3
$\begingroup$

I have run into trouble when I try to do transport calculations in TranSIESTA. I got this message in the output file, before reaching convergence: Error on inverting X6/C7 with error: 1

this is the input:

"SolutionMethod        Transiesta
SystemName          scat
SystemLabel         scat
PAO.BasisSize    DZP
PAO.EnergyShift  0.005 Ry
XC.functional   VDW
XC.authors      KBM

LatticeConstant             1.00 Ang

%block LatticeVectors
      38.6768112183         0.0000000000         0.0000000000
       0.0000000000        18.6851730347         0.0000000000
       0.0000000000         0.0000000000        30.1042270660
%endblock LatticeVectors

# ---------------------------------------------------------------------------
# Species and Atoms
# ---------------------------------------------------------------------------

NumberOfSpecies        2
NumberOfAtoms         182

%block ChemicalSpeciesLabel
  1   6    C
  2   1    H
%endblock ChemicalSpeciesLabel

# ---------------------------------------------------------------------------
# Atomic Coordinates
# ---------------------------------------------------------------------------

AtomicCoordinatesFormat Ang
%block AtomicCoordinatesAndAtomicSpecies
10.074077    9.200517    1.487844   1
12.528448    9.200517    1.487844   1
14.982818    9.200517    1.487844   1
17.437189    9.200517    1.487844   1
19.891562    9.200517    1.487844   1
 8.846891    9.200517    3.613365   1
11.301263    9.200517    3.613365   1
13.755634    9.200517    3.613365   1
16.210005    9.200517    3.613365   1
18.664376    9.200517    3.613365   1
21.118749    9.200517    3.613365   1
 8.846878    9.200517    0.779387   1
11.301249    9.200517    0.779387   1
13.755620    9.200517    0.779387   1
16.209992    9.200517    0.779387   1
18.664364    9.200517    0.779387   1
21.118736    9.200517    0.779387   1
10.074065    9.200517    2.904908   1
12.528435    9.200517    2.904908   1
14.982807    9.200517    2.904908   1
17.437178    9.200517    2.904908   1
19.891548    9.200517    2.904908   1
 7.879461    9.200519    1.284409   1
 7.879476    9.200522    3.108352   1
22.088700    9.200522    1.279482   1
22.088735    9.200524    3.113323   1
10.074077    9.200517    5.788448   1
12.528448    9.200517    5.788448   1
14.982818    9.200517    5.788448   1
17.437189    9.200517    5.788448   1
19.891562    9.200517    5.788448   1
 8.846891    9.200517    7.913969   1
11.301263    9.200517    7.913969   1
13.755634    9.200517    7.913969   1
16.210005    9.200517    7.913969   1
18.664376    9.200517    7.913969   1
21.118749    9.200517    7.913969   1
 8.846878    9.200517    5.079991   1
11.301249    9.200517    5.079991   1
13.755620    9.200517    5.079991   1
16.209992    9.200517    5.079991   1
18.664364    9.200517    5.079991   1
21.118736    9.200517    5.079991   1
10.074065    9.200517    7.205512   1
12.528435    9.200517    7.205512   1
14.982807    9.200517    7.205512   1
17.437178    9.200517    7.205512   1
19.891548    9.200517    7.205512   1
 7.879461    9.200519    5.585013   1
 7.879476    9.200522    7.408956   1
22.088700    9.200522    5.580086   1
22.088735    9.200524    7.413927   1
 10.074077   9.200517    10.089052  1
12.528448    9.200517   10.089052   1
14.982818    9.200517   10.089052   1
17.437189    9.200517   10.089052   1
19.891562    9.200517   10.089052   1
 8.846891    9.200517   12.214573   1
11.301263    9.200517   12.214573   1
13.755634    9.200517   12.214573   1
16.210005    9.200517   12.214573   1
18.664376    9.200517   12.214573   1
21.118749    9.200517   12.214573   1
 8.846878    9.200517    9.380595   1
11.301249    9.200517    9.380595   1
13.755620    9.200517    9.380595   1
16.209992    9.200517    9.380595   1
18.664364    9.200517    9.380595   1
21.118736    9.200517    9.380595   1
10.074065    9.200517   11.506116   1
12.528435    9.200517   11.506116   1
14.982807    9.200517   11.506116   1
17.437178    9.200517   11.506116   1
19.891548    9.200517   11.506116   1
 7.879461    9.200519    9.885617   1
 7.879476    9.200522   11.709560   1
22.088700    9.200522    9.880690   1
