Activity coefficient of NiO in SiO2-saturated MnO–SiO2 slag and Al2O3-saturated MnO–SiO2–Al2O3 slag at 1623 K

Guoxing Ren; Songwen Xiao; Caibin Liao; Zhihong Liu

doi:10.1007/s12613-020-2205-y

Volume 29 Issue 2

Jan. 2022

Turn off MathJax

Article Contents

Abstract

1. Introduction

2. Experimental

3. Results and discussion

4. Conclusions

Acknowledgement

Conflict of Interest

References

International Journal of Minerals, Metallurgy and Materials > 2022 > 29(2): 248-255. > DOI: 10.1007/s12613-020-2205-y

Guoxing Ren, Songwen Xiao, Caibin Liao, and Zhihong Liu, Activity coefficient of NiO in SiO₂-saturated MnO–SiO₂ slag and Al₂O₃-saturated MnO–SiO₂–Al₂O₃ slag at 1623 K, Int. J. Miner. Metall. Mater., 29(2022), No. 2, pp.248-255. https://doi.org/10.1007/s12613-020-2205-y

Cite this article as:

PDF (1190 KB)

Research Article

Activity coefficient of NiO in SiO₂-saturated MnO–SiO₂ slag and Al₂O₃-saturated MnO–SiO₂–Al₂O₃ slag at 1623 K

Guoxing Ren^{1, 2)},
Songwen Xiao^2), ,,
Caibin Liao²⁾ and
Zhihong Liu^1), ,

Author Affilications

1)
School of Metallurgy and Environment, Central South University, Changsha 410083, China
2)
Changsha Research Institute of Mining & Metallurgy Co., Ltd, Changsha 410012, China

Corresponding author:

Songwen Xiao E-mail: swinxiao@126.com

Zhihong Liu E-mail: zhliu@csu.edu.cn
Received: 17 May 2020; Revised: 27 September 2020; Accepted: 29 September 2020; Available Online: 30 September 2020

Graphical Abstract

Abstract

Abstract

As a part of the fundamental study related to the reduction smelting of spent lithium-ion batteries and ocean polymetallic nodules based on MnO–SiO₂ slags, this work investigated the activity coefficient of NiO in SiO₂-saturated MnO–SiO₂ slag and Al₂O₃-saturated MnO–SiO₂–Al₂O₃ slag at 1623 K with controlled oxygen partial pressure levels of 10⁻⁷, 10⁻⁶, and 10⁻⁵ Pa. Results showed that the solubility of nickel oxide in the slags increased with increasing oxygen partial pressure. The nickel in the MnO–SiO₂ slag and MnO–SiO₂–Al₂O₃ slag existed as NiO under experimental conditions. The addition of Al₂O₃ in the MnO–SiO₂ slag decreased the dissolution of nickel in the slag and increased the activity coefficient of NiO. Furthermore, the activity coefficient of NiO (γ_NiO), which is solid NiO, in the SiO₂ saturated MnO–SiO₂ slag and Al₂O₃ saturated MnO–SiO₂–Al₂O₃ slag at 1623 K can be respectively calculated as γ_NiO = 8.58w(NiO) + 3.18 and γ_NiO = 11.06w(NiO) + 4.07, respectively, where w(NiO) is the NiO mass fraction in the slag.
- nickel,
- equilibrium,
- MnO–SiO₂ slag,
- MnO–SiO₂–Al₂O₃ slag,
- activity coefficient,
- spent lithium-ion batteries recovery,
- polymetallic nodules

FullText(HTML)

1. Introduction

Social and economic development, especially the electrification of vehicles, will inevitably lead to an enormous increase in nickel demand [1–5]. China, as the world’s biggest producer of lithium-ion batteries (LIBs), is experiencing a serious shortage of nickel resources and thus needs to import them from Indonesia and the Philippines [2]. At the same time, Indonesia, one of the world’s largest exporters of nickel ore, has banned the export of unprocessed nickel ore as part of its effort to develop its domestic smelting industry [2]. Hence, the use of alternative nickel resources has become a realistic and urgent issue in China.

Alternative resources include end-of-use LIBs, which consume about 600 thousand tons of nickel per year [3]. A LIB is typically composed of a metal shell, a cathode, an anode, an organic electrolyte, and a polymer diaphragm. The anode is a metallic copper foil coated with graphite that uses polyvinylidene fluoride as a binder while the cathode is a metallic aluminum foil covered with lithium transition metal oxide powders, such as LiCoO₂, LiMn₂O₄, LiNiO₂, and LiCo_xMn_yNi_zO₂ [1]. The typical elemental composition of LIBs is presented in Table 1. Other potential alternative nickel resources include ocean polymetallic nodules, which are found at a depth of 3000–5000 m in the ocean bed and are called manganese nodules because of their high manganese content [6–14]. In addition to manganese, large amounts of nickel, cobalt, copper, iron, and silicon can be found in ocean polymetallic nodules (Table 2). According to the report by Das [14], nickel availability in ocean polymetallic nodules is nearly five times higher than that in land-based mineral resources. Unlike LIBs, ocean polymetallic nodules mainly contain elements in the form of oxides.

To date, many methods have been developed to extract valuable metals (Ni, Co, and Cu) from ocean polymetallic nodules. These methods include high-pressure sulfuric acid leaching, reduction–ammonia leaching, reduction roasting–ammonia leaching, reduction smelting–sulfurization (alloy-to-matte conversion)–hydrometallurgy, and reduction smelting–hydrometallurgy combinatorial processes. The reduction smelting–hydrometallurgy combinatorial process, in particular, is greatly feasible in the industry because of its relatively high productivity and minimal generation of hazardous leaching residue containing heavy metals [6–14]. For this process, the nodules are first dried and then smelt at 1623–1723 K without the addition of any flux. During smelting, the oxides of Ni, Co, Cu, and Fe in the nodules are reduced by adding cokes as reducing agents. Fe–Co–Cu–Ni alloys are subsequently produced, and silica and manganese oxides are enriched into a MnO–SiO₂ slag. The obtained slag is nearly silica saturated (sat.) due to its high SiO₂ content. After smelting, the obtained Fe–Ni–Cu–Co alloy can be further processed to recover nickel through hydrometallurgical methods, including leaching and solvent extraction.

