This is the second part of my video series on building a DIY cascade refrigeration system, where i reach (and exceed) my goal of reaching -100C.

Part I:
   • DIY Cascade Refrigeration System (Par...  

In this video, i experiment with some alternative refrigerants for second stages. I tried CO2 (R744) (which isn't viable due to freezing) and Nitrous oxide (R744A), which is a very effective refrigerant but also potentially hazardous. I reached an open-cycle temperature of -81C with CO2 and -91C with Nitrous oxide.

Later, i figured out that the source of my hydrogen contamination in my Ethylene production wasn't from excessively hot aluminum oxide catalyst, but rather from an excessive flow rate of ethanol. There was more ethanol flowing than the catalyst could decompose in a given timeframe, and when the unreacted ethylene (which was superheated) came into contact with the copper tubing feeding the condenser, it decomposed to form acetaldehyde and hydrogen. This was the primary source of contamination.

To remedy this, i built a new ethanol boiler consisting of a well insulated 250ml flask with a load resistor placed inside. The load resistor was powered by a bench power supply which allowed precise control of the ethanol flow rate. The reactor works best with under 40W into the boiler. This is low enough to ensure the catalyst can handle all of the Ethanol.

I also removed the copper lines in the reactor and replaced them with aluminum to eliminate any possibility of having copper-catalyzed decomposition that would create hydrogen.

The lowest open-cycle temperature i achieved with Ethylene was -112C, exceeding my original goal for this project of -100C. It's possible to build a cascade refrigeration system to reach liquid nitrogen temperatures, but excessively complicated to the point of not being practical, since it would require 4 stages to work.

The next phase of this project will focus on Joule-Thompson type cooling, but instead of using pure nitrogen or air, I'll be using a mixture of Propane, Ethylene, Methane, and Nitrogen. If optimized, this scheme can exceed the performance of a conventional Joule-Thompson cryocooler by an order of magnitude, while only requiring modest pressures (20-30 bar) compared to the 200+ bar normally needed.

Once again, big thanks to Exotic Chem Lab for sharing his experience in DIY refrigeration and cryocooling to help this project succeed:

   / @exotic_chem_lab  

Music Used:
Kevin MacLeod - Lobby Time