Experimental Study on Brine Storage for Overwintering by Using Salinity-Gradient Solar Pond in Zabuye Salt Lake, Tibet

Abstract

Known as the “white oil”, lithium is a key raw material to support strategic emerging industries and future industrial development. Zabuye Salt Lake is the only one in Tibet, China that has so far realized the industrialization of lithium extraction from the salt lake brine. The green and low-cost lithium extraction technology by using the salinity-gradient solar pond (SGSP) adopted has always been paid much attention by lithium-related practitioners and researchers. In order to improve the lithium yield and grade of a single crystallization pond, the cross-year brine mixing method can be used to increase the initial temperature and CO₃²⁻ concentration of the raw brine for making the SGSP. The premise is to ensure that the summer brine with low Li⁺ and high CO₃²⁻ prepared in the previous year could be stored safely for overwintering with a minimal change in brine composition, for use in brine mixing in February and March of the next year, which can be realized by using the SGSP. In this paper, two experiments of brine storage for overwintering were carried out in the Zabuye mining area, Tibet in 2020 and 2021 by using the large-scale SGSP with an area of nearly 4000 m². The results show that during the operation of the SGSP in winter, the brine temperature in the lower convective zone (LCZ) can still rise to more than 20 °C and remain relatively stable, indicating that the coverage of surface ice layer not only has an effect of heat preservation and insulation on the SGSP, but also plays a positive role in the thermal storage capacity of the SGSP. The vertical distributions of brine temperature, density and salinity in the pond showed the ideal gradient curves increasing from top to bottom, and the concentrations of Li⁺ and CO₃²⁻ in the brine only decreased slightly. The structure of the salinity-gradient layer tended to stabilize faster when the brine filling depth was larger, but the boundary between the upper convective zone (UCZ) and the non-convective zone (NCZ) was relatively blurred. It is completely feasible to store the brine for overwintering by using the SGSP in the Zabuye mining area, and the experimental results could be directly scalable to larger industrial applications. It can not only provide high-quality raw brine for cross-year brine mixing, but also reduce the pressure of brine production, and a small amount of lithium mixed salt collected is helpful to increase the output of a single crystallization pond. Additionally, the potential challenges of maintaining the SGSP system during extreme winter conditions are described, and effective measures and suggestions are proposed to make the technology feasible in diverse climates.

1. Introduction

The salinity-gradient solar pond (SGSP) is a saltwater pond with a salt concentration gradient, which also serves as a solar energy collection and storage device. Due to its simple structure, low cost, ease of large-scale construction and zero-pollution emission, it is considered to be a low-temperature heat source equipment that can store solar energy on a large scale and for a long time and has the most promising application prospect in the future [1]. In recent years, the SGSP has been increasingly favored by the majority of new energy developers and widely used in fields such as power generation, heating, salt production and seawater desalination [2,3]. Scholars from various countries have carried out a large number of theoretical, experimental and applied research on the SGSP and related technologies, and accumulated rich scientific achievements in the structural design, performance optimization and application exploration of the SGSP [4,5,6].

In recent years, with the rapid development of new energy vehicles and energy storage technology, the importance of lithium resources has become increasingly prominent. Driven by the global new energy strategy and supply–demand relationship, the price of lithium carbonate has been rapidly increasing since the end of 2020, and has skyrocketed in the second half of 2021. The price of lithium carbonate started at around USD 5500 per ton and climbed to about USD 68,000 per ton in the first quarter of 2022. In 2024, the price of lithium carbonate gradually fell and continued to hover at a low level, so far it has fallen to USD 10,000 per ton. Given the increasing global demand for lithium, especially for electric vehicle batteries and renewable energy storage, fluctuations in the price of lithium may significantly affect the feasibility and profitability of the extraction process. In order to enhance the competitiveness of the lithium industry, the economic viability of lithium extraction methods plays a crucial role and also prompts the continuous optimization and upgrading of the existing lithium extraction processes.

Zabuye Salt Lake in Tibet is the only industrial application of lithium extraction from salt lake brine by using the SGSP in China so far. Zabuye Salt Lake itself is a relatively rare high-quality carbonate-type salt lake, but it is not conducive to the industrial production of high-grade lithium mixed salt due to the lithium carbonate is easily dispersed and precipitated at various stages of solar evaporation. And the SGSP can subtly solve this problem [7]. Its main function is to extract the lithium carbonate from the lower brine by raising the brine temperature with solar energy, involving the sedimentation of varied salts and the diffusion of ions. Lithium carbonate is a kind of warm salt mineral in terms of mineralogy, whose solubility declines with the rise in temperature. The temperature rising of the lower brine and the protection of the salinity-gradient layer can make the lithium carbonate precipitate out intensively in the bottom of the pond. The salinity-gradient layer not only preserves heat but also hinders the evaporation of the lower brine, thus separating out warm salt minerals in the brine while retaining other cold salt minerals such as NaCl and KCl. Finally, the lithium carbonate could be precipitated. The SGSP in production operation in the Zabuye mining area is divided into three layers from top to bottom. The upper layer is the upper convective zone (UCZ) or freshwater layer, with a temperature similar to the ambient temperature. It is susceptible to water evaporation and other external disturbances like wind and rain and mainly plays a role in forming and protecting the middle salinity-gradient layer. The middle layer is the non-convective zone (NCZ) or salinity-gradient layer, where the salt concentration increases continuously with depth. On the one hand, it can prevent heat loss from the pond surface; on the other hand, it can also store the thermal energy in the bottom brine by using the difference in refractive index between freshwater and brine. The lower layer is the lower convective zone (NCZ) or energy storage layer, which is composed of high concentration or saturated salt solution, with a much higher temperature than that of UCZ. It mainly serves the function of collecting and storing solar energy and raising the temperature of brine [8,9]. Due to the existence of the NCZ, the brine in the LCZ evaporates slowly, and it is easier for lithium carbonate to precipitate in large quantities and improve the grade at the bottom of the pond, while other salts are difficult to precipitate. Under the natural environmental conditions of no electricity or mineral energy in the plateau lake area, the low-cost green industrial development of lithium resources in Zabuye Salt Lake brine has been successfully realized by using the lithium extraction technology of the SGSP.

The production practice of the Zabuye mining area for many years shows that if the initial temperature and CO₃²⁻ concentration of the raw brine for making the SGSP are low, the lithium yield and grade of the single crystallization pond will be seriously affected [10]. In the actual operation process, the raw brine poured into the crystallization pond is usually the winter brine produced in February and March of the year after solar evaporation and concentration of multi-stage salt fields, and the initial brine temperature is very low, which needs to undergo a long time of heat accumulation and temperature rise in the SGSP. In addition, although the concentration of Li⁺ in the brine has been enriched to a higher level at this time, the CO₃²⁻ concentration is still low, which will inevitably cause the incomplete precipitation of lithium carbonate; that is, there is still a certain concentration of Li⁺ in the tail brine discharged after salt precipitation, reducing the production efficiency and lithium yield of the single crystallization pond.

The brine mixing method is a common operation in salt field production, mainly through the mixing of brine with different components to obtain the target brine that meets the subsequent production requirements. In the Zabuye mining area, the cross-year brine mixing method is often adopted to improve the quality of raw brine. The summer brine with low Li⁺ and high CO₃²⁻ produced in the salt field before the rainy season (from July to September each year) is mixed with the winter brine with high Li⁺ and low CO₃²⁻ obtained in February and March of the next year in a certain proportion, and then poured into the crystallization pond as the raw brine to make the SGSP. The production test results show that the cross-year brine mixing method can not only significantly increase the initial temperature and CO₃²⁻ concentration of the raw brine, effectively improving the lithium yield and grade of the crystallization pond, but also alleviate the brine production pressure. However, it is worth noting that the implementation of the cross-year brine mixing method first needs to ensure that the summer brine with low Li⁺ and high CO₃²⁻ prepared in the previous year can be safely stored for overwintering and that the brine components do not change significantly, as much as possible, for use in the brine mixing operation in February and March of the next year.

Considering the characteristic of cross-season heat storage in the SGSP, even in winter, the bottom of the pond can still maintain a certain temperature, and the freshwater layer laid on the surface of the brine can prevent the evaporation of the lower brine to avoid the crystallization of salt minerals. Therefore, the SGSP can be used to store the summer brine for overwintering, which can not only maintain the relative stability of the brine components, but also effectively utilize part of the residual heat in the LCZ of the SGSP. Due to that, the brine itself has a certain initial temperature when used as the raw brine for mixing in the next year, the heat accumulation process and the production cycle of the SGSP can also be greatly shortened.

However, it is an unprecedented challenge to construct and operate the SGSP in the cold and windy climate of the Tibetan Plateau lake area in winter. In a general sense, when the ambient temperature around the SGSP drops low enough, the freshwater layer on the surface of the pond will freeze, resulting in a diminishing of the solar energy transmittance and the heat storage capacity of the SGSP [11]. But, some researchers have argued that the SGSP can still economically provide useful heat even in regions with a freezing period [12]. Li Nan has simulated the thermal performance inside the SGSP based on the actual weather and icing conditions in winter and concluded that the existence of the ice layer is not only conducive to the thermal insulation of the SGSP, but also can inhibit the increase in salinity in the UCZ caused by the salt diffusion, effectively maintaining the stability of the SGSP [13,14].

At present, the heat accumulation and temperature rise performance of the SGSP is still unclear in winter with lower solar radiation and air temperature, especially when the surface layer of the pond is iced over. Most of the studies on the effect of surface icing on the performance of the SGSP are limited to laboratory models or small solar ponds, and the actual operation of the large-scale SGSP in winter has not been systematically reported [13]. In this paper, an experimental study on brine storage for overwintering by using the large-scale SGSP was carried out in the Zabuye mining area, Tibet. The longitudinal distribution and change in physicochemical parameters such as the brine temperature, salinity and Li⁺ and CO₃²⁻ concentrations during the whole winter operation of the SGSP were measured and investigated in detail. The actual effect and feasibility of brine storage for overwintering by using the SGSP as the raw brine for use in the cross-year brine mixing operation was discussed and clarified, providing a scientific basis and beneficial suggestions for the optimization and implementation of the green lithium extraction technology from Zabuye Salt Sake brine. Compared with the existing research on using the SGSP for lithium extraction, this study innovatively uses the SGSP for brine storage overwintering, which not only makes ingenious use of the abundant solar energy and cold resources in the Tibetan Plateau, but also effectively uses part of the waste heat of SGSP, shortening the production cycle and improving the production efficiency of lithium extraction. This technology has broader implications and promotion prospects for the efficient and green utilization of salt lake resources and renewable energy in the Tibetan Plateau.

2. Overview of Study Area

Zabuye Salt Lake is located in the western hinterland of the Tibetan Plateau and belongs to Zhongba County, Shigatse Prefecture, Tibet Autonomous Region, China. Its geographical coordinates are 83°52′34″−84°23′47″ E and 31°14′47″−31°33′10″ N (Figure 1). Zabuye Salt Lake has an elevation of 4421 m and an area of 243 km². It is divided into two lakes, north and south, by a nearly east–west sandbank. The south lake is a semi-dry salt lake, and the north lake is a brine lake. The Zabuye Lake area belongs to the plateau subfrigid arid climate zone, with an annual evaporation of 2424.8 mm and an annual precipitation of 115.5 mm, of which the rainy season is mainly concentrated in July and August. The lake area has adequate sunshine, with an average annual sunshine duration of 3122 h, and a large temperature difference between day and night. Zabuye Salt Lake is the third largest lithium salt lake in the world and the largest in Asia, as well as a rare large-scale comprehensive salt lake deposit of lithium, boron, potassium and cesium. The proven reserves of lithium resource are 1.841 million tons, including 818,800 tons of liquid ore and 1.0222 million tons of solid ore. The liquid deposits contain surface brine and intergranular brine. The brine belongs to the moderate carbonate type, with a salinity fluctuating between 300 and 450 g/L, a pH of 8.7 to 9.5 and a density ranging from 1.24 to 1.26 g/cm³. The lithium grade in the brine reaches 0.12%, second only to Atacama Salt Lake of Chile (0.15%), in the world. It has unique natural solid resources of lithium carbonate and the characteristics of being lithium-rich and magnesium-poor. The magnesium–lithium ratio of 0.0036 is the lowest in the world, and the brine is close to or has reached the saturation point of lithium carbonate, making it easy to form different forms of natural lithium carbonate deposits. Therefore, it has better resource advantages than similar salt lakes [7,15].

Based on the hydrochemical characteristics of Zabuye Salt Lake brine, Zheng Mianping and his team cleverly utilized the natural environment advantages of the lake area, such as high and cold, drought and strong sunshine, and pioneered the low-cost green lithium extraction technology of “remove nitrate and alkali by freezing-evaporation and concentration-temperature rising and lithium precipitation by SGSP”, which took the lead in realizing the industrial production of lithium carbonate in domestic salt lakes. This process route makes full use of the significant advantages of abundant solar energy and geographical conditions for building salt fields on the Tibetan Plateau, which is very suitable for the Tibet Plateau where there is no electricity, no fossil energy, inconvenient transportation and a lack of fuel energy supply. Without adding any chemical reagents, lithium concentrate products with a grade of 50% to 80% can be obtained locally [17,18]. Over the years, this research team has continuously improved the process and conducted experimental research on the lithium extraction technology with the SGSP, and has successively proposed a series of optimization measures and methods, such as alkali adding and stirring, cross-year brine mixing and stereo-crystallization of the SGSP, which have been effectively applied to the actual production process of the Zabuye mining area, achieving the significant increase in the output and grade of lithium concentrates [19,20,21,22]. The production process flow of lithium extraction from Zabuye Salt Lake brine in Tibet is shown in Figure 2.

3. Materials and Methods

3.1. Materials

Two experiments were conducted using the on-site large-scale SGSP of the Zabuye mining area in 2020 and 2021, respectively. The effects of different raw brine and construction conditions of SGSP on the efficiency of brine storage for overwintering have been comparatively analyzed. The experimental raw brine was all the concentrated brine produced by multi-stage salt field evaporation in October of the same year. The specific physicochemical parameters are shown in Table 1.

3.2. Experimental Methods

The crystallization ponds S8 and S9 in the third operation area of the Zabuye mining area were selected as two experimental ponds. The specification and dimensions of these two ponds are basically the same, with a length of about 96.6 m, a width of about 37.7 m, a depth of about 4.5 m, and a slope angle of about 45°. Two batches of experimental raw brine were poured into the pond separately, and then a layer of freshwater with a certain thickness was laid on the surface of the brine to make the SGSP. Table 2 shows the construction and freshwater replenishment parameters of the SGSP for brine storage for overwintering.

Considering that the crystallizing pond in the Zabuye mining area is a composite structure with an EPDM waterproof liner (Dawang Tarpaulin Factory, Hebei Province, China)laid on the thick sandstone concrete, the thermal conductivity of concrete at 0~100 °C is 1.28 W/(m·K), and the thermal conductivity of the EPDM waterproof liner is 0.3 W/(m·K). The thermal conductivity of sodium chloride brine with a salinity of 25% at 30 °C is 0.57 W/(m·K), and its specific heat capacity is 4.03 kJ/(kg·°C). The annual average total solar radiation in the Tibet Plateau is more than 6000 MJ/m². According to previous experience and calculation, the heat capacity of the experimental scale SGSP is 3.5 × 10³ kJ, and the heat collection efficiency is about 15% [11,23]. In addition, during the operation of the SGSP, the heat transfer performance of the pond will also be affected by other factors, such as the disturbance from the external wind and rain that can destroy the internal structure of the SGSP, and the settlement of dust and other particles in the atmosphere that can reduce the turbidity of the water body. The algae and halogen insects derived from the changes in the external environmental conditions and the temperature inside the pond can affect the transmissivity of the water body, as well as the water surface icing, and so on.

After the formation of the SGSP, regular observational sampling and chemical analysis were carried out to obtain the basic operational data of the SGSP for brine storage for overwintering, including the air temperature, surface water temperature, bottom brine temperature and ice layer thickness, as well as brine temperature, salinity and density, and the concentration of Li⁺ and CO₃²⁻ at different depths. Combining with the air temperature and surface icing conditions at that time, a certain amount of freshwater was added to each of the two ponds during operation to ensure that the surface ice layer thickness remained between 2 and 3 cm [13], in order to maintain the stability of the salinity-gradient layer structure in the SGSP. The construction and operation scenario of the SGSP for brine storage for overwintering is shown in Figure 3.

Since there is no observation well in the crystallization pond of the third operation area, vertical observation and sampling at different depths inside the pond cannot be performed. The way of inclined sampling along the slope was adopted as shown in Figure 4. That is, a series of open-topped sampling bottles are fixed on the sampling rod at regular intervals, and then the sampling rod is quickly inserted into the water body along the slope and stays for a period of time for sampling. When the sampling bottles are filled, the sampling rod is carefully removed from the water surface to obtain the liquid samples at different depths inside the pond. The positions of sampling points need to be calculated by converting the distance of each sampling bottle from the bottom of the sampling rod to its vertical depth from the water surface. It should be noted that under the influence of the external environment, especially the low temperature, CO₃²⁻ in brine tends to crystallize and precipitate easily in the form of cold-phase minerals such as natronite, resulting in lower analytical results for CO₃²⁻ concentration. Therefore, once the liquid sample is obtained, it must be immediately sealed and stored, and the related analysis and testing work should be completed as soon as possible.

3.3. Analytical Methods

The monitoring, analysis and measurement work carried out during the experiment are all completed by relying on the Zabuye Salt Lake Resources and Environment Tibet Autonomous Region Field Scientific Observation and Research Station.

The longitudinal spatial distribution of brine temperature, density and salinity, with the concentration of Li⁺ and CO₃²⁻ in each layer, is measured by the stratified sampling of brine in the SGSP. The performance of SGSP is evaluated according to the idealization degree of the gradient curve of brine temperature, density and salinity. When the gradient curves of brine temperature, density and salinity gradually become stable after undergoing changes, and the concentrations of Li⁺ and CO₃²⁻ no longer change significantly with time, it can be considered that the lithium carbonate is completely precipitated from the brine under this condition, which is used as the criteria for judging the brine quality. Since part of the liquid samples is the saturated brine, it is necessary to pretreat with the deionized water for dilution and constant volume before the elemental analysis, and then use the national standard of elemental analysis method for the elemental analysis test.

