Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Reconstruction of smelting conditions during 16th‐ to 18th‐century copper ore processing in the Kielce region (Old Polish Industrial District) based on slags from Miedziana Góra, Poland

Reconstruction of smelting conditions during 16th‐ to 18th‐century copper ore processing in the... INTRODUCTIONCopper was one of the first metals used by humanity, with the oldest traces of processing dating back over 10,000 years (Davis, 2001; Kundig & Weed, 2015). Initially, fragments of native copper were transformed by forging (Davis, 2001; US Congress, Office of Technology Assessment, 1988) and later by casting and smelting (Davis, 2001; Roberts et al., 2009). An important step in metallurgy was combining copper with other metals. The production of copper–tin alloys led to the beginning of the Bronze Age. Since then, this metal has become one of the most desirable materials (Davis, 2001; Kundig & Weed, 2015; Roberts et al., 2009; Tylecote, 1992). Today, copper is also an important material. It ranks third, after iron and aluminum, in production volume. Copper owes its position to very good electrical and thermal properties, corrosion resistance, ease of processing, and aesthetic values (Davenport et al., 2002; Davis, 2001).Due to the thousand‐year history of copper smelting, in many places around the world we can observe the remains of various stages of the production processes (Bassiakos & Catapotis, 2006; Bourgarit et al., 2003; Burger et al., 2010; Derkowska et al., 2021; Kapper et al., 2017; Maldonado & Rehren, 2009; Manasse et al., 2001; Rozendaal & Horn, 2013). The most common metallurgical residues are slags—materials formed during pyrometallurgical processing of ores. This process consists of heating the previously prepared material in a metallurgical furnace. After melting, the desired metal is separated from other charge components. Thus, the end product of the process is a pure metal and a solidified silicate melt (slag) containing the remaining components (e.g., Warchulski, Gawęda, et al., 2020). In the past, slags were often stored in landfills and partially used in subsequent melts as fluxes (Agricola, 1556; US Congress, Office of Technology Assessment, 1988; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022).Analyses of old metallurgical residues are carried out to understand the issues related to historical metallurgy. Currently, slag analyses focus on several aspects. The main purpose is to understand the processes during crystallization (Kupczak et al., 2020; Warchulski, Gawęda, et al., 2020) and to reconstruct the smelting processes (e.g., Derkowska et al., 2021; Ströbele et al., 2010; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022). As slags were often stored in the vicinity of smelters, they seem to be the best source of information on the character of the metallurgical process carried out. The chemical and phase composition of slags are used to determine the smelting temperature, the viscosity of the melt, the type of oxidation–reduction conditions, and the type of added fluxes (Derkowska et al., 2021; Ettler et al., 2009; Kierczak & Pietranik, 2011; Maldonado & Rehren, 2009; Warchulski, 2016; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022). Another reason for their study is the environmental aspect. In historical times, due to the lack of tools enabling ecological protection (e.g., flue gas cleaning systems), the application of technologies using harmful elements (e.g., Hg during amalgamation; Nriagu, 1994), and the storage of waste containing potentially toxic elements (PTEs; Cabała et al., 2020; Kierczak et al., 2013; Warchulski et al., 2019) metallurgy led to the deterioration of the natural environment. Analyses of slags (Kierczak et al., 2013; Potysz et al., 2018), sediments (Aleksander‐Kwaterczak & Helios‐Rybicka, 2009; Bränvall et al., 2001; Kierczak et al., 2013), and soils (Baron et al., 2006; Ettler, 2016; Kierczak et al., 2013; Sutkowska et al., 2013; Zhou et al., 2020) provide information on the impact of historical smelting activities on the present contamination of the natural environment. These analyses allow determination of how PTEs migrate to the natural environment.The Holy Cross Mountains area was one of the most important centers of Polish copper metallurgy from the 14th to the 20th century (Paulewicz, 1992), with numerous mines and smelters (Kowalczewski & Szczecińska, 1977), but traces of most of them are becoming blurred. One of the most important copper deposits in the Holy Cross Mountains was Miedziana Góra (Old Polish Industrial District; Figure 1). This area has been gradually developed, and there is a risk of complete destruction of the traces of historical mining and metallurgy (Król & Urban, 2003). So far, many studies have been prepared on mining works carried out in Miedziana Góra (Kowalczewski, 1972; Król & Urban, 2003, 2007; Miczulski, 1972; Wojciechowski, 2002). Published data mainly concern the mined ores and the production volume (Król & Urban, 2003; Wojciechowski, 2002), and describe macroscopic features and chemical analyses of singular slag samples (Kowalczewski & Szczecińska, 1977). Some studies describe in detail the impact of mining and metallurgical works in the area of the Holy Cross Mountains on the local community (Paulewicz, 1992). Still, there is no detailed information on the phase composition of the slags and the smelting process itself. The primary aim of this research is to fill this gap. To achieve this goal, we have to answer the following questions: (i) What is the phase and chemical composition of slags from Miedziana Góra? (ii) What was the smelting temperature? (iii) What was the viscosity of the metallurgical melt during smelting? (iv) What kind of oxidation–reduction conditions existed during the process? (v) What kind of fluxes were used during smelting? For this purpose, slag samples were collected from the Kapliczna Hill, where metallurgical works were carried out in the 16th to 18th century (Figure 1). The conducted research is also crucial for future research to determine the impact of historical smelting on the natural environment. In order to assess the impact of metallurgy on the environment, the waste must first be analyzed in terms of mineralogy, petrography, and the conditions for its formation.1FIGURELocalization (a) and geological (b) geoportal.gov.pl), and a cross‐section through the Miedziana Góra ore (c) (modified from Rubinowski, 1970)LOCATION AND HISTORICAL BACKGROUNDThe research area is located in central Poland within the Holy Cross Mountains (Old Polish Industrial District; Figure 1). Miedziana Góra is located in the Świętokrzyskie voivodship, Kielce district. During fieldwork, slag samples were collected from the place where, in historical times, copper ore was mined and processed (Kowalczewski, 1972; Kowalczewski & Szczecińska, 1977; Figure 1).Historical data show that the beginnings of mining work in Miedziana Góra date back to 1590–1592. At that time, copper ore was most likely discovered during the exploitation of iron ores (Król & Urban, 2007; Miczulski, 1972). Polymetallic ore from Miedziana Góra was used for Cu and PbO production. Exploitation was carried out by vertical shafts, which were then connected with galleries (Król & Urban, 2003). The metallurgical works in Miedziana Góra were carried out intermittently up to the 18th century in a small smelter on the Kapliczna Hill (Wojciechowski, 2002). The copper obtained from these deposits was used to produce raw fittings, sheets, vessels, and alloys with lead and zinc. In addition, at the end of the 18th century, the metals produced were used to make coins in the Polish royal mint and to create weapons (Król & Urban, 2003; Molenda, 1989). One of the most famous products made of copper from Miedziana Góra was the copper sheets used to cover the roof of the Wawel Royal Castle after a fire in the 16th century (Cracow; Król & Urban, 2003). From the end of the 17th century, problems during operation (including the lack of effective drainage of the mines) caused a crisis in the mining and metallurgy in Miedziana Góra. Despite the fact that attempts were made to improve the situation, mainly through subsidies and building drainage adits, the ore was exploited only periodically (Król & Urban, 2003). Along with the region's development, large smelters processing ores from the entire area replaced the small ones next to the mines (Kowalczewski, 1972). At the end of the 18th century, copper mines were reinvested, but metallurgical works were carried out in other localities (Wojciechowski, 2002). Due to the lack of historical data (most of the documents were destroyed during World War II), the production volume in Miedziana Góra is estimated only based on geological data. According to the existing studies, the production volume between the 16th and early 18th centuries is estimated at 4750–6250 tons (Kowalczewski, 1972), or at least 6200 tons (Molenda, 1989). In the later period, only data are available for the years 1785–1788, when 26 tons of copper and 26 tons of PbO were produced (Król & Urban, 2003). Due to the high demand for copper in Poland at that time, the production did not cover the local needs, and copper was imported from abroad during the period of metallurgical activity in the Old Polish Industrial District (Molenda, 1989). Despite production mainly for the Polish market, specialists from other metallurgical centers from Poland (e.g., Olkusz, Tarnowskie Góry) and Europe (e.g., Germany, Italy, Slovakia) were involved in conducting mining and metallurgical work within the Old Polish Industrial District (Kowalczewski, 1972). Craftsmen associated with mining in the Holy Cross Mountains had numerous privileges. However, the privileges were associated with severe penalties for infractions, resulting in separating of a specific social group (Paulewicz, 1992).GeologyThe polymetallic ores in the Kielce region occur in many locations. However, apart from Miedziana Góra and Miedzianka, they form only small concentrations (Kowalczewski, 1972). The Miedziana Góra ore is located in the zone of the Miedziana Góra fold, overlapping the northern part of the Kostomłoty syncline. Lower Devonian sediments (conglomerates, quartzites, silts, schist, and ore‐bearing clays) are overlaid from the northeast by Upper Devonian formations (limestones and marbles; Wojciechowski, 2002; Figure 1). Based on complete chemical analyses that were carried out during the documentation of the ore in the 20th century, it consists mainly of SiO2 (av. 40.93 wt%), Al2O3 (av. 20.84 wt%), and Fe (over 10 wt% of Fe + FeS + FeO + Fe2O3), with Cu content up to 1.72 wt% (as Cu + CuS + CuO), Pb up to 0.07 wt% and Zn up to 0.85 wt%. However, in historical times ores with Cu content of up to 12 wt% were processed (Rubinowski, 1970). The Miedziana Góra ore can be divided into an oxidized zone and a sulfide zone (primary zone). In the primary zone, the main ore minerals are pyrite (FeS2), chalcopyrite (CuFeS2), chalcocite (Cu2S), bornite (Cu5FeS4), tetrahedrite ([Cu,Fe]12Sb4S13), sphalerite (ZnS), and galena (PbS). Sulfide minerals occur as single grains or clusters and radial clusters in clay rocks. The oxidation zone contains malachite (Cu2(CO3)(OH)2), azurite (Cu3(CO3)2(OH)2), chrysocolla ((Cu, Al)2H2Si2O5(OH)4·n(H2O)), iron oxides and hydroxides, siderite (FeCO3), smithsonite (ZnCO3), and manganese oxides (Piekarski, 1961).MATERIALS AND METHODSSamplingDuring fieldwork, in the spring of 2021 slag samples were collected from the area of Kapliczna Hill (in Miedziana Góra; Figure 1), where the smelter was most probably located (Figure 1). During the works, 62 samples of slags were collected, which were then divided into six macroscopically different types (MG1–MG6), described below.Chemical and phase analyzesThe first stage of the analysis involved macroscopic and microscopic observations to assess the slag's texture. For this purpose, an Olympus BX‐51 polarizing microscope was used (Institute of Earth Sciences, University of Silesia). The samples were then analyzed by scanning electron microscopy with energy‐dispersive spectrometry (SEM‐EDS). Analysis with SEM allowed us to more accurately determine the morphology of phases present in slags and their approximate chemical composition. A Phenom XL microscope (Faculty of Natural Sciences, University of Silesia) was used during this analysis. An electron micro‐probe (Cameca SX100, Inter‐Institutional Laboratory of Microanalysis of Minerals and Synthetic Materials, University of Warsaw) was used to determine the exact chemical composition of the phases present in the samples. EPM analyses were performed at 15 keV accelerating voltage, a 10–20.1 nA beam current, and a beam diameter of up to 5 μm. The following standards were used during the measurements: Na—albite (NaAlSi3O8); Mg, Si, Ca—diopside ([Ca,Mg,Fe]2SiO3); Al, K—orthoclase (KAlSi3O8); Ba—barite (BaSO4); Ti—rutile (TiO2); Cr—Cr2O3; Pb—crocoite (PbCrO4); Fe—Fe2O3, chalcopyrite (CuFeS2); Mn—rhodonite (CaMn3Mn[Si5O15]); Zn—sphalerite (ZnS); Cl—sodalite (Na8[Al6Si6O24]Cl2); P—YPO4; As—GaAs; Sr—celestine (SrSO4); Co—CoO; Ni—NiO; V—V2O5; Cu—cuprite (Cu2O); As—GaAs; S—chalcopyrite (CuFeS2), barite (BaSO4); Sb—stibnite (Sb2S3).Samples representing all six types of slags were also analyzed for phase composition and chemistry. The phase composition of the samples was determined using the PANalytical X'PERT PRO‐PW 3040/60 and PANalytical X'Pert PW 3710 X‐ray diffractometers equipped with CoKα1 (PW 3040/60) and CuKα1 (PW 3710) source radiation, Fe‐filter (for Co) and Ni‐filter (for Cu) to reduce the Kβ radiation. During analyses, the X'celerator detectors were used (Faculty of Natural Sciences, University of Silesia). The analyses were performed in the 5–90° 2θ angular range at 40 kV voltage (40 mA). For quantitative phase composition, a Rietveld analysis was performed. The X'PERT High Score Plus software and the PDF4+ database were used.To obtain the chemical composition, including major, minor, and trace elements, the combination of X‐ray fluorescence (XRF) spectrometry, inductively coupled plasma emission spectrometer (ICP‐ES), and inductively coupled plasma mass spectrometry (ICP‐MS) was applied. During the ICP analyses, multi‐acid digestion was used. XRF and ICP‐MS/ES analyses were performed by Bureau Veritas Mineral Laboratories (Canada). Loss of ignition was determined at 1000°C.High‐temperature experimentsExperimental melting in a laboratory furnace was used to determine the solidus and liquidus temperature of slags. Two types of slag were selected for the experiment (MG3 and MG6), representing the two most diverse slags. The MG3 sample is the most common type of slag among glassy samples (23 out of 58 samples), while the MG6 sample is the only hypocrystalline type of slag. The samples were crushed, cut into fragments with a volume of about 1 cm3, and heated in crucibles with gradually rising temperature until the entire material was melted. Each time after heating, the samples were subjected to rapid cooling. Experiments were performed at the Institute of Earth Sciences of the University of Silesia in a chamber furnace PLF 160/5 with a PC 442/18 controller, SiC heaters, and Thermocouple S with a maximum working temperature of 1550°C.SoftwareTo determine the liquidus temperatures, the MELTS‐Rhyolite v.1.0.2 software package (Ghiorso & Gualda, 2015; Gualda et al., 2012) was used. The liquidus temperature was calculated based on the bulk chemical composition (MG6) and the averaged chemical composition of the glass that forms the slags (MG1–MG4). Due to the presence of zones with increased accumulation of metallic and quartz/cristobalite phases, the MG5 sample was not used for temperature determination. For calculations by MELTS‐Rhyolite, chemical composition was normalized to 100 wt%, a pressure of 1 bar, and a Q‐Fa‐Mt oxygen buffer was used. For graphical data processing, the AUTODESK AutoCAD 2021 and CorelDRAW2021 version 22.1.1.523 (educational licenses) were used.RESULTSPetrographic characteristics of slagsSlags from Miedziana Góra were divided into glassy (MG1–MG5) and hypocrystalline (MG6) samples. X‐ray diffractometry (XRD) analyses confirmed the presence of an amorphous (glassy) phase in MG1–MG5 samples (Figure 3). Apart from the glass, MG1–MG5 slags contain SiO2 polymorphs (quartz and cristobalite; 15.7–90.4 vol%), clinopyroxenes (6.4–79.5 vol%), copper (Cu; 0.3–4.7 vol%), and litharge (PbO; up to 0.4 vol%). However, due to the lack of information on the percentage of glass, the presented results only include the relative proportions of the crystalline phases (Figure 3).The MG1 slag is black and contains quartz/cristobalite grains (up to 2–3 mm, occasionally up to 20 mm in size; 83.9 vol%; Figures 2a and 3). Imprints or fragments of charcoal appear on the surface. SEM (Figure 4a) observations confirmed that the MG1 sample comprises glass with numerous small (a few micrometers across; rarely up to several dozen micrometers; Figure 4a) metallic phases. They are composed of metallic Cu (1.4 vol%; Figure 3) and litharge (0.1 vol%; Figures 3 and 4a).2FIGUREMacroscopic images of slags occurring in Miedziana Góra3FIGUREResults of X‐ray diffraction analyses of slag samples from Miedziana Góra. Abbreviations: An, anorthite; Cpx, clinopyroxene; Cu, copper; Crs, cristobalite; Lth, litharge; Pb, lead; Qz, quartz; Wo, wollastonite group (bustamite + ferrobustamite)4FIGUREBackscattered electron images of phases occurring in slags from Miedziana Góra: (a) MG1 sample; (b) MG2 sample; (c) MG3 sample; (d) MG4 sample; (e) MG5 sample with decreasing brightness; (f) MG6 sample with magnification. Abbreviations: azu, azurite; cpx, clinopyroxene; Cu, metallic copper; gls, Glass; lit, litharge; wo, wollastonite group phasesMG2 slags have a light blue/greenish color with visible dark‐blue glassy layers with quartz/cristobalite grains up to 5 mm (83.2 vol%; Figures 2b and 3). This sample also contains metallic phases: Cu (0.3 vol%; Figure 3) and litharge (0.4 vol%; Figure 3) with a diameter up to 30 μm (Figure 4b) and skeletal clinopyroxene crystals (16.0 vol.