Layered metal sulfides: Exceptionally selective agents for radioactive strontium removal

Abstract

In this article, we report the family of robust layered sulfides K_2xMn_xSn_3-xS₆ (x = 0.5–0.95) (KMS-1). These materials feature hexagonal [Mn_xSn_3-xS₆]^2x− slabs of the CdI₂ type and contain highly mobile K⁺ ions in their interlayer space that are easily exchangeable with other cations and particularly strontium. KMS-1 display outstanding preference for strontium ions in highly alkaline solutions containing extremely large excess of sodium cations as well as in acidic environment where most alternative adsorbents with oxygen ligands are nearly inactive. The implication of these results is that simple layered sulfides should be considered for the efficient remediation of certain nuclear wastes.

Current growing interest in nuclear power as a potential solution for global energy may also raise serious environmental and health concerns due to highly radioactive nuclear waste. ⁹⁰Sr is one of the major heat producers and biohazards in nuclear wastes. The removal of radioactive strontium is essential to reducing the risk of human exposure to radiation and for the considerable cost savings due to minimization of the storage space requirements for the nuclear waste (1, 2). There is a long-standing and continuous research effort to develop highly selective strontium adsorbents for application in a variety of wastes. Inorganic ion exchangers possess a number of advantages as Sr²⁺ adsorbents over the conventional organic ion-exchange resins, such as superior chemical, thermal, and radiation stability (3). The naturally abundant ion exchangers such as clays (3, 4) and zeolites (3) are not effective as strontium adsorbents in nuclear waste solutions with extreme pH values because of their immediate decomposition (i.e., dissolution of their aluminum). Commercial inorganic adsorbents capable for Sr²⁺ adsorption in highly alkaline solutions with extremely high salt concentrations [i.e., conditions present in many nuclear waste types (1)] are mainly limited to titanates and silicotitanates (3, 5, 6). These materials, however, are not effective for Sr²⁺ capture even at mildly acidic conditions (pH < 4–5) because protons inhibit ion exchange (7). Only doped antimony silicates represent Sr²⁺ ion exchangers efficient in strongly acidic environment (pH ≤ 1) (8). Finally, solvent extraction and extraction chromatography methods have proven promising for strontium decontamination of acidic and alkaline nuclear wastes (9–11).

We show here that layered sulfide compounds with ion-exchangeable interlayer cations are not inhibited by protons and can be very efficient Sr²⁺-ion-removers over a wide pH range. Layered chalcogenides with ion-exchange properties are relatively few (12–16) and are mainly limited to alkali intercalated early transition metal dichalcogenides A_xMQ₂ (A = alkali ion; M = early transition metal from groups 4,5 and 6; Q = S, Se, Te) (13–16). Such materials, however, are not suitable for practical applications because of their hydrolytic instability (16).

Here, we report that the layered K_2xMn_xSn_3-xS₆ (x = 0.5–0.95) (KMS-1) exhibit huge selectivity for Sr²⁺ ions in both acidic and basic environments. They are especially effective in strongly alkaline environments in the presence of an enormous excess of Na⁺ ions. This property is highly relevant to the problem of nuclear waste remediation and points to the class of metal sulfide compounds as a highly promising source of materials for helping to solve it.

Results and Discussion

The KMS-1 materials can be easily prepared on a multigram scale and high purity with solid-state or hydrothermal synthesis techniques. They are extremely stable in atmosphere and water, while they display high thermal stability [see supporting information (SI) Figs. 5–7 (complete materials, instrumentation, and methods are provided in SI Materials and Methods, SI Figs. 5–12, and SI Table 2)]. Single-crystal data,^§ obtained from hexagonal-shaped crystals (Fig. 1A) synthesized hydrothermally, revealed a layered structure of K_1.9Mn_0.95Sn_2.05S₆ (CdI₂ structure type). The layer is built up by edge-sharing “Mn/Sn”S₆ octahedra with Mn and Sn atoms occupying the same crystallographic position and all sulfur ligands being three-coordinated (Fig. 1C). K⁺ ions are found between the layers and are positionally disordered, a feature that gives them high mobility and the ability to exchange with other ions particularly with strontium (Fig. 1D).

Fig. 1.

Structure and morphology. (A) SEM image of a typical crystal of K_1.9Mn_0.95Sn_2.05S₆ (KMS-1) with a hexagonal plate-like shape obtained with a hydrothermal reaction. (B) SEM image of polycrystalline K_2xMn_xSn_3-xS₆ (x = 0.5–0.95) obtained by a solid-state reaction. (C) Part of the layer framework of K_1.9Mn_0.95Sn_2.05S₆ viewed down the c-axis. The Mn-Sn and S atoms are represented by blue and yellow balls, respectively. (D) View of the structure, with a polyhedral representation of the layers, along the c-axis.

