https://orcid.org/0009-0002-7674-6979
ABSTRACT
The collection of metal resources from seawater desalination brine is a promising technology to achieve a sustainable developing society. The production of freshwater from groundwater polluted by arsenic (As) has potential to satisfy huge water demand. However, conventional methods require large energy consumption and treatment of contaminated wastes. The present study proposes the application of liquid metal tin (Sn) for collecting metallic elements such as sodium (Na), magnesium (Mg), potassium (K), calcium (Ca) and As from the brine and polluted groundwater, in which any waste is not released. The metallic elements were accumulated in liquid Sn pool in the direct contact distillation process of the brine. Each metallic element possessed its own solubility in liquid Sn, which was functioned with the liquid temperature in the range of 505–573 K. K started to precipitate at the early stage and the growth immediately stopped. In the same time, Na started to precipitate and gradually grew. Ca started to precipitate and the growth immediately stopped after K did. Mg could gradually grow. The purification of the polluted water was performed by direct contact reaction between As-polluted water and liquid Sn. The polluted water was efficiently distilled since As was captured by liquid Sn.
HIGHLIGHTS
Liquid Sn technology was newly applied for the collection of metallic elements contained in seawater desalination brine.
Na, Mg, K and Ca were separately precipitated in liquid Sn during the slow cooling process.
As-polluted water was purified by the distillation on liquid Sn at low temperature.
INTRODUCTION
Water scarcity is a critical issue to be addressed in the world (Kummu et al. 2016). Seawater desalination can satisfy the huge demand for freshwater (Curto et al. 2021). However, seawater desalination plants discharge huge amounts of brine as waste (e.g., 141.5 million m3/day (Ihsanullah et al. 2021)), which has a higher concentration of metallic elements than seawater (Jones et al. 2019).
Collection of metal resources from the brine contributes to achieving a sustainable developing society (Bello et al. 2021). However, conventional methods have significant environmental issues (Kumar et al. 2021). An electrodialysis (ED) process consumes large amounts of electricity (e.g., 16 kWh/kg-Li (Liu et al. 2020)). A carbonation process discharges hazardous wastes produced by using absorbent agents (Ihsanullah et al. 2022).
Groundwater is the largest and most reliable source of freshwater, although it is polluted by arsenic (As) (Li et al. 2021). The purification of As-polluted groundwater can satisfy the huge demand for freshwater (Nicomel et al. 2016). However, conventional methods such as absorption and electrocoagulation processes require large energy consumption and treatment of contaminated sludges (Mandal et al. 2016; Xu et al. 2018). The common issue for the distillation of seawater desalination brine and polluted groundwater is the collection and recovery of metallic elements from these feed solutions based on the requirements of environmental protection and water demand.
The seawater desalination concept with the technology of liquid metal tin (Sn: melting point 505 K) was proposed by authors in the previous study (Horikawa & Kondo in press). This concept is based on the direct contact distillation process between seawater and liquid metal Sn. The main energy source for the seawater desalination is concentrated solar power. The previous study clarified the direct contact distillation behavior between artificial seawater and liquid metal Sn at 573 K. The purity of water distilled with liquid metal Sn was sufficiently high. The metallic elements contained in seawater were dissolved into liquid Sn in the direct contact reaction. However, the way to recover the metallic elements dissolved in liquid Sn was not clarified. The necessary challenge identified in the previous study is to clarify the precipitation behavior of the metallic elements dissolved in liquid Sn. The technological limitation in the previous and present studies is in-situ recovery of the metallic elements from the liquid metal pool during the distillation process. Thus, the present study was performed to clarify the precipitation behavior of the metallic elements dissolved in liquid Sn by means of the sequential operation of distillation and cold-trapping processes.
The purpose of the present study is to clarify the accumulation and inherent precipitation behaviors of the metallic elements in liquid Sn pool. The collection and recovery technologies of metallic elements from seawater desalination brine can apply to distill As-polluted groundwater, which are available without consuming large energy and producing wastes.
