The Influence of Lead Refining Method on the Dross Content in the Metal

TECHNICAL ARTICLE The Influence of Lead Refining Method on the Dross Content in the Metal DANIEL MALECHA , 1,2,5 ROBERT ALBRECHT , 3,4 JAMES LAMB , 4 and STANISŁAW MAŁECKI 1 1.—Faculty of Non-Ferrous Metals, AGH University of Science and Technology, al. Mickiewicza 30, 30-059 Krakow, Poland. 2.—Baterpol SA, ul. Obr. Westerplatte 108, 40-335 Katowice, Poland. 3.—Institute of Materials Engineering, Faculty of Science and Technology, University of Silesia, 75 Pułku Piechoty 1a, 41-500 Chorzow, Poland. 4.—Materials Department, University of California, Santa Barbara, CA 93106, USA. 5.—e-mail: [email protected] Impurities in lead alloys affect the production yield and grid quality of leadacid batteries. In addition to contamination limits, battery manufacturers often specify the maximum level of dross that can be produced when melting a given alloy. This study presents, for the first time, an example of a method for determining the percentage of dross formed after melting an alloy. In addition, the effect of three different lead pyrorefining methods—traditional refining, traditional refining with the double addition of NaOH and NaNO 3 and an alternative method using metallic aluminum—on the percentage of dross produced during the melting of the PbSnCa alloy was evaluated. Industrialscale experiments have revealed significant differences in the amount of melt dross formed, confirming the influence of the refining method on this parameter. The aluminum refining method gave a lower dross content than the traditional approach and showed the highest process stability. Microstructural and phase analysis indicated that the resulting dross consisted mainly of metallic lead mixed with oxide phases. The results highlight the potential of lead refining using aluminum and provide valuable insights into optimizing lead recycling practices, as well as being a valuable knowledge base for lead-acid battery manufacturers. INTRODUCTION Lead acid battery (LAB) production is responsible for >80% of global lead consumption. LAB recycling is one of the best examples of implementing a closed-loop economy as the average recycling efficiency of these batteries in the EU is >80%, while in the US it is even >90%. 1–5 Recycled lead has a much less negative environmental impact and is largely what new LABs are made from. 6 Despite the increasing popularity of lithium-ion batteries, LABs remain the main type used in the automotive sector, emergency power systems and energy storage. Their popularity is due to the low cost of production, the high reliability of these batteries and their compliance with fire and building safety requirements. 7,8 In the waste lead acid battery (WLAB) recycling industry, the predominant method is to grind and separate the battery components into individual fractions such as battery paste, metallic fractions, electrolyte, spacers, polypropylene from casings and steel components. 9–11 Lead from the metallic fraction and battery paste is usually melted in tilting rotary furnaces with an oxy-gas burner and then refined to meet the technological requirements for new batteries. 12–16 The use of tilting rotary furnaces and short rotary kilns increased the efficiency of lead production from secondary raw materials but at the same time resulted in greater penetration of unwanted impurities into the base metal. Liquid lead has a high potential for dissolving many elements, further (Received March 5, 2025; accepted June 5, 2025; published online July 3, 2025) JOM, Vol. 77, No. 9, 2025 https://doi.org/10.1007/s11837-025-07558-x Ó2025 The Author(s) 6603 complicating the purification process. There are many methods for removing impurities from lead, some of which have been used for >5 centuries. Their principles remain unchanged and are based on the use of an agent that acts selectively on the impurity, the formation of a chemical or intermetallic compound between the impurity and the refining agent, the crystallization of the resulting compound at reduced temperature and its removal from the surface of the metal bath. If the scrap batteries are properly processed, the lead refining process can be limited to a few steps. In contrast, when the input raw materials do not come from WLAB processing, the metal purification process usually requires a much larger number of steps. 17 Secondary WLAB recycled lead contains elements that can harm battery performance, particularly through increased gassing (hydrogen and oxygen production). These are mainly copper, nickel, sulfur, tellurium and zinc. 18 In addition to these impurities, elements that are alloying additives, such as tin, antimony, arsenic, selenium, calcium and aluminum, are also present in lead. To produce alloys of the desired specification, they can either be removed in the refining process or used. 19,20 The most commonly used technique for the purification of lead in industry is fire refining, which allows the selective removal of impurities. Refining processes can be carried out intermittently or continuously in melting furnaces, typically holding between 20 and 300 Mg of lead. Key refining steps include decoppering using granulated sulfur to remove copper and oxidation to remove tin, antimony, arsenic, etc. 18,21–23 Other methods of lead refining are hydrometallurgical and electrochemical processes. Lead hydrometallurgy, which is a more complex process than fire refining, allows high-purity lead to be produced with minimal waste. There are also several acid leaching and lead electrolysis processes but these methods are little used on an industrial scale because of higher costs and technological limitations. 