RARE-EARTH ELEMENT AND THORIUM POTENTIAL OF HEAVY MINERAL DEPOSITS ALONG THE WEST COAST OF SOUTH AFRICA WITH SPECIAL REFERENCE TO THE NAMAKWA SANDS DEPOSIT Carlo Philander and Abraham Rozendaal* Department of Earth Sciences, Stellenbosch University, Private Bag X01, Stellenbosch 7600, South Africa Abstract The west coast of South Africa is well known for its Cenozoic unconsolidated marine and aeolian placer deposits and includes the heavy minerals mined at Namakwa Sands. Resources total ~900 million tons at a grade of 10% total heavy minerals (THMs) from which concentrates of zircon, ilmenite, rutile and leucoxene are produced. Several other resources including satellite deposits proximal and distal to the mine area, diamond mine dumps and beach placers indicate vast heavy mineral potential along the west coast. The increased global demand for rare-earth elements (REEs), uranium and thorium has initiated interest in the gangue minerals from these deposits. Monazite, impure zircon, leucoxene and garnet are common and host significant concentrations of REEs, uranium and thorium. Resources in the Namakwa Sands deposit in particular amount to 250 kt REEs, 4.7 kt uranium and 10.5 kt thorium. This suggests a vast untapped resource along the west coast.
Keywords: rare-earth element (REE), Th, heavy mineral placers, South Africa 1
INTRODUCTION Significant growth in the Th and rare-earth element (REE) markets has resulted in increased exploration for these metals in the past few years. Traditionally, these commodities have been mined and extracted from a variety of sources, including heavy mineral deposits [1]. The mineral sands industry has been well aware of the marketing potential of their secondary Th and REE co-products. Only a few operations have, however, successfully managed to offer a diverse product portfolio beyond the traditional scope of the mineral sands market [2]. The current recessionary conditions have compelled many mineral sands operators to consider unlocking greater market value from their mineral resources. Namakwa Sands forms the focus of this study and is an Exxaro TSA (Pty) mineral sands operation that has been in production since 1994. It is a leading supplier of titania slag, pig iron, premium ceramic-grade zircon and rutile concentrate to several export markets [3]. Mine production comes from an open pit where a truck-and-shovel method is employed. Several wet spiral, magnetic and electrostatic separation stages are subsequently used to produce marketable zircon and rutile products as well as ilmenite concentrates for their titanium smelter. This article considers the potential of adding value to the heavy mineral deposits along the west coast of South Africa, in particular the Namakwa Sands deposit, by investigating the geological distribution, characteristics and economic potential of minerals containing thorium and REEs.
*
Correspondence to:
[email protected]
Maarten A. T. M. Broekmans (ed.), Proceedings of the 10th International Congress for Applied Mineralogy (ICAM), DOI: 10.1007/978-3-642-27682-8_63, Ó Springer-Verlag Berlin Heidelberg 2012
531
2
METHODOLOGY This article reports on samples representing a geometallurgical study, covering 660 reverse circulation percussion exploration drillholes and an annualized plant trial [6]. Samples were air-dried and riffle-split to obtain ~2 kg of analytical sample, prior to further treatment and analysis. The analytical programme entailed X-ray fluorescence (XRF), X-ray diffraction (XRD) and, to a lesser extent, -ray radiometry of in situ samples. Derivative THM (total heavy mineral) fractions were analysed with optical methods and scanning electron microscopy (SEM) techniques, including QEMSCAN, whereas specific minerals were targeted for singlegrain analysis by electron microprobe and laser ablation techniques. Sample analysis was conducted at Namakwa Sands, the University of Cape Town, Stellenbosch University, Council for Geoscience, Anglo Research and SGS Lakefield Africa. Details of analytical instrumentation, operating parameters and procedures followed are available upon request. 3
GEOLOGY Heavy mineral concentrations are widely distributed along the relatively narrow coastal plain of the west coast of South Africa [4]. Exploration for large-scale deposits only started during the 1980s. This resulted in the discovery of several significantly large heavy mineral deposits. The deposit near Brand-se-Baai was found to have economically viable resources, and was ultimately developed into the Namakwa Sands operation [5]. Several localities along the coastal margin of the west coast of South Africa offer similar potential, but are not economically feasible at this stage. In total, the entire west coast portrays a superior heavy mineral resource of world-class dimensions, able to sustain mining well into the next millennium [6]. Many of these heavy mineral deposits are genetically related to the classic examples of fluvial, marine, and aeolian placer deposits [7]. The contemporaneous Namakwa Sands deposit is hosted by a late Cainozoic sequence of marine, terrestrial and aeolian sands. The fine- to medium-grained sediments have been deposited on a gently sloping coastal basement of metamorphites of the Mesoproterozoic Namaqualand Metamorphic Province and Neoproterozoic Gariep Supergroup and Paleozoic sediments, during the mid Miocene (~17 Ma) to late Pleistocene (0.1 Ma) [8]. The ore body displays the classical morphological and sedimentological characteristics of heavy mineral deposits located at present-day coastal margins that have been formed by the interplay of marine and terrestrial processes active in a log-spiral bay setting [9]. The economic mineralization is hosted in three distinct geological units (Figure 1). The oldest units are two localized heavy mineral-enriched strandline deposits (STR) that are tens of metres thick. These occur at 35 and 20 m amsl, respectively, and overly a non-mineralized mega-dune. The Orange Feldspathic Sand unit (OFS), comprises a 45-m thick orange–yellow reworked terrestrial sand that contains the bulk of the mineralization. A 1–3-m thick surface aeolinite, the Red Aeolian Sand unit (RAS), caps the deposit. Some parts of the deposit, in particular the OFS unit, have been negatively affected by several layers of superimposed duricrust that has semi-lithified the sands. This has resulted in the sterilization of parts of the ore body and loss of resources [10]. 4
