Oxygen and Iron Isotope Systematics of the Grängesberg Mining District (GMD), Central Sweden

Home , El Laco, Teide

Oxygen and Iron Isotope Systematics Examensarbete vid Institutionen för geovetenskaper of the Grängesberg Mining District ISSN 1650-6553 Nr 251 (GMD), Central Sweden

Franz Weis Oxygen and Iron Isotope Systematics

of the Grängesberg Mining District Iron is the most important metal for modern industry and Sweden is (GMD), Central Sweden the number one iron producer in Europe. The main sources for iron ore in Sweden are the apatite-iron oxide deposits of the “Kiruna-type”, named after the iconic Kiruna ore deposit in Northern Sweden. The genesis of this ore type is, however, not fully understood and various schools of thought exist, being broadly divided into “ortho-magmatic” versus the “hydrothermal replacement” approaches. This study focuses on the origin of apatite-iron oxide ore of the Grängesberg Mining District (GMD) in Central Sweden, one of the largest iron reserves in Sweden, employing oxygen and iron isotope analyses on Franz Weis massive, vein and disseminated GMD magnetite, quartz and metavolcanic host rocks. As a reference, oxygen and iron isotopes of magnetites from other Swedish and international iron ores as well as from various international volcanic materials were also analysed. These additional samples included both “ortho-magmatic” and “hydrothermal” magnetites and thus represent a basis for a comparative analysis with the GMD ore. The combined data and the derived temperatures support a scenario that is consistent with the GMD apatite-iron oxides having originated dominantly (ca. 87 %) through ortho-magmatic processes with magnetite crystallisation from oxide-rich intermediate magmas and magmatic fluids at temperatures between of 600 °C to 900 °C. A minor portion of the GMD magnetites (ca. 13 %), exclusively made up of vein and disseminated ore types, is in equilibrium with a high-δ18O and low-δ56Fe hydrothermal fluid at temperatures below 400 °C, indicating the existence of a hydrothermal system associated with the GMD volcano.

Uppsala universitet, Institutionen för geovetenskaper Examensarbete E1, Berggrundsgeologi, 30 hp ISSN 1650-6553 Nr 251 Tryckt hos Institutionen för geovetenskaper, Geotryckeriet, Uppsala universitet, Uppsala, 2013. Examensarbete vid Institutionen för geovetenskaper ISSN 1650-6553 Nr 251

Oxygen and Iron Isotope Systematics of the Grängesberg Mining District (GMD), Central Sweden

Franz Weis

Oxygen and iron isotope systematics of the Grängesberg Mining District (GMD), Central Sweden

Table of contents

Abstract ...... 3 Sammanfattning på Svenska ...... 3 Zusammenfassung ...... 4 1. Introduction ...... 5 1.1 Apatite-iron oxide deposits ...... 6 1.2 The origin of apatite-iron oxide ores ...... 7 2. Aim of study ...... 8 3. Samples and geological background ...... 9 3.1 Samples...... 9 3.2 Geological background...... 14 3.2.1 The geology of the Bergslagen mining district ...... 14 3.2.2 Local geology of the Grängesberg Mining District ...... 16 3.2.3 Geological background of additional Swedish ore samples ...... 18 3.2.3.1 Geology of Kiruna ...... 18 3.2.3.2 Geology of Dannemora ...... 18 3.2.3.3 The geology of Striberg ...... 19 3.2.3.4 Swedish Layered Igneous Intrusions ...... 20 3.2.4 Geological background of international magnetite samples ...... 21 3.2.4.1 Geology of El Laco (Chile) ...... 21 3.2.4.2 Layered Igneous Intrusions ...... 22 3.2.4.3 Volcanic reference materials ...... 23 4. Analytical Methods...... 26 4.1 Sample preparation and analysis ...... 26 4.2 Possible analytical errors ...... 27 5. Background on stable isotope systematics and data evaluation methodology ...... 29 5.1 Stable isotopes and the δ-value ...... 29 5.2 Stable isotope fractionation ...... 29

6. Results ...... 32 6.1 Oxygen and iron isotope data of different magnetite ores and reference materials...... 32 6.1.1 Oxygen isotopes ...... 32 6.1.2 Iron isotopes ...... 33 6.2 Isotope data for hydrothermal aureole identification ...... 35 7. Discussion ...... 37 7.1 Determining ore formation processes ...... 37 7.1.1 Comparison of magmatic and ore magnetites ...... 37 7.1.1.1 Oxygen isotopes ...... 37 7.1.1.2 Iron isotopes ...... 38 7.1.1.3 Bimodal formation ...... 40 7.1.2 Calculation of possible ore sources ...... 41 7.1.3 Geothermometric calculations ...... 49 8. Hydrothermal aureole identification ...... 51 8.1 Hypothesis and background ...... 51 8.2 Results ...... 52 8.3 Calculation of possible quartz sources ...... 54 8.4 Interpretation ...... 55 9. Summary of the data evaluation ...... 56 10. A conceptual model ...... 57 10.1 High temperature: crystallisation from magma ...... 57 10.2 Ore formation from magmatic fluids ...... 58 10.3 Low-temperature: hydrothermal re-deposition ...... 59 11. Summary and Conclusions ...... 62 12. Acknowledgments ...... 64 13. References ...... 66 14. Appendix ...... 76 14.1 Appendix 1...... 76

Abstract

Iron is the most important metal for modern industry and Sweden is the number one iron producer in Europe. The main sources for iron ore in Sweden are the apatite-iron oxide deposits of the “Kiruna-type”, named after the iconic Kiruna ore deposit in Northern Sweden. The genesis of this ore type is, however, not fully understood and various schools of thought exist, being broadly divided into “ortho-magmatic” versus the “hydrothermal replacement” approaches. This study focuses on the origin of apatite-iron oxide ore of the Grängesberg Mining District (GMD) in Central Sweden, one of the largest iron reserves in Sweden, employing oxygen and iron isotope analyses on massive, vein and disseminated GMD magnetite, quartz and meta-volcanic host rocks. As a reference, oxygen and iron isotopes of magnetites from other Swedish and international iron ores as well as from various international volcanic materials were also analysed. These additional samples included both “ortho-magmatic” and “hydrothermal” magnetites and thus represent a basis for a comparative analysis with the GMD ore. The combined data and the derived temperatures support a scenario that is consistent with the GMD apatite-iron oxides having originated dominantly (ca. 87 %) through ortho-magmatic processes with magnetite crystallisation from oxide-rich intermediate magmas and magmatic fluids at temperatures between of 600 °C to 900 °C. A minor portion of the GMD magnetites (ca. 13 %), exclusively made up of vein and disseminated ore types, is in equilibrium with a high-δ18O and low-δ56Fe hydrothermal fluid at temperatures below 400 °C, indicating the existence of a hydrothermal system associated with the GMD volcano.

Sammanfattning på Svenska

Järn är den särklass viktigaste metallen för moderna industrier och Sverige är den största järnproducenten i Europa. De huvudsakliga järnförekomsterna i landet är apatit-järnmalmer av ”Kiruna-typ”, efter den berömda Kirunajärnmalmen i norra Sverige. Ursprunget av den här malmtypen är inte helt klarlagt och det finns flera förslag som är uppdelade i den ”ortomagmatiska gruppen” och den ”hydrotermala gruppen”. Den här studien behandlar ursprunget av apatit-järnmalmfyndigheten i Grängesberg (GMD) i Bergslagen, som är en av Sveriges största järnförekomster. Syre- och järnisotoper av massiv, disseminerad och åder-typ magnetit från GMD samt kvarts och metavulkaniska bergarter analyserades. Som en jämförelse analyserades ytterligare syre- och järnisotoper från andra Svenska och internationella järnmalmförekomster och även vulkaniska magnetitkristaller som referens. De

3 ytterligare proven inkluderar både magmatiska och hydrotermala magnetiter och presentera därför en bra bas för jämförelsen med GMD magnetiterna. Data och beräknade temperaturer som presenteras i studien understödjer ett ortomagmatiskt ursprung för största delen av GMD magnetiterna (ca. 87 %). Malmen bildades från en järnrik, intermediär magma och magmatiska fluider vid temperaturer mellan 600 °C och 900 °C. Bara en liten del av GMD magnetiterna (13 %), som består av ådror och dissemineringar av magnetit, visar jämvikt med en hög-δ18O, låg-δ56Fe fluid vid temperaturer under 400 °C, som anger att det fanns även ett hydrotermalt system samtidigt med GMD vulkanen.

Zusammenfassung

Eisen bildet auch heute noch den wichtigsten Rohstoff für moderne Industrien, mit Schweden als den größten Eisenproduzenten in Europa. Den größten Teil des geförderten Eisenerzes gewinnt Schweden aus den Apatit-Eisenerz-Ablagerungen des „Kiruna-Typs“, welcher nach dem weltbekannten Kiruna-Erzvorkommen in Nordschweden benannt ist. Die Genese dieses Erztypus‘ ist jedoch bis heute umstritten und die Meinungen hierzu sind breit gefächert. Es existieren sowohl „orthomagmatische“ als auch „hydrothermale“ Lösungsansätze. Diese Studie befasst sich mit der Genese der Apatit-Eisenerz-Ablagerung des Grängesberg Bergbaugebietes (GMD) in Mittelschweden, welches eine der größten Eisenerzreserven Schwedens darstellt. Als Lösungsansatz dient eine Eisen- und Sauerstoffisotopenanalyse von sowohl massiven, Venen- und Imprägnationsmagnetiten als auch Quartz und Wirtsgestein. Als zusätzliche Referenz und zum Vergleich wurde eine weitere Eisen- und Sauerstoffisotopenanalyse an Proben von anderen schwedischen, sowie internationalen Eisenerzvorkommen und vulkanischen Magnetiten durchgeführt. Dieses zusätzlich analysierte Material repräsentiert sowohl „orthomagmatischen“ als auch „hydrothermalen“ Magnetit und stellt daher eine solide Grundlage für den Vergleich mit dem Grängesbergerz dar. Die zusammengefassten Daten und berechneten Temperaturen begünstigen die Annahme einer Erzgenese durch „orthomagmatische“ Prozesse für den größten Teil des Grängesbergerzes (ca. 87 %), welches zwischen 600 °C und 900 °C von einem eisenoxidangereicherten Magma sowie von magmatischen Flüssigkeiten kristallisierte. Lediglich ein kleiner Teil der Ablagerung (ca. 13 %), bestehend aus Venen- und Imprägnationsmagnetit, liegt im Equilibrium mit einer hydrothermalen Flüssigkeit von hohem δ18O und niedrigem δ56Fe bei Temperaturen unter 400 °C. Dies verweist auf die Existenz eines zusätzlichen hydrothermalen Systems, das vermutlich mit dem ehemaligen Grängesbergvulkan assoziiert war.

1. Introduction

“But Iron - Cold Iron - is master of them all.”

Rudyard Kipling (1865-1936)

Despite the current industrial demand (and dependence) on Rare Earth Elements (REE) and hi-tec metals, which was shown for example in the REE-crisis between USA, Japan and China of 2010 (L.A. Times, 28th December 2010), iron consistently remains the most important metal for modern industry (Vaughan 2011). In Europe, Sweden holds the leading position in the production of iron (USGS 2011) and the country’s supply of iron ore comes from two core regions; one is the Kiruna-Malmberget region in the North of Sweden and the other is the Bergslagen Mining District in Central Sweden. In both cases the most significant supply of iron-ore comes from “apatite-iron oxide ores”, also globally known as “Kiruna- type” ore, named after the largest iron oxide deposit in the North of Sweden. Central Sweden hosts the area of Grängesberg in Bergslagen, which is home to the largest apatite-iron oxide ore body in this region. GMD shows a past production of ≥156 Mt (million tons) of ore until the closure of the mine in 1989 (Fig.1). As the global demand for iron is still increasing and as market prices have increased since the late 1980s, plans are currently being made to re-open the mine at Grängesberg and exploit the likely extensive resources in the area. This is because apatite-iron oxide ores are famous for their high grades of iron and large tonnages of ore and are thus favoured by mining companies. Despite this relevance and popularity, the origin of apatite-iron oxide ores is not yet fully understood. It is thus sensible and advantageous to try to learn about the processes involved in the formation of apatite-iron oxide ore as this effort will likely aid geologists in future prospecting campaigns to locate, detect and describe this highly valuable ore type.

Fig.1 One of the main shafts at the Grängesberg iron ore mine in 1969. Source: Grängesberg Gruv AB (www.lansstyrelsen.se).

1.1 Apatite-iron oxide deposits

The ore deposit at Grängesberg is of the apatite-iron oxide type and is considered by some to belong to the group of Iron-Oxide-Copper-Gold-deposits (IOCG, Corriveau 2007). These deposits are usually sulphide-deficient and contain more than 20% iron oxides in the form of low-Ti magnetite and/or hematite as well as minor amounts of F-rich apatite, amphibole or pyroxene (Geijer 1931, Hunt et al. 2007, Corriveau 2007). The apatite-iron oxide subtype, also referred to as the “Kiruna sub-type”, is associated with low-Ti magnetite and apatite. Ore deposits of this sub-group are usually associated with calc-alkaline to mildly alkaline volcanic and plutonic host rocks such as andesites, dacites and rhyolites, which points to a subuction affinity (Frietsch & Perdahl 1995, Allen et al. 1996, Rhodes et al. 1999, Nyström et al. 2008). Apatite-iron oxide deposit genesis occurs in extensional fault zone settings such as those

6 found for example in continental volcanic arcs like the Andes and back-arc basins at subduction zones like the Japan Sea (Frietsch & Perdahl 1995, Groves et al. 2010). Deposits of the Kiruna-type vary in age from the Paleoproterozoic (e.g. Grängesberg, 1.9-1.85 Ga; Kiruna, 1.9-1.88 Ga) to the Quaternary (e.g. El Laco, Chile, 2.1 Ma; Nyström et al. 2008, Jonsson et al. 2010). The morphology of apatite-iron oxide ores consists typically of disc- or lens-shaped, concordant ore-bodies and vein systems or impregnations (Frietsch & Perdahl 1995, Sillitoe 2003, Hunt et al. 2007). Associated with this ore type is alteration such as silicification, sericitisation, albitisation, and epidotisation (Treloar & Colley 1996, Martinsson 2004). Apatite-iron oxide deposits impress with their enormous size and high grades of ore. For example, Kiruna shows a reserve of two billion tons of ore and a grade of 60 % Fe while the GMD still holds a tonnage of 150 Mt and a grade between 40 and 64% (Allen et al. 2008, Nyström et al. 2008).

1.2 The origin of apatite-iron oxide ores

The genesis of apatite-iron oxide ores is cause of a long-standing debate in the scientific community that has been ongoing for over 100 years by now. During this time, scientists studying this type of ore body divided into two broad fractions; the magmatic and the hydrothermal ‘schools of thought’. The magmatic school proposes dominantly “ortho- magmatic” ore formation by processes such as magmatic crystallisation or segregation and by exsolution of magmatic fluids, all of which would lead to intrusive and extrusive iron ores that have essentially formed by the activity of magma at high temperatures (e.g. Geijer 1931, 1967, Frietsch 1978, Lundberg & Smellie 1979, Pollard 2000, Nyström et al. 2008, Tornos et al. 2011). On the other hand, the hydrothermal school (Paràk 1975, 1984, Hitzman et al. 1992, Sheets 1997, Rhodes & Oreskes 1994, 1999, Sillitoe & Burrows 2002) puts forward formation by seafloor sedimentation, magmatic hydrothermal replacement, and hydrothermal precipitation. Both groups support their proposals with the help of field, textural and various geochemical evidences from key ore bodies like the ones at Grängesberg, El Laco and Kiruna.

2. Aim of study

Previous studies of the Grängesberg apatite-iron oxide ore have been carried out by e.g. Johansson (1910) and Magnusson (1938, 1970) who described the local geology and textural as well as morphological features of the ore. However, since the 1970s limited research has been devoted to this ore body and only recently new research efforts have been commenced (e.g. Jiao 2011, Weis 2011, Jonsson et al. 2013). These recent studies focus on petrological and geochemical approaches using oxygen isotopes and REE concentrations of ore and host rock to unravel the origin of the ore body. The results of these studies indicate that the ore formation took place in two stages, involving a magmatic as well as a hydrothermal component. Massive ore bodies are suggested to have formed from intrusion of an iron oxide- rich magma. Hydrothermal fluid circulation, originating from remnant exsolved magmatic and influxing ground- or meteoric water, caused local leaching and re-deposition of magnetite in form of veins and disseminations at low temperatures.

This project aims to further unravel the mode of formation of the Grängesberg iron ore and to provide new insight into the genesis of “Kiruna-type” ores. The current study builds on the previous work of Weis (2011) and Jonsson et al. (2013) and contributes additional oxygen isotope data (n=20) and presents new values of the iron isotope signatures of the GMD apatite-iron oxide ore (n=17). In addition, samples from other magnetite ore bodies such as El Laco (Chile, n=6) and Kiruna (Sweden, n=1), as well as regular igneous and hydrothermal magnetites are supplied for comparison (n=24). Subsequently temperature calculations are conducted with the help of the oxygen isotopes.

In a second part of the thesis an approach to identify a possible hydrothermal aureole in the Grängesberg area is presented. This section is based on whole rock oxygen isotope data derived from samples taken in the area directly surrounding the GMD ore deposit. Similar approaches were for example used in studies by Larson & Sharp (2003) on the Priest pluton in New Mexico and Ayers et al. (2006) on the Birch Creek pluton in California.

3. Samples and geological background

3.1 Samples

The massive ore samples for the current study were collected from three drillcores (DC), DC 690 (n=7), DC 717 (n=3) and DC 575 (n=3) which have been sampled by Prof. Erik Jonsson from the Geological Survey of Sweden (SGU) in 2009 at the SGU core repository in Malå. The first two drillcores come from the southern part and the centre of the ore field, whilst the third is from the northern part of the ore body (Fig.5). The three cores were sampled from the end of mine shafts that transect the ore body semi-horizontally and come from depth levels of 650 m (2x) and 570 m below the surface. Ore and host rock samples from these cores (n=44) have also been used in pilot studies on oxygen isotope analysis (Weis 2011; Jonsson et al. 2013). In addition to the massive ore samples, two samples from DC 717 and DC 690 were taken from magnetite veins and disseminations in the GMD host rocks. Another two ore samples are from waste piles at the nearby but smaller apatite-iron oxide ore deposits at Blötberget and Björnberget respectively. Besides magnetite, quartz crystals were separated from massive GMD ores (n=3).

To achieve a meaningful comparison of the GMD samples, oxygen and iron isotope values were also determined on massive magnetite ores from the El Laco ore deposit in Chile (n=2), from Kiruna (n=1), Swedish iron-skarns from Dannemora (Fig.2, n=4), the layered igenous intrusion of Panzhihua in China (n=2), the Bushveld igneous complex in South Africa (n=1), the Swedish layered igneous intrusion deposits of Taberg (n=1), Ulvön (n=1), Ruoutevare (n=1) and the Swedish Banded Iron Formation (BIF) of Striberg (n=1, Fig.2).

Samples representative of igneous magnetites were chosen from basaltic andesites from Indonesia (n=6), basalts from the Canary Islands (n=3), dacites from New Zealand (n=2) and Cyprus (n=1) and a basaltic lava bomb from Iceland (n=1). Therefore a total of 51 samples was analyzed for Part 1 of the project and detailed sample descriptions are given in Table 1.

Fig.2 Sample location of the Swedish ore samples used for the first part of this study.

