文書の過去の版を表示しています。
Class methods for calculation condition
Charge mixing method
get_charge_mixing(self) set_charge_mixing(self, method='Pulay', mixinghistory=10, mixingcoeff=0.1, precondition='No', updatecharge='Yes')
get_charge_mixing return charge mixing parameters
set_charge_mixing set density mixing method and parameters
Parameters
| argument | type | notes |
|---|---|---|
| method | string | 'Pulay' for Pulay mixing. only one supported now |
| mixinghistory | integer | number of iterations to mix [1] |
| mixingcoeff | float | Mixing coefficient for Pulay charge mixing [2] |
| precondition | string | 'Yes' or 'No' [3] |
updatecharge : string 'Yes' or 'No'
* “Yes” : Perform charge mixing according to ChargeMixing:Method setting
* “No” : Freeze charge to initial value. This setting is useful when evaluating the Harris-Foulkes density functional
[1] Number of charge residual vectors stored for generating the Pulay estimate on the self-consistent charge density, see Sec. 4.2 in Kresse/Furthmuller: Comp. Mat. Sci. 6 (1996) p34ff.))
[2] Mixing coefficient for Pulay charge mixing, corresponding to $A$ in $G^1$ in Sec. 4.2 in Kresse/Furthmuller: Comp. Mat. Sci. 6 (1996) p34ff
[3]
- “Yes” : Kerker preconditiong is used,
i.e. $q_0$ is different from zero, see eq. 82 in Kresse/Furthmuller: Comp. Mat. Sci. 6 (1996). The value of $q_0$ is fix to give a damping of 20 of the lowest $q$ vector.
- “No” : $q_0$ is zero and mixing is linear (default).
get_convergence(self) return convergence settings for Dacapo
set_convergence(self, energy=1e-05, density=0.0001, occupation=0.001, maxsteps=None, maxtime=None) set convergence criteria for stopping the dacapo calculator.
:Parameters:
energy : float set total energy change (eV) required for stopping
density : float set density change required for stopping
occupation : float set occupation change required for stopping
maxsteps : integer specify maximum number of steps to take
maxtime : integer specify maximum number of hours to run.
Autopilot not supported here.