Indeed, polycrystalline samples (Fig. 1B) of KMS-1 can completely replace the K⁺ ions with Sr²⁺ within a few hours. This was confirmed with energy dispersive spectroscopy (EDS). Powder x-ray diffraction (PXRD) measurements showed that the exchanged material is isostructural with the pristine confirming a topotactic ion exchange (Fig. 2). A shift of the (003) and (006) Bragg peaks to lower 2θ values (or higher d-spacing) was observed in the diffraction patterns of Sr²⁺-exchanged products. Alkaline earth ions have a great tendency to be hydrated, and this results in the large c-axes for the Sr²⁺-exchanged materials. For example, thermal analysis data for Sr²⁺-exchanged samples revealed the presence of ≈5 H₂O molecules per formula unit (SI Fig. 7). The expected Sr:Mn ratio in the exchanged materials should be 1 to satisfy the charge-balance requirements. Surprisingly, a Sr:Mn ratio of ≈0.5 was found because of an unusual oxidation of Mn²⁺ to Mn³⁺ occurring during the strontium exchange processes. The formation of Mn³⁺ is supported by x-ray photoelectron spectroscopy data (see SI Fig. 8 and 9 and SI Table 2).

Fig. 2.

X-ray powder diffraction patterns for the pristine K_2xMn_xSn_3-xS₆ (x = 0.5–0.95) and Sr²⁺-exchanged materials.

To assess the strontium removal capacity of KMS-1, we performed Sr²⁺ ion-exchange equilibration studies. The Sr²⁺ ion exchange-equilibrium data are shown in Fig. 3A. These data can be fitted with the Langmuir isotherm model expressed as

where q (mg/g) is the amount of the cation adsorbed at the equilibrium concentration C_e (ppm), q_m is the maximum adsorption capacity of the adsorbent, and b (liters/mg) is the Langmuir constant related to the free energy of the adsorption.

Fig. 3.

Equilibrium ion-exchange data. (A) Sr²⁺-adsorption isotherms for K_2xMn_xSn_3-xS₆·2H₂O (x = 0.95). The line represents the fitting of the data with the Langmuir isotherm model. (B) Plot of Sr uptake and K_d^Sr vs. the initial concentration of Sr²⁺.

The maximum Sr²⁺ exchange capacity q_m of KMS-1 (x = 0.95) was found to be 77 ± 2 mg/g (or 0.9 mmol/g), which compares well with those of the best strontium adsorbents [Sr²⁺ capacities = 1.0–2.0 mmol/g (7)].

The affinity of the material for Sr²⁺ can be expressed in terms of the distribution coefficient K_d (for definition, see Materials and Methods and ref. 17). In Fig. 3B are presented the plots of K_d^Sr and percent strontium uptake vs. the initial strontium concentration (0.45–79.5 ppm) in the solution. It can be seen that the percent Sr²⁺ removal (83.3–99.4%) is high over a wide range of initial concentrations. Remarkably, the maximum K_d^Sr value of 1.58 × 10⁵ ml/g (Table 1) obtained lies among the highest reported in the literature for Sr²⁺ adsorbents (18).

Table 1.

Selected data for Sr²⁺ ion exchange experiments of KMS-1

Exchanging cations	Sr2+/Na+ ratio	Conditions	Initial concentration, ppm	Final concentration, ppm	% removal	Kd, ml/g
Sr2+	—	pH ∼ 3.2, V:m ∼ 971 ml/g	4.09	0.15	96.3	2.49 × 104
Sr2+	—	pH ∼ 7, V:m ∼ 1,000 ml/g	4.60	0.03	99.3	1.52 ×105
Sr2++Na+ (0.1 M)	1/1,887	pH ∼ 13, V:m ∼ 971 ml/g	4.65	0.01	99.8	4.50 ×105
Sr2++Na+ (5 M)	1/2.17 ×105	pH ∼ 14, V:m ∼ 1,000 ml/g	2.15	0.17	92.1	1.16 ×104
Sr2++Ca2+ +Mg2++Na++Cs+	—	pH ∼ 11, V:m ∼ 990 ml/g	3.70 (Mg)	0.48 (Mg)	94.5 (Sr)	6.64 × 103 (Mg)
			11.14 (Ca)	1.17 (Ca)		8.40 × 103 (Ca)
			4.60 (Sr)	0.24 (Sr)		1.83 × 104 (Sr)
			9.17 (Cs)	3.17 (Cs)		1.87 × 103 (Cs)
			25.96 (Na)	22.42 (Na)		156.5 (Na)

The % Sr removal and strontium distribution coefficient values are in boldface type.

