1

Strategic University Programme at Department of Geology and Mineral Resources Engineering. ”THE VALUE CHAIN FROM MINERAL DEPOSIT TO BENEFICIATED PRODUCT WITH EMPHASIS ON QUARTZ” Background for the programme The production and value of industrial minerals in Norway has increased considerably in the last years. The Norwegian industrial mineral production is both characterised by the variety of different minerals and mineral products, and by the fact that Norway is the major producer in the world of some minerals such as olivine, nepheline and beneficiated carbonates (GCC) used in paper. In addition Norway produces quartz/quartzite, graphite, talc, anorthosite, feldspar, mica, dolomite and ilmenite. The production value of mineral resources in Norway , except for oil/gas, was about 7,5 thousand millions NOK in 2003. Of this, 2,6 thousand millions NOK came from the industrial minerals, 2,5 thousand millions from crushed rocks/aggregates, 1 thousand millions from natural stone and only 0,5 thousand millions NOK from metallic ores. The rest is from an increased coal production at Svalbard. The growth in industrial mineral production exceeds the decline in metallic ores, and it is recognised that there is a considerably future growth potential in industrial mineral production in Norway. The value of industrial minerals is in high degree created during the beneficiation and the production of special products, which need both competence and high technology. For instance, crushed carbonate may be priced at 35 – 40 NOK/t, while beneficiated carbonate filler used in paper may reach prices up to 1000 NOK/t. There is a gradual transition from mineral beneficiation to material technology. In accordance with recommendations given by the “National Working Group” a close co-operation with staff at the Department of Materials Technology is strongly emphasised in this project. The research activity in industrial minerals has been fairly low in Norway. Most of the research is carried out by the industry itself, and the results are generally kept in secret. However, to produce valuable and special products reaching high prices in the marked, there is a considerable need for skill and competence. Being the only university in Norway responsible for research and teaching in mineral production, the research programme is of major importance including five doctoral students and one post.doc. The programme comprises three areas of research:

• Genesis of industrially applicable high purity quartz in igneous and metamorphic environments

• Beneficiation of quartz • Advanced characterization of industrial minerals and beneficiated products.

2

It is collaboration with other universities as well as with industrial partners which also give financial support to the project. Following departments and institutions collaborate in the programme:

The department of Geology and Mineral Resources Engineering, (IGBt), NTNU The Department of Materials Technology, NTNU SINTEF Civil and Environmental Engineering, Department of Rock and Mineral Engineering The Geological Survey of Norway (NGU) The Department of Geology, The University of Oslo (UiO)

Industry partners: TITANIA A/S Hustadmarmor A/S Norwegian Crystallite North Cape Minerals Elkem A/S

Time plane: 5 years, from 01.01.2001 to 31.12.2005 Financial support: NFR 11,8 mill. NOK. In addition financial support from the industry and field contribution from researchers at NGU. Staff: 5 dr. candidates. 1 postdoc. Researchers at NTNU and NGU. Description of the research areas Research area: Genesis of industrially applicable high purity quartz in igneous and metamorphic environments Supervisors: Rune B. Larsen, Objective Our primary goal is to study the element exchange processes in quartz from igneous and sedimentary environments that are exposed to multiple episodes of high-grade metamorphic re-crystallisation. Theoretical considerations and laboratory experiments imply that this type of quartz may obtain qualities commensurable with industrially applicable high purity quartz (HPQ). South Norway comprises an excellent natural laboratory for this character of studies, because quartzites and acid meta-igneous rocks covering large areas experienced different degrees of granulite facies metamorphism.

