The immersion cooling is just about nearly pointless with Bitcoin mining. The only thing that its proponents bring to table are workarounds for lack of proper design and engineering. Immersion cooling with 3M engineered fluids only makes sense if there are real constraints on the cooled equipment: speed of light limitation, unusual complexity, high density, etc. None of that really matter for Bitcoin mining.

If your goal is low PUE then simply use typical evaporative cooling well known in chemical process engineering and refrigeration. Most of the required equipment is already available, all you'll need to do is bring a chemical/thermodynamic process engineer to the table with the PCB design engineer, who is already working for Spondoolies. Together they need to design a proper heat exchanger for the mining chips and the associated buck converters. I already asked Spondoolies if SP30 motherboard was designed for cutting into 15 smaller boards. Apparently not, probably because the PCB design engineer wasn't given this as a requirement. So they cut the heathsinks into 15 smaller pieces.

They need to conceptually invert that design: single large liquid heat exchanger and multitude of small PCBs hosting the hot chips is the way to go long term.

3M's profit margins on their immersion cooling fluids are so high, that they require scientific notation to fit in the spreadsheets. Just use common chilling/refrigeration equipment instead where the profit margins are in expressed in two digits.