This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
0361-0128/11/3937/93-25 93
IntroductionSince first applied by Walker et al. (1991), the application
of the 190Pt-186Os decay system as a tracer of deep Earthprocesses has been the focus of a surge in interest (e.g.,Walker et al., 1997a; Brandon et al., 1998, 2007; Luguet et al.,2008). However, the application of this decay system togeochronology has been limited (Walker et al., 1997b), withvery few applications to ore genesis, despite the economic rel-evance of the parent and daughter isotopes. The reasons forthis stem from the very low decay constant of 190Pt (λ = 1.477× 10–12 yr–1 ± 1%; Begemann et al., 2001) and the very smallresultant variations in the 186Os/188Os ratio. These practicaldifficulties make PGE alloys and certain other PGM clear tar-gets for study because of the wide range of potential Pt/Os ra-tios that these mineral groups may have (Harris and Cabri,1991). Nowell et al. (2008b) recently showed that it is possi-ble to produce accurate and precise 186Os/188Os measure-ments of PGM grains via laser ablation-multicollector-ICPMS (LA-MC-ICPMS) and the additional Re-Os isotopedata produced from such analyses can be used to examinemantle Os isotope variations on a global scale (Pearson et al.,2007). The former study also demonstrated that the substan-tial variation in Pt/Os within and between PGM can be usedto obtain geochronological information for samples from ul-tramafic rocks that are often difficult to date. This offers ob-vious opportunities in terms of the dating of ophiolite massifs,which is notoriously problematic when using traditionalgeochronology methods.
In this study we apply the Pt-Os chronometer to a suite ofdetrital PGM grains derived from the Meratus ophiolite, Bor-neo. The source of the PGM grains is well constrained andcomparison with existing geochronology and geochemistry al-lows us to evaluate the accuracy of the Pt-Os age. The resultillustrates the potential for the Pt-Os isotope system in con-straining the age and origin of alluvial PGM deposits.
Samples and ProvenancePGM grains from Borneo were provided for this study by
the Smithsonian Institution National Museum of Natural His-tory from sample NMNH96511. The grains were collected byA. Lacroix from the Pontyn River, Tanah Laut, in the provinceof South Kalimantan, Indonesian Borneo, and were labeled as“laurite.”
Regional geology and provenance
The Pontyn River drains the southern Meratus Mountainsand flows into the Java Sea near the town of Asemasen on thesoutheast coast of Borneo (Fig. 1). The catchment of theriver and its tributaries is geologically varied. It includes var-ious clastic sedimentary rocks, limestones, intermediate tofelsic volcanic rocks, metamorphic rocks and the ultramafic,gabbroic, and leucocratic rocks that make up the Meratusophiolite (Sikumbang, 1990; Guntoro, 1999). Trace elementabundances (particularly negative Nb-Ta anomalies and(La/Nb)N ratios of 1.7–3.5) of Meratus lavas have been inter-preted as evidence that a back-arc basin setting may haveproduced the oceanic lithosphere that forms the Meratusophiolite (Monnier et al., 1999). However, the same authorsconclude that Cr# of spinel, orthopyroxene, and clinopyrox-ene in peridotites, along with low Na2O and TiO2 contentsand depletion in incompatible elements, demonstrates that
APPLICATION OF THE 190Pt-186Os ISOTOPE SYSTEM TO DATING PLATINUM MINERALIZATION AND OPHIOLITE FORMATION: AN EXAMPLE FROM THE MERATUS MOUNTAINS, BORNEO
J. A. COGGON,1,†,* G. M. NOWELL,1 D. G. PEARSON,1 AND S. W. PARMAN2
1 Northern Centre for Isotopic and Elemental Tracing, Department of Earth Sciences, Durham University, South Road, Durham DH1 3LE, United Kingdom
2 Department of Geological Sciences, Brown University, Providence, Rhode Island 02912
AbstractThe formation age of platinum-group minerals (PGM) in placer deposits has traditionally been difficult to
constrain. We have applied the Pt-Os and Re-Os isotope systems to this problem by analyzing a suite of PGMfrom a placer deposit in southeastern Borneo that are derived, by mechanical processes, from chromitites ofthe Meratus ophiolite. Published subduction and emplacement ages and biostratigraphy of pelagic sedimentsof the ophiolite sequence define a minimum age for genesis at a spreading ridge. However, igneous compo-nents of the ophiolite have previously been undateable. Alluvial PGM grains (n = 260) from the Pontyn River,which drains the Meratus Mountains, were analyzed by laser ablation-multicollector-inductively-coupled massspectrometry (LA-MC-ICPMS). Re-Os data do not show any isochronous relationship. Despite a significant rangein 187Os/188Os (0.122–0.141), 187Re/188Os values show a very narrow range (0.000005–0.002980). In contrast, thePGM have a wide range in both 186Os/188Os (0.119801–0.120315) and 190Pt/188Os (<0.00001–1.493430), yield-ing a precise Pt-Os isochron age of 197.8 ± 8.1 Ma (2σ). This age fits well with published age constraints forthis ophiolite and we argue that it dates the crystallization of the PGM. Previous studies have shown that thePontyn PGM are derived from ophiolitic chromitite; therefore, the PGM Pt-Os isochron age also provides thefirst absolute age constraint for the genesis of igneous rocks of the Meratus ophiolite. These results highlightthe potential of the Pt-Os geochronometer as a tool for dating the crystallization age of PGM found in placerdeposits, for dating primary platinum mineralization in general, and for use in ophiolite geochronology.
Submitted: November 3, 2009Accepted: September 5, 2010
Meratus peridotites represent subcontinental lithosphericmantle that underwent low degrees of localized partial melt-ing during the final stages of an episode of continental rifting.
Burgath and Mohr (1986) and Burgath (1988) reported theoccurrence of PGE-rich chromitite seams in the serpentinizeddunite of the Meratus ultramafic rocks. Hattori et al. (1992)analyzed the osmium isotope composition of laurite grainsfrom these chromitites by SIMS. All these authors also ana-lyzed PGM from associated alluvial placers and concluded thatthe isotopic data support a detrital origin for the placer grains,which were mechanically eroded from their host chromitites.