24 The lead recovered from the WLAB is used to produce pure refined lead of the type Pb 970R 25 or lead alloys containing mainly tin, antimony and calcium. Pure lead is intended for the production of the active material of electrodes in LAB, while lead alloys are intended for the production of battery grids and connectors (bridges) used in LAB. 19,20,23,26 The grids used to make the plates at LAB can be produced using various methods. The most traditional and increasingly rare approach is gravity casting. In this method, heated lead alloy is poured into a casting mold, creating grids with thicker sections and higher mechanical strength compared to those produced by other methods. This allows for increased resistance to deep discharge. Gravity casting provides stability and resistance to cyclic loading, which is important for LABs operated in harsh environments. Another method of grid production is continuous casting. The most popular system using this method is the Wirtz Concast TM or Conroll TM . In the Wirtz process, molten lead is introduced through nozzles onto a moving mold, i.e., a rotating drum equipped with precisely shaped cavities mimicking the grid structure. On contact with the drum, the alloy solidifies to form a continuous mesh grid. Production efficiency is significantly higher than traditional gravity casting methods. 27,28 A third way of producing grids is the method known as ‘‘expanded metal.’’ This method involves cutting notches in a strip of lead alloy, which is then stretched and formed to produce a grid of the correct size and thickness. The grids produced by this method can have a thickness of<1 mm, allowing a significant increase in the number of plates in the battery and consequently increasing its starting power. 29 A final method currently gaining popularity is ‘‘punching’’ (the name ‘‘cold stamping’’ is also used). It consists of continuous casting and rolling, followed by extrusion of a lead strip grating using special molds. It is a high-performance method that enables the production of gratings with high corrosion and mechanical strength with minimal raw material consumption. The automated processes in this method reduce costs and increase the dimensional precision and mechanical properties of the gratings. During the production of gratings by the abovementioned methods, dross is formed at the melting stage, limiting production efficiency, increasing process costs and potentially reducing the quality of the lead alloy used and thus the battery grating itself. In the continuous casting process used in Wirtz Concast TM /Conroll TM and punching/coldstamping systems, the dross that enters with the lead gradually clogs the nozzles, causing production line stops, increasing production costs. In contrast, in the ‘‘expanded-metal’’ method, non-metallic inclusions, present on the cast lead strip when it is stretched, cause a break in the continuity of the grid, which is recognized as waste, generating higher production costs. The difficulties described make the issue of minimizing oxides and impurities in lead a key one for battery manufacturers. 28,30 In the nonferrous metals industry, the best described dross collection methods are in the context of aluminum metallurgy. The dross from the surface of the melt is collected using specially designed samplers. These samplers resemble the preferred cups with holes large enough to allow the molten Al to drain freely to capture the dross. 31,32 Alongside the chemical composition guidelines, battery manufacturers often specify the maximum amount of dross that can be produced when melting lead alloy. Individual LAB manufacturers have their own procedures for determining dross content. Typically, this is the ratio of the mass of dross produced by melting lead to the total mass of lead Malecha, Albrecht, Lamb, and Małecki6604 melted. Despite the prevalence of this practice, to the best of the authors’ knowledge, this aspect has been completely neglected by the available literature. Also, the issue of the influence of the method of refining lead on the amount of dross produced during melting has so far not been addressed in the literature. For this reason, the aim of this work and its innovative aspect is to describe an exemplary methodology for analyzing the amount of dross formed during metal melting and to analyze the influence of the lead refining method on this parameter. The influence of the choice of methodology and storage time of the lead alloy on the amount of dross produced during melting is also presented. These studies were carried out on an industrial scale for three different variants of lead fire refining, realized during the production of the very popular lead alloy Pb-Sn-Ca (sometimes called: PbCaSn/Pb-Ca-Sn) for positive battery collectors (grids). 20,33 The final aim of the work was to analyze the resulting dross in terms of chemical composition, phase composition and microstructure. It is hoped that the research presented will contribute to the development of lead refining techniques and be helpful to LAB producers. MATERIALS AND METHODS Materials The base material on which the tests were carried out was PbSnCa lead alloy ingot intended to produce positive LAB gratings (collectors) containing approximately 0.06 wt.