This was exactly what we had to do. We evicted the node from the cluster,

completely rebuilt the server, joined it to the cluster, then reinstalled sql and patched to correct...