# Raviart-thomas: existence of a good alternative pair for Dirichlet BCs?

Hi fenics users,

I have some difficulties in dealing with raviart-thomas pairs for a simple elliptic equation ( 11. Mixed formulation for Poisson equation — FEniCS Project).

More precisely, this pair is particularly difficult to deal with when considering a Dirichlet boundary condition on the flux J.
Indeed, one can only impose it on normal componant such as J_x*n_x or J_y.n_y where

div(J)=0 and J = D \nabla u

Do you know an alternative to this pair allowing this type of boundary condition while guaranteeing the inf-sup stability property?