http://micro.stanford.edu/mediawiki/api.php?action=feedcontributions&feedformat=atom&user=Shryu
Micro and Nano Mechanics Group - User contributions [en]
2026-07-05T11:35:50Z
User contributions
MediaWiki 1.39.7
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=5085
Seunghwa Ryu
2011-02-20T07:33:29Z
<p>Shryu: </p>
<hr />
<div>* Title<br />
** Physics Phd Candiate<br />
* Location <br />
**496 Lomita Mall (a.k.a. Durand Building), rm 224, Stanford, CA94305 <br />
* Email<br />
** shryu at stanford dot edu<br />
* Phone<br />
** 650-353-1277<br />
* Reserach Interest<br />
** Nanowire Growth Simulation (Thesis Topic)<br />
** Quantum Entanglement Computation (side work)<br />
* Education<br />
** 09/2004 - current : PhD in physics, Stanford <br />
** 09/2004 - 01/2006 : MS. in physics, Stanford <br />
** 03/2000 - 02/2004 : BS. in Physics, KAIST, Korea<br />
* CV & Homepage<br />
** https://sites.google.com/site/seunghwaweb/</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=5083
Seunghwa Ryu
2011-02-09T13:45:26Z
<p>Shryu: </p>
<hr />
<div>* Title<br />
** Physics Phd Candiate<br />
* Location <br />
**496 Lomita Mall (a.k.a. Durand Building), rm 224, Stanford, CA94305 <br />
* Email<br />
** shryu at stanford dot edu<br />
* Phone<br />
** 650-353-1277<br />
* Reserach Interest<br />
** Nanowire Growth Simulation (Thesis Topic)<br />
** Quantum Entanglement Computation (side work)<br />
* Education<br />
** 09/2004 - current : PhD in physics, Stanford <br />
** 09/2004 - 01/2006 : MS. in physics, Stanford <br />
** 03/2000 - 02/2004 : BS. in Physics, KAIST, Korea<br />
* CV<br />
** https://sites.google.com/site/seunghwaweb/</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=5082
Seunghwa Ryu
2011-02-09T13:44:49Z
<p>Shryu: </p>
<hr />
<div>* Title<br />
** Physics Phd Candiate<br />
* Location <br />
**496 Lomita Mall (a.k.a. Durand Building), rm 202, Stanford, CA94305 <br />
* Email<br />
** shryu at stanford dot edu<br />
* Phone<br />
** 650-353-1277<br />
* Reserach Interest<br />
** Nanowire Growth Simulation (Thesis Topic)<br />
** Quantum Entanglement Computation (side work)<br />
* Education<br />
** 09/2004 - current : PhD in physics, Stanford <br />
** 09/2004 - 01/2006 : MS. in physics, Stanford <br />
** 03/2000 - 02/2004 : BS. in Physics, KAIST, Korea<br />
* CV<br />
** https://sites.google.com/site/seunghwaweb/</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Research_Meeting_Schedule&diff=2256
Research Meeting Schedule
2009-09-24T01:46:35Z
<p>Shryu: /* Autumn 2009 */</p>
<hr />
<div>===Autumn 2009===<br />
<br />
Prof. Wei Cai's weekly schedule.<br />
<br />
{| class="wikitable" style="text-align:center;"<br />
!width="150pt" | Time <br />
!width="200pt" | Monday <br />
!width="200pt" | Tuesday <br />
!width="200pt" | Wednesday <br />
!width="200pt" | Thursday <br />
!width="220pt" | Friday<br />
|-<br />
|9:00-10:00 || ''work'' || ''work'' || ''work'' || ''work'' || ''work''<br />
|-<br />
|10:00-10:30|| lecture prep || (meeting) || lecture prep || (meeting) <br />
| rowspan="2" | '''ME80 office hour'''<br />
|-<br />
|10:30-11:00|| lecture prep || rowspan="2" | '''Group Meeting''' || lecture prep || (meeting)<br />
|-<br />
|11:00-12:00|| bi-weekly Eunseok,<br> Hark, Prof. Prinz|| ''work'' || ''work'' || ''work''<br />
|-<br />
|12:30-1:15 || colspan="5" align="center" bgcolor="lightgrey" | lunch break<br />
|-<br />
|1:30-2:05 || rowspan="3" | '''ME80''' || rowspan="2" | '''ME80 office hour''' <br />
| rowspan="3" | '''ME80''' || (meeting) || ''work''<br />
|-<br />
|2:10-2:45 || (meeting) || ''work''<br />
|-<br />
|2:50-3:25 || (meeting) || (meeting) || ''work''<br />
|-<br />
|3:30-4:00 || colspan="5" align="center" bgcolor="lightgrey" | coffee break <br />
|-<br />
|4:00-4:35 || Seunghwa Ryu || (meeting) || (meeting) || rowspan="2" | '''MC seminar''' || ''work''<br />
|-<br />
|4:40-6:00 || ''work'' || ''work'' || ''work'' || ''work''<br />
|}<br />
<br />
Meetings to sign-up<br />
<br />
{| class="wikitable" style="text-align:center;"<br />
|-<br />
| width="120pt" | Eunseok<br />
| width="120pt" | ILL Ryu<br />
| width="120pt" | Seokwoo<br />
| width="120pt" | Haneesh<br />
| width="120pt" | Jie<br />
| width="120pt" | Sylvie<br />
| width="120pt" | Keonwook <br />
| width="120pt" | Billy<br />
|-<br />
| colspan="2" | Hark (bi-weekly)<br />
| colspan="3" | Jie, Prof. Barnett (bi-weekly)<br />
| colspan="2" | Ibrahim (bi-weekly)<br />
| colspan="2" | Seminar speaker <br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=VASP_Computing_Generalized_Stacking_Fault_Energy_of_Au&diff=2408
VASP Computing Generalized Stacking Fault Energy of Au
2009-02-15T10:30:49Z
<p>Shryu: /* Relaxed generalized stacking fault energy */</p>
<hr />
<div><P ALIGN="CENTER"><br />
<FONT SIZE="+3" color="darkred"><STRONG><br />
VASP: Generalized Stacking Fault Energy of Au</STRONG></font></P><br />
<DIV><br />
<P ALIGN="CENTER"><STRONG> </STRONG></P><br />
</DIV><br />
<br />
<br />
==Input files==<br />
<br />
Here are the basic input files required for VASP calculation. Some of the files need to be changed since we need to perform a large number of calculations.<br />
<br />
===INCAR===<br />
<br />
<pre><br />
System = fcc Au<br />
LWAVE = .FALSE.<br />
ENCUT = 400<br />
ISMEAR = 1<br />
SIGMA = 0.1<br />
ISIF = 2<br />
</pre><br />
<br />
To increase plane wave cutoff, we manually put PREC = HIGH or PREC = Accurate.<br />
Default value is PREC = Medium<br />
<br />
===KPOINTS===<br />
<br />
<pre><br />
test convergence for 3-layer simulation cell<br />
0 0 = automatic generation of k-points<br />
Monkhorst<br />
11 9 19<br />
0 0 0<br />
</pre><br />
<br />
<pre><br />
test convergence for 6-layer simulation cell<br />
0 0 = automatic generation of k-points<br />
Monkhorst<br />
11 5 19<br />
0 0 0<br />
</pre><br />
<br />
As the y-axis dimension is doubled, we can use half K-points in y direction.<br />
<br />
===POSCAR===<br />
<br />
<pre> <br />
POSCAR for FCC Au (created manually) for 3-Layer Cell<br />
4.0605<br />
1.22474487139159 0 0<br />
0 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
You also need to put the LDA pseudopotential file as <tt>POTCAR</tt> in this directory.<br />
<br />
<pre><br />
POSCAR for FCC Au for 6 Layer Cell<br />
4.0625<br />
1.22474487139159 0 0<br />
0 3.46410161513775 0<br />
0 0 0.70710678118655<br />
12<br />
Cartesian (real coordinates r)<br />
0.00000000000000 0.00000000000000 0.00000000000000<br />
0.61237243569579 0.00000000000000 0.35355339059327<br />
0.40824829046386 0.57735026918963 0.00000000000000<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.81649658092773 1.15470053837925 0.00000000000000<br />
1.42886901662352 1.15470053837925 0.35355339059327<br />
1.22474487139159 1.73205080756888 0.00000000000000<br />
1.83711730708738 1.73205080756888 0.35355339059327<br />
1.63299316185545 2.30940107675850 0.00000000000000<br />
2.24536559755125 2.30940107675850 0.35355339059327<br />
2.04124145231931 2.88675134594813 0.00000000000000<br />
2.65361388801511 2.88675134594813 0.35355339059327<br />
</pre><br />
<br />
The height of cell is doubled compared to that of above.<br />
As different K-Points are sampled, it end up having slightly different<br />
equilibrium lattice constant.<br />
<br />
==Results==<br />
<br />
===Perfect crystal===<br />
<br />
The following table shows energy convergence with k-points. (Reference value from Au_bulk calculation: E = -4.39 eV/atom.)<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|-<br />
! width="100pt" | KPOINTS<br />
! width="100pt" | E (eV/atom)<br />
! optimal number of CPUs<br />
! computational time (second)<br />
|-<br />
| 3x3x3 || -4.4735468 || -- || --<br />
|-<br />
| 5x5x5 || -4.4121563 || -- || --<br />
|-<br />
| 7x7x7 || -4.4043722 || -- || --<br />
|-<br />
| 7x7x11 || -4.3981093 || 16 || 341<br />
|-<br />
| 7x5x13 || -4.3974777 || 16 || 291<br />
|-<br />
| 9x7x15 || -4.395454 || -- || ---<br />
|-<br />
| 11x9x19 || -4.394620 || 8 || 580 (with 3 Layer cell)<br />
|-<br />
| 11x5x19 || -4.394463 || 8 || 1339 (with 6 Layer cell)<br />
|}<br />
<br />
For the cell with small number of atoms (N~10), the intercommunication<br />
between nodes become the limiting-factor for the computation.<br />
In this example, even with much more K-Points, 8 CPU job is finished<br />
faster compared to 16 CPU job with less K-Points.<br />
<br />
===Unrelaxed stacking fault energy===<br />
<br />
For this example, we fix KPOINTS at <tt>11x9x19</tt> and use 3 Layer Cell, for which the energy of perfect crystal is <tt>E0 = -4.394620 (eV/atom)</tt>.<br />
<br />
To create a stacking fault, we shift the repeat vector '''b''' in the x-direction by sqrt(6)/6. The new POSCAR file is (notice the change in 3rd line).<br />
<br />
<pre><br />
POSCAR for FCC Au (created manually)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Unrelaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (raw data) || -26.338784 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.394629 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.028938 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.00202665 (eV/angstrom^2) = 32.487 (mJ/m^2)<br />
|}<br />
<br />
The result can be compared with previous calculations Esf = 59 (mJ/m^2) (Rosengaard and Skriver, PRB, 47, 12865, 1993) and experimental value of Esf = 42 (mJ/m^2) (Devlin, J. Phys. F: Metal Phys. 4, 1865, 1974).<br />
<br />
===Relaxed stacking fault energy===<br />
<br />
We now displace the y-component of the repeat vector '''b''' and find the minimum energy. To do so we need to submit the following PBS script.<br />
<br />
relax.isf.pbs<br />
<br />
<pre><br />
...<br />
ncpu=`cat $PBS_NODEFILE | wc -w`<br />
echo "Number of processors = $ncpu "<br />
...<br />
./auto.relax.isf.par $ncpu<br />
...<br />
</pre><br />
<br />
which calls the following shell script.<br />
<br />
auto.relax.isf.par<br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
for by in 1.730 1.735 1.740 1.745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.by.dat<br />
<br />
done<br />
</pre><br />
<br />
<br />
Here are the energy values at different <math>b_y</math>.<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Energy minimization<br />
|-<br />
| width="200pt" | <math>b_y / a_0</math> <br />
| width="250pt" | E1 (eV)<br />
|-<br />
| 1.730 || -26.341633<br />
|-<br />
| 1.735 || -26.345601<br />
|-<br />
| 1.740 || -26.345699<br />
|-<br />
| 1.745 || -26.342276<br />
|}<br />
<br />
The following Matlab script finds the mininum value of E1 = -26.3461(eV) (at <math>b_y / a_0</math> = 1.7378).<br />
<pre><br />
data=[<br />
1.730 1 -.26341633E+02 -.26341112E+02 -.156102E-02<br />
1.735 1 -.26345601E+02 -.26345033E+02 -.170179E-02<br />
1.740 1 -.26345699E+02 -.26345097E+02 -.180565E-02<br />
1.745 1 -.26342276E+02 -.26341660E+02 -.184888E-02<br />
];<br />
x = [min(data(:,1)):0.0001:max(data(:,1))];<br />
[p,s] = polyfit(data(:,1),data(:,3),2);<br />
y = polyval(p,x);<br />
[y_min ind] = min(y); <br />
x_min = x(ind);<br />
figure(1); plot(data(1:4,1),data(1:4,3),'o', x,y,'-', x_min,y_min,'r*');<br />
y_min, x_min<br />
</pre><br />
<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Relaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (min) || -26.3461 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.3974777 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.0387662 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.002715 (eV/angstrom^2) = 43.49 (mJ/m^2)<br />
|}<br />
<br />
Number of CPUs = 16. Total computation time = 20 minutes.<br />
<br />
===Unrelaxed generalized stacking fault energy===<br />
<br />
The unrelaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''' (while keeping y-component of '''b''' fixed). This is done using the following shell script.<br />
<br />
auto.unrelax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
by=1.73205080756888<br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 2 hours.<br />
<br />
The results can be plotted using the following Matlab script.<br />
<br />
<pre><br />
data2=[<br />
0.00 1.73205080756888 1 -.26384866E+02 -.26384852E+02 -.419247E-04<br />
0.05 1.73205080756888 1 -.26366785E+02 -.26366967E+02 0.547795E-03<br />
0.10 1.73205080756888 1 -.26327184E+02 -.26327283E+02 0.294942E-03<br />
0.15 1.73205080756888 1 -.26288320E+02 -.26288001E+02 -.957691E-03<br />
0.20 1.73205080756888 1 -.26269732E+02 -.26269867E+02 0.405146E-03<br />
0.25 1.73205080756888 1 -.26274866E+02 -.26274925E+02 0.176062E-03<br />
0.30 1.73205080756888 1 -.26294139E+02 -.26293966E+02 -.517175E-03<br />
0.35 1.73205080756888 1 -.26323923E+02 -.26323824E+02 -.295196E-03<br />
0.408248 1.73205080756888 1 -.26343752E+02 -.26343211E+02 -.162154E-02<br />
0.45 1.73205080756888 1 -.26331580E+02 -.26331671E+02 0.272932E-03<br />
0.50 1.73205080756888 1 -.26252512E+02 -.26252146E+02 -.109923E-02<br />
0.55 1.73205080756888 1 -.26085728E+02 -.26085280E+02 -.134254E-02<br />
0.60 1.73205080756888 1 -.25829191E+02 -.25828980E+02 -.633646E-03<br />
0.65 1.73205080756888 1 -.25506381E+02 -.25506199E+02 -.546431E-03<br />
0.70 1.73205080756888 1 -.25167205E+02 -.25167286E+02 0.242977E-03<br />
0.75 1.73205080756888 1 -.24887268E+02 -.24887237E+02 -.929289E-04<br />
0.80 1.73205080756888 1 -.24745853E+02 -.24745735E+02 -.354033E-03<br />
0.85 1.73205080756888 1 -.24780442E+02 -.24780426E+02 -.490752E-04<br />
0.90 1.73205080756888 1 -.24974975E+02 -.24974773E+02 -.604431E-03<br />
0.95 1.73205080756888 1 -.25276553E+02 -.25276421E+02 -.395463E-03<br />
1.00 1.73205080756888 1 -.25615702E+02 -.25615797E+02 0.282608E-03<br />
1.05 1.73205080756888 1 -.25927655E+02 -.25927281E+02 -.112246E-02<br />
1.10 1.73205080756888 1 -.26168023E+02 -.26167870E+02 -.459552E-03<br />
1.15 1.73205080756888 1 -.26319686E+02 -.26319608E+02 -.233191E-03<br />
1.20 1.73205080756888 1 -.26387386E+02 -.26387561E+02 0.525098E-03<br />
1.224745 1.73205080756888 1 -.26395070E+02 -.26395345E+02 0.826715E-03<br />
];<br />
bx_unrlx = data2(:,1); E1 = data2(:,4);<br />
Egsf_unrlx = (E1 - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
figure(2); plot(bx_unrlx,max(Egsf_unrlx,0),'.', x_unrlx,y_unrlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{unrlx} (mJ/m^2)');<br />
xlim([0 max(bx_unrlx)]); ylim([0 2000]);<br />
% print out results<br />
nx = length(bx_unrlx);<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_unrlx(i), Egsf_unrlx(i))); end<br />
</pre><br />
<br />
===Relaxed generalized stacking fault energy===<br />
<br />
The relaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''', while allowing the y-component of '''b''' to relax. This means for every value of <math>b_x</math> we will need to a set of calculations with different values of <math>b_y</math>. This is done using the following shell script. We need to be careful that for every <math>b_x</math> the miniminum energy is sampled within the range of <math>b_y</math>. For this purpose, we also did two separate calculations with when <math>b_x</math> = 0.60:0.05:1.20, <math>b_y</math> = 1.775:0.005:1.815, and <math>b_x</math> = 0.75:0.05:1.20, <math>b_y</math> = 1.800:0.005:1.840.<br />
<br />
auto.relax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
<br />
for by in 1.730 1.735 1.740 1.745 1.750 1.755 1.760 1.765 1.770<br />
do<br />
<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 19 hours.<br />
<br />
The following Matlab script extracts the minimized energy and plots the relaxed GSF curve.<br />
<br />
<pre><br />
data3 = load('E.bxby.dat'); <br />
N = length(data3(:,1)); ny = 9; nx = floor(N/ny);<br />
bx_rlx = data3(1:ny:nx*ny,1); by = data3(1:ny,2);<br />
E1 = reshape(data3(1:nx*ny,4),ny,nx);<br />
figure(3); plot(E1,'.-');<br />
E1_min = zeros(1,nx);<br />
for i=1:nx,<br />
E1_min(i) = min(spline(by,E1(:,i),[min(by):0.0001:max(by)]));<br />
end<br />
Egsf_rlx = (E1_min - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
figure(4); plot(bx_rlx,max(Egsf_rlx,0),'o', x_rlx,y_rlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{rlx} (mJ/m^2)');<br />
% print out results<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_rlx(i), Egsf_rlx(i))); end<br />
</pre><br />
<br />
<br />
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - <br />
<br />
For comparison, 6 layer result is presented also.<br />
With following<br />
<br />
auto.relax.gsf.par<br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
<br />
for by in 3.455 3.465 3.475 3.485 3.495 3.505 3.515 3.525 3.535 3.545 3.565 3.585 3.605 3.625<br />
do<br />
<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au<br />
4.0625<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
12<br />
Cartesian (real coordinates r)<br />
0.00000000000000 0.00000000000000 0.00000000000000<br />
0.61237243569579 0.00000000000000 0.35355339059327<br />
0.40824829046386 0.57735026918963 0.00000000000000<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.81649658092773 1.15470053837925 0.00000000000000<br />
1.42886901662352 1.15470053837925 0.35355339059327<br />
1.22474487139159 1.73205080756888 0.00000000000000<br />
1.83711730708738 1.73205080756888 0.35355339059327<br />
1.63299316185545 2.30940107675850 0.00000000000000<br />
2.24536559755125 2.30940107675850 0.35355339059327<br />
