@Rob.B - you write very much in the past tense - what happened to this? If it's as good as you say (and I don't doubt you) then this technology should be all over the place (and you should be a rich man!)?
In the early 80's my company demonstrated a very similar flow battery system, including building a 5MW system first installed in a Detroit Edison substation for initial testing, and then moved to the EPRI test facility in New Jersey. Frankly, you wouldn't believe the conversion and energy storage efficiencies we recorded for the systems. (Neither did some of our opponents and naysayers, but that's another story.)
This system used a flow cell incorporating both solid and porous graphite plates rather than a membrane. The big advantage was that the battery components didn't wear out, only occasionally requiring draining and replacing the electrolyte. Graphite, plastic, and fiberglass were the major structural components. Darned cheap!
This battery used a Zinc-Chlorine chemistry (ZnCl2) rather than bromine. Now before anyone panics over "chlorine", remember that bromine has some pretty nasty effects as well, and is potentially even more dangerous and difficult to contain. While you probably wouldn't want to drink it, the electrolyte was more like a weak chlorine-bleach solution, and wasn't going to hurt you if you spilled it on yourself.
Our primary addition to the technology of the ZnCl2 battery (First built in the 1800's) was our chlorine storage technique, so there was little chance of a massive and dangerous leak even with a major failure.
The system was designed for utility load-leveling applications, but would be perfect for wind-farms or large solar-array installations. In the case of wind-farms, the battery modules and associated equipment could easily be fit into some of the towers.
So essentially, this group has just "reinvented the wheel", using a chemistry that we considered to be more expensive and dangerous, but with the same mechanical components.
Regarding storage by pumping water uphill, here is Nate Lewis' comment (from MRS Bulletin, Oct. 2007): "...for every gallon of gasoline that we use, storing that same amount of energy in pumped water requires moving uphill by 100 m more than 50,000 gallons of water!"
I think the large-scale analysis is that this is modestly cost-effective (surprisingly efficient despite this huge volume) if the dam is already there, but not exactly a scaleable solution. The same is almost certainly true of compressing air in caves, only even worse.
This report sounds encouraging in principle; it would seem to me to come down to the details of engineering practicality. There is also a company in the U.K. (Isentropic) whose system stores energy in hot rocks, basically; a heat pump moves the energy from one to the other. "In principle" the efficiency is quite good; it all comes down to the practice.
Thanks for clarifying that $100/kwh is a capital cost. Regarding losses associated with charging/discharging and maintenance costs those are things to be explored and tuned during the optimization stage of development, which the researchers are beginning now.
do not confuse cost of storage with cost of generation. $100/kwh is a capital cost which is spread over its lifetime. What is not mentioned in the article is the cost of maintaining such a battery and any losses associated with charging/discharging.
Maybe pump it up-hill at night when demand is low (and energy is available for pumping) and flow it back thru the turbines during daytime peak demand. BTW Niagara Falls will soon have more generation capacity due to a "big-dig" to increase flow!