Sequences of alternating arenaceous and argillaceous lithologies are commonplace in the siliciclastic hydrocarbon reservoirs of the North Sea and elsewhere. Individual lithological units may vary in thickness from millimetres to tens or hundreds of metres. The first part of this introduction will discuss general aspects important for the understanding of the patterns of clay diagenesis associated with these sequences, whereas the second part will deal with the papers presented in this issue of Clay Minerals and how they fit into this general area of interest.

Analytical transmission electron microscopy (ATEM) was used to obtain chemical analyses of single illite crystals from Upper Carboniferous reservoir sandstones of the Bothamsall Oilfield, East Midlands, UK, and from Rotliegendes (Permian) sandstones of the North Sea. All samples were found to be highly aluminous, with only minor Mg and Fe, and to have near-ideal dioctahedral sheet totals. No significant variation in chemical composition of illite was found within the Bothamsall reservoir rocks, irrespective of paragenesis or stratigraphic horizon. Similar results were obtained from the Rotliegendes samples. Variation was found, however, between Bothamsall and Rotliegendes analyses populations. The Rotliegendes illites were distinctly more K-rich, which is the result of greater charge deficiency in the octahedral sheet. These results indicate that all the illite in both reservoirs precipitated in equilibrium with the reservoir pore-fluid. Furthermore, they imply that the physico-chemical composition of the pore-fluid did not evolve significantly between the different illite generations in the Bothamsall samples. These data, when compared with published analyses of authigenic illite, indicate that the compositional field for illite in sandstones is restricted in comparison with that of illite in mudrocks.

Authigenic illites and chlorites from clastic reservoirs, offshore Norway, have been studied by analytical transmission electron microscopy (ATEM). Present-day reservoir conditions range from 1600 m burial depth and 70°C to 4400 m and 160°C. The reservoir units have experienced continuous subsidence since early Cretaceous, and are presently at maximum burial. Textural and morphological evidence indicates that the authigenic illites and chlorites investigated are late diagenetic products, except for an early-formed berthierine in the shallowest reservoir. While the illites show very limited chemical variability at temperatures between 140° and 160°C the chlorites exhibit definite compositional trends with increasing temperatures. Tetrahedral Al increases substantially from 100°–160°C, and the octahedral vacancy decreases correspondingly. These observations indicate that continuous recrystallization of authigenic clays has taken place. This is also supported by the chemical composition of present-day porewaters, which are close to equilibrium with the clay mineral assemblage.

Well-cuttings and core samples from four Gulf Coast areas have been examined by high-resolution scanning-transmission electron microscopy and other techniques to detail the chemical, mineralogical and textural changes during diagenesis. Wells in each of the four areas exhibit a distinct smectite-to-illite (S-I) transition, but the estimated depths for the transition vary significantly: 5600 to 8800 ft for the onset of the transition, and 7500 to 13 600 ft for the completion (70–80% illite). Calculated temperatures also vary significantly: 58° to 92°C for the onset, and 88° to 142°C for the completion. No pattern is evident for characteristic temperatures of the S-I transition. These data suggest that it is inappropriate to use the S-I transition for determining absolute diagenetic temperatures in geological settings similar to Gulf Coast sequences. TEM data imply the presence of two distinctly different kinds of domains, interpreted as smectite (with some mixed-layer component?) and illite. Initially, smectite domains dominate and contain subparallel layers of variable thickness. As the transition proceeds, abundant dislocations provide avenues for diffusion of both necessary reactants (K and Al) and products (Na, Si, Fe, Mg); illite packets grow within a shrinking matrix of smectite. Variable temperatures for the transition may be due to several factors, including chemical heterogeneity of the original smectite, variation in water/rock ratio, porosity differences, diverse chemical character of available fluids, and the particular physical environment for diagenesis. It is suggested that actual depths and temperatures for the transition are primarily a function of kinetic factors associated with a reaction for which at least one of the principal phases is metastable; the transition therefore is highly dependent on local geological factors.