3

Background High purity quartz, i.e. quartz with extremely low concentrations of impurities, is a rare commodity that only forms under geological conditions where a narrow set of chemical and physical parameters is fulfilled. When identified, HPQ obtain very attractive prices and is applied in the communications industries and other high-technology sectors that currently are under rapid expansion. Examples of end products where HPQ is the main raw material includes photovoltaic solar cells for environmentally sound energy production, silicon metal-oxide wafers in the production of ever-faster computer chips and long distance optical fibres that are extensively used in communication networks. Marked considerations Industrial agencies forecast a solid 5-20 % annual growth in the demand for high purity granular quartz and predict a near exhaustion of raw materials. Together with environmental problems in the main quartz producing districts, a coming shortage in the supply of HPQ is implied. By far the largest proportions of HPQ (maybe as much as 90 %) come from granite pegmatites in the Spruce Pine district, North Carolina. The quality of quartz from Spruce Pine is steadily decreasing and, having an arid climate with limited water supplies, the processing of quartz is difficult and costly. Adding to these challenges are environmental concerns because the endangered species the 'Appalachian Elktoe Mussel', habit the few fresh water resources that are exploited by the mining industry. Parallel with foreseeable production shortages of HPQ, the semi-conductor industry plans at least 15 more years of development of more powerful silicon chips that depends on the productions of thicker thus more HPQ demanding silicon metal-oxide wafers. Also imposing higher demands is the fact that the production of HPQ-demanding photo voltaic devices, i.e. solar cells is forecasted to expand rapidly in the future. Geology of quartz The high-purity silica glass sector require that HPQ contain very low concentrations of structural impurities, i.e. foreign substitutional elements or charge compensator elements that are integrated as a part of the atomic lattice structure of quartz. Being bonded by the lattice structure of quartz, structural impurities are nearly impossible to remove with conventional dressing technologies. Particularly for the lighting and optical fibre industries, HPQ is required to contain very low concentrations of fluid inclusions because, expansion during melting of the fluid inclusions will generate vesicles in the silica glass melt that may be incorporated in the optical fibres. Therefore, good qualities of HPQ must crystallise under anhydrous conditions and must incorporate a minimum of structural impurities. The most important structural impurities that are easily accommodated by the quartz crystal structure includes Al, Ti, Fe, Ge, Li, Na, K, B, P, Ca and H. With lower abundance but still well accommodated we have Cr, Cu, Mg, Mn, Pb, Rb and U. For some industrial applications, low Fe and B contents are imperative (e.g. in photovoltaic cells for solar panels) whereas some Ti may be tolerated. Other applications primarily require low Ti-concentrations and yet other applications are mostly concerned with low concentrations of Li. As with many other minerals, the concentration of structural impurities rise with temperature. Quartz from diorites and monzonites, for example, comprise much higher concentrations of structural impurities than quartz from evolved granites and granite pegmatites. Recent studies of the trace element distribution in granite pegmatites in Evje-Iveland, South Norway, demonstrate that the speciation and concentration of structural impurities also depends on the degree of differentiation of

