[opsomath]

This problem has been extensively studied. Read the following glorious memoir for a ton of possibilities, and some hilarious stories. I like the one about the glove.

library.sciencemadness.org/library/books/ignition.pdf

[Enthalpy]

Excellent reading indeed. I had appreciated the experiment with a rat tail in a barrel of peroxide, and a few more.

I wish readers would not only remember the fun, but as well the thousands of attempted propellants rejected for being too dangerous. These times, people come again with N₂O, H₂O₂, Propyne and the like, while older chemists had already ruled them out.

A rocket fuel for Mars, freezing below -100°C and preferably with a flash point well over terrestrial temperatures, is not in use nor investigated as far as I know, hence the present thread.

[Enthalpy]

Phytane-like branched alkanes must be excellent as transformer oil:

• Stable at heat over time
• High dielectric strength, low losses
• Benign to wire insulation
• Flash point >155°C for Phytane, but fluid at cold
• Only biodegradability is bad

The first three points need only a strict alkane, while the fourth accepts Phytane (four-pattern chain) be blended with a little Farnesane (three-pattern) and much of longer chains.

They can also be good as lubricants and vacuum seal grease for their wide liquid range and low vapour pressure.

These two uses are an easier market than aeroplane kerosene which is extremely cheap to obtain from petroleum. All products must be free from unhealthy Pristane.

I suppose the diphosphate path that mimicks Nature, as described in the patents from Amyris Biotechnologies Inc, can produce Phytane and longer chains, not just Farnesane whose density matches kerosene. Oligomerization of Isoprene (dimer then tetramer?) is an other way if scalable and may qualify as ecological as well, if Isoprene is obtained from hevea, gutta-percha or waste tyres.

Marc Schaefer, aka Enthalpy

[Enthalpy]

Aza compounds are easier to produce than the hydrocarbon homologue. As rocket fuels, they are a tad more efficient and denser. Tertiary amines have melting and boiling points close to their hydrocarbon homologue. The following amines are not volatile, so their main drawback is the risk of ignition of leaks by contact if N₂O₄ is present in the launcher.

Propyleneamines are industrial products with liquid ranges better than ethylene amines, just like homologue hydrocarbons do; according to Huntsman, the permethylated dipropylene triamine offers excellent mp=-78°C, fp=+92°C, bp=+227°C. Polyethyleneamines are often synthesized from dichloroethane and ammonia, followed by an easy permethylation, making them cheap mass-products. Methylamine may be an alternative and avoids uncontrolled branching.

With just one ethyl instead a methyl, we get the aza versions of farnesane and phytane. The ethyl reduces the molecule's symmetry and increases the number of isomers, two properties that depress the melting point from already -78°C, possibly attaining -100°C as farnesane and phytane do.

A second ethyl at the opposite end would ease the synthesis, especially after a methylamine route. The asymmetry is lost, but the increased number of isomers is kept.

A mix of aza-farnesane and -phytane would be easier to produce and acceptable as a fuel; even better, they may form a eutectic.

Marc Schaefer, aka Enthalpy

[Enthalpy]

As a cooling liquid, phytane-like branched alkanes have advantages:

• Good heat capacity. Water is better, but:
• Electric insulator. Useful to cool computers for instance.
• No chlorine, no fluorine that let presently ban freons.
• Decent cold viscosity expected from the very low freezing point.
• Very high flash point. Tiny vapour pressure.
• Inert, as an alkane. Protects against corrosion.

Marc Schaefer, aka Enthalpy

[Enthalpy]

Compounds similar to farnesane (mp -100°C) and phytane (mp -100°C as well) but even longer should retain a reasonably low freezing point and provide a tiny vapour pressure; if liquid enough, they could make a vacuum oil, and if thicker, a vacuum grease - as an alternative to silicone and fluorosilicone oils and greases used presently.