METHODS
Experimental apparatus
Experimental apparatus used in (a) direct contact reaction process with brine and liquid Sn and (b) cold-trapping process.
Experimental apparatus used in (a) direct contact reaction process with brine and liquid Sn and (b) cold-trapping process.
List of materials of experimental apparatuses
| Components | Materials |
|---|---|
| Feed solutions | Brine, As-polluted water |
| Heated substances | Liquid Sn (Purity: higher than 99.99 wt%), Solid Sn (Purity: higher than 99.99 wt%), 316L plate (Fe-18Cr-12Ni-2Mo) |
| Crucibles (shown in Figures 1 and 2) | 316L (Fe-18Cr-12Ni-2Mo) |
| Nozzle of needle spray gun (shown in Figure 1) | 304 (Fe-18Cr-8Ni) |
| Water tank (shown in Figure 1) | 304 (Fe-18Cr-8Ni) |
| Stainless vessel (shown in Figure 2) | 304 (Fe-18Cr-8Ni) |
| Stainless lid (shown in Figure 2) | 304 (Fe-18Cr-8Ni) |
| Stainless tube (shown in Figure 2) | 316 (Fe-18Cr-12Ni-2Mo) |
| Silicon plug (shown in Figure 2) | Silicon rubber |
| Components | Materials |
|---|---|
| Feed solutions | Brine, As-polluted water |
| Heated substances | Liquid Sn (Purity: higher than 99.99 wt%), Solid Sn (Purity: higher than 99.99 wt%), 316L plate (Fe-18Cr-12Ni-2Mo) |
| Crucibles (shown in Figures 1 and 2) | 316L (Fe-18Cr-12Ni-2Mo) |
| Nozzle of needle spray gun (shown in Figure 1) | 304 (Fe-18Cr-8Ni) |
| Water tank (shown in Figure 1) | 304 (Fe-18Cr-8Ni) |
| Stainless vessel (shown in Figure 2) | 304 (Fe-18Cr-8Ni) |
| Stainless lid (shown in Figure 2) | 304 (Fe-18Cr-8Ni) |
| Stainless tube (shown in Figure 2) | 316 (Fe-18Cr-12Ni-2Mo) |
| Silicon plug (shown in Figure 2) | Silicon rubber |
Apparatus used in direct contact reaction experiments with As-polluted water and liquid Sn.
Apparatus used in direct contact reaction experiments with As-polluted water and liquid Sn.
Experimental conditions
The direct contact reaction between brine and liquid Sn was investigated by means of the experiment at the conditions presented in Table 2. The chemical composition of brine used in the current work is presented in Table 3. The brine contains chlorine (Cl), sodium (Na), magnesium (Mg), sulfur (S), potassium (K) and calcium (Ca). The purity of Sn is more than 99.99 wt%. The brine was continuously sprayed onto the free surface of liquid Sn at 573 K for 500 min under air atmosphere at atmospheric pressure. The spray rate was 0.2 mL/min, and the distilled water of approximately 100 mL was produced in each experiment. The metallic elements originally contained in the brine were dissolved into liquid Sn in this direct contact process.
