BerkeleyGW manual (version 1.2.0)
Contents
 Preliminaries
 README
 CHANGELOG
 license.txt
 CONTRIBUTORS
 LITERATURE.html
 config/README
 testsuite/README
 library/README
 examples/README
 MeanField
 MeanField/README
 ESPRESSO
 MeanField/ESPRESSO/README
 MeanField/ESPRESSO/kgrid.inp
 MeanField/ESPRESSO/version5.1/INPUT_pw2bgw.html
 MeanField/ESPRESSO/version5.1/INPUT_bgw2pw.html
 MeanField/ESPRESSO/version5.0/INPUT_pw2bgw.html
 MeanField/ESPRESSO/version5.0/INPUT_bgw2pw.html
 MeanField/ESPRESSO/version4.3.2/pw2bgw.inp
 MeanField/ESPRESSO/version4.3.2/bgw2pw.inp
 MeanField/ESPRESSO/version4.3.2/README_patch
 MeanField/ESPRESSO/patch_oldversions/README
 PARATEC
 MeanField/PARATEC/README
 SIESTA
 MeanField/SIESTA/README
 MeanField/SIESTA/patch/README
 MeanField/SIESTA/siesta2bgw.inp
 Octopus
 MeanField/Octopus/README
 EPM
 MeanField/EPM/README
 MeanField/EPM/epm.inp
 ICM
 MeanField/ICM/README
 Utilities
 MeanField/Utilities/README
 MeanField/Utilities/wfnmerge.inp
 MeanField/Utilities/fix_occ.inp
 MeanField/Utilities/wfnmix_QSGW.inp
 SAPO
 MeanField/SAPO/README
 MeanField/SAPO/sapo.inp
 Epsilon
 Epsilon/README
 Epsilon/epsilon.inp
 Epsilon/epsconv.inp
 Epsilon/epsomega.inp
 Epsilon/epsinvomega.inp
 Epsilon/epsmat_merge.inp
 Epsilon/epsmat_old2hdf5.inp
 Epsilon/epsmat_intp.inp
 Sigma
 Sigma/README
 Sigma/sigma.inp
 Sigma/sig2wan.inp
 BSE
 BSE/README_kernel
 BSE/kernel.inp
 BSE/README_absorption
 BSE/absorption.inp
 BSE/inteqp.inp
 BSE/README_inteqp
 BSE/summarize_eigenvectors.inp
 PlotXct
 PlotXct/README
 PlotXct/plotxct.inp
 NonLinearOptics
 NonLinearOptics/README
 NonLinearOptics/nonlinearoptics.inp
 Visual
 Visual/README
 Visual/surface.inp
 Visual/gsphere.inp
 Developers
 doxygen/README
 Common/README
 Common/qhull/README
 Common/voro++/README.BerkeleyGW
 MeanField/spglib1.0.9/README
About
This manual is assembled automatically from the various documentation files in the BerkeleyGW distribution.
The headings are the paths and filenames. For each executable, consult the corresponding .inp file for information about input parameters.
The homepage is http://www.berkeleygw.org.
Preliminaries
README
*** Copyright Notice ***
BerkeleyGW, Copyright (c) 2011, The Regents of the University of
California, through Lawrence Berkeley National Laboratory (subject to
receipt of any required approvals from the U.S. Dept. of Energy).
All rights reserved.
If you have questions about your rights to use or distribute this
software, please contact Berkeley Lab's Technology Transfer Department
at TTD@lbl.gov.
NOTICE. This software was developed under partial funding from the
U.S. Department of Energy. As such, the U.S. Government has been
granted for itself and others acting on its behalf a paidup,
nonexclusive, irrevocable, worldwide license in the Software to
reproduce, prepare derivative works, and perform publicly and display
publicly. Beginning five (5) years after the date permission to
assert copyright is obtained from the U.S. Department of Energy, and
subject to any subsequent five (5) year renewals, the U.S. Government
is granted for itself and others acting on its behalf a paidup,
nonexclusive, irrevocable, worldwide license in the Software to
reproduce, prepare derivative works, distribute copies to the public,
perform publicly and display publicly, and to permit others to do so.
For license information see "license.txt"

BerkeleyGW  The Berkeley GW (And More) Package
http://www.berkeleygw.org
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014) 
Version 1.0 (Aug, 2011)
J. Deslippe, G. Samsonidze, D. A. Strubbe, M. Jain
Version 0.5 (July, 2008)
J. Deslippe, G. Samsonidze, L. Yang, F. Ribeiro
Version 0.2 M. L. Tiago (2001) Original BSE Code, PlotXct
S. IsmailBeigi (2002) FFT, Documentation, General Improvements
C. Spataru (2005) Cylindrical Coulomb Truncation, Transform (obsolete)
Version 0.1 X. Blase (1998) Implemented F90, Preprocessing, Dynamical Memory Allocation
G. M. Rignanese " "
E. Chang " "
Version 0.0 M. Hybertsen (1985) Original GW Code
Report current bugs or problems and ask questions at the BerkeleyGW Forum,
at http://www.berkeleygw.org.

See more details in the README and .inp files in individual directories.

Compilation

We have tested the code extensively with various configurations, and support the following compilers and libraries:
* Operating systems: Linux, AIX, MacOS
* Fortran compilers (required): pgf90, ifort, gfortran, g95, openf90, sunf90, pathf90, crayftn, af90 (Absoft), nagfor, xlf90 (experimental)
* C compilers (required): pgcc, icc, gcc, opencc, pathcc, craycc, clang
* C++ compilers (required): pgCC, icpc, g++, openCC, pathCC, crayCC, clang++
* MPI implementation (optional): OpenMPI, MPICH1, MPICH2, MVAPICH2, Intel MPI
* LAPACK/BLAS implementation (required): NetLib, ATLAS, Intel MKL, ACML, Cray LibSci
* ScaLAPACK/BLACS implementation (required if MPI is used): NetLib, Cray LibSci, Intel MKL, AMD
* FFTW (required): versions 3.3.x
1. Architecturespecific Makefileinclude files appropriate for various supercomputers as well as
for using standard Ubuntu or Macports packages are provided in the config directory. Copy or
link one to the top directory. Example:
cp config/stampede.tacc.utexas.edu_threaded_hdf5.mk arch.mk
2. Edit it to fit your needs. Refer to config/README for details.
3. Copy flavor_real.mk or flavor_cplx.mk to flavor.mk to set flavor.
Complex may always be used. Real may be used for systems that
have both inversion (about the origin) and timereversal symmetry, and will be
faster and use less memory.
4. Stay in the root directory and run the following commands
to compile the various codes:
MAIN:
make epsilon
make sigma
make bse
MISC:
make plotxct
make nonlinearoptics
make meanfield
make visual
EVERYTHING:
make all All codes/packages (for selected flavor)
make j all parallel build (for selected flavor)
make allflavors everything, for both flavors
make j allflavors parallel build of everything, for both flavors
make install INSTDIR= install binaries, library, examples, testsuite into specified prefix
These commands will generate executables with extension .x in
the source directory and symbolic links with the same name in
the bin directory.
5. Test your build. See testsuite/README for more info.
Development

Spirit of contributions:
0. Follow guidelines at https://civet.berkeley.edu/BerkeleyGW/wiki/CodingStandards
1. These codes are to be used on various machine architectures
employing various compilers and libraries. Therefore, try not
to write machinespecific code or quick fixes that will have to
be reversed later.
2. Limit compilation changes as much as possible to arch.mk and ensure
that this is suitable for all codes in Epsilon/, Sigma/, and BSE/.
Hopefully individual Makefiles will only require the addition of
new sourcecode files.
3. Test the codes using the examples/ and please augment the examples/
with your own test cases (provided they are not too computationally
expensive).
CHANGELOG
Berkeley 1.2 (Aug/2016)
=======================
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Features marked with [*] change the default behavior of the code relative to
BGW1.1 and may cause a small change in the numbers produced by the code.
New features

1) Added new methods to deal with the frequencydependence of the polarizability:
1.1) Added GodbyNeeds (GN) plasmonpole model.
1.2) Added spectral method for realaxis (RA) fullfrequency (FF) calculations.
The spectral method allows one to compute the dielectric matrix much faster for
many frequency points on the real axis. The old method for RAFF calculations
is now refered to as the AdlerWiser method. Still, RA is no longer the default
FF method (see next item).
1.3) Added Contourdeformation (CD) fullfrequency (FF) formalism.
The CD is the recommended (and default) scheme for calculations performing an
explicit evaluation of the dynamical effects of the dielectric matrix without
plasmonpole models. It requires the evaluation of the dielectric matrix on
both real and imaginary frequencies, but with typically a much smaller number
of frequencies. The previous formalism is now refered to as realaxis (RA)
fullfrequency (FF) formalism. CD is now the default method for FF calculations.
2) Released nonlinear optics postprocessing utility.
This codes allows one to compute the interexciton transitions in nonlinearoptics
experiments.
3) Improved usability in the code output and added dynamic verbosity switch.
BerkeleyGW has now a much simpler and neater output. The code gives time estimates
for most tasks it performs  at least for those that are more timeconsuming.
There is also a runtime flag supported by all codes to switch the amount of
verbosity the code produces. The old compilation flag "DVERBOSE" is now
deprecated.
4) Improved usability in the code input.
Several input parameters, such as band occupations, are automatically detected.
If not set, parameters such as the cutoff for the bare and screened Coulomb
potential are also read from the wave function and dielectric matrix,
respectively.
5) Automatic solution of Dyson's equation in FF calculations.
BerkeleyGW now automatically solves Dyson's equation for FF calculations. It uses
a varying number of frequencies to find the graphical solution for the
quasiparticle energies, and perform extrapolation (with appropriate warning
messages) if no intersection could be found. The code will also output the files
"eqp0.dat" and "eqp1.dat" directly, containing the offshell and onshell
quasiparticle energies.
6) Added ABINIT wrapper.
It is possible to interface the ABINIT code with BerkeleyGW. For now, the
wrapper can only output complexvalued wavefunctions, even for systems with
inversion symmetry.
Improvements

1) Added support for XeonPhi Knight's Landing (KNL)based systems.
2) Improved OpenMP scaling throughout.
Using OpenMP is now recommended for largescale computations.
3) Added new and more performant HDF5 file formats for epsmat and bsemat matrices.
This is recommend as default for all builds. Note that files are generally not
compatible with BerkeleyGW 1.1 and earlier. Binary epsmat files can be
converted to the HDF5 format with the epsmat_old2hdf5 utility. Older epsmat.h5
files from BerkeleyGW1.1beta2 can be converted to the current format with the
epsmat_hdf5_upgrade utility.
4) Added parallelization over frequencies in the epsilon code.
This allows for better scalability in fullfrequency calculations. This
scheme is supported in the old AdlerWiser and in the new ContourDeformation
formalisms.
5) Improved the MonteCarlo average scheme used to compute the Coulomb potential. [*]
For bulk systems, the new default scheme is to compute the average of Coulomb
potential v(q+G) for all qpoints and Gvectors. We use a hybrid scheme to make
this evaluation faster. While the code is slightly slower for small calculations,
results should converge much faster with kpoint sampling. One can still use the
old defaults with the input option "cell_average_cutoff 1.0d12". Note that
no change was performed for 2D, 1D and 0D systems, or 3D metals.
6) Rewrote kpoint interpolation engine. [*]
The previous kpoint interpolation scheme used in the haydock, absorption, and
inteqp codes used a greedy algorithm to search for the closest coarsegrid points
around each finegrid points, which lead to discontinuities of the interpolands.
The new interpolation engine uses QHull, which is bundled with BerkeleyGW, to
perform a Delaunay tessellation of the coarsegrid kpoints. The new interpolation
scheme is much more robust. You can use the old interpolation scheme with the
"greedy_interpolation" flag.
7) Generalized scheme for kpoint interpolation.
The kpoint interpolation is used in the haydock, absorption, and inteqp codes.
The new scheme is useful for metals and systems with valenceconduction band
character mixing in the BSE.
Misc