Although spent LIBs differ from ocean polymetallic nodules in terms of form and element content, they both contain Ni, Co, Cu, and Mn. Thus, the reduction smelting–hydrometallurgy combinatorial process has also been used to extract valuable metals from spent LIBs [4–5,15–17]. During this process, the metallic aluminum foil, graphite anode, and organic materials such as electrolyte and separator are oxidized and used as reducing agents. The oxides of nickel and cobalt in cathode materials are reduced and then enriched into an alloy phase together with a metallic Cu foil at smelting temperatures of 1573–1773 K. Meanwhile, the oxides of manganese and lithium in cathode materials are enriched into a MnO–SiO₂–Al₂O₃ slag phase along with silica flux and alumina generated from the oxidation of the metallic Al foil. The slag obtained from the smelting of spent LIBs is generally alumina saturated due to the extremely high content of Al in batteries.

Table 1. Main elemental composition of lithium-ion batteries [5,15] wt%

Ni	Cu	Co	Al	Mn	C
1–10	8–16	5–20	10–33	0–21	10–17

| Show Table

DownLoad: CSV

Table 2. Typical composition of ocean polymetallic nodules [8,11] wt%

Ni	Cu	Co	Fe	Mn	Zn	Mo	SiO₂	Al₂O₃	CaO
0.90–1.36	0.66–1.17	0.073–0.37	6.20–16.99	20.03–31.23	0.08–0.15	0.04–0.06	12.64–19.22	3.79–5.79	2.27–2.97

| Show Table

DownLoad: CSV

As described previously, a smelting−reduction technology based on a MnO–SiO₂–(Al₂O₃) slag system, as a universal method, can recover nickel from spent LIBs and ocean polymetallic nodules. Maximizing nickel recovery during smelting requires knowledge on the activity coefficient of nickel oxide in SiO₂-saturated MnO−SiO₂ and Al₂O₃-saturated MnO–SiO₂–Al₂O₃ slags.

Unfortunately, most information about the activity coefficient of nickel oxide in slags is focused on FeO_x–SiO₂-based slags (fayalite slags) or FeO–CaO-based slags (calcium ferrite slags), which are commonly utilized in metallurgical processes for traditional minerals. For example, Grimsey [18] measured nickel solubility in SiO₂-saturated FeO_x–SiO₂ slags equilibrated with liquid Ni–Au–Fe alloys at 1523 and 1623 K under oxygen partial pressures in the range of 10⁻⁴ Pa to 10⁻² Pa. Reddy and Acholonu [19] determined nickel solubility in Al₂O₃-saturated FeO_x–SiO₂–Al₂O₃ slags equilibrated with Cu–Ni alloys at 1573 K and oxygen partial pressures in the range of 10⁻⁵ Pa to 10⁻³ Pa. They found that adding Al₂O₃ to iron silicate slag increases the solubility of nickel. Henao et al. [20] studied the activity coefficients of NiO in FeO_x–CaO–MgO slags at a temperature of 1773 K by equilibrating them with slags with Ni–S melt under controlled oxygen partial pressures in the range of 5.1 × 10⁻³ to 1.6 Pa. They found that the activity coefficients of NiO (relative to pure solid NiO) are in the range of 2.7–5.4, which agrees well with the results reported by Takeda et al. [21].

Many studies have also explored the activity coefficient of NiO in MgO-saturated FeO_x–MgO–SiO₂-based slags, which are widely used in the ferronickel smelting process [20,22–24]. The results of these studies show that the activity coefficient of NiO relative to solid NiO ranges from 2 to 6. Henao et al. [20] also determined the activity coefficient of NiO in FeO_x–MgO–SiO₂ slag with low S content and reported that the activity coefficients of NiO relative to solid NiO are 3.1–8.8. Pagador et al. [22] and Henao et al. [23] also studied the effect of adding Al₂O₃ and CaO into FeO_x–MgO–SiO₂ slags on the activity coefficients of NiO. They found that CaO and Al₂O₃ increase the activity coefficients of NiO in FeO_x–MgO–SiO₂ slag.

Other studies also investigated the activity coefficients of NiO in CaO–Al₂O₃-based slags, which were proposed to be applied to the deoxidation process of nickel alloy [25–27]. For instance, Henao et al. [25–26] investigated the activity coefficient of NiO in CaO–Al₂O₃ slag equilibrated with Ni–S alloy or pure nickel metal. They found that the activity coefficients of NiO in CaO–Al₂O₃ slag are greater than those in FeO_x–CaO slag because of the strong interaction between CaO and Al₂O₃.

In the current study, the activity coefficient of nickel oxide in SiO₂-saturated MnO–SiO₂ and Al₂O₃-saturated MnO–SiO₂–Al₂O₃ slags is determined to obtain fundamental data related to the reduction smelting of spent LIBs and ocean polymetallic nodules. The experiments were carried out by equilibrating the slags with Cu–Ni alloy under a controlled oxygen partial pressure with CO–CO₂ gas mixtures at 1623 K.

2. Experimental

The experimental setup used in this study mainly consists of a vertical furnace and a gas purification system (Fig. 1). The furnace was equipped with MoSi₂ heating elements and an alumina reaction tube (outer diameter: 80 mm; inner diameter: 74 mm; height: 850 mm). The temperature of the furnace was controlled within ±2 K by a B-type thermocouple (Pt–6wt% Rh/Pt–30wt% Rh), which was placed between the reaction tube and the heating element. In addition, the temperature of the samples was monitored within ±2 K by using another B-type thermocouple (Pt–6wt% Rh/Pt–30wt% Rh), which is placed close to the samples. Each thermocouple was first calibrated by determining its electromotive force at different temperatures with a multimeter. The accuracy was further checked against the melting point of pure copper, yielding values ±2 K of the accepted melting point (1356 K). CO–CO₂ gas mixtures with a total flow rate regulated at 200 cm³/min were used to control the oxygen partial pressure inside the tube. The CO and CO₂ gases were purified, dewatered, and accurately measured by passing them through a tube furnace containing magnesium chips at 873 K, a silica gel flask, molecular sieve, and a mass flow controller before they were mixed. The oxygen partial pressure was calculated according to the following reactions [24]:

Fig. 1. Schematic of experimental setup.

