The Primary Migration of Hydrocarbons


The migration of hydrocarbons presented one of the more controversial aspects of petroleum geology although the general aspects are now fairly well decided.   It was evident that hydrocarbons underwent migration but the mechanism whereby this took place was not clearly understood.

In 1980  Robert and Cordell in the AAPG published the consensus for the latter quarter of the 20th century.

"We know there is little we can prove, and we do not excuse ourselves for the assumptions or conclusions which may later prove to be erroneous ... what little we do see is biased by what we look for with our mind ... we cannot actually prove when and whence a particular show of oil or gas came or how long it will remain where we see it.  To be completely honest, we can only conjecture about oil and gas movements."

Whereas the statement is true today for most shows of hydrocarbons the reason is more because be do not bother to prove when and whence a particular show of oil or gas came.  Our technology today CAN tell the origin of a particular show but in general it is not worth either the time or expense.


There are four major processes involved in understanding hydrocarbon migration.




More than one mechanism has been suggested for these processes and in reality most of the mechanisms suggested probably operate under their own set of special circumstances.  The question to ask is which mechanism was the dominant one under the prevailing conditions?  Burial is the prime processes controlling all of the mechanisms suggested for the primary migration of hydrocarbons.


The general process is as follows.

Mechanisms of Migration

With regard to the mechanisms involved in migration there are seven main questions to answer.  These were discussed by Magara [1980:33-45].

The reason an understanding of migration is time consuming and, maybe, difficult it that there are a lot of variables involved  and their interactions may be complex. Complexity is ubiquitous in nature and although science had gained some insight into the process, the understanding of migration in a deep-seated rock is  difficulty.

Mechanisms connected with the origin in the source rock.

Three modes for migration of the organic component are normally suggested.

Baker [1980:30] noted that at shallow depths [>6,000 feet] the only hydrocarbons present in the source rock are the biological markers. Associated with these are bacterially generated gas [biogenic gas].  Liquid hydrocarbons accumulating in shallow areas will tend to have a high content of aromatic hydrocarbons, presumably because aromatics are most soluble in water and can thus be transported further.  As the thermal generation of hydrocarbons begins [>6,000 feet] heavy oil with high NSO content tends to form.  With increasing depth [10,000 feet] thermal cracking produces a lot of low molecular weight hydrocarbons. In the deepest part of the basin liquid crude i replaced by condensate and ultimately petrogenic gas. Jones [1980:56, fig.5] presents a diagram showing the sharp drop in H/C ratio during thermal maturation of un-extracted organic matter, indicating the "ejection of the kerogen - by the kerogen", of the more hydrocarbon rich generation products. Jones [p:47] concluded that "most of the major commercial oil accumulations of the world left their source rock in a continuous oil phase". We know quite well that liquid hydrocarbons can be generated from organic matter by increasing temperature and the real question to ask is if liquid hydrocarbons did form in a specific rock body, how to they get out of the source rock into the reservoir.


Very little real-world data on the solubility of hydrocarbons under differing subsurface salinity, pressure and temperature conditions is available. Hydrocarbons do have very low solubility in water and most petroleum engineers reject solution as an important mechanism, except in special circumstances such as shallow gas fields, or elevated temperature at great depth [Tissot and Welte, 1978:296; Hunt, 1979:208-213].  It is definitely true that as peak generation approaches a source rock will contain several times the hydrocarbons required for saturation of the pore water.  As a further argument against the idea McAuliffe [1979] notes that  the molecular composition of oil bears no relationship to variations in hydrocarbon molecular solubility in water.

The solution of hydrocarbon in gas has been suggested but it is unlikely that the heavier hydrocarbons could dissolve in this way. 

It is possible that much of the dissolved organic matter in a source rock is not in the form of hydrocarbon but is a pre-cursor such as alcohol or an organic acid or colloidal aggregate.  These substances are more soluble than hydrocarbons [Barker, 1980:21, Tab. II].  A mechanism involving solution of a pre-cursor chemical may be operative in some circumstances for it is undoubtedly true that such materials go into solution.  Both Dickey [1975] and Magara [1980:35] rejected the colloidal aggregate ideas, partly because this would not greatly increase the solubility of the heavier hydrocarbon molecules.

Mechanisms connected with  movement in the source rock.

There are three important considerations when one examines the movement of hydrocarbons in the source rock.



Both grain size and pore size are related to porosity. In source rocks pore diameter is extremely small 100 angstroms at 3,000 feet and 25 angstroms at 12,000 feet].  This lower size is only slightly larger than bitumoid molecules [Hinch, 1980:7] and the old argument which precludes the possibility of continuous oil flow in fine grained source rocks due to the small pore space and capillary size is clearly not valid: as the deeper fine grain reservoirs of the Gulf of Mexico testify. It is possible that the fine pore space does stop oil leaving a source rock but in oil source rocks, which have a greater amount of organic matter, the combined effects of hydrocarbon generation and compaction cause high differential pressure effects, which more the hydrocarbons our the the source rock. 


