Exploring for kimberlites in glaciated terrains using chromite in quaternary till—a regional case study from northern Finland

https://doi.org/10.1016/S0375-6742(02)00253-4Get rights and content

Abstract

Prospecting for kimberlites and related rocks in till-covered terrains requires a methodology for recovering a few small grains within tens of kilograms samples, necessitating 1 ppb sensitivity or better. As part of reconnaissance survey for the kimberlite indicator minerals, i.e. pyrope garnet, picroilmenite, chromite and chromian diopside, the Geological Survey of Finland (GTK) developed such a system by significantly modifying and augmenting a 3″ Knelson Concentrator that accomplishes nearly complete recovery of moderately heavy minerals (>0.25 mm) from till samples.

Diamondiferous kimberlites occur in the eastern Finland around the Kaavi–Kuopio and Kuhmo areas and much of the rest of the Karelian craton remains prospective based on the empirical evidence necessary for diamond preservation: thick (>200 km) lithospheric mantle, low heat flow and Archaean age rocks. A target area in Lapland, 20×50 km in size, was selected for a pilot study to test extraction of chromite for the (1) discrimination of regionally and locally derived populations, and (2) recognition of possible kimberlitic/lamproitic chromites. Area selection was based on the regional occurrence of a variety of mantle-derived rocks, the recovery of a chromian pyrope grain from till in 1996 and most importantly, the well-established Quaternary stratigraphy in the region. The sample material consisted of sixty-two 80-kg excavator and 40-kg shovel samples. Approximately 1000 chromite grains, almost exclusively 0.25–0.5 mm in diameter, were recovered and analysed by electron microprobe.

Tills in the sampling area proved to contain at least two compositional populations of chromite. The first is present in almost every sample and is apparently derived from layered mafic intrusions distal to and up-ice from the study area. The second population consists of chromites with low Ti, high Cr and Mg similar to inclusions in diamond. It is present in approximately one third of the samples, concentrated in a couple of clusters within the target area and is therefore considered to be of more local derivation. Since no high-Ti, high-Cr chromites diagnostic for kimberlites and lamproites were present in the samples, the source for the low-Ti, high-Cr, high-Mg chromite grains remains uncertain, but is probably not kimberlitic. Although this apparently is a negative outcome for diamond exploration in the target area, the main goal of the study was realised by showing the applicability of the system to heavy mineral separation from Quaternary glacial deposits.

Introduction

During the years 1996–1998, the Geological Survey of Finland (GTK) carried out reconnaissance sampling of Quaternary till for kimberlitic indicator minerals, i.e. pyrope garnet, picroilmenite, chromite and chromian diopside, in a region of eastern Lapland (Fig. 1). Apart from a single mantle derived subcalcic “G10” chromian pyrope grain (Dawson and Stephens, 1975), the only indicator mineral discovered was chromite. The recognition of possible kimberlitic/lamproitic chromites as well as other chromite populations present was set as the main objective of this study. An attempt was also made to differentiate between regionally and more locally derived populations through near quantitative extraction of chromite grains from till. The size of the sampling area was ca. 20×50 km, the long side being roughly perpendicular to the main ice flow direction.

Chromite, (Fe2+, Mg)(Cr, Al, Fe3+)2O4, was considered to be a promising mineral for this kind of study because it is a sensitive petrogenetic indicator and widely used for characterising crystallisation conditions in the host magma or from chromite-bearing mantle sources Irvine, 1965, Irvine, 1967, Dick and Bullen, 1984. However, the chromite composition can change due to hydrothermal processes, magmatic alteration and regional metamorphism (e.g. Hoffman and Walker, 1978, Jan and Windley, 1990, Kimball, 1990, Liipo et al., 1994, Vuollo and Piirainen, 1989). Chromite is a resistant mineral to weathering (Birkeland, 1974); it can therefore survive over long distances during glacial transport (Averill, 2001).

The bedrock of the study area has a number of potential source rocks for chromites, including layered mafic intrusions Alapieti, 1982, Mutanen, 1997 and their boninitic feeder dykes (Vuollo et al., 1995), komatiites (including ophiolites) Manninen, 1991, Hanski, 1997 as well as possible kimberlites and lamproites. Also importantly, the Quaternary stratigraphy in the region is well established (Hirvas, 1991).

