Hi, Thanks for your answers.
I apologize for my impatient comments yesterday. Actually I tried so hard to find those properties. And you know as graduate student we are always under pressure. I thought this as a simple problem, since Si results were presented in the papers that were referenced most, but I could not find if mentioned explicitly the number of atoms, and all other input variables necessary to reproduce the results. I found the similar question in archive but that was not answered in detail. The answer was what is written in the paper. I was not planning to work on 64-Si, rather I wanted to gain confidence reproducing the results first. However, I think an explicit example is very useful for new users. One example with detail explanation will reduce the number of questions of users about nuances behind the input choices a lot. Sophia. On Tue, Apr 1, 2008 at 3:36 AM, Marcos Verissimo Alves < [EMAIL PROTECTED]> wrote: > Sophie, > > >> > >> My calculation for 64-atoms: > > Why do you have to go to 64 atoms to get this kind of quantities? The unit > cell should suffice. Nevertheless: > > >> 1) Lattice Constant: I changed the lattice parameter and find the > energy > >> and plotting that I find 5.45 Ang, instead of 5.40 as reported in the > >> paper. Why this so different? > >> > >> 2) Bulk modulus: With B = V*Curvature = V*2c3 = 15 MPa, which is far > >> less > >> than 98.6 MPa? Why? Here, c3 is defined as E =C1 + c2*V + c3*V^2, curve > >> fit to the E vs V curve. > > This is my opinion based from previous personal experience, and doesn't > mean that the explanation is completely correct. The more experienced > users can correct me if I'm wrong. One of the possible reasons is that you > have a combination of two factors that are leading you to some error > there: using a 64 atom cell and using a quadratic equation as an equation > of state. If you are using a quadratic equation, you have to be very sure > that your total energy really behaves so, because your DFT calculations > will show terms in the total energy that would be of a higher order - by > choosing a quadratic dispersion, you are choosing to ignore these terms. > Anyway, using a quadratic dispersion is never a very good procedure. In my > experience, a cubic would be better because of the reasons stated above. > > For the supercell you are using, any small change in the lattice parameter > will lead to rather large energy differences. So you might be missing a > good description in the minimum of the curve, which is one factor that > could be affecting your results - after all, the bulk modulus is > calculated at the point of minimum energy. If you are not describing it > well through your fit, then you will have poor results. A much better > procedure would be to use the unit cell, and a Murnaghan fit to determine > directly B, B' and V_0, the equilibrium volume. Using a small cell, the > energy changes will be smaller with changes in the lattice parameters, in > comparison to the 64-atom supercell. Moreover, the Murnaghan (or > Birch-Murnaghan) equation of state can handle larger changes while giving > you rather accurate results. > > As to the cohesive energy, I can't think of anything right now, that could > be a source to the brutal discrepancies you are seeing. Maybe the k-point > sampling is not so good but, being Si and a 64-atom supercell, I would not > really guess that would be the problem. > > Best regards, > > Marcos > > >> > >> 3) Cohesive Energy: I find energy per atom 107.759eV in the bulk. Now > to > >> calculate energy per atom I tried to use the suggestions given at > >> http://www.mail-archive.com/[email protected]/msg03118.html > >> I get -7.49eV/atom as pseudopotential calculation, and -576.38eV/atom > as > >> ae. So, which one is the energy of the free atom. None is a good one > for > >> comparing with the energy I got from bulk to find the cohesive energy. > >> > >> I really tried to search the archive and find solutions to these. I > >> could > >> not find any explicit answers. I am not sure what I am missing. *Some > >> expert's simple directions can solve the problem right away*. I really > >> need