Geological Survey of Denmark and Greenland Bulletin 26, 2012, 9-12 9 Nano-quartz in North Sea Danian chalk Holger Lindgreen and Finn Jakobsen The main oil reservoir in the Central Graben in the North Sea is chalk of the Maastrichtian Tor Formation, which has high porosity and relatively high permeability. The chalk of the Danian Ekofisk Formation is an additional reservoir, but with highly variable porosity and permeability. Whereas the Tor Formation is almost pure calcite primarily consisting of coccolith debris, the Ekofisk Formation also comprises sig- nificant proportions of phyllosilicates (clay minerals) and quartz in addition to coccolith debris. For decades the quartz was assumed to be a normal crystalline α-quartz such as is present in quartz sand, and the clay fraction was assumed to consist predominantly of phyllosilicates. However, Maliva & Dickson (1992) reported the presence of presumably au- thigenic submicron-size quartz crystals arranged in clusters, and suggested that these clusters were transformed opal-CT lepispheres. Investigations by nano-structural methods (X- ray diffraction and atomic force microscopy (AFM)) revealed that the prevailing quartz component in the North Sea chalk comprises α-quartz appearing as nano-size quartz spheres ( Jakobsen et al. 2000; Lindgreen et al. 2010). Nano-quartz spheres were first observed in indurated chalk in the Ekofisk Formation in the Ekofisk Field and later in the South Arne Field. Subsequent analyses of the Ekofisk Formation in dif- ferent chalk fields showed that the content of nano-quartz varies throughout the chalk succession and to some degree ref lects the cyclic development of the chalk. The proportion of dispersed nano-quartz in the chalk is highly variable, from 10% to more than 80% in the Lower Danian (Lindgreen et al. 2010). This paper describes the nano-quartz, its forma- tion and structure and presents a model for the formation of f lint from nano-quartz in the North Sea Ekofisk chalk. Material and methods We have investigated core samples from the Ekofisk Forma- tion in the South Arne Field wells SA-1 and Rigs-1, in the Halfdan Field wells Sif-1 and Nana-1 and in the Gorm Field well N-22 (Fig. 1). Most samples contained large amounts of calcite, so calcite-free residues were prepared by dissolving the calcite in an acetate-acetic acid buffer at pH 4.5–5. In this buffer, non-calcite minerals and especially fine-grained nano-quartz and clay minerals are not corroded or dissolved. Scanning electron microscopy (SEM) is routinely used for investigations of chalk minerals and in special cases micron- sized particles can be identified (Hjuler & Fabricius 2009). However, rock samples dominated by nano-sized quartz are at the limit of resolution in the SEM and generally give poor SEM images due to poor current transmission in the fine-grained matrix. We used X-ray diffraction and AFM to characterise the ultra-fine particles in the chalk, such as nano-quartz and clay minerals. X-ray diffraction scanning using 10 s/0.1 °2Θ was ap- plied routinely to determine the mineralogical composition of both core piece samples and of non-calcite residues. High statistic scanning using 100 s/0.02 °2Θ was used to charac- terise the nano-quartz. AFM (Binnig et al. 1986) generates topographic images by scanning a sharp tip across a surface and can produce images at atomic resolution of both conductors and non- conductors. For AFM we used a Rasterscope 3000 instru- ment under room conditions with a force of 0.175 nN and a scanning speed of 500 nm/s. In the present investigation of the topography of raw surfaces, AFM was run in non-contact R in gkø b in g– Fyn H igh Gas in chalk Oil in chalk Field at other level 4°E 5°E 56°N C entral G raben 25 km National border Rigs-1 SA-1 Sif-1 Nana-1 N-22 UK NL N DK 200 km G Fig. 1. Map of the Danish Central Graben showing the locations of the investigated wells. © 2012 GEUS. Geological Survey of Denmark and Greenland Bulletin 26, 9–12 . Open access: www.geus.dk/publications/bull 1010 mode. Intact rock samples of small pieces of chalk or f lint were glued onto gold-coated sample holders. In chalk sam- ples, non-calcite minerals were identified and imaged from the insoluble residue. For such