I'm not a VASP user, I develop a different plane-wave DFT program (CASTEP; www.castep.org) but in general there are two ways you can achieve this:

a) A constrained variable-cell geometry optimisation

Here you would follow the method you set out in the original question and strain your system along one axis, e.g. the c-axis, then perform a variable-cell geometry optimisation but fix the axis you strained. You'll need to be careful to ensure your calculated stress is corrected for the finite basis set (i.e. that you've removed the Pulay Stress terms), for example using the method of Francis & Payne[1]. If you don't do that you'll have to make the cut-off energy extremely high to get reasonable results.

You'll have to look up how to do the constraints in VASP, but in CASTEP you'd do this with a standard variable-cell geometry optimisation but add:

%block cell_constraints

1 2 0     ! constraints on cell lengths a, b and c

3 4 5     ! constraints on cell angles alpha, beta and gamma

%endblock cell_constraints

to the CELL input file. The values are: "0" meaning fix, positive integer meaning "vary" and if any non-zero integers are the same it means "vary, but keep the ratio of these constant". So for example you could fix the b/a ratio as well as fixing c by using

%block cell_constraints

1 1 0

3 4 5

%endblock cell_constraints

b) An applied uniaxial stress

The other way to investigate what you're interested in is to apply a stress, and see what the resultant strain is. In some ways this is easier, since you don't have to worry about constraints. Simply apply the external stress tensor you want, and do a standard variable-cell geometry optimisation. Again, I'm afraid you'll have to look up how you specify the stress in VASP, but for reference in CASTEP you'd add:

%block external_pressure

GPa                  ! Set the unit for the pressure

0.0 0.0 0.0       ! no stress in the a-direction

       0.0 0.0       ! no stress in the b-direction

              4.0       ! 4 GPa stress in the c-direction

%endblock external_pressure

to the CELL file to apply a 4 GPa uniaxial stress along the c-axis. You only supply the upper triangle of the 3x3 stress tensor -- since the tensor must be symmetric, CASTEP fills in the lower half for you. Then you just run a normal variable-cell geometry optimisation, and read off the final lattice constant to find the corresponding strain.

Hope that helps,

Phil Hasnip

(CASTEP Developer)

[1] Article Finite Gasis Eet Corrections to Total Energy Pseudopotential...

(the typo in the title is from Research Gate, sorry about that!)

Hi Swastik, the following article can give a reference for you https://code.google.com/archive/p/ideal-strength-vasp/wikis/readme.wiki

I'm not a VASP user, I develop a different plane-wave DFT program (CASTEP; www.castep.org) but in general there are two ways you can achieve this:

a) A constrained variable-cell geometry optimisation

Here you would follow the method you set out in the original question and strain your system along one axis, e.g. the c-axis, then perform a variable-cell geometry optimisation but fix the axis you strained. You'll need to be careful to ensure your calculated stress is corrected for the finite basis set (i.e. that you've removed the Pulay Stress terms), for example using the method of Francis & Payne[1]. If you don't do that you'll have to make the cut-off energy extremely high to get reasonable results.

You'll have to look up how to do the constraints in VASP, but in CASTEP you'd do this with a standard variable-cell geometry optimisation but add:

%block cell_constraints

1 2 0     ! constraints on cell lengths a, b and c

3 4 5     ! constraints on cell angles alpha, beta and gamma

%endblock cell_constraints

to the CELL input file. The values are: "0" meaning fix, positive integer meaning "vary" and if any non-zero integers are the same it means "vary, but keep the ratio of these constant". So for example you could fix the b/a ratio as well as fixing c by using

%block cell_constraints

1 1 0

3 4 5

%endblock cell_constraints

b) An applied uniaxial stress

The other way to investigate what you're interested in is to apply a stress, and see what the resultant strain is. In some ways this is easier, since you don't have to worry about constraints. Simply apply the external stress tensor you want, and do a standard variable-cell geometry optimisation. Again, I'm afraid you'll have to look up how you specify the stress in VASP, but for reference in CASTEP you'd add:

%block external_pressure

GPa                  ! Set the unit for the pressure

0.0 0.0 0.0       ! no stress in the a-direction

       0.0 0.0       ! no stress in the b-direction

              4.0       ! 4 GPa stress in the c-direction

%endblock external_pressure

to the CELL file to apply a 4 GPa uniaxial stress along the c-axis. You only supply the upper triangle of the 3x3 stress tensor -- since the tensor must be symmetric, CASTEP fills in the lower half for you. Then you just run a normal variable-cell geometry optimisation, and read off the final lattice constant to find the corresponding strain.

Hope that helps,

Phil Hasnip

(CASTEP Developer)

[1] Article Finite Gasis Eet Corrections to Total Energy Pseudopotential...

(the typo in the title is from Research Gate, sorry about that!)

you need

Quantum Espresso 6.5