Background

There's always a lot of confusion regarding octane, octane-boosters and how they work. Typical misconceptions are evident in blank-statements like:

"Higher octane fuels burn slower, thus their higher octane number"
"Higher octane fuels burn hotter, therefore more power is generated"
Higher octane fuels explodes with more force, thus their higher power"

Both of which are untrue and are coincidental in effect, rather than causal. In actual practice, an engine has to be tuned specifically for high-octane fuels to generate extra power. If you have an engine fully-tuned and optimized for 91-octane pump gas, adding 100-octane race-gas into it will yield little if any increase. However, if you were to take that engine and increase the compression, advance the knock and/or increase the boost, then you can take advantage of the higher-octane fuel. But this precludes going back to the previous lower-octane fuels.

Three Kinds of Octane Boosters - 1. ORGANO-METALLICS

There are three primary octane-boosting additives mixed in with gasoline: organo-metallics, ethers/alcohols and aromatics. Each one has distinct chemical properties and results (along with side-effects) on octane-boosting. Some people get these families of compounds and their effects mixed up.

First, let's look at organo-metallics which is used in the little bottles of over-the-counter octane boosters, what makes them work and how they compare. By far and large, these work on the same principle as TEL-Tetra Ethyl Lead which is the principle octane-boosting component of AvGas. For automotive OTC use, a slightly less carcinogenic MMT compound is used (methylcyclopentadienyl manganese tricarbonyl); it has pretty much the same structure as TEL, but with manganese substituted for lead. These compounds have a non-linear octane-boosting curve. The initial amounts give the most boost and adding more gives decreasing benefits. Typically you get 3-4 'points' increase with these types of additives; going from 91-octane to 91.4 octane max. I've uploaded a comparison article of these types of additives: 951 RacerX: Octane Booster Comparison

As you can imagine from the metallic content, these boosters create nasty deposits in your engine. That's why they typically include a solvent such as mineral spirits to try and dissolve the deposits. Then a lubricant such as ATF or Marvel Mystery Oil is typically added to the cocktail to help your rings slide over the deposits easier and minimize the damage. If you dyno-test a car using organo-metallics (with straight-through exhaust), you can actually collect metallic pellets coming out your tailpipe. Not a good thing to be putting into your combustion chambers no less...

2. OXYGENATES

The next group of octane-boosting compounds are oxygenates: ethers & alcohols which also serves an emissions purpose by bundling their own oxygen along with the fuel. The best compound here is ethanol (CH3CH2OH) with a 115-octane (R+M)/2 rating and containing 34.73% oxygen by weight. However, its high volatility with a RVP of 18 makes it unsuitable for use in warm climates for emissions control. In which case, MTBE (CH3OC(CH3)3), ETBE (CH3CH2OC(CH3)3)__ and TAME ((CH)3CCH2OCH3) are used which has more favorable RVPs of 1.5-8.0. But they also have correspondingly lower octane of 105-110 along with lower oxygen content of 15.66-18.15 by weight.

Ethers & alcohols are basically hydrocarbons fuels with an extra hydroxyl -OH group added to one end. These fuel-additives reduce your gas-mileage due to the displacement of hydrogen and carbon atoms by the larger oxygen molecules. The increased molecular-mass of the compounds with the attahced -OH is what gives the octane-boosting effect. The -OH group also makes the compound polar, water-soluble and highly reactive chemically. They will dissolve rubber and plastic fuel lines and thus their concentration in fuels is fairly limited. Thus their octane-boosting power is also reduced. Ethers and alcohols are starting to be banned in a lot of areas because their water-solubility makes tank leaks and dispersion by ground-water a big problem

3. AROMATICS

The final group of octane-boosting compounds, aromatics show the most promise. Due to their stable benzene-ring structures, the compounds are non-polar and chemically stable (non-reactive). In fact, they are less volatile and less reactive than most other hydrocarbons in gasoline. This stability is what gives aromatics their octane-boosting powers. Normal gasoline typically contain around 25-30% aromatics, primarily toulene and xylene. Adding more will simply increase the octane rating and bring their concentrations up to what you find in higher-octane European gas (40-45% aromatics): Gasoline composition.

So by using aromatic toluene and xylene as octane-boosters, you get none of the bad side-effects of using organo-metallics (cancer and engine-deposits) or ethers & alcohols (low gas-mileage and rotting fuel-lines). By using just two gallons of xylene in a 15-gal tank of 91-octane pump gas, you've brought the octane-rating up to 94.5 and have roughly the same aromatic content as German or French gasoline. You may also notice in the Octane Booster Comparison article above, that the best octane-boosting solution was to use unleaded race-gas; the primary octane-boosting components used are toluene and xylene.

"Doesn't higher octane fuel have higher energy content and makes more power?"

