TY - JOUR
AU - Andersen, J C Ø
AB - Abstract The Skaergaard PGE–Au mineralization, aka the Platinova Reef, is a syn-magmatic Platinum Group Element (PGE) and gold (Au) mineralization that formed after crystallization of ∼74% of the bulk melt of the intrusion. It is hosted in a more than 600 m deep and bowl-shaped succession of gabbroic macro-rhythmic layers in the upper 100 m of the Middle Zone. The precious metal mineralization comprises a series of concordant, but compositionally zoned, mineralization levels identified by distinct PGE, Au and Cu peaks. They formed due to local sulphide saturation in stratiform concentrations of interstitial and evolved mush melts in six MLs over > 2000 years. The PGE–Au mineralization is compared to a stack of gold-rimmed saucers of PGE-rich gabbro of upward decreasing size. Fundamentally different crystallization and mineralization scenarios have been proposed for the mineralization, including offset reef type models based on sulphide saturation in the melt from which the silicate host crystallized, and the here argued model which restricts the same processes to the melt of the inward migrating mush zone of the magma chamber. The latter is supported by: i) a 3 D summary of the parageneses of precious metal minerals and phases (> 4000 grains) from 32 samples across the mineralization; ii) a 3 D compilation of all bulk rock assay data; and iii) a principal component analysis (PCA) of PGE, Au, Cu, and selected major and trace elements. In the main PGE-mineralization level (Pd5 alias Pd-Zone) the precious metal mineral paragenesis varies across the intrusion with precious metal sulphides and Au-alloys at the W-margin to Precambrian basement, precious metal plumbide and Au- and Ag-alloys at the E-margin to flood basalts, and skaergaardite (PdCu) and intermetallic compounds and alloys of PGE–Au and Cu in the central parts of the mineralization. Precious metal parageneses are distinct for a given sector of the intrusion, i.e. drill core (local control), rather than for a given stratigraphic or temporal interval in the accumulated gabbros. The precious metal ‘grade times width’ number (average g/t x metres) for the mineralization at an upper and a lower cut off of 100 ppb PGE or Au increases from ∼20 to ∼45 g toward the centre of the mineralization due to ponding of precious metal bearing melt. A strong increase in (Pd+Pt+Au)/Cu and dominance of (PdCu) alloys in the lower and central parts of the mineralization demonstrate the partial dissolution of droplets of Cu-rich sulphide melt and fractionation of precious metal ratios. The precious metal parageneses, the distribution of precious metals in the mineralization, and the PCA support initial accumulation of precious metals in the melt of the mush in the floor, followed by equilibration, sulphide saturation, and reactions with residual and immiscible Fe-rich silicate melt in a series of macro-rhythmic layers in the stratified and upward migrating mush zone in the floor of the magma chamber. Syn-magmatic and upward redistribution of precious metals sets the Skaergaard PGE–Au Mineralization apart from conventional reef type and offset-reef type precious metal mineralizations, and characterize ‘Skaergaard type’ precious metal deposits. INTRODUCTION The Skaergaard PGE–Au mineralization (Nielsen et al., 2015), also referred to as the Platinova Reef (Andersen et al., 1998), is a large, low grade Platinum Group Element (PGE) and gold (Au) mineralization hosted in the upper 100 m of the Middle Zone gabbros of the intrusion (Fig. 1). The mineralization is estimated to have an inferred resource of >30 million ounces (oz.) PGE and ∼9 million oz. Au (Nielsen et al., 2005; Kuo, 2007) and approximates giant mineralizations as defined by Laznicka (2006). It relates to stratiform precious metal mineralization in layered mafic intrusions for which models range from contamination-, magma mixing- or fractionation-driven sulphide saturation leading to gravitational accumulation of droplets of immiscible sulphide melt, to upwards-directed transportation of precious metals in residual melts or fluids and re-deposition, for example at redox barriers, to form precious metal reefs (e.g. review in Boudreau & Meurer, 1999; Boudreau, 2004; Barnes et al., 2017). Fig. 1 Open in new tabDownload slide Geology of the Skaergaard intrusion: (a) Map with series (LS, MBS and UBS), zones of LS, collar locations for drill cores and the sampling site ‘Toe of Forbindelsesgletscher’ (ToF), and the chip lines Midnat (Mn) and Middag (Md) Buttresses. Red star, position of the centre of the mineralization used in Fig. 10; (b) E–W section through the intrusion with subdivisions as in (a); (c) schematic representation of the correlation between zones and sub-zones of LS, MBS and UBS. After Nielsen et al. (2015) and Salmonsen &Tegner (2013). Fig. 1 Open in new tabDownload slide Geology of the Skaergaard intrusion: (a) Map with series (LS, MBS and UBS), zones of LS, collar locations for drill cores and the sampling site ‘Toe of Forbindelsesgletscher’ (ToF), and the chip lines Midnat (Mn) and Middag (Md) Buttresses. Red star, position of the centre of the mineralization used in Fig. 10; (b) E–W section through the intrusion with subdivisions as in (a); (c) schematic representation of the correlation between zones and sub-zones of LS, MBS and UBS. After Nielsen et al. (2015) and Salmonsen &Tegner (2013). Mineralization models based on sulphide saturation currently provide the most plausible explanation for the Skaergaard PGE–Au mineralization. Within this class, there are currently two different models, outlined most recently by Nielsen et al. (2015) and Holwell et al. (2016) that described the structure of the mineralization and the processes inherent in its formation. They are based on contrasting perceptions of the structure of the mineralization and the distribution of precious metals in the layered gabbros. Nielsen et al. (2015) model the mineralization as a suite of concordant but compositionally zoned mineralization levels within a > 600 m deep bowl in the floor of the intrusion, whereas Holwell & Keays (2014) build on correlation of precious metals anomalies, with increasing number of mineralization levels in the geographic centre of the mineralization (hereafter ‘centre’) and a capping gold-zone. In the centre of the mineralization the Au-zone of Holwell & Keays (2014) is above leucograbbro layer 2 (L2) and at the margin of the intrusion below leucograbbro layer 1 (L1) of the Triple Group, some 40 metres lower in the succession of gabbroic layers. Nielsen et al. (2015) and Holwell et al. (2016) argue for accumulation of immiscible droplets of sulphide melt enriched in precious metals that were subsequently subjected to reaction with Fe-rich silicate melt. Constrained by their perceptions of the structure of the mineralization, Holwell & Keays (2014) and Holwell et al. (2015, 2016) suggested that sulphide droplets sank through and scavenged precious metals from the bulk liquid accumulated and partly dissolved in Fe-rich silicate melt that ponded at the crystallization front at the floor of the magma chamber. The elemental variations up the gabbroic layers of the mineralization would, therefore, in modified form reflect fractionation and evolution of bulk liquid. The model is based and supported by observations mainly from drill cores near the margin of the intrusion (e.g. Holwell et al., 2016). Nielsen et al. (2015) alternatively argued that the bulk melt was circulated into the crystallization zone under the roof of the magma chamber, where it fractionated, reached sulphide saturation, and subsequently immiscibility between Fe- and Si-rich silicate melts. The tiny Cu-rich sulphide droplets that formed at sulphide saturation, had a density similar to Fe–Ti-oxides and were suspended in the mush melt. When the mush melt reached the two-liquid field between Fe- and Si rich silicate liquids, buoyant Si-rich melt rose and the already formed Cu-rich sulphide droplets dissolved into the dense and Fe-rich conjugate that descended along the walls to the floor of the magma chamber. Next, the now dissolved precious metals were re-deposited in macro-rhythmic layers during upward migration of the mushy crystallization zone in the floor of the intrusion. In each macro-rhythmic layers the processes were repeated: i) crystal fractionation; ii) density controlled concentration of mush melt; iii) sulphide saturation and trapping of precious metals; iv) liquid immiscibility and loss of Si-rich conjugate; v) reaction and dissolution of sulphide droplets leaving behind precious metal phases in the gabbros. It is a model that combines initial accumulation followed by an upward redistribution of precious metals in the mineralization. The main difference between the proposed sulphide saturation type models are not the processes involved (sulphide saturation, dissolution and fractionation in accordance with partition coefficients), but the timing of mineralization processes relative to the crystallization of the host rocks, and the volumes of melt to which the processes are applied. Lateral variations are significant in the intrusion and no single drill core can be representative for the mineralization. In this contribution we illustrate the 3 D distributions of the precious metals and their mineral parageneses in a search for additional constraints for the proposed mineralization models. Specifically, we compiled the information in order to evaluate the validity of the two-stage model and the upward redistribution of the precious metals in the mineralization (Nielsen et al., 2015). We include: (i) as a basis for the evaluations of the proposed models, a summary of the fundamental elemental distributions and structural observations that are critical for the perception of the structure of Skaergaard-type mineralizations (Prendergast, 2000; Miller & Andersen, 2002) and the relative timing of processes in the intrusion; (ii) intrusion-wide compilation and interpretation of the precious metal mineral parageneses from 32 samples from drill cores and bulk samples (the mineralogical data for individual samples were collected 2003–2012 but have not been compiled and discussed in any detail prior to this work); (iii) the distribution of precious metal in the mineralization using here compiled grade*width numbers (g*w, average g/t x width or height in metres) compiled for stratigraphic intervals on the basis of systematic bulk rock assays (Watts, Griffis & McOuat, 1991; Hanghøj, 2005); and (iv) a principal component analysis (PCA) of the distribution of precious metals, incompatible elements and major element oxides in the fully developed mineralization in the centre of the mineralization. The PCA is carried out to confirm combination of processes proposed to be responsible for the distribution of the precious metals. Supplementary information, including elemental correlations, paragenetic information, additional chemical information, and details on the PC analysis are provided in Supplementary Data Electronic Appendices EA1–EA14; supplementary data are available for downloading at http://www.petrology.oxfordjournals.org. The Skaergaard intrusion The Skaergaard intrusion is dated at 56 Ma (Wotzlaw et al., 2012) and was emplaced during the opening of the North Atlantic Ocean. It is a comparatively small, but well-preserved and well-exposed layered gabbro intrusion (Fig. 1a,). It is ∼7 x 11 km in surface area (Wager & Brown, 1968), ∼4 km in thickness, and has a box-like or ellipsoidal shape with a volume of ∼300 km3 (Irvine et al., 1998; Nielsen, 2004; Svennevig & Guarnieri, 2012). The magma crystallized inwards in an onion-ring type structure (Fig. 1b), with the Layered Series including Hidden Zone (HZ), Lower Zone (LZ), Middle Zone (MZ) and Upper Zone (UZ) in the bowl-shaped floor, the Marginal Border Series (MBS) along the walls, and the Upper Border Series (UBS) below the roof. The UBS and LS meet at the Sandwich horizon (SH) (Fig. 1b and c) in the upper and central part of the intrusion. All three series are further subdivided on the basis of their parallel evolution of liquidus parageneses (Wager & Brown, 1968; Salmonsen &Tegner, 2013 and references therein; see Fig. 1). Significant volumes of melanogranophyre and granophyre occur as sill-like bodies in and between the SH and UBS gabbros and represent strongly evolved compositions on the line of liquid descent (Wager & Brown, 1968; McBirney, 1989; Nielsen, 2004; Salmonsen, 2013). Detailed accounts of the intrusion are found in Wager & Brown (1968), McBirney (1996), Irvine et al. (1998) and Nielsen (2004). The mineralization The following description of the Skaergaard PGE–Au mineralization is a summary based on Nielsen (2004) and Nielsen et al. (2015), who identified and correlated elemental, lithological and density anomalies between forty-one drill cores and additional chip lines from the intrusion. N–S and E–W correlations between drill cores are provided in Andersen et al. (1998), Nielsen et al. (2015), and Holwell & Keays (2014). All publicly available primary information, including the correlation between systematic assays and lithological logs in the drill cores and density logs (which are proxies for lithology) from 13 drill cores are included in Nielsen et al. (2015, references therein, and Supplementary Data SD1-SD3, and SD5 of that publication) and Electronic Appendices EA1–EA4 of the present work. In its centre, the mineralization is hosted in the lower ∼60 m of the Triple Group (TG, Fig. 2a;Andersen et al., 1998; Holness et al., 2017c). The Triple Group is an ∼100 m thick, bowl-shaped succession of macro-rhythmic layers (MLs) located in the uppermost part of Middle Zone in the Layered Series in the floor of the intrusion (Nielsen, 2004). The TG owes its name to three prominent leucogabbro layers referred to as L1, L2 and L3, which are easily observed from a distance (Fig. 2a and b). These leucogabbro layers are lithological markers that provide constraints on the relative timing of the accumulation of the gabbroic host and the precious metals. Nielsen (2004) modelled the structure of the Layered Series in 2 D in E–W and N–S sections, and Nielsen et al. (2009) extended the models to three dimensions using the geological modelling tool Leapfrog®. These models constrain the bowl-shaped layering in the Layered Series to be ∼7000 m wide (E-W) and >600 m deep (Fig. 3; Nielsen et al., 2015; Holness et al., 2017b). The distribution of precious metals is roughly concentric and the mineralization fully developed around a centre in the south–central part of the intrusion where the MZ is at its thickest (Watts, Griffis & McOuat 1991; Andersen et al., 1998; Nielsen, 2004; Nielsen et al., 2005, 2015). Fig. 2 Open in new tabDownload slide (a) Western face of Wagertoppen (1277 m) showing the three leucogabbro layers of the Triple Group (TG). The layering in the upper left is disturbed by large blocks of gabbro assumed to be derived from the roof of the magma chamber. The layering drapes over the blocks of UBS which have sunk into the magma chamber (e.g. lower right; photo by M.B. Holness). (b) Close-up of the north-western ridge of Wagertoppen (photo by J.C. Ø. Andersen) showing L1 and L2 of the TG maintaining constant stratigraphic separation even in the most northwestern exposures and the distinct layering of the zebra-banded zone the lower part of the image. Includes several leucogabbro blocks and rafts, some coherent and some smeared out parallel to the layering in the gabbroic host. Yellow vertical bars identify the macro-rhythmic layers (ML0-ML2·2) of the mineralized section of the TG, each with a plagioclase rich top. Pd5 of the mineralization is hosted in the upper metres of ML0, and Pd1 which hosts the main Au concentration in the central parts of the intrusion is hosted in ML2·1. The section of MLs covered by the Midnat chipline (Turner & Mosher, 1989) is also shown. Fig. 2 Open in new tabDownload slide (a) Western face of Wagertoppen (1277 m) showing the three leucogabbro layers of the Triple Group (TG). The layering in the upper left is disturbed by large blocks of gabbro assumed to be derived from the roof of the magma chamber. The layering drapes over the blocks of UBS which have sunk into the magma chamber (e.g. lower right; photo by M.B. Holness). (b) Close-up of the north-western ridge of Wagertoppen (photo by J.C. Ø. Andersen) showing L1 and L2 of the TG maintaining constant stratigraphic separation even in the most northwestern exposures and the distinct layering of the zebra-banded zone the lower part of the image. Includes several leucogabbro blocks and rafts, some coherent and some smeared out parallel to the layering in the gabbroic host. Yellow vertical bars identify the macro-rhythmic layers (ML0-ML2·2) of the mineralized section of the TG, each with a plagioclase rich top. Pd5 of the mineralization is hosted in the upper metres of ML0, and Pd1 which hosts the main Au concentration in the central parts of the intrusion is hosted in ML2·1. The section of MLs covered by the Midnat chipline (Turner & Mosher, 1989) is also shown. Fig. 3 Open in new tabDownload slide Examples of available elemental and density profiles through the Skaergaard PGE–Au Mineralization (see Fig. 1a for locations). (a) From left to right: in order of increasing distance from the centre of the concentric mineralization. Compositional profiles: PGE (Pd+Pt) in blue; Au in yellow; and Cu in red in a continuous 25-cm bulk rock profile in drill core 90-22 from the centre of the intrusion, 1-m continuous averages in drill core 90-17 A located ∼ 1150 m from the western margin, in drill core 90-23 A located ∼ 900 m from the eastern margin of the intrusion, and in the Middag chip line profile almost 5 km N of the centre of the mineralization (se Fig. 1). Density profiles for drill cores 90-22 (centre) and 90-23 A (margin) and in grey the logged elevation of leucolayers L0, and L1 and L2 of the Triple Group demonstrate the extreme continuity of layering in the host rocks as well as of the mineralization levels and the upward migration relative to MLs of precious metals and Cu toward the supposed centre of the mineralization. Details and sources for the plotted data can be found in Nielsen et al. (2015) and Supplementary Data therein. (b) Cross section to scale after Supplementary Data EA2 in Nielsen et al. (2015) (see location in Fig. 1a) with elevations of intersects between drill cores and Pd5 (blue) and Pd1 (red) mineralization levels and leucogabbro layer L3 (yellow) of the Triple Group. Only drill cores for which all three markers are identified are included. In all others, one or more data points are lost in due to intersecting dykes. Mineralization levels are concordant with the lithological layering in the 7000 m wide and >600 m deep bowl-shaped succession of macro-rhythmic layers (MLs) of the upper Middle zone. Data for the compilation can be found in Supplementary Data EA1-3 in Nielsen et al. (2015). Fig. 3 Open in new tabDownload slide Examples of available elemental and density profiles through the Skaergaard PGE–Au Mineralization (see Fig. 1a for locations). (a) From left to right: in order of increasing distance from the centre of the concentric mineralization. Compositional profiles: PGE (Pd+Pt) in blue; Au in yellow; and Cu in red in a continuous 25-cm bulk rock profile in drill core 90-22 from the centre of the intrusion, 1-m continuous averages in drill core 90-17 A located ∼ 1150 m from the western margin, in drill core 90-23 A located ∼ 900 m from the eastern margin of the intrusion, and in the Middag chip line profile almost 5 km N of the centre of the mineralization (se Fig. 1). Density profiles for drill cores 90-22 (centre) and 90-23 A (margin) and in grey the logged elevation of leucolayers L0, and L1 and L2 of the Triple Group demonstrate the extreme continuity of layering in the host rocks as well as of the mineralization levels and the upward migration relative to MLs of precious metals and Cu toward the supposed centre of the mineralization. Details and sources for the plotted data can be found in Nielsen et al. (2015) and Supplementary Data therein. (b) Cross section to scale after Supplementary Data EA2 in Nielsen et al. (2015) (see location in Fig. 1a) with elevations of intersects between drill cores and Pd5 (blue) and Pd1 (red) mineralization levels and leucogabbro layer L3 (yellow) of the Triple Group. Only drill cores for which all three markers are identified are included. In all others, one or more data points are lost in due to intersecting dykes. Mineralization levels are concordant with the lithological layering in the 7000 m wide and >600 m deep bowl-shaped succession of macro-rhythmic layers (MLs) of the upper Middle zone. Data for the compilation can be found in Supplementary Data EA1-3 in Nielsen et al. (2015). The fully developed precious metal mineralization The fully developed PGE-rich part of the mineralization is in its centre hosted in eight stratigraphic intervals of host gabbro. The mineralized intervals of gabbro are referred to as Pd-levels (Fig. 3a,) and each interval has upward decreasing Pd/Pt and an easily identified Pd-dominated peak (Nielsen et al., 2005, 2015). The Pd-levels and peaks are numbered top-down from Pd1 to Pd6 and with sub-fixes “a” and “b” for levels Pd2-Pd4 (Nielsen et al., 2005). Individual Pd-levels can be correlated across the intrusion (Fig. 3a, EA1). The naming of the Pd-levels is a reminiscence of the earliest identification of mineralization levels in the early years of exploration (1986–1990). The main resource of PGEs is in level Pd5. In this work, the peak concentration of PGE (Pd+Pt) in Pd5 is used as marker and for all correlations between drill core logs and systematic assays. Au/PGE increases up the mineralization (Andersen et al., 1998; Nielsen et al., 2005; Holwell & Keays, 2014). Above Pd1, a PGE-poor, but Au-rich, stratigraphic interval identifies the top of the succession of gabbros enriched in precious metals. It is elevated ∼2 m above Pd1 and is referred to as Au/Pd1 (see, e.g. drill core 90–22 in Fig. 3a). Overlying Cu-bearing intervals with 100–200 ppb gold are numbered sequentially with increasing height from Au + 1, Au + 2 etc. Au + 1 is located ∼13 m above Au/Pd1 (see, e.g. drill core 89–09 in Andersen et al., 1998) and is identified in more drill cores despite very low levels of Au. All Pd-levels maintain near-constant elevations relative to Pd5 and can be correlated across the intrusion at the exact same stratigraphic position, irrespective of the absolute concentration of precious metals or Cu (Fig. 3a;Table 1; Electronic Appendix EA1). For example, the double peaks Pd2a and b, and Pd3a and b, can be identified near the margins of the intrusion even at very low (10–100 ppb) bulk concentrations (Fig. 3a; Electronic Appendix EA2). Pd-levels are stratigraphic intervals of gabbroic host with or without precious metal concentrations. The concentrations and elemental ratios of precious metals and Cu vary laterally in the Pd-levels. For example, peak Pd2b is PGE-rich in the centre of the mineralization (Fig. 3a, drill core 90–22), Au-rich approximately half way between the centre and margin of the concentric mineralization (Fig. 3a, drill core 90–17 A) and Cu-rich near the margin (Fig. 3a, drill core 90–23 A). In no case is the main gold-rich peak detached from a mineralization level that can be identified across the intrusion at the given stratigraphic level in the succession of macro-rhythmic layers. Table 1 Stratigraphic elevation of mineralization levels and bases of macrorhythmic layers across the intrusion and relative to elevation of Pd5 ML2·2 . base . 50·75 . Pd/Au peak 42·25 Pd1 peak 40·25 ML2·1 base 37·00 Pd2a peak 32·50 Pd2b peak 27·75 ML 2 base 23·00 Pd3a peak 18·75 Pd3b peak 16·50 ML1·2 base 12·25 Pd4a peak 10·50 Pd4b peak 5·50 ML1·1 base 2·50 Pd5·2* peak 0·75 Pd5·1* peak 0·00 ML0 base −8·75 ML2·2 . base . 50·75 . Pd/Au peak 42·25 Pd1 peak 40·25 ML2·1 base 37·00 Pd2a peak 32·50 Pd2b peak 27·75 ML 2 base 23·00 Pd3a peak 18·75 Pd3b peak 16·50 ML1·2 base 12·25 Pd4a peak 10·50 Pd4b peak 5·50 ML1·1 base 2·50 Pd5·2* peak 0·75 Pd5·1* peak 0·00 ML0 base −8·75 Based on 1-metre assays, logs and density profiles in 41 drill cores. Elevations relative to Pd5 are compiled in EA 1 and 3 in Nielsen et al. (2015)· * The Pd5 peak is, where resolution allows, divided in an upper and a lower maxima. Open in new tab Table 1 Stratigraphic elevation of mineralization levels and bases of macrorhythmic layers across the intrusion and relative to elevation of Pd5 ML2·2 . base . 