Does HC Generation Cause Over Pressure and Micro-Fracturing?

First let’s estimate the maximum rate of oil generation in a geological setting. Lets take a very good source rock, with an S2 yield of 30 mg/gRock (5% TOC and 600 mg/g HI). That is 3% in weight of HC generation potential. Converting to volume, it becomes 6 to 10% of the rock volume, depending on HC density (eg. 0.06 to 0.1 m³ per cubic meter of rock). Lets take the high end, and assume a short oil window of 10 million years (most oil windows are longer), the rate of generation is 0.1m³ /10,000,000year, or 10^-8 m³/ year, or 0.01 cc per year. Since a typical liquid drop contains 0.05 cc, this means in one cubic meter of rock, the source rock generates one drop of oil every 5 years! 

Now go on and think about how much pressure that generates, and whether this has any chance of making fractures in the source rock. Compare that one drop every 5 years rate with the 60 barrels per minute fluid injection rate we use to hydraulically frack the same rock.

Source rocks are quite "permeable" given geological time is 7 more more orders of magnitude longer than production time. Migration really does not require micro fractures. We have 10s of million years of time and a migration rate of only 0.0005 m/year is required to allow for the volume generated, creeping one pore space over a few years - at which rate, according to capillary number theory, it becomes a capillary dominated system, and viscosity (therefore permeability) does not even play a role.  

Even during gas generation, the volume increase is still minimal. Typical good marine source rocks generate only 10 to 20% of its potential as gas, and less than 30% including cracking of oil retained in the source rock. In situ gas density is lower, and volume may be 3 to 5 times of that for oil.  So we are still in the same rate range of less than 0.1 cc/year volume generated in 1 cubit meter of rock maximum.   

I do not believe these rates can cause micro fracturing, and do not believe micro-fractures are necessary for primary migration.  As HC generation happens almost uniformly everywhere in a mature source rock, fractures should be everywhere and in every mature source rock if they are required for expulsion/primary migration. We just don't see that, far from that. 

Core photo of Upper (A) and Lower (B) Eagle ford formation (Emmanuel Martin, 2013). Of all the core photos of mature source rocks I have seen, and I have been looking hard, vast majority of them do not have fractures of any kind. 

We do see micro-fractures in some source rocks, sometimes, most noticeable are those calcite filled "beef" like fractures are probably not caused by HC generation - but more likely formed during diagenesis before generation occurred. Micro-fractures are not present, or not pervasive in most mature source rocks, especially clay rich lacustrine and marine source rocks.  Occasionally we observe a few bitumen filled fractures but they tend to be a just a few, localized, not everywhere you would expect due to HC generation. Bitumen filled fractures are also not proof HC generation was the cause of the fractures. An simpler explanation of these is simply tectonic (especially shear) stress.  Some "fractures" in lab pyrolysis experiments are typically along bedding and probably caused by the thermal expansion of the rock during the experiments - even if we ignore the 13 orders of magnitude in heating (generation) rate.  Even at such high generation rate in the lab, Grohmann et al. (2021) shows that even 1 bar of vertical stress inhibits fracturing, as compared to the 100s of bars in geological conditions. 

References: 

R. Lenormand, E.  Touboul and C.  Zarcone, 1988, Numerical models and experiments on immiscible displacements in porous media, Journal of Fluid Mechanics 189(-1):165 - 187

S. Grohmann et al. Hyrous Pyrolysis of Source Rock Plugs: Geochemical and Visual Investigations and Implications for Primary Migration, 2021, IMOG conference paper. 

Does Over-Pressure Affect Seal Capacities ?

This may be the most controversial one I have ever posted. In this post I would like to challenge the widely accepted theory that over-pressure causes hydraulic fracturing in the seals and thus seal failure. I myself have taken this concept for granted during my entire career and have used it for basin modeling and petroleum system analysis for over 30 years. 

The figure below shows the reservoir pressure and HC columns in the HPHT area of the central graben in the North Sea. It is one of the most over pressured basins in the world, with over-pressures as high as 9000 psi above hydrostatic. First, there seems no correlation between pressure and the column heights. Some very large gas columns exist in the very over-pressured deep section. Pressures at top of some of the large gas columns exceed the regional LOP gradient and are very close to lithostatic!

Figure 1. Pressure and HC column data from the HPHT area of Central Graben, North Sea. Modified from Nygaard et al., 2020. Some of the reservoir pressures at top of gas column are above the minimum leak off pressure trend, and some approaches the 1 psi/ft (typical lithostatic) line.    

Many authors have mentioned hydraulic fracturing due to over-pressure may have caused some traps to have reduced columns, either not filled to spill, or have indications of a larger paleo-column, and some dry hole examples.  But such observations are common place in not so over pressured systems as well, and there can by many reasons not unique to over pressured systems. 

I have added the two dashed blue lines that connect several fields in what seems to be the centroid effects on several connected structures in a large fault block or compartment. This could explain the relatively lower pressure and large columns in the downdip parts of each centroid block. The centroid pressure transfer creates a condition that the seal has a higher pressure than the reservoir below. Several exceptionally large  (>1000 m) columns in several basins can be attributed to centroid effects.

