Empirical relationships between clay mineral transformations and temperature provide a basis for the use of clay minerals as geothermometers. Clay-mineral geothermometry has been applied mainly to diagenetic, hydrothermal, and contact- and burial-metamorphic settings to better understand the thermal histories of migrating fluids, hydrocarbon source beds, and ore and mineral formation.

Quantitatively, the most important diagenetic clay mineral reaction in sedimentary rocks is the progressive transformation of smectite to illite via mixed-layer illite/smectite (I/S). Changes in both the illite/smectite ratio and ordering of I/S, as determined from X-ray powder diffraction profiles, correlate with changes in temperature due to burial depth. Although the smectite-to-illite reaction may be influenced by several factors, reaction progress appears to be strongly controlled by temperature. Studies show that the model proposed by Hoffman and Hower in 1979 is applicable in burial diagenetic settings from about 5 to 330 Ma, and includes most rocks about Miocene to Mississippian in age. Reliability of the I/S geothermometer is, however, dependent upon a good understanding of the rock's original clay-mineral composition.

Changes in the ordering of I/S are particularly useful in the exploration for hydrocarbons because of the common coincidence between the temperatures for the conversion from random-to-ordered I/S and those for the onset of peak, or main phase, oil generation. Here, the utility of the I/S geothermometer is reviewed in hydrocarbon-bearing rocks of Miocene to Mississippian age. Using three common applications, the I/S geothermometer is compared to other mineral geothermometers, organic maturation indices, and grades of indigenous hydrocarbons. Good agreement between changes in ordering of I/S and calculated maximum burial temperatures or hydrocarbon maturity suggests that I/S is a reliable semiquantitative geothermometer and an excellent measures of thermal maturity.

Characterization of the Nonesuch Formation, middle unit of the Proterozoic Oronto Group, as a potential hydrocarbon source for the Lake Superior basin portion of the Midcontinent Rift system requires an understanding of the thermal maturity of the region and its relationship to the thermal history. Illite/smectite (I/S) expandability data were collected from the Nonesuch Formation and the overlying Freda Sandstone and compared with organic thermal maturity data; both data sets coupled with a thermal and burial history for the White Pine area of Michigan allow regional interpretation of maximum formation temperatures of the Nonesuch Formation and the Freda Sandstone with respect to time. Samples collected from drill holes in northeastern Wisconsin display nearly pure smectite within the lower Freda Sandstone trending abruptly to ordered I/S within the Nonesuch Formation. Regular trends of decreasing expandability with depth occur in four other drill holes to the northeast. Comparison of I/S expandability between similar stratigraphic intervals reveals a significant trend of increasing thermal maturity to the northeast, with the lowest thermal maturities observed in the Iron River Syncline area just west of White Pine, Michigan.

I/S geothermometry suggests maximum temperatures in the Nonesuch Formation of 140°C in Wisconsin, 115°C in the Iron River Syncline area, 160°C at White Pine, and 190°C near the southern portions of the Keweenaw Copper District. The geographic pattern of temperatures determined from I/S geothermometry is identical to that determined from organic thermal maturity indicators in the Nonesuch Formation (Imbus et al., 1988, 1990; Hieshima and Pratt, 1991; Pratt et al, 1991; Mauk and Hieshima, 1992).

Regular variations in I/S expandability with depth occur in the Freda Sandstone and the Nonesuch Formation near the southern limits of the Keweenaw Copper District. These variations suggest a fossil geothermal gradient of 55°C/km and limit the thickness of sediment above the Nonesuch Formation to approximately 3 km. In comparison, 3.6 km of Freda Sandstone are presently exposed near the Wisconsin border, and numerical modeling suggests a range of 4–6 km of sediment overlying the Nonesuch Formation. None of the data indicate the presence of the Bayfield Group sediments above the Nonesuch Formation at the time of clay diagenesis. Samples from White Pine suggest a two-stage burial history: 1) clay reaction, possible hydrocarbon maturation, and copper-sulfide mineralization at maximum temperatures above 100°C during the main rifting and burial event, followed by 2) fracturing, reverse faulting, and fluid circulation during a rift-terminating compressional event that may have allowed petroleum migration and native copper mineralization at temperatures below 100°C. Abrupt changes in I/S expandability with depth and the presence of poorly crystalline I/S (greater than 80% expandable) and kaolinite in the Freda Sandstone in Wisconsin appear to represent later overprinting of the diagenetic assemblage by fluids that were probably cooler and of differing composition than earlier diagenetic fluids. However, the authigenic assemblage from the vicinity of White Pine, Michigan, which includes up to 25% expandable I/S, appears to represent a diagenetic profile formed during the main rifting and burial event. Therefore, these expandable I/S-type clays are essentially 1.0 billion years old.

