It's not about taking on grunt work, just because you finished some assigned tasks quicker than the person (/team) who assigned them thought it would take doesn't mean the next most important things are grunt work.
If it's beneficial for the company (and I agree that this kind of experimental work is) then make it part of the tasks. Don't only do it when your estimates are wrong!
