Factors governing marble lightness in peripheral alteration haloes around carbonate-hosted Zn-Pb-Ag-(Cu-Au) deposits, Garpenberg, Sweden

(1)

Factors governing marble lightness in peripheral alteration haloes around carbonate-hosted Zn-Pb-Ag-(Cu-Au)

deposits, Garpenberg, Sweden

Marcus Eriksson

Civil Engineering, master's level 2020

Luleå University of Technology

Department of Civil, Environmental and Natural Resources Engineering

(2)

I

Abstract

A Master thesis about the Garpenberg deposit located in Bergslagen, a lithotectonic domain, with a mining history that might date back as far as 350 BC. Marble- and skarn-hosted sulfide deposits are found in the area, which creates the opportunity to mine both limestone and sulfidic ore in a single mine. Garpenberg is such a location hence this thesis, which aims to quantify the factors governing spectrophotometric lightness in marble at the Dammjön ore body. The work is mainly based on five drill cores which were logged and sampled. A total of twenty-seven samples were characterized using lithogeochemical analysis and thin-section analysis. The amount of Acid Insoluble Residue (AIR), magnetic minerals and the spectrophotometric lightness were determined for the same samples.

The calcite marble was divided into seven different varieties; 1) calcite marble breccia, 2) light, 3) grey, 4) green, 5) banded salmon pink, 6) ophicalcite and 7) spotted calcite marble. The dolomite marble is white to grey in color and skarn minerals are common and varies between 5-20 vol.%. Grey and light calcite marble are the varieties with the highest spectrophotometric lightness, and it could be shown that the lightness increases with a decreasing amount of titanium, aluminum and zirconium which are chemical proxies for mineralogical impurities of originally volcaniclastic origin.

High-quality calcite marble is a potentially economic by-product at the Garpenberg mine, the lightest samples are nearly as light as the light standard used during analysis (92.45 out of 100%). The lightest marble is also the chemically most pure which means that the calcium oxide (CaO) and total-carbon content are high.

Key geological factors detrimental to lightness and purity are the primary composition, which is determined by the admixture of volcaniclastic material in the limestone precursor. Hydrothermal alteration with the addition of silicates, sulfides and oxides forms a halo around the massive sulfide lenses. Dolomite marble, which is more proximal to ore, is richer in manganese and sulfides, and not as light as the calcite marble at Dammsjön.

(3)

II

Foreword

This MSc project was conducted within the VectOre project of the Strategic Innovation Programme for the Swedish Mining and Metal Producing Industry (STRIM) of VINNOVA, Formas and the Swedish Energy Agency, with additional financial support from Boliden and Björka Mineral.

I would like to thank Nils Jansson for his help on behalf of Luleå University of Technology throughout the entire project as my supervisor and for the valuable input he provided, but also for arranging this project.

A thanks goes out to the employees at Boliden mineral AB for helping me during my time in Garpenberg. All the geologists who assisted me and the employees at the core shed that were so friendly and made my stay very enjoyable. Thanks to Björka mineral for the support and help with sample analysis.

To the staff in the laboratories at Luleå University of technology for letting me use their facilities and equipment. Where my appreciation goes to Glacialle Tiu for her help calibrating and running the SEM and Annutam Patra for his expertise and help in the chemistry lab.

Finally, I would like to thank my family and friends for all the support and kind words that helped me incredibly during my studies. Without them I would not be here today.

(4)

III

List of appendices

Appendix I: Sample descriptions 21 pages Appendix II: Complete graphical logs 12 pages Appendix III: Lithogeochemical data sheets 7 pages

List of figures

Figure 1. Overview of the geological units and lithotectonic domain. --- 1

Figure 2. Map over the regional geology at the Garpenberg deposit. --- 5

Figure 3. Top view of orebodies at the Garpenberg deposit.. --- 7

Figure 4. Zeiss Merlin FEG-SEM at Luleå University of technology. --- 9

Figure 5. Histogram showing the number of samples in each classification group. --- 11

Figure 6. A) Hand specimen of the dolomite marble B) BSE image. --- 11

Figure 7. A) Large crystal of diopside. B) Typical pyroxene cleavage. --- 12

Figure 8. A) Altered mafic unit. B) Cluster of intergrown sphalerite. --- 12

Figure 9. Examples of the seven marble varieties found in Dammsjön.--- 13

Figure 10. A)Drill core representing calcite breccia. B) The two types of plagioclase. C) euhedral pyrite crystal and pyrrhotite. --- 14

Figure 11. Gradual transition between light calcite marble to a banded salmon pink calcite marble.-14 Figure 12. A) Pink marble in cross-polarized light. B) SBE spectrum of the fibrous chlorite.--- 15

Figure 13. A) A cluster of quartz and tremolite in grey marble. B) Tremolite alteration with inclusions of calcite. --- 15

Figure 14. Calcite dominated light calcite with a quartz cluster and tremolite. --- 15

Figure 15. A) Sericite altered plagioclase. B) Accessory minerals in the green marble. --- 16

Figure 16. A) Spotted calcite marble. B) Pyrrhotite and galena. --- 16

Figure 17. A) Broken olivine crystal. B) Intergrown sulfide crystal. --- 16

Figure 18. Drill core log 3695 --- 17

Figure 19: Drill core log 3740 --- 17

Figure 20. Drill core log 2286---18

Figure 23: Box-plot of the marble lightness.---20

Figure 24: Box-plot of the marble lightness. --- 21

Figure 25. Box-plot of the marble varieties regarding the AIR. --- 22

Figure 26. A) The CaO content in % plotted against the spectrophotometric lightness. B) The tot. Carbon content in % plotted against the spectrophotometric lightness. --- 22

Figure 27. A) Lithogeochemical plot. B) Lithogeochemical plot showing AIR vs Tot.Carbon --- 23

Figure 28. A) Ti02 B) Al2O3 C) Zr D) MnO –plotted against the spectrophotometric lightness --- 23

Figure 29. Plot that discriminate between input of hydrothermal and primary volcanoclastic component. --- 24

Figure 30. Quality specifications for limestone products used for liming lakes. --- 25

(6)

1

1 Introduction

1.1 Background

Bergslagen is a region with history of mining dating back to 500 AD. Since then over 6000 deposits with diverse character are known (Stephens et al., 2009). Banded iron-formations, different types of skarns, strata-bound and stratiform Zn-Pb-Ag (Cu, Au) sulfide ores are among these diverse ore-types (Allen et al., 1996). Today three mines are in operation, Zinkgruvan, Lovisagruvan and Garpenberg which are all sulfide deposits (Stephens et al., 2009). Non-metallic mineral deposits have also been mined throughout time but as with the metallic deposits, most of the quarries are abandoned. The geological survey of Sweden (SGU) have a record of 400 quarries from deposits of crystalline limestone with varieties of calcite and dolomite marble. (Stephens et al., 2009). The marble hosting the ore in Garpenberg have recently been appreciated for its potential for use as an industrial mineral, but no data on key quality factors for industrial minerals are available, hence this project.

