OR/12/032 Mineralogy

From Earthwise
Jump to: navigation, search
Hobbs, P R N, Entwisle, D C, Northmore, K J, Sumbler, M G, Jones, L D, Kemp, S, Self, S, Barron, M, and Meakin, J L. 2012. Engineering Geology of British rocks and soils - Lias Group. British Geological Survey, Nottingham, UK. (OR/12/032).

The following account is taken from Kemp & McKervey (2001)[1], Kemp & Hards (2000)[2] and Kemp, et al., 2005[3].

General

There are key geohazards involved in the engineering geology of clays and mudstones. Due to their high surface area, residual charge and interaction with water, clay minerals (and smectite in particular) are most frequently cited as the reason for the distinctive shrink-swell behaviour noted in many fine-grained sedimentary rocks, and consequent structural damage resulting from seasonal and long-term volume changes. Oxidation of pyrite in the environment leads to the formation of sulphuric acid, which considerably reduces the pH of groundwater. Where such acidic groundwaters make contact with concrete foundations, the main cementitious binder, calcium silicate hydrate (C-S-H), or calcium silicate hydrogel, the main component of cement paste, may be converted to thaumasite (a non-binding calcium carbonate silicate sulphate hydrate) resulting in deterioration and failure (e.g. Hobbs & Taylor, 2000[4]; Bensted, 1999[5]; Burkart, et al. 1999[6]). Greater awareness of the potential problems that thaumasite can cause has arisen with the increased use of limestone fillers in cements and limestone aggregates in concrete. Knowledge of the presence of sulphate-bearing species in Lias Group sediments is therefore crucial to foundation design and construction in such strata.

Previous mineralogical studies of the Lias Group

In view of its relatively good coastal and quarry exposure, it is perhaps surprising that studies of the mineralogy of the onshore Lias Group in England are relatively few. Several brief studies were carried out in the 1960’s (e.g. Hallam, 1960[7]; Cosgrove & Slater, 1966[8]). These indicated that clay mineral assemblages for the Lias mudstones typically contain illite, interlayered illite/smectite and smectite with minor kaolinite and chlorite.

Pye & Krinsley (1986)[9] produced a petrographic, geochemical and mineralogical study of the Whitby Mudstone Formation from the Cleveland Basin using the then recent development of backscattered scanning electron microscopy. Using a combination of techniques, they differentiated three facies; normal, restricted and bituminous. The normal facies (lower Grey Shale Member and upper Alum Shale Member) were rich in quartz, micas, chlorite and kaolinite, pyrite together with varying amounts of calcite and siderite (generally <10%) and traces of feldspar and carbonate-apatite. The restricted facies (upper Grey Shale Member) was mineralogically similar to the lower Grey Shale Member but with a lower siderite content. The same restricted facies (lower Alum Shale Member) contained more kaolinite but less quartz and chlorite than the restricted facies Grey Shales. The bituminous shale facies (Mulgrave Shale Member) was composed of quartz, kaolinite, mica, illite-smectite (I/S), chlorite, pyrite and calcite with subsidiary dolomite, feldspar and carbonate-apatite but no siderite. Textural evidence suggested that the authigenic mineralogy was predominantly early diagenetic and that the differences observed were due to changes in the prevailing sea bottom conditions.

As part of a site investigation for a low-level radioactive waste repository at Fulbeck, Lincolnshire, Bloodworth et al. (1987)[10] carried out an extensive mineralogical and lithogeochemical study of the Lower Lias (now equivalent to those deposits corresponding broadly with the Redcar Mudstone Formation of the Cleveland Basin, and Scunthorpe Mudstone/Blue Lias Formation plus Charmouth Mudstone Formation elsewhere), and Middle Lias sequence of interbedded mudstones and limestones (now corresponding broadly to the Staithes Sandstone plus Cleveland Ironstone formations in the Cleveland Basin, and Dyrham plus Marlstone Rock formations elsewhere). Clay mineral assemblages were found to be dominated by illite with subordinate kaolinite, minor chlorite and interlayered illite/smectite (I/S). Surface areas for the mudstones varied from 112 to 172 m2/g with a mean of 140 m2/g. Evolved gas and X-ray diffraction analyses revealed that pyrite was ubiquitous within the interval, typically forming 1–2% but occasionally reaching over 20% in some limestone samples. Trace amounts of gypsum were only sporadically detected.

Mitchell (1992)[11] carried out an XRD study of the clay mineralogy of a 200 m thick borehole sequence of Lias mudstones from the Copperhill Quarry, near Ancaster, Lincolnshire, in order to identify any potential clay ‘marker’ horizons or distinctive variations in clay mineral assemblage. The Lias here was found to be composed of the non-clays: quartz, mica and pyrite with traces of feldspar and calcite. Clay mineral assemblages were dominated by kaolinite with illite, chlorite and I/S. However, clay mineral abundances were based on a direct comparison of uncorrected peak intensity data. A more recent study by Bessa & Hesselbo (1997)[12] attempted to correlate the Lower Lias in southwest Britain using outcrop-based spectral gamma-ray spectroscopy. However, despite the fact that the gamma-ray signatures of such lithologies are predominantly determined by their clay mineralogy, no attempt was made to characterise the mineralogy of the mudstones. A similar study for the Cleveland Basin by Van Buchem et al. (1992)[13] presented limited clay mineralogical and geochemical data.

As a part of the ongoing BGS programme, ‘Engineering Geology of UK Rocks and Soils’, Kemp & Hards (2000)[2] investigated the mineralogy of Lias samples from two site investigation boreholes sited near the M5 motorway in Gloucestershire. They found non-clay mineral assemblages typically composed of carbonates (calcite and ankerite), quartz, feldspar (K-feldspar and albite), ‘mica’, gypsum and pyrite. Clay mineral assemblages were generally formed of illite (c. 40%), kaolinite (c. 35%), smectite (c. 20%) and chlorite (c. 5%). However, smectite contents were found to increase at certain levels to c. 30%.

This mineralogical study of a suite of Lias Group sedimentary rocks has generally confirmed the findings of previous workers. However, the wide geographic and stratigraphic distribution of the analysed samples has provided important new information, which will aid not only interpretation of the engineering behaviour of these rocks but also their diagenetic and geological histories. The engineering properties of the UK Lias will be heavily influenced by its clay mineralogy and in particular the presence of discrete smectite or illite/smectite. This study has shown that rocks from the West Midlands and southern England (the Worcester and Wessex Basins) are likely to contain smectite and therefore have greater shrink-swell potential than those rocks from the East Midlands (East Midlands Shelf) and northern England (the Cleveland Basin). However, the degree of shrink-swell is moderated by the high carbonate content typically found in southern rocks compared to those in the north. As indicated by XRD modelling, the small crystallite size of the other clay minerals; illite, kaolinite and chlorite will also significantly contribute to any volume change. The type of swelling clay species present is also useful for determining the depth of burial for the Lias Group across England. The I/S (90% illite) present in the Cleveland Basin suggests a 4 km maximum depth of burial, which corroborates earlier vitrinite reflectance, fluid inclusion and sonic velocity-based estimates. The greater proportion of smectite present in the I/S (80% illite) detected in the single sample from the East Midlands Shelf, suggests shallower burial to perhaps 3 km. However, the discrete smectite present in the Worcester and Wessex Basins indicates even shallower burial to no more than 2 km. Studies of basin maturity can therefore be used to predict likely engineering properties for the Lias Group rocks. The very common presence of pyrite, together with gypsum and jarosite in the Lias Group means that concrete engineering sited in these rocks potentially risk acid and sulphate attack and thaumasite formation. The Whitby Mudstone Formation in the Cleveland Basin together with the Blue Lias and Charmouth Mudstone formations in the Worcester and Wessex Basins show the greatest occurrence of sulphur-bearing species.

Results

Details of the samples taken for mineralogical analysis are given in Tables 3.1 and 3.2. The results of whole-rock XRD and surface area analyses are shown in Tables 3.3 and summarized in Figure 3.1. Analyses for <2 µm clay mineral XRD are given in Figure 3.2.

Whole-rock mineralogy and surface area

Up to thirteen different mineral phases were identified and quantified in the Lias Group samples. The non-clay mineralogy of the Lias Group rocks is composed of quartz, calcite, dolomite, feldspar (K-feldspar and albite), ‘mica’ (undifferentiated mica species), pyrite, gypsum and jarosite. The 'beef' sample from the Charmouth Mudstone Formation is almost totally composed of calcite with minor contaminants. In overall terms, the remaining samples are predominantly composed of quartz (6–52%, mean 30%), calcite (not detected, nd–81%, mean 18%), 'mica' (8–41%, mean 28%) and kaolinite (2–22%, mean 14%). The remaining minerals typically form <3% but may reach more elevated levels in selected samples. From whole-rock XRD analysis (Table 3.3 and Figure 3.1), it is apparent that the samples from Areas 3, 4 and 5 (East Midlands, Worcester and Wessex Basins, respectively) are highly calcareous (nd–81%, mean 31%) when compared to Area 1 samples (Cleveland Basin) samples, which often contain no carbonate species or are only poorly calcareous (nd–24%, mean 3%). Dolomite is also more common in samples from the south (nd–14%, mean 3%) than in the north (nd–3%, mean 1%). Similarly the samples from Areas 5 and 4 (Wessex and Worcester Basins) contain discrete smectite while only interlayered illite/smectite was detected in Areas 3 and 1 (East Midlands Shelf and Cleveland Basin) samples (see below). The southern batch, excluding the ‘beef’ sample, has a mean surface area of 110 m2/g but a relatively large range of values from 27 to 203 m2/g. Surface areas for the northern batch are smaller in comparison with a mean of 85 m2/g and a range of 24 to 134 m2/g. Quartz contents are approximately similar for Areas 5 and 4 (6–52%, mean 27%) and Areas 3 and 1 (22–47%, mean 33%) samples while feldspar contents (predominantly albite) are typically low (mean 2%) but reach 9% in a few samples. Of the sulphur-bearing species, pyrite appears to be commonly developed (nd–6%, mean 2%) throughout the Lias Group while weathering products gypsum and jarosite are more sporadic but form up to 12% in a few samples.