The solid and liquid samples obtained during the experiment are chemically analyzed as follows: CO₃²⁻, acid–base titration and volumetric analysis; Li⁺, atomic absorption flame photometer method [24]. In daily observation, the mercury thermometer was used to record the air temperature and the brine temperature with an accuracy of ±0.1 °C. The floating hydrometer was used to determine the specific gravity of brine with an accuracy of ±0.001 g·cm⁻³. And the electronic weighbridge was used to weigh the lithium mixed salt.

4. Results and Discussion

4.1. Experiment of Brine Storage for Overwintering in Pond S8

The experiment of brine storage for overwintering in pond S8 lasted for 118 days and ended on 2 February 2021. The variation in observed parameters, including the air temperature, surface water temperature, bottom brine temperature and ice layer thickness with the time during the operation of the SGSP, are plotted as shown in Figure 5.

Figure 5 shows the variation in the observed parameters with the time during the experiment of brine storage for overwintering in pond S8. It can be seen that since the brine filling on 6 October 2020, the air temperature has been decreasing continuously in fluctuations, from the initial 6.5 °C to the last −13.0 °C, with a decrease of 19.5 °C. Due to the direct contact with the external environment, the surface water temperature showed a similar change trend to the air temperature. That is, the surface water temperature dropped from the initial 9.5 °C to the last −3.6 °C, with a decrease of 13.1 °C, but it was always higher than the air temperature. In the early stage of operation, the temperature difference between the two was only about 3 °C. It gradually enlarged in the middle and late stages, and even approached 10 °C at the end. It was observed that when the surface water temperature dropped to −3 °C, a certain thickness of ice layer began to form. As the air temperature dropped, the thickness of the ice layer reached 2.5 cm. This is because, in the case of low air temperature, there will be heat loss on the surface of the SGSP. When the surface water temperature is below freezing point with a large amount of solidification latent heat released, the surface of the water body will freeze and form ice crystals. And when the air temperature continues to decrease, the ice layer will continue to grow downward, and its thickness will gradually increase [13]. Since evaporation is the main cause of surface heat loss in the SGSP, the addition of freshwater is required to maintain the structural stability of the SGSP [1]. Therefore, the thickness of the ice layer was kept at about 2 cm by timely replenishment of the freshwater during the experiment, so as to maintain the thermal insulation of the SGSP in winter until the ice layer disappeared at the end of the operation. Compared to the air temperature and the surface water temperature, the variation in the bottom brine temperature is quite different. Under the effect of heat accumulation and temperature rise in the SGSP, the bottom brine temperature rapidly increased from 5.5 °C of the raw brine to 15.5 °C in a short time, with an increase of 10 °C. Then, it continued to fluctuate around 15 °C and rose to more than 20 °C in the late stage of operation. The bottom brine temperature was even 25 °C higher than the surface water temperature, indicating that under the low-temperature conditions in winter, the performance of heat accumulation and temperature rise in the large-scale SGSP covered with an ice layer is still significant [8]. The coverage of the ice layer not only has an effect of heat preservation and insulation on the surface of the SGSP, but also plays a positive role in the thermal storage capacity of the SGSP.

During the experiment, the temperature, density and salinity of the brine were systematically measured, and their variations with the height from the water surface of the pond were plotted, as shown in Figure 6, Figure 7 and Figure 8. The SGSP is a typical thermo-salt double diffusion system. When the heat diffuses, the mass transfer process of salt will occur in the water body under the action of salt molecule diffusion and the Soret effect, affected by factors such as the concentration difference, temperature difference and pressure difference. In the early stage of the SGSP operation, the surface water of the pond had a certain salinity because of the salt diffusion in the NCZ. And the brine salinity and density of the UCZ gradually increased. However, due to the difference in the thermo-salt double diffusion process at the bottom of the pond, as well as the downward excretion characteristics of salt in the process of surface water freezing, the salinity gradient in the NCZ was in an unstable state [13]. In the middle and later stages, the salinity-gradient layer continuously adjusted and gradually stabilized, since the surface ice layer could inhibit the increase in salinity in the UCZ caused by the salt diffusion, thus effectively maintaining the stability of the SGSP. The salinity in the LCZ significantly decreased. In the initial stage, the brine temperature of the LCZ was close to the highest temperature of the NCZ, and the temperature difference with the UCZ was smaller, which may be caused by the heat loss at the bottom of the pond [11]. Then, the temperature difference between the UCZ and LCZ increased with time. Until nearly two months later, the longitudinal distribution of the brine temperature in the pond remained basically unchanged, forming an ideal gradient curve. After that, there was no significant change in numerical values, only a slight fluctuation within a certain range, indicating that the salinity-gradient layer has been formed and the structure of the SGSP tended to stabilize at this time.

It also can be seen from the figures that the UCZ covers from the water surface to a depth of about 40 cm, where the temperature, density and salinity of brine were all at a low level and evenly distributed with little fluctuation. The depth range between 40 cm and 100 cm is the NCZ and the salinity distribution in this layer showed an increase from top to bottom. The SGSP relies on the salinity gradient formed by this salinity distribution to offset the temperature gradient formed between the temperature of the LCZ after absorbing the solar radiation energy and the temperature of the UCZ, which inhibits the generation of convection in the pond and enables the NCZ to play a heat preservation and insulation effect. Meanwhile, there are obvious temperature gradients and density gradients positively correlated with the depth in this layer, making it a key component in preventing heat loss. The depth from 100 cm to the bottom of the pond is the LCZ, where the brine temperature is relatively higher with little longitudinal variation and relatively uniform distribution. This is because the UCZ has a certain thickness and the surface ice layer has good thermal insulation performance. Coupled with the large heat capacity of the brine in the LCZ and the better insulation effect of the pond walls, although the brine temperature of the UCZ continues to be below 0 °C, the brine temperature of the NCZ and LCZ still remains above 0 °C, especially that of the LCZ can even reach 21.4 °C. The higher brine temperature of the LCZ is expected to ensure that the chemical components of the brine with high CO₃²⁻ and low Li⁺ are relatively stable during the overwintering process, and can be safely stored until February and March of the following year, so as to provide the qualified raw brine for the cross-year brine mixing.