Devonian rocks from the Orcadian Basin show an extreme range of organic maturity. Mean vitrinite reflectance ranges from 0·6 to 10·5% and spore colour from yellow-orange to black. Illite/smectite (I/S) is abundant in the clay fraction of lacustrine mudrocks and shows variation in expandability from 30 to <5%. A correlation is observed between expandability and vitrinite reflectance, but the specific relationship is different in the Moray Firth and the Caithness-Orkney areas. Illite ‘crystallinity’ measurements in the Caithness-Orkney area are also correlated with vitrinite reflectance. Geological evidence favours an I/S origin from precursor smectite, and mineral and chemical analyses of lacustrine shales show no evidence for regional differences in composition that might affect illitization rates. The different relationships within each region are believed to result from the different relative response of organic maturity indicators and clay minerals to temperature and reaction time. The lower I/S expandabilities in the Caithness-Orkney region at a given vitrinite reflectance are attributed to relatively longer burial times.

Claystones of Kimmeridgian-Volgian(-Ryazanian) age are the main source rocks for oil in the Central Trough, North Sea, but they are too immature for oil generation in the Norwegian-Danish Basin. This paper describes the diagenesis of mixed-layer illite/smectite (I/S) in these source rocks as investigated by X-ray diffraction and high-resolution electron microscopy, interpretation for both methods being supported by computer simulations. Randomly-ordered I/S with ∼70% illite layers, together with discrete illite, kaolinite, and traces of chlorite and vermiculite, are detrital. The clay mineralogical data suggest the existence of an island at Gassum during part of the Upper Jurassic. Mixed-layer illite/smectite originating from volcanic ash is found in two wells in the Central Trough. The volcanism may be related to a period of stretching in the North Sea graben system during Jurassic-Lower Cretaceous time. Diagenesis has not altered the I/S in the Norwegian-Danish Basin. In the Central Trough, however, the amount of illite layers in I/S increases from ∼70% to ∼90% and the I/S becomes ordered. The volcanic clays have been changed to randomly ordered I/S with ∼50% illite layers. These changes occurred in the Central Trough simultaneously with oil generation.

Some of the many problems associated with the acquisition and interpretation of K-Ar isotope data for authigenic illites in porous sandstone lithologies are discussed. Difficulties arise from a lack of critical awareness of the assumptions made in deriving a K-Ar age of illite formation. Calculations are presented which imply that where sustained reservoir temperatures are high (> 150°C), erroneously low K-Ar ages could result from diffusive Ar loss. Very low levels of detrital contamination with other K-bearing minerals cause further difficulties. Even non-K-bearing contaminants may have a marked effect on apparent ages of illite ‘separates’. However, if considerable care is exercised during separation, the contamination problem is not intractable. The potential of the K-Ar technique to specify temporal development of the characteristics of reservoir rocks suggests that analytical refinements and basic experimental parameters are worth pursuit. Hypothetical examples of depth-age profiles are discussed in the context of their relevance to the timing and nature of hydrocarbon charging of reservoirs.

Subarkosic arenites of the Middle Jurassic Garn Formation are important hydrocarbon reservoirs in the Haltenbanken area. XRD and petrographic analyses of core samples from 11 wells show that a transition from incipient to extensive illitization in these sandstones is associated with a present burial depth of 3·7 km below the sea floor, corresponding to a formation temperature of 140°C. Illite has formed by reaction between K-feldspar and earlier diagenetic kaolinite and probably also by alteration of mica. Although several of the cores are filled with hydrocarbons that probably were in place before base-Eocene (57 Ma), there are no indications that illitization was inhibited by the presence of hydrocarbons in pore spaces. It is therefore suggested that 20–30% residual water saturation is sufficient for extensive illitization to occur by short-range diffusive transport in a compositionally closed system. There is an unresolved problem regarding both the timing and cause of illitization. On the one hand, the extent of illitization in the 11 wells correlates with variations in both present formation temperature and thermal maturity, implying that the present burial depth of 3·7 km below the seafloor coincides with a critical thermal threshold for illitization. According to this interpretation, the times of illitization in the different wells should correlate with variations in burial history and should in several cases be as young as 3 Ma, when deposition of one kilometer of glacially-derived sediment sharply increased temperatures throughout the Haltenbanken area. On the other hand, conventional K-Ar analyses of illite separates give dates ranging from 31–55 Ma. In general, these dates appear to be too old to fit the interpretation that the degree of illitization is a simple function of present temperature or thermal maturity. This inconsistency may reflect errors in the K-Ar dates due to a contamination problem.

