Earth-Science Reviews 158 (2016) 31–50

Contents lists available at ScienceDirect

Earth-Science Reviews

journal homepage: www.elsevier.com/locate/earscirev

Invited review

Clay mineralogy and unconventional hydrocarbon shale reservoirs in

the USA. I. Occurrence and interpretation of mixed-layer R3 ordered
illite/smectite
M.J. Wilson a,b,⁎, M.V. Shaldybin b, L. Wilson c
a
James Hutton Institute, Aberdeen, Scotland, UK
b
Tomsk Polytechnic University, Tomsk, Russia
c
Corex (UK) Ltd, Aberdeen, Scotland, UK

a r t i c l e i n f o a b s t r a c t

Article history: The mineralogy of many of the major unconventional hydrocarbon shale reservoirs in the USA, which span prac-
Received 29 January 2016 tically the whole spectrum of Phanerozoic time, is reviewed from a survey of relevant published literature. This
Received in revised form 14 April 2016 survey reveals that there is a remarkable uniformity in the mineralogy of these shales, both with regard to non-
Accepted 18 April 2016
clay minerals but particularly to the clay minerals. It was found that the clay mineralogy of practically all of the
Available online 19 April 2016
shale reservoirs older than the Upper Cretaceous are dominated by illitic clays, both in discrete form and as
Keywords:
illite-dominated, mixed-layer, illite-smectite (I/S). The layer stacking arrangement of the latter is of the long-
Shale range type described as R3, such that every smectite layer tends to be preceded and succeeded by three illite
Unconventional Reservoirs layers in a sequence like IIISIIIS. Such material is conventionally interpreted (a) as having formed from a smectite
Clay Mineralogy precursor, (b) as existing in MacEwan-type crystallites consisting of about 5 to 15 unit layers in thickness where
Illite there is three-dimensional regularity across the smectite interlayers, and (c) as having interlayers of a truly smec-
Fundamental Particles titic character. Using evidence from the fundamental particle concept of Nadeau et al. (1984b) this interpretation
Porosity is rejected. Instead, it is proposed that R3-type I/S (a) forms de novo, crystallizing from pore waters of appropriate
Permeability
chemical composition in a particular pressure and temperature stability field, as it does in conventional sand-
stone reservoirs, (b) consists primarily of thin illite crystallites or crystals b 50 Å in thickness, and (c) that the
“smectite” interlayers can be accounted for by the ability of such thin illite stacks, which have no three-
dimensional register between the fundamental particles when sedimented onto glass slides, to adsorb ethylene
glycol between the particles so leading to a false diagnosis of “smectite”. This interpretation could have major
consequences on the physicochemical properties of the shale, a matter that is examined more closely in the sec-
ond part of this review.
© 2016 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2. Review of the mineralogy of selected unconventional hydrocarbon reservoirs in the USA . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.1. Cenozoic shales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.1.1. Monterey shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.1.2. Gulf Coast shales. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.2. Upper Cretaceous shales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.2.1. Eagle Ford shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.2.2. Niobrara-Pierre shales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.2.3. Mancos shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.2.4. Tuscaloosa shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.3. Upper Jurassic shales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.3.1. Haynesville-Bossier shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.4. Permian Basin shales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.5. Mississippian shales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

⁎ Corresponding author.

http://dx.doi.org/10.1016/j.earscirev.2016.04.004
0012-8252/© 2016 Elsevier B.V. All rights reserved.
32 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50

2.5.1. Barnett shale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

2.5.2. Fayetteville shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.5.3. Woodford shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.5.4. Bakken shale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.5.5. New Albany shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
2.6. Upper Devonian shales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.6.1. Antrim shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.6.2. Chattanooga shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.7. Middle Devonian shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.7.1. Marcellus shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.8. Ordovician shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.8.1. Utica shale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
2.9. Cambrian shale. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.9.1. Conasauga shale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.1. Nature of illite and mixed-layer I/S in unconventional hydrocarbon reservoirs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.1.1. Conventional wisdom. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.1.2. Do mixed-layer I/S clays always form from a smectite precursor in sedimentary sequences? . . . . . . . . . . . . . . . . . . . . 45
3.1.3. Do the layer stacking arrangements calculated for mixed-layer I/S XRD patterns reflect physical reality? . . . . . . . . . . . . . . 46
3.1.4. Is the smectite identified in mixed-layer I/S really smectite? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

1. Introduction
Designation of a hydrocarbon reservoir as unconventional empha-
sizes the fact that the rock acts as a source, store/reservoir and seal of
the hydrocarbons it contains. In contrast, the conventional reservoir
usually involves the accumulation of hydrocarbons after their genera-
tion and migration from a different and often distant source rock, as
well as being sealed by an impermeable cap rock belonging to another
lithological formation. An unconventional hydrocarbon reservoir may,
therefore, be regarded as being self-sourced and of a highly imperme-
able nature, usually requiring stimulation through induced hydraulic
fracturing if it is to be commercially viable.
Unconventional hydrocarbon reservoirs are often described as
“shales” but the meaning of this term varies according to the context
in which it is used. For the geologist the defining characteristics of
shale are that it is a sedimentary rock, fine-grained and fissile or lami-
nated. A fairly comprehensive definition of shale is that of the
American Geological Institute (Bates and Jackson, 1980). “A fine-
grained detrital sedimentary rock formed by consolidation (especially
compression) of clay, silt or mud. It is characterized by a finely lami-
nated structure, which imparts a fissility approximately parallel to the
bedding along which the rock breaks readily into thin layers,…., and
by an appreciable content of clay minerals and detrital quartz; a thinly
laminated or fissile claystone, siltstone or mudstone. It normally con-
tains at least 50% silt with 35% clay or fine mica fraction and 15% chem-
ical or authigenic material.” In the petroleum industry, however, the
term “shale” is used in a very much broader sense and may refer to sed-
imentary rocks which are not fissile, laminated or particularly fine-
grained, as well as those rocks falling within the above geological defini-
tion. For example, the Bowland shale in the UK is a formation of Missis-
sippian age and contains a wide range of lithologies, including
calcareous mudstones, siltstones, turbiditic packstones and even sand-
stones (Clark et al., 2014). Such heterogeneity in nominal shale forma-
tions is also typical of gas shale formations in the USA. Thus, the clay
content of different lithologies of the well-known Barnett Shale varies
from 8 to 48% in siliceous mudstones, 7 to 34% in calcareous mudstones
and 8 to 24% in calcareous, turbiditic packstones (Loucks and Ruppel,
2007), and with regard to the Utica Shale play in New York State, ac-
cording to one geologist it should be more appropriately referred to as
the “Utica shale and associated organic-rich calcareous shale and inter-
bedded limestone and shale play.” Such lithological heterogeneity must
be borne in mind when considering the possible influence of clay min-
eralogy on hydrocarbon exploitation.
It is generally agreed that it is important to characterize the mineral-
ogy of the bulk rock when attempting to evaluate the potential quality
of unconventional hydrocarbon reservoir rocks, irrespective of whether
the mineralogical data have been obtained from logs, cores or cuttings.
Mineralogy is known to impinge upon a variety of petrophysical param-
eters including, for example, porosity, permeability, water saturation, as
well as attributes related to rock strength such as Young’s modulus and
Poisson’s ratio which are crucial for optimizing the potential of the for-
mation for hydraulic fracture stimulation. A recent contribution to the
influence of mineralogy on the quality of unconventional reservoirs de-
veloped a classification for organic mudstones, based on ternary plots of
the normalized contents of clay minerals, carbonate minerals and sili-
cate minerals, as estimated from geochemical logs (Gamero-Diaz et al.,
2013). This classification subdivides organic mudstones into 16 differ-
ent categories occupying separate areas on the ternary plot and was
considered to provide a qualitative means of visualizing the relationship
between overall bulk mineralogy of the rock and indicators of reservoir
and completion qualities. It was found that there was a strong correla-
tion between bulk mineralogy and completion quality, based on indica-
tors such as minimum closure stress and mineral brittleness index.
There was also a good correlation between mineralogy and reservoir
quality, based on parameters such as effective porosity, matrix perme-
ability and hydrocarbon saturation, although the correlation was not
as strong as that between mineralogy and completion quality.
However, the mineralogy of unconventional hydrocarbon shale res-
ervoirs has been less extensively characterized than conventional sand-
stone reservoirs, at least by core analyses using standard analytical
techniques such as X-Ray Diffraction (XRD) and Scanning Electron Mi-
croscopy (SEM). For the unconventional reservoirs, the usual industry
practice is to use the mineralogical analyses of selected samples that
are available to calibrate the petrophysical data acquired by downhole
logging tools. In this way, quantitative (or semi-quantitative) mineral-
ogical analyses are recorded over the complete drilled stratigraphic se-
quence. These data are regarded as extremely important as “the relative
concentrations of the (mineral) constituents have the potential to make
or break a potential resource play” (Alexander et al., 2011).
In this paper, the mineralogy, and particularly the clay mineralogy of
a wide variety of unconventional reservoirs in the USA is reviewed, with
the broad aim of trying to establish a better understanding of the rela-
tionship between individual clay minerals and petrophysical properties,
 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
particularly those properties that could relate to the commercial exploi-
tation of the hydrocarbon resource in the reservoir rock. The informa-
tion has been compiled from a variety of sources, including peer-
reviewed papers in the scientific journals, conference presentations,
university theses and occasionally commercial literature. Such a broad
approach was necessitated by the rather fragmentary information avail-
able, particularly on the clay mineralogy, of many of the shale sequences
investigated. The data were interpreted using the fundamental particle
concept for mixed-layer clay minerals of Nadeau et al. (1984b) rather
than the traditional view of these clays consisting of MacEwan-type
crystallites made up of about 5 to 15 unit layers in thickness where
there is three-dimensional regularity across the smectite interlayers.
Review will be arranged in inverse chronological order of the reservoir
rock, namely starting with the youngest formations and ending with
the oldest ones.
2. Review of the mineralogy of selected unconventional hydrocar-
bon reservoirs in the USA
The United States has by far the largest unconventional hydrocarbon
shale resources in the world that are currently commercially productive,
as well as a huge lead in the technology of their exploitation. Building on
research into shale gas conducted between partnerships of public-
funded and private organizations over several decades, commercial
shale gas production was successfully initiated from the Barnett shale
in Texas in 1998. The principal entities involved in this enterprise
were the Department of Energy (DOE) and the Gas Research Institute
(GRI) on the public-funded side and Mitchell Energy in the private sec-
tor (Trembath et al., 2012). This success prompted an explosion of activ-
ity in other potentially commercial shale prospects throughout the USA.
Boyer et al. (2011) listed the following shales as being of particular
importance from a commercial point of view, namely the Marcellus,
Barnett, Haynesville-Bossier, Fayetteville, Woodford, Antrim, Eagle
Fig. 1. Current and prospective unconventional hydrocarbon shale reservoirs in the United States (excluding Alaska). Gulf Coast shales occur both offshore and onshore. Adapted from the
Energy Information Administration (2011).
33
Ford and New Albany Shales. To these may be added the Utica, Bakken,
Pierre-Niobrara, Mancos, Conasauga, Tuscaloosa, Gulf Coast and Monte-
rey shales, all of which represent unconventional plays of great promise.
Fig. 1 shows the distribution of these shales in continental United
States and Table 1 shows their arrangement in chronological order. It
can be seen that the shale oil/gas unconventional reservoirs range
across practically the whole spectrum of Phanerozoic time, from the
Cambrian period (540 ma) in the lower Palaeozoic era to the late Mio-
cene (6 ma) or even Pleistocene epochs in the Cenozoic era. The general
nature and mineralogy of each of these unconventional reservoirs will
now be reviewed in reverse chronological order beginning with those
of youngest age.
2.1. Cenozoic shales
2.1.1. Monterey shale
The Monterey Formation represents the main source rock for hydro-
carbons in the various sedimentary basins of California and comprises a
thick sequence of marine sediments of Cenozoic age, largely consisting
of siliceous rocks, some of which are clay-rich, as well as calcareous
shales and organic-rich shales. The siliceous rocks comprise a variety
of different lithological types, including porcelanites, cherts and diato-
mites. Dolomites and phosphate-rich rocks also occur at various points
within the formation sequence.
Mineralogically, the Monterey Formation is rather unusual in that
much of the stratigraphic sequences in the various sedimentary basins
in which it occurs are made up of different forms of silica, namely
opal-A, opal-CT and quartz. It is generally accepted, following the
pioneering work of Murata and Larsen (1975) and Murata et al.
(1977), that the abundance of these silica phases is related to burial
depth and temperature.
The nature of the clay minerals in the siliceous rocks of the Monterey
Formation has been well-studied by Compton (1991). It was found that
34 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
Table 1
Major unconventional hydrocarbon shale reservoirs in the USA showing stratigraphy, age and estimated gas reserves in trillion cubic feet (tcf). (nr = not reported).
Shale formation Stratigraphic age
Monterey Cenozoic
Gulf Coast Cenozoic
Eagle Ford Upper Cretaceous
Niobrara/Pierre (Colorado) Upper Cretaceous
Mancos Upper Cretaceous
Tuscaloosa Upper Cretaceous
Haynesville-Bossier Upper Jurassic
Barnett Mississippian
Fayetteville Mississippian
Woodford (Anadarko) Late Devonian-Early Mississippian
Woodford (Arkoma) Late Devonian-Early Mississippian
Bakken Late Devonian-Early Mississippian
New Albany Late Devonian-Early Mississippian
Antrim Upper Devonian
Chattanooga Upper Devonian
Marcellus Middle Devonian
Utica Middle Ordovician
Conasauga Cambrian
the clay mineralogy of the bulk rock in the three sequences examined
consisted largely of mixed-layer illite/smectite (I/S), which could be
concentrated and examined in more detail in the b 0.1 μm fraction. A
randomly ordered mixed-layer structure (termed R0) was found in 37
out of the 41 samples examined, with only 4 samples showing a diffrac-
tion profile indicative of an ordered structure (termed R1) where the
stacking arrangement is such that every smectite layer tends to be pre-
ceded and succeeded by an illite layer in a sequence like ISISIS. Many of
the mixed-layer clays were highly smectitic and were considered to
have formed from a purely smectitic precursor derived from the alter-
ation of vitreous volcanic ash, for which there is abundant evidence in
the Monterey Formation. It was proposed that the smectite became ini-
tially illitized following the dissolution of K-feldspar or possibly mica
minerals at the same time as opal-CT transformed to quartz. In one of
the sequences examined there was little illitization of smectite. In the
other two sequences, however, extensive illitization was identified, al-
though a steady increase in the proportion of illite with depth was not
apparent and R1 stacking arrangements occurred at irregular intervals.
Diagenetic kaolinite and chlorite were also identified in the Monterey
Formation and considered to be derived from smectite where a limited
source of K ions restricted the formation of illite.
2.1.2. Gulf Coast shales
The Gulf Coast shales were deposited throughout the Cenozoic to
Upper Mesozoic eras and are represented by sediments spanning the
Cretaceous, Paleogene and Neogene periods even up to the Pleistocene
epoch. This enormously thick sequence of fine-grained sediments was
the scene of the landmark studies of Hower and his colleagues
(e.g., Perry and Hower, 1970; Perry and Hower, 1972; Hower et al.,
1976) on clay diagenesis which provided evidence for the progressive
illitization of smectite with increasing burial depth and temperature
through a layer-by-layer process, commonly referred to as solid state
transformation (SST). In this process, precursor smectite is converted
to illite, firstly through a random (R0) I/S mixed-layer structure where
the smectite component is predominant, then through a regular (R1)
I/S interstratification, and finally through a special type of long-range
ordering (R3) where every smectite layer is preceded and succeeded
by three layers of illite such that an IIISIIIS... sequence is formed. At
this stage, illite layers predominate and eventually, a pure illite is
formed following further illitization of the remaining smectite
layers. This concept was widely accepted (Środoń, 1999), elaborated
upon with regard to the detailed mechanisms involved (e.g. Altaner
and Ylagen, 1997; Cuadros and Altaner, 1998), and applied to other
fine-grained sedimentary sequences both in the Gulf Coast and else-
where (Bell, 1986; Lindgreen and Hansen, 1991; Lindgreen et al.,
Age, ma Reserves (gas, tcf)
33–1.8 nr
33–1.8 nr
100–68 21
100–68 25
100–68 5
100–68 Nr
164–145 120
359–323 53
359–323 38
372–347 6
372–347 22
372–347 6.7
372–347 86–160
383–359 31–76
383–359 12.2
383–393 354
458–444 780
509–490 Nr
1991; Cetin and Huff, 1995; Dainyak et al., 2006). McCarty et al.
(2008) recently described Gulf Coast shales in which discrete smec-
tite and R0 I/S coexist.
However, others have argued that the illitization of smectite in shale
sequences can best be accounted for by a dissolution/precipitation (D/P)
process, whereby smectite layers dissolve and thin illite precipitates as
parallel domains within the smectite particle. Such a transformation
was proposed for Gulf Coast shales by Peacor and colleagues (e.g. Ahn
and Peacor, 1986; Freed and Peacor, 1992), as well as for other fine-
grained sediment sequences (Dong and Peacor, 1996; Buatier et al.,
1992; Nieto et al., 1996) and it would appear that the SST vs D/P debate
with regard to the illitization of smectite has still not been settled
(Wilson, 2013). Whatever the mechanism involved in the conversion
of smectite to illite, it is widely accepted that the change from random
(R0) to ordered (R1) I/S stacking sequences occurs over a narrow tem-
perature range in any given well (Freed and Peacor, 1989). In the Gulf
Coast well studied by Freed and Peacor (1992) this transition occurred
at burial depths between 2134 to 2438 m at 83 to 88 °C, although
these values may be well-specific. It is found in wells located in onshore
Louisiana that at 2200 m this change corresponds to a geopressured
zone characterized by better fabric orientation, significant reduction in
porosity and permeability and an increase in fluid pressure gradient
(Kim et al., 2001).
Other clay minerals common in minor amounts in Gulf Coast shales
include kaolinite and chlorite and common non-clay minerals include
quartz, calcite, plagioclase and K-feldspar (Pollastro, 1994; Couto-
Anjos, 1986). These minerals represent a combination of detrital and
diagenetic contributions. However, where the mixed-layer I/S mineral
is of the type where the illitic component is the major phase then it is
generally agreed that some kind of smectite precursor is required for
its formation (Freed and Peacor, 1992). In other words, regularly or-
dered R1 and R3 I/S are certainly considered as diagenetic clays, as are
randomly interstratified R0 I/S clays from which they form. But al-
though I/S in Gulf Coast shales are considered to have developed from
a putative smectite precursor, the clay fractions from even the youngest
Pleistocene shales are dominated by R0 I/S (Fig. 2, a and b) and not by a
pure smectite (Totten et al., 2005). It seems likely, therefore, that the
randomly ordered R0 I/S in Gulf Coast shales is a detrital product de-
rived from weathered Cretaceous and Paleocene rocks and soils derived
from them in the Mississippi catchment, transported by the Mississippi
River, deposited in the Gulf of Mexico and subsequently buried. This in-
deed was the conclusion of Hoffman (1979) and suggests the possibility
that the formation of I/S in shales may not necessarily require a purely
smectitic precursor as is often assumed, an issue that will be addressed
later in this review.
 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50 35