Meanwhile, it can be found that the concentration of Li⁺ and CO₃²⁻ in the brine slightly reduced from the beginning to the end of the operation. From brine filling to brine discharge, the concentration of Li⁺ decreased from 1.08 g/L to 0.93 g/L, and the concentration of CO₃²⁻ decreased from 43.11 g/L to 39.75 g/L, with reductions of 0.15 g/L and 3.36 g/L, respectively. It is indicated that there was a small amount of lithium carbonate that, indeed, precipitated from the brine. According to the changes in Li⁺ concentration before and after, the theoretical precipitation amount of pure lithium carbonate was calculated to be 7.26 tons. Table 3 lists the actual collection of lithium mixed salt in pond S8. By weighing and converting the lithium mixed salt per unit area during the salt collection process, it was known that about 5 tons of pure lithium carbonate were actually collected from pond S8, which was 2.26 tons less than the theoretical value. This could be mainly due to measurement errors, incomplete salt collection and manual operation losses.

The potential losses of Li⁺ and CO₃²⁻ can be caused by other factors such as sampling errors or environmental influences in actual operation. For example, the interaction of brine at different layers during sampling may result in deviations between the measured results of Li⁺ and CO₃²⁻ concentration and the actual situation. Under low temperatures in winter, CO₃²⁻ in the obtained liquid samples is easy to precipitate in the form of cold-phase minerals such as natronite during the brine storage, which results in lower CO₃²⁻ concentration in the actual analysis. In order to ensure the chemical stability of brine during storage, firstly, the samples should be sealed and stored immediately once obtained, and the relevant work such as analysis and testing should be completed as soon as possible under the external conditions of the site; secondly, the temperature and density gradient layer distribution of brine in the solar pond should be closely monitored. Meanwhile, the freshwater should be added timely to maintain a stable salinity-gradient layer structure and keep a thin ice layer covering the surface of the brine at all times.

4.2. Experiment of Brine Storage for Overwintering in Pond S9

In order to verify the feasibility of the brine storage for overwintering by using the SGSP, and also to investigate the influence of operating conditions such as different raw brine composition and brine filling depth on the overwintering effect of brine storage in the SGSP, the experiment of brine storage for overwintering in pond S9 started on 21 October 2021 and ended on 22 February 2022, lasting for 124 days. The variation in observed parameters including the air temperature, surface water temperature, bottom brine temperature and ice layer thickness with the time during the operation of the SGSP are plotted as shown in Figure 9.

As can be seen from Figure 9, with the progress of the experiment, both the air temperature and the surface water temperature showed a gradual decreasing trend. In the early stage, the surface water temperature was close to the air temperature. But in the middle and late stages, the air temperature dropped sharply, while the surface water temperature only fluctuated slightly and was significantly higher than the air temperature. On the contrary, the bottom brine temperature kept rising under the action of heat accumulation of the SGSP in the early stage, and then basically stabilized at around 18 °C in the middle and late stages, much higher than the surface water temperature with a temperature difference of up to 25 °C. During operation, the thickness of the surface ice layer was maintained at about 3 cm for a long time by replenishing the freshwater timely. The variation in the pond S9 is basically consistent with that of pond S8, which, once again, demonstrates that the existence of the surface ice layer not only does not have a serious impact on the heat accumulation and storage performance of the SGSP under the low-temperature conditions in winter, but also helps to maintain the internal temperature of the SGSP.

The variations in brine temperature and density with depth during the operation of pond S9 are shown in Figure 10 and Figure 11. After the surface ice layer formed, the temperature and density of the brine in the pond showed a distribution regularity of gradually increasing from top to bottom. Since the brine filling depth of pond S9 was larger than that of pond S8, the salt diffusion behavior between the brine layer and the freshwater layer was more intense, resulting in the rapid formation of the NCZ. It only took about a month to form a stable structure of the SGSP, which was shorter than that of pond S8. In addition, the distribution conditions of each layer in pond S9 were also different from those in pond S8. The boundary between the UCZ and the NCZ in pond S8 was relatively clear, while it was difficult to distinguish the approximate range of them in pond S9. This is because the larger brine filling depth in pond S9 can lead to a strong upward heat transfer and salt mass transfer of the brine in the LCZ, making the UCZ and the NCZ become very thin, or even merge into one. That is, there was an obvious temperature gradient layer and density gradient layer from the water surface to the depth of about 100 cm, which showed a continuous upward trend with the increasing depth. The range from the depth of about 100 cm to the bottom of the pond was considered to be the LCZ, where the temperature and density of the brine were generally high with a relatively uniform longitudinal distribution, and did not change significantly with depth. As the running time went on, a small amount of salt crystallization would be precipitated from the brine in the LCZ under the action of heat accumulation and temperature rise in the SGSP, resulting in a slight decrease in brine density with insignificant decrease amplitude.

Figure 12 and Figure 13 showed the longitudinal distributions and variation patterns of Li⁺ and CO₃²⁻ concentrations in the brine of the pond S9 during operation in winter. The longitudinal distribution of Li⁺ concentration changed little over time and was relatively stable with a small fluctuation range. After a slight decrease in the later stage, it gradually rebounded again. But the longitudinal distribution of CO₃²⁻ concentration changed more obviously with time, with a sudden decrease before and during operation. It fluctuated greatly with the increasing depth and tended to be stable in the later stage.

Combined with the changes in brine temperature, Li⁺ and CO₃²⁻ concentrations at the beginning and end of the operation in pond S9, as shown in Table 4, it can be seen that during the overwintering process of brine storage, the concentrations of Li⁺ and CO₃²⁻ decreased by 0.09 g/L and 1.31 g/L, respectively, and the brine temperature increased by 11.0 °C, indicating that there was a certain amount of lithium carbonate precipitated from the brine. The theoretical precipitation amount of pure lithium carbonate was calculated to be 4.36 tons based on the concentration change in Li⁺ in brine. By weighing and converting the lithium mixed salt precipitated from the pond, the actual collection amount of pure lithium carbonate was obtained to be 4.69 tons, which was relatively close to the theoretical value. The actual collection of lithium mixed salt in pond S9 is shown in Table 5.

