Gasoline is a volatile mix of liquid hydrocarbons, originally distilled from crude oil in a so-called fractionating tower. Crude oil is heated to evaporate it from a pan at the bottom, and the vapor rises to condense into progressively higher pans kept at various temperature levels. Gases such as ethane and propane are taken off at the top. Then come successively less volatile products such as gasoline, kerosene, heating oil, diesel fuel, and lube oils.
Gasoline is not a pure substance but consists of a mixture of more than 100 hydrocarbon species over a specified range of volatility. Gasoline, because it evaporates easily to form combustible mixtures with air, was originally just a headache for refiners eager to sell lamp oil and lubricants. For some years after 1860 it was therefore either run into rivers or “flared off” at the refinery as not only useless but dangerous.
When the developers of the internal-combustion engine needed a fuel that would set their early, slow-turning industrial engines free of city gas mains (originally built for gas lighting), gasoline proved ideal. Being a liquid at room temperature, it could be carried in tanks, yet when mixed with air it evaporated promptly and the resulting gasoline-air mixture could be ignited by a spark over a range of mixture strengths. Chuff, said the IC engine. Chuff-chuff-chuff. Very quickly, cars, trucks, and aircraft became the essential vehicles of the 20th century.
The rapid acceptance of autos after 1910 quickly created demand for more gasoline than could be distilled out of crude oil, so process chemists looked for ways to increase the yield of gasoline from each barrel (42 gallons) of crude. One way was to break down larger, heavier, and less volatile molecules into fragments in the gasoline distillation range. This led to development of catalytic “cracking” (a catalyst is a substance that accelerates the rate of a chemical reaction without itself being changed by it). Another technique was to make desired molecular species by synthesis, joining together light gaseous molecules by polymerization to create liquids in the gasoline boiling range.
The volatility of motor gasoline is intentionally adjusted on a seasonal basis to provide a more volatile fuel in winter to make cold starting of engines easier and a less volatile fuel in summer to reduce evaporation loss. A personal experience of gasoline volatility is the common difficulty of starting the lawn mower for the first time in the spring: Through the months of winter, the most volatile parts of the fuel (called its “front end” because it evaporates easiest) have evaporated into the atmosphere, leaving only a sluggishly evaporating residue in the tank. I remember my mother saying to me as I yanked and yanked on the mower’s starter, “Just push that thing into the sun and leave it for an hour or two. Then it’ll start.” And it did.
In a warmed-up engine, the fuel is heated enough to evaporate nearly all of it on its way from carburetor or fuel injector into the engine’s cylinders. If a mixture strength of gasoline vapor between 10:1 (rich limit) and 18:1 (lean limit) results, a spark can ignite it and the engine will start. But a cold engine can evaporate only 10 to 15 percent of the fuel—its front end, the most volatile species in it. The rest of the fuel delivered will pass through the engine in un-ignitable liquid form. To make the engine start, we enrich the mixture (deliver more fuel in relation to air) until there is enough of the fuel’s front end evaporated to make an ignitable mixture. As the engine starts, we smell a pungent aroma of unburned fuel, the part of the gasoline that was too involatile to evaporate into vapor in a cold engine. It is to deal with this unburned fuel that quick light off or even preheated exhaust catalysts are being adopted.
When I was a child, most cars had a manual choke control to enrich the fuel-air mixture for cold starting. Later, automatic chokes became common, though motorcycles continued for a time to have manual chokes. In the present era of digital fuel injection, sensors for air and engine temperature are used by the ECU computer to determine how much extra fuel must be injected to create an ignitable starting mixture. Without a thought, we turn the key and are rewarded by instant starting.
Racers who use racing gasolines have often discovered that, if they aren’t careful, the last one-third of a drum of fuel “just isn’t right.” The reason is that each time the drum is opened, that “pshhh” is the sound of some of that fuel’s front end escaping. Fuel that has lost much of its front end lacks the volatility to cold-start well, gives degraded snap-throttle response, and may even lose hundreds of rpm on top. This is made worse if the fuel drum is allowed to stand in the sun at the track or in storage. Worst of all can be the “fun-size” 5-gallon pails of race gas, because each time fuel is poured from one container to another, more and more of its front end of volatile components is lost to evaporation.
Why limit the high-volatility front end to just 10 to 15 percent of the fuel? Why not make gasoline 100 percent from high volatility components like isopentane? This could be done, but it would increase evaporation loss, raise the cost of fuel (to pay for cracking heavier components down into higher-volatility light fractions), and make the resulting gasoline less resistant to “knock” (because volatile fuel components have generally low octane number).
Because vapor is constantly produced from the fuel in automobile fuel tanks, clean-air authorities mandated use of vapor-absorbing carbon canisters and have since then sought to generally reduce the vapor pressure of motor fuels, given as Reid Vapor Pressure or RVP. Now onward to another necessary quality of gasoline, its octane number, which is a measure of its resistance to the form of abnormal combustion called knock or detonation. Knock occurs when the last parts of the fuel-air mixture to burn (long after the spark) have been chemically altered by high temperature. The altered mixture then behaves like a sensitive explosive, auto-igniting and burning at sonic speed. This produces a shock wave whose collision with combustion-chamber surfaces is not only audible (knock or ping) but destructive. The resulting extreme-pressure spikes can hammer out bearings. Shock waves scour away stagnant gas from metal surfaces (the insulating “boundary layer”), accelerating heat flow into them. The combination of overheated metal and shock-wave impact erodes the outer edges of the piston crown, eventually destroying the sealing of the top piston ring. Bad, bad stuff and well worth avoiding! Today, few people have experienced detonation because so many vehicles carry systems that listen for knock and, when it is detected, retard ignition timing just enough to stop it.