%).The MG3 sample is dark blue in color with light‐blue and brown layers (Figure 2c). In MG3, quartz/cristobalite (76.4 vol%; Figure 3) grains are up to 1–2 mm, occasionally up to 5 mm in diameter (Figure 2c). In MG3 samples, clinopyroxene crystals were found in the form of dendritic crystals with lengths up to 100 μm (22.5 vol%; Figures 3 and 4c). In MG3 metallic phases were present as metallic copper (1.1 vol%). Cu sulfides (matte) and arsenides (speiss) are also present.The MG4 slag is brown in color, with visible interlacing of black slags. Quartz/cristobalite grains are present here in smaller quantities (15.7 vol%; Figure 3) than in the other glassy slag types (Figure 3) and are up to 1 mm in diameter (Figure 2). Clinopyroxenes constitute 79.5 vol% of crystalline phases present in the MG4 sample (Figure 3). In this sample, along with metallic Cu (4.7 vol%; Figure 3), metallic Pb, matte, and speiss occur (Figure 4d).Unlike the MG1–MG4 samples, the MG5 slags have a porous texture, with pores up to 20 mm in diameter. The MG5 slags are characterized by a variety of colors, but blue and greenish‐blue predominate. The MG5 sample consists of SiO2 polymorphs (90.4 vol%), clinopyroxene (6.4 vol%), metallic Cu (3.1 vol%), and litharge (0.1 vol%; Figure 3). The size of quartz/cristobalite grains reaches up to 20 mm (Figure 2e). Metallic aggregates composed of metallic Cu and litharge are up to 50 μm in diameter (Figure 4e). Occasionally present, azurite (Cu3(CO3)2(OH)2) (Figure 4e) is a secondary phase.The MG6 slags, unlike the others, are characterized by a hypocrystalline texture and a gray color. That type of slag is porous, with pore size up to 20–30 mm in diameter (Figure 2f). During SEM observations, skeletal crystals up to 400 μm long were commonly observed. These crystals are composed of phases with a structure corresponding to the wollastonite group minerals (54.6 vol% of bustamite and 5.6 vol% of ferrobustamite; Figure 3). Metallic Cu, along with Cu sulfides, are also noted up to a few micrometers in diameter (Figure 4f). The presence of anorthite (34.2 vol%), quartz (3.8 vol%), pyroxenes (1.2 vol%), and metallic phases (0.6 vol% Cu and 0.1 vol% Pb;) was revealed by XRD (Figure 3).Bulk chemical compositionThe glassy samples (MG1–MG5), despite the macroscopic differences, are characterized by similar chemical compositions (Table 1). Their main chemical components are: SiO2 (54.41–57.14 wt%; Table 1), CaO (11.25–15.36 wt%), FeO (8.42–11.03 wt%), and Al2O3 (7.77–9.14 wt%). Additionally, the presence of K2O (2.32–2.72 wt%), MgO (1.18–1.74 wt%), and MnO (0.99–1.56 wt%) was found (Table 1). Large differences among glassy slags occur in the content of Pb and Cu. The MG1, MG4, and MG5 slags contain more Pb (4.32–5.91 wt%) than the MG2 and MG3 slags (2.25 and 2.53 wt%, respectively), while samples MG1 and MG5 are characterized by higher Cu contents (0.97 and 1.61 wt%, respectively) than MG2 (0.40 wt%), MG3 (0.56 wt%), and MG4 (0.55 wt%; Table 1). Glassy slags also concentrate Zn (2168–5575 ppm), As (88–374 ppm), and Sb (72–298 ppm; Table 1). The hypocrystalline MG6 slag contains less SiO2 (49.81 wt%) and Pb (0.50 wt%), with higher amounts of CaO (21.64 wt%) and Cu (4.98 wt%) compared to glassy slags (MG1–MG5; Table 1).1TABLEChemical composition of slags from Miedziana GóraMG1MG2MG3MG4MG5MG6P2O5%0.580.430.440.590.610.38SiO255.2855.6157.1454.4155.6149.81TiO20.460.390.440.470.420.33Al2O39.147.798.808.947.777.54FeO8.4210.6111.0310.439.7810.75MnO0.991.561.471.161.051.51CaO14.8315.3611.2512.6911.5221.64MgO1.191.741.551.691.181.10Na2O0.170.150.140.150.150.10K2O2.322.462.642.722.451.67SO30.060.050.030.090.080.11BaO0.770.870.830.851.040.27TOT/C<0.02<0.020.09<0.020.040.10TOT/S0.02<0.020.020.040.030.05LOI−1.30−1.40−1.30−1.10−1.30−1.60Cu0.970.400.560.551.614.98Pb4.322.252.534.935.910.50Znppm378421684524506355752403As17288127235374471Sb1497286202298101Ni6538486811688Co20899136141234162Phase chemistryOxidesThe most common oxides in the analyzed slags are the SiO2 polymorphs. Quartz/cristobalite crystals (up to 99.1 wt% of SiO2; Table 2) are commonly found in MG1–MG5 slags and, to a lesser extent, in MG6 sample (Figures 3 and 4). Litharge (PbO) was observed in MG1, MG2, and MG5 slags in the form of phases resulting from chemical demixing of the metallic melt (Figure 4). It contains the addition of Cu (up to 1.1 wt% of CuO; Table 2).2TABLERepresentative EPMA (Electron Probe Micro‐analysis) data of the chemical composition of phases occurring in slagsqzwowocpxcpxcpxazulitbdssccCuCuSiO299.150.550.350.250.650.4na0.64S25.719.7bdlbdlTiO2bd0.100.110.21bdlbdlnabdlFe9.40.430.15bdlAl2O30.080.352.201.712.932.28na0.37Mn0.04bdlbdlbdlFeO0.065.06.412.813.812.9nabdlCo0.01bdlbdlbdlMnObdl1.901.971.411.481.60nabdlNibdlbdl0.20bdlMgObdl1.081.248.87.58.1nabdlCu61.977.693.698.0CaObdl39.336.123.522.222.5na0.14Zn0.12bdl0.15bdlK2Obdl0.100.42bdl0.320.28nabdlAs0.060.172.80.15PbOnabdlbdlnanana2.9098.1Sbbdlbdl0.50bdlCuOnananananana67.01.10Total97.2397.997.498.15Cr2O3bdlbdlbdl0.070.05bdlnabdlCoOnananananana0.12naTotal99.2498.3398.7498.798.8898.0670.02100.35a.p.f.u. (atom per formula unit)Atomic concentrationSi110.991.961.961.97‐0.02S41.1433.290.03‐Ti‐‐‐0.010.010.01‐‐Fe8.610.420.17‐Al‐0.010.050.080.130.1‐0.01Mn0.03‐‐‐Fe2+‐0.080.110.420.450.42‐‐Co0.01‐0‐Mn‐0.030.030.040.050.05‐‐Ni‐‐0.23‐Mg‐0.030.040.510.430.47‐‐Cu50.0766.1796.799.87Ca‐0.840.760.980.920.94‐0.01Zn0.09‐0.15‐K‐‐0.0100.020.01‐‐As0.040.122.440.13Pb‐‐‐‐‐‐0.040.89Sb‐‐0.27‐Cu‐‐‐‐‐‐2.950.03Total100100100100Cr‐‐‐‐‐‐‐‐Co‐‐‐‐‐‐‐o2‐23366632Abbreviations: azu, azurite; bdss, bornite–digenite solid solution; cc, chalcocite; cpx, clinopyroxene; Cu, copper; lit, litharge; qz, quartz; wo, wollastonite group; bdl, below detection limit; na, not analyzed.CarbonatesCarbonates, represented by azurite, were found only in sample MG5 and occur only in the vicinity of metallic phases (Figure 4e). Chemically, azurite comprises Cu (up to 67.0 wt% of CuO), with Pb substitutions reaching up to 2.9 wt% of PbO (Table 2).SilicatesTwo types of silicate were distinguished in the Miedziana Góra slags. The MG1–MG5 slags contain clinopyroxenes, with a chemical composition corresponding to diopside (CaMgSi2O6)–hedenbergite (CaFe2+Si2O6) solid solution, with numerous substitutions: Al (1.71–2.93 wt% of Al2O3), Mn (1.41–1.60 wt% of MnO), and minor concentrations of K2O, TiO2, and Cr2O3 (Table 2). In the MG6 sample, the wollastonite group are the only silicate phases found during EPM analyses, consisting of Si (50.3–50.5 wt% of SiO2), Ca (36.1–39.3 wt% of CaO), Fe (5.0–6.4 wt% of FeO), and Mn (1.90–1.97 wt% of MnO). The substitution of Al (up to 2.20 wt% of Al2O3), Mg (up to 1.24 wt% of MgO), and less than 0.5 wt% of oxide of Ti and K was observed (Table 2). The Fe2+ substitutions affect the unit cell size (Yamanaka et al., 1977). For this reason, in these phases, the unit cell is more closely related to the bustamite/ferrobustamite than to the wollastonite (Figure 3). The zoning visible in backscattered electron (BSE) images (Figure 4f) results from enrichment of the crystal rims in FeO with a simultaneous decrease in the CaO content.Sulfides and metallic phasesSulfides in the Miedziana Góra slags are represented by bornite (Cu5FeS4)–digenite (Cu9S5) solid solution (bdss) phases and by chalcocite. The bdss phases are mainly composed of Cu (up to 61.9 wt%), S (up to 25.7 wt%), and Fe (up to 9.4 wt%). In addition, they concentrate Zn (up to 0.12 wt%) and less than 0.1 wt% of Mn, Co, and As (Table 2). Chalcocite is composed of Cu (up to 77.6 wt%) and S (up to 19.7 wt%) with Fe (up to 0.43 wt%) and As (up to 0.17 wt%) substitution (Table 2). The metallic copper, apart from Cu (93.6–98.0 wt%), contains As (up to 2.8 wt%), Sb (up to 0.50 wt%), Ni (up to 0.20 wt%), Fe (up to 0.15 wt%), and Zn (up to 0.15 wt%; Table 2).GlassAll slag samples from Miedziana Góra contain glass. In MG1–MG5 samples, it is the main component. The glass in the MG1–MG4 slags is, on average, rich in SiO2 (52.1–57.1 wt%), Al2O3 (6.5–11.4 wt%), FeO (5.6–11.9 wt%), and CaO (15.3–16.3 wt%; Table 3). Additionally, PbO enrichment (up to 6.3 wt%) was observed. In the MG5 sample, zones with metallic phases and SiO2 polymorph concentrations are present (Figure 4e and Table 3). In these zones, glass contains more PbO (up to 21.2 wt%) and less CaO (av. 8.9 wt%) than in MG1–MG4 slags (Table 3). When compared to MG1–MG4 slags, a greater variation in SiO2 content was also observed in MG5 slags (up to 68.8 wt%; Table 3). In the MG6 slag, the average glass is also mostly made of SiO2 (av. 50.5 wt%), but the glass has a higher CaO (av. 21.9 wt%) and a lower PbO (av. 0.38 wt%) content. The proportion of other elements is similar to those observed in the MG1–MG5 slags (Table 3).3TABLEChemical composition of glasses occurring in Miedziana Góra slagsMG1MG2MG3MG4MG5MG6RangeAv. (n = 65)RangeAv. (n = 39)RangeAv. (n = 35)RangeAv. (n = 21)RangeAv. (n = 8)RangeAv. (n = 5)P2O50.08–0.710.570.36–0.560.450.36–0.560.470.44–0.700.540.26–0.670.560.29–0.370.34SiO254.9–72.457.152.3–55.353.751.4–54.052.849.8–54.052.150.6–68.856.850.3–50.950.5TiO20.32–0.650.530.29–0.450.360.32–0.450.370.24–0.430.340.29–0.850.470.31–0.370.33Al2O38.8–12.811.46.8–7.87.26.7–7.16.96.1–7.36.57.5–15.810.27.4–7.87.6FeO2.2–6.75.69.4–11.810.610.5–11.811.211.0–12.911.92.4–10.07.210.3–14.211.5MnO0.22–0.990.821.5–1.91.701.7–1.91.801.5–2.01.600.28–1.110.831.2–1.71.50CaO5.1–17.115.514.2–16.215.315.7–17.316.314.9–17.316.32.9–13.38.918.9–23.221.9MgO0.31–1.201.001.7–2.01.801.8–2.01.900.99–2.01.700.28–1.470.891.1–1.21.10Na2O0.00–0.190.100.00–2.230.110.00–0.200.090.00–0.170.070.12–0.310.200.02–0.090.05K2O2.4–5.42.802.0–2.62.302.0–2.42.201.8–2.42.101.6–5.92.901.3–1.81.50PbO0.00–4.02.001.7–4.33.02.4–3.63.02.6–6.34.22.1–21.28.80.23–0.610.38CuO0.00–0.340.030.00–0.860.030.00–0.260.070.00–5.80.370.20–0.390.260.15–0.430.30ZnO0.00–0.540.240.00–0.470.220.00–0.590.300.00–0.620.310.09–0.720.500.23–0.370.30BaO0.48–1.050.800.65–0.940.800.58–0.800.700.57–0.820.690.26–1.390.940.22–0.280.25Smelting temperatureExperimentDuring the smelting experiment, the first changes in the samples were observed at 1000°C. At this temperature, only the color of the slags was changed from blue to greenish‐gray in MG3 and from orange‐gray to greenish‐gray in MG6 samples, possibly due to the oxidation (Figure 5). At a temperature of 1100°C, sample MG3 was partially melted, leaving parts of the sample and quartz grains unchanged. Clear traces of partial melting also appeared on the surface of the MG6 sample (slag edges were smoothed; Figure 5). These changes indicate that the solidus temperature of both analyzed slags was exceeded. At a temperature of 1150°C, both types of slag were melted but not homogenized (Figure 5). The experiment was completed at a temperature of 1200°C. At this temperature, both samples were almost completely melted and homogenized. Only single quartz grains were visible on the glassy surface (Figure 5). Considering that quartz occurs in slags as grains unmelted during the smelting process, its occurrence at these temperatures could be ignored. Continuation of the experiment until the quartz melted and dissolved would lead to falsely elevated results.5FIGUREPhotos of samples after high‐temperature experimentsPhase diagramsThe chemical composition of the MG6 slag and the average chemical composition of the glass in the MG1–MG4 slags were plotted on the CaO–SiO2–Al2O3 and CaO–SiO2–MgO–10% Al2O3 phase diagrams (Muan & Osborn, 1965; Supplementary Material 1). In the case of glassy slags, the averaged glass composition was used because the chemical composition of the entire sample includes unmelted SiO2 grains (Figures 2 and 4). Due to the chemical composition of glass in the MG5 sample differing from the range for which the diagrams were designed, it was not included in plotting. Based on the CaO–SiO2–MgO–10% Al2O3 phase diagram, the liquidus temperature for the MG1–MG4 samples was in the range of 1400–1500°C (Supplementary Material 1; Table 4), while, for MG6, it was slightly below 1300°C (Supplementary Material 1; Table 4). Diagram CaO–SiO2–Al2O3 gave values of 1400–1500°C for MG2–MG4 slags and 1300–1400°C for MG1 and MG6 slags (Supplementary Material 1; Table 4).4TABLELiquidus and solidus temperatures of Miedziana Góra slags depending on the method usedMG1MG2MG3MG4MG5MG6MELTS‐rhyolite–liquidus temperature1118°C1150°C1155°C1150°C‐1139°CCaO–SiO2–MgO–10% Al2O3 phase diagram–liquidus temperature1400–1500°C1400–1500°C1400–1500°C1400–1500°C‐1200–1300°CCaO–SiO2–Al2O3 phase diagrams–liquidus temperature1300–1400°C1400–1500°C1400–1500°C1400–1500°C‐1300–1400°CExperiment–liquidus temperature‐‐1150–1200°C‐‐1150–1200°CExperiment–solidus temperature‐‐1100°C‐‐1100°CMELTS‐Rhyolite softwareThe modeling using MELTS‐Rhyolite software was another method applied to determine the smelting temperature (Ghiorso & Gualda, 2015; Gualda et al., 2012). During calculations, the average chemical composition of the glasses (MG1–MG4) and the bulk chemical composition of MG6 slag samples were used. The liquidus temperatures calculated with the MELTS‐Rhyolite software were in the range of 1118°C (MG1)–1155°C (MG3; Table 4).ViscosityOne of the most popular methods of melt viscosity determination in historical slags is the viscosity index (v.i.) proposed by Bachmann (1982) and modified by Ettler et al. (2009). This method is widely used in the analysis of smelting slags (e.g., Derkowska et al., 2021; Kupczak et al., 2020; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022) and takes into account the polymerizing to depolymerizing components ratio:v.i.=CaO+MgO+MnO+FeO+PbO+ZnOSiO2+Al2O3.$$ \mathrm{v}.\mathrm{i}.&amp;#x0003D;\frac{\mathrm{CaO}&amp;#x0002B;\mathrm{MgO}&amp;#x0002B;\mathrm{MnO}&amp;#x0002B;\mathrm{FeO}&amp;#x0002B;\mathrm{PbO}&amp;#x0002B;\mathrm{ZnO}}{\mathrm{Si}{\mathrm{O}}_2&amp;#x0002B;{\mathrm{Al}}_2{\mathrm{O}}_3}. $$To omit the unmelted SiO2 fragments present in the glassy slags for viscosity calculations, the bulk chemical composition of MG6 samples and the average chemical composition of the glass present in the MG1–MG5 samples were used. Considering this, the viscosity index for Miedziana Góra slags is in the range of 0.37–0.62 for glassy (MG1–MG5) and 0.62 for hypocrystalline (MG6) slags.Viscosity was also determined by using a model that includes both chemical composition and temperature (Giordano et al., 2006). In this method, the melt temperature is included in the calculation, and the resulting values are in SI units (Pa s). This model was proposed for natural rocks, and it has also been successfully applied for slags (Ettler et al., 2009; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022). The model is based on the following equation:logη=b1+b2*b3b3+SM+b4,$$ \log \eta &amp;#x0003D;b1&amp;#x0002B;\frac{b2\ast b3}{b3&amp;#x0002B;\mathrm{SM}}&amp;#x0002B;b4, $$where b1–b4 are temperature‐dependent parameters and SM is the structure modifier parameter. More detail on this model can be found in Giordano et al. (2006). In the case of Miedziana Góra slags, the viscosity was calculated for the temperature range of 1100–1200°C (Supplementary Material 2). Due to unmelted SiO2 fragments in glassy slags, calculations were performed for the averaged chemical composition of the glasses present in the MG1–MG5 samples. Due to the hypocrystalline nature, in the MG6 slags calculations were performed for bulk chemical composition. Viscosity calculated for glassy slags was mostly in the range of log η = 3.52–2.06 Pa s (for MG1–MG4 samples; Supplementary Material 2). Only porous glassy sample (MG5) has a higher viscosity (log η = 4.42–3.47 Pa s; Supplementary Material 2). The MG6 sample is characterized by lower viscosity (log η = 2.52–1.72 Pa s) than glassy slags (Supplementary Material 2).Due to the significant content of Cu and Pb in the slags from Miedziana Góra (Table 1; Table 3), additional viscosity calculations were applied according to the model proposed by Utigard and Warczok (UW model; 1995). Similar to the Giordano et al. (2006) method, the UW model considers the temperature and chemical composition of the slags. An additional advantage of this method is the fact that it has been proposed for metallurgical slags. Therefore, it takes into account Cu (as Cu2O) and Pb (as PbO). The calculations were made according to the formulalogηkgms=−0.49−5.1VR+−3660+12080VRTK,$$ \log \eta \left(\frac{\mathrm{kg}}{\mathrm{ms}}\right)&amp;#x0003D;-0.49-5.1\sqrt{\mathrm{VR}}&amp;#x0002B;\frac{-3660&amp;#x0002B;12080\sqrt{\mathrm{VR}}}{T(K)}, $$where VR (viscosity ratio) is the ratio of acidic oxides to basic oxides, calculated using the equation:VR=%SiO2+1.5%Cr2O3+1.2%ZrO2+1.8%Al2O31.2%FeO+0.5%Fe2O3+%PbO+0.8%MgO+0.7%CaO+2.3%Na2O+%K2O+0.7%Cu2O+1.6%CaF2.