The clay mineralogy of tuffs from Yucca Mountain, Nevada, the potential site of the nation's first high-level radioactive waste repository, has been studied in order to understand the alteration history of the rocks and to predict potential future alterations. Bulk-rock samples and clay-mineral separates from three drill holes at Yucca Mountain (USW G-1, USW G-2, and USW GU-3/G-3) were studied using X-ray powder diffraction, and supporting temperature information was obtained using fluid inclusion data from calcite. Twelve K/Ar dates were obtained on illite/smectite (I/S) separated from the tuffs from the two northernmost drill holes, USW G-1 and G-2. The predominant clay minerals in the Yucca Mountain tuffs are interstratified I/S, with minor amounts of chlorite and interstratified chlorite/smectite. The I/S reactions observed as a function of depth are similar to those observed for pelitic rocks; I/S transforms from R = 0 interstratifications through R = 1 and R ≥ 3 interstratifications to illite in USW G-2 and to R ≥ 3 I/S in USW G-1. The R = 0 I/S clays in USW GU-3/G-3 have not significantly transformed. K/Ar dates for the I/S samples average 10.4 my. These data suggest that the rocks at depth in the northern portion of Yucca Mountain were altered 10.0-11 my ago, soon after creation of the Timber Mountain caldera to the north. Both I/S geothermometry and fluid inclusion data suggest that the rocks at depth in USW G-2 were subjected to postdepositional temperatures of at least 275°C, those in USW G-1 reached 200°C, and rocks from USW GU-3/G-3 probably did not exceed 100°C. These data suggest that no significant hydrothermal alteration has occurred since Timber Mountain time, ~ 10.7 my ago.

Estimates of the temperature of formation of illite/smectites yield probable stability limits for several minerals at Yucca Mountain. Clinoptilolite apparently became unstable at about 100°C, mordenite was not a major phase above 130°C, and analcime transformed to albite above 175°-200°C. It appears that cristobalite transformed to quartz at 90°-100°C in USW G-2 but must have reacted at considerably lower temperatures (and for longer times) in USW GU-3/G-3. The reactions with increasing depth appear coupled, and clinoptilolite and cristobalite disappear approximately simultaneously, supporting aqueous silica activity as a controlling variable in the clinoptilolite-to-analcime reaction. The reaction of clinoptilolite to analcime also coincides with the appearance of calcite, chlorite, and interstratified chlorite/smectite. Although the hydrothermal fluids may have been a source for some cations, breakdown of clinoptilolite (and mordenite) probably provided the source of some of the Ca for calcite, Mg for chlorite, K for the I/S found deeper in the section, and Na for analcime and albite.

Using the rocks in USW G-1, G-2, and GU/G-3 as natural analogs to repository-induced thermal alteration suggests that the bulk of the clinoptilolite- and mordenite-bearing rocks in Yucca Mountain will not react to less sorptive phases such as analcime over the required lifetime of the potential repository. The zeolites in zeolite interval I, directly underlying the proposed repository horizon, may transform at the predicted repository temperatures. However, the reaction of clinoptilolite to analcime in interval I may require the transformation of all of the abundant opal-CT and glass to quartz, an unlikely scenario considering the unsaturated nature of these rocks and the predicted temperatures of <100°C.

The smectite-to-illite conversion during shale diagenesis has recently been used to constrain the estimate of a basin's thermal history. We have systematically investigated the kinetics for the conversion of a Na-saturated montmorillonite (SWy-1) to a mixed-layer smectite/illite as a function of KCl concentration (from 0.1 to 3 moles/liter) over a temperature range of 250° to 325°C at 500 bars in cold-seal pressure vessels using gold capsules. The results show that the conversion rate can be described by a simple empirical rate equation

-dS/dt = A · exp(-Ea/RT) · [K+] · S2

where S = fraction of smectite layers in the I/S, t = time in seconds, A = frequency factor = 8.08 × 10-4 sec-1, exp = exponential function, Ea = activation Energy = 28 kcal/mole, R = gas constant, 1.987 cal/deg-mole, T = temperature (degree Kelvin), [K+] = K+ concentration in molarity (M) in the fluid.

The results also show that Ca2+ in solutions barely affects the illitization rate, whereas Mg2+ significantly retards the rate. The retardation, however, is not as severe as previously reported. Na+ ion can significantly retard the rate only if the concentration is high.

We found that by assuming a range 0.0026-0.0052 moles/liter (100-200 ppm) of K+, concentrations similar to the value typically reported in oil field brines, the present kinetic model can reasonably predict the extent of the smectite-to-illite conversion for a number of basins from various depths and age. This narrow range of potassium concentrations, therefore, is used to model the smectite-to-illite conversion in shale when the actual chemical information of pore fluid is not available.

