The basics of the Birkland-Eyde process are pretty straightforward. Lightning strikes have enough energy to break nitrogen and oxygen bonds, and allow the freed atoms to bond together into nitric oxide, which further oxidizes into nitrogen dioxide. The difficulty in the process is mostly in developing a good method for controlling the electrical bursts and cooling down the released gases. Certain portions of the process, like generating high-voltage, high-frequency electric current, shouldn't be too difficult with our current technology, and it shouldn't take much more work to also generate a DC voltage to drive an electromagnet to control the arcs, though tuning it will probably take some time.

After we actually start to generate the nitric oxide, we can choose to split the gas into different pathways, one for making nitric acid, and one for making sulfuric acid. I'll also need vessels and components that won't oxidize, breakdown, or melt in the conditions the reactions will take place. Glass, lead, and purified lightstone are good choices for these, though steel will probably need to be used and replaced occasionally for things like the arc electrodes.

Since I collected all the components I needed in advance, I was able to speed through assembly for quite a few of my initial test parts. I used two small mana engines to drive a generator and electromagnet respectively, and used steel electrodes inside a glass reaction chamber with a transformer and a basic electromagnet outside to try to control the electric arcs. I built a heat exchanger on the downward side with lightstone, and a water based inlet check valve to ensure one direction of gas flow.

Building out that apparatus took twenty days. Of course, the first run wasn't great. Some fixes were pretty straightforward, like a larger check valve and more of an air gap between the glass reaction chamber and the electrodes inside it. Other changes to get it working to a good degree ended up taking another 44 days before I was confident that I would be capable of upscaling and still making things work. A major difficulty was properly cooling the hot gas products as well as getting the design for the controlling electromagnets to work.

The heat exchanger I'd designed didn't have nearly the total heat transfer necessary to continually cool the product gases. I ended up changing it to a closed loop water heat exchanger with a secondary copper radiator and pump to cool the gases much faster, which meant I needed a third mana engine. The electromagnets also needed to be tuned to be something approximating opposing disks partially encompassing the reaction chamber. The result was cool disk shaped electric arcs that seemed to produce the most product gases.

Once I got the design somewhat functional and capable of producing nitric acid by bubbling the gasses through water in a glass chamber, I thought it would be a fairly straightforward process to upscale production. There were a few major issues that made me realize that wouldn't be the case. With the upscaled electrical power and the need for larger electromagnets to control it, heat became a major issue.

I found that I needed to focus on keeping the electromagnet cool, and needed to change various aspects of the electrical contact design, resulting in needing to change the electromagnet design again to account for the changes to the internals. What I thought would be a relatively simple process ended up taking 106 days. The final design ended up so far removed from the initial design that I've realized I should have probably spent less time optimizing the small design once I new it was possible to produce nitric acid with an approximation of the method.

Now, however, we have a facility capable of producing a nitric oxide, nitrogen dioxide, and nitric acid at a moderate scale using a few large mana engines. The actual output, if run for a full 24 hour period is about 12 gallons of nitric acid. I haven't tried titrating it yet, but it reacts violently and for long periods of time with a sample of calcium carbonate I brought along, so it's probably at least somewhat concentrated. I'll try to do a crude titration with soda ash, which is more water soluble, to determine a more precise concentration.

The author's narrative has been misappropriated; report any instances of this story on Amazon.

The next thing to build will be the lead chamber and sulfur burner to let us make sulfuric acid in industrial quantities. After that, it's a fairly straightforward process to produce diethyl ether and nitrous oxide to use on the eagles. For diethyl ether, you just need to mix sulfuric acid with ethanol in a temperature controlled environment, and then distill the products. It'll take some testing and ingenuity to get it working well, but I don't expect it will be too difficult, at least when compared to some of the other projects I've worked on. Nitrous oxide is a decomposition product of ammonium nitrate, though we'll have to be careful there as well, since it's explosive.

Adding in the lead chamber for producing sulfuric acid was pretty straight forward, and since the nitrogen dioxide is functionally a catalyst in the reaction, though there are losses due to it's gaseous nature. Ultimately, building the lead chamber and sulfur burning reactor to produce sulfur dioxide took 33 days. Rather than using a true "lead chamber" I'm using a lead packed bed of beads, to make the process more efficient. So the beds are long cylinders that end up having their own water sheath and radiator setup to remove the excess heat the reaction produces. In the same 24 hour period that we could make 12 gallons of nitric acid, if we redirect all the product gasses into this process, we can make 70 gallons of sulfuric acid of unknown concentration.

So, the next thing I was obviously interested in was a good approximation for the concentration of the two acids. So, I began the long process of titrating them. The first step was calculating the different densities of both acids, then the saturated soda ash water and pure soda ash. I had to keep in mind that soda ash is a double base, nitric acid is a single acid, and sulfuric acid is a double acid as well. I can, at least, get an approximation for the molar masses of these acids and bases because the atoms are low enough on the periodic table that I can get close by just using double their atomic number. I probably have some error, but I can't recall the exact average atomic masses for each element anymore.

Titration was a pain because I also didn't have access to any indicator fluids, so I essentially had to repeat the experiment over and over again, while taking tiny samples of the mixture, and dripping it on a solid base to see if it reacted at all. It wasn't very precise, but ultimately, I did get numbers for the two acids that were probably within about 10% of the actual molarities.

After 20 days of testing and retesting, I came up with about 55% solution by weight for nitric acid, and 70% solution by weight for sulfuric acid. Which would mean they're pretty strong acids over all. By molarity, that'd put nitric acid at about 12 M and sulfuric acid at about 13 M, if my approximate molar mass calculation is correct for each type. If that's accurate, then these are quite concentrated acids.

The next processes I want to work on are automatic thermal regulation systems to allow me to more precisely control reactions. Until now, most of the reactions I've worked on have either been thermal excess reactions, or self regulating reactions. I didn't actually need to control them very accurately. Now, however, we're moving into that territory. So, I have a few devices that I plan to try to develop moving forward.

First, the bimetallic strip, which allows automatic triggering of certain effects at a known temperature. Second, very basic electronic circuitry to allow the bimetallic strip to complete a circuit for control purposes. Third, servo motors to allow electrically driven motion for control purposes. While we could go through the intermediates of other control processes, beginning the process to develop electrical control systems will benefit any future projects greatly, so I believe that this basic first step will probably be worth it.

If you find any errors ( broken links, non-standard content, etc.. ), Please let us know < report chapter > so we can fix it as soon as possible.