$$ \mathrm{VR}&amp;#x0003D;\frac{\left(\%\mathrm{Si}{\mathrm{O}}_2\right)&amp;#x0002B;1.5\left(\%\mathrm{C}{\mathrm{r}}_2{\mathrm{O}}_3\right)&amp;#x0002B;1.2\left(\%\mathrm{Zr}{\mathrm{O}}_2\right)&amp;#x0002B;1.8\left(\%\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3\right)}{1.2\left(\%\mathrm{FeO}\right)&amp;#x0002B;0.5\left(\%{\mathrm{F}\mathrm{e}}_2{\mathrm{O}}_3&amp;#x0002B;\%\mathrm{PbO}\right)&amp;#x0002B;0.8\left(\%\mathrm{MgO}\right)&amp;#x0002B;0.7\left(\%\mathrm{C}\mathrm{a}\mathrm{O}\right)&amp;#x0002B;2.3\left(\%\mathrm{N}{\mathrm{a}}_2\mathrm{O}&amp;#x0002B;\%{\mathrm{K}}_2\mathrm{O}\right)&amp;#x0002B;0.7\left(\%{\mathrm{Cu}}_2\mathrm{O}\right)&amp;#x0002B;1.6\left(\%\mathrm{C}\mathrm{a}{\mathrm{F}}_2\right)}. $$In this case, it was also decided to perform calculations for the average composition of the glass from glassy (MG1–MG5) slags and for the bulk chemical composition of the MG6 sample. During these calculations, the entire content of Cu was converted into Cu2O and Pb into PbO. Considering phase composition, the Fe occurs in slags as FeO rather than Fe2O3, and this was used for calculations. Calculations made with this method for glassy samples (MG1–MG5) gave results in the range of log η = 3.20–1.25 Pa s (Supplementary Material 2). The viscosity of the MG6 slag calculated by the UW model was lower (log η = 1.81–1.19 Pa s; Supplementary Material 2), but the difference was not as large as in the case of the Giordano et al. (2006) model.DISCUSSIONTemperature estimationsThe conducted experiment suggests that the solidus temperature of the studied slags was ~1100°C, while the liquidus temperature (in which it was possible to carry out smelting) was in the range of 1150–1200°C (Table 4). The temperatures obtained with MELTS software are slightly lower when compared to those obtained from the experimental method. The results obtained using phase diagrams (Supplementary Material 1) significantly exceed the temperatures determined by the experiment and calculations with the MELTS‐Rhyolite software (Table 4). The multicomponent nature of analyzed slags causes limited possibilities of effective application of phase diagrams. Most of the available diagrams take into account three to four components, and not necessarily those present in the analyzed slags in significant amounts. There are also diagrams designed for minerals that take into account more components, but in that case the chemical composition of the analyzed material should correspond to the chemical composition of the phases for which they were designed (e.g., Warchulski et al., 2022). For this reason, it is necessary to use those matching best the chemical composition of analyzed slags. Phase diagrams have been successfully used in the estimation of temperature in Rudawy Janowickie, Poland (Kierczak & Pietranik, 2011) and in Campiglia Marittima, Italy (Manasse et al., 2001), due to the less complex chemistry of the slags.The smelting temperature in the case of slags from Miedziana Góra is within the range of temperatures determined for slags after copper production in Spain (about 3000 years ago; ~1200°C; Sáez et al., 2003), in Crete in the Bronze Age (1150–1300°C; Bassiakos & Catapotis, 2006), and in Italy in the 13th to 14th century (1150–1300°C; Manasse & Mellini, 2002; Table 3), but slightly lower than in the smelters in the area of Old Copper Basin in Poland (14th to 20th century; 1200–1400°C; Derkowska et al., 2021; Kierczak & Pietranik, 2011; Kądziołka et al., 2020; Table 3).Viscosity estimationsThe melt viscosity during metal production is one of the most critical parameters that can describe the efficiency of the process. In the case of Miedziana Góra slags, the calculated values ​​of v.i. (0.37–0.62) are comparable with the slags from Leszczyna, Kondratów (18th to 20th century; 0.43–0.75; Table 5; Derkowska et al., 2021) and Okiep Copper District, South Africa (19th to 21st century; 0.22–0.69; Table 5). Higher v.i. were calculated for slags after copper production in the Rudawy Janowickie, Poland (14th to 16th century; 0.16–1.55, but mainly in the range of 0.55–1.55; Kierczak & Pietranik, 2011), Marsiliana, Italy (0.79–1.86; Table 5; Manasse & Mellini, 2002), and Itziparátzico, Mexico (14th to 16th century; 0.48–1.31; Table 5; Maldonado & Rehren, 2009). The high v.i. testifies to a better separation of copper‐containing phases from the slag and thus a more effective metallurgical process.5TABLESmelting temperatures and viscosities of Cu slags from selected locationsLocationTime (century)Temperature (°C)Viscosity (v.i.)Viscosity [log10 η = (pa s)]**Miedziana Góra16–181150–12000.37–0.621.72–4.42 (1200–1100°C)Kondratów and Leszczyna, Poland(Derkowska et al., 2021)18–201210–14000.43–0.750.23–2.19 (1400–1200°C)Rudawy Janowickie, Poland(Kierczak & Pietranik, 2011)14–161200–13000.16–1.551.37–5.79 (1200–1300°C)Itziparátzico, Mexico(Maldonado & Rehren, 2009)14–16nd0.48–1.31*‐Marsiliana, Italy(Manasse & Mellini, 2002)13–141150–13000.79–1.860.57–2.24 (1300–1150°C)Okiep Copper District, South Africa(Rozendaal & Horn, 2013)19–21nd0.22–0.69*‐Notes: nd, no data;*calculated based on chemical composition of slags;**calculated according to Giordano et al.'s (2006) model based on chemical composition of slags.The viscosity calculated according to Giordano et al. (2006) for the slags from Miedziana Góra (log η = 1.72–4.42 Pa s; Table 5) is within the viscosity range of the slags from the Rudawy Janowickie (log η = 1.37–5.79 Pa s; Table 5). The slags from Leszczyna and Kondratów (Derkowska et al., 2021) and slags from Marsiliana (Manasse & Mellini, 2002) are characterized by lower viscosity than those from Miedziana Góra (Table 5). It can be supposed that the metallurgy in Miedziana Góra was less efficient than the italian one and Leszczyna/Kondratów (Poland) metallurgy. Calculations made with the UW (Utigard & Warczok, 1995) model gave a similar range of results to the Giordano et al. (2006) method (Supplementary Material 2). Due to the lack of consideration of some components (mainly Pb and Cu) in samples with a high content of these metals, the Giordano et al. (2006) method significantly increases the viscosity (e.g., calculations made for glass in the MG5 sample; Supplementary Material 2).AtmosphereRedox conditions are one of the most important factors that should be considered in reconstructing metallurgical processes. For this purpose, Fe buffers are most often used, taking into account the temperature and the oxidation states of Fe in the samples (Sáez et al., 2003; Warchulski, Szczuka, & Kupczak, 2020). In the analyzed slags, Fe is present as Fe2+ rather than Fe3+. For this reason, the fugacity of oxygen can be described by quartz–fayalite–magnetite and wüstite–magnetite buffers (Zhao et al., 1999). Based on Fe buffers in the temperature range of 1100–1200°C, the fugacity of oxygen in the Miedziana Góra slag falls within the range of log fO2 = −12 (wüstite–magnetite buffer in 1100°C) to −8 atm. (quartz–fayalite–magnetite buffer at 1200°C; Zhao et al., 1999).When analyzing the slags after the production of metals from polymetallic deposits, apart from iron, the oxidation state of other metals can also be taken into account. The Ellingham diagram describes the transformations between the oxides and metallic phases most widely (Supplementary Material 3). The analyses show that in the MG6 slags Cu occurs mainly in metallic form (Figures 3 and 4f). For this reason, the fugacity of oxygen should be lower than log fO2 = −5 atm., while the lack of metallic Fe indicates the fugacity of oxygen greater than log fO2 = −14 atm. (Supplementary Material 3).In contrast, in glassy samples, aggregates made of Cu and PbO occur (Figure 4). The Ellingham diagram predicts the coexistence of metallic copper with PbO only in a narrow range of oxygen fugacity (log fO2 = −7 to −5 atm at 1100–1200°C; Supplementary Material 3). However, considering that the process was carried out in a sulfur‐rich atmosphere, the greater affinity of sulfur for Cu than for Pb widens the range of their coexistence (Supplementary Material 3).The diagrams (Supplementary Material 3) show that the production of Cu through matte also depends on sulfur fugacity (Supplementary Material 3). Unfortunately, no diagrams describing the relationship between the fugacity of oxygen and sulfur in the discussed temperature range (1100–1200°C) have been found.Despite the higher temperature, the diagrams proposed by Yazawa (1974) can be used to estimate the oxidation–reduction conditions during smelting in Miedziana Góra. As copper is mainly in the metallic form (except the MG5 sample, where it is in the carbonate form and results from slags weathering) during the smelting in Miedziana Góra, the oxygen fugacity can be estimated in the range of log fO2 = −4 to −12 atm. (Supplementary Material 3). The differentiation of samples in terms of chemical and phase composition makes it possible to estimate the oxidation‐reduction conditions at individual stages. During formation of the MG6 sample, oxygen fugacity should be in the range of approx. log fO2 = −6 to −12 atm. (Supplementary Material 3) and, during MG1–MG5 slag formation (to oxidize Pb), in the range of approx. log fO2 = −4 to −5 atm (Supplementary Material 3).Oxygen fugacity analysis based on Fe buffers in the case of Cu smelting showed a strongly reducing environment. In fact, the extent to which it can be carried out is wider and also depends on the fugacity of S2 (Supplementary Material 3).Summary of the metallurgical processIn Miedziana Góra, ore processing was carried out mainly to produce Cu and PbO (Król & Urban, 2003). The presence of metallic copper (Figures 3 and 4) indicates that the analyzed slags most likely come from the Cu production process. According to Agricola (1556), copper from sulfide ores was produced in a two‐stage process. The differences in chemistry (mainly in the content of Cu and Pb) and phase composition between glassy (MG1–MG5) and hypocrystalline (MG6) slags suggest they are the result of different smelting process stages. In the first smelting step, ore mixed with fluxes and witch charcoal was added to the shaft furnace and heated. In this production step, the primary goal was to heat the ore to form: (i) silicate slag; (ii) Cu‐rich speiss/matte (Davenport et al., 2002); (iii) metallic Cu. The workers controlled the amount of oxygen supplied during heating to oxidize as much Fe as possible (FeO was bonded within the slag). In order not to reduce the efficiency of the process, the workers did not allow Cu to be oxidized (Cu oxides would remain in the slags). As a result, part of Fe also did not oxidize and together with Cu, S, and As formed speiss/matte (Davenport et al., 2002). After the melt separation, the tapping hole was opened to remove and sort the metallic Cu, speiss/matte, and slag. The speiss/matte was further processed (in the second step) to obtain converted copper, while slags were discarded (Agricola, 1556; US Congress, Office of Technology Assessment, 1988).The lower Pb content in MG6 compared to MG1–MG5 slags may indicate that MG6 slags come from this stage of the process (Table 1). Due to the differences in density between Pb (11.35 g/cm3), Cu (8.94 g/cm3), and Fe (7.87 g/cm3; Atkins, 2001), the Pb should be effectively separated during this stage, where it must have accumulated at the lowest part of the furnace, leading to the formation of slags with low Pb content. The hypocrystalline nature of these slags is due to their lower viscosity (Supplementary Material 2). Under conditions of rapid cooling, low viscosity allows the formation of crystalline phases (Puziewicz et al., 2007). The presence of mainly metallic Cu and Fe2+ in wollastonite group phases marks the relatively low oxygen fugacity conditions during the formation of these slags.During the second step, the speiss/matte was remelted. At this stage, air was blown into the melt's surface to oxidize undesirable elements, and fluxes were added to bind them in the slags. In the case of Miedziana Góra, mainly Fe and Pb were removed during this step. Considering the high content of Pb and lower content of Cu, the glassy (MG1–MG5) slags most likely come from this stage of copper production (speiss/matte conversion). The presence of Pb in the form of PbO confirms the oxidative nature of this process. However, oxygen fugacity was not sufficient to oxidize Cu, which led to the formation of demix structures consisting of PbO and metallic Cu (Figure 4a,e). This was possible because of the higher sulfur affinity for Cu than for Pb and Fe (Supplementary Material 3). It also explains the absence of other sulfides (Pb or Fe) in the analyzed slags. The coexistence of metallic Cu with PbS could occur only in extremely non‐equilibrium conditions (Supplementary Material 3). The oxidation of Pb and Fe allowed them to react with fluxes, resulting in the formation of slags floating on the surface. The slag was then successively removed from the furnace. This resulted in a rapid temperature change, which, combined with the high viscosity of these slags, favors the formation of glass. The low diffusion rate and the lack of nucleation seeds could also be the reason for the low number of crystalline phases (Gawęda et al., 2013). The occasional presence of metallic Pb and Cu sulfides/arsenides in glassy slags could also be caused by the low oxygen diffusion. After all Fe and Pb were removed from the furnace, further oxidation cleaned the melt from sulfur (as SO2) and arsenic (as As2O3), leaving metallic copper in the furnace (Davis, 2001; US Congress, Office of Technology Assessment, 1988).The shaft furnace used for copper smelting in the 16th to 18th century was similar to those used in lead and gold metallurgy (Agricola, 1556; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022; Figure 6). The only difference was in the use of a different type of drive. In most cases, the smelters were located near rivers, the energy of which was used to drive the bellows (e.g., Warchulski, Szczuka, & Kupczak, 2020). In the case of Miedziana Góra, there is no river that could fulfill this function (Figure 1). For this reason, horse mills must have been used (Figure 6).6FIGUREReconstruction of the smelter that operated in Miedziana Góra, based on the description of the smelters operating in a similar period (Agricola, 1556)Ores and additivesThe documentation of the Miedziana Góra ore shows that ore minerals are found mainly in clays (Rubinowski, 1970). Chemically, these rocks are composed of SiO2, Al2O3, and Fe at different states of oxidation. The remaining elements are present in smaller amounts (on average, less than 5 wt% of oxide; Rubinowski, 1970). Comparing ore composition with the chemical composition of MG6 slags (resulting from speiss/matte production; Table 1) shows that the slags are characterized by a higher content of CaO and SiO2, with a lower content of Al2O3 (Table 1; Rubinowski, 1970). Higher SiO2 content (Table 1) may indicate the addition of sandstones occurring in the vicinity of the smelter (Figure 1). The primary aim of SiO2 addition was to produce two immiscible liquids and bond the FeO (Davenport et al., 2002). However, too much SiO2 has a negative effect, increasing the viscosity of the melt, which reduces the separation efficiency. For this reason, elements that break the silicate network were often added during the production of copper (Davenport et al., 2002; Potysz et al., 2015). In the case of Miedziana Góra, due to the increased content of CaO, limestones were most probably used for this purpose (Devonian carbonates; Figure 1).High SiO2 content (Table 1) and unmelted quartz/cristobalite (Figure 4) grains and the presence of other components (e.g., CaO, Al2O3; Table 1) in MG1–MG5 slags suggest that during the speiss/matte conversion process