The kinetic equation has been tested using field data from a large variety of geologic settings worldwide (i.e., the Gulf of Mexico, Vienna Basin, Salton Trough Geothermal Area, East Taiwan Basin, Huasna Basin, etc). The results show that the equation reasonably predicts the extent of the reaction within our knowledge of the variables involved, such as burial history, thermal gradients, and potassium concentration.

Comparing the state of reaction advancement of I/S (illite content of illite/smectite mixed layer minerals) and organic matter (vitrinite reflectance) in instances of high temperature gradients allows one to observe the evolution of geologic materials under extreme conditions. In a geothermally heated sedimentary series (Salton Sea area, California), both clays and organics have reacted to completion in the temperature gradient range of 200°-400oC/km during heating episodes of 104 years. We observed an instance of rapid heating, through magmatic intrusion into the Meso-Paleozoic eastern Paris Basin sedimentary series in late Permian time, which induced changes in the organic material, but where clays are apparently unaffected.

Authigenic kaolinite and illite are important diagenetic minerals in the Magnus Sandstone, a giant oil reservoir in the northern North Sea. These clay minerals, separated from three wells, show considerable ranges in their oxygen isotopic composition (δ8OSMOW = +9 to + 16%) and hydrogen isotopic composition (δDSMOW = - 55 to - 105%). The variations in δ18O and δD are positively linearly correlated with a high degree of statistical significance for both kaolinite and illite:

$\begin{array}{c}\mathrm{K}\mathrm{a}\mathrm{o}\mathrm{l}\mathrm{i}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{e}:\mathrm{n}=12;\delta \mathrm{D}=6.1{\delta }^{18}\mathrm{O}-169;\mathrm{r}=0.66(>95\mathrm{\%})\\ \mathrm{I}\mathrm{l}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{e}:\mathrm{n}=11;\delta \mathrm{D}=5.9{\delta }^{18}\mathrm{O}-159;\mathrm{r}=0.78(>99\mathrm{\%}).\end{array}$

Formation of the clays in a pore fluid of uniform isotopic composition over a range of temperatures appears unlikely. It is suggested that the observed relationships between clay mineral δ18O and δD are perhaps best explained by a model of precipitation at more or less constant temperature from pore fluids which varied isotopically across the oilfield. The isotopic composition of the formation waters would then lie along the line: δDw = 6.2 δl8Ow - 50. This is most plausibly interpreted as a mixing line with suggested minimal endmembers at (δ18O, δD) values of (+4, -24) and (-4, -76). The first of these represents reasonable isotopic values for Magnus Sandstone formation waters. Although δ18O of the second is compatible with an evolved Cretaceous meteoric water, its δD value is difficult to understand in the context of the model.

The importance of precipitation rate as an effective control on illite and kaolanite formation during diagenesis has been examined by measuring precipitation rates, from Al fluid concentration, in a Dickson fluid-sampling vessel at 160°-250° and 500 bars (50 MPa). These experiments are considered to be analogues of the precipitation of clays in sandstones from porewaters containing dissolved carboxylic acids, which have a transient stability and may influence aluminosilicate solubility. Precipitated illite had a lath-shaped morphology and its composition was consistent with authigenic illite in sandstones. Kaolinite formed tabular rather than vermicular shaped crystals. Kaolinite precipitation rate was two orders of magnitude faster than illite precipitation and was rate-limited by the decomposition of oxalate; kaolinite formation should be equilibrium-controlled at virtually all stages of burial. Extrapolation of illite precipitation rate to burial temperatures indicates that the first appearance of illite in a burial sequence may be kinetically controlled. A model of illite precipitation based on these experimental results has been used to predict the time required to precipitate illite during burial of a sandstone, taking into account temperature changes during burial. For northern North Sea examples, a predicted illitization threshold of -60°C occurring at 60–80 Ma corresponds to the observed initiation of authigenic illite precipitation. Times of around 2–5 Ma would be required to reach a 98% approach to equilibrium at this threshold. The main phase of illite precipitation in the northern North Sea basin is a later, hydrologically controlled event (30-50 Ma). Equilibrium would be approached in around 0.1 Ma during this phase, which is consistent with the narrow illite K/Ar age range (1-5 Ma) recorded for some sequences.