Figure 2. Central North Sea aquafer pressure (hydrocarbon column pressures are corrected to water leg pressures).Central North Sea HPHT Pressure Cell Study, by Zimmer and Farris, 2021, Oil & Gas Authority

The second part of my challenge has to do with rates again. We can fracture the rock through leak off tests (LOT), or hydraulic fracking in unconventional plays. In figure 2, which include only water leg pressures, we see many cases of reservoirs pressures close to or at lithostatic pressure (1 psi/ft) and above the regional LOT fracture curve. 

In a LOT, fluid is pumped into the well bore at a typically 0.5 barrels/min and fractures initiate when pressure exceeds the rocks tensile strength in the lateral direction (σ3). The rate of pressure increase is at about 10 psi/min. In comparison, the pressure increase in the most rapidly buried basins is on the order of 0.01 psi/year (take the nearly 10,000 psi over pressure in figure 1 and assume all that happened in one million years). To add the pressure from buoyancy, if a 500 m gas column formed over a million years, that is 0.0005 m/year and also about 0.0005 psi/year increase in buoyancy pressure. We are talking about rates different by 8 orders of magnitude! And observation tells us, there is very little lateral pressure gradient in natural systems. We know deformation in geological times scales are more ductile, especially with the typical lithologies of seals - shales.

In addition, LOT is performed in a drilled hole, which strongly affect the stress and fracture characteristic of the formation. Stress concentration factor (Kt) is about a factor of 3 based on Kirsch's solution (E.G. Kirsch, 1898) for a circular hole in an infinite plate.  In nature such holes obviously do not exist, natural fractures (faults) are likely result of tectonic stress, and stress concentration is governed by structural geometry. 

This also means basin models that assume some "fracture gradient" below lithostatic based on LOT observations to bleed pressure off cannot explain this observation in figure 2. 

Note that the HC column has a higher pressure than water, and the difference is capillary pressure (P_hc-P_w). If the reservoir is water wet, as most reservoirs are, the non-wetting phase HC pressure does not transmit to the rock matrix. This perhaps is why some of the columns can extend all the way to the lithostatic line. However, when the water pressure itself does reach lithostatic - there is no room for a HC column (such as Juno in the figure). And it would be very hard do drill in this situation.  This does not happen everywhere, perhaps only where large vertical relief structures drain from a deep depocenter. Centroid effect can cause the pressure in the shallowest structure on a connect trend to far exceed the background pressure. Structures down dip from such as high location may have a lower reservoir pressure than the overlying shale may trap larger columns than the capillary seal alone.  Fields downdip from Juno, Shearwater, Elgin, and Franklin (Fulmar, and Pentland) seem to line up on the same water pressure gradient.  The Jade (Joanne, Judy) and Jasmin trend seem to have a similar situation. 

I am also not aware of any observed hydraulic fractures in seals in over pressured areas. And I would like to hear any additional evidence that seals fail due to over pressuring. 

All petroleum accumulations are found below a seal, a layer of rock with tighter pore throats than the reservoir. This is true with tight reservoirs (unconventional) too (He and Xia, 2017). So the main mechanism of petroleum traps is capillary. The capillary force balance equation for column height , H = 2γcos(θ)[1/r-1/R]/g(ρ_o-ρ_w), has no pressure term (Purcell 1949, Berg 1975, Schowalter 1979). Any effect over pressure may have on capillary seal capacity would have to be indirectly on how it may affect pore throat size. We can assume over pressure can inhibit compaction, but it would be to a very minor degree compared to the range of pore throat sizes among the different seal rocks. 

Selected References:

Kirsch, E.G., "Die Theorie der Elastizität und die Bedürfnisse der Festigkeitslehre," Zeitschrift des Vereines deutscher Ingenieure, Vol. 42, pp. 797-807, 1898. (no I can't read German).

J. NYGAARD et all, 2020, The Culzean Field, Block 22/25a, UK North Sea, Geological Society London Memoirs · October 2020

WINEFIELD, P., GILHAM, R. & ELSINGER, R. 2005. Plumbing the depths of the Central Graben: towards an integrated pressure, fluid and charge model for the Central North Sea HPHT play. In: DORÉ, A.G. & VINING, B.A. (eds) Petroleum Geology: North-West Europe and Global Perspectives: Proceedings of the 6th Conference. Geological Society, London, 1301–1315.

Ole Christian Engdal Sollie, University of Bergen Master's thesis, 2015, Controls on hydrocarbon column-heights in the north-eastern North Sea

T. T. Schowalter, 1979, Mechanics of Secondary Hydrocarbon Migration and Entrapment; AAPG Bulletin vol. 63 (5): 723–760.

Berg, R.R. (1975) Capillary Pressures in Stratigraphic Traps. AAPG Bulletin, 59, 939-956.

Purcell, W. R., 1949, Capillary pressure--their measurements using mercury and the calculation of permeability therefrom: AIME Petroleum Trans., v. 186, p. 39-48.

He, Z and D. Xia, 2017, Hydrocarbon Migration and Trapping in Unconventional Plays, Search and Discovery Article #10968 (2017), AAPG Annual Conference Presentation.

Eva Zimmer, Matt Farris, 2021, Central North Sea HPHT Pressure Cell Study, Oil & Gas Authority.