The largest single-strand molecular wheel: Ga₂₀ from a targeted, diolate-induced size modification of the Ga₁₀‘gallic wheel’

Abstract

A Ga₂₀ single-strand wheel has been prepared by a targeted, propane-1,3-diolate-induced size modification of the known Ga₁₀‘gallic wheel’; the Ga₂₀ reverts back to Ga₁₀ on treatment with an excess of MeOH.

---

Polynuclear metal complexes with a single-strand wheel structure continue to attract a lot of interest from many groups. The prototypes were [Cr₈F₈(O₂CBu^t)₁₆]¹ and [Fe₁₀(OMe)₂₀(O₂CCH₂Cl)₁₀],² the latter dubbed the ‘ferric wheel’. Many other single-strand wheels are also now known, for Fe³ as well as other metals such as Co,⁴Cr,⁵Cu,⁶Dy,⁷Mn,⁸Ni,⁹ and V.¹⁰ More recently, we reported the Ga analogue of the ‘ferric wheel’, i.e., [Ga₁₀(OMe)₂₀(O₂CMe)₁₀] (1), which we called the ‘gallic wheel’;³ the analogous wheel with a different carboxylate was reported by Zafiropoulos and co-workers.¹¹ Single-strand molecular wheels almost always contain an even number of metal ions, and with very few exceptions possess metal nuclearities of 12 or less; these exceptions include Mn₁₆,^8b and three M₁₈ (M = Fe, Ga) wheels.^3,12 Multiple-strand wheels are also known, such as Mn₂₄¹³ and the giant torus-shaped complexes Mn₈₄,¹⁴Mo₁₅₄,¹⁴ and Mo₁₇₆.¹⁵

Wheel complexes are of interest for a number of reasons. For paramagnetic 3d metal atoms, even-membered antiferromagnetic wheels represent model systems for one-dimensional antiferromagnetism, magnetic anisotropy, and quantum effects such as coherent tunnelling of the Néel vector.¹⁶ However, there are also many wheels with large S values, such as Ni₁₂,^9cMn₁₂,^8cMn₁₆,^8b and Mn₈₄¹⁴ that in addition are single-molecule magnets (SMMs).

Our recent interest in this area has been targeted at developing new synthetic routes to large M_x (x > 12) wheels, particularly methods that might also allow some level of control of the metal nuclearity (i.e. the wheel size). Our main interest, apart from the inherent synthetic challenge, is to achieve a range of wheel sizes to allow a study of progressive changes to their properties as they approach the 1-D limit, i.e., a chain. This requires expanding the wheel size, and effectively rules out using a template approach,^5c,17 which is much less feasible for larger wheels since it requires a correspondingly larger template. Instead we have been exploring the substitution of diolates for two adjacent MeO⁻groups of the M₁₀ wheels. Since the MeO⁻groups in the latter lie above and below the central cavity, their substitution by diolates such as propane-1,3-diolate (pd²⁻) would likely affect the wheel curvature and yield a bigger wheel. A smaller one is unlikely, given the steric congestion that would result in the central hole. A previous use of pdH₂ in Ga chemistry had provided the wheel compound [Ga₁₈(O₂CR)₆(pd)₁₂(pdH)₁₂(NO₃)₆](NO₃)₆ (3)³ but the presence of both pd²⁻ and pdH⁻, as well as the additional presence of coordinated NO₃⁻ ligands, complicated matters by making unclear the exact effect of the pd²⁻. This initial study thus could not answer the important question at hand, namely what happens to the Ga₁₀ wheel size if pairs of MeO⁻ ligands are replaced by pd²⁻groups? We have now answered this question, and in doing so have discovered a complex with a record size for a single-strand wheel.