material containing SiO2, CaO, and Al2O3 (e.g., waste resulting from the enrichment of ores combined with sandstones/carbonates from the surface) was used. SiO2 polymorphs, as the most resistant to high temperature, have not been completely melted down.CONCLUSIONSThe presence of metal ores positively affected the development of the Old Polish Industrial District. It was possible thanks to the involvement of metallurgists from more experienced industrial centers, where the conditions under which smelting was carried out were similar. Based on the conducted analysis, the basic parameters that prevailed during the pyrometallurgical processing of ores from Miedziana Góra were reconstructed. Geochemical and mineralogical analyses allowed the isolation of slags from the production focused on speiss/matte and those from matte conversion. Both of these were formed under similar temperature conditions (liquidus temperature in the range of 1150–1200°C). The differences are visible in the viscosity of the metallurgical melt. The viscosity of the MG6 ranged between log η = 1.19 Pa s (UW model at 1200°C) and 2.52 Pa s (Giordano model at 1100°C). Low viscosity promoted the separation of the desired phases from the silicate melt, increasing melting efficiency. Because the second smelting step did not require low viscosity, slags after matte conversion are characterized by higher viscosity, ranging from log10 η = 1.25 Pa s (UW model at 1200°C) to 4.42 Pa s (Giordano et al., 2006 model at 1100°C). During this step, the main goal of metallurgists was to oxidize and bind undesirable elements with fluxes. The addition of SiO2 (a viscosity‐enhancing component) enabled the efficient removal of Fe from the furnace through crystallization of the Fe silicates. Differences in viscosity contributed to macroscopic differences between slags. In high‐viscosity slags, crystallization processes were hindered, which resulted in their glassy form. Based on the chemical and phase composition, the type of fluxes used during smelting was determined. The higher SiO2 and CaO content compared to the chemical composition of ores indicate the use of both carbonates and sand/sandstones during speiss/matte production, most probably from the vicinity of the smelter. In turn, during the speiss/matte conversion process, the waste resulting from the enrichment of ores combined with sandstones/carbonates from the surface was used. With the wide range of coexistence of phases found in the slags, oxidation–reduction conditions were also determined. The oxygen fugacity during speiss/matte smelting in Miedziana Góra was in the range of log fO2 = −5 (Supplementary Material 3) to −12 (wüstite–magnetite buffer at 1100°C; Zhao et al., 1999). Maintaining oxygen fugacity below the value at which copper would oxidize was necessary to limit losses due to the binding of copper oxides to slags during separation. During matte conversion, the oxygen fugacity was from log fO2 = −4 (Supplementary Material 3) to −7 (Supplementary Material 3). Oxygen fugacity in this range during matte conversion was necessary to purify the final product from Pb (through oxidation and binding to fluxes).ACKNOWLEDGEMENTSThis study was supported by the National Science Center (NCN) grant no. 2019/35/O/ST10/00313.DATA CITATIONNo dataset from external repositories was used in this study. All existing data used for this publication are cited in reference section.DATA AVAILABILITY STATEMENTThe data that support the finding of the study are given in the tables and figures within the main text. This study brought together existing data from several different sources, which are cited throughout.REFERENCESAgricola, G. (1556). De Re metallica libri XII (Ed Muzeum Karkonoskie w Jeleniej Górze, 2000), Jelenia Góra.Aleksander‐Kwaterczak, U., & Helios‐Rybicka, E. (2009). Contaminated sediments as a potential source of Zn, Pb, and cd for a river system in the historical metalliferous ore mining and smelting industry area of South Poland. Journal of Soils and Sediments, 9, 13–22. https://doi.org/10.1007/s11368‐008‐0051‐zAtkins, P. W. (2001). Chemia fizyczna. Wydaw.Bachmann, H. G. (1982). The identification of slags from archaeological sites. Institute of Archaeology Occasional Publication.Baron, S., Carignan, J., & Ploquin, A. (2006). Dispersion of heavy metals (metalloids) in soils from 800‐year‐old pollution (Mont‐Lozere, France). Environmental Science & Technology, 40(17), 5319–5326. https://doi.org/10.1021/es0606430Bassiakos, Y., & Catapotis, M. (2006). Reconstruction of the copper smelting process based on the analysis of ore and slag samples. Hesperia Supplements, 36, 329–353.Bourgarit, D., Mille, B., Prange, M., Ambert, P., & Hauptmann, A. (2003). Chalcolithic fahlore smelting at Cabrières: Reconstruction of smelting processes by archaeometallurgical finds. Archaeometallurgy in Europe, 1, 431–440.Bränvall, M. L., Bindler, R., Emteryd, O., & Renberg, I. (2001). Four thousand years of atmospheric lead pollution in northern Europe: A summary from Swedish lake sediments. Journal of Paleolimnology, 25(4), 421–435. https://doi.org/10.1023/A:1011186100081Burger, E., Bourgarit, D., Wattiaux, A., & Fialin, M. (2010). The reconstruction of the first copper‐smelting processes in Europe during the 4th and the 3rd millennium BC: Where does the oxygen come from? Applied Physics a, 100, 713–724. https://doi.org/10.1007/s00339‐010‐5651‐yCabała, J., Warchulski, R., Rozmus, D., Środek, D., & Szełęg, E. (2020). Pb‐rich slags, minerals, and pollution resulted from a medieval ag‐Pb smelting and mining operation in the Silesian‐Cracovian region (southern Poland). Minerals, 10, 28. https://doi.org/10.3390/min10010028Chirikure, S. (2015). Metals in past societies: A global perspective on indigenous African metallurgy. Springer.Davenport, W. G., King, M. J., Schlesinger, M. E., & Biswas, A. K. (2002). Extractive metallurgy of copper. Elsevier.Davis, J. R. (Ed.) (2001). Copper and copper alloys. ASM international. https://doi.org/10.31399/asm.tb.aub.t61170457Derkowska, K., Świerk, M., & Nowak, K. (2021). Reconstruction of copper smelting technology based on 18–20th‐century slag remains from the old Copper Basin, Poland. Minerals, 11(9), 926. https://doi.org/10.3390/min11090926Ettler, V. (2016). Soil contamination near non‐ferrous metal smelters: A review. Applied Geochemistry, 64, 56–74. https://doi.org/10.1016/j.apgeochem.2015.09.020Ettler, V., Červinka, R., & Johan, Z. (2009). Mineralogy of medieval slags from lead and silver smelting (Bohutín, Příbram district, Czech Republic): Towards estimation of historical smelting conditions. Archaeometry, 51, 987–1007. https://doi.org/10.1111/j.1475‐4754.2008.00455.xGawęda, A., Janeczek, J., Kierepka, M., Kądziołko‐Gaweł, M., & Krzykawski, T. (2013). Indialite‐rich paralava from a coalmine waste‐dump, Sosnowiec, Poland. Neues Jahrbuch für Mineralogie‐Abhandlungen, 190, 237–251. https://doi.org/10.1127/0077‐7757/2013/0238Geoportal. n.d. Available online: https://www.geoportal.gov.pl (accessed on 10 January 2020).Ghiorso, M. S., & Gualda, G. A. (2015). An H2O–CO2 mixed fluid saturation model compatible with rhyolite‐MELTS. Contributions to Mineralogy and Petrology, 169(6), 1, 53–30. https://doi.org/10.1007/s00410‐015‐1141‐8Giordano, D., Mangiacapra, A., Potuzak, M., Russell, J. K., Romano, C., Dingwell, D. B., & Di Muro, A. (2006). An expanded non‐Arrhenian model for silicate melt viscosity: A treatment for metaluminous, peraluminous and peralkaline liquids. Chemical Geology., 229(1–3), 42–56. https://doi.org/10.1016/j.chemgeo.2006.01.007Gualda, G. A., Ghiorso, M. S., Lemons, R. V., & Carley, T. L. (2012). Rhyolite‐MELTS: A modified calibration of MELTS optimized for silica‐rich, fluid‐bearing magmatic systems. Journal of Petrology, 53(5), 875–890. https://doi.org/10.1093/petrology/egr080Kądziołka, K., Pietranik, A., Kierczak, J., Potysz, A., & Stolarczyk, T. (2020). Towards better reconstruction of smelting temperatures: Methodological review and the case of historical K‐rich cu‐slags from the old Copper Basin, Poland. Journal of Archaeological Science, 118, 105142. https://doi.org/10.1016/j.jas.2020.105142Kapper, L., Turck, R., Goguichaishvili, A., & Morales, J. (2017). Rock – And archeomagnetic investigation of prehistoric copper slag from the Swiss Alps. Latinmag Letters, 7, 1–4.Kierczak, J., & Pietranik, A. (2011). Mineralogy and composition of historical cu slags from the Rudawy Janowickie Mountains, southwestern Poland. The Canadian Mineralogist, 49(5), 1281–1296. https://doi.org/10.3749/canmin.49.5.1281Kierczak, J., Potysz, A., Pietranik, A., Tyszka, R., Modelska, M., Néel, C., Ettler, V., & Mihaljevič, M. (2013). Environmental impact of the historical cu smelting in the Rudawy Janowickie Mountains (South‐Western Poland). Journal of Geochemical Exploration, 124, 183–194. https://doi.org/10.1016/j.gexplo.2012.09.008Kowalczewski, Z. (1972). Podstawowe problemy świętokrzyskiego górnictwa kruszcowego w świetle danych geologicznych. In Materiały do sesji naukowej odbytej z okazji jubileuszu IX wieków Kielc (pp. 57–78). Wydawnictwa Geologiczne.Kowalczewski, Z., & Szczecińska, A. (1977). Wyniki badań nad żużlami z kilku stanowisk dawnego hutnictwa metali nieżelaznych w Górach Świętokrzyskich. Rocz. Świętokrz, 5, 151–168.Król, P., & Urban, J. (2003). Kopalnie w Miedzianej Górze iw Ławęcznej oraz ochrona ich pozostałości (p. 29). Rocznik Świętokrzyski. Seria B. Nauki Przyrodnicze.Król, P., & Urban, J. (2007). Kopalnie Miedzianogórskie. Kielce.Kundig, K. J., & Weed, R. D. (2015). Copper and copper alloys. Mechanical Engineers' Handbook (3rd Edition): Materials and Mechanical Design, 1, 117–228.Kupczak, K., Warchulski, R., Dulski, M., & Środek, D. (2020). Chemical and phase reactions on the contact between refractory materials and slags, a case from the 19th century Zn‐Pb smelter in Ruda Śląska, Poland. Minerals, 10, 1006. https://doi.org/10.3390/min10111006Maldonado, B., & Rehren, T. (2009). Early copper smelting at Itziparátzico, Mexico. Journal of Archaeological Science, 36(9), 1998–2006. https://doi.org/10.1016/j.jas.2009.05.019Manasse, A., & Mellini, M. (2002). Archaeometallurgic slags from Kutná Hora. Neues Jahrbuch Für Mineralogie‐Monatshefte, 369‐384, 369–384. https://doi.org/10.1127/0028‐3649/2002/2002‐0369Manasse, A., Mellini, M., & Viti, C. (2001). The copper slags of the Capattoli Valley, Campiglia Marittima, Italy. European Journal of Mineralogy, 13(5), 949–960. https://doi.org/10.1127/0935‐1221/2001/0013/0949Miczulski, S., 1972. Początki rozwoju górnictwa i hutnictwa kruszcowego w rejonie Kielc do końca XVI w., [in:] Kowalczewski Z. (red.) Dzieje i Technika Świętokrzyskiego Górnictwa i Hutnictwa Kruszcowego. Wydawnictwa geologiczne, Warszawa, pp. 79–100.Molenda, D. (1989). Eksploatacja rud miedzi i handel miedzią w Polsce w późnym średniowieczu iw początkach nowożytności (do 1795 r.). Przegląd Historyczny, 80(4), 801‐814.Muan, A., & Osborn, E. F. (1965). Phase equilibria among oxides in steelmaking. Addison‐Wesley publishing company, INC.Nriagu, J. O. (1994). Mercury pollution from the past mining of gold and silver in the Americas. Science of the Total Environment, 149, 167–181. https://doi.org/10.1016/0048‐9697(94)90177‐5Paulewicz, M. (1992). Chęcińskie górnictwo kruszcowe (IV do poł. XVII wieku). Kieleckie Towarzystwo Naukowe, Kielce.Piekarski, K. (1961). W sprawie genezy złoża rud miedzi w Miedzianej Górze koło Kielc. Prace Geologiczne, 3, 43–58.Potysz, A., Kierczak, J., Pietranik, A., & Kądziołka, K. (2018). Mineralogical, geochemical, and leaching study of historical cu‐slags issued from processing of the Zechstein formation (old Copper Basin, southwestern Poland). Applied Geochemistry, 98, 22–35. https://doi.org/10.1016/j.apgeochem.2018.08.027Potysz, A., van Hullebusch, E. D., Kierczak, J., Grybos, M., Lens, P. N., & Guibaud, G. (2015). Copper metallurgical slags–current knowledge and fate: A review. Critical Reviews in Environmental Science and Technology, 45, 2424–2488. https://doi.org/10.1080/10643389.2015.1046769Puziewicz, J., Zainoun, K., & Bril, H. (2007). Primary phases in pyrometallurgical slags from a zinc‐smelting waste dump, Swietochłowice, upper Silesia, Poland. The Canadian Mineralogist, 45(5), 1189–1200. https://doi.org/10.2113/gscanmin.45.5.1189Roberts, B. W., Thornton, C. P., & Pigott, V. C. (2009). Development of metallurgy in Eurasia. Antiquity, 83(322), 1012–1022. https://doi.org/10.1017/S0003598X00099312Rozendaal, A., & Horn, R. (2013). Textural, mineralogical and chemical characteristics of copper reverb furnace smelter slag of the Okiep Copper District, South Africa. Minerals Engineering, 52, 184–190. https://doi.org/10.1016/j.mineng.2013.06.020Rubinowski, Z. (1970). Wyniki badań rudonośności w rejonie Miedzianej Góry i Ławeczna. Archiwum Państwego Instytutu Geologicznego.Sáez, R., Nocete, F., Nieto, J. M., Capitán, M. Á., & Rovira, S. (2003). The extractive metallurgy of copper from Cabezo Juré, Huelva, Spain: Chemical and mineralogical study of slags dated to the third millenium BC. The Canadian Mineralogist, 41(3), 627–638. https://doi.org/10.2113/gscanmin.41.3.627Ströbele, F., Wenzel, T., Kronz, A., Hildebrandt, L. H., & Markl, G. (2010). Mineralogical and geochemical characterization of high‐medieval lead–silver smelting slags from Wiesloch near Heidelberg (Germany)—An approach to process reconstruction. Archaeological and Anthropological Sciences, 2(3), 191–215. https://doi.org/10.1007/s12520‐010‐0039‐7Sutkowska, K., Czech, T., Teper, L., & Krzykawski, T. (2013). Heavy metals soil contamination induced by historical zinc smelting in Jaworzno. Ecological Chemistry and Engineering a, 20(12), 1441–1450. https://doi.org/10.2428/ecea.2013.20(12)130Tylecote, R. F. (1992). A history of metallurgy (2nd ed.). The Institute of Materials.US Congress, Office of Technology Assessment. (1988). Copper production technology. Government Printing Office.Utigard, T. A., & Warczok, A. (1995). Density and viscosity of copperhickel sulphide smelting and converting slags. In W. J. Chen, C. Dim, A. Luraschi, & P. J. Mackey (Eds.), Copper 95‐Cobre 95 proceedings of the third international conference, Vol. l V Pyrometallurgy of copper (pp. 423–437). The Metallurgical Society of CIM.Warchulski, R. (2016). Zn‐Pb slag crystallization: Evaluating temperature conditions on the basis of geothermometry. European Journal of Mineralogy, 28(2), 375–384. https://doi.org/10.1127/ejm/2015/0027‐2496Warchulski, R., Gawęda, A., Kupczak, K., Banasik, K., & Krzykawski, T. (2020). Slags from Ruda Śląska, Poland as a large‐scale laboratory for the crystallization of rare natural rocks: Melilitolites and paralavas. Lithos, 372–373, 105666. https://doi.org/10.1016/j.lithos.2020.105666Warchulski, R., Kupczak, K., Gawęda, A., & Sitko, R. (2022). Complete reconstruction of the process and conditions during gold smelting in the 15‐17th century in Złoty Stok based on metallurgical slags. Archaeometry, 64, 916–934. https://doi.org/10.1111/arcm.12752Warchulski, R., Mendecki, M., Gawęda, A., Sołtysiak, M., & Gadowski, M. (2019). Rainwater‐induced migration of potentially toxic elements from a Zn–Pb slag dump in Ruda Śląska in light of mineralogical, geochemical and geophysical investigations. Applied Geochemistry, 109, 104396. https://doi.org/10.1016/j.apgeochem.2019.104396Warchulski, R., Szczuka, M., & Kupczak, K. (2020). Reconstruction of 16th—17th century Lead smelting processes on the basis of slag properties: A case study from Sławków, Poland. Minerals, 10(11), 1039. https://doi.org/10.3390/min10111039Wojciechowski, A. (2002). Hałdy dawnego górnictwa świętokrzyskiego jako źródło metali kolorowych i szlachetnych. Przegląd Geologiczny, 50(3), 240–244.Yamanaka, T., Sadanaga, R., & Takeuchi, Y. (1977). Structural variation in the ferrobustamite solid solution. American Mineralogist, 62(11–12), 1216–1224.Yazawa, A. (1974). Thermodynamic considerations of copper smelting. Canadian Metallurgical Quarterly, 13(3), 443–453. https://doi.org/10.1179/cmq.1974.13.3.443Zhao, D., Essene, E. J., & Zhang, Y. (1999). An oxygen barometer for rutile–ilmenite assemblages: Oxidation state of metasomatic agents in the mantle. Earth and Planetary Science Letters, 166(3–4), 127–137. https://doi.org/10.1016/S0012‐821X(98)00281‐7Zhou, Y., Wang, L., Xiao, T., Chen, Y., Beiyuan, J., She, J., Zhou, Y., Yin, M., Liu, J., Liu, Y., Wang, Y., & Wang, J. (2020). Legacy of multiple heavy metal (loid)s contamination and ecological risks in farmland soils from a historical artisanal zinc smelting area. Science of the Total Environment, 720, 137541. https://doi.org/10.1016/j.scitotenv.2020.137541 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Archaeometry Wiley

Reconstruction of smelting conditions during 16th‐ to 18th‐century copper ore processing in the Kielce region (Old Polish Industrial District) based on slags from Miedziana Góra, Poland

Download PDF
23 pages

Loading next page...
