About This Item

Share This Item

The AAPG/Datapages Combined Publications Database

AAPG Bulletin


Volume: 13 (1929)

Issue: 1. (January)

First Page: 75

Last Page: 84

Title: Is Geologic Distillation of Petroleum Possible?

Author(s): William L. Russell (2)


The physical, chemical, and geologic evidence bearing on the geologic distillation of petroleum is considered critically. The hydrocarbons of petroleum heavier than gasoline could not have been distilled, for their vapor pressures are much less than the rock pressures at the corresponding temperatures, and the vast volumes of natural gas necessary for their distillation under partial pressures were generally not present. The chemical evidence is also against the extensive geologic distillation of oil, because of the comparative scarcity of unsaturated hydrocarbons. The geologic evidence points toward the same conclusion, because of the distribution of the gravity variations, the absence of the requisite volumes of gas, and the general lack of diluted brines.


It has long been customary for geologists to speak of the origin of oil by distillation. Yet these same geologists, when discussing the origin and accumulation of petroleum, generally assume that it has been at all times in the liquid state. It is important that this inconsistency should be cleared up, for the mode of migration of a gas may be quite different from that of a liquid, and the problems of oil accumulation can not be satisfactorily solved until it is known whether the oil during its migration was a liquid or a gas.

The writer is indebted to Isaac N. Beal for valuable suggestions and data, and for a criticism of the manuscript.


Though many geologists refer to the "distillation of oil" in their publications, there is, as far as the writer has been able to ascertain, no paper showing that distillation is possible under the physical conditions to which oil is subjected under burial. Many geologists have expressed doubt that geologic distillation has occurred, and Washburne (FOOTNOTE 3) long ago stated that it was impossible. The writer (FOOTNOTE 4) also made this statement in a previous paper. It seems that many geologists have used the word

FOOTNOTE 3. C. Washburne, Bull. Amer. Assoc. Petrol. Geol., Vol. 3 (1919), pp. 345-62.

FOOTNOTE 4. W. L. Russell, "The Proofs of the Carbon-Ratio Theory," Bull. Amer. Assoc. Petrol. Geol., Vol. 11 (1927), p. 979.

End_Page 75------------------------------

"distillation" in the sense of dissociation or generation, without intending to imply, necessarily, generation in a gaseous state. In a recent personal communication,(FOOTNOTE 1) Rich states that the word distillation was used in this sense in his paper entitled "Generation of Oil by Geologic Distillation during Mountain Building,(FOOTNOTE 2) and that the main conclusions of his paper do not conflict with the views here expressed by the writer. Owing to this double usage of the word, it is in many cases impossible to tell what a writer means by a given statement. In order to prevent such confusion in the future, it is suggested that the word "distillation" be employed only to express production in a gaseous state, this being the sense in which it is used in the present paper.


Obviously the best method of settling the problem under consideration is to ascertain as far as possible the physical properties of petroleum under the temperatures and pressures which exist at different depths beneath the earth's surface. The most important properties in this connection are of course the vapor pressures and the critical points.

Methods of constructing Figure 1:
Figure 1 is introduced in order to show the data graphically, and particularly to emphasize the relative order of magnitude of the vapor pressures of oil and the rock pressures It will be noticed that an increase of 1° F. for every 30 feet of depth is assumed. The actual temperature gradients observed in oil fields range from about 1° in 30 feet or slightly less to about 1° in 100 feet. The first figure is taken, not because it is the average, but because it is most favorable for distillation. It is also assumed that the rock pressure, or hydrostatic pressure on the fluid contents of the porous strata, increases at a rate of 0.40 pounds per foot of depth, this being fairly close to the average figure. The data for the vapor pressures of gasoline and kerosene, and the figur s for the critical points are taken from material assembled by Cross.(FOOTNOTE 3)

Distillation in the absence of fixed gases:
If no gases are present and the temperature is less than the critical temperature, a liquid will begin to distill only when the vapor pressure is equal to, or greater than, the hydrostatic pressure. In dealing with petroleum, the situation is complicated by the fact that the liquid consists of many hydrocarbons which differ widely in their boiling points and vapor pressures. Fortunately,

FOOTNOTE 1. J. L. Rich, personal communication, November 3, 1928.