It is considered that there are several reasons for the difference between the theoretical and actual lithium collected in pond S8 and pond S9. Firstly, the theoretical precipitation amount of pure lithium carbonate in this paper is calculated according to the concentration change in Li⁺ in the brine before and after the brine discharge. This calculation method is based on the assumption that Li⁺ in the brine could be crystallized and precipitated only in the form of lithium carbonate. Therefore, the theoretical value itself has a certain deviation from the actual value. Secondly, the actual amount of pure lithium carbonate collected from the solar pond will be affected by various factors. On the one hand, the actual precipitation amount of lithium mixed salt was calculated by separately measuring the bottom and slope of the pond. However, due to the large size of the solar pond, the lithium mixed salt at different positions on the bottom or slope of the pond varies in weight, grade and moisture content. Here, two mean values of the bottom and slope of the pond were selected for calculation, which will inevitably cause calculation errors. On the other hand, due to the tight adhesion of the dried lithium mixed salt to the bottom and slope of the pond, it is difficult to collect all the lithium mixed salt in the pond solely by manual salt collection. Moreover, it is also easy to cause measurement errors due to human operation when weighing the lithium mixed salt, resulting in significant differences between the actual value and the theoretical value. Finally, due to the limited conditions in all aspects of the Tibetan Plateau lake area, the current lithium extraction technology by using the salinity-gradient solar pond is relatively simple and extensive. Therefore, in order to minimize the losses of salt collection, the production condition, operational level and measurement method should be appropriately improved to obtain the collection amount of lithium mixed salt in the solar pond more accurately if conditions permit. And it is recommended to calculate the theoretical precipitation amount of pure lithium carbonate based on the results of identification and composition analysis of salt minerals on the basis of identifying the types of salt minerals and the concentrations of anions and cations in the solar pond.

5. Conclusions

The purpose of brine storage for overwintering by using the SGSP is to store the brine with high CO₃²⁻ and low Li⁺ safely, so as to provide the raw brine for the cross-year brine mixing process in the following year. According to the results of two experiments in 2020 and 2021, it is completely feasible to operate the SGSP in winter to realize the brine storage for overwintering in the Zabuye mining area. The ideal raw brine is the concentrated brine produced in the salt field in September or October of that year, in which the concentration of CO₃²⁻ is generally around 50 g/L, and the brine temperature is about 5 °C. During the winter operation of the SGSP, the surface ice layer can not only prevent the disturbance of strong wind to the salinity-gradient layer, but also reduce the evaporation of water, while the sunlight can still penetrate through the thin ice layer to heat the bottom brine. When the surface of the SGSP is covered by an ice layer, the surface layer is considered to be divided into two layers: the ice layer and the water layer. The top–down transfer of heat in the ice layer is mainly heat conduction rather than convection, which increases the thermal resistance of the surface of the pond. Therefore, the existence of the ice layer plays a role in the thermal insulation of the solar pond, effectively blocking the upward loss of heat at the bottom and suppressing the increase in salinity of the UCZ caused by the salt diffusion, thus better maintaining the stability of the SGSP. Under the action of heat accumulation and temperature rise in the SGSP, the brine temperature of the LCZ can rise to more than 20 °C and remain relatively stable, and only a small amount of salts precipitated from the brine with a minimal change in brine composition, so as to ensure the safe storage of brine until February and March of the following year. This can not only provide the best raw brine with high CO₃²⁻ for cross-year brine mixing, but also reduce the pressure of brine production. In addition, about 5 tons of pure lithium carbonate harvested in the winter operation process has a certain help to improve the output of a single crystallization pond. When the surface of the SGSP is covered by an ice layer, the surface layer is considered to be divided into two layers: the ice layer and the water layer. The top–down transfer of heat in the ice layer is mainly heat conduction rather than convection, which increases the thermal resistance of the surface of the pond. Therefore, the existence of the ice layer plays a role in the thermal insulation of the solar pond, effectively blocking the upward loss of heat at the bottom and suppressing the increase in salinity of the UCZ caused by the salt diffusion, thus better maintaining the stability of the SGSP.

Due to the location on the northern Tibetan Plateau and the extreme climatic condition of Zabuye Salt Lake, as well as the unique carbonate type of brine, the method described in this paper has a certain range of applicability. Although the overwintering SGSP for brine storage has worked well in the Zabuye mining area, further research and testing would be needed in other climates and geographical locations to confirm its universal applicability. The technology in this paper presents an innovative lithium extraction method that uses SGSP technology in a challenging, cold environment, and offers a promising solution to increase lithium yield while minimizing environmental impact. The research results are expected to show a clear connection to the broader goals of improving lithium extraction efficiency and sustainability. Finally, the following effective measures and suggestions for the winter operation and maintenance of the SGSP are proposed, for reference:

• (1). The running of the SGSP in cold and windy winter is different from that in summer. If the operation fails, it will result in the precipitation of a large amount of salts like borax, natronite and glauber from the brine, with a thickness of up to tens of centimeters, which brings great difficulties for cleaning the crystallization pond.

• (2). It is necessary to strictly control the thickness of the freshwater layer and regularly monitor the brine temperature changes in the NCZ and LCZ in the SGSP. When the bottom brine temperature is lower than that at the time of brine filling, this is a danger signal. If necessary, the brine should be drained in time to avoid the concentrated precipitation of salts.

• (3). The operation of the freshwater replenishment during the overwintering process of brine storage in the SGSP is very important. The thickness of the freshwater layer should be controlled at around 25 cm for the first laying, and the thickness of the freshwater layer replenishment should be controlled at around 5 cm halfway, in order to ensure that the surface of the SGSP is always covered with an ice layer of about 2–3 cm thick.

The SGSP approach has been applied to the large-scale production of lithium concentrate in Zabuye Salt Lake, which has also become one of the largest industrial production bases of salt lake lithium in China, with an annual output of 5000 tons per year. However, the resource endowment of Zabuye Salt Lake with a very low magnesium–lithium ratio is rare in the world, and the low-temperature climate and local natural conditions are difficult to replicate. Therefore, the SGSP approach has great limitations under varying conditions, and it is only applied in Northern Tibet, China and some areas of Argentina. First, the SGSP approach is only suitable for lithium extraction from the carbonate-type brine with low magnesium–lithium ratio and high concentration of Li⁺. Secondly, the SGSP approach is suitable for use in areas with sufficient sunlight, large temperature difference between day and night, and the conditions for building the salt field. Finally, the SGSP approach is generally preferred in areas with poor infrastructure, inconvenient transportation and lack of energy supply.