Complex 1 was slowly dissolved with stirring in pdH₂/MeCN (1:7, v/v) to give a colourless solution, and this was filtered and layered with Et₂O/Me₂CO (1:1, v/v). Colourless crystals of [Ga₂₀(pd)₂₀(O₂CMe)₂₀]·25MeCN (2·25MeCN) grew slowly over several days and were isolated by filtration; the yield was 20%.† Complex 2 was also obtained, and in higher yields of ∼40%, from the reaction of GaCl₃ with 3 equivs of NaO₂CMe·3H₂O in pdH₂/MeCN (1:7, v/v), followed by filtration and layering of the clear solution with Et₂O/Me₂CO (1:1, v/v). The structure‡ of 2 (Fig. 1) comprises twenty Ga^III ions linked through MeCO₂⁻ and pd²⁻ bridges to form a puckered, single-strand wheel of virtual D₅ point group symmetry. It can be conveniently described as a pentagon of {Ga₄(O₂CMe)₄(pd)₄} units (Ga20-Ga1-Ga2-Ga3, Ga3-Ga4-Ga5-Ga6, etc.) linked at each end by the O atoms of an acetate and two pd²⁻groups (Fig. 1, bottom). The resulting complex can also be described as constructed of five linear {Ga₃(η¹:η¹:μ-O₂CMe)₂(η²:η²:μ₃-pd)₂} ‘edge’ units (Ga–Ga–Ga = 177.25–179.42°) held together by five {Ga(η¹:η¹:μ-O₂CMe)₂(η²:η²:μ₃-pd)₂} ‘hinge’ units (Fig. 2, top); note that every Ga₂ pair in the molecule is thus bridged by two pd²⁻O atoms and an MeCO₂⁻group. As in 1, the Ga atoms are six-coordinate and near-octahedral; unlike 1, however, the Ga₂₀ wheel 2 is not planar. The Ga–Ga distances and Ga–O(pd)–Ga angles lie in the 2.902–2.950 Å and 94.78–101.25° ranges, respectively.

Fig. 1 The structure of the Ga₂₀ wheel 2: (top) the complete molecule with Ga atom labels; (bottom) the repeating {Ga₄(O₂CMe)₄(pd)₄} unit (Ga20–Ga1–Ga2–Ga3) and the means of attachment to Ga4 of an adjacent unit. Hydrogen atoms have been omitted for clarity. Colour code: Ga purple, O red, C grey.

Fig. 2 The five-fold symmetric Ga₂₀ structure: (top) the linkage of Ga₃‘edge’ units by Ga ‘hinge’ atoms. Black thick lines indicate the Ga–Ga vectors; (bottom) a space-filling representation. Colour code: Ga purple, O red, C grey, H white.

A space-filling representation (Fig. 2, bottom) shows that 2 has a diameter of 25.5 Å, with a central hole of 10.0 Å diameter; the corresponding values for 1 are 16.7 Å and 8.1 Å, respectively. In both complexes, there is no residual electron density in the central hole, and the wheels stack to form nanotubular channels.

Consideration of the formulas of 1 and 2 shows that they both belong to a family of ‘gallic wheels’ of general formula [Ga_n(OR)_2n(O₂CMe)_n], which is not the case for 3: for 1, n = 10 and RO⁻ = MeO⁻; for 2, n = 20 and (OR)₂ = pd²⁻. As for 2, every Ga₂ pair in 1 is bridged by two alkoxide O atoms and one MeCO₂⁻group. Thus, the complexes differ only in the identity of the alkoxide, and the resulting n value, and we can therefore answer the question posed: the replacement of two adjacent MeO⁻groups of the Ga₁₀ wheel 1 with the diolate pd²⁻ has caused a doubling of the wheel size to give the Ga₂₀ wheel 2. No doubt several factors contribute to this, including diolate bite and torsion angle restrictions, steric repulsion that would result between the pd²⁻–CH₂CH₂CH₂– backbones in the central cavity for smaller wheels, and others; all these require careful analysis and modelling.

The conversion of 1 to 2 by treatment with pdH₂ can be reversed by dissolution of 2 in MeOH, which leads to an alcohol substitution reaction and the isolation in 75% yield of colourless complex 1; the latter was confirmed by X-ray crystallography. The bi-directional interconversion between 1 and 2 is summarized in eqn (1).

