Context

Antimony is a relatively rare metal with an average concentration up to 0.4 ppm in the upper continental crust (Rudnick & Gao 2003) and up to 0.8 ppm in the oceanic crust (Jochum & Hoffmann 1997). The enrichment factor leading to a classical deposit (i.e. 1 to 3% of Sb) is about 150,000 times its crustal content, which involves very efficient extraction processes from source rocks and/or source magmas to a sink. The geochemical behaviour of Sb is comparable to the As and Hg, which may explain their similar distribution and their relatively close location to each other. Sb is relatively common in natural geothermal systems, where it is usually accompanied by As, Au and Hg (Ewers & Keays 1977; White 1981), and it can be transported within a gaseous form; for instance the Etna volcano may dissipate nearly 9 tons of Sb in only one year (Henley & Berger 2013).

The major factors, which control antimony solubility, are temperature, oxygen fugacity (fO₂), pH and the total sulfur activity (William-Jones & Normand 1997). Experimental results from Ewers (1977) show that Sb is very efficiently leached by relatively high temperature fluid (< 300°C). A simple interaction between a fluid around 300°C and a rock containing sulphides is sufficient to extract up to 60 % of Sb available from a source rock (Ewers 1977), depending on the way the Sb is held and the rate of chemical reaction. Alternatively, Sb, W and associated metals may derive from exsolving magmatic fluids, in which the main components are represented by chlorides, carbonates, fluorides, sulfates and sulfides (Fulignati et al. 2011).

Through the world, two main kinds of Sb deposits are identified:

1. the vein-type Sb mineralisation, which are mainly formed at shallow depths (i.e. 100-280°C and 0.1 to 1.2 kbar) with crustal fluids and/or meteoric fluids, and frequently associated with crustal-scale strike-slip faults. Either it is classified in the orogenic gold type deposits (e.g. Groves et al. 1998, Goldfarb et al. 2001) or epithermal-like type deposits (Schwarz-Schampera 2014);
2. the "Carbonate replacements" type of deposits subdivided in Phanerozoic stratabound deposits within carbonate and detrital series and in Archean deposits located in greenstone belts and formed by successive enrichments during massive and multistage fluid flow (Jaguin et al. 2012).

Although several processes that control the formation of Sb mineralisation are known, many questions remain unanswered, such as:

• What are the processes controlling the location and distribution of deposits at an orogen-scale?
• What is the thermal engine leading to large-scale fluid flow?
• What are the source rocks of metals and where are they originated? Magmas, fluids or a combination of both?
• Why some Sb deposits consist of Sb-only or rich in W (over 1% WO3), Hg, As or Au?
• Why carbonate-hosted Sb deposits are generally the poorest in arsenic?

Our recent works (Pochon 2017) has led to major advances on Sb mineralisation from studies performed in the Armorican massif. These Sb hydrothermal deposits mainly belong to the vein-type mineralisation and share many features with the other Sb deposits from European Variscan belt such as (1) spatial relationships with regional-scale faults, (2) localisation of Sb quartz veins within anticlinal hinges, or (3) relation with late-Variscan tectonic processes (e.g. Bouchot et al. 2005 and references therein). We have demonstrated a strong spatial relationship between Sb and mafic magmatism (Pochon et al. 2015, 2016a). In addition, we dated and highlighted an underestimated large mafic magmatic event at 360 Ma and demonstrated a genetic link between this mafic event and Sb mineralisation (Pochon et al. 2016b, 2017, 2018). This is also coincident with the age (i.e. 360 Ma, Hall et al. 1997) of important thermal processes affecting the Almadén Hg deposit. This major thermal event may be the trigger of Hg mobilisation from the stratabound deposits to epithermal deposits also present in the district (Higueras et al. 1999). Also we note in latter district a relationship between the stratabound deposits and mafic magmatism (Higueras et al. 2013), as well as the presence of numerous vein-type Sb deposits. In the Armorican Massif, this mafic magmatism is coeval to a dynamic plate change from the Gondwana-Armorica continental subduction to collision. This tectonic change likely induces a partial melting of the underlying mantle promoting a strong crustal thermal gradient and likely a mobilisation of metals (Pochon 2017).

In the Iberian massif, there are numerous Sb-Hg mineralisation (Gumiel & Arribas 1987), including the world's largest mercury deposit, Almadén (Higueras et al. 2013). This mineralisation is mainly concentrated in the Central-Iberian Zone (CIZ), which is the Armorican Massif counterpart. In the CIZ, numerous intrusions of mafic rocks are also spatially associated with Sb and Hg mineralisation (Gumiel et al. 1976). A multidisciplinary study might be therefore undertaken in the CIZ to confirm a possible link between a ca. 360 Ma mafic magmatic event and Sb mineralization. This would validate a new crustal-scale metallogenic model related to geodynamic and mantle melting processes. In addition, an extensive study on geochemistry and mineralogy of metal carriers must be performed in order to confirm the genetic relationship between mafic magmatism and Sb mineralisation.

Meanwhile, the toxicity of these metals (Sb, As, Hg) being more recognised, it is also essential to draw environmental risk maps at the scale of geologic domains (e.g. Ibero-Armorica arc). Like other toxic metals (e.g. As, Pb, Cd, Hg), its significant toxicity for human health as well as natural environment is of particular concern (Herath et al. 2017). In the past ten years, the existence of elevated levels of Sb in soils, sediments, surface/ground-water and biological systems have received considerable attention worldwide due to its adverse consequences on human food chain, water as well as agricultural crop productivity (Filella et al. 2009; Ahmad et al. 2014; Cai et al. 2016; Herath et al. 2017). Thus, elevated content of Sb in biological and geochemical systems originating from natural and anthropogenic sources are of particular global concern. To date, known EU countries affected by contamination of Sb are Spain and Slovakia. However, due to geological continuities through different countries, similar contamination might be expected in many other EU countries. It has been established that the first cause of Sb pollution is of geological origin since it is linked to the presence of deposit and/or bedrock containing Sb (Herath et al. 2017). Sb is mainly derived from the weathering, dissolution of Sb bearing minerals. The Sb mobilisation from its primary minerals to the environment is governed by oxidative and aqueous dissolution of sulfide minerals while its mobility and transformation is predominantly controlled by naturally occurring precipitation and adsorption processes. Thus, oxyhydroxides of iron, manganese and aluminum secondary minerals have been recognised as naturally occurring Sb sequestrating agents in the environment. Up to date, only few data is available regarding mobilisation processes of Sb from host mineral and its transformation through various environmental areas. Furthermore, global distribution of Sb contaminated aquifer sediments are still not fully understood and a substantial research gap exists in the release mechanisms of Sb from host minerals and related rocks as well as the analysis of toxic inorganic and organic Sb species in various environmental systems, including groundwater and geothermal fluids.

Then, an environmental risk assessment data layer based on mineral prospectivity approach and weighted by surface properties is of main importance for Europe.