[Enthalpy]

Hello nice people!

Spiropentylspiropentane is a seducing rocket fuel candidate: efficient, not too flammable - and spiropentane was synthesized in the 19th century.
http://www.chemicalforums.com/index.php?topic=79637.msg290422#msg290422
So I think at it further and would like your comments and suggestions.

The target fuel should contain supposedly:

• No or little of the too flammable spiropentane.
• Spiropentylspiropentane. It has isomers whose melting point must differ, but I ignore which one is better, and a mixture can be best, so let's say we don't care.
• Some dispiropentylspiropentane is welcome to raise the flash point and lower the melting point. 10%, 30%, no idea - a process that makes di-and tri-spiropentyl at once and adjusts the proportion would be perfect. Any mix of isomers.
  
• No or little tetra-spiropentyl, as it must precipitate at cold.

I've read that neopentane isn't available from gas, oil, coal and is expensive. Is that correct? Start from pentaerythritol then?

Basic descriptions of the (Wurtz?) coupling tell just sodium or better "zinc"... Is that a coarse-grained inert zinc powder? Or a sub-nano-mega-activated material that detonates upon air contact? In this case, metal vapour (if it works!) may be safer because a quick reaction processes less reactant at a time.
http://www.chemicalforums.com/index.php?topic=72951.0

Is Appel's reaction the proper way to the tetrahaloneopentane: cheap, big?

Do you feel any reasonable to add a fifth halogen before the Wurtz coupling as sketched? (And even a sixth one to obtain the tri-spiropentane proportion). I suppose this extra halogen will seldom react within the molecule, because bicyclobutane is even more constrained, and would be preserved for intermolecular coupling.

If adding a fifth and sometimes sixth halogen, which one? Neopentane tetrabromide doesn't look very crowded (appended image). Chlorine, for easier gem-addition and use after the bromines?

Or should plain spiropentane be made first, then coupled? Iodine and well-tuned violet and UV light could be more caring with the strained molecules. Like: violet light to dissociate I₂ and obtain spiropentyl iodide without chain reaction, then 254nm to abstract the iodine atom so two spiropentyls couple.
http://www.chemicalforums.com/index.php?topic=77307.msg283221#msg283221

Thank you!
Marc Schaefer, aka Enthalpy

[Enthalpy]

So, do you see a chance that two pentahalo and a divalent metal give di-spiropentyl at once?

[discodermolide]

You will be creating a lot of strain in your product. Probably treatment with a metal will not produce it but rather just de-halogenate the starting material. Using samarium diiodide could maybe produce the compound by some SET reaction but this, if it works, may well make any process very expensive.

[Enthalpy]

Thanks Disco!

I had hope that at least the synthesis of spiropentane from the tetrabromo and zinc isn't too difficult, since it was done over a century ago - but Sapiens were already clever then, sure.

My (poorly substantiated) guess is that if two tetrabromo and linked before the metal acts, they will make a bigger cycle instead of two spiro.

Any chance to bind two or three spiropentanes once they're formed?

[discodermolide]

Sure you can couple them all together but usually you need a halogen on one and you need a highly active iridium or ruthenium catalyst that will activate a C-H bond of the other. However exactly which C-H bond will be activated is difficult to predict. Some studies have been done on cyclohexane or decalin systems but I don't have the references to hand atm.

[Enthalpy]

Thank you!

Well, at spiropentane there aren't so many different C-H, so isomers would begin with the "trimer". Since such isomers will all burn equally well, the difference is the melting point, and I can't predict hich isomer is better.

Maybe a wide, suboptimum mix of isomers is acceptable anyway. After all, a launcher's first stage never gets cold, so the lowest freezing isomers could just be separated from the mix and kept for the more demanding uses.

[discodermolide]

The trouble with synthesising isomers is always getting the same mixture. Achieving this may be difficult and depends on so many factors, as I'm sure you are aware.
If you had an activating group, not necessarily a strong activator, just something to differentiate the C-H bonds, things would be easier. Something like a OH or a methoxy?

[Enthalpy]

[...] as I'm sure you are aware. [...]