Fig. 2. (a) XRD (CuΚα radiation) profiles of illite-rich clay separates from Gulf Coast shales; samples 23 and 28 are of Pleistocene age, 16 is of Miocene age (adapted from Totten et al., 2005).
(b) XRD profiles from smectite-rich clay fractions from Gulf Coast shales; samples 8, 23 and 24 are all of Pleistocene age (adapted from Totten et al., 2005). S = Smectite, I = Illite, K =
Kaolinite, Q = Quartz.

2.2. Upper Cretaceous shales
2.2.1. Eagle Ford shale
The Eagle Ford shale in south western and eastern Texas is of
Cenomanian to Turonian age (89–96 ma) and was deposited in marine,
deep water basins, where a combination of high organic productivity in
the uppermost waters of the basin, a limited supply of siliciclastic mate-
rial, and the deposition of organic matter to an anoxic sea floor ulti-
mately led to the formation of organic-rich, carbonate rocks
containing varying amounts of silicate minerals. In eastern Texas the
Eagle Ford Formation consists predominantly of calcareous and dolo-
mitic shales and marls, laminated to varying degrees, while in south-
western Texas the predominant marls are more carbonate-rich, and
inter-bedded limestones occur more often. According to Jennings and
Antia (2015), the average carbonate, clay and quartz contents of the
Eagle Ford shale in eastern Texas are 41, 36 and 19% respectively,
while in south western Texas the values are 56, 14 and 15% respectively.
The clay mineralogy in both areas is dominated by illite and illitic
mixed-layer minerals (I/S), although no indication is given of the pro-
portions of the components or of the type of layer stacking. The electron
microscope observations of Jennings and Antia (2015) show that many
of the minerals are authigenic, including quartz which replaces
coccolith fragments, kaolinite which fills the chambers of foraminifera,
and chlorite which replaces kaolinite. As noted previously, mixed-
layer I/S is also considered as a diagenetic clay and it is likely that at
least some illite in the Eagle Ford shale is of a similar origin.
The results of a preliminary study (Elston, 2014) of the mineralogy of
ten Eagle Ford core samples from south-eastern Texas are generally
consistent with those of Jennings and Antia (2015). The main clay min-
eral identified was illite, ranging from 7 to 47% of the bulk material,
while kaolinite was found in only one sample. Non-clay minerals in-
cluded calcite (11–80%), quartz (2–21%), pyrite (2–11%), dolomite (2–
3% in 2 samples) and albite (0–30%). The data showed a strong positive
correlation between illite and total organic carbon (TOC) and a weak
negative one between calcite and TOC. The results of Mullen (2010)
for the Eagle Ford shale in south-west Texas are also consistent with
36 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50

those of Jennings and Antia (2015), both in their general assessment of

the relative proportions of carbonate minerals, silicate minerals and clay
minerals (Fig. 3a) as well as the mineralogy of the clay fraction. With re-
gard to the latter, it is evident that illitic clays greatly predominate, ei-
ther as discrete illite or mixed-layer I/S (Fig. 3b). Apparently, there is
no record of the occurrence of smectite as a discrete phase or of the
type of I/S interstratifications in the Eagle Ford shale so that at this
stage it is not clear whether or not illite has formed through the conver-
sion of precursor smectitic material according to the classic pattern
established in the Gulf Coast shales. However, the Eagle Ford Formation
does contain many volcanic ash beds and much volcanoclastic material
which, according to the Cenomanian palaeogeography of the area de-
scribed by Jennings and Antia (2015), derive from volcanic activity to
the north east. Thus derivation through illitization of a smectite precur-
sor may be regarded as likely and certainly cannot be discounted.

2.2.2. Niobrara-Pierre shales

The Niobrara shale was deposited under marine conditions in the
Western Interior Cretaceous (WIC) Seaway which divided the North
American continent from north-to-south during late Cretaceous times.
Although the Niobrara Formation is often referred to as a “shale”, it is
made up of rocks containing variable amounts of siliciclastic and car-
bonate materials, thus encompassing a complete spectrum ranging
from true clay-rich shales to carbonate-rich chalks with intermediate
members described as calcareous shales, marls and argillaceous chalk.
These intermediate members make up the major part (~80%) of the Ni-
obrara Formation with the end-member shales and chalks making up
the remainder. In general, the calcareous rocks occur to the eastern
side of the Seaway whereas the siliciclastic materials preponderate on
the west side.
The dominant clays in the Niobrara Formation are mixed-layer I/S
with varying proportions of the two components ranging from highly
smectitic randomly interstratified I/S (R0) in samples from shallow Fig. 4. Depth vs percent (%) illite layers and ordering (R) of mixed-layer illite/smectite (I/
depths to highly illitic ordered I/S (R1, R3) in the deeper samples as S) and depth vs temperature for sandstone and shale samples from the Niobrara
Formation (adapted after Pollastro, 1993). Note the close correspondence between
seen in Fig. 4 (Pollastro, 1985). These observations were interpreted as
sandstone and shale samples.
indicating progressive illitization of the more smectitic mixed-layer
clays, similar to that found in the Gulf Coast shales by Hower et al.
(1976), albeit by an alternative reaction suggested by Boles and similar process. Pollastro (1993) also showed that comparison of the
Franks (1979). Although no discrete smectite was identified in the Nio- illitization of random mixed-layer I/S in shales and sandstones of the Ni-
brara Formation in this study, recent work by Bertog (2013) has identi- obrara Formation were remarkably similar, suggesting a common ge-
fied bentonites in these rocks indicating the influence of volcanic netic process.
activity on the western side of the Seaway and suggesting the possibility The Pierre shale directly overlies the Niobrara Formation, its stratig-
of a precursor smectite for the illitic clays. It is of interest that the mor- raphy, composition, mineralogy and geochemistry having been inten-
phology of the highly illitic I/S in the Niobrara Formation was found to sively studied by the USGS. It consists primarily of several hundred
be in the form of fibres and laths, rather similar to the morphology of feet of marine shale, with subordinate marl in the eastern part of the
authigenic illite in reservoir sandstones. Fig. 4 also seems to indicate WIC seaway, but further west is represented by several thousand feet
that the illitization in both shales and sandstones is occurring by a of volcanogenic, mainly non-marine, sediments (Schultz et al., 1980).

