origin, diagenesis, and mineralogy of chlorite …clays.org/journal/archive/volume...

Clays and Clay Minerals, Vol. 41, No. 2, 240-259, 1993.

ORIGIN, DIAGENESIS, AND MINERALOGY OF CHLORITE MINERALS IN DEVONIAN LACUSTRINE MUDROCKS,

ORCADIAN BASIN, SCOTLAND

S. HILLIER l

Department of Geology, University of Southampton, Southampton SO9, 5BP, U.K.

Laboratoire de Grologie, Ecole Normale Superieur, 24 rue Lhomond, 75231 Paris, France

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

XRD and electron microprobe analyses indicate that phases intermediate between corrensite and chlorite are probably mixed-layer chlorite/corrensite with a tendency towards segregation of layer types. Chemically, the conversion of corrensite to chlorite involves an increase in A1 for Si substitution in tetrahedral sites, but there is no change in the Fe/Mg ratio of octahedral cations. There is also no relationship of mixed-layer proportions to paleotemperature; only a general paleotemperature interval of approximately 120 ~ to 260~ in which a range of phases between corrensite and chlorite occurs. Chlorite polytypes are exclusively IIb, indicating the formation of this polytype at diagenetic temperatures.

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

Key Words--Chlorite, Corrensite, Diagenesis, Dolomite, Vitrinite Reflectance.

I N T R O D U C T I O N

Within the realm ofdiagenesis and low temperature metamorphism the geological occurrences of Mg-rich chlorite and corrensite (1:1 regularly interstratified chlorite/smectite) can be most simply divided into two types. The first includes those associated with volcaniclastic sediments (Almon et aL, 1976; Helmond and van de Kamp, 1984; Chang et aL, 1986; Inoue, 1987; Inoue and Utada, 1991) or various types of altered igneous rocks (Bettison and Schiffman, 1988; Schiff- man and Fridleifsson, 1991). The second, those that are associated with ancient marine evaporites (Droste, 1963; Jeans, 1978; Bodine, 1985; Bodine and Masden, 1987; Lippmann and Pankau, 1988), or carbonates (Peterson, 1961; Fraser et al., 1973; Rao and Bhatta- charya, 1973), or lacustrine facies (April, 1981).

For the first or 'marie' association, the origin of corrensite and chlorite is related to diagenesis or low temperature hydrothermal alteration. Essentially, this appears as a sequence of minerals from trioctahedral smectite in the lowest temperature zone to chlorite in the highest. Within this sequence, the first occurrence of the intermediate mineral corrensite is typically at about 100~ (e.g., Inoue and Utada, 1991). Some ex-

Present address: Geologisches Institut Universit~it Bern, Baltzerstrasse 1, CH3012, Bern, Switzerland.

Copyright �9 1993, The Clay Minerals Society

amples are interpreted as a more or less continuous series of mixed-layer chlorite/smectite (C/S) minerals (Helmold and van de Kamp, 1984; Chang et aL, 1986), analogous to the classic model for the dioctahedral smectite to illite reaction. Other examples are interpreted as prograde sequences of the three phases smectite, corrensite, and chlorite, but in zones that may be grossly overlapping (Inoue, 1987; Inoue and Utada, 1991) and with various interstratifications and/or in- tergrowths of these minerals being the norm. The emerging consensus seems to be with the latter interpretation, wherein corrensite is considered a discrete phase, rather than an interstratification of smectite and chlorite layers (Reynolds, 1988).

In contrast, for the "evaporite-carbonate" association, the origin of chlorite and corrensite is classically related to formation in an evaporitic environment either during, or shortly after, deposition. The proposed mechanisms for corrensite and chlorite formation in- clude neoformation from concentrated hypersaline brines or, alternatively, transformation of detrital clay minerals (Lucas, 1962; Jeans, 1978; Hauff, 1981; Fish- er, 1988; Lippmann and Pankau, 1988). The evidence for such interpretations is basically two-fold: 1) Mg- rich chlorite and corrensite occur in association with evaporitic facies from many different sedimentary basins worldwide, yet they are comparatively rare in other facies deposited under normal marine or fresh

240

Vol. 41, No. 2, 1993 Chlorite minerals in Devonian mudrocks 241

5 0 k m I

, • ~ Orkney

4 ~ 380 478

856,858 824 S_." ,132

=r

t h

Figure 1. Location of samples sites in the Devonian, Or- C~u~a d B:~nb~eOr~hen~ Sc~

samples of mixed layer chlorite minerals of Figures 4, 5, and 6, and Table 4. All samples are grey-green lacustrine mudrocks.

water conditions; 2) the distr ibution of clay minerals is often zoned, passing from dioctahedral clays at the basin margin, through an intermediate zone characterised by corrensite, and eventually to a chlorite zone in the basin centre. This zonation generally follows that o f the evaporites and other facies and, hence, the changing chemistry of the deposit ional and/or early diagenetic environment.

These studies of the 'evapori te-carbonate ' associa- t ion imply that corrensite may form at surface temperatures. However, as pointed out by Bodine and Madsen (1987), there is no evidence from any modern evaporitic environment to support this contention. The Mg-rich clay minerals which form at surface temperatures, in both alkaline saline lakes and restricted marine basins, are invariably varieties of tr ioctahedral smectite or the fibrous clay minerals palygorskite and sepiolite (Jones, 1986; Chameley, 1989). This observation led Bodine and Madsen (1987) to postulate that minerals they interpreted as C/S (with varying proport ions of expandable layers) from a Pennsylvanian evapori te cycle in the Paradox Basin, Utah, may have formed from precursor authigenic Mg-rich smectites

1,6=3.0%Ro SC 8-9 (dry gas zone) ,'~ ~,x~

0.7-1.5%Ro SC 6-7 (oil window...+) , , / , /~ ' r

~> 3.0%R0 SC 10-11 (rnetamorpho~ed) ~ < ~ 1 "~ ~'

I r 50kin

Figure 2. Generalized organic maturation map for the Or- cadian Basin, based on spore color (SC) and vitrinite reflectance (Ro) data from Hillier and Marshall (1992). Actual vitrinite reflectance values for the samples described are included in Tables 1 and 2.

during burial diagenesis. They did, however, stress that there was no direct evidence for the former presence of smectite. Such a 'uniformitar ian ' explanation suggests immedia te parallels, and greater analogy, with the smectite-corrensite-chlorite sequence o f the 'mafic ' association. It further implies that the occurrence of corrensite may always be an indicator of diagenetic grade, as originally proposed by Kiibler (1973).

This paper documents the occurrence, structure, chemistry, and origin of corrensite and chlorite and what appear to be intermediate mixed-layer minerals, in Devonian lacustrine mudrocks from the Orcadian Basin, Scotland (Figure 1). The Orcadian lacustrine sediments were deposited in lakes that were frequently desiccated and show evidence for the former presence of evaporites (Parnell, 1985; Astin and Rogers, 1991; Rogers and Astin, 1991). As such, the occurrence of corrensite and chlorite in these rocks falls into the "evapori te-carbonate" category.

242 Hillier Clays and Clay Minerals

Table 1. Whole rock mineral composition, vitrinite reflectance and clay mineralogy of the < 2 um fraction of Orcadian mudrocks.

K- Correrl- No. Quartz Albite feldspar Dolomite Calcite Siderite Dioet Trioct Pyrite Tolal Ro% Ulite Chlorite site Kaolinite

333 33 4 7 11 11 40 3 109 0.9 e 94 3 354 33 11 8 1 35 I0 98 0.9 a 73 15 12

1013 28 9 18 8 32 5 100 0.9 a 88 5 409 25 7 5 3 3 40 12 95 0.9 a 68 32 353 28 10 5 10 40 16 1 110 0.9 a 72 11 17 339 30 11 8 10 30 17 3 109 1.0 e 60 17 26 340 32 9 7 15 28 12 2 105 1.0 e 62 12 26 342 18 9 5 49 10 91 1.0 e 93 7 345 35 17 8 14 3 31 108 1.0 e 96 346 41 8 7 22 31 2 111 1.0e 93 348 27 6 5 24 41 103 1.0 e 95 385 16 10 8 19 8 46 107 1.0 e 92 389 34 14 9 15 40 112 1.0 e 93 478 22 3 4 20 53 102 1.0 a 99 1 464 18 6 8 18 7 32 8 97 1.1 a 82 18 465 15 3 8 17 2 58 103 1.2 a 100 471 26 5 4 19 52 106 1.2 a 100 928 29 3 5 28 38 103 1.2 a 100 315 40 4 1 48 2 95 1.3 a 87 870 35 5 57 97 1.3 a 85

5 14 13 8 9 3 36 13 96 1.4 e 48 52 380 15 10 10 5 1 43 12 2 98 1.4 a 55 45 535 26 13 12 4 5 30 12 102 1.5 a 50 50 448 14 10 7 3 6 41 18 99 1.8 a 60 40 449 12 10 5 7 4 37 20 95 1.9 a 52 48 856 20 11 6 4 3 45 19 108 1.9 a 55 45 858 18 8 4 4 8 41 18 101 1.9 a 63 37 968 27 17 10 1 30 10 95 2.0 a 76 24 883 13 19 17 15 3 27 14 108 2.3 a 50 27 23 969 25 13 14 30 15 97 2.4 a 49 51 795 25 10 15 8 5 26 13 102 2.9 a 50 50 876 13 20 14 l l l 24 13 1 97 2.9 a 54 46 740 23 11 8 2 6 32 13 95 3.1 a 49 51 494 25 12 17 4 19 15 1 93 3.7 a 40 60 794 17 8 4 10 11 32 16 98 3.7 a 56 44 483 22 12 16 8 1 20 17 2 98 4.3 a 28 72 432 17 9 14 6 2 28 17 4 97 4.7 a 40 60 426 11 14 22 25 24 96 5.0 e 10 90 824 25 18 14 2 20 17 3 99 6.1 a 28 72 866 29 14 13 27 21 104 7.2 a 48 52

13 15

Dioct = total dioctahedral clay. Trioct = total trioctahedral clay. Ro % = mean vitrinite reflectance, a = actual, e = estimate.

M A T E R I A L S A N D M E T H O D S

All samples d i s c u s s e d are grey-green m u d r o c k s col- lected at ou tc rop in the Orcad ian Basin and are pa r t o f a m u c h larger suite o f 444 samples o f b o t h m u d r o c k s and sands tones s tudied by Hil l ier (1989), whe re in the

full detai ls o f localit ies and m e t h o d s are given. The grey-green m u d r o c k s d i scussed here represen t a facies which was s tud ied in the m o s t detail . They are r icher in clay minera l s and poo re r in ca rbona te than the associa ted ca rbona te l amin i te facies (Donovan , 1980). The select ion o f as cons t an t a l i tho-facies as poss ib le

was d e e m e d essent ia l to facili tate subsequen t m ine r - alogical compar i son . In add i t i on , the par t icular s amples were selected because the i r d iagenet ic grade was well cons t ra ined , mos t ly by m e a s u r e m e n t o f v i t r in i te

ref lectance on the actual s ample itself, or by an es t ima te based on v i t r in i te ref lectance da ta f rom o the r s amples at the same, or a neighbor ing, locali ty (Tables 1 a n d 2). Deta i led v i t r in i te ref lectance and spore color m a p s for the Orcad ian Basin are g iven in Hil l ier and Marsha l l (1992) and Figure 2 is a s u m m a r y map .

