[Pw_forum] Doubt in workfunction example for Al100
Bipul Rakshit
bipulrr at gmail.com
Fri Apr 10 08:10:28 CEST 2015
Dear Users,
I am doing Au slab with ZnO Slab. My calculation contain 6 layers of Au and
12 Double Layer of ZnO. I run the "relax" calculation and it gives the
optimized atomic positions. Using that position when i run the "scf" run by
applying the dipole correction, then that calculation is not converging.
Following is the input of the "scf"
&CONTROL
calculation = 'scf' ,
restart_mode = 'from_scratch' ,
outdir = './OUT',
pseudo_dir = './',
prefix = 'Au-ZnO-NW',
etot_conv_thr = 2.0d-5 ,
forc_conv_thr = 1.0d-3 ,
tstress = .true. ,
tprnfor = .true. ,
wf_collect = .true.,
tefield=.true.,
dipfield=.true.,
/
&SYSTEM
ibrav = 4,
celldm(1) = 6.1413449010,
celldm(3) = 15.8461538461,
nat = 30,
ntyp = 3,
ecutwfc = 56,
ecutrho = 448,
occupations ='smearing',
smearing ='m-v',
degauss =0.03,
lda_plus_u = .true.,
Hubbard_U(1) =8.0, Hubbard_U(2)=0.0,
Hubbard_U(3) =0.0,
edir=3,
emaxpos=0.55,
eopreg=0.06,
eamp=0,
/
&ELECTRONS
mixing_beta = 0.4 ,
diagonalization = 'david' ,
diago_thr_init = 1.D-2 ,
diago_david_ndim = 10,
conv_thr = 1.0d-6,
electron_maxstep = 140,
/
&IONS
ion_dynamics ='bfgs',
bfgs_ndim = 3,
/
&CELL
cell_dynamics = 'bfgs' ,
press = 0.00,
wmass = 0.010
/
ATOMIC_SPECIES
Zn 65.39 Zn.pbe-van.UPF
O 15.9994 O.pbe-van_ak.UPF
Au 196.97 Au.pbe-van_ak.UPF
ATOMIC_POSITIONS (Angstrom)
Zn 7.499742896 5.752629371 3.230876064
Zn 5.874757485 6.690733893 5.921325325
Zn 7.499651212 5.752560151 8.611774586
Zn 5.874685478 6.690665275 11.302223847
Zn 7.499814131 5.752654773 13.985634015
Zn 5.874963177 6.690829207 16.671576079
Zn 7.499874447 5.752675885 19.356809802
Zn 5.874856769 6.690780480 22.041789006
Zn 7.499724155 5.752620086 24.729390916
Zn 5.874792904 6.690719489 27.417963638
Zn 7.499686920 5.752565718 30.107473986
Zn 5.874692982 6.690668726 32.896386195
O 5.874814089 6.690782677 2.514551440
O 7.499685927 5.752583792 5.205000701
O 5.874718327 6.690713410 7.895449962
O 7.499618111 5.752518096 10.585899223
O 5.874878823 6.690804905 13.310596326
O 7.499890469 5.752674934 15.997050233
O 5.874943952 6.690827750 18.682493540
O 7.499787626 5.752628781 21.367966551
O 5.874797778 6.690771718 24.053807177
O 7.499721925 5.752568423 26.742414673
O 5.874754539 6.690718084 29.433188626
O 7.499626837 5.752521379 32.126332530
Au 7.499703856 5.752469031 35.061709956
Au 9.124636220 6.690621085 37.151452057
Au 7.499709255 7.628772145 39.211201531
Au 7.499712252 5.752462483 41.258545283
Au 9.124644992 6.690614771 43.331292891
Au 7.499717550 7.628769078 45.387107036
K_POINTS automatic
6 6 1 1 1 1
Following is the few lines of the output file
total cpu time spent up to now is 48274.1 secs
total energy = -459.60706118 Ry
Harris-Foulkes estimate = -9029.54246705 Ry
estimated scf accuracy < 645684.43872681 Ry
iteration #105 ecut= 56.00 Ry beta=0.40
Davidson diagonalization with overlap
ethr = 1.00E-02, avg # of iterations = 14.7
negative rho (up, down): 1.422E-04 0.000E+00
Adding external electric field
Computed dipole along edir(3) :
Dipole 12.1613 Ry au, 30.9110 Debye
Dipole field 0.0481 Ry au
Potential amp. -8.7961 Ry
Total length 91.4777 bohr
total cpu time spent up to now is 48690.9 secs
total energy = -1539.58421411 Ry
Harris-Foulkes estimate = -6830.90427333 Ry
estimated scf accuracy < 340940.96928352 Ry
Please help me in this regard
On Mon, Mar 30, 2015 at 4:27 PM, Giovanni Cantele <
giovanni.cantele at spin.cnr.it> wrote:
> Provided that you have a sufficient number of layers to guarantee in both
> sides a “bulk-like” region, you can choose the innermost layers.
>
> But, because you have an interface, what is the quantity are you
> interested in? The interface barrier (in your case the Schottky barrier,
> because you have a metal-semiconductor
> junction) or the work function? Consider that, in the latter case, because
> the two sides of the slab are different:
> i) the vacuum level is different at the two sides, so you just choose the
> interplanar distance of the material you are interested in (either ZnO or
> Au)
> ii) because the vacuum level is different at the two sides, you must
> consider to add a correction for a “spurious” dipole introduced by boundary
> conditions, using tefield=.true.