2.04124145231931 2.88675134594813 0.00000000000000<br />
2.65361388801511 2.88675134594813 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
/opt/mpiexec/bin/mpiexec --comm=pmi -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
<br />
done<br />
</pre><br />
<br />
Note that, now, the y axis expansion doubled (from <math>b_y</math> = 1.775:0.005:1.815<br />
to <math>b_y</math> = 3.455:0.01:3.625 due to doubled cell height.<br />
It is important to have wide enough range for <math>b_y</math> to allow a full relaxation.<br />
<br />
And following matlab script extract the minimized energy.<br />
<pre><br />
data3 = load('E.bxby.dat'); <br />
N = length(data3(:,1)); ny = 14; nx = floor(N/ny);<br />
bx_rlx = data3(1:ny:nx*ny,1); by = data3(1:ny,2);<br />
E1 = reshape(data3(1:nx*ny,4),ny,nx);<br />
figure(3); plot(E1,'.-');<br />
E1_min = zeros(1,nx);<br />
for i=1:nx,<br />
E1_min(i) = min(spline(by,E1(:,i),[min(by):0.0001:max(by)]));<br />
end<br />
Egsf_rlx = (E1_min - (-52.733560)) / 14.2928 * 1.602e-19*1e20 * 1e3;<br />
Egsf_rlx = (E1_min - (-52.733560)) / 14.2928;<br />
pp = spline(bx_rlx, Egsf_rlx)<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
figure(4); plot(bx_rlx,max(Egsf_rlx,0),'o', x_rlx,y_rlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{rlx} (mJ/m^2)');<br />
% print out results<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_rlx(i), Egsf_rlx(i))); end<br />
x=[0:100]/100;<br />
Au_EAM_rlx=spline(bx_rlx/max(bx_rlx),Egsf_rlx,x);<br />
<br />
latticeconst=4.0625;<br />
dz=0.005;<br />
Lz=norm([ 1 -2 1])*latticeconst;<br />
tau = max(Au_EAM_rlx(2:xp)-Au_EAM_rlx(1:xp-1))/(dz*Lz) * 160.2;<br />
disp(sprintf('Ideal shear stregnth = %e(GPa)',tau));<br />
</pre><br />
<br />
===Plot final results===<br />
The following Matlab script summarizes the results and plots the relaxed and unrelaxed GSF curves together.<br />
<br />
<pre><br />
data_unrlx = [<br />
0.000000 0.000224389703<br />
0.050000 20.286175591244<br />
0.100000 64.716458956782<br />
0.150000 108.319866262614<br />
0.200000 129.174645364115<br />
0.250000 123.414561659193<br />
0.300000 101.791247821056<br />
0.350000 68.375133082561<br />
0.408248 46.128015866956<br />
0.450000 59.784373260085<br />
0.500000 148.494598889395<br />
0.550000 335.617660954438<br />
0.600000 623.438968591740<br />
0.650000 985.615170536872<br />
0.700000 1366.153181970664<br />
0.750000 1680.228084990751<br />
0.800000 1838.888435035886<br />
0.850000 1800.081357655754<br />
0.900000 1581.825346091701<br />
0.950000 1243.470355136624<br />
1.000000 862.962636312892<br />
1.050000 512.967429456257<br />
1.100000 243.286907449047<br />
1.150000 73.128828964477<br />
1.200000 -2.827085882290<br />
1.224745 -11.448138314825<br />
];<br />
<br />
data_rlx = [<br />
0.000000 -0.020670060764<br />
0.050000 19.360018794024<br />
0.100000 56.053066178278<br />
0.150000 86.173859662894<br />
0.200000 99.250871762213<br />
0.250000 96.073197595982<br />
0.300000 85.496344853148<br />
0.350000 61.842558692410<br />
0.408248 43.478576333529<br />
0.450000 54.685416520609<br />
0.500000 117.020469910803<br />
0.550000 207.709070871746<br />
0.600000 294.919590647244<br />
0.650000 369.101339702239<br />
0.700000 421.696425371603<br />
0.750000 445.021797830292<br />
0.800000 450.785540359893<br />
0.850000 441.873817315316<br />
0.900000 423.309809152175<br />
0.950000 395.882657649786<br />
1.000000 348.780068868428<br />
1.050000 276.895997333094<br />
1.100000 167.991003752519<br />
1.150000 62.813035785537<br />
1.200000 -2.921550847183<br />
1.224745 -11.456823618946<br />
];<br />
<br />
bx_unrlx = data_unrlx(:,1); Egsf_unrlx = data_unrlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_unrlx = Egsf_unrlx-Egsf_unrlx(end)/bx_unrlx(end)*bx_unrlx;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
<br />
bx_rlx = data_rlx(:,1); Egsf_rlx = data_rlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_rlx = Egsf_rlx-Egsf_rlx(end)/bx_rlx(end)*bx_rlx;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
<br />
% partial Burgers vector<br />
bp = sqrt(6)/6;<br />
figure(4); <br />
subplot(1,2,1);<br />
p41 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 max(bx_unrlx/bp)]); ylim([0 2000]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg1=legend('unrelaxed', 'relaxed', 2); set(lg1,'FontSize',10);<br />
subplot(1,2,2);<br />
p42 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 1.1]); ylim([0 200]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg2=legend('unrelaxed', 'relaxed', 2); set(lg2,'FontSize',10);<br />
</pre><br />
<br />
<br />
[[image:GSF_Au_LDA_VASP.jpg || Generalized stacking fault energy curve for LDA Au computed by VASP.]]</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=VASP_Computing_Generalized_Stacking_Fault_Energy_of_Au&diff=2407
VASP Computing Generalized Stacking Fault Energy of Au
2009-02-15T09:46:34Z
<p>Shryu: </p>
<hr />
<div><P ALIGN="CENTER"><br />
<FONT SIZE="+3" color="darkred"><STRONG><br />
VASP: Generalized Stacking Fault Energy of Au</STRONG></font></P><br />
<DIV><br />
<P ALIGN="CENTER"><STRONG> </STRONG></P><br />
</DIV><br />
<br />
<br />
==Input files==<br />
<br />
Here are the basic input files required for VASP calculation. Some of the files need to be changed since we need to perform a large number of calculations.<br />
<br />
===INCAR===<br />
<br />
<pre><br />
System = fcc Au<br />
LWAVE = .FALSE.<br />
ENCUT = 400<br />
ISMEAR = 1<br />
SIGMA = 0.1<br />
ISIF = 2<br />
</pre><br />
<br />
To increase plane wave cutoff, we manually put PREC = HIGH or PREC = Accurate.<br />
Default value is PREC = Medium<br />
<br />
===KPOINTS===<br />
<br />
<pre><br />
test convergence for 3-layer simulation cell<br />
0 0 = automatic generation of k-points<br />
Monkhorst<br />
11 9 19<br />
0 0 0<br />
</pre><br />
<br />
<pre><br />
test convergence for 6-layer simulation cell<br />
0 0 = automatic generation of k-points<br />
Monkhorst<br />
11 5 19<br />
0 0 0<br />
</pre><br />
<br />
As the y-axis dimension is doubled, we can use half K-points in y direction.<br />
<br />
===POSCAR===<br />
<br />
<pre> <br />
POSCAR for FCC Au (created manually) for 3-Layer Cell<br />
4.0605<br />
1.22474487139159 0 0<br />
0 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
You also need to put the LDA pseudopotential file as <tt>POTCAR</tt> in this directory.<br />
<br />
<pre><br />
POSCAR for FCC Au for 6 Layer Cell<br />
4.0625<br />
1.22474487139159 0 0<br />
0 3.46410161513775 0<br />
0 0 0.70710678118655<br />
12<br />
Cartesian (real coordinates r)<br />
0.00000000000000 0.00000000000000 0.00000000000000<br />
0.61237243569579 0.00000000000000 0.35355339059327<br />
0.40824829046386 0.57735026918963 0.00000000000000<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.81649658092773 1.15470053837925 0.00000000000000<br />
1.42886901662352 1.15470053837925 0.35355339059327<br />
1.22474487139159 1.73205080756888 0.00000000000000<br />
1.83711730708738 1.73205080756888 0.35355339059327<br />
1.63299316185545 2.30940107675850 0.00000000000000<br />
2.24536559755125 2.30940107675850 0.35355339059327<br />
2.04124145231931 2.88675134594813 0.00000000000000<br />
2.65361388801511 2.88675134594813 0.35355339059327<br />
</pre><br />
<br />
The height of cell is doubled compared to that of above.<br />
As different K-Points are sampled, it end up having slightly different<br />
equilibrium lattice constant.<br />
<br />
==Results==<br />
<br />
===Perfect crystal===<br />
<br />
The following table shows energy convergence with k-points. (Reference value from Au_bulk calculation: E = -4.39 eV/atom.)<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|-<br />
! width="100pt" | KPOINTS<br />
! width="100pt" | E (eV/atom)<br />
! optimal number of CPUs<br />
! computational time (second)<br />
|-<br />
| 3x3x3 || -4.4735468 || -- || --<br />
|-<br />
| 5x5x5 || -4.4121563 || -- || --<br />
|-<br />
| 7x7x7 || -4.4043722 || -- || --<br />
|-<br />
| 7x7x11 || -4.3981093 || 16 || 341<br />
|-<br />
| 7x5x13 || -4.3974777 || 16 || 291<br />
|-<br />
| 9x7x15 || -4.395454 || -- || ---<br />
|-<br />
| 11x9x19 || -4.394620 || 8 || 580 (with 3 Layer cell)<br />
|-<br />
| 11x5x19 || -4.394463 || 8 || 1339 (with 6 Layer cell)<br />
|}<br />
<br />
For the cell with small number of atoms (N~10), the intercommunication<br />
between nodes become the limiting-factor for the computation.<br />
In this example, even with much more K-Points, 8 CPU job is finished<br />
faster compared to 16 CPU job with less K-Points.<br />
<br />
===Unrelaxed stacking fault energy===<br />
<br />
For this example, we fix KPOINTS at <tt>11x9x19</tt> and use 3 Layer Cell, for which the energy of perfect crystal is <tt>E0 = -4.394620 (eV/atom)</tt>.<br />
<br />
To create a stacking fault, we shift the repeat vector '''b''' in the x-direction by sqrt(6)/6. The new POSCAR file is (notice the change in 3rd line).<br />
<br />
<pre><br />
POSCAR for FCC Au (created manually)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Unrelaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (raw data) || -26.338784 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.394629 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.028938 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.00202665 (eV/angstrom^2) = 32.487 (mJ/m^2)<br />
|}<br />
<br />
The result can be compared with previous calculations Esf = 59 (mJ/m^2) (Rosengaard and Skriver, PRB, 47, 12865, 1993) and experimental value of Esf = 42 (mJ/m^2) (Devlin, J. Phys. F: Metal Phys. 4, 1865, 1974).<br />
<br />
===Relaxed stacking fault energy===<br />
<br />
We now displace the y-component of the repeat vector '''b''' and find the minimum energy. To do so we need to submit the following PBS script.<br />
<br />
relax.isf.pbs<br />
<br />
<pre><br />
...<br />
ncpu=`cat $PBS_NODEFILE | wc -w`<br />
echo "Number of processors = $ncpu "<br />
...<br />
./auto.relax.isf.par $ncpu<br />
...<br />
</pre><br />
<br />
which calls the following shell script.<br />
<br />
auto.relax.isf.par<br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
for by in 1.730 1.735 1.740 1.745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.by.dat<br />
<br />
done<br />
</pre><br />
<br />
<br />
Here are the energy values at different <math>b_y</math>.<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Energy minimization<br />
|-<br />
| width="200pt" | <math>b_y / a_0</math> <br />
| width="250pt" | E1 (eV)<br />
|-<br />
| 1.730 || -26.341633<br />
|-<br />
| 1.735 || -26.345601<br />
|-<br />
| 1.740 || -26.345699<br />
|-<br />
| 1.745 || -26.342276<br />
|}<br />
<br />
The following Matlab script finds the mininum value of E1 = -26.3461(eV) (at <math>b_y / a_0</math> = 1.7378).<br />
<pre><br />
data=[<br />
1.730 1 -.26341633E+02 -.26341112E+02 -.156102E-02<br />
1.735 1 -.26345601E+02 -.26345033E+02 -.170179E-02<br />
1.740 1 -.26345699E+02 -.26345097E+02 -.180565E-02<br />
1.745 1 -.26342276E+02 -.26341660E+02 -.184888E-02<br />
];<br />
x = [min(data(:,1)):0.0001:max(data(:,1))];<br />
[p,s] = polyfit(data(:,1),data(:,3),2);<br />
y = polyval(p,x);<br />
[y_min ind] = min(y); <br />
x_min = x(ind);<br />
figure(1); plot(data(1:4,1),data(1:4,3),'o', x,y,'-', x_min,y_min,'r*');<br />
y_min, x_min<br />
</pre><br />
<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Relaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (min) || -26.3461 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.3974777 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.0387662 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.002715 (eV/angstrom^2) = 43.49 (mJ/m^2)<br />
|}<br />
<br />
Number of CPUs = 16. Total computation time = 20 minutes.<br />
<br />
===Unrelaxed generalized stacking fault energy===<br />
<br />
The unrelaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''' (while keeping y-component of '''b''' fixed). This is done using the following shell script.<br />
<br />
auto.unrelax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
by=1.73205080756888<br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 2 hours.<br />
<br />
The results can be plotted using the following Matlab script.<br />
<br />
<pre><br />
data2=[<br />
0.00 1.73205080756888 1 -.26384866E+02 -.26384852E+02 -.419247E-04<br />
0.05 1.73205080756888 1 -.26366785E+02 -.26366967E+02 0.547795E-03<br />
0.10 1.73205080756888 1 -.26327184E+02 -.26327283E+02 0.294942E-03<br />
0.15 1.73205080756888 1 -.26288320E+02 -.26288001E+02 -.957691E-03<br />
0.20 1.73205080756888 1 -.26269732E+02 -.26269867E+02 0.405146E-03<br />
0.25 1.73205080756888 1 -.26274866E+02 -.26274925E+02 0.176062E-03<br />
0.30 1.73205080756888 1 -.26294139E+02 -.26293966E+02 -.517175E-03<br />
0.35 1.73205080756888 1 -.26323923E+02 -.26323824E+02 -.295196E-03<br />
0.408248 1.73205080756888 1 -.26343752E+02 -.26343211E+02 -.162154E-02<br />
0.45 1.73205080756888 1 -.26331580E+02 -.26331671E+02 0.272932E-03<br />
0.50 1.73205080756888 1 -.26252512E+02 -.26252146E+02 -.109923E-02<br />
0.55 1.73205080756888 1 -.26085728E+02 -.26085280E+02 -.134254E-02<br />
0.60 1.73205080756888 1 -.25829191E+02 -.25828980E+02 -.633646E-03<br />
0.65 1.73205080756888 1 -.25506381E+02 -.25506199E+02 -.546431E-03<br />
0.70 1.73205080756888 1 -.25167205E+02 -.25167286E+02 0.242977E-03<br />
0.75 1.73205080756888 1 -.24887268E+02 -.24887237E+02 -.929289E-04<br />
0.80 1.73205080756888 1 -.24745853E+02 -.24745735E+02 -.354033E-03<br />
0.85 1.73205080756888 1 -.24780442E+02 -.24780426E+02 -.490752E-04<br />
0.90 1.73205080756888 1 -.24974975E+02 -.24974773E+02 -.604431E-03<br />
0.95 1.73205080756888 1 -.25276553E+02 -.25276421E+02 -.395463E-03<br />
1.00 1.73205080756888 1 -.25615702E+02 -.25615797E+02 0.282608E-03<br />
1.05 1.73205080756888 1 -.25927655E+02 -.25927281E+02 -.112246E-02<br />
1.10 1.73205080756888 1 -.26168023E+02 -.26167870E+02 -.459552E-03<br />
1.15 1.73205080756888 1 -.26319686E+02 -.26319608E+02 -.233191E-03<br />
1.20 1.73205080756888 1 -.26387386E+02 -.26387561E+02 0.525098E-03<br />
1.224745 1.73205080756888 1 -.26395070E+02 -.26395345E+02 0.826715E-03<br />
];<br />
bx_unrlx = data2(:,1); E1 = data2(:,4);<br />
Egsf_unrlx = (E1 - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
figure(2); plot(bx_unrlx,max(Egsf_unrlx,0),'.', x_unrlx,y_unrlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{unrlx} (mJ/m^2)');<br />
xlim([0 max(bx_unrlx)]); ylim([0 2000]);<br />
% print out results<br />
nx = length(bx_unrlx);<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_unrlx(i), Egsf_unrlx(i))); end<br />
</pre><br />
<br />
===Relaxed generalized stacking fault energy===<br />
<br />
The relaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''', while allowing the y-component of '''b''' to relax. This means for every value of <math>b_x</math> we will need to a set of calculations with different values of <math>b_y</math>. This is done using the following shell script. We need to be careful that for every <math>b_x</math> the miniminum energy is sampled within the range of <math>b_y</math>. For this purpose, we also did two separate calculations with when <math>b_x</math> = 0.60:0.05:1.20, <math>b_y</math> = 1.775:0.005:1.815, and <math>b_x</math> = 0.75:0.05:1.20, <math>b_y</math> = 1.800:0.005:1.840.<br />
<br />
auto.relax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
<br />
for by in 1.730 1.735 1.740 1.745 1.750 1.755 1.760 1.765 1.770<br />
do<br />
<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 19 hours.<br />
<br />
The following Matlab script extracts the minimized energy and plots the relaxed GSF curve.<br />
<br />
<pre><br />
data3 = load('E.bxby.dat'); <br />
N = length(data3(:,1)); ny = 9; nx = floor(N/ny);<br />
bx_rlx = data3(1:ny:nx*ny,1); by = data3(1:ny,2);<br />
E1 = reshape(data3(1:nx*ny,4),ny,nx);<br />
figure(3); plot(E1,'.-');<br />
E1_min = zeros(1,nx);<br />
for i=1:nx,<br />
E1_min(i) = min(spline(by,E1(:,i),[min(by):0.0001:max(by)]));<br />
end<br />
Egsf_rlx = (E1_min - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
figure(4); plot(bx_rlx,max(Egsf_rlx,0),'o', x_rlx,y_rlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{rlx} (mJ/m^2)');<br />
% print out results<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_rlx(i), Egsf_rlx(i))); end<br />
</pre><br />
<br />
===Plot final results===<br />
The following Matlab script summarizes the results and plots the relaxed and unrelaxed GSF curves together.<br />
<br />
<pre><br />
data_unrlx = [<br />
0.000000 0.000224389703<br />
0.050000 20.286175591244<br />
0.100000 64.716458956782<br />
0.150000 108.319866262614<br />
0.200000 129.174645364115<br />
0.250000 123.414561659193<br />
0.300000 101.791247821056<br />
0.350000 68.375133082561<br />
0.408248 46.128015866956<br />
0.450000 59.784373260085<br />
0.500000 148.494598889395<br />
0.550000 335.617660954438<br />
0.600000 623.438968591740<br />
0.650000 985.615170536872<br />
0.700000 1366.153181970664<br />
0.750000 1680.228084990751<br />
0.800000 1838.888435035886<br />
0.850000 1800.081357655754<br />
0.900000 1581.825346091701<br />
0.950000 1243.470355136624<br />
1.000000 862.962636312892<br />
1.050000 512.967429456257<br />
1.100000 243.286907449047<br />