Clay mineralogy

The Lias Group samples from Area 1 (Cleveland Basin) show relatively uniform clay mineral assemblages (Figure 3.2). A typical <2 µm fraction is composed of 48% illite/smectite (I/S), 27% illite, 19% kaolinite and 6% chlorite. However, modelling of the XRD traces is hindered by the almost complete overlap of peaks from different clay mineral species. In addition to peak overlap problems, the broad peak profiles of the I/S, produced by its relatively small crystallite size, leaves only the d001 (c. 11 Å) adequately resolved for modelling. The I/S component was therefore necessarily modelled using the air-dry diffraction trace. Based on this limited data, modelling suggests that the I/S is 90% illite and 10% smectite R0 ordered interlayered clay, which has a mean defect-free distance of 3 layers (10Å units) and a size range of 1 to 15 layers. Illite has a mean defect-free distance of 7 layers and a size range of 1 to 28 layers. Chlorite was estimated to have similar mean defect-free distance of 7 layers (14Å units) and a size range of 1 to 32 layers. Kaolinite has a mean defect-free distance of 11 layers (7Å units) but a size range of 1 to 58 layers.

The clay mineralogy of the Harbury sample (LGD1) from the southwestern edge of Area 3 (East Midlands Shelf) is also composed of I/S, illite, kaolinite and chlorite, and appears to have similar characteristics to those already described for Area 1 (Cleveland Basin). However, the air-dry profile for the Area 3 (East Midlands Shelf) sample indicates a subtle shift in the I/S d001 to c. 12Å, indicating an increased smectite component in the interlayered species to perhaps 20%.

The Lias samples from Areas 4 and 5 (Wessex and Worcester Basins) differ from their northern counterparts as they contain discrete smectite and no detectable illite/smectite (Figure 3.2). Although they are otherwise similarly composed of illite, kaolinite and chlorite, they display a greater range of clay mineral concentrations. However, modelling suggests that a typical <2 µm fraction is composed of 37% illite, 26% smectite, 25% kaolinite and 11% chlorite. Modelling of the 'southern' sample XRD traces is similarly hindered by the almost complete overlap of peaks from different clay mineral species. In addition to peak overlap problems, the broad peak profiles of smectite, produced by its relatively small crystallite size, leaves only the d001 (17.0Å) peak adequately resolved for modelling. Illite has a marginally greater mean defect-free distance of 9 layers (10Å units) and a size range of 1 to 35 layers. Kaolinite and chlorite have approximately similar mean defect-free distances and size ranges to those models produced for the 'northern' samples. In comparison, smectite has a much smaller mean defect-free distance of 1.5 layers (14.5Å units) and a size range of only 1 to 5 layers.

