Kettering assigned Thomas Midgley, Jr. to the task of finding the exact cause of knock [24]. They used a Dobbie-McInnes manograph to demonstrate that the knock did not arise from preignition, as was commonly supposed, but arose from a violent pressure rise _after_ ignition. The manograph was not suitable for further research, so Midgley and Boyd developed a high-speed camera to see what was happening. They also developed a "bouncing pin" indicator that measured the amount of knock [9]. Ricardo had developed an alternative concept of HUCF ( Highest Useful Compression Ratio ) using a variable-compression engine. His numbers were not absolute, as there were many variables, such as ignition timing, cleanliness, spark plug position, engine temperature. etc.
In 1927 Graham Edgar suggested using two hydrocarbons that could be produced in sufficient purity and quantity [11]. These were "normal heptane", that was already obtainable in sufficient purity from the distillation of Jeffrey pine oil, and " an octane, named 2,4,4-trimethyl pentane " that he first synthesized. Today we call it " iso-octane " or 2,2,4-trimethyl pentane. The octane had a high antiknock value, and he suggested using the ratio of the two as a reference fuel number. He demonstrated that all the commercially- available gasolines could be bracketed between 60:40 and 40:60 parts by volume heptane:iso-octane.
The reason for using normal heptane and iso-octane was because they both have similar volatility properties, specifically boiling point, thus the varying ratios 0:100 to 100:0 should not exhibit large differences in volatility that could affect the rating test.
Heat of
Melting Point Boiling Point Density Vaporisation
C C g/ml MJ/kg
normal heptane -90.7 98.4 0.684 0.365 @ 25C
iso octane -107.45 99.3 0.6919 0.308 @ 25C
Having decided on standard reference fuels, a whole range of engines and
test conditions appeared, but today the most common are the Research Octane
Number ( RON ), and the Motor Octane Number ( MON ).
Simply put, the octane rating of the fuel reflects the ability of the unburnt end gases to resist spontaneous autoignition under the engine test conditions used. If autoignition occurs, it results in an extremely rapid pressure rise, as both the desired spark-initiated flame front, and the undesired autoignited end gas flames are expanding. The combined pressure peak arrives slightly ahead of the normal operating pressure peak, leading to a loss of power and eventual overheating. The end gas pressure waves are superimposed on the main pressure wave, leading to a sawtooth pattern of pressure oscillations that create the "knocking" sound.
The combination of intense pressure waves and overheating can induce piston failure in a few minutes. Knock and preignition are both favoured by high temperatures, so one may lead to the other. Under high-speed conditions knock can lead to preignition, which then accelerates engine destruction [27,28].
During the oxidation of a hydrocarbon fuel, the hydrogen atoms are removed one at a time from the molecule by reactions with small radical species (such as OH and HO2), and O and H atoms. The strength of carbon-hydrogen bonds depends on what the carbon is connected to. Straight chain HCs such as normal heptane have secondary C-H bonds that are significantly weaker than the primary C-H bonds present in branched chain HCs like iso-octane [21,22].
The octane rating of hydrocarbons is determined by the structure of the molecule, with long, straight hydrocarbon chains producing large amounts of easily-autoignitable pre-flame decomposition species, while branched and aromatic hydrocarbons are more resistant. This also explains why the octane ratings of paraffins consistently decrease with carbon number. In real life, the unburnt "end gases" ahead of the flame front encounter temperatures up to about 700C due to piston motion and radiant and conductive heating, and commence a series of pre-flame reactions. These reactions occur at different thermal stages, with the initial stage ( below 400C ) commencing with the addition of molecular oxygen to alkyl radicals, followed by the internal transfer of hydrogen atoms within the new radical to form an unsaturated, oxygen-containing species. These new species are susceptible to chain branching involving the HO2 radical during the intermediate temperature stage (400-600C), mainly through the production of OH radicals. Above 600C, the most important reaction that produces chain branching is the reaction of one hydrogen atom radical with molecular oxygen to form O and OH radicals.
The addition of additives such as alkyl lead and oxygenates can significantly affect the pre-flame reaction pathways. Antiknock additives work by interfering at different points in the pre-flame reactions, with the oxygenates retarding undesirable low temperature reactions, and the alkyl lead compounds react in the intermediate temperature region to deactivate the major undesirable chain branching sequence [21,22].
The antiknock ability is related to the "autoignition temperature" of the hydrocarbons. Antiknock ability is _not_ substantially related to:
The initial knock measurement methods developed in the 1920s resulted in a diverse range of engine test methods and conditions, many of which have been summarised by Campbell and Boyd [103]. In 1928 the Co-operative Fuel Research Committee formed a sub-committee to develop a uniform knock-testing apparatus and procedure. They settled on a single-cylinder, valve-in-head, water-cooled, variable compression engine of 3.5"bore and 4.5" stroke. The knock indicator was the bouncing-pin type. They selected operating conditions for evaluation that most closely match the current Research Method, however correlation trials with road octanes in the early 1930s exhibited such large discrepancies that conditions were changed ( higher engine speed, hot mixture temperature, and defined spark advance profiles ), and a new tentative ASTM Octane rating method was produced. This method is similar to the operating conditions of the current Motor Octane procedure [12,103]. Over several decades, a large number of alternative octane test methods appeared. These were variations to either the engine design, or the specified operating conditions [103]. During the 1950-1960s attempts were made to internationally standardise and reduce the number of Octane Rating test procedures.
