Solubility of Silver in FeO_x–CaO–SiO₂ Slag Coexisting with Silver and Silver Bromide at 1473 K

Abstract

Silver, Ag, in secondary raw materials such as printed circuit boards is recovered by nonferrous metal thermochemical processes. Because printed circuit boards contain bromine, Br, which is highly reactive with Ag, as a flame retardant, there is concern that Br may affect the recovery of Ag. In this study, the FeO_x–CaO–SiO₂ slags were equilibrated with liquid AgBr and Ag in a MgO crucible at 1473 K, and the effects of concentration of Br in the slags, oxygen partial pressure, p_O2, and slag composition on the dissolution of Ag were experimentally investigated. The results showed that the concentration of Ag in the slag coexisting with Ag and AgBr was higher than that coexisting with only Ag. Moreover, the concentration of Ag in the slag increased with increasing concentration of Br in the slag coexisting with Ag and AgBr. At p_O2 = 10⁻⁹, the concentrations of Ag and Br in the slag in equilibrium with AgBr were minimum at neutral slag basicity Q = 0.45, defined as Q = (mass%CaO + mass%MgO)/(mass%CaO + mass%MgO + mass%SiO₂). The slag loss of Ag can be reduced using neutral basicity slag and by reducing the bromine partial pressure in the system to reduce the concentration of Br dissolved in the slag.

Fig. 7 Relationship between the solubilities of Ag and Br in the FeO_x–CaO–SiO₂ slag at 1473 K.

1. Introduction

The recovery of precious metals from waste electrical and electronic equipment (WEEE) has significant advantages as it reduces the consumption of natural resources and has less environmental impact than smelting ores [1]. Currently, pyrometallurgy, hydrometallurgy, or a combination of both is used for precious metals recycling [2]. The pyrochemical process can concentrate precious metals while removing halides and hazardous organics to the gas phase. Ag contained in printed circuit boards (PCBs) is recycled using pyrochemical processes. PCBs contain halogens that are highly reactive with Ag as flame retardants [3, 4]. Therefore, there is concern that halogens generated from PCBs in pyrochemical processes may affect the recovery of Ag.

The amount of valuable metal Ag dissolved in the slag must be small because the pyrochemical processes separate impurity elements to the slag. Studies on the dissolution of Ag in slag have been reported primarily for applications in copper and lead smelting. The solubility of pure Ag in slag was investigated by Wakasugi et al. in the Na₂O–B₂O₃ and Na₂O–B₂O₃–Al₂O₃ slag, by Park et al. in the CaO–B₂O₃, BaO–B₂O₃, and Na₂O–B₂O₃ slag, by Swinbourne et al. in the Na₂O–SiO₂ and PbO slag, and by Seki et al. in the FeO_x–SiO₂ slag [5–10]. The distribution behavior of Ag between slag and Cu-based alloy was reported by Seki et al., Kashima et al., Hellstén et al., and Hidayat et al. in the FeO_x–SiO₂ slag, by Takeda et al. in the FeO_x–CaO and FeO_x–CaO–SiO₂–MgO slag, Avarmaa et al. in the FeO_x–SiO₂, FeO_x–SiO₂–Al₂O₃, and FeO_x–CaO–SiO₂–Al₂O₃ slag, and Liu et al. and Ye et al. in the FeO_x–CaO–SiO₂–Al₂O₃ slag [10–20]. The distribution behavior of Ag between slag and Pb-based alloy was reportedd by Pérez et al. in the FeO_x–CaO–SiO₂ slag and by Shishin et al. in the FeO_x–SiO₂–PbO slag [21, 22]. The distribution behavior of Ag between slag and Cu₂S–FeS-based matte was determined by Kashima et al., Nagamori et al., Zapata et al., and Avarmaa et al. in the FeO_x–SiO₂ slag, by Takeda et al. in the FeO_x–CaO–SiO₂–MgO slag, by Roghani et al. in the FeO_x–CaO, FeO_x–SiO₂–MgO, and FeO_x–CaO–SiO₂–MgO slag, and Choi et al. in the FeO_x–CaO–SiO₂, FeO_x–SiO₂–MgO, and FeO_x–SiO₂–Al₂O₃ slag [11, 13, 23–30]. Many of these studies reported that the amount of Ag dissolved in slag increased with higher oxygen partial pressure in the system, leading to the dissolution of Ag in slag as AgO_0.5 because of those relationships. Liu et al. and Ye et al. studied the distribution behavior of Au, Ag, and Sn when CaF₂, CaCl₂, or CaBr₂ was added to slag and reported that the amount of Ag dissolved in the slag and its volatility increased when halogens coexisted [19, 20]. However, there are no reports on the distribution behavior of Ag in slag when the bromine partial pressure is determined.

In this study, an existing nonferrous pyrochemical process was supposed to treat PCBs containing brominated flame retardant and recover Ag. The FeO_x–CaO–SiO₂ slags were equilibrated with liquid AgBr and Ag in a MgO crucible at 1473 K. The effects of the concentration of Br in the slag, oxygen partial pressure, and slag composition on Ag dissolution were experimentally investigated.

2. Experimental Procedure

Experiments were conducted to investigate the solubility of Ag in slag in the absence of Br. A total of 8 g of a mixture of electrolytic iron (>98% purity), Fe₂O₃ (>99% purity), CaO calcined from CaCO₃ (>99.9% purity), and SiO₂ (>99.9% purity) reagents as slag and 1 g of Ag (>99.99% purity) were inserted into a MgO crucible (>99.6% purity) with an inner diameter, a height, and a wall thickness of 18, 40, and 2.5 mm, respectively. The sample was inserted into a SiO₂ reaction tube, held at 1473 K (±3 K) in a furnace, and heated using SiC heating elements. The process was performed at $p_{\text{O}_{2}} = 10^{-7}$, 10⁻⁸, or 10⁻⁹ by blowing the sample with CO and CO₂ gas mixture at a combined flow rate of 100 mL/min for 24 h. Where $p_{\text{O}_{2}}$ is the dimensionless number obtained by dividing the oxygen partial pressure by the standard pressure: 1.01325 × 10⁵ Pa. Subsequently, the sample was then quenched in Ar gas flow. The slag and alloy phases in the crucible were separated and dissolved in an aqueous solution. The concentrations of the components in the slag, alloy, and salt phases were quantified by chemical analysis using inductively coupled plasma optical emission spectrometry (ICP-OES; Agilent Technologies, 5100).

Experiments, in which Br coexisted, were conducted. A total of 8 g of the above slag reagent, 1 g of Ag, and 2 g of AgBr (>99.5% purity) were placed in a MgO crucible. The samples were held for 10 h under the conditions described above. The holding time was shortened because Br was volatilized, and the AgBr phase disappeared when the holding time was long. The concentrations of the components in each phase were determined by ICP-OES, except for Br, for which ion chromatography (IC; Thermo Fisher Scientific, ICS-2100) was used. Liquid Ag and liquid AgBr were equilibrated and the bromine partial pressure in the system was fixed. The value was calculated as $p_{\text{Br}_{2}} = 10^{-4.12}$ at 1473 K using the FactPS database of the thermodynamic computing system FactSage ver. 8.1 [31]. Although the composition of the slag varied with use of a MgO crucible, the initial composition of the slag was determined based on the FeO_x–CaO–SiO₂ phase diagram reported by Hidayat et al. [32, 33]. Figure 1 shows the FeO_x–CaO–SiO₂ ternary system at 1473 K reported by Hidayat et al. and initial compositions of the slag in this study. To confirm whether the solid phase coexisted in the slag and whether the alloy phase particles were suspended in the slag, sample No. 1 in Table 1 was observed using scanning electron microscopy (SEM; JEOL, JSM-6510A). To confirm that the samples reached equilibrium, holding times of 2, 4, 6, and 8 h were used for No. 9 of sample in Table 1. A schematic of the experimental apparatus is shown in Fig. 2.

Fig. 1

Initial compositions of the slag and FeO_x–CaO–SiO₂ ternary system at 1473 K [32, 33].

Table 1 Experimental conditions and concentrations of each element in the slag after equilibrium. ^*FeO concentration was determined from the Fe chemical analysis, assuming that all Fe in the slag is present as FeO.

Fig. 2

Schematic of the experimental apparatus.