22.088735    9.200524   11.714531   1
 10.074077   9.200517   14.389656   1
12.528448    9.200517   14.389656   1
14.982818    9.200517   14.389656   1
17.437189    9.200517   14.389656   1
19.891562    9.200517   14.389656   1
 8.846891    9.200517   16.515176   1
11.301263    9.200517   16.515176   1
13.755634    9.200517   16.515176   1
16.210005    9.200517   16.515176   1
18.664376    9.200517   16.515176   1
21.118749    9.200517   16.515176   1
 8.846878    9.200517   13.681199   1
11.301249    9.200517   13.681199   1
13.755620    9.200517   13.681199   1
16.209992    9.200517   13.681199   1
18.664364    9.200517   13.681199   1
21.118736    9.200517   13.681199   1
10.074065    9.200517   15.806720   1
12.528435    9.200517   15.806720   1
14.982807    9.200517   15.806720   1
17.437178    9.200517   15.806720   1
19.891548    9.200517   15.806720   1
 7.879461    9.200519   14.186221   1
 7.879476    9.200522   16.010164   1
22.088700    9.200522   14.181294   1
22.088735    9.200524   16.015135   1
10.074077    9.200517   18.690259   1
12.528448    9.200517   18.690259   1
14.982818    9.200517   18.690259   1
17.437189    9.200517   18.690259   1
19.891562    9.200517   18.690259   1
 8.846891    9.200517   20.815780   1
11.301263    9.200517   20.815780   1
13.755634    9.200517   20.815780   1
16.210005    9.200517   20.815780   1
18.664376    9.200517   20.815780   1
21.118749    9.200517   20.815780   1
 8.846878    9.200517   17.981802   1
11.301249    9.200517   17.981802   1
13.755620    9.200517   17.981802   1
16.209992    9.200517   17.981802   1
18.664364    9.200517   17.981802   1
21.118736    9.200517   17.981802   1
10.074065    9.200517   20.107323   1
12.528435    9.200517   20.107323   1
14.982807    9.200517   20.107323   1
17.437178    9.200517   20.107323   1
19.891548    9.200517   20.107323   1
 7.879461    9.200519   18.486824   1
 7.879476    9.200522   20.310767   1
22.088700    9.200522   18.481898   1
22.088735    9.200524   20.315738   1
 0.074077    9.200517   22.990863   1
12.528448    9.200517   22.990863   1
14.982818    9.200517   22.990863   1
17.437189    9.200517   22.990863   1
19.891562    9.200517   22.990863   1
 8.846891    9.200517   25.116384   1
11.301263    9.200517   25.116384   1
13.755634    9.200517   25.116384   1
16.210005    9.200517   25.116384   1
18.664376    9.200517   25.116384   1
21.118749    9.200517   25.116384   1
 8.846878    9.200517   22.282406   1
11.301249    9.200517   22.282406   1
13.755620    9.200517   22.282406   1
16.209992    9.200517   22.282406   1
18.664364    9.200517   22.282406   1
21.118736    9.200517   22.282406   1
10.074065    9.200517   24.407927   1
12.528435    9.200517   24.407927   1
14.982807    9.200517   24.407927   1
17.437178    9.200517   24.407927   1
19.891548    9.200517   24.407927   1
 7.879461    9.200519   22.787428   1
 7.879476    9.200522   24.611371   1
22.088700    9.200522   22.782502   1
22.088735    9.200524   24.616342   1
10.074077    9.200517   27.291467   1
12.528448    9.200517   27.291467   1
14.982818    9.200517   27.291467   1
17.437189    9.200517   27.291467   1
19.891562    9.200517   27.291467   1
 8.846891    9.200517   29.416988   1
11.301263    9.200517   29.416988   1
13.755634    9.200517   29.416988   1
16.210005    9.200517   29.416988   1
18.664376    9.200517   29.416988   1
21.118749    9.200517   29.416988   1
 8.846878    9.200517   26.583010   1
11.301249    9.200517   26.583010   1
13.755620    9.200517   26.583010   1
16.209992    9.200517   26.583010   1
18.664364    9.200517   26.583010   1
21.118736    9.200517   26.583010   1
10.074065    9.200517   28.708531   1
12.528435    9.200517   28.708531   1
14.982807    9.200517   28.708531   1
17.437178    9.200517   28.708531   1
19.891548    9.200517   28.708531   1
 7.879461    9.200519   27.088032   1
 7.879476    9.200522   28.911975   1
22.088700    9.200522   27.083106   1
22.088735    9.200524   28.916946   1                
%endblock AtomicCoordinatesAndAtomicSpecies
 # K-points
%block kgrid_Monkhorst_Pack
  20    0    0   0.0
   0    1    0   0.0
   0    0   100   0.0
%endblock kgrid_Monkhorst_Pack

