Isolation and characterization of a new polyoxometalate ligand, H₃SbW₁₄O₅₀¹⁰⁻, and its interactions with f-elements

Abstract

We report the synthesis and characterization (Raman, FTIR, DLS, single crystal XRD) of Cs₁₀H₃SbW₁₄O₅₀·6H₂O, which contains a new polyoxometalate building block: H₃SbW₁₄O₅₀¹⁰⁻ (SbW₁₄). Solution-state fluorescence, DLS, and UV-vis absorbance results with Nd³⁺, Eu³⁺, Am³⁺, and Cm³⁺ confirm that SbW₁₄ acts as an efficient complexant and fluorescence sensitizer for f-elements.

---

Polyoxometalates (POMs) represent a large family of molecules with numerous applications, and their range of compositions and structures covers most of the periodic table.^1–3 While the POM core structures are typically built upon d-block metals⁴ (W, Mo, V, Cr, Nb, Ta), their compounds often incorporate other elements and complex metal ions, and are counter-balanced by cations. POM compounds that have been characterized thus far cover over 70 elements (from hydrogen to as heavy as curium^5–8), representing one of the most diverse class of molecules.

POMs can be seen as the inorganic equivalents of organic chelators or as an ever-expanding family of molecules that uses d-block metal octahedra (e.g., WO₆²⁻) as unitary blocks, instead of carbon–carbon bonds. POM core structures include the Lindqvist⁹ (e.g., Mo₆O₁₉²⁻, H_xNb₆O₁₉^x−8 and H_xTa₆O₁₉^x−8), Peacock–Weakley¹⁰ (e.g., W₅O₁₈⁶⁻), Keggin¹¹ (e.g., PMo₁₂O₄₀³⁻, SiW₁₁O₃₉⁸⁻), Well–Dawson¹² (e.g., P₂W₁₈O₆₂⁶⁻), Anderson–Evans^13,14 (e.g., Mo₇O₂₄⁶⁻), and Preyssler¹⁵ (e.g., P₅W₃₀O₁₁₀¹⁵⁻) ions. However, the Keggin, Wells–Dawson, and Lindqvist structures are the most widely studied.^4,15–17 For instance, in the Keggin structure ([XW₁₂O₄₀]ⁿ⁻), over 25 candidates are available to be incorporated as heteroelement (X). This leads to an overwhelming number of Keggin derivatives,¹⁸ all with potentially distinct physiochemical properties.

In this context, the subset of antimono-polytungstates (Sb-POTs), has been relatively overlooked, despite having truly unique properties. In previously reported examples of Sb-POTs, the stereoactive lone pair¹⁹ of Sb(III) strongly influenced the structures, albeit still leading to Keggin derivatives.²⁰ In addition, the low charge density of Sb³⁺ shifts the POMs’ stability window to slightly higher pHs, often leading to protonated Sb-POTs.^21–23 This often results in larger POMs, yet still all derived from the Keggin structure. Several Sb-POTs have been reported in the last few decades—all based on successive condensations of [SbW₉O₃₃]⁹⁻ (SbW₉), which is a tri-lacunary Keggin ion. SbW₉ is arguably the most used Sb-POT structure and has served as precursor for larger POMs ([NaSb₉W₂₁O₈₆]¹⁸⁻, [Sb₂W₂₁O₆₉]⁶⁻, [H₂SbW₂₂O₇₆]¹⁴⁻, [Sb₈W₃₆O₁₃₂]²⁴⁻…).²⁰

In recent years, POMs have also received increasing attention for use as chelators for f-elements. In fact, POMs offer an interesting alternative relative to the more traditional organic chelators as they can act as oxygen-donor ligands but contain elements that are not easily incorporable in organic ligands. POMs can also stabilize usually unstable oxidation states of actinides, such as Am(VI).⁸ The high-molecular weight of POMs also make them useful to perform microscale crystallization tests with radioisotopes.⁵ In our quest to expand the chemistry of f-elements with POM ligands, we attempted to transpose our synthetic protocols recently used^5,7,24 for Keggin POMs with B³⁺, Ga³⁺, Si⁴⁺, Ge⁴⁺, and P⁵⁺ to antimony (Sb³⁺ – Fig. 1). Unexpectedly, the substitution of the heteroelements mentioned above for Sb³⁺ did not yield the expected Keggin POM. Not only does the behavior of antimony depart from that of the other heteroelements, but it also yielded a novel POM structure. This therefore unlocked an opportunity to explore f-element chemistry with a new type of ligand.