3. Results and Discussion

After quenching, the samples with AgBr separated into slag, salt, and alloy phases, whereas the samples without AgBr separated into slag and alloy phases. Table 1 shows the concentrations of the slag components after the experiment. ICP-OES and IC analyses showed that elements other than Ag in the alloy phase, and elements other than Ag and Br in the salt phase were both <1 mass%. Therefore, in the following discussion, the alloy phase (Ag phase) was assumed to consist only of Ag, and the salt phase (AgBr phase) consisted only of AgBr. Generally, Fe coexists in both divalent and trivalent states in slags. When metalic Fe coexists, most of it is divalent. Because the valence of iron in the slag was not quantified in this study, all Fe in the slag detected by chemical analysis was assumed to be present as FeO. SEM observations for the No. 1 of sample show that the solid phase did not coexist and that the alloy phase was not suspended in the slag.

3.1 Equilibrium time

Figure 3 shows the relationship between the holding time and the concentrations of Ag and Br in the slag for the sample with the initial composition of sample No. 9. The concentrations of Ag and Br in the slag increased over 6 h and remained constant. Therefore, a holding time of 10 h was considered sufficient for reaching equilibrium in the experiments.

Fig. 3

Relationship between the concentrations of Ag and Br in the FeO_x–CaO–SiO₂ slag and holding time at 1473 K (initial composition of sample No. 9 in Table 1).

3.2 Solubility of MgO in slags

The MgO crucible was partially dissolved in the slag and equilibrated. Figure 4 shows the relationship between the MgO concentration and the ratio of CaO to SiO₂ in the slag after equilibrium. The MgO concentration in the slag increased as the value of (mass%CaO)/(mass%SiO₂) decreased. This is because MgO and CaO are basic oxides and SiO₂ is an acidic oxide. The effects of oxygen partial pressure and AgBr on MgO solubility were not identified.

Fig. 4

Solubility of MgO in the FeO_x–CaO–SiO₂ slag at 1473 K.

3.3 Ag and Br in slags

Figure 5 shows the relationship between the concentration of Ag and basicity Q in the slag. Basicity is an index that refers to the activity of oxygen ions, which cannot be measured directly. In this study, because the amount of dissolved MgO varied with the initial slag composition, Q defined by eq. (1), was used as the basicity using the slag composition after equilibrium.

  

\begin{equation} Q = \frac{(\text{mass%CaO}) + (\text{mass%MgO})}{(\text{mass%CaO}) + (\text{mass%MgO}) + (\text{mass%SiO$_{2}$})} \end{equation} (1)

When AgBr was absent, the concentration of Ag in the slag decreased as the basicity increased. This is consistent with the results reported by Park et al. and Takeda et al. that Ag oxide behaves as a neutral oxide in slag and that the solubility of Ag was the smallest when the slag was neutral [6, 7, 13]. When AgBr was present, the amount of Ag dissolved in the slag increased under all experimental conditions compared to that without AgBr. Furthermore, the concentration of Ag in the slag showed its minimum at approximately Q = 0.45 at $p_{\text{O}_{2}} = 10^{-9}$. Wang et al. reported the effect of chlorine content in slag on the viscosity in CaO–SiO₂–Al₂O₃–MgO–CaCl₂ slag and the effect on structure from Raman spectra of quenched slag [34]. In the study, the viscosity and the average amount of the bridging oxygen decreased with increasing concentration of Cl in the slag. The results suggest that the Cl anion functioned as a modifier of silicate network like basic oxides. Although the slag and halogen are different in this study, it is possible that the Br anion behaved similarly to basic oxides in the slag, and Q, where the concentration of Ag in the slag was at a minimum, varied by the presence or absence of AgBr.

Fig. 5

Relationship between the solubility of Ag and basicity in the FeO_x–CaO–SiO₂ slag at 1473 K.

Figure 6 shows the relationship between concentration of Br and Q in the slag. The concentration of Br reached its minimum at approximately Q = 0.45 at $p_{\text{O}_{2}} = 10^{-9}$. Figure 7 is the relationship between concentrations of Br and Ag in the slag. These results show a strongly positive correlation, suggesting that Br dissolution increases the amount of Ag dissolved in the slag. Figure 8 shows the relationship between the concentrations of Ag and Br in the slag and the oxygen partial pressure when the initial composition of the slag was 40 mass%FeO_x–15 mass%CaO–45 mass%SiO₂. Regardless of the presence or absence of Br in the slag, the relationship between Ag solubility and oxygen partial pressure had a slope of approximately 1/4. This is consistent with previous studies that investigated the oxygen partial pressure dependence of Ag dissolution in slag [9, 10, 13–16, 18, 22]. The results suggest that in the absence of AgBr, Ag is dissolved in the state of AgO_0.5 in the slag. In the case of AgBr coexistence, it may also be dissolved in the state of silver oxybromide, which cannot be determined from the present results. Hirosumi et al., Miwa et al., Myoung et al., and Okeda et al. investigated the solubility of chlorine in slags consisting mainly of SiO₂–CaO–Al₂O₃ and reported that the solubility of chlorine increased with decreasing oxygen partial pressure [35–38]. However, in this study, the concentration of Br in the slag increased with increasing oxygen partial pressure. This can be attributed to the different properties of Cl and Br, the different components of the slag, and the coexistence of Ag.

