Hi Ravi, 1) I am modeling nanowires with periodic boundary conditions along the > length - which leads to modeling infinitely long wires. If I double the > length of the nanowire unit cell along the direction in which polarization > is being calculated, then should I expect to see a doubled value for > polarization as well? >
No idea whatsoever - shame on me: I know absolutely nothing of the modern theory of polarization. However, since I will most probably be doing this at some point in my new post-doc, I downloaded a pdf file from Psi-K ( www.psik.org), one of the highlights of the month. It is a reviewwritten by Raffaele Resta, who is one of the main authors of the modern theory, if I'm not mistaken. Maybe it could be a good starting point to search for some hints to the answer you want. But, if no one else (here or elsewhere) can answer this, and if you're craving for a quick answer ans are willing to play "in-silico experimentalist", you can always check this for Si, which is fast to run and well-known (or BaTiO3 - files and pseudos are ready in the siesta examples!), by calculating the polarization for a unit cell and then calculating the polarization of the doubled unit cell, for minimal values of MeshCutoff and kgrid_Monkhorst_Pack on a DZ basis set... > > 2) On straining a unit cell, sometimes I notice weird jumps and drops in > the polarization values. The slope of polarization vs strain seems to be the > same accept at the jumps leading to discontinuity. Any ideas, what might be > happening? > > I can only give you a (very) wild guess: check if, at the values where you have the jumps, you have anything worthy of notice going on with the atomic positions. However, this could well **not** be the case, especially if your strain values are not very high - could be simply changes in the electronic structure due to the strain (valence and conduction bands changing place, etc). Keep in mind that these are REALLY guesses, with not much value since I'm not really acquainted to the elements involved in the calculation of polarization. Cheers, Marcos > Thanks in advance. > > Best Regards, > Ravi > > 2009/11/30 Marcos Veríssimo Alves <[email protected]> > >> Ravi, >> >> >> As far as I know, anything with a band gap is an insulator... The term >> "semiconductor" is generally employed only to discriminate insulators with a >> band gap below a certain value, say, 4 eV. So your ZnO wires are also >> insulators - provided, of course, they have a non-zero band gap from your >> calculations :) >> >> Cheers, >> >> Marcos >> >> >> On Mon, Nov 30, 2009 at 4:14 PM, Ravi Agrawal < >> [email protected]> wrote: >> >>> Hi Chris/Andrei, >>> >>> Your ongoing discussion is interesting and quite insightful for me, as I >>> am also planning to pursue some polarization calculations. >>> >>> My system will be zinc oxide nanowires - with free surfaces and >>> periodicity induced along the length (modeling infinitely long wires). >>> I am wondering if the approach of calculating polarization using the >>> Berry-Phase approach (by employing Polarization Grids) is suitable for this >>> scenario. ZnO has a band gap and is semiconducting, but not an "insulator". >>> This question came up as the SIESTA manual suggests that Berry-phase >>> approach is suitable only for insulators. >>> >>> Please, shed some light on this issue as well, in lieu of your ongoing >>> discussion. >>> >>> Thanks, >>> Ravi >>> >>> Northwestern University >>> Evanston, IL 60208 - 3111 >>> >>> >>> >>> On Mon, Nov 30, 2009 at 9:03 AM, <[email protected]> wrote: >>> >>>> Dear Chris, >>>> I think you attempt to cheat the code and to overcome its >>>> naive protection against an attempt to calculate polarization >>>> on a metal system. What matters 12 or 13 silver atoms, >>>> a little bit metal is a metal, in the context given. >>>> Are you sure you are doing >>>> something reasonable even when "it performs the calculation!" ? >>>> Do you have a band gap in your silver-on-whatever system? >>>> The band gap is NEEDED for calculating the polarization, Berry phase or >>>> not. >>>> >>>> Best regards >>>> >>>> Andrei Postnikov >>>> >>>> > Thanks Andrei, >>>> > >>>> > I was trying to use the silver system as a simple example. What I'm >>>> truly >>>> > interested in calculating is the polarization in a polymer-metal >>>> > nanocomposite; and what I've found is that with a matrix of 400 atoms >>>> and >>>> > a >>>> > metal (i.e. silver) inclusion with an odd number of atoms (I tried >>>> with >>>> > 13) >>>> > then SIESTA won't do the polarization calculation: odd total number of >>>> > electrons. But if the inclusion is only 12 atoms, it performs the >>>> > calculation! I also tried giving the system a single positive charge >>>> but >>>> > to >>>> > no avail. >>>> > >>>> > Chris Rowan >>>> > M.Sc. Candidate >>>> > University of Victoria >>>> > Canada >>>> >>>> >>> >> > > > -- > PhD Candidate > Espinosa Research Group > Mechanical Engineering > Northwestern University > 2145 Sheridan Road, Tech Bldg. > Evanston, IL 60208 - 3111 > Phone: (847) - 467 - 7673 >