Table 1. The samples for the oxygen and iron isotope analysis for the first part of the project. Sample Sample Description Sample Location Analysis Fe O

KES090068 Massive (Ap-)magnetite ore DC 717, Grängesberg, Sweden1   KES090070 Massive Ap-magnetite ore DC 717, Grängesberg, Sweden1   KES090072 Massive (Ap-)magnetite ore DC 717, Grängesberg, Sweden1  

KES090011 Massive (Ap-)magnetite ore DC 690, Grängesberg, Sweden1   KES090012 Massive (Ap-)magnetite ore DC 690, Grängesberg, Sweden1   KES090020 Massive Ap-veined magnetite ore DC 690, Grängesberg, Sweden1   KES090024 Massive (Ap-)magnetite ore DC 690, Grängesberg, Sweden1   KES090027 Ap-veined/banded massive magnetite ore DC 690, Grängesberg, Sweden1   KES090030 Silicate-spotted massive (Ap-)magnetite ore DC 690, Grängesberg, Sweden1   KES090034 Coarse-grained Ap-spotted massive magnetite ore DC 690, Grängesberg, Sweden1   KES090034 qtz Quartz from massive (Ap-)magnetite ore DC 690, Grängesberg, Sweden1 - 

KES103003 Massive (Ap-)magnetite ore close DC 575, Grängesberg, Sweden1   KES103003 qtz Quartz from massive (Ap-)magnetite ore DC 575, Grängesberg, Sweden1 -  KES103011 Mt-dominated massive ore DC 575, Grängesberg, Sweden1   KES103016 Coarse, massive (Ap-)magnetite ore DC 575, Grängesberg, Sweden1   KES103016 qtz Quartz from massive (Ap-)magnetite ore DC 575, Grängesberg, Sweden1 - 

KES091013b Banded magnetite ore from waste pile Blötberget (near Grängesberg) Sweden1   KES091007B Calcitic-magnetite ore from waste pile Grängesberg (Björnberget) Sweden1   Disseminated magnetite in intermediate volcanic KES090044 DC 690, Grängesberg, Sweden1 rock   KES090084 Magnetite vein in intermediate volcanic rock DC 717, Grängesberg, Sweden1  

DM-1 Iron-skarn magnetite ore Botenhäll, Dannemora, Sweden2   DM-2 Iron-skarn magnetite ore Norrnäs 3, Dannemora, Sweden2   DM-3 Iron-skarn magnetite ore Konstäng, Dannemora, Sweden2   DM-4 Iron-skarn magnetite ore Strömsmalmen, Dannemora, Sweden2  

EJ-LS-11-1 Massive magnetite ore Laco Sur, El Laco, Chile1   EJ-LS-11-2 Massive magnetite ore Laco Sur, El Laco, Chile1   EJ-LS-11-3 Massive magnetite ore Laco Sur, El Laco, Chile1   EJ-LS-11-4 Massive magnetite ore Laco Sur, El Laco, Chile1   LS-2 Massive magnetite ore Laco Sur, El Laco, Chile3   LS-52 Massive magnetite ore Laco Sur, El Laco, Chile3  

Kiruna Banded massive Apatite-magnetite ore Kirunavaara, Lappland, Sweden1   Ruoutevare Ti-magnetite, layered igneous intrusion deposit DC, Kvikjokk, Norbotten, Sweden4   Ulvön Ti-magnetite, layered igneous intrusion deposit Ulvön island, Ångermanland, Sweden4   Taberg Ti-magnetite, layered igneous intrusion deposit Iron mine, Taberg, Småland, Sweden5   EJ092008 Magnetite from a banded iron formation deposit Striberg, Bergslagen, Sweden1    11

Table 1. Continued. EM419 Massive Fe-Ti magnetite ore Northern pit, Panzhihua, China6   EM424 Massive Fe-Ti magnetite ore DC, Nalahe, Panzhihua, China6   Bushveld Massive magnetite ore Upper Zone, Bushveld Complex, South Africa6 - 

TEF-NER-18 Magnetite from an ankaramite dyke NE Rift Zone, Tenerife, Spain7  - TEF-NER-57B Magnetite from an ankaramite dyke NE Rift Zone, Tenerife, Spain7   TEF-NER-70 Magnetite from a pyroxene phyric dyke NE Rift Zone, Tenerife, Spain7   MG-07 Igneous magnetite from dacite S Flank, Mt. Ruapehu, New Zealand8  - MG-09 Igneous magnetite from dacite S Flank, Mt. Ruapehu, New Zealand8  - Kelut A1 Igneous magnetite from basaltic andesite Mt. Kelut, Java, Indonesia9  - GD-D-2 Igneous magnetite from basaltic andesite Gede Dome, Java, Indonesia9  - AK-B1 Igneous magnetite from basaltic andesite SE Flank, Anak Krakatau, Java, Indonesia9  - AK-B3 Igneous magnetite from basaltic andesite SE Flank, Anak Krakatau, Java, Indonesia9  - A-BA-1 Igneous magnetite from basaltic andesite Mt. Agung, Java, Indonesia9  - M-BA06-KA-3 Igneous magnetite from basaltic andesite Kaliadem Village, Mt. Merapi, Java, Indonesia9   83/CRS/6 Igneous magnetite from a dolerite dyke Agros area, Troodos Massiv, Cyprus10  - 5 Basaltbomb Magnetite from a basaltic lavabomb NW Flank, Skjaldbreiður, Iceland  

1. Samples donated by Prof. Erik Jonsson, Geological Survey of Sweden & Uppsala University, Uppsala, Sweden

2. Samples donated by Gunnar Rauseus at Dannemora Magnetit AB, Österbybruk, Sweden

3. Samples donated by Dr. Jan-Olov Nyström, Naturhistoriska Riksmuseet, Stockholm, Sweden

4. Samples taken from the sample collection of the Geological Survey of Sweden, Uppsala, Sweden

5. Sample taken from the sample collection, Solid Earth Geology, Department of Earth Science, Uppsala University, Sweden

6. Samples collected and donated by Prof. Nicholas Arndt, Université Joseph Fourier, Grenoble, France

7. Samples donated by Prof. Valentin R. Troll, Department of Earth Science, Uppsala University (see Deegan et al. 2012)

8. Samples donated by Prof. John Gamble, Department of Geology, Victoria University of Wellington, New Zealand

9. Samples donated by Prof. Valentin R. Troll, Department of Earth Science, Uppsala University (see Gardner et al. 2012)

10. Samples donated by Prof. Christopher J. Stillman, Department of Geology, Trinity College Dublin, Ireland (see Stillman 1989)

The sample set for the second part of the thesis (hydrothermal aureole identification) contains both whole rock samples as well as quartz crystals from the ore host rock (n=23). All samples have been collected from the surface in the vicinity of the Grängesberg mine. Table 2 gives a detailed sample description and sample locations around the GMD central ore field.

Table 2. Sample set for Part II of the project. Qtz- WR- Sample Description Coordinates RT90 Analysis analysis East North Sample 29 Grey granite with mafic enclaves1 1454958 6662272  - Sample 31 Light-grey leuco gneiss1 1456585 6661782  - Sample 40 Biotite-rich, greyish granite1 1456305 6661990  - Sample 45 Mafic granite with feldspar-phenocrysts1 1454534 6661130  - Sample 48 Grey, biotite-rich granite1 1454386 6660546   Sample 49 Dark-grey pinkish granite1 1455395 6661477   Sample 52 Biotite- and amphibole-rich, grey-pinkish granite1 1456209 6662922   Sample 53 Pinkish, amphibole- and biotite-rich microgranite1 1454816 6662790 -  Sample 56 Grey-pinkish granite1 1455284 6661802 -  KPN-090007A Grey to grey-reddish dacite2 1454596 6663236 -  KPN-090026-4 Grey, porphyritic metavolcanic dacite2 1453806 6662229   KPN-090033A Light grey, porphyritic felsic volcanic rock2 1453794 6662943   KPN-090042A Grey, feldspatic porphyritic metavolcanic rock2 1453884 6663419   KHO-09003 Glacial polished, pinkish, quartz-rich porphyritic dacite3 1454400 6663551 -  KHO-09005 Pinkish, magnetite-rich dactie3 1454562 6663500   KHO-090010 Dark grey, biotite- and quartz-rich metavolcanic rock3 1454196 6661292 -  KHO-090011A Grey, biotite-rich metavolcanic rock3 1454420 6661821 -  KHO-090012 Light grey, fine grained foliated dacite3 1454571 6661728   KHO-090013 Biotite-rich, crenulated, medium-grained andesite3 1454467 6661972 -  KHO-090013G Biotite-rich, crenulated, medium-grained andesite3 1454473 6661971 -  KHO-090017 Fine-grained, banded, laminated, biotite-rich volcanic rock3 1453588 6660757  - KHO-09127b Fine-grained, light grey dacite3 1454632 6662214 -  KHO-09129A Fine-grained, biotite-poor meta-volcanic rock3 1453535 6662113  -

1 . Samples collected and donated by Prof. Valentin R. Troll, Department of Earth Science, Uppsala University

2. Samples collected and donated by Dr. Katarina Persson-Nilsson, Geological Survey of Sweden, Uppsala, Sweden

3. Samples collected and donated by Dr. Karin Högdahl, Department of Earth Science, Uppsala University

3.2 Geological background

3.2.1 The geology of the Bergslagen mining district

The Grängesberg mining district (GMD) is situated in NW-Bergslagen, a region located in South Central Sweden. Bergslagen is known for the richness of iron as well as other metals, containing more than 6000 ore deposits and mineral prospects (Fig.4; Allen et al. 2008). The different ore deposits range from sulphide mineralizations (e.g. Zn-Pb-Ag sulphides) like the ones at Garpenberg and Falun (Fig.3), Fig.3 The former copper mine at Falun in Bergslagen. The location is famous for the through granite-hosted Climax-type (very first production of the typical Swedish colour large and high-grade) molybdenum deposits, “Falun Red”, for which the copper from the mine was also used (Source: Daniels Sven to iron ore deposits of the BIF, skarn- and Olsson, www.mineralatlas.eu). apatite-rich type (Fig.4; Ripa & Kübler 2003, Stephens et al. 2009). The iron ores, however, dominate the Bergslagen area with both tonnage and occurrence. Out of the 42 largest deposits in the Bergslagen region, 32 are iron oxide deposits (e.g. BIFs, skarn and apatite-iron oxide ores) which produced 421 Mt of iron in the past. Only nine out of the 42 largest deposits are sulphide deposits with a past total production of 74 Mt of ore (Åkerman 1994, Ripa & Kübler 2003).

Beside the ore deposits, the geology of the Bergslagen district comprises relics of meta- sedimentary and meta-volcanic successions together with pre- and post-tectonic intrusions (Fig.3; Allen et al. 1996). The rocks in the area belong to the so called Svecofennian volcano- sedimentary succession, which dates back to the Palaeoproterozoic with rock formation thought to have occurred between 1.9-1.8 Ga (Oen et al. 1982, Lundström 1987, Lundström et al. 1998, Ripa and Kübler 2003, Stephens et al. 2009). Composition of the volcanic to subvolcanic rocks is predominantly dacite to rhyolite (Ripa 2001; Allen et al. 1996), however, subordinate, intermediate as well as mafic volcanic rocks occur together with chemical, epiclastic and organo-sedimentary rocks including carbonates at different stratigraphic levels in the volcanic successions (Ripa 2001). The meta-volcanic succession has earlier been interpreted as interlayered between two meta-sedimentary successions (Lundström 1995, Allen et al. 1996). The base of the meta-volcanic succession is exposed on Utö Island in the southeast of the region where it overlies a thick series of meta-sedimentary rocks (Allen et al. 14

Fig.4 Distribution of ore deposits and metallic mineralizations in the Bergslagen district, Central Sweden (Source: Stephens et al. 2009).

1996). These rocks are amongst the oldest in the region and consist of clastic meta- sedimentary facies shallowing upwards into quartzites followed by the meta-volcanic sequence (Stephens et al. 2009). The basement of the Svecofennian bedrock is not exposed in the region. However, the presence of the thick quartzites at the base and the discovery of some 2.7-1.95 Ga detrital zircons in the quartzites led to the assumption that the region is underlain by a previously eroded granitic basement (Claesson et al. 1993, Allen et al. 1996).

Between 1.9 and 1.8 Ga several large igneous intrusions were emplaced coeval with extensive volcanic activity (Wilson et al. 1984, Stephens et al. 2009, Ripa & Kübler 2003). These intrusions are divided into three types according to their composition. Present are the granitoid-dioritoid-gabbroid (GDG), the granitoid-syenitoid-dioritoid-gabbroid (GSDG) and the granite-pegmatite (GP) type (Stephens et al. 2009). The GDG along with some GSDG rocks are the oldest (1.9-1.87 Ga) of the intrusions and are pre-tectonic. The syn- and post- tectonic intrusions (1.87 Ga -1.75 Ga) are dominated by the GSDG and GP type intrusions (Stephens et al. 2009). The Bergslagen region experienced deformation and low- to medium amphibolite grade metamorphism during the Palaeoproterozoic Svekokarelian orogeny (1.9 and 1.8 Ga) and the Sveconorwegian orogeny (1.0-0.9 Ga; Lundqvist 1979, Sundblad 1994, Stephens et al. 2009, Jonsson et al. 2010). However, the latter mainly affected the western part of the Bergslagen region. Deformation of the meta-volcanic and meta-sedimentary rocks is predominantly expressed as steep, tight to isoclinal, doubly plunging synclines (Allen et al. 1996) as is also the case at GMD.

The currently available interpretation for the original geology of the Bergslagen region is that of an extensional back-arc basin inboard of an active continental margin region (Vivallo & Rickard 1984, Allen et al. 1996). The supracrustal rock succession of the region shows a continuum from deep water sedimentation (turbidites) to thermal doming and extension accompanied by shallow marine to subaerial volcanism, back to shallow marine (carbonates, volcanic silt- to sandstones) and later again deep sea sedimentation after thermal subsidence and transgression followed by compressional deformation (Allen et al. 1996, Stephens et al. 2009).

3.2.2 Local geology of the Grängesberg Mining District

The western part of the Bergslagen region is dominated by an 8 km thick meta-volcanic succession with interbedded meta-limestones (Allen et al. 1996). The apatite-iron oxide ore at Grängesberg is hosted by andesite, dacite as well as rhyolite meta-volcanic rocks that are 1.91-1.89 Ga in age (Fig. 4; Ripa & Kübler 2003, Stephens et al. 2009). Also present are some pegmatites. Directly to the west and east of the ore body the host rocks were intruded by pre- tectonic GDG-intrusions that display mixing and mingling features between felsic and mafic composition. These intrusions bear xenoliths originating from the ore body. The ore thus pre- dates the intrusion and is therefore older than the 1.85 Ga granites. Alteration zones in the host rocks and in the ore’s vicinity comprise disseminated and discrete phyllosilicate (biotite, chlorite) and amphibole-rich hydrated assemblages (Frietsch 1982). Towards the NE, a post- 16

Fig.5 Geological map of the Grängesberg Mining District. The red circles mark the positions of the drillcores (Source: SGU Kartgenerator). metamorphic intrusion of the GSDG type occurs (Fig.5). The apatite-iron oxide ore at Grängesberg is made up of dominantly magnetite, but also of hematite ore accompanied by apatite, and occurs in lense-shaped bodies as well as veins and disseminations (Magnusson 1970, Allen et al. 1996, 2008, Stephens et al. 2009). In the main lenses, banding of fine- grained fluorapatite or silicate minerals can be present as well as some skarn alteration. The main ore field, the so called “Export Field”, consists of iron oxide in the ratio of approximately 80 % magnetite and 20 % hematite ore. Hematite dominated ores occur mostly in the ore body’s footwall. The ore deposit dips between 70° and 80° towards the south-east and can be followed for more than 900 m at the surface where its width ranges between 50 and 100 m (Johansson 1910). Its principal orientation is stratiform, meaning that it is semi- 17 parallel to the internal contacts within the meta-volcanic host-rocks as well as to the main local structures. The grade of the GMD ore ranges between 40 and 64 % in iron and a tonnage of approximately 150 Mt is estimated to be held as a reserve at the GMD (Åkerman 1994, Allen et al. 2008). The phosphorous content of the ore ranges from 0.7 -1.3 % (Allen et al. 2008) and might represent a future resource for this element as well as for RE-elements.

3.2.3 Geological background of additional Swedish ore samples

3.2.3.1 Geology of Kiruna

The Kirunavaara deposit in the Kiruna mining district in Northern Sweden is hosted by trachyandesite and rhyolite ignimbrites and tuffs with an age of 1.9 to1.8 Ga (Nyström et al. 2008). These ignimbrites are underlain by older greenstones which are supposed to have formed in an extensional setting (Lindblom et al. 1996). The rocks in the area are not strongly deformed but have been affected by regional metamorphism and, together with the ore, secondary alteration which resulted in greenschist facies-grade overprint (Nyström et al. 2008). The apatite-iron oxide ore at Kiruna consists mainly of magnetite with iron contents of 60-67 % and up to 30 % apatite. The deposit holds pre-mining reserves of about 2 billion tons of ore (Nyström et al. 2008, Westerstrand & Öhlander 2011). The ore body is related to an extensive fault zone (Harlov et al. 2002) and has been interpreted to be, just like the El Laco deposit, of primarily magmatic origin on the basis of geochemical and textural observations (Nyström et al. 2008), including vesicular ore textures and oxygen isotopes. The sample used in this study represents massive magnetite ore from the Kirunavaara apatite-iron oxide ore deposit and has been donated by Prof. Erik Jonsson from the Geological Survey of Sweden.

3.2.3.2 Geology of Dannemora

The Dannemora skarn iron ore is situated in the eastern part of the Bergslagen mining district. It consists of 25 iron ore bodies and some smaller sulphide deposits, which are hosted by 1.9 Ga old meta-volcanic and meta-sedimentary rocks, such as meta-dacites and meta-rhyolites as well as calcite and dolomite meta-limestones with stromatolitic textures (Lager 2001). The volcanic rocks have been interpreted as pyroclastic flow and air-fall deposits which together with the limestones have been deposited in open marine, lagoonal and some terrestrial environments (Lager 2001). Dolomite limestones at Dannemora are very dark-coloured due to a content of fine magnetite of 5-30 % (Lager 1986, 2001). The area has been affected by greenshist facies metamorphism during the Svecokarelian orogeny and deformed at least twice, leading to isoclinal folding (Lager 2001, Dahlin & Sjöström 2010). The iron ore at 18

Dannemora consists to a great extent of massive stratabound magnetite ore dipping 65-70° to the west, within an East-South-East syncline and has an iron content of between 30 and 50 % (Lager 1986, 2001). The formation of the ore is believed to have been by the circulation of hydrothermal fluids which also altered silica-rich units in the area to skarn containing Fe- bearing silicates (Lager 2001). These ore forming fluids are supposed to have been migrating from a high relief hinterland to lower areas and accumulating within an evaporitic pan where the limestones formed (Lager 2001). The Dannemora samples used in this study are all calcite-bearing magnetite ores and come from various locations at Dannemora (see Table 1). They were provided by Gunna Rauseus from the Dannemora Mineral AB mining company. The Dannemora magnetites represent a hydrothermal ore mineralization and thusmake for a useful comparison to the GMD ore regarding the question about an “ortho-magmatic” or “hydrothermal” ore genesis.

3.2.3.3 The geology of Striberg

Banded Iron Formations such as the deposit at Striberg are found in the Western part of the Bergslagen mining district and are also hosted by the 1.91-1.89 Ga old metavolcanic rocks (Allen et al. 2008, Stephens et al. 2009). The deposit consists of alternating quartz- and magnetite-rich layers of 1-10 mm thickness and the silica content varies between 18 % and 28 % (Frietsch 1975, Allen et al. 2008, Stephens et al. 2009). Typical for BIF deposits, there is a dominance of hematite as the main Fe-bearing mineral (Fig.6), however, magnetite is also present as an alteration product of the latter (Allen et al. 2008, Stephens et al. 2009). The iron content of the deposits lies between 30 % and 55 %, and in addition they contain concentrations of phosphorous and manganese of < 0.03 % and 0.2 % respectively (Allen et al. 2008, Stephens et al. 2009). Banded Iron Formations formed solemnly from low- Fig.6 The Banded Iron temperature precipitation and, like the Dannemora Formation of Striberg in Bergsalagen, Central Sweden samples, provide a solid base for the comparison of (Source: Göran Axelsson, “ortho-magmatic” versus “hydrothermal and low- www.geology.neab.net). temperature” ore forming processes.

3.2.3.4 Swedish Layered Igneous Intrusions

Smålands Taberg: The Fe-Ti mineralization at Taberg is located in Southern Sweden, about 12 km to the south of Lake Vättern within the Protogine Zone, a 1.56 Ga old, 20 km wide and several 100 km long belt of ductile and brittle deformation (Sandecki 2000, Söderlund et al. 2004). The ore deposit consists of Ti-magnetite-rich troctolites (e.g. olivine gabbro) of about 1.2 Ga in age, which have been affected by the late Sveconorwegian amphibolite facies deformation (Sandecki 2000, Larsson & Söderlund 2005). The ore body expands about 1 km long and 400 m wide from the South East to North West and is hosted by amphibolitised gabbro-dolerite which has intruded the surrounding Smålands Granite (Hjelmqvist 1950, Sandecki 2000). The ore holds between 26 and 35 % titanomagnetite with a content of 28.7– 32.3 % FeO (Sandecki 2000). Within the ore are plagioclase-rich layers which give the appearance of a layered igneous intrusion (Hjelmqvist 1950, Sandecki 2000). The ore is supposed to have been formed as a magmatic cumulate which resulted from gravitational settling within a gabbroic magma (Sandecki 2000, Larsson & Söderlund 2005). The sample in this study is a massive Ti-magneite ore from the mine at Taberg and was taken from the mineral collection at the Department of Earth Science at Uppsala University.

Ulvön: The name Ulvöspinel for Ti-rich magnetite is derived from the layered igneous Ulvö Gabbro Complex (Mogensen 1946), located at the East coast of central Sweden, where the mineral was first described. The intrusion consists of several gently dipping lopoliths, 30-80 km in diameter and 250-300 m in thickness (Magnusson & Larson 1977, Lundqvist 1990). These gabbroic lopoliths contain alternate bands of mafic and more felsic layers, with thicknesses between 0.5 cm and 1 m, likely a result of magmatic cumulus processes (Larson 1973, Larson et al. 2008). The rocks of this area are ~1.25 Ga old and have not been affected by regional metamorphic overprint or deformation (Lundqvist 1990, Larson et al. 2008). Ti- magnetite occurs in distinct layers like for example in the Norra Ulvön Gabbro’s rhythmically layered zone which contains up to 10 cm thick bands with >50 % Fe-Ti oxides (Larson & Magnusson 1976, Larson et al. 2008). Such layers have also been mined for their metals (Larson et al. 2008). Other common minerals in the Ulvö Gabbro Complex are plagioclase (labradorite), olivine and clinopyroxene (Magnusson & Larson 1977, Larson et al. 2008). The sample used in this study is a massive Ti-magnetite ore from Norra Ulvön and was supplied from the mineral collection of the Geological Survey of Sweden.

