“…As a matter of fact, these more accurate procedures were used in the literature mostly for circularly symmetric problems, where one can take advantage of the system symmetry to reduce the number of coordinates, or to (semi-)analytically solve for the single-particle eigenstates. [16,31] Conversely, our case requires Cartesian coordinates, which allows us to investigate elliptic shells, in-plane electric fields, and systems with arbitrary shell geometry, by paying the price of having more coordinates to deal with. Thus, the main idea of the present model is to circumvent this problem by using a much simpler approach, which, in summary, consists in (i) assuming, as an approximation, that one can separate the variables, (ii) solving for the single-particles separately, and then (iii) minimizing the exciton energy as a whole by a variational procedure.…”