Correlation consistent valence basis sets for use with the Stuttgart-Dresden-Bonn relativistic effective core potentials: the atoms Ga-Kr and In-Xe

Jan M.L. Martin,Andreas Sundermann

DOI: https://doi.org/10.1063/1.1337864

2000-11-15

Abstract:We propose large-core correlation-consistent pseudopotential basis sets for the heavy p-block elements Ga-Kr and In-Xe. The basis sets are of cc-pVTZ and cc-pVQZ quality, and have been optimized for use with the large-core (valence-electrons only) Stuttgart-Dresden-Bonn relativistic pseudopotentials. Validation calculations on a variety of third-row and fourth-row diatomics suggest them to be comparable in quality to the all-electron cc-pVTZ and cc-pVQZ basis sets for lighter elements. Especially the SDB-cc-pVQZ basis set in conjunction with a core polarization potential (CPP) yields excellent agreement with experiment for compounds of the later heavy p-block elements. For accurate calculations on Ga (and, to a lesser extent, Ge) compounds, explicit treatment of 13 valence electrons appears to be desirable, while it seems inevitable for In compounds. For Ga and Ge, we propose correlation consistent basis sets extended for (3d) correlation. For accurate calculations on organometallic complexes of interest to homogenous catalysis, we recommend a combination of the standard cc-pVTZ basis set for first- and second-row elements, the presently derived SDB-cc-pVTZ basis set for heavier p-block elements, and for transition metals, the small-core [6s5p3d] Stuttgart-Dresden basis set-RECP combination supplemented by (2f1g) functions with exponents given in the Appendix to the present paper.

Chemical Physics,Computational Physics

What problem does this paper attempt to address?

The problem this paper attempts to address is the development of high-quality valence basis sets under relativistic effective core potentials (RECP) for heavy p-block elements (from gallium to krypton and indium to xenon) and the validation of these basis sets in molecular calculations. Specifically, the authors propose cc-pVTZ and cc-pVQZ quality basis sets compatible with the Stuttgart-Dresden-Bonn (SDB) relativistic effective core potentials and validate the performance of these basis sets through calculations on a series of diatomic molecules.

### Main Research Objectives:
1. **Develop high-quality basis sets**: Develop cc-pVTZ and cc-pVQZ quality basis sets compatible with SDB relativistic effective core potentials for heavy p-block elements (e.g., gallium, germanium, arsenic, selenium, etc.).
2. **Validate basis set performance**: Validate the performance of these basis sets by calculating the dissociation energies, bond lengths, harmonic frequencies, and first anharmonicities of a series of third-period and fourth-period diatomic molecules.
3. **Consider relativistic effects**: Ensure that these basis sets perform well in calculations that include relativistic effects, especially when dealing with heavy elements.
4. **Optimize extended basis sets**: Develop extended basis sets for certain elements (e.g., gallium and germanium) to include (3d) correlation, further improving computational accuracy.

### Research Background:
- **Importance of basis sets**: In electronic structure calculations, the quality of the basis set is crucial for the accuracy of the results. Although significant progress has been made in handling multi-particle correlations, the choice of basis set remains a primary factor in determining computational quality.
- **Relativistic effects**: For heavy elements, relativistic effects cannot be ignored, as they significantly impact the reliability of computational results.
- **Limitations of existing basis sets**: Existing basis sets are primarily designed for lighter elements, with relatively few basis sets developed for heavy elements, especially those considering relativistic effects.

### Research Methods:
- **Basis set optimization**: Use the MOLPRO 2000 software and the optimization program DOMIN to optimize the basis set parameters.
- **Validation calculations**: Perform calculations on a series of diatomic molecules using the CCSD(T) method and compare the results with experimental data to validate the performance of the basis sets.
- **Extended basis sets**: Develop extended basis sets for certain elements, such as gallium and germanium, to include (3d) correlation, further improving computational accuracy.

### Main Conclusions:
- **Basis set performance**: The proposed SDB-cc-pVTZ and SDB-cc-pVQZ basis sets perform comparably to, and in some cases better than, the all-electron cc-pVTZ and cc-pVQZ basis sets when calculating the properties of third-period and fourth-period diatomic molecules.
- **Relativistic effects**: These basis sets perform well in calculations that include relativistic effects, especially when dealing with heavy elements.
- **Extended basis sets**: For gallium and germanium, the extended basis sets (e.g., cc-pDVTZ and cc-pDVQZ) significantly improve computational accuracy, particularly when handling (3d) correlation.

In summary, this paper provides important tools for high-precision electronic structure calculations of heavy p-block elements by developing and validating high-quality relativistic effective core potential basis sets.