K₃MoPO₇: the first molybdenum phosphate with edge-sharing MoO₆ octahedra and PO₄ tetrahedra

Abstract

Herein we report the discovery of a novel molybdenum phosphate K₃MoPO₇, in which the basic building unit, [P₂Mo₂O₁₄]⁶⁻ ring cluster anions, contains unique edge-sharing MoO₆ and PO₄ polyhedra. The structural stability of K₃MoPO₇ is confirmed by both experimental and first-principles studies.

---

Corner-, edge- and face-sharing patterns are three basic polyhedra connection modes in inorganic chemistry.¹ These three ways of linking and their combinations significantly enrich the structural chemistry and result in diverse polymorphisms. In fact, the linkage mainly depends on the ionic charge, ionic radii and coordination number.² In inorganic oxides, the corner-sharing pattern is the most common way to be adopted by the M_xO_y (M is cation) polyhedra, such as in borates,³ silicates,⁴ aluminates⁵ and phosphates.⁶ The edge- or face-sharing patterns, on the other hand, are much less likely to exist in the above compounds. In general, both patterns could be obtained in extreme conditions only.⁷

The molybdenum phosphate system also has rich crystal chemistry with a large number of structures containing tunnels, cages and micropores.⁸ Till now hundreds of molybdenum phosphates have been synthesized, and many of them have large surface areas which can be used in ion exchange, catalysis, and separations.⁹ It is observed that in all the discovered molybdenum phosphates, the MoO₆ octahedra always share their corners with PO₄ tetrahedra to build up frameworks.⁸ According to Pauling's third and fourth rules,² polyhedra with high charge cations and low coordination numbers would be more stable if they are connected by sharing vertices while less stable if sharing edges or faces, due to the cation–cation electrostatic repulsion between centroids. Thus, for MoO₆ octahedra and PO₄ tetrahedra they are very likely to share minimal vertices (i.e., corner-shared) to build frameworks since the Mo⁶⁺ and P⁵⁺ cations have quite high positive charges. The edge- or face-sharing of MoO₆ octahedra and PO₄ tetrahedra have not been found yet. In this work we synthesize the first molybdenum phosphate K₃MoPO₇ in which the MoO₆ octahedra are sharing-edged with PO₄ tetrahedra so form the unique [P₂Mo₂O₁₄]⁶⁻ anion rings. The structural stability is investigated by thermal measures and first-principles calculations.

K₃MoPO₇ was synthesized through solid-state reaction in stoichiometric ration with K₂CO₃, MoO₃ and K₂HPO₄ as the starting materials. It can also be obtained from stoichiometric composition melt of above raw materials with slow cooling since it is a congruent compound. The proof will show in thermal characterization aspect below. Small transplant single crystals with light yellow color were obtained through spontaneous crystallization. The crystal structure was solved and refined on the basis of single-crystal data. The XRD patterns of as-synthesized samples show good agreement with the calculated one derived from the single crystal data.

K₃MoPO₇ crystallizes in a monoclinic system with the centrosymmetric space group C2/m. In an asymmetric unit, K, Mo, P and O occupy three, one, one, and five crystallographically unique positions, respectively. The structure of K₃MoPO₇ is featured by isolated [P₂Mo₂O₁₄]⁶⁻ anions, with K⁺ cations filled up the space around the anions (Fig. 1a). The [P₂Mo₂O₁₄]⁶⁻ anion is constructed by two MoO₆ octahedra and two PO₄ tetrahedra in which the Mo and P atoms are coplanar. These polyhedra alternately connect with each other via either corner-sharing or edge-sharing patterns to form an 8-membered polyanion ring with the symmetry of C_2h point group (Fig. 1b). The simple [P₂Mo₂O₁₄]⁶⁻ fundamental building unit is first found in all inorganic compounds.‡

Fig. 1 (a) Polyhedral view of the K₃MoPO₇ structure projected along the b axis. Pink octahedrons and cyan tetrahedrons stand for MoO₆ anion groups and PO₄ anion groups, respectively. (b) Ball-and-stick model for a [P₂Mo₂O₁₄]⁶⁻ anion. The K, Mo, P, O atoms are shown as gray, teal, pink and red, respectively.