In this report, the focus will be on the Garpenberg deposit which is owned 100% by Boliden Mineral AB. With a production of 2625 Kton in 2018 and estimated mineral reserves of 76 Mt of polymetallic ore lenses mined between 500 and 1200 meter below surface. (Boliden mineral AB, 2018)

The mining industry of today is facing big challenges in the ongoing change towards becoming a more sustainable industry. One major problem is the amount of waste produced for each ton of ore extracted. A possibility to solve this problem is to change the view of what is waste and what is a valuable resource in a polymetallic sulfide mine.

1.2 Objectives & aim

This project is supported by Boliden mineral AB and Björka mineral and is a part of the bigger research project, VectOre which focuses on marble-hosted mineral deposits in Bergslagen. VectOre forms part of the Strategic Innovation Program (STRIM) for the Swedish Mining and Metal Producing Industry of VINNOVA, Formas and the Swedish Energy Agency.

This project aims to quantify the factors governing the marble spectrophotometric lightness at Dammsjön, Garpenberg, where the objectives have been to:

 Sub-divide and characterize different marble varieties in terms of texture, minerology, composition and spectrophotometric lightness.

 Interpret the spatial distribution of the different marble varieties using logging and lithogeochemical data.

 Determine the key controls on marble lightness, e.g. the importance of primary composition vs hydrothermal alteration.

The work is based on five drill cores from level 1075 in Dammsjön with complimentary samples from level 700 during a mine visit. The cores were logged and sampled for lithogeochemical analysis and thin sections. The amount of Acid Insoluble Residue (AIR), magnetic minerals and the spectrophotometric lightness were determined for the same samples sent for lithogeochemical analysis. To determine the spectrophotometric lightness and determine key controls are the first steps in evaluating whether if co-mining of sulfide ore and marble should be investigated further.

(7)

2

2 Regional Geology

2.1 Fennoscandian shield

Garpenberg is located in Bergslagen, a lithotectonic domain (BLD) shown in Figure 1, formed during the Svecokarelian orogeny. Which together with the Blekinge-Bornholm and Sveconorwegian orogeny formed the Fennoscandian shield (Allen et al, 2013). The Svecokarelian orogen was formed during cyclic magmatic events that is believed to be related to a convergent continental plate margin between 2.0-1.8 Ga and accretionary activity related to subduction (Cawood et al., 2009).The cycle periods of magmatism and sedimentation is considered to be 40-50 Ma (Hermansson et al., 2008) with shorter periods of ductile deformation forming shear zones interpreted as the boundaries of the Lithotectonic domains seen in Figure 1 (Allen et al, 2013). The rocks generated during a cycle dated to 1.9-1.87 Ga is associated with many of the mineral deposits (Allen et al., 2013) found within the so-called Svecofennian domain (Gaál & Gorbatschev, 1987). A total of four magmatic cycles contributed to the formation of the Svecofennian domain where the oldest cycle is dated to 1.95 Ga within the Bothnia- Skellefteå lithotectonic domain (Allen et al., 2013).

2.2 Bergslagen

The bergslagen lithotectonic domain (BLD) has been interpreted as a continental back-arc region by Allen (1996), where bimodal volcanic activity in the western part of bergslagen as well as the abundant calc-alkaline rhyolitic volcanic rocks throughout the whole region can be seen as an indication of crustal extension. Later interpretations by Hermansson et al. (2008) and Allen et al., (2013) have evolved the

Figure 1. Overview of the geological units and lithotectonic domains forming the fennoscandian shield (modified after Koistinen, 2001, Allen et al, 2013).

(8)

3 tectonic interpretation regarding a cyclic environment during the formation of Bergslagen, which can be applied to the whole svecokarelian orogeny. A complex interplay between magmatic activity, ductile deformation and metamorphism in cycles spanning 40-50 Ma. This is a product of the rapid changes between continental back-arc extensional environments to a compression system due to the shift in direction of the subduction hinge (Hermansson et al., 2008).

This tectonic evolution has resulted in a succession of metavolcanic rocks and metasedimentary rocks together with large suites of pre- to post tectonic intrusions (Allen et al., 1996). The intrusions vary in composition and in the degree of overprint by deformation, depending on the timing. Compositionally they range from granites to gabbro but are mainly felsic in composition (Lundström, 1987).

The magmatic signature of the rocks in Berslagen indicate a single source of juvenile mantle material with no influence of Archean material, supporting the theory that a volcanic-arc crust formed and became the continental basement for the Bergslagen lithotectonic domain (Allen et al., 2013).

2.2.1 Structures

Bergslagen can be divided into four domains based on timing and grade of metamorphism as well as the timing and degree of ductile deformation (Stephens et al., 2009). The domains are referred to as the west, north, central and south domain. The north and south domain are similar in style; both comprising steeply dipping high-strain belts with a strike west-north-west-east-south-east formed under amphibolite-facies metamorphic conditions, along with more discrete zones formed under lower-grade metamorphic conditions (Stephens et al., 2009; Allen et al., 2013) In the northern part the ductile fabric was formed at 1.87 to 1.86 Ga and has later been subjected by folding.

In the eastern part of the central domain there is a high strain zone related to the zones found in the north. These ductile grain fabrics are thought to have been formed at 1.88-186 Ga under amphibole- facies metamorphic conditions prior to folding and later deformation (Stephens et al., 2009)

Complex relations between old structures and later deformation is prominent in the western domain.

The later deformation is related to the Sveconorwegian orogen at 0.98-0.95 Ga which formed north- south striking structures whilst the older structures have an east-west strike. The Sveconorwegian orogeny is responsible for the most noticeable metamorphic phase of regional greenschist metamorphic overprint in the western part of the Bergslagen lithotectonic domain (Stephens et al., 2009).

(9)

4 2.2.2 Carbonate rocks in Bergslagen

According to Allen et al. (2003), the carbonate rocks found in Bergslagen can be divided into four groups, including two major and two minor facies. Regionally extensive thick and clean dolomitic or calcitic marble with cyclic repetitions of ash-siltstone interbeds is characteristic of the first major facies.

Such units can be found at Sala and Glanshammar where medium grained dolomite marble is dominant whereas in Garpenberg it is represented by a coarse calcite marble (Allen et al., 2003). Thinner dolomite and calcite marble units with limited lateral extent is significant for the second major facies.

They range from clean beds of marble with or without interbeds of ash-siltstone and usually have zones of skarn or in some cases, they are entirely replaced by skarn (Allen et al., 2003). The minor carbonate- facies are not currently dominated by carbonates but rather by calc-silicates such as epidote, actinolite, garnet interbeds with ash-siltstone in regionally extensive 0.3-10m thick beds. Here it is inferred that the carbonates where precursor to calc-silicates, which formed by alteration and metamorphism. The fourth facies is similar, with laminated character but with limestone or magnetite/hematite quartz together with the ash-siltstone instead (Allen et al., 2003)

Stromatolite fossils have been documented in parts of the thick marble units in Bergslagen (Collini, 1965; Allen et al., 1996, 2003; Jansson 2017). This is a strong evidence that the main mechanism for building the extensive marble units found in Bergslagen was microbial activity in reefs. Most of the marble units are associated with sub-wave base conditions of the rhyolitic ash siltstone (RAS) facies.