1.150000 73.128828964477<br />
1.200000 -2.827085882290<br />
1.224745 -11.448138314825<br />
];<br />
<br />
data_rlx = [<br />
0.000000 -0.020670060764<br />
0.050000 19.360018794024<br />
0.100000 56.053066178278<br />
0.150000 86.173859662894<br />
0.200000 99.250871762213<br />
0.250000 96.073197595982<br />
0.300000 85.496344853148<br />
0.350000 61.842558692410<br />
0.408248 43.478576333529<br />
0.450000 54.685416520609<br />
0.500000 117.020469910803<br />
0.550000 207.709070871746<br />
0.600000 294.919590647244<br />
0.650000 369.101339702239<br />
0.700000 421.696425371603<br />
0.750000 445.021797830292<br />
0.800000 450.785540359893<br />
0.850000 441.873817315316<br />
0.900000 423.309809152175<br />
0.950000 395.882657649786<br />
1.000000 348.780068868428<br />
1.050000 276.895997333094<br />
1.100000 167.991003752519<br />
1.150000 62.813035785537<br />
1.200000 -2.921550847183<br />
1.224745 -11.456823618946<br />
];<br />
<br />
bx_unrlx = data_unrlx(:,1); Egsf_unrlx = data_unrlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_unrlx = Egsf_unrlx-Egsf_unrlx(end)/bx_unrlx(end)*bx_unrlx;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
<br />
bx_rlx = data_rlx(:,1); Egsf_rlx = data_rlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_rlx = Egsf_rlx-Egsf_rlx(end)/bx_rlx(end)*bx_rlx;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
<br />
% partial Burgers vector<br />
bp = sqrt(6)/6;<br />
figure(4); <br />
subplot(1,2,1);<br />
p41 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 max(bx_unrlx/bp)]); ylim([0 2000]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg1=legend('unrelaxed', 'relaxed', 2); set(lg1,'FontSize',10);<br />
subplot(1,2,2);<br />
p42 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 1.1]); ylim([0 200]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg2=legend('unrelaxed', 'relaxed', 2); set(lg2,'FontSize',10);<br />
</pre><br />
<br />
<br />
[[image:GSF_Au_LDA_VASP.jpg || Generalized stacking fault energy curve for LDA Au computed by VASP.]]</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=VASP_Computing_Generalized_Stacking_Fault_Energy_of_Au&diff=2406
VASP Computing Generalized Stacking Fault Energy of Au
2009-01-16T05:58:26Z
<p>Shryu: </p>
<hr />
<div><P ALIGN="CENTER"><br />
<FONT SIZE="+3" color="darkred"><STRONG><br />
VASP: Generalized Stacking Fault Energy of Au</STRONG></font></P><br />
<DIV><br />
<P ALIGN="CENTER"><STRONG> </STRONG></P><br />
</DIV><br />
<br />
<br />
==Input files==<br />
<br />
Here are the basic input files required for VASP calculation. Some of the files need to be changed since we need to perform a large number of calculations.<br />
<br />
===INCAR===<br />
<br />
<pre><br />
System = fcc Au<br />
LWAVE = .FALSE.<br />
ENCUT = 400<br />
ISMEAR = 1<br />
SIGMA = 0.1<br />
ISIF = 2<br />
</pre><br />
<br />
To increase plane wave cutoff, we manually put PREC = HIGH or PREC = Accurate.<br />
Default value is PREC = Medium<br />
<br />
===KPOINTS===<br />
<br />
<pre><br />
test convergence<br />
0 0 = automatic generation of k-points<br />
Monkhorst<br />
11 9 19<br />
0 0 0<br />
</pre><br />
<br />
===POSCAR===<br />
<br />
<pre><br />
POSCAR for FCC Au (created manually)<br />
4.0605<br />
1.22474487139159 0 0<br />
0 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
You also need to put the LDA pseudopotential file as <tt>POTCAR</tt> in this directory.<br />
<br />
==Results==<br />
<br />
===Perfect crystal===<br />
<br />
The following table shows energy convergence with k-points. (Reference value from Au_bulk calculation: E = -4.39 eV/atom.)<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|-<br />
! width="100pt" | KPOINTS<br />
! width="100pt" | E (eV/atom)<br />
! optimal number of CPUs<br />
! computational time (second)<br />
|-<br />
| 3x3x3 || -4.4735468 || -- || --<br />
|-<br />
| 5x5x5 || -4.4121563 || -- || --<br />
|-<br />
| 7x7x7 || -4.4043722 || -- || --<br />
|-<br />
| 7x7x11 || -4.3981093 || 16 || 341<br />
|-<br />
| 7x5x13 || -4.3974777 || 16 || 291<br />
|-<br />
| 9x7x15 || -4.395454 || -- || ---<br />
|-<br />
| 11x9x19 || -4.394620 || 8 || 580<br />
|}<br />
<br />
===Unrelaxed stacking fault energy===<br />
<br />
From now on, we fix KPOINTS at <tt>7x5x13</tt>, for which the energy of perfect crystal is <tt>E0 = -4.3974777 (eV/atom)</tt>.<br />
<br />
To create a stacking fault, we shift the repeat vector '''b''' in the x-direction by sqrt(6)/6. The new POSCAR file is (notice the change in 3rd line).<br />
<br />
<pre><br />
POSCAR for FCC Au (created manually)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Unrelaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (raw data) || -26.343752 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.3974777 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.0411142 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.0028794 (eV/angstrom^2) = 46.128 (mJ/m^2)<br />
|}<br />
<br />
The result can be compared with previous calculations Esf = 59 (mJ/m^2) (Rosengaard and Skriver, PRB, 47, 12865, 1993) and experimental value of Esf = 42 (mJ/m^2) (Devlin, J. Phys. F: Metal Phys. 4, 1865, 1974).<br />
<br />
===Relaxed stacking fault energy===<br />
<br />
We now displace the y-component of the repeat vector '''b''' and find the minimum energy. To do so we need to submit the following PBS script.<br />
<br />
relax.isf.pbs<br />
<br />
<pre><br />
...<br />
ncpu=`cat $PBS_NODEFILE | wc -w`<br />
echo "Number of processors = $ncpu "<br />
...<br />
./auto.relax.isf.par $ncpu<br />
...<br />
</pre><br />
<br />
which calls the following shell script.<br />
<br />
auto.relax.isf.par<br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
for by in 1.730 1.735 1.740 1.745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.by.dat<br />
<br />
done<br />
</pre><br />
<br />
<br />
Here are the energy values at different <math>b_y</math>.<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Energy minimization<br />
|-<br />
| width="200pt" | <math>b_y / a_0</math> <br />
| width="250pt" | E1 (eV)<br />
|-<br />
| 1.730 || -26.341633<br />
|-<br />
| 1.735 || -26.345601<br />
|-<br />
| 1.740 || -26.345699<br />
|-<br />
| 1.745 || -26.342276<br />
|}<br />
<br />
The following Matlab script finds the mininum value of E1 = -26.3461(eV) (at <math>b_y / a_0</math> = 1.7378).<br />
<pre><br />
data=[<br />
1.730 1 -.26341633E+02 -.26341112E+02 -.156102E-02<br />
1.735 1 -.26345601E+02 -.26345033E+02 -.170179E-02<br />
1.740 1 -.26345699E+02 -.26345097E+02 -.180565E-02<br />
1.745 1 -.26342276E+02 -.26341660E+02 -.184888E-02<br />
];<br />
x = [min(data(:,1)):0.0001:max(data(:,1))];<br />
[p,s] = polyfit(data(:,1),data(:,3),2);<br />
y = polyval(p,x);<br />
[y_min ind] = min(y); <br />
x_min = x(ind);<br />
figure(1); plot(data(1:4,1),data(1:4,3),'o', x,y,'-', x_min,y_min,'r*');<br />
y_min, x_min<br />
</pre><br />
<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Relaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (min) || -26.3461 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.3974777 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.0387662 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.002715 (eV/angstrom^2) = 43.49 (mJ/m^2)<br />
|}<br />
<br />
Number of CPUs = 16. Total computation time = 20 minutes.<br />
<br />
===Unrelaxed generalized stacking fault energy===<br />
<br />
The unrelaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''' (while keeping y-component of '''b''' fixed). This is done using the following shell script.<br />
<br />
auto.unrelax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
by=1.73205080756888<br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 2 hours.<br />
<br />
The results can be plotted using the following Matlab script.<br />
<br />
<pre><br />
data2=[<br />
0.00 1.73205080756888 1 -.26384866E+02 -.26384852E+02 -.419247E-04<br />
0.05 1.73205080756888 1 -.26366785E+02 -.26366967E+02 0.547795E-03<br />
0.10 1.73205080756888 1 -.26327184E+02 -.26327283E+02 0.294942E-03<br />
0.15 1.73205080756888 1 -.26288320E+02 -.26288001E+02 -.957691E-03<br />
0.20 1.73205080756888 1 -.26269732E+02 -.26269867E+02 0.405146E-03<br />
0.25 1.73205080756888 1 -.26274866E+02 -.26274925E+02 0.176062E-03<br />
0.30 1.73205080756888 1 -.26294139E+02 -.26293966E+02 -.517175E-03<br />
0.35 1.73205080756888 1 -.26323923E+02 -.26323824E+02 -.295196E-03<br />
0.408248 1.73205080756888 1 -.26343752E+02 -.26343211E+02 -.162154E-02<br />
0.45 1.73205080756888 1 -.26331580E+02 -.26331671E+02 0.272932E-03<br />
0.50 1.73205080756888 1 -.26252512E+02 -.26252146E+02 -.109923E-02<br />
0.55 1.73205080756888 1 -.26085728E+02 -.26085280E+02 -.134254E-02<br />
0.60 1.73205080756888 1 -.25829191E+02 -.25828980E+02 -.633646E-03<br />
0.65 1.73205080756888 1 -.25506381E+02 -.25506199E+02 -.546431E-03<br />
0.70 1.73205080756888 1 -.25167205E+02 -.25167286E+02 0.242977E-03<br />
0.75 1.73205080756888 1 -.24887268E+02 -.24887237E+02 -.929289E-04<br />
0.80 1.73205080756888 1 -.24745853E+02 -.24745735E+02 -.354033E-03<br />
0.85 1.73205080756888 1 -.24780442E+02 -.24780426E+02 -.490752E-04<br />
0.90 1.73205080756888 1 -.24974975E+02 -.24974773E+02 -.604431E-03<br />
0.95 1.73205080756888 1 -.25276553E+02 -.25276421E+02 -.395463E-03<br />
1.00 1.73205080756888 1 -.25615702E+02 -.25615797E+02 0.282608E-03<br />
1.05 1.73205080756888 1 -.25927655E+02 -.25927281E+02 -.112246E-02<br />
1.10 1.73205080756888 1 -.26168023E+02 -.26167870E+02 -.459552E-03<br />
1.15 1.73205080756888 1 -.26319686E+02 -.26319608E+02 -.233191E-03<br />
1.20 1.73205080756888 1 -.26387386E+02 -.26387561E+02 0.525098E-03<br />
1.224745 1.73205080756888 1 -.26395070E+02 -.26395345E+02 0.826715E-03<br />
];<br />
bx_unrlx = data2(:,1); E1 = data2(:,4);<br />
Egsf_unrlx = (E1 - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
figure(2); plot(bx_unrlx,max(Egsf_unrlx,0),'.', x_unrlx,y_unrlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{unrlx} (mJ/m^2)');<br />
xlim([0 max(bx_unrlx)]); ylim([0 2000]);<br />
% print out results<br />
nx = length(bx_unrlx);<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_unrlx(i), Egsf_unrlx(i))); end<br />
</pre><br />
<br />
===Relaxed generalized stacking fault energy===<br />
<br />
The relaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''', while allowing the y-component of '''b''' to relax. This means for every value of <math>b_x</math> we will need to a set of calculations with different values of <math>b_y</math>. This is done using the following shell script. We need to be careful that for every <math>b_x</math> the miniminum energy is sampled within the range of <math>b_y</math>. For this purpose, we also did two separate calculations with when <math>b_x</math> = 0.60:0.05:1.20, <math>b_y</math> = 1.775:0.005:1.815, and <math>b_x</math> = 0.75:0.05:1.20, <math>b_y</math> = 1.800:0.005:1.840.<br />
<br />
auto.relax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
<br />
for by in 1.730 1.735 1.740 1.745 1.750 1.755 1.760 1.765 1.770<br />
do<br />
<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 19 hours.<br />
<br />
The following Matlab script extracts the minimized energy and plots the relaxed GSF curve.<br />
<br />
<pre><br />
data3 = load('E.bxby.dat'); <br />
N = length(data3(:,1)); ny = 9; nx = floor(N/ny);<br />
bx_rlx = data3(1:ny:nx*ny,1); by = data3(1:ny,2);<br />
E1 = reshape(data3(1:nx*ny,4),ny,nx);<br />
figure(3); plot(E1,'.-');<br />
E1_min = zeros(1,nx);<br />
for i=1:nx,<br />
E1_min(i) = min(spline(by,E1(:,i),[min(by):0.0001:max(by)]));<br />
end<br />
Egsf_rlx = (E1_min - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
figure(4); plot(bx_rlx,max(Egsf_rlx,0),'o', x_rlx,y_rlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{rlx} (mJ/m^2)');<br />
% print out results<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_rlx(i), Egsf_rlx(i))); end<br />
</pre><br />
<br />
===Plot final results===<br />
The following Matlab script summarizes the results and plots the relaxed and unrelaxed GSF curves together.<br />
<br />
<pre><br />
data_unrlx = [<br />
0.000000 0.000224389703<br />
0.050000 20.286175591244<br />
0.100000 64.716458956782<br />
0.150000 108.319866262614<br />
0.200000 129.174645364115<br />
0.250000 123.414561659193<br />
0.300000 101.791247821056<br />
0.350000 68.375133082561<br />
0.408248 46.128015866956<br />
0.450000 59.784373260085<br />
0.500000 148.494598889395<br />
0.550000 335.617660954438<br />
0.600000 623.438968591740<br />
0.650000 985.615170536872<br />
0.700000 1366.153181970664<br />
0.750000 1680.228084990751<br />
0.800000 1838.888435035886<br />
0.850000 1800.081357655754<br />
0.900000 1581.825346091701<br />
0.950000 1243.470355136624<br />
1.000000 862.962636312892<br />
1.050000 512.967429456257<br />
1.100000 243.286907449047<br />
1.150000 73.128828964477<br />
1.200000 -2.827085882290<br />
1.224745 -11.448138314825<br />
];<br />
<br />
data_rlx = [<br />
0.000000 -0.020670060764<br />
0.050000 19.360018794024<br />
0.100000 56.053066178278<br />
0.150000 86.173859662894<br />
0.200000 99.250871762213<br />
0.250000 96.073197595982<br />
0.300000 85.496344853148<br />
0.350000 61.842558692410<br />
0.408248 43.478576333529<br />
0.450000 54.685416520609<br />
0.500000 117.020469910803<br />
0.550000 207.709070871746<br />
0.600000 294.919590647244<br />
0.650000 369.101339702239<br />
0.700000 421.696425371603<br />
0.750000 445.021797830292<br />
0.800000 450.785540359893<br />
0.850000 441.873817315316<br />
0.900000 423.309809152175<br />
0.950000 395.882657649786<br />
1.000000 348.780068868428<br />
1.050000 276.895997333094<br />
1.100000 167.991003752519<br />
1.150000 62.813035785537<br />
1.200000 -2.921550847183<br />
1.224745 -11.456823618946<br />
];<br />
<br />
bx_unrlx = data_unrlx(:,1); Egsf_unrlx = data_unrlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_unrlx = Egsf_unrlx-Egsf_unrlx(end)/bx_unrlx(end)*bx_unrlx;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
<br />
bx_rlx = data_rlx(:,1); Egsf_rlx = data_rlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_rlx = Egsf_rlx-Egsf_rlx(end)/bx_rlx(end)*bx_rlx;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
<br />
% partial Burgers vector<br />
bp = sqrt(6)/6;<br />
figure(4); <br />
subplot(1,2,1);<br />
p41 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 max(bx_unrlx/bp)]); ylim([0 2000]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg1=legend('unrelaxed', 'relaxed', 2); set(lg1,'FontSize',10);<br />
subplot(1,2,2);<br />
p42 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 1.1]); ylim([0 200]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg2=legend('unrelaxed', 'relaxed', 2); set(lg2,'FontSize',10);<br />
</pre><br />
<br />
<br />
[[image:GSF_Au_LDA_VASP.jpg || Generalized stacking fault energy curve for LDA Au computed by VASP.]]</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=VASP_Computing_Generalized_Stacking_Fault_Energy_of_Au&diff=2405
VASP Computing Generalized Stacking Fault Energy of Au
2009-01-16T05:54:03Z
<p>Shryu: /* Perfect crystal */</p>
<hr />
<div><P ALIGN="CENTER"><br />
<FONT SIZE="+3" color="darkred"><STRONG><br />
VASP: Generalized Stacking Fault Energy of Au</STRONG></font></P><br />
<DIV><br />
<P ALIGN="CENTER"><STRONG> </STRONG></P><br />
</DIV><br />
<br />
<br />
==Input files==<br />
<br />
Here are the basic input files required for VASP calculation. Some of the files need to be changed since we need to perform a large number of calculations.<br />
<br />
===INCAR===<br />
<br />
<pre><br />
System = fcc Au<br />
LWAVE = .FALSE.<br />
ENCUT = 400<br />
ISMEAR = 1<br />
SIGMA = 0.1<br />
ISIF = 2<br />
</pre><br />
<br />
===KPOINTS===<br />
<br />
<pre><br />
test convergence<br />
0 0 = automatic generation of k-points<br />
Monkhorst<br />
7 5 13<br />
0 0 0<br />
</pre><br />
<br />
===POSCAR===<br />
<br />
<pre><br />
POSCAR for FCC Au (created manually)<br />
4.0605<br />
1.22474487139159 0 0<br />
0 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
You also need to put the LDA pseudopotential file as <tt>POTCAR</tt> in this directory.<br />
<br />
==Results==<br />
<br />
===Perfect crystal===<br />
<br />
The following table shows energy convergence with k-points. (Reference value from Au_bulk calculation: E = -4.39 eV/atom.)<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|-<br />
! width="100pt" | KPOINTS<br />
! width="100pt" | E (eV/atom)<br />
! optimal number of CPUs<br />
! computational time (second)<br />
|-<br />
| 3x3x3 || -4.4735468 || -- || --<br />
|-<br />
| 5x5x5 || -4.4121563 || -- || --<br />
|-<br />
| 7x7x7 || -4.4043722 || -- || --<br />
|-<br />
| 7x7x11 || -4.3981093 || 16 || 341<br />
|-<br />
| 7x5x13 || -4.3974777 || 16 || 291<br />
|-<br />
| 9x7x15 || -4.395454 || -- || ---<br />
|-<br />
| 11x9x19 || -4.394620 || 8 || 580<br />
|}<br />
<br />
===Unrelaxed stacking fault energy===<br />
<br />
From now on, we fix KPOINTS at <tt>7x5x13</tt>, for which the energy of perfect crystal is <tt>E0 = -4.3974777 (eV/atom)</tt>.<br />
<br />
To create a stacking fault, we shift the repeat vector '''b''' in the x-direction by sqrt(6)/6. The new POSCAR file is (notice the change in 3rd line).<br />
<br />
<pre><br />