Taking into account that the pH of nuclear waste and contaminated ground water may vary from strongly acidic to extremely alkaline, we have also studied the effect of pH on Sr²⁺ adsorption. Sr²⁺ ion exchange experiments of KMS-1 with solutions of various pH (1.7–10) indicated significant Sr²⁺ uptake in the entire pH range tested (Fig. 4A). The capability of KMS-1 to absorb strontium from acidic solutions is particularly impressive. Specifically, the strontium adsorption at pH ∼ 4.3 did not deviate significantly from the maximum possible (≈95.9% of its maximum theoretical) and remained significant (37.2% of the maximum one) even at pH ∼ 1.7. We have also conducted ion-exchange experiments with trace concentrations of Sr²⁺ (4–6 ppm) at pH ∼ 0.4, 1.2, 2.1, and 3.2 (V:m ∼ 1,000 ml/g) that showed K_d values of 202, 383, 517, and 2.49 × 10⁴ ml/g respectively (Fig. 4B and Table 1). These K_d^Sr values revealed a remarkable strontium affinity of KMS-1 under strongly acidic conditions. For comparison, the commercial strontium adsorbent Na₂Ti₉O₂₀·xH₂O (NaTi) displays a K_d of 5,000 ml/g at pH ∼ 4 and 32 ml/g at pH ∼ 2 (for a more detailed comparison, see below) (19).

Fig. 4.

Comparison of KMS-1 with known sorbents. (A) Sr²⁺ uptake by K_2xMn_xSn_3-xS₆·2H₂O (x = 0.75) as a function of pH. The dashed line indicates the maximum theoretical Sr capacity (58 mg of Sr/g) calculated for K_xMn_xSn_3-xS₆·2H₂O (x = 0.75). (B) Representation of K_d^Sr obtained with ion-exchange reactions in acidic solutions (pH ∼ 0.4–3.2). The initial Sr²⁺ concentration was ≈4 ppm in all reactions with the exception of the reaction at pH ∼ 0.4 with an initial strontium concentration of ≈6 ppm. (Inset) K_d^Sr at pH 0.4, 1.2, and 2.1. (C) Diagram representing the dependence of K_d^Sr for various materials on the pH. A K_d value greater than 10,000 ml/g is considered excellent (see ref. 19). The data for materials NaTi (sodium titanate), CST (sodium silicotitanate), CRY-1 (cryptomelane-type manganese oxide), SOMS (Sandia octahedral molecular sieves), and SbSi were obtained from the following references: NaTi, refs. 6, 8, 19, 20, and 22; CST, refs. 8 and 20; CRY-1, ref. 23; SOMS, ref. 24; and SbSi, ref. 8.

We performed competitive Sr²⁺-Na⁺ experiments (with trace concentrations of ≈1.5–5 ppm of strontium and 1.90 × 10³ to 2.17 × 10⁵-fold excess of sodium) to examine the selectivity of KMS-1 for Sr²⁺ over Na⁺ under environmentally relevant highly alkaline conditions (pH ∼ 13–14). Sodium is the most abundant cation in highly alkaline nuclear waste (1). The results (Table 1) showed quantitative (≈99.8%) removal of Sr²⁺ from a solution of 0.1 M NaOH (pH ∼ 13, Sr²⁺:Na⁺ = 1:1,887, reaction time of ≈24 h) with an enormous K_d of 4.50 × 10⁵ ml/g. Even with much higher Na⁺ concentration (5 M) and pH ∼ 14—i.e., conditions usually present in nuclear waste—KMS-1 retained their exceptional selectivity for strontium with high K_d^Sr values of 2.62 (Sr²⁺ initial concentration of ≈2.1 ppm) to 5.84 × 10³ (Sr²⁺ initial concentration of ≈1.5 ppm) ml/g obtained in 24-h reactions. The KMS-1 continued to absorb Sr²⁺ even after 4 days' reaction^¶ with ≈92.1% removal of Sr²⁺ (Sr²⁺ initial concentration of ≈2.1 ppm) and a K_d^Sr of 1.16 × 10⁴ ml/g, which is in very good comparison with those (3.70 × 10³ to 2.26 × 10⁵ ml/g) of the most efficient inorganic adsorbents (20, 21).

We also tested the effect of a mixture of competitive cations (Na⁺, 1.1 mM; Ca²⁺, 2.8 × 10⁻¹ mM; Mg²⁺, 1.5 × 10⁻¹ mM; Cs⁺, 6.9 × 10⁻² mM) on the Sr²⁺ (0.02 mM) ion exchange in alkaline solutions (pH ∼ 11, V:m = 990 ml/g). The results (Table 1) revealed that the strontium affinity of KMS-1 is hardly affected by the presence of the competitive cations as is apparent by the high K_d^Sr value (1.83 × 10⁴ ml/g) obtained under these conditions. The selectivity order determined by the comparison of K_d values is Sr²⁺ > Ca²⁺ > Mg²⁺ > Cs⁺ ≫ Na⁺. This order shows the preference of the sulfide materials for the softer ions (e.g., Sr²⁺ vs. Ca²⁺)^‖ and the ions with the highest charge (i.e., 2+ vs. 1+ cations).