Mompers [1978] discussed the distribution of particulate organic mater in source rocks.  It is important to note that POM is often located in laminations within a source rock, and the particles are larger in size than the associated clay minerals. The organic layers thus provide localized stingers or larger connecting pores exactly in the part of the source rock where hydrocarbons or their pre-cursors originate.   The heterogeneous distribution of hydrocarbons in source rocks led McAuliffe [1980:99] and Erdman [1965] to suggest hydrocarbons are generated in this organic matter and flow through it to the reservoir.  Under this idea the lower limit of 0.5% TOC in source rock would roughly represent the value below which there would be insufficient organic matter to form a 3-D network.



Mompers [1978] clearly outlines the characteristics of a source rock which are important in the development of micro-pathways with the rock.  The mechanism also was discussed by Snarski [1970], Dickey [1975], Magara [1978] and Tissot and Welte [1978].  Mompers' basic idea  is that as hydrocarbons are generated the increase in volume causes an increase in pressure.  At some point the pressure increase causes micro-fracturing in the rock, and the hydrocarbons migrate into the micro-fractures which lead out of the source rock.  This concept allows the hydrocarbons to migrate in a liquid phase, but does not preclude the additional or alternative migration in a pre-cursor form.  This is regarded as the main mechanism for primary migration out of the source rock.


Droplets, either of hydrocarbons or water, need pore space within which to migrate and if the pore space is less than the diameter of the molecule then the molecule must either deform to pass through the pore space or will block the pore space. As temperature raises with burial, the pore space becomes aqua-thermally pressured.  An example would be a source rock with 10% porosity, heated to 115oC from 85oC [which could correspond to an increased depth of burial from 10,000 feet to 14,000 feet in the Gulf Coast Basin]. Under such a change the increase in volume of the pore water will be 0.15% of the volume of the rock [Barker, 1980:25].  This increase in volume is added to the increase caused by the generation of hydrocarbons.  The result is an increase in pore fluid pressure resulting in micro-pressuring [or deformation and pushing of the oil droplets through the pore spaces].  If the source rocks are at a critical temperature then micro-pressuring could result from a very small increase in depth of burial. Hinch [1980:1] pointed out that it is difficult to explain primary expulsion of hydrocarbons via pore water expulsion during simple compaction and flushing because by the time hydrocarbons are generated in sufficient amounts most of the pore water already has been expelled by compaction and the amount of water remaining is insufficient to flush hydrocarbons from the source rock.


Pore space chemistry.


Certain minerals contain water and with increased burial the temperature and pressure increase can cause the water to be released from the minerals.  This gives an additional source of water for flushing. The clay mineral Smectite [Montmorillonite] undergoes this process. Inter-layered water is released from smectite, the resulting material reacts with potassium in the rock to form the new clay illite. This reaction may be of significance in some cases and is probably continuous from the surface to depth.


A second possible process involving the pore space chemistry is a consequence of molecular reactions between the mineral grains and the pore fluids, resulting in the preferential adsorption of water onto the mineral grains, and the dynamic structuring of water close to the mineral grain [Hinch, 1980:11, figs 8-10].  Dynamic structuring takes place because the inter-molecular forces are electro-static.  Water readily reacts to these electro-static forces but the bulk of the hydrocarbons do not.   The water is adsorbed to the grains surface i.e. becomes part of the grain structure].  The water closest to the grain is most highly structured, the water nearer to the center of the space is lease structured.  Because the hydrocarbons tend to be excluded from the structured water area they will collect in the center of the pore spaces. This dynamic structuring can also present a minor expulsion mechanism because the smaller pores are more structured than the larger pores and the hydrocarbons are driven [diffused] to the larger pore spaces and micro-fractures.  Thus in a source rock there may be a built-in directional diffusion pathway prior to hydrocarbon generation.  Witherspoon and Saraf [1964] note the diffusion rate is directly related to the molecular size: gases being more effectively mobilized than liquids.


Mechanism pertaining to the initiation of secondary migration.


It has been suggested that for any hydrocarbons or their pre-cursors, dissolved in the pore water they will come out of solution as the water passes from the source rock into the reservoir rock due to changes in salinity and chemical characteristics of the pore water [Baharlou, 1973; Schmidt, 1973; Barker 1980]. McAuliffe [1980] noted "the secondary migration of oil as a continuous passes is probably the most efficient mechanism for moving oil and gas from source rocks to reservoir rocks, leaving a minimum of residual oil along the migration pathway".

Under conditions where the hydrocarbons are in liquid form they will tend to accumulate at the top of the reservoir bed. Most bedding planes are irregular and the droplets of hydrocarbon will tend to accumulate in the local irregularities. Once sufficient number of droplets have accumulated [20-30% of the pore space] they will move under buoyancy into and along the reservoir. Thereafter movement by either buoyancy or hydraulic pressuring will occur.  It is probable that flow in the reservoir, in geological time, will be almost instantaneous, and not much slower than oil field flooding methods. McAuliffe [1980] regarded the buoyancy flow from the source rock as taking place in slugs of hydrocarbons.  This is interrupted flow, in which hydrocarbons will flow once sufficient have built up for buoyancy to have an effect. In this situation the hydrocarbons will be replenished form the source rock only slowly and definitely too slowly to allow continuous separation of a liquid oil phase from the source rock to the trap. 

Finally:  for an oilfield to develop the reservoirs must have access, somewhere, somehow and at some time, to mature source rocks.