Experimental conditions
| Direct contact reaction process | Cold-trapping process | ||||
|---|---|---|---|---|---|
| Liquid metal | Temperature [K] | Spray rate [mL/min] | Spray duration [min] | Cooling rate [K/s] | Cooling duration [s] |
| Sn | 573 K | 0.2 | 500 | 1.4 | 4.9 × 101 |
| 1.2 × 10−1 | 6.9 × 102 | ||||
| 4.9 × 10−2 | 1.1 × 103 | ||||
| 8.3 × 10−3 | 8.6 × 103 | ||||
| Direct contact reaction process | Cold-trapping process | ||||
|---|---|---|---|---|---|
| Liquid metal | Temperature [K] | Spray rate [mL/min] | Spray duration [min] | Cooling rate [K/s] | Cooling duration [s] |
| Sn | 573 K | 0.2 | 500 | 1.4 | 4.9 × 101 |
| 1.2 × 10−1 | 6.9 × 102 | ||||
| 4.9 × 10−2 | 1.1 × 103 | ||||
| 8.3 × 10−3 | 8.6 × 103 | ||||
Chemical composition of brine used in the current work (unit: mg/L)
| Cl | Na | Mg | S | K | Ca |
|---|---|---|---|---|---|
| 3 × 104 | 1.53 × 104 | 1.98 × 103 | 1.32 × 103 | 5.82 × 102 | 4.52 × 102 |
| Cl | Na | Mg | S | K | Ca |
|---|---|---|---|---|---|
| 3 × 104 | 1.53 × 104 | 1.98 × 103 | 1.32 × 103 | 5.82 × 102 | 4.52 × 102 |
The metallic elements dissolved in liquid Sn were precipitated in the cold-trapping experiments. The total number of the cold-trapping experiments performed in the present study was four, as presented in Table 2. The liquid Sn after the direct contact reaction process with the brine was cooled from 573 to 505 K at the constant cooling rate of 1.4, 1.2 × 10−1, 4.9 × 10−2 and 8.3 × 10−3 K/s. The cooling rate of 1.4 K/s was achieved by cooling with ice water. The other cooling rates were fixed by the balance between the natural cooling and the heating power of the plate heater.
The concentration of As in artificial As-polluted water was 1,000 mg/L. As-polluted water was prepared with sodium hydroxide (NaOH) solution and hydrogen chloride (HCl) solution using arsenic trioxide (As2O3). The droplets of As-polluted water were continuously dripped onto the free surface of liquid Sn for 50 min. The experiments were performed under air atmosphere, and the temperature of liquid Sn was 523, 573 and 623 K. The dripping rate of As-polluted water was 0.1 mL/min, and the distilled water of approximately 5 mL was produced in each experiment. The direct contact experiments with solid Sn and 316L plates were also performed up to 573 K for comparison.
The Sn samples were embedded in resin for further metallurgical analysis, which were exposed to the brine in the direct contact experiments and cooled in the cold-trapping experiments. The Sn samples were cut at the center in a longitudinal direction by using a jigsaw, and its cross-section was sequentially polished with silicon carbide (SiC) papers of #500, #1200, #2000 and #4000 with ethanol as a lubricant. The cross-section of the Sn samples was observed and analyzed with an electron probe microanalyzer (EPMA, EPMA-1610; SHIMADZU; Japan).
The impurity concentration of water distilled by the direct contact between As-polluted water and liquid Sn was measured with an inductively coupled plasma optical emission spectrometer (ICP-OES, 5100 VDV ICP-OES; Agilent Technology; USA). The surface of Sn and 316L plate used in the experiments was observed and analyzed with a scanning electron microscope (SEM, JSM-7200F; JEOL Ltd; Japan) with an energy-dispersive X-ray spectrometer (EDX, EX-37270VUP; JEOL Ltd; Japan). The chemical compounds formed on the surface of Sn and 316L plate used in the experiments were identified by a thin film X-ray diffraction measurement (XRD, X’ Pert-Pro-MRD; Maven Panalytical; England).
RESULTS AND DISCUSSION
Precipitation of metallic elements in liquid Sn pool
Cross-sectional photos of Sn cooled at (a) 1.4 K/s, (b) 1.2 × 10−1 K/s, (c) 4.9 × 10−2 K/s and (d) 8.3 × 10−3 K/s after direct contact with brine.
Cross-sectional photos of Sn cooled at (a) 1.4 K/s, (b) 1.2 × 10−1 K/s, (c) 4.9 × 10−2 K/s and (d) 8.3 × 10−3 K/s after direct contact with brine.
Relationship between cooling rate and area occupied by aggregate of precipitations in cross-sectional image of solidified Sn.
Relationship between cooling rate and area occupied by aggregate of precipitations in cross-sectional image of solidified Sn.
