Hey Clefcon,
Not easy to find, unfortunately. The british caught onto this in the late fifties, and the US seems to have been at the forefront of further research in the 1960s... But despite the fact that I often see references to academic articles, etc, you can almost never get your hands on them. I'll give you what references I have in my browser's favorites, as well as a run-down, and leave you to conduct your own search. Please post anything useful you find!
ftp://imgs.ebuild.com/woc/C650369.pdf
An old thread where this was discussed, and I posted a run-down on the reaction...
(Review the SEAINT archive post for a PE's opinion on why you should be banning aluminum tools in contact with PCC)
There was extensive use of aluminum in in concrete for a long time, and by all accounts this can be quite successful. It can produce an effectively shrink-free concrete, but even though the specifics of the reaction are fairly well known, there are too many variables to reliably control the effect. Most engineers who know about it simply avoid it.
HOWVER, there are grout products that still use this reaction... Essentially in a grout you have fewer variables (no coarse aggregates to hide large quantities of undesirable elements, tightly controlled source of fine aggregates, factory mixing conditions, etc). I believe that many engineers who blindly specify "non-shrink, non-metallic" have no idea why, and probably routinely accept grouts with aluminum admixes.
Cheers,
YS
B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...