.jpg)
1. Introduction
Radionuclide migration through the underground multibarrier repository system of high-level radioactive waste (HLW) and transportation to the ecosystem should be thoroughly understood to evaluate the proof-of-safety of the repository system [1]. Groundwater flow is a major route of radionuclide migration, and the migration quantity depends on the total amount of solubilized radionuclides [2]. It is difficult to experimentally assess nuclide solubility under vast variations in spatiotemporal conditions. Geochemical modeling based on a thermodynamic database (TDB) has been employed to evaluate the solubility under diverse repository conditions [3].
The complex formation of radionuclides with groundwater components determines the solubility of radionuclide compounds and directly affects sorption and diffusion through engineered and natural barriers [4, 5]. The oxidation state of a radionuclide is an important factor affecting its chemical reactivity. Ideally, all possible chemical processes including aqueous chemical reactions, solid-phase dissolution, and redox reactions should be considered. The reliability of geochemical modeling depends on the accuracy of the thermodynamic data and the comprehensiveness of the TDB. An effective way to validate a TDB, that is, whether it can describe domestic conditions appropriately, is to directly compare the modeling results with the experimentally measured values.
In our previous study, Am solubility was measured in synthesized groundwater (Syn-DB3) by simulating groundwater conditions at the DB3 site in the KAERI underground research tunnel (KURT) [6]. Due to the shortage of Am in stock to prepare solid phases, the experiment was conducted under oversaturation conditions, where aqueous Am (241Am, t1/2 = 432.7 years) [7] was added to Syn-DB3, and the Am concentration in the supernatant was monitored using LSC. In this study, three main issues were identified: 1) The precipitated solids in the samples were not visible and thus could not be characterized. Solid-phase characterization before and after the solubility measurements is important for evaluating the experimental results. 2) The pH values of the samples were not checked at the end of the experiments despite the probable pH changes over a period of long reaction time. 3) The supernatant contained colloidal particles, which would have resulted in overestimation of the dissolved Am concentrations. More rigorous phase separation should be employed to exclude colloidal contributions to the dissolved Am concentrations.
In the thermodynamic studies of aqueous species, Nd(III) and Eu(III) are widely used as nonradioactive analogs of Am(III). The chemical behavior of trivalent actinide (An) and lanthanide (Ln) ions is highly correlated with their ionic radii [8, 9]. Am3+ (1.10 Å), Cm3+ (1.09 Å), Nd3+ (1.11 Å), and Eu3+ (1.07 Å) ions have similar ionic radii [8, 10] and differences between their activities in aqueous solutions are often less than the experimental errors [2, 11]. The NEA-TDB project has used these analogies to evaluate the thermodynamics of aqueous Am(III) species using experimental data for Cm(III), Nd(III) and Eu(III) [2, 12]. Notably, the solubility properties of the isostructural solids of Am(III) and Ln(III) can differ considerably, unlike in aqueous chemistry. Even in these cases, the NEA-TDB project discussed the solubility constants of Am(III) solids in comparison with those of isostructural Ln(III) solids, as Ln are free from alpha-radiation damage and theirs crystallinities can be relatively well characterized [2, 12].
The present study aimed to measure solubility of Ln(III) compounds of Ln2(CO3)3·xH2O(cr) (Ln = Sm, Eu) in the Syn-DB3 in under-saturation conditions. Eu was employed as the nonradioactive analog of Am. Sm was also examined because it is a trivalent Ln element included in the safety assessments of the HLW repository system conducted by the Institute for Korea Spent Nuclear Fuel (iKSNF) [13]. The solid compounds were characterized by x-ray diffraction (XRD) before and after the solubility study. Ultrafiltration was used to separate the dissolved metal ions from the solid phase. The pH and composition of the samples were analyzed at the end of the experiment. The aqueous Eu(III) species were investigated using time-resolved laserinduced fluorescence spectroscopy (TRLFS). For comparison, the solubility of Am in Syn-DB3 was also measured under oversaturated conditions in continuation of the previous study [6]. The measured solubility values were compared with geochemical modeling results based on ThermoChimie (ver. 11a) TDB [14].