FOOTNOTE 2. J. L. Rich, "Generation of Oil by Geologic Distillation during Mountain Building," Bull. Amer. Assoc. Petrol. Geol., Vol. 11 (1927), pp. 1139-50.

FOOTNOTE 3. R. Cross, Kansas City Testing Laboratory Bull. No. 17, p. 292.

End_Page 76------------------------------

however, the problem may be simplified for the present purposes by treating the different fractions of oil as units. As shown in Figure 1, the vapor pressure of gasoline is much higher than that of kerosene, just as the vapor pressure of kerosene is much higher than that of the heavy fraction of oil. Hence, if it can be shown that gasoline and kerosene could not distill under the temperatures and pressures to which they are subjected after burial, it would obviously be impossible for the heavier fractions with still lower boiling points to distill. It is evident from Figure 1 that the combined vapor pressures of gasoline and kerosene do not even remotely approach the rock pressure for the corresponding temperatures and depths. Hence, the distillation of petroleum at temperatures below the critical temperatures of its constituents would be

Fig. 1. Diagram showing relation between rock pressure, depth, temperature, and the vapor pressures of water, gasoline, and kerosene. Letters indicate critical points of water and gasoline hydrocarbons. See Table I for legend.

End_Page 77------------------------------

impossible in the absence of fixed gases. It should also be noticed that, while the vapor pressure of water is lower than that of gasoline at low temperatures, it is higher at high temperatures, and is greater than the vapor of kerosene and heavy oil at all temperatures. Hence, it is obvious that water would be distilled from the rocks before even the lighter fractions of the crude oil.

Limiting temperatures:
Before proceeding further in the discussion, it is necessary to consider what limits may reasonably be placed on the temperatures to which the petroliferous rocks have been subjected. Perhaps the coals and bituminous shales associated with the oil-bearing strata afford the best method for placing a limit on the maximum temperatures. When heated, these coals and bituminous shales begin to dissociate or change their nature at temperatures ranging from 600° to 800° F. Since these changes have not taken place in the coals and bituminous shales associated with the oil-bearing series, it is evident that the maximum temperature was never greater than 800° F.

Distillation under partial pressures:
If gases were associated with the oil, small quantities of vapors of the liquid hydrocarbons could exist in them. Under such conditions the volumes of the vapors would be proportional to their partial pressures. Hence, their relative proportions in the gas may be estimated from the vapor pressures shown in Figure 1. In order to illustrate the role which distillation under partial pressures could play in the geologic distillation of petroleum, two possible conditions will be considered, one at 400° F., the other at 600° F. At 400° F. the vapor pressure of the heavy fraction of the crude oil would be considerably less than 15 pounds per square inch, for at atmospheric pressures the part of the crude oil denser than 35° Be. generally does not begin to distill until a tem erature of about 600° F. has been reached. It will be assumed, therefore, that the absolute vapor pressure of these heavy constituents will be 5 pounds per square inch. The rock pressure corresponding to 400° F. in Figure 1 is 4,550 pounds per square inch. Under these assumptions at 400° F. the vapor pressures and relative proportions in the gas of the substances under consideration would be about as follows:


End_Page 78------------------------------

At temperatures of 600° F. the vapor pressures and proportions in the gas would be about as follows, under the same assumptions:


The rock pressure corresponding to 600° F. in Figure 1 is 7,230 pounds per square inch. The proportion of gasoline vapor is not figured, as 600° F. is above the critical temperature of most of its constituents. Under high pressures, vapors of gasoline and probably also of kerosene have the property of being absorbed in the heavy fractions of petroleum in large quantities. Hence, the proportions of gasoline and probably of kerosene would actually be much less than the figures just given. It should also be borne in mind that the temperature gradient shown in Figure 1 is exceptionally high. Another important point is that the vapors of the petroleum fractions would occupy a much smaller volume when condensed to liquids, whereas the fixed gases would expand as the rock pressure as decreased by the removal of the overburden.