Table 3.1    List of mineralogical samples from the East Midlands Shelf, Worcester Basin and Wessex Basin.
Sample No. Location NGR Area Basin Stratigraphy Detailed location Description
Formation Member (zone)
LGD1 Harbury, Warks. (quarry) SP 3862 5880 3 East Mids. Shelf Blue Lias Rugby Limestone upper waterfall near entrance Dark grey mudstone with shell frags.
LGD2 Northcot, Blockley, Gloucs. (quarry) SP 1795 3699 4 Worcester Charmouth Mdst N/A Ibex zone Dark grey mudstone with shell frags.
LGD3 Northcot, Blockley, Gloucs. (quarry) SP 1803 3404 4 Worcester Charmouth Mdst N/A Ibex zone Dark grey mudstone with shell frags.
LGD4 Ware Cliff, Lyme Regis, Dorset (coast) SY 3315 9138 5 Wessex Blue Lias (Angulata zone) below Specketty Lst band Dark grey, laminated mudstone
LGD5 Lyme Regis, Dorset (coast) SY 3337 9154 5 Wessex Charmouth Mdst Shales-with-Beef upper semicostatum Dark/pale grey, laminated mudstone
LGD5 Ware Cliff, Lyme Regis, Dorset (coast) SY 3337 9154 5 Wessex Charmouth Mdst Shales-with-Beef upper semicostatum Fibrous calcite
LGD6 Stonebarrow Hill, Dorset (coast) SY 3816 9264 5 Wessex Charmouth Mdst Belemnite Marl 2 m above base Medium grey siltstone
LGD7 Cain's Folly, Stonebarrow Hill, Dorset (coast) SY 3739 9288 5 Wessex Charmouth Mdst Black Ven Marl below lowermost Pavior Dark grey mudstone with shell frags.
LGD8 Cain's Folly, Stonebarrow Hill, Dorset (coast) SY 3739 9288 5 Wessex Charmouth Mdst Black Ven Marl 1.5 m above Pavior Limestone Dark grey, laminated mudstone
LGD9 Seatown, Dorset (coast) SY 4221 9162 5 Wessex Dyrham Eype Clay 4 m below Eype nodule bed Medium-dark grey mudstone
LGD10 Watton Cliff, Dorset (coast) SY 4529 9094 5 Wessex Bridport Sand Down Cliff Clay against Eypemouth fault Green siltstone
LGD11 A Robins Wood Hill, Gloucester (quarry) SO 3835 2149 4 Worcester Whitby Mdst N/A 5 m above Marlstone Rock Fm Green mudstone with black ‘root-like’ material
LGD11 B Robins Wood Hill, Gloucester (quarry) SO 3835 2149 4 Worcester Whitby Mdst N/A (3 m depth)(weathered) Pale green siltstone
LGD12 Robins Wood Hill, Gloucester (quarry) SO 3835 2149 4 Worcester Dyrham N/A 15 m below Marlstone Dark grey, laminated mudstone with shell frags.
Table 3.2    List of mineralogical samples from the Cleveland Basin.
Sample No. Location NGR Area Stratigraphy Detailed location Description
Formation Member (zone)
LGD13 Ravenscar (golf course) NZ 9799 0173 1 Whitby Mdst Alum Shale S. cliffs, Robin Hood’s Bay Dark grey, laminated mudstone
LGD14 Ravenscar (golf course) NZ 9829 0211 1 Whitby Mdst Mulgrave Shale S. cliffs, Robin Hood’s Bay Dark grey, laminated mudstone, fossil frags.
LGD15 Ravenscar NZ 9778 0223 1 Redcar Mdst Pyritous Shale (lower) S. cliffs, Robin Hood’s Bay Medium to dark grey, laminated mudstone
LGD16 Staithes NZ 7880 1886 1 Cleveland Irst 3 m below Avicula seam S. harbour cliffs Dark grey, laminated mudstone
LGD17 Staithes NZ 7857 1888 1 Staithes Sst S. harbour cliffs Pale/medium grey, massive slst/sst, fossil frags
LGD18 Runswick Bay (cliff) 1 Whitby Mdst Mulgrave Shale ‘jet’ workings 2 m above beach Dark grey, laminated mudstone, oxidised pyrite
LGD19 Kettleness (cliff) NZ 8318 1603 1 Whitby Mdst Grey Shales 1 m above base of W.M.F. Dark grey, laminated mudstone
LGD20 Kettleness (cliff) NZ 8317 1599 1 Whitby Mdst Grey Shales 4 m above base of W.M.F. Dark grey, laminated mudstone
LGD21 Kettleness
(former alum quarry)
NZ 8346 1586 1 Whitby Mdst Alum Shale  ?? Dark grey, laminated mudstone, oxidised pyrite
LGD22 Kettleness
(former alum quarry)
NZ 8321 1603 1 Whitby Mdst Mulgrave Shale 5 m below top of W.M.F. Dark grey, laminated mudstone
LGD23 Robin Hood’s Bay (harbour) 1 Redcar Mdst Ironstone Shale S. of current sea-wall works Dark grey, laminated mudstone
LGD24 Boggle Hole NZ 9644 0313 1 Redcar Mdst Calcareous Shales (upper) cliff near Stoupe Beck Medium grey, laminated mudstone
LGD25 Boggle Hole NZ 9631 0307 1 Redcar Mdst Calcareous Shales (lower) wave-cut plat., Stoupe Beck Dark grey, laminated mudstone
Table 3.3    Summary of whole-rock XRD and surface area analyses (nd = not detected).
Sample Location Area