50·75 . Pd/Au peak 42·25 Pd1 peak 40·25 ML2·1 base 37·00 Pd2a peak 32·50 Pd2b peak 27·75 ML 2 base 23·00 Pd3a peak 18·75 Pd3b peak 16·50 ML1·2 base 12·25 Pd4a peak 10·50 Pd4b peak 5·50 ML1·1 base 2·50 Pd5·2* peak 0·75 Pd5·1* peak 0·00 ML0 base −8·75 ML2·2 . base . 50·75 . Pd/Au peak 42·25 Pd1 peak 40·25 ML2·1 base 37·00 Pd2a peak 32·50 Pd2b peak 27·75 ML 2 base 23·00 Pd3a peak 18·75 Pd3b peak 16·50 ML1·2 base 12·25 Pd4a peak 10·50 Pd4b peak 5·50 ML1·1 base 2·50 Pd5·2* peak 0·75 Pd5·1* peak 0·00 ML0 base −8·75 Based on 1-metre assays, logs and density profiles in 41 drill cores. Elevations relative to Pd5 are compiled in EA 1 and 3 in Nielsen et al. (2015)· * The Pd5 peak is, where resolution allows, divided in an upper and a lower maxima. Open in new tab Correlation to the gabbroic host and the saucer model Systematic structural relationships are observed between mineralized levels and the host gabbros of the intrusion. For example, the peak concentration of PGE in Pd5 is always located ∼2 metres below the top of leucogabbro layer L0 (defined as the leucogabbro top to the lowermost ML in the mineralized interval, see Supplementary Data SD5 in Nielsen et al., 2015), and Pd1 is always ∼40 m above Pd5, ∼3 m above the top of the L2 leucogabbro member of the Triple Group, irrespectively of the absolute concentration of precious metals (Fig. 3a, Table 1; see also Andersen et al., 1998; and Supplementary Data SD1, and SD3 in Nielsen et al., 2015)). The near-constant stratigraphic separations between Pd-levels observed in all available drill cores, and the correlation between 13 density profiles aligned at the Pd5 peak (Supplementary Data EA3), highlight that the mineralization levels are concordant with the host rocks across the bowl-shaped floor of the intrusion (Table 1). The elevations relative to sea level of the Pd5 and Pd1 peaks and leucogabbro layer L3 in drill cores can be traced across the intrusion (Fig. 3b) and illustrate the concordant relationship between leucogabbro layers and Pd-levels in the >600 metre deep bowl of Triple Group (Fig. 3b). The stratigraphic separation between Pd5 and the main Au-peak decreases from >43 m at the central parts of the mineralization to < 1 m in the most distant profiles (Fig. 3a). In the Midnat and Middag Buttress chip lines located farthest away from the centre of the mineralization (Fig. 1a), the PGE and peak concentrations of Au overlap within the same metre of gabbro (Turner & Mosher, 1989; Andersen et al., 1998). The distribution of precious metals in the intrusion is, therefore, diachronous relative to the layering in the gabbroic host (Fig. 3a) and no single drill core can provide representative assays through the Skaergaard PGE–Au mineralization. Some represent the fully developed succession of gabbroic MLs in the centre of the mineralization (e.g. 90–22; Bernstein & Nielsen, 2004), some the little-developed mineralization at the margins (e.g. Supplementary Data EA2; Holwell et al., 2016), whilst the remainder are transitional between the two. The integration of the of the bowl-shape of the layered gabbros (Figs 1b and 3b), the concordant Pd-levels, and the variation in concentration of precious metals through the ML stratigraphy (Fig. 3a;Supplementary Data SD3 in Nielsen et al., 2015) resulted in the mineralization being described as a succession of gold-rimmed saucers of PGE-enriched gabbro that decrease in diameter up through the stratigraphy (Fig. 4). Fig. 4 Open in new tabDownload slide Schematic illustration of the structure of the precious metal mineralization. The Pd-levels are concordant with the bowl-shape of the magmatic layering in the intrusion (Nielsen et al., 2015, Supplementary Data EA2). The circular structure of the mineralization is based on modelling in Watts, Griffis & McOuat (1991), Andersen et al. (1998), Nielsen et al. (2005) and is idealised relative to the actual and more irregular structure. The colour change from blue to yellow symbolizes the lateral variation from PGE-rich to Au- and Cu-rich mineralization levels. Fig. 4 Open in new tabDownload slide Schematic illustration of the structure of the precious metal mineralization. The Pd-levels are concordant with the bowl-shape of the magmatic layering in the intrusion (Nielsen et al., 2015, Supplementary Data EA2). The circular structure of the mineralization is based on modelling in Watts, Griffis & McOuat (1991), Andersen et al. (1998), Nielsen et al. (2005) and is idealised relative to the actual and more irregular structure. The colour change from blue to yellow symbolizes the lateral variation from PGE-rich to Au- and Cu-rich mineralization levels. Macro-rhythmic layers and compositional subdivision Density (ρ) logs can be used as a proxy for lithology, and correlations across the intrusion. Leucogabbro layers are easily followed as concordant stratigraphic intervals (Supplementary Data EA3; see also Nielsen et al., 2015). The following division of the host rock lithology in drill core 90–22 (Bernstein & Nielsen, 2004; Tegner et al., 2009; Nielsen et al., 2015), therefore, applies to the host gabbros across the intrusion. This drill core was chosen out of eleven from central parts of the mineralization for the establishment of the so-called ‘Standard Profile’ in the Skaergaard intrusion (Nielsen et al., 2000; Tegner et al., 2009; Keays & Tegner, 2016) because it has a very high recovery with very few intersecting dikes within the mineralized interval (Tegner et al., 2009). The gabbros of the Triple Group divide into several MLs that encapsulate the prominent leucogabbro layers shown in Fig. 2. The MLs have dense, pyroxene-rich bases (ρ  =  3·4–3·6 g/cm3; modal abundance of plagioclase relative to clinopyroxene ≅0·5) and low density and plagioclase-rich tops (ρ  = 2·9–3·1 g/cm3; modal abundance of plagioclase relative to clinopyroxene >2) (data in Bernstein & Nielsen, 2004; Nielsen et al., 2015). Transitions between MLs are located immediately above the leucogabbro layers, in sections of gabbro with a marked upward increase in the proportion of clinopyroxene relative to plagioclase, and an accompanying increase in density. Nielsen et al. (2015) divided the stratigraphic interval enriched in precious metals in drill core 90–22 into four MLs including ML0, ML1, ML2 and ML2.1 (named with reference to the included leucogabbro layers L0, L1 and L2), where ML0 is the lowest-most ML. In this study, the lithological sub-division of the stratigraphic interval of the mineralization is revised to comprise six ML layers that each maintain near constant stratigraphic thickness across the intrusion. ML1 of Nielsen et al. (2015) was exceptionally thick (>20 vs ∼13 m for all other MLs) and, based on re-evaluation of the density logs, it is now divided into ML1·1, ML1·2, and ML2·2 is added above ML2·1. Low-density layers in the tops of ML1·1 (∼10 m above Pd5 marker) and ML2 ·1 (∼50 m above Pd5 marker) are easily followed across the intrusion in the density logs (Supplementary Data EA3). Table 1 and Fig. 5 outline the revised lithological subdivision of the gabbroic host of the mineralization. ML0 hosts precious metal mineralization levels Pd6 and Pd5 (Sub-zone and Pd-zone of Holwell & Keays, 2014), ML1·1 hosts Pd4a and b, ML1·2 hosts Pd-levels Pd3a and b, ML2 hosts Pd2a and b, ML2·1 hosts Pd1 and Au/Pd1, and ML2·2 hosts the Au + 1 (Fig. 5). Fig. 5 Open in new tabDownload slide Correlation between mineralization levels, macro-rhythmic layers (MLs) of the host gabbros, and subdivision of the mineralization as developed in the geographical centre of the multi-layered mineralization (drill core 90-22). The subdivision of the gabbros into macro-rhythmic layers follows the principles in Nielsen et al. (2015; see also text for details). Left: the density profile is divided into MLs (centre) on the basis of the midpoint of marked increases in density, i.e. half way between the low-density top of a lower ML and the density high near the base of the overlying ML. Grey sections identify ilmenite-rich intervals occurring at regular intervals but with no apparent correlation to the layering in MLs (see Nielsen et al., 2015). Right: Mineralization levels Pd6 and Pd5 in ML0, Pd4a, b in ML1·1, Pd3a, b in ML1·2, Pd2a, b in ML2, and Pd1 and Pd1/Au in ML2. Geochemical subdivision of the mineralization into LPGEM, UPGEM, UAuM and CuM in the central column. The Au-rich combined Pd1 and Pd1/Au mineralization levels have an average of 2·1 g/t precious metals over 3·6 m in drill core 90-22. Fig. 5 Open in new tabDownload slide Correlation between mineralization levels, macro-rhythmic layers (MLs) of the host gabbros, and subdivision of the mineralization as developed in the geographical centre of the multi-layered mineralization (drill core 90-22). The subdivision of the gabbros into macro-rhythmic layers follows the principles in Nielsen et al. (2015; see also text for details). Left: the density profile is divided into MLs (centre) on the basis of the midpoint of marked increases in density, i.e. half way between the low-density top of a lower ML and the density high near the base of the overlying ML. Grey sections identify ilmenite-rich intervals occurring at regular intervals but with no apparent correlation to the layering in MLs (see Nielsen et al., 2015). Right: Mineralization levels Pd6 and Pd5 in ML0, Pd4a, b in ML1·1, Pd3a, b in ML1·2, Pd2a, b in ML2, and Pd1 and Pd1/Au in ML2. Geochemical subdivision of the mineralization into LPGEM, UPGEM, UAuM and CuM in the central column. The Au-rich combined Pd1 and Pd1/Au mineralization levels have an average of 2·1 g/t precious metals over 3·6 m in drill core 90-22. Each ML includes a lower precious metal level which Nielsen et al. (2015) argue is cumulative and an upper level in mush melt accumulated in the base of the low density roof of the MLs (Fig. 5). In Fig. 5 are included sections of drill core 90–22 enriched in tiny euhedral crystals of ilmenite that cannot sink. These layers are found in all lithologies including leuco-, meso-, and melano-gabbros and show no relationship to mineralization levels or to the distribution of precious metals and Cu. They are in Nielsen et al. (2015) understood as formed due to stratbound nucleation of ilmenite in the melt of the stratified mush zone. In addition to the lithologic sub-division into MLs, Nielsen et al. (2015) sub-divided the fully developed mineralization on the basis of the distribution of the precious metals. The fully developed mineralization (e.g. in drill core 90–22) was divided into the ‘Lower PGE Mineralization’ (LPGEM) comprising ML0 and ML1·1 (Pd6, Pd5, Pd4a and b levels), and the ‘Upper PGE Mineralization’ (UPGEM) comprising ML1·2, ML2, ML2·1 (Pd3a and b, Pd2a and b, Pd1 and Au/Pd1 levels, see Fig. 5). LPGEM is rich in precious metals across the floor, whereas in the UPGEM, the precious metals are increasingly restricted to smaller and smaller gold-rimmed bowls upwards through the sequence of MLs and towards the centre of the mineralization (Figs 4 and 5). Au/PGE in bulk samples increases upwards through the mineralization levels (Bird et al., 1991; Andersen et al., 1998). Skaergaardite (PdCu) is the main precious metal phase in lower mineralization levels in the central parts of the mineralization. It is increasingly Au-rich up through the mineralization levels and is in the upper parts of the mineralization accompanied by tetra-auricupride (AuCu) (Nielsen et al., 2005; Rudashevsky et al., 2014). In addition, Au also occurs as anhedral grains on grain boundaries in already crystallized gabbros (Godel et al., 2014; Rudashevsky et al., 2014). This additional form of gold deposition is referred to as the ‘Upper Au Mineralization’ (UAuM) (Nielsen et al., 2015) and is observed in or above the uppermost Pd-level, with more than traces of PGE (Fig. 6). The upward distribution of Au in the top of the precious metal rich mineralization is illustrated by drill core 89–09 (Turner, 1990). The core was drilled near the centre of the mineralization and ∼1·5 km NW of drill core 90–22. It was wedged twice and the three parallel drill cores have very similar (g*w) numbers (Table 2), but nevertheless very different distributions of gold (Fig. 6). In drill core 89–09 the main gold peak is in the Pd1/Au level, in drill core 89–09 A in Pd1/Au as well as in a small peak c. 1 metre above, and in 89-09B the main gold peak is elevated c. 3 m above Pd1 (Fig. 6). Major local variations are seen in the distribution of gold relative to the Pd1 peak. Fig. 6 Open in new tabDownload slide Variation in PGE (Pd+Pt) and Au in drill cores 89-09, 89-09 A and 89-09B from the central parts of the mineralization. Data from Turner (1990). Drill core 89-09 was wedged twice to produce three neighbouring cores (Fig. 6a–c) through the mineralized gabbros. The elemental profiles for PGE are very similar in all three cores, whereas the details of the uppermost Au-rich mineralization (Fig. 6d–f) varies significantly, despite having very similar grade times width numbers (g*w, average grade in grams/ton times the width or height of the stratigraphic interval in metres; Table 2). The unconstrained distribution of gold is on the basis of petrographic observations (Godel et al., 2014), argued to be due to local and late redistribution (see text for further explanations). Fig. 6 Open in new tabDownload slide Variation in PGE (Pd+Pt) and Au in drill cores 89-09, 89-09 A and 89-09B from the central parts of the mineralization. Data from Turner (1990). Drill core 89-09 was wedged twice to produce three neighbouring cores (Fig. 6a–c) through the mineralized gabbros. The elemental profiles for PGE are very similar in all three cores, whereas the details of the uppermost Au-rich mineralization (Fig. 6d–f) varies significantly, despite having very similar grade times width numbers (g*w, average grade in grams/ton times the width or height of the stratigraphic interval in metres; Table 2). The unconstrained distribution of gold is on the basis of petrographic observations (Godel et al., 2014), argued to be due to local and late redistribution (see text for further explanations). Table 2 Grade times width numbers (g*w) for the gold-peak in drill cores 89-09, 89-09 A and 89-09B† Drill core . from . to . width . Au av. . PGE av. . Au av. . PGE av. . PGE+Au . . (m) . (m) . (m) . ppb . ppb . ppm . ppm . g*w . 4m interval 89-9 443·0 447·0 4·0 1311 511 1·3 0·5 7·2 89-9A 443·4 447·0 4·0 1229 495 1·2 0·5 6·8 89-9B 442·0 446·0 4·0 1106 666 1·1 0·7 7·2 0·7 g/t cut off* 89-9 444·8 446·4 1·6 2992 1093 3·0 1·1 6·6 89-9A 444·4 446·6 2·2 2277 848 2·3 0·9 7·0 89-9B 442·2 445·6 3·4 1293 747 1·3 0·7 6·8 Drill core . from . to . width . Au av. . PGE av. . Au av. . PGE av. . PGE+Au . . (m) . (m) . (m) . ppb . ppb . ppm . ppm . g*w . 4m interval 89-9 443·0 447·0 4·0 1311 511 1·3 0·5 7·2 89-9A 443·4 447·0 4·0 1229 495 1·2 0·5 6·8 89-9B 442·0 446·0 4·0 1106 666 1·1 0·7 7·2 0·7 g/t cut off* 89-9 444·8 446·4 1·6 2992 1093 3·0 1·1 6·6 89-9A 444·4 446·6 2·2 2277 848 2·3 0·9 7·0 89-9B 442·2 445·6 3·4 1293 747 1·3 0·7 6·8 * g*w, average grade in grams/ton of given stratigraphic interval multiplied by its height in metres· † Cut off based on (Pd+Pt+Au)· Data from Watts, Griffis & McOuat (1991). Open in new tab Table 2 Grade times width numbers (g*w) for the gold-peak in drill cores 89-09, 89-09 A and 89-09B† Drill core . from . to . width . Au av. . PGE av. . Au av. . PGE av. . PGE+Au . . (m) . (m) . (m) . ppb . ppb . ppm . ppm . g*w . 4m interval 89-9 443·0 447·0 4·0 1311 511 1·3 0·5 7·2 89-9A 443·4 447·0 4·0 1229 495 1·2 0·5 6·8 89-9B 442·0 446·0 4·0 1106 666 1·1 0·7 7·2 0·7 g/t cut off* 89-9 444·8 446·4 1·6 2992 1093 3·0 1·1 6·6 89-9A 444·4 446·6 2·2 2277 848 2·3 0·9 7·0 89-9B 442·2 445·6 3·4 1293 747 1·3 0·7 6·8 Drill core . from . to . width . Au av. . PGE av. . Au av. . PGE av. . PGE+Au . . (m) . (m) . (m) . ppb . ppb . ppm . ppm . g*w . 4m interval 89-9 443·0 447·0 4·0 1311 511 1·3 0·5 7·2 89-9A 443·4 447·0 4·0 1229 495 1·2 0·5 6·8 89-9B 442·0 446·0 4·0 1106 666 1·1 0·7 7·2 0·7 g/t cut off* 89-9 444·8 446·4 1·6 2992 1093 3·0 1·1 6·6 89-9A 444·4 446·6 2·2 2277 848 2·3 0·9 7·0 89-9B 442·2 445·6 3·4 1293 747 1·3 0·7 6·8 * g*w, average grade in grams/ton of given stratigraphic interval multiplied by its height in metres· † Cut off based on (Pd+Pt+Au)· Data from Watts, Griffis & McOuat (1991). Open in new tab In data sets with high stratigraphic resolution, the first Cu peak (Cu > 1000 ppm) always occurs above main the Au-peak and identifies the base of the Cu-dominated uppermost part of the mineralization (CuM, as defined in Nielsen et al., 2015). The base of CuM is the stratigraphic level at which mineralization levels change from Au-rich to Cu-rich, irrespective of the stratigraphic position of the Au-peak relative to the leucocratic layers of MLs. It is below leucogabbro L1 of the Triple Group at the margins and above L2 in the centre of the intrusion (Fig. 3a). A much simpler geochemical subdivision of the mineralization was proposed by Holwell & Keays (2014). They correlate PGE and Au anomalies across the intrusion, irrespective of the relative stratigraphic elevation, and defined a Pd-zone with an underlying Sub-zone (both PGE rich), an intermediate zone with an increasing number of Pd-levels towards the centre of the mineralization, and a capping Au-zone at the transition to the overlying Cu-rich mineralization. The proposed Au-zone would be diachronous and discordant. It rises from the margin to the centre upwards through ∼40 m of stratigraphy (including four MLs) of the Triple Group (Fig. 3a). It would encompass gold-rich parts of the PGE-saucers of Nielsen et al. (2015), as well as little-mineralized gabbros between gold-rimmed PGE-saucers in the succession of MLs (Figs 3 and 4). Reported precious metal parageneses The volumetric proportion of sulphide in the mineralization is very small and, in samples that have precious metal concentrations in the ppm range it is estimated to be ∼0·05 vol. % (Rudashevsky et al., 2014, 2015; Nielsen et al., 2015; Holwell et al., 2016). The sulphides are almost exclusively Cu-rich and are dominated by bornite, chalcocite, digenite, and minor chalcopyrite. Pyrrhotite, cobaltpentlandite, pentlandite, sphalerite, arsenopyrite and pyrite are all very rare (Rudashevsky et al., 2014, 2015; Nielsen et al., 2015; Holwell et al., 2016). Initial information on the precious metal paragenesis was provided by Bird et al. (1991), Andersen et al. (1998), Cabri et al. (2005), and Nielsen et al. (2005) who summarized the then available mineralogical information from 6 drill cores and reported the occurrence of >30 precious metal minerals and phases (Nielsen et al, 2003a, 2003b, 2003c, 2003d, 2003e; Rudashevsky et al., 2004). The mineralogy is very varied and comprises a large number of known minerals, frequently recorded stoichiometric compositions that are possible unnamed minerals or intermetallic compounds, and a suite of Cu and precious metal-bearing alloys. Subsequent mineralogical investigations presented by Rudashevsky & Rudashevsky (2005a, 2005b, 2006a, 2006b) and Rudashevsky et al. (2009a, 2009b, 2010a, 2010b, 2010c, 2010d, 2012a, 2012b, 2012c, 2012d, 2012e, 2012f, 2012g, 2012h, 2012i, 2014, 2015), Nielsen et al. (2015), McDonald et al. (2008) and Holwell et al. (2015, 2016) confirmed the complex mineralogy and add more minerals and unnamed phases to the list. All of these reports and investigations are the basis for the 3 D compilations and interpretations of the parageneses of the mineralization of this contribution. Common minerals and phases of the Skaergaard PGE–Au mineralization include skaergaardite (PdCu), nielsenite (PdCu3), (Cu, Pd) alloys, tetra-auricupride (AuCu), (Au, Cu, Pd, Ag) alloys, (Pt, Fe, Cu, Pd) alloys, electrum (Au, Ag), (Pt, Fe), seemingly stochiometric PGE and Au intermetallic compounds, a large variety of non-stoichiometric PGE and Au-rich alloys, PGE-sulphides such as vysotskite ((Pd, Ni, Cu, Fe)S), braggite ((Pt, Pd, Ni)S), and vasilite ((Pd, Cu, Fe)16S7), arsenides, such as vincentite ((Pd, Pt)3(As, Sb, Te)) and arsenopalladinite (Pd8(As, Sb)3), tellurides such as keithconnite (Pd3Te), stannides like atokite (Pd3Sn), and the plumbide zvyagintsevite (Pd3Pb). In the fully developed central part of the mineralization the precious metals are hosted in skaergaardite (PdCu) and in tetra-auricupride (AuCu) and related intermetallic compounds. Pt-rich minerals are exceedingly rare. Methods and data sources Investigation of mineral parageneses The precious metal mineral grains of the Skaergaard PGE–Au mineralization are generally so small (<5 to ∼100 microns, averages of 15–20 microns; Rudashevsky et al., 2014, 2015) that they easily escape observation under the microscope. To overcome this difficulty, the precious metal parageneses were studied from Hydroseparator® concentrates (Cabri et al., 2005) of sulphide and precious metal grains in monolayer samples (polished thin sections or mounts). A total of 32 samples were investigated (Supplementary Data EA5), and more than 4000 grains were studied using a Camscan Microspec-4DV scanning electron microscope equipped with a Link AN-10, 000 detector (Cabri et al., 2005). The samples included three bulk samples from ‘Toe of Forbindelsesgletscher’ (locality ToF in Fig. 1a; unofficial name) investigated by Skaergaard Minerals Corporation (Cabri, 2004b), and 29 drill core samples from four separate drill cores (Rudashevsky et al., 2014 and 2015, Tables 3–5). The drill cores selected for the study include 90–10 from the western margin of the intrusion, 90–18 from the SW part of the mineralization, 90–24 from the centre of the mineralization (sister core to 90–22 that was sampled by Bernstein & Nielsen (2004), Nielsen et al. (2015), and part of the sample collection used by Keays &Tegner (2016)), and 90–23 A from the eastern margin (see Fig. 1a for locations). Table 3 Identified PGE and Au native elements, minerals, unnamed minerals, intermetallic compounds and alloys of the PGE–Au mineralization of the Skaergaard intrusion Native elements . . Minerals continued . Formula . Unnammed minerals . Alloys . alloys continued . Native Ag Ag Majakite PdNiAs Au3Ag (Au, Ag, Cu) (Pd, Au, Ag, Cu) Native Pd Pd Merenskyite Pd(Te, Pb)2 Au3Cu (Au, Ag, Cu) (Pd, Cu, As, Sb) Native tellurium Te Naldrettite Pd2Sb (Cu, Pd)17S6 (Au, Cu, Ag) (Pd, Cu, Au, Pt) Nielsenite (Pd, Pt, Au)Cu3 (Pd, Au)7 (Ni, Cu)11Pb2 (Au, Cu, Ag) (Pd, Cu, Pb) Minerals Standart formula Palladoarsenide PdAs2 (Pd, Au)3(Cu, Fe)6 S3(Te, Sn)2 (Au, Ag) (Pd, Cu, Pb) Acantite Ag2S Palarstanide Pd5(As, Sn)2 AuPdCu2 (Au, Ag, Cu, Pd) (Pd, Cu, Sn) Arsenopalladinite Pd8As2·5Sb0·5 Polybasite (Ag, Cu)16Sb2S11 (Pd, Ag)2Te (Au, Cu) (Pd, Ge, Cu) Atokite Pd3Sn Skaergaardite PdCu Pd3Ag2S (Au, Cu, Fe) (Pd, Pb, Cu) Auricupride Au3Cu Sopcheite Ag4Pd3Te4 Pd3S (Au, Cu, Pd) (Pd, Pb, Cu, Sn, Te) Bogdanovite (Au, Te, Pb)3 (Cu, Fe)· Sperrylite PtAs2 (Pt, Pd)Cu3 (Au, Cu, Pd, Ag) (Pd, Pt, Cu) Braggite (Pt, Pd)S Stefanite Ag5SbS4 (Au, Pd)3(Cu, Fe) (Au, Pd, Cu, Pt, Ag) (Pd, Sn, Cu) Cabriite Pd2SnCu Telargpalite (Pd, Ag)3+xTe (Cu, Pt, Pd)3S3 (Au·Pd)Cu (Pd, Te, As) Electrum (Au, Ag) Telluropalladinite Pd9Te4 (Pd, Ag, Cu)5S (AuCu, Ag) (Pt, Cr, Pd) Froodite PdBi2 Tetra auricupride AuCu Pd2(Cu, Fe)TeBi (Cu, Au) (Pt, Cu, Fe) Vincentite Pd3As· Tetraferriplatinum PtFe (Au, Pd)3(Cu, Fe) (Au, Cu, Pd) (Pt, Cu, Fe, Pd) Hessite Ag2Te Tulameenite PtFe0·5Cu0·5 (Pd, Hg, Ag)2+xS (Cu, Au, Ni, Zn) (Pt, Fe, Cu) Hongshite PtCu Vasilite Pd16S7 Pd3Ag2S (Cu, Pd, Pt) (Pt, Fe, Pd) Isomertiete Pd11Sb2As2 Vincentite Pd3As (Au, Pd)3(Cu, Fe) (Cu, Pt, Pd) (Pt, Fe, Pd, Cu) Keithconnite Pd3, xTe (0·14<x<0·43) Vysotskite PdS (Pd, Ag, Cd, Cu, Tl)4S (Cu, Pt, Pd) (Pt, Pd, As) Kotulskite PdTe Zvyagintsevite Pd3Pb (Pd, Ag, Cu)6S (Pd, Ag, Cu) (Pt, Pd, Fe) Pd11As2Sn2 Native elements . . Minerals continued . Formula . Unnammed minerals . Alloys . alloys continued . Native Ag Ag Majakite PdNiAs Au3Ag (Au, Ag, Cu) (Pd, Au, Ag, Cu) Native Pd Pd Merenskyite Pd(Te, Pb)2 Au3Cu (Au, Ag, Cu) (Pd, Cu, As, Sb) Native tellurium Te Naldrettite Pd2Sb (Cu, Pd)17S6 (Au, Cu, Ag) (Pd, Cu, Au, Pt) Nielsenite (Pd, Pt, Au)Cu3 (Pd, Au)7 (Ni, Cu)11Pb2 (Au, Cu, Ag) (Pd, Cu, Pb) Minerals Standart formula