| 1 |
|
| 2 |
<html> |
| 3 |
<body> |
| 4 |
<pre> |
| 5 |
|
| 6 |
------------------------------------------------------------------------ |
| 7 |
INPUT FILE DESCRIPTION |
| 8 |
|
| 9 |
Program: ph.x / PWscf / Quantum Espresso (version: 6.6) |
| 10 |
------------------------------------------------------------------------ |
| 11 |
|
| 12 |
|
| 13 |
Input data format: { } = optional, [ ] = it depends, # = comment |
| 14 |
|
| 15 |
Structure of the input data: |
| 16 |
=============================================================================== |
| 17 |
|
| 18 |
title_line |
| 19 |
|
| 20 |
&INPUTPH |
| 21 |
... |
| 22 |
/ |
| 23 |
|
| 24 |
[ xq(1) xq(2) xq(3) ] # if "ldisp" != .true. and "qplot" != .true. |
| 25 |
|
| 26 |
[ nqs # if "qplot" == .true. |
| 27 |
xq(1,i) xq(2,i) xq(3,1) nq(1) |
| 28 |
... |
| 29 |
xq(1,nqs) xq(2,nqs) xq(3,nqs) nq(nqs) ] |
| 30 |
|
| 31 |
[ atom(1) atom(2) ... atom(nat_todo) ] # if "nat_todo" was specified |
| 32 |
|
| 33 |
|
| 34 |
|
| 35 |
======================================================================== |
| 36 |
Line of input: |
| 37 |
|
| 38 |
title_line |
| 39 |
|
| 40 |
|
| 41 |
DESCRIPTION OF ITEMS: |
| 42 |
|
| 43 |
+-------------------------------------------------------------------- |
| 44 |
Variable: title_line |
| 45 |
|
| 46 |
Type: CHARACTER |
| 47 |
Description: Title of the job, i.e., a line that is reprinted on output. |
| 48 |
+-------------------------------------------------------------------- |
| 49 |
|
| 50 |
===End of line-of-input================================================= |
| 51 |
|
| 52 |
|
| 53 |
======================================================================== |
| 54 |
NAMELIST: &INPUTPH |
| 55 |
|
| 56 |
+-------------------------------------------------------------------- |
| 57 |
Variable: amass(i), i=1,ntyp |
| 58 |
|
| 59 |
Type: REAL |
| 60 |
Default: 0.0 |
| 61 |
Description: Atomic mass [amu] of each atomic type. |
| 62 |
If not specified, masses are read from data file. |
| 63 |
+-------------------------------------------------------------------- |
| 64 |
|
| 65 |
+-------------------------------------------------------------------- |
| 66 |
Variable: outdir |
| 67 |
|
| 68 |
Type: CHARACTER |
| 69 |
Default: value of the ESPRESSO_TMPDIR environment variable if set; |
| 70 |
current directory ('./') otherwise |
| 71 |
Description: Directory containing input, output, and scratch files; |
| 72 |
must be the same as specified in the calculation of |
| 73 |
the unperturbed system. |
| 74 |
+-------------------------------------------------------------------- |
| 75 |
|
| 76 |
+-------------------------------------------------------------------- |
| 77 |
Variable: prefix |
| 78 |
|
| 79 |
Type: CHARACTER |
| 80 |
Default: 'pwscf' |
| 81 |
Description: Prepended to input/output filenames; must be the same |
| 82 |
used in the calculation of unperturbed system. |
| 83 |
+-------------------------------------------------------------------- |
| 84 |
|
| 85 |
+-------------------------------------------------------------------- |
| 86 |
Variable: niter_ph |
| 87 |
|
| 88 |
Type: INTEGER |
| 89 |
Default: maxter=100 |
| 90 |
Description: Maximum number of iterations in a scf step. If you want |
| 91 |
more than 100, edit variable "maxter" in PH/phcom.f90 |
| 92 |
+-------------------------------------------------------------------- |
| 93 |
|
| 94 |
+-------------------------------------------------------------------- |
| 95 |
Variable: tr2_ph |
| 96 |
|
| 97 |
Type: REAL |
| 98 |
Default: 1e-12 |
| 99 |
Description: Threshold for self-consistency. |
| 100 |
+-------------------------------------------------------------------- |
| 101 |
|
| 102 |
+-------------------------------------------------------------------- |
| 103 |
Variable: alpha_mix(niter) |
| 104 |
|
| 105 |
Type: REAL |
| 106 |
Default: alpha_mix(1)=0.7 |
| 107 |
Description: Mixing factor (for each iteration) for updating |
| 108 |
the scf potential: |
| 109 |
|
| 110 |
vnew(in) = alpha_mix*vold(out) + (1-alpha_mix)*vold(in) |
| 111 |
+-------------------------------------------------------------------- |
| 112 |
|
| 113 |
+-------------------------------------------------------------------- |
| 114 |
Variable: nmix_ph |
| 115 |
|
| 116 |
Type: INTEGER |
| 117 |
Default: 4 |
| 118 |
Description: Number of iterations used in potential mixing. |
| 119 |
+-------------------------------------------------------------------- |
| 120 |
|
| 121 |
+-------------------------------------------------------------------- |
| 122 |
Variable: verbosity |
| 123 |
|
| 124 |
Type: CHARACTER |
| 125 |
Default: 'default' |
| 126 |
Description: |
| 127 |
Options are: |
| 128 |
|
| 129 |
'debug', 'high', 'medium' : |
| 130 |
verbose output |
| 131 |
|
| 132 |
'low', 'default', 'minimal' : |
| 133 |
short output |
| 134 |
+-------------------------------------------------------------------- |
| 135 |
|
| 136 |
+-------------------------------------------------------------------- |
| 137 |
Variable: reduce_io |
| 138 |
|
| 139 |
Type: LOGICAL |
| 140 |
Default: .false. |
| 141 |
Description: Reduce I/O to the strict minimum. |
| 142 |
+-------------------------------------------------------------------- |
| 143 |
|
| 144 |
+-------------------------------------------------------------------- |
| 145 |
Variable: max_seconds |
| 146 |
|
| 147 |
Type: REAL |
| 148 |
Default: 1.d7 |
| 149 |