Fig. 1 Comparative synthesis scheme for the Keggin ion POM ligands and the new antimony POM reported here (unprotonated formula: SbW₁₄O₅₀¹³⁻). While most p-block elements react with WO₄²⁻ (at pH 4–7) to form Keggin-type POMs, the same reaction with Sb³⁺ yield a different type of POM: SbW₁₄O₅₀¹³⁻. This new POM can then be used to interact with lanthanide and actinides cations (e.g., Eu³⁺, Nd³⁺, Am³⁺, Cm³⁺).

Herein we report the single crystal XRD structure of the POM compound Cs₁₀H₃SbW₁₄O₅₀·6H₂O. This antimony(III) polytungstate is based on a new building block [H₃SbW₁₄O₅₀]¹⁰⁻ (SbW₁₄). This POM was obtained by the typical synthetic route that has consistently led to Keggin-type structures with other p-block elements.¹⁵ Additionally, we probed the potential binding of SbW₁₄ to trivalent lanthanides (Eu³⁺ and Nd³⁺) and actinides (Am³⁺ and Cm³⁺) in solution. Eu³⁺ and Cm³⁺ were used for their fluorescence properties while Nd³⁺ and Am³⁺ were used for their absorbance properties. All four f-elements also have similar ionic radii, facilitating intercomparisons. The behavior of SbW₁₄ was also contrasted with the previously known POM SbW₉. The results show that SbW₁₄ can act as an effective aqueous ligand with f-elements. This represents the first study of interactions between antimony-containing ligands and heavy actinides.

SbW₁₄ was obtained by conversion of the precursor [SbW₉O₃₃]⁹⁻ (SbW₉). This precursor was synthesized as reported by Bösing et al.²⁰ without modification. The SbW₉ to SbW₁₄ conversion reaction was found remarkably straightforward – happening at room temperature, ambient pressure, and in aqueous solution. From the synthesis by Bösing et al., a 2 mM solution of SbW₉ was prepared (30 mg in a 100 mM acetate buffer at pH 5.5). Then, an equal volume of 6 M CsCl was added (final concentration of 3 M). After 24–48 h crystals of SbW₁₄ appeared. The pH was measured at various points during the crystallization period and was stable at 5.5. For a more detailed synthesis procedure see ESI.† Single crystals of SbW₁₄ were collected for structure determination, plus Raman and FTIR analysis. Fig. S1 (ESI†) clearly shows distinct Raman and FTIR features between SbW₉ and SbW₁₄.

Interestingly, the new Sb-POT could only be crystallized through the addition of caesium counterions, as opposed to pH variation. For POMs, the most conventional manner to control their speciation is by controlling the pH, to tune their hydrolysis and condensation reactions. We assume that SbW₁₄ had thus far remained undiscovered because there is generally very little information in the literature about Sb-POTs with caesium counterions. This new crystal structure is unique in its synthesis, obtained through counterion-mediated reactions as opposed to pH-controlled reactions. The new structure, fully formulated as Cs₁₀H₃SbW₁₄O₅₀·6H₂O, crystallizes in the orthorhombic space group P2₁2₁2₁ with a unit cell volume of 6609.5 (3) A³; for more information see Table S1 (ESI†). One manner to describe the new Sb-POTs is as a combination of the [SbW₉O₃₃]⁹⁻ subunit, nicknamed the ‘basket’, with a [W₅(OH)₃O₁₄]⁻ ‘handle’ subunit (Fig. 2). The structure itself has only one symmetry element; a mirror plane along its long axis. Thus, the structure falls under the point group C_s. The ‘basket’ subunit, SbW₉, is the original precursor and itself is often described as the tri-lacunary version of parent Keggin structure ([XW₁₂O₄₀]ⁿ⁻).^25,26 In other words, removal of three adjacent tungstates from a parent but hypothetical structure [SbW₁₂O₄₀]⁵⁻, would results in the SbW₉ subunit. We suspect that SbW₉ does not convert to [SbW₁₂O₄₀]⁵⁻ due to the size of Sb³⁺ and steric hindrance created by its lone electron pair. In SbW₉, six tungstates surround the central [SbO₃]³⁻, with the remaining three tungstates opposite to the lone pair on Sb.

Fig. 2 Top. Structure overview of the isolated new polyoxometalate, SbW₁₄O₅₀¹³⁻. Color code: W (maroon), O (red), Sb (cyan). Caesium and hydrogen atoms are omitted for clarity. Bottom. Raman spectra for SbW₉ and SbW₁₄.

Bond valence sum calculations in SbW₁₄ confirms the oxidation state at +III for Sb (Table S2, ESI†). The six surrounding tungstates can be thought of as the ‘rim of the basket’ while the remaining three tungstates can be the ‘bottom’. The ‘handle’ itself, [W₅(OH)₃O₁₄]⁻, can further be subdivided and described as a [W₃(OH)₃O₁₀]⁵⁻ with two edge-sharing cis-di-oxyl-tungstens, [O=W=O]²⁺, binding to two of the three tungstates in the trimer. These two flanking tungstens then bind to the ‘basket’ completing their six-fold coordination. The ‘handle’ is attached to the basket via four out of the six tungstates surrounding Sb in the SbW₉ subunit. As a result, all tungstates have a coordination number of six, as seen for most polyoxotungstates, with rare exceptions.²⁷ Bond valence sum calculation on the O and W positions additionally reveal the potential location of three protons on the POMs, leading to the formula [H₃SbW₁₄O₅₀]¹⁰⁻, Table S3 (ESI†).

Due to its unique composition and structure, the complexation of SbW₁₄ with f-elements became of particular interest. The solution behavior of this new POT was explored through fluorescent spectroscopy (Eu³⁺vs. Cm³⁺) and UV-Visible spectrophotometry (Nd³⁺vs. Am³⁺). This represents the first study on transplutonium elements with antimony-based ligands. To first probe the physiochemical properties and complexation behavior for SbW₁₄, excitation and emission spectra were taken with Eu³⁺ at a 1 : 2 ratio (Eu³⁺ : SbW₁₄), as we expected a 1 : 1 or 1 : 2 complex based on potential binding sites of the SbW₁₄ structure (Fig. 2). Excess POM to f-element was also used to ensure full complexation. Parallel experiments were done with SbW₉, for comparison purposes and to confirm that SbW₁₄ does not convert back to its precursor upon binding to the f-elements. Both SbW₉ and SbW₁₄ were found to spontaneously bind to the trivalent lanthanides and actinides.

Fig. 3a and b show the normalized excitation and emission spectra, respectively for Eu³⁺ with SbW₉ and SbW₁₄. Both POMs bind to Eu³⁺ and sensitize its luminescence, but evidently via different pathways (Fig. 3a). The preferential excitation pathway for Eu³⁺ is via the POM band (λ_ex = 290–310 nm) in the case of SbW₁₄ but via direct excitation of Eu³⁺ (4f–4f sharp transitions at 360–410 nm, main peak λ_ex = 396 nm) in the case of SbW₉. The excitation peak via the POM has a maximum at 343 nm for SbW₉, compared to 303 nm for SbW₁₄. When comparing the emission spectra, SbW₁₄ results in more defined Eu³⁺ emission peaks, with intensities about 16-fold higher (Fig. S3, ESI†).

Fig. 3 Fluorescence studies on Eu–SbW₁₄ (purple curves) and Cm–SbW₁₄ (blue curves) compared to Eu-SbW₉ (green curves) and CmSbW₉₉ (orange curves). Eu³⁺ fluorescence: (a) excitation and (b) emission. Cm³⁺ fluorescence: (c) excitation and (d) emission. Fig. 3a is normalized to the excitation peak at 396 nm as Eu³⁺ sensitization via the POM band of SbW₉ is inefficient. See Fig. S1–S5 (ESI†) for raw intensity spectra. Solution conditions were as follows: [Eu³⁺] = 1 mM, and [Cm³⁺] = 100 μM, [SbW₉] and [SbW₁₄] at 2 mM and 200 μM, respectively. Sb-POTs were dissolved in 0.1 M acetate at pH of 5.5. *Indicates the POM excitation band.

Lifetime dependent fluorescence further shows different complexation behaviors for SbW₉versus SbW₁₄ (Table S4, ESI†). While both SbW₉ and SbW₁₄ yield mono-exponential decay curves with Eu³⁺ (suggesting only one type of complex(es) present), the lifetimes are significantly different, at 336 μs and 621 μs, respectively. Based on the Kimura equations,²⁸ which empirically correlate fluorescence lifetimes to hydration sphere, the number of coordinating water molecules decreases from three for Eu–SbW₉ to one for Eu–SbW₁₄. Although we were unable to crystallize a complex of SbW₁₄ with an f-element, the solution-state fluorescence results suggest that SbW₁₄ is more amenable to fulfil the f-element coordination sphere than SbW₉, leaving it less exposed to solvent molecules. Additional fluorescence titration experiments (Fig. S6–S8, ESI†), revealed that the Eu–SbW₁₄ likely undergoes a gradual complexation reaction when the ratio SbW₁₄ : Eu is varied. By analogy to other POMs,⁴ we tentatively ascribe the observed spectral changes to the sequential formation of 1 : 1 and 1 : 2 complexes, however the precise solution-state speciation of the f-element/SbW₁₄ systems will require further investigations.

When studies were extended from 4f to 5f-elements (Am³⁺ and Cm³⁺) some deviations could be discerned in their binding behavior and physiochemical properties. Fluorescence results (Fig. 3c, d and Fig. S4, S5, ESI†) confirmed that Cm³⁺ also binds to both SbW₉ and SbW₁₄. However, contrary to Eu³⁺, the excitation spectra revealed that, for both SbW₉ and SbW₁₄, sensitization of Cm³⁺via the POM is more efficient than direct excitation. The POM excitation band is at 276 nm for Cm–SbW₉versus 303 nm for Cm–SbW₁₄. Sensitization of Cm³⁺ is also more efficient via SbW₁₄ than SbW₉, by a factor of ∼4 (Fig. S4, ESI†). In other words, SbW₁₄ leads to brighter complexes than SbW₉ for both Eu³⁺ and Cm³⁺. The Cm³⁺ emission spectra are also distinct, with peak maximum at 606.0 nm for SbW₉ and 609.0 nm for SbW₁₄ and an increase in intensity of a factor of 1.5. Both Cm–SbW₉ and Cm–SbW₁₄ complexes exhibit a significant peak shift when compared to free Cm³⁺ under similar conditions (598.4 nm).⁵ While the Kimura equation has been shown to not be applicable for most reported curium POMs,⁵ apart from the simplest polytungstate [Cm(W₅O₁₈)₂]⁹⁻,⁶ the time-dependent fluorescence herein is somewhat consistent with the Eu³⁺ analog (i.e., lifetime longer for Cm–SbW₁₄vs. Cm–SbW₉). The measured lifetimes for Cm³⁺ are 133 μs for SbW₉ to 177 μs for SbW₁₄ (Table S4, ESI†). This corresponds to four and three water molecules, respectively. The trend is consistent with the Eu³⁺ results and indicates a lower hydration number for the Cm–SbW₁₄ complex, relative to Cm–SbW₉, but the absolute number of water molecules may not be accurate. The difference in hydration between Eu³⁺ and Cm³⁺ complexes could arise from different binding sites on SbW₁₄.

