7
$\begingroup$

I am trying to calculate the adsorption energy of benzoic acid into a PtRu surface, but the energy that I am finding is too high (around 40eV). The reported values are around -2.0eV for this. Also, I have done relaxation on the bare surface beforehand.

I have checked, and the structure does make sense visually, and can't find anything wrong with the code.

The input file is as follows:

calculation   = "vc-relax"
    forc_conv_thr =  1.00000e-05
    max_seconds   =  1.49299e+10
    nstep         = 200
    pseudo_dir    = "/gpfs/home/s362484/pseudopot"
/

&SYSTEM
    a                         =  1.06742e+01
    c                         =  4.65366e+01
    degauss                   =  1.00000e-02
    ecutrho                   =  2.25000e+02
    ecutwfc                   =  2.50000e+01
    ibrav                     = 4
    nat                       = 79
    nosym                     = .TRUE.
    nspin                     = 2
    ntyp                      = 5
    occupations               = "smearing"
    smearing                  = "gaussian"
    starting_magnetization(1) =  0.00000e+00
    starting_magnetization(2) =  2.00000e-01
/

&ELECTRONS
    conv_thr         =  1.00000e-07
    electron_maxstep = 1001
    mixing_beta      =  4.00000e-01
    startingpot      = "atomic"
    startingwfc      = "atomic+random"
/

&IONS
    ion_dynamics = "bfgs"
/

&CELL
    cell_dofree    = "xyz"
    cell_dynamics  = "bfgs"
    press_conv_thr =  5.00000e-01
/

K_POINTS {gamma}

ATOMIC_SPECIES
Pt    195.07800  Pt.pbe_v1.4.uspp.F.UPF
Ru    101.07000  Ru.pbe_v1.2.uspp.F.UPF
C      12.01070  C.pbe-v1.2.uspp.F.UPF
H       1.00794  H.pbe_v1.4.uspp.F.UPF
O      15.99940  O.pbe_v1.2.uspp.F.UPF

ATOMIC_POSITIONS {angstrom}
Pt      1.275388   0.556116   7.866917
Ru      3.864580   0.563153   7.800256
Ru     -0.072398   2.802120   7.815516
Pt      2.549131   2.792567   7.901911
Pt     -0.025204   1.306827   5.491642
Ru      2.576327   1.312502   5.580228
Ru     -1.365849   3.538932   5.574034
Pt      1.241654   3.543441   5.523319
Ru      0.000000   0.000000   3.074393  0 0 0
Pt      9.341031   2.159495   3.074393  0 0 0
Ru      1.199876   2.159495   3.074393  0 0 0
Pt     -0.133292   4.318990   3.074393  0 0 0
Pt      1.244307   0.719832   1.147382  0 0 0
Ru      3.777353   0.719832   1.147382  0 0 0
Ru     -0.088861   2.879327   1.147382  0 0 0
Pt      2.444184   2.879327   1.147382  0 0 0
Pt     -1.394925   5.032602   7.879935
Ru      1.198351   5.039890   7.818946
Ru     -2.736450   7.279528   7.776659
Pt     -0.119624   7.268719   7.860770
Pt     -2.700512   5.775223   5.478859
Ru     -0.097339   5.776998   5.570822
Ru     -4.036870   8.008524   5.502396
Pt     -1.429037   8.012987   5.462919
Ru      8.007862   4.318990   3.074393  0 0 0
Pt      6.674694   6.478487   3.074393  0 0 0
Ru     -1.466461   6.478487   3.074393  0 0 0
Pt      2.533046   0.000000   3.074393  0 0 0
Pt     -1.422030   5.038823   1.147382  0 0 0
Ru      1.111015   5.038823   1.147382  0 0 0
Ru     -2.755199   7.198318   1.147382  0 0 0
Pt     -0.222154   7.198318   1.147382  0 0 0
Pt      6.485744   0.558911   7.863799
Ru      9.073083   0.563302   7.767786
Ru      5.138507   2.804206   7.803710
Pt      7.753834   2.792301   7.880770
Pt      5.178159   1.310595   5.501473
Ru      7.782130   1.314462   5.548857
Ru      3.838276   3.544866   5.590459
Pt      6.449041   3.542363   5.483949
Ru      5.066091   0.000000   3.074393  0 0 0
Pt      3.732922   2.159495   3.074393  0 0 0
Ru      6.265967   2.159495   3.074393  0 0 0
Pt      4.932799   4.318990   3.074393  0 0 0
Pt      6.310398   0.719832   1.147382  0 0 0
Ru      8.843444   0.719832   1.147382  0 0 0
Ru      4.977229   2.879327   1.147382  0 0 0
Pt      7.510275   2.879327   1.147382  0 0 0
Pt      3.816391   5.037524   7.876924
Ru      6.406044   5.039404   7.786626
Ru      2.473515   7.279361   7.762624
Pt      5.087473   7.269748   7.831705
Pt      2.501320   5.778928   5.489335
Ru      5.108884   5.778008   5.525068
Ru      1.164620   8.012964   5.524228
Pt      3.775997   8.012012   5.423490
Ru      2.399754   4.318990   3.074393  0 0 0
Pt      1.066584   6.478487   3.074393  0 0 0
Ru      3.599630   6.478487   3.074393  0 0 0
Pt      7.599137   0.000000   3.074393  0 0 0
Pt      3.644061   5.038823   1.147382  0 0 0
Ru      6.177106   5.038823   1.147382  0 0 0
Ru      2.310892   7.198318   1.147382  0 0 0
Pt      5.839048   6.623790   1.147382  0 0 0
C       0.640571   4.162306  10.141480
C       0.347960   2.801376  10.141911
C       1.379344   1.856328  10.142238
C       2.707551   2.289810  10.142109
C       3.010109   3.650135  10.141798
C       1.974827   4.597482  10.141534
H       4.050234   3.983906  10.141745
H       3.524154   1.562031  10.142285
H       1.152836   0.784935  10.142553
H      -0.694675   2.470069  10.142019
H      -0.155340   4.911985  10.141175
C       2.245659   6.054638  10.141617
O       3.582601   6.344543  10.143125
O       1.397839   6.934842  10.140667
H       3.632729   7.329237  10.143239