Henry Ford’s Model T was limited by the low ON of early fuels to a fuel-inefficient, low-torque compression ratio of around 3:1. Raising the compression ratio to boost torque and fuel economy just pushed early engines into detonation. Engine designers and fuel chemists got busy trying to understand combustion.
In 1916, Harry Ricardo, developing an engine for British tanks in World War I, used a fast-acting combustion pressure gauge (called an engine indicator) in his work. He found out very quickly that two different phenomena were damaging engines. The first was pre-ignition, the igniting of the fuel-air mixture before the passage of the ignition spark. Usually occurring around bottom dead center as a result of something very hot in the combustion chamber, this forces an engine to compress burning mixture. In just two or three cycles, this heats the center of the piston so much that it softens and gives way. This is “holing” a piston. The hot object causing pre-ignition can be a too-hot spark-plug electrode, an overheated exhaust valve, or carbon deposits thick enough to glow at red heat.
But detonation, or knock, was completely different. On the engine indicator, pressure rise was normal during the compression stroke, and shortly after the spark, cylinder pressure began to rise smoothly as combustion transformed fuel energy into heat and pressure. Combustion pressure rise continued in normal fashion until, just as the flame front was about to consume the last parts of the mixture, violent pressure spikes occurred (or the indicator was destroyed!). Whatever detonation was, it did not occur until long after the spark had ignited the mixture and most of combustion had taken place normally. Something really bad was happening in the very last bits of charge to burn, the so-called “end gas.”
The chemists figured it out by building rapid compression machines that could simulate the range of conditions in that end gas. They found that prolonged heating of that gas caused partial preflame chemical reactions in it. Hydrogen atoms were knocked off fuel molecules—molecules whanging this way and that at high energies—by the intense thermal activity of high temperature. Some of those hydrogen atoms combined with oxygen from the air in the mixture to form hyperactive OH− radicals. When the population of those radicals reached a certain point, an autocatalytic process began that was explosive. This is detonation.
Normal combustion in engines is not explosive. It is more like a forest fire in which the heat of the flame raises the temperature of brush ahead of it, setting it afire. Chemists call this deflagration. The reaction process in detonation is completely different. Because of the instability of the heat-altered end gas, all it takes to ignite it is the sudden whack of the shock wave, propagating through it at sonic speed.
How fast is normal combustion? Flame speed in a perfectly still correct mixture of gasoline and air is about one foot per second—not fast enough to make IC engines practical. But when that mixture is eddying and swirling in turbulent motion, the flame front is shredded and mixed in, giving it greatly enlarged surface area that consumes the charge in the cylinder as if a simple flame front were moving at 50 to 150 feet per second. This is why engine people are forever talking about “combustion-chamber turbulence.” Turbulence is what makes IC engine combustion fast enough to be a practical power source.
Chemists discovered that the structure of fuels determines their detonation resistance. The three basic Tinkertoy-like structures of petroleum hydrocarbons are: 1) straight chains of carbon atoms; 2) more compact, ball-like branched chains; and 3) ring structures, usually based on the six-carbon benzene ring. In all these structures, each carbon atom has at least one hydrogen atom attached to it. Because straight chains lose their attached hydrogen atoms fairly easily, they have poor detonation resistance. But more compact forms, such as branched chains and rings, better resist loss of their hydrogens. Without free hydrogen atoms, the nasty OH− hydroxyl radicals can’t form.
There is no difference in energy content between higher- and lower-octane fuels. But we can make more power with a high-octane fuel because it can safely tolerate a higher compression ratio without detonating.
You will often read the opinion that lower-octane fuels (pump gas, for example) “burn faster” than race gas. People think this because the harder the fuel blender works to raise octane number, the less volatile the resulting fuel becomes. Low-volatility fuels can cause loss of rpm on top because they take longer to evaporate. Splash in some dreadful “Knock-O” from the gas station and the revs come back. They come back because regular pump gas is more volatile than race gas. At high revs, race gas may not have time to evaporate completely, causing the fuel mixture to lean out and the engine to lose power. Many a racer, sure that the higher the octane number the better, has poured in “118-octane turbo gas” only to find the classic symptoms of incomplete vaporization: To make a fuel that doesn’t detonate in a highly turbocharged engine, the fuel blender has to use heavy, low-volatility components. In a non-turbo engine, a fuel of slightly lower octane number but with higher volatility may very well perform better than the “Double-Throwdown 118.”
What should you do if your big-bore twin knocks on regular and you can’t find super premium? Easy does it: You protect your engine from detonation by not taking big handfuls of throttle until you can tank up on something better. This is the whole basis of the new variable-compression engines now being talked about. They can save fuel by automatically shifting to higher compression when operating on low-to-moderate throttle, yet they prevent detonation on larger throttle angles by quickly reducing compression for that condition.
It’s not that high compression ratios get us something for nothing; they just change the ratio between power delivered to the pistons and power wasted as heat out the exhaust. Ford’s Model T with its 3:1 compression generated less combustion pressure but dumped more heat out the exhaust. A modern 12:1 engine generates much higher combustion pressure but dumps less waste heat out the exhaust. Conservation of energy.
A variety of additives are used in gasoline for other purposes. When certain imported cars began to develop thick deposits on the backs of their intake valves, an additive was developed to wash those deposits away faster than they could form. Sometimes antiwear additive is put into fuel (older drivers will remember “TCP,” which stood for tricresyl phosphate). Antioxidants are used to enhance the stability of cat-cracked fuels.
Today many optimists are sure that all vehicles will shortly be electric anyway, so why bother thinking about all this 20th-century granddad stuff? First, because it’s interesting, and second, until a widespread technology for rapid, safe battery charging has been installed throughout the land, internal-combustion engines and the 168,000 gas stations that serve them will likely continue to carry the freight.