During the late 1940s - mid 1960s, the Research method became the important rating because it more closely represented the octane requirements of the motorist using the fuels/vehicles/roads then available. In the late 1960s German automakers discovered their engines were destroying themselves on long Autobahn runs, even though the Research Octane was within specification. They discovered that either the MON or the Sensitivity ( the numerical difference between the RON and MON numbers ) also had to be specified. Today it is accepted that no one octane rating covers all use. In fact, during 1994, there have been increasing concerns in Europe about the high Sensitivity of some commercially-available unleaded fuels.
The design of the engine and vehicle significantly affect the fuel octane requirement for both RON and MON. In the 1930s, most vehicles would have been sensitive to the Research Octane of the fuel, almost regardless of the Motor Octane, whereas most 1990s engines have a 'severity" of one, which means the engine is unlikely to knock if a changes of one RON is matched by an equal and opposite change of MON [32]. I should note that the Research method was only formally approved in 1947, but used unofficially from 1942 ),
Test Engine conditions Motor Octane
====================== ==================================
Test Method ASTM D2700-92 [104]
Engine Cooperative Fuels Research ( CFR )
Engine RPM 900 RPM
Intake air temperature 38 C
Intake air humidity 3.56 - 7.12 g H2O / kg dry air
Intake mixture temperature 149 C
Coolant temperature 100 C
Oil Temperature 57 C
Ignition Advance - variable Varies with compression ratio
( eg 14 - 26 degrees BTDC )
Carburettor Venturi 14.3 mm<>
Test Engine conditions Research Octane
====================== ==================================
Test Method ASTM D2699-92 [105]
Engine Cooperative Fuels Research ( CFR )
Engine RPM 600 RPM
Intake air temperature Varies with barometric pressure
( eg 88kPa = 19.4C, 101.6kPa = 52.2C )
Intake air humidity 3.56 - 7.12 g H2O / kg dry air
Intake mixture temperature Not specified
Coolant temperature 100 C
Oil Temperature 57 C
Ignition Advance - fixed 13 degrees BTDC
Carburettor Venturi Set according to engine altitude
( eg 0-500m=14.3mm, 500-1000m=15.1mm ) <>
Some fuel specifications include delta RONs, to ensure octane distribution throughout the fuel boiling range was consistent. Octane distribution was seldom a problem with the alkyl lead compounds, as the tetra methyl lead and tetra ethyl lead octane volatility profiles were well characterised, but it can be a major problem for the new, reformulated, low aromatic gasolines, as MTBE boils at 55C, whereas ethanol boils at 78C. Drivers have discovered that an 87 (RON+MON)/2 from one brand has to be substituted with an 89 (RON+MON)/2 of another, and that is because of the combination of their driving style, engine design, vehicle mass, fuel octane distribution, fuel volatility, and the octane-enhancers used.
The fuel is carefully distilled to obtain a distillate fraction that boils to the specified temperature, which is usually 100C. Both the parent fuel and the distillate fraction are rated on the octane engine using the Research Octane method [107]. The difference between these is the delta RON(100C), usually just called the delta RON. The delta RON ratings are not particularly relevant to engines with injectors, and are not used in the US.
If you are already using the proper octane fuel, you will not obtain more power from higher octane fuels. The engine will be already operating at optimum settings, and a higher octane should have no effect on the management system. Your driveability and fuel economy will remain the same. The higher octane fuel costs more, so you are just throwing money away. If you are already using a fuel with an octane rating slightly below the optimum, then using a higher octane fuel will cause the engine management system to move to the optimum settings, possibly resulting in both increased power and improved fuel economy. You may be able to change octanes between seasons ( reduce octane in winter ) to obtain the most cost-effective fuel without loss of driveability.
Once you have identified the fuel that keeps the engine at optimum settings, there is no advantage in moving to an even higher octane fuel. The manufacturer's recommendation is conservative, so you may be able to carefully reduce the fuel octane. The penalty for getting it badly wrong, and not realising that you have, could be expensive engine damage.
Attempts to mix leaded high octane to unleaded high octane to obtain higher octane are useless for most commercial gasolines. The lead response of the unleaded fuel does not overcome the dilution effect, thus 50:50 of 96 leaded and 91 unleaded will give 94. Some blends of oxygenated fuels with ordinary gasoline can result in undesirable increases in volatility due to volatile azeotropes, and some oxygenates can have negative lead responses. The octane requirement of some engines is determined by the need to avoid run-on, not to avoid knock.