Stromatolite textures have not been found in Garpenberg but is believed to have formed under similar conditions at shallow sub-wave base level which can be validated by the carbon and oxygen isotopic composition (Allen et al., 2003). Shallow marine volcanism generated a lot of pyroclastic debris that formed extensive sheets of pyroclastic material around the caldera structures. These deposits were continuously reworked and redeposited to form the RAS facies (Allen et al., 2003). This environment was not optimal for reef building which explain why many of the thickest and more extensive marble units can be found in the upper part of the stratigraphy when the volcanic activity was declining (Allen et al., 1996, 2003). The lack of primary textures can be expected due to higher metamorphic grade in the area together with the hydrothermal alteration that also reworked the calcite marble to dolomite proximal to the mineralization (Allen et al, 2003)

The town Sala, Sweden, was built during medieval times due to the important silver deposits hosted by the dolomite. The Sala deposit is a stratabound, carbonate hosted replacement type of ore with Zn- Pb-Ag, the same group of mineralization mined today at Garpenberg. (Jansson, 2017). The Sala deposit is no longer in operation, but the dolomite marble is mined as an industrial mineral today only 200 meters from the deposit. This is an example of where Zn-Pb-Ag mineralization and high quality carbonate rock co-exist in the same mining area.

(10)

5 2.3 Garpenberg

2.3.1 Geology

The Garpenberg deposit is hosted by an extensive calcite-dominated marble unit seen in Figure 2. The marble can be found in a supracrustal inlier of metavolcanic rocks and is believed to have formed during a pause in volcanic activity (Allen et al., 2003). The package of metavolcanic rocks are limited by granite intrusions to the west and a shear zone in the east called Stora Jelken. The inlier forms a tight to isoclinal syncline that is steeply dipping towards north-east (Allen et al., 2013) and the marble unit found in the north-western limb of the syncline host at least nine orebodies of polymetallic sulfide mineralization. For the stratigraphic description suggested by Allen et al. (2003) primary volcanological and sedimentological names are used to describe the rocks of Garpenberg, despite that are all affected by metamorphism. The stratigraphy is dominated by rhyolitic volcaniclastic rocks with minor basaltic and dacitic extrusive rocks as well as some syn-volcanic intrusion with compositions ranging from mafic to rhyolitic. (Jansson, 2011)

Figure 2. Map over the regional geology at the Garpenberg deposit were all rocks have undergone metamorphism, modified after Allen et al, (2003)

(11)

6 2.3.2 Stratigraphy

Lower footwall rhyolitic ash-siltstone and minor coarse volcanoclastic units.

The lower footwall is around 500 m thick and is mainly composed of rhyolitic ash-siltstone and sandstone (Allen et al., 2003), where the ash-siltstone units represent subaqueous sedimentation of vitric ash below wave base. The coarser units represent mass flow deposition of juvenile pyroclastic debris (Allen et al., 1996). The lower footwall contains two major limestone units that host magnetite iron skarn deposits as well as Zn-Pb-Ag deposits. These ash-siltstone units also contain interbeds of stratiform skarn and pods containing epidote+calcic clinoamphibole (Jansson, 2011).

Basalt lava

Mafic rocks can be found in roughly the same stratigraphic position throughout the Garpenberg syncline and define an arbitrary boundary between the lower and upper footwall. The basaltic lavas found at Garpenberg Norra, have been metamorphosed to dark green amphibolite, yet still relict pillow structures that indicate a subaqueous deposition have been observed. No information about the water depth can be interpreted (Allen et al., 2003)

Upper footwall felsic volcaniclastic succession

The upper footwall is divided into four sub-units, where the lowest stratigraphic unit in this succession is an 80 m thick feldspar-porphyritic rhyolitic pumice breccia deposited as a subaqueous mass flow deposit. This unit is overlain by a 150 m thick unit of dacitic volcaniclastic material that vary from pumice-breccia sandstones to ash-siltstone. The uppermost unit is dominated by ash-siltstone, followed by a feldspar± quartz porphyritic pumice breccia sandstone sequence. (Allen et al., 2003) Limestone (calcite-, dolomite marble, skarn)

The main limestone consists of coarse calcite marble that vary in color between white, pink/orange.

Closer to the mineralization the marble changes from the coarse-grained calcite into a fine to medium grained dolomite marble as well as medium to coarse grained skarn. The metamorphic recrystallization has destroyed most of the primary structures and deformation of the rock have replaced the primary structures with ductile deformation lineation and weak cleavages. Throughout the limestone unit, several thin interbeds of volcaniclastic material with rhyolitic to dacitic composition can be found. The interbeds often display boudinage and record strong phlogopite- or skarn alteration (Allen et al., 2003).

Lower hanging-wall limestone-volcanic breccia-conglomerate sequence

This unit is varying in thickness between 30-110m and consists of beds of breccia-conglomerate with clasts of marble and metavolcanic rocks. The conglomerates are interbedded with other volcanic rocks and occur as discontinuous lenses (Allen et al., 2003). The breccia is usually clast-supported and in Dammsjön, the clasts of marble are often replaced by skarn minerals giving the breccia a green color (Lowe, 2012).

Upper hanging-wall rhyolitic pumice breccia

The stratigraphically highest unit is a several hundred-meter-thick unit of rhyolitic pumice breccia that is chemically homogeneous with juvenile pumice and breccia texture. Feldspar and quartz crystals can also be found in the unit (Allen et al., 2003) that has been interpreted by Allen et al. (1996) to have formed during a major caldera forming pyroclastic eruption following the pause in activity which led to the formation of the main limestone unit that can be found deeper in the stratigraphy.

(12)

7 2.3.3 Garpenberg ore bodies

The Garpenberg deposit is located on the north-west limb of the regional syncline shown in Figure 2, it is comprised of several smaller pod-like lenses that together have an ore reserve of 76 Mt (Boliden summary, 2018). The deposit is classified as a Stratabound volcanic-associated limestone skarn (SVALS) Zn-Pb-Ag-(Au-Cu) sulfide deposit (Allen et al., 1996). The main ore is hosted by the marble and the adjacent felsic volcanic rock. The ore ranges from massive, semi massive to disseminated and vein network in character. The massive lenses are dominated by sphalerite and galena with some pyrite and minor pyrrhotite and chalcopyrite. Ag- rich zones are found as thin vein networks in the dolomite marble, often closely related to tremolite skarn with Gn+Sph+Py mineralization (Allen et al., 2013).

The ore occur below a thick unit of pyroclastic material in the stratigraphy that has been affected by the hydrothermal alteration, but it is not as strong as the alteration found in the footwall. The asymmetry between the strong and extensive alteration in the footwall and the moderate alteration in the hanging wall described by Allen et al. (2003) indicate that the timing of the ore formation is synchronized with the rapid formation of the pyroclastic unit (Allen et al., 1996) The Garpenberg mineralization have a high Mn/Fe ratio compared to regional skarn. A depletion in δ¹⁸O isotopes and δ¹³C iostopes close to the ore can help distinguish between regional dolomitization and ore related dolomites (Allen et al. 2003, Jansson, 2011). The Dammjö ore body is located below the lake Dammsjön at surface which is situated in the central part of the deposit, it is subvertical and closely spatially related to the main marble unit (Lowe, 2012).

Figure 3. Top view of orebodies at the Garpenberg deposit with infrastructure from level 850 and the local coordinate grid.

