I assume the ball milling stirs the material (and probably heats it up) exposing lots of surface area to interaction with the balls. I'd guess ultrasound would

I assume the ball milling stirs the material (and probably heats it up) exposing lots of surface area to interaction with the balls. I'd guess ultrasound would be blocked/absorbed by the material it hits first. But I'm not a physicist, just my first thoughts.