Crushed samples were d i spe r sed in de - ion i sed wa te r using an ul t rasonic p robe and the < 2 # m fract ion was separa ted by gravi ty settling. The samples were Mg saturated, w as h ed free o f ch lor ide and s e d i m e n t e d by centrifuge. Or ien ta t ed samples were p repa red by the smear m e t h o d and X- ray diffract ion ( X R D ) pa t t e rns

were r ecorded in the a i r -dr ied state, af ter glycolat ion (vapor pressure me thod) , a n d after heat ing to 375 ~ a n d 550~ for 1 hr. Es t imates o f the re la t ive abundance o f clay minera l s in the < 2 t tm f rac t ion were m a d e by a

Vol. 41, No. 2, 1993

Table 2.

Chlorite minerals in Devonian mudrocks

Major element XRF analyses and vitrinite reflectance of Orcadian

No SIO2 TIO2 A1203 b2E203 MNO MGO CAO NA20 K20 P205 S LOI Ro%

mudrocks.

243

333 53.98 0.63 14.24 3.68 0.08 3.40 8.39 0.83 3.96 0.11 0.86 9.93 0.9 e 354 63.51 0.75 15.99 5.38 0.04 4.02 0.85 2.22 4.04 0.12 0.00 3.07 0.9 a

1013 48.55 0.59 12.89 4.11 0.09 4.92 9.70 0.09 3.48 0.09 0.20 15.30 0.9 a 409 55.36 0.71 15.79 6.06 0.07 4.82 3.91 1.46 4.33 0.13 0.15 7.23 0.9 a 339 57.99 0.52 12.60 4.20 0.05 4.73 6.19 1.75 3.16 0.18 1.10 7.60 1.0 a 340 57.46 0.51 12.25 4.37 0.06 4.96 6.96 1.56 3.01 0.12 1.07 7.76 1.0e 342 57.04 0.92 19.43 6.06 0.05 3.56 0.69 1.99 5.30 0.21 0.00 4.75 1.0 e 345 55.43 0.54 11.94 2.98 0.05 3.22 9.09 1.36 3.26 0.11 0.24 11.80 1.0 e 346 55.90 0.55 11.57 3.78 0.05 3.92 6.52 1.50 2.94 0.11 0.99 12.29 1.0 e 348 49.75 0.57 12.96 4.87 0.07 4.80 8.45 1.35 3.48 0.12 0.00 13.57 1.0 e 353 55.01 0.70 14.74 5.74 0.07 4.44 6.16 1.86 3.44 0.11 0.44 7.33 1.0 e 385 47.36 0.87 17.45 6.49 0.06 3.91 4.74 1.68 5.34 0.12 0.00 11.98 1.0 e 389 59.20 0.73 15.00 3.95 0.04 2.57 4.19 1.86 4.62 0.10 0.00 7.74 1.0 e 478 53.47 0.71 16.75 4.41 0.08 4.65 6.23 0.71 5.47 0.11 0.00 7.40 1.0a 464 48.08 0.68 14.42 5.18 0.09 4.49 8.68 1.11 5.20 0.12 0.40 11.60 1.1 a 465 49.47 0.80 19.59 3.29 0.07 4.06 5.47 0.60 7.22 0.13 0.00 9.30 1.2 a 471 54.39 0.73 16.96 4.41 0.07 3.85 5.76 0.87 5.54 0.13 0.00 7.30 1.2 a 928 52,89 0.58 12.69 3.10 0.06 4.95 8.06 0.59 4.41 0.11 0.41 12.20 1.2 a 315 65.79 0.68 17.31 2.42 0.02 1.12 1.20 0.14 5.18 0.16 1.16 4.80 1.3 a 870 63.93 0.82 18.87 4.00 0.03 1.24 0.49 0.14 5.67 0.17 0.05 4.80 1.3 a

5 52.32 0.81 17.57 6.21 0.07 5.34 2.98 2.80 4.87 0.13 0.00 6.89 1.4 e 380 52.56 0.76 18.16 8.22 0.06 4.45 2.12 1.90 5.65 0.13 0.29 5.73 1.4 a 535 56.09 0.77 16.01 5.89 0.06 4.10 4.69 2.28 5.23 0.18 0.00 4.70 1.5 a 448 49.28 0.75 18.60 8.70 0.08 5.24 4.16 1.79 4.60 0.11 0.00 6.70 1.8 a 449 47.73 0.73 17.62 9.27 0.09 5.49 4.97 1.87 4.39 0.14 0.00 7.70 1.9 a 586 51.71 0.79 18.11 8.07 0.07 5.05 2.98 1.63 4.28 0.12 0.00 7.20 1.9 a 858 49.57 0.73 17.64 7.45 0.08 4.76 5.75 1.66 4.24 0.12 0.00 8.00 1.9 a 968 61.01 0.79 15.87 4.89 0.06 3.65 1.32 2.18 4.07 0.15 0.26 5.80 2.0 a 883 52.72 0.68 15.15 5.44 0.09 4.60 5.81 3.28 5.17 0.12 0.16 6.80 2.3 a 969 59.23 1.05 18.82 6.55 0.07 2.41 0.38 1.81 7.18 0.15 0.19 2.10 2.4 a 795 57.50 0.73 15.02 5.24 0.06 4.13 4.07 1.62 6.19 0.14 0.00 5.30 2.9 a 876 52.17 0.69 15.32 5.82 0.07 4.65 5.21 3.13 5.21 0.10 0.79 6.90 2 .9a 740 54.11 0.67 16.19 6.60 0.09 4.42 4.53 1.84 4.44 0.11 0.40 6.62 3.1 a 494 59.30 0.75 16.50 6.05 0.04 4.09 1.32 1.58 6.70 0.12 0.26 3.30 3.7 a 794 44.60 0.59 14.97 6.91 0.14 6.47 9.73 1.34 3.43 0.11 0.00 11.70 3.7 a 483 57.15 0.71 15.61 6.14 0.07 4.69 3.01 2.02 5.54 0.12 0.68 4.30 4.3 a 432 53.78 0.78 16.79 7.62 0.08 4.54 2.73 1.80 6.28 0.11 0.84 4.70 4.7 a 426 50.00 0.93 19.28 10.36 0.05 7.14 0.32 1.86 5.38 0.18 0.06 4.71 5.0 e 824 57.73 0.71 15.67 6.93 0.05 4.71 1.96 2.65 4.93 0.13 0.22 4.30 6.1 a 866 60.30 0.67 16.84 6.59 0.04 4.63 0.32 2.16 4.89 0.13 0.00 3,43 7.2 a

Mean 54.59 0.72 15.98 5.69 0.07 4.30 4.50 1.62 4.79 0.13 0.28 7.37 Stan-

dard devia- tion 4.86 0.11 2.18 1.79 0.02 1.13 2.83 0.73 1.09 0.03 0.37 3.15

LOI = Loss on ignition. % Ro = vitrinite reflectance, a = actual, e = estimate.

modi f i ed vers ion (Hillier, 1989) o f the m e t h o d de- scr ibed by J o h n s et al. (1954). These es t imates serve s imply to i l lustrate t r ends in a b u n d a n c e be tween sam- pies, no t absolute amoun t s .

Expe r imen ta l X R D pa t te rns o f mixed - l aye red m i n - erals were c o m p a r e d to pa t te rns calcula ted wi th N E W - M O D (Reynolds , 1985). The pa rame te r s used were: Soller slits 6.6 a n d 2, d ivergence slit 1, exchange ca t ion Mg, g o n i o m e t e r radius 17.3 cm, sample length 3 cm, m a s s absorp t ion coefficient (mustar) 45, and or ienta- t ion func t ion (s igma star) 12. The p rog ram was also modi f i ed to inc lude a 31.1 ~ correns i te layer to enable calculat ion o f var ious types o f inters t ra t i f ied chlor i te / correns i te minerals .

Semiquan t i t a t ive analyses o f the whole rock m i n - eralogy were m a d e by X R D o f r a n d o m p o w d e r s using a l u m i n u m p o w d e r as an in ternal s tandard . The weight

percentage o f each minera l p resen t was d e t e r m i n e d f rom the in tens i ty ra t io o f selected peaks o f the minera l to those o f the s tandard , cor rec ted by a cons t an t o f p ropor t iona l i ty d e t e r m i n e d f rom mix tures o f pure minera l s a n d the s tandard . Fo r minera l s p resen t in a m o u n t s greater t han 15%, the prec is ion and accuracy o f the d e t e r m i n a t i o n is general ly • 10% relat ive to the a m o u n t present . Chemica l analyses o f m u d r o c k s were m a d e by X- ray f luorescence spec t rome t ry (XRF) on fused glass beads using a Phi l ips P W 1400 X R F .

Elect ron m i c r o p r o b e analyses o f clay minera l s were


made using a wavelength dispersive Cameca Camebax machine with an accelerating voltage of 15 kV, beam current of 10 nA and a spot size of 2 #m. Chlorite analyses were selected using an arbi trary upper l imit o f 0.5 wt. % for total CaO + Na20 + K20, and structural formulae calculated assuming an ideal anionic framework ofOzo(OH)l 6. All Fe is assumed to be Fe 2§

RESULTS

Whole rock mineralogy and chemistry

Mineralogically, the samples are typical of many mudrocks, containing an average of about 25% quartz and 50% clay minerals (phyllosilicates) with the re- mainder divided principally between feldspars and carbonates (Table 1). Carbonate content, most ly dolomite, varies widely and some of the samples border on the composi t ional field of marls (35%-65% carbonate: Pettijohn, 1957). Notably, the most calcareous samples are low organic matur i ty samples which typically contain about 20% dolomite with little i f any calcite. Amongst the clay minerals, illite, and il l i te/smectite are dominant , followed by chlorite minerals and kaolinite. The most obvious trend in the clay mineral composi t ion of the whole rock is the frequent absence of chlorite from low organic maturi ty samples, particularly those samples which are rich in dolomite (Table 1). Chemically, all the samples appear to be similar (Table 2); the most obvious variat ion is CaO content (Table 2). The standard deviat ion for CaO is over two times that of MgO for approximately the same mean concentration.

X-ray diffraction of the <2 gm fraction

X-ray diffraction patterns of the glycolated <2 #m fraction of five of the samples listed in Tables 1 and 2 are shown in Figure 3 and estimates of the relative percentages of different clay minerals are included in Table 1. Samples 928, 471, and 409 are from areas o f relatively low organic maturi ty and samples 449 and 483 are from regions of higher maturi ty. The samples which contain chlorite show a clear increase in the "'crystallinity" of this mineral from the relatively broad peaks in sample 409 to the considerably narrower peaks in sample 483. This t rend is correlated with increasing diagenetic grade as evidenced by the vitrinite reflectance data (Table 1) and is a general t rend typical o f all the samples which have been examined (Hillier, 1989). The illitic clay minerals illustrated in Figure 3 all show fewer than 10% expandable layers and, like chlorite, there is an increase in "crystal l ini ty" from samples of low to high organic maturi ty. The samples shown (Figure 3) are from Caithness and Orkney (Fig- ure 1). Further south in the basin, low organic matur i ty samples may contain illite/smectite with up to 30% expandable layers (Hillier and Clayton, 1989). Apar t from chlorite, illite and illite/smectite, some samples

IL

IL+Q

IL 928

C

0.45) IL c /~

IL C

>35) ~ IL+O

IL C

3-29) C C IL-I- Q

I I I ~ I I

6 10 14 18 22 26 30 34

2 Theta Cu K alpha

Figure 3. Five examples of diffraction patterns of the glycolated <2 #m fractions ofOrcadian mudrocks. Samples 928, 471, and 409 are from relatively low maturity areas (Ro% 0.9 to 1.2). Samples 449 and 483 are from areas of higher maturity (Ro% 1.8 and 4.3) where chlorite is both abundant and ubiquitous. Note the increasingly "'crystaUinity" of chlorite peaks (peak width at half height adjacent to 002) in passing from low to high maturity samples (409-483). C = chlorite, I1 = illite, Q = quartz, D = dolomite.

contain kaolinite and others corrensite, but these two minerals are never observed in the same sample (Table 1). Both tend to be more common in low organic maturity samples, although corrensite may be present in samples with vitrinite reflectance values of up to 3% (Hillier, 1989).

As well as chlorite and corrensite, Orcadian mudrocks frequently contain minerals which appear to have intermediate proport ions of expandable (smectite-like) layers (Figures 4-6). All of the diffraction patterns also show the presence of discrete chlorite. Sample 816 is an example which contains corrensite. In the air-dried pattern, this gives a strong 001 reflection near 28 /~ (Figure 4) which shifts to 31/k after glycolation (Figure 5). Heating to 375~ produces two broad peaks at about 12 .~ and 8 ~ , whilst heating to 550~ gives rise to a single peak at about 12 A (Figure 6). In both the heated and the glycolated traces of sample 816, a number of peaks due to corrensite are clearly resolved from those due to chlorite. Comparison of the air-dried and the


CR

CR

"C

CR'I'C

IL

816

838

C

C

IL

J

~ 56

2 , i

6 10

2 Theta Cu K alpha

I

14

CR+C

IL

bL

8

IL

C C

IL C

I I I

2 6 10 14 18 2 2 2 6 3 0 34

2 The ta Cu K alpha

Figure 5. XRD patterns of the glycolated <2 #m fraction of samples containing mixed-layer chlorite minerals. Note that the percentages indicate the proportion of expandable layers, not the proportion of corrensite layers. CR = corrensite, C = chlorite, I1 = illite.

glycolated traces of samples 838 and 456 shows that both these samples also contain an expandable chlorite mineral. Lack of resolution, however, from the discrete chlorite in these samples makes precise identification of these minerals difficult. The only peak which is resolved is that at about 31 /~, but in both cases it is a weak reflection, much less intense than in sample 816. Nevertheless, the presence of this peak is important because it indicates either the presence of R 1 ordered C/S or of corrensite. Overall, the impression is that of a series of expandable minerals ranging between corrensite, such as in sample 816, to chlorite, as in sample 21. This last sample shows only slight changes between the air-dried and the glycolated traces, being mainly characterised by its relatively broad chlorite peak widths (Figures 4-6). Sample 409 in Figure 3 shows a very similar diffraction pattern to sample 21 and both pat-

Figure 4. XRD patterns of air-dried <2/~m fraction of samples containing mixed-layer chlorite minerals. CR = corrensite, C = chlorite, I1 = iUite.


375 550

IL

CR C

C CR IL

16

~ 3 8

IL

C

456

C

2 6 10 2 Theta Cu K alpha

14 2

838

A C

IL

21

i = s

6 10 14 2 Theta Cu K alpha

Vol. 4 l, No. 2, 1993 Chlorite minerals in Devonian mudrocks 247

terns are typical for chlorite in areas o f low organic maturi ty. Also, for the patterns obtained after heating to 375 and 550"C, the impression is that of a series from the resolved peaks of corrensite in sample 816, through the asYmmetrical broadened, composite peaks o f 838 and 456, to that which is essentially chlorite in sample 21.

At tempts to est imate the proport ion of expandable layers in the mixed layer chlorite minerals were made using peak migrat ion curves calculated for R1 ordered C/S for the peak which migrates between 8.5 and 7.1 �9 ~ (Hower, 1981 ), Because of the lack of complete resolution of this peak from that due to discrete chlorite, peak positions were obtained using a peak decomposition program (Lanson and Besson, 1992). The composite peak was assumed to consist of two peaks both with Lorentzian shapes, one due to discrete chlorite and the other due to the expandable mixed-layer mineral. With these assumptions, the results indicate close to 50% expandable layers for sample 816, 30% for sample 838, 10% for sample 456 and <5% for sample 21. In some cases, better overall fits were obtained using three peaks, two very close to the position of chlorite and a third for the expandable mineral. With three peak fits, differences in the estimates of expand- abili ty amounted to < 5% expandable layers; however, the physical significance of a three-peak fit seems more doubtful and the simpler two-peak fit is preferred.

Polytypes

Chlorite polytype analysis is in all cases hampered by the presence o f illitic clay minerals whose peaks interfere with those of chlorite. At tempts to determine the chlorite polytype were, therefore, made by obtain- ing diffraction patterns before and after t reatment of the samples with hot 1 N HCI for half an hour and subtracting the HC1 treated pattern from the untreated pattern to obtain the pattern for chlorite, which is de- stroyed by the HC1. The results of this t reatment for sample 21 are shown in Figure 7 and indicate that the dominant polytype is the IIB. In all cases the IIB polytype was found to be dominant (Hillier, 1989), although trace amounts of other polytypes cannot be ruled out.

Electron microprobe data

Petrographic examinat ion of polished thin sections using backscattered electron microscopy shows that both detrital and authigenic chlorites are present in Orcadian mudrocks (Hillier, 1989). Each type is easily identified by its morphology. Detri tal chlorites occur as elongate cleavage flakes which polish well. In con-

Figure 6. XRD patterns of samples containing mixed-layer chlorite minerals after heating to 375 and 550~ for 1 hr. CR = corrensite, C = chlorite, I1 = iUite.

trast, authigenic varieties Occur as much finer-grained, less well polished aggregates, in some cases filling mi- cropores and in other cases with outlines suggestive of grain replacement. Examples of microprobe analyses o f the two types are given in Table 3. Other analyses of authigenic examples are given in Hill ier and Velde (1991) and all analyses are plot ted in Figure 8. Mg and Fe-rich varieties occur for both authigenic and detrital chlorites, but on the whole the detrital chlorites are richer in Fe and AI. In addition, the detrital examples tend to have octahedral totals close to ideal for a trioctahedral mineral, many overlapping the fully trioctahedral line (Figure 8). This is typical o f high temperature chlorites (Laird, 1988; Hill ier and Velde, 1991). For the authigeni c examples, there is a distinct tendency towards low octahedral totals, particularly for the Mg-rich Al-poor varieties.

In addi t ion to the microprobe analysis of samples containing only chlorite (based on XRD), analyses were also made on four samples which contain mixed-layer chlorite minerals (Figures 4-6). Because o f the presence of expandable minerals, analyses were made after exchange of the interlayer cations with Cs § in the thin section (Hillier and Clayton, 1992). Use of this tech- nique confirms the expandable character of a mineral. I t also allows analyses contaminated by other Na +, K +, or Ca 2§ bearing minerals to be identified and avoided, simply by applying the same criteria used to select " g o o d " ch lor i t e ana lyses . F u r t h e r m o r e , f rom the amount of Cs + taken up it is possible to calculate the cation exchange capacity (CEC), which should be related to the proport ion of expandable layers present.

Figure 9 shows the results of these calculations as histograms of the CEC vs N, the number of analyses, selected as described. For sample 816, which contains corrensite, there is a group of analyses with CECs of about 40 meqs/100 g, Sample 838 shows both a reduced and negatively skewed distr ibution, with maximum CECs of about 20 meqs/100 g. CECs are further reduced for sample 456, and sample 21 consists almost entirely o f analyses with CECs of 5 meqs/100 g or less, a reasonable value for a pure chlorite with little or no expandable layers. In Figure 10 the same CEC data are plot ted against the Fe/(Fe + Mg) ratio. This shows that in all four samples there are both Fe and Mg-rich chlorite minerals present, but invariably the highest CECs are associated with the Mg-rich minerals. CECs for the Fe-rich minerals are consistently low. Only in sample 21, which by X R D shows little evidence for expandable layers, do both Mg-rich and Fe-rich varieties have similar CECs, all o f which are low.

Examples of individual microprobe analyses of the most expandable Mg-rich chlorite minerals from each sample are given in Table 4. Structural formulae for these analyses are calculated as chlorite, i.e., on a 28 oxygen basis, and also on a variable oxygen basis determined from estimates of the proport ion of smectite-


II I I i , I

30 32 34 36 38 40 42 44 46 48

2 Theta Cu

Figure 7. Random powder XRD patterns of the <2 ~m fraction of sample 21 A) before and B) after treatment with HCL and pattern calculated by C) difference indicating that the chlorite is the IIb polytype. Line shows position and relative intensities of diagnostic IIb peaks.

like (22 oxygen) and chlorite-like (28 oxygen) layers. Variable oxygen calculations are necessary to compare cation site occupancy and are made in two different ways. In the first case, the number of oxygens are calculated simply from the proportion of the two layer types estimated by XRD. In the second case, the oxygen basis is determined from an estimate of the proportion of layer types calculated from the sum of tetrahedral plus octahedral cations for an analysis cast as if the mineral were a chlorite. The assumptions of this second method are that both the smectite-like and chlorite-like layers are fully trioctahedral, i.e., they contain 6 and 12 octahedral cations, respectively. The proportion of expandable layers is then given by

% expandable layers = [28(S-20)]/[6(S-28)] (1)

where S is the sum of octahedral plus tetrahedral cations in the formula that is cast as chlorite.

The variable oxygen structural formulae show that there are a number of trends in the analytical data for corrensite, through the intermediate minerals, to chlorite. Tetrahedral Si decreases by up to about 0.5 atom, and, conversely, tetrahedral a luminum increases by this amount. Total a luminum increases significantly by up

8 / - . . , . / ,, / / " ' , , Serpent ine/ /

i i " - . / " i i / / " ' , . i I AI=2 / /

7 , "" ""II / . / 2 - , / / "" / " " . (

Sum(Vi)=lO " " - Sum(Vi)=11 �9 111 . �9 Sum(VI)=12

' " - / �9 �9 , t i . j S i 6 [,Sudoite / " ' . . . ~ I i I

�9 I . " ' ~ ~ D B ~ [] �9 Authigenic �9 / /

" . , / / N=37

5 / , " / [] Detrital / ",, / i I "" , i I N=22

/ AI=8 / / / " , / /

i , I A m e s i t e 4 , , "!11 , ,

4 5 6 7 8 9 10 11 12

R2+

Figure 8. Chemical composition of detrital and authigenic chlorites from Orcadian mudrocks plotted in the vector representation of chlorite composition of Weiwi6ra and Weiss (1990) as used by Hillier and Velde (1991).

to 1.5 atoms, indicating that octahedral a luminum increases by about 1 atom. Fe and Mg also increase by almost 1 atom each. Together these changes raise the total number of octahedral cations from near 9 for corrensite towards 12 for that of chlorite. There is, of course, no significant change in the Fe/(Fe + Mg) ratio because the analyses were selected on the evidence that all the expandable minerals in the series from corrensite to chlorite are Mg-rich (Figure 7). Neither is there any significant change in the Al/(A1 + Fe + Mg) ratio of the octahedral cations.

DISCUSSION

Origin of chlorite minerals

The whole rock mineralogy of the Orcadian mudrocks (Table 1) shows that they can be divided on the basis of chlorite content (absent or abundant). A similar subdivision can be made on a basin-wide scale where the distribution of chlorite defines regions where it is ubiquitous and others where it is essentially absent (Hillier, 1989). Comparing chlorite distribution to vit- finite reflectance data for the samples in Table 1 and for the Orcadian Basin as a whole (Hillier and Marshall, 1992), it is evident that the occurrence of chlorite minerals is related to diagenetic grade. In areas where vit- finite reflectance is greater than 1.3%, chlorite minerals are ubiquitous, whereas in areas of lower maturity their occurrence is either sporadic or they are absent. This relationship indicates that the majority of the chlorite minerals found throughout the Orcadian Basin are authigenic and were formed during diagenesis/metamorphism. In an early reconnaissance study of the clay mineralogy of the Orcadian Basin, Wilson (1971) con- cluded that the abundant chlorite was probably detrital because it was well crystallized. This early study, how-


Table 3. Microprobe analyses and structural formulae of detrital and authigenic chlorites from Orcadian mudrocks.

Detrital Authigenic

No 341-5F 535-11F 968-17F 824-5F 341-2313 535-14B 968-10B 824-1B SiO2 24.19 24.47 25.25 25.67 26.41 32.83 25.91 25.90 TiO2 0.08 0.05 0.06 0.06 0.07 0.03 0.14 0.05 A1203 20.77 20.11 18.75 20.46 17.40 12.90 19.12 21.89 FeO 29.73 27.18 24.15 23.04 27.55 14.70 28.66 19.15 MgO 8.57 13.75 15.00 16.35 11.84 26.85 10.72 19.47 MnO 0.28 0.25 0.40 0.06 0.29 0.00 0.19 0.28 CaO 0.00 0.07 0.06 0.02 0.10 0.05 0.13 0.00 Na20 0.16 n.d. 0.01 0.04 0.10 n.d. 0.05 0.07 K20 0.13 0.05 0.10 0.14 0.06 0.08 0.18 0.06 Sum 83.90 85.93 83.77 85.83 83.82 87.44 85.11 86.87

Si 5.49 5.35 5.57 5.46 5.91 6.52 5.75 5.33 A1 5.56 5.17 4.88 5.14 4.61 3.03 4.99 5.32 AI(T) 2.51 2.66 2.43 2.54 2.09 1.48 2.25 2.67 AI(O) 3.06 2.52 2.46 2.61 2.52 1.55 2.73 2.64 Ti 0.01 0.01 0.01 0.01 0.01 0.00 0.03 0.01 Fe2 5.64 4.96 4.46 4.11 5.16 2.45 5.32 3.30 Mg 2.89 4.48 4.94 5.18 3.96 7.95 3.55 5.97 Mn 0.06 0.05 0.08 0.01 0.05 0.00 0.04 0.05 Sum(O) 11.66 12.03 11.94 11.93 11.70 11.94 11.67 11.98 F/(FM) 0.66 0.53 0.48 0.44 0.57 0.24 0.60 0.36 Ca 0.00 0.01 0.01 0.00 0.03 0.01 0.03 0.00 Na 0.08 0.00 0.00 0.03 0.05 0.00 0.03 0.03 K 0,03 0.03 0.03 0.03 0.03 0.02 0.05 0.03

ever, was made without the benefit of organic maturity data that demonstrate that large parts of the basin have been metamorphosed. Undoubtedly, some amount of detrital chlorite is often present, as evidenced by the occurrence of large flakes of chlorite in thin-section which can be analyzed with the microprobe (Table 3). However, many low-maturity samples show no XRD evidence for the presence of chlorite, suggesting that detrital chlorite is generally not present in abundance.

In addition to demonstrating that most chlorite is authigenic, the organic maturity data combined with the whole rock semiquantitative mineralogy and chemical composition provide insight into the mechanism of chlorite authigenesis. From the mineralogical data (Table 1), it is evident there is an inverse relationship between the occurrence of chlorite and dolomite (Fig- ure 11). The organic maturity data show that this inverse relationship is a function of maturity (Figure 12). In many low maturity samples, dolomite is abundant (15-20 wt. %) and chlorite absent; whereas at high maturities, chlorite is abundant (10-15 wt. %) and dolomite is minor or absent. There is, however, no apparent change or systematic trend in the MgO content of these mudrocks across the entire range of maturity (Figure 12). With few exceptions, the amount of MgO is between 3-5 wt. %. Neither are there any obvious systematic changes in the abundance of any of the other oxides, excepting CaO, which is often lower in samples containing chlorite than those in which chlorite is absent (Table 2). These data indicate that the formation of chlorite minerals in Orcadian mudrocks occurred by a reaction which involved dolomite as the source

of Mg and Fe, and by necessity the detrital assemblage of dioctahedral clay minerals and quartz as the sources of A1 and Si.

The diagenetic formation of chlorite by a reaction that involves the breakdown of dolomite has been described previously by Muffler and White (1969) from the Pliocene and Quaternary siltstones and sandstones of the Salton Sea geothermal field, California, which are presently undergoing hydro thermal me tamor - phism. Dolomite, ankerite and kaolinite, present in the sediments at shallow depths, are replaced by an assemblage of chlorite and calcite over the depth interval from 1200 to 2300 ft. Muffler and White (1969) explained the changes in mineralogy by a metamorphic reaction similar to that originally proposed by Zen (1959).

5 Dolomite + Kaolinite + Quartz + Water = Chlorite + 5 Calcite + 5 CO2 (2)

More recently, Hutcheon et al. (1980) documented this reaction in Cretaceous sandstones of the Kootenay formation near Calgary, Canada, and later (Hutcheon, 1990) in sandstones from the Venture Field of the Sco- tian Shelf, Nova Scotia. In addition, Hutcheon et al.

(1980) postulated a similar reaction between dolomite and the muscovite component of illite.

15 Dolomite + 2 Illite + 3 Quartz + 11 Water = 3 Chlorite + 15 Calcite + 2 K +

+ 2 O H - + 15 C O 2 (3)

Generally, kaolinite is not abundant in Orcadian Ba- sin mudrocks and moreover its distribution is sporadic


111.4s8 5

10

8

6

4

2

0 5

816 (50% Expandable)

I I - - 1 1 . . 10 15 20 25 30 35 40 45 50

838 (30%Expandable)

10 15 20 25 30 35 40 45 50

10

8

6 456 (10% Expandable) N

4

2

0 5 10 15 20 25 30 35 40 45 50

0 5 10 15 20 25 30 35 40 45 50

CEC meqllOOgms

Figure 9. Cation exchange capacities determined from elec, tron microprobe analyses of chlorite minerals in thin-sections of samples 816, 838, 456, and 21 vs N the number of analyses.

(Hillier, 1989); therefore, a reaction between dolomite and kaolinite, such as in Eq. 2, probably cannot account for the ubiquitous formation o f chlorite at high maturities. Unlike kaolinite, i l l i te/smectite is ubiquitous in low maturi ty samples (Hillier and Clayton, 1989) and thus chlorite is believed to have formed mainly by a reaction between dolomite and illite/smectite, similar to Eq. 3.

Two other aspects o f the whole rock data (Table 1) are also consistent with a reaction similar to Eq. 3. Firstly, high maturi ty samples are generally richer in K-feldspar than low matur i ty ones, suggesting that the K § l iberated by such a reaction has resulted in the formation of K-feldspar. And secondly, reactions such as Eq. 2 and Eq. 3 produce calcite, which is generally absent in low maturi ty samples that are rich in dolomite but present in the majori ty of both low and high maturity samples which contain chlorite. It is also worth noting that complete reaction of about 20 wt. % do-

CEC

40 I I I �9

30 �9 l i

20

10 �9 �9

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

CEC

50 838

40

30

20

10

0 0.0 0.1

mm mmm

i N 0 �9 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

CEC

] 456

2O

10

0 0.0

[]

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

5~ t 40-

30- CEC

20

10

0 0.0

21

� 9 �9 �9 J I .,'," 0.1 0,2 0,3 0.4 0.5 0,6 0.7 0,8 0,9 1.0

Fe/(Fe + Mg)

Fe/(Fe + Mg) ratio plotted against CEC for sam- Figure 10. pies shown in Figures 4, 5, and 6, showing that in each sample there are two types of chlorite minerals. Mg-rich minerals with decreasing CECs and Fe-rich minerals whose CECs are low and fairly constant throughout. The Mg-rich minerals correspond to corrensite through mixed-layer chlorite/corrensite to Mg-rich chlorite and the Fe-rich mineral to co- existing Fe-rich chlorite.

lomite would produce approximately 1 mole of CO2 per ki logram of mudroCk. This is roughly 10 t imes the quanti ty o f CO2 that could be produced by the decar- boxylation of organic matter, as calculated by Bjor- lykke (1988), for the most favourable case o f shales rich in humic kerogen. I f reactions between clay minerals and carbonate in mudrocks and shales are more common than realised they would clearly have important consequences for diagenesis in adjacent sandstones.

Two samples (315 and 870) that have much lower MgO contents than most others (Table 2) are from lake intervals deposi ted high on an alluvial fan during a


Table 4. Microprobe analyses of expandable chlorite minerals made after Cs§ exchange, calculated as if chlorite and on variable oxygen basis calculated from estimates of the proportion of expandable layers from XRD, and microprobe analyses.

No 816-B7 816-B5 816-B6 838-B33 838-B29 838-B37 456-B52 456-B51 456-B44 21-B20 21-B16 21-B23

SiO2 33.05 32.30 31.52 31,27 30.69 32.44 30.69 TiO2 0.13 0,06 0.10 0.00 0.00 0.01 0,00 A1203 14.84 13.98 14.12 14.27 14.04 14.48 16.36 FeO 17.73 20.01 18,77 20.31 21.32 17.02 19.11 MgO 17.02 16.48 15.35 18.32 17.01 18.51 20.29 MnO 0.01 0.00 0.16 0.13 0.00 0.14 0.26 CaO 0.04 0.12 0.17 0.19 0.16 0.10 0.00 Na20 0.02 0.08 0.00 0.00 0.02 0.61 0.04 K20 0.34 0.24 0.42 0.16 0.22 0.08 0.03 Cs20 6.27 7.01 5.67 1.98 2.16 2.78 0.70 Sum 89.44 90.28 86.27 86.63 85.61 86,15 87.48 CEC 45,00 50.00 40.00 14.00 15.00 20.00 5.00

As chlorite Si 6.88 6.82 6.88 6.59 6.60 6.79 6.28 A1 3.65 3.48 3.62 3.55 3.57 3.57 3.93 Al(T) 1.12 1.18 1.12 1.42 1.40 1.21 1.72 Al(O) 2.53 2.30 2.50 2.13 2.17 2.36 2.21 Ti 0.03 0.01 0.01 0.00 0.00 0.00 0.00 Fe2 3.09 3.54 3.42 3.58 3.84 2.98 3.27 Mg 5.28 5.19 4.99 5.75 5.45 5.77 6.18 Mn 0.00 0.00 0.03 0.03 0.00 0.03 0.05 Sum(O) 10.93 11.04 10.95 11.49 11.45 11.14 11.71 F/(FM) 0.37 0.41 0.41 0.38 0.41 0.34 0.35 Ca 0.01 0.03 0.04 0.04 0.04 0.03 0.00 Na 0.00 0.03 0.00 0.00 0.00 0.25 0.03 K 0.10 0.08 0.11 0.05 0.05 0.03 0.00 Cs 0.55 0.63 0.52 0.18 0.21 0.25 0.05

Oxygen based on % expandable layers Si 6.14 6.09 6.14 6.16 6.18 6.35 6.15 AI 3,26 3.10 3,23 3.32 3.34 3.34 3.85 AI(T) 1.86 1.91 1.86 1.84 1.83 1.65 1.85 AI(O) 1.41 1.19 1.37 1.48 1.51 1.69 2.00 Ti 0.02 0.01 0.01 0.00 0.00 0.00 0.00 Fe2 2.76 3.16 3.05 3.35 3.59 2.79 3.20 Mg 4.71 4.63 4.46 5.38 5.10 5.40 6.05 Mn 0.00 0.00 0.02 0.02 0.00 0.02 0.05 Sum(O) 8.90 8.99 8.92 10.24 10.20 9.91 11.29 F/(FM) 0.37 0.41 0.41 0.38 0.41 0.34 0.35 Ca 0.01 0.02 0.04 0,04 0.04 0.02 0.00 Na 0.00 0.02 0.00 0.00 0.00 0.24 0.02 K 0.09 0.07 0.09 0.05 0.05 0.02 0.00 Cs 0.49 0.57 0.47 0.17 0.19 0.24 0.05 %Ex 50 50 50 30 30 30 10 Ox equ 25.0 25.0 25.0 26.2 26.2 26.2 27.4

28.63 27.91 29.00 28.11 26.58 0.01 0.01 0.07 0.14 0.19

18.53 17.21 15.57 17.80 18.91 22.44 24.23 22.87 21.55 20.03 15.72 15.38 17.33 16.51 17.08 0.54 0.13 0.23 0.25 0.20 0.17 0.06 0.33 0.22 0,07 0,04 0.08 0.06 0.04 0.04 0,17 0.08 0.08 0.04 0.26 1.49 0.92 0.78 0.45 0.29

87.73 86.01 86.32 85.10 83,65 11,00 7.00 6.00 3.00 2.00

6.01 6.01 6.18 6.00 5.74 4.59 4.38 3.91 4.48 4.81 t.99 1,99 1.82 2.01 2.26 2.60 2,39 2.09 2.48 2.55 0.00 0.00 0.01 0.03 0.03 3.94 4.36 4.07 3.84 3.63 4.92 4.95 5.50 5.25 5.51 0.10 0.03 0.04 0.05 0.04

11.56 11.72 11.71 11.65 11.75 0.44 0.47 0.43 0.42 0.40 0.04 0.01 0.08 0.05 0.01 0.03 0.03 0.03 0.03 0.03 0.05 0.03 0.03 0.00 0.08 0.13 0.08 0.08 0.05 0.03

from XRD 5.88 5.88 6.11 5.93 5.68 4.50 4.28 3.87 4.44 4.76 2.12 2,12 1.89 2,07 2.32 2.37 2.16 1.98 2.37 2.44 0.00 0.00 0.01 0.03 0.03 3.85 4.27 4.02 3.80 3.59 4.82 4.84 5.44 5.20 5,45 0.10 0.03 0.04 0.05 0.04

11.14 11.30 11.50 11.44 11.54 0.44 0.47 0.43 0.42 0.40 0.04 0.01 0.08 0.05 0.01 0.03 0.03 0.03 0.03 0.03 0.05 0.03 0.03 0.00 0,08 0.12 0.08 0.08 0.05 0,03

10 10 5 5 5 27.4 27.4 27.7 27.7 27.7

Oxygen based on % expandable layers from microprobe Si 6.07 6.09 6.08 6.19 6.18 6.13 6.06 5.69 5.81 5.96 5,74 5.57 AI 3.22 3,10 3.20 3.33 3.34 3.22 3.80 4.35 4.23 3.78 4,30 4.66 AI(T) 1.93 1.91 1.92 1.81 1.82 1.87 1.94 2.31 2.19 2.04 2.26 2.43 AI(O) 1.29 1.19 1.27 1.52 1.51 1.35 1.86 2,05 2.04 1.74 2.04 2.23 Ti 0.02 0.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.03 0,03 Fe2 2.73 3.16 3.02 3.37 3.59 2.69 3.16 3.73 4.22 3.93 3.68 3.52 Mg 4.66 4.63 4.41 5.40 5.10 5.21 5.97 4.66 4.78 5.31 5.03 5.34 Mn 0.00 0.00 0.02 0.02 0.00 0.02 0.05 0.10 0,03 0.04 0.05 0.04 Sum(O) 8.69 8.99 8.74 10.32 10.20 9.27 11.03 10.54 11.07 11.02 10.83 11.15 F/(FM) 0.37 0.41 0.41 0.38 0.41 0.34 0.35 0.44 0.47 0.43 0.42 0.40 Ca 0.01 0.02 0.04 0.04 0.04 0.02 0.00 0.04 0.01 0.07 0.05 0.01 Na 0.00 0.02 0.00 0.00 0.00 0.23 0.02 0.02 0.03 0.03 0.03 0.03 K 0.09 0.07 0.09 0.05 0.05 0.02 0.00 0.05 0.03 0.03 0.00 0.08 Cs 0.49 0.57 0.46 0.17 0.19 0.23 0.05 0.12 0.08 0.07 0.05 0.03 %Ex 55 50 54 28 30 45 16 24 16 16 20 14 Ox equ 24.7 25.0 24.7 26.3 26.2 25.3 27.0 26.5 27.1 27.0 26.8 27.t


30

25

20

% Dolomite 15

10

5

n , , �9 �9 �9

�9 i i �9 I l l

0 5 10 15 20 25 30

%Chlorite

Figure 11. Inverse relationship between dolomite and chlorite content of Orcadian mudrocks.

~ A

3 ~

%Ro 4

time of maximum transgression. These samples are of relatively low maturity and contain neither dolomite nor chlorite. The formation of dolomite in the Orca- dian lake sediments has been related to times of re- gression and evaporitic concentration of the lakes (Jan- away and Parnell, 1989). Therefore, the absence of dolomite in these samples may be related to their pa- leogeographic location. Such exceptions serve to emphasize that the whole rock mineralogical trends apply only to a particular facies, other facies would give different trends. For example, facies which are much richer in dolomite and poorer in clay minerals may still contain abundant dolomite even at the highest maturity.

Structure and chemistry o f chlorite minerals

Neither the structure nor the chemistry of minerals in the series between trioctahedral smectite and chlorite are well understood. Apart from corrensite, examples of intermediate minerals have only rarely been reported, and are not easily characterised by XRD. Indeed, as pointed out by Reynolds (1988), occurrences of discrete corrensite are so numerous by comparison that it is perhaps best regarded as a discrete phase in a thermodynamic sense, rather than a simple intermediate mixed-layer C/S. Others, including Meunier et aL (1988), Shau et al. (1990), and Inoue and Utada (1991) have all arrived at the same conclusion. Reyn- olds (1988) further suggested that minerals intermediate between corrensite and chlorite may be metasta- ble mixtures of these two phases in the sense of Lee et al. (1985). This HRTEM study by Lee et al. (1985) and a series of related HRTEM studies (referenced therein) have shown that in the diagenetic realm, phyl- losilicate crystallites consist of domains of discrete minerals together with domains in which the minerals

A~

�9 A A

A A

A

A

A

A

i i I i

5 10 15 20 25 30

Percent

Relationship between vitrinite reflectance and Figure 12. weight percent of chlorite (triangles), dolomite (diamonds), and MgO (line) in Orcadian mudrocks, data in Tables 1 and 2. At low maturity chlorite is absent from many samples and these same samples are rich in dolomite. At maturities higher than about 1.3% Ro all samples contain chlorite and either no dolomite or much less dolomite than low maturity samples. Although the inverse relationship between chlorite and dolomite is clearly related to maturity, the MgO content of the samples remains relatively constant throughout the maturity range.

are intimately interstratified. In fact, direct HRTEM evidence of mixed-layer chlorite/corrensite in meta- basalts was provided by Shau et al. (1990). Further- more, Shau et al. (1990) argued that the most stable kinds of interstratifications in the smectite to chlorite series should be mixed-layer smectite/corrensite and chlorite/corrensite, rather than interstratifications of smectite and chlorite layers.

The XRD diffraction patterns of mixed-layered chlorite minerals from the Orcadian Basin appear to range between corrensite and chlorite. Such a range indicates that these two minerals can be considered as end mem- bers and it follows that intermediate minerals ought to be mixtures of them. In other words, minerals that appear to be intermediate are more realistically described as interstratified chlorite/corrensite rather than as interstratified C/S. This scheme has implications for the types of layer sequences that may occur. For in-


ChloritelCorrensite(12 %)

PCR.C = 0.01

2 6 10 14 18 22 26 30 34

2 Theta Cu K alpha

Figure 13. Diffraction patterns of chlorite/corrensite (12%) calculated using the NEWMOD program: A) randomly interstratified; B) segregated; and C) completely segregated which produces a pattern like a physical mixture (junction proba- bilities adjacent to each pattern). Note that with only 12% corrensite (i.e., about 10% expandable layers) segregation produces a low angle reflection at 31 ~.

terstratifications of chlorite with corrensite an expandable (smectite-like) layer can never be directly followed by another expandable layer, simply because smectite itself is not one of the component layers (Reynolds, 1988). Thus diffraction patterns of randomly interstratified (R0) chlorite/corrensite are identical to those for maximum R1 ordered C/S with compositions in the range < 50% smectite layers�9 Therefore, estimates of the proportion of expandable layers based on XRD patterns calculated for maximum R 1 C/S are valid also for R0 chlorite/corrensite, but the proportion of expandable layers is not equal to the mole fraction of corrensite. This is given by the proportion of expandable layers divided by the proportion ofnonexpandable layers. For example, minerals such as that in sample 838, which appear to contain about 25% expandable layers, would correspond to chlorite/corrensite (33%).

According to Hower (1981 ) and Reynolds (1988) the R 1 ordering peak for C/S at 31 ~ is no longer detectable for compositions that deviate more than about 10% from the ideal 50/50 composition. This must also be the case for the statistically identical R0 chlorite/corrensite, although here the peak is not due to ordering but to the presence of a 31 /~ component layer, i.e., corrensite. The presence of a 31 /~ peak, albeit small, in samples such as 838 and 456 is, therefore, not com- patible with the estimates of 30% and 10% expandable layers for the mixed-layered minerals in these samples�9

Two explanations are favoured: Either the peak at 31 is due to discrete corrensite in the samples, or cor-

rensite occurs randomly interstratified with chlorite, but with a strong tendency towards segregation in order to produce a 31 ~ peak. The presence of this peak is certainly incompatible with an assemblage of R0 C/S plus chlorite. Corrensite, plus chlorite, plus R0 C/S is a further possible but less probable assemblage (Shau et al., 1990).

Examples of a randomly interstratified and segregated chlorite/corrensite calculated with Reynolds NEWMOD program are shown in Figure 13. Extreme segregation produces a pattern like a physical mixture. By XRD alone, it is clearly difficult to make an un- equivocal distinction between the presence of discrete corrensite and a segregated structure. Furthermore, what occurs as segregated crystallites in the rock might well be arranged quite differently by the time it is sedi- mented on a glass slide ready for XRD. The chemical analyses of the minerals, however, provide another line of evidence�9

A CEC of 40-50 (meqs/100 g) seems reasonable for a pure corrensite (Inoue, 1985), which is based on XRD present in sample 816. Reduced CEC values (Figures 9 and 10) show that it is not possible to make an analysis of a pure corrensite in any of the other three samples, only ofintergrowths or mixtures ofcorrensite and chlorite in which the proportion of corrensite is decreasing (decreasing CEC) in accordance with the XRD patterns. Therefore, at least at the scale of the microprobe, corrensite and chlorite must be intimately associated rather than occurring at separate reaction sites. This does not prove that intergrowth occurs at the scale of individual crystallites, but such a close association suggests that this is probably the case. So, rather than a physical mixture, randomly interstratified chlorite/corrensite with a strong tendency toward segregation is the preferred interpretation.

Chemically, the most significant change in the series between corrensite and chlorite is the changing Si/A1 ratio in the tetrahedral sheets. This increase in Al is interesting because the various environments in which the smectite-corrensite-chlorite series are observed are characteristically Al-poor. In addition, the assemblage corrensite plus kaolinite is rarely, if ever, observed. Usually, the formation of corrensite is explained by, or linked to, the availability of Mg. However, this trend of increasing A1 suggests that, within Mg-rich systems, the availability of Al may play an important role in determining whether it is smectite or corrensite, or chlorite that forms. Furthermore, the occurrence of relatively Fe-rich corrensite (Almon et al., 1976) and smectite-corrensite-chlorite series in which all the phases are relatively Fe-rich (Inoue, 1985) suggests that the role of Mg is not paramount. Nonetheless, it is generally the case that rocks which are relatively poor in A1 tend also to be Mg-rich�9


Shau et al. (1990) argued that, for the series smectite- corrensite-chlorite, the Fe/Mg ratio should also change, the minerals becoming progressively more Fe-rich towards the chlorite end. However, as shown here and by Bettison et al. (1991), the Fe/Mg ratio need not change. In fact, large differences in Fe /Mg are more likely to indicate separate, generically unrelated, gen- erations of chlorite. In the Orcadian examples iron- rich authigenic chlorites "coexist" with the Mg-rich phases o f the corrensite-chlorite sequence. The consistently low CECs of the Fe-rich chlorites suggest that they are completely unrelated to the corrensite-chlorite sequence of Mg-rich phases. A number of origins for these Fe-rich chlorites are possible. The composi t ion of the original dolomite, which is usually ferroan, may have been variable, or zoned, so that at certain reaction sites most ly Mg was available and at others mostly Fe. Alternatively, the Fe-r ich authigenic chlorite may be a product o f another diagenetic reaction such as the smectite-to-ill i te conversion. Regardless of their origin, however, the presence of both Fe and Mg-rich authi- genie chlorites in the same small volume of rock dem- onstrates disequil ibrium on a microscopic scale, which is probably typical of the diagenesis of many shales.

Relationships to diagenetic grade and paleotemperature

Temperature o f formation of chlorite minerals. The correlation o f chlorite diagenesis with vitrinite reflectance data indicates that temperature is an impor tant control on chlorite formation. Chlorite minerals are ubiquitous in all samples above vitrinite reflectance values of 1.3% which, according to the data of Barker and Pawlewicz (1986), corresponds to a max imum temperature o f about 170~ In geothermal wells in the Salton Sea, Muffler and White (1969) observed chlorite formation at 180~ whereas in a nearby oil well chlorite was found at temperatures as low as 125~ Similarly, from the intersection of equilibria (Eq. 1) and (Eq. 2) with the CO2-H20 miscibility surface, Hutcheon et al. (1980) obtained maximum temperatures for chlorite forma- t ion of 180"--250~ Hutcheon et al. (1980) also give a vitrinite reflectance value of 1.4% Ro (max) for the first appearance of chlorite in the Kootenay formation, although it should be noted that this value was est imated by extrapolation from a neighboring section. Thus, in both o f these studies, the temperature range o f chlorite formation (by reactions between clay minerals and carbonates) concurs with those est imated from vitrinite reflectance data for the Orcadian Basin.

Frequently, in the Orcadian Basin chlorite minerals also occur at lower vitrinite reflectance values of around 0.9 to 1% Ro. This suggests that chlorite formation may begin at temperatures as low as 120~ The apparent 50~ temperature variat ion for the the first appearance of chlorite might be due either to the influence

of various other controls on the reaction besides temperature and/or to uncertainty in the temperatures es- t imated from vitrinite reflectance due to the influence of time, which is neglected in the model of Barker and Pawlewicz (1986). The effect o f t ime would be to reduce the temperature estimates so that 120~ might be considered a conservative est imate o f the lowest temperature at which clay carbonate reactions may begin.

Depth/temperature related mineral and mixed-layer trends. Data on the depth related conversion of trioctahedral smectite to chlorite in sedimentary rocks are l imited although examples have been described by Hel- mold and van de K a m p (1984) and by Chang et aL (1986). These studies invite analogy with the well documented depth conversion of aluminous dioctahedral smectite to illite. However, the trend documented by Helmold and van de K a m p (1984) is somewhat erratic, and that shown by Chang et al. (1986) jumps rapidly from smectite to corrensite and thereafter changes little with depth.

Other examples of the conversion of tr ioctahedral smectite to chlorite have been documented by Inoue (1987) and Inoue and Utada (199 I) from hydrothermal alteration zones. Inoue (1987) and Inoue and Utada (1991) emphasize the existence o f distinct jumps from smectite to corrensite to chlorite, and often the coex- istence of these minerals in the same sample.

Some of the apparent differences between various sequences may be due to interpretat ion rather than fact because o f the difficulty of interpreting the X R D patterns. Nevertheless, the common persistence of corrensite confirms that, unlike the smectite-to-ill i te conversion, corrensite is a stable intermediate in this series, more analogous to the mineral rectorite than to a mixed- layer i l l i te/smectite mineral with an equal proport ion of two component layers.

For the sequence of minerals between corrensite and chlorite documented in the present study there is certainly no systematic relationship to paleotemperature. Instead, mixed-layered minerals with varying proportions of the component layers are often found at the same locality (Hillier, 1989), and are possibly related to minor lithological variations. Their occurrence is much like that documented by Bodine and Madsen (1987) in the Pennsylvanian evapori te cycles in the Paradox Basin, Utah, where minerals range between corrensite and chlorite over a scale of several tens of metres. Such local variat ions in the occurrence of corrensite, chlorite, and intermediate minerals indicate that, in comparison to the smectite-to-ill i te conversion, the depth/ temperature related conversion ofcorrensi te to chlorite is best regarded as only a general t rend that might be very dependent on local chemical conditions.

It follows from the above discussion that within the smectite-corrensite-chlorite sequence the only reliable indicators of paleotemperatures or relative diagenetic


grade are likely to be the first or last appearance of a part icular mineral. Moreover, the occurrence of examples (Bodine and Madsen, 1987; Schultz, 1963) where, as in the present study, the minerals range only between corrensite and chlorite suggests that the smec- rite to corrensite part of the sequence may not have occurred. That is, corrensite formation by reactions that do not involve a smectite precursor results in sequences of minerals ranging only from corrensite to chlorite.

Direct measurements made in boreholes indicate that temperatures of about 100~ are necessary for the for- mat ion of corrensite (Kfibler, 1973) and this is amply corroborated by more numerous correlations with vitrinite reflectance which are consistently between 0 .6- 0.8% Ro (Table 5). According to Barker and Pawlewicz (1986), these vitrinite reflectance values indicate temperatures between about 90 ~ and 120~ Since corrensite always occurs with chlorite in Orcadian mudrocks, the lower temperature l imit for the formation of corrensite is est imated also to be in the region of 120~ which is in good agreement with the data from the literature (Table 5).

With increasing diagenetic grade corrensite is generally believed to evolve towards chlorite, although there may be no systematic trends. The Orcadian minerals with intermediate composi t ions may be evidence of such an evolutionary sequence. Furthermore, below about 2.5% Ro, Orcadian chlorites have relatively broad diffraction patterns and show traces of expandable layers, consistent with formation, at least in part, from original corrensite. Nevertheless, at low maturi t ies obvious mixed-layer minerals always occur with discrete chlorite, indicating either that chlorite also forms directly and/or the conversion of corrensite to chlorite is very locally controlled, perhaps favouring the for- mat ion of segregated crystallites.

The upper temperature l imit for the stability o f corrensite is not well known, although Kiibler (1973) and Velde (1985) suggested similar temperatures of 250~ and 280~ respectively. More recently, Bettison and Schiffman (1988) est imated 225~ from a study of the Point Sal ophiolite and Inoue and Utada (1991) 200~ for a sequence of Miocene volcaniclastics. Although the data are limited, the distr ibution of corrensite and mixed layer chlorite/corrensite in the Orcadian Basin appears to be confined to samples with vitrinite reflectance values of < 3%. According to Barker and Paw- lewicz (1986), 3% Ro corresponds to a temperature of about 260~ and thus the Orcadian data appear to fit with previous estimates. At vitrinite reflectance values greater than 3%, Orcadian chlorites are characterised by increasingly sharper peak profiles, as illustrated by sample 483 in Figure 3. This may correspond to the final loss ofcorrensi te layers and/or increasing domain size.

In summary, in the Orcadian Basin, there is no pre-

cise trend of expandabil i ty or mixed-layer proport ions that can be related to paleotemperature, only a general paleotemperature interval of about 120"-260~ within which corrensite and a range of mixed-layered minerals variously evolved towards chlorite occur.

Chlorite polytypes. All the chlorites which were examined for polytypism were of the IIb polytype. This is the most stable polytype and is generally considered to be characteristic of high temperature metamorphic chlorites (Hayes, 1970). Recently, however, Walker (1989) showed that authigenic chlorite in very low grade metamorphic pelites and volcanic rocks from northern Maine was exclusively the IIb polytype. Walker suggested that pore pressure and time, as well as temperature, may be impor tant in controlling the occurrence of chlorite polytypes. In the Orcadian Basin, the IIb chlorite polytype has formed at sub-metamorphic temperatures, in some cases probably as low as 120~ a temperature well within the realm of diagenesis. This observation lends support to the conclusions of Walker (1989) and confirms the caution urged by Hayes (1970) in assuming that the presence of the IIb polytype in sedimentary rocks is evidence of a detrital origin.

Origin o f chlorite minerals in the 'evaporite carbonate" association

The origin of Mg-rich chlorite and corrensite in evapori te and carbonate rocks has interested clay min- eralogists ever since early studies in both Europe and North America, summarized by Millot (1964), showed them to be unusually common in these lithologies. In- variably, they were interpreted as the products o f trans- formations of detrital clay minerals, or as authigenic phases, formed by the action of the hypersaline sedimentary milieux. Indeed, syngenetic models have con- t inued to be favoured by more recent studies (Jeans, 1978; Fisher, 1988; L ippmann and Pankau, 1988) and reviews (Hauff, 1981). However, as pointed out earlier, neither corrensite nor Mg-rich chlorite has ever been described from modern evaporit ic environments (Bo- dine and Madsen, 1987). Ordered regularly interstra- tiffed corrensite-like minerals have frequently been described from the weathering environment (Johnson, 1964; Proust et al., 1986); but it is almost certainly misleading to group these minerals with diagenetic cor- rensites. For instance, Wilson and Nadeau (1985) have argued that interstratified clays formed by transfor- mat ion from coarse grained phyllosilicates during weathering are fundamental ly different from the fine grained interstratified clays formed during diagenesis. In accordance with the occurrence ofcorrensi te in other sedimentary rocks and hydrothermal alterations (Table 5), it seems more probable that corrensite is exclusively a mineral of the diagenetic or hydrothermal environment and that it first forms at temperatures of around about 100~


Table 5. Temperature and vitrinite reflectance data for the first appearance of corrensite in diagenetic and hydrothermal sequences.

T (C) Ro % Study

84-91

100 0.5-0.6

0.7-1.2

120-130(e) 0.8

0.6

0.6-0.65

160(e) 0.8

60070 0.65

0.7-0.8

100(e)

Iijima and Utada (1971). Upper Cenozoic marine tuffaceous sediments and tufts, Uetsu geosyn- dine, Japan.

Kiibler (1973), Kiibler et al., (1979). Various.

Chudaev (1978). Palaeocene and Oligocene Volcaniclastic flysch

Stalder (1979). Upper Eocene-Low- er Oligocene, volcaniclastic Tav- eyannaz sandstone.

Kisch (1981). North Range Group, Southland, southern New Zea- land.

Monnier (1982). Tertiary, Swiss Molasse Basin.

Helmold and van de Kamp (1984). Palaeogene marine arkoses, Santa Ynez mountains, California.

Chang et al. (1986). Cretaceous, sandstones and shales, Cassipore Basin, Brasil.

Pollastro and Barker (1986). Terti- ary and Upper Cretaceous, Green River Basin, Wyoming.

Inoue and Utada (1991). Miocene volcaniclastics, Kamikita Japan.

(e) = estimated.

Undoubtedly, however, there are a number of possible reaction pathways by which corrensite may form (Figure 14). Modern evaporitic environments tend to be rich in a variety of Mg-bearing carbonates and silicates, all of which represent potential reactants during burial diagenesis. Bodine and Madsen (1987) suggested that Mg-smectites may have been the precursor ofcorrensite and chlorite in the Paradox Basin, Utah. A further pathway is via reactions between Mg-bearing carbonates and detrital clay minerals, as occurs in the Orcadian Basin. However, the potential importance of such reactions in evaporite and carbonate basins has yet to be investigated. Interestingly, in cases where the distribution of carbonates also has been studied in various detail (Merrel et al., 1957; Jeans, 1978), the patterns are suggestive of clay carbonate reactions, but were interpreted otherwise. Such reactions might also explain why associated illites are often well crystallized. The least crystalline and finest grained dioctahedral clays would be the first to react, leaving behind only well-crystallized diagenetic and/or detrital coarser grained material. To test the hypothesis will require data on facies, whole rock mineralogy, geochemistry, clay mineralogy, and, above all, diagenetic grade. The latter presents the biggest problem because many of

Toe 25

100

\ /

/ \

, . . / ~ ' ~ Z . / - ~v" \

Mg-rich Mg-rich Sepiolite Carbonates Smectites Palygorskite

+ Dioctahedral Clay Minerals

Corrensite Corrensita

1 1 Chlorite Chlorite

1 ?

Figure 14. Schematic representation of pathways to corrensite and chlorite formation in the "evaporite-carbonate" association.

the occurrences are in red beds, which are usually devoid of organic matter. In the burial diagenesis hypothesis, the original presence of various Mg-minerals is directly related to the depositional environment, but the occurrence ofcorrensite and chlorite is a secondary feature that depends upon the degree of burial diagenesis. A compelling example where diagenesis is almost certainly the control was described by Chameley (1989) from the Messinian section at ODP site 652 in the Tyrrhenian Sea. Normally, the evaporite bearing Mes- sinian sediments are devoid of chlorite minerals, but at site 652 they have been subjected to much higher temperatures than normal and chlorite and corrensite are present in abundance.

CONCLUSIONS

The range of chlorite minerals which occur abun- dantly in lacustrine mudrocks from the Orcadian Basin of northern Scotland are authigenic. Chlorite authigenesis occurred by reactions between a detrital dioctahedral clay mineral assemblage and dolomite. Such reactions produce considerable amounts of CO2 and, if they are more common than presently realized, they would have important consequences for diagenesis in sandstones. XRD and microprobe data indicate that minerals which are intermediate between corrensite and chlorite are not mixed-layer chlorite/smectite but mixed-layer chlorite/corrensite with a tendency towards segregation of layer types. The most significant change in the composition of these minerals is an increase in tetrahedral A1 from corrensite to chlorite. There is no change in the ratio of Fe to Mg. Within Mg-rich systems the availability of A1 may be an im-


portant parameter in determining whether it is smectite or corrensite or chlorite that forms. In the Orcadian Basin paleotemperature estimates based on vitrinite reflectance data indicate that the formation of chlorite minerals began at temperatures of about 120~ and that corrensite may persist to temperatures of about 260~ There is no correlation of mixed-layer compositions with temperature.

Similar diagenetic reactions between clay minerals and carbonates may explain the occurrence of corrensite and Mg-rich chlorite in many other evaporite and carbonate successions. Indeed, it is probable that all occurrences of corrensite in such rocks are the result of diagenesis and that it is a reliable indicator of temperatures of about 100~ or more.

A C K N O W L E D G M E N T S

I would like to thank Trevor Clayton and John Mar- shall (Southampton University) for their enthusiastic supervision and Bruce Velde (Ecole Normale Superi- eur) for continuous encouragement and discussion about clays. John Chermak kindly took time to comment on the first version of the manuscript, and it was improved by the comments of Bob Reynolds, an anonymous referee and Reed Glassmann as editor. Work done at the University of Southampton was funded by a NERC CASE studentship with BP Petroleum Development Ltd and at the Ecole Normale Superieur by a post- Doctoral fellowship from The Royal Society, European Science Exchange Program, funded by Elf Aquitaine, both of which are gratefully acknowledged.

REFERENCES

Almon, W. R., Fullerton, k B., and Davies, D. K. (1976) Pore space reduction in Cretaceous sandstones through chemical precipitation of clay minerals: J. Sed. Petrol. 46, 89-96.

April, R. H. (1981) Trioctahedral smeetite and interstrati- fled chlorite/smectite in Jurassic strata of the Connecticut Valley: Clays & Clay Minerals 29, 31-39.

Astin, T. R. and Rodgers, D. A (1991) 'Subaqueous shrink- age cracks' in the Devonian of Scotland reinterpreted. J. Sed. Petrol. 61, 850-859.

Barker, C. E. and Pawlewicz, M. J. (1986) The correlation of vitrinite reflectance with maximum temperature in humic organic matter: in Lecture Notes in Earth Sciences 5. Palaeogeothermics, G. Buntebarth and L. Stegena, eds., Springer-Vedag, Berlin, 79-93.

Bettison, L.A. andSchiffman, P. (1988) Compositionaland structural variations of phyllosilicates from the Point Sal ophiolite, California: Amer. Mineral. 73, 62-76.

Bettison-Varga, L., Mackinnon, I. D. R., and Schiffman, P. (1991) Integrated TEM, XRD, and electron microprobe investigation of mixed-layer chlorite smectite from the Point Sal ophiolite: J. Meta. Geol. 9, 697-710.

Bjorlykke, K. (1988) Sandstone diagenesis in relation to preservation, destruction and creation of porosity: in Dia- genesis 1, Developments in Sedimentology 41, G. V. Chilin- garian and K. H. Wolf, eds., Elsevier, Amsterdam, 555- 588.

Bodine, M.J. (1985) Trioctahedral clay mineral assemblages in Paleozoic marine evaporites: in 6th International Sym- posium on Salt, Vol. 2, B. C. Schreiber and H. L. Harner, eds., Salt Institute, Arlington, Virginia, 267-284.

Bodine, M. W. and Madsen, B. M. (1987) Mixed-layer chlorite/smectites from a Pennsylvanian evaporite cycle, Grand County, Utah: in Proceedings of the International Clay Con- ference Denver, 1985, L. G. Schultz, H. van Olphen, and F. A. Mumpton, eds., The Clay Minerals Society, 85-93.

Chameley, H. (1989) Clay Sedimentology: Springer-Verlag, Berlin, 623 pp.

Chang, H. K., Mackenzie, F. T., and Schoonmaker, J. (1986) Comparisons between the diagenesis of dioctahedral and trioctahedral smectite, Brazilian offshore basins: Clays & Clay Minerals 34, 407-423.

Chudaev, O.V. (1978) Occurrence ofclay minerals in flysch- oid sediments of eastern Kamchatka (in Russian); LitoL Polezn. Iskop. 1, 105-115. (Translation in: LithoL Miner. Resour. 13, 89-97.)

Donovan, R.N. (1980) Lacustrine cycles, fish ecology and stratigraphic zonation in the Middle Devonian of Caith- ness: Scott. J. GeoL 16, 35-50.

Droste, J. B. (1963) Clay mineral composition of evaporite sequences: in Symposium on Salt, Northern Ohio Geolog- ical Survey, 56-63.

Fisher, R. S. (1988) Clay minerals in evaporite host rocks, Palo Duro Basin, Texas Panhandle: J. Sed. Petrol 58, 836- 844.

Fraser, G. C., Harvey, R. D., and Baxter, J. W. (1973) Pa- leoenvironmental significance of corrensite in middle Mis- sissippian rocks in southern Illinois: GeoL Soc. Am. Abst. with Prog. 5, 315.

Hauff, P. L. (1981) Corrensite: mineralogical ambiguities and geologic significance: U.S.G.S. Open-file Report 81- 850, 45 pp.

Hayes, J. B. (1970) Polytypism of chlorite in sedimentary rocks: Clays & Clay Minerals 18, 285-306.