Description: Maximum allowed run time before the job stops smoothly. |
| 150 |
+-------------------------------------------------------------------- |
| 151 |
|
| 152 |
+-------------------------------------------------------------------- |
| 153 |
Variable: fildyn |
| 154 |
|
| 155 |
Type: CHARACTER |
| 156 |
Default: 'matdyn' |
| 157 |
Description: File where the dynamical matrix is written. |
| 158 |
+-------------------------------------------------------------------- |
| 159 |
|
| 160 |
+-------------------------------------------------------------------- |
| 161 |
Variable: fildrho |
| 162 |
|
| 163 |
Type: CHARACTER |
| 164 |
Default: ' ' |
| 165 |
Description: File where the charge density responses are written. Note that the file |
| 166 |
will actually be saved as ${outdir}/_ph0/${prefix}.${fildrho}1 |
| 167 |
where ${outdir}, ${prefix} and ${fildrho} are the values of the |
| 168 |
corresponding input variables |
| 169 |
+-------------------------------------------------------------------- |
| 170 |
|
| 171 |
+-------------------------------------------------------------------- |
| 172 |
Variable: fildvscf |
| 173 |
|
| 174 |
Type: CHARACTER |
| 175 |
Default: ' ' |
| 176 |
Description: File where the the potential variation is written |
| 177 |
(for later use in electron-phonon calculation, see also fildrho). |
| 178 |
+-------------------------------------------------------------------- |
| 179 |
|
| 180 |
+-------------------------------------------------------------------- |
| 181 |
Variable: epsil |
| 182 |
|
| 183 |
Type: LOGICAL |
| 184 |
Default: .false. |
| 185 |
Description: If .true. in a q=0 calculation for a non metal the |
| 186 |
macroscopic dielectric constant of the system is |
| 187 |
computed. Do not set "epsil" to .true. if you have a |
| 188 |
metallic system or q/=0: the code will complain and stop. |
| 189 |
|
| 190 |
Note: the input value of "epsil" will be ignored if "ldisp"=.true. |
| 191 |
(the code will automatically set "epsil" to .false. for metals, |
| 192 |
to .true. for insulators: see routine PHonon/PH/prepare_q.f90). |
| 193 |
+-------------------------------------------------------------------- |
| 194 |
|
| 195 |
+-------------------------------------------------------------------- |
| 196 |
Variable: lrpa |
| 197 |
|
| 198 |
Type: LOGICAL |
| 199 |
Default: .false. |
| 200 |
Description: If .true. the dielectric constant is calculated at the |
| 201 |
RPA level with DV_xc=0. |
| 202 |
+-------------------------------------------------------------------- |
| 203 |
|
| 204 |
+-------------------------------------------------------------------- |
| 205 |
Variable: lnoloc |
| 206 |
|
| 207 |
Type: LOGICAL |
| 208 |
Default: .false. |
| 209 |
Description: If .true. the dielectric constant is calculated without |
| 210 |
local fields, i.e. by setting DV_H=0 and DV_xc=0. |
| 211 |
+-------------------------------------------------------------------- |
| 212 |
|
| 213 |
+-------------------------------------------------------------------- |
| 214 |
Variable: trans |
| 215 |
|
| 216 |
Type: LOGICAL |
| 217 |
Default: .true. |
| 218 |
Description: If .true. the phonons are computed. |
| 219 |
If "trans" .and. "epsil" are .true. effective charges are |
| 220 |
calculated. |
| 221 |
+-------------------------------------------------------------------- |
| 222 |
|
| 223 |
+-------------------------------------------------------------------- |
| 224 |
Variable: lraman |
| 225 |
|
| 226 |
Type: LOGICAL |
| 227 |
Default: .false. |
| 228 |
Description: If .true. calculate non-resonant Raman coefficients |
| 229 |
using second-order response as in: |
| 230 |
M. Lazzeri and F. Mauri, PRL 90, 036401 (2003). |
| 231 |
+-------------------------------------------------------------------- |
| 232 |
|
| 233 |
///--- |
| 234 |
OPTIONAL VARIABLES FOR RAMAN: |
| 235 |
|
| 236 |
+-------------------------------------------------------------------- |
| 237 |
Variable: eth_rps |
| 238 |
|
| 239 |
Type: REAL |
| 240 |
Default: 1.0d-9 |
| 241 |
Description: Threshold for calculation of Pc R |psi>. |
| 242 |
+-------------------------------------------------------------------- |
| 243 |
|
| 244 |
+-------------------------------------------------------------------- |
| 245 |
Variable: eth_ns |
| 246 |
|
| 247 |
Type: REAL |
| 248 |
Default: 1.0e-12 |
| 249 |
Description: Threshold for non-scf wavefunction calculation. |
| 250 |
+-------------------------------------------------------------------- |
| 251 |
|
| 252 |
+-------------------------------------------------------------------- |
| 253 |
Variable: dek |
| 254 |
|
| 255 |
Type: REAL |
| 256 |
Default: 1.0e-3 |
| 257 |
Description: Delta_xk used for wavefunction derivation wrt k. |
| 258 |
+-------------------------------------------------------------------- |
| 259 |
|
| 260 |
\\\--- |
| 261 |
|
| 262 |
+-------------------------------------------------------------------- |
| 263 |
Variable: recover |
| 264 |
|
| 265 |
Type: LOGICAL |
| 266 |
Default: .false. |
| 267 |
Description: If .true. restart from an interrupted run. |
| 268 |
+-------------------------------------------------------------------- |
| 269 |
|
| 270 |
+-------------------------------------------------------------------- |
| 271 |
Variable: low_directory_check |
| 272 |
|
| 273 |
Type: LOGICAL |
| 274 |
Default: .false. |
| 275 |
Description: If .true. search in the phsave directory only the |