It is worthy to note that the MoO₆/PO₄ edge-sharing configuration in the [P₂Mo₂O₁₄]⁶⁻ ring is unique and distinguished to any other microscopic pattern in molybdenum phosphates. Our comprehensive survey for all transition-metal phosphates in inorganic crystal structural database reveals that indeed the edge-sharing structure with MO₆ octahedra is very rare as the M cations are d⁰ transition-metal (M = Ti⁴⁺, Nb⁵⁺, Mo⁶⁺, W⁶⁺, Ta⁵⁺, V⁵⁺). There is only one phosphovanadate, Li₂VPO₆, having the similar edge-sharing structure.¹⁰ On the other hand, for the non-d⁰ transition-metal MO₆ octahedra (M = Co³⁺, Mn²⁺, Cr³⁺, V³⁺, V²⁺ etc.) they are relatively easy to form the edge-sharing structure with PO₄ tetrahedra; tens of these compounds containing this connection pattern have been found. This result actually agrees with the Pauling's rules since the non-d⁰ transition metals have relatively lower charge which could reduce the electrostatic repulsion between Mⁿ⁺ and P⁵⁺ cations. Therefore, the discovery of K₃MoPO₇ provides a very rare example that violates the common rules for structural stability, which has significant implications to explore the molybdenum phosphates with new structures and new functions.

In [P₂Mo₂O₁₄]⁶⁻ anion ring the Mo⁶⁺ cation of MoO₆ octahedron is located in the distorted environments due to second-order Jahn–Teller (SOJT) effect.¹¹ The centroid shifts toward a vertex of the octahedron (local C₃ [111] direction), which results in the formation of three short (1.749(2)–1.754(2) Å) and three long (2.186(2)–2.353(3) Å) Mo–O bonds, as shown in Fig. 1b, and all the three short bonds are outwards from the [P₂Mo₂O₁₄]⁶⁻ anion ring. As a comparison, in the PO₄ group P–O bond lengths are ranged from 1.52 to 1.57 Å, and this group almost keeps the regular tetrahedral shape. In K₃MoPO₇ all bond lengths are comparable to those in other reported molybdenum phosphates. The calculated total bond-valances sums on Mo and P are 5.82 and 4.85, respectively. These values match well with the expected oxidation states.¹²

It is interesting that each MoO₆ octahedron in the [P₂Mo₂O₁₄]⁶⁻ cluster anion contains three free vertices. This is also very rarely observed in molybdates. In general, the formation of the structures with MO₆ octahedra that contain three (or more) free vertices is very unfavorable due to the strong trans influence of the terminal M–O bonds (i.e., the Lipscomb restriction).¹³ In K₃MoPO₇ the ring-like [P₂Mo₂O₁₄]⁶⁻ groups are very different from all other microscopic structures in molybdenum phosphates where the MoO₆ groups are usually connected to other groups as chain, layer or three-dimensional network. The zero-dimensional topological structure of the [P₂Mo₂O₁₄]⁶⁻ cluster anion results in more dangling oxygen ions existed in the MoO₆ octahedra. Therefore, this simple group might be an ideal staring building unit to further polymerize with other groups for the construction of more complex solids, e.g., open-framework structures or polyoxometalates, for the applications of catalysis, luminescence, magnetism, medicine, etc.¹⁴

The stability of K₃MoPO₇ was experimentally demonstrated by the thermal behaviour measured with a differential scanning calorimetry (DSC). No obvious endo- or exothermic peak was observed until the melting point of the compound was reached near 490 °C (Fig. 2a). This demonstrates that the [P₂Mo₂O₁₄]⁶⁻anionic structure keeps stable in ambient atmosphere from room temperature to the melting point. After melting, the cooled solid remains were characterized by XRD and were identified that it crystallized into K₃MoPO₇ (Fig. 2b). A few weak extra peaks were observed since the volatilization of melt and component deviation. The DSC and XRD results clearly reveal that this compound melts congruent.