DM.MixSCF1   T
MaxSCFIterations      300           # Maximum number of SCF iter
DM.MixingWeight       0.05          # New DM amount for next SCF cycle
DM.Tolerance          1.d-4         # Tolerance in maximum difference
DM.UseSaveDM          true          # to use continuation files
DM.NumberPulay        6

==================================================
==================================================
# MD variables

MD.FinalTimeStep 1
MD.TypeOfRun CG
MD.NumCGsteps     000
MD.UseSaveXV      .true.
WriteMullikenPop                1
WriteBands                      .false.
SaveRho                         .true.
SaveDeltaRho                    .true.
SaveHS                          .true.
SaveElectrostaticPotential      .true. 
SaveTotalPotential              .true.
WriteCoorXmol                   .true.
WriteMDXmol                     .true.
WriteMDhistory                  .false.
WriteEigenvalues                .true.
XML.Write                       .true.
TS.HS.Save          .true.
TS.DE.Save          .true.
ParallelOverK       .true.
User.Basis          .true.
Diag.ParallelOverK  .true. 

==================================================
==================================================
#####################TRANSIESTA###################

# Bias voltage


TS.Voltage 0.0 eV



%block TS.ChemPots
  Left
  Right
%endblock TS.ChemPots

%block TS.ChemPot.Left
  mu V/2
  contour.eq
    begin
      c-Left
      t-Left
    end
%endblock TS.ChemPot.Left
%block TS.ChemPot.Right
  mu -V/2
  contour.eq
    begin
      c-Right
      t-Right
    end
%endblock TS.ChemPot.Right

TS.Elecs.Bulk true
TS.Elecs.DM.Update none
TS.Elecs.GF.ReUse false
%block TS.Elecs
  Left
  Right
%endblock TS.Elecs

%block TS.Elec.Left
  TSHS ./elec.TSHS
  chem-pot Left
  semi-inf-dir -a3
  elec-pos begin 1
%endblock TS.Elec.Left

%block TS.Elec.Right
  TSHS ./elec.TSHS
  chem-pot Right
  semi-inf-dir +a3
  elec-pos end -1
%endblock TS.Elec.Right

TS.Contours.Eq.Pole 2.5 eV
%block TS.Contour.c-Left
  part circle
   from   -40.00000 eV + V/2 to -10. kT + V/2
    points 30
     method g-legendre
%endblock TS.Contour.c-Left
%block TS.Contour.t-Left
  part tail
   from prev to inf
    points 10
     method g-fermi
%endblock TS.Contour.t-Left
%block TS.Contour.c-Right
  part circle
   from   -40.00000 eV - V/2 to -10. kT - V/2
    points 30
     method g-legendre
%endblock TS.Contour.c-Right
%block TS.Contour.t-Right
  part tail
   from prev to inf
    points 10
     method g-fermi
%endblock TS.Contour.t-Right

TS.Contours.nEq.Eta 0.0001 eV
%block TS.Contours.nEq
  neq
%endblock TS.Contours.nEq
%block TS.Contour.nEq.neq
  part line
   from -|V|/2 - 5 kT to |V|/2 + 5 kT
    delta 0.01 eV
     method mid-rule
%endblock TS.Contour.nEq.neq
#%block LocalDensityOfStates
#-2.21 0.00 eV
#%endblock LocalDensityOfStates

TBT.DOS.A.All T
TBT.DOS.Gf   T
TS.SolutionMethod btd
TS.BTD.Pivot atom+Left
TS.Elecs.Neglect.Principal true
==================================================
==================================================

After SCF calculation step; i got:

