[AlmostSurelyRob]

This is a difficult question for me to ask provided I've been using OpenFOAM for some time, but I really wonder what is the real meaning of the divergence term as given by fvm::div(phi, U). Is it

a) u \cdot grad(u)
b) \nabla \cdot uu^T

The reason why I am asking is that I want to have an fvm based Burgers equation for 1D with the flux f(u) = 0.5 u^2 and I don't know whether I should or shouldn't include the 0.5 term.

        while (simple.correctNonOrthogonal())
        {
            solve
            (
                fvm::ddt(U)
              + 0.5 * fvm::div(phi, U) //*0.5?
              - fvm::laplacian(DT, U) 
            );
        }
        phi = linearInterpolate(U) & mesh.Sf();