The green rectangle indicates the focus area of this project. Modified after Boliden Mineral AB (2011)

(13)

8

3 Method

3.1 Logging

Core logging was conducted in Garpenberg at the core archive. Five cores were logged with a length of 200-300 m each. The major rock unit as described by (Allen et al., 2003, Lowe 2012) and summarized above was the base for the logging. This project focused on the marble unit and thus intervals containing carbonate rocks were logged in more detail and further subdivided. The detailed logging was carried out by hand on graphical logs where a lithographic description, grain-size, alteration and mineralogy was described, and a graphic log was drawn based on the information (Appendix II).

Table 1. Drillhole ID for the cores logged and their length in meter.

Hole ID

Length (m) GARPN 2252 200 GARPN 3695 290 GARPN 2286 250 GARPN 2544 222 GARPN 3740 200

3.2 Whole rocks sampling and analysis

Twenty-seven samples were collected in total, mostly from drill core but two were taken as outcrop grab samples at coordinates (X 900, Y 2900, Z 680) in the underground workings of the mine. A minimum length of 1 m of homogeneous rock were selected for each sample. The sample points were marked on the core boxes and the sampled section was then sawed in half using a diamond saw. If the core had already been sampled it was cut into quarters as one part of the core must be stored as future reference.

The samples were sanded with a 60-grit sandpaper to remove dirt and impurities that drilling, and storage of the core had caused. A plastic bag was labeled with the sample ID and the core sample was put inside and sealed. The samples were given an ID were numbered in a sequence of LK20190500 to LK20190530.

The samples were sent to ALS Chemex in Öjebyn (Sweden) for sample preparation into rock pulps with method Prep-31. The pulps were sent to Bureau Veritas (ACME) in Vancouver (Canada) for whole-rock analysis. The LF200 Analytical package from ACME was ordered which include whole rock analysis using lithium borate fusion for major and minor elements by Induced coupled plasma ICP-ES and ICP-MS for trace elements. The package also includes carbon and sulfur analysis by Leco furnace (TC000). Aqua regia digestion (AQ200) using a modified 1:1:1 (HNO3, HCl, H2O) solution for digestion of easily soluble species containing metals that were analyzed with ICP-ES/MS.

Two reference samples provided from a known composition of carbonate rock were inserted in the dataset, as well as two in-house certified standards at Boliden Mineral AB. These samples provide a quality control of the lithogeochemical analysis.

3.3 Optical & scanning electron microscopy

Out of the twenty-seven samples, eighteen were chosen so that a variety of lithologies could be examined closer by making polished thin-sections. Precision Petrographics, Canada prepared the polished thin sections, which had an approximate size of 26x46 mm and a thickness of 30 µm. The thin sections were first investigated in an optical microscope to determine the mineralogical composition.

A Leica DM750 P with objectives ranging from 4X to 40X zoom at Luleå University of Technology was

(14)

9 used for the petrographical study in reflected, plane and

cross-polarized light. A Nikon ECLIPSE LV100 POL connected to a Nikon DS-Fil U2 color camera was used to take photographs of important features in each thin section.

The scanning electron microscope used was a high resolution (0.8nm) Zeiss Merlin FEG-SEM at Luleå University of Technology, Figure 4. The microscope is equipped with double detectors for back scattered electron (BSE) and secondary electron (SE) imaging, wavelength-dispersive spectroscopy (WDS) and energy dispersive system (EDS). The software used for analysis of WDS and EDS was Inca and Aztec from Oxford Instruments. A total of five thin sections were prepared with a carbon coating prior to the SEM analysis, where two sections were attached to a sample holder prior to the placement in a vacuum chamber where they were exposed to the electron beam. The beam had a working distance of approximately 8.5 mm, the voltage was set to 20 kV for the ion acceleration and a beam current of 1 nA.

3.4 Acid insoluble residue (AIR)

The test was conducted at Luleå University of Technology in the facilities of the chemistry department. Below is the course of action, and the necessary information and equipment needed to conduct the test. The guide is based on information provided by Björka Mineral (Ernström produktion AB, 1993) and was revised together with the chemistry department.

EQUIPMENT

1. Balance (accuracy, 0.001 gram)

2. Safety equipment (protective glasses, plastic gloves etc.) 3. Buchner funnel

4. Graded cylinder (250 ml) 5. Oven (105°C)

6. Beaker 500 ml

7. Flask with rubber stopper and outlet for suction apparatus 8. Suction apparatus

9. Magnetic stirrer and magnetic rod 10. Glass rod

11. Filter (pore size 3) 12. Fume cupboard

13. Small plastic trey to weigh the sample on 14. Deionized water

REAGENT

1. Hydrochloric acid (25%)

Figure 4. Zeiss Merlin FEG-SEM at Luleå University of technology.

(15)

10 METHOD

1. Add 50 ml of deionized water to the 500 ml beaker.

2. Add 50 ml hydrochloric acid (gradually), into the same beaker.

3. Place the beaker on the magnetic stirrer and turn it on.

4. Weigh 10 gram of sample (A) with two decimals.

5. Add the sample slowly into the beaker.

6. Let it react for 12 minutes, turn the stirrer of and let it sediment for 3 minutes.

7. Decant the liquid into a pre-weighed filter (B) in the Buchner funnel using the flask with suction apparatus.

8. Rinse the equipment over the funnel to make sure all the residue ends up in the filter using deionized water

9. Dry the filter with sample residue in the oven at 105ºC for 90 min.

10. Cool to room temperature and weigh the filter with the residue (C).

CALCULATION

Acid insoluble residue (wt. %) =( 𝑊𝑒𝑖𝑔ℎ𝑡 𝐶−𝑊𝑒𝑖𝑔ℎ𝑡 𝐵) 𝑊𝑒𝑖𝑔ℎ𝑡 𝐴 *100

3.5 Spectrophotometric lightness analysis

The lightness analyses were carried out by Björka Mineral in their facilities. Rejects from the lithogeochemical sample preparation were crushed using a Retch Jaw Crusher down to a grain size of less than 2 mm. The material was milled and sieved until a powder sample with a grain-size of less than 125 µm remained. This material was pressed into briquettes using a hydraulic sample press. An Elrepho 450 X Spectrophotometer was used to determine the lightness and yellowness of the sample briquettes which was calibrated against an internal standard at Björka Mineral.

3.6 Magnetic separation

A handheld electromagnet was used for the magnetic separation, in the mineral lab at Luleå University of Technology. From a pulp-sample of 50g prepared by ALS, a sub-sample of about 10 g was measured on a scale with two decimal accuracy in a non-magnetic aluminum tray. The tray was placed under the magnet as it was turned on. The magnet was guided over the fine material at 1 cm, and this was repeated three times to make sure no magnetic material was missed. A second tray was placed under the active magnet which was then turned off, resulting in that the magnetic material was released from the magnet and was collected in the second tray. The sample was weighed again to get the mass of the magnetic fraction.

3.7 Data analysis and presentation

The data produced in the project was compiled in Microsoft office Excel, which was imported into IMDEX ioGAS^TM for data analysis which produced most of the graphs presented in this report. The graphical logs were digitized in Adobe illustrator CC 2018.

(16)

11

4 Results

4.1 Lithologies

The samples were divided into different groups based on their color and physical appearances during logging. The different groups are shown in Figure 5 together with the sample count for each group.

