Phase separation in block copolymers

[rnpratoori]

Hello team,

I am new to MOOSE and C++. I want to simulate phase separation in block copolymers. For, that I want to change the split formulation to reflect these equations -

Essentially, there is one extra term in the residual of chemical potential.

I have created an auxkernel based modifying SplitCHResBase.h. Here are the new *.h and *.C files I created - SplitCHPhaseSep.zip. Is this the right way to do it? Is there a better way?

When I compile with these files, I get the error -
error: no match for 'operator+=' (operand types are 'libMesh::Real' {aka 'double'} and 'libMesh::boostcopy::enable_if_c<true, libMesh::TypeVector<double> >::type' {aka 'libMesh::TypeVector<double>'})
This error appears in computeQpResidual, computeQpWJacobian and computeQpOffDiagJacobian.

Can anyone help me identify what mistake I might be making?

Thanks in advance,
Raghu.

[GiudGiud]

Hello

error: no match for 'operator+=' (operand types are 'libMesh::Real' {aka 'double'} and 'libMesh::boostcopy::enable_if_c<true, libMesh::TypeVector<double> >::type' {aka 'libMesh::TypeVector<double>'})

means you are summing a vector and a double. You have to use a loop to do that or maybe you do not want this?

  residual += _mob[_qp] * _alpha[_qp] * _kappa[_qp] * (_c[_qp] - _bar_c) * _grad_test[_i][_qp]; //Second term

residual is a real.
The other term is a real times a vector (gradient of the i-th test function at quadrature point qp)
maybe you mean to use _test instead?

[GiudGiud]

If you are not using AD, then you could create an auxiliary variable bar_c of type CONSTANT MONOMIAL.
Then use the ProjectionAux to project c onto bar_c on every linear iterations. This computes bar_c as the average of c on each element

[rnpratoori]

That is an interesting way! Can you tell me how to initialize the _bar_c variable now?
I know I have to use - _bar_c(coupledValue()),, but unsure of what should be the variable name to be used in coupledValue

[GiudGiud]

in the object, add:

params.addCoupledVar("bar_c", ...); 

to the validParams

then you can use

...,
_bar_c(coupledValue("bar_c")),

in the initializer's list of the constructor of your object

[rnpratoori]

I have one more question. I want to calculate the total energy of the system. When it was simple CH equation, I could use TotalFreeEnergy to calculate it. Now, can I still use the same? I noticed that there is a additional_free_energy input parameter - is this for the $E_d$ term in the Basic Phase Field Equations? How can I use it with my system or is there a better way I can do it?

[GiudGiud]

I dont know. @tonkmr maybe you know?