1) Changed logic to deal with invalid GPP frequency modes. [*]
When the code performs a HLGPP or a GNGPP calculation and finds an invalid
mode with frequency with \omega_{G,G`}^2 < 0, the code will now treat that mode
within the static COHSEX approximation (i.e., move frequency to infinity).
The previous behavior was to "find" a purely complex mode frequency and relax
the causality constraint. One can switch which strategy to use with the
"invalid_gpp_mode" input flag.
2) Many bug fixes and performance improvement.
license.txt
Unless otherwise stated, all files distributed in this package
are licensed under the following terms.
BerkeleyGW, Copyright (c) 2011, The Regents of the University of
California, through Lawrence Berkeley National Laboratory (subject to
receipt of any required approvals from the U.S. Dept. of Energy).
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
(1) Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
(2) Redistributions in binary form must reproduce the above copyright
notice, this list of conditions and the following disclaimer in the
documentation and/or other materials provided with the distribution.
(3) Neither the name of the University of California, Lawrence
Berkeley National Laboratory, U.S. Dept. of Energy nor the names of
its contributors may be used to endorse or promote products derived
from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
You are under no obligation whatsoever to provide any bug fixes,
patches, or upgrades to the features, functionality or performance of
the source code ("Enhancements") to anyone; however, if you choose to
make your Enhancements available either publicly, or directly to
Lawrence Berkeley National Laboratory, without imposing a separate
written license agreement for such Enhancements, then you hereby grant
the following license: a nonexclusive, royaltyfree perpetual
license to install, use, modify, prepare derivative works, incorporate
into other computer software, distribute, and sublicense such
enhancements or derivative works thereof, in binary and source code
form.
CONTRIBUTORS
Brad A. Barker (BAB)
Xavier Blase (XAV)
Andrew Canning (AC)
Eric K. Chang (EKC)
Sangkook Choi (SKC)
Jack R. Deslippe (JRD)
Peter W. Doak (PWD)
Mark S. Hybertsen (MSH)
Sohrab IsmailBeigi (SIB)
Manish Jain (MJ)
Felipe H. da Jornada (FHJ)
JeLuen Li (JLL)
Andrei S. Malashevich (ASM)
Brad D. Malone (BDM)
Jamal I. Mustafa (JIM)
Jeffrey B. Neaton (JBN)
CheolHwan Park (CHP)
David G. Prendergast (DGP)
Filipe J. Ribeiro (FJR)
GianMarco Rignanese (GMR)
Georgy Samsonidze (GSM)
Sahar Sharifzadeh (SS)
Catalin D. Spataru (CDS)
David A. Strubbe (DAS)
Murilo L. Tiago (MLT)
Derek VigilFowler (DVF)
Li Yang (LY)
Oleg V. Yazyev (OVY)
Peihong Zhang (PZ)
Tonatiuh Rangel (TR)
Gabriel K. Antonius (GKA)
LITERATURE.html
Please let us know when your paper is published with BerkeleyGW for inclusion here (or any other additions or corrections) in the Forum, under the "Literature" topic.
I. Papers on the implementation of BerkeleyGW
 Jack Deslippe, Georgy Samsonidze, David A. Strubbe, Manish Jain, Marvin L. Cohen, and Steven G. Louie, "BerkeleyGW: A Massively Parallel Computer Package for the Calculation of the Quasiparticle and Optical Properties of Materials and Nanostructures," Comput. Phys. Commun. 183, 1269 (2012) (most uptodate on arXiV)
 Mark S. Hybertsen and Steven G. Louie, "Electron correlation in semiconductors and insulators: Band gaps and quasiparticle energies," Phys. Rev. B 34, 5390 (1986) [GW, GPP, COHSEX]
Errata: Eq. 11 should have E' instead of E in the numerator. Eq. 32 should have Ω^{2} rather than Ω in the numerator. Eq. 34a should have δ_{G,G} instead of 1 in the parentheses.
 Sheng Bai Zhang, David Tománek, Marvin L. Cohen, Steven G. Louie, and Mark S. Hybertsen, "Evaluation of quasiparticle energies for semiconductors without inversion symmetry," Phys. Rev. B 40, 3162 (1989) [GPP for systems without inversion symmetry]
Note: ε_{1} and ε_{2} in this paper refer to the Hermitian and Antihermitian components of the dielectric matrix, respectively, and not to the real and imaginary parts of ε. See D. J. Johnson, Phys. Rev. B 9, 4475 (1974).
 Michael Rohlfing and Steven G. Louie, "Electronhole excitations and optical spectra from first principles," Phys. Rev. B 62, 4927 (2000) [BSE]
Errata: Eqs. 26 and 27 should have 8π^{2} instead of 16π in the prefactor. Eq. 44 should be a sum over G ≠ 0. Eq. 45 should have ε^{1}_{G,G'} instead of ε^{1}_{G,0}.
 JeLuen Li, GianMarco Rignanese, Eric K. Chang, Xavier Blase, and Steven G. Louie, "GW study of the metalinsulator transition of bcc hydrogen," Phys. Rev. B 66, 035102 (2002) [spinpolarized GW]
 Sohrab IsmailBeigi, "Truncation of periodic image interactions for confined systems," Phys. Rev. B 73, 233103 (2006) [slab and wire truncation]
 Jeffrey B. Neaton, Mark S. Hybertsen, and Steven G. Louie, "Renormalization of Molecular Electronic Levels at MetalMolecule Interfaces," Phys. Rev. Lett. 97, 216405 (2006) [ICM]
Erratum: p. 3, left col, last paragraph. Should be 1/2zz_{0} instead of 1/4zz_{0}.
 R. Haydock, "The recursive solution of the Schrödinger equation," Comput. Phys. Commun. 20, 11 (1980)
 Loren X. Benedict and Eric L. Shirley, "Ab initio calculation of ε_{2}(ω) including the electronhole interaction: Application to GaN and CaF_{2}," Phys. Rev. B 59, 5441 (1999) [Haydock recursion in BSE]
 Georgy Samsonidze, Manish Jain, Jack Deslippe, Marvin L. Cohen, and Steven G. Louie, "Simple approximate physical orbitals for GW quasiparticle calculations," Phys. Rev. Lett. 107, 186404 (2011) [SAPO]
II. Review articles on the GW approximation and BetheSalpeter equation approaches
 Lars Hedin and Stig Lundqvist, "Effects of ElectronElectron and ElectronPhonon Interactions on the OneElectron States of Solids," Solid State Phys. 23, 1 (1970)
 Mark S. Hybertsen and Steven G. Louie, "Theory and Calculation of Quasiparticle Energies and Band Gaps," Comments on Cond. Mat. Phys. 13, 223 (1987)
 G. Strinati, "Application of the Green's functions method to the study of the optical properties of semiconductors," Riv. Nuovo Cimento 11, 1 (1988)
 Steven G. Louie, "Quasiparticle Theory of Electron Excitations in Solids," in Quantum Theory of Real Materials, eds. James R. Chelikowsky and Steven G. Louie, (Kluwer Press, Boston, 1996), p. 83
 Steven G. Louie, "FirstPrinciples Theory of Electron Excitation Energies in Solids, Surfaces, and Defects," in Topics in Computational Materials Science, ed. ChingYao Fong (World Scientific, Singapore, 1998) p. 96
 F. Aryasetiawan and O. Gunnarsson, "The GW method," Rep. Prog. Phys. 61, 237 (1998)
 Lars Hedin, "On correlation effects in electron spectroscopies and the GW approximation," J. Phys.: Condens. Matter 11, R489 (1999)
 Wilfried G. Aulbur, Lars Jönsson, and John W. Wilkins, "Quasiparticle calculations in solids," Solid State Phys. 54, 1 (1999)
 Giovanni Onida, Lucia Reining, and Angel Rubio, "Electronic excitations: densityfunctional versus manybody Green'sfunction approaches," Rev. Mod. Phys. 74, 601 (2002)
 Steven G. Louie, "Predicting Materials and Properties: Theory of the Ground and Excited State," in Conceptual Foundations of Materials: A Standard Model for Ground and ExcitedState Properties, vol. eds. Steven G. Louie and Marvin L. Cohen (Elsevier, Amsterdam, 2006) p. 9
III. Papers using BerkeleyGW
 Mark S. Hybertsen and Steven G. Louie, "First Principles Theory of Quasiparticles: Calculation of Band Gaps in Semiconductors and Insulators," Phys. Rev. Lett. 55, 1418 (1985)
 Mark S. Hybertsen and Steven G. Louie, "Electron Correlation and the Band Gap in Ionic Crystals," Phys. Rev. B 32, 7005(R) (1985) [Erratum: Phys. Rev. B 35, 9308 (1987)]
 Mark S. Hybertsen and Steven G. Louie, "Manybody Calculation of Surface States: As on Ge(111)," Phys. Rev. Lett. 58, 1551 (1987)
 John E. Northrup, Mark S. Hybertsen, and Steven G. Louie, "Theory of Quasiparticle Energies in Alkali Metals," Phys. Rev. Lett. 59, 819 (1987)
 Steven G. Louie and Mark S. Hybertsen, "Theory of Quasiparticle Energies: Band Gaps and Excitation Spectra in Solids," Int. J. Quant. Chem.: Quant. Chem. Symp. 21, 31 (1987)
 Sheng Bai Zhang, David Tománek, Steven G. Louie, Marvin L. Cohen, and Mark S. Hybertsen, "Quasiparticle Calculation of Valence Band Offset of AlAsGaAs(001)," Solid State Comm. 66, 585 (1988)
 Mark S. Hybertsen and Steven G. Louie, "Theory of Quasiparticle Surface States in Semiconductor Surfaces," Phys. Rev. B 38, 4033 (1988)
 Michael P. Surh, John E. Northrup, and Steven G. Louie, "Occupied Quasiparticle Bandwidth of Potassium," Phys. Rev. B 38, 5976 (1988)
 Xuejun Zhu, Stephen Fahy, and Steven G. Louie, "Ab Initio Calculation of Pressure Coefficients of Band Gaps of Silicon: Comparison of the LocalDensity Approximation and Quasiparticle Results," Phys. Rev. B 39, 7840 (1989) [Erratum: Phys. Rev. B 40, 5821 (1989)]
 John E. Northrup, Mark S. Hybertsen, and Steven G. Louie, "Quasiparticle excitation spectrum for nearlyfreeelectron metals," Phys. Rev. B 39, 8198 (1989)
 Sheng Bai Zhang, David Tománek, Marvin L. Cohen, Steven G. Louie, and Mark S. Hybertsen, "Evaluation of Quasiparticle Energies for Semiconductors without Inversion Symmetry," Phys. Rev. B 40, 3162 (1989)
 Sheng Bai Zhang, Mark S. Hybertsen, Marvin L. Cohen, Steven G. Louie, and David Tománek, "Quasiparticle Band Gaps for Ultrathin GaAs/AlAs(001) Superlattices," Phys. Rev. Lett. 63, 1495 (1989)
 Xuejun Zhu, Sheng Bai Zhang, Steven G. Louie, and Marvin L. Cohen, "Quasiparticle Interpretation of Photoemission Spectra and Optical Properties of GaAs(110)," Phys. Rev. Lett. 63, 2112 (1989)
 Sheng Bai Zhang, Marvin L. Cohen, Steven G. Louie, David Tománek, and Mark S. Hybertsen, "Quasiparticle Band Offset at the (001) Interface and Band Gaps in Ultrathin Superlattices of GaAsAlAs Heterojunctions," Phys. Rev. B 41, 10058 (1990)
 Hélio Chacham, Xuejun Zhu, and Steven G. Louie, "MetalInsulator Transition in Solid Xenon at High Pressures," Europhys. Lett. 14, 65 (1991)
 Hélio Chacham and Steven G. Louie, "Metallization of Solid Hydrogen at Megabar Pressures: A FirstPrinciples Quasiparticle Study," Phys. Rev. Lett. 66, 64 (1991)
 John E. Northrup, Mark S. Hybertsen, and Steven G. Louie, "ManyBody Calculation of the Surface State Energies for Si(111)2×1," Phys. Rev. Lett. 66, 500 (1991)
 Michael P. Surh, Steven G. Louie, and Marvin L. Cohen, "Quasiparticle Energies for Cubic BN, BP, and BAs," Phys. Rev. B 43, 9126 (1991)
 Xuejun Zhu and Steven G. Louie, "Quasiparticle Surface Band Structure and Photoelectric Threshold of Ge(111)2×1," Phys. Rev. B 43, 12146(R) (1991)
 Xuejun Zhu and Steven G. Louie, "Quasiparticle Band Structure of Thirteen Semiconductors and Insulators," Phys. Rev. B 43, 14142 (1991)
 Michael P. Surh, Steven G. Louie, and Marvin L. Cohen, "Band Gaps of Diamond under Anisotropic Stress," Phys. Rev. B 45, 8239 (1992)
 Hélio Chacham, Xuejun Zhu, and Steven G. Louie, "PressureInduced InsulatorMetal Transitions in Solid Xenon and Hydrogen: A FirstPrinciples Quasiparticle Study," Phys. Rev. B 46, 6688 (1992) [Erratum: Phys. Rev. B 48, 2025(E) (1993)]
 Eric L. Shirley, Xuejun Zhu, and Steven G. Louie, "CorePolarization in Semiconductors: Effects on Quasiparticle Energies," Phys. Rev. Lett. 69, 2955 (1992)
 Eric L. Shirley and Steven G. Louie, "Electron Excitations in Solid C_{60}: Energy Gap, Band Dispersions, and Effects of Orientational Disorder," Phys. Rev. Lett. 71, 133 (1993)
 Angel Rubio, Jennifer L. Corkill, Marvin L. Cohen, Eric L. Shirley, and Steven G. Louie, "Quasiparticle Band Structure of AlN and GaN," Phys. Rev. B 48, 11810 (1993)
 Steven G. Louie and Eric L. Shirley, "Electron Excitation Energies in Fullerites: ManyElectron and Molecular Orientational Effects," J. Phys. Chem. Solids 54, 1767 (1993)
 Xavier Blase, Xuejun Zhu, and Steven G. Louie, "SelfEnergy Effects on the Surface State Energies of HSi(111) 1×1," Phys. Rev. B 49, 4973 (1994)
 J. A. Carlisle, L. J. Terminello, A. V. Hamza, E. A. Hudson, Eric L. Shirley, F. J. Himpsel, D. A. LapianoSmith, J. J. Jia, T. A. Callcott, R. C. C. Perera, D. K. Shuh, Steven G. Louie, J. Stöhr, M. G. Samant, and D. L. Ederer, "Occupied and Unoccupied Orbitals of C_{60} and C_{70}," Mol. Cryst. Liq. Cryst. 256, 819 (1994)
 Oleg Zakharov, Angel Rubio, Xavier Blase, Marvin L. Cohen, and Steven G. Louie, "Quasiparticle Band Structures of Six IIVI Compounds: ZnS, ZnSe, ZnTe, CdS, CdSe, and CdTe," Phys. Rev. B 50, 10780 (1994)
 Xavier Blase, Angel Rubio, Steven G. Louie, and Marvin L. Cohen, "Quasiparticle band structure of bulk hexagonal boron nitride and related systems," Phys. Rev. B 51, 6868 (1995)
 Michael P. Surh, Hélio Chacham, and Steven G. Louie, "Quasiparticle Excitation Energies for the FCenter Defect in LiCl," Phys. Rev. B. 51, 7464 (1995)
 Eric L. Shirley and Steven G. Louie, "Photoemission and Optical Properties of C_{60} Fullerites," in Quantum Theory of Real Materials, eds. J.R. Chelikowsky and Steven G. Louie (Kluwer Press, Boston, 1996), p. 515.
 Balázs Králik, Eric K. Chang, and Steven G. Louie, "Structural Properties and Quasiparticle Band Structure of Zirconia," Phys. Rev. B 57, 7027 (1998)
 Michael Rohlfing and Steven G. Louie, "Electronhole Excitations in Semiconductors and Insulators," Phys. Rev. Lett. 81, 2312 (1998)
 Michael Rohlfing and Steven G. Louie, "Optical Excitations in Conjugated Polymers," Phys. Rev. Lett. 82, 1959 (1999)
 Michael Rohlfing and Steven G. Louie, "Excitons and Optical Spectrum of the Si(111)(2×1) Surface," Phys. Rev. Lett. 83, 856 (1999)
 Eric K. Chang, Michael Rohlfing, and Steven G. Louie, "Excitons and Optical Properties of AlphaQuartz," Phys Rev. Lett. 85, 2613 (2000)
 Eric K. Chang, Michael Rohlfing, and Steven G. Louie, "FirstPrinciples Study of Optical Excitations in AlphaQuartz," in The Optical Properties of Materials, MRS Symp. Proceed. Vol. 579, eds. Eric L. Shirley, James R. Chelikowsky, Steven G. Louie, and Gérard Martinez (Materials Research Society, Warrendale, 2000), p. 3
 Peihong Zhang, Vincent H. Crespi, Eric K. Chang, Steven G. Louie, and Marvin L. Cohen, "Computational Design of DirectBandgap Semiconductors that LatticeMatch Silicon," Nature 409, 69 (2001)
 Jeffrey C. Grossman, Michael Rohlfing, Lubos Mitas, Steven G. Louie, and Marvin L. Cohen, "High Accuracy ManyBody Calculational Approaches for Excitations in Molecules," Phys. Rev. Lett. 86, 472 (2001)
 GianMarco Rignanese, Xavier Blase, and Steven G. Louie, "Quasiparticle Effects on Tunneling Currents: A Study of C2H4 Adsorbed on the Si(001)(2×1) Surface," Phys. Rev. Lett. 86, 2110 (2001)
 Eric K. Chang, Xavier Blase, and Steven G. Louie, "Quasiparticle Band Structure of Lanthanum Hydride," Phys. Rev. B 64, 155108 (2001)
 Catalin D. Spataru, M. A. Cazalilla, Angel Rubio, Loren X. Benedict, Pedro M. Echenique, and Steven G. Louie, "Anomalous Quasiparticle Lifetime in Graphite: Band Structure Effects," Phys. Rev. Lett. 87, 246405 (2001)
 JeLuen Li, GianMarco Rignanese, Eric K. Chang, Xavier Blase, and Steven G. Louie, "GW Study of the MetalInsulator Transition of bcc Hydrogen," Phys. Rev. B 66, 035102 (2002)
 Weidong Luo, Sohrab IsmailBeigi, Marvin L. Cohen, and Steven G. Louie, "Quasiparticle Band Structure of ZnS and ZnSe," Phys. Rev. B 66, 195215 (2002)
 Loren X. Benedict, Catalin D. Spataru, and Steven G. Louie, "Quasiparticle Properties of a Simple Metal at High Electron Temperatures," Phys. Rev. B 66, 085116 (2002)
 J. A. Alford, MeiYin Chou, Eric K. Chang, and Steven G. Louie, "Firstprinciples Studies of Quasiparticle Band Structures of Cubic YH_{3} and LaH_{3}," Phys. Rev. B 67, 125110 (2003)
 Murilo L. Tiago, John E. Northrup, and Steven G. Louie, "Ab initio calculation of the electronic and optical properties of solid pentacene," Phys. Rev. B 67, 11 5212 (2003)
 Sohrab IsmailBeigi and Steven G. Louie, "ExcitedState Forces within a FirstPrinciples Green's Function Formalism," Phys. Rev. Lett. 90, 076401 (2003)
 Catalin D. Spataru, Sohrab IsmailBeigi, Loren X. Benedict, and Steven G. Louie, "Excitonic effects and optical spectra of singlewalled carbon nanotubes," Phys. Rev. Lett. 92, 077402 (2004)
 Catalin D. Spataru, Sohrab IsmailBeigi, Loren X. Benedict, and Steven G. Louie, "Quasiparticle energies, excitonic effects and optical absorption spectra of smalldiameter singlewalled carbon nanotubes," Appl. Phys. A 78, 1129 (2004)
 Murilo L. Tiago, Sohrab IsmailBeigi, and Steven G. Louie, "Effect of Semicore Orbitals on the Electronic Band Gaps of Si, Ge, and GaAs Within the GW Approximation," Phys. Rev. B 69, 125212 (2004)
 Catalin D. Spataru, Loren X. Benedict, and Steven G. Louie, "Ab Initio Calculation of BandGap Renormalization in Highly Excited GaAs," Phys. Rev. B 69, 205204 (2004)
 Murilo L. Tiago, Michael Rohlfing, and Steven G. Louie, "Bound Excitons and Optical Properties of Bulk TransPolyacetylene," Phys. Rev. B 70, 193204 (2004)
 JeLuen Li, GianMarco Rignanese, and Steven G. Louie, "Quasiparticle Energy Bands of NiO in the GW Approximation," Phys. Rev B 71, 193102 (2005)
 Catalin D. Spataru, Sohrab IsmailBeigi, Loren X. Benedict, and Steven G. Louie, "Excitonic Effects and Optical Spectra of SingleWalled Carbon Nanotubes," 27th Conference on the Physics of Semiconductors, AIP Conference Proceedings 772, 1061 (2005)
 Jeffrey B. Neaton, Koonghong Khoo, Catalin D. Spataru, and Steven G. Louie, "Electronic Transport and Optical Properties of Carbon Nanostructures from First Principles," Comput. Phys. Commun. 169, 1 (2005)
 Sohrab IsmailBeigi and Steven G. Louie, "SelfTrapped Excitons in Silicon Dioxide: Mechanism and Properties," Phys. Rev. Lett. 95, 156401 (2005)
 Catalin D. Spataru, Sohrab IsmailBeigi, Rodrigo B. Capaz, and Steven G. Louie, "Theory and Ab Initio Calculation of Radiative Lifetime of Excitons in Semiconducting Carbon Nanotubes," Phys. Rev. Lett. 95, 247402 (2005)
 Steven G. Louie and Angel Rubio, "Quasiparticle and Optical Properties of Solids and Nanostructures: The GWBSE Approach," Handbook of Materials Modeling, ed. S. Yip (Springer, Dordrecht, The Netherlands, 2005), p. 215
 CheolHwan Park, Catalin D. Spataru, and Steven G. Louie, "Excitons and ManyElectron Effects in the Optical Response of SingleWalled Boron Nitride Nanotubes," Phys. Rev. Lett. 96, 126105 (2006)
 Jeffrey B. Neaton, Mark S. Hybertsen, and Steven G. Louie, "Renormalization of Molecular Electronic Levels at MetalMolecule Interfaces," Phys. Rev. Lett. 97, 216405 (2006)
 Su Ying Quek, Jeffrey B. Neaton, Mark S. Hybertsen, E. Kaxiras, and Steven G. Louie, "Firstprinciples Studies of the Electronic Structure of Cyclopentene on Si(001): Density Functional Theory and GW Calculations," Phys. Status Solidi (b) 243, 2048 (2006)
 Rodrigo B. Capaz, Catalin D. Spataru, Sohrab IsmailBeigi, and Steven G. Louie, "Diameter and Chirality Dependence of Exciton Properties in Carbon Nanotubes," Phys. Rev. B 74, 121401 (2006)
 Takashi Miyake, Peihong Zhang, Marvin L. Cohen, and Steven G. Louie, "Quasiparticle Energy of Semicore delectrons in ZnS: Combined LDA+U and GW approach," Phys. Rev. B 74, 245213 (2006)
 Li Yang, Catalin D. Spataru, Steven G. Louie, and MeiYin Chou, "Enhanced Electronhole Interaction and Optical Absorption in a Silicon Nanowire," Phys. Rev. B 75, 201304(R) (2007)
 Li Yang, CheolHwan Park, YoungWoo Son, Marvin L. Cohen, and Steven G. Louie, "Quasiparticle Energies and Band Gaps of Graphene Nanoribbons," Phys. Rev. Lett. 99, 186801 (2007)
 Li Yang, Marvin L. Cohen, and Steven G. Louie, "Excitonic Effects in the Optical Spectra of Graphene Nanoribbons," Nano Lett. 7, 3112 (2007)
 Feng Wang, David J. Cho, Brian Kessler, Jack Deslippe, P. James Schuck, Steven G. Louie, Alex Zettl, Tony F. Heinz, and Y. Ron Shen, "Observation of Excitons in OneDimensional Metallic SingleWalled Carbon Nanotubes," Phys. Rev. Lett. 99, 227401 (2007)
 Su Ying Quek, Jeffrey B. Neaton, Mark S. Hybertsen, Efthimios Kaxiras, and Steven G. Louie, "Negative Differential Resistance in Transport through Organic Molecules on Silicon," Phys. Rev. Lett. 98, 066807 (2007)
 Catalin D. Spataru, Sohrab IsmailBeigi, Rodrigo B. Capaz, and Steven G. Louie, "Quasiparticle and Excitonic Effects in the Optical Response of Nanotubes and Nanoribbons," in Carbon Nanotubes: Advanced Topics in the Synthesis, Structure, Properties and Applications, A. Jorio, M. S. Dresselhaus, G. Dresselhaus (eds.), Topics in Applied Physics 111 (SpringerVerlag, Heidelberg, Germany 2008), p. 195
 Jack Deslippe, Catalin D. Spataru, David Prendergast, and Steven G. Louie, "Bound excitons in metallic singlewalled carbon nanotubes," Nano Lett. 7, 1626 (2007)
 CheolHwan Park, Feliciano Giustino, Marvin L. Cohen, and Steven G. Louie, "Velocity Renormalization and Carrier Lifetime in Graphene from the ElectronPhonon Interaction," Phys. Rev. Lett. 99, 086804 (2007)
 Brad D. Malone, Jay D. Sau, and Marvin L. Cohen, "Ab initio survey of the electronic structure of tetrahedrally bonded phases of silicon," Phys. Rev. B 78, 035210 (2008)
 Jack Deslippe and Steven G. Louie, "Excitons and Manyelectron Effects in the Optical Response of Carbon Nanotubes and Other Onedimensional Nanostructures," Proc. SPIE 6892, 68920U1 (2008)
 Emmanouil Kioupakis, Peihong Zhang, Marvin L. Cohen, and Steven G. Louie, "GW Quasiparticle Corrections to the LDA+U/GGA+U Electronic Structure of bcc Hydrogen," Phys. Rev. B 77, 155114 (2008)
 Jay D. Sau, Jeffrey B. Neaton, Hyoung Joon Choi, Steven G. Louie, and Marvin L. Cohen, "Electronic Energy Levels of Weakly Coupled Nanostructures: C_{60} Metal Interfaces," Phys. Rev. Lett. 101, 026804 (2008)
 Li Yang, Marvin L. Cohen, and Steven G. Louie, "Magnetic Edgestate Excitons in Zigzag Graphene Nanoribbons," Phys. Rev. Lett. 101, 186401 (2008)
 CheolHwan Park, Feliciano Giustino, Jessica L. McChesney, Aaron Bostwick, Taisuke Ohta, Eli Rotenberg, Marvin L. Cohen, and Steven G. Louie, "Van Hove Singularity and Apparent Anisotropy in the Electronphonon Interaction in Graphene," Phys. Rev. B 77, 113410 (2008)
 CheolHwan Park, Feliciano Giustino, Marvin L. Cohen, and Steven G. Louie, "Electronphonon Interactions in Graphene, Bilayer Graphene, and Graphite," Nano Lett. 8, 4229 (2008)
 CheolHwan Park, Feliciano Giustino, Catalin D. Spataru, Marvin L. Cohen, and Steven G. Louie, "Firstprinciples Study of Electron Linewidths in Graphene," Phys. Rev. Lett. 102, 076803 (2009) [Erratum: Phys. Rev. Lett. 102, 189904(E) (2009)]
 Jack Deslippe, Mario Dipoppa, David Prendergast, M. V. O. Moutinho, Rodrigo B. Capaz, and Steven G. Louie, "Electronhole Interaction in Carbon Nanotubes: Novel Screening and Exciton Excitation Spectra," Nano Lett. 9, 1330 (2009)
 Li Yang, Jack Deslippe, CheolHwan Park, Marvin L. Cohen, and Steven G. Louie, "Excitonic effects on the optical response of graphene and bilayer graphene," Phys. Rev. Lett. 103, 186802 (2009)
 Chenggang Tao, J. Sun, X. Zhang, Ryan Yamachika, Daniel Wegner, Yasaman Bahri, Georgy Samsonidze, Marvin L. Cohen, Steven G. Louie, T. Don Tilley, Rachel A. Segalman, and Michael F. Crommie, "Spatial resolution of a type II heterojunction in a single bipolar molecule," Nano Lett. 9, 3963 (2009)
 CheolHwan Park, Feliciano Giustino, Catalin D. Spataru, Marvin L. Cohen, and Steven G. Louie, "Angleresolved Photoemission Spectra of Graphene from Firstprinciples," Nano Lett. 9, 4234 (2009)
 Emmanouil Kioupakis, Murilo L. Tiago, and Steven G. Louie, "Quasiparticle electronic structure of bismuth telluride in the GW approximation," Phys. Rev. B 82, 245203 (2010)
 Feliciano Giustino, Steven G. Louie, and Marvin L. Cohen, "Electronphonon renormalization of the direct band gap of diamond," Phys. Rev. Lett. 105, 265501 (2010)
 Brad D. Malone, Steven G. Louie, and Marvin L. Cohen, "Electronic and optical properties of bodycenteredtetragonal Si and Ge," Phys. Rev. B 81, 115201 (2010)
 BiChing Shih, Yu Xue, Peihong Zhang, Marvin L. Cohen, and Steven G. Louie, "Quasiparticle band gap of ZnO: High accuracy from the conventional G_{0}W_{0} approach," Phys. Rev. Lett. 105, 146401 (2010)
 Victor W. Brar, Sebastian Wickenburg, Melissa Panlasigui, CheolHwan Park, Tim O. Wehling, Yuanbo Zhang, R. Decker, Caglar Girit, A. V. Balatsky, Steven G. Louie, Alex Zettl, and Michael F. Crommie, "Observation of CarrierDensityDependent ManyBody Effects in Graphene via Tunneling Spectroscopy," Phys. Rev. Lett. 104, 036805 (2010)
 G. S. Do, J. Kim, SeungHoon Jhi, CheolHwan Park, Steven G. Louie, and Marvin L. Cohen, "Ab Initio Calculations of PressureInduced Structural Phase Transitions of GeTe," Phys. Rev. B 82, 054121 (2010)
 Li Yang, "Firstprinciples study of the optical absorption spectra of electrically gated bilayer graphene," Phys. Rev. B 81, 155445 (2010)
 David A. Siegel, CheolHwan Park, C. Hwang, Jack Deslippe, A. V. Federov, Steven G. Louie, and Alessandra Lanzara, "Manybody Interactions in Quasifreestanding Graphene," Proc. Natl. Acad. Sci. U.S.A. 108, 11365 (2011)
 H. C. Hsueh, G. Y. Guo, and Steven G. Louie, "Excitonic Effects in the Optical Properties of a SiC Sheet and Nanotubes," Phys. Rev. B 84, 85404 (2011)
 Georgy Samsonidze, Manish Jain, Jack Deslippe, Marvin L. Cohen, and Steven G. Louie, "Simple approximate physical orbitals for GW quasiparticle calculations," Phys. Rev. Lett. 107, 186404 (2011)
 Georgy Samsonidze, Marvin L. Cohen, and Steven G. Louie, "Compensationdoped silicon for photovoltaic applications," Phys. Rev. B 84, 195201 (2011)
 Manish Jain, James R. Chelikowsky, and Steven G. Louie, "Quasiparticle Excitations and Charge Transition Levels of Oxygen Vacancies in Hafnia," Phys. Rev. Lett. 107, 216803 (2011)
 Manish Jain, James R. Chelikowsky, and Steven G. Louie, "Reliability of Hybrid Functionals in Predicting Band Gaps," Phys. Rev. Lett. 107, 216806 (2011)
 Isaac Tamblyn, Pierre Darancet, Su Ying Quek, Stanimir A. Bonev, and Jeffrey B. Neaton, "Electronic energy level alignment at metalmolecule interfaces with a GW approach," Phys. Rev. B 84, 201402(R) (2011)
 Li Yang, "Excitons in intrinsic and bilayer graphene," Phys Rev. B 83, 085405 (2011)
 Li Yang, "Excitonic Effects on Optical Absorption Spectra of Doped Graphene," Nano Lett. 11, 3844 (2011)
 Oleg V. Yazyev, Emmanouil Kioupakis, Joel E. Moore, and Steven G. Louie, "Quasiparticle effects in the bulk and surfacestate bands of Bi_{2}Se_{3} and Bi_{2}Te_{3} topological insulators," Phys. Rev. B 85, 161101(R) (2012)
 Sahar Sharifzadeh, Ariel Biller, Leeor Kronik, and Jeffrey B. Neaton, "Quasiparticle and Optical Spectroscopy of the Organic Semiconductors Pentacene and PTCDA from First Principles," Phys. Rev. B 85, 125307 (2012)
 Jesse Noffsinger, Emmanouil Kioupakis, Chris G. Van de Walle, Steven G. Louie, and Marvin L. Cohen, "PhononAssisted Optical Absorption in Silicon from First Principles," Phys. Rev. Lett. 108, 167402 (2012)
 CheolHwan Park, Feliciano Giustino, Catalin D. Spataru, Marvin L. Cohen, and Steven G. Louie, "Inelastic Carrier Lifetime in Bilayer Graphene," Appl. Phys. Lett. 100, 032106 (2012)
 Kaihui Liu, Jack Deslippe, Fajun Xiao, Rodrigo B. Capaz, Xiaoping Hong, Shaul Aloni, Alex Zettl, Wenlong Wang, Xuedong Bai, Steven G. Louie, Enge Wang and Feng Wang, "A Periodic Table of Carbon Nanotube Optical Transitions," Nature Nanotech. 7, 325 (2012)
 Sangkook Choi, Jack Deslippe, Rodrigo B. Capaz, and Steven G. Louie, "An explicit formula for optical oscillator strength of excitons in semiconducting singlewalled carbon nanotubes: family behavior", Nano Lett. 13, 54 (2012)
 Sahar Sharifzadeh, Isaac Tamblyn, Peter Doak, Pierre Darancet, and Jeffrey B. Neaton, "Quantitative Molecular Orbital Energies within a G_{0}W_{0} Approximation", Eur. Phys. J. B 85, 323 (2012)
 Jack Deslippe, Georgy Samsonidze, Manish Jain, Marvin L. Cohen and Steven G. Louie, "Coulombhole summations and energies for GW calculations with limited number of empty orbitals: A modified static remainder approach", Phys. Rev. B 87, 165124 (2013)
 Brad D. Malone and Marvin L. Cohen, "Quasiparticle semiconductor band structures including spinorbit interactions," J. Phys.: Condens. Matter 25, 105503 (2013)
 Diana Y. Qiu, Felipe H. da Jornada and Steven G. Louie, "Optical Spectrum of MoS_{2}: ManyBody Effects and Diversity of Exciton States," Phys. Rev. Lett. 111, 216805 (2013)
 Georgy Samsonidze, Marvin L. Cohen, and Steven G. Louie, "Firstprinciples study of quasiparticle energies of a bipolar molecule in a scanning tunneling microscope measurement," Comput. Mater. Sci. 91, 187 (2014)
 Georgy Samsonidze, Filipe J. Ribeiro, Marvin L. Cohen, and Steven G. Louie, "Quasiparticle and optical properties of polythiophenederived polymers," Phys. Rev. B 90, 035123 (2014)
 Manish Jain, Jack Deslippe, Georgy Samsonidze, Marvin L. Cohen, James R. Chelikowsky, and Steven G. Louie, "Improved quasiparticle wave functions and mean field for G_{0}W_{0} calculations: Initialization with the COHSEX operator," Phys. Rev. B 90, 115148 (2014)
 Georgy Samsonidze, CheolHwan Park, and Boris Kozinsky, "Insights and challenges of applying the GW method to transition metal oxides," J. Phys.: Condens. Matter 26, 475501 (2014)
 Dylan Bayerl and Emmanouil Kioupakis, "VisibleWavelength PolarizedLight Emission with SmallDiameter InN Nanowires," Nano Lett. 14, 3709 (2014)
 M. Gerosa, C. E. Bottani, L. Caramella, G. Onida, C. Di Valentin, and G. Pacchioni, "Electronic structure and phase stability of oxide semiconductors: Performance of dielectricdependent hybrid functional DFT, benchmarked against GW band structure calculations and experiments", Phys. Rev. B 91, 155201 (2015)
 Guangsha Shi and Emmanouil Kioupakis, "Electronic and Optical Properties of Nanoporous Silicon for SolarCell Applications," ACS Photonics 2, 208 (2015)
 Huashan Li, David A. Strubbe, and Jeffrey C. Grossman, "Functionalized Graphene Superlattice as a SingleSheet Solar Cell," Adv. Funct. Mater. 25, 51995205 (2015)
config/README
# Information about how to set the various options here.
# Ideally you can use a file already created for your platform in this directory.
# Otherwise, you can work from the one most similar that is here.
# Select the Fortran compiler you are using.
# if you have a different one than these, modifications in Common/compiler.h
# and Common/commonrules.mk will be necessary.
COMPFLAG = D{INTEL, PGI, GNU, PATH, XLF, G95, ABSOFT, NAG, OPEN64, SUN, CRAY}
# Use DMPI to compile Fortran in parallel with MPI. Otherwise is serial.
# Add DOMP to compile with OpenMP support. Also add appropriate compilerdependent flag to F90free:
# ifort, sunf90, absoft, pathscale, Open64: openmp. PGI: mp=nonuma. gfortran: fopenmp. XLF: qsmp=omp.
# crayftn: on by default (to turn off use h noomp). g95 and NAG do not support OpenMP.
PARAFLAG = DMPI
# DUSESCALAPACK enables usage of ScaLAPACK (required in parallel for BSE).
# DUSEESSL uses ESSL instead of LAPACK in some parts of the code.
# DUNPACKED uses unpacked rather than packed representation of the Hamiltonian
# in EPM. Packed LAPACK operations give bad results eventually in SiEPM kernel 'max x'
# test for ACML, Cray LibSci, and MKL sometimes.
# DUSEFFTW3 uses fftw3 instead of fftw2, which is recommended for better performance, and enables threading.
# DHDF5 enables usage of HDF5 for writing epsmat and bsemat files. The produced files will be
# eps0mat.h5, epsmat.h5 and bsemat.h5 and are used in place of eps0mat, epsmat, bsedmat, bsexmat
# (the latter two are combined in one .h5 file).
# Using DHDF5 gives you substantially better IO performance in many cases by taking advantage
# of HDF5's parallel IO options and other options. However, you must have the HDF5 libraries
# installed on your system and define HDF5LIB and HDF5INCLUDE below.
MATHFLAG = DUSESCALAPACK DUSEESSL DUNPACKED DHDF5
# For Fortran and C++. DDEBUG enables extra checking. DVERBOSE is an internal
# flag, and is equivalent to running all codes with the maximum verbosity level.
# Don't use the DVERBOSE option, unless a developer told you to. The DDEBUG
# option makes extra memory checks, and slows down the code by up to ~20%.
# Only use these flags if you need to develop or debug BerkeleyGW.
# The output will be much more verbose, and the code will slow down by ~20%.
DEBUGFLAG = DDEBUG DVERBOSE
# Same, but for C and C++.
C_DEBUGFLAG = DDEBUG DVERBOSE
# Command for the preprocessor. C (and not ansi) should generally be used.
# add P if the Fortran compiler will not accept indication of line numbers.
# Use gcc's (often in /lib/cpp) or a C compiler (even mpicc) with "E". Some need "x c" as well.
FCPP = cpp C
# Fortran compiler and its flags for F90 and f90 files (free source form)
# see above at PARAFLAG regarding flags for OpenMP.
# With DOMP and openmp flags here, you will get threaded BerkeleyGW.
# Without DOMP, but with openmp flags here, is appropriate for just using threaded libraries.
# Sometimes, flags such as fnosecondunderscore or assume 2underscores may be necessary
# to make linking succeed against C libraries such as FFTW, depending on what compiler built them.
F90free = {mpif90, mpiifort, ftn, ifort, pgf90, gfortran, ...}
# Fortran compiler and any flag(s) required for linking
LINK = {mpif90, mpiifort, ftn, ifort, pgf90, gfortran, ...}
# Fortran optimization flags. The levels below will generally work.
FOPTS = {fast, O3}
# Fortran optimization flags for epsilon_main and sigma_main which can have
# trouble in high optimization due to their long length. If below does not
# work, try O2.
FNOOPTS = $(FOPTS)
# Fortran compiler flag to specify location where module should be created.
# different for each compiler, refer to examples.
MOD_OPT =
# Fortran compiler flag to specify where to look for modules.
INCFLAG = I
# Use this to compile C++ in parallel with MPI. Leave blank for serial.
# MPICH, MVAPICH, Intel MPI require also DMPICH_IGNORE_CXX_SEEK, or an error may occur such as:
# "SEEK_SET is #defined but must not be for the C++ binding of MPI. Include mpi.h before stdio.h"
C_PARAFLAG = DPARA
# C++ compiling command and any needed flags
CC_COMP = {mpiCC, mpicxx, mpiicpc, CC, icpc, pgCC, g++, ...}
# C compiling command and any needed flags
C_COMP = {mpicc, mpiicc, cc, icc, pgcc, gcc, ...}
# C/C++ link command and any needed flags
C_LINK = {mpiCC, mpicxx, mpiicpc, CC, icpc, pgCC, g++, ...}
# C/C++ optimization flags
# Note: for Intel compilers, it is equivalent to use icc or icpc on C++ files.
C_OPTS = fast
# command to remove, for make clean
REMOVE = /bin/rm f
# command for making archives (.a). Default is /usr/bin/ar.
# Usually that is fine and AR does not need to be specified, but
# to do interprocedural optimizations (IPO) with ifort, you need xiar instead.
#AR = /usr/bin/env xiar
# Math Libraries
# path for FFTW library, used in lines below
FFTWPATH =
# link line for FFTW2 library. Sometimes needs to be ldfftw
FFTWLIB = L$(FFTWPATH)/lib lfftw
# link line for FFTW3 library, if DUSEFFTW3 is used. Including fftw_omp allows threaded run.
# Note: To use FFTW3 from MKL, do not link the fftw3xf interfaces.
FFTWLIB = L$(FFTWPATH)/lib {lfftw3_omp} lfftw3
# FFTW2 requires include file fftw_f77.i;
# Sometimes supercomputer installations will not have it.
# If not, find it online and copy it: e.g. http://www.fifi.org/doc/fftwdev/fortran/fftw_f77.i
# FFTW3 requires include file fftw3.f03 instead.
FFTWINCLUDE = $(FFTWPATH)/include
# Different styles will apply depending on the package used.
# Must provide BLAS as well as LAPACK: some packages will do this in one library file,
# others will not. See examples in this directory for how to link with MKL, ACML,
# ATLAS, Ubuntu packages, etc. For MKL, if you will use ScaLAPACK, it is most
# convenient to include BLACS here too. For further info on linking with MKL, see
# the linkline advisor at http://software.intel.com/sites/products/mkl/
LAPACKLIB =
# Specify below if you have DUSESCALAPACK in MATHFLAG.
# BLACS must be provided here too, if it is not provided in LAPACKLIB above. See
# examples in this directory for different packages, as with LAPACKLIB. Note that
# you may have problems if ScaLAPACK was not built with the same LAPACK used above.
SCALAPACKLIB =
# Specify below if you have DHDF5 in MATHFLAG.
# If compiling in parallel, you must have HDF5 compiled with parallel support or linking will fail.
# Path as below should generally work. Sometimes lsz too, or static linkage, might be required.
# Note that "lz" is not from HDF5, but its dependency zlib.
# libdl may be required on your system even when linking static HDF5 due to plugin support in 1.8.11 and higher
HDF5PATH =
HDF5INCLUDE = $(HDF5PATH)/include
HDF5LIB = L$(HDF5PATH)/lib lhdf5hl_fortran lhdf5_hl lhdf5_fortran lhdf5 lz
# Flags for performance profiling with an external tool such as IPM on NERSC
PERFORMANCE =
# Command to submit testsuite job script from testsuite directory.
# qsub is for PBS. If no scheduler, put make checkparallel here.
# In serial, delete this line.
TESTSCRIPT = qsub {architecture}.scr
testsuite/README
BerkeleyGW testsuite
David Strubbe 20092011
== Running tests ==
* To run the testsuite in serial, type "make check" in this directory or the main directory. "make checksave" will do the same, but retain the working directories for debugging.
* To run in parallel without a scheduler, you can just use "make checkparallel" if you do not have a scheduler (e.g. PBS). Set environment variable SAVETESTDIRS=yes to save working directories. The tests use 4 cores, so be sure that at least that many are available, or use BGW_TEST_MPI_NPROCS (see below).
* To run in parallel with a scheduler, type "make checkjobscript" or "make checkjobscriptsave", which will execute the job script for the machine you are using, as specified in arch.mk by the line TESTSCRIPT. See example job scripts for various machines in this directory called *.scr. The tests use 4 cores, so be sure to request at least that many in the job script, or use BGW_TEST_MPI_NPROCS (see below).
* Environment variables:
 TEMPDIRPATH: sets the scratch directory where temporary working directories will be created. Default is /tmp, but on some machines you may not have permission to write there.
 MPIEXEC: sets the command for parallel runs. Default is `which mpiexec`. Note that mpiexec and mpirun are generally equivalent (except for Intel MPI, in which case mpirun is recommended). Set this if you are using a different command such as 'ibrun' (SGE parallel environment), 'runjob' (BlueGene), or 'aprun' (Cray), if you need to use a different mpiexec than the one in your path, or if some options need to be appended to the command. Depending on the value of this variable, three styles of command line are used: ibrun, runjob, and mpirun/aprun.
 BGW_TEST_MPI_NPROCS: set to overrule the number of processors listed in the test files. Useful for testing correctness of parallelization, if you don't have 4 cores, or to run the testsuite faster with more cores.
[ BGW_TEST_NPROCS: deprecated, same as BGW_TEST_MPI_NPROCS, provided for compatibility with versions 1.0.x.]
 MACHINELIST: if set, will be added to command line for MPI runs. This is needed in some MPI implementations.
 EXEC: if set, will be added before name of executable. Used to run code through valgrind.
* To run just one test, in its directory, run "../run_regression_test.pl [s] [p] D ../../bin f [testname].test". Include the s flag to run in serial; otherwise it will be in parallel. Include the p flag to preserve the working directory.
Contents:
(1) run_testsuite.sh  this script runs tests from filenames ending in *.test in this directory.
(2) run_regression_test.pl  called from run_testsuite.sh for each individual test.
(3) *.scr  job scripts for running the testsuite on various parallel machines.
(4) interactive*.sh  scripts to run in parallel, in interactive mode. You have to follow the instructions in the script, not just run it.
(5) queue_monitor.pl  a Perl script for parallel tests with a SLURM or PBS scheduler, which submits a job and monitors the queue to see what happens. Used by the BuildBot.
(6) test directories.
(7) fix_testsuite.py  (for developers) can be used to create or update match reference values in test files.
(8) no_hires.sh  Some clusters will not have the Perl package Time::HiRes installed which is needed for timing in run_regression_test.pl. You can just deactivate the usage with this script.
(9) buildbot_query.pl  (for developers) can be used to find the range of values obtained on the buildslaves for a particular match, to set values in the test files.
== Writing tests ==
The test files consist of lines beginning with one of a set of tags, parsed by the Perl script. Comment lines beginning with '#' will be ignored.
Test : title
Write a title to output to identify the test. Should be the first tag in the file.
Banner : text
Write a heading in a box to identify a specific part of a test. Helpful when input files are generated by sed and do not have a distinct name.
Enabled : Yes/No
If Yes, will be run; if No, will not be run. Use to turn off a test without deleting it from the repository. Should be the second tag in the file.
TestGroups : groupname [groupname2 [...]]
The run_testsuite.sh script can be run with argument "g" and a list of groups. Then tests will only be run if there is a match between the argument list and the list in TestGroups. Examples of groups that could be used: parallel, serial, long. [This tag is actually read by run_testsuite.sh rather than run_regression_test.pl.]
Executable : program.x
Name of executable to be run (path is bin directory). Persists for multiple runs, until next Executable tag.
Processors : integer
Number of processors to use. Ignored if mpiexec is not available. May not exceed the number in the reservation in the job scripts.
Set to special value "serial" to run without mpiexec.
Command : shellcommand
Specify a shell command, which will be executed in the working directory. Useful for renaming log files that would be overwritten by a subsequent run, or preprocessing input files by sed or other utilities (after putting in working directory with Copy).
Copy : file.in [file_newname.in]
Copy a file from test directory to run directory. If new name is not supplied, original name will be used. Useful if you want to sed a file before it is run.
Unpack : data.tar.gz
The file data.tar.gz in the test directory will have "tar xzf" run on it, with resulting files going to the working directory.
Output : file.out
Output will be piped into the file named here, in the working directory.
Arguments : arg1 arg2
The current Executable will be executed (in serial) with the text given here as commandline argument(s). No files will be copied to working directory. Use redirection with ">" to capture the output if necessary.
Input : file.in [file_newname.in/PIPE/CMDLINE/NONE]
The file named here will be copied from the test directory to the working directory, for use as input. If it is followed by "PIPE", the file will be piped into the executable. If it is followed by "CMDLINE", the file will be given as a commandline argument to the executable. If it is followed by "NONE", no file will be copied (useful if an input file was already generated by sed in the working directory). If it is followed by another name, the file's name in the working directory will be the new name. This tag causes actual execution of the run that has been set up by previous tags.
Precision : 1e5
A floating point number, the tolerance for testing whether a match has passed or failed. Persists until next Precision tag. Default is 1e4.
match ; name ; COMMAND(..); referencevalue
Extracts a calculated number from a run and tests it against the reference value. The name is an identifier printed to output. The number is extracted as the standard output from the listed COMMAND, which is one of this set:
. GREP(filename, 'searchtext', field, offset)
Finds the first line in file containing 'searchtext' (which MAY NOT contain a comma), and returns the specified field of that line. "Field" is meant in the sense used by 'awk', i.e. the line is parsed into whitespace separated groups, indexed starting with 1. The optional 'offset' directs the use of the Mth line after the line containing 'searchtext'. No offset is equivalent to offset = 0. This is the most robust of the commands and should be used when possible in preference to the others.
. SIZE(filename)
Returns the size of the specified file via 'ls lt'. Useful for binary files whose contents cannot easily be checked.
. LINE(filename, line, field)
Returns the specified field of the specified line number from the file. Use GREP instead if possible.
. SHELL(shellcommand)
The result is the standard output of the listed command. Deprecated; use GREP or LINE unless absolutely necessary.
STOP TEST
For debugging purposes, to halt test after the part you are studying has run.
library/README
================================================================================
BerkeleyGW library
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014)
Version 1.0 (July, 2011) J. Deslippe, M. Jain, D. A. Strubbe, G. Samsonidze
Version 0.5 J. Deslippe, D. Prendergast, L. Yang, F. Ribeiro, G. Samsonidze (2008)
Version 0.2 M. L. Tiago, C. Spataru, S. IsmailBeigi (2004)

To build other codes with BerkeleyGW output support, type 'make library'
to create libBGW_wfn.a and wfn_rho_vxc_io_m.mod (and dependent modules),
and then compile with I library/ and link library/libBGW_wfn.a with
the other code.
An m4 macro for configure scripts is provided in this directory, for
use in linking to this library. Codes linking the library should 'use'
the module 'wfn_rho_vxc_io_m'.
To generate real wavefunctions, a GramSchmidt process should be used.
The appropriate parallelization scheme etc. will be dependent on the code,
and cannot be easily handled in a general manner here, but examples can
be found in MeanField/SIESTA/real.f90 and MeanField/ESPRESSO/pw2bgw.f90
(routine real_wfng).
examples/README
================================================================================
BerkeleyGW Examples
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.