Fig. 6

Relationship between the solubility of Br and basicity in the FeO_x–CaO–SiO₂ slag at 1473 K ($p_{\text{Br}_{2}} = -4.12$).

Fig. 7

Relationship between the solubilities of Ag and Br in the FeO_x–CaO–SiO₂ slag at 1473 K.

Fig. 8

Relationship between the solubilities of Ag and Br and the oxygen partial pressure in the FeO_x–CaO–SiO₂ slag at 1473 K.

3.4 Activity coefficient of AgO_0.5 in slag

Finally, the activity coefficients of Ag oxides in the slag were calculated. Based on the relationship between Ag solubility and the oxygen partial pressure shown in Fig. 8, here it is assumed that all of the Ag in the slag in the state of AgO_0.5 regardless of the presence or absence of AgBr. Although AgO_0.5 is liquid, the activity coefficient of AgO_0.5 (s) was calculated for comparison with the results of previous studies.

  

\begin{equation} \text{Ag (l)} + \frac{1}{4}\text{O$_{2}$ (g)} = \text{AgO$_{0.5}$ (s)} \end{equation} (2)

The equilibrium constant K for the reaction is expressed using eq. (3) as follows:

  

\begin{equation} K = \frac{a_{\text{AgO${_{0.5}}$ (s)}}}{a_{\text{Ag (l)}}\cdot p_{\text{O${_{2}}$}}{}^{\frac{1}{4}}}\quad (K = 8.55 \times 10^{-2}\ \text{at}\ 1473\,\text{K})\ [31] \end{equation} (3)

where a is the activity and $p_{\text{O}_{2}}$ is the oxygen partial pressure. K is calculated from the changes in the standard Gibbs free energy using FactPS [31]. Transforming eq. (3), the activity coefficient of AgO_0.5 (s) was obtained using eq. (4).

  

\begin{equation} \gamma_{\text{AgO${_{0.5}}$ (s)}} = \frac{K \cdot a_{\text{Ag (l)}} \cdot p_{\text{O${_{2}}$}}{}^{\frac{1}{4}}}{X_{\text{AgO${_{0.5}}$}}} \end{equation} (4)

where, γ is the activity coefficient and X is the molar fraction. In this study, a_Ag(l) = 1 because pure Ag was equilibrated. Figure 9 shows the relationship between the calculated activity coefficient and slag basicity. For comparison, the results of equilibrium experiments between 65 mass%FeO_x–35 mass%SiO₂ slag saturated with MgO and Ag at 1573 K by Seki et al. and between FeO_x–CaO–SiO₂ slag saturated with MgO and Cu–Ag alloy at 1573 K by Takeda et al. are described [10, 13]. However, the initial (mass%CaO)/(mass%CaO + mass%SiO₂) was used as the basicity because the composition of slag after equilibrium was not available for the results of Takeda et al. [13]. Although the temperature and composition of the alloy phase are different, the results of this study without AgBr are consistent with the trend reported by Takeda et al. in which activity coefficients are larger when basicity is high [13]. However, the activity coefficients of AgO_0.5 (s) under the condition of AgBr coexistence are smaller than in these studies. The concentration of CuO_0.5 in the slag, increased with increasing oxygen partial pressure in the studies by Seki et al. and Takeda et al. in which the slag and Cu–Ag alloy were equilibrated. Even if a small mass of CuO_0.5 was dissolved in the slag, the activity coefficient of AgO_0.5 (s) remained almost unchanged [10, 13]. Therefore, it is suggested that the dissolution of Br in the slag significantly changes the properties of the slag, reduces the activity coefficient of AgO_0.5 (s), and has a significant effect on the dissolution of Ag, even when the amount of Br dissolved in the slag is small less than 1 mass% Br.

Fig. 9

Activity coefficient of AgO_0.5 (s) in the FeO_x–CaO–SiO₂ slag at 1473 K [10, 13].