6. Recommendations and Way Forward

Traditional lithium extraction methods require a large amount of energy consumption and various chemical reagents according to the process requirements. This will not only increase the production cost, but also bring serious environmental pollution problems. The lithium extraction technology with the SGSP utilizes solar energy resources and does not require additional energy and chemical reagents consumption, saving nearly half of the production cost compared with the traditional lithium extraction methods. As the low-temperature conditions in winter are not conducive to the brine heating up and the lithium deposition in the SGSP, the Zabuye mining area usually mainly carries out the brine production operations in winter. At the same time as being used for temperature rising and lithium precipitation, the SGSP is also used to store the brine for overwintering, which can not only improve the production efficiency, but also recover a certain amount of lithium carbonate, slightly reducing the overall production cost. Therefore, compared with other lithium extraction methods in terms of economy and environment, this method should be considered a more environmentally friendly and efficient technology in areas with strict environmental conditions, catering to the future development trend of lithium extraction from salt lakes, and has great development space and potential.

Due to the location on the northern Tibetan Plateau and extreme climatic conditions of Zabuye Salt Lake, as well as the unique carbonate type of brine, the SGSP for brine storage overwintering described in this paper has a certain range of applicability. Although it has worked well in the Zabuye mining area, further research and testing would be needed in other climates and geographical locations to confirm its universal applicability. Only by clarifying the impact of external factors such as climate fluctuations and brine composition on this method can the technology have the possibility of being extended to other regions with different environmental conditions. To sum up, it is suggested that the application of SGSP for brine storage overwintering should be combined with the local natural conditions and superior resources as much as possible, and some effective measures and operation modes conducive to the winter operation and maintenance of SGSP should be adopted. Once the climate environment and operating conditions have changed, relevant experiments need to be carried out based on the actual situation to obtain more specific adjusted operating parameters, so as to adequately enhance the practical applicability and scalability in environments outside the Tibetan Plateau.

Author Contributions

Q.W.: experiment, visualization and writing—original draft. Y.W. and L.B.: experiment and methodology. J.Z. and K.Z.: ideas, resources and supervision. J.L. and Z.H.: reviewing and editing. J.Y.: characterization and reviewing. Z.N. and L.B.: funding acquisition and reviewing. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

Upon reasonable request, the corresponding author can provide the data of this study.

Conflicts of Interest

Authors Jintao Zhang, Ke Zhang, Juntao Li and Zhikui He were employed by the company Tibet Mining Development Co., Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Nomenclature

SGSP	salinity-gradient solar pond
LCZ	lower convective zone
NCZ	non-convective zone/salinity-gradient layer
UCZ	upper convective zone/freshwater layer
T b	brine temperature (°C)
T air	air temperature (°C)
T w	freshwater temperature (°C)
δ w	thickness of freshwater layer (cm)
D b	brine filling depth (m)
D t	total depth (m)
δ wr	thickness of freshwater replenishment (cm)
δ i	thickness of ice layer (cm)
C x	concentration of x ion in brine (g/L)
G Li +	grade of lithium ion (%)
G LC	grade of lithium mixed salt (%)
δ LC	thickness of lithium mixed salt (mm)
W m	weight per unit area of lithium mixed salt (kg/m2)
MC	moisture content (%)
W LC	weight of lithium mixed salt (t)
WP LC	weight of pure lithium carbonate (t)
d p	height from the water surface of the solar pond (cm)
ρ b	brine density (g/cm3)
S b	brine salinity (°Bé)

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Location map of Zabuye Salt Lake, Tibet [16].

Figure 2. Production process flow of lithium extraction from Zabuye Salt Lake brine in Tibet.

Figure 3. Construction and operation scenario of SGSP for brine storage for overwintering. (a) Brine filling. (b) Freshwater layer laying. (c) SGSP forming. (d) Surface icing.

Figure 4. Observation and sampling scene of SGSP for brine storage for overwintering. (a) Sampling rod with open-topped bottles fixed at regular intervals. (b) Inclined sampling way along the slope. (c) Keeping sampling rod in water body for sampling for a period of time until sampling bottles filled. (d) Removing sampling rod from water surface to obtain liquid samples.

Figure 5. Variation in observed data with time during operation of pond S8.

Figure 6. Variation in brine temperature with depth during operation of pond S8.

Figure 7. Variation in brine density with depth during operation of pond S8.

Figure 8. Variation in brine salinity with depth during operation of pond S8.

Figure 9. Variation in observed data with time during operation of pond S9.

Figure 10. Variation in brine temperature with depth during operation of pond S9.

Figure 11. Variation in brine density with depth during operation of pond S9.

Figure 12. Variation in Li+ concentration with depth during operation of pond S9.

Figure 13. Variation in CO32− concentration with depth during operation of pond S9.

Table 1

Physicochemical parameters of experimental raw brine.

Year	Salt Field No.	Tb (°C)	CLi+ (g/L)	CCO32− (g/L)
2020	#1	5.5	1.08	43.11
2021	#4	2.0	1.33	43.20

Table 2

Parameters of construction and freshwater replenishment of SGSP for brine storage for overwintering.

Pond No.	Date of Brine Filling	Tair(°C)	Db(m)	Tw(°C)	δw(cm)	Dt(m)	Date of FreshwaterReplenishment	δwr(cm)
S8	6 October 2020	6.5	2.50	9.5	25	2.75	10 November 2020	5
S9	21 October 2021	6.5	3.50	9.5	25	3.75	13 November 2021	5

Table 3

Actual collection of lithium mixed salt in pond S8.

Sampling Position	GLi+ (%)	GLC (%)	δLC (mm)
Bottom of pond	11.78	62.57	~2
Slope of pond	13.75	73.08	~2

Table 4

Changes in brine temperature, Li⁺ and CO₃²⁻ concentrations at the beginning and end of operation in pond S9.

	Date	Tb (°C)	CLi+ (g/L)	CCO32− (g/L)
Brine filling	22 October 2021	6.5	1.33	43.20
Brine discharge	22 February 2022	17.5	1.24	41.89
Difference value	124 days	11.0	−0.09	−1.31

Table 5

Actual collection of lithium mixed salt in pond S9.