This folder contains a set of example calculations for BerkeleyGW 1.2.
The meanfield portion of the examples may be done with the Empirical
Pseudopotential Method (EPM), or ab intio DFT, using the Quantum ESPRESSO,
PARATEC, or ABINIT codes. Other meanfield codes, such as Octopus and PARSEC,
are also supported, but we don't have examples for them at this point.
The examples contained here are organized according to the meanfield starting
point:
1. Example under the "DFT" folder, use ab initio DFT as the starting meanfield
picture for our GW and BetheSalpeter equation (BSE) calculations. You must
have either PARATEC, Quantum ESPRESSO, or ABINIT for run these examples.
Required pseudopotentials are stored in DFT/pseudos.
2. Examples under the "EPM" folder use the Empirical Pseudopotential Method
(EPM), and you won't need any external meanfield code to run these examples.
Calculations in these folders are much shorter to run, but are probably of
less
Each of these directories contain a number of subdirectories, each of them
containing a different example and a README file with the instructions how to
run each examples and suggestion on the number of processors to use.
Before you run anything, note that YOU MUST MODIFY ALL SCRIPTS in this folder
to set the appropriate path to BerkeleyGW (eg: $HOME/GW), your parallel job
launching command (eg: mpirun), and parallel submission flags (eg: "#SLURM ...").
Tutorial

The idea behind these examples is to provide a working set of input files
and reference output files for several calculations that span a range of
relevant materials. However, for a more pedagogical introduction to BerkeleyGW,
we suggest you first try to run our tutorials that are designed for our yearly
BerkeleyGW workshops, available at:
https://sites.google.com/site/berkeleygw2015/home
Note for ABINIT

Our ABINIT wrapper is still experimental, and not all examples have been ported
to use ABINIT.
Note for Quantum ESPRESSO

Quantum ESPRESSO examples are compatible with
Quantum ESPRESSO 5.0 and later versions. If you want to run them with Quantum
ESPRESSO 4.3.2 or earlier version, you should modify the input files for pw.x, and
change "calculation = 'bands'" to "calculation = 'nscf'" and "CELL_PARAMETERS" to
"CELL_PARAMETERS cubic".
Basic examples

 silicon: a bulk semiconductor. You should start here!
 Includes PREBUILT tarball of DFT inputs from PARATEC.
 Includes ref directories containing sample output from ESPRESSO and BerkeleyGW steps.
 Available for EPM as well as DFT.
 Shows how to use Wannier90 and sig2wan.x to obtain a GW bandstructure by
interpolation (optional step).
 Shows how to use inteqp to obtain a GW bandstructure by interpolation.
 GaAs: similar to silicon example. Runs slightly faster.
 sodium: a bulk metal
 swcnt_55: (5,5) singlewalled carbon nanotube, a metallic 1D system
 swcnt_80: (8,0) singlewalled carbon nanotube, a semiconducting 1D system
 Has runs of plotxct with visualization by volume.py and surface.x.
 CO: a small molecule
 benzene: a slightly larger molecule
 Includes run of gsphere.py and SIESTA.
Advanced examples

These examples demonstrate how to use advanced features of the code, and they
are not as tidy as the other examples. You are expected to modify submission
 Si2_sapo: another silicon example
 Shows how to perform iterative Davidson diagonalization in SAPO,
 requires pw2bgw with write_vscg capability as in espresso5.0.2.
 Zn2O2: a bulk semiconductor
 Includes ref directories containing sample output from ESPRESSO and BerkeleyGW steps.
 Shows how to use inteqp to obtain a GW bandstructure by interpolation.
 Shows how to perform iterative Davidson diagonalization in SAPO,
requires pw2bgw with write_vscg capability as in espresso5.0.2.
 Shows how to use valence only charge density in the Generalized Plasmon Pole model,
requires pw2bgw with calc_rhog capability as in espresso5.0.2.
Other examples

 Make sure you visit the tutorials presented during the BerkeleyGW workshops:
https://sites.google.com/site/berkeleygw2015/home
 The testsuite runs may also be consulted, although they are not realistic calculations.
 There are also examples in the Visual directory for the scripts there.
MeanField
MeanField/README
================================================================================
BerkeleyGW: MeanField wrappers and utilities
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.

Description:
BerkeleyGW uses a meanfield starting point, typically from DFT.
Currently, BerkeleyGW supports two planewave DFT codes,
PARATEC and Quantum ESPRESSO; one localizedorbital DFT code, SIESTA;
and two realspace codes, Octopus and PARSEC.
You can find details on how to use BerkeleyGW with these codes in
their respective directories.
For testing purposes and for crude calculations of large systems,
BerkeleyGW can be executed on top of the empirical pseudopotential
planewave code. See MeanField/EPM for details.
There is a tool for generating large numbers of unoccupied orbitals
(approximate from plane waves or exact by iterative diagonalization).
It is located in MeanField/SAPO.
You can find a simple code for computing the Coulomb integral within
an imagecharge model in MeanField/ICM.

Tricks and hints:
1. Vxc0
Vxc0 is the G=0 component of the exchangecorrelation potential Vxc.
Vxc0 is added to the eigenvalues in the wavefunction files produced
by PARATEC and ESPRESSO. Vxc0 is also included in the VXC and vxc.dat
files produced by PARATEC and ESPRESSO (the VXC file contains Vxc(G)
and the vxc.dat file contains matrix elements of Vxc). The two Vxc0
terms cancel out when calculating the quasiparticle corrections
in the Sigma code.
2. Vacuum level
To correct the DFT eigenvalues for the vacuum level take the average
of the electrostatic potential on the faces of the unit cell in the
nonperiodic directions and subtract it from the DFT eigenvalues.
The electrostatic potential is defined as Velec = Vbare + Vhartree,
while the total potential is Vtot = Vbare + Vhartree + Vxc, hence
Vtot contains Vxc0 and Velec does not. The average of Velec is
fed into the Sigma code using keywords avgpot and avgpot_outer.
The potentials can be generated with PARATEC and ESPRESSO.
In PARATEC, use elecplot for Velec or potplot for Vtot, then
convert from a3dr to cube using the Visual/volume.py script.
In ESPRESSO, use iflag=3, output_format=6, and plot_num=11
for Velec or plot_num=1 for Vtot. The averaging is done with
the Visual/average.py script.
3. Unit cell size
If you truncate the Coulomb interaction, make sure that the size
of the unit cell in nonperiodic directions is at least two times
larger than the size of the charge distribution. This is needed
to avoid spurious interactions between periodic replicas but at
the same time not to alter interactions within the same unit cell.
Run Visual/surface.x to plot an isosurface that contains 99% of the
charge density (see Visual/README for instructions on how to do this).
The code will print the size of the box that contains the isosurface
to stdout. Multiply the box dimensions in nonperiodic directions
by two to get the minimum size of the unit cell.
4. Inversion symmetry
When using the real flavor of the code, make sure the inversion
symmetry has no associated fractional translation (if it does,
shift the coordinate origin). Otherwise WFN, VXC, RHO (and VSC
in SAPO) have nonvanishing imaginary parts which are simply
dropped in the real flavor of the code. This won't do any
good to your calculation...
ESPRESSO
MeanField/ESPRESSO/README
Quantum ESPRESSO is available for download at http://www.quantumespresso.org/
The interface between Quantum ESPRESSO and BerkeleyGW consists of three programs,
kgrid.x, pw2bgw.x and bgw2pw.x. kgrid.x is compiled with BerkeleyGW package and
pw2bgw.x and bgw2pw.x are compiled with Quantum ESPRESSO.
Files for versions 4.3.2, 5.0.x, and 5.1.x are supplied here.
Starting with version 5.0, compatible files of pw2bgw.f90 and bgw2pw.f90 are
distributed with Quantum ESPRESSO.
If you are using espresso5.1, apply a patch following the
instructions in BerkeleyGW/MeanField/ESPRESSO/version5.1/README_patch.
Alternatively, you can use BerkeleyGW/MeanField/ESPRESSO/version5.1/pw2bgw.f90
(rather than espresso5.1/PP/src/pw2bgw.f90), which includes a bugfix to
the problem of an immediate segmentation fault or error
saying 'MPI_Comm_rank: Invalid communicator'. To use:
cp BerkeleyGW/MeanField/ESPRESSO/version5.1/pw2bgw.f90 espresso5.1/PP/src/
Then 'make pp' in espresso. This seems to happen for MPICH and MVAPICH (and
probably the related IntelMPI) but not OpenMPI, but the updated version should work
in all cases. This is fixed in espresso5.1.1 and later versions.
The development versions of pw2bgw.f90 and bgw2pw.f90 are available from
Quantum ESPRESSO trunk. Quantum ESPRESSO allows anonymous checkouts of its
source code. Use the command below and use 'anonymous' as the username and
a blank password to checkout the code:
svn checkout http://qeforge.qeforge.org/svn/qe/trunk/espresso
You can find more details about Quantum ESPRESSO svn on this website:
http://www.qeforge.org/gf/project/qe/scmsvn/

kgrid.x
pw.x can automatically generate a uniform grid of kpoints, either unshifted
or shifted by half a grid step (MonkhorstPack grid), and reduce it to the
irreducible wedge of the Brillouin Zone with the symmetries of the point group
of the Bravais lattice and optionally with timereversal symmetry. The same
functionality (and more) is provided by kgrid.x in this directory. Additionally,
kgrid.x can use the symmetries of the space group of the crystal instead of the
symmetries of the point group of Bravais lattice, which is needed for BerkeleyGW.
kgrid.x is not limited to the unshifted/halfstep shifted MonkhorstPack grids.
It can generate an asymmetrically shifted fine grid used to improve the
convergence in absorption calculations. Also, kgrid.x can generate a grid
of kpoints with a small qshift used in Epsilon calculation to avoid the
divergence of the Coulomb interaction. The list of kpoints generated by
kgrid.x must be manually added to the input file of pw.x. Note that timereversal
symmetry should NOT be used for BerkeleyGW (noinv = .false. if generating the
kgrid in pw.x), unless the utility wfn_time_reversal.x is run afterward.
The format of the input file for kgrid.x (along with an example for Si) is found
in kgrid.inp. See also more information in the header of kgrid.f90. You can find
the input files for kgrid.x in examples/DFT in the ESPRESSO subdirectories of
each example.

datafile2kgrid.py
Parses the atomic positions and FFT grid from a Quantum Espresso datafile.xml
file and creates the input file kgrid.inp for the kgrid.x utility.
All options to kgrid.x  such as number of kpoints, kshift, and kgrid 
are read as optional commandline arguments to datafile2kgrid.py. Type
datafile2kgrid.py help for a list of possible options.

pw2bgw.x
Converts the output files produced by pw.x to the input files for BerkeleyGW.
You cannot use USPP, PAW, or spinors in a pw.x run for BerkeleyGW.
You cannot use "K_POINTS gamma" in a pw.x run for BerkeleyGW.
Use "K_POINTS { tpiba  automatic  crystal }" even for the
Gammapoint calculation.
The format of the input file for pw2bgw.x is described
in files MeanField/ESPRESSO/version4.3.2/pw2bgw.inp,
MeanField/ESPRESSO/version5.1/INPUT_pw2bgw.html (copy of espresso5.1/PP/Doc/INPUT_pw2bgw.html),
which is generated from MeanField/ESPRESSO/version5.1/INPUT_pw2bgw.def, and
MeanField/ESPRESSO/version5.0/INPUT_pw2bgw.html (copy of espresso5.0.3/PP/Doc/INPUT_pw2bgw.html),
which is generated from MeanField/ESPRESSO/version5.0/INPUT_pw2bgw.def.
You can find the input files for pw2bgw.x in examples/DFT
in the ESPRESSO subdirectories of each example.
Version 4.3.2 and earlier:
It is recommended to run a pw.x "nscf" calculation with "K_POINTS crystal"
and a list of kpoints produced by kgrid.x.
Sometimes pw.x generates additional kpoints in a "nscf" run with an explicit
list of kpoints. If this is the case for your calculation, there are several
ways to go around this problem:
* Apply the patch provided with BerkeleyGW. It will prevent pw.x from
generating additional kpoints if they are provided explicitly, and
take care of the normalization of the weights of kpoints in a "bands"
calculation.
* Do not specify the atomic positions in the input file of kgrid.x (set
number of atoms = 0). Then pw.x will generate additional kpoints which
are the correct ones. Also set noinv = .true. in the input file of pw.x
if timereversal symmetry was not used in kgrid.x.
* Run a pw.x "bands" calculation instead of "nscf". In this case you have
to explicitly specify the occupations in the input file of pw2bgw.x (note
that this only works for insulators) and to normalize the weights of
kpoints to one in the input file of pw.x.
Version 5.0 and later:
It is recommended to run a pw.x "bands" calculation with "K_POINTS crystal"
and a list of kpoints produced by kgrid.x.
You can also run a pw.x "nscf" calculation instead of "bands", but in this
case pw.x may generate more kpoints than provided in the input file of pw.x.
If this is the case for your calculation you will get errors in BerkeleyGW.

bgw2pw.x
Converts BerkeleyGW WFN and RHO files to the format of pw.x.
This can be useful, for example, if you generate the plane waves
on top of the valence bands and want to diagonalize them in pw.x.
Look at the documentation for SAPO code in BerkeleyGW for more information.
Another possible use is to convert between different versions of pw.x.
bgw2pw.x reads common parameters from file prefix.save/datafile.xml and
writes files prefix.save/chargedensity.dat (charge density in Rspace),
prefix.save/gvectors.dat (Gvectors for charge density and potential),
prefix.save/K$n/eigenval.xml (eigenvalues and occupations for nth kpoint),
prefix.save/K$n/evc.dat (wavefunctions in Gspace for nth kpoint), and
prefix.save/K$n/gkvectors.dat (Gvectors for nth kpoint).
You must have prefix.save/K$n/eigenval.xml files present, or an error will occur,
even though their contents will not be used and will be overwritten.
The best is to run a pw.x calculation and use its prefix.save, e.g. from scf
and then get unoccupied bands from bgw2pw.x.
bgw2pw.x doesn't create restart files, so you cannot use restart_mode = 'restart'
for a subsequent pw.x run. Instead, use startingwfc = 'file'. Make sure
wf_collect = .true. and there are no prefix.wfc* files present. Also, the pw.x run
that generated the prefix.save directory originally must have wf_collect = .true.
also, for the appropriate links to the K$n files to be present.
bgw2pw.x doesn't modify file prefix.save/datafile.xml so make changes to this
file manually. For example, you will need to change the NUMBER_OF_BANDS and
NUMBER_OF_PROCESSORS tags (as well as per pool and per image) to make sure these
match the number of bands from the WFN file, as well as the number of processors
you will use in a subsequent run.
The format of the input file for bgw2pw.x is described
in files MeanField/ESPRESSO/version4.3.2/bgw2pw.inp,
MeanField/ESPRESSO/version5.1/INPUT_bgw2pw.html (copy of espresso5.1/PP/Doc/INPUT_bgw2pw.html),
which is generated from MeanField/ESPRESSO/version5.1/INPUT_bgw2pw.def, and
MeanField/ESPRESSO/version5.0/INPUT_bgw2pw.html (copy of espresso5.0.3/PP/Doc/INPUT_bgw2pw.html),
which is generated from MeanField/ESPRESSO/version5.0/INPUT_bgw2pw.def.

Notes on running pw.x
Sometimes pw.x crashes when trying to generate a LARGE number of unoccupied
states needed for GW calculations. The error messages may refer to "Cholesky
decomposition" or "diagonalization". If you run into this problem try one of
the following workarounds, or a combination of them:
1) Increase ecutwfc. Iterative diagonalization becomes inefficient and
unstable with increasing the ratio of the number of states to the size
of the Hamiltonian. Increasing ecutwfc will decrease this ratio
at the cost of computation time.
2) Split the calculation over kpoints. If one kpoint fails, it won't
affect the other kpoints. You can merge the final wavefunction file
using MeanField/Utilities/wfnmerge.x after running pw2bgw.x for each
kpoint.
3) Start with random instead of randomized atomic wavefunctions. For
this use the following parameter in the input file of pw.x:
startingwfc = 'random'
4) Split the diagonalization into several steps alternating between
different diagonalization schemes. For example, use the following
parameters in the input files for consequent runs of pw.x:
1st run:
conv_thr = 1.0d2
diagonalization = 'david'
2nd run:
conv_thr = 1.0d4
diagonalization = 'cg'
startingwfc = 'file'
3rd run:
conv_thr = 1.0d6
diagonalization = 'david'
startingwfc = 'file'
etc.
5) Run pw.x with "ndiag 1". Often serial diagonalization is more
stable than the parallel one, and the time and memory overhead
is bearable in most cases.
6) Compile pw.x without "D__SCALAPACK" in DFLAGS in make.sys
for using a custom distributedmemory diagonalization instead
of ScaLAPACK. This is useful to check if the problem
is caused by improper installation of ScaLAPACK.
Note that you will have to explicitly specify ndiag,
e.g. "mpirun np 24 pw.x ndiag 16 in prefix.in > prefix.out",
because pw.x only sets ndiag automatically in case of ScaLAPACK.
7) Last but not least, perform iterative diagonalization
in SAPO instead of using pw.x, see examples/DFT/Si2_sapo
for details.
Finally, it is recommended to always use the following parameters
in the input files for pw.x:
wf_collect = .true.
diago_full_acc = .true.
See documentation on Quantum ESPRESSO for information on these
parameters.
BEWARE: Sometimes wavefunctions may lose orthogonality during
iterative diagonalization. Neither Quantum ESPRESSO nor BerkeleyGW
checks for orthogonality of wavefunctions. This may cause erroneous
behavior like diverging head of eps0mat and possibly many other things.
This problem could be avoided by using SAPO [see 7) above] which keeps
wavefunctions orthogonal during iterative diagonalization.
MeanField/ESPRESSO/kgrid.inp
5 5 5 ! numbers of kpoints along b1,b2,b3
0.5 0.5 0.5 ! kgrid offset (0.0 unshifted, 0.5 shifted by half a grid step)
! These first two lines are the usual MonkhorstPack parameters.
0.0 0.0 0.001 ! a small qshift (0.0 unshifted, 0.001 shifted by one 1000th of b3)
! This is for WFNq in Epsilon.
0.0 0.5 0.5 ! lattice vectors in Cartesian coordinates (x,y,z)
0.5 0.0 0.5 ! in units of the lattice parameter
0.5 0.5 0.0 !
2 ! number of atoms in the unit cell
1 0.125 0.125 0.125 ! atomic species and positions in Cartesian coordinates (x,y,z)
1 0.125 0.125 0.125 ! in units of the lattice parameter
0 0 0 ! size of FFT grid
.false. ! use timereversal symmetry. Set to false for BerkeleyGW
.false. ! OPTIONAL: kpoints in the log file are in Cartesian coordinates
.false. ! OPTIONAL: output is in Octopus format
# Above: typical values for Si. Below: general description.
# nk1 nk2 nk3 ! numbers of kpoints in crystal coordinates (b1,b2,b3)
# dk1 dk2 dk3 ! kgrid offset (0.0 unshifted, 1.0 shifted by bj/nkj)
# dq1 dq2 dq3 ! a small qshift (0.0 unshifted, 1.0 shifted by bj)
# a1x a1y a1z ! lattice vectors in Cartesian coordinates (x,y,z)
# a2x a2y a2z ! in arbitrary units (bohr, angstrom, lattice parameter)
# a3x a3y a3z !
# n ! number of atoms in the unit cell
# s1 x1 y1 z1 ! atomic species and positions in Cartesian coordinates (x,y,z)
# ........... ! in the same units as the lattice vectors
# sn xn yn zn !
# nr1 nr2 nr3 ! size of FFT grid
# trs ! if set to .true., use timereversal symmetry (do not use this for BerkeleyGW)
# Cartesian ! OPTIONAL: set to .true. for kpoints in Cartesian coordinates (only in the log file)
# octopus ! OPTIONAL: set to .true. to write output file in format suitable for Octopus
MeanField/ESPRESSO/version5.1/INPUT_pw2bgw.html
Input File Description
Program: pw2bgw.x / PWscf / Quantum Espresso

TABLE OF CONTENTS
INTRODUCTION
&INPUT_PW2BGW
prefix  outdir  real_or_complex  symm_type  wfng_flag  wfng_file  wfng_kgrid  wfng_nk1  wfng_nk2  wfng_nk3  wfng_dk1  wfng_dk2  wfng_dk3  wfng_occupation  wfng_nvmin  wfng_nvmax  rhog_flag  rhog_file  rhog_nvmin  rhog_nvmax  vxcg_flag  vxcg_file  vxc0_flag  vxc0_file  vxc_flag  vxc_file  vxc_integral  vxc_diag_nmin  vxc_diag_nmax  vxc_offdiag_nmin  vxc_offdiag_nmax  vxc_zero_rho_core  vscg_flag  vscg_file  vkbg_flag  vkbg_file
INTRODUCTIONConverts the output files produced by pw.x to the input files for BerkeleyGW.
You cannot use USPP, PAW, or spinors in a pw.x run for BerkeleyGW.
You cannot use "K_POINTS gamma" in a pw.x run for BerkeleyGW.
Use "K_POINTS { tpiba  automatic  crystal }" even for the
Gammapoint calculation.
It is recommended to run a pw.x "bands" calculation with "K_POINTS crystal"
and a list of kpoints produced by kgrid.x, which is a part of BerkeleyGW
package (see BerkeleyGW documentation for details).
You can also run a pw.x "nscf" calculation instead of "bands", but in this
case pw.x may generate more kpoints than provided in the input file of pw.x.
If this is the case for your calculation you will get errors in BerkeleyGW.
Examples showing how to run BerkeleyGW on top of Quantum ESPRESSO including
the input files for pw.x and pw2bgw.x are distributed together with the
BerkeleyGW package.
Structure of the input data:
============================
&INPUT_PW2BGW
...
/
Namelist: INPUT_PW2BGW 
prefix 
STRING 
Status: 
MANDATORY

prefix of files saved by program pw.x

outdir 
STRING 
Default: 
'./'

the scratch directory where the massive datafiles are written

real_or_complex 
INTEGER 
Default: 
2

1  2
1 for real flavor of BerkeleyGW (for systems with inversion symmetry and
timereversal symmetry) or 2 for complex flavor of BerkeleyGW (for systems
without inversion symmetry and timereversal symmetry)

symm_type 
STRING 
Default: 
'cubic'

'cubic'  'hexagonal'
type of crystal system, 'cubic' for space groups 1 ... 142 and 195 ... 230
and 'hexagonal' for space groups 143 ... 194. Only used if ibrav = 0 in a
pw.x run. Written to BerkeleyGW WFN, RHO, VXC and VKB files but no longer
used (except in SAPO code in BerkeleyGW). You can use the default value for
all systems. Don't set to different values in different files for the same
system or you will get errors in BerkeleyGW.

wfng_flag 
LOGICAL 
Default: 
.FALSE.

write wavefunctions in Gspace to BerkeleyGW WFN file

wfng_file 
STRING 
Default: 
'WFN'

name of BerkeleyGW WFN output file. Not used if wfng_flag = .FALSE.

wfng_kgrid 
LOGICAL 
Default: 
.FALSE.

overwrite kgrid parameters in BerkeleyGW WFN file.
If pw.x input file contains an explicit list of kpoints,
the kgrid parameters in the output of pw.x will be set to zero.
Since sigma and absorption in BerkeleyGW both need to know the
kgrid dimensions, we patch these parameters into BerkeleyGW WFN file

wfng_nk1 
INTEGER 
Default: 
0

number of kpoints along b_1 reciprocal lattice vector.
Not used if wfng_kgrid = .FALSE.

wfng_nk2 
INTEGER 
Default: 
0

number of kpoints along b_2 reciprocal lattice vector.
Not used if wfng_kgrid = .FALSE.

wfng_nk3 
INTEGER 
Default: 
0

number of kpoints along b_3 reciprocal lattice vector.
Not used if wfng_kgrid = .FALSE.

wfng_dk1 
REAL 
Default: 
0.0

kgrid offset (0.0 unshifted, 0.5 shifted by half a grid step)
along b_1 reciprocal lattice vector. Not used if wfng_kgrid = .FALSE.

wfng_dk2 
REAL 
Default: 
0.0

kgrid offset (0.0 unshifted, 0.5 shifted by half a grid step)
along b_2 reciprocal lattice vector. Not used if wfng_kgrid = .FALSE.

wfng_dk3 
REAL 
Default: 
0.0

kgrid offset (0.0 unshifted, 0.5 shifted by half a grid step)
along b_3 reciprocal lattice vector. Not used if wfng_kgrid = .FALSE.

wfng_occupation 
LOGICAL 
Default: 
.FALSE.

overwrite occupations in BerkeleyGW WFN file

wfng_nvmin 
INTEGER 
Default: 
0

index of the lowest occupied band (normally = 1).
Not used if wfng_occupation = .FALSE.

wfng_nvmax 
INTEGER 
Default: 
0

index of the highest occupied band (normally = number of occupied bands).
Not used if wfng_occupation = .FALSE.

rhog_flag 
LOGICAL 
Default: 
.FALSE.

write charge density in Gspace to BerkeleyGW RHO file.
Only used for the GPP model in sigma code in BerkeleyGW

rhog_file 
STRING 
Default: 
'RHO'

name of BerkeleyGW RHO output file. Only used for the GPP model in sigma
code in BerkeleyGW. Not used if rhog_flag = .FALSE.

rhog_nvmin 
INTEGER 
Default: 
0

index of the lowest band used for calculation of charge density. This is
needed if one wants to exclude semicore states from charge density used
for the GPP model in sigma code in BerkeleyGW. Make sure to include the
same kpoints as in scf calculation. Selfconsistent charge density is
used if rhog_nvmin = 0 and rhog_nvmax = 0. Not used if rhog_flag = .FALSE.
BEWARE: this feature is highly experimental and may not work at all in
parallel, with pools, with spins, etc.

rhog_nvmax 
INTEGER 
Default: 
0

index of the highest band used for calculation of charge density. See
description of rhog_nvmin for more details

vxcg_flag 
LOGICAL 
Default: 
.FALSE.

write local part of exchangecorrelation potential in Gspace to
BerkeleyGW VXC file. Only used in sigma code in BerkeleyGW, it is
recommended to use vxc_flag instead

vxcg_file 
STRING 
Default: 
'VXC'

name of BerkeleyGW VXC output file. Only used in sigma code in BerkeleyGW,
it is recommended to use vxc_flag instead. Not used if vxcg_flag = .FALSE.

vxc0_flag 
LOGICAL 
Default: 
.FALSE.

write Vxc(G = 0) to text file. Only for testing, not required for BerkeleyGW

vxc0_file 
STRING 
Default: 
'vxc0.dat'

name of output text file for Vxc(G = 0). Only for testing, not required for
BerkeleyGW. Not used if vxc0_flag = .FALSE.

vxc_flag 
LOGICAL 
Default: 
.FALSE.

write matrix elements of exchangecorrelation potential to text file.
Only used in sigma code in BerkeleyGW

vxc_file 
STRING 
Default: 
'vxc.dat'

name of output text file for Vxc matrix elements. Only used in sigma code
in BerkeleyGW. Not used if vxc_flag = .FALSE.

vxc_integral 
STRING 
Default: 
'g'

'g'  'r'
'g' to compute matrix elements of exchangecorrelation potential in Gspace.
'r' to compute matrix elements of the local part of exchangecorrelation
potential in Rspace. It is recommended to use 'g'. Not used if vxc_flag = .FALSE.

vxc_diag_nmin 
INTEGER 
Default: 
0

minimum band index for diagonal Vxc matrix elements. Not used if vxc_flag = .FALSE.

vxc_diag_nmax 
INTEGER 
Default: 
0

maximum band index for diagonal Vxc matrix elements. Not used if vxc_flag = .FALSE.

vxc_offdiag_nmin 
INTEGER 
Default: 
0

minimum band index for offdiagonal Vxc matrix elements. Not used if vxc_flag = .FALSE.

vxc_offdiag_nmax 
INTEGER 
Default: 
0

maximum band index for offdiagonal Vxc matrix elements. Not used if vxc_flag = .FALSE.

vxc_zero_rho_core 
LOGICAL 
Default: 
.TRUE.

set to .TRUE. to zero out NLCC or to .FALSE. to keep NLCC when computing
exchangecorrelation potential. This flag has no effect for pseudopotentials
without NLCC. BEWARE: setting vxc_zero_rho_core to .FALSE. will produce
incorrect results. This functionality is only included for testing purposes
and is not meant to be used in a production environment

vscg_flag 
LOGICAL 
Default: 
.FALSE.

write local part of selfconsistent potential in Gspace to
BerkeleyGW VSC file. Only used in SAPO code in BerkeleyGW

vscg_file 
STRING 
Default: 
'VSC'

name of BerkeleyGW VSC output file. Only used in SAPO code in BerkeleyGW.
Not used if vscg_flag = .FALSE.

vkbg_flag 
LOGICAL 
Default: 
.FALSE.

write KleinmanBylander projectors in Gspace to BerkeleyGW VKB file.
Only used in SAPO code in BerkeleyGW

vkbg_file 
STRING 
Default: 
'VKB'

name of BerkeleyGW VKB output file. Only used in SAPO code in BerkeleyGW.
Not used if vkbg_flag = .FALSE.




This file has been created by helpdoc utility.
MeanField/ESPRESSO/version5.1/INPUT_bgw2pw.html
Input File Description
Program: bgw2pw.x / PWscf / Quantum Espresso

TABLE OF CONTENTS
INTRODUCTION
&INPUT_BGW2PW
prefix  outdir  real_or_complex  wfng_flag  wfng_file  wfng_nband  rhog_flag  rhog_file
INTRODUCTIONConverts BerkeleyGW WFN and RHO files to the format of pw.x.
This can be useful, for example, if you generate the plane waves
on top of the valence bands and want to diagonalize them in pw.x.
Look at the documentation for SAPO code in BerkeleyGW for more information.
bgw2pw.x reads common parameters from file prefix.save/datafile.xml and
writes files prefix.save/chargedensity.dat (charge density in Rspace),
prefix.save/gvectors.dat (Gvectors for charge density and potential),
prefix.save/K$n/eigenval.xml (eigenvalues and occupations for nth kpoint),
prefix.save/K$n/evc.dat (wavefunctions in Gspace for nth kpoint), and
prefix.save/K$n/gkvectors.dat (Gvectors for nth kpoint).
bgw2pw.x doesn't modify file prefix.save/datafile.xml so make changes to this
file manually (for example, you will need to change the number of bands if you
are using bgw2pw.x in conjunction with SAPO code in BerkeleyGW).
Structure of the input data:
============================
&INPUT_BGW2PW
...
/
Namelist: INPUT_BGW2PW 
prefix 
STRING 
Status: 
MANDATORY

prefix of files saved by program pw.x

outdir 
STRING 
Default: 
'./'

the scratch directory where the massive datafiles are written

real_or_complex 
INTEGER 
Default: 
2

1  2
1 for real flavor of BerkeleyGW (for systems with inversion symmetry and
timereversal symmetry) or 2 for complex flavor of BerkeleyGW (for systems
without inversion symmetry and timereversal symmetry)

wfng_flag 
LOGICAL 
Default: 
.FALSE.

read wavefunctions in Gspace from BerkeleyGW WFN file

wfng_file 
STRING 
Default: 
'WFN'

name of BerkeleyGW WFN input file. Not used if wfng_flag = .FALSE.

wfng_nband 
INTEGER 
Default: 
0

number of bands to write (0 = all). Not used if wfng_flag = .FALSE.

rhog_flag 
LOGICAL 
Default: 
.FALSE.

read charge density in Gspace from BerkeleyGW RHO file

rhog_file 
STRING 
Default: 
'RHO'

name of BerkeleyGW RHO input file. Not used if rhog_flag = .FALSE.




