 This thread is locked.Only browsing is available.
 Top Page > Browsing A question about the momentum matrix elements calculation Date: 2009/09/10 19:31 Name: Hung-Lung Huang   Dear Prof. OzakiI am a Ph.D. student of Prof. G.Y. Guo in National Taiwan University. I am trying to calculate the momentum matrix elements in OpenMX. But there are some problems I cannot overcome.According to the technical note, electric polarization is calculated by Berry Phase. The Eq. 11 shows the overlap matrix with position operator has five summation indices n,i, j, alpha and beta. However, the overlap matrix with position operator has four indices in polB.c. The overlap matrix with position operator forms as OLPpox,y,z[ct_AN][h_AN][i1][j1].The index h_AN has two meanings: lattice vector index n and atomic site index j. Theindex n sums over all lattice points of the nearest neighbors. Would you tell me howto determine the lattice points of the nearest neighbors?When I calculated the momentum matrix elements P_ij that is proportional tosummation of OLPpox,y,z in polB.c. Nevertheless, the complex conjugate of P_ij is notequal to P_ji. Would you please give me suggestions about the wrong results? Thanks and best wishes Page: Re: A question about the momentum matrix elements calculation ( No.1 ) Date: 2009/09/30 11:30 Name: T.Ozaki Hi, > The index h_AN has two meanings: lattice vector index n and atomic site index j. The> index n sums over all lattice points of the nearest neighbors. Would you tell me how> to determine the lattice points of the nearest neighbors?You can find how to obtain this informaion, for example, in "Hamiltonian_Band.c" as follows: l1 = atv_ijk[Rn]; l2 = atv_ijk[Rn]; l3 = atv_ijk[Rn]; kRn = k1*(double)l1 + k2*(double)l2 + k3*(double)l3;where l1, l2, and l3 are integer numbers, and a set of them correponds to a lattice point.> When I calculated the momentum matrix elements P_ij that is proportional to> summation of OLPpox,y,z in polB.c. Nevertheless, the complex conjugate of P_ij is not> equal to P_ji. Would you please give me suggestions about the wrong results? I assume that you are evaluating the momentum matrix elements using an anticommutation relation. In this case, we have to take account of a contributionfrom the non-local potential in the norm-conserving pseudopotential as discussed in PRB 33, 7017 (1986)(See Eq. 15 in this paper). Regards, TO  Re: A question about the momentum matrix elements calculation ( No.2 ) Date: 2011/12/24 21:28 Name: Reza Khordad  Dear Prof.,I have a request.I nrequire some information about the energy related to the momentum matrix element (E^{Gamma}_P)With best regards,R. Khordad  Re: A question about the momentum matrix elements calculation ( No.3 ) Date: 2012/01/04 18:31 Name: Sajad Ahmad  dear sir,i have some problems as i am not able to optimize the volume of the l-alanine, molecular formula (C3H7NO2). The experimental volume is 2906 au^3 whereas mine is optimizing around 3100 au^3. i have performed the calculation for both bulk and the cluster but could not optimize around the experimental. so i request to please help me in this regard..my input file parameters are as follows:Species.Number 4Even i have performed the calculations using different basis sets asSpecies.Number 4andSpecies.Number 4the space group of l-alanin is P212121kindly help mewith regards Page: