.jpg)
1. Introduction
The molten salt reactor (MSR), classified as a Generation IV nuclear reactor, utilizes molten salt as both fuel and coolant, in contrast to conventional reactors that rely on solid fuel and water coolant. MSRs operate at nearly atmospheric pressure, which reduces mechanical stress on structural components, thereby simplifying reactor design and enhancing operational safety. Furthermore, MSRs achieve significantly higher operating temperatures than traditional water-cooled reactors, resulting in superior thermodynamic efficiency while maintaining lower vapor pressure. The chemical stability of molten salt fuels enables them to dissolve a wide range of nuclear materials, including uranium and thorium. However, the high temperatures and corrosive properties of molten salts, particularly those containing fluoride and chloride, pose significant challenges, such as accelerated component degradation [1-6].
Molten salts exhibit strong hydration characteristics due to their pronounced ionic nature, which facilitates significant interactions with water molecules. As a result, they readily absorb moisture from the atmosphere, forming hydrates [7]. This absorbed moisture can increase the oxidizing potential of the salt, leading to the corrosion of transition metals such as nickel, iron, and chromium, which are key components of MSR structural materials. Furthermore, the release of corrosion products into salt can alter the properties of MSR fuel or coolant. Therefore, it is essential to supply salt with a low initial moisture concentration and maintain this condition to protect structural materials and ensure the stable operation of the MSR [4, 8].
Several methods have been reported to remove moisture from salt, including heating under vacuum with inert gas; heat sparging or purging with an inert gas; dry sparging with hydrochloric acid, hydrofluoric acid, or mixtures of these gases with hydrogen and adding reactive metals to the salt to chemically bind with moisture [9-15]. To assess the effectiveness of these methods, precise measurement of moisture concentration in the salt after treatment is crucial.
The moisture concentrations in salts have been quantitatively measured using various methods, including thermal analysis [16], oxygen analysis [17, 18], and Karl Fischer titration [18]. Among these techniques, Karl Fischer titration is recognized for its high precision and is widely employed in industries such as pharmaceuticals, food processing, and petrochemicals. This method relies on the quantitative reaction between H2O and I₂ in the presence of SO2 and a base (commonly pyridine or imidazole) within a suitable solvent, such as methanol. The primary chemical reaction can be expressed as follows [19, 20]:
In this reaction, exactly one molar equivalent of H2O reacts with I2. The endpoint of the titration is reached when I₂ is present in excess, which is typically detected potentiometrically.
When determining the moisture concentration of a salt using Karl Fischer titration, the salt is dissolved in a solvent containing SO2 to ensure complete dissolution, followed by titration with I₂. Care must be taken when analyzing highly hygroscopic salts, as their exposure to air before dissolution can lead to rapid moisture absorption, resulting in measurements that exceed the salt’s original moisture concentration. To minimize such inaccuracies, it is recommended to perform the analysis in an inert atmosphere, such as within an Ar-filled glove box, to prevent contamination from ambient moisture. However, maintaining an inert atmosphere may not always be feasible under laboratory conditions, requiring the salt to be exposed to air during the analysis.
This study explores the role of an inert atmosphere on preventing moisture contamination of chloride salts during the Karl Fischer titration. Four chloride salts with varying hygroscopicity were selected: the less hygroscopic NaCl and KCl, and the more hygroscopic MgCl2 and ZnCl2. For each salt, three sample types were prepared, including asreceived and dehydrated by heating for 6 and 12 h. Moisture concentration in each sample was independently analyzed using Karl Fischer titration under both air and inert atmospheric conditions, and the results were compared to assess the influence of atmospheric exposure.
2. Experimental
2.1 Salts and Their Dehydration
NaCl (99.5%, Junsei), KCl (anhydrous 99.5%, Samchun), MgCl2 (anhydrous 99%, Thermo Scientific) and ZnCl2 (anhydrous 98.0 %, Sigma Aldrich) were used in this study. The salts were dried by heating under vacuum in a furnace connected to a high-purity Ar-filled glove box. Table 1 lists the melting points of the salts used in this study and heating temperatures and times for dehydrating them. NaCl and KCl were dried at 400°C while MgCl2 and ZnCl2 were dried at 200°C. Drying times of 6 and 12 hours were applied for all salts. For drying the salt, the original (as-received) reagent (20 g) was placed in a graphite crucible and heated in a furnace under vacuum for the predetermined temperature and time shown in Table 1. The salt obtained at the end of drying was recovered after cooling and used for the analysis of moisture concentration in the salt.
Table 1
List of melting points of the salts used in this study and heating temperatures and times for dehydrating them
| Salts | Melting point [°C] | Dehydration conditions | |
|---|---|---|---|
|
|
|||
| Temperature [°C] | Time [h] | ||
|
|
|||
| NaCl | 801 | 400 | 6, 12 |
| KCl | 770 | 400 | 6, 12 |
| MgCl2 | 714 | 200 | 6, 12 |
| ZnCl2 | 290 | 200 | 6, 12 |
2.2 Analysis of Hygroscopicity and Moisture Concentration of Salts
The hygroscopicity of each salt reagent was evaluated by placing 5 g of the salt in a vial and monitoring its mass change over time. For each reagent, the mass of the vial containing the salt was measured using a balance in an air environment. Specifically, the hygroscopicity of MgCl2 was measured at two relative humidity (RH) levels, 18% and 50%, while for the other three salts, measurements were conducted only at 18% RH. Additionally, the mass change of the vials containing the salts was monitored over 82 h using a balance inside an Ar-filled glove box. During these experiments, the glove box environment was maintained with moisture and oxygen concentrations below 1 ppm.
The moisture concentration in the salt samples was measured using a Karl Fischer analyzer (Mettler Toledo, Titrator Compact C10S), which generates I₂ electrolytically to drive the Karl Fischer reaction (reaction 1). Initially, the moisture concentration of three samples for each salt—comprising the as-received reagent and two salt samples dried for 6 h and 12 h—was determined using the Karl Fischer analyzer in ambient air (20–25°C, 18% relative humidity), where the salts were exposed to atmospheric moisture (Fig. 1(a)). Subsequently, the Karl Fischer analyzer was relocated into an Ar-filled glove box, and the moisture concentrations of the same, as-received sample and dried salts were remeasured under the controlled atmosphere within the Ar-filled glove box (Fig. 1(b)).
3. Results and Discussion
3.1 Measurements of Hygroscopicity of the Salts
The hygroscopicity of salt is influenced by its physicochemical properties, including its affinity for water and the size and purity of the salt powder. Before measuring the moisture concentration of the salts used in this study, their inherent hygroscopicity was first evaluated. For the salt reagents used in this study, hygroscopicity was assessed by monitoring weight changes over time in both air and the Ar-filled glove box.
Fig. 2(a) shows the percentages of mass increase relative to the original (as-received) salts (NaCl, KCl, MgCl2, and ZnCl2) measured in air at 18% RH for 82 h. NaCl and KCl exhibited extremely low mass changes. In contrast, MgCl2 and ZnCl2 displayed linear increases in mass over time. Specifically, MgCl2 and ZnCl2 increased in mass by approximately 9.6% and 7.7%, respectively, after 82 h, demonstrating significantly higher hygroscopicity compared to NaCl and KCl, consistent with previous reports in the literature [21]. A linear regression analysis of the mass gain data for MgCl2 and ZnCl2 yielded equations (shown in Fig. 2(a)) to calculate the rate of moisture absorption, corresponding to increases of 18 ppm and 15 ppm per minute, respectively. These findings suggest that even brief exposure of salt to air can significantly affect their moisture concentrations. Fig. 2(b) presents the percentages of mass increase of the four salts over 82 h in the Ar-filled glove box. Unlike the results observed in air (Fig. 2(a)), even the highly hygroscopic salts of MgCl2 and ZnCl2 exhibited negligible mass changes, remaining below 0.06%. This outcome indicates that the extremely low moisture concentration in the Ar atmosphere effectively prevents moisture absorption, even for salts with high hygroscopicity.