4

the igneous melts (Larsen & Polve 1998, Larsen et al. 1998a, Larsen 1999, Larsen & Lahaye, 1999, Larsen 2000, Larsen et al. 2000). Incorporation of fluid inclusions in quartz is primarily a function of the amount of volatiles in the quartz-forming environment. Diorites and monzonites largely form under volatile undersaturated conditions whereas granite pegmatites at least partially form during volatile oversaturation. Also, the speciation of volatiles may be important for the manufacture of silica glass. H2O, for example, has a much higher solubility in the silica glass melt (several percents) than CO2, CH4 and N2 (a few hundred ppm). Dissolution of aqueous fluid inclusions during silica glass melt production may therefore hinder vesicle formation, however, being an excellent solvent, may contribute with Na, Fe, Mg, Li and several other electrolytes that are dissolved in the aqueous phase. Scope of work From the above considerations it appears that igneous quartz that formed at low temperatures may provide excellent HPQ raw materials in having low concentrations of structural impurities, but may be void because oversaturation of volatile fluids cater for high fluid inclusion abundances. Igneous quartz is not attractive because it formed at high temperatures that strongly enhances the incorporation of structural impurities HPQ with good melting behaviour, i.e. low fluid inclusion contents, is therefore very difficult to form in igneous and hydrothermal environments thus the so-called long-distance optical fibre industries where vesicle free silica glass is imperative, largely have to rely on extremely expensive man-made silicon compound glass. Granulite facies terrains Probably the only geological environment that on a large scale may produce quartz with low abundances of fluid inclusions and low concentrations of structural impurities is high-grade metamorphic terrain's and in particular, granulite facies terrain's. Factors during granulite facies metamorphism that influences the purity and quality of quartz includes: • Repetitive and massive re-crystallisation of quartz • Low density of lattice defects because of slow re-crystallisation of quartz • Homogenisation of the impurity distribution in the quartz-bearing host lithology • Continuous decrepitation of fluid inclusions • Hydrous leaching of quartz during low- to high-grade metamorphism • Regional depletion of LIL-elements during peak granulite facies metamorphism • Anhydrous conditions during peak metamorphism • Volatile fluids during peak metamorphism mostly comprises CO2, N2 and CH4 It is beyond the scope of the present proposal to discuss all these points in detail. Important for the formation of HPQ is the fact that quartz will re-crystallise repetitively during prograde and retrograde metamorphism. During prograde metamorphism, pre-metamorphic fluid inclusions will efficiently decrepitate but because metamorphic volatiles are common from diagenetic to amphibolite facies conditions, new fluid inclusions will be generated and incorporated in quartz. However, during granulite facies re-crystallisation metamorphogenic fluid inclusions will also decrepitate and because peak granulite facies metamorphism occur under anhydrous conditions, new fluid inclusions with aqueous solutions will not form. There may be CO2, CH4 and N2 fluids (or other C-O-H-N compounds) present in the system (e.g. Touret & Dietvorst, 1983; Andersen et al., 1993; Larsen et al., 1998b) however, they are rather harmless when compared to aqueous inclusions because they are easy to extract during industrial dressing processes.

5

The behaviour of structural impurities in quartz is not quite as predictable as the faith of fluid inclusions. During re-crystallisation at low temperatures, quartz with high concentrations of impurities may indeed be significantly more pure as the impurities will partition in favour of other minerals or aqueous fluids. Experiments synthesising sequential re-crystallisation of autoclave quartz at 345-380oC demonstrated 5-10 times reduction in the concentration of Al, Li, Na, K and Fe during four episodes of re-crystallisation (Armington and Balsacio, 1984 in Jung, 1992 p.194). At higher temperatures some impurities may again be incorporated, however, this depends on the availability of impurities and the distribution coefficients between fluids, quartz and other phases. Impurities that partition into fluids may no longer be available. For example, peak granulite facies metamorphism is commonly associated with pronounced LIL-element depletion enforced by a combination of metamorphic dehydration, hydrous metasomatism and partial melting. This process was documented on Tromøy in the Bamble Belt (e.g. Cooper and Field, 1977, Smalley et al, 1983) where the K and Rb concentrations of acid and intermediate gneisses are amongst the lowest ever reported for granulite facies terrain's (Touret, 1987). Therefore, leaching of LIL-elements may improve the conditions for crystallisation of HPQ although the formation of quartz from partial melts, which also form during granulite facies conditions, may enforce the formation of quartz with high impurity concentrations. To avoid quartz that formed from partial melts but still to benefit from the positive effects of granulite facies metamorphism, quartz from quartzites may be the most promising target. Quartzites, being near mono-mineralic lithologies, will not melt at granulite facies conditions because the melting point even under water saturated conditions, will be higher than granulite facies T and P. The low concentration of other minerals in quartzites also reduces the possibility of incorporation of structural impurities during repetitive recrystallisation of the quartz. Being a sedimentary lithology that probably contains compositionally contrasting quartz from multiple sources, repetitive re-crystallisation also has the positive effect of homogenising the quartz compositions throughout the quartz-bearing lithology. Finally, foreign minerals in granulite facies quartzites, being relative coarse-grained compared to lower-grade quartzites, are more easy to handle by conventional dressing techniques. Research strategy Fieldwork Detailed field studies and the main body of sampling are committed in the high grade metamorphic belt of the Bamble shear zone (SE-Norway) and the Rogaland metamorphic envelope (SW-Norway) because the general geologies of these areas are well documented throughout earlier studies. The Bamble shear zone may be divided into four metamorphic zones that from NW to SE (i.e. from A to D) comprise progressively higher metamorphic grades. Zones 'A' and 'B' reaches upper amphibolite facies whereas zones 'C' and 'D' are well within the granulite facies regime (e.g. Touret 1987). Particularly zone 'D' is void of hydrous minerals and is characterised by strong LIL element depletion. Quartzite lithologies and orthogneises are present in all metamorphic zones, but from the considerations outlined in the previous section, quartzites in zones 'C' and 'D' are most interesting. The metamorphic envelope embracing the Rogaland Intrusive Complex comprises a rich diversity of magmatic and sedimentary successions that experienced granulite facies metamorphism. Quartzites are particularly common in the Faurefjell meta-sedimentary successions that intersect different intensities of granulite facies metamorphism throughout the metamorphic envelope.

