There are two main factors to consider when designing the steam cracking section. They both relate to residence time. The first is fairly easy to consider, which is the average residence time. That has to do with the volume of the reaction vessel, and the flow rate into it. Simply dividing the volume of the vessel by the flow rate gives us the average residence time. The second factor to consider is the residence time distribution, which is normally a somewhat complicated calculation to make, and relies on a lot of heuristics. For example, stirring continuous tank of liquid tends to make the distribution closer to the average.

For our particular situation though, this is less of an issue, as we're aiming for a very low residence time. The only way to really achieve that is to have a high flow rate through a small volume. That tends to cause two things that work in our favor. First, very high flow rates tend to cause turbulent flow, which mixes the fluid, keeping the distribution closer to the average. The second is that by narrowing the flow area, we also don't need to worry as much about the geometry of the reaction vessel causing areas with eddy currents that could change the distribution.

The bad news is that we're likely going to have to clean these reaction vessels out frequently, since they'll be a relatively small volume. That means that the buildup of carbon in the form of coke can choke the reactor out and change it's properties fairly easily, which would lead to different yields than we expect.

What that meant for my testing was that I used a fixed sized flower fan that I could use to easily calculate flow rates, and various smaller pipe constrictions and lengths with heat fluorite of different sizes to operate for steam cracking. This was immediately followed by parallel flow heat exchangers to quench the output. There were a lot of potential combinations of pipe sizes and total heat input to check, and we still had to run the products again through separations to determine what products we had.

If I had to do all that testing and setup myself, it probably would have taken well over a year to complete all that testing. Over the course of the semester we produced dozens of different samples that we ran through the new distillation tower on differing settings until we isolated what I believed to be a mixture of benzene, toluene, and xylene, or BTX for short, from one output. So after figuring out what temperatures and settings for steam cracking yielded higher amounts of BTX, I then focused on separating the benzene out from the mixture. Our best yields occurred at higher temperatures in the steam cracker in the shorter end of residence times, giving us about a 45% mass yield of BTX from the initial mix. For comparison, the uncracked mix, when run through again looking for BTX only gave a 10% yield, meaning the steam cracking was improving the yield.

BTX itself has to again be separated to recover each individual chemical from the mix, and similar to when we isolated different gases from air, you have to make certain determinations about what purity you want of each product. There was one thing that I remembered that was very important to being able to determine which of the materials I was recovering, and that was that benzene has the lowest boiling point of the three chemicals in BTX by a wide enough margin to separate it easily. The trick was to initially tune the column to recover a miniscule amount of very pure benzene, then use that as a comparison for the properties of products as I tuned the column to recover more and more benzene at the cost of some purity.

What I found from those tests was that I could recover about 35% of the weight of the BTX that went in as highly pure benzene, at least from the column I was using. A more specialized column could probably recover a higher purity and a larger amount, but not by much. Everything considered from wood to final benzene yield of the current process, that's a mass yield of 1.2% benzene from the initial wood. Old growth trees on our islands can be a few tons, but farmed trees are usually about a half-ton when they're harvested. Considering we'd need to farm trees for this to work, that means we'd need about 200 trees for every ton of benzene we produce.

Stolen from its rightful place, this narrative is not meant to be on Amazon; report any sightings.

With some dedicated research, those yields can increase somewhat, with perhaps a perfect yield being closer to 2%. There are various other byproducts from the process that could all have uses as well in the future, so it isn't like it's all wasted, even the immediate usage of better waterproofing materials can be incredibly useful.

We're going to pursue three avenues of research into the other products necessary to produce artificial rubber in the form of styrene and butadiene. Styrene production requires ethene, or ethylene as it's commonly called. When reacted together with benzene in a controlled manner, we can produce styrene. There is a fairly straightforward method to produce ethene from ethanol using sulfuric acid. However, we're actually probably producing some amount of ethene already from our wood tar distillation and steam cracking. I plan on making some ethene to use as a reference sample, then seeing how much we can recover from our new benzene process at different stages. Since we only need ethene in a one-to-one molar ratio with benzene, even small yields could drastically cut down how much ethanol based production we need to do.

The other component we need is butadiene, and will compose the third research avenue. Butadiene itself can again be produced from ethanol, but it can also be recovered from steam cracking. In fact, by steam cracking heavier products, we might produce more butadiene than we need, at least compared to styrene. That will be another avenue of research moving forward. The final steps necessary for production will be polymerizing butadiene into polybutadiene, which requires a catalyst. These are generally metal oxides, but it's very important to find the right ones, as it affects the way the butadiene links to itself, which affects the output polymer's properties.

That said, I've already informed Zeb of this future production plan, even if the details are still being determined. He had a good suggestion that we do this production on our third island, which has plenty of valleys and trees on it. There, the entire production chain an be isolated to a single location, and any accidental fumes produced will hopefully stay within the valley where the facility is produced, reducing exposure for other individuals. Considering we'll likely want to be producing hundreds, if not thousands, of tons of the various plastics and rubbers a year, it would be nice to start in a location that has plenty of free land for growing crops and wood.

As it would turn out, recovery of ethene would normally be a challenging task, but we've done some of the hard work already. It really wants to stay a gas, as seen from the samples I produced from ethanol. However, it is capable of being made into a liquid at either very low temperatures, or high pressures, or both. Though even at high pressures, it seems it's critical point is below room temperature, so we still have to keep it fairly cool to be a liquid.

That said, we're no strangers to operating at cold temperatures for distillation, and these temperatures are significantly warmer than what we needed for liquid air separations. For testing, I took samples back to Kembora to use the cryogenic towers to try to isolate ethene from various byproduct stages from our benzene production line. Even preliminary recovery amounts were much higher than the amount of benzene we're producing, which was a nice find. The downside is that it does seem like we'll have to do a few additional stages of processing to ensure a pure product, since there were various other mixed hydrocarbon gases present that we'll need to be sure to have removed.

Overall, it only took me a few additional weeks of research before I had another 'good enough' system devised for the process. Since we'll definitely have a surplus of ethene, I should look into polyethylene after I finish artificial rubber research. Polystyrene is also obviously on my radar, but I suspect that styrene will be our bottleneck, so I don't know if polystyrene will be valuable enough to pursue comparatively.

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.