Marc Schaefer, aka Enthalpy

[Enthalpy]

Pentamethyl-dipropylene-triamine is already seducing as a propellant, among others to return from chilly Mars, but:
- The price of this mass product (Jeffcat ZR-40, Polycat 77) may not be exorbitant enough for space activities;
- I hope ethyls at the ends lower the freezing point further by making more isomers (4 vs 1 ?), just like the heavier phytane melts as easily (-100°C) as the lighter farnesane. Ethyls will also ease the ignition.

I naively imagine one can produce 100t/month as on the joined pictures:

• React excess di-X-propane with methylamine as hot pressured gases to get the dipropylene-thing.
• The bottom of this first reactor is a packed distillation column. Extract there the condensed dipropylene-thing.
• Some tripropylene-thing will form. Keep it, it must be beneficial to the propellant.
• In a second vessel, react with an isomer mix of ethylmethylamine until the reaction stops.
• A few methylamine ends will remain at the dipropylene-thing. Finish with ethanol or chloroethane.

Does this sound any reasonable to you?
Marc Schaefer, aka Enthalpy

[Enthalpy]

The low melting point (like -100°C) and high flash point (+100°C) of properly branched alkanes like Phytane look interesting for aeroplanes, which use to operate in air at -70°C, and appreciate materials hard to light.

These alkanes could find uses as hydraulic fluid, lubricant, cooling liquid... They lubricate better than other compounds do, are noncorrosive and protect better against water.

Marc Schaefer, aka Enthalpy

[Enthalpy]

Is it possible to get a halogen only at primary carbons?

From what I've read:

• Free-radical halogenation puts it at the alkane's most substituted carbon: bad.
• Electrophilic addition of HX puts it at the alkene's most substituted carbon: bad.
• Radical addition would put it at the alkene's least substituted carbon: good start.

BUT with myrcene (appended sketch) I have a branched conjugated diene. I imagine the most stable carbocation is at the tertiary carbon, so where would the halogen go: to the central carbon?

Also: is a Grignard reactant any reasonable in bigger amounts?
I meant zinc vapour as an alternative, but this is speculative.
http://www.chemicalforums.com/index.php?topic=72951.msg287717#msg287717

Possibly a path from myrcene to phytane (and from myrcene and isoprene to farnesane) but I suppose there are flaws in it.

[Enthalpy]

Finally I've found some papers that oligomerize isoprene to farnesene.

One with Va and Zr catalyst produces much dimer, and the trimer's structure isn't elucidated. But that one
http://yamamotogroup.uchicago.edu/yamamoto.pdf (565kB)
gets farnesene from prenyllithium on page 49, sketch appended here.

• Since a fuel needs no stereoselectivity, the reaction should not demand -75°C. Maybe some less subtle reactions exist.
• Hydrogenation to farnesane must produce the mix of stereoisomers.
• Tuning should yield geranylgeranene hopefully, to get phytane. Or even bigger molecules for a vacuum lubricant.
• Could myrcene replace isoprene, to get phytane and save reactants?
• Can some unlithiated proportion of isoprene (or myrcene) define the oligomer length?
• Recycling Li and Br at the plant would reduce the dangerous transports.

Prenylbromide synthesis is known, example with PBr₃ / SiO₂
http://www.scielo.br/pdf/jbchs/v12n5/a13v12n5.pdf (66kB)

[Enthalpy]

I've just stumbled at Wiki on an example of Suzuki-Miyaura coupling
https://en.wikipedia.org/wiki/Suzuki_reaction
starting from citronellal and others to obtain capparatriene (Wiki's picture is appended) which, after hydrogenation, is farnesane, an alkane with wide liquid range. A mix of left or right placed methyl groups is desired.

Do you see advantages to this route as compared with Yamamoto's one here above?

Can it potentially reach the $/kg range in 100t/month? If not, other uses must pay more for smaller amounts.

[Enthalpy]

Some silicon atoms in alkanes widen much the liquid range, at least in neopentane (-17°C to +10°C becomes -99°C to +27°C) and others, possibly by easing the rotations of the attached groups. Si-H are pyrophoric but quaternary silicons apparently not.

