charged point defects (F center)

[fangyongxinxi]

Dear,

study point defect in bulk material,
we often build charged point defect models (compared to neutral point defect),

can ATK build model with charged point defects ?

Thanks~~

[Umberto Martinez]

You can add/remove an extra charge to your bulk configuration with the charge parameter in the LCAOCalculator
http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.lcaocalculator.html

[fangyongxinxi]

You can add/remove an extra charge to your bulk configuration with the charge parameter in the LCAOCalculator
http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.lcaocalculator.html

thanks for your reply, it helps me alot~

[fangyongxinxi]

You can add/remove an extra charge to your bulk configuration with the charge parameter in the LCAOCalculator
http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.lcaocalculator.html

Dear, I have more questions about build the color center in bulk materials.

qeustion-1,

we can use the following script to set the charge of a system.
LCAOCalculator(
basis_set,
exchange_correlation,
charge,
numerical_accuracy_parameters,
iteration_control_parameters,
poisson_solver,
checkpoint_handler,
dynamical_matrix_parameters,
correction_extension,
algorithm_parameters,
hamiltonian_derivatives_parameters
)

the question is: the charge is positive charge or negative charge ?
eg.
charge = 1
means we put one extral electron to system, or we move one electron away from the system ?



question-2. 
how to build color center model in bulk material. (this question is more relatively to my own work)

In other people's work about MgO (see the attachment picture), the author said like follows:

"Oxygen vacancies exists in three possible variants, depending on the charge of the defect. the removal of neutral
oxygen leads to a F0 color center, with two electrons trapped in the cavity; the removal of O(-) or O(2-) ions leads
to F(+) or F(2+) centers, with only one or no electrons trapped in the cavity, respectively"

another paper said:
"color centers can also form in oxides such as MgO. When a single electron is trapped at an anion vacancy in MgO, the defect is positively charged and refferred to as F(+) center. An F center in MgO is a vacancy with two elctrons"


My understanding like this:
eg, for a 100-atom MgO supercell model,
if i want to get a F(0) center, I just need to remove one oxygen atom, nothing else to do. The " two electrons
trapped in the cavity" are from one Mg atom in the 99-atom model.
(or, I should set "charge=2" for F(0) center ?)
if i want to get a F(+) center, I need to remove one oxygen atom, and set "charge = 1", which means I put one extral
electron to the 99-atom system.
if i want to get a F(2+) center, I need to remove one oxygen atom, and set "charge = 2", which means I put two
extral electrons to the 99-atom system.
is that right?

what's more, how can I get the total number of electrons in the log file ?


looking forward to reply.

[Umberto Martinez]

1- from the link I posted before: "The charge of the system, a charge of -1 corresponds to one extra electron."

2- the two papers state the same.
You can also check this one which deals with these defects @ MgO(100) surfaces: http://dx.doi.org/10.1021/jp7108016

Remove a neutral O atom and you will have two electron in your cell (O is O2- in MgO) which supposedly they will be trapped in the vacancy : F(0)
The whole cell is still neutral so far.
Remove one electron with the charge parameters and you will get a F+ (given that you converge to that state)
Note that depending on your basis set you may need to add a ghost atom in the center of the vacancy.

[fangyongxinxi]

1- from the link I posted before: "The charge of the system, a charge of -1 corresponds to one extra electron."

2- the two papers state the same.
You can also check this one which deals with these defects @ MgO(100) surfaces: http://dx.doi.org/10.1021/jp7108016

Remove a neutral O atom and you will have two electron in your cell (O is O2- in MgO) which supposedly they will be trapped in the vacancy : F(0)
The whole cell is still neutral so far.
Remove one electron with the charge parameters and you will get a F+ (given that you converge to that state)
Note that depending on your basis set you may need to add a ghost atom in the center of the vacancy.

Dear Umberto Martinez
thanks for your help.

according to your reply, my understanding is:
remove one O atom, and set charge = 0, get F(0)
remove one O atom, and set charge =+1, (take away one electron form system) , get F(+)
remove one O atom, and set charge = +2, get F(2+)
Is that right ?

By the way, I find you published lot of good papers about point defects in oxides, and I am reading these papers now, and they help me alot for my own works.

Sincerely FangYong
2015-4-24

[Umberto Martinez]

that is right.