4. Conclusion

In this study, an existing nonferrous pyrochemical process was supposed to treat PCBs containing brominated flame retardant and recover Ag. The FeO_x–CaO–SiO₂ slag, AgBr, and Ag were equilibrated in a MgO crucible at 1473 K. The effects of the Br content, oxygen partial pressure, and slag composition on the dissolution of Ag were experimentally investigated. The experimental results showed that when the AgBr phase did not coexist, the solubility of Ag in the slag decreased with increasing basicity because AgO_0.5 is a neutral oxide and highly basic slag in this study is neutral. When the AgBr phase is present, the solubilities of Ag and Br in the slag is minimized at approximately Q = 0.45. Furthermore, the solubility of Ag increases with increasing concentration of Br in the slag. The presence of Br in the pyrochemical process appeared to increase the slag loss of Ag. It is proposed that the use of slag with a low Br solubility of approximately Q = 0.45 and pre-roasting the raw material to reduce the Br partial pressure in the furnace can reduce the dissolution of Ag into the slag.

Acknowledgments

We thank the Environmental Safety Center of Waseda University for the use of ICP-OES and IC.

REFERENCES

• 1)   M.  Bigum,  L.  Brogaard and  T.H.  Christensen: Metal recovery from high-grade WEEE: A life cycle assessment, J. Hazard. Mater. 207–208 (2012) 8–14. doi:10.1016/j.jhazmat.2011.10.001
• 2)   F.  Tesfaye,  D.  Lindberg,  J.  Hamuyuni,  P.  Taskinen and  L.  Hupa: Improving urban mining practices for optimal recovery of resources from e-waste, Miner. Eng. 111 (2017) 209–221. doi:10.1016/j.mineng.2017.06.018
• 3)   G.  Mishra,  R.  Jha,  M.D.  Rao,  A.  Meshram and  K.K.  Singh: Recovery of silver from waste printed circuit boards (WPCBs) through hydrometallurgical route: A review, Environ. Challenges 4 (2021) 100073. doi:10.1016/j.envc.2021.100073
• 4)   Y.  Chen,  J.  Li,  L.  Chen,  S.  Chen and  W.  Diao: Brominated Flame Retardants (BFRs) in Waste Electrical and Electronic Equipment (WEEE) Plastics and Printed Circuit Boards (PCBs), Procedia Environ. Sci. 16 (2012) 552–559. doi:10.1016/j.proenv.2012.10.076
• 5)   T.  Wakasugi,  A.  Hirota,  J.  Fukunaga and  R.  Ota: Solubility of Ag₂O into the Na₂O–B₂O₃–Al₂O₃ system, J. Non-Cryst. Solids 210 (1997) 141–147. doi:10.1016/S0022-3093(96)00590-X
• 6)   J.H.  Park and  D.J.  Min: Quantitative Analysis of the Relative Basicity of CaO and BaO by Silver Solubility in Slags, Metall. Mater. Trans. B 30 (1999) 689–694. doi:10.1007/s11663-999-0030-9
• 7)   J.H.  Park and  D.J.  Min: Solubility of Silver in MO (M₂O)–B₂O₃ (M = Ca, Ba and Na) Slags, Mater. Trans., JIM 41 (2000) 425–428. doi:10.2320/matertrans1989.41.425
• 8)   D.R.  Swinbourne and  X.M.  Yan: Silver solubility in sodium silicate slags, Miner. Process. Extr. Metall. 108 (1999) 59–65.
• 9)   D.R.  Swinbourne,  S.  Yan,  G.  Hoang and  D.  Higgins: Thermodynamic modelling of cupellation: activity of AgO_0.5 in molten PbO at 1000°C, Miner. Process. Extr. Metall. 112 (2003) 69–74. doi:10.1179/037195503225002754
• 10)   G.  Seki,  T.  Murata and  K.  Yamaguchi: Silver Solubility and Effect of Copper Concentration on the Activity Coefficient of Silver Oxide in the FeO_x-SiO₂ slag system at 1573K, Resources Processing 68 (2021) 70–76. doi:10.4144/rpsj.68.70
