Potential for the Recovery of Selected Metals and Critical Raw Materials from Slags from Polymineral Zn–Pb Ore Metallurgy—Part I

Abstract

Slags from the Silesia–Cracow Upland (Poland), including ten historical slags (deposited in waste dumps) and four contemporary slags (from current production), were examined to compare their chemical and mineralogical properties as well as to assess their potential for the recovery of selected metals and critical raw materials. The historical slags associated with the smelting of polymetallic ores originating from Mississippi Valley-type (MVT) deposits consisted primarily of gypsum. The contemporary slags, obtained from industrial waste rich in zinc and lead, were predominantly spinels (magnesium-aluminate and ferric) that exhibited higher iron content (up to 46.6 wt% of Fe₂O₃) compared to the historical slags (up to 26.1 wt% of Fe₂O₃). The zinc content was similar for both the slag types (3.5 wt% Zn). The average titanium and arsenic contents in the old and contemporary slags were at the same level as well, with 0.21 wt% (Ti) and 0.13 wt% (As), respectively. The contemporary slags contained higher levels of critical raw materials, such as cobalt, nickel, copper, and manganese, compared to the historical slags. Rare earth elements (REEs) were also more abundant in the contemporary slags, with an average content of 212 ppm, while the historical slags averaged 124 ppm. These findings underscore the potential for recovering valuable metals and critical raw materials from such slags, presenting opportunities for resource optimisation and environmental management.

1. Introduction

In response to the need to strengthen competitiveness and innovation in a dynamically changing global economic environment, the European Commission and EU Member States have drafted a number of strategic documents that specify priority actions aimed at ensuring the sustainable supply of critical raw materials (CRMs). Pursuant to the Act on Critical Raw Materials, adopted in March 2024 by the European Commission, this term refers to raw materials of high economic importance to the EU, the supply of which is likely to be disrupted due to the concentration of sources and the lack of good, affordable substitutes [1].

As a result, the Critical Raw Material Act (CRM Act) was proposed by the European Commission. CRM Act is a big step towards protecting the EU from the dangers of critical raw material supply disruptions and the weaknesses in the EU’s supply chains. The CRM Act will make sure the EU has access to safe and sustainable raw materials, which will help Europe reach its climate goals and digital goals, keep EU industries competitive, and keep the single market running smoothly. The Critical Raw Materials Act will ensure EU access to a secure and sustainable supply of critical raw materials, enabling Europe to meet its 2030 climate and digital objectives. One of the elements of the CRM Act is the setting of reference points (benchmarks) for domestic capacity by 2030: at least 10% of the annual EU consumption for extraction; at least 40% of annual EU consumption for processing purposes; at least 15% of annual EU consumption for recycling; no more than 65% of annual EU consumption from a single third country Achieving the set goals is extremely demanding and challenging, but it is crucial for the development prospects of the EU economy [1].

The Raw Materials Initiative [2] was adopted within this scope in 2008, while the CRM list for the European Union was updated four times in 2011, 2017, 2020 and 2023 [3,4,5,6]. Currently, the CRM list includes 34 raw materials: aluminium/bauxite, antimony, barite, beryllium, bismuth, borate, cobalt, coking coal, fluorspar, gallium, germanium, hafnium, HREE (10 REE metals), lithium, LREE (5 REE metals), magnesium, natural graphite, niobium, phosphate rock, phosphorus, PGM (5 platinum-group metals), scandium, silicon metal, strontium, tantalum, titanium metal, tungsten, vanadium, arsenic, feldspar, helium, manganese, copper and nickel (

Table 1. Current list of critical raw materials divided into groups (authors’ own review based on [6]).

METALS	METALLOIDS
heavy rare earth elements (HREEs)	germanium
light rare earth elements (LREEs)	silicon metal
platinum-group metals (PGMs)	arsenic
antimony
beryllium	NON-METALS
bismuth	helium
gallium	phosphorus
hafnium
cobalt	MINERALS AND OTHERS
lithium	barite
magnesium	borate
manganese	aluminium/bauxite
copper	phosphate rock
nickel	fluorspar
niobium	natural graphite
scandium	feldspar
strontium	coking coal
tantalum
titanium metal
vanadium
tungsten

). The raw materials are primarily listed in the form of elements. The list includes 38 metals, 3 metalloids and 2 non-metals [6].

As regards raw material source optimisation (including critical raw materials) for the development of a modern economy, in 2023, Poland adopted the document titled “The National Raw Materials Policy” with a timeframe spanning towards 2050 (PSP2050) [7]. The European Commission documents, as well as PSP2050, indicate the need to intensify action for diversifying CRM sources. One such action, listed in PSP2050 under detailed objective no. 6, is obtaining raw materials from anthropogenic deposits and supporting the development of a circular economy.

As part of this goal, projects will be supported in the following areas: (i) stocktaking of mining waste landfill sites and assessment of the potential for their use, (ii) establishing a knowledge base dedicated to the issue of waste as a source of raw materials, with the appropriate classification and indication of directions of use, and (iii) action aimed at developing a technology for raw material recovery from waste (in particular strategic and critical raw materials), including the development of technologies for processing such waste.

Activities aimed at developing technologies for the recovery of CRMs from waste material should be preceded by comprehensive mineralogical and chemical studies. The first stage of the work should be the determination of the amount of CRMS (mainly metals) in the slags and the determination of the forms of their occurrence.

A good place for such research is the Silesia–Cracow region, located in southern Poland, which is the most industrialised region in East–Central Europe. The industrial history of the region dates back to the Middle Ages when rich deposits of lead/silver ores were discovered, mined in dozens of shallow mines and processed in many local smelters [8].

During the 19th-century Industrial Revolution and in the first half of the 20th century, with the discovery of rich deposits of hard coal used in metallurgy, the Silesia–Krakow region became the world centre for zinc production [9]. Zinc was produced first from oxidised and later from sulfided zinc ores, and, by processing former waste from lead production today, mining of Zn–Pb ores in the region has ceased, while modern Zn–Pb metallurgy continues based on imported raw materials.

Several dozen zinc and lead smelters could be found over the last 200 years in the Silesia–Cracow Upland, which produced approximately 19,410,000 m³ (32,997,000 Mg) of slag from various metallurgical processes, which was, until recently, deposited in 19 waste dumps with a total area of 152 ha [10]. Currently, the volume and surface area of the slag deposited in the waste dumps is decreasing and is not precisely known. These slags originate from the smelting of polymetallic ores found in Mississippi Valley-type (MVT) deposits [8,11,12]. In addition to Zn and Pb, the slag contains many co-occurring elements as well (e.g., Cu, Ag, Mn, Tl, As, Ba, Bi, Ge, Hg, Ni, P and Sb) [13,14].

Currently, there are two zinc and lead smelting plants in the Silesia–Cracow Upland. Contemporary slags are produced in volumes of approximately 200,000 Mg/year [15]. However, the contemporary slags in this region do not originate from natural metal ore smelting. The raw material used in the smelting plants is industrial waste rich in zinc and lead, produced as part of various technological processes [16].

Knowledge regarding the content of selected metals and CRMs in the Silesia–Cracow slags, as well as the current volume of this waste, is fragmentary and dispersed. With the exception of individual cases, the vast majority of the slags have not been subjected to detailed studies [17,18,19].

Based on the available data, it is, therefore, impossible to determine the overall potential of slags associated with the smelting of polymetallic Zn–Pb ores from MVT deposits present as raw materials for the recovery of selected metals and critical raw materials. However, outside Poland, fragmentary research concerning slags as a potential source of CRMs was carried out in Belgium for the historical waste deposited in the Liege area [20,21]. Generally, there is a global information gap in this respect despite the significant volumes of polymetallic Zn–Pb MVT deposits, as well as mining and metallurgical waste in the United States, Canada, Europe and Australia [22]. It is estimated that approximately 25% of the global Zn–Pb ore mining is associated with MVT deposits [23].

Until now, there have been no comprehensive studies of slags originating from polymetallic Zn–Pb ore metallurgy. Hence, this study focuses on assessing the contents of selected metals and critical raw materials in historical slags (deposited in dumps) associated with the smelting of polymetallic Zn–Pb ores from MVT deposits. For comparison, the selected element content was also examined in contemporary slags (from current production). The research on contemporary slags produced in the Imperial Smelting Furnace process in the Miasteczko Śląskie zinc smelter, presented by Kicińska et al. [17] concerned detailed chemical and phase characteristics of slags depending on the grain fraction and environmental risk. However, this work does not provide results regarding the content of CRMs in modern slags from Zn–Pb metallurgy.

2. Materials and Methods

2.1. Materials

In the Silesia–Cracow Upland, metallurgical slags are currently deposited in 17 locations, which encompass slags from past smelting of Zn–Pb ores deposited in 15 locations and slags from contemporary metallurgical plants deposited in 2 locations. Samples were taken from five historical slag dumps and two plants producing contemporary slags. The locations of the historical and contemporary slag deposition sites are shown in Figure 1, while detailed information on the sampled waste dumps/smelters is provided in

Table 2. Locations and descriptions of sampling sites for slag testing.

Location No.	Coordinates		City	Smelter	Sample Name		Period	Technology
	N	E
Historical slags
5	50°15′59″	18°52′10″	Ruda Śląska	Hugo	H1	H2	1812–1933	B/S
7	50°18′39″	18°53′38″	Świętochłowice	ZM Silesia	S1	S2	1847–1998	B/S
8	50°19′00″	18°54′15″	Świętochłowice	Guidotto	G1	G2	1887–1937	B/S
9	50°22′04″	18°58′07″	Piekary Śląskie	KH Waryński	W1	W2	1926–1990	ZORK
11	50°20′31″	18°58′27″	Piekary Śląskie	ZGH Orzeł Biały	O1	O2	1924–1991	ZORK
Contemporary slags
16	50°16′36″	19°28′33″	Bukowno	ZGH Bolesław S.A.	B1	B2	1955–present	ZORK
17	50°30′10″	18°55′33″	Miasteczko Śląskie	HC Miasteczko Śląskie S.A.	M1	M2	1960–present	ISF

Legend: B/S—Belgian/Silesian process, ISF—Imperial Smelting Furnace, ZORK—Zinc Oxide Rotary Kiln.

. Two types of samples were collected at each site (waste dump, metallurgical plant), differing in grain size, colour (in the case of historical slags) and type of original charge (in the case of contemporary slags). The characteristics of the collected samples are presented in

Table 3. Characteristics of the collected slag samples.

Historical Slags				Contemporary Slags
Sample	Description	Sample	Description	Sample	Description
H1	grains of 0–10 mm	W1	fine grains	B1	zinc-bearing sludge charge
H2	grains over 20 mm	W2	cemented grains	B2	zinc-bearing dust charge
S1	grains of 0–10 mm	O1	grains of 0–10 mm	M1	oxide charge
S2	grains over 20 mm	O2	grains over 10 mm	M2	sulphide charge
G1	red-coloured grains
G2	red-coloured grains

.

The metallurgical slag sampling was carried out using methods accredited at the Department of Environmental Monitoring of the Central Mining Institute-National Research Institute in Katowice (GIG-PIB), according to internal research procedure SC-3.PB.03 ver. 1 [24]. The following procedures were applied:

• Slag dumps used for aggregate extraction were sampled at the finished product piles, with 15 subsamples taken to produce one test sample (dumps no. 5, 7, 11, Table 2, Figure 2);
• Unused dumps were sampled with the use of a hand auger. Fifteen boreholes were drilled in the dump to a depth of 2.0 m. The slags collected from each borehole were combined, averaged and quartered so as to obtain a representative test sample (dumps no. 8, 9, Table 2, Figure 2);
• Contemporary slags from current production were collected directly from piles in storage yards. Each test sample was composed of 15 subsamples, which were combined, averaged and quartered (sites no. 16, 17, Table 2, Figure 3).

The samples were collected in the amount of 50 kg each and stored in plastic bags under standard conditions. For further testing, the samples were air-dried and ground to a grain size of under 0.2 mm.

Suprapur acids, deionised water of first-class purity and a certified reference material (CRM) in the form of tungsten ore (NCS DC 70009) were utilised in the studies.

2.2. Slag Composition Testing Methods

The following tests were performed for the collected samples:

-

Determination of ash content by weight: the samples were incinerated at a temperature of 815 °C in a laboratory furnace;

-

Determination of mineral composition by powder X-ray diffraction (XRD) in Bragg-Brentano geometry using a D8 DISCOVER diffractometer (Bruker, Billerica, MA, USA) with a CuKα lamp, a Ni filter and a LYNXEYE XE detector; the composition was calculated based on patterns licensed in PDF-4 + 2021 RDB ICDD (International Centre for Diffraction Data) and the ICSD (Inorganic Crystal Structure Database) and NIST (National Institute of Standards and Technology) databases; the following software was used for registration and diagnostics: DIFFRAC v.4.2 and TOPAS v.4.2. Bruker AXS; the quantitative phase composition was determined by the Rietveld method; the amorphous substance content was analysed using ZnO and Al₂O₃ as internal standards [25,26,27,28];

-

Determination of primary chemical component and trace element content by wavelength dispersive X-ray fluorescence spectroscopy (WDXRF) using a ZSX PRIMUS II analyser (Rigaku, Tokyo, Japan) equipped with a 4 kW X-ray Rh tube; powdered samples were pressed into a tablet under very high pressure; binding materials (cellulose and graphite) were added to improve the quality of the tablet;

-

Determination of selected element content by inductively coupled plasma optical emission spectroscopy (ICP-OES) or mass spectroscopy (ICP-MS) using an Elmer Optima 5300 analyser (PerkinElmer, Waltham, MA, USA) or NexION 300s (PerkinElmer, Waltham, MA, USA); the analysis was performed after mineralising the material in a mixture of acids prepared according to an established procedure;

-

Analysis of the grain surface morphology and chemical composition in micro-areas by scanning electron microscopy (SEM) and X-ray energy dispersion spectroscopy (EDS) using an SU3500 SEM microscope (Hitachi, Tokyo, Japan) coupled with an UltraDry EDS Detector (ThermoFisher Scientific, Waltham, MA, USA); the images were taken after spraying the sample with gold.

2.3. Slag Sample Dissolution Methods

Spectroscopic techniques such as OES (Optical Emission Spectroscopy) and MS (Mass Spectrometry) require a sample in liquid form; therefore, the separation of the determined analytes from solid samples is a key step in this technique.

The slag samples were dissolved in the following acid mixtures:

(i)

A total of 65% nitric acid (V), 38% hydrochloric acid, 40% hydrofluoric acid and 95% sulphuric acid (VI) (denoted as dissolution procedure A);

(ii)

A total of 50% sulphuric acid (VI), 40% hydrofluoric acid, 65% nitric acid (V), 30% hydrochloric acid (denoted as dissolution procedure B).

The mineralisate was filtered through Teflon filters with 0.45 µm pores or glass filters. The filters were weighed on an analytical balance with an accuracy of ±0.0001 mg. The filters were dried at 105 °C ± 5 °C in a laboratory dryer to a constant weight.

The slag dissolution efficiency, S_D, was calculated according to the following formula:

$$S_D={\displaystyle \frac{m1-m2}{m1}}\times 100, \%$$

(1)

m1—mass of the sample;

m2—mass of the residue after filtration.

The element recovery efficiency, S_R, was calculated according to the following formula:

$$S_R={\displaystyle \frac{c2}{c1}}\times 100, \%$$

(2)

c1—certified element concentration in the CRM;

c2—measured element concentration in the solution (after dissolution).

2.3.1. Dissolution Procedure A

Dissolution procedure A is a modification of the method developed by Kashiwakura et al. [29]. This procedure was used for the high-efficiency dissolution of ash samples from hard coal combustion processes [30].

An averaged and ground sample of slag with a mass of 1 g was placed in a vessel made of Teflon, and, afterwards, a mixture of 30 cm³ of 65% nitric acid and 30 cm³ of 38% hydrochloric acid was added. Everything was heated under cover in a sand bath at a temperature of 373 K for 15 min. After that, 30 cm³ of 40% hydrofluoric acid was added with careful stirring. The Teflon vessel with the reagents was heated under cover in a sand bath at 373 K for 15 min. The reaction mixture was then cooled to 353 K and heated for another 2 h. After cooling to ambient temperature, 30 cm³ of concentrated (95%) sulphuric acid (VI) was added and heated for 1.5 h at 473 K until the fluoride ions were completely removed from the solution. The obtained sample was diluted with distilled water to a volume of 200 cm³ in a class A volumetric flask.

2.3.2. Dissolution Procedure B

Dissolution procedure B is the GIG-PIB research procedure applied for sample dissolution [31]. An averaged and ground slag sample with a mass of 0.1 g was weighed into a platinum crucible with a capacity of approximately 30 mL and moistened with a few drops of deionised water, to which 2 mL of 50% sulphuric acid (VI) and 5 mL of 40% hydrofluoric acid were added. The crucible was placed under cover in a water bath for 30 min at a temperature of 368 K. Afterwards, 1 mL of 65% nitric acid (V) was added, and the crucible was placed on a heating plate and heated until the precipitate thickened and the white fumes evaporated. The final evaporation temperature was 573 K. In the subsequent step, 3 mL of 65% nitric acid (V) and 1 mL of 30% hydrochloric acid were added to digest the resulting precipitate. The process temperature was 453 K. The content of the crucible was kept at this temperature for 40 min. The obtained sample was diluted with distilled water to a volume of 100 cm³ in a class A volumetric flask.

Due to the phase and mineral composition of slags, characterised by the presence of silicates and amorphous phases, the ions embedded in their structure are difficult to extract. In order to determine the content of selected elements, it is necessary to use hydrofluoric acid to break down the silicate structures. Based on our own experience in the mineralisation of energy and mining waste, methods utilising concentrated mineral acids (H₂SO₄, HF, HNO₃, HCl) as leaching agents were employed. The application of concentrated sulfuric acid (VI) in Procedure A, during the final stage of removing fluoride ions from the post-reaction solution, prevented the complete dissolution of the samples due to the precipitation of poorly soluble sulfates, such as those of lead, barium and calcium. In Method B, which involves preliminary mineralisation of the samples using a mixture of hydrofluoric and sulfuric (VI) acids, followed by mineralisation of the remaining residue with a mixture of hydrochloric and nitric (V) acids, a post-reaction solution was obtained containing the ions of elements in the form of soluble nitrates and chlorides.

The test procedure presented in Figure 4 was used to assess the content of selected metals and critical raw materials in 14 slag samples.

