Gasoline Blending Plus

Blends of Pure Hydrocarbons

    In some cases octanes of blends of pure hydrocarbons have huge deviations from linear blending.  A 1958 ASTM publication (Reference 9) based on API Research Project 45 provides octanes for a large number of pure hydrocarbons.  In addition, it includes octanes of blends of 20% hydrocarbons with 80% 60 octane primary reference full (60 vol% 2,2,4-trimelhylpentane  "isooctane" + 40 vol% normal heptane).  Olefins have very high blending values.  For example, 2-methyl-2-butene with a Research octane of 97.3 has a blending value of 176 and an interaction coefficient of 98.  Gasoline Components are mixtures of hydrocarbons and therefore have much smaller interaction coefficients.

    The above data from API Research Project 45 were later supplemented by data on a large variety of mixtures.  In many cases mixtures of two components to achieve the same octane level as a third component were blended with the third component.  In addition to those API Project 45 data, there exists a considerable amount of unpublished octane data on pure hydrocarbon blends.  A description of those data is provided in the Appendix.

    Over 400 unleaded octanes were obtained, mostly on 50:50 blends of pure hydrocarbons.  Octanes were obtained in duplicate. Interactions between the hydrocarbons indicate that the effect on engine knock of mixing individual hydrocarbons can be greatly different from the effect of mixing the reference fuels, normal heptane and 2,2,4-trimethyl pentane.

    The interaction coefficients between hydrocarbons range from minus 39 to plus 111.  Viewing the data from the standpoint of hydrocarbon type (paraffins, olefins, naphthenes, aromatics) there are large differences between interactions of individual hydrocarbons of a given type with a particular hydrocarbon of another type.  The differences are partially  related to octane level.  There are significant interactions between hydrocarbons of the same type.

    Based on these extensive data, hydrocarbon type is not sufficient to predict interactions between gasoline components.

    At the time these unpublished date were obtained, it was hoped that a fundamental predictor of interaction between gasoline components could be developed based on "net interaction".  The idea was to use GC analysis to predict the total interaction within each gasoline component, predict the total interaction within the 50:50 blend, then calculate the interaction coefficient between the two components.  However, after learning how well the calculations based on component types and octane levels worked, the "net interaction" study was discontinued.

    The success of the component type approach (alkylate etc.) can be explained by the fact that hydrocarbon compositions of components of a given type are quite similar.  Much of the difference in composition is related to octane level including Sensitivity (Research minus Motor).

    The unpublished octane interaction date for mixtures of individual hydrocarbons may be of use to those interested in predicting octanes from other measurements.  If you have an interest, you may call 302-475-1277.

Close Window