This file has been created by helpdoc utility.
MeanField/ESPRESSO/version5.0/INPUT_pw2bgw.html
Input File Description
Program: pw2bgw.x / PWscf / Quantum Espresso

TABLE OF CONTENTS
INTRODUCTION
&INPUT_PW2BGW
prefix  outdir  real_or_complex  symm_type  wfng_flag  wfng_file  wfng_kgrid  wfng_nk1  wfng_nk2  wfng_nk3  wfng_dk1  wfng_dk2  wfng_dk3  wfng_occupation  wfng_nvmin  wfng_nvmax  rhog_flag  rhog_file  vxcg_flag  vxcg_file  vxc0_flag  vxc0_file  vxc_flag  vxc_file  vxc_integral  vxc_diag_nmin  vxc_diag_nmax  vxc_offdiag_nmin  vxc_offdiag_nmax  vxc_zero_rho_core  vscg_flag  vscg_file  vkbg_flag  vkbg_file
INTRODUCTIONConverts the output files produced by pw.x to the input files for BerkeleyGW.
You cannot use USPP, PAW, or spinors in a pw.x run for BerkeleyGW.
You cannot use "K_POINTS gamma" in a pw.x run for BerkeleyGW.
Use "K_POINTS { tpiba  automatic  crystal }" even for the
Gammapoint calculation.
It is recommended to run a pw.x "bands" calculation with "K_POINTS crystal"
and a list of kpoints produced by kgrid.x, which is a part of BerkeleyGW
package (see BerkeleyGW documentation for details).
You can also run a pw.x "nscf" calculation instead of "bands", but in this
case pw.x may generate more kpoints than provided in the input file of pw.x.
If this is the case for your calculation you will get errors in BerkeleyGW.
Examples showing how to run BerkeleyGW on top of Quantum ESPRESSO including
the input files for pw.x and pw2bgw.x are distributed together with the
BerkeleyGW package.
Structure of the input data:
============================
&INPUT_PW2BGW
...
/
Namelist: INPUT_PW2BGW 
prefix 
STRING 
Status: 
MANDATORY

prefix of files saved by program pw.x

outdir 
STRING 
Default: 
'./'

the scratch directory where the massive datafiles are written

real_or_complex 
INTEGER 
Default: 
2

1  2
1 for real flavor of BerkeleyGW (for systems with inversion symmetry and
timereversal symmetry) or 2 for complex flavor of BerkeleyGW (for systems
without inversion symmetry and timereversal symmetry)

symm_type 
STRING 
Default: 
'cubic'

'cubic'  'hexagonal'
type of crystal system, 'cubic' for space groups 1 ... 142 and 195 ... 230
and 'hexagonal' for space groups 143 ... 194. Only used if ibrav = 0 in a
pw.x run. Written to BerkeleyGW WFN, RHO, VXC and VKB files but no longer
used (except in SAPO code in BerkeleyGW). You can use the default value for
all systems. Don't set to different values in different files for the same
system or you will get errors in BerkeleyGW.

wfng_flag 
LOGICAL 
Default: 
.FALSE.

write wavefunctions in Gspace to BerkeleyGW WFN file

wfng_file 
STRING 
Default: 
'WFN'

name of BerkeleyGW WFN output file. Not used if wfng_flag = .FALSE.

wfng_kgrid 
LOGICAL 
Default: 
.FALSE.

overwrite kgrid parameters in BerkeleyGW WFN file.
If pw.x input file contains an explicit list of kpoints,
the kgrid parameters in the output of pw.x will be set to zero.
Since sigma and absorption in BerkeleyGW both need to know the
kgrid dimensions, we patch these parameters into BerkeleyGW WFN file

wfng_nk1 
INTEGER 
Default: 
0

number of kpoints along b_1 reciprocal lattice vector.
Not used if wfng_kgrid = .FALSE.

wfng_nk2 
INTEGER 
Default: 
0

number of kpoints along b_2 reciprocal lattice vector.
Not used if wfng_kgrid = .FALSE.

wfng_nk3 
INTEGER 
Default: 
0

number of kpoints along b_3 reciprocal lattice vector.
Not used if wfng_kgrid = .FALSE.

wfng_dk1 
REAL 
Default: 
0.0

kgrid offset (0.0 unshifted, 0.5 shifted by half a grid step)
along b_1 reciprocal lattice vector. Not used if wfng_kgrid = .FALSE.

wfng_dk2 
REAL 
Default: 
0.0

kgrid offset (0.0 unshifted, 0.5 shifted by half a grid step)
along b_2 reciprocal lattice vector. Not used if wfng_kgrid = .FALSE.

wfng_dk3 
REAL 
Default: 
0.0

kgrid offset (0.0 unshifted, 0.5 shifted by half a grid step)
along b_3 reciprocal lattice vector. Not used if wfng_kgrid = .FALSE.

wfng_occupation 
LOGICAL 
Default: 
.FALSE.

overwrite occupations in BerkeleyGW WFN file

wfng_nvmin 
INTEGER 
Default: 
0

index of the lowest occupied band (normally = 1).
Not used if wfng_occupation = .FALSE.

wfng_nvmax 
INTEGER 
Default: 
0

index of the highest occupied band (normally = number of occupied bands).
Not used if wfng_occupation = .FALSE.

rhog_flag 
LOGICAL 
Default: 
.FALSE.

write charge density in Gspace to BerkeleyGW RHO file.
Only used for the GPP model in sigma code in BerkeleyGW

rhog_file 
STRING 
Default: 
'RHO'

name of BerkeleyGW RHO output file. Only used for the GPP model in sigma
code in BerkeleyGW. Not used if rhog_flag = .FALSE.

vxcg_flag 
LOGICAL 
Default: 
.FALSE.

write local part of exchangecorrelation potential in Gspace to
BerkeleyGW VXC file. Only used in sigma code in BerkeleyGW, it is
recommended to use vxc_flag instead

vxcg_file 
STRING 
Default: 
'VXC'

name of BerkeleyGW VXC output file. Only used in sigma code in BerkeleyGW,
it is recommended to use vxc_flag instead. Not used if vxcg_flag = .FALSE.

vxc0_flag 
LOGICAL 
Default: 
.FALSE.

write Vxc(G = 0) to text file. Only for testing, not required for BerkeleyGW

vxc0_file 
STRING 
Default: 
'vxc0.dat'

name of output text file for Vxc(G = 0). Only for testing, not required for
BerkeleyGW. Not used if vxc0_flag = .FALSE.

vxc_flag 
LOGICAL 
Default: 
.FALSE.

write matrix elements of exchangecorrelation potential to text file.
Only used in sigma code in BerkeleyGW

vxc_file 
STRING 
Default: 
'vxc.dat'

name of output text file for Vxc matrix elements. Only used in sigma code
in BerkeleyGW. Not used if vxc_flag = .FALSE.

vxc_integral 
STRING 
Default: 
'g'

'g'  'r'
'g' to compute matrix elements of exchangecorrelation potential in Gspace.
'r' to compute matrix elements of the local part of exchangecorrelation
potential in Rspace. It is recommended to use 'g'. Not used if vxc_flag = .FALSE.

vxc_diag_nmin 
INTEGER 
Default: 
0

minimum band index for diagonal Vxc matrix elements. Not used if vxc_flag = .FALSE.

vxc_diag_nmax 
INTEGER 
Default: 
0

maximum band index for diagonal Vxc matrix elements. Not used if vxc_flag = .FALSE.

vxc_offdiag_nmin 
INTEGER 
Default: 
0

minimum band index for offdiagonal Vxc matrix elements. Not used if vxc_flag = .FALSE.

vxc_offdiag_nmax 
INTEGER 
Default: 
0

maximum band index for offdiagonal Vxc matrix elements. Not used if vxc_flag = .FALSE.

vxc_zero_rho_core 
LOGICAL 
Default: 
.FALSE.

set to .TRUE. to zero out NLCC or to .FALSE. to keep NLCC when computing
exchangecorrelation potential. For details, see Hybertsen & Louie PRB 34,
5390 (1986), Eq. (38) and discussion afterwards

vscg_flag 
LOGICAL 
Default: 
.FALSE.

write local part of selfconsistent potential in Gspace to
BerkeleyGW VSC file. Only used in SAPO code in BerkeleyGW

vscg_file 
STRING 
Default: 
'VSC'

name of BerkeleyGW VSC output file. Only used in SAPO code in BerkeleyGW.
Not used if vscg_flag = .FALSE.

vkbg_flag 
LOGICAL 
Default: 
.FALSE.

write KleinmanBylander projectors in Gspace to BerkeleyGW VKB file.
Only used in SAPO code in BerkeleyGW

vkbg_file 
STRING 
Default: 
'VKB'

name of BerkeleyGW VKB output file. Only used in SAPO code in BerkeleyGW.
Not used if vkbg_flag = .FALSE.




This file has been created by helpdoc utility.
MeanField/ESPRESSO/version5.0/INPUT_bgw2pw.html
Input File Description
Program: bgw2pw.x / PWscf / Quantum Espresso

TABLE OF CONTENTS
INTRODUCTION
&INPUT_BGW2PW
prefix  outdir  real_or_complex  wfng_flag  wfng_file  wfng_nband  rhog_flag  rhog_file
INTRODUCTIONConverts BerkeleyGW WFN and RHO files to the format of pw.x.
This can be useful, for example, if you generate the plane waves
on top of the valence bands and want to diagonalize them in pw.x.
Look at the documentation for SAPO code in BerkeleyGW for more information.
bgw2pw.x reads common parameters from file prefix.save/datafile.xml and
writes files prefix.save/chargedensity.dat (charge density in Rspace),
prefix.save/gvectors.dat (Gvectors for charge density and potential),
prefix.save/K$n/eigenval.xml (eigenvalues and occupations for nth kpoint),
prefix.save/K$n/evc.dat (wavefunctions in Gspace for nth kpoint), and
prefix.save/K$n/gkvectors.dat (Gvectors for nth kpoint).
bgw2pw.x doesn't modify file prefix.save/datafile.xml so make changes to this
file manually (for example, you will need to change the number of bands if you
are using bgw2pw.x in conjunction with SAPO code in BerkeleyGW).
Structure of the input data:
============================
&INPUT_BGW2PW
...
/
Namelist: INPUT_BGW2PW 
prefix 
STRING 
Status: 
MANDATORY

prefix of files saved by program pw.x

outdir 
STRING 
Default: 
'./'

the scratch directory where the massive datafiles are written

real_or_complex 
INTEGER 
Default: 
2

1  2
1 for real flavor of BerkeleyGW (for systems with inversion symmetry and
timereversal symmetry) or 2 for complex flavor of BerkeleyGW (for systems
without inversion symmetry and timereversal symmetry)

wfng_flag 
LOGICAL 
Default: 
.FALSE.

read wavefunctions in Gspace from BerkeleyGW WFN file

wfng_file 
STRING 
Default: 
'WFN'

name of BerkeleyGW WFN input file. Not used if wfng_flag = .FALSE.

wfng_nband 
INTEGER 
Default: 
0

number of bands to write (0 = all). Not used if wfng_flag = .FALSE.

rhog_flag 
LOGICAL 
Default: 
.FALSE.

read charge density in Gspace from BerkeleyGW RHO file

rhog_file 
STRING 
Default: 
'RHO'

name of BerkeleyGW RHO input file. Not used if rhog_flag = .FALSE.




This file has been created by helpdoc utility.
MeanField/ESPRESSO/version4.3.2/pw2bgw.inp
&input_pw2bgw
prefix = 'silicon' ! same as in espresso
real_or_complex = 1 ! 1 for real or 2 for complex
wfng_flag = .true. ! write wavefunction in Gspace
wfng_file = 'WFN' ! wavefunction file name
wfng_kgrid = .false. ! overwrite kgrid in wavefunction file
wfng_nk1 = 4 ! ( if espresso input file contains the
wfng_nk2 = 4 ! manual list of kpoints, the kgrid
wfng_nk3 = 4 ! parameters in espresso are set to zero;
wfng_dk1 = 0.5 ! since Sigma and absorption both need to know
wfng_dk2 = 0.5 ! the kgrid dimensions, we patch these
wfng_dk3 = 0.5 ! parameters into the wavefunction file )
wfng_occupation = .true. ! overwrite occupations in wavefunction file
wfng_nvmin = 1 ! ( set min/max valence band indices; identical to
wfng_nvmax = 4 ! scissors operator for LDAmetal/GWinsulator )
rhog_flag = .true. ! write charge density in Gspace
rhog_file = 'RHO' ! charge density file name
vxcg_flag = .true. ! write exchangecorrelation potential in Gspace
vxcg_file = 'VXC' ! exchangecorrelation potential file name
vxc0_flag = .true. ! write Vxc(G=0)
vxc0_file = 'vxc0.dat' ! Vxc(G=0) file name
vxc_flag = .false. ! write matrix elements of Vxc
vxc_file = 'vxc.dat' ! Vxc matrix elements file name
vxc_integral = 'g' ! compute Vxc matrix elements in R or Gspace
vxc_diag_nmin = 0 ! min band index for diagonal Vxc matrix elements
vxc_diag_nmax = 0 ! max band index for diagonal Vxc matrix elements
vxc_offdiag_nmin = 0 ! min band index for offdiagonal Vxc matrix elements
vxc_offdiag_nmax = 0 ! max band index for offdiagonal Vxc matrix elements
input_dft = 'sla+pz' ! same as in espresso
exx_flag = .false. ! set to .true. for hybrids
vnlg_flag = .false. ! write KleinmanBylander projectors in Gspace
vnlg_file = 'KBproj' ! KleinmanBylander projectors file name
vxc_zero_rho_core = .true. ! NLCC: remove core charge when calculating Vxc
/
# above are typical values for Si. below are the defaults.
prefix = 'prefix'
real_or_complex = 2
wfng_flag = .false.
wfng_file = 'WFN'
wfng_kgrid = .false.
wfng_nk1 = 0
wfng_nk2 = 0
wfng_nk3 = 0
wfng_dk1 = 0.0
wfng_dk2 = 0.0
wfng_dk3 = 0.0
wfng_occupation = .false.
wfng_nvmin = 0
wfng_nvmax = 0
rhog_flag = .false.
rhog_file = 'RHO'
vxcg_flag = .false.
vxcg_file = 'VXC'
vxc0_flag = .false.
vxc0_file = 'vxc0.dat'
vxc_flag = .false.
vxc_file = 'vxc.dat'
vxc_integral = 'g'
vxc_diag_nmin = 0
vxc_diag_nmax = 0
vxc_offdiag_nmin = 0
vxc_offdiag_nmax = 0
input_dft = 'sla+pz'
exx_flag = .false.
vnlg_flag = .false.
vnlg_file = 'VNL'
vxc_zero_rho_core = .true.
MeanField/ESPRESSO/version4.3.2/bgw2pw.inp
&input_bgw2pw
prefix = 'silicon' ! same as in espresso
real_or_complex = 1 ! 1 for real or 2 for complex
wfng_flag = .true. ! read wavefunction in Gspace
wfng_file = 'WFN' ! wavefunction file name
wfng_nband = 0 ! number of bands to write (0 = all)
rhog_flag = .true. ! read charge density in Gspace
rhog_file = 'RHO' ! charge density file name
/
# above are typical values for Si. below are the defaults.
prefix = 'prefix'
real_or_complex = 2
wfng_flag = .false.
wfng_file = 'WFN'
wfng_nband = 0
rhog_flag = .false.
rhog_file = 'RHO'
MeanField/ESPRESSO/version4.3.2/README_patch
This directory contains a patch for espresso4.3.2 that fixes bugs
listed below. Assuming BerkeleyGW is installed in $BGWPATH/BerkeleyGW,
to apply the patch execute the following command
in the directory containing the espresso4.3.2 directory:
% patch p0 i $BGWPATH/BerkeleyGW/MeanField/ESPRESSO/version4.3.2/espresso4.3.2patch
espresso4.3.2 buglist:
(1) <<< THIS IS FIXED IN QUANTUM ESPRESSO 5.0 >>>
If you run espresso nscf calculation with startingwfc set to file
you may get error message saying "cannot read wfc : file not found".
This happens because subroutine verify_tmpdir in PW/input.f90 renames
datafile.xml to datafile.xml.bck and subroutines read_planewaves and
read_wavefunctions in PW/pw_restart.f90 try to read from datafile.xml
which doesn't exist. To get around this problem read from .xml.bck in
read_planewaves and read_wavefunctions if opening .xml returns error.
(2) <<< THIS IS FIXED IN QUANTUM ESPRESSO 5.0 >>>
Missing "CALL stop_pp ( )" at the end of PP/plan_avg.f90 causes it
to crash sometimes.
(3) <<< THIS IS FIXED IN QUANTUM ESPRESSO 5.0 >>>
Sometimes pw.x generates additional kpoints in a "nscf" run with an explicit
list of kpoints. If this is the case for your calculation, there are several
ways to go around this problem:
* Apply the patch provided with BerkeleyGW. It will prevent pw.x from
generating additional kpoints if they are provided explicitly, and
take care of the normalization of the weights of kpoints in a "bands"
calculation.
* Do not specify the atomic positions in the input file of kgrid.x (set
number of atoms = 0). Then pw.x will generate additional kpoints which
are the correct ones. Also set noinv = .true. in the input file of pw.x
if timereversal symmetry was not used in kgrid.x.
* Run a pw.x "bands" calculation instead of "nscf". In this case you have
to explicitly specify the occupations in the input file of pw2bgw.x (note
that this only works for insulators) and to normalize the weights of
kpoints to one in the input file of pw.x.
MeanField/ESPRESSO/patch_oldversions/README
This directory contains patches for older versions of espresso that fixes bugs
listed below. Assuming BerkeleyGW is installed in $BGWPATH/BerkeleyGW,
to apply the patch execute the following command
in the directory containing the espressoxxx directory:
% patch p0 i $BGWPATH/BerkeleyGW/MeanField/ESPRESSO/patch_oldversions/espressoxxxpatch
espresso buglist:
(1) <<< THIS IS FIXED IN QUANTUM ESPRESSO 5.0 >>>
If you run espresso nscf calculation with startingwfc set to file
you may get error message saying "cannot read wfc : file not found".
This happens because subroutine verify_tmpdir in PW/input.f90 renames
datafile.xml to datafile.xml.bck and subroutines read_planewaves and
read_wavefunctions in PW/pw_restart.f90 try to read from datafile.xml
which doesn't exist. To get around this problem read from .xml.bck in
read_planewaves and read_wavefunctions if opening .xml returns error.
(2) <<< THIS IS FIXED IN QUANTUM ESPRESSO 5.0 >>>
Missing "CALL stop_pp ( )" at the end of PP/plan_avg.f90 causes it
to crash sometimes.
PARATEC
MeanField/PARATEC/README
paratecSGL is available for download through SVN repository at
https://civet.berkeley.edu/svn/CODES/paratecSGL/
(only for authorized users).
To build with support for BerkeleyGW output, in arch.mk set the
line GWWFNPATH to BerkeleyGW/library, and add DBGW to M4OPTLIBS.
Full documentation for PARATEC 5.1.12:
https://drive.google.com/open?id=0BxTQVSgElL8nSnZDbEdhRVBoclE
Literature:
* B. G. Pfrommer, J. Demmel, and H. Simon, "Unconstrained Energy Functionals
for Electronic Structure Calculations," J. Comp. Phys. 150, 287 (1999).
* B. G. Pfrommer, M. Cote, S. G. Louie, and M. L. Cohen, "Relaxation of
Crystals with the QuasiNewton Method," J. Comp. Phys. 131, 233 (1997).
* Mathieu Taillefumier, Delphine Cabaret, AnneMarie Flank, and Francesco
Mauri, "Xray absorption nearedge structure calculations with
pseudopotentials: Application to the Kedge in diamond and alphaquartz,"
Phys. Rev. B 66, 195107 (2002).
* http://cmsn.drupalgardens.com/sites/cmsn.drupalgardens.com/files/CMSN_Newsletter_Vol2No2.pdf
[An older version was available at http://www.nersc.gov/projects/paratec.
PARATEC 5.1 produces only the old GW wavefunction format, incompatible
with this version of BerkeleyGW. However, you can convert them to the
new format using MeanField/Utilities/convert_old_to_new.x.]
The pseudopotentials for PARATEC can be generated with
the fhi98PP program which is available for download at
http://www.fhiberlin.mpg.de/th/fhi98md/fhi98PP/
1. Wrapper:
PARATEC output for BerkeleyGW is controlled by flags
gw_shift, gwc, gwr, gwscreening, gwcscreening, and vxc_matrix_elements.
The flags can be combined with an underscore: e.g. output_flags gwr_gwscreening
gwr and gwc are incompatible; gwscreening and gwcscreening are incompatible.
gw_shift q1 q2 q3
 generates qshifted grid, qvector is in crystal coordinates
in units of reciprocal lattice vectors (for WFNq, WFNq_fi)
This variable does the same job as the kgrid.x utility.
output_flags gwc
 writes complex wavefunctions in Gspace, for systems without
inversion symmetry about the origin, to file WFN (for all codes).
output_flags gwr
 writes real wavefunctions in Gspace, for systems with
inversion symmetry about the origin, to file WFN (for all codes)
output_flags gwscreening
 writes charge density in Gspace to file RHO,
exchangecorrelation potential in Gspace to file VXC,
and matrix elements of exchangecorrelation potential to file
vxc.dat (for Sigma). Real if possible and gwc not set, else complex.
output_flags gwcscreening
 like gwscreening, except forces complex even if real is possible.
vxc_matrix_elements diagmin diagmax offdiagmin offdiagmax
 specifies the range of bands for which to compute diagonal and
offdiagonal matrix elements of exchangecorrelation potential
(for Sigma, in conjunction with output_flags gwscreening)
Other key input flags:
pw_job {scf, nonselfcon}
 Use scf for initial calculation, nonselfcon for generating
BerkeleyGW outputs. (bandstructure does not seem to work)
energy_cutoff ecut
 Planewave cutoff for wavefunctions, in Ry.
number_kpoints
 set to 0 to use k_grid and reduce with symmetries
set to 1 to use k_grid and do not reduce with symmetries
set to any other number to read from file KPOINTS
k_grid nx ny nz
 3 integers specifying MonkhorstPack kgrid dimensions
k_grid_shift dx dy dz
 MonkhorstPack kgrid shifts (typically 0.0 or 0.5)
number_bands nb
 Number of bands to use in calculation. Fraction actually useful
or written to BerkeleyGW output determined by next variable.
eigspacefrac frac
 Fraction of bands to converge. Setting a higher number_bands
and lower eigspacefrac can make the calculation more efficient
depending on the diagonalization scheme. 0 < frac <= 1.0.
You can find the actual input files for PARATEC and BerkeleyGW in
examples/DFT, in PARATEC subdirectories for each example.
There are also bgw2para and rho2cd utilities that convert
BerkeleyGW files WFN and RHO to PARATEC format. This may be
useful, for example, if one generates the plane waves on top of the
valence and conduction bands (look into MeanField/SAPO/README for details)
and wants to diagonalize them further in PARATEC. There are no input
files; bgw2para takes as argument the wfn filename,
and it creates files WFN$n.$s and BAND needed for PARATEC.
Similarly, rho2cd requires file RHO and it creates file CD.
2. Utilities:

PARATEC: Kpoints

TOOL: kptlist.pl
Extracts a formatted list of kpoints from PARATEC file for use in the Sigma code

PARATEC: Qpoints

TOOL: qptlist.pl
Extracts a formatted list of qpoints from PARATEC file for use in the Epsilon code
SIESTA
MeanField/SIESTA/README
An example siesta2bgw calculation is provided in examples/DFT/benzene.
Please see this example for usage and example siesta and denchar input
files.
The SIESTA wrapper reads lattice vectors, atomic species and
positions from SystemLabel.XV file, kpoints from SystemLabel.KP
file, eigenvalues from SystemLabel.EIG file, wavefunctions in Rspace
from SystemLabel{.K$k}.WF$n{.UP.DOWN}{.REAL.IMAG}.cube file, symmetry
group, kinetic energy cutoffs for wavefunctions and charge density,
kgrid dimensions and offsets from the input file, generates reciprocal
lattice vectors, rotation matrices, fractional translations, FFT grid and
Gvectors, performs FFT from Rspace to Gspace, and writes the selected
range of bands to the output WFN file. Make sure to set up the real
space grid in SIESTA/DENCHAR utility same as FFT grid. If the two grids
differ the SIESTA wrapper will return an error. The output WFN file
can be used as an auxiliary wavefunction for the SAPO code.
A full GW/BSE calculation with SIESTA wavefunctions also requires RHO and
VXC files. These files can be generated from SystemLabel.RHO{.UP.DN}.cube,
SystemLabel.VT{.UP.DN}.cube and SystemLabel.VH.cube files obtained with
SIESTA/GRID2CUBE utility on SIESTA real space grid. Make sure SIESTA real
space grid is equivalent to FFT grid by setting parameter MeshCutoff
in SIESTA input file same as kinetic energy cutoff for charge density.
Input is read from file siesta2bgw.inp:
&input_siesta2bgw
systemlabel = 'ammonia'
wfng_output_file = 'wfng.lo'
rhog_output_file = ''
vxcg_output_file = ''
ecutwfn = 60.0
ecutrho = 240.0
wfng_nk1 = 1
wfng_nk2 = 1
wfng_nk3 = 1
wfng_dk1 = 0.0
wfng_dk2 = 0.0
wfng_dk3 = 0.0
wfng_ref_flag = .true.
wfng_ref_kpoint = 1
wfng_ref_spin = 1
wfng_ref_band = 4
wfng_ref_energy = 5.87
wfng_band_flag = .true.
wfng_band_min = 5
wfng_band_max = 28
wfng_energy_flag = .false.
wfng_energy_min = 0.0
wfng_energy_max = 0.0
wfng_gamma_real = .true.
/
Here, lattice vectors, kpoints, eigenvalues and wavefunctions in Rspace
are read from ammonia.XV, ammonia.KP, ammonia.EIG and ammonia.WF$n.cube
files, respectively. The kineticenergy cutoffs for
wavefunction and charge density are set to 60 and 240 Ry, respectively.
The kgrid is set to the Gammapoint. The SIESTA eigenvalues are shifted
in energy so that the HOMO state (1st kpoint, 1st spin, 4th band) appears
at 5.87 eV. The real parts of resonant SIESTA wavefunctions (from 5th to
28th band) are read from Gaussian Cube files and the imaginary parts are
set to zero (wfng_gamma_real), the wavefunctions are brought from Rspace
to Gspace and written to wfng.lo file. The SIESTA charge density and
exchangecorrelation potential files are not generated.
MeanField/SIESTA/patch/README
This directory contains a patch for denchar1.4 and
grid2cube1.1.1 in siesta3.1 that adds support
for nonorthogonal cells. The modified versions
generate Gaussian Cube files on realspace grid.
To apply the patch execute the following command
in the directory where you have siesta3.1:
% patch p0 i siesta3.1patch
Use the following keywords in siesta input file:
WaveFuncKPointsScale ReciprocalLatticeVectors
%block WaveFuncKPoints
0.000 0.000 0.000
%endblock WaveFuncKPoints
WriteDenchar T
SaveRho T
SaveElectrostaticPotential T
SaveTotalPotential T
MeshCutoff 240.0 Ry
Then run denchar for wavefunctions and grid2cube for
charge density and potentials. In denchar input file,
use keywords LatticeConstant and LatticeVectors same as
in siesta input file. Keywords Denchar.{MinMax}{XYZ}
will be ignored. In grid2cube input file, 3rd and 4th
lines (shift of the origin of coordinates and nskip)
will be ignored.
There is also irrbz.py that extracts irreducible wedge
from siesta output files prefix.KP and prefix.EIG.
This is needed because siesta reduces kpoints by
timereversal symmetry only.
MeanField/SIESTA/siesta2bgw.inp
&input_siesta2bgw
systemlabel = 'prefix' ! SystemLabel used in SIESTA calculation
wfng_output_file = 'wfng.lo' ! file to write generated wavefunctions to
rhog_output_file = 'rhog.lo' ! file to write rho(G) (blank = not written)
vxcg_output_file = 'vxcg.lo' ! file to write Vxc(G) (blank = not written)
ecutwfn = 0.0 ! kinetic energy cutoff (Ry) for wavefunctions
ecutrho = 0.0 ! kinetic energy cutoff (Ry) for rho and V
wfng_nk1 = 0 ! number of kpoints along b1
wfng_nk2 = 0 ! number of kpoints along b2
wfng_nk3 = 0 ! number of kpoints along b3
wfng_dk1 = 0.0 ! kgrid offset along b1 in units of b1/nk1
wfng_dk2 = 0.0 ! kgrid offset along b2 in units of b2/nk2
wfng_dk3 = 0.0 ! kgrid offset along b3 in units of b3/nk3
wfng_ref_flag = .false. ! shift eigenvalues to match reference state
wfng_ref_kpoint = 0 ! reference state kpoint index
wfng_ref_spin = 0 ! reference state spin index
wfng_ref_band = 0 ! reference state band index
wfng_ref_energy = 0.0 ! reference state eigenvalue (in eV)
wfng_band_flag = .false. ! choose a range of bands (0 = all)
wfng_band_min = 0 ! minimum band number
wfng_band_max = 0 ! maximum band number
wfng_energy_flag = .false. ! choose an energy window
wfng_energy_min = 0.0 ! minimum energy (in eV)
wfng_energy_max = 0.0 ! maximum energy (in eV)
wfng_gamma_real = .false. ! only real wavefunctions at the Gamma point
/
Octopus
MeanField/Octopus/README
Octopus is a realspace DFT and TDDFT code available under the GPL from http://www.tddft.org/programs/octopus.
Support for BerkeleyGW output is implemented in Octopus since version 4.1.0.
For more information, see the Octopus wiki: http://www.tddft.org/programs/octopus/wiki/index.php/BerkeleyGW.
EPM
MeanField/EPM/README
================================================================================
"The empirical pseudopotential method never lost a battle to experiment."
 Marvin Cohen, 2008
================================================================================
EPM stands for Empirical Pseudopotential Method. It is the planewave
part of TBPW1.1 modified to generate input files for BerkeleyGW.
Numerous other corrections and improvements have also been made
by Georgy Samsonidze and David Strubbe.
TBPW1.1 is available for download from http://www.mcc.uiuc.edu/software/
The input variables are described in epm.inp. Most variables are common to both
epm.x and epm2bgw.x, but a few are just for one or the other.
The real wavefunctions for systems with inversion symmetry
are obtained by applying the GramSchmidt process adapted from
paratecSGL1.1.3/src/para/gwreal.f90
To plot silicon band structure calculated with the EPM executable:
% epm.x < siliconepm.in
% gnuplot bands.plt
Uses variables NumberOfLines, NumberOfDivisions, KPointsAndLabels.
To use the explicit form factors and to compute the band gap and
the effective masses (mass only correct for Silike band structures):
% epm.x < siliconepmff.in
Uses variables gap_flag, gap_file, mass_flag, mass_file.
To generate input files for BerkeleyGW executable, use epm2bgw.x.
The scripts epm2bgw_cplx.sh and epm2bgw_cplx_spin.sh are just for use
with the testsuite. They force complex, or complex and spinpolarized.
You can find the actual input files for GW/BSE calculations on top of
EPM in directory examples/EPM, as well as testsuite directories
SiEPM, GaAsEPM, and SiWireEPM.
The kpoints for the EPM input files can be generated using utility
kgrid. Please refer to file MeanField/ESPRESSO/README for details.
Virtual crystal approximation (VCA) is implemented within EPM in
the form of two preprocessing scripts, ff2vq.py and vca.py.
ff2vq.py reads EPM form factors from file form_factors.dat, fits them
to the chosen functional form of the V(q) potential, writes potential
coefficients to file v_of_q.dat, and writes new form factors computed
from V(q) to file vq2ff.dat. This procedure is described by equations
(8) and (9) and the accompanying text in Phys. Rev. B 84, 195201 (2011).
vca.py reads V(q) potential coefficients from file v_of_q.dat,
employs the virtual crystal approximation to compute hybrid form
factors, and writes them to file vca_ff.dat. The potential mixing
is controlled by identifiers host_material and doping_level hard
coded below. This procedure is described by equations (10)  (12)
and the accompanying text in Phys. Rev. B 84, 195201 (2011).
The original form factors from file form_factors.dat or the hybrid
ones from file vca_ff.dat can be fed to epm.x and epm2bgw.x using
keywords LatticeConstant and FormFactors in the input file for epm.x
or epm2bgw.x. See example of using these keywords in siliconepmff.in.
The README file of TBPW1.1 follows below.
================================================================================
TBPW1.1
========================================
Author/Affiliation