Ruoutevare: The geology of the Ruoutevare area in Norrbotten in northern Sweden comprises ultrabasic rock types such as peridotite and pyroxenite which are associated with 20 anorthosite and gabbro of pre-Cambrian age (Stigh 1982). Associated with the gabbro intrusion is a deposit of iron ore in the form of Ti-bearing magnetite layers with a grade of 54.2 % Fe (Landergren 1948).The magnetite sample used in this study comes from this particular iron ore deposit and was taken from the mineral collection of the Geological Survey of Sweden.

3.2.4 Geological background of international magnetite samples

3.2.4.1 Geology of El Laco (Chile)

The apatite-iron oxide ores of El Laco are situated in Northern Chile at the flank of the Quaternary Pico Laco volcanic complex (Nyström et al. 2008). The area around El Laco hosts seven different deposits distributed over 30 km2 with a total amount of 500 Mt of high grade ore (Nyström et al. 2008). The ore consists mainly of magnetite, however, hematite is also present as an oxidation product. Fission track dating of apatite crystals within the El Laco ore gave an age estimation of 2.1± 0.1 Ma (Maksaev et al. 1988). The host rocks of the deposit consist of typical subduction zone andesites and dacites which have been hydrothermally altered, with alteration increasing at depth (Nyström et al. 2008, Naranjo et al. 2010, Velasco & Tornos 2012). Although hydrothermal activity was associated with the ore formation (Nyström et al. 2008, Naranjo et al. 2010), the ore at El Laco is supposed to have formed by dominantly magmatic processes, involving an iron oxide-rich magma and magmatic fluids (Henriquez & Nyström 1998). Textural analysis indicates that the ore resembles intrusive and extrusive magmatic activity, such as lava flows, pyroclastics and dykes, where the lava flow deposits are notably dominated by hematite (Nyström et al. 2008, Naranjo et al. 2010). A magmatic origin is also supported by features of lava bombs, aa and pahoehoe lavas as well as vesicle-like cavities (Fig.7) in addition to geochemical data from oxygen isotope analysis (Henriquez & Nyström 1998,

Nyström et al. 2008). However, a Fig.7 Vesicle-like structures in massive magnetite ore hydrothermal origin is put forward at El Laco, Chile. The ore is interpreted to have been a on the basis of oxygen isotopes former lava flow (Source: Henriquez et al. 2003).

(Rhodes & Oreskes 1999) arguing that the surprising isotopic homogeneity of magnetites at El Laco (~ +4.0 ‰) could not result from magmatic processes, where a wider range of values would be expected due to magma cooling and hydrothermal processes associated with volcanic activity. Instead, these authors propose an origin by hydrothermal replacement with some high-δ18O hydrothermal fluid as transport agent, possibly also in some sort of evaporitic pan. The samples for this study represent massive magnetite ore from the Laco Sur apatite- iron oxide ore deposit at El Laco and have been donated by Prof. Erik Jonsson from the Geological Survey of Sweden and by Dr. Jan Olov Nyström from the Swedish Museum of Natural History in Stockholm.

3.2.4.2 Layered Igneous Intrusions

Bushveld: The magnetite sample in this study comes from the Rustenburg Layered Suite Fig.8 A magnetite layer of the Upper Zone at (RLS) of the Bushveld complex, which is an Magnet Heights in the Eastern Bushveld 8 km thick succession of layered mafic and complex, ZA (Source: http://www.uct.ac.za/). ultramafic rocks with an age of about 2.1 Ga (Harney & Gruenewaldt 1995, Walraven et al. 1990). The RLS is divided up into the Lower, Critical, Main and Upper zone, with the Critical Zone being the economically most important one since it holds the world’s largest chromite and platinum-group element deposits (Saager 1984). However, magnetite is mined within the Upper Zone, which contains about 20 m in total thickness of pure magnetite in the form of several magnetite layers (Fig.8) within 2 km thick magnetite-bearing gabbroic rocks (Cawthorn & Molyneux 1986). The magnetite layers vary in thickness between 0.1 and 10 m and contain some silicates, mostly plagioclase feldspar (Harney & Gruenewaldt 1995). The most prominent layer is called the Main Magnetite Layer from which the magnetite used in this study originates. The magnetite layers in the Upper zone are of magmatic origin and are assumed to have been formed by cycles of magma mixing of different FeO-rich magmas and subsequent cumulate emplacement (Irvine & Sharpe 1986). The Bushveld sample in this study was donated by Prof. Nicholas Arndt from the University on Grenoble.

Panzhihua: The layered igneous intrusion of Panzhihua is located in the Panxi Mining District, Sichuan Province, in South West China and is part of the Emeishan Large Igneous Province. It is a relatively unmetamorphosed and undeformed, 2 km thick, sill-like gabbroic intrusion which dips about 50°-60° towards the NW and extends about 19 km from NE to SW 22

(Tang 1984, Zhou et al. 2005). The intrusion is concordantly emplaced within late Neoproterozoic dolomite limestones, Permian syenites and Triassic shales and coal measures and is itself 263 Ma old (Zhou et al. 2005). The intrusion is divided into four zones based on differences in internal structure and iron oxide mineralizations. These four zones are the marginal, lower, middle and upper zone. Iron ore occurs in the lower and middle zones. The mineralizations consist of both, massive lens-shaped or tabular ore bodies up to 60 m in thickness as well as disseminated ore. The massive ore bodies consist of about 80% Ti- magnetite with a weight-percentage of 43% for iron oxide and 12% titanium oxide (Zhou et al. 2005). On the basis of texture (e.g. vesicles), geochemistry and the absence of evidence for a skarn origin, the mineralization is interpreted to have formed from an oxide enriched melt (Tang 1984). The Panzhihua deposit is currently mined and holds a reserve of 1333 Mt of ore (Ma et al. 2003). The two Panzhihua samples of this study represent massive Ti-magnetite ore and were donated by Prof. Nicholas Arndt, Grenoble University.

3.2.4.3 Volcanic reference materials

Indonesia: The investigated magnetites come from the Anak Krakatau, Agung, Gede, Kelut and Merapi volcanoes on the Indonesian Islands of Java and Bali (Fig.9). These are part of the Western Sunda-Banda arc, which developed during the Cenozoic through subduction of the Indian-Australian plate under the Eurasian plate (Carlile & Mitchell 1994, Marcoux & Milési 1994). Calc-alkaline volcanism with dacites and andesites are the typical eruption products in most recent times (Van Bemmelen 1949, Marcoux & Milési 1994). Beside volcanic rocks, the

Fig.9 The Indonesian islands of Java and Bali. Shown are the locations of the volcanoes from which magnetite samples were used in this study (Source: Google Earth).

23 area also hosts several gold, tin and copper deposits associated with subduction zone volcanism (Marcoux & Milési 1994). The samples were provided by Prof. Valentin Troll and Ester Muñoz-Jolis at Uppsala University.

New Zealand: Mount Ruapehu is a 2797 m high stratovolcano at the southern end of the Taupo Volcanic Zone (TVZ) on the North Island of New Zealand (Gamble et al. 2003). It is the largest and currently active volcano on the Northern Island with the most recent eruption in 2007 and several other eruptive events during the last hundred years (Jolly et al. 2010, Price et al. 2012). Volcanic activity in the TVZ is associated with the subduction of the Pacific Plate beneath the Australian Plate along the Hikurangi-Kermadec Trench system (Gamble et al. 1999, Stern et al. 2010). Mt. Ruapehu is underlain by Mesozoic meta- greywacke, which in turn is underlain by oceanic, metamorphosed igneous crust (Price et al. 2010, 2012). The typical eruption products are subduction zone andesites and dacites of porphyritic texture (Graham & Hackett 1987, Price et al. 2005). The magnetite content of the volcanic rocks at Mt. Ruapehu varies between less than 1 % and up to 6 % and is just over 1% on average (Price et al. 2012). Samples used in this study are two dacite rocks that come from the southern Flank of Mt. Ruapehu and have been donated by Prof. John Gamble from Victoria University in Wellington, NZ.

Canary Islands: Tenerife located in the centre of the Canary archipelago is the largest (2058 km2) and highest (3718 m) of the island group which is situated over the Canary hot spot (Schmincke & Flower 1974, Ancochea et al. 1990). Volcanic activity on the island dates back to about 6.5 Ma and is today seen in several small volcanoes and the Teide-Pico Viejo edifice (Ancochea et al. 1990). Samples for this study were donated by Prof. Valentin Troll and Dr. Frances M. Deegan at Uppsala University and originate from dykes in the NE rift zone on the island. The samples comprise ankaramites and basanites (Deegan et al. 2012).

Iceland: The volcanic island of Iceland is located directly over the point in the North Atlantic, where asthenospheric flow interacts with a deep seated mantle plume (Trønes 2003), whose current plume channel lies beneath the Vatnajökull glacier and represents the plate boundaries of the Mid-Atlantic ridge (Wolfe et al. 1997). Extensive volcanism is common on the island with more than 18 active volcanoes, which are often associated with rift zones and their volcanic fissure swarms (Trønes 2003). Volcanic eruptions, often of explosive nature due to lava-snow interaction, occur every three to four years (Gudmundsson et al. 2010). The main eruption products are tholeiitic basalts as well as basaltic andesites (Sæmundsson 1979). The magnetite sample used in this study comes from a basaltic lava bomb erupted from the 24

Skjaldbreiður volcano in SW Iceland and was supplied via the Uppsala University rock collection.

Cyprus: The island of Cyprus is located in a zone of underthrusting in the Eastern part of the Mediterranean sea, where the African plate is being pushed into the Eurasian plate. It can be divided into five more or less parallel belts, which trend approximately eastwards and are convex towards the South (Gass & Masson-Smith 1963). Four of these belts are dominated by sedimentary rocks, most commonly limestones, and loose sediments ranging in age from the Triassic to recent (Henson et al. 1949, Gass & Masson-Smith 1963). The fifth belt, the Troodos igneous massif, is dominated by mafic and ultramafic igneous rocks and represents an ophiolite sequence obducted during the alpine orogeny (Gass & Masson-Smith 1963, Allerton & Vine 1991). The Troodos massif is about 11 km thick and divided up into basic and ultrabasic rocks in the center, the sheeted intrusive complex and the peripheral pillow lavas (Gass & Masson-Smith 1963). The rocks have been affected by metamorphism represented by diabase and serpentinite (Gass & Masson-Smith 1963). The sample used in this study comes from a dolerite dyke near Agros in the central part of the Troodos complex (Stillman 1989) and was donated by Prof. Christopher Stillman from Trinity College, Dublin.

4. Analytical Methods

4.1 Sample preparation and analysis

Fig.10 The Laser Fluorination Laboratory at the Department of Geology, University of Cape Town, South Africa, where the presented oxygen isotope data were processed.

The selected samples were all crushed using a hammer or a jaw crusher before separation into different grain size populations. After washing in an ultrasonic bath until no traces of dust remained, samples were dried for 24 hours at 100°C. Separation of magnetite grains or magnetite containing mineral fractions was accomplished with a magnet. The magnetic separation was applied several times until almost no impurities (i.e. silicate minerals) remained in the fraction. The samples were then examined under a stereo microscope in order to identify any ‘silicate contaminants’ and remove those from magnetites. For each sample, 5- 10 milligrams of magnetite were separated for final analysis. For three samples quartz crystals were also separated from the silicate residue as well. Due to restricted sample sizes it was not always possible to obtain a complete set of oxygen isotopes for all the samples, especially the volcanic reference materials, as clean material could not be obtained in sufficient amounts for analysis in all cases.

Oxygen isotope analysis was carried out at the University of Cape Town (South Africa) using a Finnigan DeltaXP dual inlet gas source mass spectrometer. For the oxygen analysis the quartz and magnetite samples were prepared by laser fluorination (Fig.

10 & 11; see Harris & Fig.11 Remains of evaporated magnetite samples in a sample Vogeli 2010) whereby holder after the laser fluorination process at Cape Town University. they were reacted with

10 kPa of BrF5 and the purified O2 was collected onto a 5 Å molecular sieve in a glass storage bottle. As a reference and calibration standard, Monastery garnet was used. All oxygen data was recorded in the usual δ18O notation relative to SMOW.

For the iron isotope analysis of the individual magnetite samples, which was done at the Victoria University in Wellington, New Zealand, the crystals were digested and chemically 57 purified with concentrated HF and HNO3 acid and the analysis was then done using a Fe– 58Fe double spike and a Nu Plasma MC-ICP-MS (Multicollector-Inductively Coupled Plasma Mass Spectrometer). Full details on the Fe-isotope analysis are given in Millet et al. (2012). All iron isotope data was recorded as δ56Fe, which is deviation of 56Fe/54Fe relative from the IRMM-014 standard material.

4.2 Possible analytical errors

For the first part of the project a yield was calculated for each oxygen isotope analysis. The yield indicates the quality of the analysis, meaning that it provides information on the accuracy of the obtained isotope data. The yield was calculated with the help of the Monastery garnet as a standard, the sample weight, and the amount of μmol/mg of oxygen for each mineral (i.e. the standard and the analyzed sample). A yield of precisely or close to 100 % indicates that the analyzed sample was almost fully reacted and pure. Possible errors can originate from impurities which change the amount of oxygen in the analyzed magnetite 27 samples (e.g. contamination by minerals such as calcite or quartz) or the occasion that not all the sample was reacted during the fluorination process which would change the original sample weight of the analyzed sample. This latter problem arose when the view into the reaction chamber during the laser ablation was blocked because reaction products were fogging up the covering window. Thus difficulties arose to ablate all of the mineral grains in the individual chambers of the sample holder. The loss of sight during the fluorination process occurred with different intensities. During most sample-set analyses (12 samples) the covering window was clouded after the laser ablation of the eighth sample. The analytical uncertainties for oxygen isotopes are ±0.2 ‰.

The yields obtained for the oxygen isotope analyses varied from 54.3 % up to 122.2 % (n=42) with the majority lying between and close to 90 % and 100 % (n=31) and thus the data quality seems reasonably good. The obtained oxygen data as well as the yields obtained for the analyses are shown in Table 3.

Possible analytical errors for iron isotope analysis could have mostly come from silicate impurities which were digested in acid together with the magnetite. However, analytical error arising from this problem seems less likely, since magnetite compared to most silicate minerals represents a sink for iron and thus the small amount of iron coming from impurities would be diluted. In addition, the 57Fe–58Fe double spike method is one of the most precise analytical methods and accounts for various types of analytical errors (e.g. magnet drift) producing much smaller standard errors compared to other iron isotope analytical techniques (e.g. sample standard bracketing; detailed information given in Millet et al. 2012). For the iron isotopes the uncertainties vary from ±0.027 ‰ to ±0.051 ‰ but are on average about ±0.03 ‰.

5. Background on stable isotope systematics and data evaluation methodology

5.1 Stable isotopes and the δ-value

Isotope geochemistry, including stable and radiogenic isotopes, is a very powerful tool in geosciences. They find their use in age determination and petrogenesis, in evaluating certain surface processes like weathering and even in assessment of climate change. While radiogenic isotopes are concerned with nuclear processes, stable isotope geochemistry focuses on the variations of isotopic composition due to physicochemical processes. Stable isotopes can be used as a tracer for identifying the source of a rock or mineralization as well as a geothermometer (Rollinson 1993, Hoefs 1997). The isotopes of oxygen and iron, which will be used for this purpose in this work, are the pairs of 18O/16O and 56Fe/54Fe. The relative abundance of the heavier isotope in a rock, mineral or fluid is usually given by the δ-value in units of parts per mil (‰). The δ-value, here shown with the example of oxygen, is given as:

Eq.1 δ18O = x 1000

(Mason & Moore 1982, Rollinson 1993, Hoefs 1997).

The standard value for oxygen is the Standard Mean Ocean Water value (SMOW) which has 18O/16O = 2005.2 +/- 0.43 and a δ18O of 0 ‰ (Hoefs 1997). Other standard values for oxygen isotopes used are the V-SMOW (Vienna-Standard Mean Ocean Water) or the PDB (Pee Dee Belemnite) but most commonly used in hard rock geology is the SMOW value (Hoefs 1997) and conversion between the three reference schemes is possible. The standard value used for iron isotopes is the isotopic reference material IRMM-014 based on the mole fractions of 57Fe, 56Fe and 54Fe in elemental iron which are equal to 0.021191, 0.91754 and 0.05845 respectively (Coplen et al. 2002). The purpose of the δ-value is to show whether a substance is either enriched (positive value) or depleted (negative value) in a heavy isotope (Rollinson 1993). The process which is responsible for enrichment or depletion is called stable isotope fractionation.

5.2 Stable isotope fractionation

The process of isotope fractionation means “the partitioning of isotopes between two substances or phases of the same substance with different isotope ratios” (Hoefs 1997). As a

29 consequence one of the substances can become enriched in the heavier isotope while the other is left depleted. The fractionation is expressed by the fractionation factor α which is given as

Eq.2 α1-2 =

(Javoy 1977, Hoefs 1997).

R represents the ratio heavy isotope/light isotope in each substance involved in the fractionation. Experiments showed that the fractionation can furthermore be expressed with respect to temperature. This resulted in an equation showing that

6 2 Eq.3 1000 ln α1-2 = A (10 /T ) + B (Hoefs 1997).

Here T is the absolute temperature in Kelvin. A and B are thermometric coefficients that were derived by experiments for certain mineral pairs (Javoy 1977, Chiba et al. 1989). Equation 3 shows that the isotope fractionation is temperature dependent. With T increasing, the right hand side trends towards zero and thus α gets closer to one, and this means that the fractionation will eventually stop at a certain temperature.

Moreover there is a link between the δ-values of two substances and the fractionation between them. This relationship is dependent on the size of the values and is defined as follows:

Eq.4 Δ1-2 = δ1 - δ2 ≈ 1000ln α1-2 for δ < 10 ‰

δ Eq.5 α1-2 = for δ > 10 ‰ δ

(Rollinson 1993, Hoefs 1997).

This relationship makes it possible to use the δ18O-values of two substances, such as for example a mineral and fluids, to calculate fractionation temperatures or if the fractionation at a certain temperature is known to calculate possible δ18O-values for an equilibrium substance or medium. Both ways have for example been employed in the studies of Moore & Modabberi (2003), Nyström et al. (2008) and Jonsson et al. (2013) to calculate possible magmas, fluids and temperatures from and at which ore magnetite could have formed. The same approach has been adapted for this study.

Beside the temperature effect on the fractionation, other physical properties lend their influence. Bond strength, ionic potential and oxidation state bias the isotope exchange. A general rule is that the heavier isotope goes with the stronger bond (Rollinson 1993, Hoefs 30

1997). For example quartz, with the highly charged Si4+ ion will be enriched in 18O compared to magnetite with the larger and lower charged Fe2+ ion (Rollinson 1993), the reason for this being a higher ionic potential leading to a stronger bond. The small relative mass difference between the iron stable isotopes leads to small fractionations during simple chemical processes (Anbar et al. 2000). For iron the oxidation state is the main factor for the incorporation of the heavier isotopes (Severmann & Anbar 2009). The higher oxidation state together with a lower coordination number leads to stiffer and stronger bonds which preferentially incorporate heavier iron isotopes in order to reduce the total vibrational energy of the system of which they are part (Sahar et al. 2008). A good example for this matter is the mineral magnetite. Magnetite has a 1:2 ratio of Fe2+:Fe3+ and Fe3+ in tetrahedral coordination. In comparison, other silicates may have Fe2+ in octahedral coordination such as for example the olivine-type fayalite (Sahar et al. 2008). The higher oxidation state and smaller coordination number of iron thus leads to stronger bonds and thus to enrichment in heavy iron isotopes in magnetite compared to fayalite. What is also important is that, regarding magmatic processes, more evolved magmas show a heavier isotope composition (Hoefs 1997, Sahar et al. 2008). More evolved silicate magmas (dacite to rhyolite) with high alkali contents are more capable of structurally stabilizing ferric iron (Fe3+) which, in turn, favours enrichment in the heavy isotope (Dickenson & Hess 1986, Mysen 1988, Gaillard et al. 2001). Lastly, the more mobile and isotopically lighter ferrous iron (Fe2+) is preferentially removed from magmas by volatile loss (Heimann et al 2008). In addition, more silicic magmas at lower temperatures (< 950°C) show larger fractionation factors for iron isotopes and minerals than less evolved magmas at higher temperatures (Heimann et al. 2008, Schüssler 2008).

6. Results

6.1 Oxygen and iron isotope data of different magnetite ores and reference materials

For the first part of the project, a total of 89 analyses have been conducted, with a focus on ore formation. This number includes 42 analyses for oxygen and 47 analyses for iron isotopes. All obtained isotope data, the analytical uncertainties and the yields obtained for oxygen isotope analysis are presented in Table 3.