Calcite marble is sub-divided with the purpose to establish what parameters are the source for the different characteristics. Calcite breccia, spotted calcite and ophicalcite are variations of marble with more visible impurities.

4.2.1 Dolomite marble

The dolomite marble represented by the orange column in Figure 5 is white to grey in color and skarn alteration is common and varies between 5-20 vol.%.

This can give the marble a green color at times. The thickness of the dolomite in Dammsjön is at its thickest up to 60 meters but is more commonly only a few meters thick. The dark veinlets found in the dolomite sample LK20190505 (Fig.

6A) mainly consists of galena and the sulfosalt dyscrasite which is a Silver(Ag)+Antimony(Sb) mineral. Figure 6B show the two minerals and how they are intergrown with eachother.

Figure 5. Histogram showing the number of samples in each classification group.

Dyscrasite

Galena

Figure 6. A) Hand specimen of the dolomite marble B) BSE image of a dark veinlet in sample LK20190505.

A

B

(17)

12 4.2.2 Skarn

Multiple types of skarns were encountered but only one was sampled. An extremely coarse grained diopside skarn, with crystals larger than 2 cm. The typical cleavage for pyroxene which can be seen in Figure 7 together with the light green color indicate that this is a diopside skarn with a few large calcite crystals still present. The other skarn types found within Dammsjön were actinolite+garnet skarn as well as tremolite skarn.

4.2.3 Mafic dyke

The mafic unit is fine to medium grained that has been silicified and altered (Fig 8A). The hand specimen has a homogeneous dark green color. The mafic unit is thin (4 meters) found in repetitions of three thin dykes creating a small dyke swarm. Sphalerite is present as clusters of intergrown small crystals shown in Figure 8B.

Figure 7. A) Large crystal of diopside to the left and a large calcite crystal to the right, in cross-polarized light. B) Typical pyroxene cleavage in the large diopside shown in A.

A B

Figure 8. A) Altered mafic unit with actinolite, plagioclase and quartz in a homogeneous mix. B) Cluster of intergrown sphalerite crystals found within the mafic unit.

A B

(18)

13 4.2.4 Calcite marble:

The calcite marble was divided into seven different varieties seen in Figure 9; breccia, light, grey, green, banded salmon pink, ophicalcite and spotted calcite marble, where the banded salmon pink marble will be referred to as the pink marble. They are all medium to coarse grained with no visible sulfide mineralization in the hand specimen. The main calcite marble body have a thickness of ca 50 meters and a length of 500 meters, where drill hole GARPN 3695, GARPN 3740, GARPN 2286 and GARPN 2252 all intersect both contacts.

Marble breccia: Calcite breccia with dark bands of volcanic material, where the metamorphism and deformation have established band-like structures in the breccia due to stretch lineation in one plane.

The breccia is found in contact with volcanic units either as inliers in the thick marble unit or close to its contact. Is similar to the polymict breccia-conglomerate described by (Lowe 2012) in appearance but with more dominant marble content. Figure 10A show a section from drill core 3740 which represent the calcite marble breccia. The mineral composition is dominated by calcite with zones containing significant amounts of clastic material. The clastic material is mostly plagioclase with zoning and twinning that has been affected by sericite alteration shown in Figure 10B. Minor biotite/phlogopite as well as fine-grained quartz. Euhedral crystals of pyrite and pyrrhotite are also present in the marble breccia (Fig.10C).

Pink marble Light marble

Green marble

Grey marble Marble breccia

Ophicalcite Spotted calcite marble

Figure 9. Examples of the seven marble varieties found in Dammsjön.

(19)

14 Pink calcite: The pink calcite marble often shows a banded structure where the salmon pink calcite has bands of impurities, usually rich in phyllosilicates. Around these bands, the pink color disappears and the marble attains either a green or grey tint instead. Figure 11 is a section from drill core GARPN3695 where the light calcite marble transition to the pink calcite which show the typical banded texture. The optical microscopy of the pink calcite indicates impurities consisting of various minerals. Using the SEM to analyze the minerals with WDS, it was shown that the impurities are apatite, titanite, baryte and quartz seen in Figure 12A. A fibrous mineral has a chemical signature shown in Figure 17B indicating that it is most likely a chlorite mineral.

Figure 11. Gradual transition between light calcite marble to a banded salmon pink calcite marble.

Figure 10. A) Section from drill core GARPN3740 representing calcite breccia. B) The two types of plagioclase zoned and twinning. C) Euhedral pyrite crystal and pyrrhotite.

A

B C

(20)

15 Grey calcite: The hand specimen of the grey calcite marble does not show any sulfide mineralization and microscopy confirmed that only a few very fine-grained crystals were present. Clusters of quartz together with minor tremolite were observed (Fig. 13A). Tremolite is also present as inclusions in large calcite crystals or as larger grains with uneven grain-boundaries and inclusions of calcite shown in Figure 13B.

Light calcite: The light calcite marble (Figure 14) was found in sections of only 1-3 meters, and not as massive coherent units of light marble. Varying amount of tremolite and transition to green or grey marble is the main reason for this. The microscopy show how quartz exist as clusters, often together with tremolite needles in proximity. Pyrite exists as small (5-15 micron) euhedral crystals that are not possible to see with the naked eye.

Green calcite: The green calcite marble samples from GARPN2286, GARPN3695,

GARPN3740 are all adjacent to the marble breccia and is no more than a few meters thick. The green color is pale and not very strong. The microscopy and SEM analysis show existing impurities such as plagioclase crystals with sericite alteration together with amphiboles that also show alteration. Zircon, apatite, titanite, pyrrhotite are accessory minerals found as inclusions in calcite, but also as bigger crystals as shown in Figure 15B.

Figure 14. Calcite dominated light calcite with a quartz cluster and tremolite.

Figure 12. A) Pink marble in cross-polarized light. B) SBE spectrum of the fibrous chlorite.

A B

Figure 13. A) A cluster of quartz and tremolite in calcite dominated grey marble. B) Tremolite alteration with inclusions of calcite.

A B

(21)

16 Spotted calcite marble: The spotted calcite marble is a 3 meters thick silicified variation of the calcite marble. The spots are biotite grains that are evenly distributed throughout the marble. The quartz is fine-grained and evenly distributed as well. The sulfides in Figure 16B are pyrrhotite with its reddish tint and the light euhedral crystal is galena with its typical cleavage.

Calcite marble (ophicalcite): This marble has a special texture called ophicalcite. Patches of serpentine in a matrix of calcite with broken grains of Mg-rich olivine that has been healed with sulfides shown in Figure 17A. Several sulfides were identified as a complex and intergrown crystal, including pyrrhotite, galena and alabandite (MnS) (Fig 17B).

Figure 15. A) Sericite altered plagioclase together with tremolite and minor quartz. B) Big crystals of the accessory minerals in the green marble.

Pyrrhotite

Apatite

Titanite

A B

Figure 16. A) Spotted calcite marble with fine grained quartz and biotite. B) Pyrrhotite and galena with an inclusion.

A B

µm

Galena Pyrrhotite

Alabandite

Figure 17. A) Broken olivine crystal healed with sulfides in a matrix of calcite and serpentine. B) Intergrown sulfide crystal containing pyrrhotite, galena and alabandite.

