aspects of material transport & polycrystal growth in ... habil-th… · geologie-endogene...

Aspects of material transport and

polycrystal growth in fractures

Der Fakultät für Georessourcen und Materialtechnik der Rheinisch-Westfälischen

Technischen Hochschule Aachen vorgelegte Habilitationsschrift

von

Dr. Christoph Hilgers

Aachen, im Februar 2005

i

Acknowledgments I am most indebted to my mentor Janos Urai for many discussions about science and

beyond. I would like to thank Paul Bennema, Paul Bons, Jean-Paul Breton, Stephen Cox,

Mark Jessell, David Kirschner, Win Means, Sven Sindern, and Eric Tenthorey, who

introduced me to their special expertise, for their discussions and suggestions.

Experimental work is impossible without experienced staff, and I am much obliged to

Werner Kraus for countless help in improving sample preparation techniques, and Franz-

Dieter Scherberich for many discussions on designs of experimental set-ups.

There are far more people to honour, colleagues at RWTH, people who hosted me during

my stays abroad, and people beyond science. I am most obliged to my colleagues at

Geologie-Endogene Dynamik, who contributed to this work by some means or other,

Bettina Dulle, Oliver Schenk and Zsolt Schleder. I would especially like to acknowledge

my PhD students Sofie Nollet and Maria Chatziliadou, who have to resist my restlessness

and always hear me asking “what do you want to do and why do you want to do it”.

My special thanks are due to my family, for their patience and their love.

ii

ZusammenfassungDiese Arbeit behandelt den Stofftransport und das Wachstum von Kristallen in Klüften.

Für ein besseres Prozessverständnis werden experimentelle Ansätze und detaillierte

Analysen von Geländebeobachtungen herangezogen.

Kapitel 2 beschreibt epitaxial gewachsene Mikrostrukturen anhand eines Durchlicht-

Analogexperimentes. Der advektive Stofftransport einer übersättigten Lösung führt zu

einer Versiegelung des Einströmbereiches und dem Wachstum von elongiert-blockigen

Kristallen. Diese Ergebnisse werden auf ein natürliches System adaptiert und deuten

darauf hin, daß das Wachstum von Quarz zu einer relativ homogenen Zementation der

Kluft führt. Daher ist innerhalb eines geologischen Aufschlusses kaum eine Änderung der

neugewachsenen Körner zu erwarten. Kapitel 3 präsentiert erste Hydrothermal-

experimente zur Versiegelung von Klüften in Quarzit. Bei einem Ringraumdruck von 200

MPa und Temperaturen bis 900°C wurde ein axialer Stofftransport entlang eines

Temperaturgradienten zwischen 45-125°C generiert, der zur Ausfällung von Quarz in der

Kluft führte. Ein Mineralwachstum wurde vor allem am Anfang der Kluft beobachtet. In

Kapitel 4 wird die Anordnung von Nebengesteinseinschlüssen innerhalb von drei unter-

schiedlichen, faserigen Kluftüllungen behandelt. Diese erlauben Rückschlüsse auf den

Wachstumsprozeß während der Versiegelung. Die Ergebnisse zeigen, daß die Anordnung

von Einschlüssen parallel zur Kluftwand kein hinreichendes Kriterium ist, um den Bruch-

Versiegelungs-Mechanismus abzuleiten. Alternativ wird ein kontinuierliches Wachstum

der Kluftminerale diskutiert. In Kapitel 5 werden faserige, antitaxiale Mikrostrukturen

mit stabilen ( 13C, 18O) und radiogenen (87Sr/86Sr) Isotopensignaturen verglichen.

Obwohl die Mikrostrukturen eine lokale Stoffquelle vermuten lassen, deuten die

Isotopensignaturen einen Beitrag externer Fluide an. Kapitel 6 erläutert die Versiegelung

von Klüften im Jabal Akhdar Dom, Oman. Die Raumlage der Kluftfüllungen weist auf

eine zweimalige Änderung der Hauptnormalspannungen hin. Stabile Isotope ( 13C, 18O)

deuten ein zunächst geschlossenes Fluidsystem an, was mit Fluid-Überdrücken und

Bildung von Extensionsklüften einhergeht. Während der Genese von Abschiebungen

öffnet sich das Fluidsystem und exotische Wässer mit leichteren 18O Signaturen fliessen

zu. In diesem offenen Stadium baut sich der Fluid-Überdruck ab.

iii

Contents

Chapter 1 - Introduction 1

Sealing mechanisms 1 Microstructure 3 Fracture mechanisms to form veins 4 Aims 5 References 7

Chapter 2 - Microstructural evolution of syntaxial veins formed by advective flow 11

Abstract 11 Introduction 12 Experimental setup 13 Results 14 Application to quartz veins 17 Discussion 18 References 21

Chapter 3 - Fracture sealing of quartzite under a temperature gradient: Experimental results 23

Abstract 23 Introduction 24 Experimental method 25 Results 28 Discussion 32 Conclusion 34 References 36

Chapter 4 - On the arrangement of solid inclusions in fibrous veins and the role of the crack-seal mechanism 38

Abstract 38 Introduction 39 Sample description 41 Discussion 48 Conclusion 59 References 61

iv

Chapter 5 - Textural and isotopic evidence on the fluid source and transport mechanism of antitaxial fibrous microstructures from the Alps and the Appalachians 65

Abstract 65 Introduction 66 Methods 68 Sample locations and vein microstructures 69 Stable and radiogenic isotope data 74 Discussion 77 Conclusion 83 References 84

Chapter 6 - Fracture sealing in a regional, high-pressure cell in Jabal Akhdar, Oman mountains - first results 88

Abstract 88 Introduction 89 Geological setting 91 Veins 92 Methodology of isotope analyses 98 Results of isotope analyses 98 Discussion 102 Conclusion 110 References 115

Chapter 1 - Introduction

1


Fractures play a major role in the transport of matter, because they significantly increase

the permeability within the earth’s brittle crust (Norton & Knapp 1977, Jamtveit &

Yardley 1997, Cox et al. 2001). This allows fluids to travel over long distances up to

basin scale (Oliver 1996, Gross et al. 1992). The fluid’s saturation may increase when the

pressure, temperature, host rock composition or the fluid chemistry changes, causing

precipitation of minerals (Rimstidt 1997). Such precipitates are called veins, which are of

great economic interest because they can carry ore minerals, or seal hydrocarbon

pathways (Barnes & Rose 1998, Aydin 2000). Artificial examples are the scaling of

boreholes, and the salting of hydro-fractured hydrocarbon or geothermal reservoirs

(Jasinski et al. 1998).

Fluids do not necessarily flow passively along fluid pathways, but may interact with the

host rock and induce failure (Hubbert & Rubey 1959). Fluid overpressures are frequently

used to explain fracture formation (Engelder 1990), and are well known in many

sedimentary basins (Ortoleva 1994, Law & Spencer 1998).

Veins contain information (i) on the source of fluid; derived from their composition or

isotope analyses, (ii) p-T conditions; derived from fluid inclusions or stable isotope

analyses, (iii) stresses active during vein formation; derived from the vein’s orientation

and the rheology of the host rock, and (iv) palaeo-supersaturation; derived from the vein

microstructure. This study is focused on the formation and sealing of fractures at a micro-

and mesoscale.

Sealing mechanisms

Basically, four mechanisms may cause vein formation. In the force of crystallization

model, crystals push apart the surrounding rock and form veins (Means & Li 2001), a

process well known from salt efflorence on bricks (Fitzner & Snethlage 1982, Zehnder &

Arnold 1989). It is caused by the local induced stress generated by the supersaturation

driven growth (Fletcher & Merino 2001). Crystallisation pressures are calculated to be 80

MPa (calcite) and 300 MPa (quartz) for low supersaturations of 2, where the

supersaturation is defined as the concentration in the supersaturated solution divided by

the concentration in an equilibrium solution (Wiltschko & Morse 2001). This growth


2

process is sometimes called Taber growth (Fletcher & Merino 2001, Means & Li 2001,

Elburg et al. 2002) in honour of Stephen Taber (Taber 1916a; Taber 1916b; Taber 1918),

but research on the force of crystallization and its relationship to veins started at least 70

years earlier (Bunsen 1847, Kenngott 1855, Schmidt 1911, Becker & Day 1916). Force

of crystallization growth is a continuous growth process without the vein loosing contact

to the wall. Solute migrates from the wall towards the vein and precipitates at the base of

the vein crystals (Means & Li 2001). Markers in the host rock are being displaced during

vein growth, which means that vein growth causes dilation of the rock. This requires an

opening of a void, either at the tip of an individual grain in a pore of the host rock, or a

small fluid film along the vein wall interface.

The second opening mechanism which involves dilation is the repeated fracturing of rock

(Hulin 1929, Ramsay 1980). Evidence of multiple fracturing and sealing events to form

veins was inferred from solid inclusion bands within the vein, which are aligned parallel

to the vein wall (Ramsay 1980, Fisher & Brantley 1992, Laubach et al. 2004). The

alignment of these inclusion bands matches with the morphology of the vein wall, and

each band is interpreted to represent an individual crack-seal increment (Ramsay 1980).

Some authors correlate individual crack-seal increments with earthquake events (e.g.

Remitti & Sibson 2004).

The third vein growth mechanism explains veins as rapidly ascending, mobile

hydrofractures (Bons 2001). The vertical, fluid filled fracture exposes a difference in the

pressure gradient in the fluid and the adjacent host rock. This causes a fluid overpressure

at the upper fracture tip if the fracture is sufficiently large (Walther 1990), and drives the

fluid filled fracture upwards (Bons 2001). The velocity is in the range of m/s in order to

avoid nucleation and precipitation from the fluid before the fluid filled fracture is arrested

in an overlying layer. Bons (2001) uses his model to explain a reasonable supersaturation

to form quartz veins, and suggested an upward movement of the fluid filled fracture over

a distance of app. 10 km (corresponding to a temperature gradient of 250°C). Similar to

the mechanisms outlined above, this opening mechanism is dilational. The difference is

that a closed fracture should be found at the lower end of the vein.

The fourth vein growth mechanism forms replacement veins, which is characterized by

chemical replacement of the wall rock during dissolution-precipitation creep (Dunne &


3

Hancock 1994, Janssen & Bons 1996). However, dissolution precipitation creep was also

pushed forward as mechanism to cause dilational veins (Durney 1972b, Durney &

Ramsay 1973, Durney 1976).

Microstructure

The vein microstructure provides qualitative information on the fracture width and the

supersaturation. If the fracture is wide open, crystals develop crystal facets and outgrow

their neighbours according to their crystallographic orientation (Dickson 1993). If the

open space is reduced during vein growth, the microstructure becomes elongate-blocky

and finally fibrous (Urai et al. 1991, Hilgers et al. 2001). Blocky microstructures form by

primary nucleation from a supersaturated fluid and thus require a higher supersaturation

than epitaxial overgrowths on pre-existing crystals (Hesse 1987, Mullin 1993 p.185).

The vein may also contain information on the opening kinematics of the rock during

deformation, if the microstructure is fibrous. The fibre grain boundaries are assumed to

track the direction of the fracture opening, and the temporal variation in opening direction

can be inferred from the microstructure (Durney & Ramsay 1973). This observation can

be applied to a regional scale and shows the variation of movement directions during

deformation over time (Durney 1972a, Ramsay & Huber 1983). However, such fibrous

microstructures are rare and require careful microstructural analyses. Careful

petrographic observations are needed because fibre grain boundaries do not track the

opening direction in all cases (Cox 1987, Urai et al. 1991).

Various minerals can grow in a needle-like fibrous shape, which is not the mineral’s

typical morphology, as shown for minerals such as calcite, quartz, gypsum or halite

(Mügge 1928). The minerals are aligned parallel to each other, sub-normal to the vein-

wall interface, and are characterized by the absence of significant growth competition.

This is explained by the limited space available during growth (Mügge 1925; Mügge

1928, Hilgers et al. 2001).

If fibres connect grains hosted in the wall rock, the fibrous microstructure is called

stretched crystals (Durney & Ramsay 1973) or ataxial (Passchier & Trouw 1996). The

veins have serrated grain boundaries, which are not strictly parallel to each other. It was

suggested that these microstructures grow by random fracturing within the vein


4

(Passchier & Trouw 1996), although other studies suggest growth at the vein wall

interface (Dunne & Hancock 1994, Hilgers & Urai 2002). The second type of fibrous

microstructure is called antitaxial (Durney & Ramsay 1973). These veins grow from the

vein centre towards the vein wall, and do not continue across the vein wall interface into

the wall rock. The curvature of fibrous antitaxial crystals is not caused by deformation

but crystals have grown in a curved shape, since the crystal lattice is almost undeformed

(Spencer 1991). These microstructures can be used to infer the opening directions.

Durney and Ramsay (1973) defined a third type of fibrous vein growing from both sides

of the wall into the vein. These syntaxial veins are not fibrous sensu stricto, but show

some widening of grains towards the vein centre. Thus, they are called elongate-blocky

since the microstructure shows growth competition (Fisher 1996, Bons 2000). The

widening of grains towards the vein centre indicates that the fracture was sealed by

minerals grown from the vein wall towards the vein centre and that veins grew in a larger

void. All these elongate-blocky to fibrous microstructures will be destroyed by

recrystallisation at higher p-T conditions.

Fracture mechanisms to form veins

It is not yet clear if all sealing mechanisms require fractures to form veins. For example,

some authors assume antitaxial veins to grow during a continuous growth process without

fracturing taking place (e.g. Means & Li 2001, Elburg et al. 2002, Bons & Montenari in

press). Hilgers and Urai (2002) show fractured host rock grains which imply early brittle

fracturing, and suggest that during further vein growth crack-sealing is absent (see also

Bons & Montenari in press). Such continuous vein growth may be related to ongoing

dissolution precipitation creep (Durney 1972b, Durney & Ramsay 1973, Durney 1976).

Evidence for a continuous growth process is given in chapter four and five, where we

discuss the arrangement of solid host rock inclusions and show the coexistence of an

antitaxial vein and a quartz pressure shadow. Based on the regular spacing of veins we

suggest that tectonic stresses do play a major role in the formation of veins (chapter four).

Syntectonic vein formation requires mode I or mixed-mode transitional-tensile failure

which forms open voids (e.g. Shearman et al. 1972, Cosgrove 1995, Engelder 1999,

Schultz 2000). In either case, the minimum principal effective stress 3‘ is tensile and


5

causes the opening of the rock fracture. Mode I tensile fractures form at low differential

stresses 1- 3 4T and 3‘=-T, with T the tensile strength of rock; mixed-mode (or hybrid

extension) fractures at 4T< 1- 3<5.66T and –T< 3‘<0 (Dunne & Hancock 1994, Sibson

1998, Schultz 2000). Vertical extension fractures have also been described in settings

where the minimum effective stress is compressive (Mandl 2000, p.89ff). Mandl (2000)

calls fractures formed at compressive effective stresses extensional (“cleavage”)

fractures, and explains them by the coalescence of winged cracks (see also Horii &

Nemat-Nasser 1985, Bobet 2000). Bourne (2003) doubts that such flaws may propagate

to dimensions much greater than the initial flaw. He suggests that extension fractures may

form in the stiff layer due to contrast in the elastic properties of an anisotropic, bonded

multiplayer.

Veins may also emplace during shear failure in dilational jogs, which form in the

releasing bends of small ramps along the fault plane (e.g. Cox 1987, Koehn & Passchier

2000, summary of the different mechanisms is given in Jessell et al. 1994). In this case

no fluid overpressure is required to facilitate fracturing. However, overpressure can pop

up older fractures and crystal growth may form a vein (e.g. Jolly & Sanderson 1997,

Sanderson & Zhang 2004).

Aims

In this thesis, I present new data on fracture sealing processes using microscale and

mesoscale observations.

Some of the questions I address in this thesis are:

What do vein microstructures look like if matter is transported by along-fracture transport

of supersaturated solution? May one deduce the transport process from the microstructure

observed in a vein? Can this be a tool used in the field?

Chapter two presents analogue experiments, which show the temporal evolution of the

microstructure and its variation in growth rate along-vein. Chapter three presents the first

sealing experiments of quartz at hydrothermal conditions.

Does the arrangement of solid inclusions necessarily require a crack-seal process?


6

Chapter four describes solid host rock inclusions in veins and shows that the arrangement

of solid inclusion bands cannot be explained by the crack-seal mechanism, which is

frequently applied to this type of microstructures.

Can one infer from an antitaxial microstructure to a special transport process?

Chapter five presents microstructural and isotope data ( 18O, 13C and 87Sr/86Sr) of

antitaxial veins. This study shows that the calcite to form antitaxial veins was not

completely derived locally although microstructural evidence suggests near field

transport.

How does a fluid system develop in an inverted sedimentary basin?

Chapter five presents isotope data ( 13C, 18O) sampled in Mesozoic limestones from

different calcite vein generations. The fluid system changes from a closed rock-buffered

system to an open system with the onset of brittle normal faults. Field observations are

discussed with respect to fluid overpressures.

The author contributed more that 50% to the papers presented in this thesis. Other papers,

which are related to the topic of this thesis, are

Hilgers, C, Dilg-Gruschinski, K, Urai, JL (2003) Microstructures grown experimentally from advective supersaturated solution and their implication for natural vein systems. Journal of Geochemical Exploration, 4031, 1-5.

Nollet, S, Hilgers, C, Urai, JL (subm) Sealing of fluid pathways in overpressure cells - a case study from the Buntsandstein in the Lower Saxony Basin (NW Germany). International Journal of Earth Sciences.

Nollet, S, Urai, JL, Bons, PD, Hilgers, C (in press) Numerical simulations of polycrystal growth in veins. Journal of Structural Geology.


7

References

Aydin, A (2000) Fractures, faults, and hydrocarbon entrapment, migration and flow. Marine and Petroleum Geology, 17, 797-814.

Barnes, HL, Rose, AW (1998) Origins of hydrothermal ores. Science, 279, 2064-2065. Becker, GF, Day, AL (1916) Note on the linear force of growing crystals. Journal of

Geology, 24(4), 313-333. Bobet, A (2000) The initiation of secondary cracks in compression. Engineering fracture

mechanics, 66, 187-219. Bons, PD (2000) The formation of veins and their micostructures. Journal of the Virtual

Explorer, 2.Bons, PD (2001) The formation of large quartz veins by rapid ascent of fluids in mobile

hydrofractures. Tectonophysics, 336(1-4), 1-17. Bons, PD, Montenari, M (in press) The formation of antitaxial calcite veins with well-

developed fibres, Oppaminda Creek, South Australia. Journal of Structural Geology, 1-18.

Bourne, SJ (2003) Contrast of elastic properties between rock layers as a mechanism for the initiation and orientation of tensile

failure under uniform remote compression. Journal of Geophysical Research, 108(B8),10.1029/2001JB001725.

Bunsen, R (1847) Ueber den innern Zusammenhang der pseudovulkanischen Erscheinungen Islands. Annalen der Chemie und Pharmacie, Band LXIII(1.Heft), 1-59.

Cosgrove, JW (1995) The expression of hydraulic fracturing in rocks and sediments. In: Fractography: Fracture topography as a tool in fracture mechanics and stress analysis. (ed) M.S. Ameen, pp. 187-196. Geological Society, London.

Cox, SF (1987) Antitaxial crack-seal vein microstuctures and their relationship to displacement paths. Journal of Structural Geology, 9(7), 779-787.

Cox, SF, Knackstedt, MA, Braun, J (2001) Principles of structural control on permeability and fluid flow in hydrothermal systems. Reviews in Economic Geology, 14, 1-24.

Dickson, JAD (1993) Crystal growth diagrams as an aid to interpreting the fabrics of calcite aggregates. Journal of Sedimentary Petrology, 63(1), 1-17.

Dunne, WM, Hancock, PL (1994) Paleostress analysis of small-scale brittle structures. In: Continental Deformation. (ed) P.L. Hancock pp. 101-120. Pergamon Press, Oxford.

Durney, DW (1972a) Deformation history of the western Helvetic nappes, Valais, Switzerland. PhD Thesis, Imperial College, London.

Durney, DW (1972b) Solution-transfer, an important geological deformation mechanism. Nature, 235, 315-317.

Durney, DW (1976) Pressure-solution and crystallization deformation. PhilosophicalTransaction of the Royal Society London, Series A, 283, 229-240.

Durney, DW, Ramsay, JG (1973) Incremental strains measured by syntectonic crystal growth. In: Gravity and tectonics. (eds) K.A. de Jong and R. Scholten, pp. 67-96. Wiley, New York.


8

Elburg, MA, Bons, PD, Foden, J, Passchier, CW (2002) The origin of fibrous veins: constraints from geochemistry. In: Deformation mechanisms, rheology and tectonics: Current status and future perspectives. (eds) S. de Meer, M.R. Drury, J.H.P. de Bresser and G.M. Pennock. Special Publications, pp. 103-118. Geological Society, London.

Engelder, T (1990) Smoluchowski's dilemma revisited: A note on the fluid pressure history of the central Appalachian flod-thrust belt. In: Studies in geophysics. The role of fluids in crustal processes. (ed) G.S.C.N.R. Council, pp. 140-147. National Academic Press, Washington.

Engelder, T (1999) Transitional-tensile fracture propagation: a status report. Journal of Structural Geology, 21, 1049-1055.

Fisher, DM (1996) Fabrics and veins in the forearc: A record of cyclic fluid flow at depth of <15km. In: Subduction: Top to bottom. (eds) G.E. Bebout, D.W. Scholl, S.H. Kirby and J.P. Platt, pp. 75-89. American Geophysical Union

Fisher, DM, Brantley, SL (1992) Models of quartz overgrowth and vein formation: Deformation and episodic fluid flow in an ancient subduction zone. Journal of Geophysical Research, 97(B13), 20,043-20,061.

Fitzner, B, Snethlage, R (1982) Über Zusammenhänge zwischen Salzkristallisationdruck und Porenradienverteilung. GP News Letter, 3, 13-24.

Fletcher, RC, Merino, E (2001) Mineral growth in rocks: Kinetic - rheological models of replacement, vein formation, and syntectonic crystallization. Geochimica et Cosmochimica Acta, 65(21), 3733-3748.

Gross, MR, Engelder, T, Poulson, SR (1992) Veins in the Lockport dolostone: Evidence for an Acadian fluid circulation system. Geology, 20, 971-974.

Hesse, R (1987) Selective and reversible carbonate-silica replacements in Lower Cretaceous carbonate-bearing turbidites of the Eastern Alps. Sedimentology, 34,1055-1077.

Hilgers, C, Koehn, D, Bons, PD, Urai, JL (2001) Development of crystal morphology during unitaxial growth in a progressively widening vein: II. Numerical simulations of the evolution of antitaxial fibrous veins. Journal of Structural Geology, 23, 873-885.

Hilgers, C, Urai, JL (2002) Microstructural observations on natural syntectonic fibrous veins: Implications for the growth process. Tectonophysics, 352, 257-274.

Horii, H, Nemat-Nasser, S (1985) Compression-induced microcrack growth in brittle solids: Axial splitting and shear failure. Journal of Geophysical Research, 90(B4),3105-3125.

Hubbert, MK, Rubey, WW (1959) Role of fluid pressure in mechanics of overthrust faulting. Bulletin of the Geological Society of America, 70, 115-166.

Hulin, CD (1929) Structural control of ore deposition. Economic Geology, 24(1), 15-49. Jamtveit, B, Yardley, WD (1997) Fluid flow and transport in rocks: An overview. In:

Fluid flow and transport in rocks. Mechanisms and effects. (eds) B. Jamtveit and B.W.D. Yardley, pp. 1-14. Chapman&Hall, London.

Janssen, C, Bons, PD (1996) Replacement versus crack-seal mechanism: An alternative explanation of laminated carbonates from the Görlitzer Schiefergebirge, Germany. Zeitung für geologische Wissenschaften, 24(3/4), 377-386.


9

Jasinski, R, Fletcher, P, Taylor, K, Sablerolle, W (1998) Calcite scaling tendencies for North Sea HTHP wells: Prediction, authentication and application. In: SPEAnnual Technical Conference and Exhibition. Society of Petroleum Engineers, New Orleans, Louisiana, pp. 661-669.

Jessell, MW, Willman, CE, Gray, DR (1994) Bedding parallel veins and their relationship to folding. Journal of Structural Geology, 16(6), 753-767.

Jolly, RJH, Sanderson, DJ (1997) A Mohr circle construction for the opening of a pre-existing fracture. Journal of Structural Geology, 19(6), 887-892.

Kenngott, A (1855) Mineralogische Notizen, betreffend die bekannten Species: Karstenit, Dolomit, Millerit, Turmalin, Galaktit, Wasser, Plagionit, Diopsid, Zinkit, Calcit und Felsöbanyt, sowie zwei neue: den Enstatit im Geschlechte der Augit-Spathe und den Pseudophit im Geschlechte der Serpentin-Steatite. Sitzungsberichte der Akademie der Wissenschaften XVI, Bd. I, 152-179.

Koehn, D, Passchier, CW (2000) Shear sense indicators in striped bedding-veins. Journalof Structural Geology, 22, 1141-1151.

Laubach, SE, Reed, RM, Olson, JE, Lander, RH, Bonnell, LM (2004) Coevolution of crack-seal texture and fracture porosity in sedimentary rocks: cathodoluminescence observations of regional fractures. Journal of Structural Geology, 26, 967-982.

Law, BE, Spencer, CW (Editors), 1998. Abnormal pressure in hydrocarbon environments, 70. AAPG Memoir, 264 pp.

Mandl, G (2000) Faulting in brittle rocks. 434 pp. Springer, London. Means, WD, Li, T (2001) A laboratory simulation of fibrous veins: some first

observations. Journal of Structural Geology, 23, 857-863. Mügge, O (1925) Über gehemmtes Kristallwachstum. Zeitschrift für Kristallographie,

62(5/6), 415-442. Mügge, O (1928) Ueber die Entstehung faseriger Minerale und ihrer

Aggregationsformen. Neues Jahrbuch für Mineralogie, Geologie und Paläontologie, 58(A), 303-348.

Mullin, JW (1993) Crystallization. 527 pp. Butterworth-Heinemann, Oxford. Norton, D, Knapp, R (1977) Transport phenomena in hydrothermal systems: The nature

of porosity. American Journal of Science, 277, 913-936. Oliver, NHS (1996) Review and classification of structural controls on fluid flow during

regional metamorphism. Journal of Metamorphic Geology, 14, 477-492. Ortoleva, PJ (Editor), 1994. Basin compartments and seals. AAPG Memoir 61, 477 pp. Passchier, CW, Trouw, RAJ (1996) Microtectonics. 289 pp. Springer, Berlin. Ramsay, JG (1980) The crack-seal mechanism of rock deformation. Nature, 284(5752),

135-139.Ramsay, JG, Huber (1983) Techniques in modern structural geology. Volume 1: Strain

analysis. 307 pp. Academic Press, London. Remitti, F, Sibson, RH (2004) Incremental growth of quartz slickenfibres on shearing

surfaces in an ancient subduction complex - record of paleomicroearthquakes? In: Eos Trans. AGU, 85(47), Fall Meet. Suppl., Abstract S43D-03.

Rimstidt, JD (1997) Gangue mineral transport and deposition. In: Geochemistry of hydrothermal ore deposits. (ed) H.L. Barnes, pp. 487-515


10

Sanderson, DJ, Zhang, X (2004) Stress-controlled localization of deformation and fluid flow in fractured rocks. In: The inititation, propagation, and arrest of joints and other fractures. (eds) J.W. Cosgrove and T. Engelder, pp. 299-314. Geological Society, London.