Diamondiferous kimberlites are largely confined to the stable cratons of Archaean age with a low heat flow and a thick lithospheric mantle. A low geothermal gradient combined with great thickness of the lithosphere increase the probability that an ascending kimberlite will intersect diamondiferous material present in the mantle Kennedy, 1964, Helmstaedt and Gurney, 1995, Morgan, 1995, but the necessity for the Archaean rocks is not entirely understood. It may be related to the composition of the source mantle rocks, either through an extreme depletion event (Helmstaedt and Gurney, 1995) leaving harzburgite and dunite residues, or through the effect of low fO2 allowing preferential diamond formation.

Diamondiferous kimberlites have already been found in the eastern parts of Finland, in the regions of Kaavi–Kuopio and Kuhmo (O'Brien and Tyni, 1999, European Diamonds, Press release, London, December 2000). More recently, a macrodiamond discovery from till was reported in the Kuhmo area (European Diamonds, Press release, London, June 2001).

Fig. 1 presents the Archaean area in northern Europe where the geothermal gradient is low (simplified from Kukkonen and Jõeleht, 1996) and where the lithosphere is thicker than 170 km based on seismic methods (e.g. Calcagnile, 1982). Studies on mantle xenoliths from the Kaavi–Kuopio kimberlites suggest that the petrological lithosphere–asthenosphere boundary is close to 240 km beneath the central Fennoscandian Shield (Kukkonen and Peltonen, 1999). The area where the thick lithospheric mantle, low heat flow and Archaean age rocks overlap is the most prospective for diamond exploration and covers a major portion of Finland, Russian Karelia and Kola Peninsula.

The basic exploration method for kimberlites and lamproites in glaciated terrains, particularly during reconnaissance and regional work, is to track kimberlitic indicator minerals, i.e. high pressure and temperature xenocrysts (pyrope garnet, picroilmenite, chromite and chromian diopside) dispersed in glacial sediments, especially in till (McClenaghan and Kjarsgaard, 2001). This type of exploration requires profound understanding of the glacial history in the region, e.g. recognition of various till beds each deposited during a different ice flow stage of the continental ice sheet. The indicator mineral content and the relative abundance of different indicators varies greatly between individual kimberlites (e.g. Mitchell, 1986), and the clastic portion of till deposited down-ice reflects the mineralogy of the source (e.g. Kong et al., 1999). Experience from exploration work in two eastern Finland kimberlite clusters (Kaavi and Kuhmo areas) has shown that the indicator mineral contents in the Finnish tills can be extremely low, sometimes only a few, small (0.25–1.0 mm) grains in a 40-kg sample taken less than 2 km down-ice from the source. Thus, successful indicator survey requires very sensitive processing methods.

In some cases, chromite is the only indicator mineral that can be used in exploration work. For instance, in the large Archangelsk Kimberlite Province at the eastern edge of the contiguous Karelia–Kola–Kuloi cratons (Fig. 1), many kimberlites contain little garnet or picroilmenite whereas chromite and diamond are the only indicators (Mahotkin et al., 2000). On the other hand, silicate indicator minerals are also less resistant to weathering than oxides, especially chromite, which is often the only indicator to survive extreme weathering conditions associated with laterisation (e.g. western Australia, Shee et al., 1991). This is a possibility also in the case of Lapland, which underwent extensive preglacial weathering (Hirvas et al., 1977); duricrusts that formed during several different periods still cover large areas Sarapää, 1996, Vartiainen, 1980. Based on paleomagnetic data (Mertanen and Pesonen, 1997), Finland was situated in low latitudes from the late Proterozoic to nearly the end of the Paleozoic era.

Mantle-derived chromites in kimberlitic rocks display a vast compositional spectrum. Several subpopulations of chromite can be present in a kimberlitic suite (Griffin and Ryan, 1993): (a) xenocrysts derived from dunites, harzburgites, lherzolites, pyroxenites, etc., source rocks that have been sampled from different levels of the upper mantle; (b) xenocrysts with elevated contents of TiO2 due to reaction with the host rock; and (c) magmatic, i.e. phenocryst chromites. Chromites associated with diamonds have high levels of Cr2O3 (usually >60 wt.%) and MgO (ca. 12–16 wt.%) Fipke et al., 1989, Fipke et al., 1995, Gurney and Zweistra, 1995. The Cr content in xenocryst chromites is pressure dependent and thus indicative of diamond potential (Daniels, 1991). Zn in chrome spinel is strongly negatively correlated with temperature and the “Zn thermometer” (Griffin et al., 1994) can be used to estimate the equilibration temperature of individual grains. If the TZn values for high-Cr, high-Mg, low-Zn (kimberlitic xenocrystic) spinels equilibrated with olivine are consistent with a low cratonic geothermal gradient, they indicate derivation from within the diamond stability field.

