The Setting
In a bright future, after the brutally efficient totalitarian dystopia-state that once controlled solar systems fell to its own people, the world is still getting back on its feet. They have direct fusion drives, industrial antimatter production, and a very small dyson swarm.
With this kind of orbital infrastructure and activity in space, the fuel industry is huge. Liquified hydrogen, deuterium and helium 3 are the gold standards for the space-shipping economy, with massive freighters doing arounds through the outer solar system, dropping packs of fuel at skyhooks as they loop between the sun and the gas giants.
Plasma direct fusion drives are synonymous with spaceships... with one bizarre exception... Ion engines. For some ships, having full fusion drives isn't necessary, feasible or even possible. Any craft too cheap and expendable to have a drive, or where it doesn't need its huge thrust, or where one just plain wouldn't fit, advanced hall-effect engines and fusion or solar power plants act as propulsion busses.
Some ships have plasma drives, and hall-effect engines to bridge the gap of using maneuvering thrusters and main engines. Usually they can run on a mix of atmospheric gas (nitrogen, oxygen and helium) in the high atmosphere, so it can hang in extremely low orbits in much the same way bricks don't.
The Question
But for general use, what is the best and most readily available reaction mass mixture to use in these engines? Considering that harvesting gas giants for fuel is a huge industry, what is the best reaction mass mixture thereof, considering where and how you get it, over how good it is as reaction mass, its efficiency and erosion issues you'd face.
What if its a range of mixtures? You'd choose the 'weight' of the reaction mass mix in much the same way we pick the octane of our gasoline? I picture Kitsuki and company standing in a warehouse, looking over color-coded tanks, the hue of the band on the side of the tank representing the ratios of material in it, trying to figure out which is best.
Edit
These hall-effect engines are just plain-old terran technology, in service for decades at present day. Sure, they improved them a little here and there, with super-composites used in railgun rails which makes them virtually impervious to wear and superconducting magnets for better scalability.
But they still have the same basic design and layout, and run on the same fuels as todays engines. The reason some fuels might be cheaper is that atmospheric harvesting produces noble gases and other compounds as they get distilled out of the usable fusion fuel.
But some fuels are more efficient than others, and some get produced in greater quantities than others, so an economy will dictate that one fuel should be more expensive than others baste on supply and demand. To combat this, theres a system to catalog and create specific mixtures with certain amounts of gasses, varying in price based on what it's made with, made to suit a given application.
Maybe the difference between krypton and xenon is negligible to a massive nitrogen ice freighter doing the sun-mars-titan loop, but a science probe might need to be as efficient as physically possible, regardless of the cost, and so the difference for the tiny craft is huge.