[CP2K:3024] A strange problem about hybrid-DFT combined with all-electron calculation
hut... at pci.uzh.ch
hut... at pci.uzh.ch
Mon Dec 27 14:24:57 UTC 2010
Hi
most likely (I haven't tried) the problem is the box size
together with your choice of Poisson solver. The solver "ANALYTIC"
works with a sphere and your case the maximum diameter of the sphere
is given by the minimum extension of the box.
Try the either the solver "MT" with box size 20 x 80 x 20
or the solver "WAVELET" with box size 50 x 50 x 50.
best regards
Juerg
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Physical Chemistry Institute FAX : ++41 44 635 6838
University of Zurich E-mail: hut... at pci.uzh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
---------------------------------------------------------------
-----cp... at googlegroups.com wrote: -----
To: cp2k <cp... at googlegroups.com>
From: Igor Ying Zhang <wenx... at gmail.com>
Sent by: cp... at googlegroups.com
Date: 12/27/2010 04:15AM
Subject: [CP2K:3024] A strange problem about hybrid-DFT combined with all-electron calculation
Dear all:
I am a beginner here. I started learning CP2K by investigating 6
cases in gas without periodic bondary condition which I had done in
the Gaussian03 package (G03). These cases contains CH4, C2H6, C6H14,
C8H18, C32H66, and C60. All of these calculations were performed at
the theoretical level of B3LYP/6-311G** without pesudopotential. I am
very happy to see that CP2K leads to very close results to G03 for
CH4, C2H6, C8H18, and C60. The deviations between each other are less
than 1.0E-5 (CH4), 1.0E-4 (C2H6), 1.0E-4 (C6H14), 1.0E-4 (C8H18) and
1.0E-3 a.u. (C60). However, the deviation for C32H66 is more than 5.0
a.u., where the input statement is the same as the other 5 examples.
For C32H66, the computed total energy from G03 is -1259.58294069
a.u., which is about -1264.57308160 a.u. with the cell size of
18x55x18, and -1264.65816635 a.u. with 25x55x25, and -1264.65985171
a.u. with 25x60x25. The deviations are very huge. And another puzzling
thing is that when I choose the cell size of 22x60x22, the total
energy obtained from CP2K is about -1260.48059560 a.u. which has the
deviation of more than 4 a.u. against results of the other three cell
size.
The input file of C32H66 is:
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
&FORCE_EVAL
METHOD Quickstep
&DFT
BASIS_SET_FILE_NAME /share/apps/cp2k/tests/QS/EMSL_BASIS_SETS
POTENTIAL_FILE_NAME /share/apps/cp2k/tests/QS/POTENTIAL
&MGRID
CUTOFF 300
REL_CUTOFF 50
&END MGRID
&QS
METHOD GAPW
EPS_FILTER_MATRIX 1.0E-32
EPS_PGF_ORB 1.0E-32
&END QS
&POISSON
PERIODIC NONE
PSOLVER ANALYTIC
&END
&SCF
EPS_SCF 1.0E-5
SCF_GUESS RESTART
MAX_SCF 30
&OT T
PRECONDITIONER FULL_ALL
ENERGY_GAP 0.001
LINESEARCH 2PNT
MAX_TAYLOR 5
MINIMIZER DIIS
STEPSIZE 0.2
&OUTER_SCF
EPS_SCF 1.0E-5
MAX_SCF 20
&END OUTER_SCF
&END SCF
&XC
&XC_FUNCTIONAL
&LYP
SCALE_C 0.81
&END
&BECKE88
SCALE_X 0.72
&END
&VWN
FUNCTIONAL_TYPE VWN3 # VWN3 is employed as in
Gaussian package
SCALE_C 0.19
&END
&XALPHA
SCALE_X 0.08
&END
&END XC_FUNCTIONAL
&HF
FRACTION 0.20
&MEMORY
MAX_MEMORY 800 # The memory
available to one MPI process.
