I don't want to derail the thread away from the topic in hand, but regarding CVA-01, the amount of steam required was quite large and should not be underestimated.
The BS.4 needed 1,000lb of steam at 500psi for less than 2 seconds per launch. With a launch rate of once every 30 seconds, each catapult would require 120,000lb/hr. They needed dry steam accumulators of around 2,385 cu ft (smaller if wet accumulators were used).
The BS.6 planned for CVA-01 had 2.5x the energy release, so the steam requirements would be much greater (I don't have exact lb/hr figures, 300,000lb perhaps if linear scaling is appropriate for estimating?). Wet accumulators would be used (which in themselves generated some steam). During launch operations the speed would fall from 30-28kt to 25kt, and as I say, we're talking about a 1000psi high-pressure boiler setup.
A 6 sq metre uptake and 6 m2 downtake for each Olympus would require a total of 72sq m for a sextuple setup. Plus 40 m3 for salt spray eliminators. Any obstructions or kinks causing back pressure in the air system would lose at least 100shp plus you would have 500C hot gas in those macks (the Broomstick gonna be toasty warm). That hot air might make the air pretty bumpy aft of the ship too, I'm also presuming its higher velocity hot air compared to normal boiler efflux when it exits the funnel/mack.
So I can see a fair few reasons why they kept it simple and just went with steam. In a carrier you need steam as a vital component so might as well use as much of it as you can. Even using a Rolls-Royce submarine nuclear core probably makes more sense than all-gas turbine in any AU where the Chancellor of the Exchequer is sitting in a bathtub of fifty quid notes.