MINERAL RESOURCES The Namakwa Sands ore body including its proximate satellite extensions covers more than 26,000 ha. Heavy minerals are anomalously concentrated along the log-spiral morphology of the residual STR beaches. By contrast, heavy mineral distributions of the overlying aeolian OFS and RAS are visibly aligned along a SW–NE strike, providing testimony to the dominant prevailing wind directions at the time of their deposition (Figure 2). Generally, the mineralization extends from the surface down to footwall and no overburden stripping is required. 532
Heavy mineral populations are typically diverse, but only zircon, ilmenite, rutile, leucoxene, garnet and pyroxene are present in appreciable concentrations. The rest of the heavy mineral suite comprises a range of minerals that are collectively termed as “others” and include hematite, magnetite, kyanite, monazite, chromite, cassiterite, corundum, apatite, sulfides, and titanite. The THM content decreases upwards in the sequence, but conversely the grades of the densest heavy minerals such as ilmenite, zircon, and monazite improve with increasing elevation (Table 1). Weathering and aeolian reworking possibly upgraded the denser mineral fraction at the expense of a less stable, lower-density heavy mineral population. The economic fraction comprises zircon, ilmenite, rutile, and leucoxene and accounts for 51% of the total THM resource and has an estimated distribution of 3.3% ilmenite, 0.8% zircon, 0.2% rutile, 0.3% leucoxene and 0.02% monazite in an indicated resource of 858 tonnes. Translating these grades into mineral tonnages emphasizes Namakwa Sands’ high zircon to ilmenite ratio which is claimed to be the best in the international mineral sands industry [2]. Namakwa Sands has an annual production capacity of 21 million tonnes (Mt) and produces 350 kt ilmenite, 130 kt zircon and 25 kt rutile. The deposit has a life of mine of more than 20 years. 5
MINERALOGY Most of the major minerals investigated co-host Th and REE in variable quantities (Table 2). Trace minerals such as cassiterite, hornblende, epidote, Ce-florencite, baddeyellite, tourmaline, staurolite, titanite and biogenic apatite (collophane) were also found to contain these elements. However, only zircon, monazite, leucoxene, and garnet were considered for further investigation and consideration as possible resources. 5.1
Zircon ZrSiO4 The major element chemistry of zircon is very consistent, but a whole range of trace elements can be incorporated into the silicate crystal lattice through coupled substitution. Zr4+ is commonly replaced by Hf4+, U4+, Th4+, Y3+, REE3+, Nb5+, Ta5+, Ti4+, Pb4+, Pb2+, Fe3+, Fe2+, Ca2+, Na+ and K+, and Si4+ by Al3+, P5+ and S6+. Based on chemistry, Namakwa Sands distinguishes two zircon types referred to as “pure” and “impure” zircons, respectively. Pure zircon contains low concentrations of Fe, U and Th, and is optically clear. Hafnium is the major substitute for zirconium due to similar charge balance characteristics, and on average impure zircon was found to contain a greater percentage of HfO2 than pure zircon. At trace element level, the differentiation between the two zircon types is more pronounced. Without exception, impure zircon accommodates more non-structural elements than pure zircon and may reach percentage levels. Yttrium dominates the trace element assemblage, confirming the isostructural relationship between xenotime (YPO4) and zircon. In both zircon types, U is more abundant than Th, and impure zircon contains nearly twice as many radioactive elements as pure zircon. Impure zircon displays a yellow-brown colour and may be opaque due to metamictization, a radioactive damaging process due to high concentrations of U and Th. Impure zircon hosts REE in considerably greater concentrations than pure zircon. Both zircon types are more elevated in the heavy REEs (HREEs) compared to the light REEs (LREEs). As the size of Zr4+ is closer to the heavier than to the LREEs, it has a strong preference to fractionate the HREE. Chondrite-normalized REE patterns for all zircon types are strikingly similar, characterized by a well-defined negative Eu anomaly, relatively flat LREE and smooth ascending HREE-enriched profiles (Figure 3). This tilted “birdwing” REE pattern is a typical feature of west coast zircon populations. Approximately 75% of the Namakwa Sands zircon resource consists of pure zircon. This fraction is metallurgically upgraded to a secondary zircon product with 85% >1000 ppm U + Th used in the chemical 533
industry. The prime zircon product consists of 95% 1. The relatively flat but elevated chondrite-normalized REE profiles suggest that leucoxene scavenges elements non-selectively from the soil environment. The bulk of the mineral resource consists of low-grade, poor-quality leucoxene, high in contaminant elements including U, Th and REE. Up to 20% of the rutile products from Namakwa Sands are spiked with high-grade (>85% TiO2) leucoxene to increase the total titanium content. Poor-quality leucoxene, constituting the larger proportion of the population, is discarded to non-magnetic rejects. 5.4
