We are using a high purity alumina crucible in a standard e-beam doser (we have Focus and home-made ones). We use Fe powder. The problem is that we are getting aluminum and oxygen on the surface, in addition to iron at the settings we need to get Fe (about 44W). We have used previously a self-supported iron rod, and it works fine, but for this particular experiments we need to use a crucible with Fe powder (we have some Fe57 powder). We have already bought new crucibles from another company, and the problem persists. Should we run it longer before assuming that we will always get Al and O contamination and thus need to get rid of the crucible?