It is clear, therefore, that the volume of gas required for the distillation of a given volume of oil under partial pressures would be at least 1,000 times the volume of the liquid kerosene, and at least 10,000 times the volume of the liquid heavy oil. This means that if an oil pool in a given sand covered a square mile, gas necessary for its distillation would cover at least several thousand square miles, assuming that the porosity and sand thickness remained the same.

The effect of the critical points:
Since the critical temperature is the highest temperature at which a pure substance can exist in the liquid state, and since the critical pressure is the vapor pressure of a liquid at its critical temperature, the critical points are obviously important in the present connection. The critical points of water and some of the lighter petroleum hydrocarbons are shown in Table I.(FOOTNOTE 1) It is evident that the critical temperatures increase and the critical pressures decrease with the increasing density of the hydrocarbons. If a pure liquid isolated from gases and at pressures greater than its critical pressure were heated, it would nominally become a gas at the critical temperature, but there would be no abrupt change of volume or pressure. If, on the other hand,

FOOTNOTE 1. R. Cross, op. cit., pp. 221, 292.

End_Page 79------------------------------

it was heated in contact with a gas at pressures greater than the critical pressure, the percentage of vapor of the liquid in the gas would steadily increase until the critical temperature was reached, when it would mix in all proportions with the gas. Hence, if the gasoline hydrocarbons alone were in contact with gas, they would mix in all proportions with the gas when the temperature was greater than their critical temperatures. Actually, however, appreciable quantities of heavier hydrocarbons would generally be present. According to Cross,(FOOTNOTE 1) the effect of these heavier hydrocarbons is to increase the critical temperatures of the lighter hydrocarbons associated with them, because the light constituents remain dissolved in the heavier liquid hydrocarbons. Hence, even when t e gasoline hydrocarbons were heated above their critical temperatures they would generally not mix in all proportions with the associated gases, but would for the most part remain dissolved in the heavier liquid petroleum. However, above the critical temperatures the percentage of gasoline vapors in the associated natural gas would doubtless rise rapidly with the temperature.

The possibility of polymerization:
Another hypothesis which must be considered is the possibility that oil is formed by the polymerization of gases or of gases and the gasoline hydrocarbons. As no chemical and physical data are at present available on this subject, it can not at present be discussed further. However, it should be noticed that this


FOOTNOTE 1. R. Cross, op. cit., p. 292.

End_Page 80------------------------------

could not be called distillation of oil, for the gases would not have been oil vapors, but fixed gases of different composition.


If oil were distilled and condensed again without altering its composition, it would of course be impossible to detect such distillation by chemical means, since there would be no change in its chemical characteristics. However, since it is impossible to distill the heavier constituents of many crude oils without dissociation even at atmospheric pressures, a considerable fraction of the oils would evidently be dissociated if they were distilled under the great pressures existing in the buried strata. According to Cunningham-Craig,(FOOTNOTE 1) equal parts of saturated and unsaturated hydrocarbons are formed by the cracking of saturated hydrocarbons. The absence or relative scarcity of such unsaturated hydrocarbons in many crude oils is therefore evidence that they have not been distill d, provided, of course, that the unsaturated molecules do not become saturated again by polymerization after distillation. Where igneous intrusions cut a bituminous or petroliferous formation, the containing rocks are raised to such high temperatures that any heavy oil or bituminous material remaining in it would be cracked. According to Cunningham-Craig,(FOOTNOTE 2) analyses of many oils found adjacent to igneous intrusions show large percentages of unsaturated hydrocarbons. Although this seems to indicate that the oils or parent substances have been cracked, it does not necessarily mean that they have been distilled. Whether the products of dissociation were gases or liquids at the time of their dissociation would depend on the relation of the vapor pressures to the temperature. The te peratures may well have been above the critical temperatures of the gasoline and kerosene fractions, but since the heavy constituents of oil will crack before their vapor pressures become high, it is probable that these were generated as liquids.