% mineral

S.A. m2/g
quartz
calcite
pyrite
dolomite
mica
K-feldspar
kaolin
chlorite
smectite
albite
gypsum
jarosite
illite/smectite
LGD1 Harbury, Warks. (quarry) 3 22 47 2 1 20 nd 4 nd nd 1 nd nd 3 114
LGD2 Northcot, Blockley, Gloucs.(quarry) 4 52 1 1 nd 22 nd 20 nd 1 3 nd nd nd 86
LGD3 Northcot, Blockley, Gloucs.(quarry) 4 38 1 1 nd 39 nd 21 nd nd nd nd nd nd 97
LGD4 Ware Cliff, Lyme Regis, Dorset (coast) 5 13 51 4 5 22 nd 2 nd nd nd 3 nd nd 137
LGD5 Ware Cliff, Lyme Regis, Dorset (coast) 5 22 20 5 1 31 1 9 nd 2 nd 8 1 nd 203
LGD5 'beef' Ware Cliff, Lyme Regis, Dorset (coast) 5 2 86 nd nd 6 nd 1 nd 1 nd 1 3 nd 23
LGD6 Stonebarrow Hill, Dorset (coast) 5 10 56 2 12 17 nd 2 nd nd 1 nd nd nd 77
LGD7 Cain's Folly, Stonebarrow Hill, Dorset (coast) 5 22 27 4 4 33 nd 10 nd nd nd nd nd nd 123
LGD8 Cain's Folly, Stonebarrow Hill, Dorset (coast) 5 17 23 6 14 33 nd 6 nd nd nd 1 nd nd 179
LGD9 Seatown, Dorset (coast) 5 44 3 1 nd 28 nd 15 nd 2 7 nd nd nd 98
LGD10 Watton Cliff, Dorset (coast) 5 51 21 nd nd 17 1 5 nd 2 3 nd nd nd 106
LGD11A Robins Wood Hill, Gloucester (quarry) 4 12 69 nd nd 15 nd 4 nd nd nd nd nd nd 77
LGD11B Robins Wood Hill, Gloucester (quarry) 4 6 81 nd nd 8 nd 2 nd nd nd 1 2 nd 27
LGD12 Robins Wood Hill, Gloucester (quarry) 4 40 nd 1 nd 29 nd 22 nd 3 3 2 nd nd 116
LGD13 Ravenscar (golf course) 1 30 nd 5 nd 42 nd 21 nd nd nd nd nd 2 96
LGD14 Ravenscar (golf course) 1 28 3 6 nd 38 nd 21 nd nd nd nd nd 4 85
LGD15 Ravenscar 1 29 8 3 nd 37 nd 21 nd nd nd nd nd 2 86
LGD16 Staithes 1 36 nd 2 nd 36 nd 17 nd nd 7 nd nd 2 66
LGD17 Staithes 1 40 24 1 3 13 nd 10 nd nd 9 nd nd nd 24
LGD18 Runswick Bay (cliff) 1 25 nd 3 nd 28 nd 10 nd nd 1 10 12 11 78
LGD19 Kettleness (cliff) 1 39 nd 3 nd 31 nd 21 nd nd 4 nd nd 2 66
LGD20 Kettleness (cliff) 1 31 nd 2 nd 40 nd 18 nd nd nd nd nd 9 96
LGD21 Kettleness (former alum quarry) 1 32 nd nd nd 38 nd 20 nd nd nd nd nd 10 134
LGD22 Kettleness (former alum quarry) 1 24 nd 1 nd 36 nd 17 nd nd nd 8 4 10 117
LGD23 Robin Hood's Bay (harbour) 1 24 6 2 nd 42 nd 21 1 nd nd 1 nd 3 100
LGD24 Boggle Hole 1 38 1 1 3 37 nd 17 nd nd 1 nd nd 2 79
LGD25 Boggle Hole 1 47 nd nd 1 32 nd 15 nd nd 1 nd nd 4 76
Figure 3.1    Summary of whole-rock mineralogical analysis and surface-area.
Figure 3.2    Summary of clay mineralogical analysis.

Discussion

Mineralogical analysis of Lias Group samples from various sites representing a relatively large geographic and stratigraphic range have generally similar mineralogies to those described in previous studies (e.g. Kemp & Hards, 2000[2]; Mitchell, 1992[11]; Bloodworth et al., 1987[10]; Pye & Krinsley, 1986[9]; Cosgrove & Slater, 1966[8]). Non-clay mineral assemblages are typically composed of carbonates (calcite and dolomite), quartz, feldspar (albite and occasional Kfeldspar), ‘mica’, pyrite, gypsum and jarosite. Clay mineral assemblages are generally formed of illite, smectite or illite/smectite, kaolinite and chlorite.

The relatively complex mineralogies of the Lias Group samples are difficult to quantify, even by employing state-of-the-art software modelling packages. For this reason the quoted mineral concentrations must be regarded with some caution. However, calculations using approximate values for %clay (from whole-rock XRD), the clay mineral concentrations from <2 µm XRD analysis and assuming theoretical surface area values for the individual clay minerals, reveal similar whole-rock surface area values to those determined empirically (Figure 3.3).

Figure 3.3    Calculation of theoretical surface area.

Nevertheless, this study has shown that Lias Group rocks of southern England (Wessex and Worcester Basins) and northern England (Cleveland Basin) have importantly different mineralogical characteristics. The mineralogy of the Harbury sample (Area 3 — Blue Lias F.) has an intermediate character, sharing similar characteristics to both northern and southern samples.

Southern rocks often contain large quantities of carbonate (principally as calcite with minor dolomite) while those from northern England contain little or no carbonate. [The only sample from northern England to contain appreciable calcite is the silt/sandstone from the Staithes Sandstone Formation]. The more calcareous nature of the southern rocks and sandier nature of those in the north has been noted previously, but not explained, by Anderton et al. (1979)[14]. Lithologically, both the analysed sample batches are dominated by mudstones with relatively few siltstones/sandstones and both were observed to contain fossil fragments. It would, therefore, appear that carbonate development is not lithologically controlled. Petrographic analysis is necessary to elucidate whether this apparent difference in mineralogy is due to different sediment source(s) or a different diagenetic overprint between the northern and southern rocks.

This study has also highlighted an important difference in the swelling clay species present in the Lias Group rocks. The samples from Areas 5 and 4 (Wessex and Worcester Basins) contain discrete smectite whereas illite/smectite, I/S (90% illite) is present in the Areas 3 and 1 (East Midlands Shelf and Cleveland Basin) samples. Modelling suggests that all the clay minerals present in the Lias Group have small mean defect-free distances, typically <10 layers thick. Such small crystallite sizes indicate that all species will provide an input to the surface area of the rock. However, the difference in swelling clay species does help to explain the larger surface area values for the southern batch (mean 110 m2/g) compared with the northern batch (mean 85 m2/g) despite the presence of more coarse-grained siltstone samples in the south. The smaller surface area of I/S compared to smectite might therefore be expected to produce a greater degree of swell-shrink in the southern strata. However, the high concentration of calcite in the southern samples 'dilutes' the effect of the smectite surface area. In trying to relate the mineralogy and the engineering properties of the Lias Group, it is therefore imperative not only to determine the quantity and type of clay minerals present but also the quantity of calcite present. It is also necessary to know whether the calcite is present as a cement, which will influence engineering behaviour, and/or as shell fragments, which would have a reduced effect. The only previous petrographic study of the Lias Group mudstones (Pye & Krinsley, 1986[9]) suggests that calcite and siderite are present as randomly dispersed rhombs, irregularly shape grains and patches of intergranular cement in the more silty sediments. No primary biogenic carbonate was observed as foraminifera, coccoliths or shell debris. Interestingly these authors also note that although the mudstones have a well-developed lamination and high degree of parallelism shown by micas and clay minerals, they are not notably fissile. During weathering they split into flaggy slabs rather than sheets. Such behaviour was attributed to the high proportion of authigenic minerals (carbonate, pyrite and kaolinite), which act as cements and bind adjacent laminae together.

The difference between the type of swelling clay present in the northern and southern samples also suggests differences in their burial histories (Kemp, et al., 2005[3]). During burial of sedimentary sequences, the clay minerals contained in mudstones and shales undergo diagenetic reactions in response to increasing depth and temperature. Quantitatively, the most important change is the progressive reaction of smectite to form illite via a series of intermediate illite/smectite (I/S) mixed-layer minerals. In general, progressive changes are irreversible so that where basinal sequences have been inverted clay mineral evidence of the maximum burial depth is retained and can be used to estimate the amount of uplift. According to the Basin Maturity Chart of Merriman & Kemp (1996)[15], the presence of I/S (90% illite) in the rocks from the Cleveland Basin suggests burial depths of 4 km, assuming a 'normal' geothermal gradient of 25–30°C/km. The I/S (80% illite) in the sample from Area 3 (East Midlands Shelf), therefore, suggests shallower burial to perhaps 3 km while the presence of smectite and kaolinite suggest that the mudstones from southern England are more immature and have only been buried to depths of less than c. 3 km if the same geothermal gradient is assumed. These estimates are compared with those obtained independently from geological evidence in Table 3.4. See also section 6.2.2.

Table 3.4    Estimated maximum burial depths by Area based on geological and mineralogical evidence. (Chapter 2, Geology; * Merriman & Kemp, 1996[15]).
Area 1 Area 2 Area 3 Area 4 (east) Area 4 (west) Area 5 Area 6
Burial depth (m) — based on geology # 2000 550 1000 1100 600 1500 600
Burial depth (m) — based on mineralogy* 4000 3000 <3000 <3000