6.1.1 Oxygen isotopes

The massive GMD magnetite ores show values between +0.2 ‰ and +2.8 ‰ for oxygen isotopes while the vein, disseminated and waste pile material has much lower values in the range of -1.1 ‰ to +0.1 ‰ (Fig.12). The quartzes taken from the massive iron ores have high δ18O-values of +7.6 ‰ to +8.7 ‰. For the volcanic reference material only four oxygen isotope values were obtained, which lie in or near Taylor’s (1967) range of magmatic magnetites (δ18O = +1.0 ‰ to +4.0 ‰) showing δ18O of +3.6 ‰ to +4.6 ‰. The apatite-iron oxide ores from Kiruna and El Laco show a much bigger range in oxygen values than the GMD ore samples ranging in values between -4.3 ‰ and +4.4 ‰ (Fig.14). Ores from the

Fig.12 The distribution of oxygen and iron isotope values for massive and additional GMD magnetite. The samples show much less variation for iron than for oxygen isotopes. Reference fields for common igneous magnetite isotope data are shown and were taken from Taylor (1967) and Heimann et al. (2008).

32 layered igneous intrusions, in turn, show again positive δ18O-values between +1.78 ‰ and +4.84 ‰. The hydrothermal/low-temperature ore reference materials, including those from Dannemora and Striberg, show in majority very low values for oxygen, with only one exception and their δ18O-values range from -1.2 ‰ to +2.1 ‰.

6.1.2 Iron isotopes

The iron isotopes for massive GMD magnetite ores show with few exceptions a smaller dynamic range than the oxygen isotope system with δ56Fe-values of +0.26 ‰ to +0.9 ‰ (Fig. 12, Fig.15). The additional vein, disseminated and waste pile materials show values between - 0.02 ‰ and +0.33 ‰. Apatite-iron oxide ores from Kiruna and El Laco comprise values similar to the GMD ores that range between +0.19 ‰ and +0.36 ‰. The Dannemora magnetite samples and the BIF are depleted in the heavy isotope and show δ56Fe-values from -0.57 ‰ to +0.01 ‰. Layered igneous intrusions as well as the volcanic reference materials show enrichment in 56Fe with δ56Fe-values ranging from +0.11 ‰ to +0.61 ‰ and +0.06 ‰ to +0.46 ‰ respectively. Notably, most δ56Fe-values of the analyzsed magnetites in this study lie within the range of regular igneous magnetites (+0.06 ‰ to +0.49 ‰), reported by Heimann et al. (2008).

Table 3. The dataset from the magnetite oxygen and iron isotope analysis. Sample Rock type δ18O in ‰ 2σ Yield in % δ 56Fe in ‰ 2σ GMD DC 717

KES090068 Massive magnetite ore 1.2 ±0.2 110.5 0.4 ±0.029 KES090070 Massive magnetite ore 1.8 ±0.2 62.7 0.24 ±0.028 KES090072 Massive magnetite ore 0.9 ±0.2 102.5 0.33 ±0.031

GMD DC 690

KES090011 Massive magnetite ore 2.8 ±0.2 89.8 0.31 ±0.031 KES090012 Massive magnetite ore 1.2 ±0.2 94.0 0.31 ±0.037 KES090020 Massive magnetite ore 1.1 ±0.2 100.8 0.30 ±0.038 KES090024 Massive magnetite ore 1.0 ±0.2 99.3 0.26 ±0.035 KES090027 Massive magnetite ore 1.2 ±0.2 98.5 0.29 ±0.032 KES090030 Massive magnetite ore 1.8 ±0.2 102.9 0.39 ±0.039 KES090034 Massive magnetite ore 0.5 ±0.2 100.9 0.27 ±0.035 KES090034 qtz Massive magnetite ore 8.7 ±0.2 80.3 - -

GMD DC 575

KES103003 Massive magnetite ore 0.2 ±0.2 100.2 1.0 ±0.034 KES103003 qtz Massive magnetite ore 7.6 ±0.2 54.3 - - KES103011 Massive magnetite ore 1.8 ±0.2 122.2 0.31 ±0.034 KES103016 Massive magnetite ore 1.5 ±0.2 105.8 0.27 ±0.038

Table 3. Continued KES103016 qtz Massive magnetite ore 7.9 ±0.2 77.9 - -

GMD additional

KES091013b Waste pile material 0.1 ±0.2 72.3 0.33 ±0.029 KES091007B Waste pile material -0.8 ±0.2 73.3 -0.02 ±0.032 KES090044 Disseminated magnetite -1.0 ±0.2 99.6 0.24 ±0.027 KES090084 Magnetite vein -1.1 ±0.2 99.1 0.11 ±0.028

Dannemora

DM-1 Iron-skarn magnetite ore -0.4 ±0.2 94.1 -0.36 ±0.030 DM-2 Iron-skarn magnetite ore -0.7 ±0.2 88.1 0.01 ±0.031 DM-3 Iron-skarn magnetite ore 2.1 ±0.2 91.6 -0.43 ±0.033 DM-4 Iron-skarn magnetite ore -0.6 ±0.2 94.8 -0.35 ±0.028

EL Laco

EJ-LS-11-1 Massive magnetite ore 1.9 ±0.2 96.0 0.28 ±0.035 EJ-LS-11-2 Massive magnetite ore -4.3 ±0.2 106.9 0.24 ±0.032 EJ-LS-11-3 Massive magnetite ore -1.9 ±0.2 97.0 0.36 ±0.034 EJ-LS-11-4 Massive magnetite ore 4.2 ±0.2 85.8 0.34 ±0.032 LS-2 Massive magnetite ore 4.3 ±0.2 96.4 0.27 ±0.035 LS-52 Massive magnetite ore 4.4 ±0.2 88.0 0.28 ±0.034

Kiruna Massive magnetite ore 4.1 ±0.2 116.2 0.19 ±0.030 Ruoutevare Massive magnetite ore 3.2 ±0.2 97.1 0.31 ±0.033 Ulvön Massive magnetite ore 4.0 ±0.2 102.4 0.13 ±0.032 Taberg Massive magnetite ore 4.1 ±0.2 115.9 0.23 ±0.040 EJ092008 BIF magnetite -1.2 ±0.2 102.7 -0.57 ±0.033 EM419 Massive magnetite ore 4.8 ±0.2 91.8 0.61 ±0.051 EM424 Massive magnetite ore 2.8 ±0.2 75.0 0.12 ±0.036 Bushveld Massive magnetite ore 1.8 ±0.2 96.9 - -

Volcanic Reference Material TEF-NER-18 Ankaramite - - - 0.07 ±0.046 TEF-NER-57 Ankaramite 3.7 ±0.2 89.0 0.16 ±0.024 TEF-NER-70 Ankaramite 3.7 ±0.2 94.3 0.10 ±0.022 MG-07 Dacite - - - 0.32 ±0.032 MG-09 Dacite - - - 0.29 ±0.032 Kelut A1 Basaltic andesite - - - 0.10 ±0.037 GD-D-2 Basaltic andesite - - - 0.12 ±0.030 AK-B1 Basaltic andesite - - - 0.06 ±0.030 AK-B3 Basaltic andesite - - - 0.16 ±0.029 A-BA-1 Basaltic andesite - - - 0.18 ±0.046 M-BA06-KA-3 Basaltic andesite 3.9 ±0.2 88.1 0.17 ±0.029 Basaltbomb Basalt 4.6 ±0.2 96.5 0.46 ±0.031 83/CRS/6 Dolerite - - - 0.34 ±0.027

6.2 Isotope data for hydrothermal aureole identification

In total, 31 oxygen isotope analyses have been conducted on whole rock and quartz to test for the existence of a possible hydrothermal aureole in the GMD area. Table 4 and Figure 13 (next page) show the δ-values obtained for the samples. Figure 13 also indicates the location of each sample in the Grängesberg area. Covered is the direct vicinity of the ore body and the area to the SE the direction of the ore bodies’ dip. The results range from +5.8 ‰ to +10.3 ‰, but show no recognisable systematic spatial or geochemical grouping of values.

Table 4. Dataset for the hydrothermal aureole investigation of the project. Sample Rock type Quartz Whole rock 2σ

Sample 29 Granitic host rock 7.9 - - Sample 31 Gneissic host rock 6.7 - - Sample 40 Granitic host rock 9.3 - - Sample 45 Granitic host rock 8.3 - - Sample 48 Granitic host rock 8.4 7.9 ±0.2 Sample 49 Granitic host rock 9.2 7.8 ±0.2 Sample 52 Granitic host rock 8.7 6.9 ±0.2 Sample 53 Granitic host rock - 8.4 ±0.2 Sample 56 Granitic host rock - 7.4 ±0.2 KPN-090007 Dacitic host rock - 5.8 ±0.2 KPN-090026-4 Dacitic host rock 9.1 6.8 ±0.2 KPN-090033A Felsic volcanic rock 10.3 8.6 ±0.2 KPN-090042A Metavolcanic host rock 6 7.5 ±0.2 KHO-09003 Dacitic host rock - 7.7 ±0.2 KHO-09005 Dacitic host rock 7 5.8 ±0.2 KHO-090010 Metavolcanic host rock - 6.8 ±0.2 KHO-090011A Metavolcanic host rock - 6.8 ±0.2 KHO-090012 Dacitic host rock 6.7 7.0 ±0.2 KHO-090013 Andesitic host rock - 7.7 ±0.2 KHO-090013G Andesitic host rock - 6.0 ±0.2 KHO-090017 Metavolcanic host rock 8.8 - - KHO-09127b Dacitic host rock - 8.3 ±0.2 KHO-09129A Metavolcanic host rock 8,6 - -

roughlyshowsthe position theof orebody Fig.

The distribution of distribution The

the surface samples and their oxygen isotope composition around Grängesberg. The black ellipse Grängesberg.blackaround The compositionisotope theiroxygen and surfacesamples the

(Source:Google Earth)

7. Discussion

7.1 Determining ore formation processes

7.1.1 Comparison of magmatic and ore magnetites

In order to employ the data obtained and decipher the origin of the ore at Grängesberg (e.g. high-temperature magmatic vs. low-temperature hydrothermal), several approaches are possible. The first step is to compare the individual results of ore and volcanic reference materials with each other. Not all samples have a complete pair of oxygen and iron isotope data, and so the first observation will be independently taken on their iron and their oxygen isotope values. The available data can be divided into eight groups:

1. GMD apatite-iron oxide ores (massive ores) 2. GMD additional materials (non-massive, mine waste) 3. Apatite-iron oxide ore of Kiruna 4. Dannemora iron ore 5. Striberg BIF 6. Apatite-iron oxide ores of El Laco 7. Layered igneous intrusions (Swedish and international) 8. Volcanic reference material (international)

The distribution of the iron and oxygen isotopic composition of each group is shown in Figures 7 and 8.

7.1.1.1 Oxygen isotopes

Figure 7 shows the majority of the analysed magnetites to be enriched in the heavy oxygen isotope relative to SMOW as most of them plot on the right hand side of the figure i.e. above 0 ‰. The majority of the Layered Igneous Intrusions and the volcanic reference materials fall, as maybe expected, into the range for igneous magnetites previously proposed by Taylor (1967; +1.0 to +4.0 ‰). Samples from El Laco and Kiruna, the Layered Igneous Intrusions as well as volcanic magnetites extend in part to above the Taylor range. The majority of GMD samples (n=10 out of 13) also lie within the Taylor range and overlap with samples from El Laco and the Layered Igneous Intrusions Group. The overlapping values and the fact that most oxygen values are within or above the Taylor range imply that the formation of the GMD ore is dominantly related to magmatic processes. In fact, magnetites with a δ18O ≥ +1.0 ‰ are of the ortho-magmatic type (Fig.14). There is a notable distinction between the GMD 37 magnetites and the “non-igneous” magnetites from the Striberg BIF and the Dannemora skarn-deposit. These latter two plot to the left in Figure 14 and thus show a depletion in 18O. The low-δ18O GMD samples, the GMD additional material (vein and disseminated ore as well as waste pile material) and the two El Laco ore samples that lie below the Taylor range show a depletion in 18O. They do not fit with typical “magmatic” values usually present in volcanic systems and must therefore reflect low-temperature processes such as hydrothermal precipitation. The fact that two samples from El Laco show negative δ18O-values underlines that there is a variation within oxygen isotope composition at El Laco, in contrast to the assertions made by Rhodes & Oreskes (1999).

Fig.14 The oxygen isotope signatures of the various magmatic and hydrothermal magnetites investigated in this study.

7.1.1.2 Iron isotopes

Figure 15 shows that the observed overlap in oxygen isotope values between the Layered Igneous Intrusion Group, the volcanic reference materials and the apatite-iron oxide ores is even more pronounced for iron isotopes. The match between GMD and El Laco is noteworthy. The iron isotope composition of the ore samples from the GMD, El Laco, Kiruna and the Layered Igneous Intrusion Group is virtually identical with those of the volcanic reference materials, and thus a strong link between magmatic processes and ore formation is 38 suggested. The “non-magmatic” ores from Dannemora and Striberg (cf. Lager 2001, Stephens et al. 2009), in turn, have a different isotope composition that is inconsistent with an ortho- magmatic origin in the sense that they are depleted in the heavy isotope. Most hydrothermal ores and hydrothermal precipitates show δ56Fe-values below 0.0 ‰ or an extreme enrichment in the case of some hematite ores (Anbar 2004, Markl et al. 2006, Severmann & Anbar 2009). The reason for this depletion is the high mobility of ferrous (Fe2+) iron, which, in turn, is partial to being isotopically light. Most ore and volcanic magnetites of this study, in turn, lie between δ56Fe-values of +0.1 ‰ and +0.5 ‰ and the lowest δ56Fe-value of the volcanic reference material is +0.06 ‰. Thus, samples above +0.06 ‰ are consistent with having likely formed from ortho-magmatic (high-temperature) processes. Since the GMD and other apatite-iron oxide ores show values above 0.1 ‰, a hydrothermal (i.e. low-temperature) precipitation is not a likely formation process for these ores. Interestingly, the iron isotopes of the additional GMD vein, disseminated and waste pile material as well as the El Laco ore, outside the ortho-magmatic field for oxygen isotopes, are, except for one sample, within the ortho-magmatic field for iron isotopes.

Fig.15 The iron isotope composition of various magmatic and hydrothermal magnetites of this study. 39

7.1.1.3 Bimodal formation

Most oxygen and iron data for apatite-iron oxide ore samples presented here show values that correspond to magmatic processes. However, for a small number of samples oxygen and iron isotopes seem to contradict each other. This difference, however, can be explained when a bimodal formation of the ore is assumed. Previous studies on the GMD ore (Weis 2011, Jonsson et al. 2013) propose an ore formation by high-T ortho-magmatic processes and a local secondary low-T hydrothermal re-deposition of iron. Assuming that an originally ortho- magmatic ore body is leached by a hydrothermal fluid from which then magnetite is re- deposited afterwards at much lower temperatures, the isotopic composition and fractionation would have a much greater effect on the oxygen isotope composition than on that of iron. Taking a low-temperature fluid, such as for example meteoric or surface water to leach the magnetite at depth, the δ18O of the resulting re-deposited ore would be heavily depleted since fractionation between magnetite and water is more intense at lower temperatures. This effect would be enhanced by fractionation between magnetite and a low-δ18O fluid. The same, however, would not hold true for iron isotopes, since the iron composition of the ore would be preserved in a hydrothermal fluid, which compared to the magnetite holds less iron than oxygen and thus shifts the iron isotopic composition of the new formed fluid to higher values (cf. Beard et al. 2003, Beard & Johnson 2004, Markl et al. 2006, Roonasi & Holmgren 2009). Preservation of the original iron isotope composition then occurs since quantitative leaching of iron oxides does not entail fractionation (Markl et al. 2006). More oxidising conditions would also hinder the formation of mobile and isotopically light ferrous iron. Thus, besides ferrous iron, ferric iron (Fe3+) could also be leached and re-deposited in the direct vicinity of an ortho-magmatic ore body, hence preserving the original ortho-magmatic iron isotope signature. Such hydrothermal leaching and re-deposition and the extreme oxidation associated with it would likely lead to the formation of hematite, which indeed is present at Grängesberg. A similar scenario has been proposed for the El Laco apatite-iron oxide deposit by Velasco & Tornos (2012) and by Moore & Modabberi (2003) for the Choghart apatite-iron oxide deposit in Iran.

Sample KES091007B, an ore from Björnberget near the GMD field, is the only sample that does not fit into the proposed re-depositional scenario since not only the oxygen but also the iron isotope values are negative. The formation of this sample probably involved deposition from a hydrothermal source enriched in more mobile and isotopically light Fe2+ (c.f. Anbar 2004, Severmann & Anbar 2009). The isotope composition of this sample is more in

40 agreement with the data seen for e.g. the skarn-samples from Dannemora and since KES091007B is a calcite-bearing magnetite ore, a similar origin (e.g. precipitation from an evaporitic pan, c.f. Lager 2001) may be suggested. The El Laco ore samples, for which Rhodes & Oreskes (1999) also propose an evaporitic formation scenario, do not show the same trend in the oxygen and iron isotope data as for example the presumably low- temperature Dannemora samples.

7.1.2 Calculation of possible ore sources

The second test for the processes involved in ore formation at GMD magnetites is the approach of e.g. Moore & Modabberi (2003), Nyström et al. (2008), Weis (2011) and Jonsson et al. (2013). This approach involves the use of isotope fractionation and a calculation of an equilibrium fluid or magma. The previous studies of oxygen isotopes recorded apatite-iron oxide ores at Grängesberg as well as Kiruna and El Laco in equilibrium with andesite magma or a high-T, high δ18O magmatic fluid. The same method can now be applied to the oxygen and iron isotope data obtained for this study. The required calculation builds on the relationship given by Equation 4 and is defined as follows:

Δmagnetite – source = δmagnetite - δsource ≈ 1000ln αmagnetite-source

 δmagnetite - 1000ln αmagnetite-source ≈ δsource

The potential ore sources for which iron and oxygen δ-values are re-calculated for generating a model are a) a magma and b) a high-T magmatic fluid. Since the host rocks to all three ore deposits (GMD, Kiruna and El laco) are of intermediate composition (dacite to andesite) (Allen et al. 1996, Evans 2000, Nyström et al. 2008), the fractionation factors for andesite/dacite magma are chosen. For the fluid calculation a temperature of 800°C, as proposed by Nyström et al. (2008), is assumed. The oxygen isotope fractionation factor between magnetite, andesite as well as dacite magma at normal magmatic temperatures (≥900

°C) is given by Zhao & Zheng (2003) and is Δmagnetite-magma= -4.0 ‰ and -4.3 ‰ respectively. The oxygen isotope fractionation for magnetite and water at 800 °C comes from O’Neil &

Calyton (1964), Blattner et al. (1983) and Zheng (1991) and states Δmagnetite-water = -5.2 ‰.

Finding an appropriate iron isotope fractionation factor between magnetite and andesite magma is more difficult since the values available seem controversial. A study from Huang & Lundstrom (2006) suggests magnetite being lighter than the original melt and a preliminary fractionation factor of Δmagnetite-magma = -0.26 ‰. However, no actual results were published

41 and further investigation of the matter by the authors shows the fractionation to be positive and that magnetite is in fact isotopically heavier than the magma it crystallized from. Another value for the fractionation between magnetite and an intermediate magma at 900 °C was published by Heimann et al. (2008), and is Δmagnetite-magma = +0.03 ‰. The same study provides a fractionation for magnetite and an aqueous Fe2+-chloride fluid which can be seen as the equivalent to the magmatic fluid. Such fluids do not only bear ferrous but also ferric iron in III - III 3+ the form of hexachloro (e.g. [Fe Cl4] ) and hexaquo (e.g. [Fe (H2O)6] ) complexes

(Schauble et al. 2001, Beard et al. 2003). The fractionation here is Δmagnetite-fluid = +0.24 ‰ at 800°C (Heimann et al. 2008).

As a reference for testing the calculated values, previously published values for oxygen and iron isotopic compositions of magmas and magmatic fluids are used. Andesite and dacite magmas should have a δ18O ≥ +5.7 ‰ (Taylor 1968, Bindemann 2008) and most intermediate to high silica igneous rocks have δ56Fe-values between 0.00 ‰ and +0.12 ‰, with few exceptions that show higher or lower values (Beard et al. 2003, Beard & Johnson 2004, Schüßler 2008, Heimann et al. 2008). Strong enrichment of igneous rocks in δ56Fe is possible, however, it is rather rare and mostly restricted to high-silica igneous rocks such as granites (e.g. Heimann et al. 2008). Isotope signatures of exsolved magmatic fluids are +6 to +8 ‰ for oxygen (Nyström et al. 2008) and -0.35 ‰ to -0.05 ‰ for iron (Heimann et al. 2008), respectively.

The fractionation of iron isotopes between magnetite and magma can be tested before being applied to the ore by using the volcanic reference samples which contain magnetites derived from intermediate magmas (n=8). The results of the re-calculation are shown in Table 5 and Fig.16. The majority of the re-calculated parent magmas (n=6) lies within or close to the proposed range of iron isotope compositions of intermediate igneous rocks (0.00 ‰ to +0.12 ‰) obtained from literature. Two of the samples are much higher but exceptions for stronger enrichment are possible due to 54Fe loss during degassing (see Heimann et al. 2008). This shows that the fractionation factor from Heimann et al. (2008) seems to be a good estimate for iron isotope exchange between magnetite and intermediate magmas and can thus be applied to the apatite-iron oxide ore magnetites. The results of the equilibrium re-calculation for the apatite-iron oxide ores (n=20) for both oxygen and iron isotopes are shown in Table 6.

Table 5. Magma Re-calculation for volcanic magnetites

Sample Sample description δ56Fe 2σ δ56Fe-Magma

MG-07, Ruapehu, NZ Igneous magnetite from dacite 0.32 ±0.032 0.29 MG-09, Ruapehu, NZ Igneous magnetite from dacite 0.29 ±0.032 0.26 Kelut A1, Kelut, Java Igneous magnetite from basaltic andesite 0.10 ±0.037 0.07 GD-D-2, Gede, Java Igneous magnetite from basaltic andesite 0.12 ±0.030 0.09 AK-B1, Anak Krakatau, Java Igneous magnetite from basaltic andesite 0.06 ±0.030 0.03 AK-B3, Anak Krakatau, Java Igneous magnetite from basaltic andesite 0.16 ±0.029 0.13 A-BA-1, Agung, Java Igneous magnetite from basaltic andesite 0.18 ±0.046 0.15 M-BA06-KA-3, Merapi, Java Igneous magnetite from basaltic andesite 0.17 ±0.029 0.14

Fig. 16 The results from the iron isotope magma-recalculation for volcanic-reference material. Extreme enrichment in 56Fe is rather rare but occurs.

Oxygen isotope values for the re-calculated magmas of apatite-iron oxide ores range between -0.3 ‰ and +8.4 ‰ for andesite and +0.0 ‰ and +8.7 ‰ for dacite (Table 6). Magnetites with a δ18O ≥ 1.78 ‰ (Grängesberg n=4, El Laco n=4, Kiruna n=1) are in equilibrium with both types of magma (Table 6, Fig. 10). Magma re-calculation for iron isotopes revealed δ56Fe- values from +0.16 ‰ to +0.36 ‰ for the same samples. These magma values are rather high compared to the proposed range of δ56Fe-values of intermediate magmas (0.00 ‰ to +0.12

‰), however this could be an indicator for ‘extreme’ enrichment processes of iron in some magmas supposing an isotopically heavy and oxide-rich melt. Such extreme enrichment and an oxide-rich melt could result from liquid immiscibility, i.e. the splitting of a silicate melt into two new phases. This process happens with increasing magma fractionation at high oxygen fugacities and temperatures at around 900-1000 °C and likely at shallow to moderate depth (Weidner 1982, Kolker 1982, Naslund 1983). New experiments showed that extreme enrichment of elemental iron in silicate melts, however, can already occur deep in the mantle (~1000 km) at pressures greater than 76 GPa (Nomura et al. 2011).

The results for the oxygen fractionation between magmatic fluid and magnetite showed that magnetites with δ18O ≥ +0.9 ‰ (GMD n=11, El Laco n=4, Kiruna n=1) are also in equilibrium with an aqueous magmatic fluid at 800 °C (Table 6), and the fluid-δ18O at +6.1 ‰ to +9.6 ‰. This range extends higher than the proposed range of magmatic fluids (6-8 ‰) but still lies close to the magmatic spectrum (Hoefs 1997, Nyström et al. 2008). The same magnetites are also in equilibrium with a Fe2+-chloride fluid, showing re-calculated fluid δ56Fe-values between -0.05 ‰ and +0.16 ‰. These values are above the proposed range of magmatic fluids (-0.35 ‰ to -0.05 ‰) and show that any fluids involved in the magnetite formation must have been enriched in the heavy isotope. Magmatic saline fluids are usually depleted in the heavy isotope due to concentrating the more mobile, isotopically light ferrous iron. The required enrichment in the heavy iron isotope, in turn, implies that the source of the magmatic fluid must also have been of heavy iron isotope composition. Thus, the most likely source is an oxide-rich magma with a high δ56Fe-value.

Table 6. Results for the magma/magmatic water equilibrium re-calculation δ18O Value δ18O Value δ18O Value δ56Fe δ56Fe Sample Rock type δ18O 2σ δ56Fe 2σ andesite fit dacite fit water fit water magma

KES090068 Massive ore 1.1 ±0.2 0.40 ±0.029 5.1 X 5.4 X 6.3  0.16 0.37 KES090070 Massive ore 1.8 ±0.2 0.24 ±0.028 5.8  6.1  7.0  -0.01 0.21 KES090072 Massive ore 0.9 ±0.2 0.33 ±0.031 4.8 X 5.2 X 6.1  0.09 0.30

KES090011 Massive ore 2.8 ±0.2 0.31 ±0.031 6.7  7.1  8.0  0.07 0.28 KES090012 Massive ore 1.2 ±0.2 0.31 ±0.037 5.2 X 5.5 X 6.4  0.07 0.28 KES090020 Massive ore 1.1 ±0.2 0.30 ±0.038 5.0 X 5.4 X 6.3  0.06 0.27 KES090024 Massive ore 1.0 ±0.2 0.26 ±0.035 5.0 X 5.3 X 6.2  0.02 0.23 KES090027 Massive ore 1.2 ±0.2 0.29 ±0.032 5.2 X 5.5 X 6.4  0.05 0.26 KES090030 Massive ore 1.8 ±0.2 0.39 ±0.039 5.8  6.1  7.0  0.15 0.36 KES090034 Massive ore 0.5 ±0.2 0.27 ±0.035 4.5 X 4.8 X 5.7 X 0.03 0.24

KES103003 Massive ore 0.2 ±0.2 1.00 ±0.034 4.1 X 4.5 X 5.4 X 0.76 0.97    44

Table 6. Continued. KES103011 Massive ore 1.8 ±0.2 0.31 ±0.034 5.7  6.1  7.0  0.07 0.28 KES103016 Massive ore 1.5 ±0.2 0.27 ±0.038 5.5 X 5.8 X 6.7  0.03 0.24

Waste pile KES091013b 0.1 ±0.2 0.33 ±0.029 4.1 X 4.4 X 5.3 X 0.09 0.30 material Waste pile KES091007B -0.8 ±0.2 -0.02 ±0.032 3.2 X 3.5 X 4.3 X -0.26 -0.05 material Disseminated KES090044 -1.0 ±0.2 0.24 ±0.027 3.0 X 3.3 X 4.2 X 0.00 0.21 Magnetite KES090084 Magnetite vein -1.1 ±0.2 0.11 ±0.028 2.8 X 3.1 X 4.1 X -0.13 0.08 EJ-LS-11-1 Massive ore 1.9 ±0.2 0.28 ±0.035 5.9  6.2  7.1  0.04 0.25 EJ-LS-11-2 Massive ore -4.3 ±0.2 0.24 ±0.032 -0.3 X 0.0 X 0.9 X 0.00 0.21 EJ-LS-11-3 Massive ore -1.9 ±0.2 0.36 ±0.034 2.1 X 2.4 X 3.3 X 0.12 0.33 EJ-LS-11-4 Massive ore 4.2 ±0.2 0.34 ±0.032 8.1  8.5  9.4  0.10 0.31 LS-2 Massive ore 4.3 ±0.2 0.27 ±0.035 8.2  8.6  9.5  0.03 0.24 LS-52 Massive ore 4.4 ±0.2 0.28 ±0.034 8.4  8.7  9.6  0.04 0.25 Kiruna Massive ore 4.1 ±0.2 0.19 ±0.030 8.1  8.4  9.3  -0.05 0.16    = in equilibrium with magma/magmatic water; X= not in equilibrium with common magmatic values

Magnetites with δ18O-values > 0.00 ‰ but < +0.9 ‰ (GMD n=3) do not show equilibrium with either magma or a magmatic fluid at ≥ 800 °C. The same is observed for these samples iron isotope values. These samples, however, may be compared to those magnetites published by Edfelt (2007) in a study on the apatite-iron oxide ore at Tjårrojåkka near Kiruna. Magnetite samples in Edfelt’s study are proposed to be derived from magmatic and hydrothermal fluids between 410 °C and 660 °C and show δ18O-values of -0.5 to +2.1 ‰. Testing the three GMD samples for the fractionation with magmatic fluid at a temperature of 625 °C (Table 7,

Δmagnetite-water = -6.2 ‰ and +0.35 ‰ for oxygen and iron respectively, O’Neill & Clayton 1964, Heimann et al. 2008), all three samples appear to be in equilibrium with a magmatic fluid at this temperature for oxygen isotopes, whereas for iron only one shows equilibrium (Table 7). The two samples that do not fall into equilibrium show a fluid δ56Fe-value above the proposed range for magmatic fluids (-0.35 ‰ to -0.05 ‰). One possibility for this is some sort of extreme enrichment of the heavy iron isotope involved into the fluid or, in the case of sample KES103003, the influence of leaching by a hydrothermal fluid that was preferentially removing ferrous iron (cf. Beard & Johnson 2004). The Blötberget sample’s (KES091013b) re-calculated ortho-magmatic origin is further supported by previous studies on the basis of REE analysis (see Jiao 2011).

Table 7. Fluid re-calculation for GMD magnetites at 625 °C Sample δ18O 2σ δ56Fe 2σ δ18O fluid δ56Fe fluid

KES090034 0.5 ±0.2 0.27 ±0.035 6.7  -0.08  KES103003 0.2 ±0.2 1.0 ±0.034 6.4  0.65 X KES091013b 0.1 ±0.2 0.33 ±0.029 6.3  -0.02 X

Table 8. Hydrothermal fluid re-calculation at 375°C Sample δ18O 2σ δ56Fe 2σ δ18O fluid δ56Fe fluid

KES091007B -0.8 ±0.2 -0.02 ±0.032 7.0 -0.52 KES090044 -1.0 ±0.2 0.24 ±0.027 6.8 -0.26 KES090084 -1.1 ±0.2 0.11 ±0.028 6.7 -0.39 EJ-LS-11-2 -4.3 ±0.2 0.24 ±0.032 3.5 -0.26 EJ-LS-11-3 -1.9 ±0.2 0.36 ±0.034 5.9 -0.14

Samples with δ18O < 0.0 ‰ (GMD n=3, El Laco n=2) would satisfy isotopic equilibrium with magmatic aqueous fluids at temperatures below 400°C (Δmagnetite-water < -7.6 ‰, Table 8), which is possible, however only at the very low end of the range of usual magmatic temperatures (Guilbert & Park 1986, Matthes 2001, Nyström et al. 2008). Thus these samples are more likely to be of hydrothermal re-depositional origin at low temperatures (< 400 °C) 18 56 56 with a fluid of high-δ O (≥ 3.5 ‰) and low-δ Fe composition (δ Fe < -0.2 ‰, Δmagnetite-water ≥ +0.5 ‰; typical hydrothermal fluids have δ56Fe -0.2 ‰ to -0.6 ‰, Svermann & Anbar 2009, Table 8). Such fluids could originate for example from remnant magmatic waters or from some high-δ18O meteoric or evaporitic water (c.f. Craig 1961, Rhodes & Oreskes 1999). The influence of low-δ18O fluids (< 3.5 ‰) would most likely have resulted in magnetites with even lower δ18O-values or would have required unrealistically high temperatures in order to prevent extensive isotope fractionation.

Fig.17 The composition of ore forming magmas and fluids at Grängesberg, Kiruna and El Laco compared to data available from the literature. The plot shows that ore forming magmas and fluids are enriched in 56Fe since the re-calculated values plot above the reference fields for intermediate magmas and magmatic fluids. The black line indicates the re-calculated magma δ56Fe-values for the volcanic reference materials (Table 5 & Fig.16) and shows that enrichment in 56Fe is considerable. The thicker part of the line represents the majority of re-calculated values.

To summarise the evaluation of the re-calculation, Figure 17 shows the isotopic composition of the re-calculated ore forming magmas and fluids compared to those presented in the literature. With only a few exceptions, magnetites from GMD (n=14), Kiruna (n=1) and El Laco (n=4) show equilibrium with at least one of the proposed ortho-magmatic sources (magma or magmatic fluid) for oxygen isotopes and enriched magmatic sources for iron isotopes. Only a minor fraction of GMD and El Laco magnetites (n=3 and 2 repsectively) show low-temperature (< 400 °C) equilibrium with hydrothermal fluids. The origin of the samples is summarised in Table 9 and Fig. 18. The coloured fields in Figure 18 were drawn according to the data evaluation.

Table 9. Overview of the different origins of apatite-iron oxide ore samples. δ18O δ56Fe Sample Sample description Origin Source Temperature in ‰ in ‰

KES090070 Massive magnetite ore 1.8 0.24 Ortho-magmatic Magma > 800°C KES090011 Massive magnetite ore 2.8 0.31 Ortho-magmatic Magma > 800°C KES090030 Massive magnetite ore 1.8 0.39 Ortho-magmatic Magma > 800°C KES103016 Massive magnetite ore 1.5 0.27 Ortho-magmatic Magma > 800°C EJ-LS-11-1 Massive magnetite ore 1.9 0.28 Ortho-magmatic Magma > 800°C EJ-LS-11-4 Massive magnetite ore 4.2 0.34 Ortho-magmatic Magma > 800°C LS-2 Massive magnetite ore 4.3 0.27 Ortho-magmatic Magma > 800°C LS-52 Massive magnetite ore 4.4 0.28 Ortho-magmatic Magma > 800°C Kiruna Massive magnetite ore 4.1 0.19 Ortho-magmatic Magma > 800°C

KES090068 Massive magnetite ore 1.1 0.40 Ortho-magmatic Magmatic fluid ~ 800°C KES090072 Massive magnetite ore 0.9 0.33 Ortho-magmatic Magmatic fluid ~ 800°C KES090012 Massive magnetite ore 1.2 0.30 Ortho-magmatic Magmatic fluid ~ 800°C KES090020 Massive magnetite ore 1.1 0.30 Ortho-magmatic Magmatic fluid ~ 800°C KES090024 Massive magnetite ore 1.0 0.26 Ortho-magmatic Magmatic fluid ~ 800°C KES090027 Massive magnetite ore 1.2 0.29 Ortho-magmatic Magmatic fluid ~ 800°C KES103011 Massive magnetite ore 1.8 0.31 Ortho-magmatic Magmatic fluid ~ 800°C KES090034 Massive magnetite ore 0.5 0.27 Ortho-magmatic Magmatic fluid ~ 625°C KES103003 Massive magnetite ore 0.2 1.00 Ortho-magmatic Magmatic fluid ~ 625°C KES091013b Waste pile material 0.1 0.33 Ortho-magmatic Magmatic fluid ~ 625°C

Hydrothermal EJ-LS-11-2 Massive magnetite ore -4.3 0.24 Hydrothermal < 400°C fluid Hydrothermal EJ-LS-11-3 Massive magnetite ore -1.9 0.36 Hydrothermal < 400°C fluid Hydrothermal KES091007B Waste pile material -0.8 -0.02 Hydrothermal < 400°C fluid Hydrothermal KES090044 Disseminated magnetite -1.0 0.24 Hydrothermal < 400°C fluid Hydrothermal KES090084 Vein magnetite -1.1 0.11 Hydrothermal < 400°C fluid

Fig.18 The iron and oxygen isotope values for the magnetite samples used in this study with their proposed mode of formation indicated by calculated equilibrium reference fields.

7 .1.3 Geothermometric calculations

The first two approaches suggest a bimodal formation of the Grängesberg apatite-iron oxide ore involving magmatic as well as hydrothermal components. Thus, there should be a marked temperature range showing magmatic and hydrothermal values, respectively. For this study the fractionation of oxygen isotopes involving quartz and magnetite can be used as a geothermometer to determine a temperature of formation. For three ore samples δ18O-values for quartz crystals within the ore were measured. Two of these samples are in equilibrium with a magmatic fluid at 625 °C, while the last is in equilibrium with a magmatic fluid at ~800 °C. Calculating the fractionation temperature employs Eq. 3 and 4. The thermometer coefficients A and B are taken from Chiba et al. (1989) and are 6.29 and 0 respectively. The equation is thus rewritten as:

6 2 δquartz – δmagnetite = 6.29 x (10 /T ).

Inserting the oxygen data and solving for T gives the result in Kelvin. The calculations are shown in Appendix 1 and Table 10 shows the three ore samples, their quartz and magnetite isotope values and the calculated fractionation temperature in K and °C.

Table 10. Results for the geothermometric calculations

Sample Magnetite 18 δqz – T in 2σ in Sample 18 Quartz δ O T in K description δ O δmgt °C K/°C

KES090034 Massive ore 0.5 8.7 8.2 876 603 ±21 KES103003 Massive ore 0.2 7.6 7.4 921 648 ±24 KES103016 Massive ore 1.5 7.9 6.4 991 718 ±31

The results for the drived temperatures range between 600 °C and 720 °C. The two magnetites which were tested in the previous paragraph for fractionation at a temperature of 625 °C reveal fractionation temperatures with quartz around that value, supporting the assumption that their formation involved a later stage, e.g. slightly cooled, magmatic fluid. The third temperature of 718 °C fits within the assumed temperature range of exhaled magmatic fluids (600-800 °C, Guilbert & Park 1986). Employing the same equilibrium re-calculation used in the previous paragraph for magnetite and a magmatic fluid but taking quartz instead of magnetite and a temperature of 625 °C (n=2) and 700 °C (n=1) (Δquartz-water = +1.6 ‰ and

Δquartz-water = +0.94 ‰, Matthews & Beckinsale 1979, Zheng 1991, Clayton et al. 1972), values of +6.0 ‰ and +7.1 ‰ at 625 °C and +7.0 ‰ at 700 °C are derived for the magmatic fluid (Table 11). This is in good agreement with the values obtained for the earlier calculation using magnetite (Table 6 & 7). The previous magmatic fluid calculation for magnetite sample

KES103016 was performed for 800 °C and not 700 °C. However, the difference in Δmagnetite- 18 water between these temperatures is -0.5 ‰ only and results in a magmatic fluid with δ O = +7.2 ‰, still in agreement with the value obtained when using quartz. Thus, the quartz oxygen isotope data are consistent with an ore formation from magmatic fluids at temperatures between 800 °C and 600 °C.

Table 11. Fluid re-calculation at 625°C and 700°C for quartz Sample Sample description T in °C Quartz δ18O Fluid δ18O

KES090034 qtz Quartz from massive ore 625 °C 8.7 7.1 KES103003 qtz Quartz from massive ore 625 °C 7.6 6.0 KES103016 qtz Quartz from massive ore 700 °C 7.9 7.0

8. Hydrothermal aureole identification

8.1 Hypothesis and background

In order to test for the possibility of a large-scale hydrothermal replacement, which might have had a major impact on GMD ore formation, the quartz and whole rock data from the area can be applied. Both, the quartz and the whole rocks, can be tested for possible hydrothermal alteration. Hydrothermal overprint on igneous rocks can lead to a decrease or increase in a rocks’ oxygen isotope composition by as much as 10-15 ‰, depending on fluid-δ18O, fluid- rock ratio and temperature (Taylor 1971, Rhodes & Oreskes 1999). The approach chosen in this study is analogous to that of King et al. (1997) where the authors employ quartz to show hydrothermal alteration around the Kidd Creek VMS deposit in Ontario. In this particular study it was shown that there was a hydrothermal overprint that affected quartz crystals and caused a 5 ‰ increase from the originally magmatic δ18O of the mineral. Quartz is usually unaffected by later alteration processes, however, the infilling of micro fractures within the individual crystals will change the isotopic composition (King et al. 1997). Similar studies have been carried out by Larson & Sharp (2003) and Ayers et al. (2006) where isotopic gradients with extending distances from igneous intrusion have been identified and interpreted as hydrothermal alteration. Although hydrothermal overprint from later intrusions and from regional metamorphism cannot be completely excluded for the GMD, neither of these processes seems to have had a significant influence on the isotope compositions of the GMD ores (cf. McQueen 2005, Frost et al. 2007, Hyslop et al. 2008). In addition, the rocks of the GMD have been affected by at least two phases of deformation, such as intense folding during the Svecokarelian orogeny, 1.9-1.8 Ga ago (Stephens et al. 2009). However, if a large-scale hydrothermal aureole was generated during ore formation a variation in values around the deposit or a deformed pattern in the quartz and whole rock data might be seen that could indicate a former “bull’s eye hydrothermal structure”. As most host rocks in the hanging wall towards the East of the ore body are likely younger than the ore itself, a pristine patterns should be visible East of the Export Field. Thus a potential hydrothermal pattern could be reflected by an offset in the oxygen isotope data in the Western section relative to the Eastern one and should be most pronounced in the proximity to the ore body.

8.2 Results

The quartz and whole rock data for the GMD area can be plotted along two traverses, which run from N-S and W-E over the GMD area (Figures 19 & 20). The whole rock data (n=17) range between +5.8 ‰ and +8.6 ‰ which is well within the range of normal igneous subduction zone rocks (e.g. Bindemann 2008). Quartz ranges from +6 ‰ to +10 ‰ and also lies within the range of recorded igneous quartzes (Taylor 1968). No distinct pattern of value distribution can be observed in either the distribution on the map or in the traverse (Fig.19, 20). The E-W traverse would be expected to show the strongest variation if a hydrothermal aureole to the ore was present, but shows magmatic values throughout.

Fig.19 The position of the traverses running across the GMD area. The boxes indicate the samples which have been included in the profile. No variation beyond regular magmatic values is seen, however. Please see also Fig. 20. (Source: Google Earth)

Fig.20 The oxygen isotope data of quartz crystals and whole rocks along the profiles of Traverse 1 (A) and Traverse 2 (B). Overall the data fall into the range of established magmatic values.

