JDFTx  1.7.0
Wannier Class Reference

Compute Maximally-Localized Wannier Functions. More...

#include <Wannier.h>

Classes

struct  AtomicOrbital
 
struct  TrialOrbital
 

Public Types

enum  LocalizationMeasure { LM_FiniteDifference , LM_RealSpace }
 
enum  SpinMode { SpinUp , SpinDn , SpinAll }
 
enum  FilenameType { FilenameInit , FilenameDump }
 

Public Member Functions

void setup (const Everything &everything)
 
void saveMLWF ()
 Output the Maximally-Localized Wannier Functions from current wavefunctions.
 
string getFilename (FilenameType fnType, string varName, int *spin=0) const
 

Public Attributes

std::vector< TrialOrbitaltrialOrbitals
 group of centers
 
bool needAtomicOrbitals
 
enum Wannier::LocalizationMeasure localizationMeasure
 
bool precond
 
int bStart
 
double eOuterMin
 
double eOuterMax
 outer energy window (outside which bands do not contribute)
 
double eInnerMin
 
double eInnerMax
 inner energy window (within which all bands used)
 
bool outerWindow
 
bool innerWindow
 denotes which windows are available
 
int nFrozen
 
string frozenUfilename
 number of frozen centers, and the filename to read their rotations from
 
int nCenters
 total number of centers, those being optimized and frozen
 
bool saveWfns
 whether to write wavefunctions
 
bool saveWfnsRealSpace
 whether to output Wannier functions band-by-band in real-space
 
bool saveMomenta
 whether to output momentum matrix elements
 
bool saveSpin
 whether to output spin matrix elements (non-collinear only)
 
bool saveRP
 whether to output R*P matrix elements for Wannierized angular momentum / electric quadrupole evaluation
 
string zVfilename
 filename for reading Vscloc with an applied electric field for z matrix element output
 
double zFieldMag
 magnitude of electric field difference (Eh/a0) between current calculation and the specified Vscloc
 
double z0
 
double zH
 
double zSigma
 center (lattice coords), half-width (lattice coords) and smoothness (bohrs) for slab-weight function
 
bool loadRotations
 whether to load initial rotations from previous dump
 
string initFilename
 
string dumpFilename
 filename patterns for input and output
 
string eigsFilename
 optional override for eigenvals file
 
string numericalOrbitalsFilename
 filename for reading numerical orbitals
 
vector3 numericalOrbitalsOffset
 lattice coordinates of the origin in the input
 
vector3< int > phononSup
 phonon supercell (process e-ph matrix elements on this supercell if non-zero)
 
double rSmooth
 supercell boundary width over which matrix elements are smoothed
 
enum Wannier::SpinMode spinMode
 which spin(s) to generate Wannier functions for
 
std::vector< int > iSpinArr
 set of spin indices corresponding to spinMode
 
bool polar
 whether to subtract long-range contributions in e-ph matrix elements
 
std::vector< DefectSupercelldefects
 List of defect supercells to compute Wannierized matrix elements for.
 

Friends

class WannierMinimizer
 
class DefectSupercell
 
struct CommandWannierMinimize
 

Detailed Description

Compute Maximally-Localized Wannier Functions.

Member Function Documentation

◆ getFilename()

string Wannier::getFilename ( FilenameType  fnType,
string  varName,
int *  spin = 0 
) const

Get filename for varName, based on initFilename, dumpFilename or numericalOrbitalsFilename depending on fnType Optionally include Up/Dn suffix if spin is non-null and calculation is polarized


The documentation for this class was generated from the following file: