The problem which I mentioned in the former diary is avoided by introducing a cut off wave vector. It is common procedure when we treat the Dirac cone. However, I'm not sure it is physically appropriate that the vertex correction diverges while the other terms converge. Besides that, we have other problems and our research is still in a hard situation.