EPS_STORAGE_SCALING 1.0E-1
&END MEMORY
&SCREENING
EPS_SCHWARZ 1.0E-8
#SCREEN_ON_INITIAL_P FALSE
&END SCREENING
#&INTERACTION_POTENTIAL
# POTENTIAL_TYPE TRUNCATED
# CUTOFF_RADIUS 15.000
# T_C_G_DATA /share/apps/cp2k/tests/QS/t_c_g.dat
#&END INTERACTION_POTENTIAL
&END HF
&END XC
&END DFT
&SUBSYS
&CELL
ABC 22.0 60.0 22.0
PERIODIC NONE
&END CELL
&COORD
C 1.14606000 41.16159800 1.38290300
C 1.99581200 39.88741400 1.38290300
C 1.16086200 38.60168100 1.38290300
C 2.00192600 37.32025900 1.38290300
C 1.16543600 36.03559700 1.38290300
C 2.00613200 34.75370300 1.38290300
C 1.16894800 33.46950900 1.38290300
C 2.00897600 32.18717100 1.38290300
C 1.17120900 30.90335100 1.38290300
C 2.01068900 29.62065000 1.38290300
C 1.17244700 28.33713700 1.38290300
C 2.01150800 27.05416000 1.38290300
C 1.17291300 25.77087700 1.38290300
C 2.01168800 24.48771200 1.38290300
C 1.17287900 23.20456800 1.38290300
C 2.01150800 21.92130700 1.38290300
C 1.17262600 20.63820900 1.38290300
C 2.01125500 19.35494800 1.38290300
C 1.17244600 18.07180400 1.38290300
C 2.01122100 16.78863900 1.38290300
C 1.17262600 15.50535600 1.38290300
C 2.01168700 14.22237900 1.38290300
C 1.17344500 12.93886600 1.38290300
C 2.01292500 11.65616500 1.38290300
C 1.17515800 10.37234500 1.38290300
C 2.01518600 9.09000700 1.38290300
C 1.17800200 7.80581300 1.38290300
C 2.01869800 6.52391900 1.38290300
C 1.18220800 5.23925700 1.38290300
C 2.02327200 3.95783500 1.38290300
C 1.18832200 2.67210200 1.38290300
C 2.03807400 1.39791800 1.38290300
H 1.77043400 42.05951600 1.38290300
H 0.50042500 41.20620200 2.26580600
H 0.50042500 41.20620200 0.50000000
H 2.65562800 39.88856000 2.25915700
H 2.65562800 39.88856000 0.50664900
H 0.50000000 38.60136000 0.50611800
H 0.50000000 38.60136000 2.25968800
H 2.66253300 37.32153700 2.25970500
H 2.66253300 37.32153700 0.50610100
H 0.50479700 36.03454000 0.50615900
H 0.50479700 36.03454000 2.25964700
H 2.66674500 34.75460500 2.25967700
H 2.66674500 34.75460500 0.50612900
H 0.50830200 33.46878500 0.50615100
H 0.50830200 33.46878500 2.25965500
H 2.66959500 32.18775000 2.25967600
H 2.66959500 32.18775000 0.50613000
H 0.51056300 30.90291600 0.50614800
H 0.51056300 30.90291600 2.25965800
H 2.67131300 29.62096400 2.25967400
H 2.67131300 29.62096400 0.50613200
H 0.51180300 28.33692900 0.50614600
H 0.51180300 28.33692900 2.25966000
H 2.67213500 27.05427600 2.25967200
H 2.67213500 27.05427600 0.50613400
H 0.51227200 25.77083400 0.50614400
H 0.51227200 25.77083400 2.25966200
H 2.67231900 24.48770200 2.25966900
H 2.67231900 24.48770200 0.50613700
H 0.51224100 23.20461800 0.50614200
H 0.51224100 23.20461800 2.25966400
H 2.67214200 21.92124000 2.25966700
H 2.67214200 21.92124000 0.50613900
H 0.51199200 20.63827600 0.50613900
H 0.51199200 20.63827600 2.25966700
H 2.67189300 19.35489800 2.25966400
H 2.67189300 19.35489800 0.50614200
H 0.51181500 18.07181400 0.50613700
H 0.51181500 18.07181400 2.25966900
H 2.67186200 16.78868200 2.25966200
H 2.67186200 16.78868200 0.50614400
H 0.51199900 15.50524000 0.50613400
H 0.51199900 15.50524000 2.25967200
H 2.67233100 14.22258700 2.25966000
H 2.67233100 14.22258700 0.50614600
H 0.51282100 12.93855200 0.50613200
H 0.51282100 12.93855200 2.25967400
H 2.67357100 11.65660000 2.25965800
H 2.67357100 11.65660000 0.50614800
H 0.51453900 10.37176600 0.50613000
H 0.51453900 10.37176600 2.25967600
H 2.67583200 9.09073100 2.25965500
H 2.67583200 9.09073100 0.50615100
H 0.51738900 7.80491100 0.50612900
H 0.51738900 7.80491100 2.25967700
H 2.67933700 6.52497600 2.25964700
H 2.67933700 6.52497600 0.50615900
H 0.52160100 5.23797900 0.50610100
H 0.52160100 5.23797900 2.25970500
H 2.68413400 3.95815600 2.25968800
H 2.68413400 3.95815600 0.50611800
H 0.52850600 2.67095600 0.50664900
H 0.52850600 2.67095600 2.25915700
H 1.41370000 0.50000000 1.38290300
H 2.68370900 1.35331400 2.26580600
H 2.68370900 1.35331400 0.50000000
&END COORD
&KIND C
BASIS_SET 6-311G**
POTENTIAL ALL
&END KIND
&KIND H
BASIS_SET 6-311G**
POTENTIAL ALL
&END KIND
&END SUBSYS
&END FORCE_EVAL
&GLOBAL
PROJECT C32H66
RUN_TYPE ENERGY
PRINT_LEVEL MEDIUM
&END GLOBAL
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
And in the output file, I got the warning message as follow:
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
*** 00:05:54 WARNING in
hfx_energy_potential:integrate_four_center ***
*** err=-300 The Kohn Sham matrix is not 100% occupied. This may
result ***
*** in uncorrect Hartree-Fock results. Try to decrease EPS_PGF_ORB
and ***
*** EPS_FILTER_MATRIX in the QS
section. ***
@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
I have decreased EPS_PGF_ORB and EPS_FILTER_MATREX, and also employed
truncated coulomb
interaction as suggested from the old post (http://groups.google.com/
group/cp2k/browse_thread/thread/1959ddee999d3a8e/6fae4716f12cd538?
lnk=raot&fwc=1)
Any suggestions are welcome. Best wishes!
Igor Ying Zhang
--
You received this message because you are subscribed to the Google Groups "cp2k" group.
To post to this group, send email to cp... at googlegroups.com.
To unsubscribe from this group, send email to cp2k+uns... at googlegroups.com.
For more options, visit this group at http://groups.google.com/group/cp2k?hl=en.
More information about the CP2K-user
mailing list