As an example for kimberlitic chromites, chromite analyses from two Finnish diamondiferous but subeconomic kimberlites, Lahtojoki and Seitaperä, are plotted in a Cr2O3–MgO diagram redrawn after Smith et al. (1991) (Fig. 2). Selected analyses from Seitaperä are given in Appendix A. The Lahtojoki pipe (Pipe 7) belongs to the Kaavi kimberlite cluster and exhibits a typical mineralogy for Group I kimberlites (O'Brien and Tyni, 1999). In Pipe 7, chromite is a rare accessory mineral. The Seitaperä dyke swarm in Kuhmo (Dyke 16), in contrast, is very rich in chromite. It has characteristics of both Group II kimberlite and olivine lamproite (O'Brien and Tyni, 1999). Fig. 2 shows that compared to the total number of analyses, the number of high-Cr, high-Mg chromites plotting in the diamond stability field is relatively low. Thus, for exploration purposes, it is important to be able to identify other types of kimberlitic chromites as well. In Fig. 3, the Pipe 7 and Dyke 16 chromite analyses are shown in a TiO2–Cr2O3 diagram redrawn after Fipke (1991). He has recognised three compositional types of kimberlitic/lamproitic chromites useful in diamond exploration. These include (1) chromites with similar composition to the ones that exist as diamond inclusions and intergrowths (i.e. DI-chromites), (2) high-Cr, high-Ti chromites (i.e. Cr–Ti chromites) and (3) relatively low-Cr, high-Al and/or high-Fe chromites (i.e. Al/Fe chromites). The first group is characterised by high contents of MgO (>8.7 wt.%) and Cr2O3 (>57.8 wt.%); TiO2 content is generally below 0.6 wt.% (Fig. 3). The Cr–Ti chromites are phenocrysts crystallised from TiO2-rich kimberlitic/lamproitic magmas. They contain more than 0.8 wt.% TiO2 and plot within the field unique to lamproites/kimberlites (Fig. 3), whereas the Al/Fe chromites plot within the overlap field along with chromites from rock types other than kimberlites and lamproites. Fig. 3 shows that Cr–Ti chromites are particularly abundant in Dyke 16; all of the high-Mg, high-Cr chromites plotting in the diamond field in the Cr2O3–MgO diagram (Fig. 2) are in fact phenocrysts with high levels of TiO2 and, as such, they are not indicative of diamond potential (Grütter and Apter, 1998). Al/Fe chromites are abundant in both Dyke 16 and Pipe 7 (Fig. 3); in the Cr2O3–MgO diagram, they form a low-Cr, high-Mg lherzolitic tail.

Several varieties of nonkimberlitic rocks can contain chromites with similar compositions to the DI-group described above Fipke et al., 1995, Griffin and Ryan, 1995. These rocks include komatiites and Mg-rich basalts of greenstone sequences, ophiolite-related peridotites and gabbros as well as some layered gabbro complexes. Hanski (1997), for instance, has published chromite analyses from Paleoproterozoic ophiolitic rocks of the Nuttio serpentinite belt in Finland, that have low-Ti, high-Cr, high-Mg compositions. Also, boninites may contain chromites of this type (Barnes and Roeder, 2001). However, the discovery of Cr–Ti chromites (Fipke, 1991) in regional samples is probably the most reliable evidence for the presence of kimberlitic or lamproitic sources in an area even if chromites from other sources are also present and no other indicator minerals can be found. In addition, use of the CART diagnostic software (Breiman et al., 1984) based on the combination of chromite major-and trace-elements makes it possible to discriminate (Fisher, 1990) kimberlitic and lamproitic chromites from high-Cr, high-Mg chromites derived from nonkimberlitic sources.