The diagenesis of Brent Group sandstones at Heather Field was studied to reconstruct time-dependent variations in reservoir quality, hydrodynamic history, and oil emplacement. Depositional facies, isotopic and trace-element composition of authigenic minerals, and present-day formation-water chemistry indicate several major changes in porewater chemistry related to both gravitational and compactional flow systems that significantly impacted diagenesis. Early cementation by calcite was related to influx of meteoric water and completely occludes porosity in certain areas of the Field, especially in lower reservoir zones. Geochemical, petrographic, and structural evidence indicate that average calcite precipitated at low to moderate temperature from reducing isotopically-depleted water having high levels of radiogenic Sr (40°–50°C, δ18O = −4 to −6‰, 87Sr/86Sr > 0·71). A major period of kaolinite precipitation and feldspar dissolution followed calcite cementation. The isotopic composition of pore-filling kaolinite shows Field-wide uniformity (δ18O average 13·8‰, δD average −53·2‰), suggesting thorough flushing of the reservoir by meteoric water and precipitation at low to moderate temperature (45°–60°C). Tectonic, burial, and thermal histories suggest that meteoric flushing occurred during the late-Cimmerian sea-level low, possibly in response to gravitational flow of meteoric water from exposed parts of the adjacent East Shetlands Platform. Illite and quartz diagenesis post-date kaolinite cementation, with illite K-Ar ages indicating precipitation through much of the Paleogene (55–27 Ma), coincident with migration of hydrocarbons from neighbouring sub-basins of the East Shetlands Basin. Illite stable isotopic data indicate precipitation in a system resulting from partial mixing of trapped meteoric pore-fluids with saline compaction water. The intensity of sandstone diagenesis is influenced by differences in the fluid migration history, content of detrital K-feldspar, and the time of hydrocarbon emplacement and results in spatial and temporal variations in reservoir quality.

Piper and Tartan Fields are Upper Jurassic reservoirs located in fault blocks on either side of the Witch Ground Graben. The reservoirs are divided into a series of structural units, each successively downthrown towards the graben, covering a total depth interval of 8500–15 000 ft. Mudrocks interbedded with and overlying the reservoir sequence show a progressive increase in illite content with depth from ordered I/S clays with around 40% smectite to almost pure illites with low expandibility. K-Ar analyses on 13 <0·5 µm illite separates from mudrocks covering this depth interval give ages that range from 153 to 68 Ma. Shallowest samples give the oldest ages which are considered to reflect clay neoformation in Jurassic weathering profiles. In the 40Ar/K Harper diagram this data set falls on a mixing line which reflects the amount of smectite present. Deepest samples give the youngest ages and show a major reduction of 40Ar* with increasing K content indicating formation of new illite. These data are considered to document that ‘illitization’ proceeds via a two-step reaction: initial destruction of smectite followed by neoformation of illite. According to decompacted burial curves, ‘illitization’ took place 60–50 Ma ago. Reservoir sandstones have undergone a complex diagenetic history, involving multiple illite authigenesis. K-Ar analyses of 21 <0·5 µm illite separates selected across the oil-water contact in each of the main structural units show progressively younger ages in the water zone and with increasing depth of burial. Oldest ages in the oil zone (144–120 Ma) reflect eogenetic illite authigenesis whilst progressively younger ages in the water zone result from successive mesogenetic illite generations. The youngest age of 29 Ma was obtained from an illite-rich zone at the present oil-water contact in the Tartan downthrown block. Assuming illite authigenesis in the Tartan downthrown block took place shortly after oil accumulation, ∼29–35 Ma can be considered as a maximum age for the present oil accumulation. A minimum age for oil accumulation is calculated as 22 Ma on the basis of contamination of the 29-Ma sample with up to 10% of 93-Ma authigenic illite.