samples, the residue was dis- persed ultrasonically in distilled water and the samples pre- pared by leaving a drop of the suspension to dry under room conditions on a block of highly oriented pyrolytic graphite. Structure of the nano-quartz particles AFM of non-calcite residues deposited on graphite showed that the nano-quartz consists predominantly of rather uni- form, c. 600 Å large spherical particles (Fig. 2A). AFM im- ages of intact f lint surfaces showed that the f lint consists of similar spherical particles with a diameter of c. 500 Å or more (Fig. 2B) and some irregularly shaped particles. X-ray diffrac- tion showed that the non-calcite residues and the f lint and quartz layers are composed of α-quartz having practically identical patterns and resembling the pattern of standard quartz (Fig. 3). It is remarkable that the quartz in all the ex- amined samples of dispersed quartz and f lint have almost identical unit cell a and c parameters and sizes of coherent scattering domains (Lindgreen et al. 2011). At high angles peak broadening was pronounced for the nano-quartz particles and careful recording revealed a broad and distorted pattern of the (212), (203) and (301) re- f lections compared to the ref lection from standard quartz (Fig. 4). These distortions are due to larger a and c parameters compared to those of normal quartz (Lindgreen et al. 2011). The nano-quartz spheres had colloidal properties and f loc- culated in suspensions with sufficient ionic strength, such as sea water (Fig. 5). Formation of quartz particles The nano-quartz spheres are anticipated to be of a type that might crystallise in a marine environment which is slightly enriched in silicon (Williams & Crerar 1985). The source of silicon was probably opal-A from radiolarians, which were the main silica-bearing organism in the chalk sea (Maliva & Dickson 1992). It is important that the non-crystalline Si in radiolarians will dissolve at the low concentration of Si, which is sufficient to precipitate fine quartz, and that the quartz will be the first silica phase to crystallise. A B 200 nm200 nm Standard Merck quartz SA 3344.15 m quartz in matrix Nana 2135.7 m quartz in flint 80 81 °2Θ 82 2 1 2 α 1 2 0 3 α 1 2 0 3 α 2 3 0 1 α 1 3 0 1 α 2 2 1 2 α 2 A B C Fig. 3. A: X-ray diffraction pattern of standard Merck quartz. B: of calcite- free residue from chalk, South Arne Field, well SA-1, 3344.15 m, C: and of f lint layer in Halfdan Field, well Nana-1, 2135.7 m. Co-K α radiation, 5% Si added as internal standard. Fig. 4. X-ray diffraction patterns. A: Region of (212), (203) and (301) re- f lections of standard Merck quartz. B: The same region for calcite-free residue from chalk, South Arne Field, well SA-1, 3344.15 m. C: The same region for f lint layer in Halfdan Field, well Nana-1, 2135.7 m. Co-K α ra- diation. Fig. 2. Atomic force microscopy images of spherical grains of quartz. Non- contact mode, room conditions, force 0.175 nN, scanning speed 500 nm/s. A: Calcite-free residue deposited on graphite from well SA-1, 3344.15 m. B: Intact f lint from Nana-1, 2135.7 m. 20 30 40 50 60 70 80 90 Standard Merck quartz SA 3344.15 m quartz in matrix Nana 2135.7 m quartz in flint Si Si Si °2Θ A B C 11 Data from Williams et al. (1985) indicate that the c. 500 Å diameter quartz spheres observed in the chalk and in the f lint of the North Sea Danian chalk can form at SiO2 con- centrations of c. 12 ppm. The North Sea chalk is a deep water deposit and present-day deep sea water has a concentration of 1–10 ppm SiO2 (Millot 1970; Calvert 1974). We think that only a minor increase in Si concentration would result in crystallisation of nano-quartz spheres. The colloidal quartz spheres could then have f locculated and been deposited on the sea f loor mixed with coccolith ooze. Flocculation is im- portant for sedimentation of silica and the rate of sedimenta- tion for the formation of layers rich in quartz. Chemical environment in the water column As described above, we assume that silica was not deposited as biogenic opal-A. Therefore the variation in proportion of nano-quartz cannot be caused by changes in the supply of silicon to the sea as such changes would be ref lected in changes in size and mineralogy of the