DownLoad: Full-Size Img PowerPoint

$\begin{aligned}[b] & {\rm{CO}}({\rm{g}}) + 1/2{{\rm{O}}_2}({\rm{g}}) = {\rm{C}}{{\rm{O}}_2}({\rm{g}}),\; \Delta G_1^ \ominus = - 280000\; + \\ & \quad 85.23T\;{\rm{J}}/{\rm{mol}}\end{aligned}$

(1)

${\Delta }{{G}}_{{1}}^{\ominus }=-{RT}{\rm ln}\frac{{(}{{P}}_{{{\rm CO}}_{{2}}}/{{P}}^{\ominus }{)}}{{(}{{P}}_{{\rm CO}}/{{P}}^{\ominus }{)}{{(}{{P}}_{{{\rm O}}_{{2}}}/{{P}}^{\ominus }{)}}^{{1}/{2}}}$

(2)

where ${{P}}_{{\rm O}_{{2}}}$ is the oxygen partial pressure, Pa; ${\Delta }{{G}}_{{1}}^{\ominus }$ represents the standard Gibbs free energy change of the reaction shown in Eq. (1), J/mol; $P^{\ominus}$ is standard atmospheric pressure (101325 Pa); R and T are the gas constant (J·K⁻¹·mol⁻¹) and the absolute temperature (K), respectively; ${{P}}_{{\rm CO}_{{2}}}$ and ${{P}}_{{\rm CO}}$ are the partial pressure of CO₂ (Pa) and CO (Pa), respectively.

Copper–nickel master alloys containing different amounts of nickel were produced in a tube furnace at 1623 K under argon atmosphere by using Ni powders (99.95wt%, supplied by Sinopharm Chemical Reagent Co., Ltd.) and Cu powders (99.95wt%, supplied by Sinopharm Chemical Reagent Co., Ltd.) as the starting materials. The MnO–SiO₂ and MnO–SiO₂–Al₂O₃ master slags were prepared using SiO₂ (99.99wt%, supplied by Shanghai Aladdin Biochemical Technology Co., Ltd.), Al₂O₃ (99.99wt%, supplied by Sinopharm Chemical Reagent Co., Ltd.), and MnO powders as the starting materials. MnO powders were obtained by roasting MnCO₃ powders (99.99wt%, supplied by Shanghai Aladdin Biochemical Technology Co., Ltd.) at 973 K under an argon atmosphere.

High pure silica and alumina crucibles were used for the MnO–SiO₂ slag and MnO–SiO₂–Al₂O₃ slag experiments, respectively. Before the experiments, the reaction tube was washed with purified argon gas, and the crucible containing 1.5 g of Ni–Cu alloy and 3 g of slag (49.66wt% MnO–50.34wt% SiO₂ slag or 30.22wt% MnO–34.58wt% SiO₂–35.20wt% Al₂O₃ slag) were introduced into the hot zone of the furnace at 1623 K. Then, argon flow was stopped, and the mixture of CO and CO₂ was immediately switched into the furnace. After the equilibration was established, the samples were rapidly quenched via immersion into cold water. The solidified slags were analyzed for Ni, Cu, Mn, SiO₂, and Al₂O₃; the alloys were analyzed for Ni, Mn, and Cu. The contents of Ni and Mn in the alloy samples were determined by atomic absorption spectroscopy (AAS). The Cu content in the alloy samples was determined using sodium sulfite titration. The contents of Cu and Ni in the slag samples were measured with AAS. The contents of Mn, SiO₂, and Al₂O₃ in the slag samples were measured using ammonium ferrous sulfate titration, gravimetry, and ethylenediamine tetraacetic acid (EDTA) volumetric method, respectively.

3. Results and discussion

Preliminary experiments were carried out to determine the equilibration time, and the results are shown in Fig. 2. A reaction time of 12 h was sufficient to achieve equilibrium. Thus, all the experiments in this study were set to last for 12 h. The results of this study are summarized in Table 3. Fig. 3 shows a comparison between the obtained slag composition in our study and the SiO₂ and Al₂O₃ saturation lines at 1623 K suggested by Roghani et al. [28]. The compositions of MnO–SiO₂ slag and MnO–SiO₂–Al₂O₃ slag under experimental oxygen partial pressure conditions are near the liquidus line at 1623 K, and they are saturated with silica and alumina, respectively. Some deviations can be explained by the difference between the oxygen partial pressure in this study and in the reported phase diagram [28].

Fig. 2. Content of nickel in slag or alloy as a function of time at 1623 K and an oxygen potential of 10⁻⁵ Pa.

DownLoad: Full-Size Img PowerPoint

Table 3. Compositions of equilibrated alloys and slags at different oxygen partial pressures in 1623 K