It has already been shown that in discussing the question of geologic distillation of petroleum two possibilities must be considered, namely, the distillation of the hydrocarbons in their present proportions, and distillation under partial pressures in the presence of fixed gases. As previously stated, the physical evidence indicates that oil could not have been distilled when associated with gas in about the same proportions

FOOTNOTE 1. E. H. Cunningham-Craig, Oil Finding.

FOOTNOTE 2. Op. cit.

End_Page 81------------------------------

as it now occurs, but it remains to be seen whether the geologic evidence supports this conclusion.

Figure 1 indicates that if oil with the ordinary amount of gas were subjected to heat and pressure in the presence of water, the water would become a gas before the oil, as the temperature rose, and the oil would be condensed to a liquid before the water as the temperature fell. This is because the vapor pressure of water is greater than that of petroleum at the corresponding temperatures, and because the critical temperature of water is lower than the critical temperatures of the paraffine series of hydrocarbons heavier than gasoline. Since this is true, it seems probable that if any extensive migration of distilled oil had taken place in the gaseous state it would produce recognizable changes in the distribution of the oil, gas, and water. In the gaseous state, these fluids would mi in all proportions, but as they migrated to areas of lower temperature and pressure, the heavy hydrocarbons with the lowest vapor pressure would be the first to be condensed to a liquid. Still farther away from the source the medium kerosene fractions would condense, and the gasoline fraction would be liquefied at the greatest distance from the source. The water would not condense until after the kerosene and heavy oil. It would seem, therefore, that if any extensive migration of the distilled oils in the gaseous state has taken place, the heaviest oils should be found in the areas which lay at the greatest depths or which were exposed to the greatest temperatures and pressures. The actual distribution is, however, the exact opposite. It is well known that there is a general tendency fo the specific gravity of oil to decrease with depth, and the decrease of specific gravity toward areas of greater regional alteration, where the temperatures and pressures would be presumably greater, is one of the corollaries of the carbon-ratio theory.

Even if the oils were distilled under partial pressures and existed as vapors in vastly greater volumes of gas, the same zonal distribution of oil with respect to its density should be found, for the heavier fractions would invariably be the first to condense, because of their lower vapor pressure. The absence of such a distribution is therefore a proof that the extensive migration of distilled oil in the form of vapors in vast volumes of gas did not take place.

Moreover, as has been previously shown, such distillation of oil in the form of a vapor could only take place in the presence of volumes of gas several thousand times as large as the volume of the liquid oil, even if it be assumed that no oil escaped along with the gas. Not only are such vast volumes of gas not found at present associated with the oil, but

End_Page 82------------------------------

in many areas there are reasons for thinking that such great quantities of gas were never associated with it. In some places the oil sands are overlain by porous water sands interbedded with impervious shales. If a volume of gas several thousand times as great as that of the oil has escaped from the oil sands to the surface, it would seem that considerable accumulations of gas would be entrapped in the water sands where they are on anticlines and are covered by impervious shales. The absence of such extensive accumulations of gas in many places therefore suggests that such enormous volumes of gas have not made their escape.

The occurrence of brines with much larger proportions of chlorides than sea water contains might be interpreted as an argument in favor of the escape of great volumes of gas, since it has been argued that these brines become concentrated by the evaporation of water into the escaping gas. The fact that these brines are found in close association with the oil pools, however, is an argument against the idea that the oil was carried in a vapor and condensed, for under such conditions the concentrated brines should occur where the oil and water were distilled, and where they condensed there should be a diluted brine.