Vitrinite reflectance data for Middle Jurassic coals from Area 1 (Cleveland Basin) show reflectivities of c. 0.85% and a rank equivalent to high volatile bituminous coals (Hemingway & Riddler, 1982[16]). Barnard & Cooper (1983)[17] used a combination of vitrinite reflectance and spore colouration indices to conclude that the Middle Jurassic had reached a maximum palaeotemperature of 95°C in the central part of the Cleveland Basin. Furthermore, c. 80°C palaeotemperatures for the Middle Jurassic were obtained from fluid inclusion microthermometry from sphalerite grains (Hemingway & Riddler, 1982[18]). Together these palaeotemperatures were taken to indicate a palaeo-depth of c. 2.5 km for the base of the Middle Jurassic (Hemingway & Riddler, 1982[18]). The 440 m thickness of the Lias Group in the Cleveland Basin thus produces a maximum depth of burial of c. 3 km. More recent and detailed modeling (Holliday, 1999[19]) indicate that if the time of maximum burial was end-Cretaceous, between c. 2200 and 3000 m of Late Jurassic to Late Cretaceous strata have been removed from the main basin depositional centre, assuming a lack of over pressuring. Alternatively, if the time of maximum burial was during the mid-Cenozoic, the observed palaeotemperatures indicates that between 2300 and 3200 m of Late Jurassic, Cretaceous and Palaeogene strata have been removed. Again, if the thickness of the Lias Group is added, a maximum depth of burial of c. 4 km is indicated and in agreement with the clay mineralogical data from this study. A combination of clay mineralogy, sonic log studies and palaeotemperature assessments suggest that the observed high palaeotemperatures for the Cleveland Basin can be accounted for by deep burial and there is no need to infer a local heating event.

Subsidence history plots and hydrocarbon potential studies reveal much shallower depth of burial for the Lias Group in the Worcester and Wessex Basins. Chadwick & Evans (1995)[20] used mudstone densities to suggest that 1650 m of overburden had been removed from the Mercia Mudstone Group in the Kempsey borehole, south of Worcester whereas perhaps 1200 m had been removed from the eastern part of the basin. It would, therefore, appear that the Lias Group has only been buried to perhaps 1.5 km in the Worcester Basin. Calculated organic maturity values of <0.50% R0 (Ebukanson & Kinghorn, 1986[21]) and organic geochemical analyses (Colter & Havard, 1981[22]) for the Base Lias of the Wytch Farm Oilfield, Dorset suggest organic immaturity. Although maturities are heavily influenced by the Purbeck-Isle of Wight Disturbance, burial thermal history projections based on such data suggest a maximum burial of c. 2 km and peak palaeotemperatures of c. 75°C for the locations sampled in the Wessex Basin for this study.

Mineralogical analysis also suggests that pyrite is very commonly developed throughout the Lias Group, typically forming 2% of the rock. Pye & Krinsley (1986)[9] observed authigenic pyrite occurring as framboids and larger euhedral crystals, particularly in the Mulgrave Shale Member of the Whitby Mudstone Formation. The other sulphur-bearing minerals, gypsum and jarosite are more sporadically developed but can form up to 12% of the rock. Jarosite and gypsum typically form as weathering products of pyrite. Although it is difficult to comment with such a small sample batch, stratigraphically it would appear that the Whitby Mudstone Formation together with the Blue Lias and Charmouth Mudstone Formations show the greatest occurrence of sulphur-bearing species. Concrete engineering sited in rocks from these formations therefore potentially run the greatest risk of acid and sulphate attack and thaumasite formation.

Conclusions

This mineralogical study of a suite of Lias Group sedimentary rocks has generally confirmed the findings of previous workers. However, the wide geographic and stratigraphic distribution of the analysed samples has provided important new information, which will aid not only interpretation of the engineering behaviour of these rocks but also their diagenetic and geological histories. The engineering properties of the UK Lias are heavily influenced by its clay mineralogy and in particular whether discrete smectite or illite/smectite is present. This study has shown that rocks from the West Midlands and southern England (the Worcester and Wessex Basins) are likely to contain smectite and, therefore, have greater shrink-swell potential than those rocks from the East Midlands (East Midlands Shelf) and northern England (the Cleveland Basin). However, the degree of shrink-swell is moderated by the high carbonate content typically found in southern rocks compared to those in the north. As indicated by XRD modelling, the small crystallite size of the other clay minerals; illite, kaolinite and chlorite will also significantly contribute to any volume change and shrink/swell behaviour. The type of swelling clay mineral present is also useful for determining the depth of burial for the Lias Group across England. The illite/smectite, I/S (90% illite) present in the Cleveland Basin suggests a 4 km maximum depth of burial, which corroborates earlier vitrinite reflectance-, fluid inclusion- and sonic velocity-based estimates. The greater proportion of smectite present in the I/S (80% illite) detected in the single sample from the East Midlands Shelf, suggests shallower burial to perhaps 3 km. However, the discrete smectite present in the Worcester and Wessex Basins indicates even shallower burial to no more than 2 km. Studies of basin maturity can therefore be used to predict likely engineering properties for the Lias Group rocks. The common presence of pyrite, together with gypsum and jarosite in the Lias Group means that concrete engineering works sited in these rocks potentially risk acid and sulphate attack and thaumasite formation. The Whitby Mudstone Formation in the Cleveland Basin, together with the Blue Lias and Charmouth Mudstone Formations in the Worcester and Wessex Basins, show the greatest occurrence of sulphur-bearing species.