${P_{{{\text{O}}_{\text{2}}}}}$ / Pa	Equilibrated alloy / wt%			Equilibrated slag / wt%
${P_{{{\text{O}}_{\text{2}}}}}$ / Pa	Cu	Mn	Ni	Ni	Cu	MnO	SiO₂	Al₂O₃
10⁻⁵	97.21	0.019	2.77	0.0300	0.65	48.64	50.51	—
	94.76	0.020	5.22	0.0570	0.62	48.71	50.45	—
	92.46	0.021	7.52	0.0740	0.56	47.52	51.69	—
	88.47	0.024	11.51	0.0890	0.53	48.33	50.89	—
10⁻⁶	98.02	0.035	1.95	0.0090	0.35	47.38	52.18	—
	94.33	0.041	5.63	0.0200	0.33	48.16	51.40	—
	92.51	0.042	7.45	0.0240	0.33	47.92	51.64	—
	91.90	0.058	8.04	0.0290	0.37	45.40	54.10	—
	88.96	0.059	10.98	0.0340	0.28	48.92	50.68	—
10⁻⁷	97.37	0.200	2.43	0.0035	0.27	47.90	51.75	—
	94.45	0.180	5.37	0.0060	0.22	47.57	52.15	—
	89.22	0.190	10.59	0.0110	0.21	46.41	53.32	—
	84.82	0.240	14.94	0.0130	0.19	48.58	51.17	—
10⁻⁵	97.63	0.050	2.32	0.0150	0.37	30.10	38.51	30.91
	97.54	0.039	2.42	0.0190	0.33	25.38	31.87	42.31
	90.00	0.040	9.96	0.0600	0.30	24.33	30.94	44.28
	86.01	0.066	13.92	0.0730	0.29	23.31	29.97	46.27
10⁻⁶	96.74	0.088	3.17	0.0090	0.10	26.31	32.08	41.47
	93.07	0.096	6.83	0.0180	0.09	27.97	32.44	39.45
	89.21	0.120	10.67	0.0250	0.08	26.54	31.32	42.01
	84.55	0.170	15.28	0.0280	0.15	27.42	34.80	37.56
10⁻⁷	97.37	0.200	2.43	0.0025	0.05	26.98	35.19	37.76
	94.25	0.240	5.51	0.0040	0.14	30.19	38.20	31.43
	89.98	0.250	9.77	0.0080	0.11	29.07	40.02	30.76
	85.57	0.290	14.14	0.0100	0.09	27.76	34.79	37.32

| Show Table

DownLoad: CSV

Fig. 3. Comparison between the obtained slag composition results and the SiO₂ and Al₂O₃ saturation lines at 1623 K suggested by Roghani et al. [28].

DownLoad: Full-Size Img PowerPoint

3.1 Equilibrium content of NiO in slag

The contents of nickel oxide in the MnO–SiO₂ and MnO–SiO₂–Al₂O₃ slags are shown in Fig. 4 in relation to the content of Ni in the equilibrated alloy phase at 1623 K and at various oxygen pressures. The general trend is that the content of nickel in the slags increased as the oxygen partial pressure increased. At a given oxygen partial pressure, the content of nickel oxide also increased with increasing Ni content in the alloy. Furthermore, the solubility of nickel in the MnO–SiO₂ slag was higher than that in the MnO–SiO₂–Al₂O₃ slag at the same oxygen partial pressure.

Fig. 4. Relationships between dissolution of Ni in MnO–SiO₂ and MnO–SiO₂–Al₂O₃ slags and Ni content in equilibrated alloys at 1623 K.

DownLoad: Full-Size Img PowerPoint

Many studies have found that nickel dissolution in slags presents an amphoteric behavior, i.e., nickel dissolves as Ni²⁺ and ${\text{NiO}}_2^{2 - }$ in the acidic and the relatively basic slags, respectively [29–30]. In this study, the mean MnO/SiO₂ mass ratios for the MnO–SiO₂ and MnO–SiO₂–Al₂O₃ slags equilibrated with Ni–Cu alloy under different oxygen pressures at 1623 K were 0.93 and 0.80, respectively. Thus, the nickel in the slags was assumed to be NiO under the experimental conditions. A number of studies have also proposed that if the nickel in a slag exists in the form of NiO, then its solubility should only be a function of the oxygen partial pressure and be equal to 0 at ${P}_{{\text{O}}_{\text{2}}}^{\text{1}/\text{2}}=0$ [19,31–32]. In the current study, the values of (wt% Ni in slag)/a_Ni for the MnO–SiO₂ slag and MnO–SiO₂–Al₂O₃ slag at 1623 K were plotted against ${P}_{{\text{O}}_{\text{2}}}^{\text{1}/\text{2}}$ in Fig. 5, where a_Ni is the activity of Ni in the alloy (relative to the pure liquid Ni). Both plots obviously pass through the origin; thus, the nickel in the MnO–SiO₂ and MnO–SiO₂–Al₂O₃ slags existed as NiO under the experimental conditions.

Fig. 5. Relationship between the values of (wt% Ni in slag)/a_Ni for MnO–SiO₂ slag or MnO–SiO₂–Al₂O₃ slag and the oxygen partial pressure at 1623 K.

DownLoad: Full-Size Img PowerPoint

3.2 Distribution ratio of Ni

The distribution ratio of Ni between the slag and the alloy phases, L_Ni, is defined as follows:

${L}_{\text{Ni}}=\frac{{x}_{\text{NiO}}}{{x}_{\text{Ni}}}$

(3)

where x_NiO and x_Ni denote the mole fractions of NiO in the slag and Ni in the alloy, respectively.

Fig. 6 shows the nickel distribution ratio as a function of oxygen partial pressure at 1623 K. The nickel distribution ratio obviously increases with increasing oxygen partial pressure. This result indicates a linear relationship between $\lg {L_{{\text{Ni}}}}$ and $\lg {P_{{{\text{O}}_{\text{2}}}}}$ with a gradient of about 0.5. It can be explained by the following equilibrium reaction between Ni in the alloy and NiO in the slag [23]:

${\rm{N}}{{\rm{i}}_{({\rm{l}})}} + 1/2{{\rm{O}}_{2({\rm{g}})}} = {\rm{Ni}}{{\rm{O}}_{({\rm{s}})}},\quad \Delta G_4^ \ominus = - 247750 + 92.57T$

(4)

${\Delta }{{G}}_{\text{4}}^{\ominus }=-{RT}\text{ln}\frac{{\gamma}_{\text{NiO(s)}}{x}_{\text{NiO}}}{{\gamma}_{\text{Ni(l)}}{x}_{\text{Ni}}{\text{(}{P}_{{\text{O}}_{\text{2}}}/{P}^{\ominus }\text{)}}^{\text{1}/\text{2}}}$

(5)

where γ_NiO(s) and γ_Ni(l) are the activity coefficients of NiO (relative to the pure solid NiO) and Ni (relative to the pure liquid Ni), respectively.