Palladoarsenide PdAs2 (Pd, Au)3(Cu, Fe)6 S3(Te, Sn)2 (Au, Ag) (Pd, Cu, Pb) Acantite Ag2S Palarstanide Pd5(As, Sn)2 AuPdCu2 (Au, Ag, Cu, Pd) (Pd, Cu, Sn) Arsenopalladinite Pd8As2·5Sb0·5 Polybasite (Ag, Cu)16Sb2S11 (Pd, Ag)2Te (Au, Cu) (Pd, Ge, Cu) Atokite Pd3Sn Skaergaardite PdCu Pd3Ag2S (Au, Cu, Fe) (Pd, Pb, Cu) Auricupride Au3Cu Sopcheite Ag4Pd3Te4 Pd3S (Au, Cu, Pd) (Pd, Pb, Cu, Sn, Te) Bogdanovite (Au, Te, Pb)3 (Cu, Fe)· Sperrylite PtAs2 (Pt, Pd)Cu3 (Au, Cu, Pd, Ag) (Pd, Pt, Cu) Braggite (Pt, Pd)S Stefanite Ag5SbS4 (Au, Pd)3(Cu, Fe) (Au, Pd, Cu, Pt, Ag) (Pd, Sn, Cu) Cabriite Pd2SnCu Telargpalite (Pd, Ag)3+xTe (Cu, Pt, Pd)3S3 (Au·Pd)Cu (Pd, Te, As) Electrum (Au, Ag) Telluropalladinite Pd9Te4 (Pd, Ag, Cu)5S (AuCu, Ag) (Pt, Cr, Pd) Froodite PdBi2 Tetra auricupride AuCu Pd2(Cu, Fe)TeBi (Cu, Au) (Pt, Cu, Fe) Vincentite Pd3As· Tetraferriplatinum PtFe (Au, Pd)3(Cu, Fe) (Au, Cu, Pd) (Pt, Cu, Fe, Pd) Hessite Ag2Te Tulameenite PtFe0·5Cu0·5 (Pd, Hg, Ag)2+xS (Cu, Au, Ni, Zn) (Pt, Fe, Cu) Hongshite PtCu Vasilite Pd16S7 Pd3Ag2S (Cu, Pd, Pt) (Pt, Fe, Pd) Isomertiete Pd11Sb2As2 Vincentite Pd3As (Au, Pd)3(Cu, Fe) (Cu, Pt, Pd) (Pt, Fe, Pd, Cu) Keithconnite Pd3, xTe (0·14<x<0·43) Vysotskite PdS (Pd, Ag, Cd, Cu, Tl)4S (Cu, Pt, Pd) (Pt, Pd, As) Kotulskite PdTe Zvyagintsevite Pd3Pb (Pd, Ag, Cu)6S (Pd, Ag, Cu) (Pt, Pd, Fe) Pd11As2Sn2 Alloys are identified by their dominant elements in decreasing order. Sources: Bird et al., 1991; Nielsen et al., 2003a, 2003b, 2003c, 2003d, 2003e, 2005, 2015; Cabri, 2004b; Rudashevsky & Rudashevsky, 2005a, 2005b, 2006a, 2006b; Rudashevsky et al., 2009a, 2010a, 2010b, 2010c, 2010d, 2012a, 2012b, 2012c, 2012d, 2012e, 2012f, 2012g, 2012h, 2012i, 2007, 2009, 2014, 2015; Holwell et al., 2015, 2016. Open in new tab Table 3 Identified PGE and Au native elements, minerals, unnamed minerals, intermetallic compounds and alloys of the PGE–Au mineralization of the Skaergaard intrusion Native elements . . Minerals continued . Formula . Unnammed minerals . Alloys . alloys continued . Native Ag Ag Majakite PdNiAs Au3Ag (Au, Ag, Cu) (Pd, Au, Ag, Cu) Native Pd Pd Merenskyite Pd(Te, Pb)2 Au3Cu (Au, Ag, Cu) (Pd, Cu, As, Sb) Native tellurium Te Naldrettite Pd2Sb (Cu, Pd)17S6 (Au, Cu, Ag) (Pd, Cu, Au, Pt) Nielsenite (Pd, Pt, Au)Cu3 (Pd, Au)7 (Ni, Cu)11Pb2 (Au, Cu, Ag) (Pd, Cu, Pb) Minerals Standart formula Palladoarsenide PdAs2 (Pd, Au)3(Cu, Fe)6 S3(Te, Sn)2 (Au, Ag) (Pd, Cu, Pb) Acantite Ag2S Palarstanide Pd5(As, Sn)2 AuPdCu2 (Au, Ag, Cu, Pd) (Pd, Cu, Sn) Arsenopalladinite Pd8As2·5Sb0·5 Polybasite (Ag, Cu)16Sb2S11 (Pd, Ag)2Te (Au, Cu) (Pd, Ge, Cu) Atokite Pd3Sn Skaergaardite PdCu Pd3Ag2S (Au, Cu, Fe) (Pd, Pb, Cu) Auricupride Au3Cu Sopcheite Ag4Pd3Te4 Pd3S (Au, Cu, Pd) (Pd, Pb, Cu, Sn, Te) Bogdanovite (Au, Te, Pb)3 (Cu, Fe)· Sperrylite PtAs2 (Pt, Pd)Cu3 (Au, Cu, Pd, Ag) (Pd, Pt, Cu) Braggite (Pt, Pd)S Stefanite Ag5SbS4 (Au, Pd)3(Cu, Fe) (Au, Pd, Cu, Pt, Ag) (Pd, Sn, Cu) Cabriite Pd2SnCu Telargpalite (Pd, Ag)3+xTe (Cu, Pt, Pd)3S3 (Au·Pd)Cu (Pd, Te, As) Electrum (Au, Ag) Telluropalladinite Pd9Te4 (Pd, Ag, Cu)5S (AuCu, Ag) (Pt, Cr, Pd) Froodite PdBi2 Tetra auricupride AuCu Pd2(Cu, Fe)TeBi (Cu, Au) (Pt, Cu, Fe) Vincentite Pd3As· Tetraferriplatinum PtFe (Au, Pd)3(Cu, Fe) (Au, Cu, Pd) (Pt, Cu, Fe, Pd) Hessite Ag2Te Tulameenite PtFe0·5Cu0·5 (Pd, Hg, Ag)2+xS (Cu, Au, Ni, Zn) (Pt, Fe, Cu) Hongshite PtCu Vasilite Pd16S7 Pd3Ag2S (Cu, Pd, Pt) (Pt, Fe, Pd) Isomertiete Pd11Sb2As2 Vincentite Pd3As (Au, Pd)3(Cu, Fe) (Cu, Pt, Pd) (Pt, Fe, Pd, Cu) Keithconnite Pd3, xTe (0·14<x<0·43) Vysotskite PdS (Pd, Ag, Cd, Cu, Tl)4S (Cu, Pt, Pd) (Pt, Pd, As) Kotulskite PdTe Zvyagintsevite Pd3Pb (Pd, Ag, Cu)6S (Pd, Ag, Cu) (Pt, Pd, Fe) Pd11As2Sn2 Native elements . . Minerals continued . Formula . Unnammed minerals . Alloys . alloys continued . Native Ag Ag Majakite PdNiAs Au3Ag (Au, Ag, Cu) (Pd, Au, Ag, Cu) Native Pd Pd Merenskyite Pd(Te, Pb)2 Au3Cu (Au, Ag, Cu) (Pd, Cu, As, Sb) Native tellurium Te Naldrettite Pd2Sb (Cu, Pd)17S6 (Au, Cu, Ag) (Pd, Cu, Au, Pt) Nielsenite (Pd, Pt, Au)Cu3 (Pd, Au)7 (Ni, Cu)11Pb2 (Au, Cu, Ag) (Pd, Cu, Pb) Minerals Standart formula Palladoarsenide PdAs2 (Pd, Au)3(Cu, Fe)6 S3(Te, Sn)2 (Au, Ag) (Pd, Cu, Pb) Acantite Ag2S Palarstanide Pd5(As, Sn)2 AuPdCu2 (Au, Ag, Cu, Pd) (Pd, Cu, Sn) Arsenopalladinite Pd8As2·5Sb0·5 Polybasite (Ag, Cu)16Sb2S11 (Pd, Ag)2Te (Au, Cu) (Pd, Ge, Cu) Atokite Pd3Sn Skaergaardite PdCu Pd3Ag2S (Au, Cu, Fe) (Pd, Pb, Cu) Auricupride Au3Cu Sopcheite Ag4Pd3Te4 Pd3S (Au, Cu, Pd) (Pd, Pb, Cu, Sn, Te) Bogdanovite (Au, Te, Pb)3 (Cu, Fe)· Sperrylite PtAs2 (Pt, Pd)Cu3 (Au, Cu, Pd, Ag) (Pd, Pt, Cu) Braggite (Pt, Pd)S Stefanite Ag5SbS4 (Au, Pd)3(Cu, Fe) (Au, Pd, Cu, Pt, Ag) (Pd, Sn, Cu) Cabriite Pd2SnCu Telargpalite (Pd, Ag)3+xTe (Cu, Pt, Pd)3S3 (Au·Pd)Cu (Pd, Te, As) Electrum (Au, Ag) Telluropalladinite Pd9Te4 (Pd, Ag, Cu)5S (AuCu, Ag) (Pt, Cr, Pd) Froodite PdBi2 Tetra auricupride AuCu Pd2(Cu, Fe)TeBi (Cu, Au) (Pt, Cu, Fe) Vincentite Pd3As· Tetraferriplatinum PtFe (Au, Pd)3(Cu, Fe) (Au, Cu, Pd) (Pt, Cu, Fe, Pd) Hessite Ag2Te Tulameenite PtFe0·5Cu0·5 (Pd, Hg, Ag)2+xS (Cu, Au, Ni, Zn) (Pt, Fe, Cu) Hongshite PtCu Vasilite Pd16S7 Pd3Ag2S (Cu, Pd, Pt) (Pt, Fe, Pd) Isomertiete Pd11Sb2As2 Vincentite Pd3As (Au, Pd)3(Cu, Fe) (Cu, Pt, Pd) (Pt, Fe, Pd, Cu) Keithconnite Pd3, xTe (0·14<x<0·43) Vysotskite PdS (Pd, Ag, Cd, Cu, Tl)4S (Cu, Pt, Pd) (Pt, Pd, As) Kotulskite PdTe Zvyagintsevite Pd3Pb (Pd, Ag, Cu)6S (Pd, Ag, Cu) (Pt, Pd, Fe) Pd11As2Sn2 Alloys are identified by their dominant elements in decreasing order. Sources: Bird et al., 1991; Nielsen et al., 2003a, 2003b, 2003c, 2003d, 2003e, 2005, 2015; Cabri, 2004b; Rudashevsky & Rudashevsky, 2005a, 2005b, 2006a, 2006b; Rudashevsky et al., 2009a, 2010a, 2010b, 2010c, 2010d, 2012a, 2012b, 2012c, 2012d, 2012e, 2012f, 2012g, 2012h, 2012i, 2007, 2009, 2014, 2015; Holwell et al., 2015, 2016. Open in new tab Table 4 Summary of precious metal mineralogy of the Skaergaard PGE–Au Mineralization in volume % of the precious metal paragenesis in individual samples . West Margin . South centre (SW core) . Centre . Half way to Centre . East margin . Drill core . Drill core . Drill core . Drill core . Drill core . Toe of . Drill core . 90-10 . 90-18 . 90-18 . 90-24 . 90-24 . Forbindelses . 90-23A . Upper . Lower . Upper . Lower . Upper . gletscher . Upper . mineralization . mineralization . mineralization . mineralization . mineralization . Upper . mineralization . level . level . level . level . level . mineralization levels . level . Total number of grains 208 932 1623 1092 626 ML-2·2 (Au+1) Number of grains 13 0 0 13 0 0 0 0 Au and Ag minerals and phases above no data 93·8* no data above above above PGE with Sn, Bi, Pb, Te, Se, Sb above no data – no data above above above Arsenides above no data 6·2* no data above above above Sulfides above no data – no data above above above Intermetallic alloys above no data – no data above above above ML‒2·1 (Pd1 and Pd1/Au) Number of grains 585 0 0 206 228 41 0 0 Au and Ag minerals and phases above no data 23·4 36·3 98·5 above above PGE with Sn, Bi, Pb, Te, Se, Sb above no data 0·9 1·1 0·2 above above Arsenides above no data 0·0 0·0 0·0 above above Sulfides above no data 0·0 0·6 1·4 above above Intermetallic alloys above no data 75·6 61·9 0·0 above above ML‒2 (Pd2a and b) Number of grains 438 0 152 144 47 95 0 0 Au and Ag minerals and phases above 3·8 15·0 48·2 8·8 above above PGE with Sn, Bi, Pb, Te, Se, Sb above 1·0 4·7 1·0 0·7 above above Arsenides above 0·0 6·5 0·0 0·0 above above Sulfides above 0·2 67·9 2 15·0 above above Intermetallic alloys above 94·9 5·0 48·8 75·6 above above ML‒1·2 (Pd3a and b) Number of grains 866 37 270 63 496 Au and Ag minerals and phases above no data 17·3 0·9 1·9 84·7 above PGE with Sn, Bi, Pb, Te, Se, Sb above no data 3·6 1·0 0·6 0·4 above Arsenides above no data – – 0·0 0·0 above Sulfides above no data 71·3 8·2 0·0 0·0 above Intermetallic alloys above no data 8·4 89·8 97·5 14·9 above ML‒1·1 (Pd4a and b) Above level Number of grains 482 89 25 78 122 179 Au and Ag minerals and phases 92·7 no data 6·1 0·7 1·2 no data 94·8 PGE with Sn, Bi, Pb, Te, Se, Sb 0·9 no data 0·4 0·7 0·8 no data 0·4 Arsenides 5·5 no data no data – 0·0 no data 4·8 Sulfides 1·0 no data 41·1 0·6 1·0 no data 0·0 Intermetallic alloys 0·0 no data 46·3 98·4 96·9 no data 0·0 ML‒0 (Pd5 and Pd6) Number of grains 2196 119 368 33 666 596 447 Au and Ag minerals and phases 10·9 no data 2·2 0·0 1·2 1·0 15·5 PGE with Sn, Bi, Pb, Te, Se, Sb 6·7 no data 1·8 0·6 1·2 0·4 53·4 Arsenides 28·6 no data 0·1 0·0 0·9 0·8 24·7 Sulfides 53·6 no data 18·2 0·1 0·0 3·0 4·3 Intermetallic alloys 0·4 no data 78·0 99·3 96·4 94·8 1·5 . West Margin . South centre (SW core) . Centre . Half way to Centre . East margin . Drill core . Drill core . Drill core . Drill core . Drill core . Toe of . Drill core . 90-10 . 90-18 . 90-18 . 90-24 . 90-24 . Forbindelses . 90-23A . Upper . Lower . Upper . Lower . Upper . gletscher . Upper . mineralization . mineralization . mineralization . mineralization . mineralization . Upper . mineralization . level . level . level . level . level . mineralization levels . level . Total number of grains 208 932 1623 1092 626 ML-2·2 (Au+1) Number of grains 13 0 0 13 0 0 0 0 Au and Ag minerals and phases above no data 93·8* no data above above above PGE with Sn, Bi, Pb, Te, Se, Sb above no data – no data above above above Arsenides above no data 6·2* no data above above above Sulfides above no data – no data above above above Intermetallic alloys above no data – no data above above above ML‒2·1 (Pd1 and Pd1/Au) Number of grains 585 0 0 206 228 41 0 0 Au and Ag minerals and phases above no data 23·4 36·3 98·5 above above PGE with Sn, Bi, Pb, Te, Se, Sb above no data 0·9 1·1 0·2 above above Arsenides above no data 0·0 0·0 0·0 above above Sulfides above no data 0·0 0·6 1·4 above above Intermetallic alloys above no data 75·6 61·9 0·0 above above ML‒2 (Pd2a and b) Number of grains 438 0 152 144 47 95 0 0 Au and Ag minerals and phases above 3·8 15·0 48·2 8·8 above above PGE with Sn, Bi, Pb, Te, Se, Sb above 1·0 4·7 1·0 0·7 above above Arsenides above 0·0 6·5 0·0 0·0 above above Sulfides above 0·2 67·9 2 15·0 above above Intermetallic alloys above 94·9 5·0 48·8 75·6 above above ML‒1·2 (Pd3a and b) Number of grains 866 37 270 63 496 Au and Ag minerals and phases above no data 17·3 0·9 1·9 84·7 above PGE with Sn, Bi, Pb, Te, Se, Sb above no data 3·6 1·0 0·6 0·4 above Arsenides above no data – – 0·0 0·0 above Sulfides above no data 71·3 8·2 0·0 0·0 above Intermetallic alloys above no data 8·4 89·8 97·5 14·9 above ML‒1·1 (Pd4a and b) Above level Number of grains 482 89 25 78 122 179 Au and Ag minerals and phases 92·7 no data 6·1 0·7 1·2 no data 94·8 PGE with Sn, Bi, Pb, Te, Se, Sb 0·9 no data 0·4 0·7 0·8 no data 0·4 Arsenides 5·5 no data no data – 0·0 no data 4·8 Sulfides 1·0 no data 41·1 0·6 1·0 no data 0·0 Intermetallic alloys 0·0 no data 46·3 98·4 96·9 no data 0·0 ML‒0 (Pd5 and Pd6) Number of grains 2196 119 368 33 666 596 447 Au and Ag minerals and phases 10·9 no data 2·2 0·0 1·2 1·0 15·5 PGE with Sn, Bi, Pb, Te, Se, Sb 6·7 no data 1·8 0·6 1·2 0·4 53·4 Arsenides 28·6 no data 0·1 0·0 0·9 0·8 24·7 Sulfides 53·6 no data 18·2 0·1 0·0 3·0 4·3 Intermetallic alloys 0·4 no data 78·0 99·3 96·4 94·8 1·5 Comparisons and more detailed data in EA5-EA12. * Sample 90–22 977. Open in new tab Table 4 Summary of precious metal mineralogy of the Skaergaard PGE–Au Mineralization in volume % of the precious metal paragenesis in individual samples . West Margin . South centre (SW core) . Centre . Half way to Centre . East margin . Drill core . Drill core . Drill core . Drill core . Drill core . Toe of . Drill core . 90-10 . 90-18 . 90-18 . 90-24 . 90-24 . Forbindelses . 90-23A . Upper . Lower . Upper . Lower . Upper . gletscher . Upper . mineralization . mineralization . mineralization . mineralization . mineralization . Upper . mineralization . level . level . level . level . level . mineralization levels . level . Total number of grains 208 932 1623 1092 626 ML-2·2 (Au+1) Number of grains 13 0 0 13 0 0 0 0 Au and Ag minerals and phases above no data 93·8* no data above above above PGE with Sn, Bi, Pb, Te, Se, Sb above no data – no data above above above Arsenides above no data 6·2* no data above above above Sulfides above no data – no data above above above Intermetallic alloys above no data – no data above above above ML‒2·1 (Pd1 and Pd1/Au) Number of grains 585 0 0 206 228 41 0 0 Au and Ag minerals and phases above no data 23·4 36·3 98·5 above above PGE with Sn, Bi, Pb, Te, Se, Sb above no data 0·9 1·1 0·2 above above Arsenides above no data 0·0 0·0 0·0 above above Sulfides above no data 0·0 0·6 1·4 above above Intermetallic alloys above no data 75·6 61·9 0·0 above above ML‒2 (Pd2a and b) Number of grains 438 0 152 144 47 95 0 0 Au and Ag minerals and phases above 3·8 15·0 48·2 8·8 above above PGE with Sn, Bi, Pb, Te, Se, Sb above 1·0 4·7 1·0 0·7 above above Arsenides above 0·0 6·5 0·0 0·0 above above Sulfides above 0·2 67·9 2 15·0 above above Intermetallic alloys above 94·9 5·0 48·8 75·6 above above ML‒1·2 (Pd3a and b) Number of grains 866 37 270 63 496 Au and Ag minerals and phases above no data 17·3 0·9 1·9 84·7 above PGE with Sn, Bi, Pb, Te, Se, Sb above no data 3·6 1·0 0·6 0·4 above Arsenides above no data – – 0·0 0·0 above Sulfides above no data 71·3 8·2 0·0 0·0 above Intermetallic alloys above no data 8·4 89·8 97·5 14·9 above ML‒1·1 (Pd4a and b) Above level Number of grains 482 89 25 78 122 179 Au and Ag minerals and phases 92·7 no data 6·1 0·7 1·2 no data 94·8 PGE with Sn, Bi, Pb, Te, Se, Sb 0·9 no data 0·4 0·7 0·8 no data 0·4 Arsenides 5·5 no data no data – 0·0 no data 4·8 Sulfides 1·0 no data 41·1 0·6 1·0 no data 0·0 Intermetallic alloys 0·0 no data 46·3 98·4 96·9 no data 0·0 ML‒0 (Pd5 and Pd6) Number of grains 2196 119 368 33 666 596 447 Au and Ag minerals and phases 10·9 no data 2·2 0·0 1·2 1·0 15·5 PGE with Sn, Bi, Pb, Te, Se, Sb 6·7 no data 1·8 0·6 1·2 0·4 53·4 Arsenides 28·6 no data 0·1 0·0 0·9 0·8 24·7 Sulfides 53·6 no data 18·2 0·1 0·0 3·0 4·3 Intermetallic alloys 0·4 no data 78·0 99·3 96·4 94·8 1·5 . West Margin . South centre (SW core) . Centre . Half way to Centre . East margin . Drill core . Drill core . Drill core . Drill core . Drill core . Toe of . Drill core . 90-10 . 90-18 . 90-18 . 90-24 . 90-24 . Forbindelses . 90-23A . Upper . Lower . Upper . Lower . Upper . gletscher . Upper . mineralization . mineralization . mineralization . mineralization . mineralization . Upper . mineralization . level . level . level . level . level . mineralization levels . level . Total number of grains 208 932 1623 1092 626 ML-2·2 (Au+1) Number of grains 13 0 0 13 0 0 0 0 Au and Ag minerals and phases above no data 93·8* no data above above above PGE with Sn, Bi, Pb, Te, Se, Sb above no data – no data above above above Arsenides above no data 6·2* no data above above above Sulfides above no data – no data above above above Intermetallic alloys above no data – no data above above above ML‒2·1 (Pd1 and Pd1/Au) Number of grains 585 0 0 206 228 41 0 0 Au and Ag minerals and phases above no data 23·4 36·3 98·5 above above PGE with Sn, Bi, Pb, Te, Se, Sb above no data 0·9 1·1 0·2 above above Arsenides above no data 0·0 0·0 0·0 above above Sulfides above no data 0·0 0·6 1·4 above above Intermetallic alloys above no data 75·6 61·9 0·0 above above ML‒2 (Pd2a and b) Number of grains 438 0 152 144 47 95 0 0 Au and Ag minerals and phases above 3·8 15·0 48·2 8·8 above above PGE with Sn, Bi, Pb, Te, Se, Sb above 1·0 4·7 1·0 0·7 above above Arsenides above 0·0 6·5 0·0 0·0 above above Sulfides above 0·2 67·9 2 15·0 above above Intermetallic alloys above 94·9 5·0 48·8 75·6 above above ML‒1·2 (Pd3a and b) Number of grains 866 37 270 63 496 Au and Ag minerals and phases above no data 17·3 0·9 1·9 84·7 above PGE with Sn, Bi, Pb, Te, Se, Sb above no data 3·6 1·0 0·6 0·4 above Arsenides above no data – – 0·0 0·0 above Sulfides above no data 71·3 8·2 0·0 0·0 above Intermetallic alloys above no data 8·4 89·8 97·5 14·9 above ML‒1·1 (Pd4a and b) Above level Number of grains 482 89 25 78 122 179 Au and Ag minerals and phases 92·7 no data 6·1 0·7 1·2 no data 94·8 PGE with Sn, Bi, Pb, Te, Se, Sb 0·9 no data 0·4 0·7 0·8 no data 0·4 Arsenides 5·5 no data no data – 0·0 no data 4·8 Sulfides 1·0 no data 41·1 0·6 1·0 no data 0·0 Intermetallic alloys 0·0 no data 46·3 98·4 96·9 no data 0·0 ML‒0 (Pd5 and Pd6) Number of grains 2196 119 368 33 666 596 447 Au and Ag minerals and phases 10·9 no data 2·2 0·0 1·2 1·0 15·5 PGE with Sn, Bi, Pb, Te, Se, Sb 6·7 no data 1·8 0·6 1·2 0·4 53·4 Arsenides 28·6 no data 0·1 0·0 0·9 0·8 24·7 Sulfides 53·6 no data 18·2 0·1 0·0 3·0 4·3 Intermetallic alloys 0·4 no data 78·0 99·3 96·4 94·8 1·5 Comparisons and more detailed data in EA5-EA12. * Sample 90–22 977. Open in new tab Table 5 Stratigraphic variation in relative importance of precious metal mineral groups in Pd5 in drill cores 90–23 A and 90-10 from near the E and W margins, respectively. Each sample collected at a 1-metre interval in the drill core Mineral . General formula . 90-23A samples . 90-10 samples . . . 808 . 807 . 806 . avg. . 445 . 443 . avg. . Au, Ag and Pt phases 1·9 8 36·3 15·4 0 21·7 10·8 Au-phases 1·9 8 28·3 12·7 0 21·7 10·8 (Au, Cu, Pd, Ag) alloys (Au, Cu, Pd, Ag, Pt) 0·7 8 23 10·6 0 17·3 8·65 Bogdanovite (Au, Pd, Pt)3(Cu, Fe) 1·2 n.d. 5 2·1 0 0 0 Tetra-auricupride (Au, Pd, Pt)(Cu, Fe) n.d. n.d. 0·3 0·1 0 4·4 2·2 Ag-phases 0 0 8 2·7 0 0 0 Native silver (Ag) n.d. 0 n.d. 0 0 0 0 Polybasite [(Ag, Cu)6(Sb, As)2S7][Ag9CuS4] n.d. 0 n.d. 0 0 0 0 Stephanite Ag5SbS4 n.d. n.d. 8 2·7 0 0 0 Others Plumbides, tellu tellurides and stannides 74·2 60 26·6 53·6 0 5·8 2·9 Kotulskite Pd(Te, Pb) n.d. n.d. 0·3 0·1 0 0 0 Keithconnite Pd3-x(Te, Pb, As, Sn) 0·5 2 2 1·5 0 4 0·2 Zvyagintsevite (Pd, Au, Pt)3(Pb, Sn) 73 58 24 51·7 0 1·8 0·9 Atokite (Pd, Pt)3(Sn, As, Te) n.d. 0·3 0·1 0 0 0 Arsenides 11·7 25 36·3 24·7 34·8 29·8 32·3 Arsenopalladinite (Pd, Cu, Au)8(As, Sn)3 3 11 17 10·3 0 0 0 Vincentite (Pd, Cu, Au)3(As, Sn, Te, Sb) 8 12 8 9·3 30·3 24·8 27·6 Palladoarsenide (Pd, Cu)2As 0·7 2 2 1·6 0 0 0 Isomertieite Pd11As2(Sb, Te)2 0·7 n.d. 0·3 0·3 0 0 0 Sperrylite PtAs2 0·5 0 9 3·2 0 0 0 Atokite (Pd, Pt, Au)3(Sn, Pb, As) 0 0 0 0 2·6 5 3·8 Unnamed 1·9 0 0·95 Sulphides 10 2 1 4·3 64·5 42·6 53·6 Vysotskite (Pd, Ni)S 4 0 n.d. 1·3 19·1 22·5 20·8 Vasilite Pd12Cu4S7 6 2 1 3 45·4 20·1 32·8 Pd and Pd-alloys 1 3 0 1·3 0·7 0 0·4 Native palladium (Pd) n.d. 3 n.d. 1 0 0 0 Skaergaardite PdCu 1 n.d. n.d. 0·3 0·7 0 0·4 Pt-alloy (Pt, Pd, Fe, Cu) 0·7 0 n.d. 0·2 0 0 0 Mineral . General formula . 90-23A samples . 90-10 samples . . . 808 . 807 . 806 . avg. . 445 . 443 . avg. . Au, Ag and Pt phases 1·9 8 36·3 15·4 0 21·7 10·8 Au-phases 1·9 8 28·3 12·7 0 21·7 10·8 (Au, Cu, Pd, Ag) alloys (Au, Cu, Pd, Ag, Pt) 0·7 8 23 10·6 0 17·3 8·65 Bogdanovite (Au, Pd, Pt)3(Cu, Fe) 1·2 n.d. 5 2·1 0 0 0 Tetra-auricupride (Au, Pd, Pt)(Cu, Fe) n.d. n.d. 0·3 0·1 0 4·4 2·2 Ag-phases 0 0 8 2·7 0 0 0 Native silver (Ag) n.d. 0 n.d. 0 0 0 0 Polybasite [(Ag, Cu)6(Sb, As)2S7][Ag9CuS4] n.d. 0 n.d. 0 0 0 0 Stephanite Ag5SbS4 n.d. n.d. 8 2·7 0 0 0 Others Plumbides, tellu tellurides and stannides 74·2 60 26·6 53·6 0 5·8 2·9 Kotulskite Pd(Te, Pb) n.d. n.d. 0·3 0·1 0 0 0 Keithconnite Pd3-x(Te, Pb, As, Sn) 0·5 2 2 1·5 0 4 0·2 Zvyagintsevite (Pd, Au, Pt)3(Pb, Sn) 73 58 24 51·7 0 1·8 0·9 Atokite (Pd, Pt)3(Sn, As, Te) n.d. 0·3 0·1 0 0 0 Arsenides 11·7 25 36·3 24·7 34·8 29·8 32·3 Arsenopalladinite (Pd, Cu, Au)8(As, Sn)3 3 11 17 10·3 0 0 0 Vincentite (Pd, Cu, Au)3(As, Sn, Te, Sb) 8 12 8 9·3 30·3 24·8 27·6 Palladoarsenide (Pd, Cu)2As 0·7 2 2 1·6 0 0 0 Isomertieite Pd11As2(Sb, Te)2 0·7 n.d. 0·3 0·3 0 0 0 Sperrylite PtAs2 0·5 0 9 3·2 0 0 0 Atokite (Pd, Pt, Au)3(Sn, Pb, As) 0 0 0 0 2·6 5 3·8 Unnamed 1·9 0 0·95 Sulphides 10 2 1 4·3 64·5 42·6 53·6 Vysotskite (Pd, Ni)S 4 0 n.d. 1·3 19·1 22·5 20·8 Vasilite Pd12Cu4S7 6 2 1 3 45·4 20·1 32·8 Pd and Pd-alloys 1 3 0 1·3 0·7 0 0·4 Native palladium (Pd) n.d. 3 n.d. 1 0 0 0 Skaergaardite PdCu 1 n.d. n.d. 0·3 0·7 0 0·4 Pt-alloy (Pt, Pd, Fe, Cu) 0·7 0 n.d. 0·2 0 0 0 n.d., no data. Open in new tab Table 5 Stratigraphic variation in relative importance of precious metal mineral groups in Pd5 in drill cores 90–23 A and 90-10 from near the E and W margins, respectively. Each sample collected at a 1-metre interval in the drill core Mineral . General formula . 90-23A samples . 90-10 samples . . . 808 . 807 . 806 . avg. . 445 . 443 . avg. . Au, Ag and Pt phases 1·9 8 36·3 15·4 0 21·7 10·8 Au-phases 1·9 8 28·3 12·7 0 21·7 10·8 (Au, Cu, Pd, Ag) alloys (Au, Cu, Pd, Ag, Pt) 0·7 8 23 10·6 0 17·3 8·65 Bogdanovite (Au, Pd, Pt)3(Cu, Fe) 1·2 n.d. 5 2·1 0 0 0 Tetra-auricupride (Au, Pd, Pt)(Cu, Fe) n.d. n.d. 0·3 0·1 0 4·4 2·2 Ag-phases 0 0 8 2·7 0 0 0 Native silver (Ag) n.d. 0 n.d. 0 0 0 0 Polybasite [(Ag, Cu)6(Sb, As)2S7][Ag9CuS4] n.d. 0 n.d. 0 0 0 0 Stephanite Ag5SbS4 n.d. n.d. 8 2·7 0 0 0 Others Plumbides, tellu tellurides and stannides 74·2 60 26·6 53·6 0 5·8 2·9 Kotulskite Pd(Te, Pb) n.d. n.d. 0·3 0·1 0 0 0 Keithconnite Pd3-x(Te, Pb, As, Sn) 0·5 2 2 1·5 0 4 0·2 Zvyagintsevite (Pd, Au, Pt)3(Pb, Sn) 73 58 24 51·7 0 1·8 0·9 Atokite (Pd, Pt)3(Sn, As, Te) n.d. 0·3 0·1 0 0 0 Arsenides 11·7 25 36·3 24·7 34·8 29·8 32·3 Arsenopalladinite (Pd, Cu, Au)8(As, Sn)3 3 11 17 10·3 0 0 0 Vincentite (Pd, Cu, Au)3(As, Sn, Te, Sb) 8 12 8 9·3 30·3 24·8 27·6 Palladoarsenide (Pd, Cu)2As 0·7 2 2 1·6 0 0 0 Isomertieite Pd11As2(Sb, Te)2 0·7 n.d. 0·3 0·3 0 0 0 Sperrylite PtAs2 0·5 0 9 3·2 0 0 0 Atokite (Pd, Pt, Au)3(Sn, Pb, As) 0 0 0 0 2·6 5 3·8 Unnamed 1·9 0 0·95 Sulphides 10 2 1 4·3 64·5 42·6 53·6 Vysotskite (Pd, Ni)S 4 0 n.d. 1·3 19·1 22·5 20·8 Vasilite Pd12Cu4S7 6 2 1 3 45·4 20·1 32·8 Pd and Pd-alloys 1 3 0 1·3 0·7 0 0·4 Native palladium (Pd) n.d. 3 n.d. 1 0 0 0 Skaergaardite PdCu 1 n.d. n.d. 0·3 0·7 0 0·4 Pt-alloy (Pt, Pd, Fe, Cu) 0·7 0 n.d. 0·2 0 0 0 Mineral . General formula . 90-23A samples . 90-10 samples . . . 808 . 807 . 806 . avg. . 445 . 