% of Ca and approximately 1.5 wt.% Sn. 20,33 They were produced by three different fire refining methods. The impurity content of this alloy had to meet the requirements for pure refined lead Pb 970R 25 (excluding Sn and Ca). The material used to produce the PbCaSn alloy was WLAB recycled lead. Its chemical composition is within the range shown in Table I. The production process of the PbCaSn alloy took place entirely at the LAB recycling plant located in the European Union. Caustic soda (NaOH) in flake form, sodium nitrate (NaNO 3 ) in powder form, calcium hydroxide (Ca(OH) 2 ) in powder form and sulfur (S) in granular form were used for refining on an industrial scale. All chemical reagents used during refining had >98 wt.% purity as declared by the manufacturer. Aluminum scrap with a purity of >95 wt.% was used for alternative lead refining. After the refining process, alloying additives in the form of tin with a purity of >99.85 wt.% and calcium with a purity of >98 wt.% as declared by the manufacturers were added to the lead in each case. Methods The coefficient of dross volume formed during the melting of PbSnCa alloy was studied for three different lead pyrorefining methods used during industrial-scale production. The lead refining division where the alloy was produced was equipped with refining furnaces (refining kettles), which consisted of a steel bowl, made of 40-mm-thick sheet metal, 3 m in diameter and 1.9 m high, with a capacity of approximately 11.9 m 3 , corresponding to approximately 125 Mg of lead (real maximum working capacity of approximately 110 Mg lead). The refining kettle sat on a cylinder-shaped furnace with a refractory brick lining. The kettles were fired with high-methane natural gas with a calorific value of 8200 value of 8200 kilocalories/normal cubic meter. An illustrative drawing of the refining furnace is shown in Fig. 1a. During operation, the kettles were covered with hoods connected to an extraction system integrated into the dust extraction and filtration system. These hoods were also equipped with agitators connected to 45-kW electric motors with variable speed control. The temperature of the lead in the kettle (melting furnace) was measured with two independent thermoelectric head sensor systems. Each system was equipped with a K-type (NiCr-NiAl) CTK-2000 temperature sensor. In all three refining methods, alloying additives were added to the lead in the required amounts, and then the finished PbSnCa alloy was cast on a casting machine. The ingots for testing were taken immediately after the casting process. For each of the three specified refining methods, 20 full-production cycles of PbSnCa alloy were carried out, so a total of 60 cycles were carried out on an industrial scale over a 3-year period, allowing reliable test data to be collected. The melt dross factor (Dross % in formula 1) of the PbSnCa alloy was determined in each case according to the procedure described below and shown in Fig. 1b. Two ingots were set aside from the cast lead alloy, with a piece weighing between 8 and 9 kg taken from the center of each. They were then weighed on a balance (Axis B15) with a reading accuracy of 5 g (coefficient m i in formula 1). The weighed sample was melted on a bench with a hemispherical steel bowl (internal diameter 300 mm, internal depth 75 mm, made of 8-mmTable I. Chemical composition range of secondary lead from WLAB Sb, % Sn, wt.% As, wt.% Cu, wt.% Bi, ppm Se, ppm Ag, ppm Ni, ppm Zn, ppm Cd, ppm Te, ppm 0.6–1.8 0.1–1.1 0.01–0.2 0.01–0.2 100–300 1–30 20–50 1–90 1–500 1–150 1–40 The Influence of Lead Refining Method on the Dross Content in the Metal 6605 thick sheet metal and working volume 1.2 dm 3 ) heated by a set of gas burners. After melting the sample, the temperature of the liquid metal was checked using a one-channel thermometer (1311A; TES) until it reached 480 ±5°C, and then this temperature was maintained for 10 min. After this time had elapsed, the steel punched spoon was heated to >500°C, and the dross formed on the surface of the liquid lead was collected using a steel-punched spoon with a 60mm diameter and 10-mm bowl depth at the central point, while 19 holes with 3.5 mm diameter were drilled on the surface. Dross was collected until a clean sheet of liquid metal was obtained. When the dross cooled, it was separated from the steel strainer. All the collected dross was weighed on a balance (B1.5D; Axis) to the nearest 0.5 g (coefficient m d in formula 1). An example of the appearance of the collected sample of metallic dross is shown in Fig. 2. The measurement was then repeated for the second ingot of lead, the final result being the average of the two measurements, while the difference between them plus the error due to the accuracy of the measurement on the balance was considered as the magnitude of the measurement error. Dross%¼md mi 100 ð1Þ where Dross % = percentage coefficient of the amount of dross formed when melting metal; m d = mass of dross collected from the metal surface; m i = mass of molten metal. Fig. 1. (a) Industrial lead refining furnace: (1) steel bowl made of 40-mm-thick sheet metal, 3 m in diameter and 1.9 m high with a maximum capacity of about 125 mg of lead; (2) steel propeller for mixing lead connected to a 45-kW engine with stepless control; (3) hole for mounting an oxygen/air nozzle