 
/lp/wiley/reconstruction-of-smelting-conditions-during-16th-to-18th-century-D07oWg0Oy5

References (57)

Publisher
Wiley
Copyright
© 2023 University of Oxford
ISSN
0003-813X
eISSN
1475-4754
DOI
10.1111/arcm.12837
Publisher site
See Article on Publisher Site

Abstract

INTRODUCTIONCopper was one of the first metals used by humanity, with the oldest traces of processing dating back over 10,000 years (Davis, 2001; Kundig & Weed, 2015). Initially, fragments of native copper were transformed by forging (Davis, 2001; US Congress, Office of Technology Assessment, 1988) and later by casting and smelting (Davis, 2001; Roberts et al., 2009). An important step in metallurgy was combining copper with other metals. The production of copper–tin alloys led to the beginning of the Bronze Age. Since then, this metal has become one of the most desirable materials (Davis, 2001; Kundig & Weed, 2015; Roberts et al., 2009; Tylecote, 1992). Today, copper is also an important material. It ranks third, after iron and aluminum, in production volume. Copper owes its position to very good electrical and thermal properties, corrosion resistance, ease of processing, and aesthetic values (Davenport et al., 2002; Davis, 2001).Due to the thousand‐year history of copper smelting, in many places around the world we can observe the remains of various stages of the production processes (Bassiakos & Catapotis, 2006; Bourgarit et al., 2003; Burger et al., 2010; Derkowska et al., 2021; Kapper et al., 2017; Maldonado & Rehren, 2009; Manasse et al., 2001; Rozendaal & Horn, 2013). The most common metallurgical residues are slags—materials formed during pyrometallurgical processing of ores. This process consists of heating the previously prepared material in a metallurgical furnace. After melting, the desired metal is separated from other charge components. Thus, the end product of the process is a pure metal and a solidified silicate melt (slag) containing the remaining components (e.g., Warchulski, Gawęda, et al., 2020). In the past, slags were often stored in landfills and partially used in subsequent melts as fluxes (Agricola, 1556; US Congress, Office of Technology Assessment, 1988; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022).Analyses of old metallurgical residues are carried out to understand the issues related to historical metallurgy. Currently, slag analyses focus on several aspects. The main purpose is to understand the processes during crystallization (Kupczak et al., 2020; Warchulski, Gawęda, et al., 2020) and to reconstruct the smelting processes (e.g., Derkowska et al., 2021; Ströbele et al., 2010; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022). As slags were often stored in the vicinity of smelters, they seem to be the best source of information on the character of the metallurgical process carried out. The chemical and phase composition of slags are used to determine the smelting temperature, the viscosity of the melt, the type of oxidation–reduction conditions, and the type of added fluxes (Derkowska et al., 2021; Ettler et al., 2009; Kierczak & Pietranik, 2011; Maldonado & Rehren, 2009; Warchulski, 2016; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022). Another reason for their study is the environmental aspect. In historical times, due to the lack of tools enabling ecological protection (e.g., flue gas cleaning systems), the application of technologies using harmful elements (e.g., Hg during amalgamation; Nriagu, 1994), and the storage of waste containing potentially toxic elements (PTEs; Cabała et al., 2020; Kierczak et al., 2013; Warchulski et al., 2019) metallurgy led to the deterioration of the natural environment. Analyses of slags (Kierczak et al., 2013; Potysz et al., 2018), sediments (Aleksander‐Kwaterczak & Helios‐Rybicka, 2009; Bränvall et al., 2001; Kierczak et al., 2013), and soils (Baron et al., 2006; Ettler, 2016; Kierczak et al., 2013; Sutkowska et al., 2013; Zhou et al., 2020) provide information on the impact of historical smelting activities on the present contamination of the natural environment. These analyses allow determination of how PTEs migrate to the natural environment.The Holy Cross Mountains area was one of the most important centers of Polish copper metallurgy from the 14th to the 20th century (Paulewicz, 1992), with numerous mines and smelters (Kowalczewski & Szczecińska, 1977), but traces of most of them are becoming blurred. One of the most important copper deposits in the Holy Cross Mountains was Miedziana Góra (Old Polish Industrial District; Figure 1). This area has been gradually developed, and there is a risk of complete destruction of the traces of historical mining and metallurgy (Król & Urban, 2003). So far, many studies have been prepared on mining works carried out in Miedziana Góra (Kowalczewski, 1972; Król & Urban, 2003, 2007; Miczulski, 1972; Wojciechowski, 2002). Published data mainly concern the mined ores and the production volume (Król & Urban, 2003; Wojciechowski, 2002), and describe macroscopic features and chemical analyses of singular slag samples (Kowalczewski & Szczecińska, 1977). Some studies describe in detail the impact of mining and metallurgical works in the area of the Holy Cross Mountains on the local community (Paulewicz, 1992). Still, there is no detailed information on the phase composition of the slags and the smelting process itself. The primary aim of this research is to fill this gap. To achieve this goal, we have to answer the following questions: (i) What is the phase and chemical composition of slags from Miedziana Góra? (ii) What was the smelting temperature? (iii) What was the viscosity of the metallurgical melt during smelting? (iv) What kind of oxidation–reduction conditions existed during the process? (v) What kind of fluxes were used during smelting? For this purpose, slag samples were collected from the Kapliczna Hill, where metallurgical works were carried out in the 16th to 18th century (Figure 1). The conducted research is also crucial for future research to determine the impact of historical smelting on the natural environment. In order to assess the impact of metallurgy on the environment, the waste must first be analyzed in terms of mineralogy, petrography, and the conditions for its formation.1FIGURELocalization (a) and geological (b) geoportal.gov.pl), and a cross‐section through the Miedziana Góra ore (c) (modified from Rubinowski, 1970)LOCATION AND HISTORICAL BACKGROUNDThe research area is located in central Poland within the Holy Cross Mountains (Old Polish Industrial District; Figure 1). Miedziana Góra is located in the Świętokrzyskie voivodship, Kielce district. During fieldwork, slag samples were collected from the place where, in historical times, copper ore was mined and processed (Kowalczewski, 1972; Kowalczewski & Szczecińska, 1977; Figure 1).Historical data show that the beginnings of mining work in Miedziana Góra date back to 1590–1592. At that time, copper ore was most likely discovered during the exploitation of iron ores (Król & Urban, 2007; Miczulski, 1972). Polymetallic ore from Miedziana Góra was used for Cu and PbO production. Exploitation was carried out by vertical shafts, which were then connected with galleries (Król & Urban, 2003). The metallurgical works in Miedziana Góra were carried out intermittently up to the 18th century in a small smelter on the Kapliczna Hill (Wojciechowski, 2002). The copper obtained from these deposits was used to produce raw fittings, sheets, vessels, and alloys with lead and zinc. In addition, at the end of the 18th century, the metals produced were used to make coins in the Polish royal mint and to create weapons (Król & Urban, 2003; Molenda, 1989). One of the most famous products made of copper from Miedziana Góra was the copper sheets used to cover the roof of the Wawel Royal Castle after a fire in the 16th century (Cracow; Król & Urban, 2003). From the end of the 17th century, problems during operation (including the lack of effective drainage of the mines) caused a crisis in the mining and metallurgy in Miedziana Góra. Despite the fact that attempts were made to improve the situation, mainly through subsidies and building drainage adits, the ore was exploited only periodically (Król & Urban, 2003). Along with the region's development, large smelters processing ores from the entire area replaced the small ones next to the mines (Kowalczewski, 1972). At the end of the 18th century, copper mines were reinvested, but metallurgical works were carried out in other localities (Wojciechowski, 2002). Due to the lack of historical data (most of the documents were destroyed during World War II), the production volume in Miedziana Góra is estimated only based on geological data. According to the existing studies, the production volume between the 16th and early 18th centuries is estimated at 4750–6250 tons (Kowalczewski, 1972), or at least 6200 tons (Molenda, 1989). In the later period, only data are available for the years 1785–1788, when 26 tons of copper and 26 tons of PbO were produced (Król & Urban, 2003). Due to the high demand for copper in Poland at that time, the production did not cover the local needs, and copper was imported from abroad during the period of metallurgical activity in the Old Polish Industrial District (Molenda, 1989). Despite production mainly for the Polish market, specialists from other metallurgical centers from Poland (e.g., Olkusz, Tarnowskie Góry) and Europe (e.g., Germany, Italy, Slovakia) were involved in conducting mining and metallurgical work within the Old Polish Industrial District (Kowalczewski, 1972). Craftsmen associated with mining in the Holy Cross Mountains had numerous privileges. However, the privileges were associated with severe penalties for infractions, resulting in separating of a specific social group (Paulewicz, 1992).GeologyThe polymetallic ores in the Kielce region occur in many locations. However, apart from Miedziana Góra and Miedzianka, they form only small concentrations (Kowalczewski, 1972). The Miedziana Góra ore is located in the zone of the Miedziana Góra fold, overlapping the northern part of the Kostomłoty syncline. Lower Devonian sediments (conglomerates, quartzites, silts, schist, and ore‐bearing clays) are overlaid from the northeast by Upper Devonian formations (limestones and marbles; Wojciechowski, 2002; Figure 1). Based on complete chemical analyses that were carried out during the documentation of the ore in the 20th century, it consists mainly of SiO2 (av. 40.93 wt%), Al2O3 (av. 20.84 wt%), and Fe (over 10 wt% of Fe + FeS + FeO + Fe2O3), with Cu content up to 1.72 wt% (as Cu + CuS + CuO), Pb up to 0.07 wt% and Zn up to 0.85 wt%. However, in historical times ores with Cu content of up to 12 wt% were processed (Rubinowski, 1970). The Miedziana Góra ore can be divided into an oxidized zone and a sulfide zone (primary zone). In the primary zone, the main ore minerals are pyrite (FeS2), chalcopyrite (CuFeS2), chalcocite (Cu2S), bornite (Cu5FeS4), tetrahedrite ([Cu,Fe]12Sb4S13), sphalerite (ZnS), and galena (PbS). Sulfide minerals occur as single grains or clusters and radial clusters in clay rocks. The oxidation zone contains malachite (Cu2(CO3)(OH)2), azurite (Cu3(CO3)2(OH)2), chrysocolla ((Cu, Al)2H2Si2O5(OH)4·n(H2O)), iron oxides and hydroxides, siderite (FeCO3), smithsonite (ZnCO3), and manganese oxides (Piekarski, 1961).MATERIALS AND METHODSSamplingDuring fieldwork, in the spring of 2021 slag samples were collected from the area of Kapliczna Hill (in Miedziana Góra; Figure 1), where the smelter was most probably located (Figure 1). During the works, 62 samples of slags were collected, which were then divided into six macroscopically different types (MG1–MG6), described below.Chemical and phase analyzesThe first stage of the analysis involved macroscopic and microscopic observations to assess the slag's texture. For this purpose, an Olympus BX‐51 polarizing microscope was used (Institute of Earth Sciences, University of Silesia). The samples were then analyzed by scanning electron microscopy with energy‐dispersive spectrometry (SEM‐EDS). Analysis with SEM allowed us to more accurately determine the morphology of phases present in slags and their approximate chemical composition. A Phenom XL microscope (Faculty of Natural Sciences, University of Silesia) was used during this analysis. An electron micro‐probe (Cameca SX100, Inter‐Institutional Laboratory of Microanalysis of Minerals and Synthetic Materials, University of Warsaw) was used to determine the exact chemical composition of the phases present in the samples. EPM analyses were performed at 15 keV accelerating voltage, a 10–20.1 nA beam current, and a beam diameter of up to 5 μm. The following standards were used during the measurements: Na—albite (NaAlSi3O8); Mg, Si, Ca—diopside ([Ca,Mg,Fe]2SiO3); Al, K—orthoclase (KAlSi3O8); Ba—barite (BaSO4); Ti—rutile (TiO2); Cr—Cr2O3; Pb—crocoite (PbCrO4); Fe—Fe2O3, chalcopyrite (CuFeS2); Mn—rhodonite (CaMn3Mn[Si5O15]); Zn—sphalerite (ZnS); Cl—sodalite (Na8[Al6Si6O24]Cl2); P—YPO4; As—GaAs; Sr—celestine (SrSO4); Co—CoO; Ni—NiO; V—V2O5; Cu—cuprite (Cu2O); As—GaAs; S—chalcopyrite (CuFeS2), barite (BaSO4); Sb—stibnite (Sb2S3).Samples representing all six types of slags were also analyzed for phase composition and chemistry. The phase composition of the samples was determined using the PANalytical X'PERT PRO‐PW 3040/60 and PANalytical X'Pert PW 3710 X‐ray diffractometers equipped with CoKα1 (PW 3040/60) and CuKα1 (PW 3710) source radiation, Fe‐filter (for Co) and Ni‐filter (for Cu) to reduce the Kβ radiation. During analyses, the X'celerator detectors were used (Faculty of Natural Sciences, University of Silesia). The analyses were performed in the 5–90° 2θ angular range at 40 kV voltage (40 mA). For quantitative phase composition, a Rietveld analysis was performed. The X'PERT High Score Plus software and the PDF4+ database were used.To obtain the chemical composition, including major, minor, and trace elements, the combination of X‐ray fluorescence (XRF) spectrometry, inductively coupled plasma emission spectrometer (ICP‐ES), and inductively coupled plasma mass spectrometry (ICP‐MS) was applied. During the ICP analyses, multi‐acid digestion was used. XRF and ICP‐MS/ES analyses were performed by Bureau Veritas Mineral Laboratories (Canada). Loss of ignition was determined at 1000°C.High‐temperature experimentsExperimental melting in a laboratory furnace was used to determine the solidus and liquidus temperature of slags. Two types of slag were selected for the experiment (MG3 and MG6), representing the two most diverse slags. The MG3 sample is the most common type of slag among glassy samples (23 out of 58 samples), while the MG6 sample is the only hypocrystalline type of slag. The samples were crushed, cut into fragments with a volume of about 1 cm3, and heated in crucibles with gradually rising temperature until the entire material was melted. Each time after heating, the samples were subjected to rapid cooling. Experiments were performed at the Institute of Earth Sciences of the University of Silesia in a chamber furnace PLF 160/5 with a PC 442/18 controller, SiC heaters, and Thermocouple S with a maximum working temperature of 1550°C.SoftwareTo determine the liquidus temperatures, the MELTS‐Rhyolite v.1.0.2 software package (Ghiorso & Gualda, 2015; Gualda et al., 2012) was used. The liquidus temperature was calculated based on the bulk chemical composition (MG6) and the averaged chemical composition of the glass that forms the slags (MG1–MG4). Due to the presence of zones with increased accumulation of metallic and quartz/cristobalite phases, the MG5 sample was not used for temperature determination. For calculations by MELTS‐Rhyolite, chemical composition was normalized to 100 wt%, a pressure of 1 bar, and a Q‐Fa‐Mt oxygen buffer was used. For graphical data processing, the AUTODESK AutoCAD 2021 and CorelDRAW2021 version 22.1.1.523 (educational licenses) were used.RESULTSPetrographic characteristics of slagsSlags from Miedziana Góra were divided into glassy (MG1–MG5) and hypocrystalline (MG6) samples. X‐ray diffractometry (XRD) analyses confirmed the presence of an amorphous (glassy) phase in MG1–MG5 samples (Figure 3). Apart from the glass, MG1–MG5 slags contain SiO2 polymorphs (quartz and cristobalite; 15.7–90.4 vol%), clinopyroxenes (6.4–79.5 vol%), copper (Cu; 0.3–4.7 vol%), and litharge (PbO; up to 0.4 vol%). However, due to the lack of information on the percentage of glass, the presented results only include the relative proportions of the crystalline phases (Figure 3).The MG1 slag is black and contains quartz/cristobalite grains (up to 2–3 mm, occasionally up to 20 mm in size; 83.9 vol%; Figures 2a and 3). Imprints or fragments of charcoal appear on the surface. SEM (Figure 4a) observations confirmed that the MG1 sample comprises glass with numerous small (a few micrometers across; rarely up to several dozen micrometers; Figure 4a) metallic phases. They are composed of metallic Cu (1.4 vol%; Figure 3) and litharge (0.1 vol%; Figures 3 and 4a).2FIGUREMacroscopic images of slags occurring in Miedziana Góra3FIGUREResults of X‐ray diffraction analyses of slag samples from Miedziana Góra. Abbreviations: An, anorthite; Cpx, clinopyroxene; Cu, copper; Crs, cristobalite; Lth, litharge; Pb, lead; Qz, quartz; Wo, wollastonite group (bustamite + ferrobustamite)4FIGUREBackscattered electron images of phases occurring in slags from Miedziana Góra: (a) MG1 sample; (b) MG2 sample; (c) MG3 sample; (d) MG4 sample; (e) MG5 sample with decreasing brightness; (f) MG6 sample with magnification. Abbreviations: azu, azurite; cpx, clinopyroxene; Cu, metallic copper; gls, Glass; lit, litharge; wo, wollastonite group phasesMG2 slags have a light blue/greenish color with visible dark‐blue glassy layers with quartz/cristobalite grains up to 5 mm (83.2 vol%; Figures 2b and 3). This sample also contains metallic phases: Cu (0.3 vol%; Figure 3) and litharge (0.4 vol%; Figure 3) with a diameter up to 30 μm (Figure 4b) and skeletal clinopyroxene crystals (16.0 vol.