6

Analytical techniques in addition to methods outlined under other sub-projects LA-HR-ICP-MS: the Geological Survey of Norway (NGU) recently purchased a Laser Ablation High Resolution Inductively Coupled Mass Spectrometer (LA-HR-ICP-MS). In short, the advantage with this instrument is its ability to analyse virtually any isotope in the periodic table by in situ ablation of small volumes of material directly from the sample surface. Simultaneous analysis of the ablated material by a high-resolution mass spectrometer ensures detection limits down to the sub ppm level. Finally, the laser pit may have a diameter of only 20 µm, which make it possible to obtain an exceptionally high spatial resolution. NGU has developed an analytical procedure that utilises this instrument in quantifying the trace element concentration in quartz and is considerably more rapid than conventional methods for quartz analysis. Hallimond tube micro-flotation: in order to evaluate the results from LA-HR-ICP-MS analysis it is necessary to conduct control analysis by more conventional methods. These include micro-flotation of small sample quantities by the Halimond tube technique that has proved very successful in the separation of quartz from feldspar and micas. Hallimond tube micro-flotation is mastered by NTNU (Prof. Knut Sandvik) and will be followed by conventional solution HR-ICP-MS at NGU. Fluid inclusion analysis: analysis of fluid inclusions is an essential part of the present study, because they provide important information about the P-T-X conditions that prevailed during the genesis of HPQ. The composition of the fluid inclusions will be determined with a state of the art Linkam freezing-heating stages at the Department of Geology and Mineral Resources engineering (NTNU) and will be supplemented with non-destructive raman micro-probe analysis at Free University, Amsterdam. The later method is imperative in identifying solid and fluid species (particularly C-O-H-N compounds) in the fluid inclusions. EPMA: Electron Probe Micro Analysis will be applied to selected phases co-existing with quartz in order to obtain independent P-T estimates and to calculate the principal distribution co-efficients for trace-impurities in quartz. Autoclave experiments: exchange of elements between quartz and the surrounding environment under different P-T-X conditions will be approached by autoclave experiments at University of Tromsø and University of Copenhagen where the proper instrumentation is available. Collaboration partners

• Dr. Nikos Arvanitides is an expert in both the formation and industrial applications of HPQ. NA is director at the Institute of Geology and Mineral Exploration (IGME) in Greece.

• Dr. Jens Konnerup Madsen at the Department of Geology (University of Copenhagen) is an expert in thermodynamic modelling of volatile fluids.

• Expert in metamorphic petrology and metamorphic mineral-melt-volatile reactions (individual not yet decided).

• Doctoral students

• Doctoral student in metamorphosis and mineral-fluid-melt interaction processes in quartzites • Doctoral student in mineral-chemistry and element exchange reactions of quartz based on

laboratory experiments (autoclave-experiments)

7

Research area: Quartz beneficiation Supervisors: Professor Knut L Sandvik Dept. of Geology and Mineral Resources Engineering. Professor Otto Lohne Dept. of Materials Technology and Electrochemistry. Objective Norway has had, and still has a strong mining sector, which over the last years have changed from metal to industrial minerals as products. The products are generally exported. An even stronger electro-metallurgical industry makes export products mainly from imported raw materials. However, very little refining of Norwegian raw materials takes place. In this project we aim to: • Establish a process chain producing sophisticated products from Norwegian raw materials. • Concentrate on silicon raw materials, because we assume there will be an especially strong

development in the silicone solar cell sector. Results from this project may therefore rapidly lead to practical applications in the industry.