Fig. 6. Relationship of the distribution ratio of Ni with the oxygen partial pressure at 1623 K.

DownLoad: Full-Size Img PowerPoint

Thus, the logarithm of the distribution ratio of Ni can be written as follows:

$\lg {L_{{\rm{Ni}}}} = \frac{1}{2}\lg {P_{{{\rm{O}}_2}}} - \frac{1}{2}\lg {P^ \ominus } + \lg \left[ {\exp \left( { - \frac{{\Delta G_4^ \ominus }}{{RT}}} \right)} \right] + \lg \frac{{{\gamma _{{\rm{Ni(l)}}}}}}{{{\gamma _{{\rm{NiO(s)}}}}}}$

(6)

where $\Delta {G^\ominus }$ is constant at a given temperature. Therefore, if the ratios of γ_Ni(l) to γ_NiO(s) are constant, then $\lg {L}_{\text{Ni}}$ should increase in proportion to one-half order of $\lg{P}_{{\text{O}}_{\text{2}}}$ . As shown in Fig. 6, many studies have confirmed that the ratios of γ_Ni(l) to γ_NiO(s) remain constant in many systems between slags and alloys, including Au–Ni–Fe, Ni–Fe–Co, and Cu–Ni, in equilibrium with Fe_xO–SiO₂ [18], Fe_xO–MgO(sat.)–SiO₂ [24], and Fe_xO–SiO₂–Al₂O₃ [19], respectively, under wide temperature ranges. Therefore, given the oxidation reaction of nickel and the distribution ratio in Eq. (6), the result shown in Fig. 6 is reasonable.

3.3 Activity coefficient of NiO in slag

As an important thermodynamic parameter, activity coefficient is widely used to estimate the loss of valuable metals into slags. The activity coefficient of oxide in a slag can be calculated according to Eq. (6) and is shown Table 4. The activity coefficient of Ni in Cu–Ni alloy, γ_Ni, was calculated according to Eq. (7) reported by Acholonu without consideration of the effect of Mn due to its low content [33].

${\text {ln }}\gamma _{\text{Ni}}=\frac{\text{1966}}{T}{x}_{\text{Cu}}^{\text{2}}$

(7)

Table 4. Values of activity coefficient of nickel in alloys and slags at 1623 K

Slag system	$P_{{\rm{O}}_2 }$ / Pa	Equilibrated alloy			Equilibrated slag			γ_Ni/γ_NiO
Slag system	$P_{{\rm{O}}_2 }$ / Pa	x_Ni	γ_Ni(l)	a_Ni(l)	x_NiO	a_NiO(s)	γ_NiO(s)	γ_Ni/γ_NiO
MnO–SiO₂	10⁻⁵	0.0299	3.1249	0.0935	0.00033	0.00129	3.87	0.81
		0.0563	2.9397	0.1654	0.00063	0.00227	3.60	0.82
		0.0809	2.7807	0.2250	0.00082	0.00309	3.78	0.74
		0.1234	2.5349	0.3129	0.00099	0.00430	4.36	0.58
	10⁻⁶	0.0211	3.1894	0.0672	0.00010	0.00029	2.94	1.09
		0.0607	2.9088	0.1765	0.00022	0.00077	3.47	0.84
		0.0802	2.7838	0.2232	0.00027	0.00097	3.66	0.76
		0.0865	2.7440	0.2373	0.00032	0.00103	3.23	0.85
		0.1178	2.5633	0.3020	0.00038	0.00131	3.49	0.74
	10⁻⁷	0.0263	3.1366	0.0823	0.00004	0.00011	2.93	1.07
		0.0579	2.9167	0.1688	0.00007	0.00023	3.50	0.83
		0.1136	2.5780	0.2930	0.00012	0.00040	3.32	0.78
		0.1597	2.3389	0.3736	0.00014	0.00051	3.57	0.66
MnO–SiO₂–Al₂O₃	10⁻⁵	0.0250	3.1589	0.0789	0.00019	0.00108	5.83	0.54
		0.0260	3.1519	0.0821	0.00025	0.00113	4.56	0.69
		0.1066	2.6272	0.2800	0.00079	0.00385	4.89	0.54
		0.1485	2.4031	0.3568	0.00097	0.00491	5.08	0.47
	10⁻⁶	0.0341	3.0888	0.1053	0.00012	0.00046	3.92	0.79
		0.0733	2.8233	0.2068	0.00023	0.00090	3.88	0.73
		0.1141	2.5799	0.2943	0.00033	0.00128	3.93	0.66
		0.1628	2.3283	0.3790	0.00036	0.00165	4.62	0.50
	10⁻⁷	0.0261	3.1374	0.0820	0.00003	0.00011	3.54	0.89
		0.0591	2.9038	0.1717	0.00005	0.00024	4.75	0.61
		0.1045	2.6252	0.2744	0.00010	0.00038	3.82	0.69
		0.1507	2.3794	0.3587	0.00013	0.00049	3.87	0.61

| Show Table

DownLoad: CSV

Fig. 7 shows the activity coefficients of NiO as a function of NiO content in MnO–SiO₂ and MnO–SiO₂–Al₂O₃ slags. The activity coefficients of NiO, relative to pure solid NiO, gradually increase as the content of NiO in the slag increases and are independent of the oxygen partial pressure. As expected, the activity coefficients of NiO in the slag containing Al₂O₃ are higher than those in the Al₂O₃-free slag. This result can be explained by considering the fact that manganese aluminate is chemically more stable than nickel aluminate; the respective free energies of formation from the constituent oxides are –22.33 kJ/mol for NiO·Al₂O₃ and –26.63 kJ/mol for MnO·Al₂O₃ at 1623 K. Thus, relative to manganese oxide, nickel oxide may be rejected from the slag in the presence of alumina, resulting in an increase in its activity coefficient. The relationship between the activity coefficient of NiO and the content of NiO as a weight percentage in a slag can be described by the following linear equations:

Fig. 7. Activity coefficient of nickel oxide relative to pure solid nickel oxide vs. NiO content in different slag systems at 1623 K under various oxygen pressures.

