boiling point than the nonvolatile portion of the distillate fuels. It 

 may also be due partly to the fact that the percent losses from the crude 

 oils were generally much smaller than the percent losses from the distillate 

 fuels. 



In a manner analogous to the derivation of Equation 7, it is possible 

 to derive the following equation for the portion retained, R, of any 

 particular hydrocarbon during the weathering: 



(9) 



Here C is the amount of a particular hydrocarbon present originally and 

 after weathering, and H is the peak height of that hydrocarbon and of 

 another hydrocarbon that is assumed to be nonvolatile. Octadecane (C-|g) 

 is the hydrocarbon that is assumed to have negligible volatility and the 

 ratio in parentheses is a measure of the ratio, b/a, between the overall 

 sensitivities of the two gas chromatograms . Actually, octadecane does 

 evaporate to some extent , and the calculated R is a comparative portion 

 retained as compared to the portion of octadecane retained. 



The comparative retentions of the normal alkanes from dodecane (C-io) 

 to heptadecane (C-iy) in the simulated weathering of Oils C and D are 

 shown in Figure 7. The effect of omitting the wind in the weathering of 

 Oil D and the effect of much longer weathering of Oil C are also shown. 

 The comparative retentions of the hydrocarbons decrease markedly as their 

 size and boiling points decrease. Furthermore, these differences in 

 retention are markedly increased for any particular oil as the degree of 

 weathering is Increased. 



The curves of the comparative retentions also show that appreciable 

 amounts of C15 and C-| y hydrocarbons are lost, as compared to C-13. With 

 increased weathering, it would be expected that increasing amounts of C-jg 

 and higher hydrocarbons would be lost and that Increasingly higher 

 hydrocarbons would have to be used as reasonably valid internal standards 

 for gas chromatographic comparisons. For mild weathering, the approximation 

 that certain hydrocarbons are nonvolatile may give adequate mathematical 

 comparisons; but for severe weathering, the same approximation may not be 

 suitable. Because the distillate fuels in some cases have very small 

 amounts of components higher than C-io, they do not contain good internal 

 standards. 



Smith and Maclntyre used an equation very similar to Equation 9 to 

 calculate the percentages of various normal hydrocarbons that remained 

 in oil spills after various periods of time, using the C20 peak rather 

 than the C-|g peak as a reference [2]. Qualitative trends were established 

 in the evaporation of the various hydrocarbons, but it was difficult to 

 ascertain accurate peak values. Calculations of the loss of volatiles 

 (up to C-15) were also made, but there was no possibility of measuring 

 actual losses for comparison. 



16 