443 . avg. . Au, Ag and Pt phases 1·9 8 36·3 15·4 0 21·7 10·8 Au-phases 1·9 8 28·3 12·7 0 21·7 10·8 (Au, Cu, Pd, Ag) alloys (Au, Cu, Pd, Ag, Pt) 0·7 8 23 10·6 0 17·3 8·65 Bogdanovite (Au, Pd, Pt)3(Cu, Fe) 1·2 n.d. 5 2·1 0 0 0 Tetra-auricupride (Au, Pd, Pt)(Cu, Fe) n.d. n.d. 0·3 0·1 0 4·4 2·2 Ag-phases 0 0 8 2·7 0 0 0 Native silver (Ag) n.d. 0 n.d. 0 0 0 0 Polybasite [(Ag, Cu)6(Sb, As)2S7][Ag9CuS4] n.d. 0 n.d. 0 0 0 0 Stephanite Ag5SbS4 n.d. n.d. 8 2·7 0 0 0 Others Plumbides, tellu tellurides and stannides 74·2 60 26·6 53·6 0 5·8 2·9 Kotulskite Pd(Te, Pb) n.d. n.d. 0·3 0·1 0 0 0 Keithconnite Pd3-x(Te, Pb, As, Sn) 0·5 2 2 1·5 0 4 0·2 Zvyagintsevite (Pd, Au, Pt)3(Pb, Sn) 73 58 24 51·7 0 1·8 0·9 Atokite (Pd, Pt)3(Sn, As, Te) n.d. 0·3 0·1 0 0 0 Arsenides 11·7 25 36·3 24·7 34·8 29·8 32·3 Arsenopalladinite (Pd, Cu, Au)8(As, Sn)3 3 11 17 10·3 0 0 0 Vincentite (Pd, Cu, Au)3(As, Sn, Te, Sb) 8 12 8 9·3 30·3 24·8 27·6 Palladoarsenide (Pd, Cu)2As 0·7 2 2 1·6 0 0 0 Isomertieite Pd11As2(Sb, Te)2 0·7 n.d. 0·3 0·3 0 0 0 Sperrylite PtAs2 0·5 0 9 3·2 0 0 0 Atokite (Pd, Pt, Au)3(Sn, Pb, As) 0 0 0 0 2·6 5 3·8 Unnamed 1·9 0 0·95 Sulphides 10 2 1 4·3 64·5 42·6 53·6 Vysotskite (Pd, Ni)S 4 0 n.d. 1·3 19·1 22·5 20·8 Vasilite Pd12Cu4S7 6 2 1 3 45·4 20·1 32·8 Pd and Pd-alloys 1 3 0 1·3 0·7 0 0·4 Native palladium (Pd) n.d. 3 n.d. 1 0 0 0 Skaergaardite PdCu 1 n.d. n.d. 0·3 0·7 0 0·4 Pt-alloy (Pt, Pd, Fe, Cu) 0·7 0 n.d. 0·2 0 0 0 n.d., no data. Open in new tab Whereas the concentrates of precious metal grains in the bulk samples were obtained by performing the Hydroseparator® technique on a split and sieved fraction of 1-tonne samples, the samples from drill cores represent all of the precious metal grains that were retrieved from a 1 metre section of drill core. For each drill core sample, this amounted to between 600 and 1200 g of gabbro, which varied due to lithology and previous use of the core for exploration purposes. In samples with the highest precious metal grades (>3 ppm) we commonly retrieved several hundred precious metal mineral grains from a single 1-metre sample. Information collected from the samples includes backscatter images, grain size, paragenetic information, and for each individual grain, the composition and volume, which was estimated from the size of a circle that enclosed the grain (a method comparable to that of Holwell et al., 2016). The primary mineralogical data for the present compilations are available in reports (Nielsen et al., 2003a, 2003b, 2003c, 2003d, 2003e; Rudashevsky & Rudashevsky, 2005a, 2005b, 2006a, 2006b; Rudashevsky et al., 2009a, 2009b, 2010a, 2010b, 2010c, 2010d, 2012a, 2012b, 2012c, 2012d, 2012e, 2012f, 2012g, 2012h, 2012i) on the ‘Greenland Portal’, at www.greenmin.gl (official web page of Greenland Minerals Authority). Compiled tables with lists of the precious metal minerals and phases in mineralization levels (Pd6 to Pd1/Au and Au + 1) in drill cores and bulk samples across the intrusion (E–W section) and the volumetric proportion of the minerals and phases in MLs in the mineralized intervals are provided in Supplementary Data EA5–EA12. In this contribution, we have summed up the relative volumes for: (a) PGE-rich alloys, (b) PGE-rich sulphides, (c) PGE-rich arsenides, (d) other PGE-rich minerals and phases with Sn, Bi, Pb, Te, Se, Sb, a.o., and (e) Au and Ag minerals and phases throughout the mineralization. The tables also report the depth of the Pd5 peak in the drill core from which the sample was taken, the name of the mineralization level for a given sample (see Fig. 3a), the elevation of the sample relative to nearest Pd-peak, the elevation of the sample relative to the Pd5 peak, and the number of grains that were studied from the given sample. In addition, Supplementary Data EA6–EA11 include descriptions of the precious metal parageneses of individual samples. Geochemical sources The geochemical data presented in Table 6 are compiled and refined from assays reported by Platinova A/S (Watts, Griffis & McOuat, 1991), Skaergaard Minerals Corporation (Hanghøj, 2005), Holwell & Keays (2014), and drill core 90–22 (Bernstein & Nielsen, 2004, Supplementary Data EA13). These references for the precious metal assays include information on the analytical methods used, and details of the sampling methods for those data sets. Table 6 Grade times width* for Pd, Pt and Au in the Skaergaard PGE–Au Mineralization Drill core . From metres . To metres . Distance from western margin (see Nielsen et al., 2015) . Distance from centre at at point 666 (m) (see text) . Pd . Pt . Au . Pd+Pt . Pd+Pt+Au . Pd/Pt . Au/Pd . Au/Pt . PGE . Au . % of all In core through deposit . % of all In core through deposit . Pd5, 5 metres of core with highest Pd+Pt † 89-02 201·0 206·0 2490 1559 10·45 0·84 0·31 11·29 11·60 12·51 0·030 0·37 30·1 3·8 89-03 255·0 260·0 1460 1588 8·24 0·81 0·48 9·05 9·53 10·17 0·058 0·59 34·4 6·8 89-04 272·0 277·0 1475 1588 6·43 0·76 0·39 7·20 7·58 8·42 0·060 0·51 30·8 4·6 89-06 159·0 164·0 4500 2735 9·54 0·75 0·36 10·28 10·64 12·80 0·038 0·48 39·1 5·1 89-09 484·0 489·0 1800 1324 7·68 0·73 0·36 8·41 8·77 10·46 0·047 0·50 26·9 4·4 90-10 442·6 447·6 340 2500 11·98 0·64 0·78 12·62 13·40 18·70 0·065 1·21 55·5 19·4 90-11 675·1 680·1 1970 1382 6·97 0·64 0·39 7·61 8·00 10·95 0·056 0·62 23·5 4·3 90-13 469·0 474·0 1300 1706 7·08 0·65 0·33 7·73 8·06 10·84 0·047 0·51 24·9 3·3 90-14 192·9 198·0 400 2559 11·19 0·75 0·63 11·94 12·57 14·89 0·056 0·83 62·2 15·3 90-17A 498·0 503·0 740 2118 8·64 0·75 0·26 9·39 9·65 11·52 0·030 0·35 38·6 4·9 90-18 1009·0 1014·0 1200 1824 6·15 0·76 0·33 6·91 7·24 8·11 0·054 0·44 18·6 3·5 90-19 589·0 594·0 5270 1824 6·57 0·65 0·37 7·22 7·59 10·11 0·056 0·57 24·9 3·9 90-20 975·0 980·0 1860 2588 4·68 0·88 0·33 5·56 5·89 5·33 0·071 0·38 18·3 3·3 90-22 810·0 815·0 2690 1029 9·15 0·89 0·41 10·04 10·45 10·29 0·045 0·46 26·9 3·6 90-23A 806·0 811·0 6700 3441 9·30 0·62 0·61 9·92 10·53 15·00 0·066 0·99 47·5 15·8 90-24 1055·0 1060·0 3350 294 8·80 0·74 0·35 9·54 9·89 11·89 0·040 0·47 28·1 2·7 04-28A 471·0 475·3 5147 2273 8·10 0·82 0·41 8·92 9·33 9·86 0·051 0·50 26·5 4·7 04-30 1168·0 1173·0 1500 2000 5·50 0·73 0·53 6·23 6·76 7·53 0·096 0·73 55·4 14·9 04-31 1170·0 1176·0 2400 1357 8·71 0·65 0·42 9·36 9·79 13·31 0·049 0·65 35·2 4·3 04-32 412·0 417·0 1600 1500 13·61 0·7 0·59 14·31 14·90 19·44 0·043 0·84 21·5 5·4 04·33 411·0 416·0 4559 1714 7·62 0·67 0·45 8·29 8·74 11·37 0·059 0·67 24·9 3·4 04-34 460·0 465·0 5500 3286 9·22 0·76 0·71 9·98 10·69 12·13 0·077 0·93 42·4 5·1 11-53 475·0 480·0 1765 1549 8·37 0·83 0·48 9·20 9·68 10·08 0·057 0·58 27·6 4·8 11-57 114·0 119·0 6200 3803 10·77 0·73 1·09 11·50 12·59 14·75 0·101 1·50 36·6 8·2 LPGEM (Pd6-Pd4): from a lower cut-off of 100 ppb Pd+Pt to Pd+Pt low between Pd4a and Pd3b 89-02 191·0 218·0 2490 1559 22·52 2·29 1·01 24·80 25·81 9·83 0·045 0·44 66·1 12·2 89-03 245·0 267·0 1460 1588 16·31 1·88 1·27 18·19 19·46 8·69 0·078 0·67 69·2 18·0 89-04 258·0 283·0 1475 1588 15·34 1·71 1·59 17·05 18·64 8·97 0·103 0·93 73·1 19·1 89-06 144·0 173·0 4500 2735 22·16 2·14 2·03 24·30 26·32 10·37 0·092 0·95 92·4 28·9 89-09 473·0 495·6 1800 1324 16·68 1·76 0·81 18·44 19·25 9·45 0·049 0·46 59·0 9·8 90-10 429·0 457·6 340 2500 21·07 1·68 4·00 22·75 26·75 12·54 0·190 2·38 100·0 100·0 90-11 663·5 691·5 1970 1382 16·51 1·89 0·96 18·40 19·36 8·75 0·058 0·51 56·7 10·4 90-13 459·0 483·0 1300 1706 16·22 1·83 1·07 18·04 19·12 8·88 0·066 0·59 58·0 10·6 90-14 173·5 208·0 400 2559 17·57 1·63 4·09 19·19 23·28 10·80 0·233 2·51 100·0 100·0 90-17A 484·0 512·9 740 2118 19·50 1·92 1·08 21·42 22·49 10·18 0·055 0·56 88·1 19·9 90-18 998·3 1025·7 1200 1824 15·38 2·13 0·86 17·51 18·37 7·21 0·056 0·40 47·2 8·9 90-19 571·0 603·0 5270 1824 18·33 1·83 1·22 20·17 21·39 10·01 0·067 0·67 69·5 13·1 90-20 962·0 992·0 1860 2588 14·87 2·15 0·80 17·01 17·82 6·92 0·054 0·37 56·1 8·0 90-22 1020·0 1045·0 2690 1029 20·25 2·17 1·37 22·42 23·80 9·33 0·068 0·63 60·0 12·1 90-23A 794·0 822·0 6700 3441 19·12 1·74 3·90 20·87 24·76 10·98 0·204 2·24 100·0 100·0 90-24 1045·0 1070·0 3350 294 19·80 2·02 0·87 21·82 22·69 9·80 0·044 0·43 64·4 6·9 04-28A 461·0 484·0 5147 2273 15·90 1·46 0·90 17·36 18·26 10·91 0·056 0·62 60·2 12·5 04-30 1155·0 1175·9 1500 2000 10·37 1·30 0·90 11·67 12·57 7·95 0·087 0·69 96·7 72·6 04-31 1152·0 1184·5 2400 1357 23·29 2·10 2·02 25·39 27·69 11·06 0·099 1·09 68·5 9·5 04-32 399·0 425·0 1600 1500 26·89 1·50 2·50 28·39 30·89 17·93 0·093 1·67 45·1 9·2 04-33 396·0 425·0 4559 1714 19·35 1·90 1·63 21·25 22·88 10·21 0·084 0·86 67·6 18·4 04-34 450·0 477·0 5500 3286 19·10 2·09 1·70 21·19 22·89 9·14 0·089 0·81 92·9 71·7 11-53 465·0 490·0 1765 1549 18·79 2·11 1·28 20·90 22·18 8·91 0·068 0·61 70·6 17·6 11-57 98·0 127·0 6200 3803 18·20 1·87 5·35 20·07 25·41 9·76 0·294 2·87 77·6 19·6 Pd6 to Pd1/Au: Pd+Pt+Au from a lower to an upper cut-off of 100 ppb Pd+Pt+Au‡ 89-02 163·0 218·0 2490 1559 34·45 3·06 8·25 37·51 45·76 11·26 0·239 2·70 100 100 89-03 230·0 267·0 1460 1588 24·00 2·30 7·03 26·30 33·33 10·42 0·293 3·05 100 100 89-04 251·2 283·0 1475 1588 21·19 2·15 8·31 23·34 31·64 9·85 0·392 3·86 100 100 89-06 138·0 173·0 4500 2735 24·00 2·30 7·03 26·30 33·33 10·42 0·293 3·05 100 100 89-09 444·8 495·6 1800 1324 28·87 2·41 8·04 31·28 39·32 11·97 0·279 3·34 100 100 90-10 429·0 457·6 340 2500 21·07 1·68 4·00 22·75 26·75 12·54 0·190 2·38 100 100 90-11 635·8 691·5 1970 1382 29·77 2·68 9·22 32·45 41·67 11·10 0·310 3·44 100 100 90-13 421·0 483·0 1300 1706 28·39 2·71 10·10 31·10 41·19 10·48 0·356 3·73 100 100 90-14 173·5 208·0 400 2559 17·57 1·63 4·09 19·19 23·28 10·80 0·233 2·51 100 100 90-17A 471·6 512·9 740 2118 22·13 2·17 5·41 24·30 29·71 10·18 0·244 2·49 100 100 90-18 946·0 1025·7 1200 1824 33·82 3·30 9·61 37·12 46·72 10·25 0·284 2·91 100 100 90-19 551·0 603·0 5270 1824 26·62 2·40 9·33 29·02 38·35 11·12 0·351 3·90 100 100 90-20 935·0 992·0 1860 2588 27·40 2·93 10·04 30·32 40·37 9·36 0·367 3·43 100 100 90-22 978·0 1045·0 2690 1029 34·21 3·16 11·34 37·37 48·71 10·82 0·331 3·59 100 100 90-23A 794·0 822·0 6700 3441 19·12 1·74 3·90 20·87 24·76 10·98 0·204 2·24 100 100 90-24 1018·0 1070·0 3350 294 31·15 2·75 12·67 33·89 46·56 11·34 0·407 4·61 100 100 04-28A 436·0 484·0 5147 2273 23·31 2·04 9·44 25·34 34·79 11·45 0·405 4·64 100 100 04-30 1121·0 1175·9 1500 2000 26·72 2·31 9·79 29·03 38·81 11·57 0·366 4·24 100 100 04-31 1110·0 1184·5 2400 1357 34·78 2·80 12·48 37·58 50·06 12·40 0·359 4·45 100 100 04-32 371·0 425·0 1600 1500 32·00 2·02 11·46 34·02 45·48 15·84 0·358 5·67 100 100 04-33 377·0 425·0 4559 1714 27·69 2·40 9·29 30·09 39·38 11·56 0·335 3·88 100 100 04-34 419·0 477·0 5500 3286 24·61 2·68 8·67 27·29 35·96 9·18 0·352 3·24 100 100 11-53 433·0 490·0 1765 1549 31·76 2·95 10·25 34·71 44·96 10·77 0·323 3·47 100 100 11-57 89·0 127·0 6200 3803 18·52 2·22 7·36 20·74 28·10 8·34 0·397 3·32 100 100 Drill core . From metres . To metres . Distance from western margin (see Nielsen et al., 2015) . Distance from centre at at point 666 (m) (see text) . Pd . Pt . Au . Pd+Pt . Pd+Pt+Au . Pd/Pt . Au/Pd . Au/Pt . PGE . Au . % of all In core through deposit . % of all In core through deposit . Pd5, 5 metres of core with highest Pd+Pt † 89-02 201·0 206·0 2490 1559 10·45 0·84 0·31 11·29 11·60 12·51 0·030 0·37 30·1 3·8 89-03 255·0 260·0 1460 1588 8·24 0·81 0·48 9·05 9·53 10·17 0·058 0·59 34·4 6·8 89-04 272·0 277·0 1475 1588 6·43 0·76 0·39 7·20 7·58 8·42 0·060 0·51 30·8 4·6 89-06 159·0 164·0 4500 2735 9·54 0·75 0·36 10·28 10·64 12·80 0·038 0·48 39·1 5·1 89-09 484·0 489·0 1800 1324 7·68 0·73 0·36 8·41 8·77 10·46 0·047 0·50 26·9 4·4 90-10 442·6 447·6 340 2500 11·98 0·64 0·78 12·62 13·40 18·70 0·065 1·21 55·5 19·4 90-11 675·1 680·1 1970 1382 6·97 0·64 0·39 7·61 8·00 10·95 0·056 0·62 23·5 4·3 90-13 469·0 474·0 1300 1706 7·08 0·65 0·33 7·73 8·06 10·84 0·047 0·51 24·9 3·3 90-14 192·9 198·0 400 2559 11·19 0·75 0·63 11·94 12·57 14·89 0·056 0·83 62·2 15·3 90-17A 498·0 503·0 740 2118 8·64 0·75 0·26 9·39 9·65 11·52 0·030 0·35 38·6 4·9 90-18 1009·0 1014·0 1200 1824 6·15 0·76 0·33 6·91 7·24 8·11 0·054 0·44 18·6 3·5 90-19 589·0 594·0 5270 1824 6·57 0·65 0·37 7·22 7·59 10·11 0·056 0·57 24·9 3·9 90-20 975·0 980·0 1860 2588 4·68 0·88 0·33 5·56 5·89 5·33 0·071 0·38 18·3 3·3 90-22 810·0 815·0 2690 1029 9·15 0·89 0·41 10·04 10·45 10·29 0·045 0·46 26·9 3·6 90-23A 806·0 811·0 6700 3441 9·30 0·62 0·61 9·92 10·53 15·00 0·066 0·99 47·5 15·8 90-24 1055·0 1060·0 3350 294 8·80 0·74 0·35 9·54 9·89 11·89 0·040 0·47 28·1 2·7 04-28A 471·0 475·3 5147 2273 8·10 0·82 0·41 8·92 9·33 9·86 0·051 0·50 26·5 4·7 04-30 1168·0 1173·0 1500 2000 5·50 0·73 0·53 6·23 6·76 7·53 0·096 0·73 55·4 14·9 04-31 1170·0 1176·0 2400 1357 8·71 0·65 0·42 9·36 9·79 13·31 0·049 0·65 35·2 4·3 04-32 412·0 417·0 1600 1500 13·61 0·7 0·59 14·31 14·90 19·44 0·043 0·84 21·5 5·4 04·33 411·0 416·0 4559 1714 7·62 0·67 0·45 8·29 8·74 11·37 0·059 0·67 24·9 3·4 04-34 460·0 465·0 5500 3286 9·22 0·76 0·71 9·98 10·69 12·13 0·077 0·93 42·4 5·1 11-53 475·0 480·0 1765 1549 8·37 0·83 0·48 9·20 9·68 10·08 0·057 0·58 27·6 4·8 11-57 114·0 119·0 6200 3803 10·77 0·73 1·09 11·50 12·59 14·75 0·101 1·50 36·6 8·2 LPGEM (Pd6-Pd4): from a lower cut-off of 100 ppb Pd+Pt to Pd+Pt low between Pd4a and Pd3b 89-02 191·0 218·0 2490 1559 22·52 2·29 1·01 24·80 25·81 9·83 0·045 0·44 66·1 12·2 89-03 245·0 267·0 1460 1588 16·31 1·88 1·27 18·19 19·46 8·69 0·078 0·67 69·2 18·0 89-04 258·0 283·0 1475 1588 15·34 1·71 1·59 17·05 18·64 8·97 0·103 0·93 73·1 19·1 89-06 144·0 173·0 4500 2735 22·16 2·14 2·03 24·30 26·32 10·37 0·092 0·95 92·4 28·9 89-09 473·0 495·6 1800 1324 16·68 1·76 0·81 18·44 19·25 9·45 0·049 0·46 59·0 9·8 90-10 429·0 457·6 340 2500 21·07 1·68 4·00 22·75 26·75 12·54 0·190 2·38 100·0 100·0 90-11 663·5 691·5 1970 1382 16·51 1·89 0·96 18·40 19·36 8·75 0·058 0·51 56·7 10·4 90-13 459·0 483·0 1300 1706 16·22 1·83 1·07 18·04 19·12 8·88 0·066 0·59 58·0 10·6 90-14 173·5 208·0 400 2559 17·57 1·63 4·09 19·19 23·28 10·80 0·233 2·51 100·0 100·0 90-17A 484·0 512·9 740 2118 19·50 1·92 1·08 21·42 22·49 10·18 0·055 0·56 88·1 19·9 90-18 998·3 1025·7 1200 1824 15·38 2·13 0·86 17·51 18·37 7·21 0·056 0·40 47·2 8·9 90-19 571·0 603·0 5270 1824 18·33 1·83 1·22 20·17 21·39 10·01 0·067 0·67 69·5 13·1 90-20 962·0 992·0 1860 2588 14·87 2·15 0·80 17·01 17·82 6·92 0·054 0·37 56·1 8·0 90-22 1020·0 1045·0 2690 1029 20·25 2·17 1·37 22·42 23·80 9·33 0·068 0·63 60·0 12·1 90-23A 794·0 822·0 6700 3441 19·12 1·74 3·90 20·87 24·76 10·98 0·204 2·24 100·0 100·0 90-24 1045·0 1070·0 3350 294 19·80 2·02 0·87 21·82 22·69 9·80 0·044 0·43 64·4 6·9 04-28A 461·0 484·0 5147 2273 15·90 1·46 0·90 17·36 18·26 10·91 0·056 0·62 60·2 12·5 04-30 1155·0 1175·9 1500 2000 10·37 1·30 0·90 11·67 12·57 7·95 0·087 0·69 96·7 72·6 04-31 1152·0 1184·5 2400 1357 23·29 2·10 2·02 25·39 27·69 11·06 0·099 1·09 68·5 9·5 04-32 399·0 425·0 1600 1500 26·89 1·50 2·50 28·39 30·89 17·93 0·093 1·67 45·1 9·2 04-33 396·0 425·0 4559 1714 19·35 1·90 1·63 21·25 22·88 10·21 0·084 0·86 67·6 18·4 04-34 450·0 477·0 5500 3286 19·10 2·09 1·70 21·19 22·89 9·14 0·089 0·81 92·9 71·7 11-53 465·0 490·0 1765 1549 18·79 2·11 1·28 20·90 22·18 8·91 0·068 0·61 70·6 17·6 11-57 98·0 127·0 6200 3803 18·20 1·87 5·35 20·07 25·41 9·76 0·294 2·87 77·6 19·6 Pd6 to Pd1/Au: Pd+Pt+Au from a lower to an upper cut-off of 100 ppb Pd+Pt+Au‡ 89-02 163·0 218·0 2490 1559 34·45 3·06 8·25 37·51 45·76 11·26 0·239 2·70 100 100 89-03 230·0 267·0 1460 1588 24·00 2·30 7·03 26·30 33·33 10·42 0·293 3·05 100 100 89-04 251·2 283·0 1475 1588 21·19 2·15 8·31 23·34 31·64 9·85 0·392 3·86 100 100 89-06 138·0 173·0 4500 2735 24·00 2·30 7·03 26·30 33·33 10·42 0·293 3·05 100 100 89-09 444·8 495·6 1800 1324 28·87 2·41 8·04 31·28 39·32 11·97 0·279 3·34 100 100 90-10 429·0 457·6 340 2500 21·07 1·68 4·00 22·75 26·75 12·54 0·190 2·38 100 100 90-11 635·8 691·5 1970 1382 29·77 2·68 9·22 32·45 41·67 11·10 0·310 3·44 100 100 90-13 421·0 483·0 1300 1706 28·39 2·71 10·10 31·10 41·19 10·48 0·356 3·73 100 100 90-14 173·5 208·0 400 2559 17·57 1·63 4·09 19·19 23·28 10·80 0·233 2·51 100 100 90-17A 471·6 512·9 740 2118 22·13 2·17 5·41 24·30 29·71 10·18 0·244 2·49 100 100 90-18 946·0 1025·7 1200 1824 33·82 3·30 9·61 37·12 46·72 10·25 0·284 2·91 100 100 90-19 551·0 603·0 5270 1824 26·62 2·40 9·33 29·02 38·35 11·12 0·351 3·90 100 100 90-20 935·0 992·0 1860 2588 27·40 2·93 10·04 30·32 40·37 9·36 0·367 3·43 100 100 90-22 978·0 1045·0 2690 1029 34·21 3·16 11·34 37·37 48·71 10·82 0·331 3·59 100 100 90-23A 794·0 822·0 6700 3441 19·12 1·74 3·90 20·87 24·76 10·98 0·204 2·24 100 100 90-24 1018·0 1070·0 3350 294 31·15 2·75 12·67 33·89 46·56 11·34 0·407 4·61 100 100 04-28A 436·0 484·0 5147 2273 23·31 2·04 9·44 25·34 34·79 11·45 0·405 4·64 100 100 04-30 1121·0 1175·9 1500 2000 26·72 2·31 9·79 29·03 38·81 11·57 0·366 4·24 100 100 04-31 1110·0 1184·5 2400 1357 34·78 2·80 12·48 37·58 50·06 12·40 0·359 4·45 100 100 04-32 371·0 425·0 1600 1500 32·00 2·02 11·46 34·02 45·48 15·84 0·358 5·67 100 100 04-33 377·0 425·0 4559 1714 27·69 2·40 9·29 30·09 39·38 11·56 0·335 3·88 100 100 04-34 419·0 477·0 5500 3286 24·61 2·68 8·67 27·29 35·96 9·18 0·352 3·24 100 100 11-53 433·0 490·0 1765 1549 31·76 2·95 10·25 34·71 44·96 10·77 0·323 3·47 100 100 11-57 89·0 127·0 6200 3803 18·52 2·22 7·36 20·74 28·10 8·34 0·397 3·32 100 100 * Grade times width (g*w) is the average grade over a given stratigraphic interval in grams/ton multiplied by its height in metres. † Exception: Drill core 04-28 A from which the lower part of Pd5 is missing and core 31 in which Pd5 is cut by a dike. ‡ With the exception of 90-18 that has an Au concentration in Au + 1 elevated ∼12 metres above the Au/Pd1. See text for further information. Open in new tab Table 6 Grade times width* for Pd, Pt and Au in the Skaergaard PGE–Au Mineralization Drill core . From metres . To metres . Distance from western margin (see Nielsen et al., 2015) . Distance from centre at at point 666 (m) (see text) . Pd . Pt . Au . Pd+Pt . Pd+Pt+Au . Pd/Pt . Au/Pd . Au/Pt . PGE . Au . % of all In core through deposit . % of all In core through deposit . Pd5, 5 metres of core with highest Pd+Pt † 89-02 201·0 206·0 2490 1559 10·45 0·84 0·31 11·29 11·60 12·51 0·030 0·37 30·1 3·8 89-03 255·0 260·0 1460 1588 8·24 0·81 0·48 9·05 9·53 10·17 0·058 0·59 34·4 6·8 89-04 272·0 277·0 1475 1588 6·43 0·76 0·39 7·20 7·58 8·42 0·060 0·51 30·8 4·6 89-06 159·0 164·0 4500 2735 9·54 0·75 0·36 10·28 10·64 12·80 0·038 0·48 39·1 5·1 89-09 484·0 489·0 1800 1324 7·68 0·73 0·36 8·41 8·77 10·46 0·047 0·50 26·9 4·4 90-10 442·6 447·6 340 2500 11·98 0·64 0·78 12·62 13·40 18·70 0·065 1·21 55·5 19·4 90-11 675·1 680·1 1970 1382 6·97 0·64 0·39 7·61 8·00 10·95 0·056 0·62 23·5 4·3 90-13 469·0 474·0 1300 1706 7·08 0·65 0·33 7·73 8·06 10·84 0·047 0·51 24·9 3·3 90-14 192·9 198·0 400 2559 11·19 0·75 0·63 11·94 12·57 14·89 0·056 0·83 62·2 15·3 90-17A 498·0 503·0 740 2118 8·64 0·75 0·26 9·39 9·65 11·52 0·030 0·35 38·6 4·9 90-18 1009·0 1014·0 1200 1824 6·15 0·76 0·33 6·91 7·24 8·11 0·054 0·44 18·6 3·5 90-19 589·0 594·0 5270 1824 6·57 0·65 0·37 7·22 7·59 10·11 0·056 0·57 24·9 3·9 90-20 975·0 980·0 1860 2588 4·68 0·88 0·33 5·56 5·89 5·33 0·071 0·38 18·3 3·3 90-22 810·0 815·0 2690 1029 9·15 0·89 0·41 10·04 10·45 10·29 0·045 0·46 26·9 3·6 90-23A 806·0 811·0 6700 3441 9·30 0·62 0·61 9·92 10·53 15·00 0·066 0·99 47·5 15·8 90-24 1055·0 1060·0 3350 294 8·80 0·74 0·35 9·54 9·89 11·89 0·040 0·47 28·1 2·7 04-28A 471·0 475·3 5147 2273 8·10 0·82 0·41 8·92 9·33 9·86 0·051 0·50 26·5 4·7 04-30 1168·0 1173·0 1500 2000 5·50 0·73 0·53 6·23 6·76 7·53 0·096 0·73 55·4 14·9 04-31 1170·0 1176·0 2400 1357 8·71 0·65 0·42 9·36 9·79 13·31 0·049 0·65 35·2 4·3 04-32 412·0 417·0 1600 1500 13·61 0·7 0·59 14·31 14·90 19·44 0·043 0·84 21·5 5·4 04·33 411·0 416·0 4559 1714 7·62 0·67 0·45 8·29 8·74 11·37 0·059 0·67 24·9 3·4 04-34 460·0 465·0 5500 3286 9·22 0·76 0·71 9·98 10·69 12·13 0·077 0·93 42·4 5·1 11-53 475·0 480·0 1765 1549 8·37 0·83 0·48 9·20 9·68 10·08 0·057 0·58 27·6 4·8 11-57 114·0 119·0 6200 3803 10·77 0·73 1·09 11·50 12·59 14·75 0·101 1·50 36·6 8·2 LPGEM (Pd6-Pd4): from a lower cut-off of 100 ppb Pd+Pt to Pd+Pt low between Pd4a and Pd3b 89-02 191·0 218·0 2490 1559 22·52 2·29 1·01 24·80 25·81 9·83 0·045 0·44 66·1 12·2 89-03 245·0 267·0 1460 1588 16·31 1·88 1·27 18·19 19·46 8·69 0·078 0·67 69·2 18·0 89-04 258·0 283·0 1475 1588 15·34 1·71 1·59 17·05 18·64 8·97 0·103 0·93 73·1 19·1 89-06 144·0 173·0 4500 2735 22·16 2·14 2·03 24·30 26·32 10·37 0·092 0·95 92·4 28·9 89-09 473·0 495·6 1800 1324 16·68 1·76 0·81 18·44 19·25 9·45 0·049 0·46 59·0 9·8 90-10 429·0 457·6 340 2500 21·07 1·68 4·00 22·75 26·75 12·54 0·190 2·38 100·0 100·0 90-11 663·5 691·5 1970 1382 16·51 1·89 0·96 18·40 19·36 8·75 0·058 0·51 56·7 10·4 90-13 459·0 483·0 1300 1706 16·22 1·83 1·07 18·04 19·12 8·88 0·066 0·59 58·0 10·6 90-14 173·5 208·0 400 2559 17·57 1·63 4·09 19·19 23·28 10·80 0·233 2·51 100·0 100·0 90-17A 484·0 512·9 740 2118 19·50 1·92 1·08 21·42 22·49 10·18 0·055 0·56 88·1 19·9 90-18 998·3 1025·7 1200 1824 15·38 2·13 0·86 17·51 18·37 7·21 0·056 0·40 47·2 8·9 90-19 571·0 603·0 5270 1824 18·33 1·83 1·22 20·17 21·39 10·01 0·067 0·67 69·5 13·1 90-20 962·0 992·0 1860 2588 14·87 2·15 0·80 17·01 17·82 6·92 0·054 0·37 56·1 8·0 90-22 1020·0 1045·0 2690 1029 20·25 2·17 1·37 22·42 23·80 9·33 0·068 0·63 60·0 12·1 90-23A 794·0 822·0 6700 3441 19·12 1·74 3·90 20·87 24·76 10·98 0·204 2·24 100·0 100·0 90-24 1045·0 1070·0 3350 294 19·80 2·02 0·87 21·82 22·69 9·80 0·044 0·43 64·4 6·9 04-28A 461·0 484·0 5147 2273 15·90 1·46 0·90 17·36 18·26 10·91 0·056 0·62 60·2 12·5 04-30 1155·0 1175·9 1500 2000 10·37 1·30 0·90 11·67 12·57 7·95 0·087 0·69 96·7 72·6 04-31 1152·0 1184·5 2400 1357 23·29 2·10 2·02 25·39 27·69 11·06 0·099 1·09 68·5 9·5 04-32 399·0 425·0 1600 1500 26·89 1·50 2·50 28·39 30·89 17·93 0·093 1·67 45·1 9·2 04-33 396·0 425·0 4559 1714 19·35 1·90 1·63 21·25 22·88 10·21 0·084 0·86 67·6 18·4 04-34 450·0 477·0 5500 3286 19·10 2·09 1·70 21·19 22·89 9·14 0·089 0·81 92·9 71·7 11-53 465·0 490·0 1765 1549 18·79 2·11 1·28 20·90 22·18 8·91 0·068 0·61 70·6 17·6 11-57 98·0 127·0 6200 3803 18·20 1·87 5·35 20·07 25·41 9·76 0·294 2·87 77·6 19·6 Pd6 to Pd1/Au: Pd+Pt+Au from a lower to an upper cut-off of 100 ppb Pd+Pt+Au‡ 89-02 163·0 218·0 2490 1559 34·45 3·06 8·25 37·51 45·76 11·26 0·239 2·70 100 100 89-03 230·0 267·0 1460 1588 24·00 2·30 7·03 26·30 33·33 10·42 0·293 3·05 100 100 89-04 251·2 283·0 1475 1588 21·19 2·15 8·31 23·34 31·64 9·85 0·392 3·86 100 100 89-06 138·0 173·0 4500 2735 24·00 2·30 7·03 26·30 33·33 10·42 0·293 3·05 100 100 89-09 444·8 495·6 1800 1324 28·87 2·41 8·04 31·28 39·32 11·97 0·279 3·34 100 100 90-10 429·0 457·6 340 2500 21·07 1·68 4·00 22·75 26·75 12·54 0·190 2·38 100 100 90-11 635·8 691·5 1970 1382 29·77 2·68 9·22 32·45 41·67 11·10 0·310 3·44 100 100 90-13 421·0 483·0 1300 1706 28·39 2·71 10·10 31·10 41·19 10·48 0·356 3·73 100 100 90-14 173·5 208·0 400 2559 17·57 1·63 4·09 19·19 23·28 10·80 0·233 2·51 100 100 90-17A 471·6 512·9 740 2118 22·13 2·17 5·41 24·30 29·71 10·18 0·244 2·49 100 100 90-18 946·0 1025·7 1200 1824 33·82 3·30 9·61 37·12 46·72 10·25 0·284 2·91 100 100 90-19 551·0 603·0 5270 1824 26·62 2·40 9·33 29·02 38·35 11·12 0·351 3·90 100 100 90-20 935·0 992·0 1860 2588 27·40 2·93 10·04 30·32 40·37 9·36 0·367 3·43 100 100 90-22 978·0 1045·0 2690 1029 34·21 3·16 11·34 37·37 48·71 10·82 0·331 3·59 100 100 90-23A 794·0 822·0 6700 3441 19·12 1·74 3·90 20·87 24·76 10·98 0·204 2·24 100 100 90-24 1018·0 1070·0 3350 294 31·15 2·75 12·67 33·89 46·56 11·34 0·407 4·61 100 100 04-28A 436·0 484·0 5147 2273 23·31 2·04 9·44 25·34 34·79 11·45 0·405 4·64 100 100 04-30 1121·0 1175·9 1500 2000 26·72 2·31 9·79 29·03 38·81 11·57 0·366 4·24 100 100 04-31 1110·0 1184·5 2400 1357 34·78 2·80 12·48 37·58 50·06 12·40 0·359 4·45 100 100 04-32 371·0 425·0 1600 1500 32·00 2·02 11·46 34·02 45·48 15·84 0·358 5·67 100 100 04-33 377·0 425·0 4559 1714 27·69 2·40 9·29 30·09 39·38 11·56 0·335 3·88 100 100 04-34 419·0 477·0 5500 3286 24·61 2·68 8·67 27·29 35·96 9·18 0·352 3·24 100 100 11-53 433·0 490·0 1765 1549 31·76 2·95 10·25 34·71 44·96 10·77 0·323 3·47 100 100 11-57 89·0 127·0 6200 3803 18·52 2·22 7·36 20·74 28·10 8·34 0·397 3·32 100 100 Drill core . From metres . To metres . Distance from western margin (see Nielsen et al., 2015) . Distance from centre at at point 666 (m) (see text) . Pd . Pt . Au . Pd+Pt . Pd+Pt+Au . Pd/Pt . Au/Pd . Au/Pt . PGE . Au . % of all In core through deposit . % of all In core through deposit . Pd5, 5 metres of core with highest Pd+Pt † 89-02 201·0 206·0 2490 1559 10·45 0·84 0·31 11·29 11·60 12·51 0·030 0·37 30·1 3·8 89-03 255·0 260·0 1460 1588 8·24 0·81 0·48 9·05 9·53 10·17 0·058 0·59 34·4 6·8 89-04 272·0 277·0 1475 1588 6·43 0·76 0·39 7·20 7·58 8·42 0·060 0·51 30·8 4·6 89-06 159·0 164·0 4500 2735 9·54 0·75 0·36 10·28 10·64 12·80 0·038 0·48 39·1 5·1 89-09 484·0 489·0 1800 1324 7·68 0·73 0·36 8·41 8·77 10·46 0·047 0·50 26·9 4·4 90-10 442·6 447·6 340 2500 11·98 0·64 0·78 12·62 13·40 18·70 0·065 1·21 55·5 19·4 90-11 675·1 680·1 1970 1382 6·97 0·64 0·39 7·61 8·00 10·95 0·056 0·62 23·5 4·3 90-13 469·0 474·0 1300 1706 7·08 0·65 0·33 7·73 8·06 10·84 0·047 0·51 24·9 3·3 90-14 192·9 198·0 400 2559 11·19 0·75 0·63 11·94 12·57 14·89 0·056 0·83 62·2 15·3 90-17A 498·0 503·0 740 2118 8·64 0·75 0·26 9·39 9·65 11·52 0·030 0·35 38·6 4·9 90-18 1009·0 1014·0 1200 1824 6·15 0·76 0·33 6·91 7·24 8·11 0·054 0·44 18·6 3·5 90-19 589·0 594·0 5270 1824 6·57 0·65 0·37 7·22 7·59 10·11 0·056 0·57 24·9 3·9 90-20 975·0 980·0 1860 2588 4·68 0·88 0·33 5·56 5·89 5·33 0·071 0·38 18·3 3·3 90-22 810·0 815·0 2690 1029 9·15 0·89 0·41 10·04 10·45 10·29 0·045 0·46 26·9 3·6 90-23A 806·0 811·0 6700 3441 9·30 0·62 0·61 9·92 10·53 15·00 0·066 0·99 47·5 15·8 90-24 1055·0 1060·0 3350 294 8·80 0·74 0·35 9·54 9·89 11·89 0·040 0·47 28·1 2·7 04-28A 471·0 475·3 5147 2273 8·10 0·82 0·41 8·92 9·33 9·86 0·051 0·50 26·5 4·7 04-30 1168·0 1173·0 1500 2000 5·50 0·73 0·53 6·23 6·76 7·53 0·096 0·73 55·4 14·9 04-31 1170·0 1176·0 2400 1357 8·71 0·65 0·42 9·36 9·79 13·31 0·049 0·65 35·2 4·3 04-32 412·0 417·0 1600 1500 13·61 0·7 0·59 14·31 14·90 19·44 0·043 0·84 21·5 5·4 04·33 411·0 416·0 4559 1714 7·62 0·67 0·45 8·29 8·74 11·37 0·059 0·67 24·9 3·4 04-34 460·0 465·0 5500 3286 9·22 0·76 0·71 9·98 10·69 12·13 0·077 0·93 42·4 5·1 11-53 475·0 480·0 1765 1549 8·37 0·83 0·48 9·20 9·68 10·08 0·057 0·58 27·6 4·8 11-57 114·0 119·0 6200 3803 10·77 0·73 1·09 11·50 12·59 14·75 0·101 1·50 36·6 8·2 LPGEM (Pd6-Pd4): from a lower cut-off of 100 ppb Pd+Pt to Pd+Pt low between Pd4a and Pd3b 89-02 191·0 218·0 2490 1559 22·52 2·29 1·01 24·80 25·81 9·83 0·045 0·44 66·1 12·2 89-03 245·0 267·0 1460 1588 16·31 1·88 1·27 18·19 19·46 8·69 0·078 0·67 69·2 18·0 89-04 258·0 283·0 1475 1588 15·34 1·71 1·59 17·05 18·64 8·97 0·103 0·93 73·1 19·1 89-06 144·0 173·0 4500 2735 22·16 2·14 2·03 24·30 26·32 10·37 0·092 0·95 92·4 28·9 89-09 473·0 495·6 1800 1324 16·68 1·76 0·81 18·44 19·25 9·45 0·049 0·46 59·0 9·8 90-10 429·0 457·6 340 2500 21·07 1·68 4·00 22·75 26·75 12·54 0·190 2·38 100·0 100·0 90-11 663·5 691·5 1970 1382 16·51 1·89 0·96 18·40 19·36 8·75 0·058 0·51 56·7 10·4 90-13 459·0 483·0 1300 1706 16·22 1·83 1·07 18·04 19·12 8·88 0·066 0·59 58·0 10·6 90-14 173·5 208·0 400 2559 17·57 1·63 4·09 19·19 23·28 10·80 0·233 2·51 100·0 100·0 90-17A 484·0 512·9 740 2118 19·50 1·92 1·08 21·42 22·49 10·18 0·055 0·56 88·1 19·9 90-18 998·3 