Smectite was reacted at several temperatures between 200°C and 500°C to produce interstratified illite/smectite (I/S) with different proportions of expandable layers. Dispersed and sedimented products were examined using a transmission electron microscope. Particle size and aspect ratio showed no systematic change as a function of reaction extent during R0 illitization. However, particles exhibited rounded edges during the early stages of the reaction, suggesting some dissolution of primary smectite. Additionally, increasing particle contrast in the electron beam suggests thickening of particles with increasing reaction extent. The thickening of particles is thought to be produced by the nucleation and precipitation of secondary illite layers on primary smectite layers. In the most extensively reacted I/S, particles have become aggregated into clumps or quasicrystals by lateral growth of illite layers. Internal uniformity of crystallographic alignment of individual growing crystals within each aggregate was reflected in the increasing frequency of 60° and 120° interfacial angles within each aggregate. In highly illitic I/S, these aggregates took on an overall euhedral form and became crystallographically contiguous, producing single crystal electron diffraction patterns.

Chlorite minerals, found in a great variety of rocks and geological environments, display a wide range of chemical compositions and a variety of polytypes, which reflect the physicochemical conditions under which they formed. Of particular importance for studies dealing with ore deposit genesis, metamorphism, hydrothermal alteration or diagenesis is the paleotemperature of chlorite crystallization. However, in order to understand the relationship between chlorite composition and formation temperature and hence use chlorite as a geothermometer, one must determine how other parameters influence chlorite composition. These parameters may include fO2 and pH of the solution and Fe/(Fe + Mg) and bulk mineral composition of the host rock.

Four approaches to chlorite geothermometry, one structural and three compositional, have been proposed in the past: 1) a polytype method based on the (largely qualitative) observation that structural changes in chlorite may be partly temperature-dependent (Hayes, 1970); 2) an empirical calibration between the tetrahedral aluminum occupancy in chlorites and measured temperature in geothermal systems (Cathelineau, 1988), which has subsequently been modified by several workers; 3) a six-component chlorite solid solution model based upon equilibrium between chlorite and an aqueous solution, which uses thermodynamic properties calibrated with data from geothermal and hydrothermal systems (Walshe, 1986); and 4) a theoretical method based on the intersection of chlorite-carbonate reactions and the CO2-H2O miscibility surface in temperature-XCO2 space, which requires that the composition of a coexisting carbonate phase (dolomite, ankerite, Fe-calcite or siderite) be known or estimated (Hutcheon, 1990). These four approaches are reviewed and the different calculation methods for the compositional geothermometers are applied to a selection of chlorite analyses from the literature. Results of this comparative exercise indicate that no single chlorite geothermometer performs satisfactorily over the whole range of natural conditions (different temperatures, coexisting assemblages, Fe/(Fe + Mg), fO2, etc.). Therefore, chlorite geothermometry should be used with caution and only in combination with alternative methods of estimating paleotemperatures.

Chlorite and corrensite are common clay minerals in lacustrine mudrocks from the Devonian Orcadian Basin, Scotland. The relationship of their occurrence to vitrinite reflectance data demonstrate that they are authigenic minerals, formed during burial diagenesis/metamorphism at temperatures of ≥120°C. Whole rock mineralogical and chemical analyses show that chlorite authigenesis occurred by reactions between the detrital dioctahedral clay mineral assemblage and dolomite that was formed under early evaporitic conditions in the lacustrine environment.

XRD and electron microprobe analyses indicate that phases intermediate between corrensite and chlorite are probably mixed-layer chlorite/corrensite with a tendency towards segregation of layer types. Chemically, the conversion of corrensite to chlorite involves an increase in Al for Si substitution in tetrahedral sites, but there is no change in the Fe/Mg ratio of octahedral cations. There is also no relationship of mixed-layer proportions to paleotemperature; only a general paleotemperature interval of approximately 120° to 260°C in which a range of phases between corrensite and chlorite occurs. Chlorite polytypes are exclusively IIb, indicating the formation of this polytype at diagenetic temperatures.

The occurrence of corrensite and Mg-rich chlorite in evaporite and carbonate successions is probably a reliable indicator of diagenetic alteration at temperatures of ≥ 100°C. Burial diagenetic reactions between dioctahedral clay minerals and Mg-rich carbonates may possibly explain many occurrences of corrensite and Mg-rich chlorite in such rocks.

Since the initial studies of chlorite polytypes, it has been suggested that the stability of the various polytypes may be a function of the temperature at which the mineral formed; however, few studies have been performed in which polytypes of chlorite in a specific suite of samples have been determined and correlated with temperature. A review of the reported occurrences of type I chlorite indicates that other factors, including grain size of the host rock, may be at least as important as temperature in controlling the stability of these polytypes. Results of systematic studies in areas of diagenesis and very low-grade metamorphism suggest that type-II chlorite is stable at temperatures well below 200°C and that it can form as the initial chlorite phase without passing through any intermediate polytypic stages. The conditions under which type-I polytypes occur are somewhat restricted, and cognizance of those restrictions will help to direct future studies of chlorite polytype transformations. These studies should focus on the structural details of polytype transformations; on the relationship of polytype stability to pressure, composition and kinetics; and on experimental calibration of the transformations.