2. Experimental
2.1 Ln2(CO3)3∙xH2O Solubility Sample Preparations: Under-Saturation Conditions
Solid reagents of Sm2(CO3)3∙xH2O and Eu2(CO3)3∙xH2O (Alfa Aesar, REacton 99.99%) were used as purchased. Syn-DB3 was prepared as described previously [6] and its composition is listed in Table 1. Syn-DB3 is an oxidative groundwater (Eh = 0.436 V) at pH 9.03. Samples for solubility measurements were prepared by adding solid powders to 500 mL of Syn-DB3 groundwater (PE 500 mL bottles). Three Sm repeats contained 59.4, 54.9, and 66.3 mg of Sm2(CO3)3∙xH2O powder and three Eu repeats contained 64.8, 58.1, and 53.7 mg of Eu2(CO3)3∙xH2O powder (Table 2). These values were more than 104 times higher than the expected solubility values.
Table 1
Physicochemical properties of the Syn-DB3 before and after experiments
| Element | Syn-DB3 | After Sm experiment | After Eu experiment | |
|---|---|---|---|---|
|
|
||||
| pH | 9.03 | 8.19 | 8.26 | |
| Eh (V) | 0.436 | Not measured | Not measured | |
|
|
||||
| Chemical composition (mg·L−1) | Na | 38.2 | 38.9 | 39.0 |
| Ca | 5.3 | 4.9 | 5.0 | |
| K | 0.48 | 0.52 | 0.52 | |
| Mg | 0.26 | 0.28 | 0.29 | |
| HCO3 | 84.13 | 83.0 | 82.9 | |
| SiO2 | 7.7 | 7.8 | 7.70 | |
| Cl | 1.76 | 1.80 | 1.79 | |
| SO4 | 5.51 | 5.49 | 5.55 | |
| F | 7.67 | 7.73 | 7.82 | |
Table 2
Sm and Eu solubility samples
| Sample | Added solid (mg) in 500 mL Syn-DB3 | Added [Ln]total (M) | Measured [Ln]supernatant (M) | Measured [Ln]UF (M) |
|---|---|---|---|---|
|
|
||||
| Sm1 | 59.4 | 4.9 × 10−4 | (6.21 ± 0.12) × 10−8 | (3.98 ± 0.04) × 10−8 |
| Sm2 | 54.9 | 4.6 × 10−4 | ||
| Sm3 | 66.3 | 5.5 × 10−4 | ||
|
|
||||
| Eu1 | 64.8 | 5.4 × 10−4 | (8.84 ± 0.28) × 10−8 | (6.08 ± 0.05) × 10−8 |
| Eu2 | 58.1 | 4.8 × 10−4 | ||
| Eu3 | 53.7 | 4.4 × 10−4 | ||
2.2 Determination of Dissolved Sm and Eu Concentration
Concentrations of Sm and Eu in the supernatant of the test samples were periodically measured; 10 mL of the supernatant was sampled from the top of the samples, mixed with 100 μL of high-purity HNO3 (13.14 M, Ultrapur, Merck) in 10 mL centrifuge tubes, and sent for analysis using inductively coupled plasma-mass spectrometry (ICPMS). Samples were collected at 2 h, 1 day, 7 days, 20 days, 50 days, 90 days, and 223 days.
At the end of the experiments, when no further concentration changes were observed, the final Sm and Eu concentrations were analyzed after phase separation using an ultrafiltration device (10 kDa, Amicon Ultra-15 centrifugal filter, Millipore), after which, 10 mL of the supernatant was placed in an ultrafiltration device, centrifuged (5,000 rpm, 1 h), and the filtrate was transferred to a new falcon tube containing HNO3 for ICP-MS analysis. This process was repeated four times, and the concentrations of Sm and Eu in each filtrate were analyzed using ICP-MS.
2.3 Characterizations of Residual Solids and Aqueous Phases at the end of the Experiments
At the end of the solubility measurements, the supernatants of the Sm and Eu samples were analyzed for pH and concentrations of the main components using ICP-optical emission spectrometry. After 223 days of solubility testing, the solid phase that had settled to the bottom of the solution was filtered through a membrane filter (0.2 μm), air dried, and subjected to XRD analysis.
2.4 TRLFS of Eu Solution
The Eu filtrates (0.2 μm) at the end of the solubility measurements were analyzed by time-resolved laser induced fluorescence spectroscopy (TRLFS) [15, 16]. Briefly, a nanosecond-pulsed wavelength-tunable OPO laser system (Opotek Inc. Vibrant B) was used as the excitation source at 394 nm (1.1 mJ). Luminescence spectra was measured using an intensified CCD (ICCD) detector via a spectrograph with a slit width of 300 μm and grating of 300 lines∙mm−1 (Andor Technol. SR-303i). Samples were incubated at 25℃ in quartz cuvettes during the measurements. Luminescence decay rates were measured using a kinetic mode of the ICCD detection with a gate width of 1 ms, initial gate opening delay of 1 μs, and stepwise delay by 30 μs. The number of hydration (n(H2O)) in inner-sphere of Eu(III) center was calculated according to the relation with the decay lifetime (τ) as described in Equation (1) [17, 18].
2.5 Am Solubility Measurements: Over-Saturation Conditions
Am sample preparation was described in previous study [6]. Briefly, a series of pH-varied Am(III) samples (11.8 μM) were prepared in Syn-DB3. pH values of the samples were adjusted to be in the range of 6.4–10 by adding 0.2 M HClO4 (ULTREX II, Ultrapure reagent, JT Baker) or 0.1 M NaOH (Semiconductor grade, 99.99% trace metals basis, Sigma-Aldrich). Two milliliters of each sample were stored in quartz cells with a screw cap, and their absorption spectra were monitored periodically. Absorption data between 0–70 days were already reported in our previous study [6]. In this study, the samples were kept unopened until the last day of the experiments (231 days), when the samples were subjected to pH and Am concentration analyses. Phase-separation was conducted using ultrafiltration (10 kDa, Amicon Ultra-0.5 centrifugal filter, Millipore) and the fourth filtrates were analyzed by LSC. Detailed experimental conditions for LSC radiometry are described in our previous study [6].
3. Results and Discussion
3.1 XRD Analysis on Ln2(CO3)3∙xH2O Before and After Experiments
Ln2(CO3)3∙xH2O (Ln = Sm, Eu) powders were analyzed by XRD before experiments. The measured patterns shown in Fig. 1 are almost identical to that of the crystalline Gd2(CO3)3∙xH2O(cr) (PDF#: 00-037-0559, orthorhombic). These results indicate that the Sm and Eu powders are crystalline, similar to the Gd2(CO3)3∙xH2O(cr).
Fig. 1
XRD patterns of the (a) Sm and (b) Eu solids before (black) and after (blue) solubility measurements. Red is XRD pattern of Gd2(CO3)3∙xH2O (PDF #: 00-037-0559).
After solubility measurements, the collected residual solid phases were analyzed using XRD. The XRD patterns of the solid phases before and after the tests were the same for Sm and Eu, as shown in Fig. 1. The results showed that the crystal structures of the Sm and Eu powders were maintained in the solutions during the solubility experiments.
3.2 Solubility of Sm and Eu
The results of the dissolved Sm and Eu concentration analyses are listed in Table 2. The dissolved Sm concentration reached equilibrium on reaction day of 20 and remained constant until the end of the experiment (day 223) (Fig. 2 (a)). The average Sm concentration in the supernatant during 20–223 days was (6.21 ± 0.12)×10−8 M. Similarly, Eu concentration reached equilibrium on day 7 and the average Eu concentration in the supernatant during 7–223 days was (8.84 ± 0.28)×10−8 M.
Fig. 2
Solubility of (a) Sm and (b) Eu in the Syn-DB3 groundwater. Closed and open circles are concentrations in supernatant before and after ultrafiltration (UF), respectively.
Once the samples were confirmed to be in equilibrium, the supernatants were subjected to ultrafiltration (10 kDa) for phase separation on the last day of the experiment (Day 223). A single membrane filter unit was used for repeated filtration four times (UF1–UF4) for each sample, and the Sm and Eu concentrations were tracked as the filtration was repeated. The Sm and Eu concentrations in the filtrates increased significantly from the first (UF1) to the second round (UF2) of filtration and leveled off after UF2 (Fig. 3). The low concentrations in the first two filtrates were ascribed to the adsorption of metal ions on the membrane filters and the vessel surface. After UF2, no considerable changes in the Sm and Eu concentrations were observed, indicating that the adsorption sites of the filtration unit were saturated after two rounds of filtration.
The mean values of Sm and Eu concentrations in each of UF3 and UF4 were (3.98 ± 0.04)×10−8 M and (6.08 ± 0.05)×10−8 M, respectively. These values were slightly lower than those measured without phase separation (Fig. 2 and Table 2). The pH of the Sm and Eu samples at the end of the experiments was 8.19 and 8.26, respectively. These values were lower than those obtained under the initial Syn-DB3 conditions (pH 9.03). No significant changes in the concentrations of the main components were detected after the experiment (Table 1). The carbonate content estimated from alkalinity was maintained until the end of the experiment.
3.3 Am Solubility Tested in Over-Saturation Experiments
The absorption spectra of the Am samples stored in quartz cuvettes were periodically measured to monitor their equilibrium status. The absorption data for the first 70 days in Fig. 4(a) were already reported in our previous study [6]. The absorption spectrum of the reference sample (pH 1 in Fig. 4) was consistent throughout the experiment. However, all the samples in the Syn-DB3 showed a significant decrease in absorbance in one day, after which a gradual decrease in absorbance was observed for 70 days, as shown in Fig. 4(a). No peak shifts were observed, as described in our previous study [6]. The absorption spectra of samples at 105 days are presented in Fig. 4(b). The results showed that the Am samples in Syn-DB3 reached equilibrium in 70 days, during which most of the Am precipitated. The absorbance at 505–506 nm was likely due to colloidal Am(III) compounds that are not luminescent, as discussed previously [6].
Fig. 4
Absorption properties of Am(II) in the Syn-DB3 samples. (a) Aging time-dependent absorption area changes of the samples. Data of 0−70 days are the same as in previous study [6]. (b) Absorption spectra of the samples in 105 days.
As no further change in absorbance was confirmed, the samples in the cuvettes were opened to analyze the pH and aqueous Am concentrations on reaction day 231. The results are summarized in Table 3. In comparison to the initial pH values, all the samples showed changes in pH values. The aqueous Am concentrations were measured after phase separation by ultrafiltration (10 kDa). In general, the Am concentration in the filtrates decreased as pH increased. In the range of pH 8–9 (8.45 ± 0.13 on average of S2−4), the averaged aqueous Am concentration was (5.48 ± 1.3)×10−8 (Table 3), which was in the same range as the dissolved Eu concentration ((6.08 ± 0.05)×10−8 M, Table 2) at pH 8.26.
Table 3
Am solubility samples in cuvettes
| Am in Syn-DB3 | Initial pH | Final pH (231 d) | [Am]UF (231 d) (M) |
|---|---|---|---|
|
|
|||
| S0 | 1 | 1 | Not measured |
| S1 | 6.38 | 7.27 | 2.70×10−7 |
| S2 | 7.50 | 8.30 | 6.34×10−8 |
| S3 | 8.36 | 8.48 | 3.58×10−8 |
| S4 | 8.71 | 8.56 | 6.39×10−8 |
| S5 | 10.5 | 10.1 | 2.05×10−8 |
3.4 Comparisons Between Geochemical Modeling and Experimental Results
Geochemical modeling of Sm and Eu solubility was performed under the Syn-DB3 conditions determined at the end of the experiments. The Geochemist workbench program (2023 standard) was used with the ThermoChimie (ver. 11a) TDB as primary input data [14]. Modeling was conducted by excluding low-solubility solid phases in a stepwise manner to approach the measured solubility values.
Solubility curves of Sm. Fig. 5 shows the solubility curve of Sm modeled under Syn-DB3 conditions at the end of the solubility test (see Table 1). After excluding crystalline SmCO3OH(cr), with extremely low solubility, SmCO3OH∙0.5H2O(s) was predicted as the solubility-limiting solid phase (Fig. 5(a)). When SmCO3OH∙0.5H2O(s) and Sm(OH)3(s) were additionally excluded, Sm2(CO3)3(s) was predicted as the solubility-limiting solid phase (Fig. 5(b)).
Fig. 5
Solubility curves and aqueous species of Sm in the Syn-DB3. ThermoChimie 11a TDB was used as an input database with (a) excluding formation of SmCO3OH(cr) and (b) additionally excluding formation of SmCO3OH∙0.5H2O(s), and Sm(OH)3(s). Open red circle is experimentally determined dissolved Sm concentration at pH 8.19 after ultrafiltration.
The dissolved Sm concentration after phase separation was (3.98 ± 0.004) × 10−8 M, as depicted in the solubility curve (open red circle). It is located in between the predicted concentrations of SmCO3OH∙0.5H2O(s) and Sm2(CO3)3(s) based on ThermoChimie TDB. In particular, it is slightly lower (by less than 0.5 log unit) compared to the solubility of Sm2(CO3)3(s).
The Ksp of Sm2(CO3)3(s) (−34.50 ± 2.00, see Table 4 for the reaction) in ThermoChimie TDB was adopted from SKB TR 95-35 [19], in which the Ksp was evaluated mainly based on experimental results by Firsching and Mohammadzadel [20]. In their work, Sm2(CO3)3(s) solid compounds were prepared via decomposition of rare-earth trichloroacetates. Crystalline rare-earth carbonates have been produced using this method [21]. However, the authors did not characterize the solid phases before or after the experiments. The same process was applied to Eu2(CO3)3(s), as described in the next section. Considering the uncertainty of the reported Ksp of Sm2(CO3)3(s) with unidentified crystallinity, the slightly lower solubility for the crystalline Sm2(CO3)3∙xH2O(cr) compound used in this work is reasonable.
Table 4
Thermodynamic data of trivalent Sm, Eu, and Am in ThermoChimie 11a
| Reaction | logK° (Sm) | logK° (Eu) | logK° (Am) | |
|---|---|---|---|---|
|
|
||||
| MCO3OH(cr) + H+ ⇄ M3+ + CO32− + H2O | −10.23±0.65 | −9.63±0.80 | −11.51±1.14 | |
| MCO3OH:0.5H2O(s) + H+ ⇄ M3+ + CO32− + 1.5H2O | −7.31±0.72 | −7.80±1.15 | −8.4 ± 0.50* | |
|
|
||||
| MCO3OH(s) + H+ ⇄ M3+ + CO32− + H2O | - | - | −6.2±1.00* | |
| M2(CO3)3(s) ⇄ 2M3+ + 3CO32− | −34.50±2.00** | −35.00±3.25** | −33.4±2.20* | |
|
|
||||
| M(OH)3(cr) + 3H+ ⇄ M3+ + 3H2O | - | 15.46±1.00 | 15.6±0.60* | |
| M(OH)3(s) + 3H+ ⇄ M3+ + 3H2O | 16.13±1.00 | - | - | |
| M(OH)3(am) + 3H+ ⇄ M3+ + 3H2O | 17.85±1.01 | 17.6±0.84 | 16.9±0.80* | |
|
|
||||
| M3+ + CO32− ⇄ M(CO3)+ | 7.80±0.50 | 7.90±0.50 | 7.90±0.40* | |
| M3+ + 2CO32− ⇄ M(CO3)2− | 12.80±0.60 | 12.9±0.70 | 12.60±0.60* | |
| M3+ + 3CO32− ⇄ M(CO3)33− | 14.80±0.58 | 14.8±0.58 | 14.60±1.00* | |
*Selected in NEA-TDB [12] with crystallinity as AmCO3OH:0.5H2O(s) → (cr), AmCO3OH(s) → (am,hyd), Am2(CO3)3(s) → (am)
