I intended to calculate the bandgap of TiO2. According to the literature, to obtain a bandgap in DFT close to the experimental value, it has to be incorporated with the Hubbard potential. However, during my optimization (relax, vc-relax), the input file is being aborted. I would like to inquire whether it is possible to perform other computations (scf, nscf, PDOS) with the Hubbard potential while the optimization is done without the Hubbard potential. Your suggestions will be highly appreciated.