2[Ga₁₀(OMe)₂₀(O₂CMe)₁₀] (1) + 20 pdH₂→ [Ga₂₀(pd)₂₀(O₂CMe)₂₀] (2) + 40 MeOH (1)

To probe solution species and conversions further, an ¹H NMR spectroscopic investigation was carried out. The NMR spectrum of 1 in CDCl₃ exhibited the three resonances expected for an intact Ga₁₀ wheel retaining the D_5d symmetry of the solid state: the signals are at 2.13 (singlet, 3H), 3.36 (singlet, 3H), and 3.51 ppm (singlet, 3H), corresponding to one acetate and two symmetry-inequivalent methoxide sets of ligands, respectively. The ¹H NMR of 2 in CDCl₃ shows two resonances at 1.95 (singlet, 3H) and 2.12 ppm (singlet, 3H) assignable to the two symmetry-inequivalent sets of acetates under D₅ symmetry, but is otherwise very complicated as expected for the many inequivalent pd²⁻ CH₂groups, and diastereotopic H nuclei of CH₂ pairs at the 1 and 3 positions of pd²⁻. The preliminary conclusion is that 2 also retains its solid-state structure in CDCl₃. Dissolution of 2 in CD₃OD, however, causes conversion to 1, as indicated by a white precipitate identified as 1 by IR spectral comparison with authentic material, and the presence in the NMR spectrum now of only the two resonances due to free pdD₂, a quintet at 1.75 ppm and a triplet at 3.66 ppm.

In conclusion, we have successfully converted [Ga₁₀(OMe)₂₀(O₂CMe)₁₀] (1) to the analogue in which pairs of MeO⁻groups have been replaced by the diolate pd²⁻ and found that the resulting wheel has doubled in nuclearity to [Ga₂₀(pd)₂₀(O₂CMe)₂₀] (2). This interesting result also provides the largest single-strand molecular wheel to date, and augurs well for further wheel size modifications being possible as a function of the diolate employed; such studies are currently in progress. There are also magnetic implications of this work, even though Ga(III) is diamagnetic: extension to large M₂₀ wheels for paramagnetic metals such as Cr(III), Fe(III), etc would provide larger analogues of the known wheels for these metals and thus provide a greater range of wheel sizes for the study of magnetic properties vs. size, and the relationship to the 1-D spin chain limit. Note that Cr₂₀ and Fe₂₀ analogues of 2 would contain essentially only one type (or at least very similar types) of pairwise exchange parameters, since all M₂ pairs have the same bridging ligands, as mentioned earlier. In effect, crystals of such materials would represent a collection of single-size (monodisperse) spin chains in which the magnetic properties would be specific for that chain length rather than represent the average for a distribution of chain lengths, as is the usual case in the study of spin chains.

This work was supported by the National Science Foundation (CHE-0414555).

