[CPMD-list] Monoclinic - possible solution to a problem

[Przemyslaw Dopieralski]

Hello,

After a busy night I find out that:

1. probably the geometry and everything was right
   the inversion scheme was correct.

a 10.9   alpha 90         inv            a 5.5     alpha  102
b 5.5    beta 102        ----->          b 14.6    beta    90
c 14.6   gamma 90                        c 10.9    gamma   90

and then I switched x y z ----> y z x


2. I started my calculation with this parameters

a 5.5     alpha  102
b 14.6    beta    90
c 10.9    gamma   90

switched x y z ----> y z x

what cpmd show me:

************************** SUPERCELL ***************************
 SYMMETRY:                                             MONOCLINIC
 POINT GROUP :                           [MONOCLINIC]     <1>(ci)
 LATTICE CONSTANT(a.u.):                                 10.42751
 CELL DIMENSION:  10.4275  2.6488  1.9672 -0.2227  0.0000  0.0000
 VOLUME(OMEGA IN BOHR^3):                              5759.52695
 LATTICE VECTOR A1(BOHR):           10.4275     0.0000     0.0000 <----
 LATTICE VECTOR A2(BOHR):           -6.1521    26.9264     0.0000 <----
 LATTICE VECTOR A3(BOHR):            0.0000     0.0000    20.5130 <----
 RECIP. LAT. VEC. B1(2Pi/BOHR):      0.0959     0.0219     0.0000
 RECIP. LAT. VEC. B2(2Pi/BOHR):      0.0000     0.0371     0.0000
 RECIP. LAT. VEC. B3(2Pi/BOHR):      0.0000     0.0000     0.0487
 REAL SPACE MESH:                    72          180          144
 WAVEFUNCTION CUTOFF(RYDBERG):                          100.00000
 DENSITY CUTOFF(RYDBERG):          (DUAL= 4.00)         400.00000
 NUMBER OF PLANE WAVES FOR WAVEFUNCTION CUTOFF:             48658
 NUMBER OF PLANE WAVES FOR DENSITY CUTOFF:                 389055
 ****************************************************************


Then I checked how to obtain this lattice vectors in vmd 1.8.4a22 with
script pbctools.tcl and procedure set_unitcell_vmd.

Solution: I have to set angles in different way:
a 5.5     alpha   90
b 14.6    beta    90
c 10.9    gamma   102

cause in cpmd program and in set_unitcell_vmd script we've got a
different definitions of angles.

PS Now everything looks OK :)

best regards,
Przemyslaw Dopieralski
University of Wroclaw
The Faculty of Chemistry
Theoretical Group of prof. Z.Latajka
mclar at elrond.chem.uni.wroc.pl