this help. > >> > >> > >> Thanks in advance, > >> Sophia > >> Univ. of California - Berkeley > >> > >> Attached fdf file > >> > >> > ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ > >> > >> # > >> > ----------------------------------------------------------------------------- > >> # FDF for a cubic c-Si supercell with 64 atoms > >> # > >> # E. Artacho, April 1999 > >> # > >> > ----------------------------------------------------------------------------- > >> > >> SystemName 64-atom Si > >> SystemLabel Si > >> > >> NumberOfAtoms 64 > >> NumberOfSpecies 1 > >> > >> %block ChemicalSpeciesLabel > >> 1 14 Si > >> %endblock ChemicalSpeciesLabel > >> > >> PAO.BasisSize DZP > >> PAO.EnergyShift 20 meV > >> > >> %Block PAO.Basis > >> Si 3 -0.46385 > >> n=3 0 2 E 15.42551 4.96988 > >> 7.00000 4.37722 > >> 1.00000 1.00000 > >> n=3 1 2 E 4.69636 3.83128 > >> 7.00000 4.09123 > >> 1.00000 1.00000 > >> n=3 2 1 E 11.96912 0.03131 > >> 4.55426 > >> 1.00000 > >> %EndBlock PAO.Basis > >> > >> > >> LatticeConstant 5.430 Ang > >> %block LatticeVectors > >> 2.000 0.000 0.000 > >> 0.000 2.000 0.000 > >> 0.000 0.000 2.000 > >> %endblock LatticeVectors > >> > >> %block kgrid_Monkhorst_Pack > >> 2 0 0 0.0 > >> 0 2 0 0.0 > >> 0 0 2 0.0 > >> %endblock kgrid_Monkhorst_Pack > >> > >> > >> MeshCutoff 40.0 Ry > >> > >> MaxSCFIterations 100 > >> DM.MixingWeight 0.3 > >> DM.NumberPulay 3 > >> DM.Tolerance 1.d-3 > >> DM.UseSaveDM > >> XC.functional LDA > >> XC.authors CA > >> > >> > >> SolutionMethod diagon > >> ElectronicTemperature 25 meV > >> > >> WriteForces true > >> WriteCoorStep true > >> > >> MD.TypeOfRun cg > >> MD.NumCGsteps 0 > >> MD.MaxCGDispl 0.1 Ang > >> MD.MaxForceTol 0.01 eV/Ang # earler 0.04 > >> SaveRho true > >> > >> AtomicCoordinatesFormat ScaledCartesian > >> %block AtomicCoordinatesAndAtomicSpecies > >> 0.00000 0.00000 0.00000 1 > >> 0.00000 0.50000 0.50000 1 > >> 0.25000 0.25000 0.75000 1 > >> 0.25000 0.75000 0.25000 1 > >> 0.50000 0.00000 0.50000 1 > >> 0.50000 0.50000 0.00000 1 > >> 0.75000 0.25000 0.25000 1 > >> 0.75000 0.75000 0.75000 1 > >> 0.00000 0.00000 1.00000 1 > >> 0.00000 0.50000 1.50000 1 > >> 0.25000 0.25000 1.75000 1 > >> 0.25000 0.75000 1.25000 1 > >> 0.50000 0.00000 1.50000 1 > >> 0.50000 0.50000 1.00000 1 > >> 0.75000 0.25000 1.25000 1 > >> 0.75000 0.75000 1.75000 1 > >> 0.00000 1.00000 0.00000 1 > >> 0.00000 1.50000 0.50000 1 > >> 0.25000 1.25000 0.75000 1 > >> 0.25000 1.75000 0.25000 1 > >> 0.50000 1.00000 0.50000 1 > >> 0.50000 1.50000 0.00000 1 > >> 0.75000 1.25000 0.25000 1 > >> 0.75000 1.75000 0.75000 1 > >> 0.00000 1.00000 1.00000 1 > >> 0.00000 1.50000 1.50000 1 > >> 0.25000 1.25000 1.75000 1 > >> 0.25000 1.75000 1.25000 1 > >> 0.50000 1.00000 1.50000 1 > >> 0.50000 1.50000 1.00000 1 > >> 0.75000 1.25000 1.25000 1 > >> 0.75000 1.75000 1.75000 1 > >> 1.00000 0.00000 0.00000 1 > >> 1.00000 0.50000 0.50000 1 > >> 1.25000 0.25000 0.75000 1 > >> 1.25000 0.75000 0.25000 1 > >> 1.50000 0.00000 0.50000 1 > >> 1.50000 0.50000 0.00000 1 > >> 1.75000 0.25000 0.25000 1 > >> 1.75000 0.75000 0.75000 1 > >> 1.00000 0.00000 1.00000 1 > >> 1.00000 0.50000 1.50000 1 > >> 1.25000 0.25000 1.75000 1 > >> 1.25000 0.75000 1.25000 1 > >> 1.50000 0.00000 1.50000 1 > >> 1.50000 0.50000 1.00000 1 > >> 1.75000 0.25000 1.25000 1 > >> 1.75000 0.75000 1.75000 1 > >> 1.00000 1.00000 0.00000 1 > >> 1.00000 1.50000 0.50000 1 > >> 1.25000 1.25000 0.75000 1 > >> 1.25000 1.75000 0.25000 1 > >> 1.50000 1.00000 0.50000 1 > >> 1.50000 1.50000 0.00000 1 > >> 1.75000 1.25000 0.25000 1 > >> 1.75000 1.75000 0.75000 1 > >> 1.00000 1.00000 1.00000 1 > >> 1.00000 1.50000 1.50000 1 > >> 1.25000 1.25000 1.75000 1 > >> 1.25000 1.75000 1.25000 1 > >> 1.50000 1.00000 1.50000 1 > >> 1.50000 1.50000 1.00000 1 > >> 1.75000 1.25000 1.25000 1 > >> 1.75000 1.75000 1.75000 1 > >> %endblock AtomicCoordinatesAndAtomicSpecies > >> > > > > > -- > Dr. Marcos Verissimo Alves > Post-Doctoral Fellow > Unité de Physico-Chimie et de Physique des Matériaux (PCPM) > Université Catholique de Louvain > 1 Place Croix du Sud, B-1348 > Louvain-la-Neuve > Belgique > > ------ > > Gort, Klaatu barada nikto. Klaatu barada nikto. Klaatu barada nikto. >