8.3 Calculation of possible quartz sources

Testing the quartzes (n=14) for magmatic equilibrium can be achieved by employing a magma re-calculation involving the fractionation between quartz and andesite, dacite and rhyolite magma. The fractionation factors needed for this calculation were taken from Clayton et al. (1972), Zheng (1991) and Zhao & Zheng (2003) and are:

Δquartz-andesite = +0.906 ‰

Δquartz-dacite = +0.838 ‰

Δquartz-rhyolite = +0.77 ‰

When these fractionation factors are applied to GMD quartz and whole rock data in the same way as previously to magnetites, the results show, except for one sample (KPN-090042A), that δ18O melt values lie within the magmatic spectrum of all three common magma types (Table 11). The resulting values for the calculated andesite magmas range from +5.7 to +9.4 ‰, for dacites between +5.8 ‰ and +9.5 ‰ and those for rhyolite between +5.9 ‰ and +9.5 ‰. This is in agreement with the whole rock data from the GMD area that range from +5.8 to +8.4 ‰. The obtained values also match with the oxygen isotope data obtained for the direct ore host rocks (δ18O = +6.1 ‰ - +9.0 ‰) sampled from the drillcores and published in previous studies (e.g. Weis 2011). Sample KPN-090042A shows a quartz δ18O-value of +6.0 ‰ and theoretical magma δ18O-values of +5.2 ‰ which is below the range of igneous rocks and hence not in agreement with its whole rock value of +7.5 ‰. The quartz here might have experienced some extreme high-T hydrothermal overprint (above 875 °C, e.g. magmatic

water, Δquartz-water = -0.1 ‰, Clayton et al. 1972), since high-T fractionation will result in a depletion of quartz in the heavy isotope. Alternatively, this quartz-sample may have undergone some overprint by a low-δ18O fluid, which however seems implausible since the whole rock value shows no indication for such an overprint. A high-temperature overprint by magmatic fluids (δ18O = +6.0 ‰ to +8.0 ‰), on the other hand, would have little effect on the

whole rock value since rock-water fractionation at such temperatures would be small (Δandesite-

water = 0.565 ‰ at 900 °C, Zhao & Zheng 2003), but can considerably offset quartz.

Table 11. Results from the Magma Re-calculation from the map Quartz Whole rock Re-calculated Re-calculated Re-calculated Sample Sample description 2σ δ18O δ18O dacite δ18O andesite δ18O rhyolite δ18O

Sample 29 Granitic host rock 7.9 - ±0.2 7.1 7.0 7.1 Sample 31 Gneissic host rock 6.7 - ±0.2 5.8 5.8 5.9

Table 11. Continued. Sample 40 Granitic host rock 9.3 - ±0.2 8.4 8.4 8.5 Sample 45 Granitic host rock 8.3 - ±0.2 7.4 7.4 7.5 Sample 48 Granitic host rock 8.4 7.9 ±0.2 7.6 7.5 7.6 Sample 49 Granitic host rock 9.2 7.8 ±0.2 8.4 8.3 8.5 Sample 52 Granitic host rock 8.7 6.9 ±0.2 7.9 7.8 8.0 Sample 53 Granitic host rock - 8.4 ±0.2 - - - Sample 56 Granitic host rock - 7.4 ±0.2 - - - KPN-090007 Dacitic host rock - 5.8 ±0.2 - - - KPN-090026-4 Dacitic host rock 9.1 6.8 ±0.2 8.3 8.2 8.3 KPN-090033A Felsic volcanic rock 10.3 8.6 ±0.2 9.5 9.4 9.5 KPN-090042A Metavolcanic host rock 6.0 7.5 ±0.2 5.2 5.1 5.2 KHO-09003 Dacitic host rock - 7.7 ±0.2 - - - KHO-09005 Dacitic host rock 7.0 5.8 ±0.2 6.2 6.1 6.2 KHO-090010 Metavolcanic host rock - 6.8 ±0.2 - - - KHO-090011A Metavolcanic host rock - 6.8 ±0.2 - - - KHO-090012 Dacitic host rock 6.7 7.0 ±0.2 5.9 5.8 5.9 KHO-090013 Andesitic host rock - 7.7 ±0.2 - - - KHO-090013G Andesitic host rock - 6.0 ±0.2 - - - KHO-090017 Metavolcanic host rock 8.8 - ±0.2 8.0 7.9 8.0 KHO-09127b Dacitic host rock - 8.3 ±0.2 - - - KHO-09129A Metavolcanic host rock 8.6 - ±0.2 7.7 7.6 7.8

8.4 Interpretation

The results from the “bull’s eye” test reveals that there is no complete or deformed “bull’s eye hydrothermal structure” recorded in the area around the Grängesberg ore deposit. Also no one sided drop or rise in oxygen isotopes is observed (see above). Moreover, the volcanic host rocks and quartzes show no extensive hydrothermal overprint on either side of the ore field since both the theoretical melt values and actual oxygen isotope data are in agreement with the regular range of magmatic δ18O-values intermediate to felsic associated rock types. This observation is consistent with the previous results from magma and fluid calculation (chapter 7.1.2) and does not support a formation of the massive GMD ore bodies by intense large-scale hydrothermal overprint. The alteration zones present in the direct vicinity of the ore bodies rather appear related to small scale hydrothermal activity induced by high-T magmatic processes, which is also implied by similarities in REE-patterns in both the ore and the altered rocks (Jonsson et al. 2013). The source of this hydrothermal activity may include a mixture of initially magmatic but also influxing meteoric water at a shallow level within a volcanic edifice (cf. Donoghue et al. 2010).

9. Summary of the data evaluation

The interpretation of the analysis of the presented data with various methods results in a model that favours an ortho-magmatic origin of the apatite-iron oxide ore at Grängesberg (Table 12). This conclusion is based on i) the overlap between most oxygen and iron isotope data with established magmatic iron ore deposits, ii) isotope equilibrium of the GMD ore magnetites with magma and high-temperature magmatic fluids, iii) fractionation between mineral pairs likely took place at magmatic temperatures and iv) the absence of a large-scale hydrothermal aureole in the area (or deformed traces thereof). Only a minor part of the data, consisting of non-massive (vein and disseminated) and calcite ore, show agreement with hydrothermal replacement or hydrothermal precipitation at lower temperatures (< 400 °C). Thus the majority of the GMD ore has likely formed by ortho-magmatic processes, followed by a volumetrically minor percentage of ore that was hydrothermally modified.

Table 12. Summary of the data evaluation results for the GMD. Ortho- Method Indicators Hydrothermal magmatic

Massive magnetite ore vs. vein Textural comparison and disseminated magnetites     Data overlap (with magmatic Data comparison and hydrothermal groups)    Magmatic (e.g. magma & fluid) Source re-calculation vs. hydrothermal sources    Magmatic (900 °C – 600 °C) Temperature vs. hydrothermal (< 400 °C) X calculation  temperatures  Hydrothermal aureole No large-scale hydrothermal X identification aureole detected 

10. A conceptual model

The formation of the apatite-iron oxide ore at GMD most likely took place in a subduction zone setting, perhaps a continental back arc basin, similar to the modern Andean continental arc or the Taupo Volcanic Zone in New Zealand (see Allen et al. 1996). This includes the likely presence of large volcanic systems as caldera type volcanoes (Allen et al. 1996). Ore deposits form frequently in such settings and examples are porphyry copper and VMS deposits (e.g. Galley et al. 2007, Sinclair 2007). Notably, most Apatite-iron oxide ores seem associated with this setting too and a broad connection is inferred (Hitzmann 2000, Corriveau 2007, Groves et al. 2010). On the basis of data presented, the formation of the GMD apatite- iron oxide ore itself can be explained by three key processes, involving direct magmatic, high- temperature magmatic fluids and low-temperature hydrothermal overprint. Within the framework of an evolving, large-scale volcanic centre, these processes may relate to each other through a decrease in temperature.

10.1 High temperature: crystallisation from magma

Most GMD ore was likely crystallized from an oxide-rich magma as magnetites in oxygen isotopic equilibrium with an andesite to dacite parent magma is common. Andesite and dacite, the most frequent rock type in the GMD region, is also the most common rock type in continental subduction zones (Allen et al. 1996, Price et al. 2005). These GMD andesite and dacite magmas must also have been extremely enriched in isotopic heavy ferric iron, indicated by very high re-calculated δ56Fe-values. Oxide magmas have been proposed in earlier studies such as Lundberg & Smellie 1979, Kolker (1982), Weidner (1982), Zhou et al. (2005) and Nyström et al. (2008). Such potential oxide melts have been inferred from experiments and from geological evidence such as vesicles in ore, a common feature in e.g. El Laco. Liquid immiscibility has been proposed as the mechanism of formation of such oxide melts, which means that a silicate magma could separate into two melts of different composition during cooling or crystallisation (Roedder 1978). The composition of the newly generated magmas depends on the parent magma and may be oxygen or sulfur-rich (Roedder 1978). The reasoning behind this claim is that with increasing silica content of a melt, e.g. due to fractional crystallisation, oxygen may become separated from the silicate melt together with oxyphile elements such as iron and phosphorous (Guilbert & Park 1986). Philpott (1967) and Naslund (1983) showed that an iron oxide melt could be derived from silica-saturated intermediate magmas such as an andesite or dacite magma at temperatures of around 900° - 1000 °C (Fig.13). Thus, ore genesis at GMD possibly involved the formation of an apatite- 57 iron-oxide-rich melt at depth from an intermediate parent magma, possibly by liquid immiscibility. Ascent during volcanic activity of this newly formed oxide-rich magma resulted in the formation of intrusive (lense-shaped) apatite-magnetite ore bodies at shallower depth within the previously formed intermediate volcanic rocks (e.g. as a sill and dyke complex). Ascent of these dense melts could have been facilitated by e.g. large-scale caldera unrest (cf. Nicoll et al. 2009), which may have helped to overcome the density contrast and buoyancy limits that oxide magmas may experience.

10.2 Ore formation from magmatic fluids

Deposition of massive magnetites from cumulate or from oxide magmas was followed by magnetite formation from exsolved magmatic fluids at temperatures between 600° and 800 °C (Fig.12). Intermediate magmas contain up to 7 wt. % of water plus other volatiles such as Cl,

B, F, CO2 and S (Clemens 1984, Heimann et al. 2008). With progressive crystallisation these components may accumulate and separate from the magma as a distinct fluid phase, causing the fluid pressure to increase and to form a saline (Cl-bearing) magmatic brine that can hold the metals necessary for ore formation (e.g. Nadeau et al. 2010). The exsolution of magma derived waters has also been suggested for oxide-rich magmas, as e.g. proposed by Nyström et al (2008) and Hou (2009). Their proposal is based on the presence of vesicles and holes within samples of massive ore bodies which they interpreted as considerable amounts of volatiles in the ore-forming magmas or solutions. Upon the emplacement of the iron oxide- rich melt, some iron was likely removed during the degassing process (i.e. volatile loss) and transported as metal halogen complexes (e.g. hexachloro complexes, Schauble et al. 2001). The isotope composition of the magmatic fluid reflects the oxygen isotope signature of the magma from which it was exsolved, however, the iron isotope composition of the fluid was a little more depleted in the heavy isotope compared to the magma. This is because the isotopically light ferrous iron is preferred in the fluid phase during fluid removal since it is the more mobile phase. This phenomenon, in turn, might have contributed to the ‘extreme’ 56Fe- enrichment of the melt calculated for some of the samples (cf. Heimann et al. 2008). Upon cooling, the magmatic fluid then deposited more isotopically heavy magnetite ore in the vicinity of the previously emplaced magmatic magnetites. The remaining fluids, depleted in 56Fe, might have migrated further and cooled or mixed with a lower-δ18O fluid at some stage and thus contributing to the deposition of deposits like the Björnberget deposit (calcitic magnetite ore, δ18O = -0.8 ‰, δ56Fe = -0.02 ‰). Exsolved magmatic fluids may also be

58 responsible for hydrothermal alteration of the host rocks in the vicinity of the ores as also reflected in REE data (Jonsson et al. 2010, Jonsson et al. 2013).

10.3 Low-temperature: hydrothermal re-deposition

Associated with the volcanic system was hydrothermal activity which was potentially induced by the ascending and intruding magma in the direct vicinity of emplaced ores or oxide-rich magmas (cf. Donoghue et al. 2008, 2010). This hydrothermal activity is likely also because the decreasing magmatic fluid pressure due to volatile loss and cooling, allows for increasing amounts of external fluid influx and thus heat-driven hydrothermal convection (e.g. Fig.21 & 22). Components of the hydrothermal activity likely included a high-δ18O, low-δ56Fe fluid. This hydrothermal fluid could be remnants of previously exhaled magmatic fluids or very 18O-rich meteoric waters (Craig 1961), and maybe similar to the evaporitic water proposed for the El Laco deposit in Chile (Rhodes & Oreskes 1999). A mixture of remnant magmatic and influxing meteoric fluids as the hydrothermal source appears most plausible. The magmatic induced late stage hydrothermal activity also caused local alteration of the ore host rocks and in addition began to locally redistribute ore, leading to magnetite veins and disseminations in the vicinity of massive ore bodies. This not only can explain the presence of the magnetite disseminations and veins within the surrounding rocks but also the hematite ore, which would form from more oxidizing fluids (cf. Markl et al. 2006). Hydrothermal activity most likely took place at temperatures well below 400 °C and in a shallow sub-volcanic environment.

The three main ore forming stages in the GMD are summarized in Table 13 as well as in Fig.21 and Fig.22, which give an overview of the ore forming processes, ore types, ore sources isotope composition and formation temperatures.

Table 13. Summary of the ore forming processes at Grängesberg δ56Fe δ18O δ18O δ56Fe Source of ore T in ⁰C magnetite magnetite source source

Oxide-rich > 800 ⁰C ~ +0.3 ‰ ≥ 1.8 ‰ +6 - 8 ‰ ~ +0.33 ‰ magma

Magmatic saline ~ +0.07 to 600⁰-800 ⁰C ~ +0.3 ‰ ≥ 0.2 ‰ +6 - 8 ‰ fluids +0.12 ‰

Hydrothermal ~ -0.2 to < 400 ⁰C ~ +0.2 ‰ < 0.0 ‰ > +5 ‰ fluids -0.6 ‰

Fig.21 Ore forming stages showing typical magmatic-hydrothermal evolution (modified after Matthes 2001). Stages II and III comprise ortho-magmatic ore genesis. With decreasing temperature and ongoing crystallisation of a magma, the volatile/fluid pressure will increase and magmatic fluids are being expelled into the surrounding rocks. Below ~ 600 °C the pressure begins to drop, progressively allowing external fluids to enter the system (e.g. ~ 400 °C) and initiate large-scale hydrothermal activity. Most GMD magnetites formed during ortho- magmatic evolution (Stages II and III) of the system, while vein and disseminated magnetite formed mainly during Stage V (hydrothermal re-deposition).

Fig.22 A possible scenario for ore formation at the Grängesberg volcano. A subduction zone, caldera type volcanic system produced oxide-enriched melt by e.g. liquid immiscibility at depth. Upon ascent this oxide- rich magma intrudes the previously erupted volcanic rocks. Thereby the magma is expelling oxide bearing magmatic waters. Magmatic crystallisation and magmatic fluid precipitation lead to the formation of massive magnetite ore bodies. The heat liberated by the ascending and intruding magma initiates hydrothermal circulation, which causes minor re-deposition of previous magnetite in form of veins and disseminations. Hydrothermal fluids may originate from mixtures of remnant magmatic fluids with meteoric and/or other high-δ18O surface water.

11. Summary and Conclusions

The presented isotope data lead to the proposal that the large apatite-iron oxide ore deposit at Grängesberg was formed dominantly by ortho-magmatic processes, which includes the possibility of iron-oxide rich melts that formed from originally andesite to dacite compositions, but also that of exsolved saline (e.g. Cl--rich) magmatic aqueous fluids at magmatic temperatures. A small fraction of vein and disseminated magnetites seems to have formed at low-temperatures (≤ 400 °C) from hydrothermal re-deposition. These conclusions are based on the following observations:

1. The oxygen and iron isotope compositions of massive GMD apatite-iron oxide ores show a clear distinction from ores which have been suggested to have formed from ‘non-magmatic’ processes (e.g. Dannemora, Striberg). In contrast, the GMD massive ores overlap with other “magmatic” ore deposits, with magnetites from layered igneous intrusions and with the volcanic reference materials from Indonesia, Cyprus and New Zealand.

2. The majority of the GMD apatite-iron oxide ore (n=14 out of 17) is in isotopic equilibrium with a magma of andesite to dacite composition at normal magmatic temperatures and/or magmatic aqueous fluids at temperatures between 600 and 800 °C. This group includes all massive ore types at GMD.

3. A small number of vein and disseminated magnetites (n=3) seem to have been formed by low-temperature hydrothermal processes (below 400 °C), including leaching and re-deposition of the previous ortho-magmatic deposit. This type of magnetite is confined to the vicinity of massive ores and may be considered a ‘natural progression’ of a cooling magmatic system or intrusion.

4. Geothermometric calculations for quartz and magnetite pairs from massive ore samples reveal fractionation termperatures between 600 °C and 715 °C, which independently supports a magmatic equilibrium temperature for the massive ores and their source.

5. Oxygen isotope data from quartz phenocrysts and whole rock analyses of the metavolcanic host rocks in the Grängesberg area do not support the presence of a

large-scale fossil hydrothermal aureole, which would have been expected in the ore formation, if hydrothermal activity was the main cause for ore formation.

In respect to the initial hypothesis formulated, the isotope data presented in this study and the obtained modeling results are in agreement with dominantly magmatic processes at Grängesberg as well as at other apatite-iron oxide ore deposits (e.g. Kiruna, El Laco, Blötberget, Tjårrojåkka). The obtained results and interpretations lend support to the ‘magmatic school of thought’ regarding the origin of the “Kiruna-type” ores in Sweden and possibly beyond.

12. Acknowledgments

I want to thank my supervisor Prof. Valentin R. Troll for all his help and support during the last two and a half years at Uppsala University, for giving me the opportunity to work on this project and at the Department of Earth Science and for donating a fair number of the samples used in this study. A big thank you goes also to Prof. Erik Jonsson and Dr. Katarina Persson- Nilsson from the Swedish Geological Survey and Dr. Karin Högdahl, my supervisor of my Bachelor project on the GMD at Uppsala University, for their help and support as well as providing a big part of the samples used in this project. My gratefulness also goes to my reviewer Dr. Abigail Barker at Uppsala University for her help with writing this project. At the University of Cape Town, ZA, I would like to thank Prof. Chris Harris for his help during the sample analysis in June 2012 and for his help regarding data processing. Also for analysis regarding iron isotopes, I give my gratefulness to Dr. Marc-Alban Millet at the Victoria University in Wellington, NZ.

For providing the samples and helping in sample preparation I give my gratefulness to the Dannemora Mineral AB, especially many thanks to Gunnar Rauseus, to Barbara Kern at the Technical University of Luleå, Dr. Frances M. Deegan and Kirsten Pedroza at Uppsala University, Prof. John Gamble from the Victoria University in Wellington, Prof. Nicholas Arndt at the Université Joseph Fourier in Grenoble, France, Dr. Jan-Olof Nyström at the Swedish Museum for Natural History and to Peter Dahlin and Dr. Håkan Sjöström at Uppsala University.

I wish to express further gratitude to all the people at Uppsala University who helped me during this project with sample preparation, editorial work, corrections of my essay or simply moral support and who made the time in Uppsala to the great and awesome experience that it was. To mention are (without any preference!):

Fredrik Sahlström, Sherissa Roopnarain, Klara Blomdahl, Johan Magnusson, Åke Rosén, Jon Lundh, Johannes Petrone, Stefan Andersson, Sara Eklöf, Sara Andersson, Mattias Carlsson, David Höglin, Pia Karlsson, David Nilsson, Anna Vass, Peter Nichols, Per Marklund, Sara Johansson, Christof Auer, Helena Glansholm, Maria Lindkvist, Dr. Lara Blythe, Ester Muños- Jolis, David Budd, Sylvia Berg, Carolin Ryr, Börje Dahrén, Prof. Graham Budd, Prof. Hemin Koyi, Dr. Michael Streng, Aodhan Butler, Oskar Jäckel, Heda Agic and many many more.

Last but not least I want to greatly thank my family for all their endless support and help they have given me not only during my time in Uppsala but during all my years in university and school! No matter if North, South, East or West, home and family are simply the best!

13. References

Allen, R. L., Lundström, I., Ripa, M., Simeonov, A.; Christoffersson, H. (1996): Facies analysis of a 1.9 Ga, continental margin, back-arc, felsic caldera province with diverse Zn-Pb-Ag-(Cu-Au) sulfide and Fe oxide deposits, Bergslagen region, Sweden. Economic Geology, Vol. 91: pp. 979–1008

Allen, R.; Ripa, M.; Jansson, N. (2008): Palaeoproterozoic volcanic- and limestonehosted Zn-Pb-Ag-(Cu-Au) massive sulphide deposits and Fe oxide deposits in Bergslagen, Sweden. IGC Excursion Nr. 12 (August 2008): pp. 16-21

Anbar, A. D.; Roe, J.E.; Barling, J.; Nealson, K.H. (2000): Nonbiological Fractionation of Iron Isotopes. Science Magazine, Vol. 288: pp. 126-127

Anbar, A. D. (2004): Iron stable isotopes: beyond biosignatures. Earth and Planetary Science Letters, Vol. 217: pp. 223-236.

