This website provides a convenient way to access information on CASTEP's most used pseudopotential libraries, reporting basic information as well as technical details and validation metrics for each of them. All the pseudopotentials here reported have been calculated starting with the given parameters with CASTEP 20.1.
The main page is a periodic table reporting all known elements. Clicking on them a box to the right will visualize some fundamental chemical information about the element as well as the default pseudopotential for it, and a button which can show a popup window listing all available pseudopotentials. From here, by clicking on one of the pseudopotential names, one can proceed to another page listing detailed information and plots about it. The meaning of each of these entries will be detailed in the following sections.
In this section some fundamental information about how the pseudopotential was calculated is included.
Ionic charge: number of electrons left free to act as "valence" for the pseudopotential.
XC functional: exchange-correlation functional used for the calculation of the pseudopotential.
Solver: solver method employed to calculate the potential in the free atom. CASTEP’s default is the scalar-relativistic method of Koelling and Harmon.
This table contains the suggested energy cutoffs for calculations using the given pseudopotential for different levels of precision, with descriptive names ranging from COARSE to EXTREME. When planning for a new calculation, a good estimate for the cutoff to use is, once you've decided on one of these levels, to pick the highest value from all the pseudopotentials that you're going to use.
This string is the descriptor that contains all the parameters CASTEP needs to rebuild the potential, and it can be inserted in the SPECIES_POT block of a .cell file. A more detailed description of the meaning of these strings can be found here.
States occupied by the valence electrons in the all electron solution of the atom expressed as hydrogenlike orbitals.
If available, this section gives the results of the Delta test, assessing the accuracy of pseudopotentials used in ab initio codes against an established standard. As a general rule, the lower the Delta value, the better the potential. The data may not be included for all pseudopotential libraries. All details about how the test is carried out as well as the results for different codes can be found at the website of the Center for Molecular Modeling.
A drop down menu that allows to choose which plot to visualize. These plots contain information about the final calculated pseudopotential.
Energy convergence: gives a plot of the calculated energy of the isolated, spherical atom vs. the cutoff used. Horizontal lines marking the various precision levels are plotted as well.
Beta projectors: projector functions of the non-local pseudopotential classified by momentum channel.
Partial waves: full (dashed lines) and pseudised (continuous) versions of the electronic wavefunctions corresponding to the beta projectors.
A table containing details on the projectors, including index, momentum channel, energy of the corresponding wavefunction, cutoff radius, scheme, type of the projector (Ultrasoft or Norm-conserving) and a colour legend to identify the corresponding lines in the plots.
Data for energy convergence, partial waves and beta projectors in ASCII format. Required for Gnuplot plotting as well.
Gnuplot and XMGrace source files required to reproduce the plots for energy convergence, partial waves and beta projectors.