Schmidt, R (1911) Beschaffenheit und Entstehung parallelfaseriger Aggregate von Steinsalz und von Gips. PhD Thesis, Christian Albrechts-Universität zu Kiel, Kiel, 25 pp.

Schultz, RA (2000) Growth of geologic fractures into large-strain populations: review of nomenclature, subcritical crack growth, and some implications for rock engineering. International Journal of Rock Mechanics and Mining Sciences, 37,403-411.

Shearman, DJ, Mossop, G, Dunsmore, H, Martin, M (1972) Origin of gypsum veins by hydrauilc fracture. Z1598, B149-B155.

Sibson, RH (1998) Brittle failure mode plots for compressional and extensional tectonic regimes. Journal of Structural Geology, 20(5), 65-660.

Spencer, S (1991) The use of syntectonic fibres to determine strain estimates and deformation paths: an appraisal. Tectonophysics, 194, 13-34.

Taber, S (1916a) The growth of crystals under external pressure. American Journal of Science, 12, 532-566.

Taber, S (1916b) The origin of veins of the asbestiform minerals. Proceedings of the National Academy of Sciences, 2(12), 659-664.

Taber, S (1918) The origin of veinlets in the Silurian and Devonian strata of central New York. Journal of Geology, 26, 56-73.

Urai, JL, Williams, PF, van Roermund, HLM (1991) Kinematics of crystal growth in syntectonic fibrous veins. Journal of Structural Geology, 13(7), 823-836.

Walther, JV (1990) Fluid dynamics during progressive regional metamorphism. In: Studies in geophysics. The role of fluids in crustal processes. (ed) National-Research-Council, pp. 64-71. National Academy Press, Washington.

Wiltschko, DV, Morse, JW (2001) Crystallization pressure versus "crack seal" as the mechanism for banded veins. Geology, 29(1), 79-82.

Zehnder, K, Arnold, A (1989) Crystal growth in salt efflorence. Journal of crystal growth, 97, 513-521.

Chapter 2 – Microstructural evolution of syntaxial veins formed by advective flow

11

Chapter 2 - Microstructural evolution of syntaxial veins formed by advective flow

Christoph Hilgers*

Karin Dilg-Gruschinski

Janos L. Urai

Geologie–Endogene Dynamik, RWTH Aachen, D-52056, Germany

In: Geology 32(3), 261-264

ABSTRACT

Veins are common in the Earth’s crust, formed by a wide range of processes, which lead

to crystal growth in dilation sites. The first-order processes in vein formation have been

identified, but it is much less clear how these can be diagnosed from field studies. In

order to better understand the microstructural evolution during vein growth, we grew

veins of analogue material (alum, KAl(SO4)2·12H2O) in a transmitted-light cell from an

advecting supersaturated fluid. Real-time observation shows the effects of flow rate and

supersaturation on the evolving microstructure: (1) along-vein trends in growth rate

caused by decreasing supersaturation and (2) growth competition between clear crystals

in the absence of nucleation and primary fluid inclusions. Although the overall trends in

growth rate are in agreement with previous work, the local effects at the scale of

individual grains reported here are less well understood; these new data form a basis for

better interpretation of natural microstructures. To explore the possible effects of

experimentally observed processes during vein growth, we simulate the growth kinetics

of a quartz vein at various conditions of advective flow in the Earth’s crust. Results show

that in general the along-vein changes in growth rate occur at length scales much larger

than a typical outcrop.

Keywords: vein, see-through experiments, syntaxial, sealing


12

INTRODUCTION

The growth of crystals in dilation sites is of great importance in many fields of applied

and basic earth science, because of the associated large changes in transport properties

and rock strength (Blanpied et al., 1992; McCaig et al., 1995; Cox et al., 2001). A wide

range of microscale processes has been identified, which, usually coupled, may

contribute to the evolution of veins (Fisher and Brantley, 1992; Renard et al., 2000). The

resulting microstructures are correspondingly variable, ranging from blocky, elongate-

blocky to needle-like fibrous (Oliver and Bons, 2001). Isotopic signatures indicate the

nutrients for vein growth can be supplied by long-range advective transport, or by

diffusion in a local system (Kirschner et al., 1995; Elburg et al., 2002). Vein

microstructure is interpreted to depend on parameters such as the degree of

supersaturation, growth kinetics with different degrees of anisotropy, rates of local

deformation, rates of transport and relative rates of mineral growth vs. rates of vein wall

separation. Therefore it is not surprising that diagnosis of the process of vein formation

based on microscale structural and chemical information is inherently difficult. For

example, the crack-seal mechanism has been diagnosed by distinctive microstructures

(Ramsay, 1980), but this approach has subsequently been questioned (e.g., Wiltschko and

Morse, 2001).

Most experimental studies on fracture sealing are based on dissolution, because the

dissolution rate obeys the same fist-order equation as the precipitation rate (Lasaga, 1998,

p. 87; Ngwenya et al., 2000). This technique bypasses many experimental problems, but

does not produce veins. Lee and coworkers (Lee et al., 1996; Lee and Morse, 1999) did

pioneering experiments in which calcite was precipitated in a simulated fracture during

advective flow. They identified and modeled the first-order growth process and predicted

the kinetics of calcite vein formation in nature. However, the growth rates were too low

to produce grains large enough to study the details of microstructural evolution. For such

processes, analogue experiments are a useful alternative because they allow real-time

observation at reasonable time scales (Streit and Cox, 2000).

In this study we present results from a transparent reaction cell, designed to simulate vein

growth during advective transport. The model represents a simple isothermal system with

flow of supersaturated fluid through a long, parallel-walled fracture. Fractures are sealed


13

by epitaxial crystal growth on the fracture wall from a supersaturated solution. The

results are applied to natural conditions and discussed in terms of information on the

boundary conditions contained by the vein microstructure.

EXPERIMENTAL SETUP

The apparatus is placed under a microscope, and a solution of alum (KAl(SO4)2·12H2O)

is pumped at constant flow rate from the reservoir into the transparent reaction cell (Fig.

1). We used alum as an analogue material because of its well-known growth kinetics and

high solubility (Mullin, 1993, p. 222-226). Flow rates were varied from 0.05 to 0.5 mL/h,

which corresponds to an initial flow velocity of 1.3 × 10–5 to 10.6 × 10–5 m/s.

Experiments continued until the fracture was sealed and fluid flow stopped. The initial

pressure gradient in the cell was 1.15 Pa/m. Pressure at the upstream end rose to locally

much higher values of ~0.3 MPa shortly before the fracture was sealed and the

experiment was stopped.

The reaction cell contains spacers and the model fracture. The simulated fracture is ~15

mm long, 2–3 mm wide and 0.35 mm thick. It consists of fine-grained alum crystals that

mimic the wall rock (Fig. 1).

Figure 1. Schematic drawing of experimental setup. Supersaturated fluid is pumped through a fracture located in a transparent reaction cell.

Supersaturation of the fluid was induced by starting with a saturated solution in the

reservoir; the temperature dropped by a few degrees Celsius between the reservoir and

the reaction cell. This temperature reduction resulted in a supersaturation S (= c/ceq,

where c is the concentration in the supersaturated solution and ceq is the concentration in

an equilibrium solution) of ~1.1, which is too low to induce nucleation from the solution.

Temperature was controlled to less than ±0.05 °C.


14

RESULTS

In all experiments, epitaxial growth took place on the seed crystals along the entire wall

length (Fig. 2). The grains are clear and without primary fluid inclusions.

Growth competition from the wall toward the vein center caused marked coarsening of

the microstructure away from the vein margin. The final microstructure is a result of this

growth competition, where most of the initially growing seed crystals were overgrown by

a few grains (Fig. 3). The overall vein microstructure is elongate-blocky, similar to

microstructures in syntaxial veins.

Figure 2. Time series of microstructural evolution in four experiments. In all images, advection is from left to right. A: Epitaxial growth seals vein after 17 h. Initial aperture of vein a = 1.86 mm, constant flow rate Q = 0.085 mL/h, and supersaturation S = 1.11 (experiment 0330). B: Overall growth rates appear more homogeneous along vein. Here supersaturation is kept constant, and flow rate is increased by a factor of three. Size of grains at outlet almost equals grain size at inlet (a = 3.3 mm, Q = 0.258 mL/h, S = 1.11, experiment 0421). C: Experiment at high flow rate and low supersaturation. Almost no gradient in growth rates is visible along vein (a = 3.87 mm, Q = 0.515 mL/h, S = 1.09, experiment 0425). D: At low supersaturation and low flow rate, vein is sealed close to inlet (a = 3.13 mm, Q =0.052 mL/h, S = 1.08, experiment 0505). Scale bar in all images represent 2 mm.

Details of the growth-competition process are complex. Some grains grow much faster

than their neighbors, overgrowing them and finally cutting them off from further solvent


15

supply. Grains protruding into the channel at the initial stage do not necessarily survive

during grain growth.

Crystallization and a resulting slight swelling of the matrix occur in the wall rock, as

indicated by detailed observation of the high-resolution images. This swelling is

interpreted to be due to pressure of crystallization with a small part of the supersaturated

fluid flowing in the matrix and causing growth of matrix crystals.

In all cases the average growth rate is highest at the inlet. To a first order, crystal-growth

rates vary along the fracture as a function of flow rate and supersaturation. At a low flow

rate of 0.085 mL/h, the growth rate decreases significantly toward the downstream end

(Fig. 2A). The growth is more homogeneous along the vein if the flow rate is increased

by a factor of three to 0.26 mL/h, keeping the supersaturation constant at the inlet (Fig.

2B).

Figure 3. Graph showing percentage of growing crystals over time. In all experiments, the number of grains decreases non-linearly. Low flow rates (exp. 0330 and 0505) show less growth competition because of the delayed evolution of microstructure at downstream side in these experiments.


16

The measured growth rates can be compared with results of a numerical simulation (Fig.

4), by using a finite-difference scheme, implementing the flow-velocity–dependent

growth rate of alum (Mullin, 1993, p.222–226; Hilgers and Urai, 2002). As found in

earlier studies (Lee et al., 1996; Lee and Morse, 1999), there is a reasonable agreement

between the trends in the experimental data and in the simulations.

Figure 4. Growth rate of individual crystal faces, measured perpendicular to individual crystal faces and plotted along flow path. Numerical modeling (gray lines) is based on growth kinetics of slowest-growing low-index faces of alum. This approach can explain experimental data averages’ being slightly higher than modeled growth rates. In B, the line is located at upper range of average growth rates. In this experiment there was evidence for part of fluid flow bypassing fracture through matrix. Error in individual growth rate measurements is estimated as ±3.5 × 10–7

mm/s, caused by three-dimensional effects and uncertainties in locating boundary in digital images.


17

The reason for the differences is that in our experiments, some of the flow passes through

the matrix with a resulting lower flow velocity in the fracture and by small deviations

from the growth kinetics reported in the literature.

However, superposed on these trends is a strong scatter in growth rate at the scale of

individual crystals (Fig. 4). This scatter is interpreted to be due to the faces’ having

different crystallographic indexes and thus growth rates, combined with the complex

fluid flow around the individual crystals.

Considering the final microstructures of our experiments, we can ask the questions,

Which of these microstructures is diagnostic for the single-pass fracture sealing of our

experiments, or, How can we use our results to improve interpretation of natural vein

microstructures?

First, there is a pronounced growth competition, which results in a reduction of the

number of actively growing grains with time. The number of grains vs. time in our

experiments is shown in Figure 3. The details of these trends have not been studied in any

detail, but may become useful microstructural tools to constrain conditions of vein

growth after calibration with natural veins. All crystals grow into elongate-blocky

morphologies with grain boundaries that are usually planar and irrational with respect to

the surrounding crystals. Occasionally there is a sharp change in direction of a

propagating grain boundary, associated with the disappearance of a faster-growing face.

Upon sealing of the vein, crystal faces disappear once opposing grains touch (but their

presence in the past might be detected by CL microscopy).

An along-vein gradient in the amount of precipitated material in a given time implies

transport (advective or diffusive) along the fracture and allows conclusions about the

direction of transport. Information on the transport process can only be derived if the

sealed zone is longer than can be expected from short-range along-vein diffusional

processes, typically dominant up to distances of ~10 cm.

APPLICATION TO QUARTZ VEINS

One of the characteristic features of along-vein transport is decrease of growth rate along

the transport path. On the basis of this fact, one may ask if such features could be

observable in nature and thus could be used as a tool to investigate paleo–flow


18

conditions. To test this possible application, we use our numerical model to calculate the

growth rate of a quartz vein. Because hydrodynamic effects on the growth rate of quartz

are unknown, we assume that its growth kinetics are a function of supersaturation only,

described by a linear rate law (Rimstidt and Barnes, 1980). Let us consider a fracture with

an aperture of 1 cm and a length of 250 m in which silica precipitates from supersaturated

fluid flowing at constant flow rate and temperature. Following Rimstidt and Barnes

(1980), the mass precipitated in each segment of the finite-difference grid is dependent on

the reactive surface area and the supersaturation, the resulting growth rate being a

function of the rate constant and the transport mechanism. At 2 × 10–8 m/s, a lower limit

of flow velocities of typical of crustal fracture systems (Carson and Screaton, 1998;

Oliver, 2001), the initial pressure gradient required for flow along this 250-m-long

fracture is calculated by using the cubic law (Taylor, 1999) to be 10–4 Pa. Fluid pressure

will induce widening of the fracture if it becomes larger than the normal stress, for

example due to a decrease in aperture.

Ross (1994) measured the concentration of silica in pore waters in some hydrocarbon

fields and found a supersaturation S of ~1.5 at 160–175 °C and 60–80 MPa. Thus, we

consider a S = 1.1–10 as reasonable for supersaturated hydrothermal solutions and

calculate the growth rates of quartz in this fracture at 180 MPa fluid pressure and

temperatures between 250 and 350 °C.

Similar to our experimental results and the simulations of others (Lee et al., 1996; Dijk

and Berkowitz, 1998; Lee and Morse, 1999; Giles et al., 2000), the growth rate decreases

along the fractures. More material precipitates on the upstream side, and growth rates

decrease toward the downstream end. The minimum lengths of a vein required to

diagnose sealing from advecting supersaturated solution in an outcrop by a variation in

material precipitated in the same time period along-vein are ~943 and 196 m for

calculations at 250 °C and 350 °C, respectively.

DISCUSSION

The relatively low supersaturation in our experiments is reflected by the absence of

nucleation and primary fluid inclusions in the vein microstructure. Because the formation

of primary fluid inclusions depends on the crystal’s growth rate (Mullin, 1993, p.261), we


19

propose that similar microstructures in natural veins may be used to constrain

paleosupersaturations. The system yielding steady flow is a simple, experimentally

accessible end member of many different plumbing systems. Other transport regimes

involve a constant pressure gradient or a constant flow velocity.

The aim of the experiments was a two-dimensional model of the growth process with

visible microstructural evolution during fracture sealing. Vein growth stops when two

opposing crystals touch and the fracture is sealed. In nature the solution will flow around

locally sealed spots in three dimensions, allowing growth after the first contact between

crystals growing from both sides. Therefore, longer sealed vein sections can be produced

by advective flow, whereas in our experiments, the length of sealing is very limited.

Under these conditions, the vein microstructure will always become elongate-blocky,

because the final remaining voids will form at the terminations of grain boundaries

propagating from the vein wall into the center (e.g., in Figs. 2A and 2C). Syntaxial veins

formed by this process will neither be fibrous nor produce grain boundaries that track the

opening trajectory.

Advective mass transport is a major process in the crust. It is responsible for long-

distance mass transfer and precipitation in many ore deposits (Heinrich et al., 1996; Cox

et al., 2001). The large flow rates required for vein formation owing to the low solubility

of most vein components are generally accepted (Wood and Walther, 1986; Heinrich et

al., 1996). Consequently, naturally grown microstructures should contain information on

the direction of transport. However, in many veins the diagnosis of advective flow from

the microstructures grown is not straightforward. One problem is the change in wall rock

composition and wall rock grain size influencing the grown microstructure. Other factors

are processes by which veins are completely sealed, i.e., crack collapse (Etheridge et al.,

1984; Fisher and Brantley, 1992) or a slow compaction after the decay of fluid pressure.

Our calculations of quartz veins predict that in outcrops sufficiently long, an along-vein

trend of precipitated material in a given time could be used to diagnose advective flow in

a single fracture. However, the length scale at which this decrease of along-vein growth

rate is visible is too long to be observed in typical outcrops (Fig. 5). Thus, completely

sealed microfracture arrays with small apertures frequently found in sandstones could

have been formed by the advective-flow process considered in this paper.


20

Figure 5. Characteristic length of system as a function of temperature, defined as length of system over which growth rate is less than 70% of its original value. Graph outlines the along-vein distance required to observe a change of precipitated material in a given time in the field, at different fluid temperatures of same supersaturation.

The experiments presented have explored some of the basic characteristics of vein growth

from an advecting fluid and contribute to the deduction of growth conditions from the

microstructures observed. All veins were sealed at the inlet at variable conditions, the

downstream growth rate decreasing more rapidly at low flow rates. More work is

required for a full understanding. Expanding these experiments to include higher

supersaturations and several stages of opening and sealing will allow investigation of the

effect of these complexities on vein microstructure.

Acknowledgment We would like to thank Werner Kraus for his improvement of the reaction cells. Peter

Vrojlik and Paul Bons provided valuable comments on an earlier draft of the manuscript.

Constructive reviews of Ian Main, Donald Fisher and Stephen Cox are greatly

appreciated.


21

REFERENCES

Blanpied, M., Lockner, D.A., and Byerlee, D., 1992, An earthquake mechanism based on rapid sealing of faults: Nature, v.358, p.574–576.

Carson, B., and Screaton, E.J., 1998, Fluid flow in accretionary prisms: Evidence for focused, time-variable discharge: Reviews of Geophysics, v.36, p.329–351.

Cox, S.F., Knackstedt, M.A., and Braun, J., 2001, Principles of structural control on permeability and fluid flow in hydrothermal systems: Reviews in Economic Geology, v.14, p.1–24.

Dijk, P., and Berkowitz, B., 1998, Precipitation and dissolution of reactive solutes in fractures: Water Resources Research, v.34, p.457–470.

Elburg, M.A., Bons, P.D., Foden, J., and Passchier, C.W., 2002, The origin of fibrous veins: Constraints from geochemistry, in de Meer, S., et al., eds., Deformation mechanisms, rheology and tectonics: Current status and future perspectives: Geological Society [London] Special Publication 200 , p.103–118.

Etheridge, M.A., Wall, V.J., and Cox, S.F., Vernon, R.H., 1984, High fluid pressures during regional metamorphism and deformation: Implications for mass transport and deformation mechanisms: Journal of Geophysical Research, v.89, p.4344–4358.

Fisher, D.M., and Brantley, S.L., 1992, Models of quartz overgrowth and vein formation: Deformation and episodic fluid flow in an ancient subduction zone: Journal of Geophysical Research, v.97, p.20,043–20,061.

Giles, M.R., Indrelid, S.L., Beynon, G.V., and Amthor, J., 2000, The origin of large-scale quartz cementation: Evidence from large data sets and coupled heat-fluid mass transport modelling, in Worden, R.H., and Morad, S., eds., Quartz cementation in sandstone: International Association of Sedimentologists Special Publication 29, p.21–38.

Heinrich, C.A., Walshe, J.L., and Harrold, B.P., 1996, Chemical mass transfer modelling of ore-forming hydrothermal systems: Current practice and problems: Ore Geology Reviews, v.10, p.319–338.

Hilgers, C., and Urai, J.L., 2002, Experimental study of syntaxial vein growth during lateral fluid flow in transmitted light: First results: Journal of Structural Geology, v.24, p.1029–1043.

Kirschner, D.L., Sharp, Z.D., and Masson, H., 1995, Oxygen isotope thermometry of quartz-calcite veins: Unraveling the thermal-tectonic history of the subgreenschist facies Morcles nappe (Swiss Alps): Geological Society of America Bulletin, v.107, p.1145–1156.

Lasaga, A.C., 1998, Kinetic theory in the earth sciences: Princeton, New Jersey, Princeton University Press, 811p.

Lee, Y.-J., and Morse, J.W., 1999, Calcite precipitation in synthetic veins: Implications for the time and fluid volume necessary for vein filling: Chemical Geology, v.156, p.151–170.

Lee, Y.-J., Morse, J.W., and Wiltschko, D.V., 1996, An experimentally verified model for calcite precipitation in veins: Chemical Geology, v.130, p.203–215.

McCaig, A., Wayne, D.M., Marshall, J.D., Banks, D.A., and Henderson, I., 1995, Isotopic and fluid inclusion studies of fluid movement along the Gavarnie thrust,


22

central Pyrenees: Reaction fronts in carbonate mylonites: American Journal of Science, v.295, p.309–343.

Mullin, J.W., 1993, Crystallization: Oxford, Butterworth-Heinemann, 527p. Ngwenya, B.T., Elphick, S.C., Main, I.G., and Shimmield, G.B., 2000, Experimental

constraints on the diagenetic self-sealing capacity of faults in high porosity rocks: Earth and Planetary Science Letters, v.183, p.187–199.

Oliver, N.H.S., 2001, Linking of regional and local hydrothermal systems in the midcrust by shearing and faulting: Tectonophysics, v.335, p.147–161.

Oliver, N.H.S., and Bons, P.D., 2001, Mechanisms of fluid flow and fluid-rock interaction in fossil metamorphic hydrothermal systems inferred from vein–wall rock patterns, geometry and microstructure: Geofluids, v.1, p.137–162.

Ramsay, J.G., 1980, The crack-seal mechanism of rock deformation: Nature, v.284, p.135–139.

Renard, F., Gratier, J.P., and Jamtveit, B., 2000, Kinetics of crack-sealing, intergranular pressure solution, and compaction around active faults: Journal of Structural Geology, v.22, p.1395–1407.

Rimstidt, J.D., and Barnes, H.L., 1980, The kinetics of silica-water reactions: Geochimica et Cosmochimica Acta, v.44, p.1683–1699.

Ross, T.P., 1994, Pore fluid chemistry of a pressure seal zone, Moore-Sams-Morganza gas field, Tuscaloosa Trend, Louisiana, in Ortoleva, P., ed., Basin compartments and seals: American Association of Petroleum Geologists Bulletin, v.61, p.139–149.

Streit, J.E., and Cox, S.F., 2000, Asperity interactions during creep of simulated faults at hydrothermal conditions: Geology, v.28, p.231–234.

Taylor, W.L., 1999, Fluid flow in discrete joint sets: Field observations and numerical simulations: Journal of Geophysical Research, v.104, p.28,983–29,006.

Wiltschko, D.V., and Morse, J.W., 2001, Crystallization pressure versus “crack seal” as the mechanism for banded veins: Geology, v.29, p.79–82.

Wood, B.J., and Walther, J.V., 1986, Fluid flow during metamorphism and its implications for fluid-rock ratios, in Walther, J.V., and Wood, B.J., eds., Fluid-rock interactions during metamorphism: Berlin, Springer, p.90–108.

Chapter 3 – Fracture sealing of quartzite under a temperature gradient

23

Chapter 3 - Fracture sealing of quartzite under a temperature gradient:

Experimental results

Christoph Hilgers

Geologie-Endogene Dynamik, RWTH Aachen

Lochnerstr. 4-20, 52056 Aachen, Germany

Eric Tenthorey

Research School of Earth Sciences, Australian National University

Canberra ACT 0200, Australia

In: Terra Nova, 16(4), 173- 178

Abstract

Veined lithologies are formed by fracturing and sealing processes, with the veins

representing former fluid conduits through the rock. Although detailed fieldwork and

numerical simulations have provided a better understanding of vein growth, few studies

have attempted to seal fractures and generate veins experimentally. In this pilot study, we

subjected fractured quartzite to temperature gradients of 45-125˚C under hydrothermal

conditions in a static fluid, with the aim of precipitating secondary quartz in the cooler

portions of the fracture. Results show that secondary quartz precipitates due to the

imposed temperature gradient, causing the initial fracture to seal locally. Although no

systematic sealing pattern was observed along the fracture, samples subjected to higher

temperatures exhibit a smaller fracture width and appear to have reacted more

extensively. Electron microprobe mapping visualizes the spatial distribution of secondary

quartz, which contains elevated concentrations of aluminium.


24

Introduction

Veins are common features in the Earth’s crust, providing evidence that material

was dissolved, transported and finally precipitated from a supersaturated solution within

fractures. Understanding the mechanics and kinetics of vein formation is of paramount

importance when considering transport mechanisms and metasomatism in the crust,

which may result in concentration of economically valuable metals such as gold (Cox et

al., 1991).

The location and orientation of fractures and veins is controlled primarily by the

rock’s strength and the applied effective stresses. Another important factor related to vein

genesis is the timing of fracturing with respect to fracture sealing. Fractures may be

generated before vein formation, sometimes persisting to depths as large as 9 km (Möller

et al., 1997). Vein growth may also be syntectonic, in which case precipitation of vein

material and fracture opening are coeval. Syntectonic vein growth is often driven by pore

pressure pulses exceeding the minimum principal stress (Sibson et al., 1975; Etheridge et

al., 1984), thus causing fracture aperture to widen and allow further fluid injection. This

process may be associated with the seismic cycle with the resulting vein microstructures

often being described as crack-seal textures (Ramsay, 1980).

Previous experimental work addressed crack-healing processes (Smith and Evans,

1984; Brenan, 1991; Brantley, 1992), while very little has been done on crack or fracture-

sealing mechanisms. Healing is driven by the reduction in surface energy associated with

the crack, favoring a configuration containing spherical voids (fluid inclusions) within

the original crack (Smith and Evans, 1984). Conversely, fracture sealing is driven mainly

by fluid conditions (Gratier et al., 2003), involving the transport of dissolved material

over some distance followed by mineral precipitation within a crack or on a fracture

surface. Although there is likely to be a gradation between pure crack-healing and crack-

sealing processes (Brenan, 1991), understanding the two end-members is a key to

unraveling the hydrological and physical history of veined lithologies.

The mass transport mechanisms that are active during vein formation range from

long-range advection to grain-scale diffusion processes, depending on the hydraulic

conditions in the rock. While advection has been invoked in the formation of many ore

deposits (Heinrich et al., 1996), other studies suggest that gangue precipitate such as


25

quartz was derived from the surrounding rock, indicating that diffusion was the dominant

transport mechanism (Walderhaug, 1994; Fisher et al., 1995). In either case,

precipitation is driven by a gradient in equilibrium concentration, caused by a pressure or

temperature gradient, pore size effects, or a change in the chemical composition of the

surrounding rock or the fluid (Worden and Morad, 2000; Putnis and Mauthe, 2001).

In this study, we are interested in understanding the processes controlling quartz

precipitation within a previously formed fracture. The specific goals of this study are: (i)

to characterize vein morphology and secondary mineral growth in a static fluid

throughout the fracture; (ii) determine whether precipitation of silica prevents subsequent

solutes from diffusing through the crack; (iii) determine how the magnitude of the

temperature gradient affects the rate of silica precipitation and vein formation.

Experimental Method

Experiments were conducted using blocks of Proterozoic quartzite from Cape

Sorell in W. Tasmania, which were initially fractured using a uniaxial rig. Cores were

then taken through and parallel to the two matching halves and then ground flat at both

ends. The final split rock cylinders were approximately 15 mm long and 10 mm wide,

with a fracture width of about 10-20 µm. In each experiment, specimens were jacketed

with iron tubing and a layer of quartz grains was positioned at the high temperature end

of the specimen (Fig.1). To avoid quartz grains from being squeezed into the fracture,

quartz grains with a diameter of 125 µm portion were used. The top and bottom of the

specimen were capped with split alumina spacers, and loaded into a triaxial Paterson gas

apparatus.


26

Fig.1 Specimen configuration together with typical temperature distribution curves across the specimen. Water was injected during pressure built up, so that effective confining pressure in all experiments was maintained constant at 40 MPa. Rapid dissolution in the “hot” reservoir at the base of the experimental setup was followed by diffusion through the fracture and precipitation in the cooler regions of the specimen.

The Paterson apparatus at the Australian National University’s rock physics lab is

capable of independently controlling confining pressure (Pc) and pore pressure (Pp). For

a thorough description of the specimen assembly and experimental setup, see Zhang et al.

(1994). In all experiments, Pc and Pp were raised and lowered in tandem to about 200-

240 MPa and 160-200 MPa, respectively, ensuring that an effective pressure of 40 MPa

was never exceeded (Tab. 1). This minimized cataclasis of grains in the quartz reservoir

and limited extrusion of particles into the fracture zone. Temperatures at the hot end of

the specimen were varied from 485° to 900°C, with temperature differences across the

specimen ranging from 15° to 125°C (Tab. 1, Fig. 1). Experiments lasted for 3.5 to 10

hours, at which point temperature was reduced at a quench rate of 27-48˚C/min.


27

Table 1. Summary of the experimental conditions applied. Temperatures refer to conditions in the sample assemblage over a vertical distance of 3 cm, calibrated with a Pt/Rh 87/13 thermocouple in a dummy cell. All experiments were run at an effective confining pressure of 40 MPa. Dissolution (i.e. widening of the aperture) and precipitation phenomena are described in a separate column (CL-cathodoluminescence, SEM-scanning electron microscopy).

Run # Duration (h) T min (˚C) T (˚C) microprobe observations (method)

evidence for

fracture sealing

1 4.5 440 45 no widening of inlet zone, no grain indentation at reservoir-sample interface, no significant change in aperture along the fracture (microscopy)

none

2 7.6 573 91 widening of inlet away from reservoir with length of app. 2.5 mm along-fracture, maximum inlet aperture of 130 µm, no systematic deposition around bridges visiblet (microscopy), precipitates on intruded grains at inlet (CL, see Fig.2b), sealed regions without evidence of former fracture (SEM)

spaced

3 10 650 15 lumpy quartz growth at conchoidal steps observed at inlet and central part of sample (SEM, see Fig.6)

spaced

4 7 775 125 Yes widening of inlet away from reservoir with length of app. 1.2 mm along-fracture, maximum inlet aperture of 475 µm, reservoir grains attached to fracture surface (microscopy/SEM, see Fig. 6)

continuous

5 2.5 775 125 Yes widening of inlet away from reservoir with length of app. 350 µm along-fracture, maximum inlet aperture of 85 µm, cork-like reservoir grains in inlet (microscopy/SEM, see Fig. 3), overgrowth of quartz in inlet (CL), zonation of increased Al along-fracture in central part of sample (microprobe, see Fig.5)

spaced

6 4 600 91 Yes widening of inlet away from reservoir with length of app. 400 µm, maximum inlet aperture up of 120 µm, indentation of reservoir grains at reservoir-sample interface (microscopy/SEM, see Fig.2a)

spaced

7 3.5 600 92 lumpy quartz growth at steps on fracture surface (SEM)

spaced

Since the quartz overgrowths are expected to be very thin and of the same composition as

the substrate material, identification of vein quartz will be aided by using trace amounts

of Al, which is a natural contaminant in the hydrothermal system. This novel technique is


28

based on the premise that Al should be incorporated into the quartz structure and gives

the quartz a unique chemical signature when mapped using electron microprobe. Spatial

characterization of secondary minerals using such elemental tracers has already been

applied in more complicated systems (Tenthorey and Scholz, 2002). Microstructural

characterization of the run products was conducted using thin section analysis, scanning

electron microscopy, cathodoluminescence and electron microprobe.

Results

Optical microscopy reveals that the first 2-3 mm of the fracture apertures at the hot end of

the specimens increased from 10 µm up to several hundred µm in width during the

experiments (Tab.1). In most experiments, large grains were emplaced into the fracture

and conform perfectly to the fracture morphology, irrespective of the conditions applied

to the sample.

Following each experiment, the reservoir quartz grains were well compacted, especially

in the higher temperature experiments, where little porosity remained. The porosity loss

in experiments above 573°C was likely a result of pressure solution-type processes, as

indicated by abundant grain interpenetration. At 775°C, the reservoir grains penetrate into

the quartzite surface, providing further evidence for pressure solution (Fig.2a).

Using cathodoluminescence, one can distinguish between brighter reservoir grains and

the darker quartzite of the fracture walls (Figs. 2b, 3a). These grains conforming perfectly

with the fracture wall suggests that pressure-solution effects were active at an early stage

in the experiments, with material being dissolved at the highly stressed grain-fracture

contacts and precipitated at low stress sites (Renard et al., 1999). However, some grains

near the inlet have the same luminescence as the quartzite and appear to be coated by

brighter quartz, providing possible evidence for quartz precipitation within the fracture.

Such mineral coatings are relatively thick, although they are impossible to discern using

regular SEM backscatter imaging (Fig.3b). Widespread mineral precipitation in the inlet

zone is confined to the highest temperature experiments in which a large temperature

gradient was imposed.


29

Fig.2. a) Micrograph of experiment #6 showing the indentation of reservoir grains (upper part of image) into the sample surface near the inlet (indicated by arrows). b) Micrograph of experiment #2 showing cathodoluminesence of quartz in the reservoir (left hand side) and the fractured rock sample (right hand side). Based on the difference in cathodoluminesence between the two rock halves, it is possible to discern zones of mineral precipitation. Although one large reservoir grain has been extruded into the fracture, brighter quartz coatings (G) in the fracture zone represent SiO2 that may have been transported from the reservoir zone.

Fig.3. Micrographs from inlet zone of experiment#5 showing (a) cathodoluminesence and (b) SEM backscatter intensity. Although mineral precipitation from solution is evident by brighter regions in CL (G), such zones are impossible to distinguish using back-scatter imaging. These two images point to the difficulties involved in characterizing vein growth using classical SEM techniques.

All experiments contain zones where the fracture completely closed up, as shown in thin

sections and SEM (Fig.4). Grains are optically continuous across these zones, without

any indication of a former fracture such as linear arrays of fluid inclusions. Samples


30

subjected to higher temperatures exhibit smaller average apertures of about 1-2 µm.

However, clear optical evidence of fracture sealing is absent in polished sections, with the

notable absence of fluid inclusions in epitaxially overgrown zones or crystal facets

projecting into the fracture.

Fig. 4. SEM backscatter image showing a completely sealed zone in experiment #2. In some regions the fracture is completely sealed, leaving no evidence of its former location (the two dashed curves represent the approximate pathway of the original fracture). On the left hand side of the image, the fracture is still open, its 1-2µm width being significantly lower than the original 10-20 µm aperture.

Microprobe maps of samples from the high temperature experiments provide clear

evidence of mineral overgrowth within the fractures. These maps reveal a zone of

increased Al content in quartz, which is located at the edge of many small cracks and the

along the main fracture (Fig. 5). This high Al zone can be quite wide relative to the

fracture aperture, as in figure 5, where the Al-rich zone is 15 µm wide adjacent to a 2 µm

wide opening. Measured Al2O3 concentrations in the sealed zones are as high as 0.3 wt%

compared to a starting value of 0.01 wt% in the quartzite. Al is able to readily substitute

into the quartz structure due to its similar ionic radius to Si (Götze and Zimmerle, 2000).

The Al in solution may originate either from micas in the host rock, which partially

melted during the higher temperature experiments (775°C) or from the alumina spacers

that are used to assemble the specimen and which are in direct contact with the pore fluid.

Such zones of increased Al are not homogeneously distributed along the fracture, as the

Al content depends partly on the location of mica within the host rock and the

permeability within the fracture. Compared to the inlet zones, mineral precipitation is


31

sparser and there does not appear to be any systematic variation of mineral precipitation

along the length of the fracture.

Fig. 5. Microprobe element map showing variations of Si and Al in a sealed fracture from experiment #5. Lighter zones in Al map denote elevated concentrations of Al in quartz-as shown by arrows (up to 0.3 wt% Al2O3). Narrow zone of elevated Al and lower Si indicate a melt phase.

Clear morphological evidence of mineral precipitation is revealed in SEM images of the

reacted surfaces. The freshly fractured specimens exhibit concoidal fractures with smooth

surfaces. Conversely, following reaction the fracture surfaces become highly irregular

with various precipitation textures being evident (Fig.6). Some zones within the reacted

fractures are characterized by 2-10 µm lumps of quartz coating the surfaces. These lumpy

quartz growths appear to be preferentially located at concoidal steps and edges. Land and

Milliken (2000) and Jahren and Ramm (2000) described such features as mineral

precipitate on grain surfaces. However, similar island/channel structures have been


32

observed (Cox and Paterson, 1991; Olgaard and FitzGerald, 1993) and were interpreted

as evidence for pressure solution. In some zones, the fracture surfaces have become

highly angular due to the development of faceted quartz crystals up to 5 µm in size (Fig.

6b) or acicular quartz crystals (Fig 6c).

Discussion

Veins may form in a number of ways as discussed in the introduction, and mineral

precipitation due to diffusion down a temperature gradient is one mechanism by which

mineral precipitation may occur. We assume that convection is negligible in the

stationary fluid between the fracture walls due to the small aperture and rough fracture

surfaces with bridges connecting both sides. Additionally, the hot end is placed at the

bottom of the setup, which stabilizes fluid against convection and was successfully used

by Watson and Wark (1997). Although the experimental temperatures and temperature

gradients used are drastically higher than those of many natural systems, results from

such experimental work do provide insights into natural vein growth. For example, such

experiments may shed light on the preferred zones for mineral precipitation and the type

of vein microstructures formed, and also indicate whether mineral precipitation has a

negative feedback effect on the transport process.

It was clear that precipitation was most abundant in the highest temperature experiments

in which large temperature gradients were imposed. This observation was expected since

dissolution at the hot end, diffusion along fracture and precipitation are all kinetically

controlled and should be accelerated at high temperatures (Fournier and Potter, 1982;

Manning, 1994). Although the temperature gradient would mainly affect the driving force

for precipitation, it would also affect the spatial distribution of precipitated quartz.

Rimstidt and Barnes (1980) showed that there is an optimal temperature at which a

supersaturated solution will precipitate quartz. If temperature is slightly below the

equilibrium temperature then the reaction boundary is not overstepped sufficiently for

rapid precipitation to occur.


33

Fig.6. SEM micrographs showing various precipitation textures observed on rough surfaces of reacted fractures. (a) Lumpy quartz domains on a fracture surface may be a result of quartz precipitation or may be island/channel structures typical of pressure solution (experiment# 3). (b) Image of a portion of an open fracture. Quartz precipitation is indicatedby angular textures which suggests growth of angular quartz crystals (experiment #4). (c) Growth of an acicular quartz aggregate inside a narrow portion of the fracture (experiment#4).


34

Conversely if the temperature is too far below the equilibrium temperature, then the

precipitation kinetics become very slow and Si is “quenched into solution”. At

temperatures typical of the middle crust they find that precipitation rate would be greatest

at temperatures 25-50 °C below the saturation temperature. Extrapolating the

precipitation constant for quartz (Rimstidt and Barnes, 1980) up to high temperatures and

using the solubility data of Manning (1994), we can calculate the optimal temperature for

quartz precipitation in a saturated solution cooling from 900°C. Although there is a broad

range of temperatures where mineral precipitation is rapid, precipitation rate is greatest at

approximately 750°C, indicating that larger thermal gradients in our experiments would

have accelerated mineral precipitation.

The zones of quartz precipitation in our experiments are also likely to have been

influenced by the pressure-solution processes in the reservoir and inlet zone. The

operation of such a process is indicated in the reservoir by grain interpenetration and in

the fracture zone by textures that might be interpreted as island and channel structures.

Since pressure-solution was an important factor in the reservoir then it is likely that the Si

concentration was significantly elevated as compared to the unstressed case (Elias and

Hajash, 1992). In such a case, the kinetic driving force for precipitation would also be

elevated as soon as the pressure solution derived solute migrated to zones of low normal

stress. This effect may be the cause of the abundant precipitation observed in the inlet

zone, which was otherwise difficult to explain since the inlet zone was nearly at the same

temperature as the reservoir.

Even though diffusion through the inlet may have been inhibited by precipitation,

it is clear from the microprobe element maps that secondary mineral precipitation did

occur (Fig.5). Trace concentrations of Al were used to identify secondary minerals, which

allowed us to trace the former fracture, its width corresponding to the initial fracture

aperture. The thickness of the overgrowths in Figures 3 and 5 suggest that fracture

closure was a result of sealing rather than healing processes.

Conclusion

We report the first experimental results related to vein growth in a fractured quartzitic

rock. Although precipitation kinetics were slow compared to other studies that


35

characterized vein growth in analogue or low temperature minerals (Hilgers and Urai,

2002), significant precipitation did occur by imposing a temperature gradient along the

length of a fracture. Overgrowths of quartz have been reported by different authors

(Watson and Wark 1997; Nakamura and Watson 2001), but to our knowledge, the results

presented here are the first to document fracture sealing in the relatively sluggish quartz

system. In the future, longer term experiments must be designed so that mineral

precipitation along the fracture can be readily observed and quantified.

Acknowledgements

We would like to thank H. Kokkonen for assistance with specimen preparation and J.

FitzGerald and J.L. Urai for their helpful discussions. A. Wiechowski and R. Klinghardt

are thanked for assistance during microprobe mapping. Detailed comments by S.F. Cox

greatly improved the manuscript. Constructive and thorough reviews by J.-P. Gratier and

J. Renner are very much appreciated. CH would like to thank the rock physics group at

ANU for providing the facilities needed to conduct this study, and their hospitality. This

project was partly funded by a Heitfeld foundation travel grant imparted to CH.


36

References

Brantley, S.L., 1992. The effect of fluid chemistry on quartz microcrack lifetimes. Earthand Planetary Science Letters, 113, 145-156.

Brenan, J., 1991. Development and maintenance of metamorphic permeability: implications for fluid transport. In: Contact metamorphism (Kerrick, D.M., ed.). Reviews in Mineralogy, 26, 291-320.

Cox, S.F., and Paterson, M.S., 1991. Experimental dissolution-precipitation creep in quartz aggregates at high temperatures. Geophysical Research Letters, 18, 1401-1404.

Cox, S.F., Wall, V.J., Etheridge, M.A., et al., 1991. Deformational and metamorphic processes in the formation of mesothermal vein-hosted gold deposits - examples from the Lachlan Fold Belt in central Victoria, Australia. Ore Geology Reviews,6, 391-423.

Elias, B.P., and Hajash, A., 1992. Changes in quartz solubility and porosity due to effective stress: An experimental investigation of pressure solution. Geology, 20,451-454.

Etheridge, M.A., Wall, V.J., Cox, S.F., et al., 1984. High fluid pressures during regional metamorphism and deformation: Implications for mass transport and deformation mechanisms. Journal of Geophysical Research, 89, 4344-4358.

Fisher, D.M., Brantley, S.L., Everett, M., et al., 1995. Cyclic fluid flow through a regionally extensive fracture network within the Kodiak accretionary prism. Journal of Geophysical Research, 100, 12,881-12,894.

Fournier, R.O., and Potter, R.W., 1982. An equation correlating the solubility of quartz in water from 25° to 900°C at pressures up to 10,000 bars. Geochimica et Cosmochimica Acta, 46, 1969-1973.

Götze, J., and Zimmerle, W., 2000. Quartz and silica as guide to provenance in sediments and sedimentary rocks. E. Schweizerbart'sche Verlagsbuchhandlung, Stuttgart.

Gratier, J.P., Favreau, P., and Renard, F., 2003. Modeling fluid transfer along California faults when integrating pressure solution crack sealing and compaction processes. Journal of Geophysical Research-Solid Earth, 108, 10.1029/2001JB000380.

Heinrich, C.A., Walshe, J.L., and Harrold, B.P., 1996. Chemical mass transfer modelling of ore-forming hydrothermal systems: current practise and problems. OreGeology Reviews, 10, 319-338.

Hilgers, C., and Urai, J.L., 2002. Experimental study of syntaxial vein growth during lateral fluid flow in transmitted light: first results. Journal of Structural Geology,24, 1029-1043.

Jahren, J., and Ramm, M., 2000. The porosity-preserving effects of microcrystalline quartz coatings in arenitic sandstones: examples from the Norwegian shelf. In: Quartz cementation in sandstones (Worden, R.H., and Morad, S., eds.). SpecialPublication of the International Association of Sedimentologists, 29, 271-280.

Land, L.S., and Milliken, K.L., 2000. Regional loss of SiO2 and CaCO3, and gain of K2Oduring burial diagenesis of Gulf Coast mudrocks, USA. In: Quartz cementation in sandstones (Worden, R.H., and Morad, S., eds.). Special Publication of the International Association of Sedimentologists, 29, 183-197.


37

Manning, C.E., 1994. The solubility of quartz in H2O in the lower crust and upper mantle. Geochimica et Cosmochimica Acta, 58, 4831-4839.

Möller, P., Weise, S.M., Althaus, E., et al., 1997. Paleofluids and recent fluids in the upper continental crust; results from the German Continental Deep-Drilling Program (KTB). Journal of Geophysical Research, 102, 18,233-18,254.

Nakamura, M., and Watson, E.B., 2001. Experimental study of aqueous fluid infiltration into quartzite: implications for the kinetics of fluid redistribution and grain growth driven by interfacial energy reduction. Geofluids, 1, 73-89.

Olgaard, D.L., and FitzGerald, J.D., 1993. Evolution of pore microstructures during healing of grain boundaries in synthetic calcite rocks. Contributions to Mineralogy and Petrology, 115, 138-154.

Putnis, A., and Mauthe, G., 2001. The effect of pore size on cementation in porous rocks. Geofluids, 1, 37-41.

Ramsay, J.G., 1980. The crack-seal mechanism of rock deformation. Nature, 284, 135-139.

Renard, F., Gratie, J.-P., Ortoleva, P., et al., 1999. An integrated model for transitional pressure solution in sandstones. Tectonophysics, 312, 97-115.

Rimstidt, J.D., and Barnes, H.L., 1980. The kinetics of silica-water reactions. Geochimicaet Cosmochimica Acta, 44, 1683-1699.

Sibson, R.H., Moore, J.M., and Rankin, A.H., 1975. Seismic pumping - a hydrothermal fluid transport mechanism. Journal of the Geological Society, 131, 653-659.

Smith, D.L., and Evans, B., 1984. Diffusional crack healing in quartz. Journal of Geophysical Research, 89, 4125-4135.

Tenthorey, E.A., and Scholz, C.H., 2002. Mapping secondary mineral formation in porous media using heavy metal tracers. Journal of Geophysical Research, 107,10.1029/2000JB000109.

Walderhaug, O., 1994. Temperatures of quartz cementation in Jurassic sandstones from the Norwegian continental shelf - evidence from fluid inclusions. Journal of Sedimentary Research, A64, 311-323.

Watson, E.B., and Wark, D.A., 1997. Diffusion of dissolved SiO2 in H2O at 1GPa, with implications for mass transport in the crust and upper mantle. Contributions to Mineralogy and Petrology, 130, 66-80.

Worden, R.H., and Morad, S., 2000. Quartz cementation in oil field sandstones: a review of the key controversies. In: Quartz cementation in sandstones (Worden, R.H., and Morad, S., eds.). Special Publication of the International Association of Sedimentologists, 29, 1-20.

Zhang, S., Cox, S.F., and Paterson, M.S., 1994. The influence of room-temperature deformation on porosity and permeability in calcite aggregates. Journal of Geophysical Research, 99, 15761-15775.

Chapter 4 – On the arrangement of solid inclusions in fibrous veins

38

Chapter 4 - On the arrangement of solid inclusions in fibrous veins and the role of the crack-seal mechanism

Christoph Hilgers*, Janos L. Urai

Geologie - Endogene Dynamik, RWTH Aachen, Lochnerstr. 4-20, D-52056, Germany

In: Journal of Structural Geology (in press)

Abstract

The arrangement of solid inclusions in veins was examined in three different rock types

and tectonic settings. Common to all samples is a new type of sinusoidal inclusions,

associated with inclusion bands oriented parallel to the vein wall and inclusion trails at

high angle to the vein wall.

Crosscutting relationships of inclusion bands and the sinusoidal inclusions imply that

inclusion bands do not reflect individual crack-seal increments. In our veins, inclusion

bands are not a sufficient criterion to infer the crack-seal mechanism. Further evidence

for solid inclusions not formed by the classic crack-seal mechanism is given by inclusion

bands in gypsum crystals grown in soft clay.

During incorporation, solid inclusions can become folded or boudinaged, depending on

their orientation with respect to the opening direction. The complex displacement field

within the vein is recorded by curved crystal fibres, which track the opening direction.

Based on the arrangement of solid inclusions within the vein we suggest that our veins

grew continuously during syntectonic vein formation. Solid inclusions formed by a

fluctuation of adhesion at the vein wall interface rather than by crack-sealing, and vein

growth was at least partly driven by the force of crystallisation.

Keywords

fibrous veins, crack-seal, force of crystallisation


39

1. Introduction

The growth mechanisms of fibrous veins have been much discussed because of the

special mineral habit and their potential to contain information on kinematics of

deformation in rocks (Grigor'ev, 1965, p.200; Durney and Ramsay, 1973; Cox, 1987;

Durney and Ramsay, 1973; Urai et al., 1991; Means and Li, 2001). It has been shown that

fibrous crystals grow in a curved shape, and that the fibre curvature is not caused by

deformation (Durney and Ramsay, 1973; Spencer, 1991). These curved fibre grain

boundaries are often used to infer the opening trajectory of the fracture during growth. On

the other hand, many studies have shown that the curvature itself is not a sufficient tool to

reliably reconstruct the opening kinematics of veins (Cox, 1987; Urai et al., 1991; Hilgers

et al., 2001).

Important information about the vein opening process or opening direction is provided by

solid inclusions within veins. These inclusions may be particles detached from the vein

wall, or different phases grown contemporaneously during vein growth (Cox and

Etheridge, 1983; Ramsay, 1980; Hilgers and Urai, 2002).

Based on linear bands of solid inclusions parallel to the vein wall, Ramsay, 1980)

proposed a process of repeated fracturing and sealing - the so-called crack-seal

mechanism. These crack-seal veins display saw-tooth shaped grain boundaries between

adjacent elongated grains and overgrowth of favourably oriented crystals. The texture

does not show a constant aspect ratio of grains and is thus not fibrous sensu stricto, but

columnar. Fisher and Byrne (1990) and Fisher et al. (1995) correlated the length of mica

inclusions bands with crack seal increments and inferred that discontinuous inclusion

bands are caused by partial sealing of the vein before cracking. They noted that inclusion

bands are restricted to columnar grains that outgrowth their neighbour grains.

A second type of solid inclusions forms trails at high angle to the vein wall, which are

interpreted to originate from a point source in the wall (Dunne and Hancock, 1994). Such

inclusion trails are also proposed to be associated with the crack-seal process (e.g. Cox,

1987; Heaney and Fisher, 2003). Heaney and Fisher (2003) showed evidence for fracture

events in the vein based on inclusions hosted in columnar quartz crystals, which contain a

jagged / fractured end pointing towards the vein wall and the site of crack sealing.


40

Cox (1987) noted the importance of solid inclusion trails to establish the opening

direction of veins. The alignment of inclusion trails originating from a single source in

the wall points to the opening direction regardless of the fibre grain boundary orientation.

Thus, the arrangement of solid inclusions within veins is used as an important tool to

deduce the vein opening direction, as well as the opening mechanism.

The vein opening mechanism is generally described as mode I extension failure, initiated

by an increase in fluid pressure and low differential stresses (Etheridge et al., 1984; Price

and Cosgrove, 1990; Boullier and Robert, 1992; Sibson, 2003). Following classical

fracture mechanics, the vein continues to grow as long as the fracture propagates by

dynamic growth, i.e. if the critical fracture toughness KIC is exceeded (Ingraffa, 1987,

Wiltschko and Morse, 2001). Solid inclusions will be incorporated when a fracture

reopens after crack collapse and re-sealing, i.e. if the fluid pressure increases to values

larger than the minimum principal stress normal to the fracture wall and the tensile

strength of the rock. Another model favours subcritical crack growth rather than dynamic

growth. It is based on the observation that the topography of the vein wall interface

(interpreted to be rhythmically repeated as inclusion bands within the vein) is too high on

a micrometer scale to be opened by dynamic growth (Boullier and Robert, 1992; Cox,

1995; Andreani et al., subm.). A third model explains fibrous veins as continuous growth

phenomena, the vein being opened either actively by the force of crystallisation (Fletcher

and Merino, 2001; Means and Li, 2001; Wiltschko and Morse, 2001) or passively with

space provided during ongoing deformation (e.g. around pressure shadows). All three

mechanisms require the opening of a gap, and thus cover a continuum dependent on the

gap’s size and the timing of opening increments (Hilgers et al., 2001). Although the

incorporation of host rock fragments seems unlikely during continuous growth, repeated

inclusion bands plugged off from the wall and arranged parallel to the vein wall have

been observed in fibrous veins grown by the force of crystallisation (Means and Li,

2001).

In this paper we present the result of a study of solid inclusions found in syntectonic

fibrous veins from three different settings, and discuss the evolution of the microstructure

in terms of opening mechanisms. A new type of sinusoidal solid inclusion arrangement is

described, which allows interpretation of the vein opening mechanism.


41

2. Sample description

2.1 Calcite veins

The first set of samples comes from the Upper Silurian Wills Creek formation, which was

deformed during the Alleghian orogeny. The samples (taken from an outcrop on route

147 east of Milton, Pennsylvania) consist of thin-bedded, folded shale and siltstone with

pressure solution cleavage, containing many bedding parallel fibrous calcite veins

(Pringle, 1980; Gray pers.com.). In our samples calcite is absent in the host rock.

Siltstone layers consist of quartz with grain sizes of about 20 m and white mica and are

arranged as continuous and lenticular beds in shale.

The bedding parallel veins have a lensoid shape (Fig.1), terminating between bedding

planes (Fig.1, 2). They contain abundant inclusions of host rock fragments of shale and

siltstone, and some chlorites grown from the wall into the vein. Host rock fragments are

layers, which can be traced laterally into zones without veins. These are locally folded

inside the vein (Fig.2b). At the other extreme, these layers are arranged in regularly

spaced series of inclusions, indicating stretching sub-parallel to bedding (Fig.2a). Other

series of inclusions are sinusoidal with amplitudes corresponding up to the vein’s

thickness (Fig. 2a).

Fig.1. Syntectonic calcite veins emplaced parallel to bedding and displaced fragments of host rock. Different arrangements of solid inclusions can be observed within the vein, which point to a complicated opening process. Details of two areas are highlighted and are described in figure 2.


42

The solid inclusion arrays themselves are sometimes connected by fibrous overgrowth

between the fragments. The orientation of these fibre grain boundaries is not parallel to

the fibres immediately outside the inclusions (insert in Fig.3f).

