Abstract.Since several elements between Z=20-28 are refractory in their nature, their neutron-rich isotopes are rarely available as low energy Radioactive Ion Beams (RIB) in ordinary Isotope Separator On-Line facilities [1][2][3][4]. These low energy RIBs would be especially interesting to have available under conditions which allow high-resolution betadecay spectroscopy, ion-trapping and laser-spectroscopy. As an example, availability of these beams would open a way for research which could produce interesting and important data on neutron-rich nuclei around the doubly magic 78 Ni. One way to overcome the intrinsic difficulty of producing these beams is to rely on the chemically unselective Ion Guide Isotope Separator On-Line (IGISOL) technique [5]. Quasi-and deep-inelastic reactions, such as 197 Au( 65 Cu,X)Y, could be used to produce these nuclei in existing IGISOL facilities, but before they can be successfully incorporated into the IGISOL concept their kinematics must be well understood. Therefore the reaction kinematics part of this study was first performed at the Lawrence Berkeley National Laboratory using its 88" cyclotron and, based on those results, a specialized target chamber was built, see Fig. 1 [6].The target chamber shown in Fig. 1 was recently tested on-line at the Jyväskylä IGISOL facility. Yields of massseparated radioactive projectile-like species such as 62,63 Co are about 0.8 ions/s/pnA, corresponding to about 0.06 % of the total IGISOL efficiency for the products that hit the Ni-degrader, see Fig.1. (The current maximum 443 MeV 65 Cu beam intensity at Jyväskylä is about 20 pnA.) This total IGISOL efficiency is a product of two coupled loss factors, namely inadequate thermalization and the intrinsic IGISOL efficiency. In our now tested chamber, about 9 % of the Co recoils are thermalized in the flowing He gas (p He =300 mbar) and about 0.7 % of them are converted into the mass-separated ion beams.In the future, both of these physical/chemical conditions can be suppressed by introducing Ar as a buffer gas and by relying on selective laser re-ionization. This combination will produce isobarically pure beams and it will increase the existing yields by at least a factor of 100, making this overall approach to the study of neutron rich nuclei even more attractive.