PGM mineralogy and morphology
A total of 260 placer PGM grains were analyzed in thisstudy. Electron microprobe analyses confirmed two maincompositional groups: 82 grains of Os-bearing laurite(14.3–31.1 wt % Os) and 174 grains of Pt-Fe alloy (73.8–91.5wt % Pt, 0.4–4.0 wt % Os) (Fig. 2, Tables A1, A2). Two grainsof Ir-Os-Pt alloy (iridium), with 53.0 wt percent Ir (BRN-2-076 and BRN-3-024), one of PtAs2 (sperrylite) (BRN-2-045)and one of PGE-bearing Au-Ag alloy were also identified.
Long-axis dimensions range from ca. 250 to 2,000 µm. Thethree largest grains are alloys; sulfide grains are generallysmaller, with a maximum diameter of 800 µm. Sulfides are
predominantly equidimensional and subrounded to rounded,although rare examples of more elongate and subangulargrains are present in this population. Subhedral grains showcubic systems (e.g., Fig. 3A). PGE alloys have more variedmorphologies; rounded, subspherical to ellipsoid alloys (e.g.,Fig. 3C) are uncommon, with the majority displaying irregu-lar, angular, and “nodular” forms (e.g., Fig. 3D-F). Two grainsin this group are subhedral; they have cubic crystal systems(e.g., Fig. 3B), which are typical of isoferroplatinum (Pt3Fe)(Harris and Cabri, 1991). Surface features are varied and in-clude pits (rounded and irregular, cubic, linear) ranging fromapproximately 1 to 40 µm in diameter, scratches, and frac-tures (Fig. 3G-L). These surface features, with the exceptionof euhedral pits, are typical of alluvially transported PGMgrains (Cabri et al., 1996; Oberthür et al., 2004). Euhedralpits are likely to represent negative crystals or melt inclusions,removed by weathering, that inherited the morphology of thehost PGM grain, as described by Brenker et al. (2003).
Methods
Laser ablation-multicollector-ICPMS (LA-MC-ICPMS)
Grains were mounted in 10 × 10 grids on adhesive carbonSEM tabs fixed to glass microscope slides. Samples were im-aged prior to isotopic analysis at Durham University using aHitachi TM-1000 tabletop scanning electron microscope.
Pt-Os isotope analyses were carried out at Durham Univer-sity Northern Center for Isotopic and Elemental Tracing(NCIET) using a New Wave UP 213 nm laser and ThermoFisher Neptune MC-ICPMS via the method presented inNowell et al. (2008b). Borneo samples were analyzed over sixsessions between 17-12-2007 and 18-02-2008. Laser spotsizes of 120 to 125 µm were used for Os-rich grains, while250-µm spots were used for Pt-rich/Os-poor samples. Laserrepetition rate was varied between 9 and 20 Hz, and laserpower density from 40 to 100 percent. Fixed value laser pa-rameters used are given in Nowell et al. (2008b).
At the start of each analytical session a 1 µg ml–1 DROsSstandard solution was analyzed 10 times in order to assess in-strument accuracy and reproducibility. Over the six analyticalsessions, the reproducibility was 122 ppm for 187Os/188Os and125 ppm for 186Os/188Os. Mean 187Os/188Os and 186Os/188Os ra-tios of 0.160918 ± 0.000020 and 0.119917 ± 0.000015, re-spectively (2 SD, n = 60) (Table A4) are identical to the val-ues of 0.160921 ± 0.000018 and 0.119917 ± 0.000020 (2 SD,n = 5) reported for this standard by Nowell et al. (2008b).
The 189Os/188Os ratio is free of elemental interferences so itwas used to correct for mass bias. A value of 1.21978 (Nowellet al., 2008b) was assumed for this correction.
184W, 186W, and 187Re occur as elemental isobaric interfer-ences on 184Os, 186Os and 187Os during laser ablation. To correctfor these interferences, 182W and 185Re were monitored dur-ing analyses and, assuming 182W/184W = 0.863376 ± 15, 182W/186W = 0.930700 ± 13, and 185Re/187Re = 0.598050 ± 13 (derivedfrom analyses of a 1 µg ml-1 DROsS solution doped with vary-ing concentrations of W and Re (see Nowell et al., 2008a), theappropriate level of interference was subtracted for each 1 sintegration to yield the corrected Os isotope ratios.
Interferences on 190Os and 192Os by 190Pt and 192Pt cannot becorrected in the same way since the faraday cup configuration
94 SCIENTIFIC COMMUNICATIONS
0361-0128/98/000/000-00 $6.00 94
Cretaceous flysch
Cenozoic sedimentary cover
Cretaceous volcanics
Cretaceous plutonic rocks
Ultramafic rocks, chert and melange
Schist and phyllite
E°611E°511
3° S
4° S
Pontyn R.
Java SeaLaut Island
SouthKalimantan
Mer
atus
Mou
ntain
s
Asemasem
N
Fig. 1. Map showing the geology of the Meratus Mountains, includingophiolitic rocks (DeWit et al., 1992), schists formed by subduction zone meta -morphism (purple), and location of the Pontyn River (after Wakita et al., 1998).
used does not provide a Pt monitor isotope. Instead Os mustbe treated as the interfering element, with 188Os taken as themonitor isotope. The mean 190Os/188Os ratio determined frommeasurements of the DROsS standard at the start of each ses-sion was used to subtract the Os interference on mass 190 andthereby derive the 190Pt intensity and the 190Pt/188Os ratio.
Each analysis is made up of 40 1-s integrations. After massbias and interfering element corrections were applied to eachmeasurement, the analyses were subject to a 2σ rejection.The method and corrections are discussed in greater detail byNowell et al. (2008b).