Fig.2. a) Host rock fragments (grey) show different types of arrangements within the calcite vein. Linearly oriented host rock fragments suggest bedding parallel stretching. Other fragments are arranged in a sinusoidal manner (outlined by dashed line). b) In other areas, solid inclusion can be found as typical inclusion trails, originating from a point source in the host rock. The host rock is folded and fractured, with pieces partly sliced off from the host rock (wedge-like). Scale bar refers to both images.

The calcite fibres are commonly curved and without undulose extinction, and contain

minor mechanical twins (Fig.3). Fibre curvature and the type of host rock vs. vein

composition are consistent with antitaxial growth at the vein-wall rock interface (Durney

& Ramsay 1973; Ramsay & Huber 1983). The fibres are commonly but not always

oblique to the wall rock, suggesting oblique opening during late stages of vein growth.


43

Next page

Fig.3.

a) Host rock fragment (black) have been displaced along fibrous calcite grain boundaries. White arrows show the fibre grain boundaries, which are assumed to track the opening direction of the vein. Note that the arrowhead points to the direction of solid inclusion movement, opposite to vein younging. The inclusion progressively wedges off from the host rock (top of image), and becomes fragmented within the vein.

b) A small thin layer (marked as 1) has been curves into the vein in a wedge-like manner. Note that the black grain in the top is not bedding but a calcite fibre (cc).

c) Small layers of shale become successively displaced from their country rock. Close to the host rock bedding is curved, while it forms solid inclusions displaced along fibre grain boundaries within the vein. The white arrows trace the fibre grain boundary and point to the direction of solid inclusion movement during vein growth, opposite to the fibre growth direction.

d) Within the calcite vein, three lines of solid inclusion bands are arranged at high angle to the wall rock, commonly misinterpreted as inclusion trails (see text for discussion). Next to the host rock (at top of image) a thin solid inclusion band (marked as i.b.) is arranged parallel to the vein wall interface and corresponds with the morphology of the host rock.

e) Solid inclusion band arranged at high angle to bedding (right hand side). The inclusions are homogeneously displaced parallel to fibre grain boundaries (GB).

f) Inclusion bands heterogeneously displaced from the host rock (lower part of image). The inclusion band is slightly buckled on the left hand side and fragmented / boudinaged on the right hand side. This matched with the local displacement field within the vein, indicated by the orientation of the fibre grain boundaries.

g) Thin section showing the bending of bedding (lower part of image) into the vein. Solid host rock inclusions become fragmented and are aligned highly oblique to the vein wall interface. In the upper left of the image, their orientation changes and is aligned in opposite direction.


44


45

2.2 Tiger's eye (quartz veins)

The second set of samples contain Tiger's eye hosted in banded ironstone from a mine in

the Hamersley Range in Western Australia (Powell et al., 1999; Taylor et al., 2001; Webb

et al., 2003). Layers show approximately the same thickness as in our first set of samples,

and are deformed into close to open folds (Fig.4). The tiger's eye forms bedding-parallel

veins between the iron rich layers.

The Tiger's eye consists of fibrous, bedding-parallel quartz veins and inclusions of iron

oxide/hydroxides (see also Heaney and Fisher, 2003). The fibres are oriented at high

angle to the vein wall. The vein's thickness often increases within the fold hinge,

suggesting syntectonic vein formation (Fig.4).

Fig. 4. Two rocks samples of folded, banded ironstone containing bedding-parallel quartz veins (tiger's eye, bright colours). The zones A, B are described in detail in figure 5.


46

Both the composition of vein and host rock as well as fibre curvature points to antitaxial

growth (Ramsay and Huber, 1983 p.241; Heaney and Fisher, 2003).

Solid inclusions of host rock material of coherent layers made of iron hydroxide are

included in the vein. The vein and solid inclusions are surrounded by a selvage about 20

µm wide, consisting of iron oxide. Host rock fragments are much larger than the fibre

diameter (Fig.6) and are arranged as inclusion bands and sinusoidal inclusions, the

amplitudes of which are up to the vein’s thickness (Fig.5b). The wall rock fragments can

often be made to fit together following the fibre boundaries. In some cases, the wall rock

fragments form spectacular triangular structures such as shown in figure 6.

Fig.5. a) Details of zone A in figure 4. Three different types of inclusions appear within a single vein. Solid inclusions are arranged in a sinusoidal manner, and as inclusion trails and inclusion bands. b) The microstructure of zone B includes a zone of faulting trending N-S (left side of image), causing an off-set of the veins. Veins are arranged as beefs parallel to bedding, containing sinusoidal inclusions with amplitudes close to the vein’s thickness. Note that the asymmetry of sinusoidal inclusions corresponds to the mesoscale folds (compare with figure 4).


47

Fig.6. a) Detail of solid inclusion arrangements within a single beef. Four layers of bedding have been displaced along the fibre grain boundaries, forming sinusoidal solid inclusion fragments. Note that small inclusions in the lower part of the image form Z-, S- and M-parasitic folds (dashed line, image taken from Wiersma, 2002). b) Reconstruction of solid host rock inclusions by movement along fibre grain boundaries. The original position of host rock fragments and their shape is colored in black, the different stages during movement are shown in grey (white inclusions have not been restored). Note that the movement along fibre grain boundaries results in perfect alignment as straight beds of solid host rock inclusions.

2.3 Gypsum veins

Gypsum veins were sampled in deformed Eocene carbonates and shales on the coastline

in near Ras al Hamra, Muscat, Oman. Up to several centimetres thick gypsum veins are

embedded in shales, while few millimetre-sized gypsum veins were found in adjacent

carbonates layers. The frequency of veins increases towards a faulted anticline nearby.

Veins are oriented sub-parallel to bedding and are located in rotated tension gashes (Fig.

7a) (Wilson, 1961 p. 453; Price and Cosgrove, 1990 p. 250).

The overall vein microstructure is similar to the ones described above, i.e. gypsum is

fibrous and fibres continue from one side of the wall to the other (Fig.7c,d). Within these

antitaxial veins, solid host rock inclusions are arranged as inclusion bands parallel to the


48

vein wall interface. Other inclusions are aligned in straight lines highly oblique to the

vein wall, or show a sinusoidal shape (Fig.7c,d). Inclusion bands continue across

sinusoidal inclusions, without any change in inclusion band spacing (Fig.7d).

Fig.7. a) Outcrop showing a shale layer between two limestone beds (S0 – bedding). Gypsum (gyp) tension gashes become sub-parallel to S0 at the shale-limestone interface and indicate top to the right shearing (hammer for scale). b) The vein contains numerous inclusion bands aligned parallel to the vein wall interface. On the right hand side and in the central part the vein shows minor inclusions arranged oblique to the vein wall interface.

3. Discussion

The three sets of samples come from different deformed sedimentary environments but

show very similar structures of the solid inclusions in fibrous calcite, quartz and gypsum

veins, respectively. Following the arguments given in Heaney and Fisher (2003) that

tiger's eye is a result of vein growth, and not a pseudomorphic substitution of quartz after

pre-existing crocidolite asbestos suggests that all settings described here expose primary

vein microstructures.

Bedding-parallel veins are located on both fold limbs and around the fold hinge in our

quartz and calcite veins. Their thickness has the tendency to increase in the fold hinge

(saddle reefs), pointing to vein growth during folding (Fig 4). In our gypsum veins,

saddle reefs have not been observed in the outcrop and in our samples, but the association

of veins with faults and folds nearby suggests syntectonic emplacement. These veins

emplaced in tension gashes oriented at various angles to bedding (Fig. 7 a,b).


49

Fig.7. c) Solid inclusions in gypsum veins are arranged as bands parallel to the vein wall, and highly oblique to the wall. Echelon veins wedge into the host rock causing inclusions oriented at high angle to the vein wall interface. Widely spaced solid inclusion bands may be correlated with Means and Li, 2001) type I pattern. A separate sketch outlines the overall vein microstructure. d) Sinusoidal inclusion hosted in shale with cross cutting inclusion bands parallel to the vein wall interface. Note that the inclusion bands are not displaced on top of the sinusoidal inclusion. The inclusion band spacing does not change significantly across the sinusoidal inclusion. The limb on the left hand side of the sinusoidal inclusion is boudinaged, with a fibre connecting the fragments. This fibre grain boundary connects the displaced fragments, and is slightly oblique tot the general fibre orientation. On the right hand side, sinusoidal inclusions form a continuous line without gaps. Some inclusion bands and the overall vein microstructure are highlighted in a separate sketch.


50

3.1 Vein opening mechanism

Two models have been proposed for the formation of bedding parallel veins:

(i) Fractures may have opened at depth during horizontal compression when fluid

pressure overcomes the minimum principal stress and the tensile strength across the

bedding plane (Cosgrove, 1993, Jessell et al., 1994; Cosgrove, 1995; Mandl, 2000 p.194-

205). Under such conditions, crystals will grow until fluid pressure drops due to

fracturing and the void collapses (crack-seal mechanism with incremental growth). This

may result in a feedback system, where the discharge seals off the fluid pathways and

fluid pressure repeatedly increases until the rock fails (Sibson, 1992; Ortoleva, 1994

chapter 16).

Collapse structures have been described as dark insoluble residues tracing euhedral

terminations being aligned sub-parallel to the vein wall interface (Fisher and Byrne,

1990). They may be either fragments of the wall or residue of minerals not found

precipitated in the vein. Such microstructures have not been observed in our veins.

(ii) Alternatively, veins may grow continuously, opened actively by the force of

crystallization (Bunsen, 1847; Taber, 1916; Correns, 1949; Means and Li, 2001;

Wiltschko and Morse, 2001) or passively during ongoing deformation. This mechanism

implies that the vein-wall interface does not heal but is a zone of low adhesion and

continuous accretion of new material. La Iglesia et al. (1997) measured crystallization

pressures in aqueous sulphate solutions of higher than 40 MPa in fluid saturated

dolostone at room temperature. Winkler (1973) calculated pressures for anhydrite as 40

and 230 MPa, and for halite as 66 and 380 MPa at 50˚C and supersaturations 2 and 50

(supersaturation is expressed as concentration divided by the equilibrium concentration),

respectively. High supersaturation is not uncommon in rocks. Silica supersaturation

reaching values of three have been reported in reservoir brines in Cretaceous sandstones

(Ross, 1994). Putnis et al. (1995) describe much higher supersaturation above 10 in finely

porous media without onset of crystallisation. They present an alternative model for

precipitates such as veins based on transport of fluid from very small fractures or grain

boundaries into an open fracture. The force of crystallisation has not yet been

experimentally measured for common vein minerals like quartz and calcite under

hydrothermal conditions, but calculations suggest pressures of 80 MPa (calcite) and 300


51

MPa (quartz) for low supersaturations of 2 (Wiltschko and Morse, 2001). Dewers and

Ortoleva (1990) noted that such calculations based on equilibrium thermodynamics are

valid under equilibrium conditions only, and thus represent an upper bound. Their non-

equilibrium model calculates pressures being about a factor of three smaller.

Estimating the vertical stress, we can determine whether the force of crystallisation may

have caused vein opening in our samples. Tiger’s eye grew at a depth of about 5 km at

about 180°C (Smith et al., 1982; Oliver, pers. com., based on thickness of younger

stratigraphic units and mineral assemblages). Calcite veins are assumed to have formed

with an overburden of about 6.75 km at 150-300°C (Gray pers. com., based on thickness

of younger stratigraphic units and conodont colour alteration index of 3.5-4),

corresponding to a maximum overburden stress of 175 MPa. Gypsum fibres hosted in

Eocene carbonates where overlain by less than 1 km overburden, corresponding to app.

25 MPa. The force of crystallisation stresses described above are higher than the stresses

required to open the vein, even though our conservative calculations do not take into

account elevated fluid pressure. If one assumes that the fluid pressure in the porous host

rock is close to lithostatic, the stresses required to open a vein by the force of

crystallisation will be close to zero. Thus, the force of crystallization may explain vein

opening in all three settings.

The fundamental difference of the fracture vs. force of crystallisation model is the vein

orientation. While regular vein patterns are formed by far field stresses, the orientation of

pure force of crystallisation veins only depends on local stresses governed by the strength

of the surrounding matrix (e.g. low strength across bedding planes) and the location of

other growing veins. Our gypsum veins emplaced in tension gashes which requires

tectonic stresses as kick off process for vein formation (Wiltschko and Morse, 1998). Our

calcite and quartz veins formed bedding parallel, so no firm conclusion can be drawn on

the opening mechanism from their orientation. However, their syntectonic nature

suggests the same interpretation. Better insight on the opening mechanism may be

derived from the arrangement of solid host rock inclusions.


52

3.2 Inclusion arrangement

The presence of inclusion bands and inclusion trails aligned parallel to the vein wall

interface has been frequently used to infer the crack-seal mechanism (Ramsay, 1980; Cox

and Etheridge, 1983; Cox, 1987; Fisher and Byrne, 1990; Fisher and Brantley, 1992).

Others argued for a continuous diffusional process for fibrous vein microstructures

(Durney, 1972; Durney and Ramsay, 1973; Bons and Jessell, 1997; Wiltschko and

Morse, 2001).

The detachment of solid inclusions is caused by changes in the adhesion between the host

rock, vein and inclusion. Adhesion may be changed by processes such as a variation in

fluid geochemistry resulting in a different wetting angle of the fluid film between vein

and wall rock, vein growth rate keeping pace with the opening rate of the vein, fluid

pressure causing the dilation of pore space which provides better access of growing

grains into the wall rock, or fracturing. Fracturing of an interface may trigger the

detachment of wall rock fragments, but is one of several processes possible.

In our samples, all three types of inclusion arrangements (inclusion bands, inclusion

trails, sinusoidal inclusions) were observed within a single antitaxial vein, in samples

from all three settings. We thus conclude that all different types of solid inclusions were

formed simultaneously by the detachment of fragments from the wall rock. These

inclusions can be incorporated in veins by the crack-seal mechanism and continuous

growth such as the force of crystallisation mechanism, as long as there are changes in

adhesion between the wall and the growing crystals.

In many cases, regular banding is a result of oscillating growth phenomena, as has been

shown for banded agates, Liesegang rings, banding in MVT ores, stylolites, and solution

growth (Allègre et al., 1981; Ortoleva et al., 1987; Reeder et al., 1990; Wang and Merino,

1990; Ortoleva, 1994; L'Heureux and Jamtveit, 2002). Microstructural criteria for

authigenic minerals grown by the force of crystallization were described by Maliva and

Siever (1988). They noted that µm-sized host rock inclusions within the grown

microstructure point to force of crystallisation growth. Li (2000) and Means and Li

(2001) grew fibrous veins using highly soluble material and also observed a banding

parallel to the vein wall during ongoing vein growth. They distinguished two different

types of wall-parallel inclusions; type I opaque layers caused by fluctuation in the growth


53

conditions, and type II discontinuities with much smaller spacing of some 10 µm apart.

These fibrous microstructures grow at a cohesive interface during continuous dilatant

growth (Li, 2000), the inclusion bands probably caused by an oscillation process.

Figure 8 is a natural example of non-crack-seal solid inclusion bands hosted in a large

single crystal of gypsum, which has grown in soft clay at less than 1 m depth in a salt

lake in Utah, USA. Because the crystallographic orientation of this gypsum crystal does

not change across the inclusion band, it must have grown into the porous wall rock before

inclusion bands were dragged off. A change in boundary conditions then caused further

growth at the new vein wall interface rather than growth into the wall, which formed

zones without inclusion bands. In summary, the occurrence of inclusion bands is not

sufficient to demonstrate crack-seal. Additional criteria to prove crack-seal are required

e.g. a columnar rather than fibrous texture and serrated grain boundaries.

In all samples solid inclusions form a wedge with the country rock curved into the vein

(esp. Fig. 3a-c, 7c,d). Depending on the local displacement field, these inclusions may

become folded, sheared or extended into a series of micro-boudins (Fig.3c,e, 7d). With

extreme stretching, these series of micro-boudins may become oriented at very high angle

to the vein wall. This raises an interesting question in nomenclature because fragmented

inclusion bands highly oblique to the vein-wall interface can easily misinterpreted as

inclusion trails (esp. Fig.3d).

These wedge-like solid inclusions aligned oblique to the vein wall may be interpreted as

being incorporated during bedding-parallel, distributed, echelon crack growth with bent

bridges of wall rock between overlapping fractures (Nicholson and Pollard, 1985;

Nicholson, 1991; Hilgers and Urai, 2002). However, the small size of the wedges, their

strong curvature into the vein and high topography as well as their repeated occurrence

within close distance (e.g. Fig. 3 a, c) are unlikely to be formed by dynamic fracturing

(see also Boullier and Robert, 1992). We suggest that these irregular morphologies are

more likely caused by continuous growth e.g. variations in force of crystallisation.

In our samples the reconstruction of sinusoidal solid inclusions along fibre grain

boundaries results in a perfect alignment at the vein wall (Fig.3f,6,7c). Thus, host rock

fragments have been displaced at various distances, starting from an inclusion band (e.g.

Fig.3f). Such displacement is either a result of (i) internal deformation of the vein which


54

contains an originally planar solid inclusion band by heterogeneous simple shear, (ii) a

variation of vein growth rate along-vein, (iii) a change of the location of vein growth and

total constant growth rate (defined as the summed growth rate along one vein-wall

interface at different locations), or (iv) a combination of these processes.

Fig.8. Solid clay inclusions in a large gypsum single crystal from a salt lake in Utah, USA. The crystal is undeformed and has grown app. 1 m below the surface without being exposed to tectonic stresses. Clay beds are arranged as opaque inclusion bands.

Because fibre-parallel displacements cannot be traced into the host rock, we conclude that

solid sinusoidal inclusions have not been formed by heterogeneous simple shear and

reject (i). If simple shear and growth take place simultaneously, we would still expect a


55

curvature of the vein-wall interface similar to the shape of the sinusoidal inclusion (Fig.

9a). A similar microstructure may result from a change in vein growth rate below a

sinusoidal inclusion (Fig. 9b). In both cases, inclusion bands will be offset within the

vein. This leaves (iii) as the most likely kinematics.

Crosscutting relationships of inclusion bands and sinusoidal inclusions are best displayed

in the gypsum veins (Fig. 7d). Parallel bands of solid inclusions continue above and

below sinusoidal inclusions, without a change in band spacing. Inclusion bands

crosscutting the sinusoidal inclusion are located at higher elevation outside the arc, but

continue as straight line above the arc (Fig. 7d).

This can be explained by a model in which the total growth rate along-vein is continuous,

but distributed at different growing interfaces (Fig.9c). This model explains the continuity

of solid inclusion bands above sinusoidal inclusions, and the constant inclusion band

spacing within the vein.

Crack seal opening increments are generally interpreted to represent the distance between

inclusion bands, this distance sealed by growth of vein material. This mechanism would

cause fragmentation on both limbs of sinusoidal inclusions, its spacing correlating with

the distance between inclusion bands. Such regular fragmentation has not been observed

in our veins. Therefore, solid inclusion bands cannot be used as crack-seal indicators in

our veins. Rather, a continuous growth process is required to explain the arrangement of

solid inclusions.

However, a displacement of inclusions driven actively by a continuous growth process

only is unlikely, because one would not expect a regular arrangement of sinusoidal

inclusions corresponding to the higher order fold structure, as displayed in our quartz and

calcite samples (esp. the Z-, S- and M-parasitic fold geometries in the tiger’s eye)

(Fig.5b,6). This awaits full explanation but suggests that the incorporation of solid

inclusions during syntectonic vein emplacement is not a pure growth phenomenon.

As outlined above, vein opening scenarios (fracture vs. force of crystallisation opening)

have been much discussed because of the problematic interpretation of distinctive

microstructures. The fundamental difference of both processes is the stress field in which

the vein has been formed.


56

Next page

Fig.9. Sketch illustrating the incorporation of solid inclusions using three different models. In all models, veins grow antitaxial (at the vein-wall interface, towards the host rock). Sites of growth are marked with a dark line.

a) Heterogeneous simple shear will cause slip parallel to the fibre grain boundaries, causing displacement at the vein wall interface. This interface will be bended or sheared with the same amplitude as the sinusoidal inclusion. Inclusion bands within the vein will be offset during slip. In this model, the vein emplaced before displacement takes place, e.g. by heterogeneous simple shear folding. The vein might continue to grow during shear, displayed by a dashed line.

b) Assuming variable growth rates below sinusoidal solid inclusions, the resulting microstructure is similar to the model described in a), i.e. solid inclusion bands above wall 2 will be displaced.

c) If the growth rate remains constant along-vein, inclusion bands and vein-wall interface will not be displaced. Note that there is no growth once the inclusion is surrounded by vein material. Slip only occurs parallel to the fibre axis along the wall-inclusion (marked as wall2) contact. Slip is absent between adjacent fibres as long as the growth rate along-vein is constant. This model best explains the formation of solid inclusion as observed in our rock samples.


57


58

Stress fields are variable on a local scale during force of crystallisation growth, and are

much more regular over large scales for fracture opening. Consequently, force of

crystallisation growth could be deduced from irregular vein arrays. However, as force of

crystallisation and fracturing can (and are expected to) act simultaneously, a regular vein

pattern does not indicate the absence of force of crystallisation during vein formation.

The following opening scenario is suggested for our veins (Fig. 9c,d): Initially, small

lensoid veins formed (i) parallel to bedding, probably initiated as bedding parallel tensile

fractures (quartz and calcite veins) and (ii) as oblique tension gashes (gypsum veins),

respectively. Antitaxial vein growth caused the wall rock being wedged off, causing

folding or boudinage depending on the orientation of solid inclusions relative to the local

displacement field. Inclusion bands, inclusion trails and sinusoidal inclusions are formed

by constant total growth rates at variable accretion sites, the displacement field being

illustrated by the orientation of fibre grain boundaries.


59

Fig. 9. d) Overview of the vein’s evolution and formation of sinusoidal inclusions using the constant growth rate model shown in c.

4. Conclusions

Our samples from three different lithologies and tectonic settings display similar

sinusoidal solid inclusion arrangements, originally formed as inclusion bands. Cross

cutting relationships of sinusoidal inclusions and inclusion bands require a growth model

with constant total growth rate along-vein (defined as constant growth rate at different

locations on one side of the vein). Inclusion bands themselves are not a sufficient


60

criterion to deduce on the crack-seal mechanism, and do not reflect crack seal increments

in our veins. Although clear evidence for crack-seal vs. force of crystallization

mechanisms are difficult to establish, microstructures suggest that a continuous growth

mechanism contributed significantly to vein formation in our samples.

The local displacement field of growing veins can alter the position of solid inclusion

bands caused by vein growth and slip parallel to fibre grain boundary. Due to a

heterogeneous displacement of solid inclusions, fragments of inclusion bands can be

easily misinterpreted as inclusion trails.

Acknowledgments We would like to acknowledge Mary Beth Gray, Bucknell University, Lewisburg, PA,

USA, and Nick Oliver, University of Townsville, Australia, for their constructive

comments on the local geology of the Valley and Ridge province, Pennsylvania, and the

Hamersley province, Australia. We thank Pascal Richard for introducing us to the

outcrop at Ras Al Hamra, Muscat, Oman. Peter Kukla, RWTH, kindly provided samples

containing tiger’s eye. Jochen Kolb, RWTH, helped to characterize the solid host rock

inclusions in tiger’s eye. Thanks are due to Dirk Wiersma who contributed one of his

high-quality large format photographs. Constructive reviews by Donald Fisher and David

Wiltschko led to considerable clarification of the manuscript.


61

References

Allègre, C. J., Provost, A. & Jaupart, C., 1981. Oscillatory zoning: a pathological case of crystal growth. Nature 294, 223-228.

Andreani, M., Baronnet, A., Boullier, A.-M. & Gratier, J.-P., subm. A microstructural study of a "crack-seal" type serpentine vein using SEM and TEM techniques. Journal of Structural Geology.

Bons, P. D. & Jessell, M. W., 1997. Experimental simulation of the formation of fibrous veins by localised dissolution-precipitation creep. Mineralogical Magazine 61, 53-63.

Boullier, A.-M. & Robert, F., 1992. Paleoseismic events recorded in Archean gold-quartz vein networks, Val d'Or, Abitibi, Canada. Journal of Structural Geology 14(2), 161-179.

Bunsen, R., 1847. Ueber den innern Zusammenhang der pseudovulkanischen Erscheinungen Islands. Annalen der Chemie und Pharmacie Band LXIII(1. Heft), 1-59.

Correns, C. W., 1949. Growth and dissolution of crystals under linear pressure, Discussions of the Faraday Society, No.5, Gurney and Jackson, London, 267-271.

Cosgrove, J. W., 1993. The interplay between fluids, folds and thrusts during the deformation of a sedimentary succession. Journal of Structural Geology 15(3-5), 491-500.

Cosgrove, J. W., 1995. The interplay between fluids, folds and thrusts during the deformation of a sedimentary succession: Reply. Journal of Structural Geology 17(10), 1479-1480.

Cox, S. F., 1987. Antitaxial crack-seal vein microstuctures and their relationship to displacement paths. Journal of Structural Geology 9(7), 779-787.

Cox, S. F., 1995. Faulting processes at high fluid pressures: An example of fault valve behavior from the Wattle Gully fault, Victoria, Australia. Journal of Geophysical Research 100(B7), 12,841-12,859.

Cox, S. F. & Etheridge, M. A., 1983. Crack-seal fibre growth mechanisms and their significance in the development of oriented layer silicate microstructures. Tectonophysics 92, 147-170.

Dewers, T. & Ortoleva, P., 1990. Force of crystallisation during the growth of siliceous concretions. Geology 18, 204-207.

Dunne, W. M. & Hancock, P. L., 1994. Paleostress analysis of small-scale brittle structures. In: (edited by Hancock , P. L.), Continental Deformation, Pergamon Press, Oxford, 101-120.

Durney, D. W., 1972. Solution-transfer, an important geological deformation mechanism. Nature 235, 315-317.

Durney, D. W. & Ramsay, J. G., 1973. Incremental strains measured by syntectonic crystal growth. In: (edited by de Jong, K. A. & Scholten, R.), Gravity and tectonics, Wiley, New York, 67-96.

Etheridge, M. A., Wall, V. J., Cox, S. F. & Vernon, R. H., 1984. High fluid pressures during regional metamorphism and deformation: Implications for mass transport and deformation mechanisms. Journal of Geophysical Research 89(B6), 4344-4358.


62

Fisher, D. M. & Brantley, S. L., 1992. Models of quartz overgrowth and vein formation: Deformation and episodic fluid flow in an ancient subduction zone. Journal of Geophysical Research 97(B13), 20,043-20,061.

Fisher, D. M., Brantley, S. L., Everett, M. & Dzvonik, J., 1995. Cyclic fluid flow through a regionally extensive fracture network within the Kodiak accretionary prism. Journal of Geophysical Research 100(B7), 12,881-12,894.

Fisher, D. M. & Byrne, T., 1990. The character and distribution of mineralized fractures in their Kodiak formation, Alaska: Implications for fluid flow in an underthrust sequence. Journal of Geophysical Research 95(B6), 9069-9080.

Fletcher, R. C. & Merino, E., 2001. Mineral growth in rocks: Kinetic - rheological models of replacement, vein formation, and syntectonic crystallization. Geochimica et Cosmochimica Acta 65(21), 3733-3748.

Grigor'ev, D. P. 1965. Ontogeny of minerals. Israel Program for Scientific Translation Ltd., Jerusalem.