POSCAR for FCC Au (created manually)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Unrelaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (raw data) || -26.343752 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.3974777 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.0411142 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.0028794 (eV/angstrom^2) = 46.128 (mJ/m^2)<br />
|}<br />
<br />
The result can be compared with previous calculations Esf = 59 (mJ/m^2) (Rosengaard and Skriver, PRB, 47, 12865, 1993) and experimental value of Esf = 42 (mJ/m^2) (Devlin, J. Phys. F: Metal Phys. 4, 1865, 1974).<br />
<br />
===Relaxed stacking fault energy===<br />
<br />
We now displace the y-component of the repeat vector '''b''' and find the minimum energy. To do so we need to submit the following PBS script.<br />
<br />
relax.isf.pbs<br />
<br />
<pre><br />
...<br />
ncpu=`cat $PBS_NODEFILE | wc -w`<br />
echo "Number of processors = $ncpu "<br />
...<br />
./auto.relax.isf.par $ncpu<br />
...<br />
</pre><br />
<br />
which calls the following shell script.<br />
<br />
auto.relax.isf.par<br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
for by in 1.730 1.735 1.740 1.745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.by.dat<br />
<br />
done<br />
</pre><br />
<br />
<br />
Here are the energy values at different <math>b_y</math>.<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Energy minimization<br />
|-<br />
| width="200pt" | <math>b_y / a_0</math> <br />
| width="250pt" | E1 (eV)<br />
|-<br />
| 1.730 || -26.341633<br />
|-<br />
| 1.735 || -26.345601<br />
|-<br />
| 1.740 || -26.345699<br />
|-<br />
| 1.745 || -26.342276<br />
|}<br />
<br />
The following Matlab script finds the mininum value of E1 = -26.3461(eV) (at <math>b_y / a_0</math> = 1.7378).<br />
<pre><br />
data=[<br />
1.730 1 -.26341633E+02 -.26341112E+02 -.156102E-02<br />
1.735 1 -.26345601E+02 -.26345033E+02 -.170179E-02<br />
1.740 1 -.26345699E+02 -.26345097E+02 -.180565E-02<br />
1.745 1 -.26342276E+02 -.26341660E+02 -.184888E-02<br />
];<br />
x = [min(data(:,1)):0.0001:max(data(:,1))];<br />
[p,s] = polyfit(data(:,1),data(:,3),2);<br />
y = polyval(p,x);<br />
[y_min ind] = min(y); <br />
x_min = x(ind);<br />
figure(1); plot(data(1:4,1),data(1:4,3),'o', x,y,'-', x_min,y_min,'r*');<br />
y_min, x_min<br />
</pre><br />
<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Relaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (min) || -26.3461 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.3974777 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.0387662 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.002715 (eV/angstrom^2) = 43.49 (mJ/m^2)<br />
|}<br />
<br />
Number of CPUs = 16. Total computation time = 20 minutes.<br />
<br />
===Unrelaxed generalized stacking fault energy===<br />
<br />
The unrelaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''' (while keeping y-component of '''b''' fixed). This is done using the following shell script.<br />
<br />
auto.unrelax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
by=1.73205080756888<br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 2 hours.<br />
<br />
The results can be plotted using the following Matlab script.<br />
<br />
<pre><br />
data2=[<br />
0.00 1.73205080756888 1 -.26384866E+02 -.26384852E+02 -.419247E-04<br />
0.05 1.73205080756888 1 -.26366785E+02 -.26366967E+02 0.547795E-03<br />
0.10 1.73205080756888 1 -.26327184E+02 -.26327283E+02 0.294942E-03<br />
0.15 1.73205080756888 1 -.26288320E+02 -.26288001E+02 -.957691E-03<br />
0.20 1.73205080756888 1 -.26269732E+02 -.26269867E+02 0.405146E-03<br />
0.25 1.73205080756888 1 -.26274866E+02 -.26274925E+02 0.176062E-03<br />
0.30 1.73205080756888 1 -.26294139E+02 -.26293966E+02 -.517175E-03<br />
0.35 1.73205080756888 1 -.26323923E+02 -.26323824E+02 -.295196E-03<br />
0.408248 1.73205080756888 1 -.26343752E+02 -.26343211E+02 -.162154E-02<br />
0.45 1.73205080756888 1 -.26331580E+02 -.26331671E+02 0.272932E-03<br />
0.50 1.73205080756888 1 -.26252512E+02 -.26252146E+02 -.109923E-02<br />
0.55 1.73205080756888 1 -.26085728E+02 -.26085280E+02 -.134254E-02<br />
0.60 1.73205080756888 1 -.25829191E+02 -.25828980E+02 -.633646E-03<br />
0.65 1.73205080756888 1 -.25506381E+02 -.25506199E+02 -.546431E-03<br />
0.70 1.73205080756888 1 -.25167205E+02 -.25167286E+02 0.242977E-03<br />
0.75 1.73205080756888 1 -.24887268E+02 -.24887237E+02 -.929289E-04<br />
0.80 1.73205080756888 1 -.24745853E+02 -.24745735E+02 -.354033E-03<br />
0.85 1.73205080756888 1 -.24780442E+02 -.24780426E+02 -.490752E-04<br />
0.90 1.73205080756888 1 -.24974975E+02 -.24974773E+02 -.604431E-03<br />
0.95 1.73205080756888 1 -.25276553E+02 -.25276421E+02 -.395463E-03<br />
1.00 1.73205080756888 1 -.25615702E+02 -.25615797E+02 0.282608E-03<br />
1.05 1.73205080756888 1 -.25927655E+02 -.25927281E+02 -.112246E-02<br />
1.10 1.73205080756888 1 -.26168023E+02 -.26167870E+02 -.459552E-03<br />
1.15 1.73205080756888 1 -.26319686E+02 -.26319608E+02 -.233191E-03<br />
1.20 1.73205080756888 1 -.26387386E+02 -.26387561E+02 0.525098E-03<br />
1.224745 1.73205080756888 1 -.26395070E+02 -.26395345E+02 0.826715E-03<br />
];<br />
bx_unrlx = data2(:,1); E1 = data2(:,4);<br />
Egsf_unrlx = (E1 - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
figure(2); plot(bx_unrlx,max(Egsf_unrlx,0),'.', x_unrlx,y_unrlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{unrlx} (mJ/m^2)');<br />
xlim([0 max(bx_unrlx)]); ylim([0 2000]);<br />
% print out results<br />
nx = length(bx_unrlx);<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_unrlx(i), Egsf_unrlx(i))); end<br />
</pre><br />
<br />
===Relaxed generalized stacking fault energy===<br />
<br />
The relaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''', while allowing the y-component of '''b''' to relax. This means for every value of <math>b_x</math> we will need to a set of calculations with different values of <math>b_y</math>. This is done using the following shell script. We need to be careful that for every <math>b_x</math> the miniminum energy is sampled within the range of <math>b_y</math>. For this purpose, we also did two separate calculations with when <math>b_x</math> = 0.60:0.05:1.20, <math>b_y</math> = 1.775:0.005:1.815, and <math>b_x</math> = 0.75:0.05:1.20, <math>b_y</math> = 1.800:0.005:1.840.<br />
<br />
auto.relax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
<br />
for by in 1.730 1.735 1.740 1.745 1.750 1.755 1.760 1.765 1.770<br />
do<br />
<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 19 hours.<br />
<br />
The following Matlab script extracts the minimized energy and plots the relaxed GSF curve.<br />
<br />
<pre><br />
data3 = load('E.bxby.dat'); <br />
N = length(data3(:,1)); ny = 9; nx = floor(N/ny);<br />
bx_rlx = data3(1:ny:nx*ny,1); by = data3(1:ny,2);<br />
E1 = reshape(data3(1:nx*ny,4),ny,nx);<br />
figure(3); plot(E1,'.-');<br />
E1_min = zeros(1,nx);<br />
for i=1:nx,<br />
E1_min(i) = min(spline(by,E1(:,i),[min(by):0.0001:max(by)]));<br />
end<br />
Egsf_rlx = (E1_min - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
figure(4); plot(bx_rlx,max(Egsf_rlx,0),'o', x_rlx,y_rlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{rlx} (mJ/m^2)');<br />
% print out results<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_rlx(i), Egsf_rlx(i))); end<br />
</pre><br />
<br />
===Plot final results===<br />
The following Matlab script summarizes the results and plots the relaxed and unrelaxed GSF curves together.<br />
<br />
<pre><br />
data_unrlx = [<br />
0.000000 0.000224389703<br />
0.050000 20.286175591244<br />
0.100000 64.716458956782<br />
0.150000 108.319866262614<br />
0.200000 129.174645364115<br />
0.250000 123.414561659193<br />
0.300000 101.791247821056<br />
0.350000 68.375133082561<br />
0.408248 46.128015866956<br />
0.450000 59.784373260085<br />
0.500000 148.494598889395<br />
0.550000 335.617660954438<br />
0.600000 623.438968591740<br />
0.650000 985.615170536872<br />
0.700000 1366.153181970664<br />
0.750000 1680.228084990751<br />
0.800000 1838.888435035886<br />
0.850000 1800.081357655754<br />
0.900000 1581.825346091701<br />
0.950000 1243.470355136624<br />
1.000000 862.962636312892<br />
1.050000 512.967429456257<br />
1.100000 243.286907449047<br />
1.150000 73.128828964477<br />
1.200000 -2.827085882290<br />
1.224745 -11.448138314825<br />
];<br />
<br />
data_rlx = [<br />
0.000000 -0.020670060764<br />
0.050000 19.360018794024<br />
0.100000 56.053066178278<br />
0.150000 86.173859662894<br />
0.200000 99.250871762213<br />
0.250000 96.073197595982<br />
0.300000 85.496344853148<br />
0.350000 61.842558692410<br />
0.408248 43.478576333529<br />
0.450000 54.685416520609<br />
0.500000 117.020469910803<br />
0.550000 207.709070871746<br />
0.600000 294.919590647244<br />
0.650000 369.101339702239<br />
0.700000 421.696425371603<br />
0.750000 445.021797830292<br />
0.800000 450.785540359893<br />
0.850000 441.873817315316<br />
0.900000 423.309809152175<br />
0.950000 395.882657649786<br />
1.000000 348.780068868428<br />
1.050000 276.895997333094<br />
1.100000 167.991003752519<br />
1.150000 62.813035785537<br />
1.200000 -2.921550847183<br />
1.224745 -11.456823618946<br />
];<br />
<br />
bx_unrlx = data_unrlx(:,1); Egsf_unrlx = data_unrlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_unrlx = Egsf_unrlx-Egsf_unrlx(end)/bx_unrlx(end)*bx_unrlx;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
<br />
bx_rlx = data_rlx(:,1); Egsf_rlx = data_rlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_rlx = Egsf_rlx-Egsf_rlx(end)/bx_rlx(end)*bx_rlx;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
<br />
% partial Burgers vector<br />
bp = sqrt(6)/6;<br />
figure(4); <br />
subplot(1,2,1);<br />
p41 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 max(bx_unrlx/bp)]); ylim([0 2000]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg1=legend('unrelaxed', 'relaxed', 2); set(lg1,'FontSize',10);<br />
subplot(1,2,2);<br />
p42 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 1.1]); ylim([0 200]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg2=legend('unrelaxed', 'relaxed', 2); set(lg2,'FontSize',10);<br />
</pre><br />
<br />
<br />
[[image:GSF_Au_LDA_VASP.jpg || Generalized stacking fault energy curve for LDA Au computed by VASP.]]</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=VASP_Computing_Generalized_Stacking_Fault_Energy_of_Au&diff=2404
VASP Computing Generalized Stacking Fault Energy of Au
2009-01-16T05:53:34Z
<p>Shryu: /* Perfect crystal */</p>
<hr />
<div><P ALIGN="CENTER"><br />
<FONT SIZE="+3" color="darkred"><STRONG><br />
VASP: Generalized Stacking Fault Energy of Au</STRONG></font></P><br />
<DIV><br />
<P ALIGN="CENTER"><STRONG> </STRONG></P><br />
</DIV><br />
<br />
<br />
==Input files==<br />
<br />
Here are the basic input files required for VASP calculation. Some of the files need to be changed since we need to perform a large number of calculations.<br />
<br />
===INCAR===<br />
<br />
<pre><br />
System = fcc Au<br />
LWAVE = .FALSE.<br />
ENCUT = 400<br />
ISMEAR = 1<br />
SIGMA = 0.1<br />
ISIF = 2<br />
</pre><br />
<br />
===KPOINTS===<br />
<br />
<pre><br />
test convergence<br />
0 0 = automatic generation of k-points<br />
Monkhorst<br />
7 5 13<br />
0 0 0<br />
</pre><br />
<br />
===POSCAR===<br />
<br />
<pre><br />
POSCAR for FCC Au (created manually)<br />
4.0605<br />
1.22474487139159 0 0<br />
0 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
You also need to put the LDA pseudopotential file as <tt>POTCAR</tt> in this directory.<br />
<br />
==Results==<br />
<br />
===Perfect crystal===<br />
<br />
The following table shows energy convergence with k-points. (Reference value from Au_bulk calculation: E = -4.39 eV/atom.)<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|-<br />
! width="100pt" | KPOINTS<br />
! width="100pt" | E (eV/atom)<br />
! optimal number of CPUs<br />
! computational time (second)<br />
|-<br />
| 3x3x3 || -4.4735468 || -- || --<br />
|-<br />
| 5x5x5 || -4.4121563 || -- || --<br />
|-<br />
| 7x7x7 || -4.4043722 || -- || --<br />
|-<br />
| 7x7x11 || -4.3981093 || 16 || 341<br />
|-<br />
| 7x5x13 || -4.3974777 || 16 || 291<br />
|-<br />
| 9x7x15 || -4.395454 || -- || ---<br />
|-<br />
| 11x9x19 || -4.394620 || -- || ---<br />
|}<br />
<br />
===Unrelaxed stacking fault energy===<br />
<br />
From now on, we fix KPOINTS at <tt>7x5x13</tt>, for which the energy of perfect crystal is <tt>E0 = -4.3974777 (eV/atom)</tt>.<br />
<br />
To create a stacking fault, we shift the repeat vector '''b''' in the x-direction by sqrt(6)/6. The new POSCAR file is (notice the change in 3rd line).<br />
<br />
<pre><br />
POSCAR for FCC Au (created manually)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 1.73205080756888 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
</pre><br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Unrelaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (raw data) || -26.343752 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.3974777 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.0411142 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.0028794 (eV/angstrom^2) = 46.128 (mJ/m^2)<br />
|}<br />
<br />
The result can be compared with previous calculations Esf = 59 (mJ/m^2) (Rosengaard and Skriver, PRB, 47, 12865, 1993) and experimental value of Esf = 42 (mJ/m^2) (Devlin, J. Phys. F: Metal Phys. 4, 1865, 1974).<br />
<br />
===Relaxed stacking fault energy===<br />
<br />
We now displace the y-component of the repeat vector '''b''' and find the minimum energy. To do so we need to submit the following PBS script.<br />
<br />
relax.isf.pbs<br />
<br />
<pre><br />
...<br />
ncpu=`cat $PBS_NODEFILE | wc -w`<br />
echo "Number of processors = $ncpu "<br />
...<br />
./auto.relax.isf.par $ncpu<br />
...<br />
</pre><br />
<br />
which calls the following shell script.<br />
<br />
auto.relax.isf.par<br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
for by in 1.730 1.735 1.740 1.745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
0.40824829046386 $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.by.dat<br />
<br />
done<br />
</pre><br />
<br />
<br />
Here are the energy values at different <math>b_y</math>.<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Energy minimization<br />
|-<br />
| width="200pt" | <math>b_y / a_0</math> <br />
| width="250pt" | E1 (eV)<br />
|-<br />
| 1.730 || -26.341633<br />
|-<br />
| 1.735 || -26.345601<br />
|-<br />
| 1.740 || -26.345699<br />
|-<br />
| 1.745 || -26.342276<br />
|}<br />
<br />
The following Matlab script finds the mininum value of E1 = -26.3461(eV) (at <math>b_y / a_0</math> = 1.7378).<br />
<pre><br />
data=[<br />
1.730 1 -.26341633E+02 -.26341112E+02 -.156102E-02<br />
1.735 1 -.26345601E+02 -.26345033E+02 -.170179E-02<br />
1.740 1 -.26345699E+02 -.26345097E+02 -.180565E-02<br />
1.745 1 -.26342276E+02 -.26341660E+02 -.184888E-02<br />
];<br />
x = [min(data(:,1)):0.0001:max(data(:,1))];<br />
[p,s] = polyfit(data(:,1),data(:,3),2);<br />
y = polyval(p,x);<br />
[y_min ind] = min(y); <br />
x_min = x(ind);<br />
figure(1); plot(data(1:4,1),data(1:4,3),'o', x,y,'-', x_min,y_min,'r*');<br />
y_min, x_min<br />
</pre><br />
<br />
<br />
{| class="wikitable" cellpadding="2" style="text-align:center"<br />
|- <br />
! colspan="2" | Relaxed stacking fault energy<br />
|-<br />
| width="200pt" | E1 (min) || -26.3461 (eV)<br />
|-<br />
| E0 (perfect crystal) || -4.3974777 (eV/atom)<br />
|-<br />
| Ex (excess) = (E1-E0*6) || 0.0387662 (eV)<br />
|-<br />
| Area || 14.27873 (angstrom^2)<br />
|-<br />
| Esf (unrelaxed) = Ex / Area || 0.002715 (eV/angstrom^2) = 43.49 (mJ/m^2)<br />
|}<br />
<br />
Number of CPUs = 16. Total computation time = 20 minutes.<br />
<br />
===Unrelaxed generalized stacking fault energy===<br />
<br />
The unrelaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''' (while keeping y-component of '''b''' fixed). This is done using the following shell script.<br />
<br />
auto.unrelax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
by=1.73205080756888<br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 2 hours.<br />
<br />
The results can be plotted using the following Matlab script.<br />
<br />
<pre><br />
data2=[<br />
0.00 1.73205080756888 1 -.26384866E+02 -.26384852E+02 -.419247E-04<br />
0.05 1.73205080756888 1 -.26366785E+02 -.26366967E+02 0.547795E-03<br />
0.10 1.73205080756888 1 -.26327184E+02 -.26327283E+02 0.294942E-03<br />
0.15 1.73205080756888 1 -.26288320E+02 -.26288001E+02 -.957691E-03<br />
0.20 1.73205080756888 1 -.26269732E+02 -.26269867E+02 0.405146E-03<br />
0.25 1.73205080756888 1 -.26274866E+02 -.26274925E+02 0.176062E-03<br />
0.30 1.73205080756888 1 -.26294139E+02 -.26293966E+02 -.517175E-03<br />
0.35 1.73205080756888 1 -.26323923E+02 -.26323824E+02 -.295196E-03<br />
0.408248 1.73205080756888 1 -.26343752E+02 -.26343211E+02 -.162154E-02<br />
0.45 1.73205080756888 1 -.26331580E+02 -.26331671E+02 0.272932E-03<br />
0.50 1.73205080756888 1 -.26252512E+02 -.26252146E+02 -.109923E-02<br />
0.55 1.73205080756888 1 -.26085728E+02 -.26085280E+02 -.134254E-02<br />
