Jl geol. Soc. Lond. Vol. 135, 1978, pp. 69-72, 1 fig., 2 plates. Printed in Northern Ireland.

Clay mineral diagenesis and oil migration in the Middle Jurassic Brent Sand Formation N. J. Hancock & A. M. Taylor
SUMMARY: Authigenic clay minerals, in an almost complete sequence of oil-saturated Brent Sand from an oilfield in the Viking Graben in the northern Nortll Sea, show a transition from kaolinite in the upper part into illite in the basal part of the reservoir; replacement textures indicate that the illite has replaced kaolinite. After early kaolinite formation, it is suggested that illite diagenesis took place synchronously with oil migration, and that the progressive extension of oil saturation down through the reservoir terminated diagenetic processes. Early diagenetic stages are thus 'frozen' and preserved in the upper part of the reservoir. The Middle Jurassic Brent Sand Formation (Bowen 1975) occurs over a wide area of the northern Viking Graben in the northern North Sea, and forms the reservoir for a number of important oilfields. The three members designated by Bowen (1975) are laterally persistent in a similar 'deltaic' facies, but the degree and type of diagenesis of the sands is variable from well to well. This paper deals with the clay mineral diagenesis in an almost complete sequence of the Brent Sand Formation in an oilfield well where the entire sand section is oil-saturated. Extensive coring has facilitated comprehensive analysis of clay diagenesis in the sandstones, by means of petrographic examination, scanning electron microscopy, and X-ray diffraction analysis. Authigenic clay minerals, crystallized in the pore space of sandstones, have been documented from various areas (Shelton 1964, Carrigy & Mellon 1964, Webb 1974).

Crystal habit o| clays

The authigenic kaolinite near the top of the sand section can be seen to be well crystallized, both in thin section (PI. 1A) and with the scanning electron microscope (PI. la). It consists of 'books' of stacked, pseudohexagonal platy sheets (cf. Webb 1974). All the crystal edges are razor-sharp. A little further down the section, rare illite flakes are seen to surround kaolinite 'books', and locally to penetrate into them (PI. lc, D and 2<'). This texture marks the beginning of replacement of kaolinite by illite. In the middle part of the sand section, very distinctive clay mineral textures occur. In thin section, 'books' of crystallites are seen, but they now have the typical birefringence of illite (PI. 1E). The scanning electron microscope reveals that the sharp crystal boundaries of kaolinite have gone (P1. 1F): this is a classic replacement texture, with illite pseudomorphing pre-existing kaolinite, and preserving the gross morphology of the latter (cf. Millot 1970, p. 240). Illite crystals can take a variety of forms (see Hancock 1977, for example), but this illite after kaolinite resembles none of them. Low in the Brent Sand Formation, full recrystallization has taken place, and typical illite crystal forms are seen (P1. 2A, B). Further information on clay crystallinity comes from the X R D results (Fig. 1). Kaolinite has very good crystallinity (i.e. very sharp peaks) wherever it occurs in the well section. Illite peaks, however, exhibit a downward increase in sharpness, reflecting a downhole improvement in the quality of the illite crystals (Weaver 1960).

Distribution ot clay minerals

The higher parts of the Brent Sand Formation in the well section contain very abundant authigenic kaolinite (Fig. 1). Downwards, the kaolinite decreases in abundance, and its decline is accompanied by a steady, progressive increase in the proportion of authigenic illite. This is shown both by thin section analysis and by X-ray diffraction (Fig. 1). Some of the sandstones contain detrital clay in addition to authigenic kaolinite and illite. Detrital clay is not the subject of the present investigation, and it is important to recognise and distinguish it from the authigenic clays. X-ray diffraction traces of detrital clay show a mixture of poorly crystalline clays including mixed-layer illite/montmorillonite. In thin section, the crystal habit of the clays must be considered. Detrital clay is very fine-grained, and poorly crystalline, and this helps to separate it from the better crystallized authigenic clays.