For plotting Pt-Os isochrones, total errors in 186Os /188Osratios were calculated to incorporate external reproducibility.A value of 176 ppm was used for grains with 188 beams of 1V or more; grains with 188 beams <1 V were assigned exter-nal reproducibility values of 352 ppm. These values were de-rived from repeat analyses over one year of an in-house stan-dard (Urals Os-rich PGE alloy 36720 G1, Nowell et al.,2008b). No such data is available for 190Pt/188Os since there isno Pt-rich homogeneous ablation standard. However, small-scale isotopic heterogeneity of individual samples leads to rel-atively large within-run precision, contributing a major com-ponent of the overall uncertainty on this ratio. In addition, asignificant portion of the uncertainty on 190Pt/188Os measure-ments may result from Pt-Os fractionation at the ablation site.Such interelement fractionation is poorly understood, but isestimated by Nowell et al. (2008b) to be 5 percent or less.Total uncertainty on the 190Pt/188Os ratio, therefore, includes
the within-run error and 5 percent uncertainty to account forexternal reproducibility and potential elemental fractionationthat may occur at the ablation site.
Electron microprobe and SEM
The PGM were transferred to polished sections for compo-sitional analysis using a CAMECA SX-100 electron microprobeat Brown University, Rhode Island. WDS analyses were carriedout with an accelerating voltage of 20 keV, beam current of 25nA and beam diameter of ~1 µm. Measurement times of 10 swere used on peaks, with 5-s measurements on the back-grounds above and below each peak. Pure metal standardswere used for all elements analyzed, with the exception of Sand As, for which FeS and GaAs were used, respectively. De-tection limits are 3σ and range from 0.088 to 0.589 wt percentfor analyses of sulfides. During analyses of alloys (high Pt sam-ples), Pt interference occurs on the Fe L alpha peak; therefore,Fe was analyzed on the K alpha peak for these samples witha detection limit of 1.37 wt percent. Full details of detectionlimits are given in Table A5. Silicate and base metal sulfide in-clusions were analyzed qualitatively, using EDS analyses.
Results
Inclusions and other internal features of Borneo PGM grains
BSE images of polished grains reveal distinct groups of in-ternal features. Sulfides are poorly polished since there is aconsiderable contrast in hardness between these grains and
SCIENTIFIC COMMUNICATIONS 95
0361-0128/98/000/000-00 $6.00 95
Ru
Os
Ir
BPt
Ir
Fe
DPt
Ir
Fe
Pt
Ir
Fe
F F’
F’
Pt
Os
Fe
C Pt-Fe alloysRu
Os
S
A SulfidesPt
Os
Fe
EPGE alloy rims
Os-IrOs>Ru>Ir>RhIr>Os>Pt>Rh,Ru
Pt-FeInclusions:
Fig. 2. Ternary plots showing relative (wt %) compositions of dominant elements in Borneo placer PGM and their inclu-sions, measured by electron microprobe. A, B. Sulfides. C, D. Alloys. E, F. PGE alloy rims on Pt-Fe alloy grains and inclu-sions in alloy grains. Full compositional data are displayed in Tables A1–A3.
PGE alloys, with which they are mounted. Despite this, it ispossible to identify inclusions hosted within many of the Bor-neo laurites. These appear as dark patches, approximately 5 to30 µm in diameter, that range from rounded to euhedral(hexagonal) and are crystallographically orientated (Fig. A1).The shape of euhedral inclusions and their orientation relativeto each other suggests that they may have been included as liq-uids and their morphological features are inherited from thehost grains. EDS analyses of the inclusions show a range ofphases; silicates (amphibole, epidote, clinopyroxene, serpen-tine, olivine, anorthite, and melt), base metal sulfides (pyrite,pyrrhotite, pentlandite, and chalcopyrite) and two compositeinclusions of silicate + sulfide + alloy were identified.
Silicate inclusions in alloy grains have diameters as great as75 µm and may be sub- or euhedral and crystallographicallyorientated, or anhedral—in the form of linear “blebs” (Fig.A2A, B). Additional internal features of alloys can be grouped
together: inclusions or intergrowths of Os- and Ir-dominatedPGE alloys (Fig. A2C-E, Table A3) ranging from ~20 to 100µm in length; micro-PGE alloy inclusions, up to 15 µm in diameter (Fig. A2F, H); and PGE alloy rims (Fig. A2K, L), upto ~200 µm on composite grains. BSE images also highlightsets of bright (or rarely, dark), submicron thickness linear fea-tures (Fig. A2G, I, J) which may be cogenetic or a product ofexsolution (Cabri and Genkin, 1991).
Pt-Os isotopes and age
A total of 260 PGM grains were analyzed by LA-MC-ICPMS. Values for 190Pt/188Os measured range from <0.00001to 1.493430 and values for 186Os /188Os range from 0.119801 to0.120315 (Table A6). This range in 186Os /188Os is more than 25times greater than that observed in typical convecting mantle(Brandon et al., 1999). Measured 186Os /188Os correlates posi-tively with 190Pt/188Os. The initial 186Os/188Os ratio calculated
96 SCIENTIFIC COMMUNICATIONS
0361-0128/98/000/000-00 $6.00 96
100 µm100 µm
J
30 µm30 µm
L
100 µm100 µm
K
200 µm200 µm
A
300 µm300 µm
B
200 µm200 µm
C
300 µm300 µm
D
300 µm300 µm
E F
300 µm300 µm
100 µm100 µm
G
300 µm300 µm
I
100 µm100 µm
H
BRN-1-031 BRN-1-035 BRN-1-041
BRN-1-015 BRN-1-026 BRN-2-085
BRN-2-033 BRN-2-036 BRN-1-030
BRN-2-052 BRN-1-064 BRN-1-082
Fig. 3. BSE images showing examples of grain morphologies and surface textures seen in Borneo placer sulfides (A, I, K,L) and Pt-Fe alloys (B-H, J). A. Subhedral laurite showing cubic crystal habit. B. Subhedral, cubic Pt-Fe alloy. C.Subspher-ical Pt-Fe alloy. D-F. Irregular, angular and “nodular” forms, typical of Pt-Fe alloys in this population. G, K. Irregular pits ongrain surfaces, often infilled with granular (dark) material, possibly chromite. H, I. Scratches and fractures consistent withtransport in an alluvial system. J. Rounded pits following linear trends. L. Cubic, parallel pits in the surface of a Pt-Fe alloy.
from the data is 0.119830 ± 0.000003 and the resultingisochron has an MSWD of 0.90, a probability of fit of 0.88 andan age of 197.8 ± 8.1 Ma (2σ: Fig. 4). The uncertainty on theage incorporates the 1 percent uncertainty on the decay con-stant of 190Pt as estimated by Begemann et al. (2001).