1025·7 1200 1824 15·38 2·13 0·86 17·51 18·37 7·21 0·056 0·40 47·2 8·9 90-19 571·0 603·0 5270 1824 18·33 1·83 1·22 20·17 21·39 10·01 0·067 0·67 69·5 13·1 90-20 962·0 992·0 1860 2588 14·87 2·15 0·80 17·01 17·82 6·92 0·054 0·37 56·1 8·0 90-22 1020·0 1045·0 2690 1029 20·25 2·17 1·37 22·42 23·80 9·33 0·068 0·63 60·0 12·1 90-23A 794·0 822·0 6700 3441 19·12 1·74 3·90 20·87 24·76 10·98 0·204 2·24 100·0 100·0 90-24 1045·0 1070·0 3350 294 19·80 2·02 0·87 21·82 22·69 9·80 0·044 0·43 64·4 6·9 04-28A 461·0 484·0 5147 2273 15·90 1·46 0·90 17·36 18·26 10·91 0·056 0·62 60·2 12·5 04-30 1155·0 1175·9 1500 2000 10·37 1·30 0·90 11·67 12·57 7·95 0·087 0·69 96·7 72·6 04-31 1152·0 1184·5 2400 1357 23·29 2·10 2·02 25·39 27·69 11·06 0·099 1·09 68·5 9·5 04-32 399·0 425·0 1600 1500 26·89 1·50 2·50 28·39 30·89 17·93 0·093 1·67 45·1 9·2 04-33 396·0 425·0 4559 1714 19·35 1·90 1·63 21·25 22·88 10·21 0·084 0·86 67·6 18·4 04-34 450·0 477·0 5500 3286 19·10 2·09 1·70 21·19 22·89 9·14 0·089 0·81 92·9 71·7 11-53 465·0 490·0 1765 1549 18·79 2·11 1·28 20·90 22·18 8·91 0·068 0·61 70·6 17·6 11-57 98·0 127·0 6200 3803 18·20 1·87 5·35 20·07 25·41 9·76 0·294 2·87 77·6 19·6 Pd6 to Pd1/Au: Pd+Pt+Au from a lower to an upper cut-off of 100 ppb Pd+Pt+Au‡ 89-02 163·0 218·0 2490 1559 34·45 3·06 8·25 37·51 45·76 11·26 0·239 2·70 100 100 89-03 230·0 267·0 1460 1588 24·00 2·30 7·03 26·30 33·33 10·42 0·293 3·05 100 100 89-04 251·2 283·0 1475 1588 21·19 2·15 8·31 23·34 31·64 9·85 0·392 3·86 100 100 89-06 138·0 173·0 4500 2735 24·00 2·30 7·03 26·30 33·33 10·42 0·293 3·05 100 100 89-09 444·8 495·6 1800 1324 28·87 2·41 8·04 31·28 39·32 11·97 0·279 3·34 100 100 90-10 429·0 457·6 340 2500 21·07 1·68 4·00 22·75 26·75 12·54 0·190 2·38 100 100 90-11 635·8 691·5 1970 1382 29·77 2·68 9·22 32·45 41·67 11·10 0·310 3·44 100 100 90-13 421·0 483·0 1300 1706 28·39 2·71 10·10 31·10 41·19 10·48 0·356 3·73 100 100 90-14 173·5 208·0 400 2559 17·57 1·63 4·09 19·19 23·28 10·80 0·233 2·51 100 100 90-17A 471·6 512·9 740 2118 22·13 2·17 5·41 24·30 29·71 10·18 0·244 2·49 100 100 90-18 946·0 1025·7 1200 1824 33·82 3·30 9·61 37·12 46·72 10·25 0·284 2·91 100 100 90-19 551·0 603·0 5270 1824 26·62 2·40 9·33 29·02 38·35 11·12 0·351 3·90 100 100 90-20 935·0 992·0 1860 2588 27·40 2·93 10·04 30·32 40·37 9·36 0·367 3·43 100 100 90-22 978·0 1045·0 2690 1029 34·21 3·16 11·34 37·37 48·71 10·82 0·331 3·59 100 100 90-23A 794·0 822·0 6700 3441 19·12 1·74 3·90 20·87 24·76 10·98 0·204 2·24 100 100 90-24 1018·0 1070·0 3350 294 31·15 2·75 12·67 33·89 46·56 11·34 0·407 4·61 100 100 04-28A 436·0 484·0 5147 2273 23·31 2·04 9·44 25·34 34·79 11·45 0·405 4·64 100 100 04-30 1121·0 1175·9 1500 2000 26·72 2·31 9·79 29·03 38·81 11·57 0·366 4·24 100 100 04-31 1110·0 1184·5 2400 1357 34·78 2·80 12·48 37·58 50·06 12·40 0·359 4·45 100 100 04-32 371·0 425·0 1600 1500 32·00 2·02 11·46 34·02 45·48 15·84 0·358 5·67 100 100 04-33 377·0 425·0 4559 1714 27·69 2·40 9·29 30·09 39·38 11·56 0·335 3·88 100 100 04-34 419·0 477·0 5500 3286 24·61 2·68 8·67 27·29 35·96 9·18 0·352 3·24 100 100 11-53 433·0 490·0 1765 1549 31·76 2·95 10·25 34·71 44·96 10·77 0·323 3·47 100 100 11-57 89·0 127·0 6200 3803 18·52 2·22 7·36 20·74 28·10 8·34 0·397 3·32 100 100 * Grade times width (g*w) is the average grade over a given stratigraphic interval in grams/ton multiplied by its height in metres. † Exception: Drill core 04-28 A from which the lower part of Pd5 is missing and core 31 in which Pd5 is cut by a dike. ‡ With the exception of 90-18 that has an Au concentration in Au + 1 elevated ∼12 metres above the Au/Pd1. See text for further information. Open in new tab The geochemical profile obtained from drill core 90–22 is used for the principal component analysis (Fig. 12, Table 8, Supplementary Data EA14), and is composed of a continuum of 25-cm samples (Bernstein & Nielsen, 2004; 259 samples, 258 analysed for Pd, Pt and Au, Supplementary Data EA13) and serves as ‘master profile’ for the identification of the MLs and mineralization levels in the central parts of the Skaergaard PGE–Au Mineralization. Being an integrated part of the standard profile for the intrusion (Tegner et al., 2009), the data set in Keays & Tegner (2016), drill core 90–22 allows correlation between all these studies. It is the only publicly available, continuous data set for density (Fig. 5) and full major, trace, and precious metal (Pd, Pt and Au) element geochemistry through the fully developed centre of the Skaergaard PGE–Au Mineralization. Table 8 Summary of the first six principal components from the PCA Principal component . PC1 . PC2 . PC3 . PC4 . PC5 . PC6 . Eigenvalues 6·26 3·68 1·32 0·76 0·63 0·46 Proportion of total variance (%) 44·75 26·30 9·40 5·44 4·56 3·28 Cumulative proportion of total variance (%) 44·75 71·05 80·45 85·89 90·46 93·73 Eigenvectors (loadings) Ti 0·19 −0·39 −0·08 0·06 −0·48 0·14 Fe3+ 0·18 −0·38 0·19 −0·18 −0·07 −0·08 P 0·14 0·32 −0·43 0·07 −0·44 −0·28 Pd −0·35 −0·11 −0·28 −0·14 0·01 0·10 Pt −0·34 −0·06 −0·15 0·44 0·24 −0·05 Au −0·19 0·21 0·48 −0·59 −0·22 −0·06 V 0·11 −0·46 −0·06 −0·07 0·29 0·04 Cu 0·22 0·00 0·54 0·49 −0·02 −0·46 Zn 0·26 −0·38 −0·06 −0·16 0·06 0·02 Y 0·34 0·09 −0·08 −0·17 0·46 −0·17 Zr 0·36 −0·03 −0·12 0·10 −0·31 0·11 Ce 0·30 0·26 −0·18 −0·13 0·11 −0·02 Nd 0·35 0·21 −0·14 −0·15 0·23 −0·08 Pb 0·23 0·26 0·25 0·22 0·07 0·79 Principal component . PC1 . PC2 . PC3 . PC4 . PC5 . PC6 . Eigenvalues 6·26 3·68 1·32 0·76 0·63 0·46 Proportion of total variance (%) 44·75 26·30 9·40 5·44 4·56 3·28 Cumulative proportion of total variance (%) 44·75 71·05 80·45 85·89 90·46 93·73 Eigenvectors (loadings) Ti 0·19 −0·39 −0·08 0·06 −0·48 0·14 Fe3+ 0·18 −0·38 0·19 −0·18 −0·07 −0·08 P 0·14 0·32 −0·43 0·07 −0·44 −0·28 Pd −0·35 −0·11 −0·28 −0·14 0·01 0·10 Pt −0·34 −0·06 −0·15 0·44 0·24 −0·05 Au −0·19 0·21 0·48 −0·59 −0·22 −0·06 V 0·11 −0·46 −0·06 −0·07 0·29 0·04 Cu 0·22 0·00 0·54 0·49 −0·02 −0·46 Zn 0·26 −0·38 −0·06 −0·16 0·06 0·02 Y 0·34 0·09 −0·08 −0·17 0·46 −0·17 Zr 0·36 −0·03 −0·12 0·10 −0·31 0·11 Ce 0·30 0·26 −0·18 −0·13 0·11 −0·02 Nd 0·35 0·21 −0·14 −0·15 0·23 −0·08 Pb 0·23 0·26 0·25 0·22 0·07 0·79 Details of analysis and method in text and EA14. Open in new tab Table 8 Summary of the first six principal components from the PCA Principal component . PC1 . PC2 . PC3 . PC4 . PC5 . PC6 . Eigenvalues 6·26 3·68 1·32 0·76 0·63 0·46 Proportion of total variance (%) 44·75 26·30 9·40 5·44 4·56 3·28 Cumulative proportion of total variance (%) 44·75 71·05 80·45 85·89 90·46 93·73 Eigenvectors (loadings) Ti 0·19 −0·39 −0·08 0·06 −0·48 0·14 Fe3+ 0·18 −0·38 0·19 −0·18 −0·07 −0·08 P 0·14 0·32 −0·43 0·07 −0·44 −0·28 Pd −0·35 −0·11 −0·28 −0·14 0·01 0·10 Pt −0·34 −0·06 −0·15 0·44 0·24 −0·05 Au −0·19 0·21 0·48 −0·59 −0·22 −0·06 V 0·11 −0·46 −0·06 −0·07 0·29 0·04 Cu 0·22 0·00 0·54 0·49 −0·02 −0·46 Zn 0·26 −0·38 −0·06 −0·16 0·06 0·02 Y 0·34 0·09 −0·08 −0·17 0·46 −0·17 Zr 0·36 −0·03 −0·12 0·10 −0·31 0·11 Ce 0·30 0·26 −0·18 −0·13 0·11 −0·02 Nd 0·35 0·21 −0·14 −0·15 0·23 −0·08 Pb 0·23 0·26 0·25 0·22 0·07 0·79 Principal component . PC1 . PC2 . PC3 . PC4 . PC5 . PC6 . Eigenvalues 6·26 3·68 1·32 0·76 0·63 0·46 Proportion of total variance (%) 44·75 26·30 9·40 5·44 4·56 3·28 Cumulative proportion of total variance (%) 44·75 71·05 80·45 85·89 90·46 93·73 Eigenvectors (loadings) Ti 0·19 −0·39 −0·08 0·06 −0·48 0·14 Fe3+ 0·18 −0·38 0·19 −0·18 −0·07 −0·08 P 0·14 0·32 −0·43 0·07 −0·44 −0·28 Pd −0·35 −0·11 −0·28 −0·14 0·01 0·10 Pt −0·34 −0·06 −0·15 0·44 0·24 −0·05 Au −0·19 0·21 0·48 −0·59 −0·22 −0·06 V 0·11 −0·46 −0·06 −0·07 0·29 0·04 Cu 0·22 0·00 0·54 0·49 −0·02 −0·46 Zn 0·26 −0·38 −0·06 −0·16 0·06 0·02 Y 0·34 0·09 −0·08 −0·17 0·46 −0·17 Zr 0·36 −0·03 −0·12 0·10 −0·31 0·11 Ce 0·30 0·26 −0·18 −0·13 0·11 −0·02 Nd 0·35 0·21 −0·14 −0·15 0·23 −0·08 Pb 0·23 0·26 0·25 0·22 0·07 0·79 Details of analysis and method in text and EA14. Open in new tab All cores drilled in the Skaergaard intrusion have numbers that identify the year they were drilled and a consecutive number (e.g. 90–22, drill core 22 drilled 1990). The numbering system was adopted by Platinova Resources Ltd in 1989 and has been continued by all other license holders. Drill cores 89–01 to 90–27 were drilled by Platinova Resources Ltd and Platinova A/S, drill cores 04–28 to 04–34 by Skaergaard Minerals Corporation (SMC), and drill cores 08–35 to 11–58 by Platina Resources Ltd All exploration reports and assay data in public domain are all available online in ‘Greenland Portal’ at www.greenmin.gl. Compiled information The precious metal parageneses of the Skaergaard PGE–Au Mineralization The mineralogical investigations listed above show that the parageneses of the Skaergaard PGE–Au Mineralization comprise a total of three native elements, 35 precious metal minerals, 21 repeatedly recognized precious metal phases, and about 40 alloys combining two or more of PGE, Au, Ag, Cu, Sn, Zn, Ni, Fe Sb, Pb, Ge, Te, and S (Table 3). Pt-minerals and phases are very rare and Pt is almost entirely hosted in Pd-minerals and phases. The paragenetic variations for individual drill cores and MLs (i.e. in the order that the host rocks accumulated) are summarised in Fig. 7 and Table 4. All volumetric percentages are reported relative to the total volume of the precious metal paragenesis of the studied sample. The distribution of investigated samples and detailed descriptions are found in Supplementary Data EA5–EA12. Fig. 7 Open in new tabDownload slide The volumetrically most important groups of precious metal minerals and phases encountered in samples from drill cores and bulk samples. Sulphide minerals vysotskite and vasilite dominate near the Archaean basement to the west, the plumbides and arsenides near basalts to the east, and Cu–PGE alloys and minerals, including skaergaardite (PdCu), dominate in the central parts of the intrusion. Au-rich parageneses are dominated by tetra-auricupride (AuCu) and unnamed Au3Cu. All data and information on methods and samples can found in Table 4 and Supplementary Data EA5-12. Fig. 7 Open in new tabDownload slide The volumetrically most important groups of precious metal minerals and phases encountered in samples from drill cores and bulk samples. Sulphide minerals vysotskite and vasilite dominate near the Archaean basement to the west, the plumbides and arsenides near basalts to the east, and Cu–PGE alloys and minerals, including skaergaardite (PdCu), dominate in the central parts of the intrusion. Au-rich parageneses are dominated by tetra-auricupride (AuCu) and unnamed Au3Cu. All data and information on methods and samples can found in Table 4 and Supplementary Data EA5-12. Near the margins of the intrusion, precious metals are restricted to ML0 and ML1·1. At the western margin (drill core 90–10) the precious metal assemblage is dominated by the PGE-sulphides vysotskite and vasilite, PGE-arsenide vincentite and Au-minerals including Au-rich intermetallic compounds and/or alloys. At the eastern margin (drill core 90–23 A) the paragenesis is dominated by the Pd-plumbide zvyagintsevite, Pd-arsenides vincentite and arsenopalladinite, and unnamed Au-alloys. In both drill cores the proportion of the dominant Pd-phase (as sulphides in the west and plumbide in the east) decreases upward and is volumetrically replaced first by arsenides, and then by Au-rich phases (Table 5, Fig. 7). In ML1.1 at both the east and west margins, the parageneses are strongly dominated by Au-phases (alloys of Au, Pd, Cu, Fe and Ag) and Ag-phases. The overlying mineralization levels in the same cores are Cu-rich and poor in precious metals (e.g. drill core 90–23 A in Fig. 3a). In more centrally located sampling sites (drill core 90–24 and bulk sample TOF) the PGE paragenesis of the main PGE peak in ML0 (Pd6 and Pd5) is dominated by skaergaardite (PdCu, >95 vol. %), and the remainder is accounted for by related alloys. In the SW part of the intrusion, ML0 in drill core 90–18 is characterised by a mixture of PGE sulphides (18·2 vol. %, vysotskite dominates) and skaergaardite and related alloys (78·0 vol. %), and, therefore, can be considered transitional between the skaergaardite dominated centre and the sulphide dominated W-margin (Table 4 and Fig. 7). With minor departures, the mineralogy of the precious metal mineral assemblages in overlying MLs in the central sampling sites (drill core 90–24 and TOF) continue to be dominated by skaergaardite (PdCu). With increasing height in the stratigraphy, an increasing amount of Au is substituted into skaergaardite in place of Pd (Rudashevsky et al., 2014) until skaergaardite is joined, or substituted, by tetra-auricupride (AuCu). In drill core 90–18 from the SW part of the intrusion, the relative volumes of the sulphides vysotskite and vasilite increase with stratigraphic height until the uppermost mineralization levels (those above ML2·2, five MLs above the base of the mineralization), where the assemblage is dominated by Au-phases. In parallel with the observations at the margins, the dominant Au-phases in the centre of the mineralization are tetra-auricupride and (Au, Cu, Pd) intermetallic compounds, with compositions between skaergaardite (PdCu) and tetra-auricupride (AuCu). Tetra-auricupride and Au3Cu dominate the Au-rich mineralized levels in ML2·2 in drill core 90–18, as in the uppermost Au-rich levels in drill core 08–35 A (Holwell et al., 2015) some 40 m lower in the succession of the layered gabbros (Fig. 7, Supplementary Data EA12). The uppermost part of the Au-mineralization is, irrespective of stratigraphic elevation above the main PGE peak (Pd5) and the ML it is hosted in, characterized by unnamed Au3Cu. Bulk rock PGE and Au: E–W cross section In order to provide an overview and a better understanding of the distribution of precious metals in the mineralization, we have calculated grade times width numbers (g*w) for all drill cores for which assays are publicly available. The (g*w) number for a given interval of a drill core is the average concentration in the chosen width or stratigraphic interval in grams/ton (ppm) multiplied by the width or height of the interval in metres, and is equivalent to compositing the selected stratigraphic interval of the mineralization into a 1-metre thick layer of gabbro containing all precious metal of the stratigraphic interval. We include: (1) (g*w) for the main PGE mineralization level (Pd5), which is defined as the 5 m of core with the highest 1-m (Pd+Pt) averages and equivalent to the Pd-zone of Holwell & Keays (2014); (2) (g*w) for LPGEM (Fig. 5), which includes the lower Pd6–Pd4 mineralization levels in ML0 and ML1·1, as defined by a cut off at 100 ppb (Pd+Pt) below Pd6 and the PGE-low between mineralization levels Pd4a and Pd3b of the mineralization (see Fig. 3a). Following Nielsen et al. (2015), LPGEM is the stratigraphic interval of gabbro that crystallized while precious metals were supplied to the mushy floor of the intrusion; and (3) (g*w) for the total stratigraphic interval of the precious metal mineralization, comprising the drill core section between a lower cut-off of 100 ppb at the base of Pd6, and an upper cut-off of 100 ppb at the top of the precious metal mineralization of a given drill core. For the entire precious metal mineralization (case 3 above), the (g*w) for (Pd+Pt+Au) increases from ∼25 near the margins to >45 in the centre of the mineralization (Fig. 8, Table 6). These concentrations signal results from increases in the Au and PGE (Pd+Pt) grades towards the geographic centre and up the MLs (Figs 8–10). For LPGEM (Pd6–Pd4, case 2 above), the Pd+Pt (g*w) number is 21–22 across the intrusion, with Au (g*w) decreasing from highs of ∼4 at the margins (5·4 in drill core11-57; Holwell & Keays, 2014) to a low of 0·8 at the centre of the intrusion (Table 6), and complementary to what is observed for the entire stratigraphic interval of the mineralization (Fig. 9). In Pd5 (case 1 above) the (g*w) for (Pd+Pd) decreases from a high of c.11 at the margins (note, however, Holwell et al. 2016 recorded a maximum of >14 for Pd in drill core 08–35 A, but supply no data for Pt) to < 9 in the geographically central parts of the mineralization. The Au (g*w) in Pd5 decreases from approximately 0·75 at the margins to 0·35 at the centre. Thus the Pd5 interval, just as LPGEM, displays a distribution that is complementary to the total of precious metals contained in the mineralization (case 3). Fig. 8 Open in new tabDownload slide Variation in total content of precious metals in ‘grade x width’ numbers (g*w, average grade in grams/ton x the width or height of the stratigraphic interval in metres) in a section across the intrusion (Table 6). Circles, (Pd+Pt) in the 5 metres of drill core with the highest Pd+Pt in the Pd5 mineralization level (potential ore horizon); diamonds, (Pd+Pt) in LPGEM (ML0 and ML1·1); triangles, (Pd+Pt) in the bulk mineralization from a lower cut off at Pd6 to an upper cut below off above Pd1 of 100 ppb; squares, (Pd+Pt+Au) for the bulk mineralization (Ml0 to ML2·1). Blue, Platinova Resources A/S drill cores (Watts, Griffis & McOuat, 1991); red, Skaergaard Minerals Corporation (Hanghøj, 2005); green, Holwell & Keays (2014). Open symbols, low totals in drill cores 04-30, 89-03, -04, -06 north of the section and low in 90-18 south of the section. Further explanations in the text. Fig. 8 Open in new tabDownload slide Variation in total content of precious metals in ‘grade x width’ numbers (g*w, average grade in grams/ton x the width or height of the stratigraphic interval in metres) in a section across the intrusion (Table 6). Circles, (Pd+Pt) in the 5 metres of drill core with the highest Pd+Pt in the Pd5 mineralization level (potential ore horizon); diamonds, (Pd+Pt) in LPGEM (ML0 and ML1·1); triangles, (Pd+Pt) in the bulk mineralization from a lower cut off at Pd6 to an upper cut below off above Pd1 of 100 ppb; squares, (Pd+Pt+Au) for the bulk mineralization (Ml0 to ML2·1). Blue, Platinova Resources A/S drill cores (Watts, Griffis & McOuat, 1991); red, Skaergaard Minerals Corporation (Hanghøj, 2005); green, Holwell & Keays (2014). Open symbols, low totals in drill cores 04-30, 89-03, -04, -06 north of the section and low in 90-18 south of the section. Further explanations in the text. Fig. 9 Open in new tabDownload slide Au (g*w)-numbers (average grade in grams/ton x the width or height in metres) in a profile across the intrusion as compiled from the base to the top of the mineralization and in the Pd5 mineralization level. Orange, Au (g*w) in the 5-metre section in Pd5 with the highest (Pd+Pt) of the Pd5 mineralization level (potential ore horizon, Table 6); yellow, all Au in the mineralization from <100 ppb below Pd6 (ML0) to <100 ppb above Pd1/Au (ML2·1); red, Au (g*w) number for sections through the entire mineralized succession of gabbro in drill cores from the margins of the intrusion. They provide an approximation to the average Au accumulated at the mushy floor of the magma chamber and available for redistribution up macro-MLs. See text for further explanations. Fig. 9 Open in new tabDownload slide Au (g*w)-numbers (average grade in grams/ton x the width or height in metres) in a profile across the intrusion as compiled from the base to the top of the mineralization and in the Pd5 mineralization level. Orange, Au (g*w) in the 5-metre section in Pd5 with the highest (Pd+Pt) of the Pd5 mineralization level (potential ore horizon, Table 6); yellow, all Au in the mineralization from <100 ppb below Pd6 (ML0) to <100 ppb above Pd1/Au (ML2·1); red, Au (g*w) number for sections through the entire mineralized succession of gabbro in drill cores from the margins of the intrusion. They provide an approximation to the average Au accumulated at the mushy floor of the magma chamber and available for redistribution up macro-MLs. See text for further explanations. Fig. 10 Open in new tabDownload slide Bulk precious metal contents, as in Fig. 8, but plotted relative to an assumed geographical centre for the mineralization (topographic point 666 on lower Basistoppen, see text for details, data in Table 6n). Blue, Watts, Griffis & McOuat (1991); red, Hanghøj (2005); green, Holwell & Keays (2014). Open symbols are g*w; (average grade in grams/ton x the width or height in metres) and ratios of precious metals in drill cores from the W margin of the intrusion. Left column,(g*w) number from a lower cut off at <100 ppb below Pd6 to an upper cut off at <100 ppb above Pd1; centre column,(g*w) for LPGEM that Nielsen et al. (2015) argued to have formed while precious metals were supplied from the bulk magma; right column, 5-metre section in Pd5 with the highest (Pd+Pt) of the Pd5 mineralization level (potential ore horizon). (Pd+Pt), (Pd+Pt+Au), and Pd/Pt, Au/Pt and Au/Pd ratios are shown for all. As in Fig. 8, the lower centre of the mineralization is relatively depleted in PGE and especially in Au (see also Fig. 9), whereas the upper central parts are enriched (see text for further descriptions and discussions). The mineralization is not perfectly concentric and the open symbol data at 2500 m should be compared to the samples at the E margin at 3500 to 4000 m Data from Holwell et al. (2016) have been omitted as they give no Pt values. Fig. 10 Open in new tabDownload slide Bulk precious metal contents, as in Fig. 8, but plotted relative to an assumed geographical centre for the mineralization (topographic point 666 on lower Basistoppen, see text for details, data in Table 6n). Blue, Watts, Griffis & McOuat (1991); red, Hanghøj (2005); green, Holwell & Keays (2014). Open symbols are g*w; (average grade in grams/ton x the width or height in metres) and ratios of precious metals in drill cores from the W margin of the intrusion. Left column,(g*w) number from a lower cut off at <100 ppb below Pd6 to an upper cut off at <100 ppb above Pd1; centre column,(g*w) for LPGEM that Nielsen et al. (2015) argued to have formed while precious metals were supplied from the bulk magma; right column, 5-metre section in Pd5 with the highest (Pd+Pt) of the Pd5 mineralization level (potential ore horizon). (Pd+Pt), (Pd+Pt+Au), and Pd/Pt, Au/Pt and Au/Pd ratios are shown for all. As in Fig. 8, the lower centre of the mineralization is relatively depleted in PGE and especially in Au (see also Fig. 9), whereas the upper central parts are enriched (see text for further descriptions and discussions). The mineralization is not perfectly concentric and the open symbol data at 2500 m should be compared to the samples at the E margin at 3500 to 4000 m Data from Holwell et al. (2016) have been omitted as they give no Pt values. Apart from the most westerly drill core (90–18) all open symbols in Fig. 8 refer to (g*w) numbers from drill cores that are located north of the cross section (from W to E: drill cores 04–30, 89–03, 89–04, and 89–06; see Table 6). Their low totals are artefacts of greater distances to the geographical centre of the concentric mineralization. Drill core 90–18 (1200 m from W margin) with elevated precious metal contents is located far to the SW in the mineralization (Fig. 1a), but projects too close to the western margin despite a more central location in the mineralization. It should be compared to more centrally located drill cores. Bulk rock PGE and Au relative to centre The distribution of precious metals in the mineralization is concentric (Fig. 4, Andersen et al., 1998; Watts, Griffis & McOuat, 1991) and the elemental distribution is better illustrated relative to a possible centre for the mineralization (Fig. 10). The location of the centre is, however, not well-constrained. Here we have, in part supported by Watts, Griffis & McOuat (1991) and by trial and error, used topographic fixed point 666 on the western side of Basistoppen and between the drill hole collars of drill core 90–22 and 90–24 as a proxy (Fig. 1a, see also exploration map in Watts, Griffis & McOuat, 1991). In the compilation of the available data from 24 drill cores we include (g*w) for PGE (Pd+Pt) and total precious metals (Pd+Pt+Au) as well as the three elemental ratios Pd/Pt, Au/Pt and Au/Pd for the stratigraphic interval between cut-offs of 100 ppb (Pd+Pt+Au) at base and top of mineralized gabbros, for LPGEM, and for Pd5. From the margin of the intrusion to the geographical centre of the mineralization the (g*w) for (Pd+Pt+Au) increases from ∼25 to >45 in and is accompanied by increases in Au/Pt and Au/Pd. This pattern signals a general enrichment in Au whereas Pd/Pt shows only a very limited increase toward the centre. The main PGE mineralization level (Pd5) is complementary. Drill cores near the margins of the intrusion show high (g*w) for (Pd+Pt) and (Pd+Pt+Au) compared to most of the cores from the geographical centre (Fig. 10). The Au/Pd and Au/Pt ratios decrease markedly towards the centre and are combined with a general decrease in PGE (see above), which indicates a strong relative depletion in Au in Pd5 in the centre. In LPGEM the (g*w) numbers are intermediate between numbers for Pd5 and those for the entire mineralization between 100 ppb cut offs above and below the mineralized gabbros. LPGEM shows little variation in (g*w) from margins to centre in (Pd+Pt) and (Pd+Pt+Au) and in the Pd/Pt ratio, whereas the Au/Pt and Au/Pd ratios (as in Pd5) are, in comparison, high near the margins of the intrusion and reflect general Au depletion in LPGEM toward the geographical centre. Precious metal to copper ratios in LPGEM The ratio (Pt+Pd+Au)/Cu computed from whole rock analyses (Table 7; Fig. 11) is a proxy for the average (Pt+Pd+Au)/Cu of the sulphide droplets once hosted by the bulk rock, as precious metals are heavily partitioned into the sulphide melt. Since this ratio reflects the sum of the processes that affected the composition of droplets of sulphide melt during the crystallization of the local gabbroic host, any variation of (Pt+Pd+Au)/Cu within a specific stratigraphic interval across the intrusion will likely point to differences in the local physical and chemical processes during crystallization. Figure 11 shows how (Pt+Pd+Au)/Cu varies across the intrusion for the interval between 8 m below the Pd5 peak to 4 m above it, corresponding roughly to the zone spanned by ML0. The data shows how (Pt+Pd+Au)/Cu increases from the margins towards the geographical centre of the mineralization at all depths, and that (Pt+Pd+Au)/Cu at the centre is greater than that at the margins by a factor of up to 2·2. Importantly, this difference is present despite the inwards depletion of Au (and to a lesser extent, Pd) in Pd5 and the LPGEM (Fig. 10, Table 6) and is due to a significantly lowered Cu relative to the precious metals. Fig. 11 Open in new tabDownload slide Relative Cu depletion in the geographic centre of the mineralization due to dissolution of sulphide melt hosted in immiscible droplets. The shown 12-metre section in ML0 includes the main concentration of PGE (Pd5 /Pd-zone) and below Pd6 (sub-zone). Blue, drill cores from the E margin incl. 90-23 A, 04-34, 08-35 A, and 11-57; green, drill cores from the W margin incl. 04-30 and 90-14; orange, drill cores from the SW part of the intrusion incl. 90-22 and 90-18; red, drill cores from central parts incl. 04-32, 04-33, 11-53 and 90–24, and all data in Table 7. Fig. 11 Open in new tabDownload slide Relative Cu depletion in the geographic centre of the mineralization due to dissolution of sulphide melt hosted in immiscible droplets. The shown 12-metre section in ML0 includes the main concentration of PGE (Pd5 /Pd-zone) and below Pd6 (sub-zone). Blue, drill cores from the E margin incl. 90-23 A, 04-34, 08-35 A, and 11-57; green, drill cores from the W margin incl. 04-30 and 90-14; orange, drill cores from the SW part of the intrusion incl. 90-22 and 90-18; red, drill cores from central parts incl. 04-32, 04-33, 11-53 and 90–24, and all data in Table 7. Table 7 Bulk rock ((Pd+Pt+Au)/Cu)*1000 ratios in Pd5 relative to peak concentration (1-m averages) Drill core . PRL11-57 . PRL08-35A* . 