Projector augmented wave method

The linear transformation ${\displaystyle {\mathcal {T}}}$ transforms the fictitious pseudo wavefunction ${\displaystyle |{\tilde {\Psi }}\rangle }$ to the all-electron wavefunction ${\displaystyle |\Psi \rangle }$:

${\displaystyle |\Psi \rangle ={\mathcal {T}}|{\tilde {\Psi }}\rangle }$

Note that the "all-electron" wavefunction is a Kohn–Sham single particle wavefunction, and should not be confused with the many-body wavefunction. In order to have ${\displaystyle |{\tilde {\Psi }}\rangle }$ and ${\displaystyle |\Psi \rangle }$ differ only in the regions near the ion cores, we write

${\displaystyle {\mathcal {T}}=1+\sum _{R}{\hat {\mathcal {T}}}_{R}}$,

where ${\displaystyle {\hat {\mathcal {T}}}_{R}}$ is non-zero only within some spherical augmentation region ${\displaystyle \Omega _{R}}$ enclosing atom ${\displaystyle R}$.

Around each atom, it is useful to expand the pseudo wavefunction into pseudo partial waves:

${\displaystyle |{\tilde {\Psi }}\rangle =\sum _{i}|{\tilde {\phi }}_{i}\rangle c_{i}}$ within ${\displaystyle \Omega _{R}}$.

Because the operator ${\displaystyle {\mathcal {T}}}$ is linear, the coefficients ${\displaystyle c_{i}}$ can be written as an inner product with a set of so-called projector functions, ${\displaystyle |p_{i}\rangle }$:

${\displaystyle c_{i}=\langle p_{i}|{\tilde {\Psi }}\rangle }$

where ${\displaystyle \langle p_{i}|{\tilde {\phi }}_{j}\rangle =\delta _{ij}}$. The all-electron partial waves, ${\displaystyle |\phi _{i}\rangle ={\mathcal {T}}|{\tilde {\phi }}_{i}\rangle }$, are typically chosen to be solutions to the Kohn–Sham Schrödinger equation for an isolated atom. The transformation ${\displaystyle {\mathcal {T}}}$ is thus specified by three quantities:

1. a set of all-electron partial waves ${\displaystyle |\phi _{i}\rangle }$
2. a set of pseudo partial waves ${\displaystyle |{\tilde {\phi }}_{i}\rangle }$
3. a set of projector functions ${\displaystyle |p_{i}\rangle }$

and we can explicitly write it down as

${\displaystyle {\mathcal {T}}=1+\sum _{i}\left(|\phi _{i}\rangle -|{\tilde {\phi }}_{i}\rangle \right)\langle p_{i}|}$

Outside the augmentation regions, the pseudo partial waves are equal to the all-electron partial waves. Inside the spheres, they can be any smooth continuation, such as a linear combination of polynomials or Bessel functions.

The PAW method is typically combined with the frozen core approximation, in which the core states are assumed to be unaffected by the ion's environment. There are several online repositories of pre-computed atomic PAW data.[2][3][4]

The PAW transformation allows all-electron observables to be calculated using the pseudo-wavefunction from a pseudopotential calculation, conveniently avoiding having to ever represent the all-electron wavefunction explicitly in memory. This is particularly important for the calculation of properties such as NMR,[5] which strongly depend on the form of the wavefunction near the nucleus. Starting with the definition of the expectation value of an operator:

${\displaystyle a_{i}=\langle \Psi |{\hat {A}}|\Psi \rangle }$,

where you can substitute in the pseudo wavefunction as you know ${\displaystyle |\Psi \rangle ={\mathcal {T}}|{\tilde {\Psi }}\rangle }$:

${\displaystyle a_{i}=\langle {\tilde {\Psi }}|{\mathcal {T}}^{\dagger }{\hat {A}}{\mathcal {T}}|{\tilde {\Psi }}\rangle }$,

from which you can define the pseudo operator, indicated by a tilde:

${\displaystyle {\tilde {A}}={\mathcal {T}}^{\dagger }{\hat {A}}{\mathcal {T}}}$.

If the operator ${\displaystyle {\hat {A}}}$ is local and well-behaved we can expand this using the definition of ${\displaystyle {\mathcal {T}}}$ to give the PAW operator transform:

${\displaystyle {\tilde {A}}={\hat {A}}+\sum _{i,j}|p_{i}\rangle \left(\langle \phi _{i}|{\hat {A}}|\phi _{j}\rangle -\langle {\tilde {\phi }}_{i}|{\hat {A}}|{\tilde {\phi }}_{j}\rangle \right)\langle p_{j}|}$.

Where the indices ${\displaystyle i,j}$ run over all projectors on all atoms. Usually only indices on the same atom are summed over, i.e. off-site contributions are ignored, and this is called the "on-site approximation".

In the original paper, Blöchl notes that there is a degree of freedom in this equation for an arbitrary operator ${\displaystyle {\hat {B}}}$, that is localised inside the spherical augmentation region, to add a term of the form:

${\displaystyle {\hat {B}}-\sum _{i,j}|p_{i}\rangle \langle {\tilde {\phi }}_{i}|{\hat {B}}|{\tilde {\phi }}_{j}\rangle \langle p_{j}|}$,

which can be seen as the basis for implementation of pseudopotentials within PAW, as the nuclear coulomb potential can now be substituted with a smoother one.

• Rostgaard, Carsten (2010). "The Projector Augmented-wave Method". arXiv:0910.1921v2 [cond-mat.mtrl-sci].
• Kresse, G.; Joubert, D. (1999). "From ultrasoft pseudopotentials to the projector augmented-wave method". Physical Review B. 59 (3): 1758–1775. Bibcode:1999PhRvB..59.1758K. doi:10.1103/PhysRevB.59.1758.
• Dal Corso, Andrea (2010-08-11). "Projector augmented-wave method: Application to relativistic spin-density functional theory". Physical Review B. 82 (7): 075116. Bibcode:2010PhRvB..82g5116D. doi:10.1103/PhysRevB.82.075116.

• ABINIT
• CASTEP (to calculate NMR properties)
• CP-PAW
• GPAW
• ONETEP
• PWPAW
• S/PHI/nX
• Quantum ESPRESSO
• VASP