silica. An alternative is variation in the sedimentation of coccoliths. Such variation may be due to a decrease in pH which may cause coccoliths to be partly or totally dissolved in the water column. Such a decrease in pH requires significant amounts of an acidifier. This acidifying agent was most probably atmo- spheric CO2 , which by mixing with sea water has been found to decrease the calcification of marine plankton (Riebesell et al. 2000; Feely et al. 2004). CO2 released in large quan- tities during volcanic eruptions (Holmes 1965; Zimmer & Erzinger 2003; Frondini et al. 2004; Schuiling 2004; Self et al. 2006) could be a cause of the dissolution of the coccoliths in parts of the Danian chalk deposits in the North Sea. Sensitivity analysis has indicated that only massive and short-lived volcanism could cause the CaCO3 undersatura- tion of seawater (Berner & Beerling 2007). Age determi- nations of lavas from the British Tertiary igneous province have yielded ages of 63–65 Ma (Saunders et al. 1997), cor- responding to a Danian age. We therefore propose that the pronounced quartz enrichment in the Danian chalk of the North Sea was associated with frequent volcanic eruptions in this period at and after the Cretaceous–Tertiary boundary. Our model implicates that the degree of dissolution of the coccoliths in the sedimentary environment determines the proportion between calcite and nano-quartz in the chalk. Theories for flint formation The new theory for the formation of f lint and dispersed na- no-quartz in the North Sea by crystallisation of nano-quartz in the marine environment is totally different from the gen- erally accepted theory for f lint formation in chalk based on studies of chalk from onshore outcrops (Bromley & Ekdale 1986; Clayton 1986; Zijlstra 1987; Madsen & Stemmerik 2010). According to the current theory for f lint formation, opaline tests and sponge spicules in the sediment are dis- solved during burial and the Si is recrystallised as opal-CT and quartz in hollows and by replacement of calcite. How- ever, the generally accepted theory does not agree with our results obtained for the silica in the investigated North Sea chalk. We support our point of view by observing that the type of α-quartz dispersed in chalk is identical with the type constituting the f lint nodules and f lint horizons and with the type constituting the α-quartz horizons in the chalk. We find it highly unlikely that the same size and shape of particles will crystallise in the chalk and in the hollows dur- ing dissolution or reprecipitation, whereas the particles are of the type which can possibly crystallise in the marine en- vironment that is slightly enriched in silicon (Williams & Crerar 1985). A sedimentary origin of the silica-rich chalk layers is supported by the presence of a f lint bed in well N-22. The f lint layer includes a calcite-filled burrow within Fig. 5. Flocculation of nano-quartz particles. Residue from well SA-1, 3353.0 m. Left: quartz dispersed in distilled water. Right: quartz dispersed in 0.2 M CaCl2 . Dispersed in distilled water Dispersed in 0.2M CaCl2 1212 a matrix comprising nano-quartz spheres. The occurrence of a burrow in rather pure α-quartz sediment shows that the α-quartz was soft when biological activity took place. A sedimentary origin of the f lint fits well with our results for the North Sea Tertiary chalk, which is a deposit in rela- tively deep water. However, it cannot be generally applied to other areas and deposits in different settings without further investigations. Conclusions We have proposed a new model for the formation of f lint in North Sea chalk: (1) The nano-quartz in the f lint, like the nano-quartz in the chalk matrix, has crystallised in the ma- rine chalk-sea environment. The colloidal quartz particles have then f locculated and have been deposited on the sea f loor mixed with coccolith ooze. (2) Regional variations in the concentration of nano-quartz particles in the sediment ref lect different degrees of acidification of the chalk sea. (3) This resulted in areas with a high concentration of nano- quartz spheres that could form f lint layers. In areas with lower concentration of nano-quartz spheres, indurated chalk with abundant nano-quartz