90-23A* . SKM04-34 . 90-14* . SKM04-30 . 90-18* . 90-22† . SKM04-32 . PRL11-53 . SKM04-28A . SKM04-31 . SKM04-33 . 90-24* . . East margin . West margin . SW sector . Central part of intrusion . Depth (m) relative to Pd5 peak . ((Pd+Pt+Au)/Cu) * 1000 . 4·0 3·77 3·10 1·74 2·71 – 9·17 – 2·17 4·69 4·98 4·61 6·76 7·02 – 3·0 3·66 6·30 – 1·71 – 7·84 – 3·22 7·26 8·32 5·35 11·9 6·72 – 2·0 7·58 11·00 – 4·55 – 7·47 – 7·93 12·70 15·35 9·61 8·79 10·62 – 1·0 12·33 15·00 10·48 11·10 – 11·30 – 13·54 21·26 20·63 20·43 23·52 21·36 – 0·0 13·33 16·00 – 17·21 14·29 16·11 18·15 16·53 19·35 20·62 22·09 25·15 27·67 29·49 −1·0 11·36 12·40 – 14·89 – 13·33 – 15·90 18·24 16·61 14·48 15·82 19·72 – −2·0 6·94 8·50 8·69 13·61 – 8·83 – 9·10 18·22 11·67 10·29 11·91 13·26 – −3·0 7·18 6·70 6·67 7·33 – 6·53 – 7·97 14·82 8·96 7·50 9·09 10·52 – −4·0 6·69 – – 5·15 – 7·36 – 5·79 10·64 7·61 8·66 8·41 8·66 – −5·0 6·55 – – 5·16 – 7·04 – 4·86 7·87 9·87 7·51 8·70 8·88 – −6·0 6·30 – – 4·96 – – – 4·58 6·14 8·55 7·67 8·76 1·46 – −7·0 5·65 – – 4·53 – – – 3·22 6·17 10·59 6·67 9·24 10·38 – −8·0 3·48 – – 5·00 – – – 5·08 5·51 8·75 4·54 6·94 8·33 – Drill core . PRL11-57 . PRL08-35A* . 90-23A* . SKM04-34 . 90-14* . SKM04-30 . 90-18* . 90-22† . SKM04-32 . PRL11-53 . SKM04-28A . SKM04-31 . SKM04-33 . 90-24* . . East margin . West margin . SW sector . Central part of intrusion . Depth (m) relative to Pd5 peak . ((Pd+Pt+Au)/Cu) * 1000 . 4·0 3·77 3·10 1·74 2·71 – 9·17 – 2·17 4·69 4·98 4·61 6·76 7·02 – 3·0 3·66 6·30 – 1·71 – 7·84 – 3·22 7·26 8·32 5·35 11·9 6·72 – 2·0 7·58 11·00 – 4·55 – 7·47 – 7·93 12·70 15·35 9·61 8·79 10·62 – 1·0 12·33 15·00 10·48 11·10 – 11·30 – 13·54 21·26 20·63 20·43 23·52 21·36 – 0·0 13·33 16·00 – 17·21 14·29 16·11 18·15 16·53 19·35 20·62 22·09 25·15 27·67 29·49 −1·0 11·36 12·40 – 14·89 – 13·33 – 15·90 18·24 16·61 14·48 15·82 19·72 – −2·0 6·94 8·50 8·69 13·61 – 8·83 – 9·10 18·22 11·67 10·29 11·91 13·26 – −3·0 7·18 6·70 6·67 7·33 – 6·53 – 7·97 14·82 8·96 7·50 9·09 10·52 – −4·0 6·69 – – 5·15 – 7·36 – 5·79 10·64 7·61 8·66 8·41 8·66 – −5·0 6·55 – – 5·16 – 7·04 – 4·86 7·87 9·87 7·51 8·70 8·88 – −6·0 6·30 – – 4·96 – – – 4·58 6·14 8·55 7·67 8·76 1·46 – −7·0 5·65 – – 4·53 – – – 3·22 6·17 10·59 6·67 9·24 10·38 – −8·0 3·48 – – 5·00 – – – 5·08 5·51 8·75 4·54 6·94 8·33 – * No systematic Cu determinations. Cu values from samples collected at intervals (J.C.Ø. Andersen, unpublished). † All data are based on 1-m averages and are +/- 0·5 m in elevation. In Fig. 10 the data for 90–22 have been moved up 0·5 m to align the Pd5 peak. Open in new tab Table 7 Bulk rock ((Pd+Pt+Au)/Cu)*1000 ratios in Pd5 relative to peak concentration (1-m averages) Drill core . PRL11-57 . PRL08-35A* . 90-23A* . SKM04-34 . 90-14* . SKM04-30 . 90-18* . 90-22† . SKM04-32 . PRL11-53 . SKM04-28A . SKM04-31 . SKM04-33 . 90-24* . . East margin . West margin . SW sector . Central part of intrusion . Depth (m) relative to Pd5 peak . ((Pd+Pt+Au)/Cu) * 1000 . 4·0 3·77 3·10 1·74 2·71 – 9·17 – 2·17 4·69 4·98 4·61 6·76 7·02 – 3·0 3·66 6·30 – 1·71 – 7·84 – 3·22 7·26 8·32 5·35 11·9 6·72 – 2·0 7·58 11·00 – 4·55 – 7·47 – 7·93 12·70 15·35 9·61 8·79 10·62 – 1·0 12·33 15·00 10·48 11·10 – 11·30 – 13·54 21·26 20·63 20·43 23·52 21·36 – 0·0 13·33 16·00 – 17·21 14·29 16·11 18·15 16·53 19·35 20·62 22·09 25·15 27·67 29·49 −1·0 11·36 12·40 – 14·89 – 13·33 – 15·90 18·24 16·61 14·48 15·82 19·72 – −2·0 6·94 8·50 8·69 13·61 – 8·83 – 9·10 18·22 11·67 10·29 11·91 13·26 – −3·0 7·18 6·70 6·67 7·33 – 6·53 – 7·97 14·82 8·96 7·50 9·09 10·52 – −4·0 6·69 – – 5·15 – 7·36 – 5·79 10·64 7·61 8·66 8·41 8·66 – −5·0 6·55 – – 5·16 – 7·04 – 4·86 7·87 9·87 7·51 8·70 8·88 – −6·0 6·30 – – 4·96 – – – 4·58 6·14 8·55 7·67 8·76 1·46 – −7·0 5·65 – – 4·53 – – – 3·22 6·17 10·59 6·67 9·24 10·38 – −8·0 3·48 – – 5·00 – – – 5·08 5·51 8·75 4·54 6·94 8·33 – Drill core . PRL11-57 . PRL08-35A* . 90-23A* . SKM04-34 . 90-14* . SKM04-30 . 90-18* . 90-22† . SKM04-32 . PRL11-53 . SKM04-28A . SKM04-31 . SKM04-33 . 90-24* . . East margin . West margin . SW sector . Central part of intrusion . Depth (m) relative to Pd5 peak . ((Pd+Pt+Au)/Cu) * 1000 . 4·0 3·77 3·10 1·74 2·71 – 9·17 – 2·17 4·69 4·98 4·61 6·76 7·02 – 3·0 3·66 6·30 – 1·71 – 7·84 – 3·22 7·26 8·32 5·35 11·9 6·72 – 2·0 7·58 11·00 – 4·55 – 7·47 – 7·93 12·70 15·35 9·61 8·79 10·62 – 1·0 12·33 15·00 10·48 11·10 – 11·30 – 13·54 21·26 20·63 20·43 23·52 21·36 – 0·0 13·33 16·00 – 17·21 14·29 16·11 18·15 16·53 19·35 20·62 22·09 25·15 27·67 29·49 −1·0 11·36 12·40 – 14·89 – 13·33 – 15·90 18·24 16·61 14·48 15·82 19·72 – −2·0 6·94 8·50 8·69 13·61 – 8·83 – 9·10 18·22 11·67 10·29 11·91 13·26 – −3·0 7·18 6·70 6·67 7·33 – 6·53 – 7·97 14·82 8·96 7·50 9·09 10·52 – −4·0 6·69 – – 5·15 – 7·36 – 5·79 10·64 7·61 8·66 8·41 8·66 – −5·0 6·55 – – 5·16 – 7·04 – 4·86 7·87 9·87 7·51 8·70 8·88 – −6·0 6·30 – – 4·96 – – – 4·58 6·14 8·55 7·67 8·76 1·46 – −7·0 5·65 – – 4·53 – – – 3·22 6·17 10·59 6·67 9·24 10·38 – −8·0 3·48 – – 5·00 – – – 5·08 5·51 8·75 4·54 6·94 8·33 – * No systematic Cu determinations. Cu values from samples collected at intervals (J.C.Ø. Andersen, unpublished). † All data are based on 1-m averages and are +/- 0·5 m in elevation. In Fig. 10 the data for 90–22 have been moved up 0·5 m to align the Pd5 peak. Open in new tab Principal component analysis To examine the mineralization in the context of the host rock and test models for the distribution of the precious metals, we performed a principal component analysis (PCA) on assay data from drill core 90–22. The aim of a PCA is to reduce the dimensionality of a dataset by extracting from it a set of linearly uncorrelated variables, or principal components, and retaining only those that make significant contributions to the total variance. Detailed descriptions of the PCA methodology are given by Le Maitre (1982) and Albarède (1995), and some additional, recent examples of its application to igneous geochemistry are given by, e.g. Hamelin et al. (2011) and Ueki & Iwamori (2017). The interval 978·5–1045 m within drill core 90–22 from the centre of the mineralization (Bernstein & Nielsen, 2004) was used for the PCA because it comprises the thickest and most fully developed sequences of mineralization for which both lithochemistry and precious metal assays exist. The dataset comprises 258 samples. To prepare the data for the PCA, all trace elements were converted to parts per million concentrations, and major element oxides were recast as their major element cation values. Because geochemical datasets are compositional by nature and contain only relative information, the data in their raw form are subject to spurious correlations (e.g. Aitchison, 1986; Aitchison & Egozcue, 2005; Pawlowsky-Glahn & Egozcue, 2006). To avoid these well-documented effects and reveal meaningful correlations, the data were transformed using the centred log-ratio transformation (Aitchison, 1982) and subsequently scaled and centred. To improve the performance of the PCA the data were filtered to: (i) remove variables that do not correlate significantly with other variables, and (ii) replace sets of variables that show very strong multicollinearity, such as REE, with a single variable from the set. This resulted in the selection of the following 14 variables for the PCA: Ti, Fe3+, P, Pd, Pt, Au, V, Cu, Zn, Y, Zr, Ce, Nd, and Pb (see Supplementary Data EA13). Finally, these variables were extracted from the original data set and the centred log-ratio transformation was reapplied. Table 8 summarizes the results of the first 6 principal components of the PCA (full results in Supplementary Data EA14, Table EA14-1). Application of the screen test (Cattell, 1966) indicates that the first 3 principal components (PCs) provide a meaningful representation of the input data; collectively they explain 80% of the total variance of the data. To aid analysis and interpretation, Fig. 12a–c plot for each of the three PCs the standardized loadings of the different elements considered, and Fig. EA14-3 in Supplementary Data EA14 shows how PCs 1–3 vary with depth in drill core 90–22. In Fig. 12a–c and EA14-3 and Table 8, the loadings indicate how each element correlates with the different principal components. For example, large positive values indicate that a given analyte and principal component correlate positively and strongly. Conversely a small magnitude, negative value indicates a weak, negative correlation between a component and an analyte. Fig. 12 Open in new tabDownload slide Standardized loadings of the elements considered onto: (a) principal component 1 (PC1), (b) principal component 2 (PC2), and (c) principal component 3 (PC3). Fig. 12 Open in new tabDownload slide Standardized loadings of the elements considered onto: (a) principal component 1 (PC1), (b) principal component 2 (PC2), and (c) principal component 3 (PC3). PC1 accounts for 45% of the total variance. From Fig. 12a and Table 1 it is clear that this component discriminates between aspects of the precious metal mineralization and, with the exception of Cu, elements that are typically found in silicate, oxide and phosphate minerals. That copper and the precious metals Pd, Pt and Au are anticorrelated in this component indicates that more processes are responsible in determining the distributions of these metals within the mineralized section of Skaergaard. Supplementary Data Figure EA14-3 demonstrates that local lows in PC1 correspond to mineralization levels Pd2-6, Pd1 and to lesser extend Pd1/Au. In PC2, which accounts for 26% of the total variance, Fe3+ and the metals Ti, V, Zn correlate strongly and negatively with Au, and the relatively incompatible elements P, Pb, Ce and Nd (Fig. 12b). Cu does not participate in PC2, and the remaining precious metals (Pd and Pt) and Zr are only weakly associated with this component. Systematic variation in PC2 with depth (Supplementary Data Fig. EA14-3) is less obvious than for PC1. PC2 increases to the top of ML0 and ML1·1, and in the Pd3a and b peaks of ML1·2, but has minor importance in ML2. The largest and distinct positive values of PC2 occur in ML2·1 in the top 15 m of the studied interval, and form local highs at the stratigraphic positions of Pd1 and especially at Pd/Au. PC3 accounts for 9% of the total variance of the data. Figure 12c shows that Au and Cu correlate strongly and positively with PC3; Fe3+ and Pb also load positively, but more weakly, onto this component. These metals (including Fe3+) are anticorrelated with P, which exhibits the largest negative loading in PC3, and, to a slightly lesser extent, Pd, Pt and Ce. The remaining metals (Nd, Ti, V, Y, Zn, Zr) load very weakly and negatively with PC3. Supplementary Data Fig. EA14-3 shows that the variation of PC3 with depth is complex. The variation is in general more systematic above ML1·2, corresponding to the interval spanned by UPGM, UAuM and CuM <1021 m). In ML1·2 and ML2 the PC3 loadings generally decrease with height above a Pd interval before increasing within a few metres of the next overlying Pd level. The largest positive values occur again in the top 15 m of the studied interval and correlate with elevations in Cu concentration >200 ppm). DISCUSSION Sulphide saturation as the driver for mineralization To understand the results, it is necessary to review the evidence for sulphide saturation as the driver for the mineralization process. Models proposed for origin of the Skaergaard PGE–Au Mineralization include: (1) contamination/mixing-driven sulphide saturation and precious metal deposition (Bird et al., 1991); (2) mineralization caused by upward transport and re-deposition of precious metals by rising silicate melts and/or fluids late in the solidification of the gabbros (e.g. Boudreau, 2004); and (3) closed magma chamber fractionation of melt to sulphide saturation (Andersen et al., 1998; Nielsen et al., 2005, 2015; Andersen 2006; Holwell & Keays, 2014; Holwell et al., 2015, 2016; Keays & Tegner, 2016). Together, these models cover almost the entire spectrum of processes suggested to be responsible for PGE mineralization in layered mafic intrusions. In the Skaergaard intrusion, the section of MZ beneath the mineralization includes several stratigraphic intervals that are distinguished by both: (i) elevated, but erratic Pd concentrations (Andersen et al., 1998; Nielsen et al., 2015), and (ii) the presence of many gabbroic blocks, which originate from a collapsed part of the UBS and are estimated to make up about 10 vol. % of lower MZ (Irvine et al., 1998). The variation in (Pd+Pt) and in Pd/Pt in the block zone is seemingly of local origin as it returns to low values in bulk rocks above the block zone (Nielsen et al., 2015) and the elevated precious metal grades are suggested to be caused by local contamination- or mixing-driven accumulation of precious metal-bearing droplets of sulphide melt. This was first suggested by Bird et al. (1991), who proposed that incorporation of the gabbroic UBS blocks led to an increase in crystallization of Fe–Ti-oxides and to decrease FeO* and in sulphide solubility in the silicate melt. However, the main PGE–Au mineralization contains fewer, as well as large, roof blocks (Fig. 2) but shows concordant variations across the intrusion (Supplementary Data EA1), irrespective of the frequency of UBS blocks and lateral variations in grades and elemental ratios of precious metal (Fig. 10). The Skaergaard PGE–Au mineralization demonstrates no clear evidence is for a magma mixing- or contamination-driven sulphide saturation Other models suggest that precious metals were scavenged from already-crystallized gabbros by upward migrating late-stage fluids, and then redeposited, e.g. at redox barriers, (Boudreau, 2004; Boudreau & Meurer, 1999). These models are also regarded unlikely because sulphide droplets with PGE and Au are found inside liquidus phases of the gabbroic host (Godel et al., 2014; Nielsen et al., 2015, Holwell et al., 2016). Only the late deposition of Au on grain boundaries in extensively crystallized gabbro (UAuM) could be caused by a late, residual and upward migrating, volatile-rich silicate melt and fluids (Godel et al., 2014; Rudashevsky et al., 2014; Nielsen et al., 2015). This leaves the possibility for sulphide saturation caused by closed system fractionation. This is supported by the mineralogical investigations and the PCA model. The major difference between the proposed sulphide saturation models is that Holwell et al. (2016) assume sulphide saturation in the bulk liquid, whereas Nielsen et al. (2015) restrict sulphide saturation to evolved mush melts of the crystallization zones that existed between remaining bulk liquid and already crystallized gabbros. The timing of sulphide saturation in the evolution of the bulk liquid, therefore, is a key question that needs to be addressed. Keays & Tegner (2016) suggested that sulphide saturation in the bulk liquid was reached at the evolutionary point represented by the liquidus paragenesis represented by LZc (tLZc). As noted in Nielsen et al. (2015), however, co-variations in Pd/Pt and Au/Pt in the bulk liquid prior to the mineralization in the upper MZ point to loss of Pt rather than Pd or Au, during crystallization of LZc and up MZ, e.g. in the form of ferroplatinum ((Fe, Pt); Holwell et al., 2016). If sulphide saturation had taken place from tLZc and onwards, the Pd/Au ratio should have decreased in the remaining bulk liquid throughout LZc and MZ because DPd≫DAu for coexisting sulphide and silicate liquids. With the exception of minor deposition of PGE in MZ in reaction with sunken roof blocks, a decrease in Pd/Au is not observed (Nielsen et al., 2015), and thus the initiation of sulphide saturation in the bulk liquid at tLZc is not supported. Only the closed system sulphide saturation models in Nielsen et al. (2015) and Holwell et al. (2016) seem plausible, but they are based on very different perceptions and samples sets. The Nielsen et al. (2015) model is based on drill cores from across the intrusion, and is inherently three dimensional as a result, and times mineralization processes relative to the crystallization of the silicate host rock. The study and conclusions of Holwell et al. (2016) are based mainly on data obtained from drill core 08–35 A from near the margin of the intrusion and seemingly assumed to be representative for the mineralization process throughout the intrusion. The stratigraphic variations in Fe2O3 in sections of drill core (Fig. 13) are proxies for the lithological variation in the layered host rocks of the mineralization and allow detailed correlation between the sections of drill core studied by Holwell et al. (2016) and Nielsen et al. (2015). The entire stratigraphic interval enriched in precious metals (from the Sub-zone to Au-zone of Holwell et al., 2016) in drill core 08–35 A from near the margin of the intrusion corresponds only to a 20 m succession of gabbro in ML0 and ML1·1, and to LPGEM with mineralization levels Pd6, Pd5, Pd4a and Pd4b of central drill core 90–22 (Fig. 13). Fig. 13 Open in new tabDownload slide Fe as Fe2O3 wt % up the MLs of the Skaergaard mineralization. Blue, central drill core 90–22; red, margin drill core 08-35 A. The variations overlap and demonstrate that MLs maintain near constant thickness across the intrusion. Black column, the precious metal mineralized section of drill core 08-35 A from near the margin of the intrusion in ML0 and ML1·1; grey column, mineralized ML0 to ML2·1 in drill core 90-22 in the centre of the mineralization. Upper golden field, the Au rich section in 90-22 at 0·7 ppm cut-off ; green field, the Pd1 peak. The combined average is 2·1 g/t over 3·6 m (grade x width: 7·56). The lower golden field, the Au rich interval in drill core 08-35 A (Holwell et al., 2016) with an average of 2·2 g/t over 0·8 m (grade x width: 1·76). Depth in drill core 90-22 (Bernstein & Nielsen, 2004) data are calibrated to 08-35 A by subtraction of 706 m Fig. 13 Open in new tabDownload slide Fe as Fe2O3 wt % up the MLs of the Skaergaard mineralization. Blue, central drill core 90–22; red, margin drill core 08-35 A. The variations overlap and demonstrate that MLs maintain near constant thickness across the intrusion. Black column, the precious metal mineralized section of drill core 08-35 A from near the margin of the intrusion in ML0 and ML1·1; grey column, mineralized ML0 to ML2·1 in drill core 90-22 in the centre of the mineralization. Upper golden field, the Au rich section in 90-22 at 0·7 ppm cut-off ; green field, the Pd1 peak. The combined average is 2·1 g/t over 3·6 m (grade x width: 7·56). The lower golden field, the Au rich interval in drill core 08-35 A (Holwell et al., 2016) with an average of 2·2 g/t over 0·8 m (grade x width: 1·76). Depth in drill core 90-22 (Bernstein & Nielsen, 2004) data are calibrated to 08-35 A by subtraction of 706 m In contrast the precious metal mineralization in the centre of the intrusion is in all drill cores (except drill core 90–18, which hosts an additional upper Au-rich level) contained within a stratigraphic interval 60 m thick. It includes five MLs (rather than two) with precious metal peaks in Pd-levels Pd6 to Pd1 and Pd1/Au (Fig. 5). The package of MLs maintains near constant stratigraphic thickness across the floor of the intrusion (Supplementary Data EA3) and the profile through the mineralization discussed in Holwell et al. (2016), and, therefore, is not a compressed version of the section studied by Nielsen et al. (2015). Evidence against offset reef-style mineralization Near the margin of the intrusion, in drill core 08–35 A, the uppermost PGE peak is in Pd-level Pd4b (∼5 m above the main Pd peak in Pd5, see Fig. 5 in Holwell et al., 2016) and the Au peak ∼7 m above Pd5. However, in the Middag Buttress chip line (Fig. 3a), as well as the incompletely sampled Midnat chip line (Turner & Mosher, 1989; Fig. 2b) to the north and farthest away from the centre of the deposit (see locations in map Fig. 1a), PGE and Au are accumulated in the same one-metre interval of layered gabbro that hosts Pd5 throughout the intrusion. The marked differences between the margin and the centre of the mineralization is also shown by (g*w) for the uppermost Au rich part of the mineralization (Table 6; Fig. 12). In the centre the combined Au-bearing Pd1, Pd1/Au and UAuM mineralization levels have 2·1 g/t precious metals over 3·6 m at a cut off of 0·7 g/t and a (g*w) of 7·56, whereas the drill core from the margin of the intrusion, that Holwell et al. (2016) present as representative for the Skaergaard PGE–Au Mineralization, has an average of 2·2 g/t over 0·8 m and a (g*w) of 1·76. That is less than 25% of contained gold compared to the centre of the deposit. The drill core studied in Holwell et al. (2016) does not compare with and cannot be representative for anything but the Skaergaard PGE–Au Mineralization as developed at the east margin of the intrusion. As mentioned above, the accumulation of precious metals during the formation of offset reef type mineralization is thought to be concordant with the crystallization front. Holwell et al. (2016) and Holwell & Keays (2014) attribute the formation of their Au-zone to a single, short-lived and critical event, but this scenario is difficult to reconcile with the increase in stratigraphic elevation between the peak PGE and peak Au concentrations, from <1 m farthest away to >43 m at the centre of the mineralization, whilst the host layered gabbros form a concordant succession of MLs that maintains constant stratigraphic thickness across the intrusion (Fig. 3a and b; Supplementary Data EA3). Assuming a constant crystal accumulation rate of 2 cm/y on the floor of the intrusion (Irvine, 1970; Morse, 2011), the Au and PGE (Pd5) peak are separated by >2000 years (y) in the centre of the intrusion (e.g. drill core 90–22, stratigraphic separation of 40 m, see Supplementary Data SD3 of Nielsen et al., 2015), 400 y closer to the margin (e.g. drill core 08–35 A, stratigraphic separation of 8 m, Holwell et al. 2016) and <50 y farthest away from the centre of the mineralization (e.g. Middag and Midnat Buttress chip lines, stratigraphic separation <1 m). Assuming that the crystallization rate in the upper MZ had dropped to 1 cm/y during the later stages of crystallization, when only 20% of the initial volume of bulk melt remained, the precious metal mineralization may have formed over as much as 4000 years; thus its formation would have been diachronous. In the context of an offset reef-type model, this situation is not compatible with a mineralization event triggered by a specific and short-lived event in a magma homogenized in convective currents (cf. Holwell et al., 2016). Accumulation of precious metals in the floor mush As an alternative to the classic offset reef type models, Nielsen et al. (2015) suggested that the structure of the mineralization, and temporal and elemental correlations were the result of syn-magmatic accumulation and re-distribution of precious metals in an upward migrating and stratified mush zone in the floor of the magma chamber (Nielsen et al., 2015). The diachronous distribution of PGE and Au (Fig. 3a;Supplementary Data EA1 and EA4) is regarded by these authors as critical and taken to demonstrate redistribution of precious metals up the succession of MLs. They also argued that initial accumulation of precious metals in the floor mush was due to scavenging of precious metals by immiscible droplets of sulphide melt (small and in suspension) in the crystallising and evolving bulk melt that convection carried along the mushy roof of the magma chamber. At immiscibility between Fe-rich and Si-rich silicate melts in this mush and after the buoyancy-driven loss of the Si-rich conjugate, sulphide solubility increased in the remaining Fe-rich silicate melt. In consequence, already formed droplets of sulphide melt dissolved in the Fe-rich conjugate. The dense Fe-rich melt mixed with crystals and was carried to the floor of the magma chamber by convection. The precious metals were subsequently redistributed into MLs during the upward migration of the mushy crystallization zone in the floor of the magma chamber. Precious metals were supplied to the floor during the crystallization of ML0 and ML1·1 (LPGEM) only, as MLs above ML1·1 (UPGEM) near the margins of the intrusion are Cu-rich and only have traces of precious metals (Fig. 3a, Nielsen et al., 2015). A simple upward re-deposition of precious metals that were initially evenly distributed across the floor of the magma chamber cannot account for the total of precious metals accumulated in the central parts of the mineralization (Fig. 8, Table 6). The (g*w) of the precious metal reaches a maximum of ∼50 with ∼38 for PGE and ∼12 for Au in the centre, compared to 18–20 for PGE and 4 for Au closer to the margins (Fig. 10, Table 6). These results show that distribution of precious metals in the mineralization requires pre-concentration of precious metals in the central part of the bowl in the floor of the magma chamber. Ponding of precious metal bearing silicate melt (e.g. Holwell & Keays, 2014) could potentially account for the increasingly larger (g*w) numbers and a stratigraphic separation of >40 m between PGE and Au rich mineralization levels (Fig. 3a) in the centre of the mineralization. A bowl constructed from observed Au peaks, however, would in its centre be >600 m deep relative to the rim and the ponded magma should, therefore, itself have the shape of a > 600 m deep bowl. However, the distribution of ponded melts is controlled by gravity and would not form a > 600 m deep bowl with walls thinning from forty to a few metres from the centre of the bowl and >600 m up the walls of the magma chamber. An alternative suggestion could be a bowl formed by accumulation of precious metals over thousands of years with an increase in Au, and accumulation gradually displaced to more central parts of the magma chamber. Such a scenario would, however, be in conflict with overlapping PGE and Au peaks farthest away from the centre of the mineralization and would not explain why gabbro successions that formed at the same time (same ML layer and same Pd-level, e.g. Pd2b in Fig. 3a) contain PGE in the centre, Cu-sulphides at the margin of the intrusion and Au inbetween (Fig. 3a). The explanation we suggest for the increase observed in (g*w) numbers in the central parts of the mineralization is that slurries composed of Fe-rich silicate melt and solids descended along the walls of the magma chamber and decelerated as they reached the bowl-shaped floor. During deceleration the slurries deposited carried-along solids (crystals) while Fe-rich and dense silicate melt with its load of dissolved precious metals continued into the deeper part of the bowl-shaped floor. In itself, the bowl-shape is witness to a continued accumulation of solids entrained in convection currents descending along the concomitant walls of the magma chamber. The preservation of near-constant thicknesses of MLs in the Triple Group, despite the accumulation of entrained solids near the walls, may seem as a contradiction, but is apparently the consequence of the dynamic stratification process in the MLs of the Triple Group (see section: Macro-rhythmic layers and compositional subdivision). At the time of formation of the mineralization, the density of the mush melt was less than that of pyroxene and Fe–Ti-oxides, but greater than that of plagioclase (see Fig. 26 in Nielsen et al., 2015), and in situ sorting of solids and melt(s) stratified the mush into a succession of proto-MLs in the upward migrating mush zone. The stratification process is referred to by Nielsen & Bernstein (2009) and Nielsen et al. (2015) as self-stratification, with thicknesses and cyclicity controlled by density contrasts and rheology (e.g. McKenzie, 2011, Bons et al., 2015). The density-controlled sorting does not distinguish between crystals transported from the roof zone to the floor or those that crystallized in situ. The crystals would be subjected to the same dynamic forces irrespective of their place of origin in the magma chamber, and the MLs would consequently have near-constant thicknesses across the floor of the magma chamber. Re-distribution of precious metals in upward migrating mush melt Paragenetic evidence for redistribution Concentrates of precious metal grains from drill cores near the margins of the intrusion (Table 4, Fig. 7) have parageneses that are characterized by sulphides, arsenides, plumbides, tellurides, etc., and a host rock with hydrous silicate phases. In comparison, PGE parageneses in central drill cores are dominated by skaergaardite (PdCu), and gabbros that contain no hydrous silicate phases (Fig. 7, Table 4, Fig. 15a). The more varied parageneses at the margins were suggested by Nielsen et al. (2015) to reflect the trapping in the mushy floor of melt carrying volatiles in addition to precious metals, Cu, and elements such as Pb, S, As, and Te. Trapping of residual melt and volatiles in the gabbros near the margins was suggested to lead to re-equilibration of precious metal phases to low temperature phases such as Au3Cu (see, Holwell et al., 2016). Equilibration of the precious metal phases with residual and hydrous mush melt, however, does not explain why gabbros close to the western contact against Precambrian basement (Fig. 1) are characterized by Pd-sulphides (vysotskite and vasilite) and arsenide-rich parageneses, whereas the gabbros at the eastern contact against Palaeogene basalts (Fig. 1) are characterized by the plumbide zvyagintsevite (Pd3Pb) and arsenide rich Pd-parageneses (Table 4, Fig. 7), and why the same mineralization level in the centre is entirely dominated by skaergaardite (PdCu). Skaergaardite is found as euhedral crystals within immiscible droplets of Cu-rich sulphide melt that are trapped in liquidus crystals of Fe–Ti-oxides (Godel et al., 2014; Nielsen et al., 2015). These Cu-rich sulphide droplets were already depleted in most other elements, with distribution coefficients between sulphide and silicate liquid lower than those of PGE, but were enriched in precious metals due to relative loss of the Cu-rich sulphide (Fig. 11). The process predated or was contemporaneous with the crystallization of its host mineral. A further example of paragenetic variation across the intrusion is provided by drill cores 90–18 and 90–24. In drill core 90–18, Pd-sulphides are the major PGE-bearing phases, whilst in drill core 90–24, located only 2·2 km away, the paragenesis is dominated by skaergaardite and related (Pd, Cu) alloys with only traces of precious metal sulphide. The two drill cores exhibit parallel variations in bulk rock PGE concentrations (Fig. 3a), and yet they represent very different compositional environments. The precious metal parageneses are specific to a given drill core or sector of the intrusion and are found: (i) inside sulphide droplets; (ii) as grains protected in and between minerals crystallized from the mush melt (Godel et al., 2014; Rudashevsky et al., 2014, 2015; Nielsen et al., 2015); and (iii) as un-protected grains in related to the sub-liquidus paragenesis of the gabbros (Nielsen et al., 2003a, 2003b, 2003c, 2003d, 2003e; Rudashevsky & Rudashevsky, 2005a, 2005b, 2006a, 2006b; Rudashevsky et al., 2009a, 2009b, 2010a, 2010b, 2010c, 2010d, 2012a, 2012b, 2012c, 2012d, 2012e, 2012f, 2012g, 2012h, 2012i, 2014, 2015). The observed precious metal paragenesis of a given sample, therefore, reflects the local composition of melt in the mushy floor and the melt in which the immiscible sulphide droplets formed and equilibrated. The observed paragenetic variations could not result from flow of melt or mushes across the floor of the solidifying magma chamber and cannot be explained by accumulation controlled by bulk liquid processes. Both would lead to more uniform parageneses in the same mineralization levels and MLs. Instead, the paragenetic variations, just as the 3 D distribution of precious metals in the mineralization (Figs 8–10) and upward decrease in Pd/Pt in mineralization levels (see Fig. 13 in Nielsen et al., 2015), support crystallization and equilibration in local geochemical environments that are repeated up the succession of MLs. The paragenetic variations in the precious metal mineralogy indicate that syn-magmatic processes in the silicate mush of the crystallization zone vary laterally across the floor of the magma chamber in response to local compositional variations, and that individual MLs operated as semi-closed crystallization and fractionation chambers with limited lateral communication. Compositional evidence for upward redistribution In the melt-rich zones of the MLs in the floor of the magma chamber, the crystallization is argued by Nielsen et al. (2015) to have driven the mush melt to sulphide saturation (leading to the formation of tiny, suspended immiscible droplets of sulphide melt and subsequently to the two-liquid field between Fe-rich and Si-rich silicate liquids. Reactions between the pre-existing silicate paragenesis and immiscible Fe-rich melt (Holness et al., 2011) would not have taken place unless the low-density granophyric conjugate was lost from the mush. The resorption of the silicate host may be comparable to dissolution processes more recently proposed in relationship to discordant chromite seams in the Bushveld Complex (Latypov et al., 2017). The loss of the granophyric conjugate also led to reaction between the remaining Fe-rich melt and un-protected droplets of sulphide melt. During dissolution, Pd was preferentially retained in the droplets of Cu-sulphide melt over the other precious metals, since Pd has the largest sulphide/silicate melt partition coefficient (Makovicky, 2002; Naldrett, 2011). Conversely, precious metals with lower D-values (Pt and especially Au) became available for convection- and crystallization-driven redistribution upwards to the overlying MLs (Nielsen et al., 2015, see also Holness et al., 2017a; Vukmanovic et al., 2018). The fractionation of precious metals that is argued to have occurred during dissolution of droplets of sulphide melt is a critical step in the proposed model and is supported by the relative depletion of Au in the lower and central parts of the mineralization (Figs 8–10) and in particular in ((Pd+Pt+Au)/Cu)*1000) in Pd5 (Fig. 11). In drill core 08–35 A the ratio reaches a maximum of ∼16 in the Pd5 peak (argued by Holwell et al., 2016 to be the highest among sulphide mineralizations), but in the interior of the mineralization, up to 30 (Fig. 11). Nielsen et al. (2015) used petrographic observations outlined in Holness et al. (2011) to explain the reason for this variation. Specifically, near the margins of the intrusion, Middle Zone gabbros contain paired pockets of solidified Fe-rich and Si-rich liquids (incomplete separation of immiscible melts), whereas gabbros from equivalent stratigraphic levels near the centre of the intrusion contain abundant reactive symplectites, which are a by-product of reaction between immiscible Fe-rich silicate liquid and already crystallized gabbro. These observations indicate that the separation of the conjugate silicate melts was completed in the centre of the intrusion and allowed for reaction and dissolution of unprotected, precious metal-enriched sulphide droplets and the corresponding rise in (Pd+Pt+Au)/Cu) x 1000) in bulk rock compositions in the centre of the mineralization. The marked increase in precious metals during dissolution and loss of Cu is highlighted by the PCA. PC1 shows the separation of PGE (Pd and Pt) and Au from all other elements and Cu in all mineralization levels from Pd6 to Pd1/Au (Supplementary Data EA14). As argued by Nielsen et al. (2015), the mineralization levels are melt-rich intervals in a stratified crystal mush in which sulphide saturation was followed by dissolution of first-formed droplets of sulphide melt. All intervals between mineralization levels are anti-correlated to PC1 and consequently they have no accumulation of droplets of sulphide melt formed due to sulphide saturation. Finally, in the central parts of the intrusion, the UPGEM (Pd3b, Pd3a, Pd2b, Pd2a, Pd1, Pd1/Au and UAuM in ML1·2, Ml2, Ml2·1 and ML2·2) account for ∼40% of the PGE over the full depth of the mineralization (Pd6-UAuM) and for ∼90% of that for Au (Table 6). This suggests the loss of PGE and Au from LPGEM in the lower parts of the mineralization, and redistribution of these metals to the interval spanned by UPGEM (Figs 8–10). In combination, the: (i) short-lived co-accumulation of PGE and Au in LPGEM (in Middag and Midnat Buttresses); (ii) augmented dissolution of sulphide in the central parts of the intrusion (Fig. 11); and (iii) depletion in PGE, Au and Cu in central and lowermost parts of the mineralization (Figs 8–10) demonstrate upward transport and fractionation of precious metals in a mushy crystallization zone in the bowl-shaped floor of the magma chamber. Late mineralization along grain boundaries In the central parts of the mineralization, the upwards increase of Au in immiscible sulphide melt is indicated by increasing Au-substitution into skaergaardite (PdCu) (Rudashevsky et al., 2014) and increasing proportions of tetra-auricupride (AuCu) in the uppermost Pd-level of any given drill core (Tables 3 and 4; Holwell & Keays, 2014; Nielsen et al., 2015). The clear negative value of PC1 at Pd1 (Supplementary Data EA14) suggests sulphide saturation in mineralization level Pd1, followed by dissolution and enrichment of the remaining sulphide droplets in gold. In addition to sulphide saturation related Au-accumulation in tetra auricupride in Pd1 and to lesser extend in Pd1/Au, all central drill cores have gold added to the uppermost Au-rich mineralization levels. Gold is added along grain boundaries in a mineralization event referred to as UAuM (Nielsen et al., 2015). On the basis of petrography and peak concentrations at unconstrained elevations above Pd1 (see Fig. 6) UAuM is argued (Godel et al., 2014; Nielsen et al., 2015) to be the result of mineralization processes caused by migration of residual silicate melts and fluids in already crystallized gabbro. We interpret the variance explained by PC2 as the result of this process. PC2 of the PCA (Table 8; Supplementary Data Fig. EA14-3) is closely related to the Pd1 and especially to Pd1/Au peaks and above in the gold-rich top of the mineralization in drill core 90–22 and tops ML0 and ML1·1. PC2 is anti-correlated to Cu, but correlated to P, Pb, Ce, which all are incompatible elements, at least until apatite starts to crystallise from the bulk liquid or mush melts. The marked compositional differences between the gold rich Pd1 and Pd1/Au levels and all other precious metal rich mineralization levels is highlighted in Fig. 14 that shows elevated P2O5 in most samples from Pd1 and Pd1/Au. PC2 is, therefore, on the basis of: (i) stratigraphic association; (ii) the occurrence of gold unattached to sulphide along grain boundaries; and (iii) the common occurrence of low temperature Au3Cu (Holwell et al., 2016) rather than high temperature tetra-auricupride (AuCu; Bird et al., 1991), representing the separate mineralization event referred to as (UAuM). Contrary to the conclusions of Holwell et al. (2016), the anti-correlation between Au and Cu in PC2 negates that all gold in the top of the Skaergaard PGE–Au mineralization was accumulated due to sulphide saturation and accumulation of immiscible droplets of Cu-rich sulphide melt, and negates that the Skaergaard PGE–Au Mineralization is a conventional ‘offset reef’ type mineralization. Fig. 14 Open in new tabDownload slide Correlation between FeO* (FeO total) and P2O5 (wt %) in the continuous 25-cm sample profile from the base of the mineralization below Pd6 to the Cu-rich gabbros above the precious metal rich mineralization levels in drill core 90-22. Red, all samples (# 16) related to the gold-bearing Pd1 and Au/Pd1 mineralization levels and sections of core with transgressive and late felsic veins. Blue, all other samples (# 243) in the profile demonstrate a negative correlation between FeO* and P2O5. Fig. 14 Open in new tabDownload slide Correlation between FeO* (FeO total) and P2O5 (wt %) in the continuous 25-cm sample profile from the base of the mineralization below Pd6 to the Cu-rich gabbros above the precious metal rich mineralization levels in drill core 90-22. Red, all samples (# 16) related to the gold-bearing Pd1 and Au/Pd1 mineralization levels and sections of core with transgressive and late felsic veins. Blue, all other samples (# 243) in the profile demonstrate a negative correlation between FeO* and P2O5. The association of PC2 to incompatible elements and the enrichment in the tops of ML0 and ML1·1 (EA14) strongly support that the UAuM-type redistribution of gold was related to reactions and deposition from the residual of the Fe-rich melt ponded and crystallized within the MLs. Examination of the Au-rich mineralization levels in exposures at Toe of Forbindelsesgletscher (ToF, Fig. 1a) and in drill core 90–22 reveals that the Au-bearing gabbros are rusty due to oxidation of Fe. The gabbroic host is in all other Pd-levels extremely fresh and shows no signs of alteration under the microscope (Fig. 15a), whereas gabbros rich in gold are affected by hydration, and extensive alteration and recrystallization of clinopyroxene and the presence of H2O-bearing silicates (Fig. 15b). Fig. 15 Open in new tabDownload slide Thin section photomicrographs of mineralized gabbros in the Skaergaard mineralization in drill core 90-22: (a) Pd5, peak of main PGE mineralization level (1033·25 m), and (b) Pd1, base of main Au-mineralized interval (993·5 to 990 m) showing extensive late magmatic recrystallization and alteration of clinopyroxene. Dissolution of plagioclase (greyish tones) and clinopyroxene (vivid colours) of the liquidus paragenesis and crystallization of rims, and symplectites between late crystallized masses of Fe–Ti oxides (black) and the liquidus paragenesis are common to both mineralization levels. Further descriptions in Holness et al. (2011) and in Nielsen et al. (2015). Fig. 15 Open in new tabDownload slide Thin section photomicrographs of mineralized gabbros in the Skaergaard mineralization in drill core 90-22: (a) Pd5, peak of main PGE mineralization level (1033·25 m), and (b) Pd1, base of main Au-mineralized interval (993·5 to 990 m) showing extensive late magmatic recrystallization and alteration of clinopyroxene. Dissolution of plagioclase (greyish tones) and clinopyroxene (vivid colours) of the liquidus paragenesis and crystallization of rims, and symplectites between late crystallized masses of Fe–Ti oxides (black) and the liquidus paragenesis are common to both mineralization levels. Further descriptions in Holness et al. (2011) and in Nielsen et al. (2015). The compositional fingerprint of UAuM is reminiscent of IOCG deposits sensu strictu as defined by Groves et al. (2010). We add that a volatile-bearing environment is evidenced by the common occurrence of hydrous phases in immiscible sulphide droplets as well as silicate melt inclusions (Godel et al., 2014; Holwell et al., 2016; Nielsen et al., 2015). The bulk melt of the Skaergaard intrusion was already evolved at the time of emplacement (Mg# = 0·43; Nielsen et al., 2009), and only ∼10 vol. % of the bulk liquid remained when immiscibility between Fe-rich and Si-rich melts was reached (Nielsen et al., 2015). At the time the mineralization formed, volatiles would have been concentrated in the melt of the mushy crystallization zones, and likely amounted to several wt % (dependent on loss). It would be most surprising if the residual bulk magma was not saturated in volatiles, and that no syn- to late-crystallization redistributions in the presence of a free volatile phase followed the inward migrating crystallization zone (Nielsen, 2016). Sulphide-saturated mush melt in CuM In PC3, the positive correlation between Au, Fe, Cu and Pb and anti-correlation between these elements and P and incompatibles, suggests a return to sulphide saturation. Large magnitude PC3 values occur predominantly in CuM (Supplementary Data Fig. EA14-3). Samples from CuM are characterized by Cu-rich sulphides interstitially between grains of the host rocks and described as an orthomagmatic mineralization (Holwell et al., 2016; Supplementary Data EA12). They apparently formed due to sulphide saturation, just as Pd-levels but without subsequent dissolution. The lack of dissolution accounts for the geochemical distinction between PC1 and PC3, seemingly because the immiscible Fe-rich melts that at CuM time ponded in the mushy floor were already sulphide-saturated as they formed (cf. Nielsen et al., 2015). Sulphide saturation and subsequent dissolution (PC1) is restricted to mineralization levels and PC3 seem to be of no importance in intervals between mineralization levels in LPGEM and UPGEM (Supplementary Data Fig. EA14-3). Consequently, no support is found in the PCA for sulphide saturation and accumulation of immiscible droplets of sulphide melt from the bulk liquid of the Skaergaard intrusion during the formation of its PGE–Au mineralization. Sulphide saturation was, as concluded in Nielsen et al. (2015), restricted to evolved melt in the mushy crystallization zones of the magma chamber. CONCLUSIONS The genesis of the Skaergaard PGE–Au Mineralization is very complex and results from in situ fractionation, sulphide saturation and immiscibility between Fe-rich and Si-rich silicate melts