for lead softening; (4) pipe for connecting the exhaust with a dust removal and filtering installation; (5) furnace lining made of fireclay bricks; (6) set of two independent thermoelectric head sensors for temperature measurement; (7) PbSnCa lead alloy ingot. (b) Test stand: (1) PbSnCa alloy ingot, (2) cut-out fragment of the ingot weighing 8–9 kg; (3) measuring the mass of the cut fragment; (4) melting the metal at a station equipped with a steel bowl for melting lead, heated by a gas burner, with a ventilation hood above it; (5) measuring the temperature of the liquid lead; (6) a perforated steel spoon for removing dross from the liquid lead; (7) measuring the mass of the collected dross. Malecha, Albrecht, Lamb, and Małecki6606 In addition, for one batch of alloy produced, an analysis of the dross content was carried out immediately after casting and then after 15, 30 and 60 days. This study was designed to investigate how the storage time of the melt affects its level of melt dross content. The effect of two changes in the procedure methodology on the value of the calculated ‘‘Dross % ’’ coefficient (formula 1) was also checked. In the ‘‘changed method 1,’’ measurements were taken for four samples in a manner analogous to the procedure described above, but the temperature for the lead alloy was changed from 480 ±5 to 530 ±5°C. In the ‘‘changed method 2,’’ measurements were taken for four samples also in a manner analogous to the basic procedure but using a steel crucible enclosed in a resistance furnace (PT12/7O-ST; Czylok) instead of a steel bowl heated by a gas burner, leaving the other parameters unchanged. Analysis of Chemical and Phase Composition The chemical composition of the lead alloy and metallic dross formed during the tests was determined using Spark Optical Emission Spectrometry (S-OES; ARL iSpark 8860; Thermo Scientific). The dedicated matrix for the spectrometer used was lead and lead alloys with tin and calcium, which enabled high precision of measurements. Samples from lead alloys in the form of cylinders with about 35 mm diameter were taken following the guidelines of the EN 12402:1999 standard, 35 while analyses on the spectrometer were performed following the guidelines of the ENV 12908:1997 standard. 36 The analysis performed using S-OES was performed on two samples, and the results were presented as average values. The measurement uncertainty with a 95% confidence interval was calculated using the dedicated OXSAS software (v.2.7). It considers both the standard deviations between individual partial measurements and the uncertainty ranges of certified reference materials. The analysis of aluminum scrap was performed on a spectrometer (SOES; ARL iSpark 8860; Thermo Scientific) with a matrix dedicated to aluminum according to the guidelines of the EN 14726:2019 standard; 37 the analyses were performed to verify whether the impurity content did not exceed 5 wt.% declared by the supplier. Before analysis on S-OES, the lower surface of the metal samples, both lead and aluminum, was milled on a semi-automatic milling machine (HAF2; Herzog Maschinenfabrik). The chemical composition of the resulting dross was additionally verified using an inductively coupled plasma emission spectrometer (ICP-OES; Ultima Expert; Horiba Scientific). The alloying additives added to lead were also analyzed on this spectrometer to verify whether the impurities contained in them were within the ranges declared by their manufacturers. The samples of the resulting dross were manually crushed using cutting tongs. A 0.500 ±0.001 g average sample was taken from the prepared material and weighed on an electronic balance (WAS220X; Radwag). Before ICP-OES analysis, all materials were dissolved in aqua regia (HNO 3 + 3 HCl). To include oxygen in the tested material, the sample of metal dross was also analyzed using the X-ray fluorescence spectrometer (XRF; ZSX PRIMUS II; Rigaku). The spectrometer is equipped with a rhodium anode x-ray tube operating at an accelerating voltage of 60 kV. Using ten analyzer crystals, the percentage composition of elements from beryllium to uranium was measured. Quantitative elemental analysis was carried out using the SQX fundamental parameter method. The phase composition of the resulting dross was analyzed by X-ray diffraction (XRD; Empyrean; Malvern Panalytical) using nickel-filtered Cu Ka1,2 radiation (k= 1.5406 A ˚) and equipped with a PIXcell 3D hybrid detector. The measurements were performed in the reflection mode, in the Bragg-Brentano geometry (h–hscanning technique). The divergent slit of the incident beam was 1/2° wide, and additional Soller slits of 0.04 rad were used. The latest ICDD Card PDF5 database was used to analyze the obtained diffractograms. Microstructure Characterization Metallographic Sample Preparation The collected metallic dross samples extracted from the lead alloy refining bath were prepared into metallographic sections using standard preparation techniques by placing them in a conductive resin (hot mounting). Subsequently, the mounted samples were subjected to sequential grinding using silicon carbide (SiC) abrasive papers of decreasing grit size under water lubrication. The grinding process began with 240 grit paper and continued progressively to 1200 grit. After grinding, the samples were polished using diamond suspensions of increasingly finer particle sizes, finishing with a colloidal silica Fig. 2. Example