2.4. Statistical Analysis

Data mining was performed by means of principal component analysis (PCA). The PCA method enables the preprocessing, interpretation and visualisation of experimental data. PCA decomposes the data matrix X into an object matrix S and a loading matrix D. The S matrix reveals the relationships between objects, while the D matrix reveals the relationships between parameters. The similarity of objects is defined by the Euclidean distance in the space of new variables, called principal components (PCs). The similarity of parameters is expressed through their correlation in the space defined by subsequent PCs. The defined PCs are orthogonal. This makes it possible to maximise the variance of the data on all subsequent PCs. PCA reveals similarities/differences between objects, thus enabling the identification of object classes. In this paper, the method was used to isolate slag samples with the most interesting chemical compositions. The analyses were performed using GNU Octave 9.2.0 [32].

3. Results

3.1. Chemical and Mineral Composition of Slag Samples

Table 4. Content of the primary components (XRF) and LOI in the tested slag samples.

Sample	SiO2	Al2O3	Fe2O3	CaO	MgO	Na2O	K2O	TiO2	P2O5	LOI
	wt%
H1	23.01	6.07	26.14	10.38	2.88	0.35	0.46	0.32	0.20	7.47
H2	43.74	11.54	14.42	7.29	2.57	0.49	1.37	0.56	0.23	2.58
S1	24.80	7.78	20.90	8.95	2.46	0.36	0.72	0.38	0.12	9.85
S2	41.75	14.68	13.94	5.97	2.49	0.37	1.50	0.69	0.09	4.44
G1	25.26	4.44	21.47	8.67	1.76	0.42	0.40	0.23	0.05	4.68
G2	21.10	5.18	18.80	10.30	2.56	0.38	0.50	0.25	0.05	15.94
W1	13.49	4.12	21.78	20.22	7.75	0.14	0.15	0.17	0.10	13.06
W2	14.45	2.75	17.02	25.71	7.79	0.18	0.02	0.14	0.10	12.45
O1	15.80	4.00	22.10	20.27	8.25	0.19	0.22	0.18	0.13	8.83
O2	20.80	4.60	22.56	19.76	8.07	0.26	0.26	0.21	0.13	6.54
B1	9.74	2.65	33.55	16.09	1.50	0.39	0.44	0.17	0.18	6.29
B2	8.76	2.50	45.63	19.95	3.16	0.24	0.06	0.19	0.31	6.22
M1	14.92	8.36	26.86	8.88	2.54	2.24	1.56	0.47	0.29	5.28
M2	14.55	8.22	26.88	8.73	1.99	1.49	1.39	0.47	0.33	5.46

Expanded uncertainty of ±10% for SiO₂, Al₂O₃, TiO₂, Fe₂O₃, K₂O, CaO, ±20% for MgO, P₂O₅, ±30% for Na₂O, coverage factor of 2 and significance level of 95%

and Figure 5 present the results of primary component content determination in slag samples by means of XRF.

Table 5. Phase compositions of the investigated historical slag samples (XRD).

Phase	H1	H2	S1	S2	G2	W1	O2
	wt%
Gypsum (CaSO4 × 2H2O)	68.0	56.0	69.0	26.0	38.0	77.0	46.0
Quartz (SiO2)	4.5	11.5	6.5	19.5	6.5	0.5	2.5
Hematite (Fe2O3)	4.0	2.0	3.0	2.0	1.0	3.0	6.0
Magnetite (FeFe2O4)	2.0	2.0	3.0	2.0	4.0	3.0	2.0
Calcite (CaCO3)	no	no	no	1.5	6.5	1.5	5.5
Tridymite (SiO2)	no	no	no	4.0	6.0	no	5.0
Augite (Ca, Mg, Fe, Ti, Al)2[Si2O6])	2.0	3.0	no	3.0	2.0	1.0	4.0
Plagioclase *	1.0	3.0	1.0	4.0	3.0	1.0	2.0
Anhydrite (CaSO4)	1.0	no	no	3.0	1.0	no	3.0
Dolomite (CaMg(CO3)2)	no	no	no	1.0	3.0	1.0	1.0
Ankerite (CaFe(CO3)2	no	no	no	no	2.0	no	3.0
Ferro-actinolite (Ca2Fe5[Si4O11(OH)]2)	no	no	no	4.0	no	no	no
Cristobalite (SiO2)	1.0	no	1.0	2.0	no	no	no
Mullite (3Al2O3 × 2SiO2)	no	no	no	2.0	2.0	no	no
Melanterite (FeSO4 × 7H2O)	1.0	1.0	<1.0	no	no	no	no
Illite (KAl2[AlSi3O10(OH)2])	no	no	no	1.0	no	no	no
Zincite (ZnO)	no	no	no	1.0	no	no	no
Bassanite (CaSO4 × 0.5H2O)	no	no	no	1.0	no	no	no
Sphalerite (ZnS)	no	no	no	<1.0	no	no	no
Anglesite (PbSO4)	no	no	no	<1.0	no	<1.0	no
Amorphous substance	14.5	19.5	15.5	20.5	24.5	10.5	18.5

* albite Na[AlSi₃O₈]—anorthite Ca[Al₂Si2O₈]; no—no identification in the sample.

and

Table 6. Phase compositions of the investigated contemporary slag samples (XRD).

Phase	B1	B2	M1	M2
	wt%
Spinel (MgAl2O4)	no	no	24.0	26.0
Wüstite (FeO)	5.0	21.0	2.0	2.0
Magnetite (FeFe2O4)	11.0	8.0	2.0	2.0
Quartz (SiO2)	9.5	9.5	1.5	1.5
Fluorite (CaF2)	8.5	<1.0	6.5	5.5
Augite (Ca, Mg, Fe, Ti, Al)2[Si2O6])	7.0	3.0	3.0	2.0
Gehlenite (Ca2Al[(Si, Al)2O7])	5.0	2.0	2.0	2.0
Wollastonite (CaSiO3)	6.0	2.0	2.0	no
Hardystonite (Ca2Zn[Si2O7])	6.0	3.0	no	no
Plagioclase *	3.0	3.0	3.0	3.0
Hematite (Fe2O3)	1.0	2.0	1.0	2.0
Na2Si2O5 substance	2.0	3.0	no	no
FeSiO3 substance	<1.0	4.0	1.0	no
Iron (Fe)	no	5.0	no	no
Melanterite (FeSO4 ×7H2O)	no	5.0	no	no
Maghemite (Fe2O3)	<1.0	2.0	no	no
Zincite (ZnO)	no	2.0	<1.0	no
Na2CaSi2O6 substance	<1.0	no	no	no
Amorphous substance	34.5	24.5	50.5	52.5

* albite Na[AlSi₃O₈]—anorthite Ca[Al₂Si₂O₈]; no—no identification in the sample.

and Figure 6 provide the mineral composition of the samples and examples of XRD spectra, respectively. Figure 7 displays example SEM images of selected grains of historical slags (sample H1) and slags from current production (sample M1).