• Develop further Norway’s unique situation to hold a key position in solar cell production. We have a raw material base and key industrial producers already working in related segments of the industry all the way from refined quartz to silicone wafers.

Background Quartz is found in the nature in varying purity and is traded in varying quality at strongly differentiated prices. Raw material for glass is probably the largest market, which is dominated by Belgian quartz sand. Belgian quartz sand sets the price and quality standard for such products. The same material is also used for chemical feedstock and fetches prises around 150 NOK/t. Lump quartz of reasonable purity and high thermal strength for the smelter industry may be priced at 300 NOK/t while the prices of quartz for optical purposes are not disclosed. The material chain from quartz can be drawn to metallurgical grade silicon metal from 2000 NOK/t, which also is refined by Elkem to Silgrain, trichlorosilane, silicon tetrachloride at 4900 NOK/t and semiconductor quality silicon at 600 000 NOK/t. Special products of silicon derived from quartz are often made via the expensive routes of trichlorosiane, silicon tetrachloride or sodium silicate. Volume markets are rheology control and filler in addition to the well paying semiconductor sector. The demands of purity in the last sector are in the order of 1 ppb. The chemical refinement in several steps ensures that the demands on the raw material paradoxically are not extreme, but it makes the final product expensive. A more promising and expanding market for quartz is solar cells. Here are the demands of purity much lower than for semiconductors, in the order of 1 ppm, but not yet well defined. Up till now the supply of raw material for solar cells has been based upon scrap from the semiconductor production. The price of this scrap is in the order of 150-200 NOK/kg. The demand for solar cells is outgrowing that of semiconductors and the price of silicone wafers has to be reduced in order to give a competitive solar cell energy price. This means that the cost of silicone

8

also has to be cut substantially and sufficient long term supplies of this material has to be developed. In the near future lack of sufficient amounts of cheap silicone for an expanding market may threaten the position of silicone as solar cell material no 1. A cheaper production may be achieved by using purer quartz and thereby omitting some of the expensive chemical refining steps. Norwegian Crystallite at Drag in Nordland is already producing a few thousand tonnes of refined quarts from a local deposit. (combined amount of impurities is in the range of 1 ppm). The nucleus for a industry of considerable size do therefore exist provided large reserves of quartz and reasonable methods for beneficiation are found. Project In this project we will work with Norwegian sources of quartz. A close co-operation with the ongoing NFR project “From sand to solar cells” have to be established as alternative process paths based upon purer raw materials have to be developed as an extension of existing Norwegian technology. This is the reason why professor Otto Lohne from the Department of Materials Technology and Electrochemistry is chosen as one of the supervisors for this part of the project. One possibility, which can be based upon existing processes, is to upgrade the Elkem Silgrain production, for such a purpose. In this process carbon reduced silica is purified to a certain degree. Use of improved quality raw materials may bring the product from this process to a stage where further removal of impurities may take place in a process combined with wafer production. Our project on pure quartz production should be aimed at making the kind of sand, which would give a cost efficient chain of value development from the raw material in the ground to the finished product. Quartz, one of the most common minerals, is found all over Norway, some as deposits of relatively pure quartz, but mostly associated with other rock forming minerals. Generally quartz has been mined from the mono mineral deposits, although some quartz has been a secondary product when making feldspar from pegmatite and granites. Existing Norwegian processing technology for quartz sand beneficiation is mainly based upon the Lillesand operation of North Cape Minerals, where first iron bearing minerals are removed by flotation, then feldspar is concentrated by another step of flotation. Finally the remaining iron bearing minerals are removed from feldspar is by strong field magnetic separation. The process was originally based upon knowledge from the Spruce Pine area in North Carolina. The major product from such operations is feldspar, however, but at Spruce Pine the quartz is pure enough to warrant upgrading to excellent qualities. When the process for the Drag quartz deposit, now operated by Norwegian Crystallites, was developed, our knowledge about this technology was further refined for production of pure quartz products from a high grade deposit. Regarding other quartz mineral deposits the technology has to be developed to remove other impurities. Critical elements for the use of quartz in solar cells are first of all boron and phosphorus as they are difficult to remove by metallurgical refining. Maximum values are below 1 ppm. for each. Calcium, aluminium and metal oxides are also unwanted. To obtain a physical removal of the last traces of unwanted elements, the elements have to be present in separate minerals of sufficient size to

