The metal is aluminum; anything which attacks it will do. HCl and KOH are likely candidates, with KOH discouraged as it can attack SiO2.
If the full chip (minus bond pads, fuses, etc.) is passivated, this may not be very effective, and you'll have to carefully dose (time and concentration) HF to etch the passivation.
AFAIK, HF without oxidizing power (no excess nitrate, chlorine, O2, etc. dissolved) doesn't attack Si, or at least anywhere near as fast, so maybe you left it on much too long, or had too much oxidation available. Doing a little bit at a time, and maybe adding a few granules of a reducing agent (sulfite, thiosulfate) to the solution, might help.
Tim