Nan output in optical spectrum with ALDA kernel

a.ugolotti · Post by **a.ugolotti** » Tue Oct 17, 2017 3:58 pm

Dear Yamboers,

I am defining the optical parameters for calculating the optical absorption for 2D Si-based systems and currently understanding the weight of adding the fxc contribution due to TDDFT terms. In this case the output is produced without errors, but the 2nd and 3rd colums are all Nan, while the ones for IP are ok. I get no problems in calculating the RPA-LFE spectrum. I tried to enlarge the kernel size and the cutoff but it is not helping. What I am missing?
I am using Yambo 4.1.3 with QuantumEspresso scf output.

This is the report outputted of the calculation (with input listed at its end)

Code: Select all

     __      __   ______   __       __  _______    ______       
    |  \    /  \ /      \ |  \     /  \|       \  /      \      
     \$$\  /  $$|  $$$$$$\| $$\   /  $$| $$$$$$$\|  $$$$$$\     
      \$$\/  $$ | $$__| $$| $$$\ /  $$$| $$__/ $$| $$  | $$     
       \$$  $$  | $$    $$| $$$$\  $$$$| $$    $$| $$  | $$     
        \$$$$   | $$$$$$$$| $$\$$ $$ $$| $$$$$$$\| $$  | $$     
        | $$    | $$  | $$| $$ \$$$| $$| $$__/ $$| $$__/ $$     
        | $$    | $$  | $$| $$  \$ | $$| $$    $$ \$$    $$     
         \$$     \$$   \$$ \$$      \$$ \$$$$$$$   \$$$$$$      
                                                                
                 GPL Version 4.1.2 Revision 76                  
                           MPI Build                            
                   http://www.yambo-code.org                    
 
 10/17/2017 at 18:36 YAMBO @ jigen117

 [01] CPU structure, Files & I/O Directories
 ===========================================

 * CPU-Threads     :32(CPU)-1(threads)-1(threads@X)-1(threads@DIP)-1(threads@SE)-1(threads@RT)-1(threads@K)
 * MPI CPU         :  32
 * THREADS    (max): 1
 * THREADS TOT(max):  32
 * I/O NODES       : 1
 * Fragmented WFs  :yes

 CORE databases in   .
 Additional I/O in   .
 Communications in   .
 Input file     is   opt_rpa-xc_cutoff.in
 Report file    is   ./r-rpa-xc_cutoff_e100_test_optics_chi_tddft_rim_cut_01
 Job string(main): rpa-xc_cutoff_e100_test
 Log files      in ./LOG
  
 [RD./SAVE//ns.db1]------------------------------------------
  Bands                           : 20
  K-points                        :  7
  G-vectors             [RL space]:  56689
  Components       [wavefunctions]:  7119
  Symmetries       [spatial+T-rev]: 12
  Spinor components               : 1
  Spin polarizations              : 1
  Temperature                 [ev]:  0.02585
  Electrons                       : 8.000000
  WF G-vectors                    :  8905
  Max atoms/species               : 2
  No. of atom species             : 1
  Magnetic symmetries             : no
 - S/N 008127 -------------------------- v.04.01.02 r.00076 -

 [02] CORE Variables Setup
 =========================


  [02.01] Unit cells
  ==================

  Unit cell is HCP

  ... containing 2Si atoms

  ... with scaling factors [a.u.]:  7.25709   6.28483  37.79452

  Direct Lattice(DL) unit cell [iru  /  cc(a.u.)]
  A1 = 1.000000  0.000000 -0.000000      7.257094  0.000000 -0.000000
  A2 =-0.500000  1.000000 -0.000000     -3.628547  6.284828 -0.000000
  A3 =-0.000000  0.000000  1.000000     -0.000000  0.000000  37.79452

  DL volume [au]: 1724.   