Fig. 3. Mineralogical composition of (a) bulk material and (b) clay (b2 μm) fraction of the Eagle Ford shale (adapted after Mullen, 2010).
 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
These sequences are separated by transgressive, marine sandstones and
siltstones. The Pierre shale contains many bentonite layers, particularly
in the Sharon Springs unit which occurs widely from the Dakotas to
Kansas (Gill et al., 1972). It was early recognized that the Pierre shale
was dominated by fine-grained minerals, on average about 70% of the
bulk rock consisting of material b4.0 μm in size (Tourtelot, 1962). The
clay fraction is dominated by illitic minerals, mixed-layer I/S comprising
about 70% of this fraction and illite about 16% (Schultz et al., 1980). The
mixed-layer clay is usually randomly stacked (R0), containing 20 to 60%
illite layers. The non-clay minerals in the bulk rock include quartz,
cristobalite (from radiolaria), plagioclase and K-feldspars, calcite and
dolomite. Gypsum, jarosite, pyrite, zeolites and siderite occur rarely.
The bentonite bands are characterized by a dominance of montmoril-
lonite (N 90%) and a lack of illite, either in discrete or interstratified
forms. Non-clay minerals include quartz and more calcic plagioclase
feldspar.
2.2.3. Mancos shale
The Mancos shale is of Upper Cretaceous age and is the source rock
for many of the conventional hydrocarbon reservoirs that are closely as-
sociated with it (Ridgley et al., 2013). It extends over a huge area in the
southern Rocky Mountains and the Colorado Plateau and was deposited
on the western side of the WIC Seaway under marine conditions, vary-
ing in thickness from 1500 m in the west to 60 m in the east. In the
upper part of its sequence, the Mancos shale is the time equivalent of
the Niobrara limestone, whereas the lower part of the sequence is the
equivalent of the Pierre shale (Cobban and Reeside, 1952).
For the most part, the Mancos shale is usually non-calcareous and si-
liceous, and is often rich in organic matter with a clay-rich or silty tex-
ture. In places the shale may become calcareous, containing carbonate
and gypsum concretions, while bentonites deriving from volcanic activ-
ity to the west frequently occur throughout the entire shale sequence
(Nadeau and Reynolds, 1981).
The clay mineralogy of the Mancos shale was intensively studied
by Nadeau and Reynolds (1981), including bentonite samples and
shales immediately subjacent to the bentonite beds. Within the
shale sequence in general, the major clay mineral was found to be
randomly ordered (R0) I/S with 20 to 60% illite layers. In the paired
bentonite/shale samples it was found that in most pairs I/S was
either both random or both ordered, although in a small number of
samples, I/S was ordered in the bentonites and random in the shales.
Extreme cases were found where shales and bentonites from a single
outcrop contained a variety of I/S clays, ranging from pure smectite
in calcareous shales to ordered I/S with N50% illite layers in non-
calcareous shales. The results were interpreted as generally
supporting the classical model of illitization of smectite through
burial metamorphism, although this pattern could be modified by
the thermal effects of nearby intrusive igneous bodies as well as
other factors related to pore water chemistry.
2.2.4. Tuscaloosa shale
This shale was deposited at about the same time as the Eagle Ford
shale in Texas and is sometimes considered as its eastward extension
into Louisiana and western Mississippi. The Tuscaloosa shale ranges
between 500 to 1000 ft in thickness, formed under marine condi-
tions and lies at a depth of 10,000 to 15,000 ft. It is estimated to con-
tain a reserve of 7 billion barrels of oil. The composition of the
Tuscaloosa shale is much more clay-rich than that of the eagle Ford
shale. A meeting of the Goodrich Petroleum Corporation on March
31, 2014 held to assess the economic potential of the Tuscaloosa
shale cited clay content ranging between 50 to 65%, quartz content
20 to 32%, calcite content 12 to 22% and TOC from 1 to 1.8%. So far
there appear to be no reports of the clay mineralogy of the Tusca-
loosa shale in the public domain.
37
2.3. Upper Jurassic shales
2.3.1. Haynesville-Bossier shale
These shales occur in eastern Texas, north-western Louisiana and
south western Arkansas and are made up of heterogeneous sequences
that stem from their deposition in a wide range of environments.
These have been fully described by Cicero et al. (2010) who split the
Haynesville and the overlying Bossier shale into upper and lower
units. The Lower Haynesville consists of a carbonate-dominated system
to the west, largely consisting of shallow-marine platform carbonates
and patch reefs, whereas in the east a clastic sedimentation system is
preponderant related to input from the ancestral Mississippi River. A
transition zone occurs between these two systems where thick
organic-rich shales were deposited in deep water under anoxic condi-
tions. This arrangement essentially continued into the Upper
Haynesville although with an increasing supply of clastic material to
the east, the top of this unit being marked by a major flooding surface
related to a rise in sea level. Lower Bossier time is characterized by a
westward migration of the transition zone between the carbonate-
dominated and clastic-dominated systems. Shale deposition becomes
more prominent in the basin as a whole but with decreasing amounts
of organic matter. The Upper Bossier unit continues to see widespread
shale deposition, but there are also periods of carbonate deposition
such that mixed carbonate-siliciclastic sequences are quite common.
There is a paucity of mineralogical information for the Haynesville-
Bossier shale as assessed directly by XRD analysis of cores and cuttings.
A high quality XRD pattern of the clay fraction (Figure 5) of a shale sam-
ple of this formation is entirely consistent with a predominance of R3
ordered mixed-layer I/S. However, much information has been derived
indirectly from a chemostratigraphic approach based upon analysis and
geochemical interpretation of major and trace elements. Thus, Sano
et al. (2013) used this approach to define four different units within
the Haynesville shale based on geochemical variations and calibration
by XRD data. It is of interest to note that the Haynesville shale shows a
distribution of Rb/K2O ratios which are significantly higher than those
found in the formations below and above it and that this distribution
is similar to the variations found for illitic minerals by XRD. Also plagio-
clase feldspar content is related to % Na2O, suggesting that the feldspar
mineral is primarily albite, quartz is related to % SiO2, total clays to %
Al2O3 and parts per million (ppm) thorium, and the ratio quartz/total
clay by SiO2/Al2O3. Other minerals determined by geochemical analysis
and confirmed by XRD include calcite, dolomite, chlorite and pyrite. It is
concluded that the clay mineralogy of the Haynesville-Bossier shale is
predominantly illitic.
2.4. Permian Basin shales
The Permian Basin occurs in the western part of Texas and the south
eastern part of New Mexico and is made up of several component sub-
basins, the two largest being the Midland and the Delaware Basins.
There are a number of lithological units described as shales in these
two basins but the most important members in terms of hydrocarbon
production are the Spraberry, Wolfcamp and Bone Spring. The
Wolfcamp formation is represented, and termed as such, in both the
Midland and Delaware Basins, whereas the overlying Spraberry forma-
tion occurs in the Midland Basin while the Bone Spring formation occurs
in the Delaware Basin. However, both these latter two formations are
the time equivalents of each other. The Wolfcamp formation consists
mostly of shale with some interbedded limestones and calcareous sand-
stones. It has an average thickness of ~2000 ft in the Delaware basin and
is somewhat thinner in the Midland Basin. The Spraberry formation has
a mean thickness of about 1500 ft and is made up of black shales and
calcareous mudstones interbedded with fine sandstone and coarse
siltstone, all being characterized by very low porosity and matrix per-
meability. In the Delaware Basin the Bone Spring formation consists to
a large extent of cyclic deposits of carbonates and sandstones but at
38 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
Fig. 5. XRD pattern of the glycolated clay fraction (b2.0 μm) of a sample of the Haynesville-Bossier Shale showing a predominance R3-ordered mixed-layer I/S. Note the asymmetry of the
10 Å peak towards the low angle side. (Courtesy of Dr. Ruarri Day-Stirrat).
the top of the formation impermeable shales with dark and carbona-
ceous shaly siltstones predominate. Both basins accumulated large
amounts of clastic sediment during Permian times, with deposition in
the Delaware basin occurring at considerably greater depths.
The TOC content of the Wolfcamp shale typically ranges between
2.74 to 7.24% and it is not over-pressured. It has an average porosity of
about 7% and vitrinite reflectance of 0.95–0.97% (Craft, 2013). It is also
reported to have a mineralogical composition favourable to brittleness
with cores showing a typical composition of quartz 36%, carbonate
26% and clay 25% The mineralogy of 7 samples from 2 wells of the
Wolfcamp shale was studied by WoldeGabriel et al. (undated) as part
of a project to determine its integrity as a seal in a CO2 flooded environ-
ment. The variation of the bulk mineralogy of the injector well as deter-
mined by XRD was assessed as follows: Quartz 19.3–24.6%, Illite 38.1–
65.6%, Kaolinite/Chlorite 1.1–3.6%, Pyrite 1.4–3.6%, Calcite 0.2–3.3%.
Samples from the injector/producer well yielded analyses of illite
34.8–60.9%, quartz 18.9–24.1%, kaolinite/chlorite 1.4–2.1%, calcite 0.3–
5.7 %, and pyrite 1.9–3.4%. Other minerals recorded in minor amounts
included K-feldspar, plagioclase feldspar and dolomite. Smectite was
also recorded although it was not clear whether the clay mineral was
in a discrete or mixed-layer form.
Mineralogical data on the Spraberry/Bone Spring formations is
sparse although Curtis (2013) records a mineralogical analysis for the
Avalon shale unit, which occurs at the top of these formations. Bulk
analysis for this shale was Quartz 35–60%, Carbonate 1 to 13% and
Clay 18–44%.The clays in the “Wolfberry” mudrocks of the Midland
Basin consist predominantly of mixed-layer R3-ordered I/S, with
mixtures of discrete illite and chlorite (Hamlin and Baumgardner,
2012).
2.5. Mississippian shales
2.5.1. Barnett shale
The Barnett shale is part of the Fort Worth Basin in north central
Texas and is a classic example of an unconventional shale gas reservoir.
Its successful exploitation heralded the rise of the shale gas industry in
the United States to the pre-eminent heights which this industry now
occupies. The lithofacies and depositional environment of the Barnett
shale have been well-described by Loucks and Ruppel (2007). The
shale is comprised of a lower and an upper unit, separated in the east
of the basin by an argillaceous limestone, and is composed largely of si-
liceous mudstones, marls and skeletal packstones. The siliceous mud-
stones are the predominant lithofacies in both Barnett shale units. On
the basis of sedimentary structures, facies and organic geochemistry,
as well as on other evidence, Loucks and Ruppel (2007) concluded
that the Barnett shale was deposited in a deepwater, dysaerobic to an-
aerobic basin, fed largely by settling of fine suspensions and by density
currents.
According to Montgomery et al. (2005), the Barnett shale is consid-
ered to be rich in silica (35–50%) and poor in clay minerals (b 35%). The
siliceous mudstone facies has an average composition of quartz (45%),
calcite + dolomite (8%), feldspar (5%), pyrite (5%), siderite (3%), organic
(matter (5%) and illite with minor smectite (27%). It is not clear from
this description whether the smectite occurs as a discrete phase or as
part of a mixed-layer illitic structure, but it is apparent that the clay min-
eralogy of the Barnett shale is overwhelmingly illitic.
2.5.2. Fayetteville shale
This black, marine, organic-rich shale produces natural gas from the
central portion of the Arkoma basin and underlies much of northern Ar-
kansas and eastern Oklahoma. Stratigraphically, the Fayetteville shale
conformably rests on the Chesterian Hindsville Limestone. The Fayette-
ville shale is made up of three units, a basal part consisting of black
shale, a middle part containing sandstone and an upper part where in-
terbedded shales and limestones repeat in a rhythmic pattern. Overly-
ing the Fayetteville shale is the Pitkin Limestone, containing thin
lenses of shale, above which is the Imo Formation consisting largely of
black, pyritic fossiliferous shales.
The Fayetteville shale is associated with a marine, transgressive, de-
positional phase where water depth was gradually increasing, such that
dysaerobic to anaerobic conditions became prevalent. Eventually, a re-
gressive phase became predominant, during which the Pitkin limestone
was deposited in a shore face, shallow-water environment, to be
succeeded by low-energy depositional conditions where inputs of
fine-grained terrigenous material prevailed, thus forming the Imo
shale. (Sutherland, 1988; Sanders et al., 1979).
According to Weaver (1958), the Chesterian sediments of central
USA, which would include the Fayetteville shale, contain both illite
and montmorillonite, although this paper was written before the nature
of mixed-layer I/S was clarified by Reynolds and Hower (1970). A more
 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
recent investigation by Alese (2012) presents diffraction patterns for
the clay fractions of the Fayetteville and Imo shales and shows them
to be largely made up of I/S, probably in an R3 stacking arrangement,
and chlorite. Elgamati et al. (2011) also present an XRD pattern of the
Fayetteville shale showing a predominance of I/S and quartz (Fig. 6).
2.5.3. Woodford shale
This shale formation covers a large part of Oklahoma and, although
predominantly mudstone, encompasses a complex suite of lithofacies.
These have been analysed in some detail by Caldwell (2011) in the
Anadarko Basin of west central Oklahoma, where the mudstones are de-
scribed as siliceous (75% quartz, 15% clay), argillaceous, with about 40%
of discrete illite, and up to 20% pyrite. The lowest part of the Woodford
shale consists of organic-poor, argillaceous mudstone which then be-
comes more silty, silicic and organic-rich higher up in the stratigraphic
column. This trend continues into the upper part of the Woodford
shale which becomes even more silicic towards the top of the sequence.
The Woodford shale also occurs in west central Texas and south-east
New Mexico (Comer, 1991) where it occurs in two lithofacies, namely
black, marine, organic-rich shales and siltstones. The black shale is the
most widely distributed rock type and was deposited in a deep basin
under anaerobic conditions. The siltstone formed in deep basin and
proximal shelf settings and results from turbiditic bottom flows in a
dysaerobic environment.
Al Khammali (2015) investigated the mineralogy of the clay frac-
tions of the Woodford shale in Oklahoma. Discrete illite was the domi-
nant clay mineral, with smaller amounts of mixed-layer I/S, chlorite,
and kaolinite. Non-clay minerals included quartz, dolomite, calcite, py-
rite, feldspar (albite and microcline), and apatite. Whittington (2009)
studied the mineralogy of the clay fraction of the Woodford shale in
the Arbuckle Mountains of Oklahoma and found kaolinite and mixed-
layer I/S with a very low content of smectite (b5%) suggesting long-
range order of the R3 type. Higher illite crystallinity values also sug-
gested diagenesis to anchizone conditions.
In Texas and New Mexico, Comer (1991) found abundant illite in the
black shale lithofacies with volume percentages ranging between 34
and 74% and averaging 59%.The illite in the 1.0–2.0 μm fraction was
found to be of terrigenous origin, as shown by its Rb/Sr age of 540 ma,
whereas the illite in the fine clay fraction (b 0.2 μm) was considered to
be diagenetic. Other material of authigenic origin included dolomite, py-
rite, secondary silica, glauconite, calcite, anhydrite and phosphatic
ooids. The fine fraction in the siltstones also contained abundant illite,
often in wispy laminae, and sometimes made up high proportions of
the rock.
2.5.4. Bakken shale
This shale occurs in the Williston Basin, an intracratonic basin that
occurs principally in North Dakota and Montana in the USA and in
Fig. 6. XRD profile (CuΚα radiation) of the clay (b2 μm) fraction of the Fayetteville Shale. Upper trace, air dry; lower trace, after glycol (adapted after Elgamati et al., 2011).
39
Saskatchewan in Canada. In North Dakota, the Bakken shale reaches a
maximum thickness of 46 m and consists of a sequence of lower, middle
and upper members (Pitman et al., 2001). The lower member consists of
grey to black, organic-rich shales that are occasionally calcareous, while
the upper member is lithologically similar but has a higher content of
organic matter and is generally less calcareous. The middle member of
the Bakken shale presents a more varied sequence, made up of massive
siltstones and sandstones, interbedded with sub-units of dark grey
shales and siltstones which may be calcareous and/or dolomitic
(Pitman et al., 2001). It is generally accepted that the lower and upper
members of the Bakken Shale were deposited under offshore marine
conditions and that the bottom water environment was anaerobic, as
indicated by the preservation of a high organic matter content as well
as the presence of pyrite. The middle member is also believed to have
been deposited under marine conditions, but in a coastal regime
which was well-oxygenated and occasionally dysaerobic.
Pitman et al. (2001) paid particular attention to the diagenetic his-
tory of the middle member of the Bakken shale. They showed that the
detrital matrix of quartz and K-feldspar was strongly influenced by
the secondary authigenic overgrowth of these minerals, by the crystal-
lization of authigenic carbonates, both calcite and dolomite, and by
the formation of pyritic masses of various sorts. The clay fraction was
found to be dominated by well-crystallized illite with a small proportion
(b5%) of expandable layers (Figure 7) with subsidiary amounts of iron-
rich chlorite. Both clay minerals were interpreted as being of authigenic
origin. Sonnenberg et al. (2010) also indicated that the clay mineralogy
of all three members of the Bakken shale was dominated by illite and a
similar finding was recorded by Ashu (2014) for the clay mineralogy of
the Three Forks Formation which lies directly beneath the lower mem-
ber of the Bakken shale. There is no evidence that discrete smectite has
acted as a precursor in the formation of illite or I/S in these sequences.
2.5.5. New Albany shale
This group of shales was deposited under marine conditions in the
Illinois Basin, which occupies most of Illinois and extends into Indiana
and west Kentucky. It consists of a number of different lithological
shale members including the Blocher, Selmier, Sweetland Creek, Grassy
Creek and Saverton shales, which are separated by sandstone, siltstone
and limestone units. These shales are of various colours, ranging from
black and brownish-black to gray and greenish gray but usually their or-
ganic carbon content is quite high. Thus, organic carbon ranges from 6–
10%, 5–12% and 4–15% in the Blocher, Selmier and Grassy Creek shales
respectively (Cluff et al., 1981). In addition, these shales are often calcar-
eous and pyritic and are usually finely laminated and less often biotur-
bated. They are considered to have been deposited in the various
environments characteristic of a shelf-slope-basin transition, but largely
under dysaerobic to anaerobic conditions (Cluff et al., 1981).
40 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50