Heaney, P. J. & Fisher, D. M., 2003. New interpretation of the origin of the tiger's eye. Geology 31(4), 323-326.

Hilgers, C., Koehn, D., Bons, P. D. & Urai, J. L., 2001. Development of crystal morphology during unitaxial growth in a progressively widening vein: II. Numerical simulations of the evolution of antitaxial fibrous veins. Journal of Structural Geology 23, 873-885.

Hilgers, C. & Urai, J. L., 2002. Microstructural observations on natural syntectonic fibrous veins: Implications for the growth process. Tectonophysics 352, 257-274.

Ingraffa, A. R., 1987. Theory of crack initiation and propagation in rock. In: (edited by Atkinson, B. K.), Fracture mechanics of rocks, Academic Press, London, 71-110.

Jessell, M. W., Willman, C. E. & Gray, D. R., 1994. Bedding parallel veins and their relationship to folding. Journal of Structural Geology 16(6), 753-767.

La Iglesia, A., Gonzales, V., Lopez-Acevedo, V. & Viedma, C., 1997. Salt crystallization in porous construction materials I Estimation of crystallization pressure. Journal of Crystal Growth 177, 111-118.

L'Heureux, I. & Jamtveit, B., 2002. A model of ocillatory zoning in solid solutions grown from aqueous solutions: Applications to the (Ba,Sr)SO4 system. Geochimica et Cosmochimica Acta 66(3), 417-429.

Li, T., 2000. Experimental growth of fibers and fibrous veins. Unpublished PhD thesis, State University of New York at Albany.

Maliva, R. G. & Siever, R., 1988. Diagenetic replacement controlled by force of crystallization. Geology 16, 688-691.

Mandl, G. 2000. Faulting in brittle rocks. Springer, London. Means, W. D. & Li, T., 2001. A laboratory simulation of fibrous veins: some first

observations. Journal of Structural Geology 23, 857-863. Nicholson, R., 1991. Vein morphology, host rock deformation and the origin of the

fabrics of echelon mineral veins. Journal of Structural Geology 13(6), 635-641. Nicholson, R. & Pollard, D. D., 1985. Dilation and linkage of echelon cracks. Journal of

Structural Geology 7(5), 583-590. Ortoleva, P., Merino, E., Moore, C. & Chadam, J., 1987. Geochemical self-organization

I: Reaction-transport feedbacks and modeling approach. American Journal of Science 287(10), 979-1007.


63

Ortoleva, P. J. 1994. Geochemical self-organisation. Oxford University Press. Powell, C. M., Oliver, N. H. S., Li, Z.-X., Martin, D. M. & Ronaszeki, J., 1999.

Synorogenic hydrothermal origin for giant Hamersley iron oxide ore bodies. Geology 27(2), 175-178.

Price, N. J. & Cosgrove, J. W. 1990. Analysis of geological structures. Cambridge University Press, Cambridge.

Pringle, L. R., 1980. Strain mechanisms in the upper Silurian Wills Creek formation, central Pennsylvania. Unpublished BSc thesis, Bucknell University.

Putnis, A., Prieto, M. & Fernandez-Diaz, L., 1995. Fluid supersaturation and crystallization in porous media. Geological Magazine 132(1), 1-13.

Ramsay, J. G., 1980. The crack-seal mechanism of rock deformation. Nature 284(5752), 135-139.

Ramsay, J. G. & Huber. 1983. Techniques in modern structural geology. Volume 1: Strain analysis. Academic Press, London.

Reeder, R. J., Fagioli, R. O. & MEyers, W. J., 1990. Oscillatory zoning of Mn in solution-grown calcite crystals. Earth Science Reviews 29, 39-46.

Ross, T. P., 1994. Pore fluid chemistry of a pressure seal zone, Moore-Sams-Morganza gas field, Tuscaloosa Trend, Louisiana. In: (edited by Ortoleva, P.), Basin Compartments and Seals, 61, AAPG, Tulsa, Oklahoma, 139-149.

Sibson, R. H., 1992. Implications of fault-valve behaviour for rupture nucleation and recurrence. Tectonophysics 211, 283-293.

Sibson, R. H., 2003. Brittle-failure controls on maximum sustainable overpressure in different tectonic regimes. AAPG Bulletin 87(6), 901-908.

Smith, R. E., Perdrix, J. L. & Parks, T. C., 1982. Burial metamorphism in the Hamersley basin, Western Australia. Journal of Petrology 23(1), 75-102.

Spencer, S., 1991. The use of syntectonic fibres to determine strain estimates and deformation paths: an appraisal. Tectonophysics 194, 13-34.

Taber, S., 1916. The growth of crystals under external pressure. American Journal of Science 12, 532-566.

Taylor, D., Dalstra, H. J., Harding, A. E., Broadbent, G. C. & Barley, M. E., 2001. Genesis of high-grade hematite orebodies of the Hamersley province, Western Australia. Economic Geology 96, 837-873.

Urai, J. L., Williams, P. F. & van Roermund, H. L. M., 1991. Kinematics of crystal growth in syntectonic fibrous veins. Journal of Structural Geology 13(7), 823-836.

Wang, Y. & Merino, E., 1990. Self-organizational origin of agates: Banding, fiber twisting, composition, and dynamic model. Geochimica et Cosmochimica Acta 54, 1627-1638.

Webb, A. D., Dickens, G. R. & Oliver, N. H. S., 2003. From banded iron-formation to iron ore: geochemical and mineralogical constraints from across the Hamersley Province, Western Australia. Chemical Geology 197, 215-251.

Wiersma, D. J. 2002. Exposures of mineral and rock. Inmerc bv, Wormer, Netherlands. Wilson, G., 1961. The tectonic significance of small scale structures, and their

importance to the geologist in the field. Annales de la Société Géologique de Belgique LXXXIV, 423-548.

Wiltschko, D. V. & Morse, J. W., 1998. Force of crystallization origin for veins; crystallization pressure versus fluid pressure as the driving force for vein opening,


64

Geological Society of America 30. Geological Society of America (GSA) : Boulder, CO, United States, Toronto, ON, Canada, Oct. 26-29, 1998, 197.

Wiltschko, D. V. & Morse, J. W., 2001. Crystallization pressure versus "crack seal" as the mechanism for banded veins. Geology 29(1), 79-82.

Winkler, E. M. 1973. Stone: properties, durability in man's environment. Springer, Wien.

Chapter 5 – Textural and isotopic evidence on the fluid source and transport mechanism

65

Chapter 5 - Textural and isotopic evidence on the fluid source and transport mechanism of antitaxial fibrous microstructures from the Alps and the Appalachians

Christoph Hilgers1, Sven Sindern2, 3

1Geologie – Endogene Dynamik, Lochnerstr. 4-20, RWTH Aachen, 52056 Aachen,

Germany 2Institut für Mineralogie und Lagerstättenlehre, Wüllnerstr. 2, RWTH Aachen, 52062

Aachen, Germany 3Zentrallabor für Geochronologie, Institut für Mineralogie, WWU Münster, Corrensstr.

24, 48149 Münster, Germany

in: Geofluids (in press)

Abstract

The source of fluid forming veins is of great importance in order to understand the

hydraulic system acting in the earth’s crust. Syntectonic antitaxial veins are one of the

few tools in which the opening history can be deduced from rocks, and thus are of

primary importance in determining the rock kinematics. Antitaxial veins were taken from

black shales in two different tectonic settings in the Helvetic Alps, Switzerland, and the

Taconic Appalachians, New York State. These syntectonic extension veins are regularly

spaced and are oriented sub-normal to bedding. The vein microstructure displays a

symmetry around the median line in the centre of the vein, and a symmetry in

cathodoluminescene banding parallel to the vein-wall interface, which suggests transport

along bedding-parallel dissolution planes from both vein-walls. Antitaxial veins

nucleated in transgranular fractures, but evidence for ongoing multiple crack seal

increments is lacking. Rather, veins grew continuously keeping close contact to the vein-

wall.

Radiogenic 87Sr/86Sr ratios are higher in the surrounding matrix than in the vein, and

higher than the corresponding seawater data in all samples. Variations are small and

calcite in both the vein and the host rock were derived from the same source of fluid in

the Helvetic samples. Mass balance of Sr suggests that the amount of calcite is too small


66

in the surrounding host rock to be derived locally. Stable oxygen compositions are

heavier in the host rock than in the veins, with overall low variation in both 18O and 13C values in the Mesozoic Helvetic samples. Data point to a rock buffered system, the

precipitate most likely derived from an external source.

The lower Palaeozoic Appalachian veins are lighter in 18O values than the host rock,

similar to the Helvetic veins. Radiogenic 87Sr/86Sr data and a large heterogeneity in stable

isotope values indicate an open system.

Microstructural and isotopic evidence suggests that our antitaxial veins formed by

pervasive fluid flow, the solute at least partly derived from an external source.

Introduction

Veins are former fluid conduits produced either by long-range advection of large fluid

volumes or short-range diffusional processes (e.g. Yardley 1986; Friedman 2002; Oliver

1996; Fisher 1996; Cox et al. 2001; Oliver & Bons 2001; Sibson 2004). They form as a

result of brittle failure together with void formation, followed by fluid emplacement and

precipitation. Such processes can recur and form individual veins by multiple crack-seal

events (Ramsay 1980). Crystals can also actively push apart the wall rock by the force of

crystallisation due to growth at the vein-wall interface, or grow passively within

dilational jogs (e.g. Durney & Ramsay 1973, Fletcher & Merino 2001; Means & Li 2001;

Wiltschko & Morse 2001; Hilgers & Urai in press; Bons & Montenari in press).

Different vein microstructures develop during deformation, a result of crystal growth

kinetics and vein opening rates (Mügge 1928; Urai et al. 1991; Hilgers et al. 2001).

Fibrous vein microstructures are of special interest, because they provide information on

the progressive deformation in rocks (Grigor'ev 1965; Durney & Ramsay 1973). They

consist of needle-like grains similar in shape, which are frequently used to track the

opening direction and deduce the kinematics of the geological settings (Ramsay & Huber

1983 p.241).

Three sets of syntectonic vein microstructures have been described in the literature.

Fibrous veins growing towards the vein-wall interface are called antitaxial (Durney &

Ramsay 1973). Their growth direction is indicated by their fibre width increasing towards

the vein-wall interface as well as the positioning of numerous small crystals along the


67

centre of the vein forming a median line. The composition at the vein-wall interface is

consistently different from that of the vein, additionally fibres do not continue into the

wall rock. The second type of syntectonic vein microstructure is referred to as stretched

crystals, which are columnar grains with serrated grain boundaries. In contrast to

antitaxial veins, they connect grains in the wall and consist of the same material as that of

the wall rock. Stretched crystals are assumed to have formed by repeated fracturing and

growth at alternating sites in the vein (Cox & Etheridge 1983; Passchier & Trouw 1996

p.133-136), although some stretched crystals have been shown to grow at the vein-wall

interface (Hilgers & Urai 2002). The third type of syntectonic vein microstructure is

called syntaxial, epitaxial overgrowths of the wall rock. These grains grow from the wall

into the vein, their shape not truly fibrous but rather elongate-blocky (Oliver & Bons

2001).

Considering the transport mechanism driving vein growth, experimental and numerical

studies have shown that discrete fluid flow by lateral advection in fractures commonly

results in the inlet being clogged (Lee et al. 1996; Lee & Morse 1999; Hilgers et al.

2004). However, many antitaxial veins are isolated in three dimensions and completely

sealed. This implies either diffusion or pervasive fluid flow through the rock matrix.

Generally, diffusion does not account for the mass precipitated in a geological time scale

for minerals like quartz or calcite (Walther & Orville 1982). This is dependent on the

small length scale of diffusion (Bons 2000), and the unreasonably high supersaturation

needed (Lee et al. 1996).

In order to infer the source of solute and the transport mechanism, radiogenic and stable

isotope data are used to investigate fluid transport mechanisms (e.g. Oliver et al. 1993;

McCaig et al. 1995; Badertscher et al. 2002a; Badertscher et al. 2002b). However, few

publications focus on the small-scale and describe the vein microstructures when

presenting isotope data, although the vein microstructure provides additional information

on the sealing history and may indicate the type of transport mechanism (e.g. Dietrich et

al. 1983; Rye & Bradbury 1988; Burkhard & Kerrich 1988; Kirschner et al. 1993;

Kirschner et al. 1995b; Henry et al. 1996; Badertscher et al. 2002a; Badertscher et al.

2002b; Richards et al. 2002). A first radiogenic 87Sr/86Sr study of antitaxial fibrous


68

calcite veins hosted in shale was presented by Elburg et al. (2002) from the Flinders

Ranges, Australia.

In this study, we present stable and radiogenic isotope data of monomineralic, antitaxial,

shale hosted calcite veins from three different tectonic settings. We will compare results

with other studies and outline, whether a conclusion of the transport mechanism can be

drawn solely from the vein microstructure.

Methods

We applied both stable ( 18O, 13C) and radiogenic (87Sr/86Sr) isotope techniques to our

calcite veins and host rocks. The advantage of the Rb-Sr system is the absence of

fractionation effects after vein emplacement i.e. radiogenic isotope values do not vary

with varying temperature or pressure (e.g. Faure 1986). Radiogenic 87Sr/86Sr-ratios are

suitable to distinguish the source of the material in the calcite veins, because Sr behaves

chemically very similar to Ca, fitting in the calcite lattice (e.g. McCaig et al. 1995). In

contrast, stable isotopes are extremely sensitive to fluid-rock interaction making it the

more favourable technique to investigate the transport mechanism which forms the veins.

Samples were taken from calcite veins and the surrounding host rock. A cylinder of about

1.5 mm diameter was drilled for radiogenic analyses on a rotating thick section

(Medenbach type micro-driller). In some cases, additional host rock samples were taken

at varying distances from the vein to identify lateral variations (Fig.1).

Stable isotope data ( 18O, 13C) were measured from dissolved calcite veins and host

rock calcite, using concentrated H3PO4 (50°C) according to Swart et al. (1991) (at the

Ruhr-Universität Bochum, Institut für Geologie, Mineralogie, und Geophysik). The

cryogenically cleaned CO2 fraction was then analysed for its oxygen and carbon isotopic

data with a Finnigan MAT delta S mass spectrometer. A fractionation coefficient of =

1.00925 was applied for calculation of 18O in calcite from the analysis of CO2 gas.

Reproducibility of replicate measurements was better than 0.1 per mil (2 for oxygen

and better than 0.05 per mil (2 for carbon.

Radiogenic isotope data were measured on whole rock samples and bulk powdered

samples of vein fillings. The samples were spiked with a 87Rb-84Sr mixed spike and

completely dissolved in HF-HNO3. Calcite was then leached from the powdered host


69

rock samples at room temperature with 0.1 N HCl for five minutes and consequently

centrifuged for 15 minutes in order to separate the solution and residual fraction. The

residual fraction was then treated like a normal whole rock sample. The leached dissolved

calcite was spiked, dried and re-dissolved for element separation.

Fig. 1. Sketch outlines the locations of the drilled samples for 87Sr/86Sr analyses. Note that the veins are restricted to the shale layers in the Morcles and Appalachian samples, and do not continue into the marl beds. In the Glarus samples, the veins are preferentially located in the siltstone beds, but continue into the shale beds.

Rubidium and strontium were separated by standard cation-exchange techniques using

DOWEX AG-50 W*8 resin (200-400 mesh) and 2.5 and 6 M HCl as eluants. Strontium

was measured on a VG Sector 54 multi-collector Thermal Ionization Mass Spectrometer

(TIMS) in dynamic mode and loaded with Ta-fluoride on W-filaments (Zentrallabor für

Geochronologie, Münster, Germany). The mass fractionation corrections for strontium

were based on 86Sr/88Sr = 0.1194. The 87Sr/86Sr ratio of the NBS 987 strontium standard

(n = 29) yielded 0.710278 ± 0.00003 (error is 2 for all analyses). Rubidium analyses

were carried out on a NBS design single collector solid source TIMS, loaded as chloride

on Ta-filaments. Total blanks are Rb < 50 pg and Sr < 100 pg. Age dependent 87Sr/86Sr

ratios were recalculated using the 87Rb decay constant of 1.42 x 10-11 a-1 (Neumann &

Huster 1974).

Sample locations and vein microstructures

The first sample location is situated in the Tertiary flysch of the Helvetic Morcles nappe,

at the cliffs along the SE ridge of Tour Sallière, 8 km northwest of Martigny, Switzerland


70

(Ramsay et al. 1982). The rocks are about 30-35 Ma in age (Oligocene, Rupelian) with

peak metamorphism and nappe emplacement at 20-30 Ma (Kirschner et al. 1995b;

Burkhard & Sommaruga 1998; Burkhard & Badertscher 2001).

The second set was taken in the Tertiary flysch in the footwall of the Helvetic Glarus

thrust, near Elm, Switzerland. Rocks are underlain by Mesozoic limestones, and were

exposed to peak metamorphism at about 25-30 Ma, while post-peak metamorphic

thrusting has been dated at 20-25 Ma (Hunziker et al. 1986).

The third sample location is situated in an abandoned quarry of the Cambrian Hatch Hill

Formation in the Appalachians, New York State, USA (Passchier & Urai 1988; Urai et al.

1991; Hilgers & Urai 2002). This outcrop is located in the Taconic allochthon, with

nappe emplacement / tectonic activity between 440-480 Ma (Bradley 1989; Hayman &

Kidd 2002; Kidd pers.com.).

All the sampled exposures contain interlayers of black shale beds, which have been

undergone lower greenschist facies metamorphism. Lenticular, antitaxial calcite veins are

restricted to black shale layers in the Morcles and Appalachian samples, with only a few

veins continuing into the adjacent marl beds (Fig.2a). They are regularly spaced and

aligned parallel to each other, subnormal to bedding. Regularly spaced extension veins

indicate that their formation was initiated by fracturing (e.g. Price & Cosgrove 1990, p.

54-59).

The Glarus samples are composed of interlayers of black shale and quartz rich siltstone

beds. Here, some calcite veins are restricted to the siltstone layers, while larger veins

cross-cut the shale and siltstone beds without a significant change in aperture (Fig.1).

The microstructures of the calcite veins are antitaxial, but may become blocky or

stretched when entering the marl beds (Fig.2c,d). The curvature of the fibrous grains are

often symmetrical around the vein centre (Fig.2a,g). Antitaxial calcite veins in the shale

are surrounded by a small quartz segregation band ~30 µm wide. Such bands grow

contemporaneously with and opposite to the calcite vein (Kirschner et al. 1995b; Hilgers

& Urai 2002). Small chlorite flakes were described to grow synchronous with the quartz

selvage (Hilgers & Urai 2002), and have also been identified in the shale using x-ray

diffraction. When the veins cross-cut the marl beds, the quartz selvage disappears. Here,

the vein-wall does not represent a continuous interface and is disrupted by cleavage


71

planes (Fig.2d). Quartz grains show evidence for transgranular fracturing (Fig.2d,f). The

truncation of quartz and calcite grains in the host rock and the overall absence of

intracrystalline deformation indicate that pressure solution was the dominant deformation

mechanism. The width of the calcite fibres within the individual veins corresponds with

the spacing of the cleavage, suggesting that both structures formed contemporaneously.

Intersections of quartz pressure fringes around pyrite and antitaxial calcite veins imply

the synchronous growth of quartz fringes and calcite veins (Fig.2e). Their co-existence

suggests that both quartz pressure fringes and antitaxial calcite veins formed under the

same opening and growth mechanisms, and that the quartz and calcite precipitate

originated from the same fluid. Extension in the marl beds is characterized by delocalised

overgrowths as fringes and transgranular fracture fills (Fig.2f).

The concentration of veins in the shale and in the siltstone, respectively, may be

explained by the rheology of the layers during vein formation. Whereas shales are the

favourable sites for vein formation in the marl-shale interlayers of the Appalachian and

Morcles samples, veins are more concentrated in siltstone rather than shales in the Glarus

samples. This is in accordance with published data describing an increase in rock

ductility from quartz rich layers and shales to calcite rich layers in rocks, at lower

greenschist facies (Paterson 1978, p.162-168; Bestmann et al. 2000).

Within calcite rich layers both the stylolites and the cleavage are oriented sub-parallel to

bedding, highlighted by the new grown white mica (Fig.2f). The orientation of calcite

veins sub-normal to stylolites/cleavage suggests the synchronous formation of both

structures. However, evidence for ongoing crack-seal mechanism is lacking, as repeated

inclusion bands arranged parallel to the vein-wall interface are absent in the vein

microstructure (see also Hilgers & Urai 2002; Bons & Montenari in press).

Serial sectioning of veins has shown that some veins are isolated in three dimensions (as

are pressure shadows), requiring fluid transport through the matrix. Cathodoluminescence

shows symmetrical banding within the calcite veins (Fig.2g,h) caused by different

amounts of Mg and Mn in the calcite lattice (see also figure 1 in Urai et al. 1991; figure 7

in Hilgers & Urai 2002). This points to a change in the fluid chemistry during vein

growth.


72

Next page

Fig.2.

a) Antitaxial calcite vein is restricted to the shale layer, and bounded by marl beds. Bedding planes (S0) are highlighted by a white line. Sample from the Appalachians, New York State.

b) Antitaxial calcite vein restricted to a siltstone layer. Sample from the Glarus nappe.

c) Antitaxial fibrous vein changes its microstructure and becomes blocky when entering the marl bed. Bedding is sub-horizontal in this image. Appalachian sample.

d) Continuation of an antitaxial vein into a marl bed where the microstructure changes to stretched crystals, the fibre width approximating the spacing of cleavage planes. Vein grain boundaries are highlighted as bold lines, quartz fragments are outlined by dashed lines. Morcles sample.

e) Antiaxial calcite fibres are surrounded by a thin quartz selvage showing a smooth interface between selvage and wall, and a rough interface at the vein-selvage interface. Fibre grain boundaries are often located at the peaks of the quartz selvage, which corresponds to the cleavage planes (S1) in the host rock. A pyrite grain (py) is located next to the vein and shows fibrous quartz fringes (marked by a dashed line) indicating bedding parallel stretching. Note that the quartz fibres are much longer on the right hand side, and extend into the antitaxial calcite vein. Appalachian sample.

f) Extension in the marl beds is characterised by fracturing of quartz grains. Calcite pressure fringes are aligned parallel to cleavage around quartz grains, and overgrowths between fractured quartz grains. The overgrowths coalesce and form small veins, as shown by the arrow (lower right). Dissolution in the marl beds is represented as stylolites, their planes overgrown by white mica. Morcles sample.

g+h) Micrograph of a vein crossing bedding (vertical) in crossed polarisers and cathodoluminescence (CL). The CL-image shows symmetrical bands around the centre of the vein, indicative for different growth stages at different Mg- and Mn- concentrations within the fluid. Note that the roughness of the band is similar to that of the vein-selvage interface and does not outline individual crystal facets. Appalachian sample.


73


74

Stable and radiogenic isotope data

All samples are characterised by small radiogenic and stable isotopic heterogeneities

between shale, siltstone/marl and vein. Stable and radiogenic isotope data are listed in

Table 1 and 2.

The 87Sr/86Sr ratio increases with time due to the decay of 87Rb, the application of these

data as a tracer for element transfer therefore requires a recalculation of the 87Sr/86Sr ratio

to its initial ratio, at the time of vein formation (Tab.2).

Host rock data from the Helvetic samples (Morcles and Glarus) are in good agreement

with other studies on the Tertiary flysch (Badertscher et al. 2002a) as well as Cretaceous

and Jurassic limestones (Burkhard & Kerrich 1988; Crespo-Blanc et al. 1995; Kirschner

et al. 1995a). Calcite in the host rocks of the Glarus nappe is heavier in 18O values

(+22.08, +22.14 per mil) than in the Morcles nappe (+16.94 to +19.95 per mil) (Fig. 3).

Calcite from the veins is depleted in 18O (Glarus: +19.56 to +19.96 per mil, Morcles:

+19.11, +19.19 per mil) relative to the host rock. Carbon values in host rock deviate to

lower and higher values relative to the veins. As pointed out by Hunziker et al. (1986)

and Burkhard and Kerrich (1988), the clay in the shale could drive the pore fluid to lower 18O values. This also explains the outlier measured in the Morcles nappe (Tab.1, +16.94

per mil) measured in black shale. The 18O value of the vein in the Morcles sample is

located halfway between the host rock data, therefore probably derived from the calcite in

the siltstone and shale (Fig.3). However, Morcles and Glarus samples display only small

differences between the vein and adjacent host rock (Tab.1). Such small variations can

originate in a closed system (Burkhard & Kerrich 1988; Hoefs 1997, p.162). In case of

extensive fluid flow, low isotopic variation could be explained in terms of rapid rock

buffering (Badertscher et al. 2002a).

The (87Sr/86Sr)25 and the (87Sr/86Sr)460 ratios (indices referring to the recalculated age)

from the veins are generally lower than the bulk siltstone/marl and shale samples (Fig.4).

This is due to higher Sr ratios in the silicate minerals (feldspar and sheet silicates), which

represent detritus from older metamorphic units. Especially sheet silicates being more

abundant in shales have higher Rb-Sr-ratios leading to higher Sr ratios.

In the samples from the Helvetic nappes vein calcite and calcite analysed form the

leached fractions of shale and marl range between 0.70949 and 0.70979. These values are


75

higher than typical marine carbonate values in Cretaceous to Tertiary times (Faure 1986,

p.188) and correspond with 87Sr/86Sr isotopic ratios of calc-mylonites, measured across

the detachment zone of the Glarus thrust, which consist of many small veins (0.7094 –

0.7126, Badertscher et al. 2002a). The strontium ratios may deviate (in the same shale

layer a few millimetres apart) by as much as 0.00183 (Tab.2, Fig.4 drill 11 and 12).

Tab.1. Stable isotope ( 18OSMOW, 13CPDB) data measured in antitaxial calcite veins and calcite fractions of the surrounding host rock. Samples were taken in the Glarus nappe and Morcles nappe, Switzerland, and the Taconic allochthon, New York State, USA.

sample sample-

no.

rock type µmol CO213C 18O

[‰] PDB [‰] SMOW

CH-9908B 1 shale 36.86 -1.3 16.94

CH-9908B 2a marl 20.45 -0.38 19.47

CH-9908B 2b host rock 11.85 -0.39 19.44

CH-9908B 3 calcite vein 30.88 -0.5 19.11

CH-9908B(.1) 15 calcite vein 77.54 -0.44 19.19

CH-9908B(.1) 16 marl 41.93 -0.37 19.95

CH-01C 8 calcite vein 127.46 0.25 19.96

CH-01C 10a calcite vein 92.64 0.28 19.9

CH-01C 10b calcite vein 65.25 0.27 19.86

CH-01B 11a calcite vein 77.79 0.54 19.56

CH-01B 11b calcite vein 115.09 0.53 19.75

CH-01B 12a siltst. 27.53 1.09 22.08

CH-01B 12b host rock 13.3 1.03 22.14

NY-19C 5 calcite vein 86.35 -1.6 16.55

NY-19C 6 calcite vein 33.78 -1.61 16.36

NY-38 13 calcite vein 94.52 -2.22 16.57

NY-38 14a host rock 5.32 -2.66 19.31

NY-38 14b host rock 5.08 -3.02 19.35

NY-38 18 marl 24.84 -4.17 21.62

NY-38 19 marl 17.51 -3.95 21.36

NY-38 20 calcite vein 158.13 -2.21 16.63

NY-32 21 host rock 28.46 -0.5 20.73

NY-32 22 calcite vein 187.52 -1.73 16.68

NY-38 23 host rock 27.15 -3.74 30.02

NY-38 24 calcite vein 129.56 -2.2 16.61

NY-38 25 Calcite vein 59.13 -2.18 17.05


76

The (87Sr/86Sr)25 ratios of leached and leachate samples should add to the (87Sr/86Sr)25

ratio of the corresponding bulk analysis. This is true for siltstone in the Glarus sample but

not for the Morcles samples. Here the bulk samples show higher strontium ratios than

those of the leached and leachate fractions, which supports the small scale isotopic

heterogeneity mentioned above. This is caused by thin layers of marl within the shale

sample, which were used for leached analyses.

Tab. 2. Radiogenic 87Sr/86Sr isotope data derived from antitaxial calcite veins and surrounding host rock. Vein formation in the Glarus and Morcles nappes were most likely associated with peak metamorphism at 25 Ma, the Appalachian samples were recalculated to an age of 460 Ma, which is assumed to represent the Taconic orogeny (Kidd pers. com.). Sample number CH-9908 refers to the Morcles nappe veins, CH-01 to the Glarus nappe veins, and NY to the Appalachian veins, respectively. Host rock refers to a bulk sample, which contains finely laminated marl and shale beds.


77

Appalachian samples display a more pronounced spread of 18O and 13C. Oxygen data

range from 16.36 to +17.05 per mil in the veins, and +19.31 to +30.02 per mil in the

host rock. Carbon isotopic composition ranges from –2.22 to -1.6 per mil in the calcite

veins and –4.17 to –0.5 per mil in the host rocks (Fig. 3). Similar to the Helvetic samples,

the 18O values are higher in the host rock. Radiogenic isotope data vary between the

veins and the shale (e.g. vein 0.71566, shale 0.71875 (87Sr/86Sr)460 (Fig.4). Similar to the

Helvetic samples, the values are higher than seawater.

Fig. 3. Comparison of stable isotope data with published data derived in the Mesozoic rocks in the Swiss Alps (source of data mentioned in the figure). Veins are lighter in

18O values, but overall variations are small in our samples from the Glarus and Morcles nappe. These data are in good agreement with published data. Variation is larger in the lower Palaeozoic Appalachian samples. This might be caused by fractionation, because the rock was most likely subjected to different fluid and thermal events owing to its older age.

Discussion

The transport of matter by fluid flow along a fracture would result in an elongate-blocky

vein microstructure with a decrease in growth rate and grain size along-vein (Lee et al.

1996; Lee & Morse 1999; Hilgers et al. 2004). A variation in the fluid chemistry during

growth would thus cause a decrease in width of the cathodoluminescence band along-

vein, which is not visible in our samples. The presence of fibrous grains and the uniform

bands of cathodoluminescence in our samples thus imply simultaneous growth on both


78

sides of the vein along the vein-walls. This indicates transport of matter from either side

towards the vein-wall interface.

Transgranular fractures of quartz grains indicate that our antitaxial veins nucleated by

brittle fracturing (Fig.2d,f; see also Hilgers & Urai 2002; Bons & Montenari in press). It

allows the precipitation of calcite in dilation sites, and the coalescence of nearby

fractures, forming small veins (Fig.2f).

Fig. 4. Graph presenting the variation of 87Sr/86Sr ratios of host rocks and veins. Data are re-calculated to initial values at 25 Ma (Morcles and Glarus veins) and 460 Ma (Appalachian veins), when vein formation probably took place. Radiogenic Sr isotope ratios of the veins are higher than corresponding marine carbonates and host rock samples. Note that veins are in close relationship with soluble host rock and vein data in the Morcles and Glarus samples.

The growth of our antitaxial fibrous veins may be driven by transport along dissolution

planes (stylolites/cleavage planes), which is indicated by the consistent cleavage spacings

and the fibre width of the vein (Fig.2d,e). Microstructural evidence suggests that pressure

solution is the dominant deformation mechanism in the host rock. The location of a

quartz pressure fringe next to an antitaxial vein implies the same growth process for both

microstructures (Fig.2e). Due to the absence of inclusion bands and other microstructural

evidence of repeated crack-sealing in the veins (e.g. serrated grain boundaries,

microfractures within the vein visible under cathodoluminescence) we argue that both

quartz pressure fringes and antitaxial veins formed by continuous growth during pressure

solution (see also Durney 1972; Durney & Ramsay 1973; Hilgers & Urai 2002; Bons &

Jessell 1997; Hilgers & Urai in press; Bons & Montenari in press). Although fracturing


79

nucleates antitaxial veins, evidence for repeated and multiple crack-seal increments such

as inclusion bands or serrated fibre grain boundaries are absent. Since the veins clearly

show dilational deformation such as the displacement of fractured host rock fragments,

new material must precipitate in voids at the vein-wall interface. The formation of fibrous

crystals requires growth in small voids immediately adjacent to the wall, which prevents

the development of crystal facets and the onset of growth competition (Urai et al. 1991;

Li 2000; Means & Li 2001; Hilgers et al. 2001). Consequently, the vein’s thickness does

not display the initial fracture aperture and antitaxial veins never acted as important fluid

conduits. Overall, these microstructural observations are not in favour of long-range fluid

conduits but more likely indicate pervasive flow or diffusion (see also Bons 2000; Oliver

& Bons 2001; Elburg et al. 2002).

The isotopic data of the veins and adjacent host rocks may present evidence on the

element transfer, and are important to infer the fluid transport mechanism.

In the sample from the Morcles nappe, the (87Sr/86Sr)25 ratio of calcite leached from the

shale (0.70949) is slightly lower than the (87Sr/86Sr)25 ratios of vein and calcite in marl

beds (both identical 0.70957). This variation is beyond the margins given by

reproducibility, analytical error and error due to back calculation (0.000031). The

difference between vein calcite and shale calcite indicates that calcite overgrowths in the

shale were not formed from a fluid which entered the shale through the vein, allowing the

precipitation of vein calcite and calcite in the host rock. In that case, one would have to

explain why calcite in the shale is lower radiogenic while all other minerals in the shale

which could have affected the fluid phase are much higher radiogenic (assuming that the

Sr ratio of the vein is characteristic of the fluid).

However, in the case of an opposite transport direction (from the shale towards the vein),

mobilisation of calcite from shale combined with minor contamination of Sr released

from silicate phases in the shale is in accordance with the isotopic ratios. This would

explain the (87Sr/86Sr)25 ratios in the vein. Taken the high modal abundance of chlorite in

the host rock and considering the low stability of this mineral (e.g. high ionic exchange

capacity, Steiger & Jäger 1977), it can be assumed that under greenschist facies

conditions Sr was mobilised from chlorite (Glodny 1997). Let us assume that chlorite had

a 87Sr/86Sr ratio of 0.7190 in the shale, a moderate assumption as this is the lowest


80

radiogenic whole rock ratio in the shale. It comes out that less than 1 mole % of the Sr in

the vein needs to be derived from chlorite in order to explain the 87Sr/86Sr ratio in the

vein. Consequently, 99 mole % of the Sr in the vein is derived from calcite from the shale

and marl.

The identical isotopic ratio of calcite in the marl and vein are in accordance with calcite

being precipitated from the same fluid or from the mobilisation of calcite from the marl

into the vein. The stable isotope data of that sample allow similar conclusions to be made,

as the 18O and 13C values of the vein lie between marl calcite and the shale calcite, with

the marl values being closer to those of the vein.

Therefore, the most plausible explanation of these isotopic data is a short transport

distance from the immediate vein surroundings. The stable isotopes especially indicate

that both rock types, shale and marl, contributed to the vein filling, although the marl

contribution must have dominated.

In the sample from the Glarus nappe (CH-01B) calcite leached from the siltstone hosting

the vein has a lower (87Sr/86Sr)25 ratio (0.70979) than that of the vein (0.71010). Although

the difference is small it still exceeds the analytical uncertainty given by the method

similar to the example from the Morcles nappe.

If the calcite in the vein formed by mobilisation of calcite from the siltstone, a second

more radiogenic component would have to be involved. Similar to the samples from the

Morcles nappe, sheet silicates most probably have contributed to the exchange process.

Assuming an average (87Sr/86Sr)25 ratio of 0.7160 for the silicate minerals in the siltstone,

only 5 mole % of that Sr together with 95 mole % of Sr from calcite in the siltstone

would explain the isotopic ratio of the vein.

The differences in 18O and 13C between calcite in the siltstone and calcite in the vein

may point to an additional source of calcite in the siltstone. However, 18O values are

approximately 2-3 per mil lower than the silicate rock and can be generated by

equilibration between the fluid and clay containing rocks (Burkhard & Kerrich 1988).

This means that the (87Sr/86Sr)25 as well as the 18O values in the Glarus samples could

signify local element transfer. Only the 13C in the vein requires an input of an external

source, e.g. a shale with carbonaceous material, characterised by a low 13C. As the


81

difference between the 13C in the veins to those in the siltstone is small, this input would

have been of minor importance.

In both the Morcles and Glarus samples the isotopic ratios are in accordance with the

majority of the veins being derived from the immediate surroundings. As was indicated

by the isotopic characteristics and by the modal composition of the rocks, the calcite has

been derived from the marl and siltstone.

This leads to the question whether enough calcite is present in the host rock proximal to

the vein prior to vein formation. Our data indicate that the veins Sr concentrations range

from 1350-2000 ppm (Tab.2). Assuming that a vein contains an average Sr concentration

of 1500 ppm, the volume of rock required for mobilisation of calcite consequently

depends on the modal abundance of calcite and its Sr concentration. Even if the latter was

500 ppm (Wedepohl 1978, Table 38 K-4) and calcite had an abundance of 10 wt.-%,

mass balance calculations indicate that in a rock volume about 30 times larger than the

vein volume all calcite would be dissolved and re-precipitated in the vein. Petrographic

observations indicate a lower vein-rock volume ratio of 1:3 to 1:20 in the shale.

Assuming that the bulk volume remained constant in the rock during cleavage formation

(Erslev & Ward 1994), that would mean calcium, carbonate and strontium ions were

transported over a larger distance, although the isotopic values are in accordance with a

local element exchange.

The slight isotopic heterogeneity in the rocks indicates (i) that the fluid was isotopically

not noticeably different from the isotopic ratios of the rocks or (ii) that an initially

different fluid changed its isotopic value dynamically with the ongoing fluid-rock

interaction (i.e. fast rock buffering). Both alternatives cannot be excluded or indeed a

combination of the two. Possible 87Sr/86Sr ratios of a fluid range between 0.709 and

0.7126 specified by the Mesozoic marine limestones of the Helvetic nappes, and the data

along the detachment zone of the Glarus thrust, interpreted as a pathway of focussed fluid

flow (Badertscher et al. 2002a). The close proximity of these data relative to the rock

values of our samples supports the first alternative. This means that antitaxial vein

microstructures as well as 87Sr/86Sr ratios suggest local mass transfer. However, assuming

no bulk volume change during pressure solution and cleavage formation, an additional

external source is required. These results are in agreement with data by Bardetscher et al.


82

(2002), who describe quick rock buffered fluids, advecting through the rock around the

Glarus thrust. Elburg et al. (2002) present radiogenic 87Sr/86Sr data of antitaxial fibrous

calcite veins hosted in Neoproterozoic shale, Flinders Ranges, Australia. They argue that

solute in most of their veins was transported by pervasive fluid, because for all but one

sample isotope ratios of the vein and the wall rock are significantly different.

In the Appalachian samples the large heterogeneity of the stable isotope data argues

against a closed system. The vein samples in particular are characterised by relatively

restricted values, probably pointing to a homogenising fluid flow event. The 18O data of

the veins, which are generally lower than those of the clastic host rock might in part be

explained by an equilibration process of the fluid causing calcite formation. The 13C

values in the veins could be explained by an uptake and homogenization from carbon of

the hosting clastic rocks, which have lower and higher 13C values than the vein.

However, the discrepancies suggest an influx from external sources.

The veins of the Appalachian samples have (87Sr/86Sr)460 ratios of 0.71566 and 0.71674,

which are lower than the bulk host rock data (0.71875 and 0.71763). These data are

markedly higher than 87Sr/86Sr ratios of Cambrian to Ordovician limestone of the Hatch

Hill formation (i.e. 0.71067 to 0.71135, Friedman 2002) and higher than Cambrian to

Ordovician seawater (0.70940 to 0.70782, Montanez et al. 1996; Denison et al. 1998).

Friedman (2002) presents highly radiogenic strontium ratios of 0.715096 in Devonian

rocks considered to represent fluids derived from sedimentary rocks of the Taconic

sequence (including the Hatch Hill formation) or the crystalline basement. He also

correlates his data with strontium ratios of late Cambrian continental crust in this region,

these continental crust ratios fit with our data. The difference of vein and host rock

strontium ratios corresponds with stable isotope data, indicating a fluid influx from

external sources which could be associated with Devonian or post-Devonian fluids. This

example illustrates the difficulty in interpreting isotope data and microstructural

observations in Palaeozoic rocks affected by various fluid events subsequent to the initial

vein formation.


83

Conclusion

A symmetrical grain shape in our fibrous veins as well as the three dimensional isolation

of veins indicate transport of matter through the rock matrix. The vein microstructure

depends on the host rock composition and varies laterally from antitaxial veins in shale to

stretched crystals and blocky veins in the siltstone and marl beds. Antitaxial calcite veins

nucleated in brittle fractures. Quartz pressure fringes around pyrites and antitaxial calcite

veins grew simultaneously from the same fluid by continuous growth rather than multiple

crack-seal growth.

Radiogenic (87Sr/86Sr)25 and stable isotope data ( 18O, 13C) vary only slightly between

vein, host rock and host rock calcites in the Tertiary Helvetic samples and thus could be

interpreted in terms of local mass transfer. Assuming constant bulk volume during

cleavage formation, mass balance of Sr indicates that a significant amount of vein calcite

cannot be derived from the immediate surroundings of the vein. Instead a solution

carrying calcium, carbonate and strontium ions must have been transported over a larger

distance. This shows that mass balance is a crucial parameter for the interpretation of

isotope data. External input by long-range transport characterises the Appalachian veins

hosted in the Cambrian shales.

Although the vein microstructure points to a local transport process from the surrounding

host rock to the vein, the lack of enough calcite in the immediate surroundings suggests

that the formation of our antitaxial calcite veins involves larger scale transport processes

and may have no significant local source.

Acknowledgments

Bill Kidd (SUNY, Albany) provided valuable data on the geology of the Appalachians.

Discussions with Janos Urai and Ulrich Kramm are greatly acknowleged. Janos Urai

kindly provided samples from the Glarus area. We would like to thank Ulrike Schulte and

Dieter Buhl (Geologie, Mineralogie & Geophysik, Sediment- & Isotopengeologie,

Bochum) who measured the stable isotope data. Jutta Jakobi, Gertrud Siebel (RWTH

Aachen) and Heidi Baier (WWU Münster) are thanked for assistance in analytical work,

and Amanda Rogers (RWTH Aachen) for her assistance in improving the English text.

Comments by Marlina Elburg and an anonymous reviewer significantly improved the

manuscript.


84

References

Badertscher, NP, Abart, R, Burkhard, M, McCaig, A (2002a) Fluid flow pathways along the Glarus overthrust derived from stable and Sr-Isotope patterns. AmericanJournal of Science, 302, 517-547.

Badertscher, NP, Beaudoin, G, Therrien, R, Burkhard, M (2002b) Glarus overthrust: A major pathway for the escape of fluids out of the Alpine orogen. Geology, 30(10),875-878.

Bestmann, M, Kunze, K, Matthews, A (2000) Evolution of a calcite marble shear zone complex on Thassos Island, Greece: microstructural and textural fabrics and their kinematic significance. Journal of Structural Geology, 22, 1789-1807.

Bons, PD (2000) The formation of veins and their micostructures. Journal of the Virtual Explorer, 2.

Bons, PD, Jessell, MW (1997) Experimental simulation of the formation of fibrous veins by localised dissolution-precipitation creep. Mineralogical Magazine, 61, 53-63.

Bons, PD, Montenari, M (in press) The formation of antitaxial calcite veins with well-developed fibres, Oppaminda Creek, South Australia. Journal of Structural Geology, 1-18.

Bradley, DC (1989) Taconic plate kinematics as revealed by foredeep stratigraphy, Appalachian orogen. Tectonics, 8(5), 1037-1049.

Burkhard, M, Badertscher, NP (2001) Finite strain has no influence on the illite crystallinity of tectonized Eocene limestone breccias of the Morcles nappe, Swiss Alps. Clay Minerals, 36, 171-180.

Burkhard, M, Kerrich, R (1988) Fluid regimes in the deformation of the Helvetic nappes, Switzerland, as inferred from stable isotope data. Contributions to Mineralogy and Petrology, 99, 416-429.

Burkhard, M, Sommaruga, A (1998) Evolution of the western Swiss Molasse basin: structural relations with the Alps and the Jura belt. In: Cenozoic foreland basins of Western Europe. (eds) A. Mascle, C. Puidefàbregas, H.P. Luterbach and M. Fernàndez. Geological Society Special Publication, pp. 279-298. Geological Society, London.

Cox, SF, Etheridge, MA (1983) Crack-seal fibre growth mechanisms and their significance in the development of oriented layer silicate microstructures. Tectonophysics, 92, 147-170.

Cox, SF, Knackstedt, MA, Braun, J (2001) Principles of structural control on permeability and fluid flow in hydrothermal systems. Reviews in Economic Geology, 14, 1-24.

Crespo-Blanc, A, Mason, H, Sharp, Z, Cosca, M, Hunziker, J (1995) A stable and 40Ar/39Ar isotope study of a major thrust in the Helvetic nappes (Swiss Alps): Evidence for fluid flow and constraints on nappe kinematics. GSA Bulletin,107(10), 1129-1144.

Denison, RE, Koepnick, RB, Burke, WH, Hetherington, EA (1998) Construction of the Cambrian and Ordovician seawater 87Sr/86Sr curve. Chemical Geology, 152,325-340.

Dietrich, D, McKenzie, JA, Song, H (1983) Origin of a calcite in syntectonic veins as determined from carbon-isotope ratios. Geology, 11, 547-551.


85

Durney, DW (1972) Solution-transfer, an important geological deformation mechanism. Nature, 235, 315-317.

Durney, DW, Ramsay, JG (1973) Incremental strains measured by syntectonic crystal growth. In: Gravity and tectonics. (eds) K.A. de Jong and R. Scholten, pp. 67-96. Wiley, New York.

Elburg, MA, Bons, PD, Foden, J, Passchier, CW (2002) The origin of fibrous veins: constraints from geochemistry. In: Deformation mechanisms, rheology and tectonics: Current status and future perspectives. (eds) S. de Meer, M.R. Drury, J.H.P. de Bresser and G.M. Pennock. Special Publications, pp. 103-118. Geological Society, London.

Erslev, EA, Ward, DJ (1994) Non-volatile element and volume flux in coalesced slaty cleavage. Journal of Structural Geology, 16, 531-554.

Faure, G (1986) Principles of isotope geology. 589 pp. Wiley, New York. Fisher, DM (1996) Fabrics and veins in the forearc: A record of cyclic fluid flow at depth

of <15km. In: Subduction: Top to bottom. (eds) G.E. Bebout, D.W. Scholl, S.H. Kirby and J.P. Platt, pp. 75-89. American Geophysical Union

Fletcher, RC, Merino, E (2001) Mineral growth in rocks: Kinetic - rheological models of replacement, vein formation, and syntectonic crystallization. Geochimica et Cosmochimica Acta, 65(21), 3733-3748.

Friedman, GM (2002) Highest Phanerozoic strontium isotopic ratios of pre-rift Late Cambrian passive margin in New York State, USA: products of continental weathering and orogenesis. Sedimentary Geology, 147, 143-153.

Glodny, J (1997) Der Einfluß von Deformation und fluidinduzierter Diaphtorese auf radioaktive Zerfallssysteme in Kristallingesteinen. PhD Thesis, WWU Münster, Münster, 262 pp.

Grigor'ev, DP (1965) Ontogeny of minerals. 250 pp. pp. Israel Program for Scientific Translation Ltd., Jerusalem.

Hayman, NW, Kidd, WSF (2002) Reactivation of prethrusting, synconvergence normal faults as ramp within the Ordovician Champlain-Taconic thrust system. GSABulletin, 114(4), 476-489.

Henry, C, Burkhard, M, Goffe, B (1996) Evolution of synmetamorphic veins and their wallrocks trough a Western Alps transect: no evidence for large-scale fluid flow. Stable isotope, major- and trace-element systematics. Chemical Geology, 127, 81-109.

Hilgers, C, Dilg-Gruschinski, K, Urai, JL (2004) Microstructural evolution of syntaxial veins formed by advective flow. Geology, 32(3), 261-264.

Hilgers, C, Koehn, D, Bons, PD, Urai, JL (2001) Development of crystal morphology during unitaxial growth in a progressively widening vein: II. Numerical simulations of the evolution of antitaxial fibrous veins. Journal of Structural Geology, 23, 873-885.

Hilgers, C, Urai, JL (2002) Microstructural observations on natural syntectonic fibrous veins: Implications for the growth process. Tectonophysics, 352, 257-274.

Hilgers, C, Urai, JL (in press) On the arrangment of solid inclusions in fibrous veins and the role of the crack-seal mechanism. Journal of Structural Geology.

Hoefs, J (1997) Stable isotope geochemistry. 201 pp. Springer, Berlin.


86

Hunziker, JC et al. (1986) The evolution of illite to muscovite; mineralogical and isotopic data from the Glarus Alps, Switzerland. Contributions to Mineralogy and Petrology, 92, 157-180.

Huon, S, Burkhard, M, Hunziker, J-C (1994) Mineralogical, K-Ar, stable and Sr isotope systematics of K-white micas during very low-grade metamorphism of limestones (Helvetic nappes, western Switzerland). Chemical Geology, 113(3-4), 347-376.

Kirschner, DL, Masson, H, Cosca, M, Sharp, Z, Hunziker, J (1995a) An integrated deforamtion-temperature-time (D-T-t) history for the Morcles nappe (Helvetic Alps, Switzerland): Constraints from stable isotope geothermometry and argon 40/39 geochronology. Journal of the Czech Geological Society, 40(3), A-22.

Kirschner, DL, Sharp, ZD, Masson, H (1995b) Oxygen isotope thermometry of quartz-calcite veins: Unraveling the thermal-tectonic history of the subgreenschist facies Morcles nappe (Swiss Alps). GSA Bulletin, 107, 1145-1156.

Kirschner, DL, Sharp, ZD, Teyssier, C (1993) Vein growth mechanisms and fluid sources revealed by oxygen isotope laser microprobe. Geology, 21, 85-88.

Lee, Y-J, Morse, JW (1999) Calcite precipitation in synthetic veins: implications for the time and fluid volume necessary for vein filling. Chemical Geology, 156, 151-170.

Lee, Y-J, Morse, JW, Wiltschko, DV (1996) An experimentally verified model for calcite precipitation in veins. Chemical Geology, 130, 203-215.

Li, T (2000) Experimental growth of fibers and fibrous veins. PhD Thesis, State University of New York at Albany, Albany, 420 pp.

McCaig, A, Wayne, DM, Marshall, JD, Banks, DA, Henderson, I (1995) Isotopic and fluid inclusion studies of fluid movement along the Gavarnie thrust, Central Pyrenees: Reaction fronts in carbonate mylonites. American Journal of Science,295, 309-343.

Means, WD, Li, T (2001) A laboratory simulation of fibrous veins: some first observations. Journal of Structural Geology, 23, 857-863.

Montanez, IP, Banner, JL, Osleger, DA, Borg, LE, Bosserman, PJ (1996) Integrated Sr isotope variations and sea-level history of Middle to Upper Cambrian platform carbonates: Implications for the evolution of Cambrian seawater 87Sr/86Sr. Geology, 24(10), 917-920.

Mügge, O (1928) Ueber die Entstehung faseriger Minerale und ihrer Aggregationsformen. Neues Jahrbuch für Mineralogie, Geologie und Paläontologie, 58(A), 303-348.

Neumann, W, Huster, E (1974) The half life of 87Rb, measured as difference between the isotopes 87Rb and 85Rb. Zeitschrift für Physik, 270, 121 - 127.

Oliver, NHS (1996) Review and classification of structural controls on fluid flow during regional metamorphism. Journal of Metamorphic Geology, 14, 477-492.

Oliver, NHS, Bons, PD (2001) Mechanisms of fluid flow and fluid-rock interaction in fossil metamorphic hydrothermal systems inferred from vein-wall rock patterns, geometry and microstructure. Geofluids, 1(2), 137-162.

Oliver, NHS, Cartwright, I, Wall, VJ, Golding, SD (1993) Stable isotope signature of kilometre-scale fracture-dominated metamorphic fluid pathways, Mary Kathlee, Australia. Journal of Metamorphic Geology, 11, 705-720.

Passchier, CW, Trouw, RAJ (1996) Microtectonics. 289 pp. Springer, Berlin.


87

Passchier, CW, Urai, JL (1988) Vorticity and strain analysis using Mohr diagrams. Journal of Structural Geology, 10(7), 755-763.

Paterson, MS (1978) Experimental rock deformation - the brittle field. Minerals and Rocks, 13. 254 pp. Springer, Berlin.

Price, NJ, Cosgrove, JW (1990) Analysis of geological structuresCambridge University Press, Cambridge.

Ramsay, JG (1980) The crack-seal mechanism of rock deformation. Nature, 284(5752),135-139.

Ramsay, JG, Dietrich, D, Casey, M (1982) Excursions A & B Western Helvetic nappes. Field guide International Conference on planar and linear fabrics of deformed rocks.

Ramsay, JG, Huber (1983) Techniques in modern structural geology. Volume 1: Strain analysis. 307 pp. Academic Press, London.

Richards, IJ, Connelly, JB, Gregory, RT, Gray, DR (2002) The importance of diffusion, advection, and host-rock lithology on vein formation: A stable isotope study from the Paleozoic Ouachita orogenic belt, Arkansas and Oklahoma. GSA Bulletin,114(11), 1343-1355.

Rye, DM, Bradbury, HJ (1988) Fluid flow in the crust: An example from a Pyrenean thrust ramp. American Journal of Science, 288, 197-235.

Sibson, RH (2004) Controls on maximum fluid overpressure defining conditions for mesozonal mineralisation. Journal of Structural Geology, 26, 1127-1136.

Steiger, RH, Jäger, E (1977) Subcommission on Geochronology: Convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36, 359-362.

Urai, JL, Williams, PF, van Roermund, HLM (1991) Kinematics of crystal growth in syntectonic fibrous veins. Journal of Structural Geology, 13(7), 823-836.

Walther, JV, Orville, PM (1982) Volatile production and transport in regional metamorphism. Contributions to Mineraleralogy and Petrology, 79, 252-257.

Wedepohl, KH (1978) Handbook of Geochemistry, Vol. II/4Springer, Berlin, Heidelberg, New York.

Wiltschko, DV, Morse, JW (2001) Crystallization pressure versus "crack seal" as the mechanism for banded veins. Geology, 29(1), 79-82.