  Reciprocal Lattice(RL) unit cell [iku  /  cc]
  B1 = 1.000000  0.500000  0.000000      0.865799  0.499869  0.000000
  B2 = 0.000000  1.000000  0.000000     -0.000000  0.999739 -0.000000
  B3 = 0.000000  0.000000  1.000000      0.000000  0.000000  0.166246


  [02.02] Symmetries
  ==================

  DL (S)ymmetries [cc]
  [S1] 1.000000  0.000000  0.000000  0.000000  1.000000  0.000000  0.000000  0.000000  1.000000
  [S2]-0.500000  0.866025  0.000000 -0.866025 -0.500000  0.000000  0.000000  0.000000  1.000000
  [S3]-0.500000 -0.866025  0.000000  0.866025 -0.500000  0.000000  0.000000  0.000000  1.000000
  [S4]-1.000000  0.000000  0.000000  0.000000  1.000000  0.000000  0.000000  0.000000  1.000000
  [S5] 0.500000  0.866025  0.000000  0.866025 -0.500000  0.000000  0.000000  0.000000  1.000000
  [S6] 0.500000 -0.866025  0.000000 -0.866025 -0.500000  0.000000  0.000000  0.000000  1.000000

  [SYMs] Time-reversal derived K-space symmetries:  7  12
  [SYMs] Spatial inversion 7 is NOT a symmetry
  [SYMs] Group table built correctly

  [02.03] RL shells
  =================

  Shells, format: [S#] G_RL(mHa)

   [S946]:16443(0.3397E+5) [S945]:16431(0.3397E+5) [S944]:16419(0.3395E+5) [S943]:16395(0.3395E+5)
   [S942]:16371(0.3393E+5) [S941]:16347(0.3387E+5) [S940]:16323(0.3384E+5) [S939]:16311(0.3383E+5)
   [S938]:16275(0.3383E+5) [S937]:16251(0.3382E+5) [S936]:16227(0.3374E+5) [S935]:16215(0.3372E+5)
   [S934]:16191(0.3370E+5) [S933]:16167(0.3368E+5) [S932]:16155(0.3366E+5) [S931]:16143(0.3364E+5)
   [S930]:16119(0.3361E+5) [S929]:16095(0.3359E+5) [S928]:16071(0.3355E+5) [S927]:16059(0.3354E+5)
   [S926]:16035(0.3354E+5) [S925]:16011(0.3351E+5) [S924]:15999(0.3350E+5) [S923]:15975(0.3348E+5)
   [S922]:15963(0.3347E+5) [S921]:15939(0.3347E+5) [S920]:15915(0.3345E+5) [S919]:15903(0.3340E+5)
   ...
   [S12]:57( 677.1240) [S11]:55( 624.1085) [S10]:43( 555.0141) [S9]:31( 513.5575)
   [S8]:19( 499.7387) [S7]:13( 497.4788) [S6]:11( 345.4714) [S5]:9( 221.1017)
   [S4]:7( 124.3697) [S3]:5( 55.27542) [S2]:3( 13.81886) [S1]:1( 0.000000)

  [02.04] K-grid lattice
  ======================

  Compatible Grid is 2D
  B1 [rlu]= 0.000000 -0.166667  0.000000
  B2      =-.1667    0.7451E-8  0.000   
  Grid dimensions               :  6   6
  K lattice UC volume       [au]:   0.0040

  [02.05] Energies [ev] & Occupations
  ===================================

  Fermi Level        [ev]: -2.864155
  VBM / CBm          [ev]:  0.00      0.00    
  Electronic Temp. [ev K]: 0.2585E-1  300.0   
  Bosonic    Temp. [ev K]: 0.2585E-1  300.0   
  El. density      [cm-3]: 0.313E+23
  States summary         : Full        Metallic    Empty
                           0001-0003   0004-0005   0006-0020

  [WARNING]Metallic system
   
  N of el / N of met el:  8.000000  1.999999
  Average metallic occ.:             0.500000
  X BZ K-points :  36

  Energy unit is electronVolt [eV]