of the collected sample of metallic dross. The Influence of Lead Refining Method on the Dross Content in the Metal 6607 suspension (particle size: 0.06 lm) to achieve a mirror-like surface free of scratches and deformation. The sample surface was not etched. Scanning Electron Microscopy (SEM) and Chemical Mapping (EDS) SEM observations and chemical element mapping were conducted using scanning electron microscopy (SEM; Apreo C; Thermo Scientific and JSM-7100F; Jeol). Both microscopes were equipped with highstability Schottky field emission guns, ensuring consistent, high-quality imaging. The operating voltage during SEM analysis was set between 15 and 20 keV. Images were captured using a standard secondary electron (SE) detector, which provides detailed topographical information. Chemical element mapping was performed using the Apreo C, while the chemical composition in the marked area was analyzed using the JSM-7100F with energydispersive spectroscopy (EDS). Lead Fire-Refining (Pyro-Refining)—A Literature Analysis of Available Techniques The traditional WLAB method of pyrorefining recycled lead starts with the so-called ‘‘drossing’’ process, removing excess impurities from the lead. The copper residue is then removed using sulfur. It then moves on to oxidative refining, which proceeds in two ways: by directly introducing air and/or oxygen into the lead and by using oxidizing chemicals such as NaOH and NaNO 3 . Lead that enters refining from WLAB rarely requires operations such as removing silver, bismuth or thallium because of sufficiently low levels of these elements already in the process charge. 9,34 Primary Decoppering by Drossing The drossing process is based on the low solubility of Cu, As, Se, Ni and Co in lead at low temperatures<350°C. This operation is carried out in melting furnaces (refining kettles) usually with a capacity of 20–300 mg in a semi-circular shape with a rolled section. The kettle is heated by gas or electricity. Liquid or solid lead is supplied to the furnace. First, the lead is heated to a temperature of 500–600°C. To accelerate the flow of the impurities separated from the lead to the surface of the bath, a mechanical stirrer is placed on the kettle by means of a gantry to mix the lead. After the bath has cooled to 330–350°C, the so-called ‘‘dross,’’ i.e. crystals of copper and other elements, mainly As, Se, Ni, Co and to a small extent Sb, is collected from its surface. Arsenic and antimony form hardly fusible intermetallic compounds with copper, such as Cu 5 As 2 and Cu 3 Sb, which are lighter than lead and float to the surface. The main constituent of the dross after drossing is lead (55–70 wt.%), because when the dross is removed from the surface of the bath with large punched shovels using overhead cranes, the shovel also collects some lead content along with the resulting dross, some of which does not manage to flow back through its punched bottom into the bath. The entire process usually takes between 1 and 3 h. 16,17,35 Decoppering with Sulfur Decoppering is carried out in the same melting furnace (refining kettle). To remove copper from lead, pure granulated sulfur is used, which is applied to the refined lead and mixed using a mechanical stirrer set up on the kettle. Some refineries still use pyrite (FeS 2 ) or galena (PbS) instead of pure sulfur for economic reasons, but copper removal efficiency is then lower. Sulfur is poured into the kettle in the amount resulting from the ratio of 1 kg sulfur per 1 kg copper. Sulfur reacts with lead and copper to form sulfides, which are solid at the process temperature (330–340°C). Due to the greater chemical affinity of sulfur with copper than with lead, the following reaction takes place: Cu½ PbþS½ Pb!CuS ð2Þ The product of this reaction is CuS, which does not dissolve in lead, is lighter than lead and does not melt at the process temperature, flowing to the surface of the metal to form dross. After a period of time, an increase in the concentration of copper in the lead is observed because of the system moving towards equilibrium. Given the above, dross from the lead surface should be removed as quickly as possible to reduce copper concentrations below the 5 ppm level. Decoppering takes between 3 and 5 h. The total amount of dross generated in the process varies within wide limits: from 8 to 25% of the raw lead amount. 17,21,35 To increase the efficiency of the lead decoppering process with sulfur, caustic soda (NaOH) can be added before the sulfur is added. Copper is also known to be removed from lead using metallic aluminum, which is introduced into the molten lead. As a result of this process, intermetallic compounds of aluminum with copper are brought to the surface of the lead. 36,37 Oxidative Refining with Air and/or Oxygen In this process (called lead softening), elements such as Sn, As and Sb are removed from lead by blowing compressed air and/or oxygen into the lead bath at a temperature of 650–750°C. This is possible because the free energy of formation of Sn, As and Sb oxides is more negative than the free energy of formation of lead oxide, shown in Fig. 3. Blowing air into the bath results in the oxidation of Sn, As, Sb and, to a lesser extent, Pb, successively, according to reactions 3–6. Sn½ PbþO2!SnO2ð3Þ Malecha, Albrecht, Lamb, and Małecki6608 2As½ Pbþ3=2O2!As2O3ð4Þ 2Sb½ Pbþ3=2O2!Sb2O3ð5Þ Pb þO2!2PbO ð6Þ If the process is discontinued at the appropriate time, the stripped dross will contain mainly SnO 2 , so it can be used as raw material for tin recovery. 