| 276 |
quantities requested in input. |
| 277 |
+-------------------------------------------------------------------- |
| 278 |
|
| 279 |
+-------------------------------------------------------------------- |
| 280 |
Variable: only_init |
| 281 |
|
| 282 |
Type: LOGICAL |
| 283 |
Default: .false. |
| 284 |
Description: If .true. only the bands and other initialization quantities are calculated. |
| 285 |
(used for GRID parallelization) |
| 286 |
+-------------------------------------------------------------------- |
| 287 |
|
| 288 |
+-------------------------------------------------------------------- |
| 289 |
Variable: qplot |
| 290 |
|
| 291 |
Type: LOGICAL |
| 292 |
Default: .false. |
| 293 |
Description: If .true. a list of q points is read from input. |
| 294 |
+-------------------------------------------------------------------- |
| 295 |
|
| 296 |
+-------------------------------------------------------------------- |
| 297 |
Variable: q2d |
| 298 |
|
| 299 |
Type: LOGICAL |
| 300 |
Default: .false. |
| 301 |
Description: If .true. three q points and relative weights are |
| 302 |
read from input. The three q points define the rectangle |
| 303 |
q(:,1) + l (q(:,2)-q(:,1)) + m (q(:,3)-q(:,1)) where |
| 304 |
0< l,m < 1. The weights are integer and those of points two |
| 305 |
and three are the number of points in the two directions. |
| 306 |
+-------------------------------------------------------------------- |
| 307 |
|
| 308 |
+-------------------------------------------------------------------- |
| 309 |
Variable: q_in_band_form |
| 310 |
|
| 311 |
Type: LOGICAL |
| 312 |
Default: .false. |
| 313 |
Description: This flag is used only when qplot is .true. and q2d is |
| 314 |
.false.. When .true. each couple of q points q(:,i+1) and |
| 315 |
q(:,i) define the line from q(:,i) to q(:,i+1) and nq |
| 316 |
points are generated along that line. nq is the weigth of |
| 317 |
q(:,i). When .false. only the list of q points given as |
| 318 |
input is calculated. The weights are not used. |
| 319 |
+-------------------------------------------------------------------- |
| 320 |
|
| 321 |
+-------------------------------------------------------------------- |
| 322 |
Variable: electron_phonon |
| 323 |
|
| 324 |
Type: CHARACTER |
| 325 |
Default: ' ' |
| 326 |
Description: |
| 327 |
Options are: |
| 328 |
|
| 329 |
'simple' : |
| 330 |
Electron-phonon lambda coefficients are computed |
| 331 |
for a given q and a grid of k-points specified by |
| 332 |
the variables nk1, nk2, nk3, k1, k2, k3. |
| 333 |
|
| 334 |
'interpolated' : |
| 335 |
Electron-phonon is calculated by interpolation |
| 336 |
over the Brillouin Zone as in M. Wierzbowska, et |
| 337 |
al. arXiv:cond-mat/0504077 |
| 338 |
|
| 339 |
'lambda_tetra' : |
| 340 |
The electron-phonon coefficient \lambda_{q \nu} |
| 341 |
is calculated with the optimized tetrahedron method. |
| 342 |
|
| 343 |
'gamma_tetra' : |
| 344 |
The phonon linewidth \gamma_{q \nu} is calculated |
| 345 |
from the electron-phonon interactions |
| 346 |
using the optimized tetrahedron method. |
| 347 |
|
| 348 |
'epa' : |
| 349 |
Electron-phonon coupling matrix elements are written |
| 350 |
to file prefix.epa.k for further processing by program |
| 351 |
epa.x which implements electron-phonon averaged (EPA) |
| 352 |
approximation as described in G. Samsonidze & B. Kozinsky, |
| 353 |
Adv. Energy Mater. 2018, 1800246 doi:10.1002/aenm.201800246 |
| 354 |
arXiv:1511.08115 |
| 355 |
|
| 356 |
'ahc' : |
| 357 |
Quantities required for the calculation of phonon-induced |
| 358 |
electron self-energy are computed and written to the directory |
| 359 |
"ahc_dir". The output files can be read by postahc.x for |
| 360 |
the calculation of electron self-energy. |
| 361 |
Available for both metals and insulators. |
| 362 |
"trans"=.false. is required. |
| 363 |
|
| 364 |
For metals only, requires gaussian smearing (except for 'ahc'). |
| 365 |
|
| 366 |
If "trans"=.true., the lambdas are calculated in the same |
| 367 |
run, using the same k-point grid for phonons and lambdas. |
| 368 |
If "trans"=.false., the lambdas are calculated using |
| 369 |
previously saved DeltaVscf in "fildvscf", previously saved |
| 370 |
dynamical matrix, and the present punch file. This allows |
| 371 |
the use of a different (larger) k-point grid. |
| 372 |
+-------------------------------------------------------------------- |
| 373 |
|
| 374 |
+-------------------------------------------------------------------- |
| 375 |
Variable: el_ph_nsigma |
| 376 |
|
| 377 |
Type: INTEGER |
| 378 |
Default: 10 |
| 379 |
Description: The number of double-delta smearing values used in an |
| 380 |
electron-phonon coupling calculation. |
| 381 |
+-------------------------------------------------------------------- |
| 382 |
|
| 383 |
+-------------------------------------------------------------------- |
| 384 |
Variable: el_ph_sigma |
| 385 |
|
| 386 |
Type: REAL |
| 387 |
Default: 0.02 |
| 388 |
Description: The spacing between double-delta smearing values used in |
| 389 |
an electron-phonon coupling calculation. |
| 390 |
+-------------------------------------------------------------------- |
| 391 |
|
| 392 |
///--- |
| 393 |
VARIABLES FOR "ELECTRON_PHONON" = 'AHC': |
| 394 |
|
| 395 |
+-------------------------------------------------------------------- |
| 396 |
Variable: ahc_dir |
| 397 |
|
| 398 |
