Gold economic resources on Earth result from an exceptional concentration phenomenon yielding metal contents in ore of thousand to million times higher than those in common rocks. This process is thought to be controlled by aqueous fluids enriched in salt, sulfur and CO2, which flow through the Earth’s crust, extract the metal from rocks or magmas and transport and precipitate it in the right place at the right moment. Yet, the factors responsible for this transfer remain enigmatic in the face of the extraordinary chemical inertness of gold and the weak capacities of the major fluid components, chloride (Cl), hydrogen sulfide (H2S/HS) and sulfate (SO42–), to solubilise this noble metal. In particular, all gold and sulfur speciation models ignore sulfur radical ions (S3) recently shown to be stable in aqueous fluid phase at elevated temperature (from 200 to 700°C) and pressure (to 30 kbar) [1,2].

To quantify the effect of the radical ions on Au behaviour in hydrothermal fluids, we combined in situ X-ray absorption spectroscopy and solubility measurements with first-principles molecular dynamics (FPMD) and thermodynamic modelling of Au structure and speciation in aqueous solutions saturated with gold metal and containing hydrogen sulfide, sulfate and S3. These experimental solutions are representative of fluids that formed major types of gold deposits in the crust (temperatures to 500°C, pressures to 200 MPa equivalent to ~ 7 km depth, sulfur contents to 3 wt%, and NaCl-KCl salt contents to 20 wt%).

Direct evidence for gold-trisulfur ion complexes

Fig. 97: Direct evidence for gold-trisulfur ion complexes from in situ solubility measurements of gold in a sulfur-bearing aqueous solution as a function of temperature at 600 bar.

XAS experiments were carried out at beamline BM30B (FAME) using a unique spectroscopic cell that enables simultaneous measurement of both total metal concentration in the fluid and its local atomic structure [3]. We found that the presence of S3 in the fluid yields Au solubility enhancement by a factor of 10 to 100 compared to the traditional Au chloride and sulfide complexes such as Au(HS)2 and AuCl2 (Figure 97). XAS spectra, aided by FPMD simulations, indicate that S3 binds Au by forming complexes of the type Au(HS)S3 (Figure 98). These data, complemented by Au solubility measurements using a flexible-cell hydrothermal reactor, were analysed with a thermodynamic model that allowed the stability of the new Au-trisulfur ion species to be constrained across a wide range of geological conditions, from deep subduction-zone magmas to thermal springs at the surface.

XANES spectra at Au L3-edge of S-bearing experimental solutions

Fig. 98: (a) XANES spectra at Au L3-edge of S-bearing experimental solutions and reference compounds indicating the formation of Au-S3 bonds in S3-rich solutions. (b) Experimental EXAFS spectrum at 400°C and 600 bar of the solution from Figure 97 (red curve) compared with FPMD simulated spectra (black curves) of different Au-H-O-S clusters (Au = pink, S = yellow, H = grey, O = red), showing the best match between experiment and theory for clusters [HS-Au-S3].

Applying this model to natural fluids shows that sulfur radical ions, even though less abundant than sulfide or chloride, are capable of extracting large amounts of gold from magmas or rocks at depth and transporting the metal in high concentrations through the Earth’s crust. When these hot fluids rise to the surface, cool down or encounter a rock of different composition (e.g., carbonate or organic-rich), the sulfur radicals break down and deposit the metal in veins and cavities. As such, the discovery of soluble and mobile Au-S3 complexes helps explain the enigma of gold deposit formation and offers new possibilities for resource prospecting. Furthermore, such complexes may find applications in ore processing and hydrothermal synthesis of Au-based nanomaterials. This study shows that old gold known from Antiquity has yet to reveal all its secrets.


Principal publication and authors

Sulfur radical species form gold deposits on Earth, G.S. Pokrovski (a), M.A. Kokh (a), D. Guillaume (a), A.Y. Borisova (a), P. Gisquet (a),  J.-L. Hazemann (b), E. Lahera (c), W. Del Net (c),  O. Proux (c), D. Testemale (b), V. Haigis (d, e),  R. Jonchière (d,e), A.P. Seitsonen (d), G. Ferlat (e), R. Vuilleumier (d), A.M. Saitta (e),  M.-C. Boiron (f) and J. Dubessy (f), Proc. Nat. Acad. Sci. USA (PNAS) 112(42), 13484-13489 (2015); doi: 10.1073/pnas.1506378112.
(a) Groupe Métallogénie Expérimentale, Géosciences Environnement Toulouse (GET), Observatoire Midi-Pyrénées, Université de Toulouse, CNRS, IRD, Toulouse (France)
(b) CNRS, Université Grenoble Alpes, Institut NEEL, Grenoble (France)
(c) Observatoire des Sciences de l’Univers de Grenoble, CNRS, Université Grenoble Alpes, Saint Martin d’Hères (France)
(d) École Normale Supérieure, PSL Research University, Département de Chimie, Sorbonne Universités, UPMC, Université Paris 06, CNRS, UMR 8640 Pasteur, Paris (France)
(e) Sorbonne Universités, UPMC, Université Paris 06 and CNRS, UMR 7590, Institut de Minéralogie, de Physique des Matériaux et de Cosmochimie (IMPMC), Paris (France)
(f) Université de Lorraine, CNRS, CREGU, GeoRessources, Vandoeuvre lès Nancy (France)



[1] G.S. Pokrovski and L.S. Dubrovinsky, Science 331, 1052-1054 (2011).
[2] G.S. Pokrovski and J. Dubessy, Earth Planet. Sci. Lett. 411, 298-309 (2015).
[3] D. Testemale et al., Rev. Sci. Instrum. 76, 043905-043909 (2005).