Out of each five gallon process, I could reliably pull out 9 half-gallon containers before it was difficult to pull more material out of the column. Due to the nature of a distillation column, the composition of the material I pulled out would depend on both which end of the column I extracted from, as well as what had been extracted prior. For example, if I pulled all nine samples from the top of the column, I would get different compositions than if I pulled from the bottoms.
To attempt to get my bearings on the composition gases, I pulled a sample from the top, then the bottom, and tried sticking a flaming stick into a small amount of each. The top sample didn't react, but the bottom sample flared up, likely meaning that liquid is rich in liquid oxygen. It's very likely then that the top is liquid nitrogen rich, but I'd need a different test to verify that for certain. I would expect argon to make up between 0.1% to 2% of the atmosphere, dependent on various factors like atmospheric thickness, and planetary age.
To get a rough estimate of oxygen concentration, I recharged the column with a fresh batch, and pulled quarter-gallon samples from the bottom, and repeated the stick test until the stick no longer flared up. By the fourth sample, it was barely flaring up, and the fifth sample no longer flared. With a little room for error plus the small percent of oxygen likely remaining in the column, I'd say that we have between 15% and 22% oxygen in our atmosphere.
I then resumed using the balance to try to determine the presence of a third gas. Results, while somewhat small, did seem to indicate a denser gas that has a boiling point between the two other gases. I tested this by drawing out different samples of gases from the tops and bottoms in a row, and testing them against each other. At the point where the oxygen starts to disappear, the density rises slightly, before falling quickly and remaining low. Considering it's a small spike in density in about one, sometimes two samples, followed by the remaining 4 gallons or so being lower density, I'm guessing that it's argon.
At this point, I've got about a month worth of water left in the reservoir to use before I'd need to wait until next spring to resume testing, so I really need to make things count. I could easily try to charge oxygen and nitrogen into a crystal growth apparatus to grow a crystal on the mountain currently. I currently am incapable of charging argon. What I would like to do is collect multiple of the argon rich samples, and throw them back into the distillation tower and enrich them further. For that, however, I'd need another small stirling cryocooler, with a cold side head that could be inserted into a dewar flask and kept relatively sealed, to prevent evaporation of collected samples while I distill them.
Given we're only actually making about a third of a gallon of liquid air a day, I don't think we'll actually make enough to re-distill out any reasonable amount of argon. We might, however, be able to pull a rich enough amount out of the column to at least attempt to charge a crystal apparatus with it. To achieve this, I'll initially pull off about two and a half gallons from the top of the column, then pull about a gallon and a half from the bottom, after reaching equilibrium again.
That gallon and a half will get mixed with liquid air again, and put into the column. Some basic fractional math will let me calculate how much more to pull off the top next time, and so on. After three cycles, I'll pull off some of the oxygen from the bottom as well. In doing this, I'm basically using the column as a cooled container where I'm slowly concentrating the argon, if it is argon.
----------------------------------------
As I hauled a dewar flask up the mountain, I wished it wasn't summer. I filled three flasks with what I considered to be the best samples I could for each of the three gases. I have a goblin on either side of me carrying one other flask each. Each flask is actually fairly small, only holding about a cup of liquid. Since I broke my smallest crystal growth apparatus, I'm going to be modifying the next largest size, and testing growth with what few crystals fit in that apparatus that I have left.
This tale has been pilfered from Royal Road. If found on Amazon, kindly file a report.
I plan on testing the oxygen last, given how high risk it is to actually use. I plan on starting with nitrogen, then moving on to the argon. That said, the argon test is actually still majority oxygen. If the argon test yields any results, the oxygen test should inform me of which of the gasses actually contributed. If none succeed, then I think I'll put the project on hold.
----------------------------------------
Well, I ruined this crystal growth apparatus as well, but not before getting some promising results. I modified the apparatus to have a gas charging chamber attached, where the liquid gases could boil and then enter the chamber as a gas through a stone-shaping operated valve. I had the two goblins I brought along remelt the old crystal slag, and then would slowly try to grow a crystal like I did previously. The nitrogen was useless, as expected.
The argon rich oxygen produced a very small amount of growth. The pure oxygen, however, did not produce any growth, and in fact, irreparably weakened the container, causing it to collapse when I tried to pull a vacuum to empty the chamber at the very end of work. I'm sure the argon test contributed, given it's large percentage of oxygen. Oxygen is so reactive, that I'm guessing it modified and weakened the structure to the point of failure.
The argon test being successful, even if minutely, was a huge win. It means I can move forward with the next stage of this whole process. First, I'll need to fix the two broken crystal growth apparatuses. Next, I have a few ideas for how I'd like to proceed with processing the argon to make it more pure. First, I'll need to make two more small cryocoolers, designed so that their cold-side piston fits snuggly in a five gallon dewar flask. That way I can use the cryocooler to keep whatever liquid inside in that state. Then, as I use the distillation column, I can pull of the quarter of a gallon of oxygen that was rich in argon.
By repeating that process 20 times, I'll have enough of the argon rich oxygen to repeat the process in the column again, and this time, I'll pull the argon from the top of the column. Though we might run into production issues at that pace. If everything works perfectly, we only produce about 60 gallons of liquid air in the operational months, but I want about a hundred for processing.
Even then, the total volume we produce on the tail end of that isn't that high. Depending on how the argon actually forms in the crystal structure, it probably won't add much more to the volume of a crystal than it possesses as a liquid. Adding a gallon of argon to a crystal apparatus, for example, probably won't grow the crystal by more than a few gallons itself. Considering a gallon is only about an eighth of a cubic foot, we'll need a lot of argon.
I'll eventually be able to check this, but for the sake of argument, if I had a gallon of liquid argon, and it grew a crystal by a cubic foot, I'd still need nearly 700 gallons of argon to get a crystal the size of ParTor. In short, our facility would need to be scaled up considerably for that to work. Even if it actually grows the crystal by ten times that amount, we'd still need 70 years to reach our target.
That is, it would take that long if we only relied on this argon production and the old crystal waste material. We do have the mana crystals that the miners are currently harvesting as well as the handful of leftover crystals in the cave. Next year, I'll probably also bring on some stone shaping goblins or lesser earth demons to take over the argon liquid gas production process, freeing me up to begin work on other processes.
As byproducts of the argon production, we'll have a decent amount of liquid nitrogen and liquid oxygen. The liquid oxygen actually would be quite useful to aid in our ore roasting process. We could simply put a container of liquid oxygen to evaporate near the air intake for the roasting ovens, and the extra oxygen content should roast the ores more thoroughly, removing a larger percentage of the sulfur. The liquid nitrogen would also be a fairly pure source of nitrogen to use for the previously discussed Haber process for making ammonia. In fact, the oxygen would also be useful for attempting to make the iron oxide catalyst for the process as well, though the Haber process is still quite low on my priority list.