Clay diagenetic history

Kaolinite growth in sandstones is commonly attributed to the effects of percolating acid water (e.g. Fairbridge 1967), altering labile detrital minerals, especially feldspars (Pettijohn et al. 1972). Feldspar is

70

N . J . H a n c o c k & A . M. Taylor closely linked with the late Cimmerian earth-movements near the Jurassic-Cretaceous boundary. However, the time of the alkaline (illite) diagenesis is much less certain. The evidence of the well section studied suggests contemporaneity of illite diagenesis and filling of the reservoir with hydrocarbons, but does not in itself determine the date of either process. The best evidence possibly comes from consideration of generation oi~'the hydrocarbons. These only begin to be released in significant quantities from the organic matter of sediments after the temperature exceeds 140-150F (Tissot et al. 1971), and there is evidence that pronounced oil production takes place at around 220240F (Philippi 1965, Burst 1969). Study of palaeotemperature gradients suggests that these temperatures were probably not reached until the Tertiary, perhaps the Miocene (Cooper et al. 1975). This, therefore, may represent the time of illite formation in the Brent Sand Formation. The clay diagenesis-hydrocarbon migration relationships described here are not unique to this well, and they may prove to have relatively wide applications. Dewatering of montmorillonite has been shown to be of great importance in hydrocarbon exploration (Burst 1969). More recently, a relationship between clay diagenesis and hydrocarbon migration has been demonstrated in the Mackenzie Delta (Barefoot & Van Elsberg 1975) which may have considerable exploration potential. The clay/hydrocarbon field needs further investigation. ACKNOWLEDGEMENTS. The board of directors of Robertson Research International Ltd. have given permission for this publication. PLATE 1 (facing) Photomicrographs (A. C, E) and scanning electron micrographs (B, D, F) of authigenic clay mineral textures in the Middle Jurassic Brent Sand Formation. FIGS. A, B. Abundant interlocked 'books' of authigenic kaolinite fill the pore space near the top of the Brent Sand. Note the pseudo-hexagonal platy crystals. A, photomicrograph, x200. B, scanning electron micrograph, x 1500. FIGs. c, D. An 'island' of authigenic kaolinite is surrounded by later-formed illite, and some penetration of illite into the kaolinite occurs. The shape of this kaolinite aggregate suggests that it may represent a replaced feldspar grain, c, photomicrograph, x 175. D, scanning electron micrograph, x730. FIGS. E, F. Authigenic illite pseudomorphing pre-existing kaolinite. The 'book' texture characteristic of kaolinite is clearly displayed (cf. A, a), but it is now composed of illite. Birefringence colours in thin section are those of illite. Note at high power the ragged nature of individual (illite) crystal plates (V) as compared to B. E, photomicrograph, x 400. F, scanning electron micrograph, 1500.

now almost totally absent from the well section under study, although our knowledge of the Brent Sand Formation suggests it was originally present, and also some of the kaolinite occupies 'grain-shaped' areas which may have once been feldspars (P1. lc). The kaolinite growth in the Brent Sand Formation may be associated with the late Jurassic-early Cretaceous Cimmerian earth-movements, when the crests of the structures in many Viking Graben oilfields were eroded (Bowen 1975, Ziegler 1975), and probably exposed to rain water. Subsequent growth of authigenic iUite, and conversion of kaolinite to illite, reflects the appearance of alkaline conditions at raised temperatures. These may be expected to develop with deeper burial, associated with the compaction and dewatering of shales (MiiUer 1967). Under these conditions, cations become concentrated in the pore fluids (Degens & Chilingar 1967). In the well section under study, kaolinite formation took place freely throughout the Brent Sand. Subsequent illite diagenesis, however, was inhibited in the higher parts of t h e reservoir, and only proceeded to completion low in the Brent Sand. Relation of oil migration to day diagenesis It is suggested that oil migration into the reservoir took place synchronously with illite diagenesis, as amplified below. The clay mineral relationships indicate that, following late diagenesis, the stable clay mineralogy consists of illite. It follows, therefore, that the kaolinite preserved unaltered in the upper part of the reservoir is a disequilibrium assemblage. Diagenetic processes take place in an aqueous medium, and consequently, as rocks become oilsaturated, diagenesis will be arrested. Inhibition of diagenesis by hydrocarbon accumulation has been described from various localities (e.g. Millot 1970, Webb 1974, Marie 1975). In the middle part of the reservoir, illite began to form before diagenesis was arrested. This demonstrates that progressive filling of the reservoir took place downwards from the top. As the hydrocarbon column extended progressively further downwards, water was excluded and diagenesis halted. The 'frozen' clay assemblages demonstrate that the alkaline waters forming illite must have been present, but they were expelled by the hydrocarbon accumulation before equilibrium could be attained. Oil migration was thus synchronous with the illite diagenesis; this has also been suggested by Sommer (1975). Only near the base of the sandstones did aqueous diagenetic reactions continue long enough for the development of abundant, well crystallized illite.