Notes and references

1. V. Gerbeleu, Yu. T. Struchkov, G. A. Timko, A. S. Batsanov, K. M. Indrichan and G. A. Popovich, Dokl. Akad. Nauk SSSR, 1990, 313, 1459.
2. K. L. Taft and S. J. Lippard, J. Am. Chem. Soc., 1990, 112, 9629.
3. P. King, T. C. Stamatatos, K. A. Abboud and G. Christou, Angew. Chem., Int. Ed., 2006, 45, 7379 and references cited therein.
4. (a) J. K. Beattie, T. W. Hambley, J. A. Klepetko, A. F. Masters and P. Turner, Chem. Commun., 1998, 45; (b) E. K. Brechin, O. Cador, A. Caneschi, C. Cadiou, S. G. Harris, S. Parsons, M. Vonci and R. E. P. Winpenny, Chem. Commun., 2002, 1860.
5. (a) I. M. Atkinson, C. Benelli, M. Murrie, S. Parsons and R. E. P. Winpenny, Chem. Commun., 1999, 285; (b) M. Eschel, A. Bino, I. Felner, D. C. Johnston, M. Luban and L. L. Miller, Inorg. Chem., 2000, 39, 1376; (c) M. Affronte, S. Carretta, G. A. Timco and R. E. P. Winpenny, Chem. Commun., 2007, 1789 , and references therein.
6. (a) G. A. Ardizzoia, M. A. Angaroni, G. LaMonica, F. Cariati, M. Moret and N. Masciocchi, J. Chem. Soc., Chem. Commun., 1990, 1021; (b) G. Mezei, P. Baran and R. G. Raptis, Angew. Chem., Int. Ed., 2004, 43, 573.
7. L. G. Westin, M. Kritikos and A. Caneschi, Chem. Commun., 2003, 1012.
8. (a) G. L. Abbati, A. Cornia, A. C. Fabretti, A. Caneschi and D. Gatteschi, Inorg. Chem., 1998, 37, 1430; (b) M. Murugesu, W. Wernsdorfer, K. A. Abboud and G. Christou, Angew. Chem., Int. Ed., 2005, 44, 892; (c) D. Foguet-Albiol, T. A. O’Brien, W. Wernsdorfer, B. Moulton, M. J. Zaworotko, K. A. Abboud and G. Christou, Angew. Chem., Int. Ed., 2005, 44, 897.
9. (a) A. J. Blake, C. M. Grant, S. Parsons, J. M. Rawson and R. E. P. Winpenny, J. Chem. Soc., Chem. Commun., 1994, 2363; (b) A. L. Dearden, S. Parsons and R. E. P. Winpenny, Angew. Chem., Int. Ed., 2001, 40, 151; (c) C. Cadiou, M. Murrie, C. Paulsen, V. Villar, W. Wernsdorfer and R. E. P. Winpenny, Chem. Commun., 2001, 2666.
10. (a) H. Kumagai and S. Kitagawa, Chem. Lett., 1996, 471; (b) Y. Chen, Q. Liu, Y. Deng, H. Zhu, C. Chen, H. Fan, D. Liao and E. Gao, Inorg. Chem, 2001, 40, 3723.
11. G. S. Papaefstathiou, A. Manessi, C. P. Raptopoulou, A. Terzis and T. F. Zafiropoulos, Inorg. Chem., 2006, 45, 8823.
12. S. P. Watton, P. Fuhrmann, L. E. Pence, A. Caneschi, A. Cornia, G. L. Abbati and S. J. Lippard, Angew. Chem., Int. Ed. Engl., 1997, 36, 2774.
13. Th. C. Stamatatos, K. A. Abboud, W. Wernsdorfer and G. Christou, Angew. Chem., Int. Ed., 2008, 47, 6694.
14. A. J. Tasiopoulos, A. Vinslava, W. Wernsdorfer, K. A. Abboud and G. Christou, Angew. Chem., Int. Ed., 2004, 43, 2117.
15. (a) A. Müller and C. Serain, Acc. Chem. Res., 2000, 33, 2 , and references cited therein; (b) B. Salignac, S. Riedel, A. Dolbecq, F. Sécheresse and E. Cadot, J. Am. Chem. Soc., 2000, 122, 10381.
16. (a) A. Chiolero and D. Loss, Phys. Rev. Lett., 1998, 80, 169; (b) F. Meier and D. Loss, Phys. Rev. B., 2001, 64, 224411; (c) F. Meier and D. Loss, Phys. Rev. Lett., 2001, 86, 5373.
17. (a) R. W. Saalfrank, I. Bernt, E. Uller and F. Hampel, Angew. Chem., Int. Ed. Engl., 1997, 36, 2482; (b) G. A. Timco, A. S. Batsanov, F. K. Larsen, C. A. Muryn, J. Overgaard, S. J. Teat and R. E. P. Winpenny, Chem. Commun., 2005, 3649; (c) E. J. L. McInnes, S. Piligkos, G. A. Timco and R. E. P. Winpenny, Coord. Chem. Rev., 1995, 139, 313.

---

Footnotes

† CCDC 692717. For crystallographic data in CIF or other electronic format see DOI: 10.1039/b812721b

‡ Vacuum-dried solid analysed as 2·2MeCN. Calcd. (found): C, 30.18 (30.06); H, 4.53 (4.45); N, 0.68 (0.67%). Crystal data for 2·25MeCN: C₁₅₀H₂₅₅Ga₂₀N₂₅O₈₀, M_w = 5083.22, monoclinic, space groupP2/n with a = 22.768(5) Å, b = 33.543(7) Å, c = 29.161(7) Å, β = 97.201(4), V = 22095(9) ų, T = 173(2) K, Z = 4, 89990 reflections collected, 28911 unique (R_int = 0.2296), R1 [I > 2σ(I)] = 0.1007, wR2 = 0.2210 (F², all data). Many crystals from multiple preparations were screened and all were found to be weak diffractors of X-rays, no doubt due to the large amount of disordered solvent in the crystal. This limited the usable data collected, and thus only the Ga atoms were refined anisotropically. CCDC 692717.

---