Yardley, BWD (1986) Fluid migration and veining in the Connemara schists, Ireland. In: Fluid-rock interactions during metamorphism. (eds) J.V. Walther and B.J. Wood, pp. 110-131. Springer, 218pp

Chapter 6 – Fracture sealing in a regional, high-pressure cell in Jabal Akhdar, Oman mountains

88

Chapter 6 - Fracture sealing in a regional, high-pressure cell in Jabal Akhdar, Oman mountains - first results

Christoph Hilgers1, David L. Kirschner2, Jean-Paul Breton3, Janos L. Urai1

1Geologie-Endogene Dynamik, RWTH Aachen, Lochnerstr. 4-20, D-52056 Aachen,

Germany 2Department of Earth and Atmospheric Science, St. Louis University, MO

3BRGM Oman Branch, PO Box 3785, Ruwi 112, Oman

subm to Geofluids

Abstract

Fractures are important conduits for fluid flow in the Earth’s crust. To better understand

the spatial and temporal relations among fracturing, fracture sealing, and fluid flow, we

have studied fractures, faults, and veins in a large dome (Jabal Akhdar) in the Oman

mountains. Our work combines the results of meso- and microstructural analyses and

stable isotope analyses.

Seven generations of fractures and veins have been identified in the carbonate-rich dome.

The earliest generations of veins developed during extension and subsidence of the

Mesozoic basin. These veins formed in the inclined segments of bedding-parallel

stylolites and in extensional veins that are perpendicular to bedding (#1 and #2,

respectively). These extension-related veins are truncated by bedding-parallel veins (#3)

that developed prior to displacement on some large normal faults. These veins are

consistent with a change in stress regime (from brittle transtension to ductile

compressional shear) and are related to strongly deformed earlier pinch-and-swell veins

on the southern limb of the dome (#4). Normal faults contain a set of en-echelon tension

gashes (#5) and veins emplaced in dilational jogs along the fault planes (#6). In the

northern part of the dome, veins associated with thrusts post-date the normal faults (#7).

Samples of these vein sets and their host rocks were analyzed to provide information on

fluid-rock interaction in the dome and the scale(s) of fluid movement. Oxygen isotope

values range from ca. +16.2 to +29.3 ‰; carbon isotope values range from 0 to +3.6 ‰.


89

The results of the structural and isotopic analyses are consistent with the early veins (#2-

#5) having precipitated from overpressured fluid in a rock-buffered system. During

normal faulting (#5 and #6), a more open system allowed external fluid to infiltrate the

dome at drained conditions and precipitate the youngest sets of veins (#6 and #7).

Introduction

The Oman mountains of the Alpine-Himalayan chain formed during northeast-directed

subduction of Arabia below the Eurasian plate (Fig.1). Intra-oceanic subduction started in

the Cenomanian and continued into the Middle Turonian to early Campanian with the

obduction of the Samail ophiolite (Boudier et al. 1985; Beurrier et al. 1989; Hacker 1994;

Hacker et al. 1996; Breton et al. 2004 and references therein). Several tectonic windows

below the ophiolite are present in the region including the Jabal Akhdar, which is the

focus of this study. Metamorphic facies associated with subduction range from upper

anchizone in the Jabal Akhdar to blueschist and eclogite facies in the Saih Hatat tectonic

window. In the Saih Hatat, eclogites were subducted to depths of 70 km at approximately

80 Ma (Hacker 1994; Gray et al. 2000; Breton et al. 2004; Searle et al. 2004) (Fig.1). The

increase of the metamorphic gradient towards the north indicates that the subduction was

ocean-ward verging (le Métour 1988; Breton et al. 2004; Searle et al. 2004).

The Jabal Akhdar dome, which is located in the central part of the Oman mountains, is a

large 2500 km2 box-shaped fold with an amplitude of 3 km and a wavelength of 70 km

(Searle 1985). The dome formed in the Tertiary (Glennie et al. 1974; Glennie 1995; Gray

et al. 2000) (Fig.2). Exposed in the core of the Jabal Akhdar dome are pre-Permian rocks

that are unconformably overlain by Middle Permian to Cenomanian carbonates (Glennie

et al. 1974). These 2.5-km-thick carbonates were deposited on the subsiding southern

passive margin of the Tethyan ocean (Mann et al. 1990; Hanna 1990; Pratt & Smewing

1993; Masse et al. 1997; Masse et al. 1998; Hillgärtner et al. 2003). The carbonates are

very well exposed laterally and vertically, with excellent 1-km-high exposures in the

wadis that transect the dome.

The deformation and fluid-related histories of the dome can be deciphered from the

numerous crosscutting faults, fractures, and veins in the carbonates. To document the

deformation and fluid-related histories, we combine the observations and analyses of our


90

field study with microstructural observations of the veins, and stable isotope analysis of

host rocks and veins to derive the deformation and fluid histories of the carbonates during

deformation.

Fig. 1. Three tectonic windows are present in the Oman mountains. The Jabal Akhdar tectonic window is a dome with Infracambrian sedimentary rocks in its core. The Hawasina nappes and the Samail ophiolite were obducted southward on autochthonous basement and Mesozoic platform carbonates (modified after Glennieet al. 1973; Hanna 1990; McL. Miller et al. 2002). (inset) Arrows indicate orientations of recent stress field in regional map. The Zagros mountains and Makran subduction zones are the suture zone between Eurasian and Arabian plates (modified after Konert et al. 2001; Breton et al. 2004; stress orientations from http://www.world-stress-map.org).


91

Geological Setting

Exposed in the Jabal Akhdar dome are Infracambrian sedimentary rocks that are

unconformably overlain by Mesozoic autochtonous platform carbonates. To date, we

have focused our study on the Cretaceous limestones and clayey limestones that are

exposed in the Wadi Mistal and Wadi Bani Awf of the north limb and Wadi Nakhar of

the southern limb (Fig.2). These platform carbonates, which transition into deep-water

facies northeast of the dome (Masse et al. 1997; Hillgärtner et al. 2003), have been

described in detail by Breton et al. 2004).

Fig.2. Geologic map of Jabal Akhdar dome. The outer limb consists of autochthonous Mesozoic carbonates, which were sampled in Wadi Bani Awf and Wadi Mistal on the northern flank, and Wadi Nakhar on the southern flank. (modified after Bechennec et al. 1993).

The uppermost autochthonous carbonate unit is comprised of Turonian to Santonian

megabreccias and olistoliths of the Muti formation. These sediments were derived from

an outer carbonate platform to the northeast and document the transition from a passive

continental margin to a foredeep basin (Robertson 1987; Breton et al. 2004). The pelagic

matrix of the megabreccias is Middle-Late Turonian in the northern part of Jabal Akhdar

and Coniacian-Santonian in the southwest part of Jabal Akhdar (le Métour 1988; Rabu


92

1988; Breton et al. 2004), consistent with migration of the depocenter towards the south

during thrusting (Robertson 1987; Breton et al. 2004).

The Muti formation is unconformably overlain by allochthonous Permian-Cretaceous

volcano-sedimentary rocks of the Hawasina nappes and the Samail ophiolite. The Samail

ophiolite comprises two magmatic sequences. The first one is dated Albian to early

Cenomanian (Beurrier 1988) and formed along a 500-km-long paleo-ridge (Nicolas et al.

1988). The second sequence is dated at 97 to 94 Ma (Tilton et al. 1981; Beurrier 1988)

and is related to the intra-oceanic subduction. Oduction of the Samail ophiolite onto the

continental margin north of Jabal Akhdar started in the Middle Turonian. The Hawasina

and Samail tectonic pile was then thrust southwest over the Jabal Akhdar area in Late

Santonian (Bechennec et al. 1988; Breton et al. 2004). Almost 450 km of displacement of

the Semail ophiolite obduction occurred between 95 and 80 Ma (Warburton et al. 1990;

Breton et al. 2004). Shear indicators at the base of the ophiolite are consistent with top-

to-south shearing over the Jabal Akhdar dome (Boudier et al. 1988).

The thrust stack is unconformably overlain by autochthonous Maastrichtian and early

Tertiary limestones, which were deposited after nappe emplacement (Glennie et al. 1973;

Hanna 1990). The distribution and onlap of Tertiary sediments occurred during Eocene to

Miocene uplift of Jabal Akhdar (Searle et al. 2004), which resulted in the erosion of the

overlying Hawasina and ophiolite nappes in the central part of Jabal Akhdar. The Jabal

Akhdar dome formed during multi-phase deformation from Early Paleocene to Miocene-

Pliocence (Searle 1985; Poupeau et al. 1998; Breton 2002).

Veins

We have identified seven generations of calcite-rich veins in three wadis of the Jabal

Akhdar dome. The veins and overprinting relationships are similar in the three wadis, and

are thus discussed together. The relative timing of the different vein generations is based

on their relationships to normal faults (Tab.1). Normal faults are consistent across Jabal

Akhdar and formed prior to doming.

Stylolite veins (#1)

The first generation of calcite veins formed in steeply inclined segments of stylolite

seams. The stylolite seams are parallel to bedding, enriched in clay, are up to one


93

centimeter thick, have irregular shape indentations, and would be described as composite

seams according to the nomenclature of Guzzetta (1984). In Wadi Nakhar, these stylolite

seams have accommodated up to ten centimeters of shortening by dissolution (Fig. 3a).

Bedding-normal veins (#2)

A set of vertical calcite veins crosscut the stylolites, are perpendicular to bedding (Fig.

3a), and are truncated by normal faults. On the north limb of the dome, bedding-normal

veins form in the necks of boudin layers (Fig. 3b). These are absent in Wadi Nakhar,

where the same generation is present in apparently undeformed beds. The veins strike

north-south to northwest-southeast, parallel to the veins in the stylolite seams.

Bedding-parallel veins (#3)

Two types of bedding-parallel veins have been observed. Thin (< 5cm thick) layered

bedding-parallel veins (#3.1) extend for several tens of meters, truncate subvertical veins

(set #2) (Fig.3c), and are offset locally by en echelon vein arrays and associated normal

faults (#5 and #6) (Fig. 3d,e). The second type of bedding-parallel vein formed in

dilational jogs (#3.2) that opened several centimeters wide during bedding-parallel shear

(Fig.4a). On both limbs, this second type of bedding-parallel vein formed during top-to-

NNE shearing. A relative chronology of these bedding-parallel veins could not been

established by cross-cutting relationships.

Pinch-and-swell veins (#4)

On the southern limb of Jabal Akhdar, veins are locally stretched to form pinch-and-swell

structures in cleaved marls (Fig.4b). The fracture cleavage dips towards the W and

contains small pressure shadows around pyrites that are oriented sub-parallel to cleavage.

Grooves on the pinch-and-swell vein surface are consistent with top to the E shear.

Staining of the pinch-and-swell veins has revealed several stages of fracture-sealing

events. Calcite breccia within a quartz vein indicates that the vein was first sealed with

calcite (Fig.4c). The quartz also becomes brecciated and the vein is sealed by a second

phase of calcite (Fig.4d).

Intersections of pinch-and-swell and other veins have not been observed, thus a relative

age relationship is difficult to establish. However, the pinch-and-swell veins are truncated


94

by brittle normal faults and thus pre-date normal faulting. These veins may be associated

with bedding-parallel veins.

Normal faults and associated veins (#5 and 6)

Normal faults, with throws up to several hundreds of meters on both sides of the dome,

are associated with drag folds, slickensides, grooves, and two vein sets (Fig.3d,e). The

dips of conjugate normal faults change across the Jabal Akhdar and are consistent with

faulting having occurred prior to dome formation (Breton 2002). The first set of veins

(#5) is comprised of en-echelon tension gashes that are perpendicular to bedding and

parallel to the normal faults. The en-echelon veins are offset by normal faults and

bedding planes that slipped during drag folding (Fig.4e,f). The second vein set (#6)

occurs in dilational jogs along the normal faults. These veins contain sub-vertical and

sub-horizontal grooves. Cross-cutting relationships are consistent with sub-horizontal

movement having occurred after sub-vertical movement.

Veins in late thrust faults (#7)

Small thrusts, associated folds, and reverse faults are present in the northern limb of the

dome at Wadi Mistal (Fig.3f). The thrusts, which offset the normal faults, contain blocky

calcite veins in the thrust plane. Breton (2002) interprets these late thrusts to have

formed during one phase of doming of Jabal Akhdar in the Tertiary (Glennie et al. 1974,

Gray et al. 2000).


95

Table 1. Overview of vein sets and evidence for relative time relationship (MPS-maximum principal stress).

vein generation description

evidence for relative age

relationship regime

#1 stylolite vein

cross cut by extension veins

(#2) MPS normal to bedding

#2

extension vein,

boudinage vein

truncated by bedding-parallel

vein (#3) MPS normal to bedding

#3 bedding-parallel vein

displaced by normal faults (#5),

cross-cutting relationship of

#3.1 and #3.2 unclear MPS oblique to bedding

#4 pinch & swell vein

undeformed extension vein

nearby, truncated by normal

fault (#5), reltive timing to #3

unclear MPS oblique to bedding

#5

en echelon extension

vein arrays

aligned along and displaced by

normal faults (#6) MPS normal to bedding

#6

veins in dilation sites

of normal faults truncate en echelon veins (#5) MPS normal to bedding

#7 thrust veins cross cuts #5/#4 and #2 MPS oblique to bedding


96

Fig.3. a) Bedding-parallel stylolites contain calcite veins in their steep limbs of the teeth (arrow). Some stylolites are cross-cut by vertical calcite veins. Wadi Nakhar. b) Extensional veins formed in boudin necks in the northern limb of Jabal Akhdar (indicated by arrows). Veins are normal to bedding. c) Thin dark rock slivers within a horizontal vein crosscut a vertical vein. Wadi Nakhar. d) Normal faults displace two horizontal veins (indicated by white arrows). The normal fault on the left contains a thick vein that formed in a dilational jog of the fault. Wadi Nakhar. e) En echelon tension gashes are associated with normal faulting. Slip of the normal fault is dragged (folded) adjacent limestone beds. Rectangular area is enlarged in Figure 4. Wadi Nakhar. f) Reverse faults and thrusts displace earlier normal faults and boudins, and are associated with fault-bend folding. Arrows denote two limestone beds with regularly spaced boudin veins that are oriented normal to bedding. Wadi Mistal


97

Fig.4. a) Bedding-parallel calcite veins form in dilational jogs displace some bedding-normal veins (#2). Shear criteria show top-to-the NNE shear sense. Wadi Nakhar. b) Pinch-and-swell veins are oriented oblique to bedding and cleavage and indicate intense stretching. The dip sub-parallel to cleavage (bedding is horizontal). Wadi Nakhar. c) Stained vein sample from the pinch-and-swell structure. Calcite (dark color) is located along the vein-host rock interface. Central part of vein is filled with quartz (light color). Locally, angular fragments of calcite “float” within quartz vein consistent with quartz precipitation after calcite. d) Overview of stained pinch-and-swell vein. Note angular quartz clasts (light color) within the calcite vein (dark color). Quartz fragments are located in both the pinch and swell regions, indicating that the quartz vein filled the whole vein prior to stretching. Wadi Nakhar. e) Detail of figure 3e. Tension gashes crosscut bedding-parallel veins and are displaced along the bedding plane close to the normal fault. Steeper parts along the normal faults form dilation jogs (#5) and are filled with blocky calcite veins. The normal fault off-sets bedding-parallel veins (#3.1). Wadi Nakhar


98

Methodology of Isotope Analyses

Stable isotope analyses of host rocks and veins were made to determine the history of

fluid-rock interaction and fluid flow during the deformation history. Samples were

collected from approximately 50 outcrops primarily in the three wadis of the Jabal

Akhdar dome. The samples were slabbed with a water-lubricated saw and micro-drilled.

The resulting powders were then analyzed on an automated carbonate reaction device that

is connected to a continuous-flow, gas-source, isotopic ratio mass spectrometer at Saint

Louis University. Approximately 0.5 mg of each powder was reacted with H3PO4 at

90°C for several hours prior to analysis. Approximately one in-house standard was

analyzed for every five unknowns. The in-house standard’s 13C value of -2.33 ‰ and 18O of 24.72 ‰ were calibrated to NBS-19 standard values of 1.95 and 28.64‰,

respectively. The standard deviations of the in-house standard were between 0.01 and

0.08 ‰ for 13C and 0.08 and 0.14 ‰ for 18O (N=58). Some unknown powders were

analyzed two to four times; the reported values are averages of these analyses. The 18O

values were converted to VSMOW according to 18OVSMOW=1.03091 18OPDB+30.91 of

Coplen et al. (1983).

Results of Isotopic Analyses

Host rocks have 13C values between 0.5 and 3.6 ‰ and 18O values between 23.4 and

28.4 ‰ (Tab. 2, Fig.5). Similar values have been obtained from Aptian to Cenomanian

age carbonates that were deposited in the paleo-Tethys sea and similar age oceans (cf.

Scholle & Arthur 1980; Renard 1986; Burkhard & Kerrich 1988; Weissert et al. 1998;

Jenkyns & Wilson 1999; Ghisetti et al. 2001; Skelton 2003 p.169, 267).

Veins have 13C values between 0.1 and 3.6 ‰ and 18O values between 16.2 and 29.3

‰ (Fig.5). Most of the early veins (#1-5) have 13C and 18O values similar to their

adjacent host rock values (Fig.6), consistent with a relatively closed, host-rock-buffered

system during veining. Some veins of generations #6 and #7 have 18O values that differ

from their host rock values, consistent with a more open fluid system with fluids coming

from outside of the carbonate-dominated stratigraphy of the dome itself. The transition

from a more closed to open fluid system can be seen in the increased variability of the


99

18O data that occurs from vein sets #5 to #6 (Fig.7). The isotopic values vary slightly

within individual veins (Fig.8), though much less than the range in values among vein

generations #1 to #7 (Fig.7).

Fig.5. Comparison of stable isotope values measured from wall rock and vein ( 18OVSMOW, 13CPDB). The data of the wall rock are overall consistent with the signatures of the veins. Stable isotopes measured in Wadi Mistal (Cenomian) are heavier in carbon than data from the other exposures (up to Aptian). Samples from different vein sets were taken in Wadi Mistal(e.g. UTM/WGS84 0570788/2578399), Wadi Bani Awf (e.g. 0549726/2579085), Wadi Nakhar (0521913/2565767) and in one exposure on the southern slope of the dome (0542101/2559120).


100

Fig. 6. Isotopic data from early formed veins and adjacent wall rock pairs. Similarly, in both 18O and 13C are consistent with a host-rock buffered system.


101

Fig.7. Variation of stable isotopes for different vein generations (grey dots) compared to the host rock (black dots). Similarity in 18O values for most of the host rock - veins pairs are consistent with a host-rock-buffered fluid system. The system became open to external fluids only during formation of some late veins, normal faults, and thrust faults.


102

Fig.8. Isotopic variations in individual veins, measured normal (across-vein) and parallel (along-vein) to the vein wall. The very limited variation in isotopic values within the veins is consistent with the fluid remaining isotopically constant during formation of individual veins.

Discussion

A complex deformation history was responsible for the formation of these seven

generations of veins. The first phase of deformation caused vertical shortening, which

resulted in the formation of the bedding-parallel stylolites and bedding-normal veins (#1).

This probably occurred during basin subsidence when the maximum principal stress 1

was vertical to bedding and equal to the overburden stress ( 1 gz , where is the

density, g the gravitational acceleration, z the overburden). Assuming the stylolites

started forming at a depth of 1.5 to 2 km (cf. Bjørlykke 1989), the overburden would have


103

been approximately 50 MPa and effective vertical stress would have been 30 MPa, if

pore-fluid pressure was equal to hydrostatic pressure. The differential stress would have

been approximately 18 MPa if it is assumed that the effective horizontal stress, ' '3h ,

can be inferred using the coefficient of earth pressure at rest, K0, to be 0.4 and equal to '

' 0v

h

K , which takes into account the inelastic behavior of rock due to compaction

and diagenesis (Jones et al. 1991; Mandl 2000, p.179; Fig.9a).

The vertical orientation of the second generation of extensional veins (#2) is consistent

with the 1 having been vertical, similar to its orientation during formation of veins #1

and basin subsidence. The sediments hosting the second generation of veins were

cohesive since the fractures were planar and transgranular. The formation of these tensile

fractures must have required a high internal fluid pressure pf > N + T, at a low

differential stress 1 3 4T and 3 = N = -T ( N is the stress acting normal to the

fracture surface; T the tensile strength of the rock). Breton et al. (2004) estimated

maximum temperatures of about 200°C and maximum overburden pressures of 200 to

400 MPa that increased from south to north across the Jabal Akhdar dome. Assuming a

temperature gradient of 25°C/km, the maximum overburden of the limestones was about

8 km at the southern flank. Assuming K0 to be 0.4 and a tensile strength of 10 MPa for

limestone (Thuro et al. 2001), the formation of the tensile fractures would have required a

differential stress of less than 40 MPa, which would correspond to an maximum

overburden of about 4.5 km depth and a fluid pressure of 36 MPa (Fig.9b). The fluid

pressure and the differential stress change if we assume a consolidated rock, which may

be approximated as elastic material. The horizontal stress is derived as 1h v . The

differential stress increases by 5 MPa (for =0.25) and fluid overpressure will induce

shear failure rather than tensile failure. Assuming poroelastic deformation, the effective

horizontal stress decreases more slowly than the pore pressure increases, causing a

decrease in differential stress with increasing overpressure (Engelder 1994; Engelder &

Fischer 1994; Mandl 2000, p.165-188) (Fig.9c). This can change the overall failure mode

and may promote tensile over shear failure during burial of sedimentary basins at greater

depth (Hillis 2001). In isotropically and laterally confined layers the horizontal stress is


104

increased by 1 21 1h v fp , assuming zero horizontal strain (e.g. Teufel et

al. 1991; Segall 1989; Engelder 1994; Engelder & Fischer 1994; Mandl 2000, p.165-

188). The term is the Poisson ratio (0.25 dimension less), the Biot coefficient of

effective stress ( =0.5 as a strong rock, Mandl 2000, p. 171). This means that tensile

fractures may form at much lower depth than 4.5 km.

Bedding-parallel veins (#3) formed after the stylolites (#1) and vertical extension (#2)

vein. Four general groups of mechanical models have been formulated to explain the

formation of bedding-parallel horizontal veins: (i) tensile fractures form along bedding

planes due to high fluid pressures with high pore-fluid factors 1

1fp (e.g.

Henderson et al. 1990; Cosgrove 1993; Jessell et al. 1994), (ii) coalescence of dilational

jogs during progressive shear along bedding planes (e.g. Cox 1987; Koehn & Passchier

2000), and (iii) coalescence of en echelon crack arrays (Teixell et al. 2000 and references

therein). Gratier (1987) pointed out that (iv) extension veins may also form normal to the

maximum compressive stress. The fluid pressures invoked in these models vary widely.

Model (i) is similar to the formation of bedding-normal veins due to internal fluid

pressure outlined above except 1 became horizontal after formation of vein sets #1 and

#2. This resulted in tensile fractures forming parallel to the maximum principal stress.

Alternatively, dilational extension-shear (or mixed-mode) fractures may have formed

oblique to the maximum principal stress 1, if the differential stress was between 4T and

5.66T (e.g. Sibson 1998; Sibson 2000; Schultz 2000). In either case, the fluid pressure

would have been supra-hydrostatic.

Model (ii) forms at high differential stresses 1- 3>5.66T, and thus requires much lower

fluid pressures or voids passively opened in dilational jogs along a shear fracture to form

veins. It has been shown in some studies that these veins contain inclusions trails that are

parallel to the steps/ramps (Jessell et al. 1994; Koehn & Passchier 2000).

Model (iii) is a variation of model (ii) in that extensional fractures form under

compressive effective confining pressures by the alignment of conjugate en echelon

extension arrays at 1- 3>5.66T (Teixell et al. 2000) (the compressive strength C

generally approximated as 2C T , see Sibson 1998 and reference therein).


105

Model (iv) is a variation of model (i) and takes into account the variation in tensile

strength across and along bedding or foliation planes. If the variation in tensile strength

parallel and normal to bedding/foliation is larger than the differential stress 1- 3 (i.e.

3 1h vT T fractures may open normal to the maximum principal stress 1 with

1fp T (Gratier 1987; Cosgrove 1995; example given in Jolly & Lonergan 2002).

In this study, the continuous bedding-parallel veins are tens of meters long (#3.1), do not

contain significant jogs, steps, and inclusion bands that are oriented perpendicular to

bedding. Thus, vein formation by mechanism (ii) is unlikely. Vein formation by

mechanism (iii) is unlikely because the veins are parallel to bedding over long distances,

do not cut across the surrounding host rock, and do not contain host rock “bridges”. If we

thus assume that bedding parallel veins formed by supra-lithostatic fluid pressures

pf> 3+T with the maximum principal stress 1 acting horizontally, tensile failure still

requires a small differential stress of <40MPa to form #3.1. An increase of fluid pressure

would be required to form veins by this mechanism (Fig.9d). This causes a decrease in

differential stress, which at failure condition would from randomly oriented tensile

fractures (Cosgrove 1995). Such fractures have not been observed to be associated with

vein set #3.1. On the other hand, bedding-parallel veins (#3.2) and sheared pinch-and-

swell veins (#4) are consistent with simple shear during vein growth, so that the

maximum principal stress was oriented oblique to bedding. Pinch-and-swell veins (#4)

are interpreted as extension veins rotated into the extension field during progressive

shear. This may indicate that vein #3.1, #3.2 and #4 might all have formed during

compression oblique to bedding at constant fluid pressures (Fig.9e). In this case, the

differential stress first decreases due to increase of horizontal stress. After hydrostatic

stress conditions and a switch of the principal stresses the differential stress increases.

Failure can be induced at different stress conditions if fluid overpressure is increased, or

may cause shear failure at very high differential stress (black circle in Fig.9e).

The next generation of fracture sealing is represented by en echelon vein arrays, which

show the nucleation of brittle normal faults (e.g. Mazzoli & di Bucci 2003). Because

tension gashes and brittle normal faults have formed during one continuous process, we

suggest these veins formed according to model (iii) with the maximum principal stress

oriented vertical to bedding. The fluid pressure is interpreted to have dropped to almost


106

hydrostatic, consistent with the influx of fluids external to the dome during brittle normal

faulting (Fig.9f).

The last generation of veins formed in the thrusts and are associated with the final

exhumation of the Jabal Akhdar dome. The isotopic values of the veins are consistent

with external fluids infiltrating the dome through an interconnected fracture network due

to fluid pressure being slightly elevated above hydrostatic (Fig.9f).

Cause for overpressures

Overpressures may be generated by stress-related processes (e.g. disequilibrium

compaction), an increase in fluid volume (e.g., gas generation), fluid movement (e.g.,

transfer of pore pressure from overpressured rocks), and stress-insensitive diagenetic

processes (e.g., quartz cementation) (Plumley 1980; Swarbrick & Osborne 1998;

Nordgard Bolas et al. 2004). Each of these processes might have contributed to the

overpressures that occurred within the dome. In limestones, compaction and cementation

are the most likely processes to form overpressures. The Samail ophiolite and Hawasina

nappe were emplaced on the autochtonous limestones of Jabal Akhdar between late

Santonian and early Maastrichtian (Breton et al. 2004), which added a minimum of 8

kilometers to the sediments of the dome. Thermal expansion due to rapid burial of the

limestones and mineral dehydration may have contributed to fluid release. Stable isotopes

show no evidence that significant amounts of hydrocarbons contributed to overpressure

generation.