Type: CHARACTER |
| 399 |
Default: outdir // 'ahc_dir/' |
| 400 |
Description: Directory where the output binary files are written. |
| 401 |
+-------------------------------------------------------------------- |
| 402 |
|
| 403 |
+-------------------------------------------------------------------- |
| 404 |
Variable: ahc_nbnd |
| 405 |
|
| 406 |
Type: INTEGER |
| 407 |
Status: REQUIRED |
| 408 |
Description: Number of bands for which the electron self-energy is to be computed. |
| 409 |
+-------------------------------------------------------------------- |
| 410 |
|
| 411 |
+-------------------------------------------------------------------- |
| 412 |
Variable: ahc_nbndskip |
| 413 |
|
| 414 |
Type: INTEGER |
| 415 |
Default: 0 |
| 416 |
Description: Number of bands to exclude when computing the self-energy. Self-energy |
| 417 |
is computed for bands with indices from "ahc_nbndskip"+1 to |
| 418 |
"ahc_nbndskip"+"ahc_nbnd". "ahc_nbndskip"+"ahc_nbnd" cannot |
| 419 |
exceed nbnd of the preceding SCF or NSCF calculation. |
| 420 |
+-------------------------------------------------------------------- |
| 421 |
|
| 422 |
+-------------------------------------------------------------------- |
| 423 |
Variable: skip_upperfan |
| 424 |
|
| 425 |
Type: LOGICAL |
| 426 |
Default: .false. |
| 427 |
Description: If .true., skip calculation of the upper Fan self-energy, which |
| 428 |
involves solving the Sternheimer equation. |
| 429 |
+-------------------------------------------------------------------- |
| 430 |
|
| 431 |
\\\--- |
| 432 |
|
| 433 |
+-------------------------------------------------------------------- |
| 434 |
Variable: lshift_q |
| 435 |
|
| 436 |
Type: LOGICAL |
| 437 |
Default: .false. |
| 438 |
Description: Use a wave-vector grid displaced by half a grid step |
| 439 |
in each direction - meaningful only when ldisp is .true. |
| 440 |
When this option is set, the q2r.x code cannot be used. |
| 441 |
+-------------------------------------------------------------------- |
| 442 |
|
| 443 |
+-------------------------------------------------------------------- |
| 444 |
Variable: zeu |
| 445 |
|
| 446 |
Type: LOGICAL |
| 447 |
Default: zeu="epsil" |
| 448 |
Description: If .true. in a q=0 calculation for a non metal the |
| 449 |
effective charges are computed from the dielectric |
| 450 |
response. This is the default algorithm. If "epsil"=.true. |
| 451 |
and "zeu"=.false. only the dielectric tensor is calculated. |
| 452 |
+-------------------------------------------------------------------- |
| 453 |
|
| 454 |
+-------------------------------------------------------------------- |
| 455 |
Variable: zue |
| 456 |
|
| 457 |
Type: LOGICAL |
| 458 |
Default: .false. |
| 459 |
Description: If .true. in a q=0 calculation for a non metal the |
| 460 |
effective charges are computed from the phonon |
| 461 |
density responses. This is an alternative algorithm, |
| 462 |
different from the default one (if "trans" .and. "epsil" ) |
| 463 |
The results should be the same within numerical noise. |
| 464 |
+-------------------------------------------------------------------- |
| 465 |
|
| 466 |
+-------------------------------------------------------------------- |
| 467 |
Variable: elop |
| 468 |
|
| 469 |
Type: LOGICAL |
| 470 |
Default: .false. |
| 471 |
Description: If .true. calculate electro-optic tensor. |
| 472 |
+-------------------------------------------------------------------- |
| 473 |
|
| 474 |
+-------------------------------------------------------------------- |
| 475 |
Variable: fpol |
| 476 |
|
| 477 |
Type: LOGICAL |
| 478 |
Default: .false. |
| 479 |
Description: If .true. calculate dynamic polarizabilities |
| 480 |
Requires "epsil"=.true. ( experimental stage: |
| 481 |
see example09 for calculation of methane ). |
| 482 |
+-------------------------------------------------------------------- |
| 483 |
|
| 484 |
+-------------------------------------------------------------------- |
| 485 |
Variable: ldisp |
| 486 |
|
| 487 |
Type: LOGICAL |
| 488 |
Default: .false. |
| 489 |
Description: If .true. the run calculates phonons for a grid of |
| 490 |
q-points specified by "nq1", "nq2", "nq3" - for direct |
| 491 |
calculation of the entire phonon dispersion. |
| 492 |
+-------------------------------------------------------------------- |
| 493 |
|
| 494 |
+-------------------------------------------------------------------- |
| 495 |
Variable: nogg |
| 496 |
|
| 497 |
Type: LOGICAL |
| 498 |
Default: .false. |
| 499 |
Description: If .true. disable the "gamma_gamma" trick used to speed |
| 500 |
up calculations at q=0 (phonon wavevector) if the sum over |
| 501 |
the Brillouin Zone includes k=0 only. The gamma_gamma |
| 502 |
trick exploits symmetry and acoustic sum rule to reduce |
| 503 |
the number of linear response calculations to the strict |
| 504 |
minimum, as it is done in code phcg.x. |
| 505 |
+-------------------------------------------------------------------- |
| 506 |
|
| 507 |
+-------------------------------------------------------------------- |
| 508 |
Variable: asr |
| 509 |
|
| 510 |
Type: LOGICAL |
| 511 |
Default: .false. |
| 512 |
Description: Apply Acoustic Sum Rule to dynamical matrix, effective charges |
| 513 |
Works only in conjunction with "gamma_gamma" tricks (see above) |
| 514 |
+-------------------------------------------------------------------- |
| 515 |
|