  *X* K [1] : 0.000000  0.000000  0.000000 ( cc) * Comp.s 7053 * weight   0.02778
              0.000000  0.000000  0.000000 (rlu)
  E -11.13778  -3.21879  -1.18705  -1.18705   2.04991   2.52153   3.06019   3.06019
  E  4.328033  4.416226  5.412890  5.637754  6.903212  7.972698  8.552167  8.771686
  E   8.82500   8.82500  11.01532  11.33317
   
  *X* K [2] : 0.000000  0.166623  0.000000 ( cc) * Comp.s 7095 * weight  0.166667
              0.000000  0.166667  0.000000 (rlu)
  E -10.81022  -3.28464  -2.65837  -1.67256   2.20105   2.70812   4.16744   4.24279
  E  4.693976  4.725660  5.169508  6.002145  6.321871  8.121449  8.142830  8.342927
  E   9.48366  10.30769  11.24870  11.57309
   
  *X* K [3] : 0.000000  0.333246  0.000000 ( cc) * Comp.s 7100 * weight  0.166667
              0.000000  0.333333  0.000000 (rlu)
  E -9.867507 -5.642589 -2.340160 -1.933834  1.352883  2.035538  4.523439  5.790285
  E  5.962943  6.347785  6.973099  7.117610  7.228760  7.718422  8.133737  9.416586
  E   9.53086   9.73574  10.44994  10.58146
   
  *X* K [4] : 0.000000 -0.499869  0.000000 ( cc) * Comp.s 7100 * weight   0.08333
              0.000000 -0.500000  0.000000 (rlu)
  E -8.869076 -7.201611 -2.598315 -1.032564  0.626264  1.173529  5.556201  6.454691
  E  6.658648  7.375817  7.638889  7.834033  7.855898  8.263299  8.968507  9.030438
  E   9.07777   9.39492  10.31019  10.65354
   
  *X* K [5] : 0.144300  0.249935  0.000000 ( cc) * Comp.s 7063 * weight  0.166667
              0.166667  0.250000  0.000000 (rlu)
  E -10.16987  -4.81749  -2.76978  -2.19634   2.38611   3.41510   3.63350   4.51815
  E  5.365953  5.601500  6.099938  6.831061  6.871687  7.617240  8.862434  9.105433
  E  10.50354  10.57207  11.44216  11.72388
   
  *X* K [6] : 0.144300  0.416558  0.000000 ( cc) * Comp.s 7097 * weight  0.333333
              0.166667  0.416667  0.000000 (rlu)
  E -9.010526 -6.731303 -3.610681 -1.108508  1.003822  3.011045  4.852134  5.532335
  E  6.402900  6.992079  7.081530  7.714928  8.279523  8.314942  9.179128  9.325556
  E   9.81306  10.00363  10.18405  10.49881
   
  *X* K [7] : 0.288600  0.499869  0.000000 ( cc) * Comp.s 7119 * weight   0.05556
              0.333333  0.500000  0.000000 (rlu)
  E -7.827    -7.827    -4.484    -.1068E-3 0.1070E-3  3.998     5.744     6.237   
  E  6.237344  7.242418  7.242505  8.260670  8.755470  8.755647  8.962394  8.962548
  E   9.02068   9.06795  10.04447  10.04505

 [03] Transferred momenta grid
 =============================

 [RD./SAVE//ndb.kindx]---------------------------------------
  Fragmentation                    :no
  Polarization last K   :  7
  QP states             :  1   7
  X grid is uniform     :yes
  BS scattering         :yes
  COLL scattering       :no
 - S/N 008127 -------------------------- v.04.01.02 r.00076 -

 IBZ Q-points :  7
 BZ  Q-points :  36

 Q [00001] : 0.00      0.00      0.00     (iku) * weight   0.02778
 Q [00002] : 0.000000  0.166667  0.000000 (iku) * weight  0.166667
 Q [00003] : 0.000000  0.333333  0.000000 (iku) * weight  0.166667
 Q [00004] : 0.000000 -0.500000  0.000000 (iku) * weight   0.08333
 Q [00005] : 0.166667  0.250000  0.000000 (iku) * weight  0.166667
 Q [00006] : 0.166667  0.416667  0.000000 (iku) * weight  0.333333
 Q [00007] : 0.333333  0.500000  0.000000 (iku) * weight   0.05556
  
