In a recent publication we have argued that using two very successful n-9 Be optical potentials (A. Bonaccorso and R. J. Charity, Phys. Rev. C89 (2014) 024619) and microscopic projectile densities, it is possible to build a single-folded (light-) nucleus-9 Be imaginary optical potential which is more accurate than a double-folded optical potential. By comparing to experimental reaction cross sections, we showed for 8 B, 8 Li and 8 C projectiles, that a very good agreement between theory and data could be obtained with such a "bare" potential, at all but the lowest energies where a small semi-microscopic surface term was added to the single-folded potential to take into account projectile breakup. In this paper we extend this study to the case of 9 C projectiles and assess the sensitivity to the projectile density used. We then obtained the modulus of nucleus-nucleus S-matrix and parametrize it in terms of a strong-absorption radius Rs and finally extracted the phenomenological energy dependence of this radius. This approach could be the basis for a systematic study of optical potentials for light exotic nuclei scattering on light targets and/or parametrizations of the S-matrix. Furthermore our study will serve to make a quantitative assessment of the description of the coretarget part of knockout reactions, in particular their localization in terms of impact parameters.