Variation in 190Pt/188Os is dominated by mineralogy (Pt-Fealloys vs. Os-rich laurites), whereas 186Os /188Os ratio variationis minor and is most likely dominated by radiogenic in-growth. This is addressed further in the Discussion section.Pt-Fe alloys and sperrylite have ~76 to 91 wt percent Pt, 0.4to 4.0 wt percent Os, 190Pt/188Os ratios that range from0.00098 to 1.494, and 186Os /188Os values of 0.119801 to0.120315; these phases are predominant in defining the slopeof the isochron. Laurite grains in this population have <0.2 wtpercent Pt, 14.3 to 31.1 wt percent Os and 190Pt/188Os valuesup to only 0.0743 and therefore contribute to defining the ini-tial 186Os/188Os ratio.
Re-Os isotopes
Despite a very restricted range in 187Re/188Os values(0.000005–0.002980), there is considerable variation in 187Os/188Os (0.122117–0.140674). The extremely low Re/Os ratiosresult in calculated initial 187Os/188Os values that are almostidentical to measured 187Os/188Os, illustrating that these sam-ples do not share a common initial 187Os/188Os value. Hence,Re-Os isotope systematics are scattered and do not exhibit anisochronous relationship. Studies of other ophiolite-relatedPGM have shown considerable variation in 187Os/188Os that isnot related to Re/Os variation (Meibom et al., 2002; Walkeret al., 2002; Pearson et al., 2007).
Variations in 187Re/188Os and 187Os/188Os ratios may be re-lated to R-factor fractionation (Campbell and Naldrett, 1979)during PGM growth. Alternatively, these variations may re-sult from the PGM population being derived from several dif-ferent chromitite pods, essentially formed from isolated meltswith differing initial 187Os/188Os.
Discussion
PGM genesis
There has been much debate over the formation of alluvialor detrital PGM grains. Bowles (1988) has proposed a sec-ondary origin for PGM associated with the Freetown layeredintrusion in Sierra Leone by lateritic-type crystallizationwithin placers in a tropical environment. Perhaps morewidely applicable on a global scale is the proposal of a mag-matic origin for these minerals. Several scenarios have beensuggested for crystallization of PGM in the deep mantle orcore (Bird and Bassett, 1980; Bird et al., 1999); however,many workers now conclude that ophiolitic PGM grains areformed during chromitite genesis (Ahmed, 2007; Pearson etal., 2007; Tsoupas and Economou-Eliopoulos, 2008) in thelithospheric upper mantle (podiform chromitites) or base ofthe crust (stratiform chromitites; Paktunc, 1990).
In the case of Borneo and the Meratus grains, the tropicalclimate provides the potential for secondary crystallization ofPGM by the weathering mechanism described by Bowles(1988). If the PGM grains analyzed in this study were formedby a secondary process involving recrystallization within lat-erites, then the Pt-Os isochron must represent a mixing lineand not a true or meaningful age because lateritization is wellbelow any likely diffusional reequilibration temperature forthe Pt-Os (or Re-Os) isotope system. In such a scenario, PGMformation would be late in the geologic history of the Mera-tus ophiolite, occurring after high-level exhumation of themantle section. This greatly reduces the time available for ra-diogenic Os in-growth and means that any proposed precur-sors to the PGM grains, such as ultramafic rocks or chromi-tite veins, would need to have evolved to high 186Os/188Osratios rapidly and with unusually high Pt/Os ratios. Becausethe initial 186Os/188Os ratio of typical mantle materials variesso little, both spatially and throughout geologic time, such ascenario seems very unlikely on this basis alone.
Further evidence against a secondary origin for the Mer-atus PGM grains comes from Re-Os isotope systematics andS isotope data. The similarity of 187Os/188Os ratios in placerand in situ chromite-hosted PGM in the Meratus area ledHattori et al. (1992) to conclude that mechanical processesare responsible for transporting the PGE (in the form of al-loys and sulfides) to the placer deposits and that the placergrains are direct samples of the mantle-hosted PGM grains.Sulfur isotopes along with arsenic and selenium contents ofplacer laurite (RuS2) from the Pontyn and Tambanio Rivers,southeast Borneo, confirm that these grains are derivedfrom their host chromites by purely mechanical processes.The sulfur isotope ratio of the PGM grains originating fromS-bearing inclusions such as laurite has a clear mantle sig-nature as opposed to the fractionated and variable S isotopesignatures expected if these grains had formed in near-sur-face environments (Hattori et al., 2004). Hence, there is
Fig. 4. Pt-Os isochron diagram for 260 Borneo placer PGM. Fill color oferror ellipse indicates mineralogy: gray = Pt-Fe alloy, black = laurites and oneAu-Ag alloy; white = grains not exposed by or lost during polishing. Errors on186Os/188Os incorporate within run uncertainties and long term external re-producibility on this ratio based on repeat analyses of an in-house standardgrain. 190Pt/188Os errors include 5 percent uncertainty to account for poten-tial elemental (Pt/Os) fractionation that may occur at the ablation site (Now-ell et al., 2008b). The uncertainty quoted on the isochron age incorporates anuncertainty of 1 percent on the decay constant of 190Pt (Begemann et al.,2001). Plotted using Isoplot version 3.1 (Ludwig, 2003).
powerful evidence against a low-T secondary origin for theMeratus alluvial PGM.
Silicate, sulfide, and multiphase or composite inclusions areobserved in the Borneo PGM grains. Composite inclusionsmay represent inclusions of trapped melt that later under-went fractional crystallization. Peck et al. (1992) and Brenkeret al. (2003) have used the presence of mineral and melt in-clusions, respectively, within PGM from alluvial deposits toargue strongly for a magmatic origin for such grains. A mag-matic origin is further supported by the presence of PGM(both sulfides and PGE alloys) as inclusions in chromite grains(Hattori et al., 2004; Ahmed, 2007; Tsoupas and Economou-Eliopoulos, 2008; Petrou and Economou-Eliopoulos, 2009).Ophiolitic chromitite formation may occur by crystallizationfrom a partial mantle melt after melt-rock interaction (Pak-tunc, 1990; Zhou et al., 1998) or by magma mixing (Ballhaus,1998). In either case, significant volumes of partial meltingmust occur to produce the parent magmas; thus, chromititemineralization is interpreted as a near-ridge process (Paktunc,1990) that takes place relatively soon after generation ofoceanic lithosphere.