Cross-sectional EPMA analysis on metallic elements precipitated in liquid Sn cooled at 8.3 × 10−3 K/s: (a) Sn, (b) Na, (c) Mg, (d) K, (e) Ca and (f) Cl.
Cross-sectional EPMA analysis on metallic elements precipitated in liquid Sn cooled at 8.3 × 10−3 K/s: (a) Sn, (b) Na, (c) Mg, (d) K, (e) Ca and (f) Cl.
Table 4 indicates the enrichment behavior of Mg, K and Ca in the precipitations obtained after the cold-trapping process at the cooling rate of 8.3 × 10−3 K/s. The atomic ratio between Ca and Na in the Ca-rich precipitations (at the point 1 in Figure 5(a)) was approximately 1.9 × 101. This Ca/Na ratio was 8.6 × 102 times larger than that in brine. Ca and Na were separately precipitated in the cooling process. Mg was involved in the Ca-rich precipitations. The atomic ratio of Ca/Sn was approximately 1.9. The data indicated that Sn was slightly involved in the Ca-rich precipitations.
Atomic ratio of metallic elements based on results of EPMA point analysis to explain concentration of Mg, K and Ca precipitated in liquid Sn after cold-trapping process at cooling rate of 8.3 × 10−3 K/s (EPMA analysis point was indicated in Figure 5(a))
| Point | Ca/Na | Mg/Na | K/Na | Na/Sn | Ca/Sn | Mg/Sn | K/Sn | |
|---|---|---|---|---|---|---|---|---|
| Brine | – | 2.2 × 10−2 | 1.2 × 10−1 | 1.6 × 10−2 | – | – | – | – |
| Precipitations | 1 (Ca-rich) | 1.9 × 101 | 5.3 | 3.3 × 10−1 | – | 1.9 | – | – |
| 2 (Na-rich) | 4.4 × 10−2 | 1.2 × 10−1 | 4.4 × 10−3 | 1.5 × 101 | – | – | – | |
| 3 (Mg-rich) | 3.4 | 1.9 × 102 | 7.1 × 10−1 | – | – | 1.7 × 101 | – | |
| 4 (K-rich) | 6.3 | 4.7 × 101 | 1.8 × 101 | – | – | – | 2.0 |
| Point | Ca/Na | Mg/Na | K/Na | Na/Sn | Ca/Sn | Mg/Sn | K/Sn | |
|---|---|---|---|---|---|---|---|---|
| Brine | – | 2.2 × 10−2 | 1.2 × 10−1 | 1.6 × 10−2 | – | – | – | – |
| Precipitations | 1 (Ca-rich) | 1.9 × 101 | 5.3 | 3.3 × 10−1 | – | 1.9 | – | – |
| 2 (Na-rich) | 4.4 × 10−2 | 1.2 × 10−1 | 4.4 × 10−3 | 1.5 × 101 | – | – | – | |
| 3 (Mg-rich) | 3.4 | 1.9 × 102 | 7.1 × 10−1 | – | – | 1.7 × 101 | – | |
| 4 (K-rich) | 6.3 | 4.7 × 101 | 1.8 × 101 | – | – | – | 2.0 |
The concentration of Ca, Mg and K was low in the Na-rich precipitations (at the point 2 in Figure 5(a)). The atomic ratio of Na/Sn was approximately 15, and this data indicated that Sn was not involved in the Na-rich precipitations.
The atomic ratio between Mg and Na in the Mg-rich precipitations (at point 3 in Figure 5(a)) was approximately 1.9 × 102. This Mg/Na ratio was 1.6 × 103 times higher than that of brine. The atomic ratio of Mg/Sn was approximately 17. The Ca-rich precipitations could slightly involve CaSn3 according to the phase diagram of the Ca–Sn system (Palenzona et al. 2000). The Mg-rich precipitations could slightly involve Mg2Sn according to the phase diagram of Mg–Sn system (Nayeb-Hashemi & Clark 1984). The atomic ratio of Sn/Mg in Mg2Sn is 0.5, which is 6 times lower than that of Sn/Ca in CaSn3. Thus, the Mg-rich precipitations did not involve Sn.