Another important point is that there is no real need for any theory involving the distillation of oil, for the migration and accumulation of petroleum may be explained by other processes.


1. Geologic distillation of petroleum in the absence of relatively enormous volumes of fixed gases is impossible except near igneous intrusions.

2. Although small quantities of the lighter constituents of crude oil may have been distilled as vapors in large volumes of natural gas, it is highly improbable that important quantities of crude oil have been distilled in this manner.

3. There is no necessity for assuming the occurrence of geologic distillation, for other processes are capable of explaining the migration and accumulation of oil.


In presenting quantitative data to show that the vaporization of petroleum is impossible under ordinary geological conditions, Russell has rendered a timely service. He has focused attention on the confusion

FOOTNOTE 1. Consulting geologist, Ottawa, Kansas.

End_Page 83------------------------------

which has resulted from the prevailing use of the word "distillation" loosely and with a double meaning, and has urged that henceforth the word be used only in the sense of production in a gaseous state. Such a restriction in the use of the term is certainly desirable and necessary if confusion is to be avoided.

Authors who have previously used the word "distillation" in connection with the generation of oil by geologic processes such as regional metamorphism and devolatilization, have rarely used the word in the narrow and strictly correct sense which Russell proposes. They have used it most commonly in the sense of any process whereby oil and gas are generated from their mother substances in the rocks under the influence of heat and pressure. Whether the hydrocarbons were formed as oils or as oil vapors is a distinction which most of the authors with whose works the writer is familiar have failed to make.

In the writer's paper on "Generation of Oil by Geologic Distillation During Mountain Building"(FOOTNOTE 1) the word distillation was used in the broader sense. Although care was taken to emphasize the generation of oil by cracking directly from its mother substances, oil vapors were also referred to because, at the time, it was the writer's impression that part, at least, of the petroleum would be generated in the form of vapors. The arguments presented in that paper, however, in no wise depend on the generation of petroleum in the gaseous form.

In the light of the data presented by Russell, it may be that vaporization has no place in the process of the generation of oil. "Geologic distillation" of oil, in the sense of geologic vaporization of oil, under ordinary conditions may be an impossibility, as the title and negative conclusion of Russell's paper indicate, but the large-scale generation of oil under the influence of the dynamic forces and temperatures associated with mountain building and regional metamorphism is not thereby ruled out.

In spite of the data presented by Russell, the writer believes that it would be unsafe to conclude that true distillation or vaporization of petroleum on a considerable scale is impossible until more is known about the temperatures which accompanied regional metamorphism, such as that which devolatilized the rocks of the eastern Appalachian province. The temperatures given in Russell's Figure 1 are those due to depth alone. Even making allowance for the fact that he has used a temperature gradient nearly twice as steep as the probable average, it may be that in regions undergoing metamorphism, temperatures, for the various depths, far exceed those shown. The writer does not know of any data on the probable temperatures once endured by the metamorphosed areas, but it seems possible tha they may have been high enough to have exceeded the critical points of the light oils which would have resulted from the cracking of the mother substances of petroleum under those temperatures, and thus to have permitted the mingling of oil vapors with the gases in all proportions.

FOOTNOTE 1. John L. Rich, Bull. Amer. Assoc. Petrol. Geol., Vol. 11 (1927), pp. 1139-49.

End_of_Article - Last_Page 84------------


(2) Geologist, Creek Drilling Company. Permanent address, 430 Temple Street.

Copyright 1997 American Association of Petroleum Geologists

Pay-Per-View Purchase Options

The article is available through a document delivery service. Explain these Purchase Options.

Watermarked PDF Document: $14
Open PDF Document: $24

AAPG Member?

Please login with your Member username and password.

Members of AAPG receive access to the full AAPG Bulletin Archives as part of their membership. For more information, contact the AAPG Membership Department at [email protected].