# Input

To run a CAC simulation, one may create/modify `cac.in`

, in which the commands provide all input parameters for a CAC simulation.

The `cac.in`

file, along with the potential files (`embed.tab`

, `pair.tab`

, and `edens.tab`

for the EAM potential; `lj.para`

for the LJ potential), are read by the Fortran CAC code to run the CAC simulation.

The potential files for some FCC metals are provided in the `potentials`

directory.

### EAM potential

The EAM formulations for potential energy E and the force on atom k, \mathbf{f}_k, are

where

Note that the force formulation above only holds for monatomic pure materials.

The first line of each `*.tab`

file is

```
N first_val last_val
```

where `N`

is a positive integer that equals the number of data pair (each line starting from the second line), `first_val`

and `last_val`

are non-negative real numbers suggesting the first and the last datum in the first column (starting from the second line), respectively.

- In
`embed.tab`

, the first column is the unitless host electron energy \bar{\rho}; the second column is the embedded energy F, in eV. - In
`pair.tab`

, the first column is the interatomic distance r, in Angstrom; the second column is the pair potential V, in eV. - In
`edens.tab`

, the first column is the interatomic distance r, in Angstrom; the second column is the unitless local electron density \rho.

For example, the first few lines of `potentials/eam/Ag/williams/edens.tab`

are

```
3000 0.5018316703334310 5.995011000293092
0.5018316703334310 8.9800288540000004E-002
0.5036633406668621 9.0604138970000001E-002
0.5054950110002930 9.1404200869999990E-002
0.5073266813337241 9.2200486049999988E-002
```

In CAC simulations, an approximation is introduced to calculate the host electron density \bar{\rho} of the integration points in the coarse-grained domain. For more information, read chapter 3 of Shuozhi Xu's Ph.D. dissertation.

The readers may find EAM potential files in these database:

Note that most of these files do not have the format that suits the CAC simulation.

### LJ potential

The LJ formulation for potential energy is

where \epsilon and \sigma are two parameters. In the PyCAC code, the interatomic force, not the energy, is shifted such that the force goes continuously to zero at the cut-off distance r_\mathrm{c}, i.e., if r < r_\mathrm{c}, f = f(r) - f(r_\mathrm{c}); otherwise, f = 0.

In `lj.para`

, a blank line or a line with the "#" character in column one (a comment line) is ignored; three positive real numbers (\epsilon, \sigma, and r_\mathrm{c}) and one non-negative real number (r_0) are given in any sequence, where r_0 is a place holder that should always be 0.0 for the LJ potential. Note that for the EAM potential, r_0 equals the minimum interatomic distance, i.e., the smaller `first_val`

given in `pair.tab`

and `edens.tab`

.

For example, `potentials/lj/Cu/kluge/lj.para`

reads

```
# parameters for the LJ potential
epsilon 0.167
sigma 2.315
rcmin 0.
rcoff 5.38784
```

where `epsilon`

= \epsilon, `sigma`

= \sigma, `rcmin`

= r_0, and `rcoff`

= r_\mathrm{c}.

### Other files

When `boolean_restart`

= *t*, a `cac_in.restart`

file needs to be provided. This file is renamed from one of the `cac_out_#.restart`

files, where `#`

is a positive integer.

When `restart_group_number`

> 0, or `boolean_restart_refine`

= *t* and `refine_style`

= *group*, one or more `group_in_*.id`

files need to be provided, where `*`

is a positive integer. These files are renamed from `group_out_*_#.id`

files, which are created automatically when the total number of groups > 0. Note that if the `#`

here does not match that in the `cac_out_#.restart`

file, the information of the restart group may be incorrect.

When `modify_number`

> 0 and at least one of the `modify_style`

= *add_atom*, one or more LAMMPS data files `lmp_*.dat`

need to be provided, where `*`

is the id of the current modify command in `cac.in`

.