The atomic ratio between K and Na in the K-rich precipitations (at the point 4 in Figure 5(a)) was approximately 1.8 × 101. This K/Na ratio was 1.1 × 103 times higher than that of brine. The atomic ratio of K/Sn was approximately 2.0. Sn was involved in the K-rich precipitations. These data indicated that Na, Mg, K and Ca were separately precipitated in the cooling process.
Discussion on inherent precipitation behavior of mineral elements in liquid Sn pool
Values of A and B for solubility equation of Na, Mg, K and Ca in liquid Sn
| Elements | A | B | References |
|---|---|---|---|
| Na | 4.8 | − 2.4 × 103 | Sangster & Bale (1998a) |
| Mg | 2.2 | − 9.1 × 102 | Nayeb-Hashemi & Clark (1984) |
| K | 4.1 | − 2.7 × 103 | Sangster & Bale (1998b) |
| Ca | 3.3 | − 2.1 × 103 | Palenzona et al. (2000) |
| Elements | A | B | References |
|---|---|---|---|
| Na | 4.8 | − 2.4 × 103 | Sangster & Bale (1998a) |
| Mg | 2.2 | − 9.1 × 102 | Nayeb-Hashemi & Clark (1984) |
| K | 4.1 | − 2.7 × 103 | Sangster & Bale (1998b) |
| Ca | 3.3 | − 2.1 × 103 | Palenzona et al. (2000) |
Schematic illustration to explain possible precipitation behavior of Na, Mg, K and Ca in liquid Sn pool: (a) accumulation of Na, Mg, K and Ca in liquid Sn during direct contact process with brine, (b) precipitation of K in cooling process, (c) formation and growth of Na precipitations in cooling process, (d) formation and growth of Mg precipitations in cooling process and (e) precipitation of Ca in cooling process.
Schematic illustration to explain possible precipitation behavior of Na, Mg, K and Ca in liquid Sn pool: (a) accumulation of Na, Mg, K and Ca in liquid Sn during direct contact process with brine, (b) precipitation of K in cooling process, (c) formation and growth of Na precipitations in cooling process, (d) formation and growth of Mg precipitations in cooling process and (e) precipitation of Ca in cooling process.
When all the metallic elements in the brine sprayed onto liquid Sn are assumed to be dissolved into the melt during the direct contact process, the concentrations of Na, Mg, K and Ca (i.e., CNa, CMg, CK and CCa) are estimated as 6.9, 0.93, 0.28 and 0.21 wt%, respectively. The possible quantities of the Na, K and Ca precipitations (i.e., ΔmNa, ΔmK and ΔmCa) in maximum are estimated as 1.2 × 103, 4.2 × 101 and 1.3 × 101 mg, respectively. The concentration of Mg in liquid Sn might not be higher than its solubility based on the theoretical model and the solubility. The total quantity of the precipitations estimated by Equations (1) and (2) indicated that Na was dominant and the quantity of Ca and K precipitations was small. This estimation agreed well with that recognized in the results of EPMA analysis as shown in Figure 5.
The metallic elements started to precipitate at the temperature when their concentrations became higher than their inherent solubilities in liquid Sn due to the temperature decrease. K started to precipitate at the early stage since its concentration became higher than the solubility during the cooling process. The growth immediately stopped as shown in Figure 6(b) according to its low concentration in liquid Sn. In the same time, Na started to precipitate as shown in Figure 6(c). The Na-rich precipitations gradually grew during the cooling process as shown in Figure 6(d). The boundary between the precipitations and Sn was shifted according to their growth. Small K-rich precipitations were then located at a distance from the boundary. The concentration of Ca was lower than that of K in liquid Sn. The solubility of Ca was higher than that of K in liquid Sn. Ca started to precipitate after the precipitation of K and the growth stopped immediately as shown in Figure 6(e) according to its low concentration in liquid Sn. The volume of the Mg-rich precipitations was as large as that of Na-rich precipitations. The Mg-rich precipitations were detected between the K-rich and Ca-rich precipitations. Mg could start to precipitate and grow during the cooling process as shown in Figure 6(d) and 6(e).