3.1.1. Primary Component Content

The content of SiO₂ (Si content expressed as SiO₂) in the analysed samples ranges from 8.8 wt% (sample B2) to 43.7 wt% (sample H2). The average content of SiO₂ in historical and contemporary slags is 24.4 wt% and 12.0 wt%, respectively. Slags from the current production of the same manufacturer have similar SiO₂ content. For example, the content of SiO₂ in samples B1 and B2 amounts to 9.7 wt% and 8.8 wt%, respectively. Historical slags vary in their silicon content. Samples with coarse grains (grain size > 10 mm or cemented grains) are characterised by a higher silica content than samples with fine grains. For example, the SiO₂ content in samples H1 (grain size < 10 mm) and H2 (grain size > 20 mm) reaches 23.0 wt% and 43.7 wt%, respectively. The largest differences in SiO₂ content are observed between samples collected from the dump located in Ruda Śląska (samples H1 and H2) and those collected from the dump in Świętochłowice (samples S1 and S2).

The Al₂O₃ content (with Al expressed as Al₂O₃) in the analysed samples ranges from 2.5 wt% (sample B2) to 14.7 wt% (sample S2). The average Al₂O₃ content in historical and contemporary slags is similar, with 6.5 wt% and 5.4 wt%, respectively. Contemporary slags from the same manufacturer present comparable Al₂O₃ content; for example, the Al₂O₃ content in samples B1 and B2 is 2.6 wt% and 2.5 wt%, respectively. Historical samples exhibit variability in the aluminium content. Coarse-grained samples (grain size > 20 mm) collected from the dump in Ruda Śląska (sample H2) and from the dump in Świętochłowice (sample S2) have considerably higher aluminium content than the fine-grained samples from these dumps. For instance, the Al₂O₃ content in samples H1 (grain size < 10 mm) and H2 (grain size > 20 mm) is 6.1 wt% and 11.5 wt%, respectively. The Al₂O₃ content in the remaining historical slags (G1, G2, W1, W2, O1, O2) does not exceed 5.2 wt%.

The content of CaO (Ca content expressed as CaO) in the analysed samples ranges from 6.0 wt% (sample S2) to 25.7 wt% (sample W2). The average CaO content in historical and currently produced slags is at the same level and amounts to 13.8 wt% and 13.4 wt%, respectively. The samples differ in CaO content depending on the sampling site (waste dump, manufacturer). For example, the average CaO content in samples collected from the dump in Świętochłowice (samples S1 and S2) and the dump in Piekary Śląskie (samples W1 and W2) is 7.5 wt% and 23.0 wt%, respectively.

The highest content of MgO (Mg content expressed as MgO) is found in samples collected from the dumps in Piekary Śląskie (samples W1, W2, O1, O2), where the average MgO content reaches 8.0 wt%. In the remaining samples (H1, H2, S1, S2, G1, G2, and all samples of contemporary slags), the MgO content does not exceed 3.2 wt%.

The content of Fe₂O₃ (Fe content expressed as Fe₂O₃) in the analysed samples ranges from 13.9 wt% (sample S2) to 45.6 wt% (sample B2). The average Fe₂O₃ content in historical and currently produced slags is 19.9 wt% and 33.2 wt%, respectively. In the case of historical slags, a relationship is generally observed between the iron content and the grain size. For example, the Fe₂O₃ content in sample H1 (grain size < 10 mm) and sample H2 (grain size > 20 mm) are 26.1 wt% and 14.4 wt%, respectively. Slags from current production are characterised by a higher iron content than historical slags. The highest content is found in contemporary slags from ZGH Bolesław S.A., as the Fe₂O₃ content in samples B1 and B2 reaches 33.6 wt% and 45.6 wt%, respectively. These samples are also characterised by the lowest silicon and aluminium content among the analysed samples.

3.1.2. Mineral Composition of Historical Slags

The historical and contemporary slags differ in terms of mineral composition.

The primary mineral component of the historical slags is gypsum, with its content ranging from 26.0 wt% (sample S2) to 77.0 wt% (sample W1). Quartz content ranges from 0.5 wt% (sample W1) to 19.5 wt% (sample S2). Silicon oxide is also present in some samples, in forms such as tridymite (max. 6 wt%—sample G2) or cristobalite (max. 2.0 wt%—sample S2). The amorphous substance content ranges from 10.5 wt% (sample W1) to 24.5 wt% (sample G2). Hematite content is 6 wt% in sample O2, whereas, in other samples, it ranges from 1.0 wt% (sample G2) to 4.0 wt% (sample H1). In samples G2 and O2, calcite is present at 6.5 wt% and 5.5 wt%, respectively, while, in the other samples, the calcite content is at 1.5 wt% (samples S2, W1) or is not found at all. The mineral composition of the historical slag samples is complemented by minerals such as magnetite, augite, plagioclase, anhydrite, dolomite, ankerite, ferro-actinolite, mullite, melanterite, illite, zincite, bassanite, sphalerite and anglesite. These minerals are present in the analysed samples in amounts ranging from <1.0 wt% to 4.0 wt%.

Four groups of crystalline phases can be distinguished in the historical slags:

• The first group consists of sulphates in the system: gypsum–anhydrite–bassanite–melanterite–anglesite. Gypsum is present in all the analysed samples, while the other sulphates are found only in some samples (e.g., bassanite was identified only in sample S2). The gypsum in historical slags is likely a product of the reaction between sulphates and dolomite remnants in an aqueous medium:

  CaMg(CO₃)₂ + 2SO₄²⁻ + 4H₂O → CaSO₄ × 2H₂O + MgSO₄ +2CO₂ + 4(OH)⁻
  This reaction produces gypsum and magnesium sulphate, which does not retain a crystalline form under slight environment alkalisation [33].

2.

The second group of mineral phases consists of oxides in the system: quartz–tridymite–cristobalite–mullite–hematite–magnetite–zincite. Among these minerals, quartz, hematite and magnetite are present in all the samples, whereas zincite is only found in sample S2.

3.

