GW2X

References: YShigeta2001

Specifications for the GW2X calculation of core ionization potentials. Note that donor states need to be actively localized using the LOCALIZE keyword in DONOR_STATES. N_SEARCH should be kept to the minimum, such that only core states are localized. [Edit on GitHub]

Keywords

SECTION_PARAMETERS
BATCH_SIZE
C_OS
C_SS
EPS_GW2X
MAX_GW2X_ITER
PSEUDO_CANONICAL
XPS_ONLY

Keyword descriptions

SECTION_PARAMETERS: logical = F 

Lone keyword: T

Usage: &GW2X {logical}

Enables the GW2X correction of the core ionization potentials [Edit on GitHub]

BATCH_SIZE: integer = 64 : MOs batch size for batched tensor contraction. Larger size is faster, but uses more memory. Default should be safe in most cases. [Edit on GitHub]

C_OS: real = 1.00000000E+000 : Opposite-spin scling factor. SCS => 6/5, SOS => 1.3 [Edit on GitHub]

C_SS: real = 1.00000000E+000 : Same-spin scling factor. SCS => 1/3, SOS => 0.0 [Edit on GitHub]

EPS_GW2X: real = 1.00000000E-002 : Convergence threshold for GW2X iterations (in eV) [Edit on GitHub]

MAX_GW2X_ITER: integer = 10 : Maximum number of iterations for GW2X [Edit on GitHub]

PSEUDO_CANONICAL: logical = T 

Aliases: PSEUDO_CANO

Whether the pseudo-canonical version of GW2X should be used (versus only using the diagonal of the generalized Fock matrix) [Edit on GitHub]

XPS_ONLY: logical = F 

Lone keyword: T

If set to .TRUE., only run GW2X calculations for XPS spectroscopy and ignore all XAS calculations. It is still required to define the DONOR_STATES and KERNEL%EXACT_EXCHANGE subsections. [Edit on GitHub]