Sampling Position	Wm (kg·m−2)	MC (%)	GLC (%)	WLC (t)	WPLC (t)
Bottom of pond	2.48	35.05	21.68	19.84	4.30
Slope of pond	0.93	30.46	28.61	1.29	0.39
Total	/	/	/	21.13	4.69

References

1. Guo, Y.; Zhang, T.; Chen, L.; Si, Y.; Chen, X.; Mei, S. Review of key technologies and development status of salt gradient solar pond. Sol. Energy; 2020; 11, pp. 9-23.

2. Kasaeian, A.; Sharifi, S.; Yan, W.M. Novel achievements in the development of solar ponds: A review. Sol. Energy; 2018; 174, pp. 189-206. [DOI: https://dx.doi.org/10.1016/j.solener.2018.09.010]

3. Ma, H.; Wang, H. A review of heat storage performance improvement and power generation application of salt gradient solar pond. Chin. J. Process Eng.; 2023; 23, pp. 972-986.

4. Wang, G.; Zhu, Y.; Xu, G.; Li, H.; Wang, C. Studies on solar pond based on heat storage performance of salt lake brine. Chem. World; 2019; 60, pp. 514-518.

5. Beiki, H.; Soukhtanlou, E. Determination of optimum insulation thicknesses for salinity gradient solar pond’s bottom wall under different climate conditions. SN Appl. Sci.; 2020; 2, 1284. [DOI: https://dx.doi.org/10.1007/s42452-020-3078-4]

6. Sayer, A.H.; Al-hussaini, H.; Campbell, A.N. Experimental analysis of the temperature and concentration profiles in a salinity gradient solar pond with and without a liquid cover to suppress evaporation. Sol. Energy; 2017; 155, pp. 1354-1365. [DOI: https://dx.doi.org/10.1016/j.solener.2017.08.002]

7. Zheng, M.; Xiang, J.; Wei, X.; Zheng, Y. Saline Lake on the Qinghai-Xizang (Tibet) Plateau; Science Press: Beijing, China, 1989.

8. Zhang, H.; Xie, M.; Sun, W.; Ge, S. Simulation and experiment of thermal performance of a salt-gradient solar pond. J. Eng. Thermophys; 2006; 27, pp. 65-68.

9. Nie, Z.; Bu, L.; Zheng, M.; Huang, W. Experimental study of natural brine solar ponds in Tibet. Sol. Energy; 2011; 85, pp. 1537-1542. [DOI: https://dx.doi.org/10.1016/j.solener.2011.04.011]

10. Nie, Z. Technological Research on Depositing Lithium Salt from Brine of Carbonate Saline Lakes. Ph.D. Thesis; Chinese Academy of Geological Sciences: Beijing, China, 2001.

11. Huang, W.; Wang, X.; Sun, Z.; Nie, Z.; Sha, Z. Construction and operational of solar pond in Zabuye saline lake, Tibet. Acta Energiae Solaris Sin.; 2010; 31, pp. 163-167.

12. Gao, C.; Cao, D.; Zhang, Y. Research progress of solar pond technology. J. Salt Chem. Ind.; 2012; 41, pp. 14-19. [DOI: https://dx.doi.org/10.4028/www.scientific.net/AMR.724-725.180]

13. Li, N. Water Object Study of Salt Gradient Solar Pond and Effect of Surface Ice on Solar Pond Performance. Ph.D. Thesis; Dalian University of Technology: Dalian, China, 2010.

14. Li, N. Effect study of surface ice on solar pond performance. Acta Energiae Solaris Sin.; 2011; 32, pp. 1200-1205.

15. Zheng, X.; Zhang, M.; Xu, C.; Li, B. Annals of Salt Lakes in China; Science Press: Beijing, China, 2002.

16. Wang, Y.; Zheng, M.; Yan, L.; Bu, L.; Qi, W. Influence of the regional climate variations on lake changes of Zabuye, Dangqiong Co and Bankog Co salt lakes in Tibet. J. Geogr. Sci.; 2019; 29, pp. 1895-1907. [DOI: https://dx.doi.org/10.1007/s11442-019-1995-x]

17. Zheng, M. Study advances in saline lake resources on the Qinghai-Tibet plateau. Acta Geosci. Sin.; 2001; 22, pp. 97-102.

18. Nie, Z.; Bu, L.; Zheng, M. Lithium resources industrialization of salt lakes in China: A case study of the Xitaijinaier salt lake and the Zabuye salt lake. Acta Geosci. Sin.; 2010; 31, pp. 95-101.

19. Wu, Q.; Yu, J.; Bu, L.; Nie, Z.; Wang, Y.; Renchen, N.; He, T.; Zhang, K.; Zhang, J.; He, Z. The application of an enhanced salinity-gradient solar pond with nucleation matrix in lithium extraction from Zabuye salt lake in Tibet. Sol. Energy; 2022; 244, pp. 104-114. [DOI: https://dx.doi.org/10.1016/j.solener.2022.08.031]

20. Yu, J.; Wu, Q.; Bu, L.; Nie, Z.; Wang, Y.; Zhang, J.; Zhang, K.; Renchen, N.; He, T.; He, Z. Experimental study on improving lithium extraction efficiency of salinity-gradient solar pond through sodium carbonate addition and agitation. Sol. Energy; 2022; 242, pp. 364-377. [DOI: https://dx.doi.org/10.1016/j.solener.2022.07.027]

21. Nie, Z.; Wu, Q.; Ding, T.; Bu, L.; Wang, Y.; Yu, J.; Hou, X. Research progress on industrialization technology of lithium extraction from salt lake brine in China. Inorg. Chem. Ind.; 2022; 54, pp. 1-12.

22. Wu, Q.; Bu, L.; Zhang, J.; Zhang, K.; Yu, J.; Li, J.; Solange, D.; He, Z.; Nie, Z. Study of the optimization of the stereo-crystallization process with enhanced salinity-gradient solar pond for lithium extraction from Zabuye salt lake in Tibet. Carbonates Evaporites; 2024; 39, 10. [DOI: https://dx.doi.org/10.1007/s13146-024-00921-0]

23. Li, N.; Xue, Z.; Liu, H.; Zhang, L. Experiment and analysis of thermal performance for salt-gradient solar pond with solar collector. J. Therm. Sci. Technol.; 2017; 5, pp. 392-397.

24. Qinghai Institute of Salt LakeChinese Academy of Sciences. Analysis Methods for Brines and Salts; Science Press: Beijing, China, 1988; pp. 160-200.