Fig. 7. XRD powder pattern of the clay fraction from the Middle Bakken Formation showing the dominance of illite with subsidiary chlorite (adapted after Pitman et al., 2001).

Throughout the various members of the New Albany shale, the main 2.6.2. Chattanooga shale
clay mineral is illite with a small percentage of expandable layers, as is This shale and its stratigraphic equivalents occur in several states in
characteristic of mixed-layer I/S with an R3 type of stacking (Cluff the eastern part of the USA including Tennessee, Kentucky, Ohio and Al-
et al., 1981). There is no evidence that this arrangement has developed abama. The depositional conditions in which the Chattanooga shale ac-
through the illitization of discrete smectite, which has not been identi- cumulated have long remained contentious with Conant and Swanson
fied as a clay fraction component in these rocks. However, Gharrabi (1961) favouring a shelf setting and marine shallow water (~100 ft) or-
and Velde (1995) confirmed the predominance of illite and mixed- igin, while Ettensohn et al. (1988) proposed a euxinic basin model
layer I/S with small smectite contents in the New Albany shale of the Il- where water depth ranged from 100 to 700 ft.
linois Basin but interpreted this as part of the final stages of the smectite Bates and Strahl (1957) described batches of the shale as faintly lam-
to illite transformation. Chlorite occurs in minor amounts and kaolinite inated, chocolate-brown sediment containing occasional nodules and
is usually absent. Silt-sized quartz and feldspar are clearly of detrital or- lenses of pyrite and flakes of mica. Thin sections revealed sand-silt-
igin, whereas calcite, dolomite and pyrite evidently originated through size grains of quartz and feldspar within a matrix of fine organic parti-
authigenesis. cles and clay minerals. XRD and chemical studies estimated the rock
to comprise 22% quartz, 9% feldspar, 31% illite and kaolinite, 22% organic
matter and 11% pyrite, with the remainder consisting of iron-rich and
2.6. Upper Devonian shales heavy minerals. Conant and Swanson (1957) in a preliminary report
on the Chattanooga shale in central Tennessee described the shale as
2.6.1. Antrim shale consisting of two sequences termed the Dowellton and the Gassaway
These marine shales occur in the Michigan Basin. They have a high Members respectively. The Dowellton Member consists of a lower unit
content of organic matter, are highly siliceous and pyritic and have evi- of black shales and an upper unit of gray claystones which towards
dently formed under anoxic conditions. The Antrim shale formation oc- the top of the unit contains a thin bentonite layer about one inch in
cupies about a quarter of the total area of the Michigan Basin and was thickness. The overlying Gassaway Member is described as a nearly ho-
deposited as part of a large Devonian-Mississippian “Black Shale Sea” mogeneous succession of massive black shale consisting of 20–25%
that existed between the Transcontinental arch in the west and the Ap-
palachian Basin in the east (Goodman and Maness, 2008). The Antrim
shale formation is sub-divided into various units, the most important
of which are, in descending chronological order, the Upper Antrim, La-
chine, Paxton and Norwood, all of which have high organic matter con-
tents sometimes reaching up to 20% (Patel, 2014).
The mineralogy of the Antrim shale is somewhat variable depending
upon sample location within the Michigan Basin. Ruotsala (1980) re-
ported that typical Antrim shale was composed of quartz (50–60%), il-
lite (20–35%), kaolinite (5–10%), with minor amounts of chlorite and
pyrite (0–5%). Illite and quartz contents were highest in the deepest
parts of the basin where the shale was thickest. Matthews (1993) also
described the average mineralogy of the Antrim shale (Fig. 8). The illitic
clay in the Antrim shale was characterized in detail by Dong et al.
(1997) using Transmission Electron Microscopy (TEM) and Analyti-
cal Electron Microscopy (AEM) and shown to be mixed-layer I/S
with an illite content of 90%. It is noteworthy that although the
mixed-layer I/S was attributed to the illitization of a precursor smec-
tite, such smectitic material has not been identified in the sediments Fig. 8. Bulk mineralogy of the Antrim Shale Formation showing that the clay minerals are
of the Michigan Basin. dominated by illite (adapted after Matthews, 1993).
 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
quartz, 10% feldspar, 25–30% clay and mica, 10–15% pyrite and 15–20%
organic matter. Leventhal and Hosterman (1982) analysed correlatives
of the Chattanooga shale in Kentucky, Ohio, Virginia and Tennessee
and found an average composition of 30% quartz silt, 65% clay, and 5%
pyrite for the Cleveland Member of the Ohio shale. The clay fraction
was dominated by illitic minerals with 60% illite, 30% illite-smectite
and 10% chlorite. Similar results were recorded for the Chagrin shale
and its equivalent in Tennessee, West Virginia and Ohio. The predomi-
nance of illite and mixed-layer I/S in the clay fraction of the Chattanooga
shale in Alabama is confirmed in a report by Pashin et al. (2011) for the
Alabama Geological Survey.
XRD traces of bulk samples of the Chattanooga shale from boreholes
in Alabama and Tennessee were illustrated by Rheams and Neathery
(1984). The illitic nature of the shale is clearly evident from these traces,
as is the calcareous and dolomitic nature of the shale at a certain depth
in the Alabama borehole (Fig. 9). The broad and asymmetrical nature of
the illite 10 Å peak is strongly suggestive of mixed-layer I/S where the
illite component is predominant. SEM observations identified lath-like
illitic forms bridging open pores within the shale (Fig. 10, a and b).
2.7. Middle Devonian shale
2.7.1. Marcellus shale
This shale was deposited in the Appalachian basin and occurs princi-
pally in New York and Pennsylvania, as well as in contiguous states. It is
a black shale, becoming lighter coloured towards the top of the forma-
tion and is frequently inter-bedded with limestones. The Marcellus
Fig. 9. X-ray diffractograms (CuΚα radiation) of Chattanooga Shale, core hole #8, Madison County, Alabama (adapted after Rheams and Neathery, 1984).
41
shale contains concentrations of iron pyrites and siderite and is thought
to have been deposited in an anoxic marine basin that formed parallel to
the mountain chain and offshore from the continent during the Acadian
orogeny. The sediments were laid down, possibly as distal turbidites, far
from shore at depths of about 500 ft. The Marcellus Formation consists
of two sequences that contain composites of both transgressive and
high stand tracts and generally coarsens upwards. The average TOC con-
tent of much of the Marcellus shale is N 6.5% and frequently is in the 8–
16% range (Wang and Carr, 2013).
The most complete study of the clay mineralogy of the Marcellus
shale is that of Hosterman and Whitlow (1983) who found that illite
is uniformly present in practically all samples and is the most abundant
of the clay minerals ranging from 40 to 90% (Fig. 11 and 12). Mixed-
layer I/S is most abundant in the youngest samples and least abundant
in the oldest. Chlorite is a common component of the Marcellus shale,
averaging about 10 to 15%, while kaolinite only occurs in about 30% of
samples at a maximum level of 20%. A mixed-layer illite-chlorite com-
ponent was also identified indirectly from heating experiments and
was interpreted as having formed from mixed-layer I/S. Both illite and
I/S were also interpreted as being of authigenic so that kaolinite was
the only clay mineral that was considered to be of a detrital nature.
Bulk samples of the Marcellus shale were found to contain detrital
quartz, plagioclase feldspar, K-feldspar and biotite, with quartz being
by far the most abundant mineral. Non-clay authigenic minerals in-
cluded calcite, dolomite, siderite, pyrite, gypsum and other minerals.
Quartz occurs in all samples and calcite and pyrite in most. Other studies
have amply confirmed these generalizations.
42 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
Fig. 10. SEMs showing (a) illite laths bridging pores and associated with platy illite and
nodular pyrite in black Devonian shale at 8317 ft depth in Greene County, Alabama and
(b) illite laths bridging pores within clay matrix of black shale at 1038 ft depth in the
Shell Burke well. (after Rheams and Neathery, 1984).
Fig. 11. XRD trace (CuΚα radiation) of bulk sample of Devonian black shale of the Marcellus Formation at 7680–7703 ft. showing its illitic nature (adapted after Hosterman and Whitlow,
1983). C = Chlorite, I = Illite, F = Feldspar, Cal = Calcite, PF = Plagioclase Feldspar.
Fig. 12. XRD pattern of the glycolated clay fraction (b2.0 μm) of a sample of the Marcellus
Shale showing a predominance R3-ordered mixed-layer I/S. Note the asymmetry of the
10 Å peak towards the low angle side. (Courtesy of Dr Ruarri Day-Stirrat).
2.8. Ordovician shale
2.8.1. Utica shale
This Upper Ordovician black shale occurs beneath the Marcellus
shale and is found largely in Ohio, Pennsylvania, New York and West
Virginia. It is predominantly a mixed sequence of carbonates and clay-
rich shales and has a total organic carbon content ranging from about
2.5 to 4.0%.The thickness of the Utica shale typically ranges between
150 to 350 ft in New York but can be much thicker in areas of Pennsyl-
vania. The Utica shale was deposited under marine conditions in a fore-
land basin setting (the Appalachian Basin) associated with tectonic
activity of the Taconic orogeny. At the same time, carbonate deposition
was taking place on stable shelf platforms and on basin slopes, correla-
tion between the different facies enabling Willan et al. (2012) to iden-
tify stratigraphic sequences that could be interpreted as a series of
transgressive and high stand systems tracts.
The mineralogy of the Utica shale appears to be remarkably uniform,
being dominated by quartz, carbonate minerals (calcite and dolomite)
and clay minerals. Pyrite is also a common constituent testifying to
the anoxic environment in which the sediment formed. The clay miner-
alogy appears to consist mainly of illitic minerals and chlorite. Yang and
Hesse (1991) investigated the clay mineralogy of Lower Palaeozoic
 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50 43

