Silver isotope analysis of gold nuggets: An appraisal of instrumental isotope fractionation effects and potential for high-resolution tracing of placer gold
Graphical abstract
Introduction
Silver has two stable isotopes, 107Ag (51.4%) and 109Ag (48.6%). Natural variations of 109Ag/107Ag are of interest in cosmochemistry, planetary core formation and volatile element budgets (Chen and Wasserburg, 1996; Hauri et al., 2000; Woodland et al., 2005; Schönbächler et al., 2007; Schönbächler et al., 2008, Schönbächler et al., 2010) and as a ‘stable isotope’ tracer in economic geology (Hauri et al., 2000; Chugaev and Chernyshev, 2012; Tessalina, 2015;s Argapadmi et al., 2018; Mathur et al., 2018), archaeometry (Desaulty et al., 2011) and environmental research (Desaulty and Albarede, 2013; Yang et al., 2009). The known natural isotopic variation in 109Ag/107Ag found in basaltic rocks, clastic sediments, pyrite, and in silver and gold deposits, is almost 3‰. By comparison, Ag isotope measurements with external precision in the range ±0.01–0.03‰ can routinely be obtained using multi-collector inductively coupled plasma mass spectrometry (MC-ICP-MS, e.g. (Schönbächler et al., 2007; Luo et al., 2010; Desaulty et al., 2011), allowing resolution of relatively small Ag isotopic differences.
Silver is a ubiquitous component in natural gold (lode and detrital, Hough et al., 2009; Hough et al., 2007) and in archaeological gold artefacts (e.g., Schlosser et al., 2012), varying in abundance from trace-levels to % levels, and reaching >30% in some gold nuggets (Hough et al., 2009). Both elements are often transported together in hydrothermal systems and surficial fluids (Saunders et al., 2014; Seward et al., 2014). Silver isotope data in natural gold may thus be used to trace the source rocks of gold (e.g. Tomkins, 2013) and/or help constrain physicochemical processes that lead to gold deposition. Limited published data for natural gold (18 samples) from diverse mineralization styles and locations show a range of 0.9‰ (δ109Ag −0.42–+0.5) relative to the NIST SRM978a silver reference material (Chugaev and Chernyshev, 2012; Argapadmi et al., 2018). Small variations within this relatively large range are readily resolved with current MC-ICPMS techniques, suggesting that Ag isotope systematics may have considerable potential as an additional tracer of gold, complementing other isotopic (Pb and Cu isotopes, e.g., Bendall et al., 2009; Kamenov et al., 2013; Desaulty and Albarede, 2013; Standish et al., 2013) and microchemical (Outridge et al., 1998; Chapman et al., 2006; Chapman et al., 2010; Brostoff et al., 2009) techniques.
Here we describe a new method for the extraction, purification and high-precision isotope analysis of Ag in natural and processed gold, with particular emphasis on the control of Ag yield and effects of instrumental mass bias. Isotopic analyses on the MC-ICP-MS employed the combination of Pd-doping and standard bracketing used by other authors (e.g. Schönbächler et al., 2007) to produce 109Ag/107Ag ratios with an external precision of <20 ppm, 2 sd (or <0.02‰). A solution prepared from several gold nuggets was characterised isotopically for use as a silver-in‑gold reference material (CEZAg) in Ag isotope studies of natural and processed gold.
The method was used to investigate chemically diverse detrital gold from placers in Central Africa, Bolivia and the Rhine River in Germany. One of the main objectives of this work was to examine possible effects of supergene processing which are known to modify the morphology, surface texture, internal crystal structure and chemical composition of gold grains (Groen et al., 1990; Hough et al., 2007; Fairbrother et al., 2012; Stewart et al., 2017). Silver depletion in rims and along crystal boundaries within grains are common features of gold particles and possibly involve abiotic and biotic redox processes. In hydrothermal settings, Ag isotopic variation may be produced through redox changes that occur during deposition of dissolved Ag to form native metal or metal alloy (Mathur et al., 2018; Argapadmi et al., 2018). Preliminary low-temperature experiments suggest that redox processes also cause significant Ag isotope variations in the surficial environment where gold nuggets are formed and modified (Mathur et al., 2018), consistent with observations for other metals (Cu and Fe: Teng et al., 2017). We report Ag isotope data for different sections in single gold nuggets to examine if these processes produce resolvable intra-grain Ag isotope gradients or step changes. Studies of this type are essential to establish if and how Ag isotopes respond to supergene alteration processes in detrital gold. This in turn affects the potential use of Ag isotopes to trace particular placers to primary sources, an aspect we address here by examining the Ag isotope range within a single gold placer.