The third group of minerals consists of silicates in the system: anorthite + albite (present in all samples) − augite (present in samples H1, H2, S2, G2, W1, O2) − ferro-actinolite (present only in sample S2) − illite (present only in sample S2).

4.

The fourth group of minerals consists of carbonates, represented by calcite and dolomite.

The mineral composition of historical slags indicates that the material originated as a result of similar thermal processes, as the contents of the primary components (gypsum, quartz and amorphous substance) exhibit significant correlations. The Pearson correlation coefficient between the gypsum content and the content of crystalline SiO₂ (quartz + tridymite + cristobalite) is −0.78, while the Pearson correlation coefficient between the gypsum content and amorphous substance content is −0.73 (n = 7).

3.1.3. Mineral Composition of Contemporary Slags

For contemporary slags, the primary mineral component is the MgAl₂O₄ spinel (samples M1, M2), wüstite or magnetite (samples B1, B2). The amorphous substance content ranges from 24.5 wt% (sample B2) to 52.5 wt% (sample M2). The amorphous substance is predominant in slags from HC Miasteczko Śląskie S.A. (samples M1 and M2). In samples from ZGH Bolesław S.A. (samples B1, B2), the amorphous substance is at nearly half that level.

Three groups of crystalline phases can be distinguished in the contemporary slags:

• The first mineral system, including spinels, iron minerals and oxides of other metals, is as follows: MgAl₂O₄ (magnesium-aluminate spinel)–magnetite (iron spinel)–wüstite–iron–hematite–maghemite–melanterite–zincite. The highest contents of phases belonging to the first group are found in sample B2 (45.0 wt%). The contents in samples B1, M1 and M2 are 18.0 wt%, 30.0 wt% and 32.0 wt%, respectively. In samples M1 and M2, spinels (magnesium-aluminate and iron) account for 26.0 wt% and 28.0 wt%, respectively, while the sum of other minerals from this group is only 4.0 wt%. Spinels are the main crystalline phase in slags produced by HC Miasteczko Śląskie S.A., with over 50 wt% of amorphous material. An interesting feature is the presence of 5.0% of metallic iron (Fe) in sample B2. Such an amount of metallic iron, with 21.0 wt% of wüstite (FeO), is likely due to limited oxygen access or excess CO₂ in the thermal process.
• The second group consists of quartz-silicate minerals, including quartz–augite–wollastonite–hardystonite–gehlenite–plagioclase. Samples B1 and B2 (ZGH Bolesław S.A) contain 36.5 wt% and 22.5 wt% of minerals from this group, respectively. Samples M1 and M2 (HC Miasteczko Śląskie S.A.) contain 11.5 wt% and 8.5 wt% of these minerals, respectively. Among these minerals, quartz, augite, gehlenite and plagioclase are present in all the samples, while hardystonite is only found in samples from ZGH Bolesław S.A.
• The third system consists only of fluorite. Fluorite is present in all the analysed contemporary slag samples, ranging from under 1.0 wt% (sample B2) to 8.5 wt% (sample B1). Its presence in the slags is likely due to the use of fluorine-containing fluxes in the thermal processes [34].

3.2. Selecting the Method for Slag Sample Dissolution in Acid Mixtures

Two slag samples from different categories, H1 (historical slag) and M1 (contemporary slag), were dissolved using the procedures described in Section 2.3.

Table 7. Slag dissolution efficiency assessment.

Sample	Slag Dissolution Efficiency, SR
	Procedure A *	Procedure B *
	%
H1	84.4	99.4
M1	83.2	99.0

* procedures as described in Section 2.3.

presents the dissolution efficiency. The dissolution process was carried out according to the two procedures. In each of them, the solvent mixture contained hydrofluoric acid, which is necessary due to the high silicon content (maximum SiO₂ content is 43.7 wt%), especially in historical slag. Procedure B is effective for the slag sample dissolution, as it ensures the transfer of 99% of the sample mass to the solution.

3.3. Efficiency Assessment of Selected Element Recovery from Slags

3.3.1. Rare Earth Elements (REEs)

The recovery efficiency of rare earth elements from slags was assessed with reference to a certified reference material (CRM) in the form of tungsten ore. The CRM was dissolved according to procedure B (Section 2.3), and the content of individual REEs in the solution was determined by MS-OES.

Table 8. Efficiency assessment of rare earth element recovery from the CRM.

Element	REE Content			Recovery Efficiency, SR
	Certified (In CRM)	Detection Limit	Measured (In Solution)
	ppm	ppm	ppm	%
La	23.7 ± 0.2	<0.01	23.1 ± 4.6	97.5
Ce	60.3 ±1.8	<0.01	67.1 ± 13.4	111.3
Pr	7.9 ± 0.6	<0.01	7.9 ± 1.6	100.0
Nd	32.9 ± 2.2	<0.01	32.0 ± 6.4	97.3
Sm	12.5 ± 0.8	<0.01	12.5 ± 2.5	100.0
Eu	0.16 ± 0.03	<0.01	0.15 ± 0.03	93.8
Gd	14.8 ± 0.3	<0.01	15.3 ± 3.1	103.4
Tb	3.3 ± 0.2	<0.01	3.2 ± 0.6	97.0
Dy	20.7 ± 1.1	<0.01	21.7 ± 4.3	104.8
Ho	4.5 ± 0.2	<0.01	4.6 ± 0.9	102.2
Er	13.1 ± 0.9	<0.01	14.4 ± 2.9	109.9
Tm	2.2 ± 0.2	<0.01	2.3 ± 0.5	104.5
Yb	14.9 ± 1.0	<0.01	16.8 ± 3.4	112.8
Lu	2.4 ± 0.2	<0.01	2.5 ± 0.5	104.2
Sc	5.4 ± 0.5	<5	7.3 ± 1.5	135.2
Y	128 ± 25	<0.01	141 ± 28	110.2

presents the efficiency of rare earth element recovery from the CRM. The dissolution procedure applied is effective in the recovery of lanthanide elements from the matrix. The recovery efficiency ranges from 95% to 113%. An efficiency of over 120% is observed in the case of scandium, which may be related to the low content of the element in the tested CRM as well as equipment limitations, such as the detection limit. The absolute detection error is below 2 ppm, so the efficiency level is acceptable.

3.3.2. Lead, Zinc and Selected Critical Elements

The content of selected elements in the slag samples was examined by means of XRF (solid sample) as well as by ICP-OES or MS-OES (dissolved sample). The results were compared with each other.