Well, it's not so simple. Really depends upon what you mean by 'higher' and 'energy content'. 'Octane' does not directly relate to 'energy content' or 'power' in anyway. There are many, many components and properties of gasoline that is custom-tailored by the refinery, such as specific-gravity, octane, oxygenates (ethers & alcohols), RVP-reid vapor pressure (volatility), D86-distillation curve, combustion-temperature, sensitivity, flame-front speed, VL-vapor/liquid ratio, etc. Just about each and every one of these properties can be tailored and are sometimes dependent, and sometimes independent upon each other.

One of the basic measures of energy-content is BTU/gallon or Calories/gal. The amount of heat released by any given volume of fuel is directly related to the number of Hydrogen and Carbon atoms in that gallon. Oxygenated fuels that use MTBE or alcohols to have extra Oxygen onboard deliver much less energy per gallon because the oxygen atoms are simply HUGE compared to a hydrogen or carbon. Such fuels tend to deliver less mileage per gallon than non-oxygenated fuels. BUT, they do not deliver less power, because that's more of a function of air-mass ingested into the engine per 4-stroke cycle than fuel (air is tough to cram in, fuel is simple to inject).

Compared to gasoline's specfic-gravity of 0.751-g/cc, toluene is 0.881-g/cc and xylene (most likely a mixture of m-xylene; o-xylene; p-xylene) is around 0.871-g/cc. This means they have more hydrogens and carbons to combust per gallon with the O2 in the air that's being pumped through the engine. The results of using large-percentage mixtures of these aromatics in your fuel is a richer mixture than before with just pure pump gas (without re-jetting). This will be safer than using the other common additive, 100LL AvGas which is lighter than gasoline and will result in lean mixtures and melted catalytics and O2-sensors. (LowLead AvGas is still contains several times more organometallics than leaded auto gasoline). I've known of several people that have destroyed some very expensive engines because they ran a large amount AvGas without re-tuning their air-fuel ratios. Besides, 100LL AvGas is only about 98-MON anyway, so it's not as effective as toluene or xylene.

Octane Doesn't Predict...

Another factor that octane doesn't predict is combustion temperature which may or may not relate to the power produced. It's possible to blend two mixtures of branched-chain paraffins along with aromatics to create two concoctions both of which have higher octane than pump gas, and one of them will have higher combustion temps than pump-gas, and yet the other will have lower combustion temps.

A lot of people also confuse octane with flame-front propagation speed which is yet another independent factor. Take the old-days measurement of octane-ratings with iso-octane (2,2,4-trimethylpentane) with a octane-100 rating and n-heptane with a 0-octane rating. They both have the exact same flame-front speed, yet one of them has a fairly high anti-knock index. The other, n-heptane, has such low knock-resistance that you can just tap the beaker and the stuff would explode!

Octane Does Predict...

In the end, all that octane predicts is AKI-Anti Knock Index as measured on a knock engine. These are variable-compression single-cylinder engines that can vary their compression between about 7.0:1 to 15.0:1. There's a highly-sensitive and accurate knock-sensor and computer hooked up to this engine that gives a readout of knock. The engine is run with the mystery fuel and starts at a low-compression. Then the compression is increased gradually while knock is monitored. Various levels of compression-ratios are used and the corresponding knock measured. This is looked-up on standardized tables and the MON-octane rating of the fuel is then determined. In the end, that's ALL that the octane predicts, is how much resistance the fuel has to knock.

So what's the point of all this? Just use the xylene to increase your fuel's octane-level!!! Two to three gallons in a 15-gal tank won't change the specific gravity by so much that it'll mess up your AF-ratios. By itself, the resultant 96-octane mix won't automatically give you any more or less power. But will allow you to TUNE your car for higher power by increasing ignition advance, increasing compression or turning up the boost!

I've accumulated a couple of good posts on octane-blending on my RacerX website under the Fuel-FAQ section. There's the obligatory 4-part Gasoline-FAQ, and the F1-Rocket Fuel and DIY Octane Boosters FAQ.

Also the toxicity of xylene and toluene is actually lower than gasoline (due to their stable ring-structures):
Toxicity Profile: Toluene
Toxicity Profile: Xylene

REFERENCES:

The Chemistry of Hydrocarbon Fuels - Harold H. Schobert_ -_ Butterworth-Heinemann Ltd.
Automotive Fuels Reference Book - Keith Owen, Trevor Coley - SAE#R151
Mixture Formation in Spark-Ignition Engines - H.P. Lenz - Springer-Verlag
Fuel Injection - Jeff Hartman - Motorbooks International
Lean Combustion in Spark-Ignited Internal Combustion Engines - Germane, Wood, Hess - SAE#831694
An Introduction to Thermal Fluid Engineering - Z. Warhaft - Cambridge University Press