shales and slates in the Canadian Appalachians, including the equiva-

lents of the Utica shale and recorded XRD patterns showing a predomi-
nance of mixed-layer I/S with R3-type (Kalkberg) ordering (Fig. 13).
SEM images of the Utica shale also show tiny lath-like particles strongly
reminiscent of mixed layer I/S (Fig. 14, a and b) in sandstones.

2.9. Cambrian shale

2.9.1. Conasauga shale

This formation consists principally of shales, limestones and
dolostones and occurs in thick successions (over 15,000 ft in some
areas), in Alabama and north-west Georgia. This thickening is the result
of thrusting and deformation consequent upon successive orogenies
that affected the southern Appalachians. A black shale facies occur in
structural basins that have been much affected by these tectonic activi-
ties. Pashin et al. (2011) report that inter-bedded shales and limestones
are the source of most of the gas in the Conasauga shale and that the
shale, while brittle in dry samples, swells and retains water when wet.
The non-clay fraction of the Conasauga shale consists principally of
quartz, calcite and dolomite, with mean values of 16, 28 and 13% respec-
tively (Pashin et al., 2011). Dolomite increases with depth and may be as
high as 25%. Both K-feldspar and plagioclase feldspar occur with mean
values of 4 and 5% respectively and with plagioclase becoming increas-
ingly predominant with depth. Small amounts of pyrite and apatite are
also found. According to Pashin et al. (2011), the clay mineral compo-
nent of the Conasauga shale amounts to an average of 31%, with a
range of 12 to 50% and with mixed-layer I/S and illite/mica being the
predominant constituents with mean values of 10 and 13% respectively.
Mean values of chlorite, kaolinite and smectite were found to be 7, 1 and
1% respectively.
XRD profiles after Lee et al. (1989) for the 2.0–0.2 μm and b0.2 μm
fractions for the Nolichucky and Pumpkin Valley shales of Tennessee,
both of which belong to the Conasauga shale, are shown in Fig. 15.
The patterns are dominated by broad reflections at about 10, 5 and
3.33 Å, representing the first three basal reflections of illite and/or
mixed-layer illite-smectite where the proportion of smectite is very
low (5–10%).
The mineral content of the Conasauga shale appears to be largely
authigenic in origin. This includes quartz, much of which appears to
have a biogenic origin, the carbonate minerals, pyrite, mixed-layer I/S
and additionally a proportion of that identified as illite/mica. On the
other hand, Pashin et al. (2011) found that thin section observations re-
vealed an important detrital component but that its quantitative assess- Fig. 14. SEM images showing (a) lath-like illite growing in Utica shale and (b) Utica shale
ment required further research. fabric showing an intimate mixture of platy and tiny lath-like particles (lower left corner),
quartz grains and pyrite framboids (lower right corner) (after Daniels et al., 2011)

3. Discussion

The above review has established that there is a remarkable unifor-

Fig. 13. Mixed-layer I/S in the Utica Shale of the Canadian Appalachians. (after Yang and
Hesse, 1991).
mity in the mineralogy of the unconventional hydrocarbon reservoirs of
the USA, both with regard to non-clay minerals as well as clay minerals.
In brief, the non-clay minerals of these formations usually consist
largely of quartz, calcite and dolomite, although the relative proportion
of siliciclastic to carbonate minerals is highly variable. For example the
Eagle Ford shale is essentially calcareous whilst the Barnett shale is
highly siliceous. K-feldspar and plagioclase feldspar are usually present
in subordinate quantity in most of the reservoir rocks reviewed, and de-
trital mica can often be seen in thin sections. Pyrite is of frequent occur-
rence in minor amounts. The Monterey Formation in California
constitutes the only exception to this mineralogical uniformity in that
poorly ordered opal-A and opal-CT are often the major silica compo-
nents in this formation rather than quartz. Thin section and SEM obser-
vations confirm that non-clay mineralogy of nearly all the shale
formations reviewed consists of contributions from both terrigenous
and authigenic sources. The universal occurrence of pyrite in these
44 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
Fig. 15. XRD traces (CuΚα radiation) of the first and second fractionations of the 2.0–
0.2 μm and b 0.2 μm fractions of the (a) the Nolichucky and (b) Pumpkin Valley Shales
(adapted after Lee et al., 1989).
sediments indicates that deposition occurred in generally anaerobic
environment.
With regard to the clay minerals, illitic material consisting of a mix-
ture of discrete illite and mixed-layer I/S is predominant, in practically
all the unconventional reservoirs reviewed. Chlorite and kaolinite
occur in minor amounts and discrete smectite is found only rarely, usu-
ally in thin bentonite beds which occur only in the youngest of the hy-
drocarbon reservoirs. This review is focused mainly on the clay
mineralogy of these reservoir rocks, specifically on the question of
why illitic material is so predominant in these strata, both in discrete
and mixed-layer forms, taking into account the physical nature of
these clay minerals, their mode of origin, as well as their significance
in the exploitation of the hydrocarbon resource.
3.1. Nature of illite and mixed-layer I/S in unconventional hydrocarbon
reservoirs
3.1.1. Conventional wisdom
Following the landmark papers of Hower and his colleagues (Hower
and C., 1966; Perry and Hower, 1970; Hower et al., 1976) on burial dia-
genesis in Gulf Coast shales and on the nature of illite and mixed-layer I/
S, it has become almost axiomatic that authigenic illite in pelitic sedi-
ments develops through a smectitic precursor. Many pelitic sequences
throughout the world have been described in which the layer stacking
arrangement of mixed-layer I/S progresses from a highly smectitic and
randomly ordered structure to a more highly illitic and more ordered
structure with increasing burial depth. Using the concept of
“Reichweite” to describe types of ordering (Reynolds, 1980), a typical
mixed-layer I/S sequence of R0 → R1 → R3, where R0 equates to random
interstratifications of I and S layers, R1 to regular IS alternation and R3 to
a four layer ordered structure such as ISII or IIIS, has been observed in
many pelitic successions throughout the world, and is also characteristic
of some of the unconventional hydrocarbon reservoirs described above.
The mechanism for this transformation has remained controversial,
with many favouring a solid state transformation (SST) whereby pre-
cursor smectite converted to illite by a layer-by-layer process, involving
K-fixation in the smectite interlayers with concomitant replacement of
Si4+ by Al3+ in adjacent tetrahedral sheets, thus increasing the layer
charge (Hoffman and Hower, 1979; Bethke and Altaner, 1986; Altaner
and Ylagen, 1997; Cuadros and Altaner, 1998; Lindgreen and Hansen,
1991; Lindgreen et al., 1991; Dainyak et al., 2006). Others have pre-
ferred a dissolution-precipitation (D/P) process where the precursor
smectite gradually decomposes and yields crystals of newly formed il-
lite in increasing amounts (Dong and Peacor, 1996; Dong et al., 1997;
Buatier et al., 1992; Nieto et al., 1996; Środoń, 1999; Środoń et al.,
2000). It may be noted, however, that both SST and D/P processes re-
quire a precursor smectite (presumably montmorillonite) to initiate
the illitization process through the various stages of mixed-layer I/S.
The first question to be critically examined in this review will be, there-
fore, the evidence for a smectite precursor in the authigenic formation
of mixed layer I/S and illite in the unconventional hydrocarbon shale
reservoirs described above.
The second question to be critically discussed is whether the layer
stacking arrangements of mixed-layer I/S in terms of the various ran-
dom and ordered layer structures described above, truly reflect the
physical reality of these clay materials. At present, there is a virtual una-
nimity of opinion amongst clay scientists that it does. The strongest ev-
idence to support this view comes from the seminal papers of Reynolds
who developed computer programs for calculating one-dimensional X-
ray diffraction profiles of mixed-layer I/S with various compositions and
layer stacking arrangements that exactly matched the experimental
XRD profiles obtained in the laboratory (Reynolds, 1965; Reynolds,
1967; Reynolds and Hower, 1970; Reynolds, 1980). In brief, the calcu-
lated diffraction profiles simulated the experimental profiles so well
that it seemed reasonable to think that the compositional and structural
parameters used in calculating the modelled profiles must represent the
physical reality of the clay material under investigation. That this may
not necessarily be so was well understood by Reynolds (1980, p. 293)
who cautioned that the particle size (or the coherent scattering domain
(CSD) of mixed-layer I/S materials was an important factor to take into
account. His calculated XRD profiles assumed that I/S crystallites
consisted of 7 to 13 I and S layers, but he cautioned that the real sample
could be made up of much thinner crystallites (2 to 5 layers), although
he considered that this would change the diffraction profile in terms of
peak positions and intensities. In general, however, it was and is as-
sumed that mixed-layer I/S clays are made up of so-called MacEwan
crystallites consisting of coherently diffracting domains of about 5 to
15 unit layers in thickness and where three-dimensional organization
is maintained across the expandable smectite interlayers. Here it will
be argued from an evidential basis, stemming largely from the funda-
mental particle concept of Nadeau et al. (1984a,b,c; 1985), that
mixed-layer I/S clays are in fact made up of crystallites that are very
much thinner than those assumed in the calculation of one-
dimensional XRD profiles of these clay materials. Particular attention
will be focused on R3 ordered mixed-layer I/S as this type of illitic clay
is most common in the unconventional hydrocarbon shale reservoirs
being reviewed here.
A third and final question to be critically addressed bears on the
question of the identification of the smectite, or to be more specific,
the montmorillonite component that is presumed to exist in R3-
ordered mixed-layer I/S structures. It should be noted that in most in-
stances this identification is based solely upon the response to ethylene
glycol treatment as judged by a change in the XRD profile from that re-
corded in the air-dried state. Put simply, if ethylene glycol treatment
 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50 45

brings about such a change then it is assumed that the clay must contain
interstratified smectite layers. Usually no further evidence in support of
this identification is provided. Here it is argued, again from an evidential
basis, that this is not necessarily the case where illitic clays consist of
very thin crystallites.