 Q [00001] : 0.00      0.00      0.00     (cc ) * weight   0.02778
 Q [00002] : 0.000000  0.166623  0.000000 (cc ) * weight  0.166667
 Q [00003] : 0.000000  0.333246  0.000000 (cc ) * weight  0.166667
 Q [00004] : 0.000000 -0.499869  0.000000 (cc ) * weight   0.08333
 Q [00005] : 0.144300  0.249935  0.000000 (cc ) * weight  0.166667
 Q [00006] : 0.144300  0.416558  0.000000 (cc ) * weight  0.333333
 Q [00007] : 0.288600  0.499869  0.000000 (cc ) * weight   0.05556

 [04] Coloumb potential Random Integration (RIM)
 ===============================================

 [RD./rpa-xc_cutoff_e100_test//ndb.RIM]----------------------
  Brillouin Zone Q/K grids (IBZ/BZ):   7   36    7   36
  Coulombian RL components        : 1
  Coulombian diagonal components  :yes
  RIM random points               : 1000000
  RIM  RL volume             [a.u.]: 0.143802
  Real RL volume             [a.u.]: 0.143898
  Eps^-1 reference component       :0
  Eps^-1 components                : 0.00      0.00      0.00    
  RIM anysotropy factor            : 0.000000
 - S/N 008127 -------------------------- v.04.01.02 r.00076 -

 Summary of Coulomb integrals for non-metallic bands |Q|[au] RIM/Bare:

 Q [1]:0.1000E-40.9886 * Q [2]: 0.166623 1.047394
 Q [5]: 0.288600 1.018670 * Q [3]: 0.333246 1.014024
 Q [6]: 0.440843 1.008480 * Q [4]: 0.499869 1.006546
 Q [7]: 0.577199 1.005252

 [05] Coloumb potential CutOff :box z
 ====================================

 [RD./rpa-xc_cutoff_e100_test//ndb.cutoff]-------------------
  Brillouin Zone Q/K grids (IBZ/BZ):   7   36    7   36
  CutOff Geometry                 :box z
  Coulomb cutoff potential        :box z 20.000
  Box sides length            [au]:  0.00000   0.00000  20.00000
  Sphere/Cylinder radius      [au]: 0.000000
  Cylinder length             [au]: 0.000000
  RL components                   :  8907
  RL components used in the sum   : 16443
  RIM corrections included        :yes
  RIM RL components               : 1
  RIM random points               : 1000000
 - S/N 008127 -------------------------- v.04.01.02 r.00076 -

 [06] External corrections
 =========================


 [07] Optics
 ===========

 [WF] Performing Wave-Functions I/O from ./SAVE

 [FFT-Rho] Mesh size:  15   15   81

 [WF loader] Normalization (few states)  min/max  :0.26711E-8  1.0000   

 [xc] Functional Perdew, Burke & Ernzerhof(X)+Perdew, Burke & Ernzerhof(C)
 [xc] LIBXC used to calculate xc functional
 [RD./rpa-xc_cutoff_e100_test//ndb.dip_iR_and_P]-------------
  Brillouin Zone Q/K grids (IBZ/BZ):   7   36    7   36
  RL vectors                   (WF):  8907
  Fragmentation                    :yes
  Electronic Temperature        [K]: 300.00000
  Bosonic    Temperature        [K]: 300.00000
  X band range           :  1  20
  X band range limits    :  5   4
  X e/h energy range [ev]:-1.0000000 -1.0000000
  RL vectors in the sum  :  8907
  [r,Vnl] included       :yes
  Using shifted grids    :no
  Using covariant dipoles:no
  Using G-space approach :yes
  Using R-space approach :no
  Direct v evaluation    :no
  Field momentum norm    :0.10000E-4
  Wavefunctions          :Perdew, Burke & Ernzerhof(X)+Perdew, Burke & Ernzerhof(C)
 - S/N 008127 -------------------------- v.04.01.02 r.00076 -