Fig. 2 (a) DSC curves of K₃MoPO₇. (b) X-ray powder diffraction patterns of K₃MoPO₇. The middle curve is powder of as-synthesized. The top one is powder after melting. The bottom curve is simulated XRD derived from the single crystal data.

To further investigate the mechanism of the structurally stability of the edge-sharing [PO₄] and [MoO₆] in the [P₂Mo₂O₁₄] ring, the first-principles studies on the electronic density difference and Mulliken atomic/bond populations¹⁵ were performed¹⁶ by the plane-wave pseudopotential method implemented in the CASTEP package.¹⁷ The electronic density difference calculation produces a density difference field which shows the changes in the electron distribution as all chemical bonds are formed in the system. Fig. 3a and b exhibit the electronic density difference in the edge-sharing and corner-sharing MoO₆/PO₄ groups, respectively, which shows a different charge redistribution among the edge-sharing O(3) and corner-sharing O(2). In detail, the O(3) atoms obtain more electronic charges from the neighboring Mo atoms compared with the O(2) atoms, but the charge transfer from the neighboring P atoms to the O(3) is less than to the O(2) atoms. This indicates a necessary migration of electronic charges from P–O bond to Mo–O bond due to the modification of the chemical environment around the oxygen atoms as they are charged from corner-sharing to edge-sharing.

Fig. 3 Contour plots of the electronic density difference on the planes formed by (a) Mo, P and edge-sharing O(3) atoms and (b) Mo, P and corner-sharing O(2) atoms. The off-plane O atoms are represented by small red balls.

The more quantitative results come from the Mulliken analysis; P–O(3) and P–O(2) bond populations are 0.59 and 0.74, respectively, while those on the Mo–O(3) and Mo–O(2) population is 0.28 and 0.23, respectively. Thus, the increase (decrease) of bond covalence between Mo (P) and edge-sharing O compared with those around the corner-sharing O account for the stability of the whole [Mo₂P₂O₁₄] group. It should be noted, however, that the Mulliken analysis reveals that the effective charges on O(3) (−0.96) is less than that on O(2) (−1.03). This implies that indeed the edge-sharing oxygen atoms are less stable than the corner-sharing oxygen, in consistence with the Pauling's second rule (i.e., the oxygen anion having more charge tends to be more stable). We suggest that the zero-dimensional closed-ring configuration of the [Mo₂P₂O₁₄] group would be beneficial to the formation of edge-sharing structure compared with the chain, layer or three-dimensional network.

Conclusions

A unique [P₂Mo₂O₁₄]⁶⁻ ring-shape anion containing edge-sharing MoO₆ octahedra and PO₄ tetrahedra has been discovered in a novel molybdenum phosphate K₃MoPO₇. The basic building unit with edge-sharing MoO₆ and PO₄ polyhedra was first found. K₃MoPO₇ is a congruent compound with melting point at about 490 °C. The first-principles calculations reveal that the unusual edge-sharing feature attributes to the migration of electronic charges from P–O bond to Mo–O bond. In other words, the edge-sharing Mo–O bond is of larger covalence than that of the corner-sharing Mo–O bond. Our studies presented in this work, therefore, would greatly prompt the development of molybdenum phosphates and have implications on the progress of structural chemistry.

Acknowledgements

This work was supported by the National Natural Science Foundation of China under Grant nos 11174297 and 91022036, and the National Basic Research Project of China (nos 2010CB630701 and 2011CB922204), and Foundation of the Director of Technical Institute of Physics and Chemistry, CAS.