Garnet (Fe, Mg)3·Al2(SiO4)3 Garnet geochemistry shows that the dominant pinks to red coloured varieties are generally enriched in Fe, Al and Mg, and belong to the almandine-pyrope group. Its consistent bulk chemistry is a reflection of its homogeneous internal structure which is generally devoid of inclusions. Garnet is depleted in U and Th, and has the lowest REE content of all the minerals analysed. Yttrium dominates the REE assemblage, and analyses furthermore show that garnet incorporates HREE more readily than LREE, which is concordant with comparative studies [13]. The normalized REE patterns are marked by LREE-depleted and HREE-
534
enriched ascending profiles with pronounced negative Eu anomalies, similar to the results of other garnet studies [14]. In the ilmenite beneficiation process, more than 99% of the garnet fraction is rejected. The coarser fraction is further refined by a local customer to produce sandblasting material. 6
DISCUSSION In the past 3 years, the global demand for rare earths has increased beyond production capacities, as applications for these exotic elements increased significantly. The last decade also saw a noticeable market shift away from the LREE in favour of the HREE [15]. The comparatively low abundance of HREE in world-wide exploitable deposits contributed to steep rises of HREE prices the past few years In 2008, China was responsible for more than 95% of the 124 kt of rare earths produced globally and is considering to significantly reduce their REE exports to conserve and consolidate their resource base in the future [16]. In response, many dependant companies are pursuing alternative sources to satisfy their own growing needs, which can lead to increased production outside of China. As a global REE resource, heavy mineral deposits rank very low and at best are low-grade, highvolume deposits [17]. Like the Namakwa Sands ore body, many other sizeable heavy mineral deposits are expected to host REE below the 0.1% level, mainly due to their similar granite-gneiss-dominated provenance, which has limited REE-bearing minerals, and their common placer characteristics. Another concern related to heavy mineral deposits is their tendency to co-host REE with refractory U and Th. The mining and distribution of minerals with these naturally occurring radioactive elements are strictly regulated worldwide particularly from an environmental perspective. This has eliminated monazite as a significant source of REE due to its high Th content, and has shifted preference to low-Th minerals such as bastnäsite. Recently, renewed interest in thorium as a non-proliferative nuclear fuel as well as in non-energy applications might revive the depressed thorium and related monazite market. Based on the technical complexity and associated cost to extract the REEs, mineral sand operations are probably better positioned as suppliers of REE-enriched mineral concentrates than the traditional source. There are a few aspects favouring mineral sand producers as alternative REE suppliers. The considerable volume of their mineral resources offers appreciable quantities of contained REE. The Namakwa Sands ore body hosts about 250 kt of REE of which 80% is contained by LREE-enriched monazite. Zircon is the chief contributor to the HREE component together with Y (Figure 3). Up to 10% of the total REE, particularly the HREE, present in the Namakwa Sands, however, report to saleable rutile and zircon products in refractory form and the bulk is rejected as by-products. In addition, heavy mineral deposits host a diversity of minerals with low U and Th but high HREE such as garnet and leucoxene which are rejected and can be blended into a customer-specific REE feedstock concentrate. A significant advantage over hard rock deposits is that the REE-containing minerals in heavy mineral deposits are highly liberated and easily upgradeable, eliminating the need for expensive comminution or other forms of pre-concentration. Trails performed with Namakwa Sands rejects showed relatively good rare-earth mineral grade–recovery relationships under laboratory-scale processing conditions. These upgraded concentrates – in essence mixtures of phosphate, oxide, and silicate minerals – would be amenable to hydrometallurgical extraction processes currently used to recover rare earths. Enticing low-cost, high-return opportunities exist for mineral sands producers like Namakwa Sands in their U, Th and REE-enriched by-products which could increase the in situ value and revenue of their mineral resources and reduce the cost of managing the stockpiled by-products.
535
Relaxing or revising the tight regulatory nuclear legislation is, however, required to further develop these prospects past the conceptual stage. Optimistically, the growing global demand for Th as replacement nuclear fuel might influence a positive outcome in this regard. Namakwa Sands’ good track record as an environmentally responsible company, particularly underlined by its world leading radiation management programme, could successfully support possible application to amend the conditions of their nuclear license. 7
CONCLUSIONS The Namakwa Sands heavy mineral deposit has a grade of ~500 ppm U, Th and rare-earth oxides in its in situ resources of 858 Mt. Thorium and rare earths are present in nearly all minerals analysed including the economic zirconium–titanium mineral assemblage, gangue silicates such as garnet and pyroxene as well as a range of other trace minerals including monazite. Cerium monazite is the only rare-earth mineral present and is also the most significant carrier of thorium, whereas all the other noteworthy minerals contain on average