However, we posit that the Kimura equation fails to accurately describe the Cm³⁺ hydration for the Sb-POTs and starts to be less valid for Eu³⁺ as well, in this particular context. We attribute this to the alkali counterions, which are typically considered inert but have been shown to form strong ion pairs with certain POM complexes. Herein, SbW₉ uses Na⁺ whereas SbW₁₄ uses Cs⁺, which could impact fluorescence properties. Based on the emission spectra for SbW₉, the observed species are 1 : 1 complexes, [Eu(SbW₉O₃₃)(H₂O)₃]⁶⁻ and [Cm(SbW₉O₃₃)(H₂O)₄]⁶⁻, but the speciation for Cm/Eu–SbW₁₄, remains ambiguous as we do not know its coordination mode. We also performed dynamic light scattering (DLS) experiments on SbW₁₄, and SbW₉, with and without Eu³⁺ and Cm³⁺ (Fig. S9 and Table S5, ESI†). The measured hydrodynamic diameters could support the formation of 1 : 1 complexes for both POMs under the tested conditions, although the size difference between 1 : 1 and 1 : 2 complexes may not be enough to discriminate with this technique. The role of counterions on the apparent hydrodynamic size measured by DLS also adds some uncertainty to the applicability of DLS to POM complexes.

To gain more insight into the speciation, UV-Vis absorbance experiments with Nd³⁺ and Am³⁺ were performed (Fig. 4). Both metals previously demonstrated sensitive absorbance peak shifting for different Nd³⁺/Am³⁺-POT species.^5,8 Am³⁺ typically exhibit significant peak shifting with the free ion absorbance at 504 nm, 1 : 1 Am³⁺ : POT complexes at 506–512 nm, and 1 : 2 complexes at 514–520 nm.⁵ For both Nd³⁺ and Am³⁺, there is evidence of complexation with both SbW₉ and SbW₁₄ (Fig. 4). The absorbance bands of Nd³⁺ and Am³⁺ are both affected in the presence of SbW₉ and SbW₁₄, relative to their free ion state. In the case of Am³⁺, SbW₉ leads to asymmetric broadening of the peak, likely indicative of a lower symmetry relative to the aqua ion. For Am³⁺ with SbW₁₄, the peak exhibits a more obvious change with a dominant peak appearing at 508.0 nm. Note that the Am³⁺–Sb-POT experiments were limited to 20 μM concentrations, as opposed to 1 mM for Nd³⁺ or 100 μM for Cm³⁺, which may have limited the complexation efficacy of the POMs. The results nonetheless confirm that both SbW₉ and SbW₁₄ bind to trivalent lanthanides and actinides in solution. The similar extinction coefficient also supports the formation of 1 : 1 complexes under these conditions.

Fig. 4 UV-Vis spectra for Nd³⁺ (top) and Am³⁺ (bottom) with SbW₉ and SbW₁₄. Solution conditions: [Nd³⁺] = 1 mM. [Am³⁺] = 20 μM. [SbW₉] and [SbW₁₄] = 2 mM and 40 μM, respectively. Sb-POTs dissolved in 0.1 M acetate buffer (pH 5.5).

In conclusion, this study reports a new antimono-polytungstate structure, SbW₁₄, and its solution behavior with lanthanide and actinide elements. The Cs₁₀H₃SbW₁₄O₅₀·6H₂O structure was synthesized by an unconventional route of counterion-mediated conversion (versus traditional hydrolysis and condensation reactions). This compound was characterized via single crystal XRD, FTIR, Raman microscopy and compared to its precursor, SbW₉. Solution-state fluorescence, UV-Vis spectroscopy, and DLS experiments with Eu³⁺vs. Cm³⁺, and Nd³⁺vs. Am³⁺, revealed consistent complexation of SbW₉ and SbW₁₄ to the lanthanides or actinides. Furthermore, we demonstrate that the new structure functions as a more efficient sensitizer and for both Eu³⁺ and Cm³⁺, relative to SbW₉. Future work will focus on elucidating the solution-state speciation and expand the counterion conversion of other Sb-POTs and interactions with other f-elements.