Chemical interaction between As-polluted water and liquid Sn
(a) Concentrations of As and Na in water distilled by direct contact experiments with As-polluted water and (b) ratio between concentration of As to that of Na.
(a) Concentrations of As and Na in water distilled by direct contact experiments with As-polluted water and (b) ratio between concentration of As to that of Na.
Surface SEM/EDX analysis on Sn and 316L plate used in direct contact experiments with As-polluted water: (a) Sn used at 573 K and (b) 316L plate used at 573 K.
Surface SEM/EDX analysis on Sn and 316L plate used in direct contact experiments with As-polluted water: (a) Sn used at 573 K and (b) 316L plate used at 573 K.
XRD analysis on surface of Sn and 316L plate used in direct contact experiments with As-polluted water: (a) Sn used at 573 K and (b) 316L plate used at 573 K.
XRD analysis on surface of Sn and 316L plate used in direct contact experiments with As-polluted water: (a) Sn used at 573 K and (b) 316L plate used at 573 K.
As and O atoms could be dissolved into liquid Sn. As was captured in liquid Sn and did not evaporate during the distillation experiments. Thus, the As concentration in the distilled water became low when the distillation process was performed with liquid Sn. However, the contamination of the distilled water with As was detected. The oxidation of liquid Sn surface was promoted according to Equation (4) and the chemical reaction with As2O3 became weak. As2O3 could not be reduced. As2O3 evaporated at higher temperature and transferred to the distilled water due to its low boiling point at 738 K. The As captured in liquid Sn may be separately precipitated in the cold-trapping method explained in the previous chapter.
CONCLUSIONS
Major conclusions are follows:
(1) The enrichment and precipitation of mineral elements such as Na, Mg, K and Ca contained in brine were demonstrated by means of the cold-trapping experiments with liquid Sn pool. Na, Mg, K and Ca contained in brine were dissolved and accumulated in liquid Sn during the direct contact reaction process. The metallic elements were precipitated from liquid Sn cooled at a lower cooling rate. The metallic elements of Na, Mg, K and Ca were separately precipitated in the liquid Sn pool.
(2) The distillation of As-polluted water was demonstrated by means of the direct contact reaction experiments with liquid Sn. The distillation process at higher temperature induced the advection of As according to the evaporation. However, the evaporation of As was suppressed when As contained in water was captured by liquid Sn.
ACKNOWLEDGEMENTS
Authors would acknowledge Mr Toshihiko Hara of Institute of Science Tokyo for his technical support on the preparation of the specimens and the experimental apparatus. Authors would acknowledge the staff of Materials Analysis Division, Core Facility Center of Institute of Science Tokyo for their technical support on the analysis. Authors would like to thank Dr Kazuki Matsuo of EX-Fusion Co., Ltd for his fruitful comments from industrialization of the distillation systems. Authors would like to acknowledge Asahi Organic Chemicals Industry Co., Ltd for providing seawater desalination brine used in the current work.
FUTURE REMARKS
The steady-state precipitation of metallic elements in liquid Sn circulating systems will be clarified by means of the experiments using a non-isothermal flowing apparatus. The optimum temperature distribution in the cold-trapping system will also be clarified to improve the separation and recovery of metallic elements dissolved in the fluid. The pilot plant based on this liquid Sn direct contact process will be designed, which can contribute to the large-scale distillation of brine and polluted water.
FUNDING
This paper is partially based on results obtained from a project, JPNP20004, subsidized by the New Energy and Industrial Technology Development Organization (NEDO). This work was supported by JST SPRING, Japan Grant Number JPMJSP2106 and JPMJSP2180.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.
