Slags from Ruda Śląska, Poland as a large-scale laboratory for the crystallization of rare natural rocks: melilitolites and paralavas

Highlights

• Chemistry and phase composition of slags correspond to the melilitolites/paralavas.

• Crystallization started in the temperature range 1250–1300 °C.

• Thermal gradient only partly influences the phase differentiation.

• Glass domination is a result of solidus dependences in the åk-gh solid solution.

• Similar crystallization process is observed in the melilitolites/paralavas.

Abstract

Zinc and lead smelting slags from Ruda Śląska are unique in their chemistry and phase composition, which resemble rare natural rocks such as paralavas and melilitolites. Moreover, considering its size, we can treat a pyrometallurgical slag dump as a geological body. In slags from Ruda Śląska the assemblage melilite ± pseudowollastonite ± wollastonite ± plagioclase was discovered in glassy slag. High-temperature experiments were performed to determine the temperature conditions and to reconstruct the crystallization of such an assemblage. Two slag samples were subjected to complete melting and crystallization with controlled thermal gradients of: 53.25 °C/h, 15.20 °C/h and 7.60 °C/h. The results showed that crystal nucleation started at temperatures of 1250-1300 °C depending on the fluctuations of chemical composition. In both samples the thermal gradient only partly influenced the phase differentiation, being rather responsible for the disappearance of the primary glass. Moreover, even slight differences in chemical composition resulted in changes in phase assemblages under the same temperature conditions (mll + gls vs pwol+pl + mll + gls). It was proven that it is due to the combination of undercooling conditions and solidus dependences in the åkermanite – gehlenite solid solution. The occurrence of such phenomena should be considered in natural rocks with similar composition. In the case of the Ruda Śląska slags it explains the dominance of glassy slag in that location. The experiments gave us an opportunity to observe and precisely analyze crystallization in real time, providing new insights into the creation of slags and their natural analogues. However, the study has also shown that possible variations of the original crystallization should always be assessed.

Introduction

Studies of pyrometallurgical slags has focused on four main aspects: (i) the mechanical and chemical properties of slags in the context of their commercial use (Fisher and Barron, 2019; Mikoda et al., 2019; Pan et al., 2019; Potysz et al., 2018); (ii) the environmental impact of slag accumulation due to the Potentially Toxic Elements (PTEs) mobility in the environment (Mendecki et al., 2020; Piatak and Seal II, 2010; Saikia et al., 2018; Tyszka et al., 2014; Warchulski et al., 2015; Warchulski et al., 2019); (iii) the recreation of the historical smelting process (Cabała et al., 2020; Ettler et al., 2009; Ettler et al., 2015; Manasse and Mellini, 2002; Scheinert et al., 2009; Warchulski et al., 2018); (iv) slags as a laboratory of melt crystallization, with unique chemical compositions and textures (Ettler et al., 2000; Puziewicz et al., 2007; Warchulski et al., 2016; Sobanska et al., 2016). The last aspect is the least studied, despite the fact that slags may be treated as analogues of uncommon, but natural, rocks.

Zn-Pb smelting slags are the result of crystallization of silicate melts with a high calcium content and a significant amount of aluminum, iron, and magnesium. As a consequence, their main components are high-temperature silicates and aluminosilicates, usually rich in calcium, iron and magnesium (e.g., olivine, pyroxene, melilite, feldspar), as well as oxides (e.g., spinel group minerals, hematite) accompanied by sulfides (e.g., sphalerite/wurtzite, galena; Puziewicz et al., 2007; Piatak and Seal II, 2010; Warchulski et al., 2015; Warchulski et al., 2018; Tyszka et al., 2018). ZnPb slags are usually fine- to medium-grained, with subordinate amount of glass (Warchulski et al., 2015). Rarely slags are coarse grained, with unusual unique miarolitic texture and mineral sizes up to several centimetres (Warchulski et al., 2016).

Slags from Ruda Śląska are composed of melilite + glass ± pseudowollastonite ± wollastonite ± plagioclase (Warchulski et al., 2019). A unique feature of the Ruda Śląska slags is the predominance of glass, with dispersed crystalline phases, including pseudowollastonite. A similar assemblage has not yet been described in such material. The chemical and phase compositions of the slags (Warchulski et al., 2019) classify them as the anthropogenic analogues of natural magmatic rocks – melilitolites (e.g. Isakova et al., 2017; Panina et al., 2013; Stoppa et al., 2003; Stoppa and Sharygin, 2009), pyrometamorphic rocks – paralavas (e.g., Grapes, 2006; Melluso et al., 2003; Seryotkin et al., 2012; Sokol et al., 2008; Vapnik et al., 2007) or rare meteorites (Caillet Komorowski et al., 2010). Unlike their natural counterparts, with the origin and crystallization condition often being subjects of dispute (e.g. Capitanio, 2005; Melluso et al., 2003; Melluso et al., 2005), in metallurgical slags these processes are relatively well known, due to the well described details of the metallurgical process, including historical accounts (e.g. Warchulski et al., 2018 and references therein). That fact enables a reliable reconstruction of the crystallization process and comparative analysis with natural rocks, which will give insights into their formation. Unlike small-scale crystallization simulations, we can treat a pyrometallurgical slag dump as a geological body, with an original area of about 8.5 h and a volume of 1,860,000 m³ (Jończy, 2006), thus better reflecting the crystallization conditions of natural, volcanic or pyrometamorphic rocks.

Achieving this goal is performed on the basis of high-temperature heating experiments, and a multi-factor comparative analysis of the original slags and experimental slags crystallized in controlled, yet varied conditions and finally natural rocks. Heating experiments have already been proven to be useful tools in analysing the crystallization of similar anthropogenic (e.g. Ciesielczuk et al., 2015) and natural (e.g. Stoppa and Sharygin, 2009) materials. To achieve our main aim, it is also necessary to find the reason for the dominance of glassy slags and the observed sequence of phase crystallization.