• 11)   M.  Kashima,  M.  Eguchi and  A.  Yazawa: Distribution of Impurities between Crude Copper, White Metal and Silica-Saturated Slag, Trans. JIM 19 (1978) 152–158. doi:10.2320/matertrans1960.19.152
• 12)   Y.  Takeda,  S.  Ishiwata and  A.  Yazawa: Distribution Equilibria of Minor Elements between Liquid Copper and Calcium Ferrite Slag, Trans. JIM 24 (1983) 518–528. doi:10.2320/matertrans1960.24.518
• 13)  Y. Takeda and G. Roghani: DISTRIBUTION EQUILIBRIUM OF SILVER IN COPPER SMELTING SYSTEM, First International Conference on Processing Materials for Properties, (TMS, 1993) pp. 357–360.
• 14)  K. Avarmaa, H. O’Brien and P. Taskinen: Equilibria of Gold and Silver between Molten Copper and FeO_x-SiO₂-Al₂O₃ Slag in WEEE Smelting at 1300 °C, Proceedings of The 10th International Conference on Molten Slags, Fluxes and Salts, (TMS, 2016) pp. 193–202.
• 15)   K.  Avarmaa,  L.  Klemettinen,  H.  O’Brien and  P.  Taskinen: Equilibria of Gold and Silver between Molten Copper and FeO_x-SiO₂-Al₂O₃ Slag in WEEE Smelting at 1300 °C, Miner. Eng. 133 (2019) 95–102. doi:10.1016/j.mineng.2019.01.006
• 16)   K.  Avarmaa,  H.  O’Brien,  L.  Klemettinen and  P.  Taskinen: Precious metal recoveries in secondary copper smelting with high-alumina slags, J. Mater. Cycles Waste Manag. 22 (2020) 642–655. doi:10.1007/s10163-019-00955-w
• 17)   N.  Hellstén,  L.  Klemettinen,  D.  Sukhomlinov,  H.  O’Brien,  P.  Taskinen,  A.  Jokilaakso and  J.  Salminen: Slag Cleaning Equilibria in Iron Silicate Slag–Copper Systems, J. Sustainable Metall. 5 (2019) 463–473. doi:10.1007/s40831-019-00237-7
• 18)   T.  Hidayat,  J.  Chen,  P.C.  Hayes and  E.  Jak: Distributions of Ag, Bi, and Sb as Minor Elements between Iron-Silicate Slag and Copper in Equilibrium with Tridymite in the Cu-Fe-O-Si System at T = 1250 °C and 1300 °C (1523 K and 1573 K), Metall. Mater. Trans. B 50 (2019) 229–241. doi:10.1007/s11663-018-1448-8
• 19)   Z.  Liu,  F.  Ye,  R.  Chen,  H.  Wang and  L.  Xia: The Partitioning Behaviors of Au, Ag, and Sn in the Cu-FeO_X-SiO₂-CaO-Al₂O₃ Smelting System in the Presence of Halogen Elements, JOM 74 (2022) 622–633. doi:10.1007/s11837-021-04987-2
• 20)   F.  Ye,  Z.  Liu and  L.  Xia: Slag chemistry, element distribution behaviors, and metallurgical balance of e-waste smelting process, Circular Economy 2 (2023) 100062. doi:10.1016/j.cec.2023.100062
• 21)   M.  Pérez,  A.  Romero,  A.  Hernández,  I.  Almaguer and  R.  Benavides: Distribution of lead and silver under lead blast furnace conditions, Rev. Metal. 48 (2012) 213–222. doi:10.3989/revmetalm.1150
• 22)   D.  Shishin,  T.  Hidayat,  U.  Sultana,  M.  Shevchenko and  E.  Jak: Experimental Study and Thermodynamic Calculations of the Distribution of Ag, Au, Bi, and Zn Between Pb Metal and Pb–Fe–O–Si slag, J. Sustainable Metall. 6 (2020) 68–77. doi:10.1007/s40831-019-00257-3
• 23)   M.  Nagamori and  P.J.  Mackey: Thermodynamics of Copper Matte Converting: Part II. Distribution of Au, Ag, Pb, Zn, Ni, Se, Te, Bi, Sb and As Between Copper, Matte and Slag in the Noranda Process, Metall. Trans. B 9 (1978) 567–579. doi:10.1007/BF03257205
