Currently, I believe that most commercial D is extracted by fractional distillation when making Ammonia (NH3), since ND3 has a different BP.
One can also separate it from cryogenic hydrogen.
In either case, you might use electrolysis to separate the H2 (and HD, and D2) from water, but commercially, cracking methane is much more popular (and cheaper..)
I don't think anyone uses electrolysis any more. It was used in early days, before the large scale manufacture of ammonia (didn't really get going until after the 30's), and large scale cryogen manufacturing (LH2 wasn't made in significant quantities until the 50's for the H-bomb program... now, I see a tank truck of LH2 every morning on the freeway..)
D2 gas is pretty inexpensive now. $100 or so will buy a lecture bottle of D2 which will last you a lifetime of amateur fusion experiments (unless you are careless and let it leak out). The amount of work required to extract D2 from anything
The "fusion power will save us" folks have always publicized the fact that "the fuel comes from seawater", as opposed to fission which requires Uranium, because there is a perception that Uranium is rare and hard to find. Mind you, uranium is pretty common, and cheap, and the biohazard of a breeder reactor making Pu from the U238 isn't much different than the hazard from a full tilt power generation fusion reactor: both generate literally tons of really "hot" waste.
Created on Monday, March 12, 2001 3:32 PM EDT by James Lux