A B

(22)

17 4.2 Logging

This part of the report summarizes the result from the core logging and how the lithologies presented in chapter 4.1 are spatially related and will focus on the marble varieties found at Dammsjön. The position and sample ID for each sample is shown in the graphical logs below. The complete graphic logs can be found in Appendix I. Figure 5-9 are the digitized logs illustrating the complex alterations and variations in the bedrock. Figure 10 is a geological interpretation of the horizontal level in Dammsjön at level 1075 based on the logged cores.

4.2.1 GARPN 3695

A horizontal drillcore at the 1075 level in Garpenberg seen in Figure 5 intersect the periphery of the Dammsjön ore body. The first 100 meters of this drill core were not logged due to time limitations and the lack of marble in that section. A metamorphosed and hydrothermally altered volcanic rock, quartzite with minor pyrite was the first unit in logged in GARPN3695. A clear but gradational contact between the quartzite and skarn altered dolomite with an inlier of sericite quartzite, followed down hole. Between the dolomite and the next marble variety - calcite breccia - there is a unit of actinolite/garnet skarn. The calcite breccia transition into a volcanic breccia with fragments of marble, where the marble fragments in the breccia have a light-colored alteration rim which is typical for the breccia found at Dammjön (Lowe, 2012). The volcanic breccia is followed by a sericite-altered volcanic rock with zones of phlogopite schist with poor rock quality. The schist have a clear and sharp contact towards tremolite skarn that transition into calcite marble. The calcite marble is a 54-meter-thick unit that is varying in colour between light and grey, and in the amount of skarn minerals. At 283-289 meters, there is a banded salmon pink calcite marble zone which is important for the interpretation of the spatial distribution since it is present in four of the logged drill cores. At the contact between the metavolcanic rock and the calcite marble, there is a thin dolomite unit. The metavolcanic unit vary in

Figure 18. Drill core log of GARPN3695, Digitized from

graphical logs with sample positions included. Figure 19. Drill core log of GARPN3740, Digitized from graphical logs with sample positions included.

(23)

18 character from Qz-phyric sericite quartzite to fine-grained, silicified quartzite. From 320- 360 meters there is a dolomite marble with abundant tremolite and a matrix of talc. This marble type is associated with an inlier of a cordierite-rich volcanic rock. Calcite marble with patches of serpentine (“ophicalcite”) was logged as two minor units where one is in gradual contact against dolomite marble and the other occur as an inlier in the last unit logged in this core. The last unit observed is a metamorphosed hydrothermally altered volcanic rock, sericite quartzite with cordierite porphyroblasts.

4.2.2 GARPN3740

The core GARPN3740 seen in Figure 19 intersect first with a tremolite skarn unit, with an inlier of metamorphosed hydrothermally altered volcanic rock with quartz-phyric texture and pyrite porphyroblasts. The skarn has a gradual contact to the calcite marble breccia which is the main variety in this marble unit. A ten-meter-thick tremolite+garnet skarn is located in the center of the calcite breccia and a volcanic breccia with fragments of calcite marble follows the calcite breccia down-core.

With a clear but gradual contact, the volcanic breccia transition in to a 50-meter-thick sericite+phlogopite quartzite that contain stringer mineralization of sphalerite. At the contact between quartzite and calcite marble in Figure 19 downcore, there is diopside skarn and marble breccia. The marble unit extends from 138 to 179 meters and varies in character between grey, light, banded salmon pink, grey, marble breccia, grey, banded salmon pink and the thickness of each unit varies as well. The calcite marble has a clear but gradual contact with the following dolomite marble that contain dark veinlets of sulfides and sulfosalts and varying tremolite alteration. The last unit logged was a phlogopite quartzite transitioning into a phlogopite schist down core.

Figure 20: Drill core log of GARPN2286, Digitized from graphical logs with sample positions included.

(24)

19 4.2.3 GARPN 2286

GARPN2286 in Figure 8 show a thin calcite marble breccia that is in contact with calcite marble. The calcite marble is shifting in character in a similar matter as the cores 3695 and 3740. Grey marble shift to a salmon pink marble, and back again to grey calcite marble where boudinage of phyllosilicate-rich bands is present. At the contact, significant skarn alteration which is typical for almost all contacts in the area between a marble and rocks of volcanic origin. Here the contact skarn is composed of pyroxene and amphibole. The marble unit is about 30 meters thick and is followed by quartz-rich, hydrothermally altered volcanic rock, with pyrite in the quartzose groundmass associated with fractures similar to stockwork. At 73 meters in Figure 8 there is an 18-meter-thick tremolite+diopside skarn with garnet porphyroblasts that contains impregnation of pyrite and pyrrhotite. Sericite quartzite with minor garnet porphyroblast after metamorphosed hydrothermally altered volcanic follow after the skarn. A thin mafic unit were the contact is slightly discordant to regional foliation is altered but not as extensive as the host sericite quartzite, suggesting a later deposition of the unit. The sericite quartzite extends to 139 meters where a sharp contact between it and calcite marble can be seen. Here, cordierite porphyroblasts are found in the volcaniclastic unit at the contact as well as minor skarn containing clinozoisite. The calcite marble is about 53 meters thick with four inliers of volcanic rocks and the type of marble vary, with thin units of green and grey calcite marble and two zones of spotted calcite marble. The dominating variety is calcite marble breccia and the calcite marble unit end with a banded salmon pink marble that contain minor veinlets of sphalerite and galena. The pink marble has a clear but gradational contact with a dolomite marble with extensive skarn alteration. A mineralized quartzite is the final unit logged in 2286. It contains a stockwork of sphalerite, galena and pyrite with the richest zone between 201-203 meters.

4.2.4 GARPN2252

A tremolite+diopside skarn with minor garnet porphyroblasts is the first unit observed in GARPN2252 (Figure 7). It is followed by a metamorphosed, quartz-rich hydrothermally altered volcanic rock. The quart-rich rock is pyrite-impregnated and interbedded with breccia-conglomerate beds, where the marble fragments have been completely replaced by tremolite skarn. Remnants of the marble fragments can be found in the breccia and skarns. The alteration in the metavolcanic rock shifts from silicification to sericite+phlogopite between 49-65 meters as well as some zones with quartz+feldspar phyric-texture suggesting minor alteration. At 118 meter there is a mineralized skarn with sphalerite and pyrite at the contact between the metavolcanic rock and a dolomite marble. The marble contains skarn minerals that vary between 5-20 vol.%. This is a thick marble unit with inliers of diopside skarn and sericite quartzite as well as some calcite marble. The salmon pink calcite can also be found in this drill core at 174 meters alongside with grey calcite marble. The dolomite marble in contact with the metamorphosed hydrothermally altered volcanic rock, chlorite+phlogopite schist is mineralized with sphalerite that is associated with the skarn alteration in the marble. The schist is also mineralized with sphalerite and chalcopyrite.

(25)

20 4.2.5 GARPN2544

The drill core shown in Figure 9 is dominated by a metamorphosed hydrothermally altered volcanic rock with variations in alteration the first 95 meters. Strong silicification and sericite is common, but some section have a local feldspar-quartz phyric texture, suggesting weaker alteration. The contact with dolomite marble at 95 meters has a two-meter-thick skarn zone with minor chalcopyrite and clinozoizite. The dolomite marble has varying tremolite abundance, 10-40 vol.% and at the lower contact the marble contains around 5 vol.% dark veinlets. The contact is sharp towards the metamorphosed, silica-rich hydrothermally altered volcanic rock, showing hydrothermal brecciation where pyrite and phlogopite are associated to the fractures.