Kimberlitic (and lamproitic) chromite grains have certain diagnostic morphological features, which can be used for exploration purposes. These include, for instance, a satin-like sheen and fine layering, matte pitted or smooth glossy surfaces, and bevelled edges (Muggeridge, 1995). However, these surface features can be lost through physical wear, e.g. glacial transport (Golubev, 1995). Some kimberlitic chromites can also have a unique internal structure. In this type of grains, the outer rim or “corona” usually exhibits enrichment in TiO2 compared to the grain core Berryman et al., 1999, Shee et al., 1991, Wyatt et al., 1999. The increase of TiO2 in the coronas probably reflects progressive reequilibration of xenocrystic chromite with the host, TiO2-rich, kimberlitic magma and/or the overgrowth by high TiO2 phenocrystic spinel crystallising from the magma (Grütter and Apter, 1998). The corona chromites are fragile and cannot survive over long transportation distances in unglaciated terrains (Berryman et al., 1999); nevertheless, it can be assumed that they last somewhat longer in glacial transport.

Fig. 4A–D presents backscattered electron (BSE) images of four different types of chromite grains extracted from two Finnish kimberlites. Grains 4A and 4B are from the Lahtojoki pipe (Pipe 7) of the Kaavi kimberlite cluster, while grains 4C and 4D are from the Seitaperä dyke swarm in Kuhmo (Dyke 16). In Pipe 7, chromite is a rare mineral and hardly ever occurs as grains larger than 0.5 mm in diameter. Dyke 16 is rich in chromite with many grains up to 1 mm in size. Grain 4A is a small rounded octahedron (ca. 0.4 mm in diameter) with a smooth, well-polished surface. Under a binocular microscope, it exhibits a satin-like sheen. Grain 4B is also a small octahedron (ca. 0.3 mm) but it is not very rounded and has an unusual surface texture. The type of growth forms on its surface is sometimes seen on diamonds (Bulanova, 1995) but rarely on spinels. Grain 4C is a relatively large fragment (greatest length >1 mm) completely irregular in shape. Its original surface is somewhat resorbed and exhibits matte lustre under a binocular microscope. Grain 4D is an irregular fragment (ca. 0.7 mm) like 4C but it displays the corona texture described above. The lighter colour spinel phase on the grain surface is enriched in TiO2 and depleted in MgO compared to the darker grain core. A closer look at grains 4C and 4D reveals that they are both somewhat fractured. During glacial transport, these grains would probably start breaking along the preexisting fractures resulting in multiple smaller fragments until the terminal grade grain size (Dreimanis, 1982) was reached and the particles began to float in the matrix.

Section snippets

The study area and regional geology

The bedrock of Lapland contains a variety of mantle derived rocks, e.g. mafic layered intrusions and dykes, gabbros and diabases aged 2.44–2.00 Ga Alapieti, 1982, Hanski, 1982, Hanski, 1984, Kouvo, 1977, Mutanen, 1997, Vuollo, 1994, Vuollo et al., 1995, Proterozoic komatiites including ophiolites Manninen, 1991, Hanski, 1997 and, as the youngest mantle derived magmas, a 1.15 Ga diabase dykes Väänänen, 1965, Manninen, 1991. Apart from previous diamond claims and claim reservations as an indirect

Sampling

In 1997, GTK made preliminary till sampling around the discovery site of the chromian pyrope grain. Quaternary samples, 30 in number, were taken by shovel. They represented the uppermost part of the basal till horizon, i.e. till bed II (Hirvas, 1991) and the regional indicator dispersion. The initial sample size was ca. 40 kg. The samples failed to recover new pyrope grains, but instead revealed a few high-Cr, high-Mg chromites, suggesting a potential for kimberlites or lamproites.

Stage 2

Mineralogical observations of the till samples

The till samples turned out to be heterogeneous regarding their heavy mineral content (Lehtonen, 1999). Laboratory tests showed significant variation in the relative proportions of heavy minerals in the Knelson preconcentrates as well as in the fine fraction distribution and magnetite content of the HMS concentrates (Fig. 7). Two deep excavator samples (M12B, M13B) (Fig. 5) differed most from the others: they both had considerably less magnetite than those from other test pits as well as

Discussion

This study suggests that, methodologically, the discrimination of regional and more local chromite populations from till samples is realistic. However, the volume of the bedrock source has its impact on the results. The layered intrusions in northern Finland are voluminous geological formations that host abundant chromite mineralisations at various stratigraphic levels Alapieti, 1982, Mutanen, 1997. Consequently, it can be assumed that tills derived from these intrusions contain chromites in