Time of day diagenesis

It seems most likely that the growth of kaolinite is

Jl geol. Soc. Lond. v o l . . 1 3 5 ,

1978

HANCOCK

& TAYLOR

?~?~ii

~i!ii i':~~

........iji!~, ~i ~ :~[
i~~

~,,~,i~!!. . . . .~i~i :' ~ii!~ ~ . .': . .

i,!
'il

. . . .

....~:~,

.....

........

~;~L,~

~i'~

PLATE

1 70

HANCOCK & TAYLOR

.It geol. Soc. Lond. vol. 1 3 5 , 1 9 7 8

~i~iiii!;i

... ~ii ~!iii '............

PLATE2

Photomicrograph (A) and scanning electron micrographs (B-D) showing replacement textures of authigenic clay minerals in the Middle Jurassic Brent Sand Formation.
FI6s. A, B. Authigenic illite fully recrystallized from pre-existing kaolinite characterizes the lower part of the Brent Sand. Pseudomorph textures (cf. PI. 1) have disappeared leaving a typical illite morphology. A, photomicrograph, 160. B, scanning electron micrograph, 900. FIG. . Authigenic kaolinite crystals being attacked by filamentous illite (right). Scanning electron micrograph, x 1700. FIc. D. Detail of the ecige of a partly pseudomorphed kaolinite 'book', showing illite growths developing on the rims of larger kaolinite sheets. Scanning electron micrograph, x 5700.

Clay mineral diagenesis and oil migration in the Middle Jurassic Brent Sand Formation
PERCENT OFROCK Kaolinite Illite UPPER BRENT SAND XRD TRACES (untreated clay fraction,<2pm)

71

I0 T

30
Feet

MIDDLE BRENT SAND

LOWER

BRENT SAND IIIite peak I0 ~

I I I

Kaolinite peak 7,16 ~,

I0

20

I0

20

FIG. 1. Clay mineralogy of the Brent Sand in the well section studied. Petrographic data show that kaolinite is replaced downwards by increasingly abundant illite. Representative X-ray diffraction traces (right) show changing emphasis from kaolinite to illite downhole, together with downward improvement in illite crystallinity (sharper peak). Note some detrital mixed-layer illite/montmorillonite in sample 2 (secondary peak on flank of 10/~ peak).

References
BAREFOOT, R. R. & VAN ELSBERG, J. N. 1975. The role of
adsorbed or structured water in the diagenesis of the Tertiary sediments in the Mackenzie Delta: a preliminary report. Geol. Suro. Can. Paper 75-1C, 303-9. BOWEN, J. M. 1975. The Brent Oil-Field. In: WOOKLAND,A. W. (q.v.), 353-61. BURST, J. F. 1969. Diagenesis of Gulf Coast clayey sediments and its possible relation to petroleum migration. Bull. A m . Ass. Petrol. Geol. 53, 73-93.

CARRIGY, M. A. & MELLON, G. B. 1964. Authigenic clay mineral cements in Cretaceous and Tertiary sandstones of Alberta. J. sedim. Petrol. 34, 461-72. COOPER, B. S., COLEMAN, S. H., BARNARD, P. C. & BUTI~RWORTH, J. S. 1975. Palaeotemperatures in the Northern North Sea Basin. In: WOODLAND, A. W. (q.v.), 487-92. DEGENS, E. T. & CHILINGAR, G. V. 1967. Diagenesis of subsurface waters. In: LARSEN, G. & CHILINGAR, G. V. (q.v.), 477-502.

