[gpaw-users] Transport calculation--atom "too close to boundary"

[Jens Jørgen Mortensen]

On 20-02-2012 12:40, Jouko Nieminen wrote:
> Hi,
>
> Our group has also been struggling with this very same cut-off problem
> in STM-simulations:
>
>> RuntimeError: Atom at 0.506 0.503 0.496 too close to boundary (beg. of box [1 1  1], end of box [36 35 35])
> and useful ideas to deal with this are very welcome.
>
> We have in principle just run the scripts of the following tutorial
> https://wiki.fysik.dtu.dk/gpaw/tutorials/negfstm/negfstm.html
> but changed from Aluminium to Copper or Palladium, and the geometry from
> fcc(100) to fcc(111) and changing the thickness of the slab.
>
>
> Jussi suggests that centering the atoms in the cell would help.
> Referring to the tutorial above, where exactly the centerings should be
> done? After setting up the surface or surface principal layers? Or the
> tip or the principal layers?
>
> This cut-off issue seems to be a persistent problem, and browsing the
> discussions some years backwards, it was somewhere suggested there is a
> bug in the stm code
> https://listserv.fysik.dtu.dk/pipermail/gpaw-users/2010-July/000236.html
>
> Is there anyone developing STM codes at the moment? Since this seems to
> be a recurrent problem, it would be nice to get this fixed.

The code was written by Janosch Michael Rauba for his masters project  - 
he is no longer on the mailing list.  I don't think there is anyone 
around who is familiar enough with the code to fix bugs in it :(

Jens Jørgen

>
> with kind regards,
>
>                            Jouko Nieminen
>
> PS.  Here is the error message in total:
>
>> Traceback (most recent call last):
>>    File "STMsimu.py", line 54, in<module>
>>      stm.initialize()
>>    File "/v/linux26_x86_64/appl/nano/gpaw/0.7.2/lib/python/gpaw/transport/jstm.py
>> ", line 352, in initialize
>>      self.tip_cell.initialize(tip_indices, tip_atom_index)
>>    File "/v/linux26_x86_64/appl/nano/gpaw/0.7.2/lib/python/gpaw/transport/jstm.py
>> ", line 1265, in initialize
>>      f = AtomCenteredFunctions(self.gd, [phit], spos_c, i)
>>    File "/v/linux26_x86_64/appl/nano/gpaw/0.7.2/lib/python/gpaw/transport/jstm.py
>> ", line 160, in __init__
>>      smallgd.N_c)
>>    File "/v/linux26_x86_64/appl/nano/gpaw/0.7.2/lib/python/gpaw/lfc.py", line 260
>> , in set_positions
>>      movement |= sphere.set_position(spos_c, self.gd, self.cut)
>>    File "/v/linux26_x86_64/appl/nano/gpaw/0.7.2/lib/python/gpaw/lfc.py", line 89,
>>   in set_position
>>      for beg_c, end_c, sdisp_c in gd.get_boxes(spos_c, rcut, cut):
>>    File "/v/linux26_x86_64/appl/nano/gpaw/0.7.2/lib/python/gpaw/grid_descriptor.p
>> y", line 262, in get_boxes
>>      (tuple(spos_c) + (beg_c, end_c)))
>> RuntimeError: Atom at 0.506 0.503 0.496 too close to boundary (beg. of box [1 1
>> 1], end of box [36 35 35])
>> srun: error: c235: task0: Exited with exit code 1
>