Fig. 2
Percentages of mass increase relative to the original dry single salt after exposure: (a) air and (b) Ar in a certain time at room temperature of 20–25℃ and relative humidity of 18%.
The percentages of mass increase relative to the original MgCl2 measured in air at 18% RH, as shown in Fig. 2(a), is compared to that at 50% RH, as presented in Fig. (3). Its percentages of mass increase at 50% RH exhibits a significantly steeper increase, reaching approximately 80% of the initial mass after 82 h. This result demonstrates that the hygroscopic nature of MgCl2 leads to greater water absorption as ambient humidity increases, even for the same salt sample.
Fig. 3
Percentages of mass increase relative to the original dry single salt after Air exposure for different humidity.
3.2 Moisture Concentration Measured in Salts in air and Ar-filled Glove box
Fig. 4 presents the moisture concentrations in NaCl salt before and after drying, measured in both air and the Ar-filled glove box. The average moisture concentration of the undried NaCl salt (as-received reagent) was 1,156 ppm when measured in air, whereas it was 649 ppm in the Ar-filled glove box, representing nearly twofold difference. A similar trend was observed in salt samples that underwent dehydration by drying at 400°C for 6 h, with average moisture concentrations of 768 ppm in air and 345 ppm in the Ar-filled glove box. These findings suggest that even low-hygroscopic NaCl is influenced by the moisture content of the surrounding gas during measurement. Specifically, exposure to air increases the surface moisture of salt particles, leading to higher measured concentrations. For samples dried for 12 h at 400°C, the average moisture concentrations were 364 ppm in air and 336 ppm in the Ar-filled glove box, with the air-measured value being slightly higher. Additionally, the standard deviations of the moisture concentrations measured in air—both before and after drying—were generally larger than those measured inside the glove box. This indicates greater variability in the air-exposed measurements, likely due to inconsistent moisture absorption from the surrounding environment. Overall, these results highlight the sensitivity of moisture concentration measurements to the ambient gas conditions, emphasizing the need for controlled environments when analyzing salt moisture content.
Fig. 4
Comparison of moisture concentration in NaCl salts measured by Karl Fischer titrator in air and in the Ar-filled glove box.
The moisture concentration of KCl salts with low hygroscopicity, such as NaCl, was analyzed before and after drying under two different conditions: in air and within the Ar-filled glove box. The results are presented in Fig. 5. The average moisture concentration of the undried KCl salt (raw reagent) was 802 ppm, whereas the concentration measured in the Ar-filled glove box was significantly lower, at 365 ppm—less than the value observed in air. This trend was consistent with the findings for NaCl shown in Fig. 4, where exposure to atmospheric moisture led to a higher moisture concentration in air compared to the controlled Ar environment. The difference in moisture concentration between the two conditions became even more pronounced after drying at 400°C. For NaCl dried for 6 h, the moisture concentration measured in the Ar-filled glove box was 63 ppm, while the concentration measured in air was 490 ppm—over seven times higher. Similarly, after 12 h of drying, the moisture concentration in the Ar-filled glove box decreased further to 19 ppm. However, when measured in air, the moisture concentration remained significantly elevated at 370 ppm, nearly 20 times higher than the corresponding value in the Ar-filled glove box. Additionally, the standard deviation of moisture concentration was substantially greater for salt samples analyzed in air compared to those measured in the Ar-filled glove box. This indicates that exposure to ambient moisture caused considerable variability in the measured values.
Fig. 5
Comparison of moisture concentration in KCl salts measured by Karl Fischer titrator in air and in the Ar-filled glove box.