 [WARNING] The system is a metal but Drude term not included.
 [WF] Performing Wave-Functions I/O from ./SAVE

 [FFT-X] Mesh size:  15   15   81

 [X-CG] R(p) Tot o/o(of R)  :  21  114  100

 CpuTiming [Min/Max/Average]: 01m-03s/01m-07s/01m-05s

 [08] Game Over & Game summary
 =============================

 10/17/2017 at 18:36 YAMBO @ jigen117 [start]
 10/17/2017 at 18:38                  [end]

 Timing   [Min/Max/Average]: 01m-03s/01m-23s/01m-21s

 .-ACKNOWLEDGMENT
 |
 | The users of YAMBO have little formal obligations with respect to
 | the YAMBO group (those specified in the GNU General Public
 | License, http://www.gnu.org/copyleft/gpl.txt). However, it is
 | common practice in the scientific literature, to acknowledge the
 | efforts of people that have made the research possible. In this
 | spirit, please find below the reference we kindly ask you to use
 | in order to acknowledge YAMBO:
 |
 | Yambo: An ab initio tool for excited state calculations
 | A. Marini, C. Hogan, M. Gr\"uning, D. Varsano
 | Computer Physics Communications  180, 1392 (2009).
 |
  
 .-Input file : opt_rpa-xc_cutoff.in
 | optics                       # [R OPT] Optics
 | chi                          # [R CHI] Dyson equation for Chi.
 | tddft                        # [R   K] Use TDDFT kernel
 | rim_cut                      # [R RIM CUT] Coulomb potential
 | X_q_0_CPU= "16.2.1"          # [PARALLEL] CPUs for each role
 | X_q_0_ROLEs= "c.v.k"         # [PARALLEL] CPUs roles (k,c,v)
 | RandQpts= 1000000            # [RIM] Number of random q-points in the BZ
 | RandGvec= 1            RL    # [RIM] Coulomb interaction RS components
 | CUTGeo= "box z"              # [CUT] Coulomb Cutoff geometry: box/cylinder/sphere X/Y/Z/XY..
 | % CUTBox
 |   0.00000 |  0.00000 | 20.00000 |        # [CUT] [au] Box sides
 | %
 | CUTRadius= 0.000000          # [CUT] [au] Sphere/Cylinder radius
 | CUTCylLen= 0.000000          # [CUT] [au] Cylinder length
 | Chimod= "ALDA"               # [X] IP/Hartree/ALDA/LRC/BSfxc
 | FxcGRLc= 1             RL    # [TDDFT] XC-kernel RL size
 | NGsBlkXd= 109          RL    # [Xd] Response block size
 | % QpntsRXd
 |  1 | 1 |                     # [Xd] Transferred momenta
 | %
 | % BndsRnXd
 |   1 | 20 |                   # [Xd] Polarization function bands
 | %
 | GrFnTpXd= "R"                # [Xd] Green`s function (T)ordered,(R)etarded,(r)senant,(a)ntiresonant [T, R, r, Ta, Ra]
 | % EnRngeXd
 |   0.00000 | 10.00000 | eV    # [Xd] Energy range
 | %
 | % DmRngeXd
 |   0.10000 |  0.10000 | eV    # [Xd] Damping range
 | %
 | ETStpsXd= 1000               # [Xd] Total Energy steps
 | % LongDrXd
 | 0.1000E-4 | 0.000    | 0.000    |        # [Xd] [cc] Electric Field
 | %

What I am missing?

