16 April 1990 | B. W. H. van Beest, G. J. Kramer, R. A. van Santen
This paper presents a force field for silicas and aluminophosphates derived from ab initio calculations on small clusters. The authors show that force fields cannot be determined solely from cluster data and that incorporating bulk-system information is essential. They derive a force field combining microscopic (ab initio) and macroscopic (experimental) data, ensuring accuracy and transferability to other polymorphs. The force field is based on the Buckingham form, with parameters determined from ab initio calculations on clusters. The interatomic potentials include Coulomb and covalent contributions. The authors demonstrate that the force field can accurately describe the potential-energy surface of silica clusters and predict the properties of quartz. They also show that incorporating bulk information into the fitting procedure is essential for developing a high-quality force field. The force field is applied to other systems, including aluminophosphates, and is shown to be more accurate than previous force fields. The paper concludes that accurate force fields can be developed for solids starting from ab initio calculations on small clusters, ensuring proper microscopic basis for nearest-neighbor interactions and applicability to different polymorphs. The approach is applicable to other systems, such as amorphous carbon or silicon.This paper presents a force field for silicas and aluminophosphates derived from ab initio calculations on small clusters. The authors show that force fields cannot be determined solely from cluster data and that incorporating bulk-system information is essential. They derive a force field combining microscopic (ab initio) and macroscopic (experimental) data, ensuring accuracy and transferability to other polymorphs. The force field is based on the Buckingham form, with parameters determined from ab initio calculations on clusters. The interatomic potentials include Coulomb and covalent contributions. The authors demonstrate that the force field can accurately describe the potential-energy surface of silica clusters and predict the properties of quartz. They also show that incorporating bulk information into the fitting procedure is essential for developing a high-quality force field. The force field is applied to other systems, including aluminophosphates, and is shown to be more accurate than previous force fields. The paper concludes that accurate force fields can be developed for solids starting from ab initio calculations on small clusters, ensuring proper microscopic basis for nearest-neighbor interactions and applicability to different polymorphs. The approach is applicable to other systems, such as amorphous carbon or silicon.