%).The MG3 sample is dark blue in color with light‐blue and brown layers (Figure 2c). In MG3, quartz/cristobalite (76.4 vol%; Figure 3) grains are up to 1–2 mm, occasionally up to 5 mm in diameter (Figure 2c). In MG3 samples, clinopyroxene crystals were found in the form of dendritic crystals with lengths up to 100 μm (22.5 vol%; Figures 3 and 4c). In MG3 metallic phases were present as metallic copper (1.1 vol%). Cu sulfides (matte) and arsenides (speiss) are also present.The MG4 slag is brown in color, with visible interlacing of black slags. Quartz/cristobalite grains are present here in smaller quantities (15.7 vol%; Figure 3) than in the other glassy slag types (Figure 3) and are up to 1 mm in diameter (Figure 2). Clinopyroxenes constitute 79.5 vol% of crystalline phases present in the MG4 sample (Figure 3). In this sample, along with metallic Cu (4.7 vol%; Figure 3), metallic Pb, matte, and speiss occur (Figure 4d).Unlike the MG1–MG4 samples, the MG5 slags have a porous texture, with pores up to 20 mm in diameter. The MG5 slags are characterized by a variety of colors, but blue and greenish‐blue predominate. The MG5 sample consists of SiO2 polymorphs (90.4 vol%), clinopyroxene (6.4 vol%), metallic Cu (3.1 vol%), and litharge (0.1 vol%; Figure 3). The size of quartz/cristobalite grains reaches up to 20 mm (Figure 2e). Metallic aggregates composed of metallic Cu and litharge are up to 50 μm in diameter (Figure 4e). Occasionally present, azurite (Cu3(CO3)2(OH)2) (Figure 4e) is a secondary phase.The MG6 slags, unlike the others, are characterized by a hypocrystalline texture and a gray color. That type of slag is porous, with pore size up to 20–30 mm in diameter (Figure 2f). During SEM observations, skeletal crystals up to 400 μm long were commonly observed. These crystals are composed of phases with a structure corresponding to the wollastonite group minerals (54.6 vol% of bustamite and 5.6 vol% of ferrobustamite; Figure 3). Metallic Cu, along with Cu sulfides, are also noted up to a few micrometers in diameter (Figure 4f). The presence of anorthite (34.2 vol%), quartz (3.8 vol%), pyroxenes (1.2 vol%), and metallic phases (0.6 vol% Cu and 0.1 vol% Pb;) was revealed by XRD (Figure 3).Bulk chemical compositionThe glassy samples (MG1–MG5), despite the macroscopic differences, are characterized by similar chemical compositions (Table 1). Their main chemical components are: SiO2 (54.41–57.14 wt%; Table 1), CaO (11.25–15.36 wt%), FeO (8.42–11.03 wt%), and Al2O3 (7.77–9.14 wt%). Additionally, the presence of K2O (2.32–2.72 wt%), MgO (1.18–1.74 wt%), and MnO (0.99–1.56 wt%) was found (Table 1). Large differences among glassy slags occur in the content of Pb and Cu. The MG1, MG4, and MG5 slags contain more Pb (4.32–5.91 wt%) than the MG2 and MG3 slags (2.25 and 2.53 wt%, respectively), while samples MG1 and MG5 are characterized by higher Cu contents (0.97 and 1.61 wt%, respectively) than MG2 (0.40 wt%), MG3 (0.56 wt%), and MG4 (0.55 wt%; Table 1). Glassy slags also concentrate Zn (2168–5575 ppm), As (88–374 ppm), and Sb (72–298 ppm; Table 1). The hypocrystalline MG6 slag contains less SiO2 (49.81 wt%) and Pb (0.50 wt%), with higher amounts of CaO (21.64 wt%) and Cu (4.98 wt%) compared to glassy slags (MG1–MG5; Table 1).1TABLEChemical composition of slags from Miedziana GóraMG1MG2MG3MG4MG5MG6P2O5%0.580.430.440.590.610.38SiO255.2855.6157.1454.4155.6149.81TiO20.460.390.440.470.420.33Al2O39.147.798.808.947.777.54FeO8.4210.6111.0310.439.7810.75MnO0.991.561.471.161.051.51CaO14.8315.3611.2512.6911.5221.64MgO1.191.741.551.691.181.10Na2O0.170.150.140.150.150.10K2O2.322.462.642.722.451.67SO30.060.050.030.090.080.11BaO0.770.870.830.851.040.27TOT/C<0.02<0.020.09<0.020.040.10TOT/S0.02<0.020.020.040.030.05LOI−1.30−1.40−1.30−1.10−1.30−1.60Cu0.970.400.560.551.614.98Pb4.322.252.534.935.910.50Znppm378421684524506355752403As17288127235374471Sb1497286202298101Ni6538486811688Co20899136141234162Phase chemistryOxidesThe most common oxides in the analyzed slags are the SiO2 polymorphs. Quartz/cristobalite crystals (up to 99.1 wt% of SiO2; Table 2) are commonly found in MG1–MG5 slags and, to a lesser extent, in MG6 sample (Figures 3 and 4). Litharge (PbO) was observed in MG1, MG2, and MG5 slags in the form of phases resulting from chemical demixing of the metallic melt (Figure 4). It contains the addition of Cu (up to 1.1 wt% of CuO; Table 2).2TABLERepresentative EPMA (Electron Probe Micro‐analysis) data of the chemical composition of phases occurring in slagsqzwowocpxcpxcpxazulitbdssccCuCuSiO299.150.550.350.250.650.4na0.64S25.719.7bdlbdlTiO2bd0.100.110.21bdlbdlnabdlFe9.40.430.15bdlAl2O30.080.352.201.712.932.28na0.37Mn0.04bdlbdlbdlFeO0.065.06.412.813.812.9nabdlCo0.01bdlbdlbdlMnObdl1.901.971.411.481.60nabdlNibdlbdl0.20bdlMgObdl1.081.248.87.58.1nabdlCu61.977.693.698.0CaObdl39.336.123.522.222.5na0.14Zn0.12bdl0.15bdlK2Obdl0.100.42bdl0.320.28nabdlAs0.060.172.80.15PbOnabdlbdlnanana2.9098.1Sbbdlbdl0.50bdlCuOnananananana67.01.10Total97.2397.997.498.15Cr2O3bdlbdlbdl0.070.05bdlnabdlCoOnananananana0.12naTotal99.2498.3398.7498.798.8898.0670.02100.35a.p.f.u. (atom per formula unit)Atomic concentrationSi110.991.961.961.97‐0.02S41.1433.290.03‐Ti‐‐‐0.010.010.01‐‐Fe8.610.420.17‐Al‐0.010.050.080.130.1‐0.01Mn0.03‐‐‐Fe2+‐0.080.110.420.450.42‐‐Co0.01‐0‐Mn‐0.030.030.040.050.05‐‐Ni‐‐0.23‐Mg‐0.030.040.510.430.47‐‐Cu50.0766.1796.799.87Ca‐0.840.760.980.920.94‐0.01Zn0.09‐0.15‐K‐‐0.0100.020.01‐‐As0.040.122.440.13Pb‐‐‐‐‐‐0.040.89Sb‐‐0.27‐Cu‐‐‐‐‐‐2.950.03Total100100100100Cr‐‐‐‐‐‐‐‐Co‐‐‐‐‐‐‐o2‐23366632Abbreviations: azu, azurite; bdss, bornite–digenite solid solution; cc, chalcocite; cpx, clinopyroxene; Cu, copper; lit, litharge; qz, quartz; wo, wollastonite group; bdl, below detection limit; na, not analyzed.CarbonatesCarbonates, represented by azurite, were found only in sample MG5 and occur only in the vicinity of metallic phases (Figure 4e). Chemically, azurite comprises Cu (up to 67.0 wt% of CuO), with Pb substitutions reaching up to 2.9 wt% of PbO (Table 2).SilicatesTwo types of silicate were distinguished in the Miedziana Góra slags. The MG1–MG5 slags contain clinopyroxenes, with a chemical composition corresponding to diopside (CaMgSi2O6)–hedenbergite (CaFe2+Si2O6) solid solution, with numerous substitutions: Al (1.71–2.93 wt% of Al2O3), Mn (1.41–1.60 wt% of MnO), and minor concentrations of K2O, TiO2, and Cr2O3 (Table 2). In the MG6 sample, the wollastonite group are the only silicate phases found during EPM analyses, consisting of Si (50.3–50.5 wt% of SiO2), Ca (36.1–39.3 wt% of CaO), Fe (5.0–6.4 wt% of FeO), and Mn (1.90–1.97 wt% of MnO). The substitution of Al (up to 2.20 wt% of Al2O3), Mg (up to 1.24 wt% of MgO), and less than 0.5 wt% of oxide of Ti and K was observed (Table 2). The Fe2+ substitutions affect the unit cell size (Yamanaka et al., 1977). For this reason, in these phases, the unit cell is more closely related to the bustamite/ferrobustamite than to the wollastonite (Figure 3). The zoning visible in backscattered electron (BSE) images (Figure 4f) results from enrichment of the crystal rims in FeO with a simultaneous decrease in the CaO content.Sulfides and metallic phasesSulfides in the Miedziana Góra slags are represented by bornite (Cu5FeS4)–digenite (Cu9S5) solid solution (bdss) phases and by chalcocite. The bdss phases are mainly composed of Cu (up to 61.9 wt%), S (up to 25.7 wt%), and Fe (up to 9.4 wt%). In addition, they concentrate Zn (up to 0.12 wt%) and less than 0.1 wt% of Mn, Co, and As (Table 2). Chalcocite is composed of Cu (up to 77.6 wt%) and S (up to 19.7 wt%) with Fe (up to 0.43 wt%) and As (up to 0.17 wt%) substitution (Table 2). The metallic copper, apart from Cu (93.6–98.0 wt%), contains As (up to 2.8 wt%), Sb (up to 0.50 wt%), Ni (up to 0.20 wt%), Fe (up to 0.15 wt%), and Zn (up to 0.15 wt%; Table 2).GlassAll slag samples from Miedziana Góra contain glass. In MG1–MG5 samples, it is the main component. The glass in the MG1–MG4 slags is, on average, rich in SiO2 (52.1–57.1 wt%), Al2O3 (6.5–11.4 wt%), FeO (5.6–11.9 wt%), and CaO (15.3–16.3 wt%; Table 3). Additionally, PbO enrichment (up to 6.3 wt%) was observed. In the MG5 sample, zones with metallic phases and SiO2 polymorph concentrations are present (Figure 4e and Table 3). In these zones, glass contains more PbO (up to 21.2 wt%) and less CaO (av. 8.9 wt%) than in MG1–MG4 slags (Table 3). When compared to MG1–MG4 slags, a greater variation in SiO2 content was also observed in MG5 slags (up to 68.8 wt%; Table 3). In the MG6 slag, the average glass is also mostly made of SiO2 (av. 50.5 wt%), but the glass has a higher CaO (av. 21.9 wt%) and a lower PbO (av. 0.38 wt%) content. The proportion of other elements is similar to those observed in the MG1–MG5 slags (Table 3).3TABLEChemical composition of glasses occurring in Miedziana Góra slagsMG1MG2MG3MG4MG5MG6RangeAv. (n = 65)RangeAv. (n = 39)RangeAv. (n = 35)RangeAv. (n = 21)RangeAv. (n = 8)RangeAv. (n = 5)P2O50.08–0.710.570.36–0.560.450.36–0.560.470.44–0.700.540.26–0.670.560.29–0.370.34SiO254.9–72.457.152.3–55.353.751.4–54.052.849.8–54.052.150.6–68.856.850.3–50.950.5TiO20.32–0.650.530.29–0.450.360.32–0.450.370.24–0.430.340.29–0.850.470.31–0.370.33Al2O38.8–12.811.46.8–7.87.26.7–7.16.96.1–7.36.57.5–15.810.27.4–7.87.6FeO2.2–6.75.69.4–11.810.610.5–11.811.211.0–12.911.92.4–10.07.210.3–14.211.5MnO0.22–0.990.821.5–1.91.701.7–1.91.801.5–2.01.600.28–1.110.831.2–1.71.50CaO5.1–17.115.514.2–16.215.315.7–17.316.314.9–17.316.32.9–13.38.918.9–23.221.9MgO0.31–1.201.001.7–2.01.801.8–2.01.900.99–2.01.700.28–1.470.891.1–1.21.10Na2O0.00–0.190.100.00–2.230.110.00–0.200.090.00–0.170.070.12–0.310.200.02–0.090.05K2O2.4–5.42.802.0–2.62.302.0–2.42.201.8–2.42.101.6–5.92.901.3–1.81.50PbO0.00–4.02.001.7–4.33.02.4–3.63.02.6–6.34.22.1–21.28.80.23–0.610.38CuO0.00–0.340.030.00–0.860.030.00–0.260.070.00–5.80.370.20–0.390.260.15–0.430.30ZnO0.00–0.540.240.00–0.470.220.00–0.590.300.00–0.620.310.09–0.720.500.23–0.370.30BaO0.48–1.050.800.65–0.940.800.58–0.800.700.57–0.820.690.26–1.390.940.22–0.280.25Smelting temperatureExperimentDuring the smelting experiment, the first changes in the samples were observed at 1000°C. At this temperature, only the color of the slags was changed from blue to greenish‐gray in MG3 and from orange‐gray to greenish‐gray in MG6 samples, possibly due to the oxidation (Figure 5). At a temperature of 1100°C, sample MG3 was partially melted, leaving parts of the sample and quartz grains unchanged. Clear traces of partial melting also appeared on the surface of the MG6 sample (slag edges were smoothed; Figure 5). These changes indicate that the solidus temperature of both analyzed slags was exceeded. At a temperature of 1150°C, both types of slag were melted but not homogenized (Figure 5). The experiment was completed at a temperature of 1200°C. At this temperature, both samples were almost completely melted and homogenized. Only single quartz grains were visible on the glassy surface (Figure 5). Considering that quartz occurs in slags as grains unmelted during the smelting process, its occurrence at these temperatures could be ignored. Continuation of the experiment until the quartz melted and dissolved would lead to falsely elevated results.5FIGUREPhotos of samples after high‐temperature experimentsPhase diagramsThe chemical composition of the MG6 slag and the average chemical composition of the glass in the MG1–MG4 slags were plotted on the CaO–SiO2–Al2O3 and CaO–SiO2–MgO–10% Al2O3 phase diagrams (Muan & Osborn, 1965; Supplementary Material 1). In the case of glassy slags, the averaged glass composition was used because the chemical composition of the entire sample includes unmelted SiO2 grains (Figures 2 and 4). Due to the chemical composition of glass in the MG5 sample differing from the range for which the diagrams were designed, it was not included in plotting. Based on the CaO–SiO2–MgO–10% Al2O3 phase diagram, the liquidus temperature for the MG1–MG4 samples was in the range of 1400–1500°C (Supplementary Material 1; Table 4), while, for MG6, it was slightly below 1300°C (Supplementary Material 1; Table 4). Diagram CaO–SiO2–Al2O3 gave values of 1400–1500°C for MG2–MG4 slags and 1300–1400°C for MG1 and MG6 slags (Supplementary Material 1; Table 4).4TABLELiquidus and solidus temperatures of Miedziana Góra slags depending on the method usedMG1MG2MG3MG4MG5MG6MELTS‐rhyolite–liquidus temperature1118°C1150°C1155°C1150°C‐1139°CCaO–SiO2–MgO–10% Al2O3 phase diagram–liquidus temperature1400–1500°C1400–1500°C1400–1500°C1400–1500°C‐1200–1300°CCaO–SiO2–Al2O3 phase diagrams–liquidus temperature1300–1400°C1400–1500°C1400–1500°C1400–1500°C‐1300–1400°CExperiment–liquidus temperature‐‐1150–1200°C‐‐1150–1200°CExperiment–solidus temperature‐‐1100°C‐‐1100°CMELTS‐Rhyolite softwareThe modeling using MELTS‐Rhyolite software was another method applied to determine the smelting temperature (Ghiorso & Gualda, 2015; Gualda et al., 2012). During calculations, the average chemical composition of the glasses (MG1–MG4) and the bulk chemical composition of MG6 slag samples were used. The liquidus temperatures calculated with the MELTS‐Rhyolite software were in the range of 1118°C (MG1)–1155°C (MG3; Table 4).ViscosityOne of the most popular methods of melt viscosity determination in historical slags is the viscosity index (v.i.) proposed by Bachmann (1982) and modified by Ettler et al. (2009). This method is widely used in the analysis of smelting slags (e.g., Derkowska et al., 2021; Kupczak et al., 2020; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022) and takes into account the polymerizing to depolymerizing components ratio:v.i.=CaO+MgO+MnO+FeO+PbO+ZnOSiO2+Al2O3.$$ \mathrm{v}.\mathrm{i}.&amp;#x0003D;\frac{\mathrm{CaO}&amp;#x0002B;\mathrm{MgO}&amp;#x0002B;\mathrm{MnO}&amp;#x0002B;\mathrm{FeO}&amp;#x0002B;\mathrm{PbO}&amp;#x0002B;\mathrm{ZnO}}{\mathrm{Si}{\mathrm{O}}_2&amp;#x0002B;{\mathrm{Al}}_2{\mathrm{O}}_3}. $$To omit the unmelted SiO2 fragments present in the glassy slags for viscosity calculations, the bulk chemical composition of MG6 samples and the average chemical composition of the glass present in the MG1–MG5 samples were used. Considering this, the viscosity index for Miedziana Góra slags is in the range of 0.37–0.62 for glassy (MG1–MG5) and 0.62 for hypocrystalline (MG6) slags.Viscosity was also determined by using a model that includes both chemical composition and temperature (Giordano et al., 2006). In this method, the melt temperature is included in the calculation, and the resulting values are in SI units (Pa s). This model was proposed for natural rocks, and it has also been successfully applied for slags (Ettler et al., 2009; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022). The model is based on the following equation:logη=b1+b2*b3b3+SM+b4,$$ \log \eta &amp;#x0003D;b1&amp;#x0002B;\frac{b2\ast b3}{b3&amp;#x0002B;\mathrm{SM}}&amp;#x0002B;b4, $$where b1–b4 are temperature‐dependent parameters and SM is the structure modifier parameter. More detail on this model can be found in Giordano et al. (2006). In the case of Miedziana Góra slags, the viscosity was calculated for the temperature range of 1100–1200°C (Supplementary Material 2). Due to unmelted SiO2 fragments in glassy slags, calculations were performed for the averaged chemical composition of the glasses present in the MG1–MG5 samples. Due to the hypocrystalline nature, in the MG6 slags calculations were performed for bulk chemical composition. Viscosity calculated for glassy slags was mostly in the range of log η = 3.52–2.06 Pa s (for MG1–MG4 samples; Supplementary Material 2). Only porous glassy sample (MG5) has a higher viscosity (log η = 4.42–3.47 Pa s; Supplementary Material 2). The MG6 sample is characterized by lower viscosity (log η = 2.52–1.72 Pa s) than glassy slags (Supplementary Material 2).Due to the significant content of Cu and Pb in the slags from Miedziana Góra (Table 1; Table 3), additional viscosity calculations were applied according to the model proposed by Utigard and Warczok (UW model; 1995). Similar to the Giordano et al. (2006) method, the UW model considers the temperature and chemical composition of the slags. An additional advantage of this method is the fact that it has been proposed for metallurgical slags. Therefore, it takes into account Cu (as Cu2O) and Pb (as PbO). The calculations were made according to the formulalogηkgms=−0.49−5.1VR+−3660+12080VRTK,$$ \log \eta \left(\frac{\mathrm{kg}}{\mathrm{ms}}\right)&amp;#x0003D;-0.49-5.1\sqrt{\mathrm{VR}}&amp;#x0002B;\frac{-3660&amp;#x0002B;12080\sqrt{\mathrm{VR}}}{T(K)}, $$where VR (viscosity ratio) is the ratio of acidic oxides to basic oxides, calculated using the equation:VR=%SiO2+1.5%Cr2O3+1.2%ZrO2+1.8%Al2O31.2%FeO+0.5%Fe2O3+%PbO+0.8%MgO+0.7%CaO+2.3%Na2O+%K2O+0.7%Cu2O+1.6%CaF2.