Age of the Meratus ophiolite
The varied Pt-Os fractionation within the Meratus PGMgrains clearly offers possible geochronological constraints onthe timing of formation of the Meratus ophiolite. The firstissue to consider in evaluating this potential is the likelihoodof a single source for the PGM grains and the possibility thatthe correlation observed on the Pt-Os isochron diagram (Fig.4) may be a mixing line. Addressing the first of these issues, wespecifically selected the Meratus samples because their prove-nance is well constrained. There is good consensus that thePGM grains were derived directly from the single ophiolitebody based on close proximity and lack of any other plausiblehost rock units in the area (Hattori et al., 1992; Monnier et al.,1999). These factors mean that it is unlikely that our PGMgrain population is derived from a mixture of ultramafic bodieswith differing ages. Nonetheless, it is possible that several dif-ferent chromitite bodies within the single massif may have con-tributed to the population of grains that we have analyzed.Analyses of different chromitite bodies within a single ophioliteconfirm that such bodies can have variable 187Os/188Os ratios(e.g., Walker et al., 1996). This variability in initial 187Os/188Osratios, combined with subsequent postformation in-growth,may be a major reason why the Re-Os data for Meratus PGMare scattered, with no correlation on the isochron diagram. Incontrast to the Re-Os decay system, any variability in initialisotopic ratios due to the derivation of grains from multiplechromitite bodies would be masked in the Pt-Os system be-cause the variation in 186Os/188Os of the convecting mantlethroughout the whole of the Phanerozoic is less than 100 ppm(Brandon et al., 2006). Hence, the Pt-Os isochron correlation,in this instance, is unlikely to reflect mixing phenomena be-tween bodies with significantly different initial 186Os/188Osand seems best explained as reflecting radiogenic in-growthfrom variable Pt-Os since the time of PGM formation.
Taking the Pt-Os isochron age for Meratus PGM grains asa genesis age allows us to examine its potential accuracy in thecontext of other geochronologic data. Published plate recon-structions for the tectonically complex Java Sea allow imprecise
but independent age estimates for the Meratus ophiolite thatrange from Jurassic (Monnier et al., 1999) to Cretaceous(Parkinson et al., 1998). The application of traditional geo -chronological techniques to constrain the emplacement of theMeratus ophiolite body has proven problematic. K-Ar datingof terrigenous sediments within the infra-ophiolitic sole givesan estimated age of ophiolite accretion to the continental mar-gin of ~145 Ma, and obduction at ~90 Ma (Monnier et al.,1999). Uncertainty estimates are not provided and we note theinherent problems of dating the deposition of sediments usingthe K-Ar system (Dickin, 2005; Selby, 2009). Perhaps the mostextensive work on dating the Meratus ophiolite is that ofWakita et al. (1998), who used the K-Ar system to analyzemicas from the Hauren schist. This work yielded K-Ar ages of102 to 190 Ma, suggesting that subduction of oceanic lithos-phere was occurring as early as 190 m.y. ago; thus, bothoceanic plates at this convergent margin must be at least 190Ma. Chromitite ore forms relatively early in the history of thehost lithosphere, and certainly before the slab is subducted;thus, a subduction age for this lithosphere provides a mini-mum age for the chromitite and hence the PGM grains.Wakita et al. (1998) provide support for an Early Jurassic orlatest Triassic age for the ophiolite in the form of radiolarianbiostratigraphy of the chert that was originally deposited as asiliceous ooze overlying the igneous succession of the ophio-lite. They observed an assemblage that is assigned as earlyMiddle Jurassic to late Early Cretaceous (equiv to ~180–100Ma). Assuming that sedimentation began immediately afterformation of the crust at a spreading center, and completepreservation of the pelagic assemblage, this gives an estimatedformation age for the oceanic lithosphere at about 180 Ma.
Our Pt-Os isochron age of 197.8 ± 8.1 Ma is conformablewith the available age constraints. The initial 186Os/188Os ratioof the isochron is well within error of that expected for de-rivation from a convecting mantle source (Brandon et al.,2006). These factors, together with the low MSWD and highprobability of fit, indicate a viable age for the Pt-Os isochron;hence, we propose that the precise Pt-Os isochron age deter-mined on the Meratus PGM grains is an accurate representa-tion of the genesis of the PGE mineralization.
Further interpretation of the Pt-Os age is possible when weconsider the event that it most likely represents. The likelyformation of PGM grains during chromitite genesis is aclearly definable event in the early evolution of the oceaniclithosphere. The crystallization of the PGM grains from mag-mas, with each grain acquiring a distinct Pt-Os ratio depend-ing on its mineralogy, most likely occurred at this time. If eachgrain is isolated from others via chromite or olivine, it is likelythat the radiometric clock is set at this instant, despite the rel-atively high temperatures of this part of the oceanic lithos-phere. This early formation of PGM grains means that any Pt-Os isochron derived from them will give an age that isequivalent to the genesis age of the lithosphere itself anddates one of the earliest events recorded in the evolution ofthe rocks of the ophiolite complex.
Conclusions and Prospects for Dating Detrital PGM Deposits
We have analyzed a suite of PGM grains collected from al-luvial deposits derived from the Meratus ophiolite massif for
98 SCIENTIFIC COMMUNICATIONS
0361-0128/98/000/000-00 $6.00 98
their Pt-Os and Re-Os isotope systematics. These grains dis-play a wide variety of Pt/Os fractionations while Re/Os valuesare restricted to a narrow range. In the case of the Re-Os sys-tem, no correlation exists on a Re-Os isochron diagram andno chronological information can be derived. We attributethis largely to inherent Os isotope heterogeneity in the origi-nal magmas from which the PGM grains formed. The meanvalue of the initial 187Os/188Os(197.8 Ma) ratios of the grains ana-lyzed is similar to the average value found in other PGMsuites of Mesozoic age (Hattori and Hart, 1991; Büchl et al.,2002; Meibom et al., 2004). This composition most likely rep-resents that of the depleted upper convecting mantle at thetime of lithosphere formation.
In contrast to the nonsystematic Re-Os isotope characteris-tics, a well-defined correlation exists on a Pt-Os isochron dia-gram, defining an age of 197.8 ± 8.1 Ma that we interpret asthe age of formation of the PGM grains in the lower oceaniclithosphere. This age is consistent with radiometric and bio -stratigraphic age constraints available for the Meratus ophio-lite (Wakita et al., 1998). This result, along with other exam-ples recently documented by Nowell et al. (2008b), illustratesthe potential of the laser ablation Pt-Os isotope method indating PGM mineralization associated with ultramafic rocks.This type of LA-MC-ICPMS analysis can be applied to detri-tal or in situ populations but a given sample type needs toyield a large population of PGM grains, preferably >100, witha variety of Pt/Os ratios. Not all PGM suites possess this rangeof Pt-Os ratios. Alternatively, Nowell et al. (2008b) haveshown that there may be sufficient Pt/Os variation in exsolvedor polycrystalline PGM to enable geochronological data to beobtained. This approach thus provides a new way of datingPGE mineralization both within ophiolites and within placeror detrital deposits derived from these sources.
AcknowledgementsThis research was supported by a National Environment
Research Council (NERC) research studentship to JAC and aNERC research grant (NE/F006497/1). We thank Paul Poh -wat of the Smithsonian Institution National Museum of Nat-ural History for the provision of the Meratus samples. JoeDevine and Kassandra Costa are thanked for their assistancewith electron microprobe analyses. KH and BE are thankedfor providing valuable critical reviews of the manuscript.
REFERENCESAhmed, A.H., 2007, Diversity of platinum-group minerals in podiform
chromitites of the Late Proterozoic ophiolite, Eastern Desert, Egypt: Ge-netic implications: Ore Geology Reviews, v. 32, p. 1–19.
Ballhaus, C., 1998, Origin of podiform chromite deposits by magma min-gling: Earth and Planetary Science Letters, v. 156, p. 185–193.
Begemann, F., Ludwig, K.R., Lugmair, G.W., Min, K., Nyquist, L.E., Patch-ett, P.J., Renne, P.R., Shih, C.Y., Villa, I.M., and Walker, R.J., 2001, Call foran improved set of decay constants for geochronological use: Geochimicaet Cosmochimica Acta, v. 65, p. 111–121.
Bird, J.M., and Bassett, W.A., 1980, Evidence of a deep mantle history in ter-restrial osmium-iridium-ruthenium alloys: Journal of Geophysical Re-search, v. 85, p. 5461–5470.
Bird, J.M., Meibom, A., Frei, R., and Nägler, T.F., 1999, Osmium and leadisotopes of rare OsIrRu minerals: Derivation from the core-mantle bound-ary region?: Earth and Planetary Science Letters, v. 170, p. 83–92.
Bowles, J.F.W., 1988, Further studies of the development of platinum-groupminerals in the laterites of the Freetown Layered Complex, Sierra Leone,in Prichard, H.M., Potts, P.J., Bowles, J.F.W. and Cribb, S.J., ed., Geo-Plat-inum 87, New York, Elsevier Applied Science, p. 273–280.
Brandon, A.D., Walker, R.J., Morgan, J.W., Norman, M.D. and Prichard,H.M., 1998, Coupled Os-186 and Os-187 evidence for core-mantle interac-tion: Science, v. 280, 1570–1573.
Brandon, A.D., Norman, M.D., Walker, R.J. and Morgan, J.W., 1999, 186Os-187Os systematics of Hawaiian picrites: Earth and Planetary Science Let-ters, v. 174, p. 25–42.
Brandon, A.D., Walker, R.J., and Puchtel, I.S., 2006, Platinum-osmium iso-tope evolution of the Earth‘s mantle: Constraints from chondrites and Os-rich alloys: Geochimica et Cosmochimica Acta, v. 70, p. 2093–2103.
and Os-187 enrichments and high-He-3/He-4 sources in the Earth’s mantle:Evidence from Icelandic picrites: Geochimica et Cosmochimica Acta, v. 71,p. 4570–4591.
Brenker, F.E., Meibom, A., and Frei, R., 2003, On the formation of peri-dotite-derived Os-rich PGE alloys.: American Mineralogist, v. 88, p.1731–1740.
Büchl, A., Brügmann, G., Batanova, V.G., Münker, C., and Hofmann, A.W.,2002, Melt percolation monitored by Os isotopes and HSE abundances: Acase study from the mantle section of the Troodos Ophiolite: Earth andPlanetary Science Letters, v. 204, p. 385–402.
Burgath, K.P., 1988, Platinum-group minerals in ophiolitic chromitites andalluvial placer deposits, Meratus-Bobaris area, southeast Kalimantan, inPrichard, H.M., Potts, P.J., Bowles, J.F.W., and Cribb, S.J., ed., Geo-Plat-inum ‘87, New York, Elsevier Applied Science, p. 383–403.
Burgath, K.P., and Mohr, M., 1986, Chromitites and platinum-group miner-als in the Meratus-Bobaris ophiolite zone, south-east Borneo, in Gallagher,M.J., Ixer, R.A., Neary, C.R., and Prichard, H.M., ed., Metallogeny of basicand ultrabasic rocks: London, The Institution of Mining and Metallurgy, p.333–349.
Cabri, L.J., and Genkin, A.D., 1991, Reexamination of Pt-alloys from lodeand placer deposits, Urals: Canadian Mineralogist, v. 29, p. 419–425.
Cabri, L.J., Harris, D.C., and Weiser, T.W., 1996, Mineralogy and distribu-tion of platinum-group mineral (PGM) placer deposits of the world: Ex-ploration and Mining Geology, v. 5, p. 73–167.
Campbell, I.H., and Naldrett, A.J., 1979, The influence of silicate:sulfide ra-tios on the geochemistry of magmatic sulfides: ECONOMIC GEOLOGY, v. 74,p. 1503–1506.
DeWit, M.J., Roering, C., Hart, R.J., Armstrong, R.A., Deronde, C.E.J.,Green, R.W.E., Tredoux, M., Peberdy, E., and Hart, R.A., 1992, Formationof an Archean continent: Nature, v. 357, p. 553–562.
Guntoro, A., 1999, The formation of the Makassar strait and the separationbetween SE Kalimantan and SW Sulawesi: Journal of Asian Earth Sciences,v. 17, p. 79–98.
Harris, D.C., and Cabri, L.J., 1991, Nomenclature of platinum-group-ele-ment alloys—Review and Revision: Canadian Mineralogist, v. 29, p.231–237.
Hattori, K., and Hart, S.R., 1991, Osmium-isotope ratios of platinum-groupminerals associated with ultramafic intrusions: Os-isotopic evolution of theoceanic mantle: Earth and Planetary Science Letters, v. 107, p. 499–514.
Hattori, K., Burgath, K.P., and Hart, S.R., 1992, Os-isotope study of plat-inum-group minerals in chromitites in alpine-type ultramafic intrusionsand the associated placers in Borneo: Mineralogical Magazine, v. 56, p.157–164.
Hattori, K.H., Cabri, L.J., Johanson, B., and Zientek, M.L., 2004, Origin ofplacer laurite from Borneo: Se and As contents, and S isotopic composi-tions: Mineralogical Magazine, v. 68, p. 353–368.
Ludwig, K., 2003, Isoplot/Ex, version 3: A geochronological toolkit for Mi-crosoft Excel, Geochronology Centre, Berkeley,California.
Luguet, A., Pearson, D.G., Nowell, G.M., Dreher, S.T., Coggon, J.A., Spet-sius, Z.V., and Parman, S.W., 2008, Enriched Pt-Re-Os isotope systematicsin plume lavas explained by metasomatic sulfides: Science, v. 319, p.453–456.
Meibom, A., Sleep, N.H., Chamberlain, C.P., Coleman, R.G., Frei, R., Hren,M.T., and Wooden, J.L., 2002, Re-Os isotopic evidence for long-lived het-erogeneity and equilibration processes in the Earth’s upper mantle.: Na-ture, v. 419, p. 705–708.
Meibom, A., Frei, R., and Sleep, N.H., 2004, Osmium isotopic compositionsof Os-rich platinum group element alloys from the Klamath and SiskiyouMountains: Journal of Geophysical Research, p. 109, p. B02203.
Monnier, C., Polve, M., Girardeau, J., Pubellier, M., Maury, R.C., Bellon, H.,and Permana, H., 1999, Extensional to compressive Mesozoic magmatism
SCIENTIFIC COMMUNICATIONS 99
0361-0128/98/000/000-00 $6.00 99
at the SE Eurasia margin as recorded from the Meratus ophiolite (SE Bor-neo, Indonesia): Geodinamica Acta, v. 12, p. 43–55.
Nowell, G.M., Luguet, A., Pearson, D.G., and Horstwood, M.S.A., 2008a,Precise and accurate 186Os/188Os and 187Os/188Os measurements by multi-collector plasma ionisation mass spectrometry (MC-ICP-MS) part I: Solu-tion analyses: Chemical Geology, v. 248, p. 363–393.
Nowell, G.M., Pearson, D.G., Parman, S.W., Luguet, A., and Hanski, E.,2008b, Precise and accurate 186Os/188Os and 187Os/188Os measurements bymulti-collector plasma ionisation mass spectrometry, part II: Laser ablationand its application to single-grain Pt-Os and Re-Os geochronology: Chem-ical Geology, v. 248, p. 394–426.
Oberthur, T., Melcher, F., Gast, L., Wohrl, C., and Lodziak, J., 2004, Detri-tal platinum-group minerals in rivers draining the eastern Bushveld com-plex, South Africa: Canadian Mineralogist, v. 42, p. 563–582.
Paktunc, A.D., 1990, Origin of podiform chromite deposits by multistagemelting, melt segregation and magma mixing in the upper mantle: Ore Ge-ology Reviews, v. 5, p. 211–222.
Parkinson, C.D., Miyazaki, K., Wakita, K., Barber, A.J., and Carswell, D.A.,1998, An overview and tectonic synthesis of the pre-Tertiary very high pres-sure metamorphic and associated rocks of Java, Sulawesi and Kalimantan,Indonesia: The Island Arc, v. 7, p. 184–200.
Pearson, D.G., Parman, S.W., and Nowell, G.M., 2007, A link between largemantle melting events and continent growth seen in osmium isotopes: Na-ture, v. 449, p. 202–205.
Peck, D.C., Keays, R.R., and Ford, R.J., 1992, Direct crystallization of re-fractory platinum-group element alloys from boninitic magmas: Evidencefrom western Tasmania: Australian Journal of Earth Sciences, v. 39, p.373–387.
Petrou, A.L., and Economou-Eliopoulos, M., 2009, The activation energyvalues estimated by the Arrhenius equation as a controlling factor of plat-inum-group mineral formation: Geochimica et Cosmochimica Acta, v. 73,p. 1625–1636.
Selby, D., 2009, U-Pb zircon geochronology of the Aptian/Albian boundaryimplies that the GL-O international glauconite standard is anomalouslyyoung: Cretaceous Research, v. 30, p. 1263–1267.
Sikumbang, N., 1990, The geology and tectonics of the Meratus mountains,South Kalimantan, Indonesia: Geologi Indonesia, v. 13, p. 1–31.
Tsoupas, G., and Economou-Eliopoulos, M., 2008, High PGE contents andextremely abundant PGE-minerals hosted in chromitites from the Veriaophiolite complex, northern Greece: Ore Geology Reviews, v. 33, p. 3–19.
Wakita, K., Miyazaki, K., Zulkarnain, I., Sopaheluwakan, J., and Sanyoto, P.,1998, Tectonic implications of new age data for the Meratus complex ofsouth Kalimantan, Indonesia: Island Arc, v. 7, p. 202–222.
Walker, R.J., Morgan, J.W., Naldrett, A.J., Li, C., and Fassett, J.D., 1991, Re-Os isotope systematics of Ni-Cu sulfide ores, Sudbury Igneous Complex,Ontario—evidence for a major crustal component: Earth and PlanetaryScience Letters, v. 105, p. 416–429.
Walker, R.J., Hanski, E., Vuollo, J., and Liipo, J., 1996, The Os isotopic com-position of Proterozoic upper mantle: Evidence for chondritic upper man-tle from the Outokumpu ophiolite, Finland: Earth and Planetary ScienceLetters, v. 141, p. 161–173.
Walker, R.J., Morgan, J.W., Hanski, E.J., and Smolkin, V.F., 1997a, Re-Os sys-tematics of Early Proterozoic ferropicrites, Pechenga Complex, northwest-ern Russia: Evidence for ancient Os-187-enriched plumes: Geochimica etCosmochimica Acta, v. 61, p. 3145–3160.
Walker, R.J., Morgan, J.W., Beary, E.S., Smoliar, M.I., Gzamanske, G.K., andHoran, M.F., 1997b, Applications of the 190Pt-186Os isotope sytem to geo-chemistry and cosmochemistry: Geochimica et Cosmochimica Acta, v. 61,p. 4799–4807.
Walker, R.J., Prichard, H.M., Ishiwatari, A., and Pimentel, M., 2002, The os-mium isotopic composition of convecting upper mantle deduced fromophiolite chromites: Geochimica et Cosmochimica Acta, v. 66, p. 329–345.
Zhou, M.-F., Sun, M., Keays, R.R., and Kerrich, R.W., 1998, Controls onplatinum-group elemental distributions of podiform chromitites: A casestudy of high-Cr and high-Al chromitites from Chinese orogenic belts:Geochimica et Cosmochimica Acta, v. 62, p. 677–688.
100 SCIENTIFIC COMMUNICATIONS
0361-0128/98/000/000-00 $6.00 100
SCIENTIFIC COMMUNICATIONS 101
0361-0128/98/000/000-00 $6.00 101
TABLE A1. Compositions of Borneo Placer PGM Sulfides Analyzed by Electron Microprobe at Brown University, RI, USA
(wt %)
Grain Ru Rh Pd Re Ir Pt Os Fe Co Ni Cu W S As Total
1 mean (n = 2)2 mean (n = 9)3 mean (n = 11)4 mean (n = 3)5 mean (n = 4)b.d.l. = below detection limits; detection limits are 3 sigma - given in Table A5
TABLE A2. (Cont.)
(wt %)
Grain Ru Rh Pd Re Ir Pt Os Fe Co Ni Cu Mo W Total
106 SCIENTIFIC COMMUNICATIONS
0361-0128/98/000/000-00 $6.00 106
TABLE A3. Compositions of Alloy Inclusions and Composite Grains in Borneo Placer PGE Alloy Grains Analyzed by Electron Microprobe at Brown University, RI, USA
TABLE A4. Mean Os Compositions and Reproducibility for 1 ppm DROsS Reference Material Solution Measured at the Start of MC-ICPMS Sessions during which Borneo PGM Were Analyzed
All MC-ICP-MS data corrected for mass bias using an exponential law and a 189Os/188Os ratio of 1.21978
TABLE A5. Detection Limits for EMP Analyses of Borneo PGM
Element Peak Detection limit (%)
Ru La 0.120Rh La 0.088Pd Lb 0.220Re La 0.308Ir La 0.352Pt La 0.473Os Ma 0.163
Fe (low Pt) La 1.37Fe (high Pt) Ka 0.120
Co Ka 0.135Ni Ka 0.097Cu La 0.297W La 0.589S Ka 0.090As Lb 0.482Mo La 0.109
Detection limits are 3 sigma standard deviation; for high Pt samples, Fewas analyzed on the K alpha peak since Pt interference occurs on the Fe Lalpha peak
108 SCIENTIFIC COMMUNICATIONS
0361-0128/98/000/000-00 $6.00 108
100µm
100µm
50µm
20µm
A B
C D
BRN-3-35BRN-3-35 BRN-3-65BRN-3-65
OlOl
EpEp
CpyCpyAmphAmph
AmphAmph
EpEp
Fig. A1. BSE images of inclusions in laurite grains. Dashed lines highlight the parallel orientation of different inclusionswithin the same grain, indicating that the habit of inclusions is likely dominated by the host PGM grain. Boxes in A and Bshow areas enlarged in C and D, respectively. Irregular pits and fractures are an artifact of poor polishing, due to polishingboth sulfides and PGE alloys in the same mount.
SCIENTIFIC COMMUNICATIONS 109
0361-0128/98/000/000-00 $6.00 109
100µm
100µm
100µm
100µm
100µm100µm100µm
500µm
200µm200µm
20µm
20µm
LKJ
IHG
FED
CBA
BRN-3-071 BRN-3-088 BRN-3-021
BRN-1-009 BRN-2-075 BRN-1-034
BRN-1-100 BRN-2-071 BRN-2-033
BRN-2-022 BRN-3-067 BRN-3-036
Fig. A2. BSE images of internal features in Pt-Fe alloy grains show that internal heterogeneity in this population spans arange of scales and compositions. A, B. Silicate inclusions. C-E. PGE alloy inclusions or intergrowths dominated by Os andIr. F and H. Micro-PGE alloy inclusions. G, I and J. Linear features that may be cogenetic or products of exsolution. K, L.Irregular PGE alloy rims.
110 SCIENTIFIC COMMUNICATIONS
0361-0128/98/000/000-00 $6.00 110
TABLE A6. Os Isotope Compositions of Borneo Placer PGM
Mineralogy is based on EMP analyses (see Tables A1 and A5)Within run uncertainties are quoted as 2SE; total absolute error for 186Os/188Os ratios incorporates an estimate of external reproducibility based on re-
peat analyses, over a period of one year, of an in-house LA standard (Urals Os-rich PGE alloy 36720 G1 (Nowell et al., 2008); for 190Pt/188Os ratios Total ab-solute error includes 5% uncertainty on elemental (Pt/Os) fractionation that may occur at the ablation site