References

  1. Kemp, S J, and McKervey, J A. 2001. The mineralogy of mudrocks from the Lias Group of England. British Geological Survey Internal Report, IR/01/124.
  2. 2.0 2.1 2.2 Kemp, S J, and Hards, V L. 2000. The mineralogy of Lower Jurassic (Lias) mudstones from the M5, Gloucestershire. British Geological Survey Technical Report WG/00/11
  3. 3.0 3.1 Kemp, S J, Merriman, R J, and Bouch, J E. 2005. Clay mineral reaction progress — the maturity and burial history of the Lias Group of England and Wales. Clay Minerals, 40, pp.43–61.
  4. Hobbs, D W, and Taylor, M G. 2000. Nature of the thaumasite sulfate attack mechanism in field concrete. Cement and Concrete Research, 30 (4), 529–533.
  5. Bensted, J. 1999. Thaumasite — background and nature in deterioration of cements, mortars and concretes. Cement and Concrete Composites 21 (2), 117–121.
  6. Burkart, B, Goss, G C, and Kern, J P. 1999. The role of gypsum in production of sulfate-induced deformation of lime-stabilized soils. Environmental and Engineering Geoscience, 5 (2), 173–187.
  7. Hallam, A. 1960. A sedimentary and faunal study of the Blue Lias of Dorset and Glamorgan. Philosophical Transactions of the Royal Society, London, Vol. B243, 698, 1–44.
  8. 8.0 8.1 Cosgrove, M E, and Slater, D L. 1966. The stratigraphical distribution of kaolinite in the post-Armorican formations of South-West England. Proceedings of the Ussher Society, 1, 5, 249–252.
  9. 9.0 9.1 9.2 9.3 Pye, K, and Krinsley, D H. 1986. Microfabric, mineralogy and early diagenetic history of the Whitby Mudstone Formation (Toarcian), Cleveland Basin, UK. Geological Magazine, 123, 191–203.
  10. 10.0 10.1 Bloodworth, A J, Kemp, S J, Inglethorpe, S D J, and Morgan, D J. 1987. Mineralogy and lithochemistry of strata beneath proposed low-level radioactive waste site at Fulbeck, Lincolnshire; Report to Sir Alexander Gibb and Partners, Site Investigation Consultants. British Geological Survey, Mineralogy and Petrology Technical Report No. 87/15/C.
  11. 11.0 11.1 Mitchell, C J. 1992. Clay mineralogy of the Lias, Copperhill Quarry, Grantham. British Geological Survey Technical Report, WG/92/9.
  12. Bessa, J L, and Hesselbo, S P. 1997. Gamma-ray character and correlation of the Lower Lias, SW Britain. Proceedings of the Geologists’ Association, 108, 113–129.
  13. Van Buchem, F S P, Melnyk, D H, and McCave, I N. 1992. Chemical cyclicity and correlation of Lower Lias mudstones using gamma ray logs, Yorkshire, UK. Journal of the Geological Society, London, 149, 991–1002.
  14. Anderton, R, Bridges, P H, Leeder, M, and Selwood, B W. 1979. A Dynamic Stratigraphy of the British Isles: A Study in Crustal Evolution. George Allen & Unwin Ltd., London.
  15. 15.0 15.1 Merriman, R J, and Kemp, S J. 1996. Clay minerals and sedimentary basin maturity. Mineralogical Society Bulletin, 111, 7–8.
  16. Hemingway, J E, and Riddler, G P. 1982. Basin inversion in North Yorkshire. Transactions of the Institution of Mining & Metallurgy, B91, 175–186.
  17. Barnard, C E, and Cooper, B S. 1983. A review of geochemical data related to the northwest European Gas Province. 19–33 in Petroleum geochemistry and exploration of Europe. BROOKS, J. (editor). Geological Society (London) and Blackwell Scientific Publishers, Oxford.
  18. 18.0 18.1 Hemingway, J E, and Riddler, G P. 1982. Basin inversion in North Yorkshire. Transactions of the Institution of Mining & Metallurgy, B91, 175–186.
  19. Holliday, D W. 1999. Palaeotemperatures, thermal modelling and depth of burial studies in northern and eastern England. Proceedings of the Yorkshire Geological Society, 52 (4), 337–352.
  20. Chadwick, R A, and Evans, C J. 1995. The timing and direction of Permo-Triassic extension in southern Britain. 161–192 in Permian and Triassic rifting in Northwest Europe. BOLDY, S A R. (editor). Geological Society Special Publication No. 91.
  21. Ebukanson, E J, and Kinghorn, R R F. 1986. Maturity of organic matter in the Jurassic of southern England and its relation to the burial history of the sediments. Journal of Petroleum Geology, 9, 259–280
  22. Colter, V S, and Havard, D J. 1981. The Wytch Farm oilfield, Dorset. 494–503 in Petroleum Geology of the Continental Shelf of NW Europe. ILLING, L V, and HOBSON, D G. (editors). Hayden & Sons, London.