0.60 1.73205080756888 1 -.25829191E+02 -.25828980E+02 -.633646E-03<br />
0.65 1.73205080756888 1 -.25506381E+02 -.25506199E+02 -.546431E-03<br />
0.70 1.73205080756888 1 -.25167205E+02 -.25167286E+02 0.242977E-03<br />
0.75 1.73205080756888 1 -.24887268E+02 -.24887237E+02 -.929289E-04<br />
0.80 1.73205080756888 1 -.24745853E+02 -.24745735E+02 -.354033E-03<br />
0.85 1.73205080756888 1 -.24780442E+02 -.24780426E+02 -.490752E-04<br />
0.90 1.73205080756888 1 -.24974975E+02 -.24974773E+02 -.604431E-03<br />
0.95 1.73205080756888 1 -.25276553E+02 -.25276421E+02 -.395463E-03<br />
1.00 1.73205080756888 1 -.25615702E+02 -.25615797E+02 0.282608E-03<br />
1.05 1.73205080756888 1 -.25927655E+02 -.25927281E+02 -.112246E-02<br />
1.10 1.73205080756888 1 -.26168023E+02 -.26167870E+02 -.459552E-03<br />
1.15 1.73205080756888 1 -.26319686E+02 -.26319608E+02 -.233191E-03<br />
1.20 1.73205080756888 1 -.26387386E+02 -.26387561E+02 0.525098E-03<br />
1.224745 1.73205080756888 1 -.26395070E+02 -.26395345E+02 0.826715E-03<br />
];<br />
bx_unrlx = data2(:,1); E1 = data2(:,4);<br />
Egsf_unrlx = (E1 - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
figure(2); plot(bx_unrlx,max(Egsf_unrlx,0),'.', x_unrlx,y_unrlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{unrlx} (mJ/m^2)');<br />
xlim([0 max(bx_unrlx)]); ylim([0 2000]);<br />
% print out results<br />
nx = length(bx_unrlx);<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_unrlx(i), Egsf_unrlx(i))); end<br />
</pre><br />
<br />
===Relaxed generalized stacking fault energy===<br />
<br />
The relaxed generalized stacking fault energy is obtained by displacing the x-component of the repeat vector '''b''', while allowing the y-component of '''b''' to relax. This means for every value of <math>b_x</math> we will need to a set of calculations with different values of <math>b_y</math>. This is done using the following shell script. We need to be careful that for every <math>b_x</math> the miniminum energy is sampled within the range of <math>b_y</math>. For this purpose, we also did two separate calculations with when <math>b_x</math> = 0.60:0.05:1.20, <math>b_y</math> = 1.775:0.005:1.815, and <math>b_x</math> = 0.75:0.05:1.20, <math>b_y</math> = 1.800:0.005:1.840.<br />
<br />
auto.relax.gsf.par <br />
<br />
<pre><br />
#!/bin/bash<br />
<br />
ncpu=$1<br />
<br />
rm WAVECAR <br />
<br />
for bx in 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 \<br />
0.408248 0.45 0.50 0.55 0.60 0.65 0.70 \<br />
0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 \<br />
1.15 1.20 1.224745<br />
do<br />
<br />
for by in 1.730 1.735 1.740 1.745 1.750 1.755 1.760 1.765 1.770<br />
do<br />
<br />
cat > POSCAR << FIN<br />
POSCAR for FCC Au (created by tcl)<br />
4.0605<br />
1.22474487139159 0 0<br />
$bx $by 0<br />
0 0 0.70710678118655<br />
6<br />
Cartesian (real coordinates r)<br />
0 0 0<br />
0.40824829046386 0.57735026918963 0<br />
0.81649658092773 1.15470053837925 0<br />
0.61237243569579 0 0.35355339059327<br />
1.02062072615966 0.57735026918963 0.35355339059327<br />
0.20412414523193 1.15470053837925 0.35355339059327<br />
FIN<br />
<br />
echo "bx=$bx by=$by ncpu=$ncpu"<br />
<br />
mpiexec -np $ncpu ../../../../bin/vasp.mpotts.mva2 >& vasp.log<br />
<br />
E=`tail -1 OSZICAR`<br />
echo $bx $by $E | sed -s 's/F=//; s/E0=//; s/d E =//;' >> E.bxby.dat<br />
<br />
timeused=`grep "Total CPU time used" OUTCAR`<br />
echo $bx $by $timeused >> time.bxby.dat<br />
<br />
done<br />
<br />
done<br />
</pre><br />
<br />
Number of CPUs = 16. Total computation time = 19 hours.<br />
<br />
The following Matlab script extracts the minimized energy and plots the relaxed GSF curve.<br />
<br />
<pre><br />
data3 = load('E.bxby.dat'); <br />
N = length(data3(:,1)); ny = 9; nx = floor(N/ny);<br />
bx_rlx = data3(1:ny:nx*ny,1); by = data3(1:ny,2);<br />
E1 = reshape(data3(1:nx*ny,4),ny,nx);<br />
figure(3); plot(E1,'.-');<br />
E1_min = zeros(1,nx);<br />
for i=1:nx,<br />
E1_min(i) = min(spline(by,E1(:,i),[min(by):0.0001:max(by)]));<br />
end<br />
Egsf_rlx = (E1_min - (-4.3974777*6)) / 14.27873 * 1.602e-19*1e20 * 1e3;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
figure(4); plot(bx_rlx,max(Egsf_rlx,0),'o', x_rlx,y_rlx,'-');<br />
set(gca,'FontSize',16);<br />
xlabel('b_x / a_0'); ylabel('\gamma_{gsf}^{rlx} (mJ/m^2)');<br />
% print out results<br />
for i=1:nx, disp(sprintf('%12.6f %24.12f', bx_rlx(i), Egsf_rlx(i))); end<br />
</pre><br />
<br />
===Plot final results===<br />
The following Matlab script summarizes the results and plots the relaxed and unrelaxed GSF curves together.<br />
<br />
<pre><br />
data_unrlx = [<br />
0.000000 0.000224389703<br />
0.050000 20.286175591244<br />
0.100000 64.716458956782<br />
0.150000 108.319866262614<br />
0.200000 129.174645364115<br />
0.250000 123.414561659193<br />
0.300000 101.791247821056<br />
0.350000 68.375133082561<br />
0.408248 46.128015866956<br />
0.450000 59.784373260085<br />
0.500000 148.494598889395<br />
0.550000 335.617660954438<br />
0.600000 623.438968591740<br />
0.650000 985.615170536872<br />
0.700000 1366.153181970664<br />
0.750000 1680.228084990751<br />
0.800000 1838.888435035886<br />
0.850000 1800.081357655754<br />
0.900000 1581.825346091701<br />
0.950000 1243.470355136624<br />
1.000000 862.962636312892<br />
1.050000 512.967429456257<br />
1.100000 243.286907449047<br />
1.150000 73.128828964477<br />
1.200000 -2.827085882290<br />
1.224745 -11.448138314825<br />
];<br />
<br />
data_rlx = [<br />
0.000000 -0.020670060764<br />
0.050000 19.360018794024<br />
0.100000 56.053066178278<br />
0.150000 86.173859662894<br />
0.200000 99.250871762213<br />
0.250000 96.073197595982<br />
0.300000 85.496344853148<br />
0.350000 61.842558692410<br />
0.408248 43.478576333529<br />
0.450000 54.685416520609<br />
0.500000 117.020469910803<br />
0.550000 207.709070871746<br />
0.600000 294.919590647244<br />
0.650000 369.101339702239<br />
0.700000 421.696425371603<br />
0.750000 445.021797830292<br />
0.800000 450.785540359893<br />
0.850000 441.873817315316<br />
0.900000 423.309809152175<br />
0.950000 395.882657649786<br />
1.000000 348.780068868428<br />
1.050000 276.895997333094<br />
1.100000 167.991003752519<br />
1.150000 62.813035785537<br />
1.200000 -2.921550847183<br />
1.224745 -11.456823618946<br />
];<br />
<br />
bx_unrlx = data_unrlx(:,1); Egsf_unrlx = data_unrlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_unrlx = Egsf_unrlx-Egsf_unrlx(end)/bx_unrlx(end)*bx_unrlx;<br />
pp = spline(bx_unrlx, Egsf_unrlx);<br />
x_unrlx = [min(bx_unrlx):0.001:max(bx_unrlx)]; y_unrlx = ppval(pp, x_unrlx);<br />
<br />
bx_rlx = data_rlx(:,1); Egsf_rlx = data_rlx(:,2);<br />
% apply offset<br />
Egsf_unrlx = Egsf_unrlx - Egsf_unrlx(1);<br />
Egsf_rlx = Egsf_rlx-Egsf_rlx(end)/bx_rlx(end)*bx_rlx;<br />
pp = spline(bx_rlx, Egsf_rlx);<br />
x_rlx = [min(bx_rlx):0.001:max(bx_rlx)]; y_rlx = ppval(pp, x_rlx);<br />
<br />
% partial Burgers vector<br />
bp = sqrt(6)/6;<br />
figure(4); <br />
subplot(1,2,1);<br />
p41 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 max(bx_unrlx/bp)]); ylim([0 2000]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg1=legend('unrelaxed', 'relaxed', 2); set(lg1,'FontSize',10);<br />
subplot(1,2,2);<br />
p42 = plot(x_unrlx/bp,y_unrlx,'r-', x_rlx/bp, y_rlx,'b-', ...<br />
bx_unrlx/bp,max(Egsf_unrlx,0),'m.', bx_rlx/bp, max(Egsf_rlx, 0),'k.');<br />
set(gca,'FontSize',12);<br />
xlim([0 1.1]); ylim([0 200]);<br />
xlabel(' b_x / b_p '); <br />
ylabel('\gamma_{GSF} (mJ/m^2)');<br />
lg2=legend('unrelaxed', 'relaxed', 2); set(lg2,'FontSize',10);<br />
</pre><br />
<br />
<br />
[[image:GSF_Au_LDA_VASP.jpg || Generalized stacking fault energy curve for LDA Au computed by VASP.]]</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Group_Presentation_Schedule&diff=968
Group Presentation Schedule
2009-01-14T19:38:56Z
<p>Shryu: </p>
<hr />
<div>Please contact '''Seunghwa Ryu''' to sign up for your presentation at the group meeting.<br />
<br />
{| border="1"<br />
|+ '''2008-2009 Winter Quarter (1/6-3/20) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || 1/21 || Qual practice talk<br />
|}<br />
<br />
{| border="1"<br />
|+ '''2008-2009 Autumn Quarter (9/23-12/12) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
|Enseok Lee || 9/25 || ESC Meeting Practice, Topic TBD<br />
|-<br />
| || 10/2 || <br />
|-<br />
| || 10/9 || <br />
|-<br />
| Chris Weinberger || 10/16 || Dislocation 2008 Conference Talk Practice <br />
|-<br />
| Seunghwa Ryu || 10/30 || Nucleation in 2-D and 3-D Ising Systems I<br />
|-<br />
| Seunghwa Ryu || 11/6 || Nucleation in 2-D and 3-D Ising Systems II<br />
|-<br />
| Jie Yin || 11/13 || Elastic Fields of Dislocations in Anisotropic Media I<br />
|-<br />
| Keonwook Kang || 11/20 || Brittle and Ductile Failures of Silicon Nanowires under Tension Simulation<br />
<br />
|-<br />
| Chris Weinberger || 11/20|| Torsion and Bending Simulations of Metallic Nanowires<br />
|-<br />
| || 12/4 || <br />
|-<br />
| || 12/11 || <br />
|}<br />
<br />
<br />
{| border="1"<br />
|+ '''Tentative topics'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || TBD || Qual practice talk <br />
|}<br />
<br />
Past Group meetings<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Summer Quarter (6/24-9/14) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
|Alfredo Correa || 7/3 || An introduction to Quantum Mechanics I<br />
|-<br />
| Alfredo Correa || 7/31 || An introduction to Quantum Mechanics II<br />
|-<br />
| Alfredo Correa || 8/14 || An introduction to Quantum Mechanics III<br />
|-<br />
| Chris Weinberger || 8/28 || Basic Differential Geometry I<br />
|-<br />
| Chris Weinberger || 9/4|| Basic Differential Geometry II<br />
|-<br />
| Seokwoo Lee || 9/11 || Uniaxial Compression Tests of FCC Au Nanopillars I<br />
|-<br />
| Seokwoo Lee || 9/18 || Uniaxial Compression Tests of FCC Au Nanopillars II<br />
|}<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Spring Quarter (4/2-6/14)Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
|Alfredo Correa || 4/14 || Quantum Corral Wave-function engineering<br />
|-<br />
| Seunghwa Ryu || 4/22 || Binary Phase Diagram from Atomistic Simulation I (Related Physics Review)<br />
|-<br />
| Seunghwa Ryu || 4/29 || Binary Phase Diagram from Atomistic Simulation II (Realization via Simulation)<br />
|-<br />
| Seunghwa Ryu || 5/6 ||Fast Quantum Entanglement Minimization Algorithm (Pure mathematical approach)<br />
|}<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Winter Quarter(1/8-3/16) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Yongxing Shen || 1/17 || Modeling electrostatic force microscopy and related technique (PhD thesis defense practice)<br />
|-<br />
| Sylvie Aubry || 1/23 || Calculation of Si thermal boundary resistance<br />
|-<br />
| Seokwoo Lee || 1/30 || Dislocation Dynamics in thin films<br />
|-<br />
| Jie Yin || 2/6 || Anisotropic elasticity I<br />
|-<br />
| Hark Lee || 2/13 || TBD<br />
|-<br />
| Billy Cash || 2/20 || Fatigue in metals<br />
|-<br />
| Steven Fitzgerald || 2/27 || An extended Frenkel-Kontorova model for crowdions<br />
|-<br />
| Yongxing Shen || 3/5 || Euler-Bernoulli beam theory applied to AFM<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Group_Presentation_Schedule&diff=967
Group Presentation Schedule
2009-01-12T20:01:14Z
<p>Shryu: </p>
<hr />
<div>Please contact '''Seunghwa Ryu''' to sign up for your presentation at the group meeting.<br />
<br />
{| border="1"<br />
|+ '''2008-2009 Winter Quarter (1/6-3/20) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || ?? || Qual practice talk<br />
|}<br />
<br />
{| border="1"<br />
|+ '''2008-2009 Autumn Quarter (9/23-12/12) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
|Enseok Lee || 9/25 || ESC Meeting Practice, Topic TBD<br />
|-<br />
| || 10/2 || <br />
|-<br />
| || 10/9 || <br />
|-<br />
| Chris Weinberger || 10/16 || Dislocation 2008 Conference Talk Practice <br />
|-<br />
| Seunghwa Ryu || 10/30 || Nucleation in 2-D and 3-D Ising Systems I<br />
|-<br />
| Seunghwa Ryu || 11/6 || Nucleation in 2-D and 3-D Ising Systems II<br />
|-<br />
| Jie Yin || 11/13 || Elastic Fields of Dislocations in Anisotropic Media I<br />
|-<br />
| Keonwook Kang || 11/20 || Brittle and Ductile Failures of Silicon Nanowires under Tension Simulation<br />
<br />
|-<br />
| Chris Weinberger || 11/20|| Torsion and Bending Simulations of Metallic Nanowires<br />
|-<br />
| || 12/4 || <br />
|-<br />
| || 12/11 || <br />
|}<br />
<br />
<br />
{| border="1"<br />
|+ '''Tentative topics'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || TBD || Qual practice talk <br />
|}<br />
<br />
Past Group meetings<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Summer Quarter (6/24-9/14) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
|Alfredo Correa || 7/3 || An introduction to Quantum Mechanics I<br />
|-<br />
| Alfredo Correa || 7/31 || An introduction to Quantum Mechanics II<br />
|-<br />
| Alfredo Correa || 8/14 || An introduction to Quantum Mechanics III<br />
|-<br />
| Chris Weinberger || 8/28 || Basic Differential Geometry I<br />
|-<br />
| Chris Weinberger || 9/4|| Basic Differential Geometry II<br />
|-<br />
| Seokwoo Lee || 9/11 || Uniaxial Compression Tests of FCC Au Nanopillars I<br />
|-<br />
| Seokwoo Lee || 9/18 || Uniaxial Compression Tests of FCC Au Nanopillars II<br />
|}<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Spring Quarter (4/2-6/14)Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
|Alfredo Correa || 4/14 || Quantum Corral Wave-function engineering<br />
|-<br />
| Seunghwa Ryu || 4/22 || Binary Phase Diagram from Atomistic Simulation I (Related Physics Review)<br />
|-<br />
| Seunghwa Ryu || 4/29 || Binary Phase Diagram from Atomistic Simulation II (Realization via Simulation)<br />
|-<br />
| Seunghwa Ryu || 5/6 ||Fast Quantum Entanglement Minimization Algorithm (Pure mathematical approach)<br />
|}<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Winter Quarter(1/8-3/16) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Yongxing Shen || 1/17 || Modeling electrostatic force microscopy and related technique (PhD thesis defense practice)<br />
|-<br />
| Sylvie Aubry || 1/23 || Calculation of Si thermal boundary resistance<br />
|-<br />
| Seokwoo Lee || 1/30 || Dislocation Dynamics in thin films<br />
|-<br />
| Jie Yin || 2/6 || Anisotropic elasticity I<br />
|-<br />
| Hark Lee || 2/13 || TBD<br />
|-<br />
| Billy Cash || 2/20 || Fatigue in metals<br />
|-<br />
| Steven Fitzgerald || 2/27 || An extended Frenkel-Kontorova model for crowdions<br />
|-<br />
| Yongxing Shen || 3/5 || Euler-Bernoulli beam theory applied to AFM<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Group_Presentation_Schedule&diff=966
Group Presentation Schedule
2009-01-12T20:00:47Z
<p>Shryu: </p>
<hr />
<div>Please contact '''Seunghwa Ryu''' to sign up for your presentation at the group meeting.<br />
<br />
{| border="1"<br />
|+ '''2008-2009 Winter Quarter (1/6-3/20) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || ?? || Qual practice talk<br />
<br />
<br />
{| border="1"<br />
|+ '''2008-2009 Autumn Quarter (9/23-12/12) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
|Enseok Lee || 9/25 || ESC Meeting Practice, Topic TBD<br />
|-<br />
| || 10/2 || <br />
|-<br />
| || 10/9 || <br />
|-<br />
| Chris Weinberger || 10/16 || Dislocation 2008 Conference Talk Practice <br />
|-<br />
| Seunghwa Ryu || 10/30 || Nucleation in 2-D and 3-D Ising Systems I<br />
|-<br />
| Seunghwa Ryu || 11/6 || Nucleation in 2-D and 3-D Ising Systems II<br />
|-<br />
| Jie Yin || 11/13 || Elastic Fields of Dislocations in Anisotropic Media I<br />
|-<br />
| Keonwook Kang || 11/20 || Brittle and Ductile Failures of Silicon Nanowires under Tension Simulation<br />
<br />
|-<br />
| Chris Weinberger || 11/20|| Torsion and Bending Simulations of Metallic Nanowires<br />
|-<br />
| || 12/4 || <br />
|-<br />
| || 12/11 || <br />
|}<br />
<br />
<br />
{| border="1"<br />
|+ '''Tentative topics'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || TBD || Qual practice talk <br />
|}<br />
<br />
Past Group meetings<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Summer Quarter (6/24-9/14) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
|Alfredo Correa || 7/3 || An introduction to Quantum Mechanics I<br />
|-<br />
| Alfredo Correa || 7/31 || An introduction to Quantum Mechanics II<br />
|-<br />
| Alfredo Correa || 8/14 || An introduction to Quantum Mechanics III<br />
|-<br />
| Chris Weinberger || 8/28 || Basic Differential Geometry I<br />
|-<br />
| Chris Weinberger || 9/4|| Basic Differential Geometry II<br />
|-<br />
| Seokwoo Lee || 9/11 || Uniaxial Compression Tests of FCC Au Nanopillars I<br />
|-<br />
| Seokwoo Lee || 9/18 || Uniaxial Compression Tests of FCC Au Nanopillars II<br />
|}<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Spring Quarter (4/2-6/14)Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
|Alfredo Correa || 4/14 || Quantum Corral Wave-function engineering<br />
|-<br />
| Seunghwa Ryu || 4/22 || Binary Phase Diagram from Atomistic Simulation I (Related Physics Review)<br />
|-<br />
| Seunghwa Ryu || 4/29 || Binary Phase Diagram from Atomistic Simulation II (Realization via Simulation)<br />