| 516 |
+-------------------------------------------------------------------- |
| 517 |
Variable: ldiag |
| 518 |
|
| 519 |
Type: LOGICAL |
| 520 |
Default: .false. |
| 521 |
Description: If .true. forces the diagonalization of the dynamical |
| 522 |
matrix also when only a part of the dynamical matrix |
| 523 |
has been calculated. It is used together with "start_irr" |
| 524 |
and "last_irr". If all modes corresponding to a |
| 525 |
given irreducible representation have been calculated, |
| 526 |
the phonon frequencies of that representation are |
| 527 |
correct. The others are zero or wrong. Use with care. |
| 528 |
+-------------------------------------------------------------------- |
| 529 |
|
| 530 |
+-------------------------------------------------------------------- |
| 531 |
Variable: lqdir |
| 532 |
|
| 533 |
Type: LOGICAL |
| 534 |
Default: .false. |
| 535 |
Description: If .true. ph.x creates inside outdir a separate subdirectory |
| 536 |
for each q vector. The flag is set to .true. when "ldisp"=.true. |
| 537 |
and "fildvscf" /= ' ' or when an electron-phonon |
| 538 |
calculation is performed. The induced potential is saved |
| 539 |
separately for each q inside the subdirectories. |
| 540 |
+-------------------------------------------------------------------- |
| 541 |
|
| 542 |
+-------------------------------------------------------------------- |
| 543 |
Variable: search_sym |
| 544 |
|
| 545 |
Type: LOGICAL |
| 546 |
Default: .true. |
| 547 |
Description: Set it to .false. if you want to disable the mode |
| 548 |
symmetry analysis. |
| 549 |
+-------------------------------------------------------------------- |
| 550 |
|
| 551 |
+-------------------------------------------------------------------- |
| 552 |
Variables: nq1, nq2, nq3 |
| 553 |
|
| 554 |
Type: INTEGER |
| 555 |
Default: 0,0,0 |
| 556 |
Description: Parameters of the Monkhorst-Pack grid (no offset) used |
| 557 |
when @ref ldisp=.true. Same meaning as for nk1, nk2, nk3 |
| 558 |
in the input of pw.x. |
| 559 |
+-------------------------------------------------------------------- |
| 560 |
|
| 561 |
+-------------------------------------------------------------------- |
| 562 |
Variables: nk1, nk2, nk3, k1, k2, k3 |
| 563 |
|
| 564 |
Type: INTEGER |
| 565 |
Default: 0,0,0,0,0,0 |
| 566 |
Description: When these parameters are specified the phonon program |
| 567 |
runs a pw non-self consistent calculation with a different |
| 568 |
k-point grid thant that used for the charge density. |
| 569 |
This occurs even in the Gamma case. |
| 570 |
nk1,nk2,nk3 are the parameters of the Monkhorst-Pack grid |
| 571 |
with offset determined by k1,k2,k3. |
| 572 |
+-------------------------------------------------------------------- |
| 573 |
|
| 574 |
+-------------------------------------------------------------------- |
| 575 |
Variable: diagonalization |
| 576 |
|
| 577 |
Type: CHARACTER |
| 578 |
Default: 'david' |
| 579 |
Description: |
| 580 |
Diagonalization method for the non-SCF calculations. |
| 581 |
|
| 582 |
'david' : |
| 583 |
Davidson iterative diagonalization with overlap matrix |
| 584 |
(default). Fast, may in some rare cases fail. |
| 585 |
|
| 586 |
'cg' : |
| 587 |
Conjugate-gradient-like band-by-band diagonalization. |
| 588 |
Slower than 'david' but uses less memory and is |
| 589 |
(a little bit) more robust. |
| 590 |
+-------------------------------------------------------------------- |
| 591 |
|
| 592 |
+-------------------------------------------------------------------- |
| 593 |
Variable: read_dns_bare |
| 594 |
|
| 595 |
Type: LOGICAL |
| 596 |
Default: .false. |
| 597 |
Description: If .true. the PH code tries to read three files in the DFPT+U |
| 598 |
calculation: dns_orth, dns_bare, d2ns_bare. |
| 599 |
dns_orth and dns_bare are the first-order variations of |
| 600 |
the occupation matrix, while d2ns_bare is the second-order |
| 601 |
variation of the occupation matrix. These matrices are |
| 602 |
computed only once during the DFPT+U calculation. However, |
| 603 |
their calculation (especially of d2ns_bare) is computationally |
| 604 |
expensive, this is why they are written to file and then can be |
| 605 |
read (e.g. for restart) in order to save time. |
| 606 |
+-------------------------------------------------------------------- |
| 607 |
|
| 608 |
+-------------------------------------------------------------------- |
| 609 |
Variable: ldvscf_interpolate |
| 610 |
|
| 611 |
Type: LOGICAL |
| 612 |
Default: .false. |
| 613 |
Description: If .true., use Fourier interpolation of phonon potential |
| 614 |
to compute the induced part of phonon potential at each |
| 615 |
q point. Results of a dvscf_q2r.x run is needed. |
| 616 |
Requires "trans" = .false.. |
| 617 |
+-------------------------------------------------------------------- |
| 618 |
|
| 619 |
///--- |
| 620 |
OPTIONAL VARIABLES FOR DVSCF INTERPOLATION: |
| 621 |
|
| 622 |
+-------------------------------------------------------------------- |
| 623 |
Variable: wpot_dir |
| 624 |
|
| 625 |
Type: CHARACTER |
| 626 |
Default: outdir // 'w_pot/' |
| 627 |
Description: Directory where the w_pot binary files are written. |
| 628 |
Must be the same with wpot_dir used in dvscf_q2r.x. |
| 629 |
The real space potential files are stored in wpot_dir |
| 630 |
with names ${prefix}.wpot.irc${irc}//"1". |
| 631 |
+-------------------------------------------------------------------- |