DownLoad: Full-Size Img PowerPoint

$\begin{aligned}[b] &{\gamma _{{\rm{NiO(s)}}}} = 8.58w({\rm{NiO}}) + 3.18, {{\rm{\; for\; SiO}}_2}\; {\rm saturated}\;{\rm{MnO}}-{{\rm{SiO}}_2}\; \\ & \quad {\rm slag}\; {\rm at} \;1623\;{\rm{K}}\\[-13pt] \end{aligned}$

(8)

$\begin{aligned}[b] &{{\gamma }_{\rm{NiO(s)}}}=11.06 w({\rm{NiO}}) + 4.07, \; {\rm {for}}\; {\rm Al_2}{\rm O_3}\; {\rm {saturated}}\; {\rm {MnO}}-\\ & \quad {{\rm SiO}_2}-{{\rm Al}_2}{{\rm{O}}_3} \;{\rm slag}\; {\rm at} \;1623\; {\rm K} \\[-12pt] \end{aligned}$

(9)

where w(NiO) is the NiO content in the slag, wt%.

As mentioned above, alumina is a welcome addition to the MnO–SiO₂ slag for the smelting reduction of ocean polymetallic nodules because the higher the activity coefficient of NiO is, the lower the losses to the slag will be. Considering the large amount of metallic Al in spent LIBs, the reduction smelting of ocean polymetallic nodules could be optimized by adding spent LIBs to produce a MnO–SiO₂–Al₂O₃ slag rather than a MnO–SiO₂ slag.

4. Conclusions

In this paper, phase equilibrias between Cu–Ni alloy and SiO₂-saturated MnO–SiO₂ or Al₂O₃-saturated MnO–SiO₂–Al₂O₃ slags were experimentally conducted under controlled oxygen partial pressure levels of 10⁻⁷, 10⁻⁶, and 10⁻⁵ Pa at 1623 K, which is closely related to the reduction smelting of spent LIBs and ocean polymetallic nodules. The main findings can be summarized as follows.

(1) The distribution ratios of Ni and the solubility of nickel oxide in the slags increased with increasing oxygen partial pressure. Furthermore, the solubility of nickel in the MnO–SiO₂ slag was higher than that in the MnO–SiO₂–Al₂O₃ slag at the same oxygen partial pressure.

(2) The addition of Al₂O₃ in the MnO–SiO₂ slag decreased the dissolution of Ni in the slag and increased the activity coefficient of NiO.

(3) The activity coefficient of NiO, which is solid NiO, in the SiO₂ saturated MnO–SiO₂ slag and Al₂O₃ saturated MnO–SiO₂–Al₂O₃ slag at 1623 K can be respectively calculated as γ_NiO(s) = 8.58w(NiO) + 3.18 and γ_NiO(s) = 11.06w(NiO) + 4.07, respectively, where w(NiO) is the NiO content in the slag, wt%.

Acknowledgement

This work was financially supported by the National Natural Science Foundation of China (No. 51704038).

Conflict of Interest

The authors declare no potential conflict of interest.

References (33)

