How Do You Make Rocket Fuels?

There's a huge variety of rocket fuels in use - RP-1, LOX, LH2, Methane, Hydrazine etc - Large quantities are needed, usually shipped to the launch sites in tanker trucks, but, how do the fuels actually get made?
The industrial processed involved use chemistry, physics and even biology to create the final product and they're the result of decades of research and experience.

Follow me on Twitter for more updates:

I have a discord server where I regularly turn up:

If you really like what I do you can support me directly through Patreon

[RageDavis]

Well, I´m a chemical engineer, and basically, you just broke down the essentials of the different processes:
Mixing the chemicals according to the recipe, add catalyst as required (e.g. if an iron oxide catalyst in required, a rusty piece of junk from an old car will do the job, yet it will not be as effective as smaller particles with a good dispersion in the reactor, in other words: the reaction will take longer), set pressures and temperatures, give it some time for the chemistry to happen and proceed to downstream processing (rectification...).
On the process of steam reforming for hydrogen production: yes, it does release CO2, but burning fossile fuels to make electricity and use the electric energy to do water electrolysis would be less efficient, releasing even more CO2 than steam reforming itself.
The details of processing can be summarized as following:
- handling toxic chemicals (hydrazine) requires a vast amount of savety equipment: overengineered vessels and pipes, sensors to monitor unwanted releases, protective gear for the workers
- handling corrosive substances does require expensive materials that are able to withstand the corrosive stuff and savety equipment for the workers of course
- handling cryogenic substances requires loads of insulating materials, special lubricants, pumps and valves with extremely tight tolerances and protective gear, hydrogen production and storage in particular requires loads of sensors to detect hydrogen and special precautions to prevent the formation of an explosive mixture with oxidizers including plain air
- handling pressurized substances requires the vessels and pipes to be strong enough, which usually means beefy wall thickness of the vessels, piping and instrumentation.
The tricky part is to develop sensors for the different chemicals, that are sensitive enough to detect contaminations at a rather low level, but are insensitive for other stuff.
The rest (designing a plant that is save, from fire extinquishers to the girth of electrical wiring, from escape routes to ease of maintainance) is a matter of experience that has been accumalated over the last century and is appied on any chemical plant to a certain extend, as required by local laws and regulations.
So yes, when you look into the details, it gets complicated very soon, but that is exactly why there are experts like engineers after all:
Anyone can build a hut from branches that were cut from trees, but it requires lots of knowledge to build a skyscraper people enjoy living in.

Last but not least: Yet another great video for those that are interested in rocket science.
Two thumbs up, Scott!

[donhull]

I am a chemist, retired now, and you did an excellent job of summarizing for the non-chemists how these rocket fules are made.
The chemistry is well known and the really hard part of making rocket fuels is not the chemistry, it is the chemical engineering, the mechanics if you will, for . Your diagrams and explinations did a great job of explaining what the fuels are and how the mechanicl processes work.

[xyzzyx]

Scott, I as a PhD chemist, am impressed how layman-y you managed to make all the process sound. If I had to explain all to a layperson I could NOT be able to simplify all so much, yet I think you managed to keep pretty much all chemically accurate yet clear to a non-expert. You even got the parahydrogen thingy accurate. Bravo, from someone with a PhD from a top-A-tier university in the US. You are a great science communicator

one minute point I would have made at the end is that all these chemical steps pile up such that at the end you pretty much compress as much chemical energy you can in a certain mass/volume so it can become rocket-relevant. all those steps are like walking up a stairway, and after you do it enough, you reach the top of the Empire State Building, ready to release the potential energy you stockpiled. you kinda need the chemical equivalent of that for rocket fuels.

[Fragaut]

4:40 The hydrotreating happens AFTER fractionation. Then, you fractionate a second time to eliminate the light ends which are typically produced from hydrogenating molecules like sulfides (R-S-R' + 2.H2 -> H2S + R-H + R'-H) or thiols (R-SH + H2 -> R-H + H2S) and from the inevitable cracking of a part of the feedstock (few % typ.). If you want a precise cut, you may also want to cut the bottom at high reflux, to eliminate heavy ends that were either created by the odd polymerization (very small fraction << %) or deliberately let through the first fractionation, for plant optimization reasons or when coprocessing kerosene and diesel fractions in the same unit. Some of the hydrotreating feedstock also comes from cracking heavier fractions through a FCC or a HCU and is typ. already partially treated. It all depends on the refinery setup. Those plants are optimized to an insane degree and some of the trade-offs and feedstock routings can be VERY counter-intuitive.

Then - not sure if they do that for RP1 - but if you want a truly super-clean product - like high quality food-grade mineral oil - you hydrotreat a second time with special hydrosaturation catalysts, typ. using platinum, ideally at comparatively low temperature and high pressures, to displace the equilibrium as much as possible towards saturation of aromatic rings with hydrogen. Whatever double bonds are still present are also saturated, it's essentially irreversible and there's very little of them in already hydrotreated feedstock, anyway.

