[Enthalpy]

Hello dear friends!

Rocket use the fuel to cool the engine walls, where polymer deposits would be deadly. This precludes unsaturated fuels up to now. Though, cooling the engine with oxygen would enable alkenes and alkynes, and for instance plain ethylene is very efficient. This significant change looks easier than the existing oxygen-rich staged combustion and turbine.

William P. Dailey reported in 1992 a promising synthesis for spirohexene
www.dtic.mil/dtic/tr/fulltext/u2/a248465.pdf
just 1s short of cubane's performance, so I checked his cyclopropenes and similar alkenes.

Would you know data about the oxygen-free detonation risk of alkynes? Sensitivity to shocks, detonators, heat... I found
www.dtic.mil/dtic/tr/fulltext/u2/028238.pdf
but the result pages 28 and 75 are classified since 1953. Butyne is known to detonate, bigger compounds are unclear to me.

====================

In the appended list, the safe and affordable "kerosene" RG-1 is the performance reference. Methane engines are being developped currently. Spiropentane (and cyclopropane) are alternatives to methane, and cubane the fantasy of rocket designers.

The Isp is 99% of what Propep computes at optimum mass ratio. The gain in seconds versus RG-1 is divided by G. Flammability is "bad" up to C₉. Heat sensitivity indicates the °C where 10g decompose from 1000 moles in 300s; for ethylene it's at too easy 0.13bar.

Tetravinylmethane and divinylcyclobutane don't justify to cool a rocket engine with oxygen, but they outperform cyclooctatetraene for ramjets. Tetravinylmethane is less flammable, and divinylcyclobutane is the dimer of butadiene over a catalyst.

[3]-Ladderdiene is storable and efficient. To produce it, just try to make cyclobutadiene, and you get the dimer - looks scalable to 100t.
https://en.wikipedia.org/wiki/Ladderane#Dimerization_of_cyclobutadiene
Seems interesting for Ramjets too. [4] or [5]-Ladderdiene would be less flammable.

Cyclobutene looks simple to produce, but less so than ethylene which is the prime incentive for oxygen-cooled engines.
https://de.wikipedia.org/wiki/Cyclobuten (or French or Spanish)

Dailey gives synthesis paths to dimethylcyclopropene (rather trimethyl, and with a hydrazine, yuk) and to spirohexene (looks sounder and scalable if recycling the byproducts). To evaluate Hf=+333kJ/mol for liquid spirohexene, I observe that spiropentane has 34kJ more strain energy than two cyclopropanes, add arbitrarily half that at spirohexene, and neglect any spiro-ene interaction. My estimate gives a bigger performance advantage than Dailey's to spirohexene over dimethylpropene; spirohexene is also easier to produce and less volatile.

The saturated but fragile bicyclobutane fits in the list as I wouldn't cool an engine with it.

And because Dailey achieved cyclopropenes in spirohexene in dispirodecadiene, I imagine to adapt his synthesis to dispirooctadiene. The start is simpler (heat propadiene over glass beads), the result would outperform cubane. Full debromination would afford dispirooctane or "soctane", a rocket fuel in the +5s range. One alternative debromination there
http://pubs.acs.org/doi/abs/10.1021/acscatal.6b01914
it's a question of reactant cost and ease of recycling.

Dailey told that spiropentene is "unstable" and spiropentadiene "very unstable" (on a chemist's scale) while the fuels above are at least unknown.

Marc Schaefer, aka Enthalpy

[Enthalpy]

A logical upgrade to the dispirooctadiene would have one (spiro) cyclopropene more, or trispirodecatriene, to be less flammable... The precursor trimethylenecyclobutane does exist and has a synthesis path described by US patent 2917499
https://www.google.com/patents/US2917499 (click on Pdf for drawings)
This precursor can be stored at -80°C, oh good. At room temperature it dimerizes and isomerizes. Whether a cyclopropanation can be carried on it?

[Enthalpy]

I have found no data about the detonation risk of alkyne fuel candidates. Just because pentyne is not reputed to detonate without air, I've checked a few unstrained alkynes with 5 carbons per triple bond. They are only 5s better than RG-1.

Saturated strained amines bring the same performance, look easy to mass-produce (diazetidylcyclopropane), should resist some heat, can offer a high flash point
http://www.chemicalforums.com/index.php?topic=86972.0
so for me, unstrained alkynes are out.

[Enthalpy]

The oligomers of allene are almost rocket fuels, just waiting for a transformation.

Weinstein and Fenselau analyzed the oligomers of allene, both by bare heat and with Ni(0), and found not only dimethylenecyclobutane, but also dimethylenespiro[3,3]heptanes, trimethylenecyclohexane and more.

Trimethylenecyclohexane would give trispirododecatriene, if Dailey's synthesis applies with one methylene more. Better performance, but access to the middle methylene isn't very clear. The major intermediate is 1,2,4- mixed with little 1,3,5- that gives more likely a solid, possibly soluble.

The proportions of dimethylenespiro[3,3]heptanes wasn't clear on the abstracts but at least propadiene is cheap. As the access to the methylenes and strain resemble methylenecyclobutane much, raising hope for trispiroundecadiene less volatile, denser and slightly more efficient.

I recomputed the Hf of dispirooctadiene and find even +22s now.

----------

And since Sherburn et al could cyclopropanate dendralenes to ivyanes
https://cen.acs.org/articles/89/i6/Meet-Ivyanes.html
maybe the dihalocyclopropanation for Dailey's synthesis applies too? Limited access is a question. The resulting ivyenes would have excellent performance and probably the best melting point here.

----------

All these heat-sensitive molecules are meant for engines cooled by the oxygen, not by the fuel. Excepted dispirooctadiene, they are little volatile and might have a safe flash point - that is, if a flame contact doesn't decompose and light the liquid directly. +15s is the performance of cubane, wow.