Section snippets
Gold samples
Natural gold nuggets were sourced from Central Africa, Bolivia and the Rhine River in Germany. A large (3.2 g) gold nugget from Central Africa, collected near the border between Burundi and the Democratic Republic of Congo and probably sourced from a primary Au deposit within a pan-African (~500 Ma) shear zone, was provided by Prof. B. Lehmann (Technical University of Clausthal, Germany). This specimen was large enough to be sectioned with a steel scalpel, to produce nine fragments representing
Results
The mean results of the analyses of the established Ag reference materials (SRM 978a, CertiPUR), the laser standard, the newly prepared CEZAg gold solution, and the Ag content of the natural gold nuggets from different placers are summarized in Table 3. Data of individual measurements including different parts of single gold nuggets are given in the electronic Annexes (Table S1, Table S2, Table S3, Table S4, Table S5, Table S6). The mean δ109Ag value for the gold nuggets is 0.02 ± 0.62‰ (2 s).
Mass bias behaviour of Ag and Pd
For consistency with earlier studies, we used 108Pd/105Pd to constrain mass bias factors for 107Ag/109Ag. However, other Pd isotope ratios are potentially better alternatives. For example, 110Pd/106Pd has the same average isotope mass (108) as 109Ag/107Ag and should thus be more suitable for mass bias correction than 108Pd/105Pd (average mass 106.5). In our sample and reference solution, 111Cd+ intensities were never larger than the background noise (Electronic Annex Table S1, Table S2, Table S3
Concluding remarks
The technique described here allows Ag isotope ratios to be determined with an intermediate precision in the range 0.01–0.04‰ (2 s) on natural and archaeological artefact gold samples as small as 0.03 mg. Gold nuggets show an Ag isotopic range (δ109Ag −0.57‰–+0.83) larger than the currently known range for primary gold (~0.8‰) reported in Chugaev and Chernyshev (2012) and Argapadmi et al. (2018). Most of the data cluster near −0.2-0.0‰, similar to basaltic rocks, but more extreme compositions
Acknowledgment
We are grateful to Prof. B. Lehmann (Technical University of Clausthal, Germany) for providing the gold samples from the Tipuani placer in Bolivia and from Burundi/Democratic Republic of Congo. We appreciate very much the assistance of Nicole Lockhoff and Sigrid Klaus in the laboratory. The original manuscript benefited substantially from thoughtful comments by two anonymous reviewers.
References (50)
- et al.
Trace element fingerprinting of ancient Chinese gold with femtosecond laser ablation-inductively coupled mass spectrometry
J. Archaeol. Sci.
(2009) - et al.
Microchemical characterisation of natural gold and artefact gold as a tool for provenancing prehistoric gold artefacts: a case study in Ireland
Appl. Geochem.
(2006) - et al.
Supergene gold transformation: biogenic secondary and nano-particulate gold from arid Australia
Chem. Geol.
(2012) - et al.
Copper fractionation in acid mine drainage
Geochim. Cosmochim. Acta
(2009) - et al.
Precise analysis of copper and zinc isotopic compositions by plasma-source mass spectrometry
Chem. Geol.
(1999) - et al.
Fractionation of silver isotopes in native silver explained by redox reactions
Geochim. Cosmochim. Acta
(2018) - et al.
Geological and geochemical controls on the silver content (fineness) of gold in gold-silver deposits
Ore Geol. Rev.
(1991) - et al.
Determination of trace elemental signatures in placer gold by laser ablation–inductively coupled plasma–mass spectrometry as a potential aid for gold exploration
J. Geochem. Explor.
(1998) - et al.
The Witwatersrand Basin, South Africa: geological framework and mineralization processes
Ore Geol. Rev.
(1995) - et al.
Ca isotope fractionation on the Earth and other solar system materials
Geochim. Cosmochim. Acta
(1978)