* Dyutiman Das, UIUC
* William Mattson, UIUC
* Nichols A. Romero, UIUC
* Richard M. Martin, UIUC
========================================
Author email

nromero@uiuc.edu
========================================
Description of Software

TBPW is an electronic structure code primarily intended for pedagogical purposes. It is written from the groundup in a modular style using Fortran 90. This code is composed of two distinct parts: a tightbinding (TB) and plane wave (PW). Additionally, there is a plane wave density (PWD) code which outputs the electron density on a grid.
The main characteristics of these codes are:
* Readily provides band structure plots
* TB implemented using a rotation matrix formalism allows the use of orbitals with arbitrary angular momentum l
* PW implemented using the option of diagonalisation via directinversion
Send email to tbpwsubscribe@mcc.uiuc.edu to subscribe to the mailing list, tbpw@mcc.uiuc.edu
This material is based upon work supported by the NSF under Award No. 9976550 and the DOE under Award No. DEFG96ER45439
========================================
Submitted

20030131
========================================
Disclaimer

Developed by Electronic Structure Group, University of Illinois, Department of Physics, http://www.physics.uiuc.edu/research/ElectronicStructure/ Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the Software), to deal with the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
* Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimers.
* Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimers in the documentation and/or other materials provided with the distribution.
* Neither the names of the Electronic Structure Group, the University of Illinois, nor the names of its contributors may be used to endorse or promote products derived from this Software without specific prior written permission.
THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE CONTRIBUTORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS WITH THE SOFTWARE.
========================================
Copyright

University of Illinois Open Source License. Copyright (c) 2003, University of Illinois Board of Trustees. All rights reserved. Users also agree to terms of License Agreement available at http://www.mcc.uiuc.edu/
================================================================================
MeanField/EPM/epm.inp
# BEWARE: you cannot comment out tags with # in EPM!
# any occurrence of a tag will be found and read.
# These parameters are only for epm2bgw
real_or_complex 1 # 1 = real, 2 = complex. default is complex.
wfng_flag T # Write wavefunction file (default F)
wfng_file WFN # filename (default WFN)
KPointNumbers 5 5 5 # number of points in kgrid in each direction b1,b2,b3
KPointOffsets 0.5 0.5 0.5 # shift in kgrid in each direction b1,b2,b3
rhog_flag T # Write density file (default F)
rhog_file RHO # filename (default RHO)
vxcg_flag T # Write xc potential file (default F)
vxcg_file VXC # filename (default VXC)
vxc_flag T # Write VXC matrix element file (default F)
vxc_file vxc.dat # filename (default vxc.dat)
vxc_diag_nmin 1 # minimum band to write diagonal matrix elements (default 0)
vxc_diag_nmax 8 # maximum band to write diagonal matrix elements (default 0)
vxc_offdiag_nmin 1 # minimum band to write offdiagonal matrix elements (default 0)
vxc_offdiag_nmax 8 # maximum band to write offdiagonal matrix elements (default 0)
disable_symmetries F # set to true to write no symmetries (default F)
# EPM does not really do a spinpolarized calculation, but can write two copies of everything
nspin 1 # set to 2 to write spin up and spin down (default 1)
temperature 0 # kT in eV for FermiDirac distribution occupations (only for BerkeleyGW output)
# The Fermi level is taken as the midgap.
FFTGrid 16 16 16 # FFT grid for computing density (default: determined from EnergyCutoff)
# These parameters are common to epm and epm2bgw
# Set form factor parameters (see examples in form_factors.dat),
# or default is to use V(q) from atomPotentialMod (only for H,Si,Ga,As)
FormFactors 5.43 0.224 0.055 0.072 0.000 0.000 0.000
# 0 is for LAPACK direct diagonalization (default), 1 for conjugate gradients
DiagonalizationSwitch 0
AbsoluteTolerance 1.0d0 # tolerance for LAPACK diagonalization (default 1 means machine precision)
CGIterationPeriod 5 # parameter for conjugate gradients (default 3)
CGTolerance 1.0d10 # parameter for conjugate gradients (1d5)
InputEnergiesInEV # Rydberg is default
EnergyCutoff 11.0 # planewave cutoff for the potential (and wavefunctions)
NumberOfDimensions 3 # dimensionality of the system (default 3)
#LatticeConstantFormat Angstrom # default is Bohr
LatticeConstant 10.2612 # in units given above
LatticeVectors # in lattice constant units
0.0 0.5 0.5
0.5 0.0 0.5
0.5 0.5 0.0
NumberOfAtoms 2
NumberOfSpecies 1
ChemicalSpeciesLabel # index, atomic number, name
1 14 Si
AtomicCoordinatesFormat ScaledByLatticeVectors
# ScaledByLatticeVectors (default) is in crystal coordinates
# ScaledCartesian is in units of LatticeConstant
AtomicCoordinatesAndAtomicSpecies # x, y, z, species index
0.125 0.125 0.125 1
0.125 0.125 0.125 1
NumberOfBands 8
NumberOfOccupiedBands 4 # note: you cannot really do metals, occupations are fixed.
KPointsScale ReciprocalLatticeVectors # default; other choice is TwoPi/a
KPointsList 19 # number of kpoints supplied
0.100000000 0.100000000 0.100000000 2.0 # kx, ky, kz, weight (will be renormalized)
0.100000000 0.100000000 0.300000000 6.0
0.100000000 0.100000000 0.500000000 6.0
0.100000000 0.100000000 0.700000000 6.0
0.100000000 0.100000000 0.900000000 6.0
0.100000000 0.300000000 0.300000000 6.0
0.100000000 0.300000000 0.500000000 12.0
0.100000000 0.300000000 0.700000000 12.0
0.100000000 0.300000000 0.900000000 12.0
0.100000000 0.500000000 0.500000000 6.0
0.100000000 0.500000000 0.700000000 12.0
0.100000000 0.500000000 0.900000000 6.0
0.100000000 0.700000000 0.700000000 6.0
0.300000000 0.300000000 0.300000000 2.0
0.300000000 0.300000000 0.500000000 6.0
0.300000000 0.300000000 0.700000000 6.0
0.300000000 0.500000000 0.500000000 6.0
0.300000000 0.500000000 0.700000000 6.0
0.500000000 0.500000000 0.500000000 1.0
# for plotting band structures; used if KPointsList not present
NumberOfLines 4
NumberOfDivisions 20
KPointsScale ReciprocalLatticeVectors # same as above
KPointsAndLabels
0.500 0.500 0.500 L
0.000 0.000 0.000 G
0.500 0.000 0.500 X
0.625 0.250 0.625 U
0.000 0.000 0.000 G
# These parameters are only for epm. Find band gap and effective mass.
gap_flag F # default F
gap_file bandgap.dat # filename (default gap.dat)
# Note: effective mass only works for siliconlike band structures.
mass_flag F # default F
mass_file effmass.dat # filename (default mass_file)
ICM
MeanField/ICM/README
This program reads the wavefunction file in Gaussian Cube or
XCrySDen XSF format and computes the Coulomb integral within
an image charge model.
It is based on Visual/surface.cpp code, so the input parameter
file formats for the two codes are quite similar. You can find
more details in Visual/surface.inp and Visual/README.
Input is read from file icm.inp:
inputfilename C6H6.b_15.cube
inputfileformat cube
threshold 0.99
threshold_power 1
coulomb_power 1
mirrorplaneorigin
0.0 0.0 2.0
mirrorplanenormal
0.0 0.0 1.0
mirrorplaneunit angstrom
uc F
uco
0.0 0.0 0.0
ucv
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
ucu latvec
sc T
sco
0.5 0.5 0.5
scv
1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
scu latvec
In the above example, the HOMO wavefunction of benzene is read from
Gaussian Cube file. The wavefunction is placed in the center of the
supercell (see the meaning of uc, uco, ucv, ucu, sc, sco, scv, scu
parameters in Visual/surface.cpp). The parts of the wavefunction
outside an isosurface that contains 99% of the charge density are
dropped (parameters threshold and threshold_power have the same
meaning as isovalue and power in Visual/surface.cpp). Parameter
coulomb_power tells the code whether the wavefunction in the Coulomb
integral needs to be squared. Set both powers to 1 if the wavefunction
file contains the squared amplitude as produced by ESPRESSO, and to 2
for the linear amplitude as in PARATEC or SIESTA. The mirror plane
is defined by parameters mirrorplaneorigin and mirrorplanenormal,
in the above example it is parallel to the xy plane crossing the
z axis at 2 Angstrom. Problems may occur with nonorthogonal unit cells;
use of cubic cells is recommended.
Utilities
MeanField/Utilities/README
================================================================================
MeanField utilities
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014)
Version 1.0 (July, 2011) J. Deslippe, M. Jain, D. A. Strubbe, G. Samsonidze
Version 0.5 J. Deslippe, D. Prendergast, L. Yang, F. Ribeiro, G. Samsonidze (2008)
Version 0.2 M. L. Tiago, C. Spataru, S. IsmailBeigi (2004)

degeneracy_check.x:
Determine which numbers of bands do not break a degenerate subspace, to avoid warnings or
errors in Epsilon, Sigma, and BSE codes. Supply any number of binary WFN files as command
line arguments, and a list of numbers that are acceptable for all kpoints in all files
will be written to standard output.
mf_convert_wrapper.sh:
Converts WFN/RHO/VXC files between binary and ASCII. This can be useful for moving files
correctly between big and littleendian machines, or for examining the contents of a
file (though wfn_rho_vxc_info.x is more userfriendly). Using this wrapper script, the
flavor (real/complex), type of file, and binary/ASCII is determined automatically.
wfn_rho_vxc_info.x:
Prints the contents of the header of a (binary) WFN, RHO, or VXC file in a clearly labeled
format, for human inspection.
wfnmerge.x:
Merges many WFN files into one. It assumes that all input files have the
same number and same ordering of Gvectors for the charge density.
The number and name of input files is read from "wfnmerge.inp", as well
is the kgrid and the kshift.
FAQ:
Q: Why would someone want to use this?
A: For example, maybe to do a converged calculation one needs to include a
large number of unoccupied states. There may not be the resources (either
CPUs or wallclock) to do this all in one shot, so it may be beneficial to
split up a calculation of kpoints (e.g., 4x4x4 MP grid) into smaller
pieces and then add up the final wavefunctions.
Q: What is the deal with kpoint weights?
A: Quantum Espresso renormalizes the kpoint weights that you use in a
calculation. If you are splitting up a MP grid into different groups of
kpoints, and each group is normalized, the relative weights between
kpoints in different groups is no longer correct. BerkeleyGW does not
make use of the kpoint weights (except for in determining whether the
Fermi level is reasonable for metallic calculations). So for most uses
the fact that these weights are incorrect does not matter. If it matters
to you, you can simply modify wfnmerge to read in the kpoint weights of
your choosing.
scissors2eqpx:
Write an eqp.dat file based on a WFN file and scissors parameters.
For testing equivalence of eqp_corrections and scissors.
fix_occ.x:
Fix the occupations of a WFN file. This is useful if you have split a
large nscf calculations into smaller blocks, and you want the occupations
of the merged WFN to be consistent. It can also fix inconsistencies in the
occupations and in the Fermi energy for metals.
wfn_dotproduct.x
Form the overlap between the bands in two wavefunction files.
If they are copies of the same file, you can check orthonormality.
Only bands at corresponding kpoints are considered, since
the overlap is zero by Bloch`s theorem if the kpoints differ.
Warning: a known issue is possible errors from gmap when relating
kpoints by symmetry operations.
wfn_time_reversal.x
Give a WFN file with a kgrid reduced by timereversal symmetry, and get a new WFN file
with the kpoints unfolded by timereversal symmetry. This can save up to half the time
in the nonselfconsistent DFT calculations for input to BerkeleyGW. It can be used
after calculating kpoints from kgrid.x with timereversal symmetry, or from 'automatic'
kpoints from Quantum ESPRESSO. Either use VXC, or add the new kpoints to vxc.dat by
hand according to the rule = *.
MeanField/Utilities/wfnmerge.inp
WFN.OUT ! name of output WFN file
4 4 4 ! final kgrid
0.0 0.0 0.0 ! final kshift
2 ! total number of input WFN files
8 ! total number of kpoints in all input WFN files
wfn_0/WFN ! name of 1st input WFN file
wfn_1/WFN ! name of 2nd input WFN file
1.0 ! relative weight of 1st input WFN file
1.0 ! relative weight of 2nd input WFN file
MeanField/Utilities/fix_occ.inp
WFN_in ! input WFN
WFN_out ! output WFN
0 ! original number of electrons in WFN_in (0 to autodetect)
0 ! extra charge to put in
MeanField/Utilities/wfnmix_QSGW.inp
1.032 !slope
0.117 !intercept
SAPO
MeanField/SAPO/README
The SAPO (Simple Approximate Physical Orbitals) code reads DFT
wavefunctions from WFN file, generates additional wavefunctions on top
of them, and writes both to another WFN file. It can generate plane waves,
decompose them into irreducible representations of the space group of the
Bravais lattice or the crystal, and orthonormalize them with respect to
DFT wavefunctions. It can insert SIESTA wavefunctions read from auxiliary
WFN file in between plane waves and orthonormalize them altogether.
It can apply a random variation to the plane waves and SIESTA wavefunctions
or generate completely random wavefunctions and orthonormalize them.
It can correct eigenvalues during orthonormalization or perform subspace
diagonalization. It can drop the plane waves and SIESTA wavefunctions which
have a large overlap with DFT wavefunctions. It can keep wavefunctions with
the eigenvalues in a given energy range. It can extract eigenvalues and
plane wave coefficients from WFN file for plotting purposes.
The SAPO method is described in Phys. Rev. Lett. 107, 186404 (2011)
20121224 (gsm): added iterative Davidson diagonalization, see
examples/DFT/Si2_sapo for details
Note: When using iterative Davidson diagonalization, it is recommended
to generate VXC rather than vxc.dat in pw2bgw run, then compute matrix
elements of VXC between SAPO wavefunctions. This is more consistent
than using vxc.dat because vxc.dat is computed between ESPRESSO
wavefunctions, and those might be slightly different from SAPO
wavefunctions.
Note: wfng_input_file should contain all occupied orbitals (and may or
may not contain some unoccupied orbitals), otherwise occupations written
to wfng_output_file will be wrong.
Note: Symmetrization of planewaves (sapo_symmetry .gt. 0) has no effect
on a SAPO calculation since it is just a linear combination in a degenerate
subspace. It is currently disabled because it requires symmetry subroutines
from Quantum ESPRESSO which are not compatible with our BSDtype license.
To enable it: open MeanField/SAPO/pw.f90; comment out 'die'; uncomment
calls to 's_axis_to_cart', 'find_group', 'set_irr_rap', and 'divide_class';
and link the corresponding routines from Quantum ESPRESSO.
Input is read from file sapo.inp:
&input_sapo
wfng_input_file = 'wfng.in'
wfng_aux_file = ''
vlr_input_file = ''
vnlg_input_file = ''
wfng_output_file = 'wfng.out'
sapo_band_number = 0
sapo_planewave_min = 1
sapo_planewave_max = 580
sapo_energy_shift = 0.0
sapo_energy_match = .true.
sapo_symmetry = 2
sapo_print_ir = .true.
aux_flag = .false.
aux_band_min = 0
aux_band_max = 0
aux_energy_shift = 0.0
sapo_random = 1
sapo_random_ampl = 1.0d3
sapo_overlap_flag = .false.
sapo_overlap_max = 0.0
sapo_orthonormal = .true.
sapo_ortho_block = 0
sapo_ortho_order = 0
sapo_ortho_energy = .false.
sapo_energy_sort = .false.
sapo_hamiltonian = .false.
sapo_energy_range = .false.
sapo_energy_min = 0.0
sapo_energy_max = 0.0
sapo_check_norm = .false.
sapo_plot_kpoint = 0
sapo_plot_spin = 0
sapo_plot_bandmin = 0
sapo_plot_bandmax = 0
sapo_plot_pwmin = 0
sapo_plot_pwmax = 0
sapo_eigenvalue = .false.
sapo_projection = 0
sapo_amplitude = .false.
sapo_ampl_num = 0
sapo_ampl_del = 0.0
sapo_ampl_brd = 0.0
/
Here, the wavefunctions are read from wfng.in file and written to wfng.out
file. The parameter sapo_band_number specifies the number of wavefunctions
to be read from the input file (if set to 0 all the wavefunctions will be
read). The PW wavefunctions will be constructed from plane waves ranging
from 1 (sapo_planewave_min) to 580 (sapo_planewave_max). The kinetic energies
of the plane waves will be shifted to match the eigenvalues of the input
wavefunctions (sapo_energy_match). The plane waves will be decomposed into
irreducible representations of the space group of the Bravais lattice
(sapo_symmetry = 2), and the decomposition will be printed to the standard
output (sapo_print_ir). The auxiliary wavefunctions will not be inserted in
between the PW wavefunctions (aux_flag). A random variation (sapo_random = 1)
with a small amplitude (sapo_random_ampl = 1.0d3) will be added to the PW
wavefunctions. The PW wavefunctions will be orthonormalized with respect
to the valence and conduction bands (sapo_orthonormal) in the ascending
order with respect to energy (sapo_ortho_order = 0). The orthonormality
check will not be performed (sapo_check_norm). The energy eigenvalues,
the projections of wavefunctions onto plane waves, and the squared
absolute values of amplitudes of the wavefunctions with respect to
kinetic energies of plane waves will not be plotted (sapo_eigenvalue,
sapo_projection, and sapo_amplitude).
The auxiliary wavefunctions can be used in two different ways:
(1) Use hybrid valence wavefunctions as wfng_input_file and LDA conduction
wavefunctions as wfng_aux_file, set sapo_planewave_min and sapo_planewave_max
to 0, sapo_orthonormal to .true. and sapo_ortho_block to 1. This way you will
get a good starting point for nscf hybrid calculations in PARATEC or ESPRESSO.
(2) Use Siesta wrapper in MeanField/SIESTA directory to construct the resonant
states from SIESTA wavefunctions (look into MeanField/SIESTA/README for details
on how to use siesta2bgw), feed them to the SAPO code through wfng_aux_file
parameter, and generate continuum states from plane waves by setting
sapo_planewave_max to a large number. Also set sapo_orthonormal,
sapo_ortho_block and sapo_ortho_order to orthonormalize the resonant and
continuum states in the desired order with respect to the bound states
from PARATEC or ESPRESSO. You may want to set sapo_ortho_energy to .true.
for correcting the eigenvalues during the orthonormalization, or to set
sapo_hamiltonian to .true. for correcting both the wavefunctions and
the eigenvalues by diagonalizing the Hamiltonian. The latter requires
the local potential file in Gaussian Cube format (vlr_input_file),
and optionally the nonlocal pseudopotential file (vnlg_input_file).
Finally, set sapo_energy_sort to .true. for sorting the resonant
and continuum states by their eigenvalues before writing them
to the output file. You may also set sapo_energy_range to .true.
for keeping the eigenvalues in the energy range of sapo_energy_min
to sapo_energy_max.
PS: If you don't know or don't want to manually set
sapo_planewave_min and aux_band_min, set both of them to 1,
sapo_overlap_flag to .true., and sapo_overlap_max to 0.3.
This will automatically throw away planewaves and auxiliary
states that have large overlaps (scalar products > 0.3)
with DFT states and among themselves. The scalar products
of PW and AUX states with DFT states will be written to files
overlap_dft_k#_s#.dat, and the scalar products of PW states
with AUX states will be written to files overlap_aux_k#_s#.dat,
so you can inspect what states were thrown away after the run.
This useful feature is informally known as boverlap, in honor
of Brad Malone who first proposed this idea. This confirms
once again that evolution is a byproduct of laziness.
MeanField/SAPO/sapo.inp
&input_sapo
wfng_input_file = 'wfng.pw' ! file to read DFT wavefunctions from
wfng_aux_file = 'wfng.lo' ! file to read auxiliary wavefunctions from
vscg_input_file = 'vscg' ! file to read selfconsistent potential
! in Gspace for constructing Hamiltonian
! (generated by pw2bgw.x from Quantum ESPRESSO)
vkbg_input_file = 'vkbg' ! file to read KleinmanBylander projectors
! in Gspace for constructing Hamiltonian
! (generated by pw2bgw.x from Quantum ESPRESSO)
wfng_output_file = 'wfng' ! file to write generated wavefunctions to
sapo_band_number = 1 ! number of DFT wavefunctions to use
! (set to 1 to use all wavefunctions found
! in wfng_input_file)
sapo_planewave_min = 0 ! indices of the lowest and highest plane waves
sapo_planewave_max = 0 ! to construct wavefunctions from (0 = none)
sapo_energy_shift = 0.0 ! energy shift for PW kinetic energies (eV)
sapo_energy_match = .false. ! match PW kinetic energies & DFT eigenvalues
sapo_symmetry = 0 ! 0 = none, 1 = crystal, 2 = lattice symmetry
sapo_print_ir = .false. ! print irreducible representations of PW
aux_flag = .false. ! insert auxiliary wavefunctions in between PW
aux_band_min = 0 ! minimum and maximum band indices of auxiliary
aux_band_max = 0 ! wavefunctions (0 = all)
aux_energy_shift = 0.0 ! energy shift for auxiliary eigenvalues (eV)
sapo_random = 0 ! 0 = none, 1 = small variation, 2 = random
sapo_random_ampl = 0.0 ! amplitude of the random variation
sapo_random_norm = .false. ! normalize randomized wavefunctions
sapo_overlap_flag = .false. ! drop PW & AUX states that have a large
! overlap with DFT states
sapo_overlap_max = 0.0 ! maximum scalar product of PW & AUX with DFT
sapo_orthonormal = .false. ! orthonormalize PW & AUX wavefunctions wrt DFT
sapo_ortho_block = 0 ! 0 = PW/AUX, 1 = AUX/PW, 2 = ordered in energy
sapo_ortho_order = 0 ! 0 = ascending, 1 = descending order in energy
sapo_ortho_energy = .false. ! correct PW & AUX eigenvalues by perturbative
! approach during orthonormalization
sapo_energy_sort = .false. ! sort PW & AUX eigenvalues in ascending order
sapo_hamiltonian = .false. ! correct PW & AUX wavefunctions & eigenvalues
! by diagonalizing subspace Hamiltonian or
! by iteratively diagonalizing Hamiltonian
sapo_ham_nrestart = 0 ! maximum number of iterative diagonalization
! restarts with random initial wavefunctions
! for kpoints/spins that didn`t converge
sapo_ham_ndiag = 0 ! maximum number of subspace diagonalization
! steps during iterative diagonalization
sapo_ham_ndim = 0 ! maximum number of basis functions in units
! of the total number of wavefunctions,
! same as diago_david_ndim in pw.x
sapo_ham_tol = 0.0 ! tolerance on the maximum norm of residual
! vectors for iterative diagonalization (Ry)
sapo_ham_resinc = .false. ! stop expanding the basis set and update
! the wavefunctions when the maximum norm
! of residual vectors starts increasing
sapo_do_all_bands = .false. ! orthonormalize and diagonalize DFT & PW & AUX
! wavefunctions or PW & AUX wavefunctions
sapo_energy_range = .false. ! keep PW & AUX eigenvalues in the energy range
sapo_energy_min = 0.0 ! minimum energy for PW & AUX eigenvalues (Ry)
sapo_energy_max = 0.0 ! maximum energy for PW & AUX eigenvalues (Ry)
sapo_check_norm = .false. ! check orthonormality of wavefunctions
sapo_plot_kpoint = 0 ! kpoint index (0 = all) for output plots
sapo_plot_spin = 0 ! spin index (0 = all) for output plots
sapo_plot_bandmin = 0 ! range of band indices (0 = all) for output
sapo_plot_bandmax = 0 ! eigenvalue, projection and amplitude plots
sapo_plot_pwmin = 0 ! range of plane wave indices (0 = all) for
sapo_plot_pwmax = 0 ! output projection plots
sapo_eigenvalue = .false. ! plot energy eigenvalues wrt band index (eV)
sapo_projection = 0 ! plot projections of wavefunctions onto PW
! 0 = none, 1 = wrt band, 2 = wrt PW, 4 = both
sapo_amplitude = .false. ! plot squared absolute values of amplitudes of
! wavefunctions wrt kinetic energies of PW (Ry)
sapo_ampl_num = 0 ! number of points on the kinetic energy scale
sapo_ampl_del = 0.0 ! separation between the adjacent points (Ry)
sapo_ampl_brd = 0.0 ! broadening for wavefunction amplitudes (Ry)
/
Epsilon
Epsilon/README
================================================================================
GW code: Epsilon
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014)
Version 1.0 (September, 2011)
Version 0.5 J. Deslippe, D. Prendergast, L. Yang, F. Ribeiro, G. Samsonidze (2008)
Version 0.2 S. IsmailBeigi (2002)
Version 0.1 G. M. Rignanese, E. Chang, X. Blase (1998)
Version 0.0 M. Hybertsen (1985)

Description:
Epsilon is the name of the code that generates the polarizability matrix and
the inverse dielectric matrix for a bulk or nanoscale system. The main
result of the Epsilon code is the generation of epsmat which can be used
in a Sigma or BSE calculation.

Required Input:
epsilon.inp Input parameters. See example in this directory.
WFN This is linked to the unshifted grid.
WFNq This is linked to the shifted grid.
When a shift is not required, this can be the same file as WFN
(i.e. for semiconductor or graphene screening with spherical or
box truncation).
Auxiliary Files: (output files from previous runs, used as input to speed up calculation)
chimat(.h5) Polarizability matrix. No need to recalculate matrix elements.
The file has a ".h5" extension if the code is compiled
with HDF5 support (for specification, see epsmat.h5.spec).
chi0mat(.h5) Polarizability matrix at q=0. No need to recalculate matrix elements.
The file has a ".h5" extension if the code is compiled
with HDF5 support (for specification, see epsmat.h5.spec).
The files below are used if eqp_corrections is set in epsilon.inp.
The corrected eigenvalues are used for constructing the polarizability matrix.
eqp.dat A list of quasiparticle energy corrections for the bands in WFN.
eqp_q.dat A list of quasiparticle energy corrections for the bands in WFNq.

epsilon.inp Please see example epsilon.inp in this directory
for more complete options.

Output Files:
espmat(.h5) Inverse dielectric matrix (q<>0).
The file has a ".h5" extension if the code is compiled
with HDF5 support (for specification, see epsmat.h5.spec).
eps0mat(.h5) Inverse dielectric matrix (q>0).
The file has a ".h5" extension if the code is compiled
with HDF5 support (for specification, see epsmat.h5.spec).
epsilon.log The log file containing values of chimat and epsmat.
chi_converge.dat Convergence chi with respect to empty orbitals.
Columns: number of conduction bands,
Re chi(G=0,G'=0,q), extrapolated Re chi(G=0,G'=0,q),
Re chi(G=Max,G'=Max,q), extrapolated Re chi(G=Max,G'=Max,q)

Tricks and hints:
1. Comments on convergence of epsilon_cutoff
There are many ways to check convergence of this parameter:
Ideally, the inverse epsilon matrix is almost diagonal for large
G vectors. So, for small ( G , G' ) it is usually large but, for
G or G' close to G_max, it is smaller by a factor of 100, more or
less. Check if the ratio between the greatest and lowest matrix
elements is large.
Alternatively, one can check if epsinv( G_max , G'_max ) is close
to 1 for G = G'. It should be about 0.999 or so. Note that 0.9 isn't
"close to 1" in this case! Also check if epsinv( G_max , G'_max ) is
indeed small for G != G'.
2. Comments on the null point
The null point (Gamma point) is treated in a special manner. In input,
declare it as slightly shifted from the true Gamma point:
0.0000 0.0050 0.0050 1.0 1
This is the only kpoint with itestq=1 instead of 0!
The magnitude of this vector must be small: of order 0.01 or smaller
for less than 100 points in full Brillouin zone. If the density
of points in BZ increases, that magnitude should decrease accordingly.
3. Macroscopic dielectric constant
In GW, the screened interaction gives the formula:
eps(G,G';q) = delta_GG'  [4pi e^2/(q+G)^2 ] chi(G,G';q)
where chi(G,G';q) is the polarizability (P in the literature).
if G=0 and q=0
then we let
eps(0,G';q) = delta_0G'  lim(q>0) [4pi/(q)^2] chi(0,G';q)
always check chi(0,0;q>0) and eps^(1)(0,0;q>0)
in epsilon.log
1/eps^(1)(0,0;q>) should be the macroscopic epsilon,
localfield effects
1/(14*pi*chi(0,0;q>0)*V_c(0) should also be (roughly) the macroscopic epsilon,
without localfield effects
4. Merging epsmat files
The Epsilon code is complicated and involves lots of computation. Frequently,
it is useful to calculate the epsilon matrix for a small number
of qpoints, and later put all those pieces of matrix into a big
epsmat file. Use epsmat_merge to do this. See below.
Notes:
epsilon cutoff *must* be the same for all files to be merged
The ordering of qpoints in the input file must correspond to the ordering of
epsmat files: the first file has the first set of points, the second file
has the second and so on.
5. Utilities:

Epsilon: binary/ASCII conversion

TOOLS: epsbinasc, epsascbin
USAGE: simply type name of executable.
The input file is named epsconv.inp.

 WFN/RHO/VXC binary/ASCII conversion 

TOOL: mf_convert_wrapper.sh
USAGE: ./mf_convert_wrapper.sh infile outfile
Real/complex, bin/asc, file type, is automatically detected.