** Crystallinity not identified, but assumed as hydrated crystalline phases according to the synthesis protocol [20]
Solubility curves of Eu and aqueous species. Fig. 6 shows the solubility curves of Eu modeled under the Syn- DB3 groundwater conditions at the end of the solubility test (see Table 1). When crystalline EuCO3OH(cr), with extremely low solubility, was excluded, EuCO3OH∙0.5H2O(s) was predicted as the solubility-limiting solid phase (Fig. 6(a)). When EuCO3OH∙0.5H2O(s) was additionally excluded, Eu2(CO3)3(s) was predicted as the solubility-limiting solid phase (Fig. 6(b)). The dissolved Eu concentration after ultrafiltration was (6.08 ± 0.05)×10−8 M), as depicted with open red circle on the solubility curve in Fig. 6. The measured Eu concentration was 10 times higher than the predicted concentration by EuCO3OH∙0.5H2O(s) based on ThermoChimie TDB. The experimental results obtained with crystalline Eu2(CO3)3∙xH2O(cr) compound here closely follow the Ksp of Eu2(CO3)3(s) in ThermoChimie, although the ThermoChimie-selected Ksp of Eu2(CO3)3(s) (−35.00 ± 3.25) has the same issue regarding the unidentified solid phase as described for the case of Sm [19, 20].
Fig. 6
Solubility curves and aqueous species of Eu in the Syn-DB3. ThermoChimie 11a TDB was used as an input database with (a) excluding formation of EuCO3OH(cr) and (b) additionally excluding formation of EuCO3OH∙0.5H2O(s), and Eu(OH)3(s). Open red circle is experimentally determined dissolved Eu concentration at pH 8.26 after ultrafiltration.
As Eu(III) shows strong luminescence properties, the filtered Eu solutions (0.2 μm filter) after the last solubility measurements were examined using TRLFS. Fig. 7 shows the luminescence spectra of the Eu(III) samples and their intensity decay rates. These spectra correspond to the 5D0–7FJ (J = 0–4) electronic transitions of Eu(III) [22, 23]. All three samples exhibited the same luminescence spectra, where the J = 2 peak, which is a hypersensitive peak, was significantly enhanced relative to the J = 1 peak. The strength of the hypersensitive peak was highly sensitive to inner-sphere environmental changes in the Eu(III) center [22]. Additionally, the appearance of the J = 0 peak is indicative of a low-symmetry Eu (III) site due to complex formation. These spectral properties are well-known features of strongly complex Eu ions [16, 23].
Fig. 7
Aqueous Eu(III) species properties. (a) Representative luminescence spectra of Eu filtrates. They correspond to electronic transitions of 5D0 → 7FJ (J = 0–4). (b) Luminescence decays as a function of ICCD delay time. Lifetimes were determined from inverse of the slopes and the inner-sphere hydration number, n(H2O), was calculated from the lifetime according to equation (1). (c) pH-dependent aqueous Eu(III) species distributions under Syn-DB3 condition based on ThermoChimie TDB. The dotted line is at pH 8.26, the final pH value of the Eu solubility samples.
The geochemical modeling results assess that Eu(CO3)+ and Eu(CO3)2− complexes are the main species in Syn-DB3 at pH 8.26 and they are equally distributed (Fig. 7(c)). The observed luminescence spectral properties are similar to those reported for Eu-CO3 complexes [24, 25]. The luminescence decay lifetime (τ) was 164 ± 20 μs on average (Fig. 7(b)). According to Equation (1), the inner-sphere hydration number was 6.0 ± 0.8. We considered nine coordination numbers around the Eu(III) center [22, 23]. These results indicate that on average, three water molecules are displaced from the Eu(III) center under the Syn-DB3 condition. The results are in line with the modeling results; mixture of equal amount of Eu(CO3)+ and Eu(CO3)2− would result in displacing three water molecules on average, as carbonate is typically a bidentate ligand [25].