• 24)   G.  Roghani,  J.  Font,  M.  Hino and  K.  Itagaki: Distribution of Minor Elements between Calcium Ferrite Slag and Copper Matte at 1523 K under High Partial Pressure of SO₂, Mater. Trans., JIM 37 (1996) 1574–1579. doi:10.2320/matertrans1989.37.1574
• 25)   G.  Roghani,  M.  Hino and  K.  Itagaki: Phase Equilibrium and Minor Elements Distribution between SiO₂–CaO–FeO_x–MgO Slag and Copper Matte at 1573 K under High Partial Pressures of SO₂, Mater. Trans., JIM 38 (1997) 707–713. doi:10.2320/matertrans1989.38.707
• 26)   G.  Roghani,  Y.  Takeda and  K.  Itagaki: Phase equilibrium and minor element distribution between FeO_X-SiO₂-MgO-based slag and Cu₂S-FeS matte at 1573 K under high partial pressures of SO₂, Metall. Mater. Trans. B 31 (2000) 705–712. doi:10.1007/s11663-000-0109-9
• 27)   N.  Choi and  W.D.  Cho: Distribution of gold and silver between nickel-copper matte and silica-saturated iron silicate slag, Min. Metall. Explor. 15 (1998) 23–29. doi:10.1007/BF03403220
• 28)  H.M. Heneo, K. Yamaguchi and S. Ueda: Distribution of precious metals (Au, Pt, Pd, Rh and Ru) between copper matte and iron-silicate slag at 1573 K, Sohn International Symposium, (TMS, 2006) pp. 723–729.
• 29)   K.  Avarmaa,  H.  O’Brien,  H.  Johto and  P.  Taskinen: Equilibrium Distribution of Precious Metals Between Slag and Copper Matte at 1250–1350 °C, J. Sustainable Metall. 1 (2015) 216–228. doi:10.1007/s40831-015-0020-x
• 30)   K.  Avarmaa,  H.  Johto and  P.  Taskinen: Distribution of Precious Metals (Ag, Au, Pd, Pt, and Rh) Between Copper Matte and Iron Silicate Slag, Metall. Mater. Trans. B 47 (2016) 244–255. doi:10.1007/s11663-015-0498-4
• 31)   C.W.  Bale et al.: FactSage thermochemical software and databases, 2010–2016, Calphad 54 (2016) 35–53. doi:10.1016/j.calphad.2016.05.002
• 32)   T.  Hidayat,  P.C.  Hayes and  E.  Jak: Experimental Study of Ferrous Calcium Silicate Slags: Phase Equilibria at P_O2 Between 10⁻⁵ atm and 10⁻⁷ atm, Metall. Mater. Trans. B 43 (2012) 14–26. doi:10.1007/s11663-011-9569-3
• 33)   T.  Hidayat,  P.C.  Hayes and  E.  Jak: Experimental Study of Ferrous Calcium Silicate Slags: Phase Equilibria at P_O2 Between 10⁻⁸ atm and 10⁻⁹ atm, Metall. Mater. Trans. B 43 (2012) 27–38. doi:10.1007/s11663-011-9572-8
• 34)   C.  Wang,  J.L.  Zhang,  H.S.  Zhang,  K.X.  Jiao,  J.Q.  Yang and  K.C.  Chou: Effect of chlorine on the viscosities and structures of blast furnace slags, Ironmak. Steelmak. 43 (2016) 769–774. doi:10.1080/03019233.2016.1234541
• 35)   T.  Hirosumi and  K.  Morita: Solubility of Chlorine in Aluminosilicate Slag Systems, ISIJ Int. 40 (2000) 943–948. doi:10.2355/isijinternational.40.943
• 36)   M.  Miwa and  K.  Morita: Chloride Capacities of CaO–SiO₂–Al₂O₃(–FeO, MgO, MnO) Slags and Their Application in the Incineration Processes, ISIJ Int. 42 (2002) 1065–1070. doi:10.2355/isijinternational.42.1065
• 37)   K.J.  Myoung,  Y.S.  Lee,  D.J.  Min,  U.C.  Chung and  S.H.  Yi: Thermodynamic Behaviour of Chlorine in CaO-SiO₂-MgO-Al₂O₃(-CaF₂) Slags, Steel Res. Int. 75 (2004) 783–791. doi:10.1002/srin.200405843
• 38)   M.  Okeda,  M.  Hasegawa and  M.  Iwase: Solubilities of Chlorine in CaO-SiO₂-Al₂O₃-MgO Slags: Correlation Between Sulfide and Chloride Capacities, Metall. Mater. Trans. B 42 (2011) 281–290. doi:10.1007/s11663-010-9465-2