Table 9. Pearson correlation coefficients between the results obtained by XRF and ICP-OES.

Element	Pearson Correlation Coefficients, R2	Element	Pearson Correlation Coefficients, R2
Pb	0.86	Mn	0.99
Zn	0.87	Ni	0.95
Ti	0.97	Sb	0.73
As	0.94	V	0.77
Cu	0.98	Co	0.78

presents the Pearson correlation coefficients between the results obtained by these two methods.

The conducted tests indicate that dissolution procedure B ensures an acceptable recovery level of REE (minimum recovery efficiency reaches 95%) and of the following elements: zinc, lead, titanium, arsenic, copper, manganese, nickel, antimony, tungsten and cobalt (correlation coefficient above 0.7) from slag samples.

3.3.3. Other Critical Raw Materials

The content of other critical raw materials such as lithium, germanium, bismuth, strontium, tantalum and tungsten was also determined in the analysed slags (dissolved sample). However, these results were not validated by comparison with other methods, and, thus, they may be subject to errors arising from the sample dissolution process.

Table 10. Minimum and maximum contents of selected elements in the analysed slag samples.

Element	Content
	Min.	Max.
	ppm
Li	27 ± 6	427 ± 85
Sr	125 ± 25	622 ± 124
Bi	0.0	160 ± 32
W	1.4 ± 0.3	61 ± 12
Ge	3.5 ± 0.7	51 ± 10
Ta	0.3 ± 0.1	1.1 ± 0.2

presents the minimum and maximum contents of the above elements measured in the analysed samples. Only lithium and strontium were identified in quantities above 200 ppm. The maximum detected content of germanium in the analysed samples is 51 ppm. The actual content of this critical raw material in the slags may be higher, as germanium forms volatile compounds such as GeCl₄, GeBr₄ and GeF₄, which might also be partially evaporated during the sample dissolution process [35].

4. Discussion of the Potential to Use Polish Slags from Pb-Zn Metallurgy as a Source of Selected Metals and Critical Raw Materials

Figure 8 presents the contents of zinc, lead, summed rare earth metals and other selected critical raw materials, such as titanium, arsenic, copper, manganese, nickel, antimony and vanadium, in 14 analysed slag samples.

Table 11. Statistical analysis of the selected element contents in analysed samples of historical and contemporary slags.

Element	Historical Slags			Contemporary Slags
	Content
	Minimum	Maximum	Average	Minimum	Maximum	Average
	ppm
Zn	8643	60,676	35,195	7174	51,626	35,550
Pb	1524	25,663	8310	520	17,631	10,056
REEs	60	251	124	79	272	212
Ti	1093	4253	2141	1085	3005	2009
As	129	2318	1359	106	2268	1302
Co	117	286	206	518	637	578
Mn	1609	5991	3648	22,615	53,590	34,848
Cu	167	2138	1069	4213	13,572	9097
Ni	63	117	99	1611	3498	2715
Sb	8	164	60	276	375	321
V	29	161	68	79	241	163

Historical slags (n = 10), Contemporary slags (n = 4).

summarises the minimum, maximum and average contents of these elements in the analysed samples, divided into historical slags and slags from current production.

The average zinc content in the analysed historical slags and contemporary slags is similar, at 3.5 wt%. The average lead content in historical slags and slags from current production is 0.83 wt% and 1.00 wt%, respectively. The average titanium and arsenic contents in deposited and contemporary slags are also at the same level, i.e., 0.21 wt% (Ti) and 0.13 wt% (As), respectively. For historical slags collected from a dump located in Ruda Śląska (samples H1, H2) and Świętochłowice (samples S1 and S2), there is a noticeable correlation between the grain size and the content of zinc, lead, titanium and arsenic. For instance, in samples H1 (grain size < 10 mm) and H2 (grain size > 20 mm), the zinc contents are 5.09 wt% and 3.11 wt%, respectively; the lead contents are 0.84 wt% and 0.55 wt%, respectively; the titanium contents are 0.21 wt% and 0.37 wt%, respectively; and the arsenic contents are 0.23 wt% and 0.03 wt%, respectively.

Contemporary slags contain higher quantities of critical raw materials such as cobalt, nickel, copper and manganese compared to historical ones. The average contents of these metals in the analysed slags from current production are 578 ppm (Co), 2715 ppm (Ni), 9097 ppm (Cu) and 34,848 ppm (Mn). In contrast, the average contents of cobalt, nickel, copper and manganese in the analysed historical slags are 206 ppm (Co), 99 ppm (Ni), 1069 ppm (Cu) and 3648 ppm (Mn). On average, the analysed contemporary slags contain approximately 3 times more cobalt, 27 times more nickel, 9 times more copper and 10 times more manganese than historical ones. These differences are likely due to the type of metallurgical charge used as well as the process conditions. The physicochemical processes that occurred during the years of historical slag storage, such as oxidation, dissolution of compounds contained in the slags and their transfer to the water–soil environment, strongly shaped the slags’ current composition [36,37].

The content of rare earth elements (REEs) varies in the analysed samples, ranging from 60 ppm (sample W1) to 272 ppm (sample B1). The REE content exceeds 200 ppm in three of the four analysed contemporary slag samples. In the analysed historical slags, the average rare earth metal content is 124 ppm, with only one sample (sample S2) containing more than 200 ppm of REEs. Cerium, lanthanum, neodymium, yttrium and scandium are the primary rare earth elements found in the analysed samples. Figure 9 and Figure 10 present their proportions in the total rare earth metal content in the analysed samples. In historical slags, cerium constitutes from 28% to 31% of the total rare earth element mass. The second most common REE in historical slags is yttrium, which comprises from 14% to 17% of the total rare earth element mass. The sum of cerium, lanthanum, neodymium, yttrium and scandium in historical slags constitutes from 82% to 83% of the REE mass. In contemporary slags, lanthanium constitutes from 23% to 45% of the total rare earth element mass, while cerium accounts for from 25% to 34% of the total REE mass. Lanthanum is the dominant element in samples B1 and B2, whereas cerium is dominant in samples M1 and M2. The sum of cerium, lanthanum, neodymium, yttrium and scandium in contemporary slags constitutes from 86% to 89% of the REE mass.