|-<br />
| Seunghwa Ryu || 5/6 ||Fast Quantum Entanglement Minimization Algorithm (Pure mathematical approach)<br />
|}<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Winter Quarter(1/8-3/16) Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Yongxing Shen || 1/17 || Modeling electrostatic force microscopy and related technique (PhD thesis defense practice)<br />
|-<br />
| Sylvie Aubry || 1/23 || Calculation of Si thermal boundary resistance<br />
|-<br />
| Seokwoo Lee || 1/30 || Dislocation Dynamics in thin films<br />
|-<br />
| Jie Yin || 2/6 || Anisotropic elasticity I<br />
|-<br />
| Hark Lee || 2/13 || TBD<br />
|-<br />
| Billy Cash || 2/20 || Fatigue in metals<br />
|-<br />
| Steven Fitzgerald || 2/27 || An extended Frenkel-Kontorova model for crowdions<br />
|-<br />
| Yongxing Shen || 3/5 || Euler-Bernoulli beam theory applied to AFM<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Research_Meeting_Schedule&diff=2230
Research Meeting Schedule
2009-01-09T21:22:43Z
<p>Shryu: /* Winter 2009 */</p>
<hr />
<div>===Winter 2009===<br />
<br />
Prof. Wei Cai's weekly schedule.<br />
<br />
{| class="wikitable" style="text-align:center;"<br />
!width="150pt" | Time <br />
!width="200pt" | Monday <br />
!width="200pt" | Tuesday <br />
!width="200pt" | Wednesday <br />
!width="200pt" | Thursday <br />
!width="200pt" | Friday<br />
|-<br />
|9:00-10:00|| bi-weekly Eunseok,<br> Hark, Prof. Prinz <br />
| rowspan="3" | ''work'' <br />
| ''work'' <br />
| rowspan="3" | ''work'' <br />
| ''work''<br />
|-<br />
|10:00-10:45|| lecture prep. || lecture prep. || lecture prep.<br />
|-<br />
|11:00-12:15||'''ME346A'''||'''ME346A'''||'''ME346A'''<br>(make-up lecture)<br />
|-<br />
|12:30-1:15 || colspan="5" align="center" bgcolor="lightgrey" | lunch break<br />
|-<br />
|1:30-2:05 ||Billy ||Keonwook ||Seok-Woo ||Eunseok Lee <br />
| rowspan="5" align="center" | ''work'' <br />
|-<br />
|2:10-2:45 ||research meeting ||research meeting ||Chris || Sylvie<br />
|-<br />
|2:50-3:25 ||research meeting || research meeting ||Jie || Seunghwa<br />
|-<br />
|3:30-4:00 || colspan="3" align="center" bgcolor="lightgrey" | coffee break <br />
| ''work''<br />
|-<br />
|4:00-6:00 || ''work'' || ''work'' || ''work'' || '''MC seminar'''<br />
|}<br />
<br />
Meetings to sign-up<br />
<br />
{| class="wikitable" style="text-align:center;"<br />
|-<br />
| width="120pt" | <br />
| width="120pt" | <br />
| width="120pt" | <br />
| width="120pt" | <br />
| width="120pt" | <br />
| width="120pt" | Seunghwa <br />
|-<br />
<br />
| width="120pt" | Alfredo<br />
| width="120pt" | <br />
| width="120pt" | Hark<br />
| width="120pt" | Haneesh <br />
| width="120pt" | <br />
|-<br />
| colspan="2" | '''Group lunch''' <br />
| colspan="2" | '''ME346A Office hour''' <br />
| colspan="2" | Seminar speaker <br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Computing_Melting_Point_by_Free_Energy_Method&diff=2100
Computing Melting Point by Free Energy Method
2008-11-24T09:13:17Z
<p>Shryu: /* Submit jobs */</p>
<hr />
<div><H1 ALIGN="CENTER><br />
Computing Melting Point by Free Eneregy Method</H1><br />
<br />
<P ALIGN="CENTER"><STRONG>Seunghwa Ryu and Wei Cai</STRONG></P><br />
<br />
This tutorial describes how to compute melting points from free energy methods. We provide the MD++ script files and describe how to use them in detail. The theoretical background is published in <br />
<br />
Comparison of Thermal Properties Predicted by Interatomic Potential Models, ''Modelling and Simulation in Materials Science and Engineering'', '''16''', 085005 (2008). [http://micro.stanford.edu/~caiwei/papers/Ryu08msmse-meltingT.pdf (PDF)].<br />
<br />
===Download files===<br />
<br />
First, you need to install MD++ on your computer by following the instructions on [[MD++_Manuals | MD++ Manuals]].<br />
<br />
Second, copy files [[media:melting_cubic.tcl.txt | melting_cubic.tcl]] and<br />
[[media:melting_noncubic.tcl.txt | melting_noncubic.tcl]] to your input file directory of MD++. You can do so by<br />
<br />
<pre><br />
mkdir -p ~/Codes/MD++/scripts/work/melting<br />
cd ~/Codes/MD++/scripts/work/melting<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_cubic.tcl.txt -O melting_cubic.tcl<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_noncubic.tcl.txt -O melting_noncubic.tcl<br />
</pre><br />
<br />
Both files contain a big comment section in the beginning that describes in detail the (16) steps to compute the melting point. These steps are fully automated. In this tutorial, we describe how to run these scripts on a parallel cluster (using wcr.stanford.edu as an example) and what kind of results you should expect.<br />
<br />
===Compile excutable file===<br />
<br />
For each different potential, MD++ has corresponding excutable files.<br />
<pre><br />
Stillinger-Weber Silicon - sw<br />
Stillinger-Weber Germanium - swge<br />
MEAM - meam-lammps, meam-baskes, meam<br />
EAM - eam<br />
Finnis-Sinclair - fs<br />
Tersoff Silicon - tersoff <br />
</pre><br />
<br />
For example, you can compile sworig by doing<br />
<pre><br />
cd ~/Codes/MD++/<br />
make sworig build=R SYS=wcr<br />
</pre><br />
then you will have 'sworig_wcr' file in the directory '~/Codes/MD++/bin'<br />
<br />
For other potentials, refer to the explanations in melting_cubic.tcl<br />
or email to shryu@stanford.edu for any question.<br />
<br />
===Submit jobs===<br />
<br />
Here are the PBS scripts for submitting the jobs on wcr.stanford.edu.<br />
<br />
<pre><br />
cd ~/Codes/MD++/<br />
mkdir runs/Single_Elem_Tm<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt1.pbs.txt -O melt1.pbs<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt2.pbs.txt -O melt2.pbs<br />
qsub melt1.pbs<br />
qsub melt2.pbs<br />
</pre><br />
<br />
Then you'll get free energy data in <br />
<pre><br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Solid/<br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Liquid/<br />
</pre><br />
<br />
===Analyse data===<br />
<br />
At the end of the calculation, data files and Matlab files will be automatically generated. Transfer these files to a location where you have Matlab installed. Run the following commands in Matlab and you will get the free energy plot (for solid and liquid phases). The melting point is the temperature at which the two free energy curves cross.<br />
<br />
The analysis files are generated in<br />
~/Codes/MD++/runs/Single_Elem_Tm/<br />
<pre><br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
tar_SW_Si<br />
</pre><br />
<br />
to zip necessary data <br />
<pre><br />
cd ~/Codes/MD++/runs/Single_Elem_Tm<br />
./tar_SW_Si<br />
</pre><br />
then you'll have<br />
'inter_SW_Si.tar'<br />
<br />
copy this file to your local computer and unzip.<br />
Then, execute<br />
<pre><br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
</pre><br />
you'll see following output files including all free energy data and error analysis<br />
<pre><br />
Solid Free Energy Info<br />
lattice const = 5.43095<br />
strain = 0.00518<br />
E_0 = -2218.74337 eV ( -4.33348 eV/atom)<br />
F_w = -253.34582 eV ( -0.49482 eV/atom)<br />
F_ha = E_0 + F_w = -2472.08919 eV ( -4.82830 eV/atom)<br />
W_re_to_ha = 5.99169 eV ( 0.01170 eV/atom)<br />
W_ha_to_re = -7.30068 eV ( -0.01426 eV/atom)<br />
F_re - F_ha = (W_ha_to_re - W_re_to_ha)/2 = -6.64619 eV ( -0.01298 eV/atom)<br />
F_re at 1600.000000 = -2478.73538 eV ( -4.84128 eV/atom)<br />
W_T1_to_T11 = 417.87095 eV ( 0.81615 eV/atom)<br />
W_T11_to_T1 = -417.87508 eV ( -0.81616 eV/atom)<br />
dW1 = (W_T1_to_T11 - W_T11_to_T1)/2 = 417.87302 eV ( 0.81616 eV/atom)<br />
F_re at 2000.000000 = -2605.55266 eV ( -5.08897 eV/atom)<br />
<br />
Liquid Free Energy Info<br />
strain = -0.01894<br />
F_ideal = -919.36940 ( -1.79564 eV/atom)<br />
W_re_to_ga = 1888.21102 eV ( 3.68791 eV/atom)<br />
W_ga_to_re = -1887.53306 eV ( -3.68659 eV/atom)<br />
F_re - F_ga = (W_ga_to_re - W_re_to_ga)/2 = -1887.87204 eV ( -3.68725 eV/atom)<br />
W_ga_to_id = -256.29295 eV ( -0.50057 eV/atom)<br />
W_id_to_ga = 256.52177 eV ( 0.50102 eV/atom)<br />
F_ga - F_id = (W_id_to_ga - W_ga_to_id)/2 = 256.40736 eV ( 0.50080 eV/atom)<br />
F_re at 1800.000000 = -2550.83408 eV ( -4.98210 eV/atom)<br />
W_T2_to_T22 = -583.02785 eV ( -1.13873 eV/atom)<br />
W_T22_to_T2 = 583.06542 eV ( 1.13880 eV/atom)<br />
dW2 = (W_T2_to_T22 - W_T22_to_T2)/2 = -583.04664 eV ( -1.13876 eV/atom)<br />
F_re at 1384.615385 = -2386.68534 eV ( -4.66149 eV/atom)<br />
<br />
Error Analysis<br />
<br />
Solid Side<br />
Err_re_to_ha = 0.02769 eV ( 0.00005407 eV/atom)( 0.628 K)<br />
Err_ha_to_re = 0.02815 eV ( 0.00005498 eV/atom)( 0.638 K)<br />
Err_T1_to_T11 = 0.00374 eV ( 0.00000730 eV/atom)( 0.085 K)<br />
Err_T11_to_T1 = 0.00441 eV ( 0.00000861 eV/atom)( 0.100 K)<br />
<br />
Liquid Side<br />
Err_re_to_ga = 0.04793 eV ( 0.00009362 eV/atom)( 1.086 K)<br />
Err_ga_to_re = 0.06601 eV ( 0.00012893 eV/atom)( 1.496 K)<br />
Err_ga_to_id = 0.02557 eV ( 0.00004993 eV/atom)( 0.580 K)<br />
Err_id_to_ga = 0.03421 eV ( 0.00006682 eV/atom)( 0.776 K)<br />
Err_T2_to_T22 = 0.01488 eV ( 0.00002906 eV/atom)( 0.337 K)<br />
Err_T22_to_T2 = 0.01183 eV ( 0.00002311 eV/atom)( 0.268 K)<br />
Max dT = 2.315<br />
Tm= 1695.08000 (K), err= 1.044 Fm = -4.89779 (eV/atom)<br />
L= 165.8833097832 (eV) Ss=0.308452 Sl=0.406313<br />
L= 0.3239908394 (eV/atom) ss=0.000602 sl=0.000794<br />
L= 31206.7976529696 (J/mol) ss=58.027496 sl=76.437718<br />
L= 1111115.7748689589 (J/kg) ss=2066.064819 sl=2721.559412<br />
</pre></div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Computing_Melting_Point_by_Free_Energy_Method&diff=2099
Computing Melting Point by Free Energy Method
2008-11-24T09:08:56Z
<p>Shryu: </p>
<hr />
<div><H1 ALIGN="CENTER><br />
Computing Melting Point by Free Eneregy Method</H1><br />
<br />
<P ALIGN="CENTER"><STRONG>Seunghwa Ryu and Wei Cai</STRONG></P><br />
<br />
This tutorial describes how to compute melting points from free energy methods. We provide the MD++ script files and describe how to use them in detail. The theoretical background is published in <br />
<br />
Comparison of Thermal Properties Predicted by Interatomic Potential Models, ''Modelling and Simulation in Materials Science and Engineering'', '''16''', 085005 (2008). [http://micro.stanford.edu/~caiwei/papers/Ryu08msmse-meltingT.pdf (PDF)].<br />
<br />
===Download files===<br />
<br />
First, you need to install MD++ on your computer by following the instructions on [[MD++_Manuals | MD++ Manuals]].<br />
<br />
Second, copy files [[media:melting_cubic.tcl.txt | melting_cubic.tcl]] and<br />
[[media:melting_noncubic.tcl.txt | melting_noncubic.tcl]] to your input file directory of MD++. You can do so by<br />
<br />
<pre><br />
mkdir -p ~/Codes/MD++/scripts/work/melting<br />
cd ~/Codes/MD++/scripts/work/melting<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_cubic.tcl.txt -O melting_cubic.tcl<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_noncubic.tcl.txt -O melting_noncubic.tcl<br />
</pre><br />
<br />
Both files contain a big comment section in the beginning that describes in detail the (16) steps to compute the melting point. These steps are fully automated. In this tutorial, we describe how to run these scripts on a parallel cluster (using wcr.stanford.edu as an example) and what kind of results you should expect.<br />
<br />
===Compile excutable file===<br />
<br />
For each different potential, MD++ has corresponding excutable files.<br />
<pre><br />
Stillinger-Weber Silicon - sw<br />
Stillinger-Weber Germanium - swge<br />
MEAM - meam-lammps, meam-baskes, meam<br />
EAM - eam<br />
Finnis-Sinclair - fs<br />
Tersoff Silicon - tersoff <br />
</pre><br />
<br />
For example, you can compile sworig by doing<br />
<pre><br />
cd ~/Codes/MD++/<br />
make sworig build=R SYS=wcr<br />
</pre><br />
then you will have 'sworig_wcr' file in the directory '~/Codes/MD++/bin'<br />
<br />
For other potentials, refer to the explanations in melting_cubic.tcl<br />
or email to shryu@stanford.edu for any question.<br />
<br />
===Submit jobs===<br />
<br />
Here are the PBS scripts for submitting the jobs on wcr.stanford.edu.<br />
<br />
<pre><br />
cd ~/Codes/MD++/<br />
mkdir runs/Single_Elem_Tm<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt1.pbs.txt -O Melt1.pbs<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt2.pbs.txt -O Melt2.pbs<br />
qsub Melt1.pbs<br />
qsub Melt2.pbs<br />
</pre><br />
<br />
Then you'll get free energy data in <br />
<pre><br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Solid/<br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Liquid/<br />
</pre><br />
<br />
===Analyse data===<br />
<br />
At the end of the calculation, data files and Matlab files will be automatically generated. Transfer these files to a location where you have Matlab installed. Run the following commands in Matlab and you will get the free energy plot (for solid and liquid phases). The melting point is the temperature at which the two free energy curves cross.<br />
<br />
The analysis files are generated in<br />
~/Codes/MD++/runs/Single_Elem_Tm/<br />
<pre><br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
tar_SW_Si<br />
</pre><br />
<br />
to zip necessary data <br />
<pre><br />
cd ~/Codes/MD++/runs/Single_Elem_Tm<br />
./tar_SW_Si<br />
</pre><br />
then you'll have<br />
'inter_SW_Si.tar'<br />
<br />
copy this file to your local computer and unzip.<br />
Then, execute<br />
<pre><br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
</pre><br />
you'll see following output files including all free energy data and error analysis<br />
<pre><br />
Solid Free Energy Info<br />
lattice const = 5.43095<br />
strain = 0.00518<br />
E_0 = -2218.74337 eV ( -4.33348 eV/atom)<br />
F_w = -253.34582 eV ( -0.49482 eV/atom)<br />
F_ha = E_0 + F_w = -2472.08919 eV ( -4.82830 eV/atom)<br />
W_re_to_ha = 5.99169 eV ( 0.01170 eV/atom)<br />
W_ha_to_re = -7.30068 eV ( -0.01426 eV/atom)<br />
F_re - F_ha = (W_ha_to_re - W_re_to_ha)/2 = -6.64619 eV ( -0.01298 eV/atom)<br />
F_re at 1600.000000 = -2478.73538 eV ( -4.84128 eV/atom)<br />
W_T1_to_T11 = 417.87095 eV ( 0.81615 eV/atom)<br />
W_T11_to_T1 = -417.87508 eV ( -0.81616 eV/atom)<br />
dW1 = (W_T1_to_T11 - W_T11_to_T1)/2 = 417.87302 eV ( 0.81616 eV/atom)<br />
F_re at 2000.000000 = -2605.55266 eV ( -5.08897 eV/atom)<br />
<br />
Liquid Free Energy Info<br />
strain = -0.01894<br />
F_ideal = -919.36940 ( -1.79564 eV/atom)<br />
W_re_to_ga = 1888.21102 eV ( 3.68791 eV/atom)<br />
W_ga_to_re = -1887.53306 eV ( -3.68659 eV/atom)<br />
F_re - F_ga = (W_ga_to_re - W_re_to_ga)/2 = -1887.87204 eV ( -3.68725 eV/atom)<br />
W_ga_to_id = -256.29295 eV ( -0.50057 eV/atom)<br />
W_id_to_ga = 256.52177 eV ( 0.50102 eV/atom)<br />
F_ga - F_id = (W_id_to_ga - W_ga_to_id)/2 = 256.40736 eV ( 0.50080 eV/atom)<br />
F_re at 1800.000000 = -2550.83408 eV ( -4.98210 eV/atom)<br />
W_T2_to_T22 = -583.02785 eV ( -1.13873 eV/atom)<br />
W_T22_to_T2 = 583.06542 eV ( 1.13880 eV/atom)<br />
dW2 = (W_T2_to_T22 - W_T22_to_T2)/2 = -583.04664 eV ( -1.13876 eV/atom)<br />
F_re at 1384.615385 = -2386.68534 eV ( -4.66149 eV/atom)<br />
<br />
Error Analysis<br />
<br />
Solid Side<br />
Err_re_to_ha = 0.02769 eV ( 0.00005407 eV/atom)( 0.628 K)<br />
Err_ha_to_re = 0.02815 eV ( 0.00005498 eV/atom)( 0.638 K)<br />
Err_T1_to_T11 = 0.00374 eV ( 0.00000730 eV/atom)( 0.085 K)<br />
Err_T11_to_T1 = 0.00441 eV ( 0.00000861 eV/atom)( 0.100 K)<br />
<br />
Liquid Side<br />
Err_re_to_ga = 0.04793 eV ( 0.00009362 eV/atom)( 1.086 K)<br />
Err_ga_to_re = 0.06601 eV ( 0.00012893 eV/atom)( 1.496 K)<br />
Err_ga_to_id = 0.02557 eV ( 0.00004993 eV/atom)( 0.580 K)<br />
Err_id_to_ga = 0.03421 eV ( 0.00006682 eV/atom)( 0.776 K)<br />
Err_T2_to_T22 = 0.01488 eV ( 0.00002906 eV/atom)( 0.337 K)<br />
Err_T22_to_T2 = 0.01183 eV ( 0.00002311 eV/atom)( 0.268 K)<br />
Max dT = 2.315<br />
Tm= 1695.08000 (K), err= 1.044 Fm = -4.89779 (eV/atom)<br />
L= 165.8833097832 (eV) Ss=0.308452 Sl=0.406313<br />
L= 0.3239908394 (eV/atom) ss=0.000602 sl=0.000794<br />
L= 31206.7976529696 (J/mol) ss=58.027496 sl=76.437718<br />
L= 1111115.7748689589 (J/kg) ss=2066.064819 sl=2721.559412<br />
</pre></div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Computing_Melting_Point_by_Free_Energy_Method&diff=2098
Computing Melting Point by Free Energy Method
2008-11-24T08:56:36Z
<p>Shryu: </p>
<hr />
<div><H1 ALIGN="CENTER><br />
Computing Melting Point by Free Eneregy Method</H1><br />
<br />
<P ALIGN="CENTER"><STRONG>Seunghwa Ryu and Wei Cai</STRONG></P><br />
<br />
This tutorial describes how to compute melting points from free energy methods. We provide the MD++ script files and describe how to use them in detail. The theoretical background is published in <br />
<br />
Comparison of Thermal Properties Predicted by Interatomic Potential Models, ''Modelling and Simulation in Materials Science and Engineering'', '''16''', 085005 (2008). [http://micro.stanford.edu/~caiwei/papers/Ryu08msmse-meltingT.pdf (PDF)].<br />
<br />
===Download files===<br />
<br />
First, you need to install MD++ on your computer by following the instructions on [[MD++_Manuals | MD++ Manuals]].<br />
<br />
