There are many reasons a bath is a better method:
Direct contact with the dry ice would work.....the key word is direct. The reason you can hold a piece of dry ice bare handed ( don't do it, but you can) is because it never directly contacts your hand. There is a thin space of sublimating gas between you and it. This gas is a pretty good insulator.
Cooling is the transfer of heat from one material to another. The cold doesn't go in from the Dry Ice, it is the heat from the blade that has to come out. It needs a material that can transfer heat. Gasses are the poorest of heat transfer materials. Liquids are hundreds of times better. You can harmlessly reach inside a 400F oven and set a pan on the shelf. If you reach into a 200F pot of water, you come out with a severely damaged hand.
When a blade is placed between two blocks of DI, the blocks and the blade are bouncing back and forth in this gas layer, thus the screaming. Personally, I would not like to subject any untempered blade to severe vibration.
Some places are cooling faster than others. The cooling being done is the end of the quench, and just like quenching in a tank of oil, water, or air... it should be even throughout the blade and consistent in drop rate.
Placing a blade between two pieces of dry ice will also require a lot more DI than a slurry will.
Putting the DI in an alcohol bath will lower the bath to within 10° of the DI temp...which is low enough for the Mf conversion to finish.
Summation - The liquid in the bath assures even heat transfer across all blade surfaces at the same time and being liquid allows a smooth drop rate. Isn't that what quenching is all about?