particles are now preserved. (4) The acidification may have been caused by enhanced atmo- spheric CO2 linked to massive volcanic eruptions. References Berner R.A. & Beerling D.J. 2007: Volcanic degassing necessary to pro- duce a CaCO 3 undersaturated ocean at the Triassic–Jurassic boundary. Palaeogeography, Palaeoclimatolog y, Palaeoecolog y 244, 368–373. Binnig, G., Quate, C.F. & Gerber, Ch. 1986: Atomic force microscope. Physics Revue Letters 56, 930–933. Bromley R.G. & Ekdale, A.A. 1986: Flint and fabric in the European chalk. In: Sieveking, G.D.G. & Hart, M.B. (eds): The scientific study of f lint and chert, 71–82. Cambridge: Cambridge University Press. Calvert S.E. 1974: Deposition and diagenesis of silica in marine sediments. International Association of Sedimentologists, Special Publication 1, 273–300. Clayton C.J. 1986: The chemical environment of f lint formation in Up- per Cretaceous chalk. In: Sieveking, G.D.G. & Hart, M.B. (eds): The scientific study of f lint and chert, 43–54. Cambridge: Cambridge Uni- versity Press. Feely, R.A., Sabine C.L., Lee, K., Berelson, W., Kleypas, J., Fabry, V.J. & Millero, F. J. 2004: Impact of anthropogenic CO 2 on the CaCO 3 sys- tem in the oceans. Science 305, 362–366. Authors’ address Geological Survey of Denmark and Greenland, Øster Voldgade 10, DK-1350 Copenhagen K, Denmark. E-mail: hl@geus.dk Frondini, F., Chiodini, G., Caliro, S., Cardellini, C., Granieri, D. & Ven- tura, G. 2004: Diffuse CO 2 degassing at Vesuvio, Italy. Bulletin of Vol- canolog y 66, 642–651. Hjuler, M.L. & Fabricius, I.L. 2009: Engineering properties of chalk re- lated to diagenetic variations of Upper Cretaceous onshore and offshore chalk in the North Sea area. Journal of Petroleum Science and Engi- neering 68, 151-170. Holmes, A. 1965: Principles of physical geolog y, 1288 pp. London: Nelson. Jakobsen, F., Lindgreen, H. & Springer, N. 2000: Precipitation and f loc- culation of spherical nano silica in North Sea chalk. Clay Minerals 35, 175–184. Lindgreen, H., Jakobsen, F. & Springer, N. 2010: Nano-size quartz ac- cumulation in reservoir chalk, Ekofisk Formation, South Arne Field, North Sea. Clay Minerals 45, 171–182. Lindgreen, H., Drits, V.A., Salyn, A.L., Jakobsen, F. & Springer, N. 2011: Formation of f lint horizons in North Sea chalk through marine sedi- mentation of nano-quartz. Clay Minerals 46, 525-537. Madsen, H.B. & Stemmerik, L. 2010: Diagenesis of f lint and porcellanite in the Maastrichtian chalk at Stevns Klint, Denmark. Journal of Sedi- mentary Research 80, 578–588. Maliva, R.G. & Dickson, J.A.D. 1992: Microfacies and diagenetic controls of porosity in Cretaceous/Tertiary chalks, Eldfisk Field, Norwegian North Sea. A APG Bulletin 76, 1825–1838. Millot, G. 1970: Geolog y of clays, 429 pp. New York: Springer. Riebesell, U., Zondervan, I., Rost, B., Tortell, P.D., Zeebe, R.E. & Morel, F.M.M. 2000: Reduced calcification of marine plankton in response to increase atmospheric CO2 . Nature 407, 364–367. Saunders, A.D., Fitton, J.G., Kerr, A.C., Norry, M.J. & Kent, R.W. 1997: The North Atlantic Igneous Province. In: Mahoney, J.J. & Coffin, M.F. (eds): Large igneous provinces. Geophysical Monograph Series 100, 45–97. Schuiling, R.D. 2004: Thermal effects of massive CO 2 emissions associ- ated with subduction volcanism. Comptes Rendus Geoscience 336, 1053–1059. Self, S., Widdowson, M., Thordarson, T. & Jay, A.E. 2006: Volatile f luxes during f lood basalt eruptions and potential effects on the global envi- ronment: a Deccan perspective. Earth and Planetary Science Letters 248, 518–532. Williams, L.A. & Crerar, D.A. 1985: Silica diagenesis, II. General mecha- nisms. Journal of Sedimentary Petrolog y 55, 312–321. Williams, L.A., Parks, G.A. & Crerar, D.A. 1985: Silica diagenesis, I. Solu- bility controls. Journal of Sedimentary Petrolog y 55, 301–311. Zijlstra, H.J.P. 1987: Early diagenetic silica precipitation, in relation to re- dox boundaries and bacterial metabolism, in Late Cretaceous chalk of the Maastrichtian type locality. Geologie en Mijnbouw 66, 343–355. Zimmer, M. & Erzinger, J. 2003: Continuous H 2 O, CO 2 , 222Rn and tem- perature measurements on Merapi Volcano, Indonesia. Journal of Vol- canolog y and Geothermal Research 125, 25–38.