Wells 13/28-2, 13/28-3 and 13/29-1, located in the Inner Moray Firth, penetrated a predominantly arenaceous marine to non-marine sedimentary complex of Oxfordian age banked up against an argillized Caledonian granite. These sediments exhibit a close relationship between the pattern of mineral diagenesis and the physico-chemical conditions of the original depositional environment. Both marine and non-marine lithofacies have undergone a complex alternation of mineral cementation and grain dissolution. The marine lithofacies is dominated by silicate cements. The earliest cements consist of chert, phosphate and pyrite; these were followed, sequentially, by fine-grained quartz and minor mica, quartz overgrowth, dolomite (non-ferroan to ferroan), and calcite (non-ferroan to ferroan). The non-marine lithofacies is dominated by clay mineral cements. The earliest cement is kaolinite, followed by quartz overgrowth cement, then by mica cement and, finally, by a late phase of kaolinite cement. The porewaters responsible for the non-marine mica cement have caused extensive argillization of the granite occurring adjacent to the non-marine sediments in Well 13/29-1. K-Ar radioisotope dating suggests that the precipitation of the mica cement and the granite argillization both took place sometime during the late Early Cretaceous (95–115 Ma, late Aptian to Albian). The close correlation between style of mineral diagenesis and depositional environment in this sedimentary complex developed during burial within an essentially sealed and overpressured reservoir in which the different facies stewed in their own porewaters. At between 4000 and 6000 ft of burial, minor tectonic perturbations caused the limited invasion of the marine sediments by non-marine porewaters and of non-marine sediments by marine porewaters. These were responsible for the presence locally of (i) allochthonous non-marine cements (mica, late kaolinite) within the marine lithofacies, and (ii) allochthonous marine cement (ferroan calcite) within the non-marine lithofacies. Prior to the entry of hydrocarbons during the Early Tertiary at a burial depth of between 6000 and 9000 ft, the sandstones experienced a major loss of pore-fluid pressure, localized collapse, and the precipitation of allochthonous baryte cement in both the marine and non-marine lithofacies.

The NW Hutton Field lies within the Brent Province of the East Shetland Basin and is located in Block 211/27, 130 km NE of the Shetland Islands. Oil of 37° API, the source of which is the Upper Jurassic Kimmeridge Clay Formation, is reservoired in Brent Group sandstones in a complex, highly faulted and tilted fault-block structure at an average depth of 12000 ft subsea. The Brent Group Reservoir shows a complex diagenetic sequence. Early calcite concretions showing partial replacement by siderite occur in the marine-facies sandstones of the Broom and Rannoch Formations. The fluvio-deltaic and marine-facies sandstones of the Etive, Ness and Tarbert Formations show a different diagenetic sequence (post-dating carbonate cementation) of secondary porosity formation by K-feldspar dissolution, followed by quartz overgrowth cementation and kaolinite and illite authigenesis. Development of kaolinite was concentrated in the crestal areas of the structure while both illite and quartz cements increase with depth in the downflank areas. The marine sandstones of the Broom Formation are the exception, being kaolinite-rich with little illite even below the oil-water contact. An early phase of oil migration into the structure appears to have been concurrent with the later stages of quartz overgrowth and clay mineral cementation, the later main phase of migration causing diagenesis to cease in the crestal part of the field. In the down-flank areas illite cementation appears to have ceased soon after the main migration of hydrocarbon fluids into the structure.

A study of the ion-exchange properties of montmorillonite has been performed in order to facilitate computer predictions of the chemical properties of natural fluids and mineral assemblies. Clay/electrolyte interactions can be described using a technique based on the concept of hypothetical surface complex formation. This technique, which is compatible with ion-association models such as GEOCHEM, can be used to simulate simultaneous ion-exchange, hydrolysis of clay edges and anion adsorption on clay surfaces. Effects such as variable cation-exchange capacity and compositionally dependent exchange constants, normally indicating non-ideal behaviour, can be simulated using different combinations of ideal reactions involving charged surfaces and complexing groups representing clay edges. The modelling procedures are flexible and thermodynamically self-consistent. The techniques were applied to data on the ion-exchange characteristics of Wyoming bentonite to yield thermodynamic data for the reactivity of this clay with alkali metals, alkaline earths and a range of first-row transition metals at 25°C.

