theories of pseudopotentials
play

Theories of pseudopotentials 1. OPW method 2. PK type - PowerPoint PPT Presentation

Theories of pseudopotentials 1. OPW method 2. PK type pseudopotential 3. Norm-conserving pseudopotential by TM 4. Ultra-soft pseudopotential by Vanderbilt 5. MBK pseudopotential 6. Solving the 1D Dirac eq. 7. What we can do if we generate PPs


  1. Theories of pseudopotentials 1. OPW method 2. PK type pseudopotential 3. Norm-conserving pseudopotential by TM 4. Ultra-soft pseudopotential by Vanderbilt 5. MBK pseudopotential 6. Solving the 1D Dirac eq. 7. What we can do if we generate PPs by ourselves 8. On the vps file Taisuke Ozaki (ISSP, Univ. of Tokyo) The Summer School on DFT: Theories and Practical Aspects, July 2-6, 2018, ISSP

  2. Intuitive ideas of pseudopotentials Electronic structure of Si bulk 1. Since core electrons is situated at energetically very deeper states, they are inert chemically. In molecules and solids, they do not change so largely. 2. Is there a way of constructing an effective potential consisting of the nucleus potential and Coulomb potential given by the core electrons states calculated in advance ? 3. If the effective potential is much shallower than that of the true nucleus potential, it is expected that the calculation will become quite easier. Science 351 , aad3000 (2016).

  3. OPW (Orthogonalized Plane Wave Method) method C. Herring, Phys. Rev. 57, 1169 (1940) ˆ    i  H E c,v i i i valence electrons oscillate near the  It is assumed that has been solved in advance. vicinity of nucleus because of the c     orthogonality with core electrons. , q exp q r PW i is orthogonalized with by c      , , , OPW q PW q PW q c c c It is easy to verify that         , , , OPW q PW q PW q c' c' c' c c c      , , 0 PW q PW q c' c' By using the OPW as basis set, the number of basis functions can be reduced

  4. Phillips-Kleinman (PK) method Phys. Rev. 116, 287 (1959)    , C PW G Smooth part of wave funtion G G         Orthogonalize it with core electrons c c  Let’s write Eq. by ˆ .    H E L.H.S R.H.S   ˆ ˆ ˆ         H H H       E E E c c c c c   ˆ c     H E c c c c Features of V One can get by equating L.H.S with R.H.S. eff      ˆ        1. Non-local potential  H E E  E c c c   c 2. Energy dependent  This gives a new view that feels the 3. For a linear transformation following effective potential.           v v E E       ' eff ext c c c c c c c Positive in general the form of Eq. is invariant. V eff is shallower than v.

  5. Scattering by a spherical potential Incident wave  q r i e 1   2 2 q Scattered wave iqr e          i q r  i sca ( , ) (2 1) sin (cos ) r e l e P l l l qr l d    ( ) | ( ) ( ) r j kr D j kr Phase shift 0 0 l r l l dr 0 l    tan ( , ) r l 0 d   ( ) | ( ) ( ) r n kr D n kr 0 l r l l 0 dr 0  2 r      0 2 2 ( , ) | | ( ) | D r drr r Logarithmic derivative of ψ   l r l 2 ( ) d 0 r r 0 0 0 d l    ( , ) ln ( ) D r r r If the norm of pseudized wave is conserved within r 0 and the l l dr logarithmic derivative coincides with that for the all electron case, the phase shift coincides with the all electron case to first order.

  6. Norm-conserving pseudopotential by Troullier and Matins N. Troullier and J. L. Martins, Phys. Rev. B 43 , 1993 (1991).    2 1 ( 1) d l l ( ) u r        ( )  ( ) ( ) V r u r u r l ( ) R r l l l 2 2   2 2 dr r l r ( ) For , the following form is used. u r l   6 (AE)    ( ) u r r r 2 i   ( ) p r c r l cl ( ) u r 2 i  l   1 l   r r exp[ ( )] r p r 0 i cl Putting u l into radial Schroedinger eq. and solving it with respect to V, we have  ( 1) 1 '' ( ) l l u r     (scr) l ( ) V r l 2 2 2 ( ) r u r l  ( 1) '( ) 1 l l p r        2 ''( ) [ '( )] p r p r   l 2 r c 0 ~ c 12 are determined by the following conditions: • Norm-conserving condition within the cutoff radius • The second derivatives of V (scr) is zero at r=0 Equivalence of the derivatives up to 4 th orders of u l at the cutoff radius •

  7. Unscreeing and partial core correction (PCC) Unscreeing Since V (scr) contains effect of valence electrons, the ionic pseudopotential is constructed by subtracting the effects.       (ps) (scr) ( ) ( ) ( ) [ ( ) ( )] V r V r V r V r r Hartree xc pcc l l v Partial Core Correction(PCC) Valence and PCC charges of carbon atom In order take account of the non-linearity of exchange- correlation term, it would be better to include the partial core correction.