Epsilon: merging

TOOL: epsmat_merge
USAGE: epsmat_merge.[real/cplx].x
Merges the content of multiple binary epsmat files into an output file
called 'epsmat', concatenating data. See input file epsmat_merge.inp,
for specification of cutoff, qpoints, and input epsmat filenames.
Epsilon/epsilon.inp
# epsilon.inp
# Energy cutoff for the dielectric matrix, in Ry. The dielectric matrix
# epsilon_{GG`} will contain all Gvectors with kinetic energy q+G^2
# up to this cutoff.
epsilon_cutoff 35.0
# Total number of bands (valence+conduction) to sum over. Defaults to the
# number of bands in the WFN file minus 1. FIXME: consider degeneracy.
#number_bands 1000
# Specify the Fermi level (in eV), if you want implicit doping
# Note that value refers to energies AFTER scissor shift or eqp corrections.
#fermi_level 0.0
# The Fermi level is treated as an absolute value
# or relative to that found from the mean field (default)
#fermi_level_absolute
#fermi_level_relative
# RECOMMENDED fast FFTW truncation schemes
# The Coulomb Interaction is cutoff on the edges of
# the WignerSeitz Cell in the nonperiodic directions
# Periodic directions are a1,a2 for slabs and a3 for wires
#cell_box_truncation
#cell_wire_truncation
#cell_slab_truncation
# Analytic, but nonWignerSeitz Cell Truncation
#spherical_truncation
# For Spherical Truncation, radius in Bohr
#coulomb_truncation_radius 10.00
# Frequency dependence of the inverse dielectric matrix.
# Set to 0 to compute the static inverse dielectric matrix (default).
# Set to 2 to compute the full frequency dependent inverse dielectric matrix.
# Set to 3 to compute the two frequencies needed for GodbyNeeds GPP model.
#frequency_dependence 0
# Full frequency dependence method for the polarizability, if frequency_dependence==2:
# 0: RealAxis formalism, AdlerWiser formula.
# 1: RealAxis formalism, spectral method (PRB 74, 035101, (2006))
# 2: ContourDeformation formalism, AdlerWiser formula.
#frequency_dependence_method 2
# Plasma frequency (eV) needed for ContourDeformation method. The exact value
# is unimportant, especially if you have enough imaginary frequency points. We
# recommend you keep this value fixed at 2 Ry.
#plasma_freq 27.21138506d0
# For frequency_dependence 3, the value of the purely imaginary frequency, in eV:
#imaginary_frequency 27.21138506d0
# Parameters for fullfrequencydependent calculations, in eV.
#
#
# Method 0  RealAxis formalism with AdlerWiser formula.
# 
# Default values, which are good for typical semiconductors. The main
# convergence knobs are:
#  broadening: should the energy resolution due to kpoint sampling, or
# a number as small as possible if you have a molecule.
#  delta_frequency: should be converged (the smaller, the better). For
# molecules, delta_frequency should be the same as broadening.
#  low_frequency_cutoff: should be larger than the maximum transition, i.e., the
# energy difference between the highest conduction band and the lowest valence band.
#
#broadening 0.1
#delta_frequency (= broadening)
#frequency_low_cutoff 200.0
#frequency_high_cutoff (= 4*frequency_low_cutoff)
#delta_frequency_step 1.0
#
#
# Method 1  RealAxis formalism with spectral method.
# 
# Default parameters. Same remarks as for method 1 apply here.
#
#broadening 0.1
#delta_frequency (= broadening)
#frequency_low_cutoff 200.0
#frequency_high_cutoff (= 4*frequency_low_cutoff)
#delta_frequency_step 1.0
#delta_sfrequency (= delta_frequency)
#delta_sfrequency_step (= delta_frequency_step)
#sfrequency_low_cutoff (= frequency_low_cutoff)
#sfrequency_high_cutoff (= sfrequency_low_cutoff)
#
#
# Method 2  ContourDeformation formalism with AdlerWiser formula (default).
# 
# Default values, which are good for typical semiconductors. The main
# convergence knobs are:
#  broadening: should the energy resolution due to kpoint sampling, or
# a number as small as possible if you have a molecule.
#  delta_frequency: should be converged (the smaller, the better). For
# molecules, delta_frequency should be the same as broadening.
#  low_frequency_cutoff: you should increase this valid if you wish to use
# the Sigma code and look into QP states deep in occupied manifold or high in
# the unoccupied manifold.
#
#broadening 0.25
#delta_frequency (= broadening)
#frequency_low_cutoff 10.0
#number_imaginary_freqs 15
#
#
# Description of the flags
# 
#
#  Each val>cond transition is broadened by "broadening".
#  A uniform frequency grid is set up from "init_frequency" (defaults to 0),
# up to "low_frequency_cutoff", with a spacing of "delta_frequency" between
# two frequencies.
# For methods 1 and 2:
#  A nonuniform frequency grid is setup from "low_frequency_cutoff" to
# "high_frequency_cutoff", where the frequency spacing gets increased by
# "delta_frequency_step".
#
# For method 2:
#  A separate frequency grid is setup for the spectral function. The variables
# "init_sfrequency", "delta_sfrequency", "delta_sfrequency_step",
# "sfrequency_low_cutoff", and "sfrequency_high_cutoff" define this grid, in
# an analogy to the flags used to define the grid for the polarizability matrix.
#
# For method 3:
#  A frequency grid with "number_imaginary_freqs" is setup on the imag axis.
# Logging convergence of the head & tail of polarizability matrix with respect to conduction bands.
# Set to 1 for no convergence test
# Set to 0 for the 5 column format including the extrapolated values (default).
# Set to 1 for the 2 column format, real part only.
# Set to 2 for the 2 column format, real and imaginary parts.
#full_chi_conv_log 1
# qx qy qz 1/scale_factor is_q0
# scale_factor is for specifying values such as 1/3
# is_q0 = 0 for regular, nonzero qvectors (read val WFNs from WFN)
# is_q0 = 1 for a small qvector in semiconductors (read val WFNs from WFNq)
# is_q0 = 2 for a small qvector in metals (read val WFNs from WFN)
# if present the small qvector should be first in the list
# You can generate this list with kgrid.x: just set the shifts to zero and use
# same grid numbers as for WFN. Then replace the zero vector with q0.
begin qpoints
0.000000 0.000000 0.005000 1.0 1
0.000000 0.000000 0.062500 1.0 0
0.000000 0.000000 0.125000 1.0 0
0.000000 0.000000 0.187500 1.0 0
0.000000 0.000000 0.250000 1.0 0
0.000000 0.000000 0.312500 1.0 0
0.000000 0.000000 0.375000 1.0 0
0.000000 0.000000 0.437500 1.0 0
0.000000 0.000000 0.500000 1.0 0
0.000000 0.000000 0.562500 1.0 0
0.000000 0.000000 0.625000 1.0 0
0.000000 0.000000 0.687500 1.0 0
0.000000 0.000000 0.750000 1.0 0
0.000000 0.000000 0.812500 1.0 0
0.000000 0.000000 0.875000 1.0 0
0.000000 0.000000 0.937500 1.0 0
end
# Scissors operator (linear fit of the quasiparticle
# energy corrections) for the bands in WFN and WFNq.
# For valenceband energies:
# ev_cor = ev_in + evs + evdel * (ev_in  ev0)
# For conductionband energies:
# ec_cor = ec_in + ecs + ecdel * (ec_in  ec0)
# Defaults below. evs, ev0, ecs, ec0 are in eV.
# If you have eqp.dat and eqp_q.dat files
# this information is ignored in favor of the eigenvalues
# in eqp.dat and eqp_q.dat.
#evs 0.0
#ev0 0.0
#evdel 0.0
#ecs 0.0
#ec0 0.0
#ecdel 0.0
# or
#cvfit 0.0 0.0 0.0 0.0 0.0 0.0
# Set this to use eigenvalues in eqp.dat and eqp_q.dat
# If not set, these files will be ignored.
#eqp_corrections
# Write the bare Coulomb potential V(q+G) to file
#write_vcoul
# Matrix Element Communication Method (Chi Sum Comm). Default is gcomm_matrix
# which is good if nk*nc*nv > nmtx*nfreq. If nk*nc*nv < nfreq*nmtx
# (nk*nv < nfreq since nc~nmtx), use gcomm_elements. Only "gcomm_elements"
# is supported with the spectral method.
#gcomm_matrix
#gcomm_elements
# Number of pools for distribution of valence bands
# The default is chosen to minimize memory in calculation
#number_valence_pools 1
# By default, the code computes the polarizability matrix, constructs
# the dielectric matrix, inverts it and writes the result to file epsmat.
# Use keyword skip_epsilon to compute the polarizability matrix and
# write it to file chimat. Use keyword skip_chi to read the polarizability
# matrix from file chimat, construct the dielectric matrix, invert it and
# write the result to file epsmat.
#skip_epsilon
#skip_chi
# Use traditional simple binary format for epsmat/eps0mat instead of HDF5 file format.
# Relevant only if code is compiled with HDF5 support.
#dont_use_hdf5
# Verbosity level, options are:
# 1 = default
# 2 = medium  info about kpoints, symmetries, and eqp corrections.
# 3 = high  full dump of the reduced and unfolded kpoints.
# 4 = log  log of various function calls. Use to debug code.
# 5 = debug  extra debug statements. Use to debug code.
# 6 = max  only use if instructed to, severe performance downgrade.
# Note that verbosity levels are cumulative. Most users will want to stick
# with level 1 and, at most, level 3. Only use level 4+ if debugging the code.
#verbosity 1
# (Full Frequency only) Calculates several frequencies in parallel. No "new"
# processors are used here, the chi summation is simply done in another order
# way to decrease communication. This also allows the inversion of multiple
# dielectric matrices simultaneously via ScaLAPACK, circumventing ScaLAPACK's
# scaling problems. Can be very efficient for system with lots of Gvectors and when
# you have many frequencies. In general gives speedup. However, in order to
# calculate N frequencies in parallel, the memory to store pol%gme is currently
# multiplied by N as well.
#nfreq_group 1
# EXPERIMENTAL FEATURES FOR TESTING PURPOSES ONLY
# 'unfolded BZ' is from the kpoints in the WFN file
# 'full BZ' is generated from the kgrid parameters in the WFN file
# See comments in Common/checkbz.f90 for more details
# Replace unfolded BZ with full BZ
#fullbz_replace
# Write unfolded BZ and full BZ to files
#fullbz_write
# The requested number of bands cannot break degenerate subspace
# Use the following keyword to suppress this check
# Note that you must still provide one more band in
# wavefunction file in order to assess degeneracy
#degeneracy_check_override
# Instead of using the RHO FFT box to perform convolutions, we automatically
# determine (and use) the smallest box that is compatible with your epsilon
# cutoff. This also reduces the amount of memory needed for the FFTs.
# Although this optimization is safe, you can disable it by uncommenting the
# following line:
#no_min_fftgrid
# Uncomment this flag if you would like to restart your Epsilon calculation
# instead of starting it from scratch. Note that we can only reuse qpoints
# that were fully calculated. This flag is ignored unless you are running
# the code with HDF5.
#restart
# Qgrid for the epsmat file. Defaults to the WFN kgrid.
#qgrid 0 0 0
Epsilon/epsconv.inp
#
# Example epsconv.inp for epsbinasc/epsascbin.
# number of qpoints
# output filename
# number of input files (will be merged)
# input filenames
1
epsmat.out
1
epsmat.in
Epsilon/epsomega.inp
eps0mat ! epsmat file, fullfrequency or static
RHO ! RHO file for PlasmonPole
0.0 0.0 0.0 ! qvector in crystal coordinates
1 0 0 ! Gvector in crystal coordinates
1 0 0 ! G`vector in crystal coordinates
201 ! nFreq, for static epsmat. For fullfreq., overwritten from file
0.5 ! dDeltaFreq in eV, in case of static epsmat. See above.
2.0 ! dBrdning in eV, in case of static epsmat. See above.
epsPP.dat ! PlasmonPole epsilon, Re and Im parts
epsR.dat ! Retarded epsilon, Re and Im parts
epsA.dat ! Advanced epsilon, Re and Im parts
Epsilon/epsinvomega.inp
eps0mat(.h5) ! epsmat file, fullfrequency or static
RHO ! RHO file for PlasmonPole
0.0 0.0 0.0 ! qvector in crystal coordinates
1 0 0 ! Gvector in crystal coordinates
1 0 0 ! G`vector in crystal coordinates
201 ! nFreq, for static epsmat. For fullfreq., overwritten from file
0.5 ! dDeltaFreq in eV, in case of static epsmat. See Above.
2.0 ! dBrdning in eV, in case of static epsmat. See Above.
0.1 ! Exciton binding energy to use to calculate the effective eps^1 head for BSE.
epsInvPP.dat ! PlasmonPole epsilon inverse, Re and Im parts
epsInvR.dat ! Retarded epsilon inverse, Re and Im parts
epsInvA.dat ! Advanced epsilon inverse, Re and Im parts
Epsilon/epsmat_merge.inp
44 7 ! ecuts1, nqtot
0.0000 0.0000 0.5000 1.0 ! qx qy qz div
0.0000 0.2500 0.0000 1.0
0.0000 0.2500 0.5000 1.0
0.0000 0.5000 0.0000 1.0
0.0000 0.5000 0.5000 1.0
0.2500 0.5000 0.0000 1.0
0.2500 0.5000 0.5000 1.0
1 ! nfiles
epsmat ! filename(i), i=1,nfiles
Epsilon/epsmat_old2hdf5.inp
epsmat ! Name of input file
epsmat.h5 ! Name of output file
WFN ! Name of wave function file to extract header from
icutv ! truncation flag (0=no truncation, 2=spherical, 4=wire, 5=box, 6=slab)
truncval ! truncation radius, for spherical truncation. Otherwise, ignored
Epsilon/epsmat_intp.inp
WFN !wavefunction
epsmat !input epsmat file
epsmat_out !output epsmat file
7 ! number of interpolated points to produce
0.00000000 0.00000000 0.00000000 1.0
0.00000000 0.00781250 0.00000000 1.0
0.00000000 0.01562500 0.00000000 1.0
0.00000000 0.03125000 0.00000000 1.0
0.00000000 0.04687500 0.00000000 1.0
0.02343750 0.49218750 0.00000000 1.0
0.02343750 0.02343750 0.00000000 1.0
0.00000000 0.50781250 0.00000000 1.0
Sigma
Sigma/README
================================================================================
GW code: Sigma
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014)
Version 1.0 (September, 2011)
Version 0.5 J. Deslippe, D. Prendergast, L. Yang, F. Ribeiro, G. Samsonidze (2008)
Version 0.2 S. IsmailBeigi (2002)
Version 0.1 G. M. Rignanese, E. Chang, X. Blase (1998)
Version 0.0 M. Hybertsen (1985)

Description:
Sigma is the second half of the GW code. It should be run after
calculating epsmat/esp0mat in Epsilon. It gives the quasiparticle
selfenergies and dispersion relation for quasielectron and
quasihole states. The main result is written to sigma.log file.

Required Input:
sigma.inp Input parameters. See example in this directory.
epsmat(.h5) Inverse dielectric matrix (q<>0). Created using Epsilon.
The file has a ".h5" extension if the code is compiled
with HDF5 support (for specification, see epsmat.h5.spec).
eps0mat(.h5) Inverse dielectric matrix (q>0). Created using Epsilon.
The file has a ".h5" extension if the code is compiled
with HDF5 support (for specification, see epsmat.h5.spec).
WFN_inner Real or complex wavefunctions. kgrid should be the same as
the qgrid of epsmat, though there can be a shift.
This is usually the reduced points from an
unshifted grid.
RHO Charge density (only needed for generalized plasmonpole model),
from SCF calculation on which WFN_inner is based.
vxc.dat Matrix elements of the exchangecorrelation potential,
from a meanfield calculation or a previous Sigma run.
VXC file may be used instead via keyword dont_use_vxcdat.
Additional Input:
WFN_outer Outer wavefunctions between which the
selfenergy operator and exchangecorrelation
potential are sandwiched. If absent inner
wavefunctions will be used instead. Note that
VXC should be consistent with outer wfn while
RHO with inner wfn. If outer wfn is generated
using a hybrid functional, VXC contains the
local part of exchangecorrelation potential.
In this case set bare_exchange_fraction in
sigma.inp to compensate for the nonlocal part,
or use vxc.dat instead of VXC.
The file below will be read only if eqp_corrections is set in sigma.inp.
eqp.dat A list of quasiparticle energy corrections
for the bands in WFN_inner.
The corrected eigenvalues are used for
constructing the selfenergy operator.
The file below will be read only if eqp_outer_corrections is set in sigma.inp.
eqp_outer.dat A list of quasiparticle energy corrections for
the bands in WFN_outer.
The corrected eigenvalues determine
the energies at which the selfenergy
operator is calculated.
vxc2.dat Matrix elements of the exchangecorrelation potential,
from a meanfield calculation or a previous Sigma run. This
is for a oneshot hybrid functional like calculation instead
of a Sigma calculation. VXC2 file may be used instead via
keyword dont_use_vxc2dat.
Auxiliary Files:
x.dat Matrix elements of the bare exchange, generated by a previous
Sigma run to speed up subsequent calculations.
Read if use_xdat keyword is present, written otherwise.
VXC Exchangecorrelation potential (whose matrix elements
are subtracted from DFT eigenvalues). Produced by
meanfield code, from SCF calculation on which WFN_inner is based
(or WFN_outer, if present).
Used only if vxc.dat is not present or keyword dont_use_vxcdat is set.
VXC2 Exchangecorrelation potential (whose matrix elements
are added to the hybrid functional like exchange). Produced by
meanfield code, from SCF calculation on which WFN_inner is based
(or WFN_outer, if present).
Used only if vxc2.dat is not present or keyword dont_use_vxc2dat is set.

sigma.inp Please see example sigma.inp in this directory for complete options.

Output Files:
The file below will be read only if eqp_corrections is set in sigma.inp.
eqp0.dat Contains the onshell QP energies = Emf  Vxc  Sig(Eo)
This is *not* the recommended quantity to use for QP properties.
This is Eqn. 36 from Hybertsen & Louie PRB 34 5390.
eqp1.dat Contains the offshell solution to the linearized Dyson`s equation
= Eqp0 + (dSig/dE) / (1  dSig/dE) * (Eqp0  Eo),
or a full linear interpolation if more freq. points where computed.
This is the recommended quantity to use for QP properties.
If you only evaluate Sigma at two frequencies, this is
Eqn. 37 from Hybertsen & Louie PRB 34 5390. If there are
more frequencies available, the code uses them to find
a better linearized solution. For GW calculations without
plasmonpole models, the code reports how many solutions
to Dyson`s equation were found.
sigma_hp.log The log file containing quasiparticle energy values
for desired states. For a fullfrequency calculation
only the value for Sigma calculated at energy closest
to the outer wavefunction eigenvalue is shown. The file
spectrum.dat contains the full Sigma(E) spectra.
ch_converge.dat Convergence of Coulombhole term with respect to empty orbitals.
spectrum.dat The real and imaginary parts of Sigma as a function of
energy. The energy grid is specified in sigma.inp. This
file is only output in fullfrequency calculations.
Some general notes:
* IM(SIGMA) was calculated using the same broadening
as RE(SIGMA), while IM(SIGMA2) was calculated without
any broadening.
There are 2 Sigmas currently reported if doing a real axis
calculation (frequency_dependence_method 0)
1. Sigma = SX + CH with no Broadening. We do the energy
denominator integral analytically.
2. Sigma = X + COR with no broadening. We do the energy
denominator integral analytically.
For contour deformation (frequency_dependence_method 2)
only X + COR is given. But spectrum.dat contains columns
for contributions from the poles and imaginary axis
integral
* Ew is not the same as the frequencies specified in
sigma.inp. Ew is the absolute offshell quasiparticle
energy, and it is *not* measured wrt the Fermi energy.
The (physically meaningful) onshell quasiparticle
energy EQP is the solution of Dyson's equation:
Ew = E_LDA  Vxc + Re[Sig(Ew)].

A Note About the Wings of Epsilon (for Semiconductors Only):
The wings of Chi have terms of the following form:
< vk  e^(i(G+q)r)  ck+q >< ck+q  e^(iqr)  vk > (1)
The matrix element on the right is < u_ck+q  u_vk > where u is the periodic part
of the Bloch function. From k.p perturbation theory, this matrix element is proportional
to q. The matrix element on the left with a nonzero G is typically roughly
a constant as a function of q for small q (q being a small addition to G).
Thus for a general Gvector, Chi_wing(G,q) \propto q. This directly leads
to the wings of the screened untruncated Coulomb interaction being proportional
to 1/q. Note that this function changes sign as q > q. Thus, when treating the
q=0 point, we set the value of the wing to zero (the average of its value in the
miniBrillouin zone (mBZ).
For Gvectors on highsymmetry lines, some of the matrix elements on the left of (1)
will be zero for q=0, and therefore proportional to q. For such cases,
Chi_wing(G,q) \propto q^2, and the wings of the screened Coulomb interaction
are constant as a function of q. However, setting the q>0 wings to zero still
gives us, at worst, linear convergence to the final answer with increased kpoint
sampling, because the q>0 point represents an increasingly smaller area in the
BZ. Thus, we still zero out the q>0 wings, as discussed in
A Baldereschi and E Tosatti, Phys. Rev. B 17, 4710 (1978).
In the future, it may be worthwhile to have the user calculate chi / epsilon at
two qpoints (a third qpoint at q=0 is known) in order to compute the linear
and quadratic coefficients of each chi_wing(G,q) so that all the correct analytic
limits can be taken. This requires a lot of messy code and more work for the user
for only marginally better convergence with respect to kpoints (the wings tend
to make a small difference, and this procedure would matter for only a small set
of the Gvectors).
It is important, as always, for the user to converge their calculation
with respect to both the coarse kpoint grid used in sigma and kernel as well
as with the fine kpoint grid in absorption.

Tricks and hints:
1. Comments on convergence of scc (screened_coulomb_cutoff)
and bcc (bare_coulomb_cutoff)
The parameters scc and bcc are independent to each other and are
related to the number or terms taken into account in the
planewave expansion of Sigma_(SEX + COH) and V_xc, respectively.
Since Sigma_(SEX) is related to epsilon, scc should be equal to
or less than the epsilon_cutoff used to compute chi0 and epsilon.
If you are to use large values for scc and bcc (say 50 Ry or so),
a good suggestion is run the sigma code twice: first with large
bcc and small scc, just to get x.dat and vxc.dat; then with large
scc and any bcc, using the x.dat and vxc.dat files generated before.
The second run is usually much faster than a run with large scc
and bcc.
2. Metals
If you are doing a metal you should report the shift used in sigma.inp
whether using truncation or not.
3. Offdiagonal
If you are doing offdiagonal matrix elements of Sigma use the utility
offdiag in this directory to diagonalize the Sigma matrix.
4. Linear extrapolation for eqp1
If eqp0  ecor > finite_difference_spacing linear extrapolation
for eqp1 may be inaccurate. You should test the validity of eqp1
by rerunning calculation with selfenergy evaluated at the eqp0
energies. For that, use the eqp_outer.dat file created
with eqp.py script and point WFN_outer to WFN_inner (if you were
not already using WFN_outer), i.e. run
i) ln s WFN_inner WFN_outer
ii) ln s eqp1.dat eqp_outer.dat
and add eqp_outer_corrections to sigma.inp. If you get the same eqp1,
the linearization looks good; if not, you can continue iterations like
this to converge to the solution of the Dyson equation.
5. Utilities:

Sigma: QP shift and scissors paramters

TOOL: qp_shifts.py
USAGE: qp_shift.py sigma_hp.log
From sigma_hp.log file, writes two files which contain the LDA
energies and the GW shifts (E_qpE_lda) for that energy. Useful
for determining scissor shifting parameters ecdel, evdel, etc.

Sigma: QP extraction

TOOL: eqp.py
USAGE: eqp.py sigma_hp.log
Extracts QP energies from sigma_hp.log file and generates file
eqp.dat. This utility is currently deprecated, since the code
writes the eqp0.dat and eqp1.dat automatically.
Sigma/sigma.inp
# sigma.inp
# Energy cutoff for the screened Coulomb interaction, in Ry. The screened
# Coulomb interaction W_{GG`}(q) will contain all Gvectors with kinetic energy
# q+G^2 up to this cutoff. Default is the epsilon_cutoff used in the
# epsilon code. This value cannot be larger than the epsilon_cutoff or
# the bare_coulomb_cutoff.
#screened_coulomb_cutoff 35.0
# Energy cutoff for the bare Coulomb interaction, in Ry. The bare Coulomb
# interaction v(G+q) will contain all Gvectors with kinetic energy
# q+G^2 up to this cutoff. Default is the WFN cutoff.
#bare_coulomb_cutoff 60.0
# Total number of bands (valence+conduction) to sum over. Defaults to the
# number of bands in the WFN file minus 1.
#number_bands 1000
# Specify the Fermi level (in eV), if you want implicit doping
# Note that value refers to energies AFTER scissor shift or eqp corrections.
#fermi_level 0.0
# The Fermi level is treated as an absolute value
# or relative to that found from the mean field (default)
#fermi_level_absolute
#fermi_level_relative
# The matrix elements of Sigma
# < psi_n  Sigma(E_l)  psi_m >
# E_l is irrelevant for frequency_dependence 1 and 0
# The diagonal matrix elements of Sigma (n .eq. m .eq. l)
#number_diag ndiag
#begin diag
# n_1
# n_2
# ...
# n_ndiag
#end
# or specify a range of bands spanned by n
band_index_min 15
band_index_max 22
# The offdiagonal matrix elements of Sigma
# band_index_min .le. n .le. band_index_max
# band_index_min .le. m .le. band_index_max
# band_index_min .le. l .le. band_index_max
# NOTE: the offdiag utility only works with the sigma_matrix syntax, below.
#number_offdiag noffdiag
#begin offdiag
# n_1 m_1 l_1
# n_2 m_2 l_2
# ...
# n_noffdiag m_noffdiag l_noffdiag
#end
# or select a specific value of l and let n and m vary
# in the range from band_index_min to band_index_max
# Set l to 0 to skip the offdiagonal calculation (default)
# If l = 1 then l_i is set to n_i (i = 1 ... noffdiag)
# i.e. each row is computed at different eigenvalue
# If l = 2 then l_i is set to m_i (i = 1 ... noffdiag)
# i.e. each column is computed at different eigenvalue
# For l > 0, all elements are computed at eigenvalue of band l.
# Set t to 0 for the full matrix (default)
# or to 1/+1 for the lower/upper triangle
#sigma_matrix l t
# The range of spin indices for which Sigma is calculated
# The default is the first spin component
#spin_index_min 1
#spin_index_max 1
# What screening is present? (default = semiconductor)
# Does not apply for frequency_dependence = 1.
screening_semiconductor
#screening_metal
#screening_graphene
# this is for a system with linear DOS at the Fermi level
# RECOMMENDED fast FFTW truncation schemes
# The Coulomb Interaction is cutoff on the edges of
# the WignerSeitz Cell in the nonperiodic directions
# Periodic directions are a1,a2 for slabs and a3 for wires
#cell_box_truncation
#cell_wire_truncation
#cell_slab_truncation
# Analytic, but non WignerSeitz cell, truncation
#spherical_truncation
# For Spherical Truncation, radius in Bohr
#coulomb_truncation_radius 10.00
# Cutoff energy for averaging the Coulomb Interaction
# in the miniBrillouin Zones around the Gammapoint
# without Truncation or for Cell Wire or Cell Slab Truncation.
# The value is in Rydbergs, the default is 10^{12}
#cell_average_cutoff 1.0d12
# Frequency dependence of the inverse dielectric matrix.
# Set to 1 for the HartreeFock and hybrid functional approximation.
# Set to 0 for the static COHSEX approximation.
# Set to 1 for the Generalized Plasmon Pole model (default).
# Set to 2 for the full frequency dependence.
# Set to 3 for the Generalized Plasmon Pole model in the GodbyNeeds flavor.
# Note: does not work with parallelization in frequencies.
#frequency_dependence 1
# Full frequency dependence method for the polarizability.
# set to 0 for the RealAxis Integration method.
# set to 2 for the ContourDeformation method (default).
#frequency_dependence_method 2
# For Full Frequency Calculations.
# In order to evaluate Sigma(omega), you have to specify a frequency grid
# for the frequencies omega. You can control where to center/shift the grid
# with the flag frequency_grid_center. The options are:
# 0 if you don`t want any shift, i.e., omega is an absolute energy
# 1 if you want to shift the first frequency omega=0 to the Fermi energy
# (this was the default behavior in BerkeleyGW1.0)
# 2 if you want to center the frequency grid around each meanfield QP energy (default)
#frequency_grid_shift 2
# If frequency_grid_style is 2 (default), specify the frequency spacing
# delta_frequency_eval in eV (default=0.2) and the width, in eV, of the
# evaluation grid centered around each QP state (default=2.0). The code
# will generate a frequency grid from max_frequency_eval to max_frequency_eval.
# You should increase the value of max_frequency_eval if you expect the
# absolute QP corrections to be large, such as when dealing with molecules
# (the code will warn if the frequency grid is too small).
#delta_frequency_eval 0.2
#max_frequency_eval 2.0
# Note for advanced users: the grids are actually centered around the outer
# meanfield energies, so you can use WFN_outer and eqp_outer_corrections or
# outer scissors parameters to finetune the frequency grid.
# If frequency_grid_style is 0 or 1, specify the initial frequency
# init_frequency_eval (before the shift), in eV, the frequency spacing
# delta_frequency_eval, in eV, and the number of frequency points
# number_frequency_eval.
#init_frequency_eval 0.0
#delta_frequency_eval 0.1
#number_frequency_eval 501
# For Contour Deformation calculations, specify the integration method on the
# imaginary axis (default is 0). Options are:
# 0 for piecewise constant matrix elements centered at each imaginary frequency
# 2 for piecewise quadratic matrix elements over each integration segment
# 3 for cubic Hermite matrix elements over each integration segment
#cd_integration_method 0
# For Generalized Plasmon Pole.
# The matrix element of the selfenergy operator is
# expanded to first order in the energy around Ecor.
# Finite difference form for numerical derivative of Sigma.
# grid = 3 : Calculate Sigma(w) on a grid, using the same frequency grid as
# in fullfrequency calculations.
# none = 2 : dSigma/dE = 0 [skip the expansion].
# backward = 1 : dSigma/dE = (Sigma(Ecor)  Sigma(EcordE)) / dE
# central = 0 : dSigma/dE = (Sigma(Ecor+dE)  Sigma(EcordE)) / (2*dE)
# forward = 1 : dSigma/dE = (Sigma(Ecor+dE)  Sigma(Ecor)) / dE
# default = 2 : forward for diagonal and none for offdiagonal
#finite_difference_form 2
# dE is finite difference spacing given in eV
# The default value is 1.0
#finite_difference_spacing 1.0
# For HartreeFock/hybrid functional, no epsmat/eps0mat files are needed.
# Instead provide a list of qpoints and the grid size.
# The list of qpoints should not be reduced with time
# reversal symmetry  because BerkeleyGW never uses time
# reversal symmetry to unfold the q/kpoints. Instead,
# inversion symmetry does the job in the real version of
# the code.
# qx qy qz 1/scale_factor is_q0
# scale_factor is for specifying values such as 1/3
# You can generate this list with kgrid.x: just set the shifts to zero and use
# same grid numbers as for WFN_inner.
#begin qpoints
# 0.0 0.0 0.0 1.0 1
#end
#qgrid 1 1 1
# What should we do when we perform a HLGPP or a GNGPP calculations and we
# find an invalid mode frequency with \omega_{G,G`}^2 < 0? Options are:
# 1: Default, same as 3.
# 0: Skip invalid mode and ignore its contribution to the self energy.
# 1: "Find" a purely complex mode frequency. This was the default behavior in BGW1.x.
# 2: Set the mode frequency to a fixed value of 2 Ry.
# 3: Treat that mode within the static COHSEX approximation (move frequency to infinity).
#invalid_gpp_mode 1
# Add remainder from tail of epsilon for full frequency.
#use_epsilon_remainder
# Logging the convergence of the CH term with respect to the number of bands
# in the output file "ch_converge.dat". Options are:
# 0 to log only the real part of VBM, CBM and the gap (default).
# 1 to log the real and imag. parts of all bands for which we compute Sigma.
#full_ch_conv_log 0
# Use precalculated matrix elements of bare exchange from x.dat.
# The default is not to use them.
#use_xdat
# Do not use precalculated matrix elements of exchangecorrelation from vxc.dat.
# The default is to use them.
#dont_use_vxcdat
# Read from traditional simple binary format for epsmat/eps0mat instead of HDF5 file format.
# Relevant only if code is compiled with HDF5 support.
#dont_use_hdf5
# Fraction of bare exchange.
# Set to 1.0 if you use the exchangecorrelation matrix elements
# read from file vxc.dat. Set to 1.0 for local density functional,
# 0.0 for HF, 0.75 for PBE0, 0.80 for B3LYP if you use the local
# part of the exchangecorrelation potential read from file VXC.
# For functionals such as HSE whose nonlocal part is not some
# fraction of bare exchange, use vxc.dat and not this option.
# This is set to 1.0 by default.
#bare_exchange_fraction 1.0
# Broadening for the energy denominator in CH and SX within GPP.
# If it is less than this value, the sum is better conditioned than
# either CH or SX directly, and will be assigned to SX while CH = 0.
# This is given in eV, the default value is 0.5
#gpp_broadening 0.5
# Cutoff for the poles in SX within GPP.
# Divergent contributions that are supposed to sum to zero are removed.
# This is dimensionless, the default value is 4.0
#gpp_sexcutoff 4.0
# kx ky kz 1/scale_factor
# scale_factor is for specifying values such as 1/3
begin kpoints
0.0000 0.0000 0.0000 1.0
end
# Scissors operator (linear fit of the quasiparticle
# energy corrections) for the bands in WFN_inner
# For valenceband energies:
# ev_cor = ev_in + evs + evdel * (ev_in  ev0)
# For conductionband energies:
# ec_cor = ec_in + ecs + ecdel * (ec_in  ec0)
# Defaults below. evs, ev0, ecs, ec0 are in eV
#evs 0.0
#ev0 0.0
#evdel 0.0
#ecs 0.0
#ec0 0.0
#ecdel 0.0
# or
#cvfit 0.0 0.0 0.0 0.0 0.0 0.0
# Scissors operator (linear fit of the quasiparticle
# energy corrections) for the bands in WFN_outer.
# Has no effect if WFN_outer is not supplied.
# For valenceband energies:
# ev_cor = ev_in + evs_outer + evdel_outer * (ev_in  ev0_outer)
# For conductionband energies:
# ec_cor = ec_in + ecs_outer + ecdel_outer * (ec_in  ec0_outer)
# Defaults below. evs_outer, ev0_outer, ecs_outer, ec0_outer are in eV
#evs_outer 0.0
#ev0_outer 0.0
#evdel_outer 0.0
#ecs_outer 0.0
#ec0_outer 0.0
#ecdel_outer 0.0
# or
#cvfit_outer 0.0 0.0 0.0 0.0 0.0 0.0
# Set this to use eigenvalues in eqp.dat
# If not set, this file will be ignored.
#eqp_corrections
# Set this to use eigenvalues in eqp_outer.dat
# If not set, this file will be ignored.
# Has no effect if WFN_outer is not supplied.
#eqp_outer_corrections
# The average potential on the faces of the unit cell
# in the nonperiodic directions for the bands in WFN_inner
# This is used to correct for the vacuum level
# The default is zero, avgpot is in eV
#avgpot 0.0
# The average potential on the faces of the unit cell
# in the nonperiodic directions for the bands in WFN_outer
# This is used to correct for the vacuum level.
# Has no effect if WFN_outer is not supplied.
# The default is zero, avgpot_outer is in eV
#avgpot_outer 0.0
# Write the bare Coulomb potential V(q+G) to file
#write_vcoul
# Number of pools for parallel sigma calculations
# The default is chosen to minimize memory in calculation
#number_sigma_pools 1
# Threshold for considering bands degenerate, for purpose of
# making sure all of degenerate subspaces are included,
# for bandaveraging, and for setting offdiagonals to zero by symmetry. (Ry)
tol_degeneracy 1e6
# Verbosity level, options are:
# 1 = default
# 2 = medium  info about kpoints, symmetries, and eqp corrections.
# 3 = high  full dump of the reduced and unfolded kpoints.
# 4 = log  log of various function calls. Use to debug code.
# 5 = debug  extra debug statements. Use to debug code.
# 6 = max  only use if instructed to, severe performance downgrade.
# Note that verbosity levels are cumulative. Most users will want to stick
# with level 1 and, at most, level 3. Only use level 4+ if debugging the code.
#verbosity 1
# EXPERIMENTAL FEATURES FOR TESTING PURPOSES ONLY
# 'unfolded BZ' is from the kpoints in the WFN_inner file
# 'full BZ' is generated from the kgrid parameters in the WFN_inner file
# See comments in Common/checkbz.f90 for more details
# Replace unfolded BZ with full BZ
#fullbz_replace
# Write unfolded BZ and full BZ to files
#fullbz_write
# The requested number of bands cannot break degenerate subspace
# Use the following keyword to suppress this check
# Note that you must still provide one more band in
# wavefunction file in order to assess degeneracy
#degeneracy_check_override
# The sum over qpoints runs over the full Brillouin zone.
# For diagonal matrix elements between nondegenerate bands
# and for spherically symmetric Coulomb potential (no truncation
# or spherical truncation), the sum over qpoints runs over
# the irreducible wedge folded with the symmetries of
# a subgroup of the kpoint. The latter is the default.
# In both cases, WFN_inner should have the reduced kpoints
# from an unshifted grid, i.e. same as qpoints in Epsilon.
# With no_symmetries_q_grid, any calculation can be done;
# use_symmetries_q_grid is faster but only diagonal matrix elements
# of nondegenerate or bandaveraged states can be done.
#no_symmetries_q_grid
#use_symmetries_q_grid
# Offdiagonal elements are zero if the two states belong to
# different irreducible representations. As a simple proxy,
# we use the size of the degenerate subspaces of the two states:
# if the sizes are different, the irreps are different, and the
# matrix element is set to zero without calculation.
# Turn off this behavior for testing by setting flag below.
# Using WFN_outer effectively sets no_symmetries_offdiagonals.
#no_symmetries_offdiagonals
# Rotation of the kpoints may bring Gvectors outside of the sphere.
# Use the following keywords to specify whether to die if some of
# the Gvectors fall outside of the sphere. The default is to die.
# Set to die in case screened_coulomb_cutoff = epsilon_cutoff.
# Set to ignore in case screened_coulomb_cutoff < epsilon_cutoff.
#die_outside_sphere
#ignore_outside_sphere
# Dealing with the convergence of the CH term.
# Set to 0 to compute a partial sum over empty bands.
# Set to 1 to compute the exact static CH.
# In case of exact_static_ch = 1 and frequency_dependence = 1 (GPP) or 2 (FF),
# the partial sum over empty bands is corrected with the static remainder
# which is equal to 1/2 * (exact static CH  partial sum static CH),
# additional columns in sigma_hp.log labeled ch', sig', eqp0', eqp1'
# are computed with the partial sum without the static remainder,
# and ch_converge.dat contains the static limit of the partial sum.
# In case of exact_static_ch = 0 and frequency_dependence = 0 (COHSEX),
# columns ch, sig, eqp0, eqp1 contain the exact static CH,
# columns ch', sig', eqp0', eqp1' contain the partial sum static CH,
# and ch_converge.dat contains the static limit of the partial sum.
# For exact_static_ch = 1 and frequency_dependence = 0 (COHSEX),
# columns ch', sig', eqp0', eqp1' are not printed and
# file ch_converge.dat is not written.
# Default is 0 for frequency_dependence = 1 and 2;
# 1 for frequency_dependence = 0;
# has no effect for frequency_dependence = 1.
#
# It is important to note that the exact static CH answer
# depends not only on the screened Coulomb cutoff but also on the bare Coulomb cutoff
# because GG' for G's within the screened Coulomb cutoff can be outside the screened
# Coulomb cutoff sphere. And, therefore, the bare Coulomb cutoff sphere is used.
#exact_static_ch 0
# Do not average W over the minibz, and do not replace the head of eps0mat with
# averaged value.
#skip_averagew
Sigma/sig2wan.inp
sigma_hp.log ! Sigma output file to read kpoints, eigenvalues and symmetries from
1 ! spin component to read from sigma_hp.log file
1 ! set to 0 or 1 to read eqp0 or eqp1 from sigma_hp.log file
prefix.nnkp ! Wannier90 input file to read kpoints from
prefix.eig ! file where the output of sig2wan is written
nbands ! number of bands to write out
BSE
BSE/README_kernel
================================================================================
BSE code: Kernel
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014)
Version 1.0 (July, 2011) J. Deslippe, M. Jain, D. A. Strubbe, G. Samsonidze.
Version 0.5 J. Deslippe, D. Prendergast, L. Yang, F. Ribeiro, G. Samsonidze (2008)
Version 0.2 C. Spataru, S. IsmailBeigi (2004)
Version 0.1 M. L. Tiago, E. Chang, G. M. Rignanese (1999)

