 all-electron theory by incorporating CEP.
• The G3/05 test set, comprising 440 thermochemical data points for atoms and compounds from the first three periods, was used for benchmarking.
• Calculations focused on molecular geometries and thermochemical properties.

Main Results:

• G4CEP achieved a mean absolute error of 1.09 kcal mol⁻¹, slightly higher than the G4 all-electron method's 0.83 kcal mol⁻¹.
• Optimized molecular geometries showed high similarity between G4 and G4CEP, suggesting core-electron effects influence pseudopotential accuracy.
• G4CEP demonstrated a significantly lower uncertainty (±2.72 kcal mol⁻¹) compared to G3CEP (±4.06 kcal mol⁻¹).

Conclusions:

• G4CEP represents a significant advancement in combining composite methods with pseudopotentials, offering improved reliability.
• The method is particularly reliable for calculating enthalpies of formation and electron affinities.
• Further investigation into core-electron effects and basis set adjustments is warranted to fully understand discrepancies with experimental data.