  8. Pseudopotentials by the TM method Red: All electron calculation Blue: Pseudopotential Radial wave function of C 2s Pseudopotential for C 2s and – 4/r

  9. Separable pseudopotentials Since the pseudopotential depends on the angular momentum l , it is non-local.     (PS) ( ) ( , ') ( ') ( , ') V r V r r r r V r r oc NL l      ˆ ˆ (PS) m m ( , ') ( ) ( ) ( ) ( ) r r r r V Y V r V r Y NL l l loc l lm   ˆ ˆ (NL) m m ( ) ( ) ( ) Y r V r Y r l l l lm   1 (NL) (NL) m m V R Y Y R V   l l l l l l c   lm l The non-local potential is usually used as a separable form due to the simplicity of calculations. L Kleinman and D. M. Bylander, PRL 48, 1425 (1982). P. E. Blöchl, Phys. Rev. B 41, 5414 (1990).

  10. Ultrasoft pseudopotential by Vanderbilt D. Vanderbilt, PRB 41, 7892 (1990). The phase shift is reproduced around multiple reference energies by the following non-local operator. If the following generalized norm conserving condition is fulfilled, the matrix B is Hermitian. Thus, in the case the operator V NL is also Hermitian. If Q=0, then B-B*=0

  11. How the non-local operator works? Operation of the non-local operator to pseudized wave function Note that It turns out that the following Schroedinger equation is satisfied.

  12. The matrix B and the generalized norm conserving condition The matrix B is given by Thus, we have By integrating by parts ・・・ (1) By performing the similar calculations, we obtain for the all electron wave functions ・・・ (2) By subtracting (2) from (1), we have the following relation.

  13. Norm-conserving pseudopotential by MBK I. Morrion, D.M. Bylander, and L. Kleinman, PRB 47, 6728 (1993). If Q ij = 0, the non-local operator can be transformed to a diagonal form. The form is exactly the same as that for the Blöchl expansion, resulting in no need for modification of OpenMX. To satisfy Q ij =0, the pseudized wave function is written by The coefficients can be determined by matching up to the third derivatives to those for the all electron, and Q ij =0 . Once c ’s are determined, χ is given by

  14. The form of MBK pseudopotentials The pseudopotential is given by the sum of a local term V loc and non-local term V NL .   (ps) ( ) V V r V loc NL The local term V loc is independent of the angular channel l . On the other hand, the non-local term V NL is given by projectors       | | V NL i i i i The projector consists of radial and spherical parts, and depends on atomic species, energy-channel, and l-channel.

  15. Relativistic pseudopotentials By using the eigenfunctions of the spherical operator for the Dirac Eq., one can introduce a relativistic pseudopotential as

  16. Optimization of pseudopotentials (i) Choice of parameters Optimization of PP typically takes a half 1. Choice of valence electrons (semi-core included?) week. 2. Adjustment of cutoff radii by monitoring shape of pseudopotentials 3. Adustment of the local potential 4. Generation of PCC charge (ii) Comparison of logarithm derivatives No good If the logarithmic derivatives for PP agree well with those (iii) Validation of quality of PP by performing good of the all electron potential, a series of benchmark calculations. go to the step (iii), or return to the step (i). good No good Good PP

  17. Comparison of logarithmic derivatives Logarithmic derivatives of wave functions for s, p, d, and f channels for Mn atom. It is found that the separable MBK is well compared with the all-electron. If there is a deviation in the logarithmic derivatives, the band structure will not be reproduced.

  18. OpenMX vs. Wien2k in fcc Mn

  19. 1D-Dirac equation with a spherical potential 1-dimensional radial Dirac equation for the majority component G is given by Minority component The mass term is given by By expressing the function G by the following form, One obtain a set of equations: L の満たすべき条件 The charge density is obtained from

  20. Solving the 1D-Dirac equation By changing the variable r to x with , and applying a predictor and corrector method, we can derive the following equations: For a given E, the L and M are solved from the origin and distant region, and they are matched at a matching point. M L In All_Electron.c, the calculation is performed.

Recommend


More recommend