The levels of vanadium and antimony content in the analysed samples are similar to the content of REEs. The vanadium content ranges from 29 ppm (sample W2) to 241 ppm (sample M2). Only one analysed sample has a vanadium content of over 200 ppm (sample M2). The antimony content ranges from 8 ppm (sample S1) to 375 ppm (sample B1). The antimony content exceeds 200 ppm only in contemporary slags.

Table 12. Potential for using Polish slags from Pb-Zn metallurgy as a source of selected metals and critical raw materials.

Sample	Content
	Zn + Pb	REE	Ti + As + Mn + Cu	Co + Ni + Sb + V
	wt%	ppm	wt%	ppm
H1	5.93	126	0.82	496
H2	3.67	194	0.80	365
S1	6.58	174	0.66	427
S2	4.17	251	0.67	402
G1	8.63	104	0.67	472
G2	5.64	100	0.64	354
W1	1.22	60	0.93	370
W2	3.36	60	0.78	278
O1	2.01	80	1.12	587
O2	2.29	88	1.14	585
B1	4.95	272	6.93	2702
B2	0.77	79	4.55	4180
M1	5.96	258	3.87	4482
M2	6.56	238	3.55	3740

summarises the analysis of the potential for using Polish Pb-Zn metallurgy slags as a resource, classifying the raw materials contained in the slags into four categories: the sum of zinc and lead content (non-critical metals), the sum of rare earth metals (critical raw materials with a high supply risk index) and the sum of Ti, As, Mn and Cu (critical raw materials with an average content of over 1000 ppm in historical slags), as well as the sum of Co, Ni, Sb and V (critical raw materials with an average content of under 1000 ppm in historical slags).

Table 13. PCA parameters (19 parameters).

Item	Parameter	Unit	Item	Parameter	Unit
1	Si content	ppm	10	Pb content	ppm
2	Al content	ppm	11	Zn content	ppm
3	Fe content	ppm	12	Ti content	ppm
4	Ca content	ppm	13	As content	ppm
5	Mg content	ppm	14	Cu content	ppm
6	Na content	ppm	15	Mn content	ppm
7	K content	ppm	16	Ni content	ppm
8	S content	ppm	17	Sb content	ppm
9	P content	ppm	18	V content	ppm
			19	REE content	ppm

presents the PCA parameters used for the analysis. The results of the PCA analysis are displayed in Figure 11. The first three main factors of the autoscaled data described 83.45% of the total data variance, while the first two PCs described over 70%. The object and parameter projections onto the planes determined by PC-1 and PC-2 indicate two main groups: historical slags (marked by red stars) and contemporary slags (marked by blue squares). The object projections onto the plane defined by PC-1 and PC-2 (score plot) reveal a similarity between samples 13 and 14. Analysing the parameters projected onto the plane determined by the same PCs (loading plot) indicates that samples 13 and 14 are rich in REEs (parameter 19). Sample 13 (M1) was recommended for the next step of the research. Samples 11 (B1) and 12 (B2) exhibit similarities as well. Sample 1 (H1) was also selected as a historical sample representative of typical physicochemical properties.

Slags from Pb-Zn metallurgy show potential as a source of valuable elements. The feasibility of recovering selected metals from this waste using both physical and chemical methods will be discussed in the second part of this paper [38]. One sample of historical slag (H1) and two samples of contemporary slag (B1 and M1) were selected for further analysis.

5. Conclusions

Fourteen slag samples were collected and analysed for their chemical compositions. Ten samples associated with the smelting of polymetallic Zn–Pb ores from MVT deposits were collected from dumps in the Silesia–Cracow Upland. Additionally, four samples of contemporary slags from this region were collected as well. However, the origin of the contemporary slags is not related to natural metal ore smelting. The charge for the contemporary smelters consists of industrial waste rich in zinc and lead, produced as part of various technological processes.

Detailed physicochemical and mineral analyses of the collected slag samples were conducted, leading to the following conclusions:

• Mineral composition: The primary mineral component of the historical slags is gypsum. Contemporary slags consist primarily of spinels (magnesium-aluminate and ferric). The gypsum in historical slags is likely a product of sulphate reactions with residual dolomite in an aqueous medium.
• Iron content: Contemporary slags exhibit a higher iron content compared to historical slags. The maximum iron content (expressed as Fe₂O₃) in the analysed historical slags and slags from current production is 26.1 wt% and 46.6 wt%, respectively. Various iron minerals, such as wüstite, magnetite, hematite, melanterite and maghemite, as well as metallic iron, were identified in the contemporary slags. The simultaneous presence of metallic iron (Fe) and wüstite (FeO) is likely a result of limited oxygen access or excess CO₂ during the thermal processing in the furnace.
• Zinc content: The average zinc content in the historical and contemporary slags is similar, at 3.5 wt%.
• Critical raw materials: Contemporary slags contain higher amounts of critical raw materials such as cobalt, nickel, copper and manganese compared to historical slags. The average contents of cobalt, nickel, copper and manganese in the analysed contemporary slags are 578 ppm, 2715 ppm, 9097 ppm and 34,848 ppm, respectively. In contrast, the average contents of cobalt, nickel, copper and manganese in the analysed historical slags are 206 ppm, 99 ppm, 1069 ppm and 3648 ppm, respectively. It is likely that the physicochemical processes that occurred during the years of historical slag storage, such as oxidation, dissolution of compounds contained in the slags and their transfer to the water–soil environment, strongly shaped the current composition of the historical slag.
  The content of rare earth elements (REEs) varies in the analysed samples, ranging from 60 ppm to 272 ppm. The REE content in the majority of the contemporary slag samples exceeds 200 ppm. In the analysed historical samples, the average rare earth metal content is 124 ppm, with only one sample containing over 200 ppm of REEs.
• Potential for raw material recovery: The slags from Zn + Pb metallurgy in the Silesia–Cracow Upland constitute a promising material for recovering selected metals (e.g., Fe, Zn) and critical raw materials (e.g., REEs). The presented findings can add to the European knowledge base regarding the content of CRMs in Zn–Pb metallurgical waste that can potentially be recovered. This, however, is the subject of the second part of this study.
• Prospects for germanium recovery: The presence of germanium was detected in the analysed slag samples, with a maximum concentration of 51 ppm. The content of this critical raw material in the slags may be higher, as germanium forms volatile compounds that may be lost during sample dissolution. The accurate determination of germanium content in slags requires the application of alternative sample dissolution procedures, which will be the subject of further analysis.