Minor sphalerite and pyrite mineralization are found between 173-196 meters in the silica-rich rock.

The geological interpretation is presented in Figure 23, which is based on the previous described logs. It shows two large calcite marble bodies in the peripheral parts of the Dammsjön ore body, and laterally extensive pink calcite and dolomite closer to mineralization in the adjacent metavolcanic rock.

Figure 23. Geological interpretation of the peripheral parts of level 1075 in Dammsjön.

Dammsjön

(26)

21 4.3 Marble Lightness

The eight lightest samples have a lightness value above 90 on the 0-100 scale relative to the light standard of Björka mineral. All these samples belong to the varieties logged as ‘grey’ or ‘light calcite marble’. The green and pink calcite marble have median values of 85.65 and 83.75 respectively. The median is preferred over mean values in this study due to the limited amount of data since it minimizes the effect of outliers. The color of the calcite marble is not retained after it is grinded down into a fine powder with a particle size of less than 125 microns. The boxplot in Figure 24A shows the distribution of the marble lightness for each group. The table inset in Figure 24B show all the measured values for both lightness (R457) and yellowness.

4.4 Magnetic separation

Only sample LK20190524 contained a magnetic fraction of 0.01g out of a 10.05 g sub-sample. The sample is logged as ophicalcite, the other samples did not have a magnetic fraction that could be measured nor seen during ocular observation.

Sample ID R 457 Yellowness LK20190500 82.5 3.46 LK20190502 83.3 3.5 LK20190503 86.7 2.72 LK20190504 87.7 1.87 LK20190505 84.45 1.96 LK20190506 90.05 1.64 LK20190507 91.2 1.03 LK20190508 91.8 0.9 LK20190509 84.78 3.82 LK20190510 83.75 5.01 LK20190511 92.45 0.91 LK20190512 87.85 1.77 LK20190513 80.7 2.54 LK20190514 72.5 4.14 LK20190516 85.65 2.43 LK20190517 90.85 1.25 LK20190518 83.45 3.9 LK20190519 74.8 1.56 LK20190520 87.45 1.71 LK20190521 90.8 1.24 LK20190522 91.4 1.53 LK20190523 43.2 6.26 LK20190524 42.4 0.12 LK20190526 87.4 0.82 LK20190527 92.4 0.96 LK20190529 76.35 2.75 LK20190530 80.75 2.07

Figure 24. A) Box-plot of the marble lightness for each of these groups. The colored bar indicates the first quartile, the black dot represents the mean value while the line represents the median value. B) Measured lightness and yellowness for each sample.

A

F i g u r e 2 4 . A ) B o x - p l o t o f t h e m a r b l e l i g h t n e

B

(27)

22 4.5 Acid insoluble residue (AIR)

The result from the test show the amount of minerals in weight-percent that is resistant to the acid used in the test. The individual value for each sample is represented in Figure 25 as well as a box-plot showing how the AIR corresponds to each of the marble types. The amount of residue is not a direct indicator of the carbon content of the samples, because other minerals other that carbonates are soluble in 25 % HCl.

4.6 Lithogeochemical data

The lithogeochemical data includes all the major rock forming elements, REE, total carbon, total sulfur as well as most ore forming metals. The complete data sheet can be found in Appendix III. Chemical purity in calcite is when the crystals only contain CaCO3. Such a crystal has a theoretical maximum content of 56.03 vol.% CaO and 12.00 vol.% total-carbon during chemical analysis. In Figure 25 A) and B) the CaO and total-carbon content are plotted against the lightness respectively. Each point represents one sample which are color-coded based on the different marble varieties. The trendlines are flattening towards the theoretical maximum content of CaO and tot C.

Sample ID AIR (wt%) Sample ID AIR (wt%) LK20190500 7.940 LK20190516 6.458 LK20190502 16.950 LK20190517 3.468 LK20190503 4.256 LK20190518 10.307 LK20190504 2.623 LK20190519 70.427 LK20190505 2.467 LK20190520 4.281 LK20190506 1.877 LK20190521 5.085 LK20190507 3.571 LK20190522 5.725 LK20190508 2.444 LK20190523 LK20190509 7.966 LK20190524 53.471 LK20190510 3.403 LK20190526 0.906 LK20190511 3.109 LK20190527 0.959 LK20190512 5.251 LK20190529 24.403 LK20190513 10.899 LK20190530 6.244 LK20190514 33.908

Figure 25. A) Box-plot of the marble varieties regarding the AIR. B) Measured AIR (wt.%) for each sample.

A

F i g u r e 2 5 . A ) B o x - p l o t o f t h e m a r b l e v a r i e t i e s r

B

Figure 26. A) The CaO content in % plotted against the spectrophotometric lightness. B) The Tot C content in % plotted against the spectrophotometric lightness.

A B

(28)

23 In Figure 27A, the samples are plotted together with chemically pure calcite and dolomite represented by the larger grey points in Fig. 27A. Two distinct trends are formed in the plot. In Figure 27B the tot.

Carbon is plotted versus the AIR and a vey clear linear correlation appears.

The elements Titanium (Ti), Aluminium (Al), Zirconium (Zr) and manganese (Mn) are plotted against the spectrophotometric lightness (R457) in Figure 28. The major elements were analyzed as oxides in percent while Zr is expressed in ppm. The Ti, Al, and Zr are immobile elements associated with minerals found in volcaniclastic material while manganese is associated with hydrothermal alteration (Allen et al., 2003). The three immobile elements show a similar trend where the lightness increases with a decreasing amount of these elements.

The origin on the impurities can be further validated, using a geochemical plot shown in Figure 28 developed by Allen et al. (2003) which can discriminate between the volcaniclastic and hydrothermal component in rocks from Bergslagen. Fe+Mn/TiO2 represent the hydrothermal component, and Al2O3

in wt.% represent the volcaniclastic component. Skarn and dolomite plot along the Y-axis indicating a

Figure 28. A) Ti02 B) Al2O3 C) Zr D) MnO –plotted against the spectrophotometric lightness

A B

C D

Figure 27. A) Lithogeochemical plot of the samples compared to chemically pure calcite and dolomite. B) Lithogeochemical plot showing AIR vs Tot.Carbon.

A B

(29)

24 hydrothermal alteration while calcite marble plot along the X-axis confirming what was indicated in Figure 29, that the amount of volocaniclastic material govern the lightness in calcite marble.

4.7 Quality Assurance/ Quality Control

The data of the reference material is shown in table 2. A comparison of the major elements from previous studies of the reference material indicate that the lithogeochemical analysis from ACME have the required accuracy and precision for adequate interpretations of the data, as the samples LK20190525 (darkblue) and LK20190501(orange) are within the margins of error.