Second, copy files [[media:melting_cubic.tcl.txt | melting_cubic.tcl]] and<br />
[[media:melting_noncubic.tcl.txt | melting_noncubic.tcl]] to your input file directory of MD++. You can do so by<br />
<br />
<pre><br />
mkdir -p ~/Codes/MD++/scripts/work/melting<br />
cd ~/Codes/MD++/scripts/work/melting<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_cubic.tcl.txt -O melting_cubic.tcl<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_noncubic.tcl.txt -O melting_noncubic.tcl<br />
</pre><br />
<br />
Both files contain a big comment section in the beginning that describes in detail the (16) steps to compute the melting point. These steps are fully automated. In this tutorial, we describe how to run these scripts on a parallel cluster (using wcr.stanford.edu as an example) and what kind of results you should expect.<br />
<br />
===Compile excutable file===<br />
<br />
For each different potential, MD++ has corresponding excutable files.<br />
Stillinger-Weber Silicon - sw<br />
Stillinger-Weber Germanium - swge<br />
MEAM - meam-lammps, meam-baskes, meam<br />
EAM - eam<br />
Finnis-Sinclair - fs<br />
Tersoff Silicon - tersoff <br />
<br />
For example, you can compile sworig by doing<br />
<pre><br />
cd ~/Codes/MD++/<br />
make sworig build=R SYS=wcr<br />
</pre><br />
then you will have 'sworig_wcr' file in the directory '~/Codes/MD++/bin'<br />
<br />
For other potentials, refer to the explanations in melting_cubic.tcl<br />
or email to shryu@stanford.edu for any question.<br />
<br />
===Submit jobs===<br />
<br />
Here are the PBS scripts for submitting the jobs on wcr.stanford.edu.<br />
<br />
<pre><br />
cd ~/Codes/MD++/<br />
mkdir runs/Single_Elem_Tm<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt1.pbs.txt -O Melt1.pbs<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt2.pbs.txt -O Melt2.pbs<br />
qsub Melt1.pbs<br />
qsub Melt2.pbs<br />
</pre><br />
<br />
Then you'll get free energy data in <br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Solid/<br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Liquid/<br />
<br />
===Analyse data===<br />
<br />
At the end of the calculation, data files and Matlab files will be automatically generated. Transfer these files to a location where you have Matlab installed. Run the following commands in Matlab and you will get the free energy plot (for solid and liquid phases). The melting point is the temperature at which the two free energy curves cross.<br />
<br />
The analysis files are generated in<br />
~/Codes/MD++/runs/Single_Elem_Tm/<br />
<br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
tar_SW_Si<br />
<br />
to zip necessary data <br />
<pre><br />
cd ~/Codes/MD++/runs/Single_Elem_Tm<br />
./tar_SW_Si<br />
</pre><br />
then you'll have<br />
inter_SW_Si.tar <br />
<br />
copy this to your local computer unzip.<br />
then execute<br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
you'll see following output</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Computing_Melting_Point_by_Free_Energy_Method&diff=2097
Computing Melting Point by Free Energy Method
2008-11-24T08:54:35Z
<p>Shryu: /* Compile excutable file */</p>
<hr />
<div><H1 ALIGN="CENTER><br />
Computing Melting Point by Free Eneregy Method</H1><br />
<br />
<P ALIGN="CENTER"><STRONG>Seunghwa Ryu and Wei Cai</STRONG></P><br />
<br />
This tutorial describes how to compute melting points from free energy methods. We provide the MD++ script files and describe how to use them in detail. The theoretical background is published in <br />
<br />
Comparison of Thermal Properties Predicted by Interatomic Potential Models, ''Modelling and Simulation in Materials Science and Engineering'', '''16''', 085005 (2008). [http://micro.stanford.edu/~caiwei/papers/Ryu08msmse-meltingT.pdf (PDF)].<br />
<br />
===Download files===<br />
<br />
First, you need to install MD++ on your computer by following the instructions on [[MD++_Manuals | MD++ Manuals]].<br />
<br />
Second, copy files [[media:melting_cubic.tcl.txt | melting_cubic.tcl]] and<br />
[[media:melting_noncubic.tcl.txt | melting_noncubic.tcl]] to your input file directory of MD++. You can do so by<br />
<br />
<pre><br />
mkdir -p ~/Codes/MD++/scripts/work/melting<br />
cd ~/Codes/MD++/scripts/work/melting<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_cubic.tcl.txt -O melting_cubic.tcl<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_noncubic.tcl.txt -O melting_noncubic.tcl<br />
</pre><br />
<br />
Both files contain a big comment section in the beginning that describes in detail the (16) steps to compute the melting point. These steps are fully automated. In this tutorial, we describe how to run these scripts on a parallel cluster (using wcr.stanford.edu as an example) and what kind of results you should expect.<br />
<br />
===Compile excutable file===<br />
<br />
For each different potential, MD++ has corresponding excutable files.<br />
Stillinger-Weber Silicon - sw<br />
Stillinger-Weber Germanium - swge<br />
MEAM - meam-lammps, meam-baskes, meam<br />
EAM - eam<br />
Finnis-Sinclair - fs<br />
Tersoff Silicon - tersoff <br />
<br />
For example, you can compile sworig by doing<br />
<pre><br />
cd ~/Codes/MD++/<br />
make sworig build=R SYS=wcr<br />
</pre><br />
then you will have 'sworig_wcr' file in the directory '~/Codes/MD++/bin'<br />
<br />
For other potentials, refer to the explanations in melting_cubic.tcl<br />
or email to shryu@stanford.edu for any question.<br />
<br />
===Submit jobs===<br />
<br />
Here are the PBS scripts for submitting the jobs on wcr.stanford.edu.<br />
<br />
<pre><br />
cd ~/Codes/MD++/<br />
mkdir runs/Single_Elem_Tm<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt1.pbs.txt -O Melt1.pbs<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt2.pbs.txt -O Melt2.pbs<br />
qsub Melt1.pbs<br />
qsub Melt2.pbs<br />
</pre><br />
<br />
Then you'll get free energy data in <br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Solid/<br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Liquid/<br />
<br />
===Analyse data===<br />
<br />
At the end of the calculation, data files and Matlab files will be automatically generated. Transfer these files to a location where you have Matlab installed. Run the following commands in Matlab and you will get the free energy plot (for solid and liquid phases). The melting point is the temperature at which the two free energy curves cross.<br />
<br />
The analysis files are generated in<br />
~/Codes/MD++/runs/Single_Elem_Tm/<br />
<br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
tar_SW_Si<br />
<br />
to zip necessary data <br />
<pre><br />
cd ~/Codes/MD++/runs/Single_Elem_Tm<br />
./tar_SW_Si<br />
</pre><br />
then you'll have<br />
inter_SW_Si.tar <br />
<br />
copy this to your local computer unzip.<br />
then execute<br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
<br />
you'll see following output</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Computing_Melting_Point_by_Free_Energy_Method&diff=2096
Computing Melting Point by Free Energy Method
2008-11-24T08:53:58Z
<p>Shryu: </p>
<hr />
<div><H1 ALIGN="CENTER><br />
Computing Melting Point by Free Eneregy Method</H1><br />
<br />
<P ALIGN="CENTER"><STRONG>Seunghwa Ryu and Wei Cai</STRONG></P><br />
<br />
This tutorial describes how to compute melting points from free energy methods. We provide the MD++ script files and describe how to use them in detail. The theoretical background is published in <br />
<br />
Comparison of Thermal Properties Predicted by Interatomic Potential Models, ''Modelling and Simulation in Materials Science and Engineering'', '''16''', 085005 (2008). [http://micro.stanford.edu/~caiwei/papers/Ryu08msmse-meltingT.pdf (PDF)].<br />
<br />
===Download files===<br />
<br />
First, you need to install MD++ on your computer by following the instructions on [[MD++_Manuals | MD++ Manuals]].<br />
<br />
Second, copy files [[media:melting_cubic.tcl.txt | melting_cubic.tcl]] and<br />
[[media:melting_noncubic.tcl.txt | melting_noncubic.tcl]] to your input file directory of MD++. You can do so by<br />
<br />
<pre><br />
mkdir -p ~/Codes/MD++/scripts/work/melting<br />
cd ~/Codes/MD++/scripts/work/melting<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_cubic.tcl.txt -O melting_cubic.tcl<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melting_noncubic.tcl.txt -O melting_noncubic.tcl<br />
</pre><br />
<br />
Both files contain a big comment section in the beginning that describes in detail the (16) steps to compute the melting point. These steps are fully automated. In this tutorial, we describe how to run these scripts on a parallel cluster (using wcr.stanford.edu as an example) and what kind of results you should expect.<br />
<br />
===Compile excutable file===<br />
<br />
For each different potential, MD++ has corresponding excutable files.<br />
Stillinger-Weber Silicon - sw<br />
Stillinger-Weber Germanium - swge<br />
MEAM - meam-lammps, meam-baskes, meam<br />
EAM - eam<br />
Finnis-Sinclair - fs<br />
Tersoff Silicon - tersoff <br />
<br />
For example, you can compile sworig by doing<br />
<pre><br />
cd ~/Codes/MD++/<br />
make sworig build=R SYS=wcr<br />
</pre><br />
then you will have sworig_wcr file in the directory '~/Codes/MD++/bin'<br />
<br />
For other potentials, refer to the explanations in melting_cubic.tcl<br />
or email to shryu@stanford.edu for any question.<br />
<br />
===Submit jobs===<br />
<br />
Here are the PBS scripts for submitting the jobs on wcr.stanford.edu.<br />
<br />
<pre><br />
cd ~/Codes/MD++/<br />
mkdir runs/Single_Elem_Tm<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt1.pbs.txt -O Melt1.pbs<br />
wget http://micro.stanford.edu/mediawiki-1.11.0/images/Melt2.pbs.txt -O Melt2.pbs<br />
qsub Melt1.pbs<br />
qsub Melt2.pbs<br />
</pre><br />
<br />
Then you'll get free energy data in <br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Solid/<br />
~/Codes/MD++/runs/Single_Elem_Tm/SW_Si_Liquid/<br />
<br />
===Analyse data===<br />
<br />
At the end of the calculation, data files and Matlab files will be automatically generated. Transfer these files to a location where you have Matlab installed. Run the following commands in Matlab and you will get the free energy plot (for solid and liquid phases). The melting point is the temperature at which the two free energy curves cross.<br />
<br />
The analysis files are generated in<br />
~/Codes/MD++/runs/Single_Elem_Tm/<br />
<br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
tar_SW_Si<br />
<br />
to zip necessary data <br />
<pre><br />
cd ~/Codes/MD++/runs/Single_Elem_Tm<br />
./tar_SW_Si<br />
</pre><br />
then you'll have<br />
inter_SW_Si.tar <br />
<br />
copy this to your local computer unzip.<br />
then execute<br />
Hessian_SW_Si.m<br />
Free_Energy_of_SW_Si.m<br />
<br />
you'll see following output</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Recommended_Books&diff=1041
Recommended Books
2008-02-22T08:43:23Z
<p>Shryu: /* Experimental Data */</p>
<hr />
<div>==Dislocations==<br />
{|<br />
|-<br />
|<br />
* J. P. Hirth and J. Lothe, ''Theory of Dislocations'', (Wiley, New York, 1982). This is the "Bible" for research in dislocations. It may be somewhat difficult to read if you do not have background in dislocations.<br />
|<br />
|-<br />
|<br />
* D. H. Hull and D. J. Bacon, ''Introduction to Dislocations'', 4th ed. paperback. This book provides a good entry-level introduction to dislocations. It is recommended as a first read if you do not have previous background in dislocations.<br />
|[[Image:Hull_Bacon.JPG|78px]]<br />
|-<br />
|<br />
* J. Weertman and J. R. Weertman, ''Elementary Dislocation Theory'',(Oxford University Press, 1992, paperback). Another reasonable introductory text to dislocations. Chris and Keonwook both started learning dislocation theory from this book. This book costs less than $40!<br />
|[[Image:Weertman.JPG|78px]]<br />
|-<br />
|<br />
* V. V. Bulatov and W. Cai, [http://micro.stanford.edu ''Computer Simulations of Dislocations''] (Oxford University Press, 2006). This book provides numerical models and algorithms useful for modeling dislocations at atomistic and continuum level.<br />
|[[Image:Bulatov_Cai.gif|78px]]<br />
|}<br />
<br />
==Materials Science==<br />
{|<br />
|-<br />
|<br />
* Y.-M. Chiang, W. D. Kingery, D. P. Birnie, ''Physical Ceramics''. This book provides the basic materials science background in ceramics. It is important if your research is in solid oxide fuel cells.<br />
|[[Image:Chiang_Kingery.JPG|78px]]<br />
|-<br />
|<br />
* S. Suresh, ''Fatigue of Metals'', 2nd ed. (Cambridge University Press, 1998). A must read if your research is related to metal fatigue.<br />
|[[Image:SureshFC.JPG|78px]]<br />
|}<br />
<br />
==Computer Simulations==<br />
{|<br />
|-<br />
|<br />
* M. P. Allen and D. J. Tildesley, ''Computer Simulation of Liquids'', (Oxford University Press, 1989). This is the classic text on molecular simulations. But since it is about 20 years old, several important methods are not discussed in this book.<br />
|[[Image:Allen_tildesley.JPG|78px]]<br />
|-<br />
|<br />
* D. Frenkel and B. Smit, ''Understanding Molecular Simulation: From Algorithms to Applications'', 2nd ed. Academic Press, San Diego, CA, 2002. This book provides an update to Allen and Tildesley's book with discussions on more advanced algorithms. It is a good reference but may be too advanced if you did not have any background in molecular simulations. <br />
|[[Image:Frenkel_smit.gif|78px]]<br />
|}<br />
<br />
==Statistical Mechanics==<br />
{|<br />
|-<br />
|<br />
* J. P. Sethna, ''Statistical Mechanics: Entropy, Order Parameters and Complexity'', (Oxford University Press, 2006). [http://pages.physics.cornell.edu/sethna/StatMech/ PDF available]<br />
|[[Image:Sethna.JPG|78px]]<br />
|-<br />
|<br />
* D. Chandler, ''Introduction to Modern Statistical Mechanics'', (Oxford University Press). Somewhat advanced, but very interesting to read. This is the book that Spider man uses when studying nanotechnology.<br />
|[[Image:Chandler.JPG|78px]]<br />
|}<br />
<br />
==Experimental Data==<br />
{|<br />
|-<br />
|<br />
* ''CRC Handbook of Chemistry and Physics'', 88th Edition, 2007-2008 http://208.254.79.26/ whole contents available in PDF form] Include tables of wide-range experimental data including reference papers.<br />
|<br />
|-<br />
|<br />
* ''NIST-JANAF Thermochemical Tables'', 4th Edition, 1998 http://www.nist.gov/srd/PDFfiles/jpcrdM9.pdf whole contents available in PDF form] Include tables of heat capacity, entropy of solids and liquids, including experimental methods and reference papers.<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Recommended_Books&diff=1039
Recommended Books
2008-02-21T00:49:08Z
<p>Shryu: </p>
<hr />
<div>==Dislocations==<br />
{|<br />
|-<br />
|<br />
* J. P. Hirth and J. Lothe, ''Theory of Dislocations'', (Wiley, New York, 1982). This is the "Bible" for research in dislocations. It may be somewhat difficult to read if you do not have background in dislocations.<br />
|<br />
|-<br />
|<br />
* D. H. Hull and D. J. Bacon, ''Introduction to Dislocations'', 4th ed. paperback. This book provides a good entry-level introduction to dislocations. It is recommended as a first read if you do not have previous background in dislocations.<br />
|[[Image:Hull_Bacon.JPG|78px]]<br />
|-<br />
|<br />
* J. Weertman and J. R. Weertman, ''Elementary Dislocation Theory'',(Oxford University Press, 1992, paperback). Another reasonable introductory text to dislocations. Chris and Keonwook both started learning dislocation theory from this book. This book costs less than $40!<br />
|[[Image:Weertman.JPG|78px]]<br />
|-<br />
|<br />
* V. V. Bulatov and W. Cai, [http://micro.stanford.edu ''Computer Simulations of Dislocations''] (Oxford University Press, 2006). This book provides numerical models and algorithms useful for modeling dislocations at atomistic and continuum level.<br />
|[[Image:Bulatov_Cai.gif|78px]]<br />
|}<br />
<br />
==Materials Science==<br />
{|<br />
|-<br />
|<br />
* Y.-M. Chiang, W. D. Kingery, D. P. Birnie, ''Physical Ceramics''. This book provides the basic materials science background in ceramics. It is important if your research is in solid oxide fuel cells.<br />
|[[Image:Chiang_Kingery.JPG|78px]]<br />
|-<br />
|<br />
* S. Suresh, ''Fatigue of Metals'', 2nd ed. (Cambridge University Press, 1998). A must read if your research is related to metal fatigue.<br />
|[[Image:SureshFC.JPG|78px]]<br />
|}<br />
<br />
==Computer Simulations==<br />
{|<br />
|-<br />
|<br />
* M. P. Allen and D. J. Tildesley, ''Computer Simulation of Liquids'', (Oxford University Press, 1989). This is the classic text on molecular simulations. But since it is about 20 years old, several important methods are not discussed in this book.<br />
|[[Image:Allen_tildesley.JPG|78px]]<br />
|-<br />
|<br />
* D. Frenkel and B. Smit, ''Understanding Molecular Simulation: From Algorithms to Applications'', 2nd ed. Academic Press, San Diego, CA, 2002. This book provides an update to Allen and Tildesley's book with discussions on more advanced algorithms. It is a good reference but may be too advanced if you did not have any background in molecular simulations. <br />
