[CPMD-list] Re: Fw: Gradient correction functional

[Axel Kohlmeyer]

>>> "SL" == Sergey Lisenkov <proffess at yandex.ru> writes:


sergey,

it should work by just using. 

&DFT
FUNCTIONAL REVPBE
&END

i'd also recommend using a GC-CUTOFF of 1.0e-6 to 
improve numerical stability, energy conservation 
and convergence, if you have areas of low electronic 
density in your system (e.g. an isolated molecule).

other than that you should get the same effect by
changing the code in dftin.F around line 135

from:

          IF(INDEX(LINE,'GGAX').NE.0) MGCX=2
          IF(INDEX(LINE,'PBEX').NE.0) MGCX=3
          IF(INDEX(LINE,'REVPBEX').NE.0) MGCX=4

to:

          IF(INDEX(LINE,'GGAX').NE.0) MGCX=2
          IF(INDEX(LINE,'REVPBEX').NE.0) MGCX=4
          IF(INDEX(LINE,'PBEX').NE.0) MGCX=3

best regards,
     axel.

SL> Dear CPMD authors and users,

SL> I put in my input file in section DFT the following:

SL> &DFT
SL> FUNCTIONAL REVPBE
SL> GRADIENT CORRECTION  REVPBEX  REVPBEC
SL>  LDA CORRELATION PZ
SL> SLATER 
SL>  0.666666666667
SL> NEWCODE
SL> &END

>> From pseudopotentials from Boero library I got the information:
SL> &INFO  
SL>     ============================================================
SL>     |    Pseudopotential Report                                |
SL>     ------------------------------------------------------------
SL>     |  Atomic Symbol                   :   H                   |
SL>     |  Atomic Number                   :   1                   |
SL>     |  Number of core states           :   0                   |
SL>     |  Number of valence states        :   1                   |
SL>     |  Exchange-Correlation Functional :                       |
SL>     |     Slater exchange :  0.6667                            |
SL>     |     LDA correlation : Ceperley-Alder                     |
SL>     |     Exchange GC     : revPBE (Z-Y, PRL 1998)             |
SL>     |     Correlation GC  : Perdew-Burke-Ernzerhof             |
SL>     |  Electron Configuration :   N   L  Occupation            |
SL>     |                             1   S    1.0000              |
SL>     |  Full Potential Total Energy    -0.464927                |
SL>     |  Trouiller-Martins normconserving PP                     |
SL>     |     n    l        rc       energy                        |
SL>     |     1    S    0.4950     -0.24065                        |
SL>     |     2    P    0.4032     -0.24065                        |
SL>     |  Number of Mesh Points :   511                           |
SL>     |  Pseudoatom Total Energy   -0.464911                     |
SL>     ============================================================

SL> But in output file I see:
SL> ...
SL>  EXCHANGE CORRELATION FUNCTIONALS 
SL>     LDA EXCHANGE:                        SLATER (ALPHA = 0.66667)
SL>     LDA CORRELATION:                              PERDEW & ZUNGER
SL>        [J.P. PERDEW AND A ZUNGER, PRB 23 5048 (1981)]
SL>     GRADIENT CORRECTED FUNCTIONAL
SL>     DENSITY THRESHOLD:                                1.00000E-08
SL>     EXCHANGE ENERGY
SL>        [A.D. BECKE, PHYS. REV. A 38, 3098 (1988)]
SL>        PARAMETER BETA:                                   0.004200
SL>     CORRELATION ENERGY
SL>        [J.P. PERDEW, PHYS. REV. B 33, 8822 (1986)]
 
SL>  ***     DETSP| THE NEW SIZE OF THE PROGRAM IS    1352/  20992 kBYTES ***

SL>  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
SL>  WARNING! XC FUNCTIONALS INCONSISTENT FOR C_MT_revPBE.psp
SL>  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

SL>  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
SL>  WARNING! XC FUNCTIONALS INCONSISTENT FOR H_MT_revPBE.psp
SL>  !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

SL> ....
SL> You can see that Exchange and correlation are not revPBE, as I put.

SL> What is incorrect in my input file?

SL>  Thank you very much for your help.
SL> Best wishes,
SL>   Sergey



--

=======================================================================
Axel Kohlmeyer       e-mail: axel.kohlmeyer at theochem.ruhr-uni-bochum.de
Lehrstuhl fuer Theoretische Chemie          Phone: ++49 (0)234/32-26673
Ruhr-Universitaet Bochum - NC 03/53         Fax:   ++49 (0)234/32-14045
D-44780 Bochum  http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.