Fig. 6 shows the moisture concentration in MgCl2 salt before and after drying, measured both in air and inside the Ar-filled glove box. The melting point of MgCl2 is 714°C, which is slightly lower than that of NaCl and KCl (Table 1). However, during heating, MgCl2 may generate MgOHCl or HCl gas [22], necessitating a lower heating temperature. Consequently, MgCl2 was heated at 200°C, whereas NaCl and KCl were heated at 400°C. The average moisture concentration of MgCl2 before drying was 3,939 ppm in air and 4,672 ppm inside the Ar-filled glove box, with the latter being slightly higher. These values are significantly greater than the moisture concentrations of undried NaCl and KCl, as shown in Figs. 4 and 5, highlighting the high hygroscopicity of MgCl2. The average moisture concentration in the salt of MgCl2 dried at 200°C for 6 h was 916 ppm when analyzed inside the Ar-filled glove box, which is more than five times lower than before drying. However, when analyzed in air, the moisture concentration remained at 3,463 ppm, which is 3.8 times higher than the value obtained in the Ar-filled glove box. Similarly, for MgCl2 dried for 12 h, the moisture concentration measured in air and inside the Ar-filled glove box was 3,442 ppm and 747 ppm, respectively, indicating a difference of more than fourfold. Notably, the standard deviations of the moisture measurements for MgCl2 samples analyzed in air were generally larger than those of the samples analyzed inside the Arfilled glove box. Additionally, the standard deviations of the MgCl2 moisture concentrations measured in air were higher than those observed for NaCl and KCl under the same conditions, as shown in Figs. 4 and 5. These results suggest that the high hygroscopicity of MgCl2 makes its moisture concentration more susceptible to variations in ambient humidity compared to NaCl and KCl, which exhibit lower hygroscopicity.
Fig. 6
Comparison of moisture concentration in MgCl2 salts measured by Karl Fischer titrator in air and in the Ar-filled glove box.
For ZnCl2, a highly hygroscopic salt, the moisture concentration before and after drying was measured both in air and within the Ar-filled argon glove box, as shown in Fig. 7. Before drying, the average moisture concentration of ZnCl2 was 2,974 ppm when measured in air and 1,954 ppm in the Ar-filled glove box. The higher moisture concentration in air is attributed to the salt’s exposure to ambient humidity during the measurement process. Given its low melting point of 290°C, ZnCl2 was dried at 200°C. After drying for 6 h, the average moisture concentration remained at 2,546 ppm when analyzed in air and 1,651 ppm in the Ar-filled glove box, showing a difference of approximately 1.5 times. A similar trend was observed after 12 h of drying, with average moisture concentrations of 2,147 ppm in air and 1,625 ppm in the Ar-filled glove box. The consistently higher moisture values in air compared to the Ar environment align with the hygroscopic nature of ZnCl2, as observed in other salts. However, unlike the other salts, ZnCl2 did not exhibit a significant reduction in moisture concentration even after 6 and 12 h of drying at 200°C, suggesting a stronger retention of water molecules.
Fig. 7
Comparison of moisture concentration in ZnCl2 salts measured by Karl Fischer titrator in air and in the Ar-filled glove box.
4. Conclusions
This study demonstrates the significant influence of an inert gas (Ar) atmosphere on the accurate measurement of moisture concentrations in chlorinated molten salts. The results consistently show that Karl Fischer titration conducted in an Ar-filled glove box yields lower and more stable moisture readings compared to measurements taken in air. This finding underscores the necessity of minimizing atmospheric exposure to obtain precise data, as ambient moisture can lead to overestimated values. Furthermore, the differences in hygroscopic behavior among the tested salts emphasize the importance of tailored dehydration strategies. By ensuring accurate moisture analysis, this research contributes to the development of more reliable corrosion mitigation strategies for molten salt reactors, ultimately supporting their commercialization and longterm viability.