9

be liberated. The characterisation and analysis of the unwanted elements and how they are distributed in the minerals will be important. Our department has access to new advanced analysis instruments, which makes such characterisation possible. Alternatives to traditional purification to purer sand do exist. Mechanical activaton, that is disturbance of the crystal lattice of the mineral by crushing may give a more reactive quartz, which may be cleaned by weak acids such as CO2. Other alternatives to increase the reactivity of quartz at relatively low temperatures and pressures may also exist. The traditional smelting/reducing technologies for silicon, which is a path for silicon production, depends today on lumpy quartz, which is heat resistant without cracking. Little is known about the mechanism, which gives competent lumps, or how to determine without full-scale tests, which deposits are suited for this purpose. This itself is an interesting point for investigation. Furthermore the most likely product from a beneficiation process will be in the form of quartz sand. Fine grained feed will blind a smelter furnace and has therefore to be agglomerated to give the required lumpy material provided the Silgrain process should be chosen. Development of competent and cheap agglomerates including binders not adding unwanted impurities will also be an important task if this path is to be followed. The fact is that surprisingly little research has been done to assess the different quartz types of Norway or quartz generally. The way to a process for solar cell raw material therefore possesses many challenges of a basic nature. Collaboration with expertise from the geological side to the metallurgical side is required. Because the project “From sand to solar cells” already is established, formation of such contacts is already established for this project. Because mineral processing is a wide subject, some tasks that are of importance to other parts of the value chain may also be included.. Industrial contacts We have discussed this application with Norwegian Crystallites, which at present is the only producer of highly refined quartz in Norway. The company is very interested in co-operating with us in the project. Norwegian Crystallites is backed by Hustadkalk A/S, which should have the capacity to commercialise viable research results. Other interested parties may be Elkem, which has quartz quarries and is a major user of quartz for ferrosilicone and silicone metal production and North Cape Minerals, which has the plant in Lillesand. Elkem is well connected to the “Sand to solar cells” programme. Doctoral students Under the programme Erik Larsen is engaged as a doctoral student and Rolf Arne Kleiv as a post doctoral student for the last part of the project.

10

Research area: Mineral Characterisation “Development of quantitative methods to describe raw materials and mineral products”. Supervisors Professor Terje Malvik, Department of Geological and Mineral Resource Engineering Professor Jarle Hjelen, Departement of Materials Technology Dr.candidate Kari Moen Objective The project aims at developing automatic SEM - based quantitative techniques in order to:

• quantify the mineral content, both major and trace minerals in rocks/ores and in milled products,

• quantify the mineral texture • quantify the occurrence of minerals in different types of particles in milled materials and

mineral products (PTA: Particle Texture Analysis) Background All mineral processes are treating particles, and not chemical elements. The particles are containing one or more minerals. To achieve high quality products meeting the different requirements from the customers, and with a minimum waste, a thorough and detailed knowledge concerning the occurrence of the minerals in the raw materials and in the mill products is needed. In investigations of ore minerals, particularly the base metal sulphides, there has for a long time been used automatic SEM-based image analysers to quantify the occurrence of the minerals. Special emphasis has been laid on determining the liberation properties of the economic minerals to avoid under- or over grinding of the minerals. It is recognized that the behaviour of particles in separation processes is a function of different properties of the particles of which the most important are: Mineralogical parameters; types of minerals making up the particles, mineral chemistry,..; Textural properties; size, shape, grain boundaries, type of intergrowths (particle texture)…; Specific properties of the minerals including specific gravity, surface property, magnetic property…; In Norway, there has been a strong and positive development within industrial minerals. Due to this the project aims at developing methods which can be used on industrial minerals. In additions the methods also can be used on all kinds of other materials.