Description:
This code constructs the direct and exchange Kernel matrix on
the coarse grid. This is done essentially by computing Eqs 34,
35 and 4246 of Rohlfing and Louie.

Required Input:
kernel.inp Input parameters. See detailed explanation below.
WFN_co Wavefunctions on coarse grid. Recommended: use an unshifted
grid of the same size as the qgrid in epsmat.
Shift will increase number of qvectors needed in epsmat.
epsmat(.h5) Inverse dielectric matrix (q<>0). Created using Epsilon.
Must contain the all q=kk' generated from WFN_co, including
with symmetry if use_symmetries_coarse_grid is set.
The file has a ".h5" extension if the code was built
with HDF5 support.
eps0mat(.h5) Inverse dielectric matrix (q>0). Created using Epsilon.
The file has a ".h5" extension if the code was built
with HDF5 support.
Note Kernel does not require quasiparticle eigenvalues. It
may be run in parallel with Sigma.

kernel.inp Please see example kernel.inp in this directory
for more complete options.

Output Files:
bsedmat Direct kernel matrix elements on unshifted coarse grid.
bsexmat Exchange kernel matrix elements on unshifted coarse grid.
or
bsemat.h5 Includes data from both of above if compiled with HDF5 support.
For specification, see bsemat.h5.spec. Note that
you can use the bsemat_hdf5_upgrade.py utility to
upgrade a bsemat.h5 file generated with BerkeleyGW
prior to 1.2 (r6974).

A Note About the Wings of Epsilon (for Semiconductors Only):
The wings of Chi have terms of the following form:
< vk  e^(i(G+q)r)  ck+q >< ck+q  e^(iqr)  vk > (1)
The matrix element on the right is < u_ck+q  u_vk > where u is the periodic part
of the Bloch function. From k.p perturbation theory, this matrix element is proportional
to q. The matrix element on the left with a nonzero G is typically roughly
a constant as a function of q for small q (q being a small addition to G).
Thus for a general Gvector, Chi_wing(G,q) \propto q. This directly leads
to the wings of the screened untruncated Coulomb interaction being proportional
to 1/q. Note that this function changes sign as q > q. Thus, when treating the
q=0 point, we set the value of the wing to zero (the average of its value in the
miniBrillouin zone (mBZ).
For Gvectors on highsymmetry lines, some of the matrix elements on the left of (1)
will be zero for q=0, and therefore proportional to q. For such cases,
Chi_wing(G,q) \propto q^2, and the wings of the screened Coulomb interaction
are constant as a function of q. However, setting the q>0 wings to zero still
gives us, at worst, linear convergence to the final answer with increased kpoint
sampling, because the q>0 point represents an increasingly smaller area in the
BZ. Thus, we still zero out the q>0 wings, as discussed in
A Baldereschi and E Tosatti, Phys. Rev. B 17, 4710 (1978).
In the future, it may be worthwhile to have the user calculate chi / epsilon at
two qpoints (a third qpoint at q=0 is known) in order to compute the linear
and quadratic coefficients of each chi_wing(G,q) so that all the correct analytic
limits can be taken. This requires a lot of messy code and more work for the user
for only marginally better convergence with respect to kpoints (the wings tend
to make a small difference, and this procedure would matter for only a small set
of the Gvectors).
It is important, as always, for the user to converge their calculation
with respect to both the coarse kpoint grid used in sigma and kernel as well
as with the fine kpoint grid in absorption.
JRD+MJ

Tricks and hints:
1. To optimize distribution of work among PEs, do the following:
Choose processors to divide:
nk^2 (if npes < nk^2)
nk^2*nc^2 (if npes < nk^2*nc^2)
nk^2*nc^2*nv^2 (if npes < nk^2*nc^2*nv^2)
2. All input and output files (except kernel.inp) are in
binary format.
3. The interaction matrices are calculated in full! i.e. not only the
upper triangle. In diag, currently only the upper triangle is used.
4. The Brillouin zone is built using a WignerSeitz construction,
this way the head matrix elements are easily calculated.
5. The dielectric matrix is by default stored in memory.
6. The parameters scc (screened_coulomb_cutoff) and bcc
(bare_coulomb_cutoff) are the same as scc and bcc used in Simga.
The parameters scc and bcc should be equal to or less than the
epsilon_cutoff and wavefunction_cutoff used in Epsilon and DFT,
respectively. See Sigma/README for more details.
Converters from old versions of file formats to current version are available in version 2.4.
BSE/kernel.inp
# kernel.inp
# specify the number of valence bands (counting down from HOMO)
number_val_bands 3
# conduction bands (counting up from LUMO)
number_cond_bands 3
# Energy cutoff for the screened Coulomb interaction, in Ry. The screened
# Coulomb interaction W_{GG`}(q) will contain all Gvectors with kinetic energy
# q+G^2 up to this cutoff. Default is the epsilon_cutoff used in the
# epsilon code. This value cannot be larger then the epsilon_cutoff or
# the bare_coulomb_cutoff.
#screened_coulomb_cutoff 35.0
# Energy cutoff for the bare Coulomb interaction, in Ry. The bare Coulomb
# interaction v(G+q) will contain all Gvectors with kinetic energy
# q+G^2 up to this cutoff. Default is the WFN cutoff.
#bare_coulomb_cutoff 60.0
# Specify the Fermi level (in eV), if you want implicit doping
# Note that value refers to energies AFTER scissor shift or eqp corrections.
#fermi_level 0.0
# The Fermi level is treated as an absolute value
# or relative to that found from the mean field (default)
#fermi_level_absolute
#fermi_level_relative
# What screening is present? (default = semiconductor)
screening_semiconductor
#screening_metal
#screening_graphene
# this is for a system with linear DOS at the Fermi level
# RECOMMENDED fast FFTW truncation schemes
# The Coulomb Interaction is cutoff on the edges of
# the WignerSeitz Cell in the nonperiodic directions
# Periodic directions are a1,a2 for slabs and a3 for wires
#cell_box_truncation
#cell_wire_truncation
#cell_slab_truncation
# Analytic, but non WignerSeitz cell, truncation
#spherical_truncation
# For Spherical Truncation, radius in Bohr
#coulomb_truncation_radius 10.00
# Flag for coarse grid of kpoints in the BZ.
# Should we unfold using symmetries? either
# no_symmetries_coarse_grid (default) or
# use_symmetries_coarse_grid. If you calculated
# epsmat on a reduced q grid, you should use
# symmetries here!
#use_symmetries_coarse_grid
no_symmetries_coarse_grid
# Write the bare Coulomb potential V(q+G) to file
#write_vcoul
# Low communication
# The default behavior of the code is to distribute the dielectric matrix
# among the processors. While this minimizes memory usage, it also
# increases the communication. By using the low_comm flag, each processor
# will store the whole dielectric matrix. It is advisable to use this flag
# whenever each PE has enough memory to hold the whole epsmat file.
#low_comm
# Low Memory option
# Calculate matrix elements separately for each k,c,v,k',c',v' pair
#low_memory
# Use traditional simple binary format for bsedmat/bsexmat instead of HDF5 file format.
# Relevant only if code is compiled with HDF5 support.
#dont_use_hdf5
# Verbosity level, options are:
# 1 = default
# 2 = medium  info about kpoints, symmetries, and eqp corrections.
# 3 = high  full dump of the reduced and unfolded kpoints.
# 4 = log  log of various function calls. Use to debug code.
# 5 = debug  extra debug statements. Use to debug code.
# 6 = max  only use if instructed to, severe performance downgrade.
# Note that verbosity levels are cumulative. Most users will want to stick
# with level 1 and, at most, level 3. Only use level 4+ if debugging the code.
#verbosity 1
# EXPERIMENTAL FEATURES FOR TESTING PURPOSES ONLY
# High Memory option
# Save all wavefunction FFTs when calculating the BSE kernel.
# Overwrites the low_memory option.
#high_memory
# 'unfolded BZ' is from the kpoints in the WFN file
# 'full BZ' is generated from the kgrid parameters in the WFN file
# See comments in Common/checkbz.f90 for more details
# Replace unfolded BZ with full BZ
#fullbz_replace
# Write unfolded BZ and full BZ to files
#fullbz_write
# Rotation of the kpoints may bring Gvectors outside of the sphere.
# Use the following keywords to specify whether to die if some of
# the Gvectors fall outside of the sphere. The default is to ignore.
# Set to die in case screened_coulomb_cutoff = bare_coulomb_cutoff.
# Set to ignore in case screened_coulomb_cutoff < bare_coulomb_cutoff.
#die_outside_sphere
#ignore_outside_sphere
# Flag to read kpoints from the 'kpoints' file.
# The default is to read kpoints from the wfn file.
#read_kpoints
# Calculates the kernel for all n1n2,n1'n2' transitions, and not only the pairs
# vc,v'c' given by the TammDancoff approximation (TDA). Notes:
# 1) With this option BerkeleyGW will calculate transitions beyond those needed
# by the full BSE (such as vv,cc).
# 2) In principle, one should *always* calculate these extra kernel blocks,
# because the interpolation routine in absorption mixes wavefunction
# characters, in particular for metallic and narrowgap semiconductors.
# 3) You do not need these extended kernel blocks *only* if you are *not
# interpolating* later on. However, the only way to do nonTDA absorption
# calculation right now is with the extended kernel flag (i.e., there's no
# intermediate option to include only vc,c'v' transitions but not vv,cc).
# 4) Only parallelization over k^2 is supported.
#extended_kernel
BSE/README_absorption
================================================================================
BSE code: Absorption
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014)
Version 1.0 (July, 2011) J. Deslippe, M. Jain, D. A. Strubbe, G. Samsonidze
Version 0.5 J. Deslippe, D. Prendergast, L. Yang, F. Ribeiro, G. Samsonidze (2008)
Version 0.2 M. L. Tiago, C. Spataru, S. IsmailBeigi (2004)

Description:
This code is the second half of the BSE code. It interpolates the coarse grid
electronhole kernel onto the fine grid and then diagonalized the BSE equation.
The output is the electronhole eigenfunctions and eigenvalues as well the
absorption spectra.

Required Input:
absorption.inp Input parameters. See detailed explanation below.
WFN_fi Wavefunctions in unshifted fine grid
(conduction and valence for momentum operator,
conduction for velocity operator).
WFNq_fi Wavefunctions in shifted fine grid
(not needed for momentum operator,
valence for velocity operator).
WFN_co Wavefunctions on unshifted coarse grid.
Must be the same as used for Kernel.
eps0mat(.h5) Must be same as used in Kernel.
The file has a ".h5" extension if the code is compiled
with HDF5 support (for specification, see epsmat.h5.spec).
epsmat(.h5) Must be same as used in Kernel.
The file has a ".h5" extension if the code is compiled
with HDF5 support (for specification, see epsmat.h5.spec).
bsedmat BSE matrix elements in coarse grid, direct part. This
should be generated with Kernel code using same WFN_co.
bsexmat BSE exchange matrix elements. This should be generated
with Kernel code using same WFN_co.
or
bsemat.h5 Includes data from both of above if compiled with HDF5 support.
For specification, see bsemat.h5.spec. Note that
you can use the bsemat_hdf5_upgrade.py utility to
upgrade a bsemat.h5 file generated with BerkeleyGW
prior to 1.2 (r6974).
Additional Input:
eqp.dat A list of quasiparticle energy corrections for the bands in WFN_fi.
Used if eqp_corrections is set in absorption.inp.
eqp_q.dat A list of quasiparticle energy corrections for the bands in WFNq_fi.
Used if eqp_corrections is set in absorption.inp.
eqp_co.dat A list of quasiparticle energy corrections for the bands in WFN_co.
Used if eqp_co_corrections is set in absorption.inp.
kpoints A list of kpoints in unshifted fine grid. EXPERIMENTAL.
If absent kpoints from WFN_fi file are used.
kpoints_co A list of kpoints in unshifted coarse grid. EXPERIMENTAL.
If absent kpoints from WFN_co file are used.
Auxiliary Files: (output files from previous runs  used as input to speed up calculation)
dtmat Transformation matrices, dcc/dvv use for interpolation
between coarse and fine grid. This file must be consistent
with your bsedmat and bsexmat files and corresponding
coarse and fine wavefunctions.
NOTE: the file format for dtmat was changed in BerkeleyGW 1.1.0 (r5961)
vmtxel Optical matrix elements (velocity or momentum) between
single particle states.
epsdiag.dat Diagonal elements of dielectric matrix on the qgrid.
Must be consistent with epsmat and eps0mat.
eigenvalues.dat Contains electronhole eigenvalues and transition matrix elements.

absorption.inp Please see example absorption.inp in this directory
for more complete options.

Output Files:
eigenvalues.dat Has eigenvalues/transition matrix elements of eh states,
eigenvalues in eV, mtxels in atomic units.
eigenvalues_noeh.dat As above but for noninteracting eh transitions, showing
kpoint, bands, and spin for each.
eigenvectors Has the excitonic wavefunctions in Bloch space: A_svck
absorption_eh.dat Dielectric function and density of excitonic states.
Four Columns
energy (in eV)  epsilon_2  epsilon_1  DOS
DOS is normalized (\int (DOS) d(omega) = 1)
absorption_noeh.dat Noninteracting (RPA) dielectric function and joint density
of states. Four columns:
energy (in eV)  epsilon_2  epsilon_1  JDOS
JDOS is normalized (\int (JDOS) d(omega) = 1)
dvmat_norm.dat The norms of the dvv overlap matrices between the valence
band k on the fine grid and the closest kpoint on the coarse grid
dcmat_norm.dat The norms of the dcc overlap matrices between the conduction
band k on the fine grid and the closest kpoint on the coarse grid
eqp.dat, eqp_q.dat Quasiparticle corrections for WFN_fi and WFNq_fi interpolated from
the coarse grid if eqp_co_corrections is used.
bandstructure.dat Same as eqp.dat and eqp_q.dat but in a format suitable for plotting
as a bandstructure.

Tricks and hints:
1. To optimize distribution of work among PEs, choose the number of
PEs so that Nk*Nc*Nv in the fine grid is a multiple of the
number of PEs. The parallelization is first done over over kpoints.
2. The Brillouin zone is built using a WignerSeitz construction;
this way the head matrix elements are easily calculated.
3. Check if the transformation matrices have norm close to 1! They
are usually normalized (look at the end of intwfn.f90). The norms are in
the files dvmat_norm.dat and dcmat_norm.dat.
4. Unfolding of irreducible BZ: if you want to skip the unfolding and
use the set of kpoints in WFN_fi as a sampling of the whole BZ, specify
the no_symmetries_* options in absorption.inp.
5. The "number_eigenvalues" keyword: using this keyword tells the code
to store only the first nn eigenvalues/eigenvectors. So far, this option
is implemented only with the SCALAPACK diagonalization routines.
6. Analyzing eigenvectors. We have a tool called summarize_eigenvectors to read in
and analyze eigenvectors for a group of exciton states. For specific states
specified, it sums the total contribution from each kpoint.
Please see the example input summarize_eigenvectors.inp.
BSE/absorption.inp
# absorption.inp
##########################################
# OPTIONS FOR BOTH ABSORPTION AND INTEQP #
##########################################
# Number of occupied bands on fine (interpolated) kpoint grid
number_val_bands_fine 3
# Number of occupied bands on coarse (input) kpoint grid
number_val_bands_coarse 3
# Number of unoccupied bands on fine (interpolated) kpoint grid
number_cond_bands_fine 3
# Number of unoccupied bands on coarse (input) kpoint grid
number_cond_bands_coarse 3
# In metallic systems, PARATEC often outputs incorrect occupation
# levels in wavefunctions. Use this to override these values.
# lowest_occupied_band should be 1 unless you have some very
# exotic situation.
#lowest_occupied_band vmin
#highest_occupied_band vmax
# Specify the Fermi level (in eV), if you want implicit doping
# Note that value refers to energies AFTER scissor shift or eqp corrections.
# If the Fermi level is moved, then you must use both or neither
# of eqp_corrections and eqp_co_corrections, or no interpolation.
# NOTE: If you are using inteqp.x to interpolate eqp_co.dat into a eqp.dat
# file that will be later used in absorption.x, shift the Fermi level only
# in absorption.x, but not in inteqp.x.
#fermi_level 0.0
# The Fermi level is treated as an absolute value
# or relative to that found from the mean field (default)
#fermi_level_absolute
#fermi_level_relative
# Scissors operator (linear fit of the quasiparticle
# energy corrections) for the bands in WFN_fi and WFNq_fi.
# For valenceband energies:
# ev_cor = ev_in + evs + evdel * (ev_in  ev0)
# For conductionband energies:
# ec_cor = ec_in + ecs + ecdel * (ec_in  ec0)
# Defaults below. evs, ev0, ecs, ec0 are in eV
#evs 0.0
#ev0 0.0
#evdel 0.0
#ecs 0.0
#ec0 0.0
#ecdel 0.0
# or
#cvfit 0.0 0.0 0.0 0.0 0.0 0.0
# These flags define whether to use symmetries to unfold
# the Brillouin zone or not in files WFN_fi (unshifted fine
# grid), WFNq_fi (shifted fine grid), WFN_co (coarse grid).
# Warning, the default is always not to unfold!
# If your unshifted fine grid is reduced (most cases),
# you will want to use symmetries here!
#no_symmetries_fine_grid
#use_symmetries_fine_grid
# If your shifted fine grid is reduced (most cases),
# you will want to use symmetries here! Note that
# if you use symmetries in the unshifted grid you will
# need to generate the shifted grid using the same
# procedure as generating the shifted grid for an
# epsilon calculation. i.e. use gw_shift in paratec
# and a nonzero small qshift in kgrid.inp
#no_symmetries_shifted_grid
#use_symmetries_shifted_grid
# If you calculated the coarse grid on a reduced qgrid,
# you should use symmetries here!
#no_symmetries_coarse_grid
#use_symmetries_coarse_grid
# Regular grid used to calculate qpt_averages.
# Default is the kgrid in the WFN_fi file.
# It matters only if you want to perform minicell averages.
#regular_grid n1 n2 n3
# How to calculate optical transition probabilities.
# No default for absorption, one must be specified.
# For inteqp, default is momentum.
# For momentum and JDOS, only WFN_fi is used.
# For velocity, the valence bands come from WFNq_fi.
use_velocity
#use_momentum
# For Haydock there is also a third option to calculate
# Joint density of states. Diagonalization computes this for free.
#use_dos
# Verbosity level, options are:
# 1 = default
# 2 = medium  info about kpoints, symmetries, and eqp corrections.
# 3 = high  full dump of the reduced and unfolded kpoints.
# 4 = log  log of various function calls. Use to debug code.
# 5 = debug  extra debug statements. Use to debug code.
# 6 = max  only use if instructed to, severe performance downgrade.
# Note that verbosity levels are cumulative. Most users will want to stick
# with level 1 and, at most, level 3. Only use level 4+ if debugging the code.
#verbosity 1
# EXPERIMENTAL FEATURES FOR TESTING PURPOSES ONLY (ABSORPTION AND INTEQP)
# 'unfolded BZ' is from the kpoints in the WFN file
# 'full BZ' is generated from the kgrid parameters in the WFN file
# See comments in Common/checkbz.f90 for more details
# Replace unfolded BZ with full BZ
#fullbz_replace
# Write unfolded BZ and full BZ to files
#fullbz_write
# Flag to read kpoints from the 'kpoints' file.
# The default is to read kpoints from the wfn file.
#read_kpoints
# This is needed only with read_kpoints and if you selected velocity operator above.
# This should match the shift between WFN_fi and WFNq_fi.
q_shift s1 s2 s3
# The requested number of valence or conduction bands cannot break degenerate subspace.
# Use the following keyword to suppress this check.
# Note that you must still provide one more band in
# wavefunction file in order to assess degeneracy.
# For inteqp, only the coarse grid is checked.
#degeneracy_check_override
#####################################
# OPTIONS ONLY FOR ABSORPTION BELOW #
#####################################
# Which algorithm do we use to solve the BSE? We can either do a full
# diagonalization with ScaLAPACK or LAPACK) or use the
# Haydock iterative solver.
diagonalization
#haydock
# What screening is present? (default = semiconductor)
screening_semiconductor
#screening_metal
#screening_graphene
# this is for a system with linear DOS at the Fermi level
# Compute only the lowest neig eigenvalues/eigenvectors.
# The default is to calculate all nv*nc*nk of them.
# Only for diagonalization, not for haydock.
#number_eigenvalues neig
# If you are not performing full diagonalization (eg: if using Haydock),
# you must specify the number of iterations.
#number_iterations 100
# Set this to use eigenvalues in eqp.dat and eqp_q.dat
# If not set, these files will be ignored.
#eqp_corrections
# Set this to use eigenvalues in eqp_co.dat
# These quasiparticle corrections will be interpolated to
# the shifted and unshifted fine grids and written to eqp.dat,
# eqp_q.dat, and bandstructure.dat. If not set, this file will be ignored.
#eqp_co_corrections
# RECOMMENDED fast FFTW truncation schemes
# The Coulomb Interaction is cutoff on the edges of
# the WignerSeitz Cell in the nonperiodic directions
# Periodic directions are a1,a2 for slabs and a3 for wires
#cell_box_truncation
#cell_wire_truncation
#cell_slab_truncation
# Analytic, but non WignerSeitz cell, truncation
#spherical_truncation
# For Spherical Truncation
#coulomb_truncation_radius 10.00
# Cutoff energy for averaging the Coulomb Interaction
# in the Mini Brillouin Zones around the Gammapoint
# without Truncation or for Cell Wire or Cell Slab Truncation.
# The value is in Rydbergs, the default is 10^{12}
#cell_average_cutoff 1.0d12
# This is needed if you selected momentum operator above.
# This vector will be normalized in the code.
#polarization p1 p2 p3
# If we have already calculated the optical matrix elements,
# do we read them from file to save time?
#read_vmtxel
# Read 'eps2_moments' generated during previous run (Haydock only)
#read_eps2_moments
# Reduce cost of calculation if we've already computed
# the eigensolutions. This reads the eigenvalues and
# transition matrix elements from the file 'eigenvalues.dat'
#read_eigenvalues
# If we only want the noninteracting spectra (i.e. no electron
# hole interaction, RPA). Only WFN_fi and WFNq_fi are needed as inputs.
# With the 'read_eigenvalues' flag, both absorption_eh.dat and
# absorption_noeh.dat will be created.
#noeh_only
# If already calculated, read the interpolation matrices dvv
# and dcc from file dtmat.
#read_dtmat
# Numerical broadening width and type for generating absorption spectrum.
# Haydock only does Lorentzian. epsilon_1 is always Lorentzian.
# The width is controlled by energy_resolution.
energy_resolution 0.1
# epsilon_2 with diagonalization is Gaussian by default. You can also choose Lorentzian.
#lorentzian_broadening
gaussian_broadening
# Frequency step (eV) for generating absorption spectrum.
#delta_frequency 0.01
# The Gaussian and Voigt options are only available for epsilon_2 with diagonalization.
# Voigt becomes Lorentzian at sigma > 0 and Gaussian at gamma = 0.
# Note that the parametrization of Voigt function diverges at sigma = 0.
#voigt_broadening
#energy_resolution_sigma 0.1
#energy_resolution_gamma 0.01
# This specifies what kind of epsilon we read.
# Default is to read epsmat and eps0mat, but if
# this flag is present we read file epsdiag.dat instead.
# epsdiag.dat contains only the diagonal part of eps(0)mat
# which is all that is needed by absorption. It is always created
# by an absorption run for use in future runs, to save time reading.
#read_epsdiag
# Determines output of eigenvectors (eigenvectors)
# eig = 0 : do not write eigenvectors (default)
# eig < 0 : write all eigenvectors
# eig > 0 : write eig eigenvectors
#write_eigenvectors eig
# Flag the type of kernel to calculate:
# spin_triplet: direct kernel only (no exchange).
# WARNING: triplet transition matrix elements would be exactly for electricdipole transitions,
# so we give instead the spin part of magneticdipole interaction.
# spin_singlet (default): direct + exchange. appropriate for spinpolarized too.
# local_fields: localfields + RPA (exchange only)
# Note that triplet kernel only applies to a spinunpolarized system;
# for a spinpolarized system, the solutions naturally include both singlets and triplets
# (or other multiplicities for a magnetic system).
#spin_triplet
#spin_singlet
#local_fields
# Write the bare Coulomb potential V(q+G) to file
#write_vcoul
# If your two grids are the same, use this flag to skip interpolation and
# reduce the time/memory you need. DO NOT use this if you have a different
# fine grid from the coarse grid, as for velocity operator.
# Then you should use eqp_corrections with eqp.dat (and eqp_q.dat).
#skip_interpolation
# The new interpolation algorithm is based on the Delaunay tessellation of the
# kpoints. This guarantees that the interpolant is a continuous function, and
# that we are always interpolating, and never extrapolating. (default)
#delaunay_interpolation
# You can also use the previous interpolation method, which might give
# interpolants that are not continuous.
#greedy_interpolation
# Read from traditional simple binary format for bsedmat/bsexmat instead of HDF5 file format.
# Relevant only if code is compiled with HDF5 support.
#dont_use_hdf5
# EXPERIMENTAL FEATURES FOR TESTING PURPOSES ONLY (ABSORPTION ONLY)
# Average the head of W in addition to the head of V using a model
# for the inverse dielectric matrix. This is done for insulators for
# all truncation schemes. The W average is limited only to the first
# minibz even if cell_average_cutoff != 0. Currently it is always done
# regardless whether average_w is included in the input file or not.
#average_w
# Multiply kernel by arbitrary factor. Default is 1.0 of course.
kernel_scaling 1.0
# Use the same flag as in kernel.inp.
#extended_kernel
# Perform the co/fi WFN transformation and eqp interpolation without restricting
# the character of the coefficients, i.e., write:
# v> = \sum_n d_vn` n`>, where n`> is either v`> or c`>.
# This is a very good idea for metallic systems, but it does not work so well
# for semiconductors because of the inherit approximation used in the
# interpolation. Notes:
#  If extended_kernel is not set, we restrict the d_vn` and d_cn` coefficients
# to the subspaces of d_cc` and d_vv` after the eqp interpolation.
#  If extended_kernel is set, this flag is automatically turned on, and we
# keep the unrestricted d_vn` and d_cn` coefficients for the kernel interpolation.
#unrestricted_transformation
# Zero out all dvc'/dcv' coefficients. Use this flag together with extended_kernel
# to reproduce the result of a restricted kernel calculation.
#zero_unrestricted_contribution
# The averaging scheme performed for the q=0 term of the BSE Hamiltonian
# yields exciton binding energies that do not converge variationally. One
# can fix this issue by zeroing out the q=0 term of the interaction, either
# before or after the interpolation of the kernel of the BSE.
# See Figure 7 of PRB 93, 235435 (2016).
# zero_q0_element n
# where n is:
# 0: don`t zero the W(q=0) term (default)
# 1: zero W(q=0) term after interpolation.
# 2: zero W(q=0) before interpolation (not recommended)
BSE/inteqp.inp
# inteqp.inp
# inteqp.inp uses a subset of the parameters from absorption.inp
# Please see absorption.inp for examples.
BSE/README_inteqp
================================================================================
BSE code: Inteqp
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.0 (July, 2011) J. Deslippe, D. A. Strubbe

Description:
This code takes the eqp_co.dat file from Sigma and
interpolates it from the coarse to the fine grid using wavefunction
projections (to resolve band crossings) and linear interpolation
in the Brillouin zone.