                  ***************************
                     *  WELCOME TO TRANSIESTA  *
                     ***************************

ts-Vha: -0.11730E-01 eV

transiesta: created H and S sparsity pattern:
  <sparsity:(M of: G ((UC of: T T T sparsity for geom step 0)))
    nrows_g=2366 nrows=2366 sparsity=.1333 nnzs=746036, refcount: 1>

transiesta: created update sparsity pattern:
  <sparsity:G ((UC of: (M of: T T T sparsity for geom step 0)))
    nrows_g=2366 nrows=2366 sparsity=.1072 nnzs=599924, refcount: 1>

transiesta: Determining an optimal tri-matrix using: atom+Left
transiesta: Established a near-optimal partition for the tri-diagonal matrix.
BTD partitions (9): 
  [ 312, 364, [338] * 3, 325, 338, 1, 12 ]
transiesta: Matrix elements in % of full matrix:  38.63664 %

transiesta: mem of electrodes (static):                     2.86MB
transiesta: mem of global update arrays (static):          41.08MB
transiesta: mem of tri-diagonal matrices:                  66.01MB
transiesta: Total memory usage:                           109.94MB

transiesta: Charge distribution, target =    728.00000
Total charge                  [Q]  :   728.00000
Device                        [D]  :   514.03356
Left                          [E1] :    98.05875
Left                 / device [C1] :     5.94659
Right                         [E2] :    98.05467
Right                / device [C2] :     5.94168
Other                         [O]  :     5.96475

Error on inverting X6/C7 with error: 1
Stopping Program from Node:    7
Error on inverting X6/C7 with error: 1
Stopping Program from Node:   18

What is causing this error?

$\endgroup$
3
  • $\begingroup$ If possible, please provide your input for this calculation, so we can diagnose if there is some problem with your input structure/settings. $\endgroup$
    – Tyberius
    Feb 17, 2023 at 16:34
  • 1
    $\begingroup$ Just a note to the close voters, since there's one vote for "opinion-based closure" and 1 vote for "needs more focus". Opinion-based closures are usually for questions like "do you think John Pople really deserved the Nobel Prize in chemistry?" or "should Don Truhlar get a Nobel Prize?". The answers would be much more opinion-based than fact-based. "Needs more focus" is for when there's more than one question in the post; the subheading says "This question currently includes multiple questions in one. It should focus on one problem only." What is causing the error is probably something that $\endgroup$ Mar 3, 2023 at 23:28
  • $\begingroup$ is a matter of "fact" rather than opinion, although in this case the answer by Nick Papior (a Siesta developer by the way, so knowledgeable for this question that has the Siesta tag) and the comments on it suggest that the question could have had better detail in it, so "needs more detail / clarity" would be more appropriate than "opinion-based" or "needs more focus". But, we have a convention that we keep questions open if they have an answer with a positive net score, see that post for the reason why. My vote is keep open @Tyberius $\endgroup$ Mar 3, 2023 at 23:32

1 Answer 1

3
$\begingroup$

Inversion errors in TranSiesta happens when the system has 0's in the diagonal of the matrices. In this case the error either stems from a call to:

   zgetrf
   zgetri

in which case the error means that the U(1,1) element is zero and thus the matrix is singular.

From the question it is unclear whether this is related to some settings in the input, or whether it is a peculiar artefact from the geometry. Generally the matrices should not be singular as one adds a fictitious imaginary value that ensures some broadening in the complex plane.

However, it may happen when two parts of the system are physically disconnected, in which case the imaginary part does not enter through the self-energies and then the chances of hitting a singular energy-point becomes much higher.

I would highly suggest you to understand all options and their implications.

  1. you are using k-points along directions which are non-periodic
  2. the electrodes are using PDOS k-points along non-periodic directions
  3. the mesh-cutoff are different in the two calculations
  4. the basis sets are different in the two calculations

All in all you are not being consistent and this may be the root cause of your problems. Always ensure that electrode settings are equivalent to the electrode part of the device region!

$\endgroup$
5
  • $\begingroup$ Thank you, Mr. Nick, I add the input file above. I would appreciate it if you could take a look and provide your feedback., If the error is due to the geometric structure or some input command setting. ( for other size structures, the calculation is done without any problem, except for this) $\endgroup$
    – Hananab
    Feb 19, 2023 at 10:13
  • $\begingroup$ I can't immediately see anything wrong. One thing though is that you have k-points along the x-direction, why? Also, it might be easier to check if you have a way to upload the pseudo's. $\endgroup$
    – nickpapior
    Feb 20, 2023 at 7:46
  • $\begingroup$ Could you also add the electrode inputs? $\endgroup$
    – nickpapior
    Feb 21, 2023 at 7:40
  • $\begingroup$ It is not working, I can't access it. $\endgroup$
    – nickpapior
    Feb 21, 2023 at 11:43
  • $\begingroup$ @ABHanan see my updated response, I will not test this anymore, there are too many inconsistencies between your electrode and device region calculations. $\endgroup$
    – nickpapior
    Feb 22, 2023 at 8:18

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .