Platinum, and Gold Distribution in Serpentinite Seamounts in the Mariana and Izu-bonin Forearcs: Evidence from Leg 125 Fluids and Serpentevites

Palladium, platinum, and gold were analyzed for 20 interstitial water samples from Leg 125. No Pd or Pt was detected in fluids from serpentinite muds from Conical Seamount in the Mariana forearc, indicating that low-temperature seawater-peridotite interaction does not mobilize these elements into the serpentinizing fluids to levels above 0.10 parts per billion (ppb) in solution. However, Au may be mobilized in high pH solutions. In contrast, fluids from vitric-rich clays on the flanks of the Torishima Seamount in the Izu-Bonin forearc have Pd values of between 4.0 and 11.8 nmol/L, Pt values between 2.3 and 5.0 nmol/L and Au values between 126.9 and 1116.9 pmol/L. The precious metals are mobilized, and possibly adsorbed onto clay mineral surfaces, during diagenesis and burial of the volcanic-rich clays. Desorption during squeezing of the sediments may produce the enhanced precious metal concentrations in the analyzed fluids. The metals are mobilized in the fluids probably as neutral hydroxide, bisulfide, and ammonia complexes. Pt/Pd ratios are between 0.42 and 2.33, which is much lower than many of the potential sources for Pt and Pd but is consistent with the greater solubility of Pd compared with Pt in most natural low-temperature fluids. INTRODUCTION During dredging and submersible studies of the Mariana forearc, scientists discovered a series of serpentinite seamounts in the outerhalf of the forearc terrane (Fryer et al., 1985; Ishii, 1985; Fryer and Fryer, 1987). Detailed A/Wn submersible studies of Conical Seamount in 1987 found that the seamount was made up of serpentinite flows, and was actively venting cold, dense fluids through aragonite chimneys from its summit (Fryer et al., unpubl. data). The fluids sampled had high pH and were depleted in chloride, calcium, and magnesium and enriched in silica, methane and sulfide relative to the ambient seawater (Fryer et al, unpubl. data). Experimental work by Seyfried and Dibble (1980) and Janecky and Seyfried (1987) produced fluids of similar chemistry during serpentinization of harzburgite. Moreover, the fluids are similar to some of the present-day serpentinizing fluids analyzed from the Coastal Ranges ophiolites of the western United States (Barnes et al., 1970; 1972). During Ocean Drilling Program (ODP) Leg 125, five sites were drilled on two of these serpentinite seamounts (Fig. 1): two flank sites (778 and 779) and one summit site (780) on Conical Seamount in the Mariana forearc; and one upper flank (783) and one lower flank (784) site on Torishima Forearc Seamount in the Izu-Bonin forearc. Conical Seamount was shown predominantly to comprise serpentinite-rich muds in which serpentinized ultramafic and metabasalt clasts were entrained. In contrast, Torishima Forearc Seamount was found to be made up predominantly of vitric rich muds and sands overlying a lower serpentinite mud section. The thick sediment cover and the lack of recent serpentinite flows suggest that Torishima forearc Seamount is no longer actively venting fluids. Detailed lithologic sections of the holes samples are shown in Figure 2. Shipboard and post-cruise studies of the interstitial waters squeezed from the sediments revealed a similar chemistry to that of fluids sampled by the Alvin dives, with low chloride, high pH (up to 12.6), and high concentrations of carbonate, bicarbonate, sulfate ions, and the light hydrocarbons, methane, ethane, and propane (Mottl and Haggerty, this volume). Preliminary work by shipboard scientists suggested that three different sources contributed to the unusual 'Fryer, P., Pearce, J. A., Stokking, L. B., et al., 1992. Proc. ODP, Sci. Results, 125: College Station, Texas (Ocean Drilling Program). department of Geological Sciences, Science Laboratories, South Road, Durham, DH1 3LE, U.K. Geological Survey of Canada Sector, 601 Booth Street, Ottawa, K1AOE8, Canada. chemistry of these fluids. Seawater reaction with harzburgites and metabasics, diagenetic pore fluids, and a possible deep-sourced fluid from the downgoing Pacific slab (Leg 125 Shipboard Party, 1989). A detailed discussion of the fluid geochemistry is presented by Mottl and Haggerty (this volume). The mobility of platinum-group elements (PGE) in hydrothermal solutions has only recently received attention by geologists. This was mainly because of the beliefs (1) that PGEs are inert under most geological environments, (2) that PGE deposits can be explained by magmatic processes, and (3) the scarcity of proven hydrothermal PGE deposits. However, recent studies of classic magmatic PGE sulfide deposits in the Bushveld and Stillwater complexes suggest that latestage chloride-rich fluids may be important in the redistribution of Pt and Pd (Boudreau et al., 1986). Also, several stratiform sedimenthosted base metal deposits, such as the Kupferschiefer (Kucha, 1982), have elevated Pt and Pd values, which are thought to be related to hydrothermal mobilization of these elements. Work on laterites by Bowles (1988) suggested that PGEs were mobile in the surficial environment, a conclusion also reached by Wood and Vlassopoulos (1990). In contrast, many Au deposits are hydrothermal in origin and a considerable body of work exists about the transport and deposition of Au in hydrothermal systems (Seward, 1984). Recent research on the mantle sections of ophiolites has shown some rocks to have high PGE values. The highest values are from ultramafic rocks in the Unst ophiolite in the Shetland Isles, which has been suggested to have a supra-subduction zone (SSZ) origin (Prichard and Lord, 1988). The peridotites drilled during Leg 125 are petrographically similar to SSZ ophiolite peridotites (Ishii and Robinson, this volume). Serpentinization has long been suggested as a potential process for mobilizing precious metals (e.g., Stumpf!, 1974), and both serpentinized peridotites and fluids from Leg 125 have been analyzed for this study. The interstitial pore fluids collected during Leg 125 have allowed us to undertake a preliminary study of the concentration of Pd, Pt, and Au in serpentinizing and diagenetic fluids. Fluids from the actively venting Conical Seamount and the "extinct" Torishima forearc Seamount have been analyzed to study the effects of fluid composition and source rock on the solubility of Pd, Pt, and Au. This research allows us to study (1) the absolute concentrations of Pd, Pt, and Au in the fluids, (2) how these elements are mobilized in solution in comparison to the thermodynamic models of Mountain and Wood (1988a,b), Wood et al (1989), Vlassopoulos and Wood (1990), and Vlassopoulos et al. (1990), and (3) the source of the precious metals.