9,16,17 Oxidative Refining with NaOH and NaNO 3 Based on the Harris Method The Harris method began to be used to refine lead more than 100 years ago, but the process is now being modified. Harris’s original method involved liquid lead, heated to 420–450°C and then passed through a molten mixture of sodium hydroxide (NaOH), sodium nitrate (NaNO 3 ) and sodium chloride (NaCl). The main oxidizing agent was sodium nitrate. The oxygen produced from the decomposition of saltpeter is a very active oxidant for impurities (As, Sn, Sb). In contrast, the main function of NaOH is to absorb As, Sn and Sb compounds formed in the process, with sodium chloride increasing this capacity Today, the refining process based on the Harris method (also known as lead softening) is carried out in the same melting furnace (refining kettle) as in the earlier refining stages. A mechanical stirrer is placed on the kettle to mix the lead, and sodium hydroxide (NaOH) and sodium nitrate (NaNO 3 ) are introduced into the funnel created by the mixing. The process starts at 430–450°C, but heating the lead in the kettle is only necessary at the beginning of the process because the oxidation reactions of the impurities are exothermic and give off enough heat to maintain the required temperature. Consequently, the energy consumption of the Harris process is low. In both the original and current Harris methods, the following reactions 7–9are likely to take place. 2As½ Pbþ2NaNO3þ4NaOH !2Na3AsO4þN2þ2H2Oð7Þ 5Sn½ Pbþ4NaNO3þ6NaOH !5Na2SnO3þN2þ3H2Oð8Þ 2Sb½ Pbþ2NaNO3þ4NaOH !5Na3SbO4þ2N2þ2H2Oð9Þ During this process, zinc is also removed, probably according to the following reaction 10. Zn½ Pbþ2NaNO3þ2NaOH !2Na2ZnO2þN2þH2Oð10Þ The order of oxidation of the impurities in this process is as follows: As, Sn and Sb, so it differs from that in the oxygen/air oxidation process, where the order is as follows: Sn, As and Sb. Fig. 3. Relationship between DGand temperature of the reactions: (3) red, (4) orange, (5) blue, (6) green. The data presented in the graph were obtained using HSC Chemistry v.6.1 software (Cold figure online). The Influence of Lead Refining Method on the Dross Content in the Metal 6609 The duration of the process depends on the impurity content of the lead and the dosing rate of NaOH and NaNO 3 . The removal rate is typically around 0.2% per hour. In this process, in addition to the elements described and their reactions 7–10, elements such as selenium, calcium, aluminum and barium are also removed. Bismuth and precious metals such as silver and gold remain almost entirely in lead during this refining process. There is also research into replacing the sodium hydroxide used in this process with a cheaper equivalent in the form of calcium carbonate (CaCO 3 ). 17,21,22,35 Sulfur Removal Sulfur from lead can be removed to<0.0001 wt.% by using sodium hydroxide NaOH, added at approximately 400°C, in amounts ranging from two to three times the weight content of sulfur. Sulfur in molten lead in the presence of oxygen can react with NaOH according to reaction 11. S½ Pbþ2NaOH þ3O½ Pb!Na2SO4þH2Oð11Þ Removal of Remaining Elements The methodology for removing other elements such as nickel, silver, gold, bismuth, zinc, tellurium and thallium, which was not used in the presented studies, is described in Table S-I. RESULTS AND DISCUSSION Production of Pb-Sn-Ca Alloy—Using Traditional Refining The first 20 production cycles of the PbSnCa alloy were carried out by the traditional fire refining of lead according to the scheme shown in Fig. 4. The process input was WLAB recycled lead (range of chemical composition is shown in Table I) with an initial mass of approximately 100 Mg. As described in Sect. ‘‘Primary Decoppering by Drossing’’, the first stage was so-called, drossing, where the lead was cooled to 340–450°C, and approximately 100 kg of sawdust and 50 kg of NaOH were added to reduce the dross formed. The average time for this process was 6 h. In each case analyzed, the copper content was too high and required decoppering with granular sulfur to bring it below 20 ppm (Sect. ‘‘Decoppering with Sulfur’’). It was calculated that during the de-gassing carried out for 100 Mg of lead, an average of approximately 100 kg of granular sulfur was used for every 1000 ppm of copper removed from the lead, while the process took an average of 2h. The next step was oxidative refining with oxygen, i.e., the purification of lead from tin, arsenic and antimony by blowing oxygen/air with a lance at a metal temperature of approximately 550–650°C (Sect. ‘‘Oxidative Refining with Air and/or Oxygen’’). This process was carried out each time until the Sb content fell below 0.2 wt.%. It was calculated that an average of 25 m 3 of oxygen/air was consumed per 1000 ppm of the sum of Sn, Sb and As contaminants in 100 Mg of lead. The average duration of this process is 10 h. The next step was a process based on the so-called Harris method, which involved heating the lead to approximately 430°C and then adding sodium hydroxide and sodium nitrate (Sect. ‘‘Oxidative Refining with NaOH and NaNO 3 Based on the Harris Method’’). This process was carried out until the Sb, As and Sn content was reduced to<1 ppm, using an average of 100 kg of NaOH and 200 kg of NaNO 3 for every 1000 ppm of total Sb, As and Sn impurities removed in 100 Mg lead. The process took an average of 18 h. The analyzed production cycles were chosen so that no additional lead refining operations such as silver, nickel, tellurium, thallium removal, etc., were required. Thus, each time after the abovementioned processes, the content of impurities in lead met the requirements for lead-grade Pb970R. 25 After each refining step, the resulting dross was pulled off the lead surface using a perforated shovel attached to a gantry until a completely clean sheet of lead was achieved. The final stage of production was the addition of alloying additives, namely Sn, to a level of approximately 1.5 wt.