3.1.2. Do mixed-layer I/S clays always form from a smectite precursor in

sedimentary sequences?
In considering this question, it is first necessary to discuss the likely
origin of the putative smectite precursor. Most authorities attribute the
formation of montmorillonitic smectite in sediments to the alteration of
the glassy component of volcanic ash, and where the sedimentary se-
quence is interbedded with bentonites, which clearly demonstrate
such an origin, then this seems entirely reasonable. It may be noted,
however, that Hoffman (1979) concluded that the smectite precursor
of the mixed-layer I/S in the Gulf of Mexico shales was not from a volca-
nic source but from weathered rocks and soils in the catchment area of
the ancient Mississippi River. Similarly, Chalmley (1989) doubted that
the smectite in the Meso-Cenozoic sediments in the Atlantic region of
Europe had a volcanogenic genesis, attributing the origin of the clay
mineral to weathering and erosion of rocks and soils in the source area.
In this brief literature review of the geology and mineralogy of many
of the shale formations designated as unconventional hydrocarbon res-
ervoirs in the USA, it became apparent that a volcanogenic origin for
mixed-layer I/S, as demonstrated by frequent inter-bedded highly
smectitic bentonites, was not of universal occurrence. In fact, this rela-
tionship was found only in the youngest of these hydrocarbon reser-
voirs, namely the Cenozoic Monterey Formation and the Upper
Cretaceous Eagle Ford, Niobrara-Pierre, Mancos and, possibly, the Tus-
caloosa Formations. No evidence was found for a volcanogenic smectite
precursor in any of the older shale formations, with the possible excep-
tion of the Utica shale. In fact, no evidence was found for the occurrence
of randomly ordered (R0) I/S in any of these formations, whose clay
mineralogy was almost invariably dominated by illite and ordered
highly illitic (R3) I/S. It could, of course, be argued that the reason for
the absence of smectite and smectitic I/S in these older reservoir rocks
is that these clay minerals have been converted to illite and ordered
highly illitic I/S with the passage of geological time. This would, how-
ever, presuppose either a more or less constant deposition of volcanic
ash, or similarly constant contributions of smectitic sediment from
source areas, throughout the time that the fine-grained reservoir rocks Fig. 16. Filamentous illite (a) surrounding quartz grains in the Rotliegend sandstone of the
were forming. Both scenarios seem extremely unlikely. An alternative North Sea after critical point drying. (Images of clay, courtesy Mineralogical Society) and
explanation for the common occurrence of highly illitic R3 I/S in these (b) occupying pores in a North Sea sandstone (courtesy Corex, UK).
rocks is that the formation of such clay does not require a smectitic pre-
cursor, a proposition which will now be discussed.
The most striking evidence for the de novo formation of R3 type I/S, of illite and mixed-layer I/S in reservoir sandstones cannot be attributed
where this type of clay is simply regarded as crystallizing from pore wa- to conversion of a precursor smectite, bearing in mind the unique fila-
ters of appropriate chemical composition in a particular pressure and mentous morphology of the illitic clays as well as the lack of evidence
temperature stability field, comes from conventional sandstone hydro- for volcanogenic contributions to the sandstones themselves or to sed-
carbon reservoirs. Scanning electron microscope (SEM) observations iments higher up in the sequence. Thus, a smectitic precursor is not a
very often show illitic clays (both illite and R3 I/S) forming in the universal requirement for the authigenic formation of illite in sand-
pores of these rocks and showing a unique morphology variously de- stones and so the question arises as to whether this might be true for
scribed as filamentous, hairy or lath-like (Wilson et al., 2014a; Wilson finer-grained sediments too.
et al., 2014b). Consideration of this morphology (Figure 16) makes it Turning to the unconventional hydrocarbon reservoirs reviewed
abundantly clear that formation from a precursor montmorillonite, here, Figures 10 and 14 show SEM images of lath-like illitic clays grow-
which usually occurs in poorly shaped, flake-like particles, via a solid ing in the pores of Chattanooga and Utica shales respectively. These
state, layer-by-layer conversion, is out of the question. A dissolution/ clays look essentially similar to those found in conventional sandstone
precipitation might be considered as a possibility, but there is rarely reservoirs, but as a rule they would be much more difficult to observe
evidence for the existence of a more smectitic clay higher up in the se- by SEM because of the finer-grained fabric and much smaller pore
quence that might have served as a precursor for the illitic clays at sizes that are characteristic of shales. Although requiring further confir-
greater depth. This is not to deny that illitic clays do not form from mation, it is suggested here that such observations are consistent with
smectitic precursors, as they clearly do with increasing depth of burial de novo crystallization of illitic clays in shales, as in sandstones, in condi-
in Gulf Coast shales and in many other shale sequences. In these in- tions where no precursor smectitic materials are required. It may be
stances, it is evident that the smectitic clays have decomposed and con- added parenthetically that while changes in provenance in sandstones
tributed at least some of their chemical constituents to the can be assessed by point counting, this is much more difficult in mud-
crystallization of more illitic clays. The point here is that the formation stones and requires the use of other techniques.
46 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50

3.1.3. Do the layer stacking arrangements calculated for mixed-layer I/S
XRD patterns reflect physical reality?
Up until 1984, this question was almost invariably answered in the
affirmative, even though direct observational evidence to support this
conclusion was lacking. The almost unanimous view then, and probably
even now, was that mixed-layer I/S occurred in MacEwan-type crystal-
lites which as described above are thought to consist of 5 to 15 unit
layers in thickness and where there was three-dimensional regularity
across the smectite interlayers. However, following the work of
Nadeau et al. (1984a,b,c), who developed a fundamental particle con-
cept that could be applied to mixed-layer minerals in general and to
mixed-layer I/S in particular (Nadeau et al., 1985), the application of
the MacEwan-type crystallite concept to these particular clay materials
was thrown into question, even though it was clear that this concept
was valid for other types of mixed-layer minerals such as hydrobiotite
(Wilson and Nadeau, 1985). Using electron microscopy, much effort
was subsequently made to show either that the stacking arrangement
of mixed-layer I/S in samples of various formations did consist of
MacEwan-type crystallites, or alternatively, did not consist of thinner
fundamental-type particles (reviewed in Wilson, 2013). Although
these works generally concluded that their findings were only compat-
ible with the view that I/S was made up of MacEwan-type crystallites,
this conclusion was not accepted as necessarily valid in a critical review
of the subject by Wilson (2013). In addition, there was little consider-
ation of how the evidence that was inconsistent with the idea that
mixed-layer I/S clays were comprised of MacEwan-type crystallites
could be explained.
Evidence that clay materials yielding a mixed-layer I/S XRD profile
could not possibly be made up of MacEwan-type crystallites came ini-
tially from a combined SEM, TEM and XRD study by McHardy et al.
(1982) of authigenic illitic clays in the Magnus sandstone of the North
Sea. This study used critical point drying on preserved samples and
showed that the clays consisted of a network of very fine filamentous,
lath-like particles that practically filled the pore space of the sandstone

Fig. 17. SEMs of Magnus sandstone after critical point drying and showing filamentous illitic clay (a) occupying the sandstone pores, (b) and (c) attached and embedded in authigenic
quartz overgrowths and (e) and (f) the very fine-grained nature of the filaments (after McHardy et al., 1982).
 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50 47

and were anchored by authigenic quartz overgrowths (Fig. 17). TEM

measurement of the thickness of these lath-like particles when dried
down after mild ultrasonic treatment onto an electron microscope
grid revealed that on average they were only 20 to 30 Å thick. XRD pat-
terns of this material dried down onto glass slides after similar mild ul-
trasonic treatment were those of fully ordered R2 mixed-layer I/S clay
containing about 20% smectite layers (Fig. 18).These results are difficult
to reconcile with a stacking arrangement of crystallites comprising up to
15 unit layers in thickness, although it is possible to argue that such an
arrangement had existed but had been disrupted prior to the ultrasonic
treatment used before TEM examination, but had remained intact after
the ultrasonic treatment prior to XRD examination. However, further
experiments by Nadeau et al. (1984a) revealed the unreality of this
interpretation.
Nadeau et al. (1984a) recorded XRD patterns of oriented aggregates
from mixed suspensions of the b0.1 μm fractions consisting of various
ratios of Wyoming montmorillonite and mixed-layer (R1) I/S
(Figure 19). TEM measurements determined that 83% of particles of
the Wyoming montmorillonite had a thickness of 10Å and that 79% of

Fig. 19. XRD patterns of ethylene glycol solvated sedimented aggregates (b0.1 μm
fractions) on glass slides consisting of physical mixtures (b to e) of (a) montmorillonite
and (f) regular mixed-layer I/S. Mean layer thickness of montmorillonite is 10 Å and of
I/S is 20 Å. Ratios by weight of mixtures are (b) 1:1, (c) 2:1, (d) 4:1 and (e) 8:1 (after
Nadeau et al., 1984a).

the particles of R1 I/S were 20 Å thick. When these two suspensions

Fig. 18. XRD patterns of oriented aggregate of filamentous clay from the Magnus
Sandstone of the North Sea (a) air-dried, (b) Mg-saturated and (c) after ethylene glycol.
Unmarked peaks are from kaolinite (after McHardy et al., 1982).
were mixed and dried onto glass slides, the result was to produce XRD
patterns that are almost identical to those of random mixed-layer I/S
with 90 to 50% smectite layers and showing the characteristic peak mi-
gration curves for 001I/002S and 002I/003S illustrated by Reynolds
48 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50

These observations do not necessarily imply that all mixed-layer I/S

clays are artificial XRD constructs and do not exist in nature. Certainly
where precursor smectite clays with coherent scattering domains
(CSDs) of variable thickness are being illitized by a dissolution/precipi-
tation process then it is readily conceivable that crystallites of R0 I/S
may be built up in this way. Such a process may even lead to a tendency
for repetition of I and S units where their CSDs are of extremely thin,
thus leading to R1 I/S. For mixed-layer R3 I/S, however, it is indeed ar-
gued that this may very well be an XRD construct, especially when the
reality of the presence of smectite layers is questionable as discussed
below.

3.1.4. Is the smectite identified in mixed-layer I/S really smectite?

Fig. 20. TEM of randomly oriented lath-like illite particles separated from Rotliegend
Sandstone, North Sea (after Nadeau et al., 1984a).
(1980). In contrast, oriented aggregates of mixtures of b0.2 μm suspen-
sions of Wyoming montmorillonite and Rotliegend illite in the same ra-
tios to those in Fig. 19 yielded XRD patterns of mixtures, with no
indication of mixed layering. The results of the experiments of Nadeau
et al. (1984a) can only imply that the mixed-layer I/S clays are not
necessarily made up of MacEwan-type crystallites, even though
modelled XRD profiles assuming a stacking arrangement based upon
such crystallites yield excellent matches with experimental profiles. It
is noteworthy that Reynolds (1992), from an examination of I/S clays
with a range of expandability and stacking arrangements, showed that
the degree of turbostratic disorder increased linearly with the amount
of expandable layers, indicating that such disorder occurred only at ex-
pandable layers, contrary to what would be expected if I/S clays were
comprised mainly of MacEwan-type crystallites.
The question here is whether a response to ethylene glycol treat-
ment, in terms of changes induced in the XRD profile of clay from an
air-dried state, necessarily means that smectitic material is present in
the clay sample. In the particular case of long-range ordered, mixed-
layer I/S, with putative stacking arrangements like IIIS (R3) or IIS (R2),
there is evidence to indicate that the response to glycol is a result of
inter-particle diffraction, whereby thin (b 50 Å in thickness) illite parti-
cles, quickly dried down on glass slides, are able to adsorb ethylene gly-
col between the thin particles and so yield XRD profiles that have
apparently changed from the air-dried state. Even though such thin illite
particles would nominally possess a higher layer change than that of
smectite, the lack of three-dimensional register between the
randomly-oriented, dried down particles (Fig. 20), would enable ethyl-
ene glycol to be adsorbed between them and so yield a changed XRD
pattern that apparently indicated smectite interlayers, even though
smectite interlayers were not actually present.
As described above, the phenomenon of inter-particle diffraction
was first broadly described by McHardy et al. (1982) as a way of ac-
counting for apparently irreconcilable XRD and TEM results with