> dipfield=.true., and the related variables (see
> PW/DOC/INPUT_PW.txt for explanation) edit, eamp=0.D0, eopreg and examples
>
> In the case you are instead interested in the interface barrier, you
> easily figure out that by choosing, as the window for the macroscopic
> average either the ZnO or the Au interplanar distance you get rid of
> microscopic oscillations only on one side. In this case, you can compute
> the averages with both windows, and plot them on the same plot. If needed,
> you might want to have a look at Fig. 6 of N.R. D’amico, et al, J. Phys.:
> Condens. Matter 27 (2015) 015006
>
> Hope this helps,
>
> Giovanni
>
> PS Warning: the ZnO orientation you are studying is polar, so the
> macroscopic average does not give you a constant value of the potential in
> the innermost part of the ZnO slab,
> but a linearly varying one. There are examples in the literature on how
> the handle the electrostatic potential in polar slabs, see references in
> the above mentioned paper.
>
>
>
> On 30 Mar 2015, at 11:47, Bipul Rakshit <bipulrr at gmail.com> wrote:
>
> Really thanks for your suggestion Giovanni,
> I just have few more doubts that if i want to find the work function of
> Au-ZnO Slab. I have created Au(111) and ZnO(0001) Slab. So it has an
> interface of Zn-Au and a vacuum of 10 Angstrom. I took 6 layers of Au and 4
> double layers of ZnO. Then how to proceed.
> Means
> 1) Which two consecutive atomic planes I have to choose Au-Au, Au-Zn or
> the Zn-O planes. for the microscopic average calculation?
>
> 2) Also in-order to choose the inner part of the slab, so that part is
> inside the Au-Slab or the ZnO Slab?
>
> regards
>
>
> On Wed, Mar 18, 2015 at 5:02 PM, Giovanni Cantele <
> giovanni.cantele at spin.cnr.it> wrote:
>
>> The average electrostatic potential you are calculating/plotting does
>> show microscopic oscillations, revealing the atomic planes
>> (plot the 2d column of reference/Al100.avg.out as a function of the 1st
>> one).
>>
>> The calculation of the work function requires a “constant” energy level
>> to compare the bulk and the slab calculation. For this purpose,
>> you run a macroscopic average of the x-y averaged electrostatic
>> potential. The window you choose for the macroscopic average
>> is just the distance between two consecutive atomic planes (in a.u.).
>> Because the input positions are in alat units, you get
>>
>> (2.8284271247461898 - 2.1213203435596428) * 5.4235090117 = 3.835
>>
>> In this way, microscopic oscillations with period 3.835 are averaged and
>> a constant level (in the inner region of the slab) is obtained.
>>
>> Concerning 17.8087, in order to get this constant value, you can choose
>> any position in the inner part of the slab where the
>> macroscopic average does not show significant variations.
>>
>> Giovanni
>>
>>
>> On 18 Mar 2015, at 10:49, Bipul Rakshit <bipulrr at gmail.com> wrote:
>>
>> In espresso, there is an example to find the workfunction of Al. In the
>> run_example, the input for the the macroscopic average is the following
>> cat > Al100.avg.in <http://al100.avg.in/> <<EOF
>> 1
>> Al100.pot
>> 1.D0
>> 1440
>> 3
>> 3.835000000
>> EOF
>>
>> In this file the quantity "3.835" i saw in average.f90 as "awin !
>> the size of the window for macroscopic averages"
>>
>> So my doubt is how we can choose this no. Can we get the information from
>> another file prerun file, like Al100.pot, or something else.
>>
>> Also in run_example there is another quantity vSlab
>>
>> vSlab=`grep "17.8087" Al100.avg.out | cut -d \ -f 10`
>>
>> So how the value correspond to "17.8087" is assign as vSlab?
>>
>> Kindly help me in this matter.
>>
>> regards
>> --
>> Dr. Bipul Rakshit
>> Research Associate,
>> Institute of Physics (IOP),
>> Bhubaneswar- 751 005
>> Orissa
>> India
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>>
>>
>> --
>>
>> Giovanni Cantele, PhD
>> CNR-SPIN
>> c/o Dipartimento di Fisica
>> Universita' di Napoli "Federico II"
>> Complesso Universitario M. S. Angelo - Ed. 6
>> Via Cintia, I-80126, Napoli, Italy
>> e-mail: giovanni.cantele at spin.cnr.it
>> Phone: +39 081 676910
>> Skype contact: giocan74
>>
>> ResearcherID: http://www.researcherid.com/rid/A-1951-2009
>> Web page: http://people.na.infn.it/~cantele
>>
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>>
>
>
>
> --
> Dr. Bipul Rakshit
> Research Associate,
> Institute of Physics (IOP),
> Bhubaneswar- 751 005
> Orissa
> India
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>
>
> --
>
> Giovanni Cantele, PhD
> CNR-SPIN
> c/o Dipartimento di Fisica
> Universita' di Napoli "Federico II"
> Complesso Universitario M. S. Angelo - Ed. 6
> Via Cintia, I-80126, Napoli, Italy
> e-mail: giovanni.cantele at spin.cnr.it
> Phone: +39 081 676910
> Skype contact: giocan74
>
> ResearcherID: http://www.researcherid.com/rid/A-1951-2009
> Web page: http://people.na.infn.it/~cantele
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>
--
Dr. Bipul Rakshit
Research Associate,
Institute of Physics (IOP),
Bhubaneswar- 751 005
Orissa
India
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150410/6d2312b2/attachment.html>
More information about the users
mailing list