Principals of PTA - Particle Texture Analysis Purpose Describe mineral liberation, mineral associations and how the minerals are grown together by means of backscatter electron images and X-ray data in the scanning electron microscope (SEM).

11

Performance • Polished sections of narrow grain size fractions are analysed by Oxford Inca Feature • The backscatter electron images are thresholded by grey levels and analysed by means of X-

rays aiming to obtain analyses of every single mineral grain. • The images and the classified mineral analyses are stored. • Inputs for the PTA software are the Inca database and the uncompressed images from each

field. • Composite particles are found by means of image analysis. • Results from the image analysis is stored in a database • Queries are then performed to get desired information • The results can be presented as distributions of

o relative amount of minerals in particles o mineral associations in particles o free grinding distributions o statistics

Figure 1 Examples of classified minerals, identified grains, free grinding decisions and mineral associations

12

Principals of EBSD - Electron BackScatter Diffraction lattice and grain texture

Electron backscatter diffraction The EBSD technique is based on the weak diffraction pattern that forms when a focused, stationary, primary electron beam strikes a polished sample, backscatters and diffracts. The diffraction pattern is formed on a fluorescent screen and transferred by a camera to the computer (Hjelen 1990). Rapid developments in both hardware and software in the past 10 years have made EBSD easy to use and ideal for the rapid analysis of microstructures of crystalline materials. The diffraction pattern is therefore characteristic for the crystal structure and space orientation of the crystal.

Figure 2 Origin of Kikuchi lines from EBSD (Schwartz et.al. 2000)

Figure 3 EBSD patterns from 2 orientations of calcite

13

The EBSD software automatically locates the positions of individual diffraction bands, compares these to theoretical data about the relevant phase and rapidly calculates the 3-D crystallographic orientation. The whole process from start to finish can take from 0.025 sec. to more than one second dependent on the phases analysed. For each pixel the phase and orientation are stored and can be visualised by maps or plots. Different kinds of plots can show misorientation, grain size, poles etc. The spatial resolution of this technique is superior to x-rays, since elastic backscattered and diffracted electrons have a smaller interaction volume than x-rays. The interaction volume is dependent on acceleration voltage and atomic number of the analysed minerals, but is usually in the sub micron range. (Moen et. al. 2003) (Moen et. al. 2004)

Grain orientation distribution (14)

Orientation

0 20 40 60 80 100 120 140 160 180 200

Share

0

2

4

6

8

1 0

1 2

1 4

V0V1V2

14

Figure 4 EBSD orientation map, (0001) and (11-20) pole figures and grain shape orientationdistribution for 3 orthogonal directions of a marble block (Moen et. al. 2004).

Cooperation with the industry The importance of being able to quantify the mineralogical parameters is fully recognized by the industry. Several industry partners (TITANIA, Hustadmarmor, Norwegian Crystallite) support the project in order to benefit from the developments being made. In the future, an on line measurement system using the web, and where the industry partner can take an active part in the measurements, is aimed. Combined with flexible solutions and measurement settings this will lead to techniques being able to describe most mineralogical and textural phenomena occurring in rocks/ores and in mineral products. By use of new and advanced laboratory equipment determine physical and chemical properties of industrial minerals and beneficiated products of industrial mineral with special emphasis on thermal, mechanical and surface properties. The results may be used to define areas for new applications of minerals, establish correlations between material properties and geological processes leading to the different mineral raw materials, and modelling of micro processes occurring during heating and mechanical impact of minerals. Electron Microscopy Laboratory facilities The Scanning Electron Microscopy laboratory is the largest SEM-lab. in Scandinavia, with a suite of 7 conventional, low-vacuum and field emission SEMs. In addition a Transmission Electron Microscope (TEM) and an Electron Probe Micro-Analyser (EPMA) belong to the EM-lab. (http://www.material.ntnu.no/lab/material/index.html) These analytical electron microscopes are designed for material characterization using a range of imaging and analytical methods. (Pdf). The instruments are equipped with Energy Dispersive Spectrometers (EDS), Wavelength Dispersive Spectrometers (WDS), Electron BackScatter Diffraction (EBSD) detectors, CL etc.Automatic EBSD analysis is combined with deformation- and heat treatment experiments in a special in-situ substage to study dynamic processes.

