Frankly I don't think any team is using anything ultra exotic as far as fuel is concerned. The actual gains being made are more as follows.
Large Petrochemical Company(LPC): Here are all the detailed fuel kinetics analysis you wanted.
Large Engine Manufacturer F1 team(LEM): Thanks LPC, and thanks for the free predictive simulations, we could have never done it without you. Because it costs billions to do this type of research.
LPC: No problem, just make sure you plaster our names big and bright on the side of your car, team clothing, and name drop us like you're breathing air.
LEM: Absolutely, we'll even claim the fuel you give us which is the same stuff going around the world because it all comes from about 16 refineries is extra special. No one needs to know that the predictive models based on all the research you've done is what's helping us design our combustion chamber. By the time the others catch on it'll be 3 years too late for them and they'll be fighting an uphill battle trying to catch up. In the mean time everyone can think up crazy conspiracy theories like oil burning, we can have a chuckle and get back to mopping the floor with them.
People are worried about all sorts of exotic mechanical bits that will save the day, thermodynamics has been well studied, and any gains simply come from better engineering. Combustion is the most important aspect of the power units, would it not help you get the most out of the combustion if you understand exactly what it is you're combusting, and how it's being combusted? Fuel isn't just burn hot not hot, there is a lot that goes on, fuel reformulates itself several times during combustion. Fuel itself is a very complex mixture of chemicals, during combustion these chemicals are "broken down" mix with oxygen, form oxygen rich compounds called radicals, and there's a chain reaction too long to describe with words, and this is before we even get to the exothermic side of it. These radicals are essentially what cause combustion, they're here ionizing and robbing and giving electrons all over the place. Eventually in this dance certain molecules get created during the reformulation process which are unstable and create the exothermic reaction, and during this period more stuff that can be described happens again.
In an ideal world you'd just get water and CO2, but the world isn't perfect, and reality is more complex than our puny brains can comprehend. The more you understand this combustion process the better you can design your hardware to accommodate this process, to exploit it for your needs.
If I had to peg my finger on it, this predictive software has been available in "rough draft" versions since 2010, today it's just so much better. The resolution is much better, but you still need massive super computer clusters to do this type of research. Luckily, or unluckily for their competitors LPC's aren't bound by the sporting regulations of CFD on-time.
I guess that's why no new manufacturers want to join?
Who else is there besides Esso/Mobil1:Honda, Castrol/BP:Renault, Pennzoil/Shell:Ferrari, Petronas:Mercedes, besides Total is there any other LPC that can compete? There's not a lot of LPC's that could partner with large manufacturers that want to enter the sport. Maybe Sunoco/BMW