References

[1]	C. Hanisch, J. Diekmann, A. Stieger, W. Haselrieder, and A. Kwade, Recycling of lithium-ion batteries, [in] Handbook of Clean Energy Systems, John Wiley & Sons, Ltd., Chichester, 2015, p. 1.
[2]	A. Home, LME Stock Surge Grounds High-Flying Nickel, But for How Long?, Jan Harve ed. Glacier Media Group, 2020 [2020-1-17]. https://www.mining.com/web/lme-stock-surge-grounds-high-flying-nickel-but-for-how-long/
[3]	NetworkNewsWire, Electric Vehicle Growth Creates East-Asian Battery Mineral Boom, NetworkNewsWire, New York, 2020 [2020-4-10]. https://www.prnewswire.com/news-releases/electric-vehicle-growth-creates-east-asian-battery-mineral-boom-301014990.html
[4]	M.Y. Chen, X.T. Ma, B. Chen, R. Arsenault, P. Karlson, N. Simon, and Y. Wang, Recycling end-of-life electric vehicle lithium-ion batteries, Joule, 3(2019), No. 11, p. 2622. DOI: 10.1016/j.joule.2019.09.014
[5]	G.X. Ren, S.W. Xiao, M.Q. Xie, B. Pan, Y.Q. Fan, F.G. Wang, and X. Xia, Recovery of valuable metals from spent lithium-ion batteries by smelting reduction process based on MnO–SiO₂–Al₂O₃ slag system, [in] Reddy R.G., Chaubal P., Pistorius P.C., Pal U. eds, Advances in Molten Slags, Fluxes, and Salts: Proceedings of the 10th International Conference on Molten Slags, Fluxes and Salts 2016. Springer, Cham, 2016, p. 211.
[6]	P.K. Sen, Metals and materials from deep sea nodules: An outlook for the future, Int. Mater. Rev., 55(2010), No. 6, p. 364. DOI: 10.1179/095066010X12777205875714
[7]	G. Senanayake, Acid leaching of metals from deep-sea manganese nodules – A critical review of fundamentals and applications, Miner. Eng., 24(2011), No. 13, p. 1379. DOI: 10.1016/j.mineng.2011.06.003
[8]	N.S. Randhawa, J. Hait, and R.K. Jana, A brief overview on manganese nodules processing signifying the detail in the Indian context highlighting the international scenario, Hydrometallurgy, 165(2016), p. 166. DOI: 10.1016/j.hydromet.2015.09.013
[9]	E.H. Jeong, C.W. Nam, K.H. Park, and J.H. Park, Sulfurization of Fe–Ni–Cu–Co alloy to matte phase by carbothermic reduction of calcium sulfate, Metall. Mater. Trans. B, 47(2016), No. 2, p. 1103. DOI: 10.1007/s11663-016-0590-4
[10]	S. Agarwal, K.K. Sahu, R.K. Jana, and S.P. Mehrotra, Recovery of Cu, Ni, Co and Mn from sea nodules by direct reduction smelting, [in] Proceedings of the Eighth (2009) ISOPE Ocean Mining Symposium, Chennai, 2009, p. 131.
[11]	D. Friedmann, A.K. Pophanken, and B. Friedrich, Pyrometallurgical treatment of high manganese containing deep sea nodules, J. Sustainable Metall., 3(2017), No. 2, p. 219. DOI: 10.1007/s40831-016-0070-8
[12]	K.D. Mehta, C. Das, and B.D. Pandey, Leaching of copper, nickel and cobalt from Indian Ocean manganese nodules by Aspergillus niger, Hydrometallurgy, 105(2010), No. 1-2, p. 89. DOI: 10.1016/j.hydromet.2010.08.002
[13]	R. Barik, K. Sanjay, B.K. Mishra, and M. Mohapatra, Micellar mediated selective leaching of manganese nodule in high temperature sulfuric acid medium, Hydrometallurgy, 165(2016), p. 44. DOI: 10.1016/j.hydromet.2015.12.005
[14]	S.C. Das, Extraction of metals from polymetallic ocean nodules, [in] Proceeding National Symposium on Chemical and Allied Materials from Ocean, Calcutta, 1989, p. 9.
[15]	S.W. Xiao, G.X. Ren, M.Q. Xie, B. Pan, Y.Q. Fan, F.G. Wang, and X. Xia, Recovery of valuable metals from spent lithium-ion batteries by smelting reduction process based on MnO–SiO₂–Al₂O₃ slag system, J. Sustainable Metall., 3(2017), No. 4, p. 703. DOI: 10.1007/s40831-017-0131-7
[16]	N.S. Randhawa, R.K. Jana, and N.N. Das, Silicomanganese production utilising low grade manganese nodules leaching residue, Miner. Process. Extr. Metall., 122(2013), No. 1, p. 6. DOI: 10.1179/1743285512Y.0000000022
[17]	M. Sommerfeld, D. Friedmann, T. Kuhn, and B. Friedrich, “zero-waste”: A sustainable approach on pyrometallurgical processing of manganese nodule slags, Minerals, 8(2018), No. 12, art. No. 544. DOI: 10.3390/min8120544
[18]	E.J. Grimsey, The effect of temperature on nickel solubility in silica saturated fayalite slags from 1523 to 1623 K, Metall. Trans. B, 19(1988), No. 2, p. 243. DOI: 10.1007/BF02654208
[19]	R.G. Reddy and C.C. Acholonu, Distribution of nickel between copper–nickel and alumina saturated iron silicate slags, Metall. Trans. B, 15(1984), No. 1, p. 33. DOI: 10.1007/BF02661060
[20]	H.M. Henao, M. Hino, and K. Itagaki, Phase equilibrium between Ni–S melt and FeO_X–SiO₂ or FeO_X–CaO based slag under controlled partial pressures, Mater. Trans., 43(2002), No. 9, p. 2219. DOI: 10.2320/matertrans.43.2219
[21]	Y. Takeda, S. Ishiwata, and A. Yazawa, Distribution equilibria of minor elements between liquid copper and calcium ferrite slag, Trans. Jpn. Inst. Met., 24(1983), No. 7, p. 518. DOI: 10.2320/matertrans1960.24.518
[22]	R.U. Pagador, M. Hino, and K. Itagaki, Distribution of minor elements between MgO saturated FeO_x–MgO–SiO₂ or FeO_x–CaO–MgO–SiO₂ slag and nickel alloy, Mater. Trans., JIM, 40(1999), No. 3, p. 225. DOI: 10.2320/matertrans1989.40.225
[23]	H. Henao, M. Hino, and K. Itagaki, Distribution of Ni, Cr, Mn, Co and Cu between Fe–Ni alloy and FeO_x–MgO–SiO₂ base slags, Mater. Trans., 42(2001), No. 9, p. 1959. DOI: 10.2320/matertrans.42.1959
[24]	G.Q. Li and F. Tsukihashi, Distribution equilibria of Fe, Co and Ni between MgO-saturated FeO_x–MgO–SiO₂ slag and Ni alloy, ISIJ Int., 41(2001), No. 11, p. 1303. DOI: 10.2355/isijinternational.41.1303
[25]	H. Henao, M. Hino, and K. Itagaki, Phase equilibrium between Ni–S melt and CaO–Al₂O₃ based slag in CO–CO₂–SO₂ gas mixtures at 1773 K, Mater. Trans., 43(2002), No. 11, p. 2873. DOI: 10.2320/matertrans.43.2873
[26]	H.M. Henao and K. Itagaki, Phase equilibrium and distribution of minor elements between Ni–S melt and Al₂O₃–CaO–MgO slag at 1873 K, Metall. Mater. Trans. B, 35(2004), No. 6, p. 1041. DOI: 10.1007/s11663-004-0060-2
[27]	X. Lu, T. Miki, and T. Nagasaka, Activity coefficients of NiO and CoO in CaO–Al₂O₃–SiO₂ slag and their application to the recycling of Ni–Co–Fe-based end-of-life superalloys via remelting, Int. J. Miner. Metall. Mater., 24(2017), No. 1, p. 25. DOI: 10.1007/s12613-017-1375-8
[28]	G. Roghani, E. Jak, and P. Hayes, Phase equilibrium studies in the “MnO”–Al₂O₃–SiO₂ system, Metall. Mater. Trans. B, 33(2002), No. 6, p. 827. DOI: 10.1007/s11663-002-0066-6
[29]	S.H. Lee, S.M. Moon, D.J. Min, and J.H. Park, Thermodynamic behavior of nickel in CaO–SiO₂–Fe_tO slag, Metall. Mater. Trans. B, 33(2002), No. 1, p. 55. DOI: 10.1007/s11663-002-0085-3
[30]	J.G. Park, H.S. Eom, W.W. Huh, Y.S. Lee, D.J. Min, and I. Sohn, A study in the thermodynamic behavior of nickel in the MgO–SiO₂–FeO slag system, Steel Res. Int., 82(2011), No. 4, p. 415. DOI: 10.1002/srin.201000151
[31]	E.J. Grimsey and X.L. Liu, The activity coefficient of cobalt oxide in silica-saturated iron silicate slags, Metall. Mater. Trans. B, 26(1995), No. 2, p. 229. DOI: 10.1007/BF02660963
[32]	B. Derin and O. Yücel, The distribution of cobalt between Co-Cu alloys and Al₂O₃–FeO–Fe₂O₃–SiO₂ slags, Scand. J. Metall., 31(2002), No. 1, p. 12. DOI: 10.1034/j.1600-0692.2002.310103.x
[33]	C.C. Acholonu, Distribution of Copper, Cobalt, Nickel, Between Alloys and Silica-Unsaturated Iron Slags [Dissertation], University of Nevada, Reno, 1983, p. 9.