Fig. 21. XRD diagrams after ethylene glycol treatment of the b0.2 μm fractions and particle thickness distribution histograms as measured by TEM from (a) Jurassic North Sea sandstone,
(b) Devonian bentonite and (c) Permian North Sea sandstone with accompanying particle thickness histograms. Samples (a) and (b) are typical of R3 mixed-layer I/S, with about 90% illite
layers, and consist of particles which are significantly thinner than those of sample (c) (after Nadeau et al., 1985).
 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
respect to illitic clays in the Magnus Sandstone of the North Sea. This
concept was examined in more detail by Nadeau et al. (1984a, 1984b),
applied to a wide range of mixed-layer I/S samples (Nadeau et al.,
1985), and generalized to a broader fundamental particle concept for
mixed-layer minerals (Nadeau et al., 1984c). Here it is important to
note that it was concluded that “long range regularly interstratified
illite-smectite of the IIS and IIIS types are composed primarily of funda-
mental ‘illite’ particles 20–50 Å thick. Conventional illite is composed
primarily of particles N50 Å thick” (Nadeau et al., 1984c). Fig. 21
shows some of the data on which this conclusion is based. If this conclu-
sion is correct, the implication here with regard to the unconventional
hydrocarbon shale reservoirs older than Upper Cretaceous in the USA,
where these long-range ordered, mixed-layer I/S clays form a significant
part of the clay fraction, is that these clays are primarily thin illite and
that the smectite interlayers do not actually exist in reality.
4. Conclusions
Given that the arguments presented above concerning the physical
nature of what has previously been termed regularly ordered (R3)
mixed-layer I/S are correct, that is that such clay material in reality is
simply thin illite, then it would be legitimate to question whether or
not this really matters in the context of the influence of this clay mate-
rial on the overall properties of the shale formations in which it occurs.
In order to clarify this issue, it is necessary to discuss the physico-
chemical characteristics of such illitic material, if possible as it occurs
in highly compressed and lithified shale formations, and to compare
these characteristics with what might be expected if such material
consisted of mixed-layer I/S phase as presently understood. It will
then be considered how these characteristics could impact on the
physico-chemical properties of the unconventional hydrocarbon reser-
voirs as a whole, including porosity, permeability, pore pressure and
formation damage. These matters are discussed further in Part II of
this paper.
Acknowledgments
The authors are grateful to Paul Nadeau, Ruarri Day-Stirrat and
Andrew Hurst for their constructive reviews and suggestions of this
paper, all of which resulted in its material improvement.
References
Ahn, H., Peacor, D.R., 1986. Transmission and analytical electron microscopy of the
smectite-to-illite transformation. Clays and Clay Minerals. 34, 165–179.
Al Khammali, S., 2015. Geochemical and clay mineralogical characteristics of the
Woodford Shale, Payne County, Oklahoma. MSc thesis. Kansas State University.
Alese, A.O., 2012. Gamma spectroscopy and geochemical investigation of the Mississip-
pian (Chesterian) Fayetteville Shale and Imo Shale, Arkoma Basin, Arkansas. MSc
Thesis. Oklahoma State University.
Alexander, T., Baihly, J., Boyer, C., Clark, B., Waters, G., Jochen, V., Le Calvez, J., Lewis, R.,
Miller, C.K., Thaeler, J., Toelle, B.E., 2011. Shale gas revolution. Oilfield Review. 23
(3), 40–55.
Altaner, S.P., Ylagen, R.F., 1997. Comparison of structural models of mixed-layer illite/
smectite and reaction mechanisms of smectite illitization. Clay Clay Miner. 45,
517–533.
Ashu, R., 2014. A petrographic study of the Three Forks Formation (Upper Devonian),
Williston Basin, North Dakota, based on thin section analysis, XRD and SEM. Journal
of Geological Research, 264170 http://dx.doi.org/10.1155/2014/264170 (9pp).
Bates, R.L., Jackson, J.A., 1980. Glossary of Geology. second ed. American Geological Insti-
tute, Leesburg Pike, Falls Church, VA, USA.
Bates, T.F., Strahl, E.O., 1957. Mineralogy, petrography, and radioactivity of representative
samples of Chattanooga Shale. Bulletin of the Geological Society of America. 68,
1305–1314.
Bell, T.E., 1986. Microstructure in Mixed-layer Illite/Smectite and Its Relationship to the
Reaction of Smectite to Illite.
Bertog, J.L., 2013. Timing of onset of volcanic centers in the Campanian of western North
America as determined by distal ashfalls. OPG 3, 121–133.
Bethke, C.M., Altaner, S.P., 1986. Layer-by-layer mechanism of smectite illitization and ap-
plication to a new rate law. Clay Clay Miner. 34, 136–145.
49
Boles, J.R., Franks, S.G., 1979. Clay diagenesis in Wilcox sandstones of south west Texas:
implications of smectite diagenesis on sandstone cementation. J. Sediment. Petrol.
49, 55–70.
Boyer, C., Clark, B., Jochen, V., Lewis, R., Miller, C.K., 2011. Shale gas: a global resource.
Oilfield Review 23 (3), 28–39.
Buatier, M.D., Peacor, D.R., O’Neil, J.R., 1992. Smectite-illite transition in Barbados accre-
tionary wedge sediments; TEM and AEM evidence for dissolution/crystallization at
low temperature. Clay Clay Miner. 40, 65–80.
Caldwell, C.D., 2011. Lithostratigraphy of the Woodford Shale, Anadarko Basin, West Cen-
tral Oklahoma. www.ogs.ou.edu/MEETINGS/.../ShalesMoving2011/CaldwellLitho.pd.
Cetin & Huff, 1995. Layer charge of the expandable component of illite-smectite in K-
bentonite as determined by alkyl ammonium exchange. Clay Clay Miner. 43, 150–158.
Chalmley, H., 1989. Clay Sedimentology. Springer, Berlin.
Cicero, A.D., Steinhoff, I., MaClain, T., Koepke, K.A., Dezelle, J.D., 2010. Sequence stratigra-
phy of the Upper Jurassic mixed carbonate/siliciclastic Haynesville and Bossier Shale
depositional systems in east Texas and north Louisiana. Gulf Coast Association of
Geological Societies Transactions 60, 133–148.
Clark, H., Turner, P., Bustin, M.R., 2014. Unlocking the resource potential of the Bowland
Basin, NW England. SPE 167776, 1–11.
Cluff, R. M., Reinbold, M. L. & Lineback, J. A. 1981. The New Albany Shale Group of Illinois.
Illinois State Geological Survey. Circular 518. Champaign, Illinois. 83p.
Cobban, W.A., Reeside Jr., J.B., 1952. Correlation of the Cretaceous formations of the Western
Interior of the United States. Geological Society of America Bulletin. 63, 1011–1044.
Comer, J.B., 1991. Stratigraphic analysis of the Upper Devonian Woodford Formation,
Permian basin, west Texas and south-eastern New Mexico. Report of Investigations
201. Indiana Geological Survey.
Compton, J.S., 1991. Origin and diagenesis of clay minerals in the Monterey Formation,
Santa Maria Basin area California. Clay Clay Miner. 39, 440–466.
Conant, L.C., Swanson, V.E., 1957. Chattanooga shale and related rocks of central Tennessee
and nearby areas. Trace Elements Investigations Report 682. U.S. Geol. Surv pp. 1-13.
Conant, L.C., Swanson, V.E., 1961. Chattanooga Shale and related rocks of central Tennes-
see and nearby areas. U.S. Geol. Surv. Prof. Pap. 357 91p.
Couto-Anjos, S.M., 1986. Absence of clay diagenesis in Cretaceous-Tertiary marine shales
Campos Basin Brazil. Clay Clay Miner. 34, 424–434.
Craft, J.R., 2013. Wolfcamp shale graduates to world class play. Unconventional Oil and
Gas Report 10/01/2013.
Cuadros, J., Altaner, S., 1998. Characterization of mixed-layer illite-smectite from bentonites
using microscopic, chemical and X-ray methods; constraints on the transformation of
the the smectite-to-illite transformation mechanism. Am. Mineral. 83, 762–774.
Curtis, M.E., 2013. Influence of thermal maturity on organic shale microstructure. Pre-
sented at the Oklahoma Shale Gas and Oil Workshop (November 20, 2013. Power
Point Presentation).
Dainyak, L.D., Drits, V.A., Zviagina, V.V., Lindgreen, H., 2006. Cation redistribution in the
octahedral sheet during diagenesis of illite-smectites from Jurassic and Cambrian
source rock shales. Am. Mineral. 91, 589–603.
Daniels, J.J., Cole, D., Murphy, M., Welch, S., Sheets, J., 2011. Mineralogical and physical prop-
erty characteristics of the Utica-Point Pleasant shale from core and geophysical log anal-
ysis, in Taking a deeper look at shales: geology and potential of the Upper Ordovician
Utica Shale in the Appalachian basin. Petroleum Technology Transfer Council, Ohio Geo-
logical Survey, and Ohio Geological Society, June 21, 2011 New Philadelphia, OH.
Dong, H., Peacor, D.R., 1996. TEM observations of coherent stacking relations in smectite I/
S and illite of shales: evidence for MacEwan crystallites and dominance of 2M1
polytypism. Clay Clay Miner. 44, 257–275.
Dong, H., Peacor, D.R., Freed, R.L., 1997. Phase relations among smectite, R1 illite-smectite
and illite. Am. Mineral. 82, 379–391.
Elgamati, M., Zhang, H., Bai, B., Flori, R., Qi, Q., 2011. Some micron pore characteristics of
shale gas plays. SPE 144050. Presentation at SPE America’s Unconventional Gas Con-
ference and Exhibition. 12 16 June, 2011.
Elston, H.W., 2014. Mineralogical and Geochemical Assessment of the Eagle Ford Shales.
Senior Thesis for BSc Degree at Ohio State University (31pp).
Ettensohn, F.R., Miller, M.L., Dillman, S.B., Elam, T.D., Geller, K.L., Swager, D.R., Markowitz,
G., Woock, R.D., Barron, L.S., 1988. Characterization and implications of the Devonian-
Mississippian black shale sequence, eastern and central Kentucky, U.S.A.: pycnoclines,
transgression, regression, and tectonism. In: McMillan, N.J., Embry, A.F., Glass, D.J.
(Eds.), Devonian of the World, Proceedings of the Second International Symposium
on the Devonian System: Canadian Society of Petroleum Geologists. Memoir. 14
(2), pp. 323–345.
Freed, R.L., Peacor, D.R., 1989. Variability in temperature of the smectite-illite reaction in
Gulf Coast sediments. Clay Miner. 24, 171–180.
Freed, R.L., Peacor, D.R., 1992. Diagenesis and the formation of illite-rich I/S crystallites in
Gulf Coast shales: TEM study of clay separates. J. Sediment. Petrol. 62, 220–-234.
Gamero-Diaz, H., Miller, C., Lewis, R., Contreras Fuentas, C., 2013. Evaluating the impact of
mineralogy on reservoir quality and completion quality of organic shale plays. Search
and Discovery Article#41221. AAPG Rocky Mountain Section Meeting Salt Lake City,
Utah, September 22-24, 2013.
Gharrabi, H., Velde, B., 1995. Clay mineral evolution in the Illinois Basin and its causes.
Clay Miner. 30, 353–-364.
Gill, J.R., Cobban, W.A., Schultz, L.G., 1972. Stratigraphy and composition of the Sharon
Springs Member of the Pierre Shale in Western Kansas. US Geological Survey Profes-
sional Paper 728.
Goodman, W.R., Maness, T.R., 2008. Michigan's Antrim Gas Shale Play: a Two-Decade
Template for Successful Devonian Gas Shale Development. American Association of
Petroleum Geologists.
Hamlin, H.S., Baumgardner, R.W., 2012. Wolfberry (Wolfcampiam Leonardian) deep
water depositional systems in the Midland Basin: stratigraphy, lithofacies, reservoirs
and source rocks (Report of Investigation 277). Texas Bureau of Economic Geology.
50 M.J. Wilson et al. / Earth-Science Reviews 158 (2016) 31–50
Hoffman, J.C., 1979. An valuation of potassium uptake by Mississippi River-borne clays
following deposition in the Gulf of Mexico. PhD thesis. Case Western Reserve Univer-
sity, Cleveland, Ohio.
Hoffman, J., Hower, J., 1979. Clay Mineral Assemblages as Low Grade Metamorphic