Ancochea, E.; Fuster, J. M.; Ibarrola, E.; Cendrero, A.; Coello, J.; Hernan, F.; Cantagrel, J.M.; Jamond, C. (1990): Volcanic evolution of the island of Tenerife (Canary Islands) in the light of new K-Ar data. Journal of Volcanology and Geothermal Research, Vol.44: pp. 231 -249.

Ayers, J. C.; Loflin, M.; Miller, C.R.; Barton, M.D.; Coath, C.D. (2006): In situ oxygen isotope analysis of monazite as a monitor of fluid infiltration during contact metamorphism: Birch Creek Pluton aureole, White Mountains, eastern California. Geology, Vol. 34, Nr. 8: pp. 653-656.

Åkerman, C. (1994): Ores and mineral deposits. In: C.Fredén (ed.): Geology. National Atlas of Sweden: pp. 55– 66.

Allerton, S.; Vine, F. J. (1991): Spreading evolution of the Troodos ophiolite, Cyprus. Geology, Vol. 19, No. 6: pp. 637-640.

Beard, B. L.; Johnson, C. M.; Skulan, J. L.; Nealson, K. H.; Cox, L.; Sun, H. (2003): Application of Fe isotopes to tracing the geochemical and biological cycling of Fe. Chemical Geology, Vol. 195: pp. 87-117.

Beard, B. L.; Johnson, C. M. (2004): Fe Isotope Variations in the Modern and Ancient Earth and Other Planetary Bodies. Reviews in Mineralogy & Geochemistry, Vol. 55: pp. 319-357.

Bindeman, I. (2008): Oxygen Isotopes in Mantle and Crustal Magmas as revealed by Single Crystal Analysis. Reviews in Mineralogy & Geochemistry, Vol. 69: pp. 445-478.

Blattner, P.; Braithwaite, W. R.; Glover, R. B. (1983): New evidence on magnetite oxygen isotope geothermometers at 175° and 112° C in Wairakei steam pipelines (New Zealand). Isotope Geoscience, Vol. 1: pp. 195-204.

Carlile, J. C.; Mitchell, A.H.G. (1994): Magmatic arcs and associated gold and copper mineralization in Indonesia. In: van Leeuwen, T.M.; Hedenquist, J.W.; James L.P.; Dow, J.A.S. (Editors), Indonesian Mineral Deposits -- Discoveries of the Past 25 Years. Journal of Geochemical Exploration,Vol. 50: pp. 91-142.

Cawthorn, R. G.; Molyneux, T. G. (1986): Vanadiferous magnetite deposits of the Bushveld Complex. In: Anhaeusser, C. R.; Maske, S. (eds.) Mineral Deposits of Southern Africa. Vol. 2: pp. 1267-1286. Geological Society South Africa, Johannesburg.

Chiba, H.; Chacko, T.; Clayton, Robert N.; Goldsmith, J. R. (1989): Oxygen isotope fractionations involving diopside, forsterite, magnetite, and calcite: Application to geothermometry. Geochimica et Cosmochimica Acta, Vol. 53 (1989): pp. 2985-2995

Claesson, S.; Huhma, H.; Kinney, P.D.; Williams, I.S. (1993): Svecofennian detrital zircon ages implications for the Precambrian evolution of the Baltic Shield. Precambrian Research, Vol. 64: pp. 109-130.

Clayton, R. N.; O’Neil, J. R; Mayeda, T. K. (1972): Oxygen Isotope Exchange between Quartz and Water. Journal of Geophysical Research, Vol. 77, No. 17: pp. 3057-3067.

Clemens, J. D. (1984): Water contents of silicic to intermediate Magmas. Lithos, Vol. 17: pp. 273-287.

Coplen, T. B.; Böhlke, J. K.; De Bièvre, P.; Ding, T.; Holden, N. E.; Hopple, J. A.; Krouse, H. R.; Lamberty, A.; Peiser, H. S.; Revesz, K.; Rieder, S. E.; Rosman, K. J. R.; Roth, E.; Taylor, P. D. P.; Vocke, R. D.; Xiao Jr. and Y. K. (2002): Isotope-abundance variations of selected elements. Pure and Applied Chemistry Vol. 74, No.10: pp. 1987-2017.

Corriveau, L. (2007): Iron oxide copper-gold (±Ag ±Nb ±P ±REE ±U) deposits: A Canadian perspective. In: Goodfellow, W. D., ed., Mineral deposits of Canada: A synthesis of major deposit- types, district metallogeny, the evolution of geological provinces, and exploration methods: Geological Association of Canada, Mineral Deposits Division Special Publication, No. 5: pp. 307–328.

Craig, H. (1961): Isotopic Variations in Meteoric Waters. Science, Vol. 133, No. 3465: pp. 1702-1703.

Dahlin, P.; Sjöström, H. (2010): Structure and stratigraphy of the Dannemora inlier, eastern Bergslagen Region. Department of Earth Sciences, Uppsala University, 91 p.

Deegan, F. M.; Troll, V. R.; Barker, A.K.; Harris, C.; Chadwick, J. P.; Carracedo, J. C.; Delcamp, A. (2012): Crustal versus source processes recorded in dykes from the Northeast volcanic rift zone of Tenerife, Canary Islands. Chemical Geology, Vol. 334: pp. 324-344.

Dickenson M. P.; Hess P. C. (1986): The structural role and homogeneous redox equilibria of iron in peraluminous, metaluminous and peralkaline silicate melts. Contributions to Mineralogy and Petrology, Vol. 92: pp. 207-217.

Donoghue, E.; Troll, V. R.; Harris, C.; O’Halloran, A.; Walter, T. R.; Pérez Torrado, F. J. (2008): Low- temperature hydrothermal alteration of intra-caldera tuffs, Miocene Tejeda caldera, Gran Canaria, Canary Islands. Journal of Volcanology and Geothermal Research, Vol. 176: pp. 551-564.

Donoghue, E.; Troll, V. R.; Harris, C. (2010): Hydrothermal alteration of the Miocene Tejeda Intrusive Complex, Gran Canaria, Canary Islands: insights from petrography, mineralogy and O- and H- isotope geochemistry. Journal of Petrology, Vol. 51, pp. 2149-2176.

Edfelt, Å. (2007): The Tjårrojåkka Apatite-Iron and Cu (-Au) Deposits, Northern Sweden - Products of One Ore Forming Event. PhD-Thesis, Department of Chemical Engineering and Geosciences, Luleå University of Technology, Sweden, 167.p

Evans, A. M. (2000): Ore Geology and Industrial Minerals – An Introduction. Blackwell Science, 3. Edition, Oxford, 403 p.

Frietsch, R. (1975): Brief outline of the metallic mineral resources of Sweden. Geological survey of Sweden, C718: 64 p.

Frietsch, R. (1978): On the magmatic origin of iron ores of the Kiruna type. Economic Geology, Vol.73: pp. 478-485.

Frietsch, R. (1982): Alkali metasomatism in the ore-bearing metavolcanics of central Sweden. Geological Survey of Sweden, C 79, 54 p.

Frietsch, R.; Perdahl, J.-A. (1995): Rare earth elements in apatite and magnetite in Kiruna-type iron ores and some other iron ore types. Ore Geology Reviews, Vol. 9: pp. 489-510.

Frost, C. D.; von Blanckenburg, F.; Schoenberg, R.; Frost, B. R.; Swapp, S.M. (2007): Preservation of Fe isotope heterogeneities during diagenesis and metamorphism of banded iron formation. Contributions to Mineralogy and Petrology, Vol. 153: pp. 211-235.

Gaillard F.; Scaillet B.; Pichavant M.; Beny J.-M. (2001): The effect of water and fO2 on the ferric-ferrous ratio of silicic melts. Chemical Geology, Vol. 174: pp. 255-273.

Galley, A. G.; Hannington, M. D.; Jonasson, I. R. (2007): Volcanogenic massive sulphide deposits. In: Goodfellow, W. D., ed., Mineral Deposits of Canada: A Synthesis of Major Deposit-Types, District Metallogeny, the Evolution of Geological Provinces, and Exploration Methods. Geological Association of Canada, Mineral Deposits Division, Special Publication No. 5: p. 141-161.

Gamble, J. A.; Wood, C. P.; Price, R. C.; Smith, I. E. M.; Stewart, R. B.; Waight, T. (1999): A fifty year perspective of magmatic evolution on Ruapehu Volcano, New Zealand: verification of open system behaviour in an arc volcano. Earth and Planetary Science Letters, Vol. 170: pp. 301-314.

Gamble, J. A.; Price, R. C.; Smith, I. E. M.; Mc Intosh, W. C.; Dunbar, N.W. (2003): 40Ar/39Ar geochronology of magmatic activity, magma flux and hazards at RuapehuVolcano, Taupo Volcanic Zone, New Zealand. Journal of Volcanology and Geothermal Research, Vol. 120: pp. 271-287.

Gardner, M. F.; Troll, V. R.; Gamble, J. A.; Gertisser, R.; Hart, G. L.; Ellam, R. M.; Harris, C.; Wolff, J. A. (2012): Crustal differentiation processes at Krakatau volcano, Indonesia. Journal of Petrology, doi:10.1093/petrology/egs066, 34 p.

Gass, I. G.; Masson-Smith, D. (1963): The Geology and Gravity Anomalies of the Troodos Massif, Cyprus. Philosophical Transactions of the Royal Society of London, Series A, Mathematical and Physical Sciences, Vol. 255, No. 1060: pp. 417-467.

Geijer, P. (1931): The iron ores of the Kiruna type. Sveriges Geologiska Undersökning C367: pp. 1-39.

Geijer, P. (1967): Internal features of the apatite-bearing magnetite ores. Sveriges Geologiska Undersökning C624: pp. 1-32.

Graham, I. J.; Hackett,W. R. (1987): Petrology of calc-alkaline lavas from Ruapehu volcano and related vents, Taupo Volcanic Zone, New Zealand. Journal of Petrology, Vol. 28: pp. 531-567.

Groves, D. I.; Bierlein, F. P.; Meinert, L. D.; Hitzman, M. W. (2010): Iron Oxide Copper-Gold (IOCG) Deposits through Earth History: Implications for Origin, Lithospheric Setting, and Distinction from Other Epigenetic Iron Oxide Deposits. Economic Geology, Vol. 105: pp. 641-654.

Gudmundsson, M.T,; Pedersen, R.; Vogfjörd, K.; Thorbjarnardottir, B.; Jakobsdottir, S.; Roberts, M.J. (2010): Eruptions of Eyjafjallajökull Volcano, Iceland. EOS, Vol. 91: pp. 190-191.

Guilbert, J. M.; Park, C.F. (1986): The Geology of Ore Deposits. W.H. Freeman and Company, New York, 985 p.

Harlov, D.-E.; Andersson, U.-B.; Foerster, H.; Nyström, J.-O.; Dulski P.; Broman, C. (2002): Apatite-monazite relations in the Kiirunavaara magnetite-apatite ore, northern Sweden. Chemical Geology, Vol. 191: pp. 47-72.

Harney, D. M. W.; von Gruenewaldt, G. (1995): Ore-forming processes in the upper part of the Bushveld complex, South Africa. Journal of African Earth Sciences, Vol. 20, No. 2: pp. 77-89.

Harris, C.; Vogeli, J. (2010): Oxygen isotope composition of garnet in the Peninsula Granite, Cape Granite Suite, South Africa: constraints on melting and emplacement mechanisms. South African Journal of Geology, Vol. 113: pp. 401-412.

Heimann, A.; Beard B.; Johnson C. (2008): The role of volatile exsolution and sub-solidus fluid/rock interactions in producing high 56Fe/54Fe ratios in siliceous igneous rocks. Geochimica et Cosmochimica Acta, Vol.72: pp. 4379–4396.

Henriquez, F.; Nyström, J. O. (1998): Magnetite bombs at El Laco volcano, Chile. GFF, Vol. 120: pp. 269-271.

Henriquez, F.; Naslund, H. R.; Nyström, J. O.; Vivallo, W.; Aguirre, R,.; Dobbs, F. M.; Llédo, H. (2003): New field evidence bearing on the origin of the El Laco magnetite deposit, Northern Chile - A discussion.Economic Geology, Vol. 98: pp. 1497-1500.

Henson, F. R. S.; Browne, R. V.; McGinty, J. (1949): A synopsis of the stratigraphy and geological history of Cyprus. Quarterly Journal of the Geological Society, Vol. 105: pp. 1-41.

Hitzman, M. W.; Oreskes N.; Einaudi M.T. (1992): Geological characteristics and tectonic setting of Proterozoic iron oxide (CU-U-Au-REE) deposits. Precambrian Research, Vol.58: pp. 241-287.

Hitzman, M. W. (2000): Iron oxide-Cu-Au deposits: what, where, when, and why. In: Porter, T. M. (ed) Hydrothermal iron oxide copper-gold and related deposits: A global perspective. Australian Mineral Foundation, Adelaide, pp 9–25.

Hjelmqvist, S. (1950): The titaniferous iron–ore deposit of Taberg in the South of Sweden. Geological Survey of Sweden, C 512: pp. 1 –55.

Hoefs, J. (1997): Stable Isotope Geochemistry. Springer-Verlag, Berlin Heidelberg New York, 201 p.

Hou, T.; Zhang, Z.; Du, Y.; Li, S. (2009): Geology of the Gushan iron oxide deposit associated with dioritic porphyries, eastern Yangtze craton, SE China. International Geology Review, Vol. 51: pp. 520-541.

Huang F.; Lundstrom C. C. (2006): Iron isotopic fractionation factor between magnetite and hydrous silicic melt. American Geophysical Union (AGU) Fall Meeting abstract, V21B- 0575.

Hunt, J. A.; Baker, T.; Thorkelson, D. J. (2007): A Review of Iron Oxide Copper-Gold Deposits, with Focus on the Wernecke Breccias, Yukon, Canada, as an Example of a Non-Magmatic End Member and Implications for IOCG Genesis and Classification. Exploration and Mining Geology, Vol. 16: pp. 209–232.

Hyslop, E. V.; Valley, J. W.; Johnson, C. M.; Beard, B. L. (2008): The effects of metamorphism on O and Fe isotope compositions in the Biwabik Iron Formation, northern Minnesota. Contributions to Mineralogy and Petrology, Vol. 155: pp. 313-328.

Irvine, T. N.; Sharpe, M.R. (1986): Magma mixing and the origin of stratiform oxide ore zones in the Bushveld and Stillwater Complexes. In: Gallagher, M.J., Ixer, R.A., Neary, C.R. and Prichard, H.M. (eds.) Metallogeny of Basic and Ultrabasic Rocks) pp. 183-198. Institute Mining Metallurgy, London.

Javoy, M. (1977): Stable Isotopes and Geothermometry. Journal of the Geological Society, Vol.133: pp. 609- 636.

Jiao, J. (2011): Petrography and geochemistry of a section through the Blötberget apatite-iron oxide deposit, Bergslagen, south central Sweden. Master Thesis, Department of Earth Science, Uppsala University, Sweden, 79 p.

Johansson, H. (1910): Die eisenerzführende Formation in der Gegend von Grängesberg. Geologiska Föreningens i Stockholm Förhandlingar Vol.32, Issue 2 (February 1910): pp.239-410.

Jolly, A. D.; Sherburn, S.; Jousset, P.; Kilgour, G. (2010): Eruption source processes from seismic and acoustic observations of the 25 September 2007 Ruapehu eruption-North Island, New Zealand. Journal of Volcanology and Geothermal Research, Vol. 191: pp. 33-45.

Jonsson, E.; Nilsson, K. P.; Hallberg, A.; Högdahl, K. (2010): The Palaeoproterozoic apatite-iron oxide deposits of the Grängesberg area: Kiruna-type deposits in central Sweden. In: Nakrem, H. A.; Harstad, A. O.; Haukdal, G. (eds.) NGF abstracts and proceedings 1, pp. 88-89.

Jonsson, E.; Troll, V.R.; Högdahl, K.; Harris, C.; Weis, F.; Nilsson, K.P.; Skelton, A. (2013): Ortho-magmatic origin of giant ‘Kiruna-type’ apatite-iron-oxide-ores in Central Sweden. Unpublished article, in preparation for submission to Nature Scientific Reports.

King, E. M.; Barrie, C. T.; Valles, J. W. (1997): Hydrothermal alteration of oxygen isotope ratios in quartz phenocrysts, Kidd Creek mine, Ontario: Magmatic values are preserved in zircon. Geology, Vol. 25, No.12: pp. 1079-1082.

Kolker, A. (1982): Mineralogy and Geochemistry of Fe-Ti-Oxide and Apatite (Nelsonite) deposits and evaluation of the liquid immiscibility hypothesis. Economic Geology, Vol. 77: pp.1146-1158.

Lager, I., 1986: The Dannemora iron ore deposit. In: Lundström, I. & Papunen, H. (eds.): Mineral deposits of southwestern Finland and the Bergslagen province, Sweden. 7th IAGOD symposium and Nordkalott project meeting, excursion guide No. 3. Geological Survey of Sweden Ca 61: pp. 26–30.

Lager, I. (2001): The geology of the Palaeoproterozoic limestone-hosted Dannemora iron deposit, Sweden. Sveriges Geologiska Undersökning Rapporter och meddelanden 107, 49 p.

Landergren, S. (1948): On the Geochemistry of Swedish iron ores and associated rocks – A study on iron-ore formation. Geological Survey of Sweden, Series C, No. 469, Vol. 42, No. 5: 182 p.

Larson, S. Å. (1973): Igneous layering in the Ulvö dolerite, Ångermanland, Central Sweden. GFF, Vol. 95: pp. 407–409.

Larson, S. Å.; Magnusson, K.-Å. (1976): The magnetic and chemical character of Fe-Ti oxides in the Ulvö dolerite, Central Sweden. Geological Survey of Sweden, C723, 29 p.

Larson, T. E.; Sharp, Z. D. (2003): Stable isotope constraints on the Al2SiO5‘triple‐point’ rocks from the Proterozoic Priest pluton contact aureole, New Mexico, USA. Journal of Metamorphic Geology, Vol. 21, No. 8: pp. 785-798.

Larson, S. Å.; Hogmalm, K. J.; Meurer, W. P. (2008): Character and significance of spectacular layering features developed in the thin, alkali-basaltic sills of the Ulvö Gabbro Complex, Sweden. Mineralalogy and Petrology, Vol. 92: pp. 427–452.

Larsson, D.; Söderlund, U. (2005): Lu–Hf apatite geochronology of mafic cumulates: An example from a Fe–Ti mineralization at Smålands Taberg, southern Sweden. Chemical Geology, Vol. 224: pp. 201-211.

Lindblom, S.; Broman, C.; Martinsson, O. (1996): Magmatic-hydrothermal fluids in the Pahtohavare Cu-Au deposit in greenstone at Kiruna, Sweden. Mineralium-Deposita, Vol. 31: pp. 307-318.

Lundberg, B.; Smellie, J. A. T. (1979): Painirova and Mertainen Iron Ores: Two Deposits of the Kiruna Iron Ore Type in Northern Sweden. Economic Geology, Vol. 74: pp. 1131-1152.

Lundström, I. (1987): Lateral variations in supracrustal geology within the Swedish part of the Southern Svecokarelian Volcanic Belt. Precambrian Research, Vol. 35: pp. 353–365.

Lundström, I. (1995): Beskrivning till berggrundskartorna Filipstad SO och NO. Geological survey of Sweden Af 177, No. 185, 218 p.

Lundström, I.; Allen, R. L.; persson, P. O.; Ripa, M. (1998): Stratigraphies and depositional ages of Svecofennian, Paleoproterozoic metavolcanic rocks in E. Svealand and Bergslagen, south central Sweden. GFF, Vol. 120: pp. 315−320.

Lundqvist, T. (1979): The Precambrian of Sweden. Geological Survey of Sweden, C 768, 87 p. 70

Lundqvist, T. (1990): Beskrivning till berggrundskartan över Västernorrlands län (English summary). Geological Survey of Sweden, Ba 31, 429 p.

Ma, Y. X.; Ji, X. T.; Li, J. C.; Huang, M.; Min, Z. Z. (2003): Mineralresources of Panzhihua, Sichuan Province, SW China. Chengdu: Chengdu University of Technology, 275 p.

Magnusson, N. H. (1938): Neue Untersuchungen innerhalb des Grängesbergfeldes .SGU ser. C, Nr. 401 (1938)

Magnusson, N. H. (1970): The origin of the iron ores in central Sweden and the history of their alterations. SGU ser. C, Nr. 643: 127 p.

Magnusson, K.-Å.; Larson, S. Å. (1977): A paleomagnetic investigation of the Ulvö dolerite, Ångermanland, Central Sweden. Lithos, Vol. 10: pp. 205–211.

Maksaev, V.; Gardeweg, M.; Ramírez, C.F.; Zentilli, M. (1988): Aplicación del método trazas de fisión (fission track) a la datación de cuerpos de magnetita de El Laco e Incahuasi en el altiplano de la Región de Antofagasta. In: Congreso Geológico Chileno, No. 5, Actas 1: B1-B23. Santiago.