% and Ca to a level of approximately 0.06 wt.% and then casting the lead alloy into ingot form. Production of Pb-Sn-Ca alloy—Using Traditional Refining with Double Use of NaOH and NaNO 3 The subsequent refining method followed a similar path to the previous one, according to the scheme shown in Fig. 4. The only difference was the use of additional refining based on the Harris method (Sect. ‘‘Oxidative Refining with NaOH and NaNO 3 Based on the Harris Method’’). The process flow remained unchanged until the end of the refining step with NaOH and NaNO 3 . Once the dross had been collected from the lead surface, the process was repeated, adding a fixed 100 kg of NaOH and 250 kg of NaNO 3 to the already pure lead. After 3 h, the dross was collected from the metal surface. The next step, according to the scheme in Fig. 4, was to introduce alloying additives and then cast the lead alloy on a casting machine. Pb-Sn-Ca Alloy Production—When Refining with Aluminum The last 20 production cycles of the PbSnCa alloy were carried out using an alternative method of refining lead using aluminum with tin retained in the lead; the scheme is shown in Fig. 5. 38 As in the other two processes, the feedstock was WLAB recycled lead with an initial mass of approximately Malecha, Albrecht, Lamb, and Małecki6610 110 Mg. The first stage was drossing (Sect. ‘‘Primary Decoppering by Drossing’’); the second was decoppering using granulated sulfur (Sect. ‘‘Decoppering with Sulfur’’). The next step was to remove the sulfur. Sulfur from the lead was removed to <0.001 wt.% using sodium hydroxide NaOH, added at 390–410°C, in an amount of 2–3 weight quantities of sulfur still in the bath (Sect. ‘‘Sulfur Removal’’). The lead was then heated to approximately 670°C, and aluminum scrap was added to it in an amount of 0.2–0.5 of the sum of the amounts by weight of the antimony, arsenic, selenium, nickel and tellurium impurities while stirring the lead continuously until all the aluminum scrap had dissolved. The probable reactions that occur during this process are described by Eqs. 12–15. 39,40 Sb½ PbþAl !AlSb ð12Þ As½ PbþAl !AlAs ð13Þ Fig. 4. Schematic of the traditional lead refining process with the addition of alloying elements and casting of the finished PbSnCa lead alloy into ingot. The Influence of Lead Refining Method on the Dross Content in the Metal 6611 (3) Change in the methodology for analyzing the amount of dross formed translates into a derived value for the ratio of the amount of dross to the mass of metal melted. (4) The metallic dross formed by melting the PbSnCa alloy contains mainly lead, tin and calcium. The largest difference in chemical composition between the dross and the base metal was recorded for the tin content, indicating its high tendency to oxidize during the process. (5) Impurities such as iron, sulfur, sodium, selenium, cadmium, tellurium, chromium and magnesium, although present in small quantities in the PbSnCa alloy, accumulate in the dross and evidently increase in their amounts. (6) Phase analysis of the metallic dross showed the presence of metallic lead and lead oxides such as PbO, PbO 2 ,Pb 2 O 3 , tin oxides SnO, SnO 2 and calcium oxide CaO. (7) Microstructure studies of the metallic dross have shown that there are localized clusters of tin oxides and calcium oxides, which are unevenly distributed in the metallic lead. SUPPLEMENTARY INFORMATION The online version contains supplementary material available at https://doi.org/10.1007/s11837-02507558-x. ACKNOWLEDGEMENTS This work was supported by the Polish Ministry of Education and Science (Applied Doctorate Program, no. DWD/5/0149/2021). FUNDING Ministerstwo Edukacji i Nauki, DWD/5/0149/ 2021, Daniel Malecha. CONFLICT OF INTEREST The authors declare that they have no conflict of interest. OPEN ACCESS This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit h ttp://creativecommons.org/licenses/by/4.0/. REFERENCES 1. S. Arnout, B. Blanpain, D.R. Swinbourne, M. Chintinne, and M. Stevenson, in Handbook of Recycling, edited by C. Meskers, E. Worrell, and M. A. Reuter, Second Edition (Elsevier, Amsterdam, The Netherlands, 2024), pp. 355–370. 2. M.H. Morcali, ChemistrySelect 5, 7183 https://doi.org/10.10 02/slct.202001521 (2020). 3. J. Bu¨ngeler and B. Riegel, in Emerging Battery Technologies to Boost the Clean Energy Transition. The Materials Research Society Series. Springer, edited by Passerini S., Barelli L., Baumann M., Peters J., and Weil M. (Springer, Cham, 2024), pp. 121–141. 4. Eurostat, https://ec.europa.eu/eurostat/statistics-explained/i ndex.php?title=Waste_statistics_-_recycling_of_batteries_a nd_accumulators (2025). 5. US Environmental Protection Agency, https://www.epa.go v/facts-and-figures-about-materials-waste-and-recycling/nat ional-overview-facts-and-figures-materials (2024). 