although it can be a pain in the rear, and digging through documents like this might be confusing at first, you can often find the answers to nearly any question about sound isolation (or at least many questions) at www.nrc.ca
. that's the site of the National Research Council of Canada, and they have probably the largest collection of publicly available data on earth, and almost certainly the largest collection of data including low frequency information (lower than the STC range).
for 2.5" steel studs, using
-boring old fiberglass (0.7 pounds/cubic foot), STC=39, OITC=25, M&T=26.4
-normal mineral fiber (2-2.2 pcf), STC=38, OITC=25, M&T=26.2
-higher density mineral fiber (2.75-3pcf), STC=36, OITC=24, M&T=25.4
-extremely high density mineral fiber (6pcf), STC=37, OITC=25, M&T=26.3
so none outperformed plain old boring off-the-shelf fiberglass for all-around performance, although all had clearly superior mid/high absorption characteristics.
in other comparisons from the NRC's document IR-761, the same basic trend is observed: the low frequency resonance shifts up as density goes too high, and that negates whatever advantage mid/high frequency gains yield, and the glass tends to perform at least as well as anything.
hence i recommend just using fiberglass. That said, mineral fiber of normal density doesn't prove itself WORSE than fiberglass, so use that if you like, but avoid extremely high density materials like specialty fibers that are very dense... use those for room treatments!