$$ \mathrm{VR}&amp;#x0003D;\frac{\left(\%\mathrm{Si}{\mathrm{O}}_2\right)&amp;#x0002B;1.5\left(\%\mathrm{C}{\mathrm{r}}_2{\mathrm{O}}_3\right)&amp;#x0002B;1.2\left(\%\mathrm{Zr}{\mathrm{O}}_2\right)&amp;#x0002B;1.8\left(\%\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3\right)}{1.2\left(\%\mathrm{FeO}\right)&amp;#x0002B;0.5\left(\%{\mathrm{F}\mathrm{e}}_2{\mathrm{O}}_3&amp;#x0002B;\%\mathrm{PbO}\right)&amp;#x0002B;0.8\left(\%\mathrm{MgO}\right)&amp;#x0002B;0.7\left(\%\mathrm{C}\mathrm{a}\mathrm{O}\right)&amp;#x0002B;2.3\left(\%\mathrm{N}{\mathrm{a}}_2\mathrm{O}&amp;#x0002B;\%{\mathrm{K}}_2\mathrm{O}\right)&amp;#x0002B;0.7\left(\%{\mathrm{Cu}}_2\mathrm{O}\right)&amp;#x0002B;1.6\left(\%\mathrm{C}\mathrm{a}{\mathrm{F}}_2\right)}. $$In this case, it was also decided to perform calculations for the average composition of the glass from glassy (MG1–MG5) slags and for the bulk chemical composition of the MG6 sample. During these calculations, the entire content of Cu was converted into Cu2O and Pb into PbO. Considering phase composition, the Fe occurs in slags as FeO rather than Fe2O3, and this was used for calculations. Calculations made with this method for glassy samples (MG1–MG5) gave results in the range of log η = 3.20–1.25 Pa s (Supplementary Material 2). The viscosity of the MG6 slag calculated by the UW model was lower (log η = 1.81–1.19 Pa s; Supplementary Material 2), but the difference was not as large as in the case of the Giordano et al. (2006) model.DISCUSSIONTemperature estimationsThe conducted experiment suggests that the solidus temperature of the studied slags was ~1100°C, while the liquidus temperature (in which it was possible to carry out smelting) was in the range of 1150–1200°C (Table 4). The temperatures obtained with MELTS software are slightly lower when compared to those obtained from the experimental method. The results obtained using phase diagrams (Supplementary Material 1) significantly exceed the temperatures determined by the experiment and calculations with the MELTS‐Rhyolite software (Table 4). The multicomponent nature of analyzed slags causes limited possibilities of effective application of phase diagrams. Most of the available diagrams take into account three to four components, and not necessarily those present in the analyzed slags in significant amounts. There are also diagrams designed for minerals that take into account more components, but in that case the chemical composition of the analyzed material should correspond to the chemical composition of the phases for which they were designed (e.g., Warchulski et al., 2022). For this reason, it is necessary to use those matching best the chemical composition of analyzed slags. Phase diagrams have been successfully used in the estimation of temperature in Rudawy Janowickie, Poland (Kierczak & Pietranik, 2011) and in Campiglia Marittima, Italy (Manasse et al., 2001), due to the less complex chemistry of the slags.The smelting temperature in the case of slags from Miedziana Góra is within the range of temperatures determined for slags after copper production in Spain (about 3000 years ago; ~1200°C; Sáez et al., 2003), in Crete in the Bronze Age (1150–1300°C; Bassiakos & Catapotis, 2006), and in Italy in the 13th to 14th century (1150–1300°C; Manasse & Mellini, 2002; Table 3), but slightly lower than in the smelters in the area of Old Copper Basin in Poland (14th to 20th century; 1200–1400°C; Derkowska et al., 2021; Kierczak & Pietranik, 2011; Kądziołka et al., 2020; Table 3).Viscosity estimationsThe melt viscosity during metal production is one of the most critical parameters that can describe the efficiency of the process. In the case of Miedziana Góra slags, the calculated values ​​of v.i. (0.37–0.62) are comparable with the slags from Leszczyna, Kondratów (18th to 20th century; 0.43–0.75; Table 5; Derkowska et al., 2021) and Okiep Copper District, South Africa (19th to 21st century; 0.22–0.69; Table 5). Higher v.i. were calculated for slags after copper production in the Rudawy Janowickie, Poland (14th to 16th century; 0.16–1.55, but mainly in the range of 0.55–1.55; Kierczak & Pietranik, 2011), Marsiliana, Italy (0.79–1.86; Table 5; Manasse & Mellini, 2002), and Itziparátzico, Mexico (14th to 16th century; 0.48–1.31; Table 5; Maldonado & Rehren, 2009). The high v.i. testifies to a better separation of copper‐containing phases from the slag and thus a more effective metallurgical process.5TABLESmelting temperatures and viscosities of Cu slags from selected locationsLocationTime (century)Temperature (°C)Viscosity (v.i.)Viscosity [log10 η = (pa s)]**Miedziana Góra16–181150–12000.37–0.621.72–4.42 (1200–1100°C)Kondratów and Leszczyna, Poland(Derkowska et al., 2021)18–201210–14000.43–0.750.23–2.19 (1400–1200°C)Rudawy Janowickie, Poland(Kierczak & Pietranik, 2011)14–161200–13000.16–1.551.37–5.79 (1200–1300°C)Itziparátzico, Mexico(Maldonado & Rehren, 2009)14–16nd0.48–1.31*‐Marsiliana, Italy(Manasse & Mellini, 2002)13–141150–13000.79–1.860.57–2.24 (1300–1150°C)Okiep Copper District, South Africa(Rozendaal & Horn, 2013)19–21nd0.22–0.69*‐Notes: nd, no data;*calculated based on chemical composition of slags;**calculated according to Giordano et al.'s (2006) model based on chemical composition of slags.The viscosity calculated according to Giordano et al. (2006) for the slags from Miedziana Góra (log η = 1.72–4.42 Pa s; Table 5) is within the viscosity range of the slags from the Rudawy Janowickie (log η = 1.37–5.79 Pa s; Table 5). The slags from Leszczyna and Kondratów (Derkowska et al., 2021) and slags from Marsiliana (Manasse & Mellini, 2002) are characterized by lower viscosity than those from Miedziana Góra (Table 5). It can be supposed that the metallurgy in Miedziana Góra was less efficient than the italian one and Leszczyna/Kondratów (Poland) metallurgy. Calculations made with the UW (Utigard & Warczok, 1995) model gave a similar range of results to the Giordano et al. (2006) method (Supplementary Material 2). Due to the lack of consideration of some components (mainly Pb and Cu) in samples with a high content of these metals, the Giordano et al. (2006) method significantly increases the viscosity (e.g., calculations made for glass in the MG5 sample; Supplementary Material 2).AtmosphereRedox conditions are one of the most important factors that should be considered in reconstructing metallurgical processes. For this purpose, Fe buffers are most often used, taking into account the temperature and the oxidation states of Fe in the samples (Sáez et al., 2003; Warchulski, Szczuka, & Kupczak, 2020). In the analyzed slags, Fe is present as Fe2+ rather than Fe3+. For this reason, the fugacity of oxygen can be described by quartz–fayalite–magnetite and wüstite–magnetite buffers (Zhao et al., 1999). Based on Fe buffers in the temperature range of 1100–1200°C, the fugacity of oxygen in the Miedziana Góra slag falls within the range of log fO2 = −12 (wüstite–magnetite buffer in 1100°C) to −8 atm. (quartz–fayalite–magnetite buffer at 1200°C; Zhao et al., 1999).When analyzing the slags after the production of metals from polymetallic deposits, apart from iron, the oxidation state of other metals can also be taken into account. The Ellingham diagram describes the transformations between the oxides and metallic phases most widely (Supplementary Material 3). The analyses show that in the MG6 slags Cu occurs mainly in metallic form (Figures 3 and 4f). For this reason, the fugacity of oxygen should be lower than log fO2 = −5 atm., while the lack of metallic Fe indicates the fugacity of oxygen greater than log fO2 = −14 atm. (Supplementary Material 3).In contrast, in glassy samples, aggregates made of Cu and PbO occur (Figure 4). The Ellingham diagram predicts the coexistence of metallic copper with PbO only in a narrow range of oxygen fugacity (log fO2 = −7 to −5 atm at 1100–1200°C; Supplementary Material 3). However, considering that the process was carried out in a sulfur‐rich atmosphere, the greater affinity of sulfur for Cu than for Pb widens the range of their coexistence (Supplementary Material 3).The diagrams (Supplementary Material 3) show that the production of Cu through matte also depends on sulfur fugacity (Supplementary Material 3). Unfortunately, no diagrams describing the relationship between the fugacity of oxygen and sulfur in the discussed temperature range (1100–1200°C) have been found.Despite the higher temperature, the diagrams proposed by Yazawa (1974) can be used to estimate the oxidation–reduction conditions during smelting in Miedziana Góra. As copper is mainly in the metallic form (except the MG5 sample, where it is in the carbonate form and results from slags weathering) during the smelting in Miedziana Góra, the oxygen fugacity can be estimated in the range of log fO2 = −4 to −12 atm. (Supplementary Material 3). The differentiation of samples in terms of chemical and phase composition makes it possible to estimate the oxidation‐reduction conditions at individual stages. During formation of the MG6 sample, oxygen fugacity should be in the range of approx. log fO2 = −6 to −12 atm. (Supplementary Material 3) and, during MG1–MG5 slag formation (to oxidize Pb), in the range of approx. log fO2 = −4 to −5 atm (Supplementary Material 3).Oxygen fugacity analysis based on Fe buffers in the case of Cu smelting showed a strongly reducing environment. In fact, the extent to which it can be carried out is wider and also depends on the fugacity of S2 (Supplementary Material 3).Summary of the metallurgical processIn Miedziana Góra, ore processing was carried out mainly to produce Cu and PbO (Król & Urban, 2003). The presence of metallic copper (Figures 3 and 4) indicates that the analyzed slags most likely come from the Cu production process. According to Agricola (1556), copper from sulfide ores was produced in a two‐stage process. The differences in chemistry (mainly in the content of Cu and Pb) and phase composition between glassy (MG1–MG5) and hypocrystalline (MG6) slags suggest they are the result of different smelting process stages. In the first smelting step, ore mixed with fluxes and witch charcoal was added to the shaft furnace and heated. In this production step, the primary goal was to heat the ore to form: (i) silicate slag; (ii) Cu‐rich speiss/matte (Davenport et al., 2002); (iii) metallic Cu. The workers controlled the amount of oxygen supplied during heating to oxidize as much Fe as possible (FeO was bonded within the slag). In order not to reduce the efficiency of the process, the workers did not allow Cu to be oxidized (Cu oxides would remain in the slags). As a result, part of Fe also did not oxidize and together with Cu, S, and As formed speiss/matte (Davenport et al., 2002). After the melt separation, the tapping hole was opened to remove and sort the metallic Cu, speiss/matte, and slag. The speiss/matte was further processed (in the second step) to obtain converted copper, while slags were discarded (Agricola, 1556; US Congress, Office of Technology Assessment, 1988).The lower Pb content in MG6 compared to MG1–MG5 slags may indicate that MG6 slags come from this stage of the process (Table 1). Due to the differences in density between Pb (11.35 g/cm3), Cu (8.94 g/cm3), and Fe (7.87 g/cm3; Atkins, 2001), the Pb should be effectively separated during this stage, where it must have accumulated at the lowest part of the furnace, leading to the formation of slags with low Pb content. The hypocrystalline nature of these slags is due to their lower viscosity (Supplementary Material 2). Under conditions of rapid cooling, low viscosity allows the formation of crystalline phases (Puziewicz et al., 2007). The presence of mainly metallic Cu and Fe2+ in wollastonite group phases marks the relatively low oxygen fugacity conditions during the formation of these slags.During the second step, the speiss/matte was remelted. At this stage, air was blown into the melt's surface to oxidize undesirable elements, and fluxes were added to bind them in the slags. In the case of Miedziana Góra, mainly Fe and Pb were removed during this step. Considering the high content of Pb and lower content of Cu, the glassy (MG1–MG5) slags most likely come from this stage of copper production (speiss/matte conversion). The presence of Pb in the form of PbO confirms the oxidative nature of this process. However, oxygen fugacity was not sufficient to oxidize Cu, which led to the formation of demix structures consisting of PbO and metallic Cu (Figure 4a,e). This was possible because of the higher sulfur affinity for Cu than for Pb and Fe (Supplementary Material 3). It also explains the absence of other sulfides (Pb or Fe) in the analyzed slags. The coexistence of metallic Cu with PbS could occur only in extremely non‐equilibrium conditions (Supplementary Material 3). The oxidation of Pb and Fe allowed them to react with fluxes, resulting in the formation of slags floating on the surface. The slag was then successively removed from the furnace. This resulted in a rapid temperature change, which, combined with the high viscosity of these slags, favors the formation of glass. The low diffusion rate and the lack of nucleation seeds could also be the reason for the low number of crystalline phases (Gawęda et al., 2013). The occasional presence of metallic Pb and Cu sulfides/arsenides in glassy slags could also be caused by the low oxygen diffusion. After all Fe and Pb were removed from the furnace, further oxidation cleaned the melt from sulfur (as SO2) and arsenic (as As2O3), leaving metallic copper in the furnace (Davis, 2001; US Congress, Office of Technology Assessment, 1988).The shaft furnace used for copper smelting in the 16th to 18th century was similar to those used in lead and gold metallurgy (Agricola, 1556; Warchulski, Szczuka, & Kupczak, 2020; Warchulski et al., 2022; Figure 6). The only difference was in the use of a different type of drive. In most cases, the smelters were located near rivers, the energy of which was used to drive the bellows (e.g., Warchulski, Szczuka, & Kupczak, 2020). In the case of Miedziana Góra, there is no river that could fulfill this function (Figure 1). For this reason, horse mills must have been used (Figure 6).6FIGUREReconstruction of the smelter that operated in Miedziana Góra, based on the description of the smelters operating in a similar period (Agricola, 1556)Ores and additivesThe documentation of the Miedziana Góra ore shows that ore minerals are found mainly in clays (Rubinowski, 1970). Chemically, these rocks are composed of SiO2, Al2O3, and Fe at different states of oxidation. The remaining elements are present in smaller amounts (on average, less than 5 wt% of oxide; Rubinowski, 1970). Comparing ore composition with the chemical composition of MG6 slags (resulting from speiss/matte production; Table 1) shows that the slags are characterized by a higher content of CaO and SiO2, with a lower content of Al2O3 (Table 1; Rubinowski, 1970). Higher SiO2 content (Table 1) may indicate the addition of sandstones occurring in the vicinity of the smelter (Figure 1). The primary aim of SiO2 addition was to produce two immiscible liquids and bond the FeO (Davenport et al., 2002). However, too much SiO2 has a negative effect, increasing the viscosity of the melt, which reduces the separation efficiency. For this reason, elements that break the silicate network were often added during the production of copper (Davenport et al., 2002; Potysz et al., 2015). In the case of Miedziana Góra, due to the increased content of CaO, limestones were most probably used for this purpose (Devonian carbonates; Figure 1).High SiO2 content (Table 1) and unmelted quartz/cristobalite (Figure 4) grains and the presence of other components (e.g., CaO, Al2O3; Table 1) in MG1–MG5 