This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Heavy Element Chemistry program at Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. Release number: LLNL-JRNL-870910.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. M. K. Kinnan, W. R. Creasy, L. B. Fullmer, H. L. Schreuder-Gibson and M. Nyman, Eur. J. Inorg. Chem., 2014, 2361–2367.
2. J.-J. Chen, M. D. Symes and L. Cronin, Nat. Chem., 2018, 10, 1042–1047.
3. A. Gaita-Ariño, F. Luis, S. Hill and E. Coronado, Nat. Chem., 2019, 11, 301–309.
4. N. I. Gumerova and A. Rompel, Chem. Soc. Rev., 2020, 49, 7568–7601.
5. I. Colliard, J. R. I. Lee, C. A. Colla, H. E. Mason, A. M. Sawvel, M. Zavarin, M. Nyman and G. J.-P. Deblonde, Nat. Chem., 2022, 14, 1357–1366.
6. I. Colliard and G. J.-P. Deblonde, Chem. Commun., 2024, 60, 5999–6002.
7. I. Colliard and G. J.-P. Deblonde, JACS Au, 2024, 4, 2503–2513.
8. H. Zhang, A. Li, K. Li, Z. Wang, X. Xu, Y. Wang, M. V. Sheridan, H.-S. Hu, C. Xu, E. V. Alekseev, Z. Zhang, P. Yan, K. Cao, Z. Chai, T. E. Albrecht-Schönzart and S. Wang, Nature, 2023, 616, 482–487.
9. I. Lindqvist, Ark. Kemi, 1953, 5, 247–250.
10. R. D. Peacock and T. J. R. Weakley, J. Chem. Soc. A, 1971, 1836–1839.
11. J. F. Keggin and W. L. Bragg, Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character, 1997, 144, 75–100.
12. B. Dawson, Acta Crystallogr., 1953, 6, 113–126.
13. J. S. Anderson, Nature, 1937, 140, 850.
14. H. T. Evans, J. Am. Chem. Soc., 1948, 70, 1291–1292.
15. M. T. Pope, Heteropoly and Isopoly Oxometalates, Springer, Berlin, Heidelberg, 1st edn, 1983.
16. N. I. Gumerova and A. Rompel, Sci. Adv., 2023, 9, eadi0814.
17. A. Misra, K. Kozma, C. Streb and M. Nyman, Angew. Chem., Int. Ed., 2020, 59, 596–612.
18. C. A. Ohlin, Phys. Chem. Chem. Phys., 2020, 22, 4043–4050.
19. B. M. Benin, K. M. McCall, M. Wörle, D. Borgeaud, T. Vonderach, K. Sakhatskyi, S. Yakunin, D. Günther and M. V. Kovalenko, Chem. Mater., 2021, 33, 2408–2419.
20. M. Bösing, I. Loose, H. Pohlmann and B. Krebs, Chem. – Eur. J., 1997, 3, 1232–1237.
21. H. Naruke and T. Yamase, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 1992, 48, 597–599.
22. Z. Tang, M. Wang, X. Jia, S. Xie, P. Chen, D. Wang, L. Chen and J. Zhao, Inorg. Chem., 2022, 61, 14648–14661.
23. X. Xin, Y. Ma, L. Hou, Y. Wang, X. Xue, J. Lin and Z. Han, Inorg. Chem., 2019, 58, 9567–9571.
24. I. Colliard and G. J.-P. Deblonde, Inorg. Chem., 2024, 63, 16293–16303.
25. J. F. Keggin, Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character, 1934, 144, 75–100.
26. C. Falaise, G. Mpacko Priso, N. Leclerc, M. Haouas and E. Cadot, Inorg. Chem., 2023, 62, 2494–2502.
27. G. Deblonde and I. Colliard, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2024, 80, 667–670.
28. T. Kimura, G. R. Choppin, Y. Kato and Z. Yoshida, Radiochim. Acta, 1996, 72, 61–64.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2326569. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4cc05558f

---