Sample no. SiO2_PCT TiO2_PCT Al2O3_PCT Fe2O3_PCT MnO_PCT MgO_PCT CaO_PCT Na2O_PCT K2O_PCT P2O5_PCT tot C_PCT S_PCT LOI_PCT Raw total_PCT LC201720023.14 0.01 0.79 0.51 0.06 20.06 30.35 0.01 0.14 <0.01 12.28 <0.02 44.6 99.66

LC201800523.06 0.01 0.77 0.51 0.06 20.52 30.38 0.01 0.12 0.01 12.6 <0.02 44.2 99.65

LC201800563.08 0.01 0.78 0.51 0.06 20.22 30.46 0.01 0.13 <0.01 12.65 <0.02 44.4 99.66 LC201800023.14 0.01 0.8 0.52 0.06 20.32 30.38 <0.01 0.13 <0.01 12.87 <0.02 44.3 99.65

LC201800313.2 0.01 0.81 0.49 0.06 20.16 30.12 0.01 0.14 <0.01 12.86 <0.02 44.6 99.65

LC201800713.05 0.01 0.8 0.5 0.06 20.4 30 0.01 0.14 <0.01 11.97 <0.02 44.7 99.65

LC-mean 3.112 0.010 0.792 0.507 0.060 20.280 30.282 0.010 0.133 0.010 12.538 44.467 99.653

Std 0.053 0.000 0.013 0.009 0.000 0.153 0.164 0.000 0.007 0.000 0.321 0.180 0.005

LK20190525 3.15 0.01 0.8 0.52 0.06 20.13 30.13 0.01 0.15 <0.01 12.38 <0.02 44.7 99.65

LC2017201613.25 0.03 1.64 2.07 0.26 18.44 29.16 <0.01 <0.01 0.02 9.01 0.38 34.7 99.52 LC2017203213.49 0.03 1.65 2.04 0.26 18.38 29.13 <0.01 <0.01 0.02 9.03 0.38 34.5 99.53

LC2018004313.86 0.04 1.99 2.25 0.25 18.87 28.76 <0.01 <0.01 0.03 8.9 0.4 33.5 99.53

LC2018004512.4 0.04 1.81 2.1 0.26 18.28 30.59 <0.01 <0.01 0.03 9.17 0.4 34.3 99.81

LC2018001714.07 0.04 1.95 2.19 0.25 18.92 28.92 <0.01 <0.01 0.03 8.98 0.4 33.2 99.51

LC2018008612.4 0.04 1.69 2.03 0.26 18.95 29.24 <0.01 0.01 0.03 9.38 0.41 34.9 99.52

LC-mean 13.245 0.0367 1.788 2.113 0.257 18.640 29.300 0.010 0.027 9.078 0.395 34.183 99.570

Std 0.652 0.005 0.140 0.081 0.005 0.278 0.599 0.000 0.005 0.157 0.011 0.623 0.108

LK20190501 12.66 0.03 1.71 2.07 0.26 18.57 29.59 <0.01 <0.01 0.03 9.12 0.35 34.6 99.53

Table 2. Lithogeochemical data from two materials, Blue series coupled with the analysis for this project (dark blue). Green series coupled analysis shown in orange.

Figure 29. Plot that discriminate between input of hydrothermal and primary volcanoclastic component.

(30)

25 4.8 Comparison against quality specifications for products

During a year, 68 000 tons of limestone (calcite marble) powder is released into Swedish lakes to prevent acidification form acid rain fall. All products used for this need to pass a rigorous quality control by the research institute of Sweden (RISE) (Naturvårdsverket, 2010). Figure 30 show a list of elements with quality specifications that the product cannot exceed if the lime is to be used for liming lakes. Twelve of the twenty-four marble samples did exceed the limits for at least one of the elements in table 3, many of them are dolomite.

In table three, the samples marked in red have at least one element that exceeds the quality specifications for using it as a pH-regulator in lakes. The best samples on the other hand are well below the quality requirements.

Table 4. Samples compared with the quality specifications for limestone used for liming lakes in Sweden, Samples marked in red exceeded the limit for at least one element.

Figure 30. Quality specifications for limestone products used for liming lakes in Sweden (Naturvårdsverket,2010)

(31)

26

5. Discussion and Analysis

Marbles at Dammsjön were divided into seven varieties based mostly on color, but also based on the amount and type of impurities. Compared to the description in Allen et al. (2003) - who divided the marble into white, pink and orange calcite marble - the orange and pink marble were instead interpreted as one unit termed the banded salmon pink calcite marble. This unit is elevated in Fe (Figure 28) and its color could be a result of Nano inclusions of Fe³⁺ -oxides. Hematite inclusions in white marble could be the reason giving it a pink color, while marble with the same amount of Fe wt.

%, but in another redox state does not attain any color. This mean that marble with elevated Fe content can have color but there is no simple relation in the marble at Garpenberg between pink color and Fe content. The marble breccia was included into the marble units instead of the lower hanging wall of polymict breccias to allow it as well to be evaluated for purity and lightness, and thus its potential for use as an industrial mineral. There is no clear evidence on what causes the color change in the calcite crystals, but there are some indications that the green color might be a result of a reflection of other green minerals like tremolite+serpentine in the otherwise clear carbonate crystals. In the pink marble, the color might be a product of inclusions of minerals such as titanite, apatite and baryte. The colors themselves do not seem to be a significant contributor to the varying lightness, but is indicator of impurities which are detrimental for the marble lightness.

As shown in Figure 27B, there is a strong correlation between the total carbon content and AIR, especially for the calcite marble samples. This indicate that the AIR is a test that could potentially be completely replaced by Leco total carbon analysis, thus cutting down on the analysis cost when carbon analysis is part of a standard analysis package from lithogeochemical laboratories.

As mentioned in the results, only the sample LK20190522 had enough magnetic minerals in the form of pyrrhotite to be detected. The magnetic separation was important to investigate due to the significant impact oxides and sulfides have on the lightness. Thus, if there would have been a substantial magnetic fraction in the samples, these could easily be removed during processing to yield a higher purity.

To determine the hydrothermal versus the volcaniclastic component in the marble, the lithogeochemical analysis was important. During logging it was extremely hard to distinguish and determine the amount of these components, something which Allen et al. (2003) also described as very difficult. They also pointed out that aluminum is a good indicator for determine the volcaniclastic primary component, which along with the immobile element zirconium and titanium can work as a proxy for this component in the geochemical data. The SEM data show that aluminum is mostly found in phyllosilicates while zirconium occur in zircon and titanium in titanite.

This study suggest that chemical purity is the single most important factor controlling marble lightness.

This can be seen in Figure 26A and B, where the total carbon and CaO are reveal a positive trend against the lightness. But as the total carbon and CaO increase, the curve starts to flatten and other factors such as what type of impurity start to control the lightness, such as impurity in primary composition and different types and intensity of hydrothermal alteration. The described alteration types are key controls for marble lightness in calcite- and dolomite marble respectively. The primary composition is related to the conditions where the marble precursor was deposited. Calm conditions over a long period of time would be ideal for the formation of extensive pure and light carbonate rock (Allen et al., 2003). Aluminum, Zr, Ti are three immobile elements that are exclusively found in the volcanic protolit. Low content of these elements can thus be used as a proxy to identify light marble with low amounts of volcaniclastic material.

Factors governing marble lightness in peripheral alteration haloes around carbonate-hosted Zn-Pb-Ag-(Cu-Au) deposits, Garpenberg, Sweden