[CPMD-list] Help for uspp and model!

[Axel Kohlmeyer]

On Tue, 10 May 2005, [gb2312] shyma wrote:

dear shyma,


S> Dear CPMD user:               
S>      I have some questions need your help:
S>      Firstly, I skimed through some papers and find the uspp file is used widely,but the uspp files I find was not compatible with CPMD, I couldn't run the cpmd.x program with K-points successfully when I use the Vanderbilt pesudopotentials that are generated with vesion 7.3.5 of vanderbil
S> ts atomic code, the out-file showed that 
S> "
S>  PROGRAM STOPS IN SUBROUTINE SETSYS| K-POINTS WITH VANDERBILT NOT IMPLEMENTED [PR
S> OC=   0]
S> "

S> could anyone help me on pesudopontential files which are consistent
S> with GGA(does GGA can get better valuse than LDA, many people said so)
S> and CPMD. So could you e-mail to me if you have these pesudopential?

there is a BP titanium potential in the Goedecker library. please note,
that this is a semi-core potential and will require a _very_ high plane
wave cutoff (most likely 200ry or more), but for a compound like TiN
this is probably needed, anyway. this is _why_ people mostly use uspps
for that kind of system...

S>      Secondly, I can't view the structure directly so I'm not sure the
S> model in the cpmd input-file is what I want(the same model of 5 slabs
S> TiN + 10 angstroms vacuum in my mind was used in CASTEP and CPMD, but

the simples trick to see all atoms and the box, would be to start a
wavefunction optimization with maxstep set to 1, a small plane wave 
cutoff and using the keyword RHOOUT to create a DENSITY file.
run this file through cpmd2cube.x (without -trim !) and view the 
resulting cube, e.g. in VMD. for examples see:
http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/cpmd-vmd/part5.html#chap7_sect1
or
http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/cpmd-tutor/part5.html#chap6_sect1

S> the total energy of latter is just 1/3 of the former, so I think there
S> some wrong with my cpmd-input-files because I could view the intial
S> model in the castep). I mean that need I use other keyword in the

that may also depend on the pseudopotential. if you were using a semicore
potential (e.g. ZV=12) in CASTEP but a 'minimal' one (e.g. ZV=4) in cpmd
the total energy should be very different.

S> cpmd-input-file and how to constrcut it if I want to construct the
S> model of 2*2 or 3*3 atom on surface in the periodic boundary condition?
S> The following input-file is based on your advices in your last reply,
S> please check and correct it if it doesn't meet the 2*2*5 slabs:

coming back to the VMD visualization. you can also activate periodic
images there and thus can see the effect of the PBC and whether
your input looks correct. note, that you need at least the 1.8.3 release
to get proper representation of the periodic images of non-orthogonal
unitcells.

best regards,
	axel.

S> 
S> &INFO
S>  (001) slab of TiN, 5 layers + 8 layers of vacuum
S> &END
S> 
S> &CPMD
S>  OPTIMIZE GEOMETRY
S>  UNITE HESSIAN
S>  BFGS
S>  FREE ENERGY FUNCTIONAL
S>  LANCZOS DIAGONALISATION
S>  LANCZOS PARAMETERS
S>  1  6  20  1.0D-18        !( I don't understand it clearly and imitate from other slab exemple)
S>  TROTTER FACTOR
S>  0.001
S>  ORTHOGONALISATION LOWDIN
S>  CONVERGENCE ORBITAL
S>  1.0D-5
S>  CONVERGENCE GEOMETRY
S>  5.0D-4
S>  MAXSTEP
S>  1000
S>  MEMORY BIG
S>  ELECTRON TEMPERATURE
S>  1000
S>  COMPRESS WRITE32
S>  STRUCTURE BONDS
S> &END
S> 
S> &SYSTEM
S>  POINT GROUP
S>  AUTO
S>  SYMMETRY
S>  8
S>  ANGSTROM
S>  SCALE
S>  CELL
S>  3.182  1.0  9.2  0.0  0.0  0.0
S>  CUTOFF
S>  50.OO
S>  STATES
S>  30
S>  TESR  !( I don't understand it clearly and imitate from other slab exemple)
S>  3
S>  KPOINTS MONKHORST-PACK FULL
S>  8  8  1
S> &END
S> 
S> &DFT
S>  FUNCTIONAL LDA
S> &END
S> 
S> &ATOMS
S> *Ti_MT_LDA_NLCC.psp KLEINMAN-BYLANDER
S>  LMAX=D
S>  5
S>  0    0    0
S>  1/2  1/2  1/13
S>  0    0    2/13
S>  1/2  1/2  3/13
S>  0    0    4/13
S> 
S> *N_GO_LDA.psp KLEINMAN-BYLANDER
S>  LMAX=P
S>  5
S>  1/2  1/2  0
S>  0    0    1/13
S>  1/2  1/2  2/13
S>  0    0    3/13
S>  1/2  1/2  4/13
S> &END
S> 
S> Best wishes!
S> Thanks in advance!
S>                       yours  shyma                                             

-- 

=======================================================================
Dr. 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.