107

Next page

Fig. 9. Mohr circle evolution to generate the observed vein pattern.

a) Stylolite veins formed early during basin subsidence, when the sediments were unconsolidated or slightly consolidated. Increase in rock strength is indicated by changing color of the failure envelope from grey to black. The grey Mohr circle displays the stress conditions at a hydrostatic fluid pressure at 2 km depth. The black Mohr circle shows the conditions during tensile failure forming veins parallel to the maximum principal stress. The arrow outlines the fluid overpressure required to form tensile fractures.

b) Extension veins also formed during basin subsidence. The strength of the rock increased due to consolidation. In order to form extensional veins, the fluid pressure was supra-hydrostatic. The grey Mohr circles show the stress conditions at hydrostatic fluid pressure, the black Mohr circle the conditions at tensile failure. The corresponding arrow outlines the fluid overpressure. We plotted two Mohr circles at hydrostatic conditions (grey) to display the difference in differential stress for elastic and inelastic stress conditions (using a Possion ratio of =0.25 and K0=0.4,respectively). The differential stress of 40 MPa is governed by the tensile strength of limestone. Once fractured, the fluid will drain and seal the fluid pathway, causing new overpressures.

c) An increase of fluid pressure may cause a decrease in differential stress, if the deformation mechanism is poroelastic. This may cause tensile failure in settings where the total stresses generally would cause shear failure. However, this means that fluid overpressures in a porous rock was larger than expected at elastic conditions. A leakage of the seal would release the overpressure, increase differential stress, and ultimately cause shear failure.

d) Bedding-parallel veins formed in a compressional setting. We consider two models: (i) The fluid pressure increased during re-orientation of the principal stress, (ii) the fluid pressure remained constant during reorientation of the principal stress. These are exemplified for hybrid extension shear (mixed mode) failure. Model (i) describes the switch of the principal stresses and formation of bedding-parallel veins while fluid pressure increases. An increase in fluid pressure caused a reduction of the effective vertical stress ’v until the differential stress became zero.

e) Model (ii) forces the differential stress to decrease until the horizontal stress equals the vertical stress at constant fluid pressure. Then the differential stress increased until shear failure fractured the rock (black Mohr circle). Fluid overpressures may have caused failure at lower differential stresses.

f) Veins associated with normal faults and thrusts formed at high differential stresses during shear failure. The graph sketches the stress conditions at 8 km depth for elastic and inelastic stress conditions during compression, and displays the amount of overpressure required to cause failure. Our data do not allow to determine the actual depth during shear failure.


108


109

Relation to regional geology

The vein orientations are consistent with at least two rotations of the maximum principal

stress from normal to bedding (#1, #2, #5, #6) to oblique to bedding (esp. #3.2, #4, #7).

Several deformation phases have been observed in the Mesozoic platform of the Jabal

Akhdar. Early conjugate normal faults trend WNW-ESE to NW-SE and die out in the

Muti formation or are sealed at its base (Rabu 1988; Boote et al. 1990). They mark a

transtensional regime with maximal extension striking NNE-SSW, potentially due to the

pulling down of continental lithosphere during subduction of autochthonous

Infracambrian basement and Meozoic cover in Turonian-Santonian time (Breton et al.

2004). We suggest that bedding-normal veins (#1 and #2) formed in this period during

burial of the sediments.

This phase of deformation was followed by a regional cleavage formation in shaly

lithologies, which evolved from a fracture cleavage on the southern flank into a

penetrative slaty cleavage on the northern flank. Cleavage formation was coeval with

bedding-parallel sliding with subtractive and additive ramps (Breton 2002), due to top-to-

the-NNE shearing during the Campanian and Early Maastrichtian. This resulted from

exhumation of the more buoyant subducted autochthonous Infracambrian basement and

Mesozoic cover in a compressive regime (Breton et al. 2004). This deformation phase

probably produced the bedding-parallel (#3) and pinch-and-swell veins (#4), of which

#3.2 displays the regional top-to-the NNE shear. In contrast, Breton et al. (2004)

described some normal faults that formed prior to regional bedding-parallel slip in Wadi

Mistal. They correlated this slip to the formation of a regional cleavage. Consequently,

either (i) our normal faults (#5 and #6) formed after regional cleavage formation because

they are not off-set by bedding parallel slip in Wadi Nakhar, (ii) different generations of

normal faults formed prior and after regional cleavage formation, (iii) the offset of

normal faults by bedding-parallel slip in Wadi Mistal is not related to the regional top-to-

the NNE shear, but to a later slip system (e.g. flexural slip folding of the dome), or (iv)

bedding-parallel veins in Wadi Nakhar are not related to bedding-parallel slip.

All these deformation events are related to the Eoalpine geodynamic evolution of the

northern Oman Mountains. In contrast, the relative timing of veins that formed after the

cleavage is unclear because the Tertiary cover is completely absent in Jabal Akhdar. Only


110

a local relative chronology was established, which is tentatively related to the regional

events by Breton (2002).

The sub-vertical fracture cleavage in the Jabal Nakhl and the Rustaq area might have

formed during transpression with compression striking NW-SE, possibly coeval to

formation of Jabal Nakhl anticline (Breton 2002). The necessary overburden for the

formation of the cleavage indicates that this phase of deformation occurred before erosion

of the allochthonous units and occurred during early Paleocene NW-SE compression

(Gnos et al. 1997). This phase was followed by new NW-SE striking normal faults and

dextral movement on the early WNW-ESE trending normal faults on the southern flank

(Breton 2002).

Reverse faults with W-E trending grooves mark the reactivation of normal faults and can

be linked with the Miocene (Burdigalian) W-E transpression that triggered the uplift of

the Oman mountain chain and deposition of a molasse (Barzaman formation) (Breton

2002). Reverse faults and additive thrusts with NNE-SSW trending grooves mark another

reactivation associated with tightening of the previous NNE-SSW faults and the doming

of the central part of Jabal Akhdar along a WSW- ENE axis. The folding of the northern

and southern limbs seems to be related to the NNE-SSW Plio-Quaternary transpression

(Breton 2002). Subhorizontal grooves on veins (#6) are consistent with reactivation of

normal faults, but our there is no clear sequence of deformation events in our data set.

The thrust related veins (#7) probably formed during doming of Jabal Akhdar.

Thus, the different vein sets indicate several changes in the stress regime and may be

associated with different regional deformation phases. According to the region geology

outlined above, early vein sets can be related to transtension during subsidence (#1),

subduction (#2) and compressional shear during exhumation (#3, #4), while later veins

may represent a transtension (#5,6) and transpression (#7) phase.

Conclusion

The Jabal Akhdar dome contains multiple calcite vein sets, indicating an extension phase

with stylolite (#1) and subvertical veins (#2) having formed at high overpressures. This

phase was followed by a phase with compression oblique to bedding at high fluid

pressures, resulting in the formation of bedding parallel veins (#3.1, #3.2). Pinch-and-


111

swell veins record a complex deformation history that involved both calcite and quartz

cementation (#4) formed during progressive simple shear. The stable isotope data of

calcite ( 13C and 18O) show that the early veins formed in a rock-buffered environment.

The following extension phase is displayed as en echelon vein arrays and normal faults.

During normal faulting, the overpressure cell drained and allowed infiltration of external

fluids. The minimum principal stress then rotated from vertical to sub-horizontal to

bedding, resulting in the formation of thrusts and associated veins during doming of Jabal

Akhdar.

Further work will include a more systematic study of the different vein generations and

the heterogeneity of veins associated with faults along-strike, and will better constrain the

p-T conditions during vein formation.

Acknowledgments

One of us (CH) would like to thank the German Academic Exchange Service DAAD and

the German Research Council (project no. Hi-816/2) for funding parts of this project.

Stimulating discussions with Mohammed Al-Wardi (University of Leeds) are greatly

acknowledged.

Appendix

Tab. 2. Stable isotope data derived from wall rock (w) and veins (v). The different vein types were classified according to the relative timing. Runs with high deviation from the in-house standard were re-measured and the data of the different runs averaged.

18O 13C outcrop # sample # rock type vein type run date location

25.47 3.38 1 a v normal fault 41013 Wadi Mistal

26.20 3.36 2 v normal fault averaged data Wadi Mistal

25.89 3.57 2 w averaged data Wadi Mistal


25.76 3.51 4 1 v boudin vein 41014 Wadi Mistal

26.04 3.61 4 v boudin vein averaged data Wadi Mistal

25.74 3.41 4 w 41014 Wadi Mistal

25.66 3.59 5 2 v thrust 41015 Wadi Mistal




112


25.31 3.33 5 v thrust averaged data Wadi Mistal



26.14 3.53 6 v parallel S0 41013 Wadi Mistal


25.46 3.53 8 c v tension gash 41014 Wadi Mistal

25.48 3.43 8 a v tension gash 41014-2 Wadi Mistal

26.07 3.45 8 d v cleavage parallel (?) 41014-2 Wadi Mistal

25.79 3.51 8 b v tension gash 41014-2 Wadi Mistal

25.25 3.03 8 c w 41014-2 Wadi Mistal

16.23 1.66 9 a v thrust pull apart averaged data Wadi Bani Awf

19.38 1.87 9 b v thrust pull apart averaged data Wadi Bani Awf

27.55 2.22 9 b w 41014 Wadi Bani Awf

27.96 2.22 9 a w 41014-2 Wadi Bani Awf

27.93 2.00 9 a w 41020 Wadi Bani Awf

27.57 2.02 9 b w 41020 Wadi Bani Awf

26.91 2.01 11 v boudin vein 41014 Wadi Bani Awf

27.50 2.00 11 v paralllel S0 41014 Wadi Bani Awf

27.80 1.96 11 2 v boudin vein 41016 Wadi Bani Awf




27.75 2.34 11 w 41014 Wadi Bani Awf

27.45 1.90 11 w 41014-2 Wadi Bani Awf

24.28 1.88 12 v normal S0 41014 Wadi Bani Awf

27.14 0.75 13 v parallel S0 averaged data Wadi Bani Awf

24.64 3.20 14 v normal S0 averaged data southern slope

25.05 3.30 14 w 41014 southern slope

25.32 3.08 15 v oblique to S0 averaged data southern slope

25.33 2.00 15 2 v oblique to S0 averaged data southern slope

27.63 2.68 16 2 v pinch & swell 41014 Wadi Nakhar

27.78 2.67 17 v normal S0 vein 41014 Wadi Nakhar

25.87 3.62 18 v - 41013 Wadi Nakhar

27.24 2.66 18 v - 41013 Wadi Nakhar

28.00 2.63 18 8 v - 41015 Wadi Nakhar


113

27.92 2.78 18 5 v - 41015 Wadi Nakhar

27.43 2.43 18 7 v - 41015 Wadi Nakhar

27.77 2.66 18 3 v - 41015 Wadi Nakhar

27.68 2.73 18 6 v - 41015 Wadi Nakhar

27.52 2.55 18 4 v - 41015 Wadi Nakhar

28.12 2.49 18 2 v - 41016 Wadi Nakhar

27.79 2.62 18 1 v - 41016 Wadi Nakhar

27.86 2.24 18 w 41013 Wadi Nakhar

26.85 2.03 19 v normal S0 41013 Wadi Nakhar

27.73 2.81 19 w 41014 Wadi Nakhar

28.07 2.71 20 v normal fault 41014 Wadi Nakhar

27.28 1.57 21 1 v normal fault 41026 Wadi Nakhar

27.43 1.58 21 2 v normal fault averaged data Wadi Nakhar

27.67 1.65 21 3 v normal fault averaged data Wadi Nakhar

27.38 1.61 21 v normal fault averaged data Wadi Nakhar

27.58 1.66 22 v normal fault 41014-2 Wadi Nakhar

27.62 1.80 23 v normal S0 averaged data Wadi Nakhar

27.45 1.46 23 w 41014-2 Wadi Nakhar

26.50 1.72 24 a v

normal S0 fault

breccia 41014-2 Wadi Nakhar

27.45 1.95 25 v pinch & swell 41013 Wadi Nakhar




26.17 2.68 29 v oblique to S0 averaged data Wadi Nakhar

21.96 1.36 29 2 v oblique to S0 averaged data Wadi Nakhar

25.41 2.30 29 w 41014-2 Wadi Nakhar

22.22 1.85 30 1 v normal S0 breccia 41016 Wadi Nakhar

24.09 1.98 30 2 v normal S0 breccia 41016 Wadi Nakhar

24.26 2.03 30 v normal S0 breccia averaged data Wadi Nakhar

25.81 2.22 30 w averaged data Wadi Nakhar

18.89 1.72 31 - averaged data Wadi Mistal

26.40 1.33 31 1 v - 41014 Wadi Nakhar

26.47 0.89 31 v - 41020 Wadi Nakhar

25.03 1.63 33 w 41013 Wadi Nakhar



114

25.93 1.90 35 w 41014 Wadi Nakhar

27.65 1.83 36 v normal S0 41015 Wadi Nakhar

27.06 0.17 36 v - 41020 Wadi Nakhar

26.92 0.06 36 2 v normal S0 vein averaged data Wadi Nakhar


27.59 2.04 38 v - 41013 Wadi Nakhar

28.03 1.94 38 w - 41013 Wadi Nakhar



27.12 1.74 41 w 41014-2 Wadi Nakhar

27.35 1.38 42 v parallel S0 41015 Wadi Nakhar


25.41 1.26 43 w 41014-2 Wadi Nakhar

27.25 1.55 43 w 41015 Wadi Nakhar


25.11 2.08 45 3 v tension gash 41016 Wadi Nakhar




24.54 2.32 45 v tension gash averaged data Wadi Nakhar

25.48 2.17 45 w 41014 Wadi Nakhar

27.81 1.71 46 v paralllel S0 41014 Wadi Nakhar


28.04 1.22 47 v normal S0 stylolite 41014 Wadi Nakhar

28.35 2.06 47 w 41015 Wadi Nakhar

26.38 1.73 48 v limestone 41013 Wadi Nakhar


28.05 2.10 49 3 v normal S0 vein A Wadi Nakhar



29.23 2.87 49 v normal S0 vein A averaged data Wadi Nakhar


26.56 2.18 50 v normal S0 vein B 41013 Wadi Nakhar

27.75 2.24 50 w 41013 Wadi Nakhar


115

References

Bechennec, F, le Metour, J, Platel, JP (1993) Geological Map of Oman, scale 1:1,000,000Ministry of Petroleum and Minerals, Sultanate of Oman.

Bechennec, F, le Metour, J, Rabu, D, Villey, M, Beurrier, M (1988) The Hawasina basin: A fragment of a starved passive continental margin, thrust over the Arabian platform during obduction of the Sumail nappe. Tectonophysics, 151, 323-343.

Beurrier, M (1988) Géologie de la Nappe ophiolitique de Samail dans les parties orientale et centrale des Montagnes d'Oman. PhD Thesis, 412 pp.

Beurrier, M et al. (1989) Géochimie des filons doléritiques et des roches volcaniques ophiolitiques de la nappe. Bull Soc Geol France, 8(2), 205-219.

Bjørlykke, K (1989) Sandstone diagenesis and porosity modification during basin evolution. Geologische Rundschau, 78(1), 243-268.

Boote, DRD, Mou, D, Waite, RI (1990) Structural evolution of the Suneinah foreland, Central Oman Mountains. In: The Geology and Tectonics of the Oman Region.(eds) A.H.F. Robertson, M.P. Searle and A.C. Ries, pp. 397-418. Geological Society, London.

Boudier, F, Ceuleneer, G, Nicolas, A (1988) Shear zones, thrusts and related magamtism in the Oman ophiolite: Initiation of thrtusting on an oceanic ridge. Tectonophysics, 151, 275-296.

Boudier, F et al. (1985) Kinematic of oceanic thrusting in the Oman Ophiolite: model for plate convergence. Earth and Planetary Science Letters, 75, 215-222.

Breton, JP (2002) Structural field study of Jabal Akhdar, Sultanate of Oman. CDG/CG/2002/128, BRGM.

Breton, JP, Bechenne, F, le Metour, J, Moen-Maurel, L, Razin, P (2004) Eoalpine (Cretaceous) evolution of the Oman Tethyan continental marin: Insights from a structural field study in Jabal Akdhar (Oman mountains). GeoArabia, 9(2), 1-18.

Burkhard, M, Kerrich, R (1988) Fluid regimes in the deformation of the Helvetic nappes, Switzerland, as inferred from stable isotope data. Contributions to Mineralogy and Petrology, 99, 416-429.

Coplen, TB, Kendall, C, Hopple, J (1983) Comparison of stable isotope reference samples. Nature, 302(5905), 236-238.

Cosgrove, JW (1993) The interplay between fluids, folds and thrusts during the deformation of a sedimentary succession. Journal of Structural Geology, 15(3-5),491-500.

Cosgrove, JW (1995) The expression of hydraulic fracturing in rocks and sediments. In: Fractography: Fracture topography as a tool in fracture mechanics and stress analysis. (ed) M.S. Ameen, pp. 187-196. Geological Society, London.

Cox, SF (1987) Antitaxial crack-seal vein microstuctures and their relationship to displacement paths. Journal of Structural Geology, 9(7), 779-787.

Engelder, T (1994) Brittle crack propagation. In: Continental deformation. (ed) P.L. Hancock, pp. 43-52. Pergamon Press Ltd, Oxford.

Engelder, T, Fischer, MP (1994) Influence of poroelastic behavior on the magnitude of minimum horizontal stress, Sh, in overpressured parts of sedimentary basins. Geology, 22, 949-952.


116

Ghisetti, F, Kirschner, DL, Vezzani, L, Agosta, F (2001) Stable isotope evidence for contrasting paleofluid circulation in thrust faults and normal faults of the central Apennines, Italy. Journal of Geophysical Research, 106(B5), 8811-8825.

Glennie, KW (1995) The geology of the Oman mountains. 92 pp. Scientific Press Ltd. Glennie, KW et al. (1973) Late Cretaceous nappes in Oman mountains and their

geological evolution. AAPG Bulletin, 75(1), 5-27. Glennie, KW, Hughes-Clarke, MW, Boeuf, MG, Pilaar, WF, Reinhardt, BM (1974)

Geology of the Oman mountains. Verhandelingen Koninklijk Nederlands geologisch mijnbouwkundig Genootschap, 31, 1-423.

Gnos, E, Immenhauer, A, Peters, T (1997) Late Cretaceous/early Tertiary convergence between the Indian and Arabian plates recorded in ophiolites and related sediments. Tectonophysics, 271, 1-19.

Gratier, JP (Editor), 1987. Pressure solution-deposition and associated tectonic differentiation in sedimentary rocks. Geological Society Special Publication, No.29. Geological Society, London, 25-38 pp.

Gray, DR, Gregory, RT, McL. Miller, J (2000) A new structural profile along the Muscat-Ibra transect, Oman: Implications for emplacement of the Samail ophiolite. In: Ophiolites and oceanic crust: New insights from field studies and the ocean drilling program. (eds) Y. Dilek, E.M. Moores, D. Elthon and A. Nicolas. Geological Society of America Special Paper, pp. 513-523. Geological Society of America, Boulder.

Guzzetta, G (1984) Kinematics of stylolite formation and physics of the pressure-solution process. Tectonophysics, 101, 383-394.

Hacker, BR (1994) Rapid emplacement of young oceanic lithosphere: Argon geochronology of the Oman ophiolite. Science, 265, 1536-1565.

Hacker, BR, Mosenfelder, JL, Gnos, E (1996) Rapid emplacement of the Oman ophiolite: Thermal and geochronological constraints. Tectonics, 15(6), 1230-1247.

Hanna, SS (1990) The Alpine deformation of the Central Oman mountains. In: Geological Society Special Publication. (eds) A.H.F. Robertson, M.P. Searle and A.C. Ries, pp. 341-359. Geological Society, London.

Henderson, JR, Henderson, MN, Wright, TO (1990) Water-sill hypothesis for the origin of certain veins in the Meguma Group, Nova Scotia, Canada. Geology, 18, 654-657.

Hillgärtner, H et al. (2003) The Barremian-Aptian evolution of the eastern Arabian carboante platform margin (northern Oman). Journal of Sedimentary Research,73(5), 756-773.

Hillis, RR (2001) Coupled changes in pore pressure and stress in oil fields and sedimentary basins. Petroleum Geoscience, 7, 419-425.

Jenkyns, HC, Wilson, PA (1999) Stratigraphy, paleoceanography, and evolution of Cretaceous Pacific guyots: Relics from a greenhouse Earth. American Journal of Science, 299(5), 341-392.

Jessell, MW, Willman, CE, Gray, DR (1994) Bedding parallel veins and their relationship to folding. Journal of Structural Geology, 16(6), 753-767.

Jolly, RJH, Lonergan, L (2002) Mechanism and controls on the formation of sand intrusions. Journal of the Geological Society, 159, 605-617.


117

Jones, ME, Leddra, MJ, Goldsmith, AS, Yassir, N (1991) Mechanisms of compaction and flow in porous sedimentary rocks. In: Neotectonics and resources. (eds) J.W. Cosgrove and M. Jones, pp. 16-42. Belhaven Press

Koehn, D, Passchier, CW (2000) Shear sense indicators in striped bedding-veins. Journalof Structural Geology, 22, 1141-1151.

Konert, G, Afifi, AM, Al-Hajri, SA, Droste, H (2001) Paleozoic stratigraphy and hydrocarbon habitat of the Arabian plate. GeoArabia, 6(3), 407-442.

le Métour, J (1988) Géologie del'autochtone des Montagnes d'Oman: la fenêtre du Saih Hatat. PhD Thesis, Pierre and Marie Curie University, Paris, 420 pp.

Mandl, G (2000) Faulting in brittle rocks. 434 pp. Springer, London. Mann, A, Hanna, SS, Nolan, SC (1990) The post-Campanian evolution of the Central

Oman Mountains: Tertiary extension of the Eastern Arabian margin. In: Thegeology and tectonics of the Oman region. (eds) A.H.F. Robertson, M.P. Searle and A.C. Ries, pp. 549-563. Geological Society, London.

Masse, J-P, Borgomano, J, Al Maskiry, S (1997) Stratigraphy and tectonosedimentary evolution of a late Aptian-Albian carboante margin: the northeastern Akdhar (Sultanate of Oman). Sedimentary Geology, 113, 269-280.

Masse, J-P, Borgomano, J, Al Maskiry, S (1998) A platform-to-basin transition for lower Aptian carbonates (Shuaiba Formation) of the northeastern Jebel Akhdar (Sultanate of Oman). Sedimentary Geology, 119, 297-309.

Mazzoli, S, di Bucci, D (2003) Critical displacement for normal fault nucleation from en-echelon vein arrays in limestones: a case study from the southern Apennines (Italy). Journal of Structural Geology, 25, 1011-1020.

McL. Miller, J, Gray, DR, Gregory, RT (2002) Geometry and significance of internal windows and regional isoclinal folds in northeast Saih Hatat, Sultanate of Oman. Journal of Structural Geology, 24, 359-386.

Nicolas, A, Ceuleneer, G, Boudier, F, Misseri, M (1988) Structural mapping in the Oman ophiolites: Mantle diapirism along an oceanic ridge. Tectonophysics, 151, 27-56.

Nordgard Bolas, HM, Hermanrud, C, Teige, GMG (2004) Origin of overpressures in shales: Constriants form basin modeling. AAPG Bulletin, 88(2), 193-211.

Plumley, WJ (1980) Abnormally high fluid pressure: Survey of some basic principles. AAPG Bulletin, 64(3), 414-430.

Poupeau, G, Saddiai, O, Michard, A, Goffe, B, Oberhänsli, R (1998) Late thermal evolution of the Oman Mountains subophiolitic windows: Apatite fission-track thermochronology. Geology, 26(12), 1139-1142.

Pratt, BR, Smewing, JD (1993) Early Cretaceous platform-margin configuration and evolution in the central Oman mountains, Arabian peninsula. AAPG Bulletin,77(2), 225-244.

Rabu, D (1988) Géologie de l'autochtone des Montagnes d'Oman la fenêtre du Jabal Akdhar. La semelle métamorphique de la Nappe ophiolitique de Semail dans les parties orientale et centrale des Montagnes d'Oman: une revue, BRGM Orléans.

Renard, M (1986) Pelagic carbonate chemostratigraphy (Sr, Mg, 18O, 13C). MarineMicropaleontology, 10, 117-164.

Robertson, AHF (1987) Upper Cretaceous Muti formation: transition of a Mesozoic nate plateform to a foreland basin in the Oman mountains. Sedimentology, 34, 1123-1142.


118

Scholle, PA, Arthur, MA (1980) Carbon isotope fluctuations in Cretaceous pleagic limestones: Potential stratigraphic and petroleum exploration tool. AAPG Bulletin,64(1), 67-87.

Schultz, RA (2000) Growth of geologic fractures into large-strain populations: review of nomenclature, subcritical crack growth, and some implications for rock engineering. International Journal of Rock Mechanics and Mining Sciences, 37,403-411.

Searle, MP (1985) Sequence of thrusting and origin of culminations in the northern and central Oman Mountains. Journal of Structural Geology, 7(2), 129-143.

Searle, MP, Warren, CJ, Waters, DJ, Parrish, RR (2004) Structural evolution, metamorphism and restoration of the Arabian continental margin, Saih Hatat region, Oman Mountains. Journal of Structural Geology, 26, 451-473.

Segall, P (1989) Earthquakes triggered by fluid extraction. Geology, 17, 942-946. Sibson, RH (1998) Brittle failure mode plots for compressional and extensional tectonic

regimes. Journal of Structural Geology, 20(5), 65-660. Sibson, RH (2000) Fluid involvement in normal faulting. Journal of Geodynamics, 29,

469-499.Skelton, PW (Editor), 2003. The Cretaceous world. Cambridge University Press,

Cambridge, 360 pp. Swarbrick, RE, Osborne, MJ (1998) Mechanisms that generate abnormal pressures: an

overview. In: Abnormal pressures in hydrocarbon environments. (eds) B.E. Law, G.F. Ulmishek and V.I. Slavin, pp. 13-33. AAPG MEMOIR

Teixell, A, Durney, DW, Arboleya, M-L (2000) Stress and fuid control on décollement within competent limestone. Journal of Structural Geology, 22, 349-371.

Teufel, LW, Rhett, DW, Farrell, HE (1991) Effect of reservoir depletion and pore pressure drawdown on in situ stress and deformation in the Ekofisk Field, North Sea. In: Rock Mechanics as multidisciplinary science. (ed) J.C. Roegiers, pp. 63-72. Balkema, Rotterdam.

Thuro, K, Plinninger, RJ, Zäh, S, Schütz, S (2001) Scale effects in rock strength properties. Part I: Unconfined compressive test and Brazilian test. In: Rockmechanics - a challenge for society (Editors) P. Särkka and P. Eloranta. Balkema, Rotterdam, Espoo, Finland, pp. 169-174.

Tilton, GR, Hopson, CA, Wright, JE (1981) Uranium-lead isotopic ages of the Samail ophiolite, Oman, with application to Tethyan ridge tectonics. Journal of Geophysical Research, 86, 2736-2776.

Warburton, J, Burnhill, TJ, Graham, RH, Isaac, KP (1990) The evolution of the Oman Mountains foreland Basin. In: The Geology and Tectonics of the Oman Region.(eds) A.H.F. Robertson, M.P. Searle and A.C. Ries, pp. 419-427. Geological Society, London.

Weissert, H, Lini, A, Föllmi, KB, Kuhn, O (1998) Correlation of Early Cretaceous carbon isotope stratigraphy and platform drowning events: a possible link? Palaeogeography, Palaeoclimatology, Palaeoecology, 137, 189-203.

aspects of material transport & polycrystal growth in ... habil-th… · geologie-endogene...

Documents