Marc Schaefer, aka Enthalpy

[Enthalpy]

Preliminary trials of an oxygen-cooled chamber can be very simple.

In a first check or demo, one can just fold a thin copper sheet as a tub, fill it with liquid oxygen, and put an acetylene+oxygen burner below, as sketched below on left side. I did it with water, it worked immediately. Just observe that copper shows no reaction.

A more refined model, as on right side, can represent roughly a combustion chamber by winding a rectangular tube as the wall of a cylinder. The width, height, thickness, length, oxygen throughput of the tube can match one cooling channel of a known or projected chamber. The liquid oxygen pressure can optionally match the chamber's operation, so evaporation occurs under identical conditions, if it does. The transferred heat per unit area resembles already the real chamber; this is best observed in the next step, a true chamber, scaled down or not. Usual materials for cooling jackets, low alloyed copper and nickel, are recommended for oxygen, even at heat. Cobalt is easily co-deposited when electroforming and it hardens nickel; molybdenum can be worth trying.

==========

1-methylcyclopropene is available commercially to stop fruit ripening. It exists pure but serves as a 0.14% solution, so its cost per 100t is unclear. From my quick estimate, performance as a fuel resembles the astounding bicyclobutane. Msds don't mention the stability. The already cited Dailey reports some sooting for similar carbon-rich compounds.

Marc Schaefer, aka Enthalpy

[Enthalpy]

Ever worse... Dicyclopropyldiazetidine, but with a double bond in the 4-ring. 1s Isp gained if oxygen cools the engine walls. Of course I ignore if the compound exists and can be produced and is safe enough as a rocket fuel. For cyclobutene it's yes, for diisopropyldiazeditine too, but the combination? At least:

• Cis-ethene exposes both Br simultaneously to the N-H, which may help this step.
• The tension in the N-N bond eases.
• In case this step eases, it might improve the path to dicyclopropyldiazetidine.

I display the recycling of HBr too, because many people claim that BrEtBr is expensive. Electrolysis produces efficient HOBr. Good for many productions.

Marc Schaefer, aka Enthalpy

[Enthalpy]

[2+2] photocycloadditions work for many varied compounds
  ncbi.nlm.nih.gov
but fuels need ultimately the parent product while published additions deal with easier substituted compounds.

If the parent products are inaccessible, then back to halogenated ethylene and to dehalogenation and dehydrohalogenation. Nothing tragic, this eases the wavelengths, and halogens recycled on site cost only electricity.

I described an LED source of 0.5MW UV light, at 245nm that's 1mol/s photons, and at 150€/MWh, photons cost 0.06€/mol
  chemicalforums
Measuring 0.5m×1.1m×0.6m, the source is scalable.

========== Cyclobutene, [3] and [4]-ladderanes

The cited compilation includes alkenes+alkynes that produce cyclobutenes. That would be a dirt-cheap path to efficient cyclobutene and to [3] and [4]-ladderanes.

========== Ladderdienes

Maybe cheap acetylene and light make ladderdienes in one pot.

C₈ [3]-ladderdiene has low boiling hence flash points and must react further. Estimated Bp: [3] +129°C, [4] +168°C, [5] +205°C. [4] is as reluctant to flame as turpentine and can contribute to a mix centred around [5]. A mix should also improve the Mp, an isomer mix of syn, anti, cis, trans too.

Marc Schaefer, aka Enthalpy

[Enthalpy]

Could someone kindly provide me a UV absorption spectrum for acetylene, or a link, please? As usual, I squandered my time to get nothing.

I suspect the spectrum for acetylene resembles much ethylene, which doesn't help, and the excitation from Homo to the next state needs a short wavelength uneasy for LED, while suitable older lamps are less efficient. With some luck, a sensitizer accepts longer wavelengths. If the compounds absorb light too slowly, liquefying them helps. It's also a matter of quantum efficiency and of consumed light and electricity.

In case the direct photocycloaddition of acetylene is impractical, consider some dihaloethylene instead. Longer wavelengths will bring tetrahalocyclobutane, then a metal is to make the transient cyclobutadiene that dimerizes to 3-ladderdiene. But if the metal produces a chain of bicyclobutyl, keep it, it's better booze than ladderanes.

Cyclobutene and 3-ladderane synthesis could just illuminate dihaloethylene diluted in ethylene, restore a double bond using a metal, and optionally photodimerize the cyclobutene. 4-ladderane would add an acetylene step.

[Enthalpy]

Propadiene provides methylenecyclobutanes, just by heating over a catalyst, quickly and with good yields. So would the very cheap compounds make decent propellants?

They are unsaturated, the most abundent is a conjugated diene. Maybe some additive stabilizes it, but oxygen must cool the engine walls. The dimers are very volatile liquids, the trimers are easily flammable too, undesired. But the performance exceeds fuels difficult to produce.

So could the unsaturated dimers provide safe efficient C₁₂ tetramers?

One team failed with metathesis because 1,2-dimethylenecyclobutane poisoned all catalysts as a conjugated diene. The product would be a marginally flammable C₁₀ anyway.

I hope photocycloaddition does provide a wide mix of tetramers, "tetrallene" collectively. The oxygen must still cool the engine walls, but the C₁₂ would offer an excellent flash point and density. The melting point can't be forecast, the molecule is stiff but spiro help.

Alas, in 2023 nobody has measured the enthalpy of formation of spiroheptane. Papers give useless software estimates, or compare them, or deduce bizarre models, and journals publish that. So my attempt telling 3s over Rg-1 is not credible. Intuitively, it must be +5s, maybe +6s.

Marc Schaefer, aka Enthalpy