Finally, it would be useful to compare the Sr²⁺ affinity and selectivity of KMS-1 with those of the state-of-the art inorganic adsorbents. In Fig. 4C, the dependence of reported K_d^Sr values of various materials (including KMS-1) with the pH of the solution is represented. KMS-1 outperforms commercial adsorbents like sodium silicotitanate (7, 8, 20) and sodium titanate (6–8, 19, 20, 22), as well as others like CRY-1 (cryptomelane-type manganese oxide) (23), under strongly acidic conditions. The KMS-1 contain soft basic sites, the S²⁻ ligands, which display small affinity for hard proton ions. Instead, the strontium exchange of the majority of oxygen-containing materials strongly interferes with proton ions having high affinity for the hard O²⁻ ligands. At the same time, KMS-1 is nearly as effective as the commercial exchangers in highly alkaline environments (pH ∼ 14). Sandia octahedral molecular sieves (SOMS) also seem efficient under acidic and alkaline conditions (24), but their selectivity for strontium is dramatically reduced even with the presence of low Na⁺ concentrations (24). Sb silicate (SbSi) is the only inorganic adsorbent highly selective for strontium in pH as low as 1 (8), but its ability to be efficient at pH ≥ 14 (pH of many nuclear waste types) is questionable because of its appreciable dissolution at pH ≥ 13 (25).

Conclusion

The KMS-1 compounds are capable of highly selective interactions with Sr²⁺ or SrOH⁺ [this is the main form of Sr²⁺ at pH > 12.8 (19)] and strong discrimination against hard ions such as Na⁺ and H⁺. They possess this function despite the fact that their layers have no special features for structural preference (e.g., pore or window openings or rings of sizes specific for Sr). To the best of our knowledge, they represent the only examples ever reported of nonoxidic inorganic ion exchangers showing exceptional selectivity for strontium. The KMS-1 compounds with their soft negatively charged sulfide layers, therefore, point to a new paradigm in terms of the chemical classes that can be considered as selective adsorbents under extreme pH conditions for the effective treatment of radioactive wastes. The importance of metal sulfides in this area is poised to grow.

Materials and Methods

Polycrystalline K_2xMn_xSn_3-xS₆·yH₂O (x = 0.5–0.95; y = 2–5) (KMS-1) can be synthesized by solid-state reaction of Sn (1.9 mmol), Mn (1.1 mmol), K₂S (2 mmol), and S (16 mmol) at 500°C or by hydrothermal reaction of Sn (60 mmol), Mn (30 mmol), K₂CO₃ (30 mmol), and S (180 mmol) at 200°C. The yield for both preparation methods was found to be ≈80%. Single crystals of KMS-1, suitable for x-ray analysis, were obtained by a hydrothermal reaction of K₂S (0.40 mmol), MnCl₂ (0.20 mmol), Sn (0.40 mmol), and S (0.40 mmol) at 220°C. Energy dispersive spectroscopy (EDS) and inductively coupled plasma (ICP)–atomic emission (AES) analyses were used to determine the composition of KMS-1. The purity of various samples was verified by x-ray powder diffraction.

A typical ion-exchange experiment of KMS-1 with Sr²⁺ is as follows. In a suspension of KMS-1(0.07 mmol, 40 mg) in water (20 ml), an excess of SrCl₂·6H₂O (1.0 mmol) was added as a solid. The mixture was kept under magnetic stirring or constant shaking for ≈12 h. Then, the dark brown polycrystalline material was isolated by filtration, washed several times with water, acetone, and ether, and dried in air. The distribution coefficient K_d, used for the determination of the affinity and selectivity of compounds KMS-1 for Sr²⁺, is given by the equation K_d = (V[(C₀ − C_f)/C_f])/m, where C₀ and C_f are the initial and equilibrium concentration of Mⁿ⁺ (ppm), V is the volume (ml) of the testing solution, and m is the amount of the ion exchanger (g) used in the experiment.

The Sr²⁺ uptake from solutions of various concentrations was studied by the batch method at V:m ∼ 1,000 ml/g, at room temperature and 24-h contact. The solids were then separated by centrifugation and filtration [through filter paper (Whatman No. 1)]. The Sr²⁺ content of the solutions was determined with ICP-AES. The data obtained were used for the determination of Sr²⁺ adsorption isotherms. Competitive ion-exchange experiments of KMS-1 were also carried out with the batch method at V:m ratio (1,000 ml/g), at room temperature and contact time of 1–7 days.