[Pw_forum] Questions regarding max_irr_dim

[Miguel Martínez]

Dear espreso lovers,

I've been lately wondering why max_irr_dim is set to 4 in
espresso/PH/phcom.f90. Space group no 223 (Pm-3n) for example can have
6-fold degeneracy in R, and I'm pretty sure other space groups also may
display high degeneracy. I know it's just a swift change and a quick
recompile away, but maybe there's an additional reason why this
parameter is set to 4. Memory usage? Lost NaNs? Warning sign to the
interested ones?

Another side question is how espresso determines the system symmetry.
Performing calculation on AlH3 at ~110 GPa (2 formula units per cell,
space grop Pm-3n) it just happens that with some magic parametres the
phonon degeneracy at R splits in half, even after fixing nq(i). Maybe
this is related to the line 26 comment in PH/find_mode_sym.f90:

 "It does NOT work at zone border in non symmorphic space groups"

Finally, the REPRES tool at www.cryst.ehu.es states that the Pm-3n
group at R (0.5 0.5 0.5) can have three dim2 irreps and one dim6 irrep,
while ph.x finds one 4-fold degenerate mode. I've consulted one of the
mantainers and he checked those values. Is there any reason why
espresso would get a false degeneracy?

Regards,

Miguel

PS: In case one is interested in the input files...

============  alh3.scf.in ===============
&control
	calculation	=	'scf'
	restart_mode	=	'from_scratch'
	pseudo_dir 	=	'./'
	outdir		=	'/tmp/espresso-miguel/'
	prefix		=	'alh3'
 /
&system
        ibrav           =       1
        celldm(1)       =       5.8204  
        nat             =       8
        ntyp            =       2
        ecutwfc         =       30. 
        ecutrho         =       200.    
        nbnd            =       20
        occupations     =       'smearing'
        smearing        =       'marzari-vanderbilt'
        degauss         =       0.02
 /
&electrons
	diagonalization	=	'david'
	conv_thr        =       1.D-8
        mixing_beta     =       0.7D0
 /
ATOMIC_SPECIES
Al 26.981538    Al.pbe-n-van.UPF
H  1.00794      H.pbe-van_ak.UPF
ATOMIC_POSITIONS crystal
Al      0.000000000   0.000000000   0.000000000
Al      0.500000000   0.500000000   0.500000000
H       0.250000000   0.000000000   0.500000000
H       0.750000000   0.000000000   0.500000000
H       0.500000000   0.250000000   0.000000000
H       0.500000000   0.750000000   0.000000000
H       0.000000000   0.500000000   0.250000000
H       0.000000000   0.500000000   0.750000000
K_POINTS automatic
12 12 12  1 1 1


============= alh3.ph.in ==============
phonons of AlH3
&inputph
        tr2_ph  =       1.0d-14
        prefix  =       'alh3'
        fildyn  =       'mat.dyn'
        outdir  =       '/tmp/espresso-miguel'
        amass(1)=       26.981538
        amass(2)=       1.00794
	lnscf	=	.true.
/
-0.5 -0.5 -0.5





-- 
----------------------------------------
Miguel Martínez Canales
   Condensed Matter Physics Dpt.
   UPV/EHU
   Faculty of Science and Technology
   Apdo. 644
   48080 Bilbao (Spain)
Fax:  +34 94 601 3500
Tlf:  +34 94 601 5326
----------------------------------------

 "If you have an apple and I have an apple and
 we exchange these apples then you and I will
 still each have one apple. But if you have an
 idea and I have an idea and we exchange these
 ideas, then each of us will have two ideas."

 George Bernard Shaw