Thanks for your support,

Aldo Ugolotti
PhD student
Department of Materials Science
University of Milano-Bicocca
via Cozzi, 55, 20125 Milano
Italy

Post by **Daniele Varsano** » Tue Oct 17, 2017 4:35 pm

Dear Aldo,
can you upload the input/report and outputs of your calculations.
Such a problem can arise in the evaluation of the ALDA when the density approach to zero.
Please note that you can also calculate the ALDA in the e-h basis.
This way is somehow desirable as you include automatically all g-vecs of the kernel.
The input is generated by:

Code: Select all

yambo -o b -k alda -y d

In this way the kernel is evaluated in real space, and the problem of zero density should be mitigated.

Finally please pay attention to this:

Code: Select all

 Coulomb cutoff potential        :box z 20.000

This means you are cutting the Coulomb interaction after 10 a.u.
a safe way is to set it just a bit smaller than your cell: e.g. 37, cutting the interaction at half of the cell.

Best,

Daniele

a.ugolotti · Post by **a.ugolotti** » Wed Oct 18, 2017 11:08 am

Dear Daniele,

thanks for your quick reply. Switching to the e-h space the calculation is working (no NaN values or errors), however the output is showing a very different spectrum than RPE-LFE. Actually I am not yet proficient in this part of the theory enough to understand if this is expected or not.

The input is generated by:
Code: Select all
    yambo -o b -k alda -y d
In this way the kernel is evaluated in real space, and the problem of zero density should be mitigated.

I tried this option earlier, without specifying the solver method, and I got no output (There is a not-so-visible-error):

Code: Select all

 [03] Transferred momenta grid
 =============================

 [RD./SAVE//ndb.kindx]---------------------------------------
  Fragmentation                    :no
  Polarization last K   :  7
  QP states             :  1   7
  X grid is uniform     :yes
 *ERR* BS scattering         :no
  COLL scattering       :no
 - S/N 008127 -------------------------- v.04.01.02 r.00076 -
 [RL indx] Q-grid:User defined / from ndb.kindx.

I am attaching a file with the report and the output of the run in e-h space

Thanks,

Aldo Ugolotti
PhD student
Department of Materials Science
University of Milano-Bicocca
via Cozzi, 55, 20125 Milano
Italy

a.ugolotti · Post by **a.ugolotti** » Wed Oct 18, 2017 11:10 am

And here I am attaching a file with the report and the output of the non-working calculation in reciprocal space.

Aldo Ugolotti

Post by **Daniele Varsano** » Thu Oct 19, 2017 11:08 am

Dear Aldo,
you can insert your signature in your forum profile, in order to avoid to copy it in very post.

I tried this option earlier, without specifying the solver method, and I got no output (There is a not-so-visible-error):

If you do not specify the diagonalization, Yambo builds the excitonic (ALDA) matrix only so without diagonalization you will have the the matrix stored in the database, but you cannot calculate the spectrum, so no output.

however the output is showing a very different spectrum than RPE-LFE

The RPA-LF calculation in g space and eh space should be the same provided that the same parameters are used, you are just changing the base.
If you do a calculation:

Code: Select all

yambo -o b -k sex -y d

setting

BSENGexx

to the same value of

Code: Select all

NGsBlkXd

in the G space and considering the same bands as you are doing you should have the same results.
Most probably the spectra are different because these two values in the two calculations are very different, more than the effect of the ALDA kernel.
Increasing BSENGexx in eh space is much more lighter than doing it in Gspace.

PS: I can see form your report that your system is a metal, so I would not expect big difference from the non interacting picture. Local field could play a role in the direction perperndicular to the plane where the system lies.

Best,

Daniele

a.ugolotti · Post by **a.ugolotti** » Fri Oct 20, 2017 4:32 pm

Dear Daniele,

I think I found the problem. I was considering too few empty states, now the spectra are behaving as expected.
Thanks for your support.

Yambo Community Forum

Nan output in optical spectrum with ALDA kernel

Nan output in optical spectrum with ALDA kernel

Re: Nan output in optical spectrum with ALDA kernel

Re: Nan output in optical spectrum with ALDA kernel

Re: Nan output in optical spectrum with ALDA kernel

Re: Nan output in optical spectrum with ALDA kernel

Re: Nan output in optical spectrum with ALDA kernel