12.3.1.4.3. ewaldbackgroundΒΆ

Converges the self-consistent polarized background of a periodic system with Ewald summation The following table contains the defaults input options for the calculator, The default OPTIONAL means this option is switched off, if no input is given. REQUIRED arguments have to be specified, otherwise an error is thrown.

Property Name

Default Value

Description

Valid Input

multipoles

system.xml

MPS mapping file (xml) of the system

output.checkpoint

ewaldbackground.hdf5

HDF5 file for the converged background (used by qmmm Ewald
regions)

coulombmethod.alpha

OPTIONAL

Ewald splitting parameter; derived from the box and screening_factor
if not set

float+

coulombmethod.k_max

OPTIONAL

Reciprocal-space cutoff; derived from alpha and screening_factor
if not set

float+

coulombmethod.shape

cube

Shape of the macroscopic sample (surface term): cube or
sphere (same term), or slab
cube
sphere
slab

realspace.r_min

1.0

Real-space sum over periodic images: minimal shell radius
before convergence may be declared

float+

realspace.field_tol

1e-8

Real-space sum: convergence threshold on the field contribution
of one shell (atomic units)

float+

realspace.screening_factor

6.0

Accuracy of the Ewald sums (dimensionless): sets the real-space
cutoff screening_factor/alpha and a reciprocal cutoff
with the same error (erfc(screening_factor)); lower it
only deliberately

float+

polarmethod.thole_a

0.39

Thole damping parameter

float+

polarmethod.induce

true

Compute induced dipoles (false: static multipoles only)

bool

polarmethod.max_iter

200

Maximal number of iterations of the induction solver

int+

polarmethod.tolerance

1e-8

Convergence tolerance of the induction solver

float+

polarmethod.use_jor

false

Use Jacobi over-relaxation (as the legacy code) instead
of preconditioned conjugate gradients

bool

polarmethod.jor_omega

0.35

Relaxation parameter for use_jor

float+

polarmethod.use_block_jacobi_preconditioner

false

Block-Jacobi (per segment) preconditioner for the conjugate
gradient solver

bool

polarmethod.debug_match_legacy_first_step

false

Debugging: replicate the two-phase first step of the legacy
code

bool

Note

An xml file containing the defaults for the ewaldbackground calculator can be created via -p ewaldbackground -o FILENAME command line options `