6. J. Hong, Z. Yu, W. Shi, J. Hong, C. Qi, and L. Ye, Int. J. Life Cycle Assess. 22, 909 https://doi.org/10.1007/s11367-016-12 09-3 (2017). 7. P.P. Lopes and V.R. Stamenkovic, Science 369(6506), 923– 924 https://doi.org/10.1126/science.abd3352 (2020). 8. B.B. McKeon, J. Furukawa, and S. Fenstermacher, Proc. IEEE 102, 951 https://doi.org/10.1109/JPROC.2014.2316823 (2014). 9. M. Altiner, S. Top, and S. Kursunoglu, in Recycling Technologies for Secondary Zn-Pb Resources. The Minerals, Metals & Materials Series., edited by M. Kaya (Springer Cham, 2023), pp. 235–289. 10. X. Jie, Z. Yao, C. Wang, D. Qiu, Y. Chen, Y. Zhang, B. Ma, and W. Gao, J. Sustain. Metall. 8, 978 https://doi.org/10. 1007/s40831-022-00576-y (2022). 11. M. Li, J. Yang, S. Liang, H. Hou, J. Hu, B. Liu, and R.V. Kumar, J. Power. Sources.https://doi.org/10.1016/j.jpow sour.2019.226853 (2019). 12. R. D. Prengaman and A. H. Mirza, in Lead-Acid Batteries for Future Automobiles (Elsevier Inc., 2017), pp. 575–598. 13. A.G. Morachevskii, Russ. J. Appl. Chem. 87, 241 https://doi. org/10.1134/S107042721403001X (2014). 14. H. Y. Sohn and M. Olivas-Martinez, in Treatise on Process Metallurgy (Second Edition), edited by S. Seetharaman, R. Guthrie, A. McLean, S. Seetharaman, and H. Y. Sohn, Second Edition (Elsevier, 2024), pp. 605–624. 15. M. W. Stevenson, in Encyclopedia of Electrochemical Power Sources (Second Edition), edited by Garche Ju¨rgen (Elsevier, 2025), pp. 563–584. 16. H. Y. Sohn and M. Olivas-Martinez, in Treatise on Process Metallurgy, edited by S. Seetharaman (Elsevier, 2014), pp. 671–700. 17. M. Kucharski, Recycling of non-ferrous metals (in Polish) (AGH University Press, Krakow, 2010). 18. T.W. Ellis and A.H. Mirza, J. Power. Sources 195, 4525 h ttps://doi.org/10.1016/j.jpowsour.2009.12.118 (2010). 19. D. Pavlov, Lead-Acid Batteries: Science and Technology (Second Edition), edited by D. Pavlov, Second Edition (Elsevier, 2017), pp. 169–243. 20. R. D. Prengaman, Lead-Acid Batteries for Future Automobiles (Elsevier Inc., 2017), pp. 269–299. 21. C. Fleuriault, 9th International Symposium on High-Temperature Metallurgical Processing, edited by Jiann-Yang Hwang, Tao Jiang, Mark William Kennedy, Dean Gregurek, Shijie Wang, Baojun Zhao, Onuralp Yu¨ cel, Ender Keskinkilic, Jerome P. Downey, Zhiwei Peng, and Rafael Padilla Malecha, Albrecht, Lamb, and Małecki6618 (TMS 2018. The Minerals, Metals & Materials Series. Springer, Cham, 2018), pp. 187–197. 22. S. Asante and P. Taylor, New Directions in Mineral Processing, Extractive Metallurgy, Recycling and Waste Minimization, edited by R. G. Reddy, A. Anderson, C. G. Anderson, C. Fleuriault, E. D. Spiller, M. Strauss, E. E. Vidal, and M. Zhang (TMS 2023. The Minerals, Metals & Materials Series. Springer, Cham, 2023), pp. 257–274. 23. M. W. Stevenson, Encyclopedia of Electrochemical Power Sources, edited by Garche Ju¨rgen (Elsevier, 2009), pp. 165– 178. 24. L.-A. Ouyang, Y. He, P. He, J. Zhou, H. Huang, and Z. Guo, J. Sustain. Metall. 10, 1905 https://doi.org/10.1007/s40831024-00928-w (2024). 25. EN 12659:1999 Lead and Lead Alloys—Lead (European Committee for Standardization (CEN), Brussels, 1999). 26. R. D. Prengaman, Encyclopedia of Electrochemical Power Sources, edited by J. Garche (Elsevier, 2009), pp. 648–654. 27. E. Cattaneo, H. Stumpf, H.G. Tillmann, and G. Sassmannshausen, J. Power. Sources 67, 283 https://doi.org/10.1016/ S0378-7753(97)02558-5 (1997). 28. D.A.J. Rand, D.P. Boden, C.S. Lakshmi, R.F. Nelson, and R.D. Prengaman, J. Power. Sources 107, 280 https://doi.org/ 10.1016/S0378-7753(01)01083-7 (2002). 29. L. Prout, J. Power. Sources 50, 193 https://doi.org/10.1016/ 0378-7753(94)01896-0 (1994). 30. M. Vonc ˇina, J. Medved, G. S ˇegel, and T. Balas ˇko, Mater. Tehnol. https://doi.org/10.17222/mit.2024.1119 (2025). 31. A. Kvithyld, S. Bao, A. Nordmark, M. Syvertsen, M. Schlesinger, and A. Johansson, Light Metals 2014 (Springer International Publishing, Cham, 2014), pp931–935. 32. C.K.W. Solem, S. Deledda, G. Tranell, and R.E. Aune, J. Sustain. Metall. 9, 95 https://doi.org/10.1007/s40831-022-00 624-7 (2023). 33. A.G. Dos Santos, M.R. Vieira, F.J. da Silva, and N. Bouchonneau, Heliyon.https://doi.org/10.1016/j.heliyon.2024.e3 8536 (2024). 34. G. Rojek, K. Regulski, P. Jarosz, J. Gabryel, and J. Kusiak, Comput. Methods Mater. Sci. 15, 78 https://doi.org/10.7494/c mms.2015.1.0506 (2015). 35. F. Habashi, in Handbook of Extractive Metallurgy, edited by F. Habashi (Wiley-VCH, Heidelberg, 1997), pp. 581–714. 36. A.G. Morachevskii and T.V. Butukhanova, Russ. J. Appl. Chem. 82, 1317 https://doi.org/10.1134/S1070427209070301 (2009). 37. A. Cybulski and R. Prajsnar, World Metall. ERZMETALL 66, 340 (2013). 38. D. Malecha, S. Małecki, P. Jarosz, R. Kowalik, and P. _ Zabin ´ski, Materials 16, 5882 https://doi.org/10.3390/ma161 75882 (2023). 39. D. Malecha, M. Zubko, P. _ Zabin ´ski, and S. Małecki, Metall. Mater. Trans. A.https://doi.org/10.1007/s11661-025-07813-5 (2025). 40. D. Malecha, P. S ´wiec, R. Albrecht, P. Jarosz, and S. Małecki, JOM.https://doi.org/10.1007/s11837-024-06892-w (2024). 41. T. Szyndler, R. Gnida, D. Malecha, S. Małecki, and P. Jarosz, PL 443802 A1 (2023). 42. T. Szyndler, R. Gnida, D. Malecha, S. Małecki, and P. Jarosz, EP 4417719 A1 (2023). 43. T. Szyndler, R. Gnida, D. Malecha, S. Małecki, and P. Jarosz, PL 439991 A1 (2021). Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. The Influence of Lead Refining Method on the Dross Content in the Metal 6619