Required Input:
inteqp.inp Input parameters. See example absorption.inp in this directory
for complete options (noting which sections apply to inteqp).
WFN_fi Wavefunctions in unshifted fine grid.
WFN_co Wavefunctions on coarse grid.
eqp_co.dat A list of quasiparticle energy corrections
for the bands in WFN_co.
Optional Input:
WFNq_fi Wavefunctions in shifted fine grid
(used for valence bands if use_velocity is selected)
Auxiliary Files: (output files from previous runs  used as input to speed up calculation)
dtmat Transformation matrices, dcc/dvv use for interpolation
between coarse and fine grid. This file must be consistent
with your bsedmat and bsexmat files and corresponding
coarse and fine wavefunctions.
NOTE: the file format for dtmat was changed in BerkeleyGW 1.1.0 (r5961)

Output Files:
bandstructure.dat The GW bandstructure on the fine grid.
eqp.dat Quasiparticle energy corrections for the bands in WFN_fi.
eqp_q.dat Quasiparticle energy corrections for the bands in WFNq_fi (if use_velocity).
dvmat_norm.dat The norms of the dvv overlap matrices between the valence
band k on the fine grid and the closest kpoint on the coarse grid
dcmat_norm.dat The norms of the dcc overlap matrices between the conduction
band k on the fine grid and the closest kpoint on the coarse grid
BSE/summarize_eigenvectors.inp
.true. ! Whether you used TDA. This will usually be .true.
20 ! Number of eigenvectors in file. Set to zero to use default ns*nv*nc*nk
1.5 2.1 ! Emin Emax (eV). Energy window to print information about all states in
3 ! Number of specific states to print A(k). Output files will be exciton_01 ... exciton_99
1.56783332 ! Energy values (eV) of the states to print A(k)
1.67345203
1.96328918
PlotXct
PlotXct/README
================================================================================
PlotXct code
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014)
Version 1.0 (July, 2011)
Version 0.5 F. Ribeiro (2008), S. IsmailBeigi
Version 0.2 M. L. Tiago (20022007)

Description:
PlotXct plots the exciton wavefunction in real space based on
output of BSE/absorption (with full diagonalization).
Reads files WFN_fi/WFNq_fi (from mean field) and eigenvectors (output
from BSE/absorption code when using diagonalization, not Haydock) and
plots selected exciton state in real space according to:
Psi(r,r`,s) = Sum_cvk A_svck * phi_sck(r) * exp(ik.r) * conjg( phi_svk(r`) * exp(i[k+q].r`)
where r`, hole coordinate, is fixed at some point and r, electron coordinate,
runs over a supercell of given size (typically, equivalent to the fine kgrid),
with a realspace mesh defined by the FFT grid of the wavefunction files.
A separate plot can be generated for each spin s.
Output is written in format suitable for Data Explorer (DX).
The input option 'restrict_kpoints' reduces the sum over kpoints
above to a sum over the ones that give most of the contribution
to the norm of eigenvectors. This is handy if there are many kpoints
but only a few of them give sizable contribution.
input: WFN_fi WFNq_fi wavefunction files
eigenvectors output from BSE/absorption code
plotxct.inp input parameters

1. Copy/edit plotxct.inp.
You'll need:
. index of the exciton state you want to plot
. hole position in lattice coordinates
. supercell dimensions (or kgrid)
. spin component you want to plot (if spinpolarized)
2. Run plotxct.x
The ASCII file 'xct.[state]_s[spin].a3Dr' will be created.
'a3Dr' means that the file is in ASCII and contains 3D data in 'r' real space.
The header of this file contains information on
. state index
. state energy in eV
. hole position in atomic units (a.u.)
. spin component plotted
. supercell lattice vectors in a.u.
. number of discretization points
followed by three columns of data corresponding to the complex values of the
electronic part of the excitonic wavefunction, and its magnitude squared.
The values are written with very low precision due to filesize concerns.
3. Convert to a format readable by the plotting utility of your choice.
Use the 'volume.py' utility:
Usage: volume.py imfn imff ivfn ivff ovfn ovff phas cplx [hole]
imfn = input matter file name
imff = input matter file format
(matparatecvaspespressosiestatbpwxyzxsf)
ivfn = input volumetric file name
ivff = input volumetric file format (a3dr)
ovfn = output volumetric file name
ovff = output volumetric file format (cubexsf)
phas = remove wavefunction phase (truefalse)
cplx = complex wavefunction (reimabsabs2)
hole = plot hole (truefalse)
You can specify whether you want a .xsf (XCrysDen) or .cube (Gaussian Cube)
file format. You can remove or keep the arbitrary phase of the wavefunction
by setting parameter "phas" to true or false. You can plot the re, im, abs,
or abs2 part of the wavefunction.
You must specify a path to the input file (example: "../01scf/input")
in one of the supported formats (DFT and others) so that the script can
obtain the atomic positions from the file.
You can choose whether to display or not the position of the hole by setting
parameter "hole" to true or false.
The resulting .xsf or .cube file can be viewed in XCrysDen or converted to
POVRay script format using Surface code (for more details on the latter
option, see Visual/README).
****************
Example:
% ~/BerkeleyGW/Visual/volume.py ../01scf/input paratec xct.000001_s1.a3Dr a3dr \
xct.000001.abs2.xsf xsf false abs2 false
% xcrysden xsf xct.000001.abs2.xsf
****************
PlotXct/plotxct.inp
#
# Index of state to be plotted, as it appears in eigenvectors
#
plot_state 20
#
# Index of spin component of the exciton to plot. Default is 1.
#
plot_spin 1
#
# Size of supercell
#
supercell_size 1 1 60
#
# coordinates of the hole in crystal coordinates, in units of supercell lattice vectors
# (usually, the hole is near the center of the supercell)
# (e.g. center of 1 1 60 supercell should be written as 0.5 0.5 30.0)
#
hole_position 0.8025 0.3487 30.00
# input option 'restrict_kpoints' reduces the sum over kpoints
# above to a sum over the specified number that give most of the contribution
# to the norm of eigenvectors. This is handy if there are many kpoints
# but only a few of them give sizable contribution.
#restrict_kpoints 5
# qshift used in the calculation of valence bands (WFNq_fi file)
# Only needed if restrict_kpoints is used, otherwise determined automatically.
q_shift 0.00000 0.00000 0.00000
# Verbosity level, options are:
# 1 = default
# 2 = medium  info about kpoints, symmetries, and eqp corrections.
# 3 = high  full dump of the reduced and unfolded kpoints.
# 4 = log  log of various function calls. Use to debug code.
# 5 = debug  extra debug statements. Use to debug code.
# 6 = max  only use if instructed to, severe performance downgrade.
# Note that verbosity levels are cumulative. Most users will want to stick
# with level 1 and, at most, level 3. Only use level 4+ if debugging the code.
#verbosity 1
# EXPERIMENTAL FEATURES FOR TESTING PURPOSES ONLY
# 'unfolded BZ' is from the kpoints in the WFN file
# 'full BZ' is generated from the kgrid parameters in the WFN file
# See comments in Common/checkbz.f90 for more details
# Replace unfolded BZ with full BZ
#fullbz_replace
# Write unfolded BZ and full BZ to files
#fullbz_write
NonLinearOptics
NonLinearOptics/README
================================================================================
NonLinearOptics code
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.0 (Aug, 2011) (Alpha)
Version 0.5 (July, 2008) (Alpha)
Version 0.2 J. Deslippe (2008) (Alpha)

Description:
This code is intended to be run after a full GW/BSE calculation and
generates the UltraFast or 2photon spectrum of a material by calculating
the optical transition strength between the lowest excited exciton state
(user inputed) and higher excited exciton states.
WARNING  This code should be considered alpha quality. It has been tested
on only a few systems. Use at your own risk and validate.

Required Input:
nonlinearoptics.inp Input parameters. See detailed explanation below.
eigenvectors The exciton wavefunctions in Bloch space.
WFN_fi Bloch wavefunctions on the unshifted fine grid.
WFNq_fi Bloch wavefunctions on the shifted fine grid.
Only needed if velocity operator is used.
Additional Input:
eqp.dat This is not yet implemented (energies are read from A_svck)
If it were to work, it would probably be:
A list of quasiparticle energy corrections for the bands
in WFN_fi and WFNq_fi. If absent the scissors operator from
nonlinearoptics.inp is used.
Auxiliary Files: (output files from previous runs  used as input to speed up calculation)
vmtxel_nl The interband optical matrix elements. This is different
from the file vmtxel created by absorption in that it contains
valencevalence and conductionconduction matrix elements.
Specify read_vmtxel_uf in the nonlinearoptics.inp file and files
WFN_fi and WFNq_fi are not needed.

nonlinearoptics.inp Please see example nonlinearoptics.inp in this directory
for more complete options.

Output Files:

Tricks and hints:
NonLinearOptics/nonlinearoptics.inp
# nonlinearoptics.inp
# Select type of run
#two_photon_job (Two Photon Calculation)
#ultrafast_job (UltraFastSpec)
# fine grid is used to actually do the diagonalization; here the bands
# count down/up from the Fermi level (as appropriate)
number_val_bands_fine 10
number_cond_bands_fine 10
# What type of operator do we use. See explanation in BSE directory
#use_velocity
#q_shift 0.0 0.00 0.00
use_momentum
polarization 0.0 0.0 1.0
# If we have already calculated matrix elements
# (stored in file vmtxel_nl) read them in.
#read_vmtxel_nl
# In eV, width of Lorentzian broadening
energy_resolution 0.025
# Scissors operator (linear fit of the quasiparticle
# energy corrections) for the bands in WFN_fi and WFNq_fi.
# For valenceband energies:
# ev_cor = ev_in + evs + evdel * (ev_in  ev0)
# For conductionband energies:
# ec_cor = ec_in + ecs + ecdel * (ec_in  ec0)
# Defaults below. evs, ev0, ecs, ec0 are in eV
#evs 0.0
#ev0 0.0
#evdel 0.0
#ecs 0.0
#ec0 0.0
#ecdel 0.0
# or
#cvfit 0.0 0.0 0.0 0.0 0.0 0.0
# Are the grids (fine from diag calc) to be expanded using symmetries?
#use_symmetries_fine_grid
#use_symmetries_shifted_grid
# EXPERIMENTAL FEATURES FOR TESTING PURPOSES ONLY
# 'unfolded BZ' is from the kpoints in the WFN file
# 'full BZ' is generated from the kgrid parameters in the WFN file
# See comments in Common/checkbz.f90 for more details
# Replace unfolded BZ with full BZ
#fullbz_replace
# Write unfolded BZ and full BZ to files
#fullbz_write
# TWOPHOTON ONLY
# One can limit the number of final states to consider in the twophoton process
# to a subset of neig (the default value used).
#number_of_final_states 500
Visual
Visual/README
================================================================================
Visualization tools
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014)
Version 1.0 (July, 2011) J. Deslippe, M. Jain, D. A. Strubbe, G. Samsonidze
Version 0.5 J. Deslippe, D. Prendergast, L. Yang, F. Ribeiro, G. Samsonidze (2008)
Version 0.2 M. L. Tiago, C. Spataru, S. IsmailBeigi (2004)

Description:
This directory contains a set of visualization tools designed to
simplify your work with the DFT codes and the BerkeleyGW package.
Here you will find the following scripts and codes:
1. Surface is a C++ code for generating an isosurface of a volumetric
scalar field (such as the wave function, charge density, or local
potential). The scalar field is read from Gaussian Cube or XCrySDen
XSF file, the surface triangulation is performed using marching cubes
or marching tetrahedra algorithm, and the isosurface is written in
the POVRay scripting language. The final image is rendered using
the raytracing program POVRay.
Running the code requires a fairly complicated input parameter file,
described with an example in surface.inp.
2. Matter is a python library for manipulating atomic structures with
periodic boundary conditions. It can translate and rotate the atomic
coordinates, generate supercells, assemble atomic systems from fragments,
and convert between different file formats. The supported file formats
are mat, paratec, vasp, espresso, siesta, xyz, xsf, and povray.
mat is the native file format of the library. paratec, vasp, espresso,
and siesta represent the formats used by different planewave and
localorbital DFT codes. xyz is a simple format supported by many
molecular viewers, and xsf is the internal format of XCrySDen viewer.
povray stands for a scripting language used by the raytracing
program POVRay.
The core of the library consists of files common.py, matrix.py, and
matter.py. Script convert.py is a commandline driver that performs
the basic operations supported by the library. Script link.py is a
molecular assembler that can be used to rotate and link two molecules
together. Script gsphere.py generates a realspace grid and a sphere
of Gvectors given lattice parameters and a kinetic energy cutoff.
This helps to estimate the number of unoccupied states needed in GW
calculations. Script average.py takes an average of the scalar field
on the faces or in the volume of the unit cell. This is used to
determine the vacuum level in DFT calculations. Script volume.py
converts the a3Dr file produced by PARATEC or BerkeleyGW to Gaussian
Cube or XCrySDen XSF format.
Each script requires a different set of commandline arguments.
Running individual scripts without arguments displays a list of
all possible commandline arguments and a short description of
each argument.

Examples:
1. Benzene Charge Density
This example demonstrates how to plot isosurfaces.
Go to directory ./benzene and run ESPRESSO:
$ sh link
Submit script. Suggested ncpu = 36, walltime = 0:30:00
This will produce the Gaussian Cube file rho.cube containing
the charge density of the benzene molecule. For your convenience,
the compressed file rho.cube.tgz is included in the ./benzene
directory.
Let us generate an isosurface that contains 90% of the charge
density using the marching cubes algorithm with the smooth triangles
and render it in POVRay:
$ ../surface.x rho_uc.inp
$ povray A0.3 +W640 +H480 +Irho_uc.pov
Examine the rho_uc.gif file to find that the pieces of the benzene
molecule are placed in the corners of the unit cell. To assemble the
pieces together we define a supercell in the rho_sc.inp file. This
is done by setting the parameter sc to T and by placing the supercell
origin (sco) at position (0.5 0.5 0.5) in crystal coordinates
(scu which stands for supercell units is set to latvec). The lattice
vectors of the supercell are not changed (scv or supercell vectors
in the parameter file). Let us render a new image:
$ ../surface.x rho_sc.inp
$ povray A0.3 +W640 +H480 +Irho_sc.pov
Now the charge density comes up nice and centered in the middle
of the supercell.
2. Nanotube Exciton Wavefunction
In this example we will plot an isosurface of the exciton
wavefunction. Run ../examples/DFT/swcnt_80/ fully. Then
go to directory ../examples/DFT/swcnt_80/8absorption/.
PlotXct will produce the a3Dr file xct.020_s1.a3Dr containing
a wavefunction of the 20th exciton in the (8,0) nanotube.
Then volume.py will convert the a3Dr file to Gaussian Cube.
$ volume.py ../ESPRESSO/1scf/in espresso xct.020_s1.a3Dr a3dr xct020.cube cube false abs2 true
The command will produce the Gaussian Cube file xct020.cube that
contains the squared absolute value of the wavefunction
(cplx = abs2) without the sign (phas = false) and the
position of the hole (hole = true). Then surface will generate an
isosurface that contains 90% of the charge density.
$ ../surface.x xct020.inp
You can then render it in POVRay (which you must install separately):
$ povray A0.3 +W2560 +H1920 +Ixct020.pov
The resulting image xct020.gif is included in the example directory.
Alternately, you can visualize the Cube file directly in XCrySDen.
3. RockSalt Lattice
Here you will learn how to make large supercells of bulk crystals.
Go to directory ./rocksalt where you will find the nacl.mat file.
Render it in POVRay:
$ ../convert.py nacl.mat mat nacl.pov povray
$ povray A0.3 +W640 +H480 +Inacl.pov
The unit cell contains only two atoms. Let us make a supercell that
contains a fractional number of unit cells. The supercell is described
by the sc1 file which defines the origin of the supercell, the lattice
vectors of the supercell, and the units in which these quantities
are given. These are equivalent to sco, scv, and scu in the surface
parameter file in the first example. The last line in the sc1 file
corresponds to sct (supercell translation) in the surface parameter
file. We will get back to it later on. Let us render the supercell:
$ ../convert.py nacl.mat mat nacl_sc1.pov povray supercell sc1
$ povray A0.3 +W640 +H480 +Inacl_sc1.pov
Note the broken bonds on the faces of the supercell.
That is because the supercell contains a fractional number
of the unit cells, so the translational symmetry is broken
which results in the NaNa and ClCl bonds. We disable the
translational symmetry of the supercell by setting sct to
false (the last line in the sc2 file). This enforces the
translational symmetry of the underlying unit cell structure.
Render the new supercell in POVRay:
$ ../convert.py nacl.mat mat nacl_sc2.pov povray supercell sc2
$ povray A0.3 +W640 +H480 +Inacl_sc2.pov
Now the bonds on the faces of the supercell come out right.
4. Organic Molecule Synthesis
In this example you will assemble the bipolar molecule, bithiophene
naphthalene diimide, from the donor and acceptor units, bithiophene
and naphthalene diimide. Go to directory ./btnd and open files
nd.xyz and bt.xyz in any molecular viewer, for example Jmol.
Identify the atoms you want to link together, these are the two
carbon atoms, # 15 in nd.xyz and # 8 in bt.xyz. You need to remove
the two hydrogen atoms, # 21 in nd.xyz and # 14 in bt.xyz, and place
# 15 and # 8 at the distance of 1.49 Angstrom from each other along
the 1521148 line. This is done by invoking the following command:
$ ../link.py nd.xyz xyz bt.xyz xyz btnd.xyz xyz 15 21 8 14 0.0 degree 1.49 angstrom
You can also rotate bt.xyz around the 1521148 axis by an arbitrary
angle, although the rotation angle is set to zero in the above command.
5. Estimate the number of unoccupied states for Epsilon and Sigma
Go to directory ../examples/DFT/benzene/0gsphere/ and run the following:
$ gsphere.py g_eps.in g_eps.out
Examine input file g_eps.in. It contains the lattice vectors in bohr,
the cutoff energy in Ry (epsilon_cutoff from ../5gw/epsilon.inp or
screened_coulomb_cutoff from ../5gw/sigma.inp), '0 0 0' for the FFT
grid to be determined automatically, and 'true' to sort the Gvectors
by their kinetic energies. Examine output file g_eps.out. Look for the
number of Gvectors, ng. You will find 'ng = 2637' meaning that you
will need at least that many unoccupied states. Of course the number
of unoccupied states is a convergence parameter, but gsphere.py gives
you an idea in which range of values should you check the convergence.
Now run the following:
$ gsphere.py g_rho.in g_rho.out
Input file g_rho.in is similar to g_eps.in but it has a different
cutoff energy which is 4 times ecutwfc from ../ESPRESSO/1scf/in.
This is the charge density cutoff or ecutrho in espresso documentation.
Note that the last line in g_rho.in is set to 'false'. This is because
using builtin Python sort function on that many Gvectors (up to
ecutrho) will freeze your python interpreter. Examine output file
g_rho.out. You should see the following:
grid = ( 128 72 140 )  minimal
( 128 72 140 )  factors 2, 3, 5, 7, 1*11, 1*13
( 128 72 144 )  factors 2, 3, 5
The third line is the size of FFT grid in espresso calculation.
(Well, not necessarily, because espresso algorithm starting with
version 5 is smarter than what is implemented in gsphere.py.)
The zcomponent is equal to 144 meaning that for an optimal
loadbalancing you should run espresso on a number of cores
which is a divisor of 144 (that is, 144, 72, 36, 18, etc. cores).

Notes:
Matter library may lack support for some advanced features
of PARATEC, VASP, ESPRESSO and SIESTA formats. For example,
LatticeParameters and ZMatrix are not implemented for SIESTA.
This can be added to functions paratec_read, paratec_write,
vasp_read, vasp_write, espresso_read, espresso_write,
siesta_read and siesta_write in file matter.py.
Also note that all the additional information not related
to the crystal structure (kpoints, energy cutoffs, etc.)
will be lost and replaced with the default parameters when
converting between PARATEC, VASP, ESPRESSO and SIESTA formats.
Visual/surface.inp
parameter file  example (HOMO of benzene)

header text  header C6H6_band_15
inputfilename file  inputfilename C6H6.b_15.cube
inputfileformat cubexsf  inputfileformat cube
outputfilename file  outputfilename C6H6.b_15.pov
outputfileformat povray  outputfileformat povray
isovalue (0.0,1.0)  isovalue 0.9
sign positivenegativeboth  sign both
power 012  power 1
algorithm cubetetrahedron  algorithm cube
smooth TF  smooth T
box TF  box F
basis TF  basis F
uc TF  uc F
uco  uco
ucox ucoy ucoz  0.0 0.0 0.0
ucv  ucv
ucv1x ucv1y ucv1z  1.0 0.0 0.0
ucv2x ucv2y ucv2z  0.0 1.0 0.0
ucv3x ucv3y ucv3z  0.0 0.0 1.0
ucu bohrangstromlatvec  ucu latvec
sc TF  sc T
sco  sco
scox scoy scoz  0.5 0.5 0.5
scv  scv
scv1x scv1y scv1z  1.0 0.0 0.0
scv2x scv2y scv2z  0.0 1.0 0.0
scv3x scv3y scv3z  0.0 0.0 1.0
scu bohrangstromlatvec  scu latvec
sct TF  sct T

the isosurface is defined by isovalue * max psi for power = 0 and by
integral_v psi^power = isovalue * integral psi^power for power > 0
where psi is the scalar field and v is the volume inside the isosurface
note that espresso already outputs psi^2, setting power = 2 yields
psi^4 in this case, to produce psi^2 isosurface set power = 1
algorithm stands for marching cubes and marching tetrahedra
smooth defines whether to average normals for smooth rendering
box specifies whether to generate supercell box in povray script
basis specifies whether to generate coordinate axes in povray script
sfo = scalar field origin sfv = scalar field vectors
uco = unit cell origin ucv = unit cell vectors ucu = unit cell units
sco = supercell origin scv = supercell vectors scu = supercell units
sct = supercell translational symmetry
sfo/sfv are read from volumetric file
if uc = T then uco/ucv are read from parameter file
else uco/ucv are set to sfo/sfv
if sc = T then sco/scv are read from parameter file
else sco/scv are set to uco/ucv
if uc = T and ucu = latvec then uco/ucv are scaled by sfv
if sc = T and scu = latvec then sco/scv are scaled by ucv
if sct = T then scv is used for translational symmetry
else ucv is used for translational symmetry
in most cases, set uc = F (uc = T is only needed if volumetric data
do not span the whole unit cell and the supercell is being used, so
the correct unit cell must be defined)
use sc = T to construct the supercell spanning several unit cells or
to assemble the unit cell around the origin of the coordinate system
as in benzene example above (sco = (0.5 0.5 0.5) in latvec units)
the supercell may span a fractional number of unit cells, in this case
set sct = F to produce the correct bonds at the faces of the supercell
Visual/gsphere.inp
25.917768986 0.000000000 0.000000000
0.000000000 14.496126116 0.000000000
0.000000000 0.000000000 28.284379996
240.0
0 0 0
false
0.0 0.0 0.0
# example input file for gsphere.py, for benzene
# a1x a1y a1z [= lattice vectors (bohr)]
# a2x a2y a2z
# a3x a3y a3z
# energy cutoff (Ry)
# FFT_grid_x FFT_grid_y FFT_grid_z (set to zero to determine automatically)
# whether to sort Gvectors by kinetic energy (true) or leave unsorted (false)
# kx ky kz (kpoint at which to generate Gvectors, crystal coordinates)
Developers
doxygen/README
================================================================================
BerkeleyGW Doxygen documentation
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.
Version 1.1 (June, 2014)
Version 1.0 (July, 2011) J. Deslippe, M. Jain, D. A. Strubbe, G. Samsonidze
Version 0.5 J. Deslippe, D. Prendergast, L. Yang, F. Ribeiro, G. Samsonidze (2008)
Version 0.2 M. L. Tiago, C. Spataru, S. IsmailBeigi (2004)

Description

This directory automatically generates documentation for BerkeleyGW
using Doxygen. This is aimed towards developers, and endusers might
as well just ignore this directory.
To create the documentation, simply type `make`. You`ll need doxygen and
graphviz installed. Then open doxygen/html/index.html in a web browser.
Obtaining Doxygen and Graphviz

Doxygen and graphviz are available at:
http://www.doxygen.org/
http://www.graphviz.org/
Alternatively, if you have a Debianbased distribution, such as
Ubuntu, you can ask your administrator to install the following packages:
aptget install doxygen graphviz
or on MacOS with MacPorts (www.macports.org):
port install doxygen graphviz
Common/README
================================================================================
BerkeleyGW: Common directory (developer info)
================================================================================
Version 1.2 (Aug, 2016)
F. H. da Jornada, J. Deslippe, D. VigilFowler, J. I. Mustafa, T. Rangel,
F. Bruneval, F. Liu, D. Y. Qiu, D. A. Strubbe, G. Samsonidze, J. Lischner.

Description:
This directory contains common files shared by different components of
the BerkeleyGW package. The most important files are described below:
1. typedefs.f90
This file contains the derived data types that are used throughout
the code.
2. nrtype.f90
This file contains some general constants and the definitions of
single and doubleprecision real and complex types following the
convention of the Numerical Recipes book. The definitions of DP
and DPC are used throughout the code.
It also contains the fundamental physical constants. The units used
throughout the code are Rydberg atomic units with a few exceptions,
such as eV in Sigma and Hartree atomic units in EPM. The difference
between Hartree and Rydberg atomic units is sketched below.
Hartree atomic units:
hbar=1 e=1 m=1 a0=hbar^2/(m*e^2)=1 E=hbar^2/(2*m*a0^2)=0.5
Rydberg atomic units:
hbar=1 e=sqrt(2) m=0.5 a0=hbar^2/(m*e^2)=1 E=hbar^2/(2*m*a0^2)=1
3. f_defs.h
This file contains preprocessor definitions for the builtin Fortran
functions. This helps to avoid accidental conversion from double to
single precision.
Common/qhull/README
DISCLAIMER
==========
This directory contains source code from Qhull. The file COPYING.txt
has the terms for distributing these files.
The following files are part of the BerkeleyGW package, and not Qhull:
 libtile_qhull.c
 libtile_qhull.h
 test_delaunay_f.f90
 test_delaunay_c.c
 Makefile
 README
The files listed above are distributed under the BerkeleyGW license
found in the topmost directory.
CHANGES TO QHULL
================
The following change had to be made to Qhull:
 qhull_a.h: line 105
Remove macro for intel compilers.
Source: http://mail.scipy.org/pipermail/scipyuser/2011July/029908.html
Common/voro++/README.BerkeleyGW
DISCLAIMER
==========
This directory contains source code from voro++0.4.6. The file LICENSE has
the terms for distributing these files.
The following files are part of the BerkeleyGW package, and not voro++:
 libtile_voro.cc
 libtile_voro.h
 test_voro.f90
 README.BerkeleyGW
The following files were removed from the original voro++ distribution:
 examples/*
 html/*
 man/*
 scripts/*
 src/README
 src/Doxyfile
 src/Makefile.dep
The following files were changed to better integrate with BerkeleyGW:
 Makefile
 Makefile.dep
 src/Makefile
MeanField/spglib1.0.9/README
* Imported spglib1.0.9 from http://spglib.sourceforge.net. This is a BSDlicensed C library for symmetries by Atsushi Togo. DAS
* Added src/spglib_f.c which is not actually part of spglib 1.0.9 although it appears to have been forgotten by accident since it is mentioned in the documentation and was present in earlier version. I obtained it by emailing the author. DAS
* Removed test directory which we do not need and contains a large number of files. DAS
* Removed examples, python, etc., blank files, and automake/libtoolrelated files. DAS
..o.
.oxxo.
.oxxxxo...
oxxxxxxo.
.oxxxxxxx.
.ooooooxxo..
.oooooooxo..
.oooooxxo...
.........oxooo......
............................
.................................
....................................
. ..oo. .... .................................oooxxxxxxxo.
.............oxxxx@ox@@@x@x.....................o...........ooooooooooxx.
.o.........oox@x.oo........xxx@@............ooxxxxo..........ooooxxxxxoxo
.x........x@xxo...............o@xxo........oxxx@@@xoooooooooooooooxxxo...
.o......ox@@o..................oox@o.....ooxxx@xoooxxxxxxxoooooooooooo....
o..ooooo@@xoooo....ooo...........x@o.....ooxxxxo .oxxxxxxxxxxooooooo....
. .oooo@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@....ooooox. .oxx@@@@@xxoo........
.ooooxxxxxxxooooooxooooooooooooooxo...oooooxx. ..ox@xxxoo.........
.ooooooxxxx@xoooooooooooooxxoooooooooooxxxxx. .oxxooooooooxxo.
.oooooxxxxx@@@xxxxxxxxxxxxxxxxxxxxxxxxoxxxxo. .oxxxxxxxxo.
....oooxxxxx@@@xo..oxxx@@@@@@xxxxoxxoooooooxx. .oxxxoo..
.....ooxxxx@@xo. ........ .ooooooooxxo
..oooxxxx@@@o .oooooooxoo.
....oxooxxxxx. .ooo..oooo.
.....o.ooxxxxxo. .oooooooxo.
......ooooxxxxxxo. .ooooooxoo..
........ooxxxxxxxo.. .o....oxoo...
.......ooooxxxxxxxo. ........oooo.
.ooooooo..ooxxxxoooo. .........ooo...
..oxo...ooooxxxoooo.. .ooo......oooo...
.ooooo....o. .oxxxoo....ooo....
.oooooo... ...ooooo...ooo..
... .oo.......
....ooo...
__ __
______ [  [  ._____ _ _
_ _ \   _   / ___ \   
 _)  .. _. _.  /  ..   .. _ _ / / \_\ \ /\ / /
 __'./ /__\\[ /`\_ '' < / /__\\  / /__\\ \    _____ \ \/ \/ /
_ __  \__.    `\ \  \___.   \___. \ \/ / \ \.___  \ /\ /
_______/ \.__./[_] [__ \_] \.__./[___]\.__./ \ / \.____./ \/ \/
/ /
/_/