|[[Image:Frenkel_smit.gif|78px]]<br />
|}<br />
<br />
==Statistical Mechanics==<br />
{|<br />
|-<br />
|<br />
* J. P. Sethna, ''Statistical Mechanics: Entropy, Order Parameters and Complexity'', (Oxford University Press, 2006). [http://pages.physics.cornell.edu/sethna/StatMech/ PDF available]<br />
|[[Image:Sethna.JPG|78px]]<br />
|-<br />
|<br />
* D. Chandler, ''Introduction to Modern Statistical Mechanics'', (Oxford University Press). Somewhat advanced, but very interesting to read. This is the book that Spider man uses when studying nanotechnology.<br />
|[[Image:Chandler.JPG|78px]]<br />
|}<br />
<br />
==Experimental Data==<br />
{|<br />
|-<br />
|<br />
* ''CRC Handbook of Chemistry and Physics'', 88th Edition, 2007-2008<br />
[http://208.254.79.26/ whole contents available in PDF form]<br />
Include tables of wide-range experimental data including reference papers.<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Group_Presentation_Schedule&diff=934
Group Presentation Schedule
2008-02-14T21:14:29Z
<p>Shryu: </p>
<hr />
<div>Please contact '''Jie Yin''' to sign up for your presentation at the group meeting.<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Winter Quarter Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Yongxing Shen || 1/17 || Modeling electrostatic force microscopy and related technique (PhD thesis defense practice)<br />
|-<br />
| Sylvie Aubry || 1/23 || Calculation of Si thermal boundary resistance<br />
|-<br />
| Seokwoo Lee || 1/30 || Dislocation Dynamics in thin films<br />
|-<br />
| Jie Yin || 2/6 || Anisotropic elasticity I<br />
|-<br />
| Hark Lee || 2/13 || TBD<br />
|-<br />
| Billy Cash || 2/20 || Fatigue in metals I<br />
|-<br />
| Steven Fitzgerald || 2/27 || Fusions<br />
|-<br />
| || 3/5 ||<br />
|-<br />
| || 3/12 ||<br />
|-<br />
| Seunghwa Ryu || 3/19 || Accurate Melting Point Computation I (Co-exist Method)<br />
|-<br />
| || 3/26* || <br />
|-<br />
| Seunghwa Ryu || 4/2 || Accurate Melting Point Computation II (Free Energy Method pt. 1)<br />
|-<br />
| Seunghwa Ryu || 4/9 || Accurate Melting Point Computation III (Free Energy Method pt. 2)<br />
<br />
|}<br />
<br />
*: The week of spring break. The spring quarter starts on Apr 1st.<br />
<br />
<br><br />
<br />
<br />
{| border="1"<br />
|+ '''Tentative topics'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || TBD || Qual practice talk <br />
|-<br />
| Jie Yin || TBD || Anisotropic Elasticity<br />
|-<br />
| Hark Lee || TBD, W Qrt ? || Qual practice talk<br />
|-<br />
| Yongxing Shen || TBD, W Qrt || Euler-Bernoulli beam theory applied to AFM<br />
|-<br />
| Bill Cash || TBD || Fatigue in metals II<br />
|-<br />
| Seunghwa Ryu || S Qrt || Binary Phase Diagram from Atomistic Simulation I (Related Physics Review)<br />
|-<br />
| Seunghwa Ryu || S Qrt || Binary Phase Diagram from Atomistic Simulation II (Realization via Simulation)<br />
|-<br />
| Seunghwa Ryu || S Qrt || Fast Quantum Entanglement Minimization Algorithm (Pure mathematical approach)<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Group_Presentation_Schedule&diff=933
Group Presentation Schedule
2008-02-13T10:18:03Z
<p>Shryu: </p>
<hr />
<div>Please contact '''Jie Yin''' to sign up for your presentation at the group meeting.<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Winter Quarter Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Yongxing Shen || 1/17 || Modeling electrostatic force microscopy and related technique (PhD thesis defense practice)<br />
|-<br />
| Sylvie Aubry || 1/23 || Calculation of Si thermal boundary resistance<br />
|-<br />
| Seokwoo Lee || 1/30 || Dislocation Dynamics in thin films<br />
|-<br />
| Jie Yin || 2/6 || Anisotropic elasticity I<br />
|-<br />
| Hark Lee || 2/13 || TBD<br />
|-<br />
| Billy Cash || 2/20 || Fatigue in metals I<br />
|-<br />
| Steven Fitzgerald || 2/27 || Fusions<br />
|-<br />
| || 3/5 ||<br />
|-<br />
| || 3/12 ||<br />
|-<br />
| Seunghwa Ryu || 3/19 || Accurate Melting Point Computation I (Co-exist Method)<br />
|-<br />
| Seunghwa Ryu || 3/26* || Accurate Melting Point Computation II (Free Energy Method pt. 1)<br />
|-<br />
| Seunghwa Ryu || 4/2 || Accurate Melting Point Computation III (Free Energy Method pt. 2)<br />
|}<br />
<br />
*: The week of spring break. The spring quarter starts on Apr 1st.<br />
<br />
<br><br />
<br />
<br />
{| border="1"<br />
|+ '''Tentative topics'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || TBD || Qual practice talk <br />
|-<br />
| Jie Yin || TBD || Anisotropic Elasticity<br />
|-<br />
| Hark Lee || TBD, W Qrt ? || Qual practice talk<br />
|-<br />
| Yongxing Shen || TBD, W Qrt || Euler-Bernoulli beam theory applied to AFM<br />
|-<br />
| Bill Cash || TBD || Fatigue in metals II<br />
|-<br />
| Seunghwa Ryu || S Qrt || Binary Phase Diagram from Atomistic Simulation I (Related Physics Review)<br />
|-<br />
| Seunghwa Ryu || S Qrt || Binary Phase Diagram from Atomistic Simulation II (Realization via Simulation)<br />
|-<br />
| Seunghwa Ryu || S Qrt || Fast Quantum Entanglement Minimization Algorithm (Pure mathematical approach)<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Group_Presentation_Schedule&diff=932
Group Presentation Schedule
2008-02-13T10:12:42Z
<p>Shryu: </p>
<hr />
<div>Please contact '''Jie Yin''' to sign up for your presentation at the group meeting.<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Winter Quarter Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Yongxing Shen || 1/17 || Modeling electrostatic force microscopy and related technique (PhD thesis defense practice)<br />
|-<br />
| Sylvie Aubry || 1/23 || Calculation of Si thermal boundary resistance<br />
|-<br />
| Seokwoo Lee || 1/30 || Dislocation Dynamics in thin films<br />
|-<br />
| Jie Yin || 2/6 || Anisotropic elasticity I<br />
|-<br />
| Hark Lee || 2/13 || TBD<br />
|-<br />
| Billy Cash || 2/20 || Fatigue in metals I<br />
|-<br />
| Steven Fitzgerald || 2/27 || Fusions<br />
|-<br />
| || 3/5 ||<br />
|-<br />
| || 3/12 ||<br />
|-<br />
| Seunghwa Ryu || 3/19 || Accurate Melting Point Computation I (Co-exist Method)<br />
|-<br />
| Seunghwa Ryu || 3/26* || Accurate Melting Point Computation II (Free Energy Method pt. 1)<br />
|-<br />
| Seunghwa Ryu || 4/2 || Accurate Melting Point Computation III (Free Energy Method pt. 2)<br />
|}<br />
<br />
*: The week of spring break. The spring quarter starts on Apr 1st.<br />
<br />
<br><br />
<br />
<br />
{| border="1"<br />
|+ '''Tentative topics'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || TBD || Qual practice talk <br />
|-<br />
| Jie Yin || TBD || Anisotropic Elasticity<br />
|-<br />
| Hark Lee || TBD, W Qrt ? || Qual practice talk<br />
|-<br />
| Yongxing Shen || TBD, W Qrt || Euler-Bernoulli beam theory applied to AFM<br />
|-<br />
| Bill Cash || TBD || Fatigue in metals II<br />
|-<br />
| Seunghwa Ryu || Apr or May || Construction of Binary Phase Diagram from Atomistic Simulation I<br />
|-<br />
| Seunghwa Ryu || Apr or May || Construction of Binary Phase Diagram from Atomistic Simulation II<br />
|-<br />
| Seunghwa Ryu || May or June || Fast Quantum Entanglement Minimization Algorithm<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Group_Presentation_Schedule&diff=931
Group Presentation Schedule
2008-02-13T10:11:33Z
<p>Shryu: </p>
<hr />
<div>Please contact '''Jie Yin''' to sign up for your presentation at the group meeting.<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Winter Quarter Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Yongxing Shen || 1/17 || Modeling electrostatic force microscopy and related technique (PhD thesis defense practice)<br />
|-<br />
| Sylvie Aubry || 1/23 || Calculation of Si thermal boundary resistance<br />
|-<br />
| Seokwoo Lee || 1/30 || Dislocation Dynamics in thin films<br />
|-<br />
| Jie Yin || 2/6 || Anisotropic elasticity I<br />
|-<br />
| Hark Lee || 2/13 || TBD<br />
|-<br />
| Billy Cash || 2/20 || Fatigue in metals I<br />
|-<br />
| Steven Fitzgerald || 2/27 || Fusions<br />
|-<br />
| || 3/5 ||<br />
|-<br />
| || 3/12 ||<br />
|-<br />
| Seunghwa Ryu || 3/19 || Accurate Melting Point Computation from Atomistic Simulation I<br />
|-<br />
| Seunghwa Ryu || 3/26* || Accurate Melting Point Computation from Atomistic Simulation II<br />
|-<br />
| Seunghwa Ryu || 4/2 || Accurate Melting Point Computation from Atomistic Simulation III<br />
|}<br />
<br />
*: The week of spring break. The spring quarter starts on Apr 1st.<br />
<br />
<br><br />
<br />
<br />
{| border="1"<br />
|+ '''Tentative topics'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || TBD || Qual practice talk <br />
|-<br />
| Jie Yin || TBD || Anisotropic Elasticity<br />
|-<br />
| Hark Lee || TBD, W Qrt ? || Qual practice talk<br />
|-<br />
| Yongxing Shen || TBD, W Qrt || Euler-Bernoulli beam theory applied to AFM<br />
|-<br />
| Bill Cash || TBD || Fatigue in metals II<br />
|-<br />
| Seunghwa Ryu || Apr or May || Construction of Binary Phase Diagram from Atomistic Simulation I<br />
|-<br />
| Seunghwa Ryu || Apr or May || Construction of Binary Phase Diagram from Atomistic Simulation II<br />
|-<br />
| Seunghwa Ryu || May or June || Fast Quantum Entanglement Minimization Algorithm<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=428
Seunghwa Ryu
2008-02-10T11:13:20Z
<p>Shryu: </p>
<hr />
<div>* Title<br />
** Physics Phd Candiate<br />
* Location <br />
**496 Lomita Mall (a.k.a. Durand Building), rm 202, Stanford, CA94305 <br />
* Email<br />
** shryu at stanford dot edu<br />
* Phone<br />
** 650-353-1277<br />
* Reserach Interest<br />
** Nanowire Growth Simulation (Thesis Topic)<br />
** Quantum Entanglement Computation (side work)<br />
* Education<br />
** 09/2004 - current : PhD in physics, Stanford <br />
** 09/2004 - 01/2006 : MS. in physics, Stanford <br />
** 03/2000 - 02/2004 : BS. in Physics, KAIST, Korea<br />
* CV<br />
** http://www.stanford.edu/~shryu/Seunghwa_CV.pdf</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=427
Seunghwa Ryu
2008-02-10T10:13:10Z
<p>Shryu: </p>
<hr />
<div>* Title<br />
** Physics Phd Candiate<br />
* Location <br />
**496 Lomita Mall (a.k.a. Durand Building), rm 202, Stanford, CA94305 <br />
* Email<br />
** shryu at stanford dot edu<br />
* Phone<br />
** 650-353-1277<br />
* Reserach Interest<br />
** Nanowire Growth Simulation (Thesis Topic)<br />
** Quantum Entanglement Computation (side work)<br />
* Education<br />
** 09/2004 - current : PhD in physics, Stanford <br />
** 09/2004 - Jan/2006 : MS. in physics, Stanford <br />
** 03/2000 - 02/2004 : BS. in Physics, KAIST, Korea<br />
* CV<br />
** http://www.stanford.edu/~shryu/Seunghwa_CV.pdf</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=426
Seunghwa Ryu
2008-02-10T10:12:35Z
<p>Shryu: </p>
<hr />
<div>* Title<br />
** Physics Phd Candiate<br />
* Location <br />
**496 Lomita Mall (a.k.a. Durand Building), rm 202, Stanford, CA94305 <br />
* Email<br />
** shryu at stanford dot edu<br />
* Phone<br />
** 650-353-1277<br />
* Reserach Interest<br />
** Nanowire Growth Simulation (Thesis Topic)<br />
** Quantum Entanglement Computation (side work)<br />
* Education<br />
** 09/2004 - current : PhD in physics, Stanford <br />
** 09/2004 - Jan/2006 : MS. in physics, Stanford <br />
** 03/2000 - 02/2004 : BS. in Physics, KAIST<br />
* CV<br />
** http://www.stanford.edu/~shryu/Seunghwa_CV.pdf</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=425
Seunghwa Ryu
2008-02-10T10:11:19Z
<p>Shryu: </p>
<hr />
<div>* Title<br />
** Physics Phd Candiate<br />
* Location <br />
**496 Lomita Mall (a.k.a. Durand Building), rm 202, Stanford, CA94305 <br />
* Email<br />
** shryu at stanford dot edu<br />
* Phone<br />
** 650-353-1277<br />
* Reserach Interest<br />
** Nanowire Growth Simulation <br />
** Quantum Entanglement Computation<br />
* Education<br />
** 09/2004 - current : PhD in physics, Stanford <br />
** 09/2004 - Jan/2006 : MS. in physics, Stanford <br />
** 03/2000 - 02/2004 : BS. in Physics, KAIST<br />
* CV<br />
** http://www.stanford.edu/~shryu/Seunghwa_CV.pdf</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=424
Seunghwa Ryu
2008-02-10T10:10:41Z
<p>Shryu: </p>
<hr />
<div>* Title: Physics Phd Candiate<br />
* Location <br />
**496 Lomita Mall (a.k.a. Durand Building), rm 202, Stanford, CA94305 <br />
* Email: shryu at stanford dot edu<br />
* Phone: 650-353-1277<br />
* Reserach Interest<br />
** Nanowire Growth Simulation <br />
** Quantum Entanglement Computation<br />
* Education<br />
** 09/2004 - current : PhD in physics, Stanford <br />
** 09/2004 - Jan/2006 : MS. in physics, Stanford <br />
** 03/2000 - 02/2004 : BS. in Physics, KAIST<br />
* CV<br />
** http://www.stanford.edu/~shryu/Seunghwa_CV.pdf</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Group_Presentation_Schedule&diff=928
Group Presentation Schedule
2008-02-10T10:06:41Z
<p>Shryu: </p>
<hr />
<div>Please contact '''Jie Yin''' to sign up for your presentation at the group meeting.<br />
<br />
<br />
{| border="1"<br />
|+ '''2007-2008 Winter Quarter Weekly Presentation Schedule'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Yongxing Shen || 1/17 || Modeling electrostatic force microscopy and related technique (PhD thesis defense practice)<br />
|-<br />
| Sylvie Aubry || 1/23 || Calculation of Si thermal boundary resistance<br />
|-<br />
| Seokwoo Lee || 1/30 || Dislocation Dynamics in thin films<br />
|-<br />
| Jie Yin || 2/6 || Anisotropic elasticity I<br />
|-<br />
| Hark Lee || 2/13 || TBD<br />
|-<br />
| Billy Cash || 2/20 || Fatigue in metals I<br />
|-<br />
| Billy Cash || 2/27 || Fatigue in metals II<br />
|-<br />
| || 3/5 ||<br />
|-<br />
| || 3/12 ||<br />
|-<br />
| Seunghwa Ryu || 3/19 || Melting Point Calculation I<br />
|-<br />
| Seunghwa Ryu || 3/26* || Melting Point Calculation II<br />
|-<br />
| Seunghwa Ryu || 4/2 || Melting Point Calculation III<br />
<br />
|}<br />
<br />
*: The week of spring break. The spring quarter starts on Apr 1st.<br />
<br />
<br><br />
<br />
<br />
{| border="1"<br />
|+ '''Tentative topics'''<br />
|-<br />
| Name || Date || Topic <br />
|-<br />
| Jie Yin || TBD || Qual practice talk <br />
|-<br />
| Jie Yin || TBD || Anisotropic Elasticity<br />
|-<br />
| Hark Lee || TBD, W Qrt ? || Qual practice talk<br />
|-<br />
| Yongxing Shen || TBD, W Qrt || Euler-Bernoulli beam theory applied to AFM<br />
|-<br />
| Steven Fitzgerald || TBD, W Qrt || Fusions<br />
|-<br />
| Seunghwa Ryu || TBD || Construction of Binary Phase Diagram from Atomistic Simulation I<br />
|-<br />
| Seunghwa Ryu || TBD || Construction of Binary Phase Diagram from Atomistic Simulation II<br />
|-<br />
| Seunghwa Ryu || TBD || Fast Quantum Entanglement Minimization Algorithm<br />
|}</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=423
Seunghwa Ryu
2008-02-10T10:01:34Z
<p>Shryu: </p>
<hr />
<div>* Title: Physics Phd Candiate<br />
* Location: <br />
**496 Lomita Mall (a.k.a. Durand Building)<br />
**rm 202, Stanford, CA94305 <br />
* Email: shryu at stanford dot edu<br />
* Phone: 650-353-1277<br />
* Reserach Interest: <br />
** Nanowire Growth Simulation <br />
** Quantum Entanglement Computation<br />
* Education<br />
** 09/2004 - current : PhD in physics, Stanford <br />
** 09/2004 - Jan/2006 : MS. in physics, Stanford <br />
** 03/2000 - 02/2004 : BS. in Physics, KAIST<br />
* CV<br />
** http://www.stanford.edu/~shryu/Seunghwa_CV.pdf</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=422
Seunghwa Ryu
2008-02-10T10:00:47Z
<p>Shryu: </p>
<hr />
<div>* Title: Physics Phd Candiate<br />
* Location: <br />
**496 Lomita Mall (a.k.a. Durand Building)<br />
**rm 202, Stanford, CA94305 <br />
* Email: shryu at stanford dot edu<br />
* Phone: 650-353-1277<br />
* Reserach Interest: <br />
** Nanowire Growth Simulation <br />
** Quantum Entanglement Computation<br />
* Education<br />
** 09/2004 - current : PhD Stanford <br />
** 09/2004 - Jan/2006 : MS. Stanford <br />
** 03/2000 - 02/2004 : BS. KAIST<br />
* CV<br />
** http://www.stanford.edu/~shryu/Seunghwa_CV.pdf</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=421
Seunghwa Ryu
2008-02-10T09:49:40Z
<p>Shryu: </p>
<hr />
<div>Title: Physics Phd Candiate<br />
Reserach Interest: <br />
* Nanowire Growth Simulation <br />
* Quantum Entanglement Computation<br />
Education</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Seunghwa_Ryu&diff=420
Seunghwa Ryu
2007-11-29T21:23:57Z
<p>Shryu: New page: My research topic is ....</p>
<hr />
<div>My research topic is ....</div>
Shryu
http://micro.stanford.edu/mediawiki/index.php?title=Main_Page&diff=194
Main Page
2007-11-29T21:23:29Z
<p>Shryu: /* Group Members */</p>
<hr />
<div>Welcome to the Micro and Nano Mechanics Group at Stanford<br />
<br />
==Group Members==<br />
<br />
[[Wei Cai]]<br />
<br />
[[Keonwook Kang]]<br />
<br />
[[Seunghwa Ryu]]<br />
<br />
==Research Projects==<br />
<br />
[[Nanowire Mechanics]]<br />
<br />
[[Dislocation Core]]<br />
<br />
==Tutorials==<br />
[[Dislocations]]<br />
<br />
[[Atomistic Calculations]]<br />
<br />
[[Unix Basics]]<br />
<br />
[[Computer Setup]]<br />
<br />
[[MD++ Manuals]]<br />
<br />
==Fun==<br />
<br />
[[Local Restaurants]]</div>
Shryu