Such silanes would be bad rocket or jet fuels, but may excel as aeroplane hydraulic fluid, transformer oil, coolant fluid for electronic equipment, vacuum oil and grease, and all uses needing a high flash or boiling point combined with a low melting point.

Synthesis seems easier (at least to me!) with silicon, as Si-H add to multiple bonds with anti-Markovnikov orientation.
(thanks Wildfyr, that was the missing bit! http://www.chemicalforums.com/index.php?topic=91666.msg327437#msg327437)
Appended is an example of silane inspired by farnesane, keeping three carbons between the branchings and unsymmetric groups that permit isomers.

• Alkynes could replace the alkenes if it helps, followed by hydrogenation.
• Other alkenes are possible, for instance isobutene.
• A continuous process with permanent separation of the products allows a wide imbalance of reactants.
• Anyway, a bit of diethyl rather than ethylpropyl is acceptable in an oil.
• The fused last step on the sketch results in dibutyl and dipentyl mixed with butylpentyl. Have two steps if the mix is less good.
• If propadiene fails, replace by 1,4-pentadiene, or by 3-halopropene to recreate a double bond after the first silane addition.
• Longer molecules, with a third silicon, may still be liquid, with better flash point and vapour pressure.
• If well available, methyl- or ethylmethyl-silane can replace SiH₄ to produce molecules shorter or with more silicons.

Marc Schaefer, aka Enthalpy

[Enthalpy]

Compounds with a wide liquid range could also protect steel against corrosion, especially in mechanical workshops. Many steel tools can't accept a paint or protective layer, which would be inaccurate or too soft, so their functional surfaces remain blank and get covered with oil. Though, these petrol cuts evaporate within weeks or are unpleasantly viscous to the fingers. Compounds with a lower melting point shall remain mobile but not evaporate.

Candidates must be non-toxic, water repellant, excellent at wetting steel, with low odour, preferably good lubricants at low speed and high pressure,  and non-flammable. 500g might retail at 10€, so production should cost under 2€/kg.

Wind music instruments need lubricants too, which are already special productions. Slide trombones, rotary and slide valves need very pure mobile lubricants that don't evaporate, while sealant corks need half-thick greases.

These must be totally innocuous, absolutely water-repellant, benign to wood and varnishes and Cu-Ni-Zn-Ag alloys over decades, resist carbonic and some hydrochloric acid, with low pleasant odour or rather none, good lubricants at low speed and low pressure, excellent at wetting metals. 50g can retail for 20€, so production should cost under 20€/kg.

Marc Schaefer, aka Enthalpy

[wildfyr]

Don't poly- or oligosiloxanes already fulfill the roles of these last two posts?

[Enthalpy]

Hi Wildfyr and the others, thanks for your interest!

For some reason (beyond the price), silicones are not used against corrosion on workshop tools nor as lubricant for music instruments. The drawbacks I see:
- They are horrible lubricants. As a thin film (or rather at high "shear number") their viscosity vanishes, so they build no gliding wedge. The parts go in contact and wear quickly.
- They don't wet metals as well as hydrocarbons do - but that's only my impression.

Silicone oils do have a remarkable liquid range, which I suppose (with big doubts) results from their ease of rotation at the Si-O bonds. While hydrogen atoms hinder the C-C rotations of alkanes, methyls are out of the way for dimethylsiloxanes. The resulting irregular and changing shape must favour the liquid.

This explanation, if correct, would apply to SiCH too, telling why tetramethylsilane melts at -99°C versus -17°C for neopentane. That's why I suggest such SiCH. Since silane looks cheaper to obtain than siloxanes, and addition of alkenes too, maybe such SiCH are affordable for many uses.

From estimations I got from Am1 (beware software), methyl branches help the C-C rotations of alkanes too, by making the easiest conformation less favourable. This can be an element of explanation too, in addition to the unfavourable stacking.