Notes and references

1. (a) U. Mueller, Inorg. Struc. Chem, Wiley, 1993; (b) I. Lindqvist, O. Hassel, M. Webb and M. Rottenberg, Acta Chem. Scand., 1950, 4, 1066; (c) A. F. Wells, Struct. Inorg. Chem, Oxord, 1962; (d) D. Long, E. Burkholder and L. Cronin, Chem. Soc. Rev., 2007, 36, 105; (e) N. V. Izarova, M. T. Pope and U. Kortz, Angew. Chem., Int. Ed., 2012, 51, 9492.
2. L. Pauling, J. Am. Chem. Soc., 1929, 51, 1010.
3. (a) J. D. Grice, P. C. Burns and F. C. Hawthorne, Can. Mineral., 1999, 37, 731; (b) C. T. Chen, T. Sasaki, R. K. Li, Y. C. Wu, Z. S. Lin, Y. Mori, Z. G. Hu, J. Y. Wang, G. Aka and M. Yoshimura, Nonlinear Optical Borate Crystals, Wiley-VCH, 2012; (c) P. C. Burns, J. D. Grice and F. C. Hawthorne, Can. Mineral., 1995, 33, 1131; (d) F. C. Hawthorne, P. C. Burns and J. D. Grice, Rev. Mineral., 1996, 33, 41; (e) X. Yan, S. Luo, Z. Lin, Y. Yue, X. Wang, L. Liu and C. Chen, J. Mater. Chem. C, 2013, 1, 3616; (f) X. Yan, S. Luo, Z. Lin, J. Yao, R. He, Y. Yue and C. Chen, Inorg. Chem., 2014, 53, 1952; (g) C. Chen, S. Luo, X. Wang, G. Wang, X. Wen, H. Xu, X. Zhang and Z. Xu, J. Opt. Soc. Am. B, 2009, 26, 1519; (h) S. C. Wang, N. Ye, W. Li and D. Zhao, J. Am. Chem. Soc., 2010, 132, 8779.
4. (a) C. Meade, R. J. Hemley and H. K. Mao, Phys. Rev. Lett., 1992, 63, 1387; (b) G. V. Gibbs, Am. Mineral., 1982, 67, 421; (c) M. Grimsditch, Phys. Rev. Lett., 1984, 52, 2379; (d) L. Levien, C. T. Prewitt and D. J. Weidner, Am. Mineral., 1980, 65, 920; (e) R. M. Hazen, L. W. Finger, R. J. Hemley and H. K. Mao, Solid State Commun., 1989, 72, 507.
5. (a) C. Y. Chen, H. X. Li and M. E. Davis, Microporous Mater., 1993, 2, 17; (b) Z. Luan, C. Cheng, W. Zhou and J. Klinowski, J. Phys. Chem., 1995, 99, 1018; (c) Z. Luan, M. Hartmann, D. Zhao, W. Zhou and L. Kevan, Chem. Mater., 1999, 11, 1621; (d) P. Mondal and J. W. Jeffery, Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1975, 31, 689.
6. (a) A. K. Cheetham, G. Ferey and T. Loiseau, Angew. Chem., Int. Ed., 1999, 38, 3268; (b) M. Estermann, L. B. Mccusker, C. Baerlocher, A. Merrouche and H. Kesler, Nature, 1991, 352, 320; (c) P. Y. Feng, X. H. Bu, S. H. Tolbert and G. D. Stucky, J. Am. Chem. Soc., 1997, 119, 2497; (d) R. K. Brow, R. J. Kirkpatrick and G. L. Turner, J. Am. Chem. Soc., 1993, 76, 919.
7. (a) S. Jin, G. Cai, W. Wang, M. He, S. Wang and X. Chen, Angew. Chem., Int. Ed., 2010, 49, 4967; (b) H. Huppertz and B. von der Eltz, J. Am. Chem. Soc., 2002, 124, 9376; (c) H. Emme and H. Huppertz, Z. Anorg. Allg. Chem., 2002, 628, 2165; (d) H. Emme and H. Huppertz, Chem.–Eur. J., 2003, 9, 3623; (e) H. Huppertz and H. Emme, J. Phys.: Condens. Matter, 2004, 16, S1283; (f) H. Emme and H. Huppertz, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2005, 61, i29; (g) H. Huppertz and W. Schnick, Chem.–Eur. J., 1997, 3, 249.