in mushy crystallization zones in the magma chamber. The model, first developed in Nielsen et al. (2015) on the basis of a structural model that compares the mineralization to a stack of gold-rimmed saucers with upward decreasing diameter, is here further supported by: i) 3 D paragenetic variations of precious metal minerals; ii) 3 D distribution of precious metals in the mineralization; and iii) PC1-PC3 of the principal component analysis. The precious metals concentrated in droplets of sulphide melt in the bulk melt circulated to the mushy roof. They dissolved and were entrained in immiscible Fe-rich mush melt that descended to the floor of the magma chamber. Flow differentiation in mushes along the bowl-shaped floor concentrated melt with its cargo of dissolved precious metals in the deepest and central parts of the bowl shaped floor. The co-accumulation of PGE and Au in more distant parts of the mineralization, the diachronous distribution of precious metals, and the lateral variations in bulk composition and precious metal parageneses in the >600 m deep bowl exclude, in our view, a classic offset-reef type model for the origin of the mineralization. In our model, the precious metals were re-distributed up the MLs of the Triple Group due to repetition in stratified mush of: i) fractionation of mush melt; ii) density stratification; iii) sulphide saturation in remaining mush melt; iv) immiscibility between Fe-rich and Si-rich silicate liquids; iv) loss of Si-rich conjugate; vi) reaction and equilibration between formed droplets of sulphide melt and ponded Fe-rich melt; vii) fractionation of Fe-rich melt and loss of its residual taking dissolved PGE, P, REE, HFSE, Cu, Au and volatiles along to the mush of the overlying ML. The fully developed Skaergaard PGE–Au Mineralization is neither a classic reef nor an offset reef type precious metal deposit, but a three-dimensional distribution of precious metals and the result of prolonged crystallization, syn-depositional and syn-magmatic processes in a crystal mush. Therefore we argue that the mineralization should not be referred to as the ‘Platinova Reef’, but to the ‘Skaergaard PGE–Au Mineralization’ (Nielsen et al., 2015) to avoid little-supported associations to reef-type mineralizations to which its structure and genesis cannot and should not be compared. The Skaergaard PGE–Au Mineralization is a mineralization type in its own right. ACKNOWLEDGEMENTS The investigations would not have been possible without the dedicated support of Leif Thorning, former head of Department at the Geological Survey of Denmark and Greenland. Most helpful reviews by Jim Mungall, Roger Scoon, David Holwell and Allan Wilson of earlier version of the manuscript are much appreciated. Susanne Rømer is thanked for the preparation of illustrations. FUNDING These investigations have been supported entirely by the Geological Survey of Denmark and Greenland over a period of more than 10 years. REFERENCES Aitchison J. ( 1982 ). The statistical analysis of compositional data (with discussion) . Journal of the Royal Statistical Society: Series B (Methodological) 44 , 139 – 177 . Google Scholar Crossref Search ADS WorldCat Aitchison J. ( 1986 ). The statistical analysis of compositional data. In: Monographs on Statistics and Applied Probablity . Chapman & Hall Ltd , London, UK; p. 416 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Aitchison J. , Egozcue J. J. ( 2005 ). Compositional Data Analysis: Where are we and where should we be heading? Mathematical Geology 37 , 829 – 850 . Google Scholar Crossref Search ADS WorldCat Albarède F. ( 1995 ). Introduction to Geochemical Modeling . Cambridge, England : Cambridge University Press , p. 543 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC Andersen J. C. Ø. ( 2006 ). Postmagmatic sulfur loss in the Skaergaard Intrusion: implications for the formation of the Platinova Reef . Lithos 92 , 198 – 221 . Google Scholar Crossref Search ADS WorldCat Andersen J. C. Ø. , Rasmussen H., Nielsen T. F. D., Rønsbo J. C. ( 1998 ). The Triple Group and the Platinova gold and palladium reefs in the Skaergaard Intrusion: stratigraphic and petrographic relations . Economic Geology 93 , 488 – 509 . Google Scholar Crossref Search ADS WorldCat Barnes S. J. , Holwell D. A., Le Vaillant M. ( 2017 ). Magmatic sulfide ore deposits . Elements 13 , 89 – 95 . Google Scholar Crossref Search ADS WorldCat Bernstein S. , Nielsen T. F. D. ( 2004 ). Chemical stratigraphy in the Skaergaard intrusion. GEUS Report 2004/123, 31 pp. + CD. Bird D. K. , Brooks C. K., Gannicott R. A., Turner P. A. ( 1991 ). A gold-bearing horizon in the Skaergaard Intrusion . Economic Geology 86 , 1083 – 1092 . Google Scholar Crossref Search ADS WorldCat Bons P. D. , Baur A., Elburg M. A., Lindhuber M. J., Marks M. A. W., Soesoo A., van Milligen B. P., Walte N. P. ( 2015 ). Layered intrusions and traffic jams . Geology 43 , 71 – 74 . Google Scholar Crossref Search ADS WorldCat Boudreau A. ( 2004 ). PALLADIUM – A program to model the chromatographic separation of the platinum-group elements, base metals and sulfur in a solidifying igneous crystal pile . The Canadian Mineralogist 42 , 393 – 403 . Google Scholar Crossref Search ADS WorldCat Boudreau A. E. , Meurer W. P. ( 1999 ). Chromatographic separation of the platinum-group elements, gold, base metals and sulfur during degassing of a compacting and solidifyingcrystal pile . Contributions to Mineralogy and Petrology 134 , 174 – 185 . Google Scholar Crossref Search ADS WorldCat Cabri L. J. ( 2004a ). New developments in process mineralogy of platinum-bearing ores. In: Proceedings of the Canadian Mineral Processors, 36th Annual Meeting, Ottawa, Canada, pp. 189 – 198 . Cabri L. ( 2004b ). A mineralogical evaluation of two samples for Skaergaard Minerals Corp, License No. 2000/10. In: Skaergaard Minerals Corporation (2004): Report on exploration activities on Skaergaard Mineral license during 2003, EA 4, 65pp. (in archive of the Geological Survey of Denmark and Greenland, GRF no. 21840; available on-line). Cabri L. J. , Beattie M., Rudashevsky N. S., Rudashevsky V. N. ( 2005 ). Process mineralogy of Au, Pd and Pt ores from the Skaergaard intrusion, Greenland, using new technology . Minerals Engineering 18 , 887 – 897 . Google Scholar Crossref Search ADS WorldCat Cattell R. B. ( 1966 ). The scree test for the number of factors . Multivariate Behavioral Research 1 , 245 – 276 . Google Scholar Crossref Search ADS PubMed WorldCat Godel B. , Rudashevsky N. S., Nielsen T. F. D., Barnes S. J., Rudashevsky V. N. ( 2014 ). Constraints on the origin of the Skaergaard intrusion Cu-Pd mineralization: insights from high-resolution X-ray computed tomography . Lithos 190-191 , 27 – 36 . Google Scholar Crossref Search ADS WorldCat Groves D. I. , Bierlein F. P., Meinert L. D., Hitzman M. W. ( 2010 ). Iron oxide copper-gold (IOCG) deposits through Earth History: Implications for origin, lithospheric setting, and distinction from other epigenetic iron oxide deposits . Economic Geology 105 , 641 – 654 . Google Scholar Crossref Search ADS WorldCat Hamelin C. , Dosso L., Hanan B. B., Moreira M., Kositsky A. P., Thomas M. Y. ( 2011 ). Geochemical portray of the Pacific Ridge: new isotopic data and statistical techniques . Earth and Planetary Science Letters 302 , 154 – 162 . Google Scholar Crossref Search ADS WorldCat Hanghøj K. ( 2005 ). Report on exploration activities in 2004 on Skaergaard license no. 2005/09. Internal report, Skaergaard Minerals., Corp, 51 pp, 1 EA: Geochemical analyses, core recovery, drill hole survey, 175 pp. (In archive of the Geological Survey of Denmark and Greenland, report GRF 21895; (available online at www.greenmin.gl). Holness M. B. , Nielsen T. F. D., Tegner C. ( 2017 ). The Skaergaard intrusion of east Greenland: paradigms, problems and new perspectives . Elements 13 , 391 – 396 . Google Scholar Crossref Search ADS WorldCat Holness M. B. , Stripp G., Humphreys M. C. S., Veksler I. V., Nielsen T. F. D., Tegner C. ( 2011 ). Silicate liquid immiscibility within the crystal mush: late-stage magmatic microstructures in the Skaergaard intrusion, east Greenland . Journal of Petrology 52 , 175 – 222 . Google Scholar Crossref Search ADS WorldCat Holness M. B. , Tegner C., Nielsen T. F. D., Charlier B. ( 2017 ). The thickness of the mushy layer on the floor of the Skaergaard magma chamber at apatite saturation . Journal of Petrology 58 , 909 – 932 . Google Scholar Crossref Search ADS WorldCat Holness M. B. , Vukmanovic Z., Mariani E. ( 2017 ). Assessing the role of compaction in the formation of adcumulates: a microstructural perspective . Journal of Petrology 58 , 643 – 673 . Google Scholar Crossref Search ADS WorldCat Holwell D. , Keays R. R. ( 2014 ). The formation of low-volume, high-tenor magmatic PGE–Au sulfide mineralization in closed systems: evidence from precious and base metal geochemistry of the Platinova reef, Skaergaard Intrusion, East Greenland . Economic Geology 109 , 387 – 406 . Google Scholar Crossref Search ADS WorldCat Holwell D. A. , Barnes S. J., Le Vaillant M., Keays R. R., Fisher L. A., Prasser R. ( 2016 ). 3D textural evidence for the formation of ultra-high tenor precious metal bearing sulfide microdroplets in offset reefs: an extreme example from the Platinova Reef, Skaergaard Intrusion, Greenland . Lithos 256-257 , 55 – 74 . Google Scholar Crossref Search ADS WorldCat Holwell D. A. , Keays R. R., McDonald I., Williams M. R. ( 2015 ). Extreme enrichment of Se, Te, PGE and Au in Cu sulfide microdroplets: evidence from LA-ICP-MS analysis of sulfidesin the Skaergaard Intrusion, east Greenland . Contributions to Mineralogy and Petrology 170 , 53 . Google Scholar Crossref Search ADS WorldCat Irvine T. N. ( 1970 ). Heat transfer during solidification of layered intrusions. I. sheets and sills . Canadian Journal of Earth Sciences 7 , 1031 – 1061 . Google Scholar Crossref Search ADS WorldCat Irvine T. N. , Andersen J. C. Ø., Brooks C.K. ( 1998 ). Included blocks (and blocks within blocks) in the Skaergaard Intrusion: geological relations and the origins of rhythmic modally graded layers . Geological Society of America Bulletin 110 , 1398 – 1447 . Google Scholar Crossref Search ADS WorldCat Keays R. R. , Tegner C. ( 2016 ). Magma chamber processes in the formation of the low-sulphide magmatic Au–PGE mineralization of the Platinova Reef in the Skaergaard intrusion, East Greenland . Journal of Petrology 56 , 2319 – 2340 . Google Scholar Crossref Search ADS WorldCat Kuo C. S. ( 2007 ). The Mineral Industries of Denmark, the Faroe Islands, and Greenland. US Geological Survey: Mineral Yearbook 2005 , 8.1 – 8.4 . Latypov R. , Chistyakova S., Mukherjee R. ( 2017 ). A novel hypothesis for origin of massive chromitites in the Bushveld igneous complex . Journal of Petrology 58 , 1899 – 1940 . Google Scholar Crossref Search ADS WorldCat Laznicka ( 2006 ). Giant Metallic Deposits: Future Sources of Industrial Metals . Berlin, Heidelberg : Springer-Verlag , p. 732 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Le Maitre R. W. ( 1982 ). Numerical Petrology: Statistical Interpretation of Geochemical Data . Elsevier , p. 281 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McBirney A. R. ( 1989 ). The Skaergaard layered series: 1. Structure and average compositions . Journal of Petrology 30 , 363 – 397 . Google Scholar Crossref Search ADS WorldCat McBirney A. R. ( 1996 ). The Skaergaard Intrusion. In: Cawthorn R. G. (ed.) Layered Intrusions . Amsterdam, Holland: Elsevier , pp. 147 – 180 . Google Scholar Crossref Search ADS Google Preview WorldCat COPAC McDonald A. M. , Cabri L. J., Rudashevsky N. S., Stanley C. J., Rudashevsky V. N., Ross K. C. ( 2008 ). Nielsenite, PdCu3, a new platinum group intermetallic mineral species from the Skaergaard intrusion, Greenland . The Canadian Mineralogist 46 , 709 – 716 . Google Scholar Crossref Search ADS WorldCat McKenzie D. ( 2011 ). Compaction and crystallization in magma chambers: toward a model for the Skaergaard intrusion . Journal of Petrology 52 , 905 – 930 . Google Scholar Crossref Search ADS WorldCat Makovicky E. ( 2002 ). Ternary and quaternary phase systems with PGE. In Cabri J. P. (ed.) The Geology, Geochemistry, Mineralogy and Mineral Beneficiation of Platinum-Group Elements . Montreal : Canadian Institute of Mining, Metallurgy and Petroleum, Montreal , pp. 131 – 175 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Miller J. D. , Andersen J. C. Ø. ( 2002 ). Attributes of Skaergaard-type PGE reefs. In: Boudreau, A. (ed.) Extended Abstracts, 9th Internat. Platinum Conference, Billings, Montana, USA, 2125 July 2002. pp. 305 – 308 . Morse S. A. ( 2011 ). The fractional latent heat of crystallizing magmas . American Mineralogist 96 , 682 – 689 . Google Scholar Crossref Search ADS WorldCat Naldrett A. J. ( 2011 ). Fundamentals of magmatic sulphide deposits . Reviews in Economic Geology 17 , 1 – 50 . Google Scholar OpenURL Placeholder Text WorldCat Nielsen T. F. D. ( 2004 ). The shape and volume of the Skaergaard intrusion, Greenland: implications for mass balance and bulk composition . Journal of Petrology 45 , 507 – 530 . Google Scholar Crossref Search ADS WorldCat Nielsen T. F. D. ( 2016 ). In situ fractionation and inward migration of the solidification front in the Skaergaard intrusion , East Greenland. Geological Survey of Denmark and Greenland Bulletin 35 , 59 – 62 . Google Scholar OpenURL Placeholder Text WorldCat Nielsen T. F. D. , Andersen J. C., Brooks C. K. ( 2005 ). The Platinova reef of the Skaergaard intrusion. In: Mungal J. E. (ed.) Exploration for Platinum Group Element Deposits , 35 . Ottawa : Mineralogical Association of Canada, Short Course , pp. 431 – 455 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nielsen T. F. D. , Andersen J. C., Holness M. B., Keiding J. K., Rudashevsky N. S., Rudashevsky V. N., Salmonsen L. P., Tegner C., Veksler I. V. ( 2015 ). The Skaergaard PGE and gold deposit: the result of in situ fractionation, sulphide saturation, and magma chamber scale precious metal redistribution by immiscible Fe-rich malt . Journal of Petrology 56 , 1643 – 1976 . Google Scholar Crossref Search ADS WorldCat Nielsen T. F. D. , Bernstein S. ( 2009 ). Chemical stratigraphy of the Triple Group and mineralization of the Skaergaard intrusion: insight in the crystallization process. AGU Fall meeting 2009, Abstract V21A-1956. Nielsen T. F. D. , Olsen S. D., Stensgaard B. M. ( 2009 ). Developing a 3-D model for the Skaergaard intrusion in East Greenland: constraints on structure, mineralization and petrogenetic models . Geological Survey of Denmark and Greenland Bulletin 17 , 61 – 64 . Google Scholar OpenURL Placeholder Text WorldCat Nielsen T. F. D. , Rasmussen H., Rudashevsky N. S., Kretser Y. L., Rudashevsky V. N. ( 2003a ). PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 1: sample 90-23A, 807. GEUS report 2003/47, 46 pp. (available online at www.greenmin.gl) Nielsen T. F. D. , Rasmussen H., Rudashevsky N. S., Kretser Y. L., Rudashevsky V. N. ( 2003b ). PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 2: sample 90–24, 1057. GEUS report 2003/48, 83 pp. (available online at www.greenmin.gl) Nielsen T. F. D. , Rasmussen H., Rudashevsky N. S., Kretser Y. L., Rudashevsky V. N. ( 2003c ). PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 3: sample 90–18, 1010. GEUS report 2003/52, 67 pp. (available online at www.greenmin.gl) Nielsen T. F. D. , Rasmussen H., Rudashevsky N. S., Kretser Y. L., Rudashevsky V. N. ( 2003d ). PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 4: sample 90-23A, 806. GEUS report 2003/53, 40 pp. (available online at www.greenmin.gl) Nielsen T. F. D. , Rasmussen H., Rudashevsky N. S., Kretser Y. L., Rudashevsky V. N. ( 2003e ). PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 5: sample 90-23A, 808. GEUS report 2003/54, 37 pp. (available online at www.greenmin.gl) Nielsen T. F. , Tegner C., Thy P., Fonseca A. K., Jakobsen J. K., Kristensen M., Simpson J. A., Brooks C., Kent A. J., Peate D. W., Lesher C. E. ( 2000 ). Retrieval of Platinova drill cores: a new Skaergaard initiative. Eos Trans. AGU 81 (48), Fall Meet, Suppl. abstract V21E-15 Pawlowsky-Glahn V. , Egozcue J. J. ( 2006 ). Compositional data and their analysis: an introduction . Geological Society, London, Special Publications 264 , 1 – 10 . Google Scholar Crossref Search ADS WorldCat Prendergast M. D. ( 2000 ). Layering and precious metals mineralization in the Rincón del Tigre complex, Eastern Bolivia . Economic Geology  95 , 113 – 130 . Google Scholar Crossref Search ADS WorldCat Rudashevsky N. S. , Kretser Y. L., Rudashevsky V. N., Nielsen T. F. D. ( 2009a ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 7, sample 90–18 958. GEUS report 2009/68, 16 pp plus tables and plates. Rudashevsky N. S. , Kretser Y. L., Rudashevsky V. N., Nielsen T. F. D. ( 2009b ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 6, sample 90-23A 798. GEUS report 2009/54, 17 pp plus tables and plates. Rudashevsky N. S. , McDonald A. M., Cabri L. J., D. Nielsen T. F., Stanley C. J., Kretzer Y. L., Rudashevsky V. N. ( 2004 ). Skaergaaardite, PdCu, a new platinum group intermetallic mineral from the Skaergaard intrusion, Greenland . Mineralogical Magazine 68 , 615 – 632 . Google Scholar Crossref Search ADS WorldCat Rudashevsky N. S. , Rudashevsky V. N. ( 2005a ). Gold, PGE, and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Sample 90–24, 1018. Part 29, sample 90–24 1018. Unpublished GEUS report, 19pp plus tables and plates (available on request). Rudashevsky N. S. , Rudashevsky V. N. ( 2005b ). PGE, gold, and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Sample 90–24, 1022. Part 28, sample 90–24 1022. Unpublished GEUS report, 21pp plus tables and plates (available on request). Rudashevsky N. S. , Rudashevsky V. N. ( 2006a ). PGE, gold, and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Sample 90–24, 1045N. Part 21, sample 90–24 1045N, Unpublished GEUS report, 17pp plus tables and plates (available on request). Rudashevsky N. S. , Rudashevsky V. N. ( 2006b ). PGE, gold and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Sample 90–24, 1034. Part 24, sample 90–24 1034. Unpublished GEUS report, 25pp plus tables and plates (available on request). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2010a ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 8, sample 90–18 972. GEUS report 2010/73, 19 pp plus tables and plates. Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2010b ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 9, sample 90–18 978. GEUS report 2010/74, 21 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2010c ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 10, sample 90–18 988. GEUS report 2010/85, 16 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2010d ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 11, sample 90–24 1062. GEUS report 2010/138, 17 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2012a ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 12, sample 90–18 1001. GEUS report 2012/91, 13 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2012b ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 13, sample 90–18 1012. GEUS report 2012/92, 20 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2012c ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 14, sample 90–10 434, GEUS report 2012/93, 19 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2012d ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 15, sample 90–10 443, GEUS report 2012/94, 19 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2012e ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 16, sample 90–10 445, GEUS report 2012/95, 16 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2012f ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 17, sample 90-24 1059. GEUS report 2012/96, 17 pp plus tables and plates (available online at www.greenmin.gl) Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2012g ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 18, sample 90-24 1056. GEUS report 2012/97, 20 pp plus tables and plates (available online at www.greenmin.gl) Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2012h ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 19, sample 90–24 1053 GEUS report 2012/98, 15 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2012i ). Gold, PGE and sulphide phases of the precious metal mineralization of the Skaergaard intrusion. Part 20, sample 90–24 1048. GEUS report 2012/104, 20 pp plus tables and plates (available online at www.greenmin.gl). Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D. ( 2015 ). Intermetallides and alloys of copper and palladium in ores of the Skaergaard massif . Proceedings of the Russian Mineralogical Society 144 , 30 – 53 . Google Scholar OpenURL Placeholder Text WorldCat Rudashevsky N. S. , Rudashevsky V. N., Nielsen T. F. D., Shebanov A. D. ( 2014 ). Au-Cu alloys and inter-metallides in Pd-Au ores of the Skaergaard massif . Proceedings of the Russian Mineralogical Society 143 , 1 – 23 . Google Scholar OpenURL Placeholder Text WorldCat Salmonsen L.-P. ( 2013 ). Petrology of the upper border series of the Skaergaard intrusion. Ph.D. thesis, Aarhus University, p. 96 . Salmonsen L. P. , Tegner C. ( 2013 ). Crystallization sequence of the Upper Border series of the Skaergaard intrusion: revised subdivision and implications for chamber-scale magma homogeneity . Contributions to Mineralogy and Petrology 165 , 1155 – 1171 . Google Scholar Crossref Search ADS WorldCat Svennevig K. , Guarnieri P. ( 2012 ). From 3D mapping to 3D modelling: a case study from the Skaergaard intrusion, southern East Greenland . Geological Survey of Denmark and Greenland Bulletin 26 , 57 – 60 . Google Scholar Crossref Search ADS WorldCat Tegner C. , Thy P., Holness M. B., Jakobsen J. K., Lesher C. E. ( 2009 ). Differentiation and compaction in the Skaergaard Intrusion . Journal of Petrology 50 , 813 – 840 . Google Scholar Crossref Search ADS WorldCat Turner P. A. ( 1990 ). Report on the 1989 field season Skaergaard concession. Platinova Resources Ltd./ Corona Corporation Joint venture, 47 pp., plus EA' es and maps (in archive of Geological Survey of Denmark and Greenland, GRF 20843. (available online at http://www.greenmin.gl) Turner P. A. , Mosher W. ( 1989 ). Report of 1988 field season: Skaergaard concession and East Greenland concession license. Platinova Resources Ltd/Corona Corporation Joint Venture, 71 pp. plus EA' es and maps (in archive of Geological Survey of Denmark and Greenland, GRF 20845; available online at www.greenmin.gl). Ueki K. , Iwamori H. ( 2017 ). Geochemical differentiation processes for arc magma and the Sengan volcanic cluster, Northeastern Japan, constrained from principal component analysis . Lithos 290-291 , 60 – 75 . Google Scholar Crossref Search ADS WorldCat Vukmanovic Z. , Holness M. B., Monks K., Andersen J. C. Ø. ( 2018 ). The Skaergaard trough layering: sedimentation in a convecting magma chamber . Contributions to Mineralogy and Petrology 173 , 43 . Google Scholar Crossref Search ADS PubMed WorldCat Wager L. R. , Brown G. M. ( 1968 ). Layered Igneous Rocks . Edinburgh and London : Oliver and Boyd , p. 588 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Watts Griffis , McOuat Limited ( 1991 ). 1990 Skaergaard project, Platinova/Corona concession, East Greenland. Exploration report, 55 pp. and EA’es (in archive of the Geological Survey of Denmark and Greenland, GRF no. 20848, available online at www.greenmin.gl). Wotzlaw J.-F. , Bindeman I. N., Schaltegger U., Brooks C. K., Naslund H. R. ( 2012 ). High-resolution insights into episodes of crystallization, hydrothermal alteration and remelting in the Skaergaard intrusive complex . Earth and Planetary Science Letters 355 , 199 – 212 . Google Scholar Crossref Search ADS WorldCat © The Author(s) 2019. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) © The Author(s) 2019. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
TI - Elemental Distributions and Mineral Parageneses of the Skaergaard PGE–Au Mineralization: Consequences of Accumulation, Redistribution, and Equilibration in an Upward-Migrating Mush Zone
JF - Journal of Petrology
DO - 10.1093/petrology/egz057
DA - 2019-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/elemental-distributions-and-mineral-parageneses-of-the-skaergaard-pge-tbaVlx8gZP
SP - 1903
EP - 1934
VL - 60
IS - 10
DP - DeepDyve
ER -