72

N.J. Hancock & A. M. Taylor

FAIRBRIDGE, R. W. 1967. Phases of diagenesis and authigenesis. In: LARSEN, G. & CHILINGAR, G. V. (q.v.), 19-89. HANCOCK,N. J. 1978. Possible causes of Rotliegend sandstone diagenesis in northern West Germany. Jl geol. Soc. Lond. 135, 35--40. LARSEn, G. & CHIUNGAR, G. V. (eds) 1976. Diagenesis in Sediments: Developments in Sedimentology 8, Elsevier, Amsterdam, 551 pp. MARm, J. P. P. 1975. Rotliegendes stratigraphy and diagenesis. In: WOODLAND, A. W. (q.v.), 205-11. MILLOT, G. 1970. Geology of Clays. Springer-Verlay, Berlin, 429 pp. MOLLER, G. 1967. Diagenesis in argillaceous sediments. In: LARSEN, G. & CHILINGAR,G. V. (q.v.), 127-77. PETTIJOHN, F. J., POTTER, P. E. & SIEVER, R. 1972. Sand and Sandstone. Springer-Verlag, Berlin, 618 pp. PHtLrPPt, G. T. 1965. On the depth, time and mechanism of petroleum generation. Geochim. cosmochim. Acta 29, 1021-49,

SHELTON, J. W. 1964. Authigenic kaolinite in sandstone. J. sedim. Petrol. 34, 102-11. SOMMER, F. 1975. Histoire diagrnrtique d'une srrie grrseuse de Mer du Nord. Datation de l'introduction des hydrocarbures. Revue Inst. ft. Prtrole 30, 729-41. TISSOT, B., CALIFET-DEBYSER,Y., DEROO, G. & OUDIN, J. L. 1971. Origin and evolution of hydrocarbons in early Toarcian Shales, Paris Basin, France. Bull. Am. Ass. Petrol. Geol. 55, 2177-93. WEAVER, C. E. 1960. Possible uses of clay minerals in the search for oil. Bull. Am. Ass. Petrol. Geol. 44, 1505-18. WEBB, J. E. 1974. Relation of oil migration to secondary clay cementation, Cretaceous sandstones, Wyoming. Bull. Am. Ass. Petrol. Geol. 58, 2245-9. WOODLAND, A. W. (ed.) 1975. Petroleum and the Continental Shelf of North West Europe. 1, Geology. Applied Science Publishers, Barking, Essex, 501 pp. ZIEGLER, P. A. 1975. North Sea Basin history in the tectonic framework of North-Western Europe. In: WOODLAND, A. W. (q.v.), 131-49.

Received and read 9 March 1977; revised typescript received 6 May 1977. NIGEL J. HANCOCK& ALAN M. TAYLOR, Robertson Research International Ltd., 'Ty'n-y-Coed', Llanrhos, Llandudno, Gwynedd, North Wales

Discussion
DR. A. PARKER WRITES: Dr Hancock claims that in the Brent Sand there is an inverse correlation between authigenic illite and authigenic kaolinite, and considers that the illite is forming by progressive alteration of the kaolinite with increasing depth. He does not believe there to be a significant amount of detrital clay present. Except in the case of the Upper Brent sand, this correlation is far from obvious by simple visual inspection of the data, and some quantification would appear necessary. Furthermore, Dr Hancock has not determined the mineralogy of the shales that are interbedded with the sandstones. In a similar investigation, I have found that the clay mineralogies of adjacent shales and sandstones are often identical, both qualitatively and quantitatively. Although there may be some clay authigenesis in these Jurassic shales, it is reasonable to assume that most of their clay mineral component is detrital in origin, the Scandinavian platform having contributed fair amounts of kaolinite. I have found that it is very easy to overestimate the proportion of authigenic clay in such studies; when using the SEM in particular, it is far easier to identify wellcrystallized material than poorly crystalline or amorphous material, and one tends to be influenced by this in making visual estimations of relative proportions. In my experience, partial recrystallization and

re-distribution of detrital clay-sized material, particularly in pore throats, are usually more significant in permeability reduction than authigenic kaolinite, which generally forms in actual pores. A PARKER, Sedimentology Research Laboratory, The University, Whiteknights, Reading, R G 6 2AB.

DR N. J. HANCOCK & A. M. TAYLOR REPLY" Dr Parker's main point seems to be a concern that we have failed to take detrital clay into account, in our study of authigenic clays. In reply, we cannot usefully do more than refer him to the section on the distribution of clay minerals in our paper. Dr Parker points out that it is easy to over-estimate well crystallized authigenic clay with the SEM. We agree entirely: any single analytical technique is liable to introduce its own bias, and we would commend to Dr Parker the integrated approach which we have used, taking into account thin section petrography, SEM and XRD. Dr Parker's comments on the clay mineralogy of interbedded shales are most interesting. Unfortunately, we cannot comment for the well described in our paper, but we have not observed in shales the sort of depth sequence which we have described for the clays within the sandstones.