You mis-under-estimate me.

[Enthalpy]

Well, at spiropentane there aren't so many different C-H, so isomers would begin with the "trimer".

Ahum. Two groups of equivalent positions on the spiropentane skeleton, so the dimer has already isomers. Keep hope, I improve.

[Enthalpy]

For the synthesis of plain spiropentane from tetrahalopentaerythrityl,
https://de.wikipedia.org/wiki/Spiropentan
gives only historical methods with tetrabromide and zinc in ethanol. Unfortunately,
http://hdl.handle.net/2060/19930081859 (739kB)
Nasa's ancestor - and everyone - obtained mainly methylenecyclobutane that way, and at best 20% spiropentane.

Sodium in dioxane improved the yield, so supposing that Zn²⁺ catalyzes the path to methylenecyclobutane,
Applequist, Fanta and Henrikson in "An improved synthesis of spiropentane"
used tetrasodium ethylenediaminetetraacetate to trap Zn²⁺ as it formed and obtained 76% spiropentane.

Less old papers are paywalled.

Vapours of K, Na or Zn would remove ZnX₂ quickly and carry no Zn²⁺, so they might provide the spiropentane.
http://www.chemicalforums.com/index.php?topic=72951.msg298581#msg298581
Excess of metal should favor the internal coupling.

Have one haloatom more on the neopentane to obtain bis-spiropentyl directly, or at least have it ready on spiropentane for the dimerization?

Could pentaerythritol be reacted directly, without halogen? It's volatile enough, but the heat of reaction is unfavourable with K, Na and not quite good with Rb. The heat of reaction would enable a Grignard-like with Zn vapour, but many other reasons will probably let it fail. Worth a try?

Marc Schaefer, aka Enthalpy

[Enthalpy]

Sure you can couple [spiropentanes] together but usually you need a halogen on one and you need a highly active iridium or ruthenium catalyst that will activate a C-H bond of the other. [...]

That's even more useful since the halogenation of spiropentane has a limited yield. The unmodified spiropentane provides half the output without this loss.

[Enthalpy]

Maybe dispiropentylmethane is easier to produce than bis-spiropentyl. 1s less efficient than bis-spiropentyl, it's still excellent 7s over Rg-1 "kerosene", and the software-estimated +174°C boiling point makes it hard to light. The melting point isn't predictable.

My (poorly substantiated) suggestion sketched below is to cyclize methylerythritol (from acetaldehyde methylated somewhere and formaldehyde?) in methylspiropentane, which is known to halogenate easily and selectively - more so than vinyl. Then, maybe after X' is made OH, it could react with excess spiropentane.

Maybe the ring widens instead of coupling with spiropentane, and then it's a path to spirohexane and suggests one to spiroheptane. Or it makes anything unwanted.

Marc Schaefer, aka Enthalpy.

[Enthalpy]

Would you have opinions and lights about this tentative path to spiropentane over azo bridges?

Pentaerythritol shall become the tetramine by some brute's method similar to ethylamines production but with ammonia excess. If some azetidine rings form, keep them, it's good booze too.

Then, oxidative coupling within the molecule is to build two azo bridges. Removing N₂ by heat or light hopefully brings the new bonds and avoids rearrangements, as for bicyclopentane.

I believe to have seen a paper few years old that builds the N=N bridge from two primary amines on one molecule using H₂O₂ but don't find it anymore.
Houben-Weyl, "Science of Synthesis", 41.6.1.1
cites oxidizers like IF₅, supposedly expensive to buy or recycle, but also NaClO and others. N₂O maybe?

The molecule shape optimized by Am1 (beware software!) shows that as soon as two nitrogens on neopentane are bonded, they are farther away from the others. This might favour two double bonds - or not. And to my understanding, hydrogens drift at an amine, so which ones react first shouldn't be an obstacle.

A route over the reductive bonding of nitro groups instead of amines seems more common. Maybe dilution prevents boom.

Methylspiropentane, which shall help make bigger molecules, is hopefully obtained from pentaerythritol with one methyl more, itself resulting from one longer reactant.

Marc Schaefer, aka Enthalpy