| 632 |
|
| 633 |
+-------------------------------------------------------------------- |
| 634 |
Variable: do_long_range |
| 635 |
|
| 636 |
Type: LOGICAL |
| 637 |
Default: .false. |
| 638 |
Description: If .true., add the long-range part of the potential |
| 639 |
to the Fourier interpolated potential as in: |
| 640 |
S. Ponce et al, J. Chem. Phys. 143, 102813 (2015). |
| 641 |
Reads dielectric matrix and Born effective charges from |
| 642 |
the ${wpot_dir}/tensors.dat file, written in dvscf_q2r.x. |
| 643 |
Currently, only the dipole (Frohlich) part is implemented. |
| 644 |
The quadrupole part is not implemented. |
| 645 |
+-------------------------------------------------------------------- |
| 646 |
|
| 647 |
+-------------------------------------------------------------------- |
| 648 |
Variable: do_charge_neutral |
| 649 |
|
| 650 |
Type: LOGICAL |
| 651 |
Default: .false. |
| 652 |
Description: If .true., impose charge neutrality on the Born effective |
| 653 |
charges. Used only if "do_long_range" = .true.. |
| 654 |
+-------------------------------------------------------------------- |
| 655 |
|
| 656 |
\\\--- |
| 657 |
|
| 658 |
///--- |
| 659 |
SPECIFICATION OF IRREDUCIBLE REPRESENTATION |
| 660 |
|
| 661 |
+-------------------------------------------------------------------- |
| 662 |
Variable: start_irr |
| 663 |
|
| 664 |
Type: INTEGER |
| 665 |
Default: 1 |
| 666 |
See: last_irr |
| 667 |
Description: Perform calculations only from "start_irr" to "last_irr" |
| 668 |
irreducible representations. |
| 669 |
|
| 670 |
IMPORTANT: |
| 671 |
* "start_irr" must be <= 3*nat |
| 672 |
* do not specify "nat_todo" together with |
| 673 |
"start_irr", "last_irr" |
| 674 |
+-------------------------------------------------------------------- |
| 675 |
|
| 676 |
+-------------------------------------------------------------------- |
| 677 |
Variable: last_irr |
| 678 |
|
| 679 |
Type: INTEGER |
| 680 |
Default: 3*nat |
| 681 |
See: start_irr |
| 682 |
Description: Perform calculations only from "start_irr" to "last_irr" |
| 683 |
irreducible representations. |
| 684 |
|
| 685 |
IMPORTANT: |
| 686 |
* "start_irr" must be <= 3*nat |
| 687 |
* do not specify "nat_todo" together with |
| 688 |
"start_irr", "last_irr" |
| 689 |
+-------------------------------------------------------------------- |
| 690 |
|
| 691 |
+-------------------------------------------------------------------- |
| 692 |
Variable: nat_todo |
| 693 |
|
| 694 |
Type: INTEGER |
| 695 |
Default: 0, i.e. displace all atoms |
| 696 |
Description: Choose the subset of atoms to be used in the linear response |
| 697 |
calculation: "nat_todo" atoms, specified in input (see below) |
| 698 |
are displaced. Can be used to estimate modes for a molecule |
| 699 |
adsorbed over a surface without performing a full fledged |
| 700 |
calculation. Use with care, at your own risk, and be aware |
| 701 |
that this is an approximation and may not work. |
| 702 |
IMPORTANT: |
| 703 |
* "nat_todo" <= nat |
| 704 |
* if linear-response is calculated for a given atom, it |
| 705 |
should also be done for all symmetry-equivalent atoms, |
| 706 |
or else you will get incorrect results |
| 707 |
+-------------------------------------------------------------------- |
| 708 |
|
| 709 |
+-------------------------------------------------------------------- |
| 710 |
Variable: modenum |
| 711 |
|
| 712 |
Type: INTEGER |
| 713 |
Default: 0 |
| 714 |
Description: For single-mode phonon calculation : modenum is the index of the |
| 715 |
irreducible representation (irrep) into which the reducible |
| 716 |
representation formed by the 3*nat atomic displacements are |
| 717 |
decomposed in order to perform the phonon calculation. |
| 718 |
Note that a single-mode calculation will not give you the |
| 719 |
frequency of a single phonon mode: in general, the selected |
| 720 |
"modenum" is not an eigenvector. What you get on output is |
| 721 |
a column of the dynamical matrix. |
| 722 |
+-------------------------------------------------------------------- |
| 723 |
|
| 724 |
\\\--- |
| 725 |
|
| 726 |
///--- |
| 727 |
Q-POINT SPECIFICATION |
| 728 |
|
| 729 |
+-------------------------------------------------------------------- |
| 730 |
Variable: start_q |
| 731 |
|
| 732 |
Type: INTEGER |
| 733 |
Default: 1 |
| 734 |
See: last_q |
| 735 |
Description: Used only when ldisp=.true.. |
| 736 |
Computes only the q points from "start_q" to "last_q". |
| 737 |
|
| 738 |
IMPORTANT: |
| 739 |
* "start_q" must be <= "nqs" (number of q points found) |
| 740 |
* do not specify "nat_todo" together with |
| 741 |
"start_q", "last_q" |
| 742 |
+-------------------------------------------------------------------- |
| 743 |
|
| 744 |
+-------------------------------------------------------------------- |
| 745 |
Variable: last_q |
| 746 |
|
| 747 |
Type: INTEGER |
| 748 |
Default: number of q points |
| 749 |
See: start_q |
| 750 |
Description: Used only when "ldisp"=.true.. |
| 751 |
Computes only the q points from "start_q" to "last_q". |
| 752 |
|
| 753 |
IMPORTANT |
| 754 |
* "last_q" must be <= "nqs" (number of q points) |
| 755 |
* do not specify "nat_todo" together with |
| 756 |
"start_q", "last_q" |
| 757 |
+-------------------------------------------------------------------- |
| 758 |
|
| 759 |
+-------------------------------------------------------------------- |