Solubility curves of Am. Fig. 8 shows the solubility curves of Am modeled under Syn-DB3 groundwater conditions. As the compositions of Syn-DB3 could not be measured after solubility measurements owing to the shortage of samples, modeling was conducted with the initial Syn- DB3 condition. For Sm and Eu, the solution conditions did not significantly change at the end of the experiment. The AmCO3OH compounds (crystalline, hydrated, and amorphous) were the solubility-limiting solid phase. The solubility decreased with crystallinity in the order of Am CO3OH(s) > AmCO3OH∙0.5H2O(s) > AmCO3OH(cr). When AmCO3OH(cr), AmCO3OH∙0.5H2O(s) and Am(OH)3(cr) solid phases were excluded, Am2(CO3)3(s), AmCO3OH(s) and Am(OH)3(am) appeared as solubility-limiting solid phases depending on the pH.
Fig. 8
Solubility curves of Am(III) in Syn-DB3 groundwater condition. ThermoChimie 11a TDB was used as an input database with (a) excluding formation of crystalline AmCO3OH(cr) and (b) additionally excluding crystalline AmCO3OH∙0.5H2O(s) and Am(OH)3(cr). Open red circles are experimentally determined dissolved Am concentrations after ultrafiltration.
The Am concentrations in the filtrates were much higher than the solubility of AmCO3OH:0.5H2O(s) (Fig. 8(a)) and slightly lower than that of AmCO3(OH)(s) (Fig. 8(b)). These values are much lower than those for Am2(CO3)3(s) alone (Fig. 8(b)). According to NEA-TDB, Vol. 5, Am CO3OH∙0.5H2O(s), AmCO3OH(s), and Am2(CO3)3(s) are classified as hydrated crystalline AmCO3OH∙0.5H2O(cr, hyd), amorphous AmCO3OH(am), and amorphous Am2 (CO3)3(am), respectively [2]. However, there are no selected thermodynamic data for dissolution of hydrated crystalline Am2(CO3)3∙xH2O(s), which would be less soluble than the amorphous Am2(CO3)3(am). As similar concentrations were observed in both Am ((5.48 ± 1.3)×10−8 M) and Eu ((6.08 ± 0.05)×10−8 M), formation of the hydrated crystalline Am2(CO3)3∙xH2O(s) phase should be considered as in the case of Eu. More experimental data are required for Am–carbonate systems to better evaluate their solubility in natural water environments.
4. Conclusions
The solubilities of trivalent Sm, Eu, and Am were measured in synthetic KURT-DB3 groundwater, Syn-DB3, and evaluated in comparison with geochemical modeling results based on the ThermoChimie TDB. When Ln2(CO3)3·xH2O(cr) (Ln = Sm, Eu) solid compounds were equilibrated in the Syn-DB3, the dissolved Sm and Eu concentrations were (3.98 ± 0.04) × 10−8 and (6.08 ± 0.05)×10−8 M at pH 8.19 and 8.26, respectively. These values were comparable to those of Sm2(CO3)3(s) and Eu2(CO3)3(s) as evaluated by geochemical modeling. XRD analysis before and after experiments confirmed that the crystallinities of the solid compounds of Ln2(CO3)3·xH2O(cr) were maintained throughout the experiments. TRLFS results support the modeling results on aqueous Eu species, mixtures of Eu(CO3)+ and Eu(CO3)2−, as dominant aqueous species in the Syn-DB3 condition. When Am solubility in Syn-DB3 was measured under over-saturation conditions, the aqueous Am concentrations converged to (5.48 ± 1.3)×10−8 M under pH 8.45 ± 0.13 condition. This is comparable to the dissolved Eu concentrations in equilibration of Eu2(CO3)·xH2O(cr). Our results show that more experimental data are required for Ln/An-carbonate systems, with careful characterization of the solid phases, to better evaluate their solubilities in domestic natural water systems.