slags suggest that during the speiss/matte conversion process material containing SiO2, CaO, and Al2O3 (e.g., waste resulting from the enrichment of ores combined with sandstones/carbonates from the surface) was used. SiO2 polymorphs, as the most resistant to high temperature, have not been completely melted down.CONCLUSIONSThe presence of metal ores positively affected the development of the Old Polish Industrial District. It was possible thanks to the involvement of metallurgists from more experienced industrial centers, where the conditions under which smelting was carried out were similar. Based on the conducted analysis, the basic parameters that prevailed during the pyrometallurgical processing of ores from Miedziana Góra were reconstructed. Geochemical and mineralogical analyses allowed the isolation of slags from the production focused on speiss/matte and those from matte conversion. Both of these were formed under similar temperature conditions (liquidus temperature in the range of 1150–1200°C). The differences are visible in the viscosity of the metallurgical melt. The viscosity of the MG6 ranged between log η = 1.19 Pa s (UW model at 1200°C) and 2.52 Pa s (Giordano model at 1100°C). Low viscosity promoted the separation of the desired phases from the silicate melt, increasing melting efficiency. Because the second smelting step did not require low viscosity, slags after matte conversion are characterized by higher viscosity, ranging from log10 η = 1.25 Pa s (UW model at 1200°C) to 4.42 Pa s (Giordano et al., 2006 model at 1100°C). During this step, the main goal of metallurgists was to oxidize and bind undesirable elements with fluxes. The addition of SiO2 (a viscosity‐enhancing component) enabled the efficient removal of Fe from the furnace through crystallization of the Fe silicates. Differences in viscosity contributed to macroscopic differences between slags. In high‐viscosity slags, crystallization processes were hindered, which resulted in their glassy form. Based on the chemical and phase composition, the type of fluxes used during smelting was determined. The higher SiO2 and CaO content compared to the chemical composition of ores indicate the use of both carbonates and sand/sandstones during speiss/matte production, most probably from the vicinity of the smelter. In turn, during the speiss/matte conversion process, the waste resulting from the enrichment of ores combined with sandstones/carbonates from the surface was used. With the wide range of coexistence of phases found in the slags, oxidation–reduction conditions were also determined. The oxygen fugacity during speiss/matte smelting in Miedziana Góra was in the range of log fO2 = −5 (Supplementary Material 3) to −12 (wüstite–magnetite buffer at 1100°C; Zhao et al., 1999). Maintaining oxygen fugacity below the value at which copper would oxidize was necessary to limit losses due to the binding of copper oxides to slags during separation. During matte conversion, the oxygen fugacity was from log fO2 = −4 (Supplementary Material 3) to −7 (Supplementary Material 3). Oxygen fugacity in this range during matte conversion was necessary to purify the final product from Pb (through oxidation and binding to fluxes).ACKNOWLEDGEMENTSThis study was supported by the National Science Center (NCN) grant no. 2019/35/O/ST10/00313.DATA CITATIONNo dataset from external repositories was used in this study. All existing data used for this publication are cited in reference section.DATA AVAILABILITY STATEMENTThe data that support the finding of the study are given in the tables and figures within the main text. This study brought together existing data from several different sources, which are cited throughout.REFERENCESAgricola, G. (1556). De Re metallica libri XII (Ed Muzeum Karkonoskie w Jeleniej Górze, 2000), Jelenia Góra.Aleksander‐Kwaterczak, U., & Helios‐Rybicka, E. (2009). Contaminated sediments as a potential source of Zn, Pb, and cd for a river system in the historical metalliferous ore mining and smelting industry area of South Poland. Journal of Soils and Sediments, 9, 13–22. https://doi.org/10.1007/s11368‐008‐0051‐zAtkins, P. W. (2001). Chemia fizyczna. Wydaw.Bachmann, H. G. (1982). The identification of slags from archaeological sites. Institute of Archaeology Occasional Publication.Baron, S., Carignan, J., & Ploquin, A. (2006). Dispersion of heavy metals (metalloids) in soils from 800‐year‐old pollution (Mont‐Lozere, France). Environmental Science & Technology, 40(17), 5319–5326. https://doi.org/10.1021/es0606430Bassiakos, Y., & Catapotis, M. (2006). Reconstruction of the copper smelting process based on the analysis of ore and slag samples. Hesperia Supplements, 36, 329–353.Bourgarit, D., Mille, B., Prange, M., Ambert, P., & Hauptmann, A. (2003). Chalcolithic fahlore smelting at Cabrières: Reconstruction of smelting processes by archaeometallurgical finds. Archaeometallurgy in Europe, 1, 431–440.Bränvall, M. L., Bindler, R., Emteryd, O., & Renberg, I. (2001). Four thousand years of atmospheric lead pollution in northern Europe: A summary from Swedish lake sediments. Journal of Paleolimnology, 25(4), 421–435. https://doi.org/10.1023/A:1011186100081Burger, E., Bourgarit, D., Wattiaux, A., & Fialin, M. (2010). The reconstruction of the first copper‐smelting processes in Europe during the 4th and the 3rd millennium BC: Where does the oxygen come from? Applied Physics a, 100, 713–724. https://doi.org/10.1007/s00339‐010‐5651‐yCabała, J., Warchulski, R., Rozmus, D., Środek, D., & Szełęg, E. (2020). Pb‐rich slags, minerals, and pollution resulted from a medieval ag‐Pb smelting and mining operation in the Silesian‐Cracovian region (southern Poland). Minerals, 10, 28. https://doi.org/10.3390/min10010028Chirikure, S. (2015). Metals in past societies: A global perspective on indigenous African metallurgy. Springer.Davenport, W. G., King, M. J., Schlesinger, M. E., & Biswas, A. K. (2002). Extractive metallurgy of copper. Elsevier.Davis, J. R. (Ed.) (2001). Copper and copper alloys. ASM international. https://doi.org/10.31399/asm.tb.aub.t61170457Derkowska, K., Świerk, M., & Nowak, K. (2021). Reconstruction of copper smelting technology based on 18–20th‐century slag remains from the old Copper Basin, Poland. Minerals, 11(9), 926. https://doi.org/10.3390/min11090926Ettler, V. (2016). Soil contamination near non‐ferrous metal smelters: A review. Applied Geochemistry, 64, 56–74. https://doi.org/10.1016/j.apgeochem.2015.09.020Ettler, V., Červinka, R., & Johan, Z. (2009). Mineralogy of medieval slags from lead and silver smelting (Bohutín, Příbram district, Czech Republic): Towards estimation of historical smelting conditions. Archaeometry, 51, 987–1007. https://doi.org/10.1111/j.1475‐4754.2008.00455.xGawęda, A., Janeczek, J., Kierepka, M., Kądziołko‐Gaweł, M., & Krzykawski, T. (2013). Indialite‐rich paralava from a coalmine waste‐dump, Sosnowiec, Poland. Neues Jahrbuch für Mineralogie‐Abhandlungen, 190, 237–251. https://doi.org/10.1127/0077‐7757/2013/0238Geoportal. n.d. Available online: https://www.geoportal.gov.pl (accessed on 10 January 2020).Ghiorso, M. S., & Gualda, G. A. (2015). An H2O–CO2 mixed fluid saturation model compatible with rhyolite‐MELTS. Contributions to Mineralogy and Petrology, 169(6), 1, 53–30. https://doi.org/10.1007/s00410‐015‐1141‐8Giordano, D., Mangiacapra, A., Potuzak, M., Russell, J. K., Romano, C., Dingwell, D. B., & Di Muro, A. (2006). An expanded non‐Arrhenian model for silicate melt viscosity: A treatment for metaluminous, peraluminous and peralkaline liquids. Chemical Geology., 229(1–3), 42–56. https://doi.org/10.1016/j.chemgeo.2006.01.007Gualda, G. A., Ghiorso, M. S., Lemons, R. V., & Carley, T. L. (2012). Rhyolite‐MELTS: A modified calibration of MELTS optimized for silica‐rich, fluid‐bearing magmatic systems. Journal of Petrology, 53(5), 875–890. https://doi.org/10.1093/petrology/egr080Kądziołka, K., Pietranik, A., Kierczak, J., Potysz, A., & Stolarczyk, T. (2020). Towards better reconstruction of smelting temperatures: Methodological review and the case of historical K‐rich cu‐slags from the old Copper Basin, Poland. Journal of Archaeological Science, 118, 105142. https://doi.org/10.1016/j.jas.2020.105142Kapper, L., Turck, R., Goguichaishvili, A., & Morales, J. (2017). Rock – And archeomagnetic investigation of prehistoric copper slag from the Swiss Alps. Latinmag Letters, 7, 1–4.Kierczak, J., & Pietranik, A. (2011). Mineralogy and composition of historical cu slags from the Rudawy Janowickie Mountains, southwestern Poland. The Canadian Mineralogist, 49(5), 1281–1296. https://doi.org/10.3749/canmin.49.5.1281Kierczak, J., Potysz, A., Pietranik, A., Tyszka, R., Modelska, M., Néel, C., Ettler, V., & Mihaljevič, M. (2013). Environmental impact of the historical cu smelting in the Rudawy Janowickie Mountains (South‐Western Poland). Journal of Geochemical Exploration, 124, 183–194. https://doi.org/10.1016/j.gexplo.2012.09.008Kowalczewski, Z. (1972). Podstawowe problemy świętokrzyskiego górnictwa kruszcowego w świetle danych geologicznych. In Materiały do sesji naukowej odbytej z okazji jubileuszu IX wieków Kielc (pp. 57–78). Wydawnictwa Geologiczne.Kowalczewski, Z., & Szczecińska, A. (1977). Wyniki badań nad żużlami z kilku stanowisk dawnego hutnictwa metali nieżelaznych w Górach Świętokrzyskich. Rocz. Świętokrz, 5, 151–168.Król, P., & Urban, J. (2003). Kopalnie w Miedzianej Górze iw Ławęcznej oraz ochrona ich pozostałości (p. 29). Rocznik Świętokrzyski. Seria B. Nauki Przyrodnicze.Król, P., & Urban, J. (2007). Kopalnie Miedzianogórskie. Kielce.Kundig, K. J., & Weed, R. D. (2015). Copper and copper alloys. Mechanical Engineers' Handbook (3rd Edition): Materials and Mechanical Design, 1, 117–228.Kupczak, K., Warchulski, R., Dulski, M., & Środek, D. (2020). Chemical and phase reactions on the contact between refractory materials and slags, a case from the 19th century Zn‐Pb smelter in Ruda Śląska, Poland. Minerals, 10, 1006. https://doi.org/10.3390/min10111006Maldonado, B., & Rehren, T. (2009). Early copper smelting at Itziparátzico, Mexico. Journal of Archaeological Science, 36(9), 1998–2006. https://doi.org/10.1016/j.jas.2009.05.019Manasse, A., & Mellini, M. (2002). Archaeometallurgic slags from Kutná Hora. Neues Jahrbuch Für Mineralogie‐Monatshefte, 369‐384, 369–384. https://doi.org/10.1127/0028‐3649/2002/2002‐0369Manasse, A., Mellini, M., & Viti, C. (2001). The copper slags of the Capattoli Valley, Campiglia Marittima, Italy. European Journal of Mineralogy, 13(5), 949–960. https://doi.org/10.1127/0935‐1221/2001/0013/0949Miczulski, S., 1972. Początki rozwoju górnictwa i hutnictwa kruszcowego w rejonie Kielc do końca XVI w., [in:] Kowalczewski Z. (red.) Dzieje i Technika Świętokrzyskiego Górnictwa i Hutnictwa Kruszcowego. Wydawnictwa geologiczne, Warszawa, pp. 79–100.Molenda, D. (1989). Eksploatacja rud miedzi i handel miedzią w Polsce w późnym średniowieczu iw początkach nowożytności (do 1795 r.). Przegląd Historyczny, 80(4), 801‐814.Muan, A., & Osborn, E. F. (1965). Phase equilibria among oxides in steelmaking. Addison‐Wesley publishing company, INC.Nriagu, J. O. (1994). Mercury pollution from the past mining of gold and silver in the Americas. Science of the Total Environment, 149, 167–181. https://doi.org/10.1016/0048‐9697(94)90177‐5Paulewicz, M. (1992). Chęcińskie górnictwo kruszcowe (IV do poł. XVII wieku). Kieleckie Towarzystwo Naukowe, Kielce.Piekarski, K. (1961). W sprawie genezy złoża rud miedzi w Miedzianej Górze koło Kielc. Prace Geologiczne, 3, 43–58.Potysz, A., Kierczak, J., Pietranik, A., & Kądziołka, K. (2018). Mineralogical, geochemical, and leaching study of historical cu‐slags issued from processing of the Zechstein formation (old Copper Basin, southwestern Poland). Applied Geochemistry, 98, 22–35. https://doi.org/10.1016/j.apgeochem.2018.08.027Potysz, A., van Hullebusch, E. D., Kierczak, J., Grybos, M., Lens, P. N., & Guibaud, G. (2015). Copper metallurgical slags–current knowledge and fate: A review. Critical Reviews in Environmental Science and Technology, 45, 2424–2488. https://doi.org/10.1080/10643389.2015.1046769Puziewicz, J., Zainoun, K., & Bril, H. (2007). Primary phases in pyrometallurgical slags from a zinc‐smelting waste dump, Swietochłowice, upper Silesia, Poland. The Canadian Mineralogist, 45(5), 1189–1200. https://doi.org/10.2113/gscanmin.45.5.1189Roberts, B. W., Thornton, C. P., & Pigott, V. C. (2009). Development of metallurgy in Eurasia. Antiquity, 83(322), 1012–1022. https://doi.org/10.1017/S0003598X00099312Rozendaal, A., & Horn, R. (2013). Textural, mineralogical and chemical characteristics of copper reverb furnace smelter slag of the Okiep Copper District, South Africa. Minerals Engineering, 52, 184–190. https://doi.org/10.1016/j.mineng.2013.06.020Rubinowski, Z. (1970). Wyniki badań rudonośności w rejonie Miedzianej Góry i Ławeczna. Archiwum Państwego Instytutu Geologicznego.Sáez, R., Nocete, F., Nieto, J. M., Capitán, M. Á., & Rovira, S. (2003). The extractive metallurgy of copper from Cabezo Juré, Huelva, Spain: Chemical and mineralogical study of slags dated to the third millenium BC. The Canadian Mineralogist, 41(3), 627–638. https://doi.org/10.2113/gscanmin.41.3.627Ströbele, F., Wenzel, T., Kronz, A., Hildebrandt, L. H., & Markl, G. (2010). Mineralogical and geochemical characterization of high‐medieval lead–silver smelting slags from Wiesloch near Heidelberg (Germany)—An approach to process reconstruction. Archaeological and Anthropological Sciences, 2(3), 191–215. https://doi.org/10.1007/s12520‐010‐0039‐7Sutkowska, K., Czech, T., Teper, L., & Krzykawski, T. (2013). Heavy metals soil contamination induced by historical zinc smelting in Jaworzno. Ecological Chemistry and Engineering a, 20(12), 1441–1450. https://doi.org/10.2428/ecea.2013.20(12)130Tylecote, R. F. (1992). A history of metallurgy (2nd ed.). The Institute of Materials.US Congress, Office of Technology Assessment. (1988). Copper production technology. Government Printing Office.Utigard, T. A., & Warczok, A. (1995). Density and viscosity of copperhickel sulphide smelting and converting slags. In W. J. Chen, C. Dim, A. Luraschi, & P. J. Mackey (Eds.), Copper 95‐Cobre 95 proceedings of the third international conference, Vol. l V Pyrometallurgy of copper (pp. 423–437). The Metallurgical Society of CIM.Warchulski, R. (2016). Zn‐Pb slag crystallization: Evaluating temperature conditions on the basis of geothermometry. European Journal of Mineralogy, 28(2), 375–384. https://doi.org/10.1127/ejm/2015/0027‐2496Warchulski, R., Gawęda, A., Kupczak, K., Banasik, K., & Krzykawski, T. (2020). Slags from Ruda Śląska, Poland as a large‐scale laboratory for the crystallization of rare natural rocks: Melilitolites and paralavas. Lithos, 372–373, 105666. https://doi.org/10.1016/j.lithos.2020.105666Warchulski, R., Kupczak, K., Gawęda, A., & Sitko, R. (2022). Complete reconstruction of the process and conditions during gold smelting in the 15‐17th century in Złoty Stok based on metallurgical slags. Archaeometry, 64, 916–934. https://doi.org/10.1111/arcm.12752Warchulski, R., Mendecki, M., Gawęda, A., Sołtysiak, M., & Gadowski, M. (2019). Rainwater‐induced migration of potentially toxic elements from a Zn–Pb slag dump in Ruda Śląska in light of mineralogical, geochemical and geophysical investigations. Applied Geochemistry, 109, 104396. https://doi.org/10.1016/j.apgeochem.2019.104396Warchulski, R., Szczuka, M., & Kupczak, K. (2020). Reconstruction of 16th—17th century Lead smelting processes on the basis of slag properties: A case study from Sławków, Poland. Minerals, 10(11), 1039. https://doi.org/10.3390/min10111039Wojciechowski, A. (2002). Hałdy dawnego górnictwa świętokrzyskiego jako źródło metali kolorowych i szlachetnych. Przegląd Geologiczny, 50(3), 240–244.Yamanaka, T., Sadanaga, R., & Takeuchi, Y. (1977). Structural variation in the ferrobustamite solid solution. American Mineralogist, 62(11–12), 1216–1224.Yazawa, A. (1974). Thermodynamic considerations of copper smelting. Canadian Metallurgical Quarterly, 13(3), 443–453. https://doi.org/10.1179/cmq.1974.13.3.443Zhao, D., Essene, E. J., & Zhang, Y. (1999). An oxygen barometer for rutile–ilmenite assemblages: Oxidation state of metasomatic agents in the mantle. Earth and Planetary Science Letters, 166(3–4), 127–137. https://doi.org/10.1016/S0012‐821X(98)00281‐7Zhou, Y., Wang, L., Xiao, T., Chen, Y., Beiyuan, J., She, J., Zhou, Y., Yin, M., Liu, J., Liu, Y., Wang, Y., & Wang, J. (2020). Legacy of multiple heavy metal (loid)s contamination and ecological risks in farmland soils from a historical artisanal zinc smelting area. Science of the Total Environment, 720, 137541. https://doi.org/10.1016/j.scitotenv.2020.137541

Journal

Archaeometry – Wiley

Published: Jun 1, 2023

Keywords: copper; oxygen fugacity; process reconstruction; slags; smelting; temperature; viscosity

There are no references for this article.