[QE-users] [SUSPICIOUS MESSAGE] strange behavior of assume_isolated = 'esm' in QE 6.4.1
Giuseppe Mattioli
giuseppe.mattioli at ism.cnr.it
Tue Nov 26 14:22:04 CET 2019
Dear Users and Developers
I report a strange behavior of my calculations using the effective
screening medium (ESM) in a vacuum-slab-vacuum framework. The very
same input file (I can provide all details to whom interested in
investigating the problem)
&control
calculation = 'relax'
restart_mode='from_scratch',
/
&system
ibrav= 8, celldm(1)=16.376100, celldm(2)=0.577350, celldm(3)=6.633333
nat=66, ntyp=4,
ecutwfc=40.0,
ecutrho=320.0,
occupations='smearing', smearing='gaussian', degauss=0.01,
nspin=1,
input_dft='vdw-df-c09'
assume_isolated = 'esm',
esm_bc='bc1',
/
&electrons
mixing_mode='plain'
mixing_beta=0.1
conv_thr=1.0d-8
electron_maxstep=200
/
&ions
ion_dynamics='bfgs'
/
ATOMIC_SPECIES
Au 107.868 Au-pbe-n-van.UPF
S 32.064 S.pbe-n-rrkjus_psl.1.0.0.UPF
C 12.011 C_pbe.van.UPF
H 1.008 H_pbe.van.UPF
ATOMIC_POSITIONS {angstrom}
...
K_POINTS {automatic}
2, 4, 1, 0, 0, 0
produced very different results in versions 5.2.0 (and also 6.1) with
respect to 6.4.1. In detail, even if energies and forces are
practically identical, a strange earthquake affects the prefix.esm1
file which contains the most useful information about the asymptotic
behavior of the potential. This is the 5.2.0 version of the file
#z (A) Tot chg (e/A) Avg v_hartree (eV) Avg v_local (eV) Avg
v_hart+v_loc (eV)
-28.70 0.0000 -27662.1284310 27662.4560939 0.3276628
-28.60 0.0000 -26919.5682331 26920.1817195 0.6134864
-28.51 0.0000 -26361.3370690 26362.1527348 0.8156658
-28.42 0.0000 -26091.0720006 26091.9643464 0.8923458
-28.33 0.0000 -25999.3298072 26000.2221594 0.8923522
-28.24 0.0000 -25907.5916317 25908.4839785 0.8923468
-28.14 0.0000 -25815.8486654 25816.7410174 0.8923520
-28.05 0.0000 -25724.1112606 25725.0036079 0.8923472
-27.96 0.0000 -25632.3675265 25633.2598782 0.8923518
-27.87 0.0000 -25540.6308863 25541.5232340 0.8923477
-27.78 0.0000 -25448.8863911 25449.7787427 0.8923515
-27.68 0.0000 -25357.1505079 25358.0428561 0.8923482
-27.59 0.0000 -25265.4052602 25266.2976115 0.8923513
-27.50 0.0000 -25173.6701248 25174.5624734 0.8923486
-27.41 0.0000 -25081.9241344 25082.8164854 0.8923510
-27.32 0.0000 -24990.1897362 24991.0820853 0.8923491
-27.22 0.0000 -24898.4430146 24899.3353653 0.8923507
-27.13 0.0000 -24806.7093413 24807.6016908 0.8923495
-27.04 0.0000 -24714.9619013 24715.8542518 0.8923505
-26.95 0.0000 -24623.2289394 24624.1212894 0.8923500
-26.86 0.0000 -24531.4807954 24532.3731456 0.8923502
-26.76 0.0000 -24439.7485298 24440.6408802 0.8923504
-26.67 0.0000 -24347.9996976 24348.8920475 0.8923499
-26.58 0.0000 -24256.2681118 24257.1604627 0.8923509
-26.49 0.0000 -24164.5186085 24165.4109582 0.8923496
-26.40 0.0000 -24072.7876848 24073.6800361 0.8923513
-26.30 0.0000 -23981.0375289 23981.9298782 0.8923494
This is the 6.4.1 version, containing very different, puzzling values.
#z (A) Tot chg (e/A) Avg v_hartree (eV) Avg v_local (eV) Avg
v_hart+v_loc (eV)
-28.70 -0.0000057 -94168.0655664 26367.1846760
-67800.8808904
-28.60 0.0000056 -93866.2448433 26275.4442915
-67590.8005518
-28.51 -0.0000053 -93564.4241206 26183.7039198
-67380.7202007
-28.42 0.0000052 -93262.6033992 26091.9635363
-67170.6398629
-28.33 -0.0000049 -92960.7827325 26000.2231637
-66960.5595689
-28.24 0.0000048 -92658.9622545 25908.4827811
-66750.4794734
-28.14 -0.0000045 -92357.1418416 25816.7424075
-66540.3994341
-28.05 0.0000043 -92055.3211652 25725.0020259
-66330.3191392
-27.96 -0.0000040 -91753.5000324 25633.2616513
-66120.2383811
-27.87 0.0000039 -91451.6786141 25541.5212708
-65910.1573433
-27.78 -0.0000036 -91149.8573231 25449.7808951
-65700.0764280
-27.68 0.0000034 -90848.0364491 25358.0405156
-65489.9959335
-27.59 -0.0000031 -90546.2159214 25266.3001389
-65279.9157824
-27.50 0.0000029 -90244.3954296 25174.5597604
-65069.8356692
-27.41 -0.0000026 -89942.5747620 25082.8193827
-64859.7553792
-27.32 0.0000024 -89640.7539973 24991.0790053
-64649.6749920
-27.22 -0.0000021 -89338.9333601 24899.3386265
-64439.5947336
-27.13 0.0000020 -89037.1129077 24807.5982501
-64229.5146576
-27.04 -0.0000016 -88735.2924009 24715.8578703
-64019.4345305
-26.95 0.0000015 -88433.4715111 24624.1174949
-63809.3540162
-26.86 -0.0000012 -88131.6501622 24532.3771142
-63599.2730480
-26.76 0.0000010 -87829.8286568 24440.6367398
-63389.1919171
-26.67 -0.0000007 -87528.0074506 24348.8963580
-63179.1110926
-26.58 0.0000005 -87226.1867804 24257.1559846
-62969.0307958
-26.49 -0.0000002 -86924.3664924 24165.4156018
-62758.9508906
-26.40 0.0000000 -86622.5462160 24073.6752294
-62548.8709867
-26.30 0.0000003 -86320.7257023 23981.9348457
-62338.7908566
I might try to calculate the same values with average.x, but even if
they were correct, the automatic inclusion of correct values into the
prefix.esm1 file would be far better.
Thanks in advance for suggestions
Giuseppe
GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: <giuseppe.mattioli at ism.cnr.it>
More information about the users
mailing list