| 760 |
Variable: dvscf_star |
| 761 |
|
| 762 |
Type: STRUCTURE |
| 763 |
Default: disabled |
| 764 |
Description: It contains the following components: |
| 765 |
|
| 766 |
dvscf_star%open (logical, default: .false.) |
| 767 |
dvscf_star%dir (character, default: outdir//"Rotated_DVSCF" or the |
| 768 |
ESPRESSO_FILDVSCF_DIR environment variable) |
| 769 |
dvscf_star%ext (character, default: "dvscf") the extension to use |
| 770 |
for the name of the output files, see below |
| 771 |
dvscf_star%basis (character, default: "cartesian") the basis on which |
| 772 |
the rotated dvscf will be saved |
| 773 |
dvscf_star%pat (logical, default: false) save an optional file with the |
| 774 |
displacement patterns and q vector for each dvscf file |
| 775 |
|
| 776 |
IF dvscf_star%open is .true. use symmetry to compute and store the variation |
| 777 |
of the self-consistent potential on every q* in the star of the present q. |
| 778 |
|
| 779 |
The rotated dvscf will then be stored in directory dvscf_star%dir with name |
| 780 |
prefix.dvscf_star%ext.q_name//"1". Where q_name is derived from the coordinates |
| 781 |
of the q-point, expressed as fractions in crystalline coordinates |
| 782 |
(notice that ph.x reads q-points in cartesian coordinates). |
| 783 |
E.g. q_cryst= (0, 0.5, -0.25) -> q_name = "0_1o2_-1o4" |
| 784 |
|
| 785 |
The dvscf can be represented on a basis of cartesian 1-atom displacements |
| 786 |
(dvscf_star%basis='cartesian') or on the basis of the modes at the rotated q-point |
| 787 |
(dvscf_star%basis='modes'). Notice that the el-ph wannier code requires 'cartesian'. |
| 788 |
Each dvscf file comes with a corresponding pattern file with an additional ".pat" |
| 789 |
suffix; this file contains information about the basis and the q-point of the dvscf. |
| 790 |
|
| 791 |
Note: rotating dvscf can require a large amount of RAM memory and can be i/o |
| 792 |
intensive; in its current implementation all the operations are done |
| 793 |
on a single processor. |
| 794 |
Note2: this feature is currently untested with image parallelisation. |
| 795 |
+-------------------------------------------------------------------- |
| 796 |
|
| 797 |
+-------------------------------------------------------------------- |
| 798 |
Variable: drho_star |
| 799 |
|
| 800 |
Type: STRUCTURE |
| 801 |
See: dvscf_star |
| 802 |
Default: disabled |
| 803 |
Description: It contains the following components: |
| 804 |
|
| 805 |
drho_star%open (logical, default: .false.) |
| 806 |
drho_star%dir (character, default: outdir//"Rotated_DRHO" or the |
| 807 |
ESPRESSO_FILDRHO_DIR environment variable) |
| 808 |
drho_star%ext (character, default: "drho") the extension to use |
| 809 |
for the name of the output files, see below |
| 810 |
drho_star%basis (character, default: "modes") the basis on which |
| 811 |
the rotated drho will be saved |
| 812 |
drho_star%pat (logical, default: true) save an optional file with the |
| 813 |
displacement patterns and q vector for each drho file |
| 814 |
|
| 815 |
Like "dvscf_star", but for the perturbation of the charge density. |
| 816 |
Notice that the defaults are different. |
| 817 |
+-------------------------------------------------------------------- |
| 818 |
|
| 819 |
\\\--- |
| 820 |
|
| 821 |
===END OF NAMELIST====================================================== |
| 822 |
|
| 823 |
|
| 824 |
________________________________________________________________________ |
| 825 |
* IF ldisp != .true. and qplot != .true. : |
| 826 |
|
| 827 |
======================================================================== |
| 828 |
Line of input: |
| 829 |
|
| 830 |
xq(1) xq(2) xq(3) |
| 831 |
|
| 832 |
|
| 833 |
DESCRIPTION OF ITEMS: |
| 834 |
|
| 835 |
+-------------------------------------------------------------------- |
| 836 |
Variables: xq(1) xq(2) xq(3) |
| 837 |
|
| 838 |
Type: REAL |
| 839 |
Description: The phonon wavevector, in units of 2pi/a0 |
| 840 |
(a0 = lattice parameter). |
| 841 |
Not used if "ldisp"=.true. or "qplot"=.true. |
| 842 |
+-------------------------------------------------------------------- |
| 843 |
|
| 844 |
|
| 845 |
===End of line-of-input================================================= |
| 846 |
|
| 847 |
|
| 848 |
|
| 849 |
* ELSE IF qplot == .true. : |
| 850 |
|
| 851 |
SPECIFICATION OF Q POINTS WHEN "QPLOT" == .TRUE. |
| 852 |
|
| 853 |
======================================================================== |
| 854 |
CARD: |
| 855 |
|
| 856 |
///////////////////////////////////////// |
| 857 |
// Syntax: // |
| 858 |
///////////////////////////////////////// |
| 859 |
|
| 860 |
nqs |
| 861 |
xq1(1) xq2(1) xq3(1) nq(1) |
| 862 |
xq1(2) xq2(2) xq3(2) nq(2) |
| 863 |
. . . |
| 864 |
xq1(nqs) xq2(nqs) xq3(nqs) nq(nqs) |
| 865 |
|
| 866 |
///////////////////////////////////////// |
| 867 |
|
| 868 |
DESCRIPTION OF ITEMS: |
| 869 |
|
| 870 |
+-------------------------------------------------------------------- |
| 871 |
Variable: nqs |
| 872 |
|
| 873 |
Type: INTEGER |
| 874 |
Description: Number of q points in the list. Used only if "qplot"=.true. |
| 875 |
+-------------------------------------------------------------------- |
| 876 |
|
| 877 |
+-------------------------------------------------------------------- |
| 878 |
Variables: xq1, xq2, xq3 |
| 879 |
|
| 880 |
Type: REAL |
| 881 |
Description: q-point coordinates; used only with @ref ldisp=.true. and qplot=.true. |