[1]	Hendrik Setiawan, Himawan Tri Bayu Murti Petrus, Indra Perdana. Reaction kinetics modeling for lithium and cobalt recovery from spent lithium-ion batteries using acetic acid [J]. International Journal of Minerals, Metallurgy and Materials, 2019, 26(1): 98-107. DOI: 10.1007/s12613-019-1713-0
[2]	Kai-lin Cheng, Dao-bin Mu, Bo-rong Wu, Lei Wang, Ying Jiang, Rui Wang. Electrochemical performance of a nickel-rich LiNi_0.6Co_0.2Mn_0.2O₂ cathode material for lithium-ion batteries under different cut-off voltages [J]. International Journal of Minerals, Metallurgy and Materials, 2017, 24(3): 342-351. DOI: 10.1007/s12613-017-1413-6
[3]	Xin Lu, Takahiro Miki, Tetsuya Nagasaka. Activity coefficients of NiO and CoO in CaO-Al₂O₃-SiO₂ slag and their application to the recycling of Ni-Co-Fe-based end-of-life superalloys via remelting [J]. International Journal of Minerals, Metallurgy and Materials, 2017, 24(1): 25-36. DOI: 10.1007/s12613-017-1375-8
[4]	Tao Li, Juan-yu Yang, Shi-gang Lu. Effect of modified elastomeric binders on the electrochemical properties of silicon anodes for lithium-ion batteries [J]. International Journal of Minerals, Metallurgy and Materials, 2012, 19(8): 752-756. DOI: 10.1007/s12613-012-0623-1
[5]	Xue-min Yang, Wei-jie Zhao, Han-jie Guo, Qiang Zhang, Jian Zhang. A thermodynamic model of calculating mass action concentrations for structural units or ion couples in NaClO₄-H₂O and NaF-H₂O binary solutions and NaClO₄-NaF-H₂O ternary solution [J]. International Journal of Minerals, Metallurgy and Materials, 2010, 17(5): 546-557. DOI: 10.1007/s12613-010-0356-y
[6]	Yan-bin Chen, Yang Hu, Fang Lian, Qing-guo Liu. Synthesis and characterization of spinel Li_1.05Cr_0.1Mn_1.9O_4-zF_z as cathode materials for lithium-ion batteries [J]. International Journal of Minerals, Metallurgy and Materials, 2010, 17(2): 220-224. DOI: 10.1007/s12613-010-0217-8
[7]	Chuangang Lin, Weihua Qiu, Zhimin Li, Jialing Hong, Huanyu Hu, Zhongqin Lu. Comparative study of LiNiO₂, LiNi_0.8Co_0.2O₂ and LiNi_0.75Al_0.25O₂ for lithium-ion batteries [J]. International Journal of Minerals, Metallurgy and Materials, 2002, 9(6): 448-452.
[8]	Bo Song, Seon-Hyo Kim, Jinghong Mao, Qing Lin. Thermodynamic Behavior of Manganese and Phosphorus between Liquid Iron and CaO-MgO-SiO₂-Al₂O₃-Fe_tO-MnO-P₂O₅ Slags [J]. International Journal of Minerals, Metallurgy and Materials, 2001, 8(3): 171-176.
[9]	Yan Feng, Haoran Li, Wenming Zhang. Future Trends of Deep Sea Bed Mining Technology [J]. International Journal of Minerals, Metallurgy and Materials, 1999, 6(1): 4-7.
[10]	ZHANG Fan, XIE Fanyou, CAI Wenjuan, CHOU Kuochih. New formulae for calculating activities from binary system phase diagrams containing solid solution by introducing the parameter θ [J]. International Journal of Minerals, Metallurgy and Materials, 1997, 4(3): 1-4.

Cited By

Cited by

Periodical cited type(3)

1.	Yu Yamashita, Junichi Takahashi, Katsunori Yamaguchi. Distribution Behavior of Nickel, Cobalt, and Copper between Al<sub>2</sub>O<sub>3</sub>–Li<sub>2</sub>O–CaO Slag or Al<sub>2</sub>O<sub>3</sub>–Li<sub>2</sub>O–SiO<sub>2</sub> Slag and Molten Ni–Co–Cu Alloy at 1623 K. MATERIALS TRANSACTIONS, 2025, 66(1): 60. DOI:10.2320/matertrans.MT-M2024120
2.	Minghui Wu, Ying Ren, Lifeng Zhang. In-Situ Observation of Collision and Agglomeration of La2O3–La2S3 Inclusions on Surface of Molten 304 Stainless Steels. Metallurgical and Materials Transactions B, 2024, 55(4): 2324. DOI:10.1007/s11663-024-03095-1
3.	Fan Li, Siyu Xu, Dongjie Zhou, et al. Optimized Conditions for Extracting Fe–Ni–Co–Cu Alloy from Polymetallic Nodules. Journal of Sustainable Metallurgy, 2023, 9(3): 1363. DOI:10.1007/s40831-023-00732-y

Other cited types(0)

Figures(7) / Tables(4)

Cite this Article

PDF

XML

SpringerLink

中文信息

Article Metrics

Article views (1963) PDF downloads (79) Cited by(3)

Activity coefficient of NiO in SiO₂-saturated MnO–SiO₂ slag and Al₂O₃-saturated MnO–SiO₂–Al₂O₃ slag at 1623 K