Helmold, K. P. and van de Kamp, P. C. (1984) Diagenetic mineralogy and controls on albitisation and laumontite formation in Paleogene arkoses, Santa Ynez Mountains, Cal- ifornia: in Clastic Diagenesis, D. A. McDonald and R. C. Surdam, eds., Am. Assoc. Petroleum Geol. Mem. 27, Tulsa, Oklahoma, 239-276.

Hillier, S. (1989) Clay mineral diagenesis and organic maturity indicators in Devonian lacustrine mudrocks from the Orcadian Basin, Scotland: Ph.D. thesis, University of Southampton, United Kingdom, 298 pp.

Hillier, S. and Clayton, 1. (1989) Illite/smectite diagenesis in Devonian lacustrine mudrncks from northern Scotland and its relationship to organic maturity indicators: Clay Miner. 24, 181-196.

Hillier, S. and Velde, B. (1991) Octahedral occupancy and the chemical composition of diagenetic (low temperature) chlorites: Clay Miner. 26, 149-168.

Hillier, S. and Clayton, T. (1992) Cation exchange staining of clay minerals in thin section for electron microscopy: Clay Miner. 27, 379-384.

Hillier, S. and Marshall, J. E.A. (1992) Organic maturation, thermal history and hydrocarbon generation in the Orca- dian Basin, Scotland: J. Geol. Soc. Lond. 149, 491-502.

Hower, J. (1981) X-ray diffraction identification of mixed- layer clay minerals: in Clays and the Resource Geologist, F. J. Longstaffe, ed., Mineral. Assoc. Canada Short Course Handbook 7, 39-59.

Hutcheon, I. (1990) Clay-carbonate reactions in the Venture area, Scotian Shelf, Nova Scotia, Canada: in Fluid-Mineral Interactions: A Tribute to H. P. Eugster, R. J. Spencer and I-Ming Chou, eds., The Geochemical Society Special Pub- lication 2, 199-212.


Hutcheon, I., Oldershaw, A., and Ghent, E. D. (1980) Dia- genesis of Cretaceous sandstones of the Kootenay Forma- tion at Elk Valley (southeast British Columbia) and MI. Allan (southwestern Alberta): Geochim. Cosmochim. Acta 44, 1425-1435.

Ijima, A. and Utada, M. (1971) Present-day zeolitic diagenesis of the Neogene geosynclinal deposits in the Niigata oil field, Japan: in Molecular sieve zeolites" L Advances in chemistry series 101, 342-348.

Inoue, A. (1985) Chemistry of corrensite: A trend in composition of trioctahedral chlorite/smectite during diagenesis: Journal of the College of Arts and Sciences, Chiba Uni- versity B-18, 69-82.

Inoue, A. (1987) Conversion of smectite to chlorite by hydrothermal and diagenetic alterations, Hokuroku Kuroko mineralization area, northeast Japan: in Proceedings of the International Clay Conference, Denver. 1985, L. G. Schultz, H. van Olphen, and F. A. Mumpton, eds., The ClaY Min- erals Society 158-164.

Inoue, A. and Utada, M. (1991) Smectite to chlorite transformation in thermally metamorphosed volcanoclastic rocks in the Kamikita area, northern Honshu, Japan: Amer. Min- eral 76, 628-640.

Janaway, T. M. and Parnell, J. (1989) Carbonate production within the Orcadian Basin, Northern Scotland: A petrographic and geochemical study: Palaeogeography, Palaeo- climatology, Palaeoecology 70, 89-105.

Jeans, C. V. (1978) The origin of the Triassic clay assemblages of Europe with special reference to the Keuper Marl and Rhaetic of parts of England: Phil Trans. R. Soc. Lond. 289, 549-639.

Jones, B. F. (1986) Clay mineral diagenesis in lacustrine sediments. Studies in Diagenesis: U.S.G.S. Bull. 1578, 29 i-- 300.

Johns, W. D., Grim, R. E., and Bradley, W.F. (1954) Quan- titative estimations of clay minerals by diffraction methods: J. Sed. Petrol. 24, 242-255.

Johnson, L.J. (1964) Occurrenceofaregularlyinterstratified chlorite-vermiculite as a weathering product of chlorite in a soil: Amer. Mineral. 49, 556-572.

Kisch, H. J. (1981) Coal rank and iilite crystallinity ass0, cialed with the zeolite facies of Southland and the pum- pellyite-bearing facies of Otago, southern New Zealand: N. Zeal J. Geol. Geophys. 24, 349-360.

Kiibler, B. (1973) La corrensite, indicateur possible de milieux de srdimentation et du degr6 de transformation d'un srdiment: Bull. Centre Rech. Pau-S.N.P.A. 7, 543-556.

Kiibler, B., Pittion, J. L., Heroux, Y., Charollais, J., and Wied- man, M. (1979) Sur le pouvoir reflecteur de la vitrinite dans quelques roches du Jura, de la Molasse et de la Nappes prralpines, helvrtiques et penniques: Eclogae Geol. Heir. 72, 347-373.

Laird, J. (1988) Chlorites: Metamorphic petrology: in Hy- drous Phyllosilicates (Exclusive of Micas), S. W. Bailey, ed., Reviews in Mineralogy 19. Mineralogical Society of Amer- ica, Washington, D.C., 725 pp.

Lanson, B. and Besson, G. (1992) Characterisation of the end of smectite to illite transformation: Decomposition of X-ray patterns: Clays & Clay Minerals 40, 40-52.

Lee, J, H., Ahn, J. H., and Peacor, D. R. (1985) Textures in layered silicates: Progressive changes through diagenesis and low temperature metamorphism: J. Sedim. Petrol. 55, 532-540.

Lippmann, F. andPankau, H-G. (1988) DerMineralbestand des Mittleren Mushelkalkes yon Nalgold, Warttemberg: Neues Jahrbuch Miner. Abh. 158, 257-292.

Lucas, J. (1962) La transformation de minrraux argileux clans la srdimentation: l~.tudes sur les argiles du Trias. M(m. Surv. Carte G~ol. lsace Lorraine 23, 1-202.

Merrel, H. W., Jones. D. J., and Sand. L. B. (1957) Sedi- mentation features in paradox shales, southeastern Utah. Bull. GeoL Soc. Am. 68, 1766.

Meunier, A., Clement, J., Bouchet, A., and Beaufort, D. (1988) Chlorite-calcite and corrensite-dolomite crystalli- zation during two superimposed events of hydrothermal alteration in the "les Cretes" granite, Vosge, France: Ca- nadian Miner. 26, 413--422.

Millot, G. (1964) G~ologie des Argiles: Masson et Cie, Paris, 499 pp.

Monnier, F. (1982) Thermal diagenesis in the Swiss molasse Basin: Implications for oil generation: Can. J. Earth Sci. 19, 328-342.

Muffler, L. J. P. and White, D. E. (1969) Active metamorphism of upper Cenozoic sediments in the Salton Sea geothermal field and the Salton Trough, Southeast California: Geol. Soc. Am. Bull. 80. 157-182.

Parnell, J. (1985) Evidence for replaced evaporites in the ORS of northern Scotland: Replaced gypsum horizons in Easter Ross: Scott. J. GeoL 21. 377-380,

Peterson, M. N. A. (1961) Expandable chloritic clay minerals from the Upper Mississippian carbonate rocks of the Cumberland Plateau in Tennessee: Amer. Mineral 46, 1245- t269.

Pettijohn, F .J . (1957) Sedimentary rocks: 2nd ed., Harper, New York, 718 pp.

Pollastro, R. M. and Barker, C. E. (1986) Application of clay mineral, vitrinile reflectance and fluid inclusion studies to the thermal and burial history of the Pinedale anticline, Green River Basin, Wyoming: in Roles of Organic Matter in Sediment Diagenesis, D. L. Gautier, ed., Soc. Econ. Pa- leontol. Mineral. Spec. Publ. 38, 73-83.

Proust, D., Eymery, j.-P., and Beaufort, D. (1986) Super- gene vermiculitization of a magnesian chlorite: Iron and magnesian removal processes: Clays & Clay Minerals 34, 572-580.

Rao, C. G. and Bhattacharya, N. (1973) Corrensite from the Sirban limestone of Riasi, Jammu, and Kashmir slate, In- dia: J. Geol. Soc. India 14, 193-196.

Reynolds Jr., R. C. (1985) NEWMOD �9 A computer program for the calculation of the one-dimensional pattern s of mixed-layered clays: R. C. Reynolds, 8 Brook Rd., Hano- ver, New Hampshire.

Reynolds Jr., R. C. (1988) Mixed-layer chlorite minerals: in Hydrous Phyllosilicates (Exclusive of Micas), S. W. Bai- ley, ed., Reviews in Mineralogy 19. Mineralogical Society of America, 601-629.

Rogers, D. A. and Astin, T. R. (1991) Ephemeral lakes, mud pellet dunes and wind blown sand and silt: Reinter- pretations of Devonian lacustrine cycles in north Scotland: in Lacustrine Facies Analysis, P. Anadon, L. Cabrera, and K. Kelts, eds., International Association of Sedimentolo- gists, Special Publication 13, 201-223.

Schiffman, P. and Fridleifsson, G. O. (1991) The smectite- chlorite transition in driUhole NJ-15, Nesjavellir geothermal field, Iceland: XRD, BSE and electron microprobe in: vestigations. J. Meta. GeoL 9, 679-696.

Schultz, L. G. (1963) Clay minerals in the Triassic rocks of the Colorado Plateau: USGS Bull. 1147-C, 71 pp.

Shau, Y. H., Peacor, D. R., and Essene, E. J. (1990) Cor- rensite and mixed-layer chlorite/corrensite in metabasalt from northern Taiwan: TEM/AEM, EMPA, XRD, and op- tical studies: Contrib. Miner. Petrol 105, 123-142.

Stalder, P. (1979) Organic and inorganic metamorphism in the Taveyannaz sandstone of the Swiss Alps and equivalent sandstones in France and Italy: J. Sed. Petrol 49, 463--482.

Velde, B. (1985) Clay minerals: A physico chemical explanation of their occurrence. Developments in Sedimentology No. 40: Elsevier, Amsterdam, 421 pp.


Walker, J.R. (1989) Polytypism ofchlorite in very low grade metamorphic rocks. Arner. Mineral. 74, 738-743.

Wiewi6ra, A. and Weiss, Z. (1990) Crystallochemical clas- sifications of phyllosilicates based on the unified system of projection of chemical composition: II. The chlorite group: Clay Miner. 25, 83,92.

Wilson, M.J. (1971) Clay mineralogy of the Old Red Sand- stone (Devonian) of Scotland: J. Sed. Petrol 41,995-1007.

Wilson, M. J. and Nadeau, P. H. (1985) Interstratified clay

minerals and weathering processes: in The Chemistry of Weathering. J. I. Drever, ed., NATO ASI Series D, Reidel, Dordrecht, Holland, 97-118.

Zen, E.A. (1959) Clay mineral carbonate relations in sedimentary rocks: Am. J. Sci. 257, 29-43.

(Received 18 March 1993," accepted 19 March 1993; Ms. 2352)

origin, diagenesis, and mineralogy of chlorite …clays.org/journal/archive/volume...

Documents