8. (a) R. C. Haushalter, K. G. Strohmaier and F. W. Lai, Science, 1989, 246, 1289; (b) G. Costentin, A. Leclaire, M. M. Borel, A. Grandin and B. Raveau, Rev. Inorg. Chem., 1993, 13, 77; (c) R. C. Haushalter and L. A. Mundi, Chem. Mater., 1992, 4, 31; (d) A. Leclaire, A. Guesdon, M. M. Borel, F. Berrah and B. Raveau, Solid State Sci., 2001, 3, 877.
9. (a) S. Natarajan and S. Mandal, Angew. Chem., Int. Ed., 2008, 47, 4798; (b) R. Murugavel, A. Choudhury, M. G. Walawalkar, R. Pothiraja and C. N. R. Rao, Chem. Rev., 2008, 108, 3549.
10. V. C. Korthuis, R. D. Hoffmann, J. Huang and A. W. Sleight, J. Solid State Chem., 1993, 105, 294.
11. P. S. Halasyamani, Chem. Mater., 2004, 16, 3586.
12. I. D. Brown and D. Altermatt, Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem., 1985, 41, 244.
13. M. T. Pope and A. Muller, Angew. Chem., Int. Ed., 1991, 30, 34.
14. (a) C. L. Hill, Chem. Rev., 1998, 98, 1; (b) P. Gouzerh and A. Proust, Chem. Rev., 1998, 98, 77; (c) N. Mizuno and M. Misono, Chem. Rev., 1998, 98, 199; (d) A. Muller, F. Peters, M. T. Pope and D. Gatteschi, Chem. Rev., 1998, 98, 239; (e) J. T. Rhule, C. L. Hill, D. A. Judd and R. F. Schinazi, Chem. Rev., 1998, 98, 327; (f) T. Yamase, Chem. Rev., 1998, 98, 307; (g) D. E. Katsoulis, Chem. Rev., 1998, 98, 359; (h) A. M. Khenkin, L. Weiner, Y. Wang and R. Neumann, J. Am. Chem. Soc., 2001, 123, 8531; (i) A. Dolbecq, E. Dumas, C. R. Mayer and P. Mialane, Chem. Rev., 2010, 110, 6009.
15. R. S. Mulliken, J. Chem. Phys., 1955, 23, 1833.
16. The local density approximation and ultrasoft pseudopotentials are adopted in these calculations. O 2s²2p⁴, P 3s²3p³, K 3s²3p⁶4s¹ and Mo 4s²4p⁶4d⁵5s¹ are treated as valence electrons. The kinetic energy cutoff of 500 eV and Monkhorst–Pack 2 × 4 × 3 k-point meshes are used. The choice of these computational parameters is good enough to ensure the accuracy of present purpose.
17. (a) M. C. Payne, M. P. Teter, D. C. Allan, T. A. Arias and T. A. J. D. Joannopoulos, Rev. Mod. Phys., 1992, 64, 1045; (b) S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. J. Probert, K. Refson and M. C. Payne, Z. Kristallogr., 2005, 220, 567.

---

Footnotes

† CCDC 996800. For crystallographic data in CIF or other electronic format see DOI: 10.1039/c4ra03612c

‡ Crystal data for K₃MoPO₇: M = 712.42 g mol⁻¹, monoclinic, a = 13.990(3) Å, b = 5.8600(12) Å, c = 9.6157(19) Å, β = 111.68(3)°, Z = 2, V = 732.54(87) ų, space group C2/m. 5379 reflections measured, 1545 independent reflections (R_int = 0.0261). The final R₁ value was 0.029. The final wR(F²) value was 0.0776 (I > 2σ(I)). The final R₁ value was 0.0311 (all data). The final wR (F²) value was 0.0790 (all data). The goodness of fit on F² was 1.064.

---