Sampling and sample processing

The processing methods used turned out to be applicable in tracking source rocks up-ice by identifying characteristic heavy minerals dispersed in till. A relatively high number of chromite grains was extracted from till where they existed at an extremely dilute concentration. The crucial part of the processing line was the GTK modified 3″ Knelson concentrator, which reduced the originally 20–40 kg (<2 mm) samples to 400–600 g. Another important application was the HIMS for heavy mineral

Acknowledgements

This study was conducted at GTK and supported also by the Academy of Finland. The authors would like to thank especially Dr. Hugh O'Brien for discussions and useful recommendations. Dr. Sven Monrad Jensen is appreciated for revising the manuscript and providing valuable suggestions for improvements. Mr. Bo Johanson and Mr. Lassi Pakkanen from the GTK E-beam Laboratory provided the microanalytical data. The support of Professor Ilmari Haapala of the University of Helsinki and is appreciated. We

References (65)

  • M.T. Muggeridge

    Pathfinder sampling techniques for locating primary sources of diamond: recovery of indicator minerals, diamonds and geochemical signatures

    Diamond Exploration: Into the 21st Century

    J. Geochem. Explor.

    (1995)
  • T. Alapieti

    The Koillismaa layered igneous complex, Finland—its structure, mineralogy and geochemistry, with emphasis on the distribution of chromium

    Geol. Surv. Finl., Bull.

    (1982)
  • S.A. Averill

    The application of heavy indicator mineralogy in mineral exploration with emphasis on base metal indicators in glaciated terrains

    Drift Exploration in Glaciated Terrain

    Spec. Publ.-Geol. Soc. Lond.

    (2001)
  • S.J. Barnes et al.

    The range of spinel compositions in terrestial mafic and ultramafic rocks

    J. Petrol.

    (2001)
  • A.K. Berryman et al.

    The discovery and geology of the timber creek kimberlites, Northern Territory, Australia

  • P.W. Birkeland

    Pedology, Weathering and Geomorphological Research

    (1974)
  • L. Breiman et al.

    Classification and Regression Trees

    (1984)
  • Chernet, T., Marmo, J.S., 2001. Mechanical and digital size analysis of Kemi chromite ore treated by electric shock...
  • Daniels, L.R.M., 1991. Diamonds and related minerals from the Dokolwayo kimberlite, kingdom of Swaziland. Unpubl. PhD...
  • J.B. Dawson et al.

    Statistical analysis of garnets from kimberlites and associated xenolith

    J. Geol.

    (1975)
  • H.J.B. Dick et al.

    Chromian spinel as a petrogenetic indicator in abyssal and alpine-type peridotites and spatially associated lavas

    Contrib. Mineral. Petrol.

    (1984)
  • A. Dreimanis

    Work group (1)—Genetic classification of tills and criteria for their differentation: progress report and activities 1977–1982 and definition of glacigenic terms

  • N.I. Fisher

    On the application of some computer-intensive statistical methods in the earth sciences

  • Fipke, C.E., Significance of chromite, ilmenite, G5 Mg-almandine garnet, zircon and tourmaline in heavy mineral...
  • Fipke, C.E., Gurney, J.J., Moore, R.O., Nassichuk, W.W., 1989. The development of advanced technology to distinguish...
  • C.E. Fipke et al.

    Diamond exploration techniques emphasising indicator mineral geochemistry and Canadian examples

    Geol. Surv. Can., Bull.

    (1995)
  • G. Gaál et al.

    An outline of the Precambrian evolution of the Baltic Shield

    Precambrian Res.

    (1987)
  • W.L. Griffin et al.

    Trace elements in garnets and chromites: evaluation of diamond exploration targets

  • W.L. Griffin et al.

    Chromite macrocrysts in kimberlites and lamproites: geochemistry and origin

  • D.I. Groves et al.

    Spinel phases associated with metamorphosed volcanic-type iron–nickel sulfide ores from Western Australia

    Econ. Geol.

    (1977)
  • D.I. Groves et al.

    Exploration significance of chrome–spinels in mineralized ultramafic rocks and nickel–copper ores

    Spec. Publ.-Geol. Soc. S. Afr.

    (1983)
  • H.S. Grütter et al.

    Kimberlite and lamproite-borne chromite phenocrysts with “diamond-inclusion” type chemistries

  • Cited by (0)

    View full text