Marcoux, E.; Milési, J.-P. (1994): Epithermal gold deposits in West Java, Indonesia: geology, age and crustal source. Journal of Geochemical Exploration, Vol. 50: pp. 393-408.

Markl, G.; von Blanckenburg, F.; Wagner, T. (2006): Iron isotope fractionation during hydrothermal ore deposition and alteration. Geochimica et Cosmochimica Acta, Vol. 70: pp. 3011–3030.

Martinsson, O. (2004): Geology and Metallogeny of the Northern Norrbotten Fe-Cu-Au Province. In: Allen, R. L.; Martinsson, O.; Weihed, P., eds., Svecofennian Ore-Forming Environments: Volcanic-Associated Zn-Cu-Au- Ag, Intrusion-Associated Cu-Au, Sediment-Hosted Pb- Zn, and Magnetite-Apatite Deposits of Northern Sweden, Society of Economic Geologists, Guidebook Series, 33., p. 131-148.

Matthes, S. (2001): Mineralogie – Eine Einführung in die spezielle Mineralogie, Petrologie und Lagerstättenkunde. 6th Edition, Springer-Verlag, Berlin Heidelberg New York, 507 p.

Matthews, A.; Beckinsale, R.D. (1979): Oxygen isotope equilibration systematics between quartz and water. American Mineralogist, Vol. 64: pp. 232-240.

Mason, B.; Moore, C. B. (1982): Principles of Geochemistry Fourth Edition. John Wiley & Sons, 344 p.

McQueen, K.G. (2005): Ore Deposit Types and their primary expressions. In: Butt, C. R. M.; Scott, K. M.; Cornelius, M.; Robertson, I. D. M., eds., Regolith expression of Australian ore systems: a compilation of exploration case histories with conceptual dispersion, process and exploration models. CRC LEME, Australian National University, Canberra, Australia, 423 p.

Millet, M.-A.; Baker, J.A.; Payne, C.E. (2012): Ultra-precise stable Fe isotope measurements by high resolution multiple-collector inductively coupled plasma mass spectrometry with a 57Fe–58Fe double spike. Chemical Geology, Vol. 304-305: pp. 18-25.

Mogensen, F. (1946): A ferro-ortho-titanate ore from Södra Ulvön. Geologiska Föreningens i Stockholm Förhandlingar 68: pp. 578–588.

Moore, F.; Modabberi, S. (2003): Origin of Choghart iron oxide deposit, Bafq mining district, Central Iran: new isotopic and geochemical evidence. Journal of Sciences, Islamic Republic of Iran, Vol. 14: pp. 259-269.

Mysen B. O. (1988): Structure and properties of silicate melts. Elsevier, 354 p.

Nadeau, O.; Williams-Jones, A. E.; Stix, J. (2010): Sulphide magma as a source of metals in arc-related magmatic hydrothermal ore fluids. Nature Geoscience, Vol. 3: pp. 501-505.

Naranjo, J. A.; Henriquez, F.; Nyström, J. O. (2010): Subvolcanic contact metasomatism at El Laco Volcanic Complex, Central Andes. Andean Geology, Vol. 37, No. 1: pp. 110-120.

Naslund, H. R. (1983): The effect of oxygen fugacity on liquid immiscibility in iron-bearing silicate melts. American Journal of Science, Vol. 283: pp. 1034-1059.

Nicoll, G. R.; Holness, M. B.; Troll, V. R.; Donaldson, C.H.; Holohan, E. P.; Emeleus, C. H.; Chew, D. (2009): Early mafic magmatism and crustal anatexis on the Isle of Rum: evidence from the Am Màm intrusion breccias. Geological Magazine, Vol. 146: pp. 368-381.

Nomura, R.; Ozawa, H.; Tateno, S.; Hirose, K.; Hernlund, J.; Muto, S.; Ishii, H.; Hiraoka, N. (2011): Spin crossover and iron-rich silicate melt in the Earth’s deep mantle. Nature, Vol. 473: pp. 199-202.

Nyström, J. O.; Billström, K.; Henríquez, F.; Fallick, A. E.; Naslund, H. R. (2008): Oxygen isotope composition of magnetite in iron ores of the Kiruna type in Chile and Sweden. GFF Vol.130, Issue 4: pp. 177-188.

Oen, I. S.; Helmers, H.; Verschure, R. H.; Wiklander, U.; (1982): Ore deposition in a Proterozoic incipient rift zone environment: a tentative model for the Filipstad– Grythyttan–Hjulsjö region, Bergslagen, Sweden. Geologische Rundschau, Vol. 71: pp. 182–194.

O'Neil, J. R.; Clayton, R. N. (1964): Oxygen isotope geothermometry. In: Craig, H., Miller, S. L. & Wasserburg, G.J. (eds) Isotopic and Cosmic Chemistry, pp. 157-168. North-Holland Publishing Co., Amsterdam.

Paràk, T. (1975): Kiruna iron ores are not "intrusive-magmatic ores of the Kirunatype". Economic Geology, Vol.70: pp. 1242-1258.

Paràk, T. (1984): On the magmatic origin of iron ores of the Kirunatype – a discussion. Economic Geology, Vol. 79: pp. 1945-1949.

Philpotts, A. R. (1967): Origin of certain iron titanium oxide and apatite rocks. Econimic Geology, Vol. 62, No. 3: pp. 303–316.

Pollard, P. J. (2000): Evidence of a magmatic fluid and metal source for Fe-oxide Cu-Au mineralisation. In: Porter, T. M. (ed) Hydrothermal iron oxide copper-gold and related deposits: A global perspective. Australian Mineral Foundation, Adelaide, pp. 27–41.

Price, R. C.; Gamble, J. A.; Smith, I. E. M.; Stewart, R. B.; Eggins, S.; Wright, I. C. (2005): An integrated model for the temporal evolution of andesites and rhyolites and crustal development in New Zealand’s North Island. Journal of Volcanology and Geothermal Research, Vol. 140: pp. 1-24.

Price, R. C.; Turner, S.; Cook, C.; Hobden, B.; Smith, I. E. M.; Gamble, J. A.; Handley, H.; Maas, R.; Möebis, A. (2010): Crustal influences and U-Th-Ra disequilibrium in andesitic lavas of Ngauruhoe volcano, New Zealand. Chemical Geology, Vol. 244: pp. 437-451.

Price, R. C.; Gamble, J. A.; Smith, I. E. M.; Maas, R.; Waight, T.; Stewart, R. B.; Woodhead, J. (2012): The Anatomy of an AndesiteVolcano: a Time-Stratigraphic Study of Andesite Petrogenesis and Crustal Evolution at RuapehuVolcano, New Zealand. Journal of Petrology, Vol. 53: pp. 2139-2189.

Rhodes, A. L.; Oreskes, N. (1999): Oxygen isotope composition of magnetite deposits at El Laco. Chile: Evidence of formation from isotopically heavy fluids. Society of Economic Geologists Special Publication Nr.7: pp. 333-351.

Rhodes, A. L.; Oreskes, N.; Sheets, S. (1999): Geology and Rare Earth Element Geochemistry of Magnetite Deposits at El Laco, Chile. In: Skinner, B. J., ed., Geology and ore deposits of the Central Andes. Special Publication 7, Society of Economic Geologists. pp. 299-332.

Ripa, M. (2001): A review of the Fe oxide deposits of Bergslagen, Sweden and their connection to Au mineralization. In: Economic Geology Research Vol. 1 (1999-2000) In: Sveriges Geologiska Undersökning C 833: pp. 132-136

Ripa, M.; Kübler, L. (2003): Apatite-bearing iron ores in the Bergslagen region of south-central Sweden. Economic Geology Research, Vol. 2 (2001–2002) In: Sveriges geologiska undersökning Rapporter och meddelanden 113: pp. 49–54.

Roedder, E. (1978): Silicate liquid immiscibility in magmas and in the system K2O-FeO-Al2O3-Sio2: an example of serendipity. Geochimica et Cosmochimica Acta, Vol. 4: pp. 1597-1617.

Rollinson, H. (1993): Using geochemical data: evaluation, presentation, interpretation. Longman Group UK Limited, 352 p.

Roonasi, R.; Holmgren, A. (2009): A study on the mechanism of magnetite formation based on iron isotope fractionation. In: Howard, S. M. (ed) EPD Congress 2009 : proceedings of sessions and symposia sponsored by the Extraction and Processing Division (EPD) of The Minerals, Metals & Materials Society (TMS),Warrendale

Saager, R. (1984): Metallische Rohstoffe von Antimon bis Zirkonium. Bank Vontobel, Zürich, 176 p.

Sahar, A.; Young, E. D.; Manning, C. E. (2008): Equilibrium high-temperature Fe isotope fractionation between fayalite and magnetite: An experimental calibration. Earth and Planetary Science Letters, Vol. 268: pp. 330-338.

Sæmundsson, K. (1979): Outline of the geology of Iceland. Jokull, Vol. 29: pp. 7-28.

Sandecki, J. (2000): Mineralogical and genetical aspects of the Smålands Taberg Fe–Ti–V ore, Protogine Zone of Southern Sweden. GFF, Vol. 122: pp. 351– 358.

Schauble, E. A.; Rossman, G. R.; Taylor, H. P. (2001): Theoretical estimates of equilibrium Fe-isotope fractionations from vibrational spectroscopy. Geochim. Cosmochim. Acta, Vol. 65: pp. 2487– 2497.

Schmincke, H.-U.; Flower, M. J. F. (1974): Magmenevolution auf atlantischen Vulkaninseln. Naturwissenschaften, Vol. 61: pp. 288-297.

Schüßler, J. (2008): Controls on stable iron isotope variations in magmatic systems - Significance of mineral- melt isotopic fractionation in experiments and nature. PhD-Thesis, Faculty of Natural Science, Gottfried Wilhelm Leibniz Universität Hannover, Germany, 178 p.

Severmann, S.; Anbar, A. D. (2009): Reconstructing Palaeoredox Conditions through a Multitracer Approach: The Key to the past is the Present. Elements Vol. 5, Nr.6: pp.359-364

Sillitoe, R. H.; Burrows, D.R. (2002): New field evidence bearing on the origin of the El Laco magnetite deposit, northern Chile. Economic Geology, Vol. 97: pp. 1001-1109.

Sillitoe, R. H. (2003): Iron oxide-copper-gold deposits: an Andean view. Mineralium Deposita, Vol. 38: pp. 787- 812.

Sinclair, W. D. (2007): Porphyry deposits. In: Goodfellow, W. D., ed., Mineral Deposits of Canada: A Synthesis of Major Deposit-Types, District Metallogeny, the Evolution of Geological Provinces, and Exploration Methods. Geological Association of Canada, Mineral Deposits Division, Special Publication No. 5: pp. 223-243.

Söderlund, P.; Söderlund, U.; Möller, C.; Gorbatschev, R.; Rodhes, A. (2004): Petrology and ion microprobe U- Pb chronology applied to a metabasic intrusion in southern Sweden: A study on zircon formation during metamorphism and deformation. Tectonics, Vol. 23: TC5005, doi:10.1029/2003TC001498

Stephens, M. B.; Ripa, M.; Lundström, I.; Persson, L.; Bergman, T.; Ahl, M.; Wahlgren, C-H.; Persson, P-O.; Wickström, L. (2009): Synthesis of the bedrock geology on the Bergslagen region, Fennoscandian Shield, south- central Sweden. Sveriges Geologiska Undersökning, Ba 58, 259 p.

Stern, T.; Stratford, W.; Seward, A.; Henderson, M.; Savage, M.; Smith, E.; Benson, A.; Greve, S.; Salmon, M. (2010): Crust-mantle structure of the central North Island, New Zealand based on seismological observations. Journal of Volcanology and Geothermal Research, Vol. 190: pp. 58-74.

Stigh, J. (1982): Ultramafiter i svenska delen av fjällkedjan – Möjligt slutförvaringsplats för radioaktivt avfall? Rapport till Svensk Kärnbränsleförsörjning AB, Geological Survey of Sweden, 57 p.

Stillman, C. J. (1989): Some dyke-Gabbro relationships in the Troodos Ophiolite, Cyprus. In: Gibson, I. L.; Malpas, J.; Robinson, P. T.; Xenophontos, C. (eds.): Cyprus Crustal Study Project: initial report, hole CY-4. Geological Survey of Canada, Ottawa, Ontario, Canada: pp. 107-113.

Sundblad, K. (1994): A genetic reinterpretation of the Falun and Åmmeberg ore types, Bergslagen, Sweden. Mineralium Deposita, Vol. 29: pp. 170-179.

Tang, H. S. (1984): The Panzhihua V–Ti–magnetite deposits. In: Panxi Geological Team of Sichuan Geological Survey (eds) Research Report of Metallogeny and its Prediction (unpublished) (in Chinese).

Taylor, H. P. (1967): Oxygen isotope studies of hydrothermal mineral deposits. In: Barnes, H. L. (ed.): Geochemistry of Hydrothermal Ore Deposits: pp.109-142 , Holt, Rinehart and Winston Inc., New York.

Taylor, H. P. JR. (1968): The oxygen isotope geochemistry of igneous rocks. Contributions to Mineralogy and Petrology, Vol. 19: pp. 1 -17.

Taylor, H. P. (1971): Oxygen Isotope Evidence for Large-Scale Interaction between Meteoric Ground Waters and Tertiary Granodiorite Intrusions, Western Cascade Range, Oregon. Journal of Geophysical Research, Vol. 76, Nr. 32: pp. 7855-7874.

Tornos, F.; Velasco, F.; Morata, D.; Barra, F.; Rojo, M. (2011): The magmatic hydrothermal evolution of the El Laco deposit as tracked by melt inclusions and isotope data In: “Let’s Talk Ore deposits”. Barra, F.; Reich, M.; Campos, E.; Tornos, F. eds. Antofagasta: pp. 345-348.

Treloar, P. J.; Colley, H. (1996): Variations in F and Cl contents in apatites from magnetite-apatite ores in northern Chile, and their ore-genetic implications. Mineralogical Magazine, Vol. 60: pp. 285-301.

Trønnes, R. G. (2003): Geology and geodynamics of Iceland. Nordic Volcanological Institute, University of Iceland, 19 p.

U. S. Geological Survey. (2011): Mineral commodity summaries 2011. U.S. Geological Survey, 198 p.

Van Bemmelen, R. W. (1949): The geology of Indonesia. Government Printing Office, The Hague.

Vaughan, D. J. (2011): The mastery of iron. Elements Vol. 7: pp. 75.

Velasco, F.; Tornos, F. (2012): Insights on the Effects of the Hydrothermal Alteration in the El Laco Magnetite Deposit (Chile). Macla, Vol. 16: pp. 210-211.

Vivallo, W.; Rickard, D. (1984): Early Proterozoic ensialic spreadingsubsidence evidence from the Garpenberg enclave, Central Sweden. Precambrian Research, Vol. 26: pp. 203-221.

Walraven, F.; Armstrong, R. A.; Kruger, F. J. (1990): A chronostratigraphic framework for the northcentral Kaapvaal craton, the Bushveld Complex and the Vredefort structure. Tectonophysics, Vol. 171: pp. 23-48.

Weidner, J. R. (1982): Iron-oxide magmas in the system Fe-C-O. Canadian Mineralogist, Vol. 20: pp. 555-566.

Weis, F. (2011): Oxygen Isotope Signatures of the Apatite-Iron Oxide Ore at Grängesberg. Bachelor Thesis, Department of Earth Science, Uppsala University, 35 p.

Westerstrand, M.; Öhlander, B. (2011): Geochemical Effects of Increased Production on Recirculated Process Water at the Kiirunavaara Iron Mine, Northern Sweden. Mine Water and the Environment, Vol. 30: pp. 252- 262.

Wilson, M. R.; Hamilton, P. J.; Fallick, A. E.; Aftalion, M.; Michard, A. (1984): Granites and early Proterozoic crustal evolution in Sweden: evidence from Sm-Nd, U-Pb and O isotope systematic. Earth and Planetary Science Letters Nr. 72 (1985): pp.376-388.

Wolfe, C. J.; Bjarnason, I. T.; Van Decar, J. C.; Solomon, S. C. (1997): Seismic structure of the Iceland plume. Nature, Vol. 385: pp. 245-247.

Zhao, Z.-F.; Zheng, Y.-F. (2003): Calculation of Oxygen Isotope Fractionation in Magmatic Rocks. Chemical Geology, Vol. 193: pp. 59-80.

Zheng, Y.-F. (1991): Calculation of oxygen isotope fractionation in metal oxides. Geochimica et Cosmochimica Acta, Vol. 55: pp. 2299-2307.

Zhou, M.-F.; Robinson, P. T.; Lesher, C. M.; Keays, R. R.; Zhang, C.-J.; Malpas, J. (2005): Geochemistry, Petrogenesis and Metallogenesis of the Panzhihua Gabbroic Layered Intrusion and Associated Fe–Ti–V Oxide Deposits, Sichuan Province, SW China.. Journal of Petrology, Vol. 46, No. 11: pp. 2253–2280.

Internet resources: http://articles.latimes.com/2010/dec/28/business/la-fi-china-rare-earth-20101228 (last accessed: 02.01.2013) http://web.uct.ac.za/depts/geolsci/dlr/301s/bv_field/bv43.jpg (last accessed: 01.01.2013) http://www.geology.neab.net/travel/2004/bergslagen/Striberg-1.jpg (last accessed: 01.01.2013) http://www.lansstyrelsen.se/dalarna/SiteCollectionImages/Sv/samhallsplanering-och-kulturmiljo/skyddad- bebyggelse/Centralanl%25201969.jpg (last accessed: 02.01.2013) http://www.mineralatlas.eu/lexikon/index.php/Schweden/Dalarna,%20Provinz/Falun/Falu%20Gruva (last accessed: 01.01.2013) http://maps2.sgu.se/kartgenerator/maporder_sv.html (last accessed: 08.01.2013)

14. Appendix

14.1 Appendix 1

The thermometer calculations for quartz magnetite pairs were conducted on three massive ore bodies at Grängesberg. The calculations were done by using Eq. 3 & 4 and rearranging in order to solve for the temperature T.

Combining Eq. 3 and 4 gives

6 2 δqz – δmgt = A (10 /T ) + B, where A = 6.29 and B = 0.

Rearranging and solving for T (in Kelvin) gives

δ δ

Sample KES090034:

18 18 δ Omagnetite = +0.5 ±0.2 ‰ δ Oquartz = +8.7 ±0.2 ‰ δqz – δmgt = +8.20 ±0.4 ‰



 T = 876 K = 603 ± 21°C

Sample KES103003:

18 18 δ Omagnetite = +0.2 ±0.2 ‰ δ Oquartz = +7.6 ±0.2 ‰ δqz – δmgt = +7.4 ±0.4 ‰



 T = 921 K = 649 ± 24°C

Sample KES103016:

18 18 δ Omagnetite = +1.5 ±0.2 ‰ δ Oquartz = +7.9 ±0.2 ‰ δqz – δmgt = +6.4 ±0.4 ‰



 T = 991 K = 718 ± 31°C

Tidigare utgivna publikationer i serien ISSN 1650-6553

Nr 1 Geomorphological mapping and hazard assessment of alpine areas in Vorarlberg, Austria, Marcus Gustavsson

Nr 2 Verification of the Turbulence Index used at SMHI, Stefan Bergman

Nr 3 Forecasting the next day’s maximum and minimum temperature in Vancouver, Canada by using artificial neural network models, Magnus Nilsson

Nr 4 The tectonic history of the Skyttorp-Vattholma fault zone, south-central Sweden, Anna Victoria Engström

Nr 5 Investigation on Surface energy fluxes and their relationship to synoptic weather patterns on Storglaciären, northern Sweden, Yvonne Kramer

Nr 241 Re-processing of Shallow and Deep Crustal Reflection Seismic Data along BABEL Line7, Central Sweden, Hanieh Shahrokhi, August 2012

Nr 242 Usability of Standard Monitored Rainfall-Runoff Data in Panama, Juan Diaz River Basin, José Eduardo Reynolds Puga, August 2012

Nr 243 Numerical Model of a Fossil Hydrothermal System in the Southern East Pacific Rise Exposed at Pito Deep, Páll Halldór Björgúlfsson, September 2012

Nr 244 Regional Precipitation Study in Central America, Using the WRF Model Tito Maldonado, September 2012

Nr 245 Reduktion av järn, mangan och CODMn i dricksvatten – Ett pilotförsök vid Högåsens vattenverk, Tommy Olausson, September 2012

Nr 246Short-term Variations in Ice Dynamics During the Spring and Summer Period on Storglaciären, Kebnekaise, Sweden, Helena Psaros, September 2012

Nr 247 Reassessment of the ‘last’ Goniopholidid Denazinosuchus kirtlandicus Wiman, 1932 from the Late Cretaceous of New Mexico, Oskar Bremer, October2012

Nr 248 Variability of Intermittent Headwater Streams in a Boreal Landscape Influence of different discharge conditions, Tum Nihm, October2012

Nr 249 2D Reflection Seismic Imaging at a Quick-clay Landslide Site, in Southwest Sweden, Muhammad Umar Saleem, November 2012

Nr 250 A Strategy for Earthquake Catalog Relocations Using a Maximum Likelihood Method, Ka Lok Li, December 2012