15

Project: “Quartz raw-material for metallurgical production of FeSi and Si-metal”

Image 1: Furnace no. 1 at Elkem Thamshavn (Photo: Elkem ASA, Silicon Division)

Project overview

Doctoral candidate: Siv. ing. Kurt Aasly (Department of Geology and Mineral Resources Engineering, NTNU) Supervisor Professor Terje Malvik (NTNU) Co-supervisor is Dr. ing. Edin Myrhaug (Elkem Silicon Division ASA) The Project period is scheduled to June 1st, 2003 to May 31st, 2007 and financing is through a University scholarship at NTNU with Elkem ASA, Silicon Division as a partial sponsor for the project, through project expenses. The project is integrated in the SUP financed by the Norwegian Research Council (NFR) “The value chain from deposit to beneficiated product with emphasis on quartz”.

Project – outline

The main focus is on the development of knowledge concerning quartz used as a raw material for the metallurgical production of ferrosilicon and silicon metal. The main objective for this PhD project is: - by using systematic investigations and modern analytical equipment

- to obtain a basic knowledge about the properties of quartz raw material that are important for the production of silicon

- and further, develop methods to test these properties

The work will mainly be carried out on typical quartz used for FeSi and Si-production.

16

Quartz properties

The raw materials for Si for the metallurgical production of ferrosilicon and silicon metal, are Quartz and Quartzites. These are in operational terms separated into two types: Rock quartz and Gravel quartz, which are blasted rock and non-blasted, sedimentary deposited materials respectively. The high purity of the products and the nature of the production process necessitate the need of some requirements to the specifications for the raw materials. The most important property requirements for quartz/quartzite, are: - purity (pollution and inclusion)

- mechanical strength

- thermal strength

- softening properties

Because of Elkems involvement in the project, the focus will mainly be related to problems and challenges described by Elkem: - What causes too much fines to be generated from the raw-material?

- Which properties make the quartz decrepitate when shock heated in the furnace?

- Which properties are important for the melting- (softening) properties of the quartz resulting in a certain melting progress?

Earlier work

Little published research has been carried out on the properties of quartz related to the metallurgical production of ferrosilicon and silicon metal prior to this project. However, it appears that company-internal research has been more frequent, unfortunately results from such work is difficult, if not impossible, to reach. Operational experience and none scientific tests seem to have been the working method in most (Fe)Si plants.

Analytical techniques in the project

Several analytical techniques will be evaluated in this project. Examples of techniques which are more or less known and understood, for application on quartz for (Fe)Si production is listed below: Optical microscopy investigations

• Polarization microscopy

– Mineralogy

– Texture (e.g. grain size)

• Fluorescence microscopy

– Porosity

– Grain boundaries

• Linkam TS1500 High Temperature System

– Microscopic technique

– In situ heating effects (up to 1500 ºC)

– Causes of these effects

Scanning Electron Microscope (SEM) techniques:

17

• Cathodoluminescence (CL)

– classify different generations

– internal zoning in minerals

– distribution of trace minerals

– Micro cracks

• Electron Backscatter Diffraction (EBSD)

– Crystal structure

– Orientation map

• Energy-dispersive x-ray Spectrometry (EDS)

– Semi quantitative in situ element analysis

Microprobe – quantitative in situ element analysis

XRD – Mineralogy (crystal structure)

DTA – Mineralogy

– Phase transitions

Dilatometry and Segercone – Softening properties (melting properties)

And other techniques relevant for testing mechanical and thermal properties of quartz.