The operation does not normally modify the boiling range enough to justify another fractionation, only a stripping to recover the excess H2 and the odd C1/C2 produced. A small purge on the hydrogen recycle is normally all that's needed to control the light-ends. If anymore is produced, there's a big problem with the feedstock or the process control and the very expensive catalyst in the reactor is likely dead.

Note : those hydrosaturation catalysts are very sensitive to so-called "poisoning", when contaminants - mostly sulfur - react irreversibly with the catalyst and kill its activity. So the feedstock must already be very clean, which is why it cannot be done in one shot, directly from crude oil fractions. Those units also typ. have so-called guard beds of specialized sacrificial catalyst to capture remaining contaminants before the feedstock reaches the hydrosaturation catalyst proper.

6:58 The regenerative counter-current heat exchange between the incoming high pressure gas and the cold low pressure excess gas from the expansion stage is not just an optimization in the Linde cycle. It's its key feature. The core idea of this cycle is that you use the gas you are liquefying as its own refrigerant. The first drawing shown from Wikipedia is truly awful 😄

11:47 At the reforming conditions, it's mostly carbon monoxide that's produced. CH4 + H2O -> CO + 3.H2. It's also a deeply endothermic reaction. There's much more energy in CO + 3.H2 than in the original CH4 molecule. It's done by running the reaction at high temperature (800-1000°C) in long tubes radiatively heated inside a furnace. So, in most plants, there is a second reactor to make more hydrogen from this carbon monoxide using the so-called WGS (Water Gas Shift) reaction : CO + H2O <-> CO2 + H2. The lower the temperature, the more the reaction goes towards making hydrogen, but also the slower it goes and the less active the catalysts are. So, it's often done either in two steps or only partially, using the leftover CO, unreacted CH4 and some of the H2 to heat the furnace, after separating the product hydrogen.

So, there. Ok, that's enough chem^H^H^H^H pedantry for today 🤣

[MrHichammohsen]

Man even your way in explaining chemistry is so clear, we can't thank you enough.

[brad_marston]

Small correction: it is the nuclear (proton) spins that distinguish the para and ortho forms of H2, not the atomic spins which usually refers to the electron spin. In the H2 molecule the electron spins are in a spin singlet, it is the total spin of the two protons that distinguishes the two spin isomers.

[MeteorMark]

Thank you for explaining all this Scott!
At our integrated Steel Factory we have several Linde Gas Air separation plants, supplying us with mainly Oxygen for several production processes.
Big tanks of LOX, LN2 and Argon, also transported off site to other Linde Gas customers in the food industry.
We also have big pressure tanks for fire suppression in certain areas, filled with either Argon or the latest Argonite A2N2 mixture.
Smarter everyday! 😉

[SuperR]

Making me think a lot about Ignition! An informal history of liquid rocket propellants by John D Clarke. Fantastic read and fun video!

[Doping]

The hydrazine synthesis sends chills down my spine. NaOCl and ammonia easily gives you NCl3, an absolutely terrifying explosive. I don't want to know how many injuries were incured to dial in the process parameters to make it work.

[bobharris]

I took a year of organic chemistry in college. I almost understood what you just said. Very impressive. 👍

[PaladinofRealm]

Chemistry student here: It is relatively easy to synthesise things like Hydrazine on a large scale. The hard part was getting there, and optimising the process. And the challenge of keeping the production affordable in the face of economic factors.

[arie]

Great reference to John D.Clark's Ignition! It is an absolutely fascinating read (though a bit technical at times for non-chemists, I presume). The gist of it is that it's not really difficult to synthesize some of these things, but handling is an entirely other matter

[AsbestosMuffins]

never thought H2 would have an ortho or para allignment state, i only ever thought things with larger structure could have different alignments

[lucasmoore]

Ph.D. chemist here. Pretty good explanation!

The hard parts of these processes are the exact pressure, temperatures, catalysts, and equipment used to run them. Almost all of which are industrial secrets, patented, or a mix. So if you somehow had more details to share, you’d probably get some strongly-worded letters from attorneys. Best to let them keep their secrets.

[blindsniper]

You forgot about the solid fuels. They're pretty interesting as well.

[chasewilbur]

You explained the chemistry really well. There are a lot of little things to do and get right for every different type of chemical reaction. For this type of video and audience it would be way too much to get into that level of detail.

[zeke]

I worked in a cryogenic production plant in the US Air Force many years ago. We produced liquid oxygen and liquid nitrogen. The two are separated through distillation. It's basically the compression of air then the expansion, either through an expansion valve (high pressure plant) or through a turbo expander (low pressure plant). The now liquid air was directed to a distillation tower where the heavier oxygen settled in the bottom and the nitrogen on top.

[peterallen]

Honestly, something that never ceases to amaze me is just how much chemical engineering has to go into different colors of dyes and paints.

[skister]

I'm so glad I found your channel.

Your delivery is amazing, its light and breezy not preachy and boring.
You take complicated topics and brake them down into easily digested chunks.

I wish you all the best with the channel.

[mortisCZ]

Hi Scott, chemist here. You have explained it rather nicely without going into specific details.
I am glad that chemical industry gets its fair share of appreciation because we tend to get only more than fair share of cancer.