The shale from the Witch Ground Graben in the North Sea consists mostly of clay and calcite with relatively small amounts of quartz and plagioclase. The clay mineralogy is dominated by smectite and illite with varying amounts of chlorite and kaolinite. The ion composition was determined by ion chromatography, the porewater anions and cations being removed by leaching the shale with water, while the exchange cations were removed by reacting the shale with a large excess of tetramethylammonium ions. The clay mineralogy from XRD is consistent with the measured values of CEC. The water content and wireline log conductivity of the shale is controlled largely by the CEC (i.e. clay mineralogy). The cation content of the shale section is dominated by Na, with only small concentrations of K, Mg and Ca ions. The total anion concentration, which is dominated by chloride, shows an inverse relation to the concentration of exchange sites in the shale, suggesting that the compaction of the shale is to some extent controlled by a Donnan or salt-exclusion mechanism. Application of the Donnan equilibrium led to the reasonable conclusion that the shale is in compactional equilibrium with an external reservoir of about four times the anion content of seawater. The extent of salt exclusion in the shale is relatively low—a consequence of the high salt concentration in the external formation and the relatively low concentration of exchange sites in the shale.

Existing data on interatomic potentials have been used to study the interactions between an uncharged clay sheet and a water molecule. Calculations show that most of the clay surface is relatively hydrophobic, with binding energies for a water molecule in the range 1·0–4·5 kcal mol−1. There is, however, a low-energy site for an oriented water molecule above the layer OH group and within the ring of six SiO4 tetrahedra. Using two different models for the interactions, the binding energy in this position is found to be either 13·2 or 21·8 kcal mol−1. The existence of the low-energy site accounts for the formation of the hydrated ‘10 Å’ phase of talc, which is known from high-pressure experiments. Data on the PT stability of this phase can be used to estimate its energy of dehydration. This quantity is shown to be consistent with the value of 21·8 kcal mol−1 for the binding energy of a water molecule and the energy associated with the expansion of the layers from the 9·35 Å phase.

Late Triassic sandstones from Quadrant 22 in the North Sea contain two generations of chlorite cement. Grain-coating chlorites consisting of curved, irregular crystal plates formed very early in diagenesis and are largely confined to marginal marine beds. Later diagenetic pore-filling chlorites, consisting of euhedral platy crystals, occur throughout the sequence in fluvial as well as marginal marine beds. Both generations of cement show similar XRD patterns; the 001 basal reflection of 14·0–14·06 Å is broad and of low intensity, whereas the 002 reflection at 7·05 Å is sharp and intense. Slight contraction of the 14 Å basal spacing to 13·6 Å occurs after heating to 600°C. Electron microprobe analyses show the cements to be Fe-rich clinochlores with Fe/(Fe + Mg) ratios varying between 0·41 and 0·50. They are more Mg- and Si-enriched than most other reported North Sea authigenic chlorites. The grain-coating chlorites have an average composition of (Fe3·06Al3·93Mg3·32) [(Si6·76Al1·24)O20] (OH)16. The pore-filling chlorites tend to show more variable compositions. The high Si values probably reflect both minor interstratification with smectite or vermiculite and contamination of the analyses by a closely associated illite/smectite clay. An origin related to alteration of a precursor swelling chlorite or corrensite is attractive on chemical and morphological grounds for the early grain-coating chlorites. However, direct precipitation from porewaters is the favoured origin for the later pore-filling chlorites, with the required ions being supplied by mineral decomposition reactions involving detrital silicate minerals and dolomite cements.

Authigenic illite samples have been isolated from preserved reservoir core and characterized by XRD, XRF, SEM and ATEM. Reactivity towards aqueous solutions containing acids, alkalis, complexing agents and reducing agents has been evaluated using both static (flasks) and dynamic (flow-cell) experiments at 80°C and atmospheric pressure. In flask experiments with dilute reagents, reaction appears to be simple stoichiometric dissolution. The same pattern extends to higher concentrations (molar), except for alkalis where Si is preferentially leached leaving an Al-rich residual phase. Under the more vigorous conditions of flow (continuous leaching by fresh reactant solutions), stronger acid and alkaline solutions both cause substantial dissolution although by different mechanisms. Acids displace K+ and cause illite to swell. In many of these experiments, gels were seen to precipitate downstream from the reaction site. These reactions are rapid, taking place in hours at 80°C. Similar reactions can be expected to take place in consequence of a range of drilling, stimulation and production procedures. Clay dissolution, modification and, possibly, migration would be anticipated close to the well-bore and damage from scaling (gels) somewhat further away.