Geothermometers: Application to the Thrust Faulted Disturbed Belt of Montana,
USA. Society of Economic Palaeontologists and Mineralogists. 26. Special Publica-
tions, Tulsa, OK, pp. 55–79.
Hosterman, J.W., Whitlow, S.I., 1983. Clay mineralogy of Devonian shales in the Appala-
chian Basin. USGS Professional Paper. 1298 30 pp.
Hower, J., C., Mowatt.T., 1966. The mineralogy of illite and mixed-layer illite/montmoril-
lonite. American Mineralogist 51, 825–854.
Hower, J., Eslinger, E.V., Hower, M.E., Perry, E.A., 1976. Mechanism of burial metamor-
phism of argillaceous sediments I. Mineralogical and chemical evidence. I. Mineralog-
ical and chemical evidence. Geol. Soc. Am. Bull. 87, 725–737.
Jennings, D.S., Antia, J., 2015. Petrographic characterization of the Eagle Ford shale, south
Texas: Mineralogy, common constituents and distribution of nanometer scale pore
types. In: Camp, W.K., Diaz, E., Wawak, B. (Eds.), Electron Microscopy of Shale Hydro-
carbon Reservoirs. AAPG. Memoir 102, pp. 101–115.
Kim, J.W., Berg, R.R., Watkins, J., Tieh, T.T., 2001. Texture, mineralogy, and petrophysical
properties of geopressured shalesGulf of Mexico. Gulf Coast Association of Geological
Societies Transactions 51, 161–172.
Lee, S.Y., Hyder, L.K., Baxter, P.M., 1989. Mineralogical characterization of selected shales
in support of nuclear waste repository studies: progress Report for October 1987–
September, 1988. Environmental Sciences Division. Publication No. 3225, p. 36 pp.
Leventhal, J.S., Hosterman, J.W., 1982. Chemical and mineralogical analysis of Devonian
black-shale samples from Martin County, Kentucky; Carroll and Washington
counties, Ohio; Wise County, Virginia; and Overton County, Tennessee, U.S.A.
Chem. Geol. 37, 239–-264.
Lindgreen, H., Hansen, P.L., 1991. Ordering of illite-smectite in Upper Jurassic claystones
from the North Sea. Clay Minerals. 26, 105–125.
Lindgreen, H., Jacobsen, H., Jakobsen, H.J., 1991. Diagenetic structural transformations in
North Sea Jurassic illite/smectite. Clays and Clay Minerals. 39, 54–69.
Loucks, R.G., Ruppel, S.C., 2007. Mississippian Barnett shale: lithofacies and depositional
setting of a deep-water shale-gas succession in the Fort Worth Basin Texas. Associa-
tion of American Petroleum Geologists Bulletin. 91, pp. 579–-601.
Matthews, R.D., 1993. Petroleum Geology of the Devonian and Mississippian Black Shale
of Eastern North America. United States of America Geological Survey.
McCarty, D.K., Sakharov, B.A., Drits, V.A., 2008. Early clay diagenesis in Gulf Coast sedi-
ments: new insights from XRD profile modeling. Clay Clay Miner. 56, 359–379.
McHardy, W.J., Wilson, M.J., Tait, J.M., 1982. Electron microscope and X-ray diffraction
studies of filamentous illitic clay from sandstones of the Magnus Field. Clay Miner.
17, 23–39.
Montgomery, S.L., Jarvie, D.M., Bowker, K.A., Pollastro, R.M., 2005. Mississippian Barnett
Shale, Fort Worth basin, north central Texas: Gas-shale play with multi trillion
cubic foot potential. Association of American Petroleum Geologists Bulletin. 89,
pp. 155–175.
Mullen, J., 2010. Petrophysical characterization of the Eagle Ford shale in south Texas.
CSUK/SPE 138145 (19 pp).
Murata, K.J., Larsen, R.R., 1975. Silica mineralogy and structure of Monterey shale, Tem-
blor Range, California. USGS Journal of Research. 3, 553–566.
Murata, K.J., Friedman, I., Gleason, J.D., 1977. Oxygen isotope relations between diagenetic
silica minerals in Monterey Shale, Temblor Range, California. Am. J. Sci. 277, 259–272.
Nadeau, P.H., Reynolds Jr., R.C., 1981. Burial and contact metamorphism in the Mancos
Shale. Clay Clay Miner. 29, 249–259.
Nadeau, P.H., Tait, J.M., McHardy, W.J., Wilson, M.J., 1984a. Interstratified XRD character-
istics of physical mixtures of elementary particles. Clay Miner. 19, 67–76.
Nadeau, P.H., Wilson, M.J., McHardy, W.J., Tait, J.M., 1984b. Interstratified clays as funda-
mental particles. Science 225, 923–925.
Nadeau, P.H., Wilson, M.J., McHardy, W.J., Tait, J.M., 1984c. Interparticle diffraction; a new
concept for interstratified clays. Clay Miner. 19, 757–769.
Nadeau, P.H., Wilson, M.J., McHardy, W.J., Tait, J.M., 1985. The conversion of smectite to
illite during diagenesis; evidence from some illitic clays from bentonites and sand-
stones. Mineral. Mag. 49, 393–400.
Nieto, F., Ortega-Huertas, M., Peacor, D.R., Arostegui, J., 1996. Evolution of illite/smectite
from early diagenesis to incipient metamorphism in sediments of the Basque-
Cantabrian Basin. Clay Clay Miner. 44, 304–323.
Pashin, J.C., Kopaska-Merkel, D.C., Arnold, A.C., McIntyre, M.R., 2011. Geological Founda-
tion for Production of Natural Gas from Diverse Shale Formations. RHSEA Final Re-
port. Contract 07122-17 175 pp.
Patel, H.J., 2014. Geologic assessment of drilling, completion, and stimulation methods in
selected gas shale plays worldwide. Honours Undergraduate Thesis. Texas A & M
University.
Perry, E.A., Hower, J., 1970. Burial diagenesis in Gulf Coast pelitic sediments. Clay Clay
Miner. 18, 165–177.
Perry, E.A., Hower, J., 1972. Late stage dehydration in deeply buried pelitic sediments. Am.
Assoc. Pet. Geol. Bull. 56, 2013–2021.
Pitman, J.K., Price, L.C., LeFever, J.A., 2001. Diagenesis and Fracture Development in the
Bakken Formation, Williston Basin: Implications for Reservoir Quality in the Middle
Member. USGS paper. 1653 pp. pp 1–pp19.
Pollastro, R.M., 1985. Mineralogical and morphological evidence for the formation of illite
at the expense of illite/smectite. Clay Clay Miner. 33, 265–274.
Pollastro, R.M., 1993. Considerations and applications of the illite/smectite
geothermometer in hydrocarbon-bearing rocks of Miocene to Mississippian age.
Clay Clay Miner. 41, 119–133.
Pollastro, R.M., 1994. Clay diagenesis and mass balance – the forest and the trees. Clay
Clay Miner. 42, 93–97.
Reynolds, R.C., 1965. An X-ray study of an ethylene glycol-montmorillonite complex. Am.
Mineral. 50, 990–1001.
Reynolds, R.C., 1967. Interstratified clay systems. Calculation of the total one-dimensional
diffraction function. Am. Mineral. 52, 661–672.
Reynolds, R.C., 1980. Interstratified clay minerals. In: Brindley, G.W., Brown, G. (Eds.),
Crystal Structure of Clay Minerals and Their X-Ray Identification. Mineralogical Soci-
ety, London, pp. 249–304.
Reynolds, R.C., 1992. X-ray diffraction studies of illite/smectite from rocks, b1 μm oriented
powders and b1 μm randomly oriented powder aggregates: the absence of labora-
tory induced artefacts. Clay Clay Miner. 40, 387–396.
Reynolds, R.C., Hower, J., 1970. The nature of interlayering in mixed layer illite-
montmorillonite. Clay Clay Miner. 18, 25–36.
Rheams, K.F., Neathery, T.L., 1984. Characterization and geochemistry of Devonian oil
shale north Alabama-south central Tennessee. Symposium on Characterization and
Chemistry of Oil Shales. Division of Fuel Chemistry and Petroleum Chemistry,
American Chemical Society St Louis, April 8–13, 1984.
Ridgley, J.L., Condon, S.M., Hatch, J.R., 2013. Geology and Oil and Gas Assessment of the
Mancos-Menefee Composite Total Petroleum System. Chapter 4 in USGS Digital
Data Series 69-F. pp. p.1–p97.
Ruotsala, A.P., 1980. Mineralogy of the Antrim Shale. www.researchgate.net/.../
236555468.
Sanders, W.B., Ramsbottom, W.H.C., Manger, W.L., 1979. Mesothemic cyclicity in the mid
Carboniferous of the Ozark shelf region. Geology 7, 293–296.
Sano, J.L., Ratcliffe, K.T., Spain, D.R., 2013. Chemostratigraphy of the Haynesville Shale. In:
Hammes, U., Gale, A. (Eds.), In Geology of the Haynesville Gas Shale in East Texas and
West Louisiana. American Association of Petroleum Geologists, Memoir 105,
pp. 137–154.
Schultz, L.G., Tourtelot, H.A., Gill, J.R., Boerngen, J.G., 1980. Geochemistry of the Pierre
Shale and equivalent rocks of Late Cretaceous age: composition and properties of
the Pierre Shale and equivalent rocks, Northern Great Plains Region. US Geological
Survey Professional Paper 1064.
Sonnenberg, S.Z., Appleby, S.K., Sarg, J.R., 2010. Quantitative mineralogy and
microfractures in the Middle Bakken Formation, Williston Basin, North Dakota.
Search and Discovery. Article #40628. Poster presentation, AAPG Annual Convention
and Exhibition New Orleans, Louisiana, April 11–15, 2010.
Środoń, J., 1999. Nature of mixed-layer clays and mechanisms of their formation and al-
teration. Annu. Rev. Earth Planet. Sci. 27, 19–53.
Środoń, J., Eberl, D.D., Drits, V.A., 2000. Evolution of particle size during illitization of
smectite and implications for reaction mechanism. Clay Clay Miner. 48, 446–458.
Sutherland, P.K., 1988. Late Mississippian and Pennsylvanian depositional history in the
Arkoma basin area, Oklahoma and Arkansas. Geol. Soc. Am. Bull. 100, 1787–1802.
Totten, M.W., Dixon, M., Hanan, M.A., 2005. Diagenesis of mixed-layer clay minerals in the
South Timbalier area. Gulf of Mexico. Gulf Coast Association of Geological Societies
Transactions. 55, pp. 821–829.
Tourtelot, H.A., 1962. Preliminary investigation of the geological setting and chemical
composition of the Pierre Shale, Great Plains Region. US Geological Survey Profes-
sional Paper 390.
Trembath, A., Jenkins, J., Nordhaus, T. & Shellenberger, M. 2012. Where the shale gas revolu-
tion came from; Government's role in the development of hydraulic fracturing in shale.
Breakthrough Institute Energy and Climate Program. http://thebreakthrough.org/
energy.shtml
Wang, T., Carr, T.R., 2013. Organic-rich Marcellus Shale lithofacies modelling and distribu-
tion pattern analysis in the Appalachian basin). Am. Assoc. Pet. Geol. Bull. 97,
2173–2205.
Weaver, C.E., 1958. Geologic interpretation of argillaceous sediments Part II. Clay petrol-
ogy of the Upper Mississippian – Lower Pennsylvanian sediments of central United
States. Am. Assoc. Pet. Geol. Bull. 42, 272–309.
Whittington II, R.A., 2009. Clay Mineralogy and Illite Crystallinity in the Late Devonian to
Early Mississippian Woodford Shale in the Arbuckle Mountains, Oklahoma, USA.
Georgia State University, MSc Thesis.
Willan, C.G., McCallum, S.D., Warner, T.B., 2012. Regional Interpretation of the Middle-
Upper Ordovician Utica Shale Play in the Appalachian Basin. Search and Discovery Ar-
ticle #80239. AAPG Annual Convention and Exhibition Long Beach, California, USA,
April 22–25, 2012.
Wilson, M.J., 2013. Illite/smectite (I/S). Deer, Howie and Zussman, Rock-Forming Min-
erals, Volume 3C Sheet Silicates: Clay Minerals. Geological Society of London,
pp. 478–523.
Wilson, M.J., Nadeau, P.H., 1985. Interstratified clay minerals and weathering processes.
In: Drever, J.I. (Ed.), The Chemistry of Weathering. Reidel Publishing Company,
pp. 97–118.
Wilson, M.J., Wilson, L., Patey, I., 2014a. The influence of individual clayminerals on for-
mation damage of reservoir sandstones: a critical review with some new insights.
Clay Miner. 49, 127–145.
Wilson, L., Wilson, M.J., Green, J., Patey, I., 2014b. The influence of clay mineralogy on for-
mation damage in North Sea reservoir sandstones: a review with illustrative exam-
ples. Earth-Sci. Rev. 134, 70–80.
WoldeGabriel, G., Wehner, S., Raines, M., Chipera, S., Fittipaldo, M. & Guthrie, G. (undated)
Mineralogical and Textural Variations in Shale and Its Seal Integrity in a CO2-flooded En-
vironment. LA-UR-06-0812 https://www.netl.doe.gov/publications/proceedings/.../
Poster%20190.pd
Yang, C., Hesse, R., 1991. Clay minerals as indicators of diagenetic and anchimetamorphic
grade in an overthrust belt, external domain of southern Canadian Appalachians. Clay
Miner. 26, 211–231.