Thank you

$\endgroup$
8
  • $\begingroup$ +1 Welcome to our community! $\endgroup$
    – Camps
    Commented Apr 11 at 16:21
  • 1
    $\begingroup$ How many calculations did you do? There are necessary at least three calculations and to use the BSSE correction. $\endgroup$
    – Camps
    Commented Apr 11 at 16:25
  • 2
    $\begingroup$ You need to do three calculations with the same conditions (same type of pseudos, parameters, cell size, etc.). One for the bare slab (energy $E_{slab}$), one for the molecule ($E_{mol}$) and one for the complex slab+molecule ($E_{slab+mol}$). Then you calculate the adsorption energy as $E_{ads}=E_{slab+mol}-E_{slab}-E_{mol}$. About the BSSE, take a look here. $\endgroup$
    – Camps
    Commented Apr 11 at 23:34
  • 1
    $\begingroup$ @CésarCatizanethis question is about why the energy is too high. Please ask questions about CP and BSSE in separate posts. $\endgroup$ Commented Apr 12 at 11:35
  • 1
    $\begingroup$ @CésarCatizane It would help if you attached logs of all three calculations. For a while, "vc-relax" looks suspicious, Usually you don't relax cell for slabs with adsorbates, $\endgroup$
    – user36313
    Commented Apr 12 at 12:02

1 Answer 1

4
$\begingroup$

To compute adsorption energy you need to to do the following:

  1. Do 'vc-relax' for a bulk PtRu crystal. A small unit cell should be enough. But make sure that you have enough k-points (test convergence). Write down the obtained energy per unit cell as $E_{bulk}$

  2. Using the relaxed bulk crystal from Step 1, create a slab of the required facet. (Facet (111) in your case, it looks good). Probably you don't need so much vacuum (you took about 40 A, 20 A is usually enough for a first run). But probably need more metal layers (you took only 4, this is too little). Fix a few bottom layers (you have fixed two layers, it is OK). Relax using simple calculation="relax" (not "vc-relax"). Write down the relaxed slab energy as $E_{slab}$. By the way, now you can get surface energy, that you may want to compare with literature data if it is available: $E_{surf} = (E_{slab} - N E_{bulk}) / S$, where $N$ - is the number of crystal units in the slab, and $S$ is the surface area of the slab (keep in mind, that you slab has two sides ;-) ).

  3. Place your molecule on the relaxed slab from Step 2. Your configuration looks reasonable, aside from above noted too thick vacuum and too thin slab. Good job. Do relaxation with calculation="relax". Write down the obtained relaxed energy of the slab with adsorbate as $E_{total}$.

  4. Separately relax with calculation="relax" the same molecule in the same cell, but without slab. Write down the obtained relaxed energy of the molecule as $E_{mol}$.

  5. Your adsorption energy will be $E_{ads} = E_{total} - E_{mol} - E_{slab} $

  6. And now you may start thinking about effects of lateral interactions, vibrational energy, BSSE, solvent effects, etc.

$\endgroup$

You must log in to answer this question.

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