| 882 |
The phonon wavevector, in units of 2pi/a0 (a0 = lattice parameter). |
| 883 |
The meaning of these q points and their weights nq depend on the |
| 884 |
flags q2d and q_in_band_form. (NB: nq is integer) |
| 885 |
+-------------------------------------------------------------------- |
| 886 |
|
| 887 |
+-------------------------------------------------------------------- |
| 888 |
Variable: nq |
| 889 |
|
| 890 |
Type: INTEGER |
| 891 |
Description: The weight of the q-point; the meaning of nq depends |
| 892 |
on the flags q2d and q_in_band_form. |
| 893 |
+-------------------------------------------------------------------- |
| 894 |
|
| 895 |
===END OF CARD========================================================== |
| 896 |
|
| 897 |
|
| 898 |
|
| 899 |
ENDIF |
| 900 |
________________________________________________________________________ |
| 901 |
|
| 902 |
________________________________________________________________________ |
| 903 |
* IF nat_todo was specified : |
| 904 |
|
| 905 |
======================================================================== |
| 906 |
Line of input: |
| 907 |
|
| 908 |
atom(1) atom(2) ... atom(nat_todo) |
| 909 |
|
| 910 |
|
| 911 |
DESCRIPTION OF ITEMS: |
| 912 |
|
| 913 |
+-------------------------------------------------------------------- |
| 914 |
Variables: atom(1) atom(2) ... atom(nat_todo) |
| 915 |
|
| 916 |
Type: INTEGER |
| 917 |
Description: Contains the list of indices of atoms used in the |
| 918 |
calculation if "nat_todo" is specified. |
| 919 |
+-------------------------------------------------------------------- |
| 920 |
|
| 921 |
|
| 922 |
===End of line-of-input================================================= |
| 923 |
|
| 924 |
|
| 925 |
|
| 926 |
ENDIF |
| 927 |
________________________________________________________________________ |
| 928 |
|
| 929 |
|
| 930 |
:::: ADDITIONAL INFORMATION |
| 931 |
|
| 932 |
NB: The program ph.x writes on the tmp_dir/_ph0/{prefix}.phsave directory |
| 933 |
a file for each representation of each q point. This file is called |
| 934 |
dynmat.#iq.#irr.xml where #iq is the number of the q point and #irr |
| 935 |
is the number of the representation. These files contain the |
| 936 |
contribution to the dynamical matrix of the irr representation for the |
| 937 |
iq point. |
| 938 |
|
| 939 |
If "recover"=.true. ph.x does not recalculate the |
| 940 |
representations already saved in the tmp_dir/_ph0/{prefix}.phsave |
| 941 |
directory. Moreover ph.x writes on the files patterns.#iq.xml in the |
| 942 |
tmp_dir/_ph0/{prefix}.phsave directory the displacement patterns that it |
| 943 |
is using. If "recover"=.true. ph.x does not recalculate the |
| 944 |
displacement patterns found in the tmp_dir/_ph0/{prefix}.phsave directory. |
| 945 |
|
| 946 |
This mechanism allows: |
| 947 |
|
| 948 |
1) To recover part of the ph.x calculation even if the recover file |
| 949 |
or files are corrupted. You just remove the _ph0/{prefix}.recover |
| 950 |
files from the tmp_dir directory. You can also remove all the _ph0 |
| 951 |
files and keep only the _ph0/{prefix}.phsave directory. |
| 952 |
|
| 953 |
2) To split a phonon calculation into several jobs for different |
| 954 |
machines (or set of nodes). Each machine calculates a subset of |
| 955 |
the representations and saves its dynmat.#iq.#irr.xml files on |
| 956 |
its tmp_dir/_ph0/{prefix}.phsave directory. Then you collect all the |
| 957 |
dynmat.#iq.#irr.xml files in one directory and run ph.x to |
| 958 |
collect all the dynamical matrices and diagonalize them. |
| 959 |
|
| 960 |
NB: To split the q points in different machines, use the input |
| 961 |
variables start_q and last_q. To split the irreducible |
| 962 |
representations, use the input variables "start_irr", "last_irr". Please |
| 963 |
note that different machines will use, in general, different |
| 964 |
displacement patterns and it is not possible to recollect partial |
| 965 |
dynamical matrices generated with different displacement patterns. A |
| 966 |
calculation split into different machines will run as follows: A |
| 967 |
preparatory run of ph.x with "start_irr"=0, "last_irr"=0 produces the sets |
| 968 |
of displacement patterns and save them on the patterns.#iq.xml files. |
| 969 |
These files are copied in all the tmp_dir/_ph0/{prefix}.phsave directories |
| 970 |
of the machines where you plan to run ph.x. ph.x is run in different |
| 971 |
machines with complementary sets of start_q, last_q, "start_irr" and |
| 972 |
"last_irr" variables. All the files dynmat.#iq.#irr.xml are |
| 973 |
collected on a single tmp_dir/_ph0/{prefix}.phsave directory (remember to |
| 974 |
collect also dynmat.#iq.0.xml). A final run of ph.x in this |
| 975 |
machine collects all the data contained in the files and diagonalizes |
| 976 |
the dynamical matrices. This is done requesting a complete dispersion |
| 977 |
calculation without using start_q, last_q, "start_irr", or "last_irr". |
| 978 |
See an example in examples/GRID_example. |
| 979 |
|
| 980 |
On parallel machines the q point and the irreps calculations can be split |
| 981 |
automatically using the -nimage flag. See the phonon user guide for further |
| 982 |
information. |
| 983 |
|
| 984 |
|
| 985 |
This file has been created by helpdoc utility on Wed Dec 02 08:04:43 CET 2020 |
| 986 |
|
| 987 |
</pre> |
| 988 |
</body> |
| 989 |
</html> |