Low-valent vanadium catecholate clusters

Abstract

A simple synthetic route to low-valent vanadium catecholate clusters, via solvothermal reaction of [V(acac)₃] with catechols in alcohols, is presented. The syntheses and X-ray structures of [V^III₆O(dbcat)₄(OMe)₈(MeOH)₂] (1; which has the Lindqvist structure), [V^III₆O₂(dbcat)₂(dbsq)₂(acac)₂(PhCO₂)₄(EtO)₂] (2; which could alternatively be formulated as [V^III₄V^IV₂O₂(dbcat)₄(acac)₂(PhCO₂)₄(EtO)₂]) and [V^IV₂(dbcat)₄(EtOH)₂] (3) are reported (dbcat²⁻ = 3,5-^tBu₂-catecholate; dbsq⁻ = 3,5-^tBu₂-semiquinone). The X-ray structures of the primary products of the reactions of 1 and 3 with atmospheric oxygen, [V₆O₃(“dbcat”)₄(OMe)₈] (4) and [V₂O₂(“dbcat”)₄] (5), are also reported—the assignments of redox states for these four-electron oxidised species are more ambiguous.

---

Vanadium catechol chemistry is very rich given the redox active nature of both metal ion and ligand.^1–3 It was recognized early on that catecholates could stabilize low metal-ion oxidation states (exploited by nature in V^III sequestering tunichromes)² and early studies on simple tris-chelates¹ addressed their complicated redox chemistry and ambiguity in electronic structure. The low-valent species are reducing, and reactions with oxygen give Pierpont's mixed-valence ligand dimer³ [(V^VO)(dbsq)(dbcat)]₂ (dbcat²⁻ = 3,5-^tBu₂-catecholate; dbsq⁻ = 3,5-^tBu₂-semiquinone), which has received renewed attention since its identification as the catalyst resting state in very efficient vanadium-based catechol dioxygenation catalysis.⁴ Shilov et al.'s discovery that V^II catecholates efficiently reduce dinitrogen in protic media,⁵ due to the probably tetrametallic nature of the as yet unidentified catalyst, gives further impetus to low valent vanadium catecholate research. Despite this very few oligomeric examples are known.⁶ Given that solvothermal methods, well established in polyoxovanadate synthesis,⁷ have proved useful in preparing large V^III clusters⁸ particularly in reducing solvents such as alcohols,^8,9 we have now begun to explore such reactions to prepare low valent vanadium catecholates. Here we report that this is a viable route, with the product very sensitive to small changes in reaction conditions: we report the isolation of two types of hexavanadium tetracatecholates, one of which has the Lindqvist-type structure, and dimetallic tetracatecholates where the latter is a reduced form of Pierpont's dimer. We also report the isolation of the products of reaction of two of these species with atmospheric oxygen.

A simple solvothermal reaction of [V^III(acac)₃] (acac⁻ = acetylacetonate) with dbcatH₂ (3 : 2) in MeOH at 150 °C under anaerobic conditions gives dark brown crystals of [V₆O(dbcat)₄(OMe)₈(MeOH)₂] (1) on cooling (60%).‡ The EtOH version of this material can be made equivalently, but the crystallography is much poorer. 1 crystallises in P1̄, and the molecules lie on an inversion centre (Fig. 1). The core consists of six vanadium ions defining an octahedron, centred on a μ₆-oxide, O1. Eight μ₂-methoxides bridge V1 and symmetry equivalent (s.e.) to the remaining four metal ions, spanning eight of the edges of the {V₆} octahedron. The hexacoordinations of V1,1A are completed by terminal MeOH. Four doubly deprotonated dbcat ligands lie in the equatorial plane defined by V2,3,2A,3A, each chelating one metal ion and bridging to a second via the 1-oxygen, the 2-position being trans to O1. Redox assignment for 1 is straightforward: bond valence sum (BVS) analysis gives all metal ions as V^III (mean 2.94; Table S1†). Charge balance then requires all the dbcat ligands to be in their catecholate (dbcat²⁻) form; this is consistent with their C–O and C–C(ind) [between the C(O) atoms] bond lengths.¹⁰ Hence, 1 is formulated as [V^III₆O(dbcat)₄(OMe)₈(MeOH)₂], and this is confirmed by its magnetic properties which can be fit to a spin Hamiltonian involving only V^III (s = 1) ions (Fig. S1†). The {V^III₆O₁₉} core of 1 is a stabilised form of an all V^III Lindqvist ion (Fig. S2†). Lindqvist structures incorporating vanadium(III) are extremely rare.¹¹

Fig. 1 Crystal structure of 1 viewed perpendicular to (top; ^tBu groups removed for clarity) and down (bottom) the V1⋯V1A vector. Scheme: V (green), O (red), C (black), H omitted for clarity. Selected bond lengths (Å): V–O1 2.1614(7)–2.2096(8), V1–O6(ROH) 2.053(2), V–O(Me) 1.9635(19)–1.994(2), V–O(cat) 1.9055(18)–2.0420(19), C–O(cat) 1.356(3)–1.374(3), C–C(ind) 1.411(4) and 1.417(4).

If an excess of PhCO₂H is added as a co-ligand to the reaction that gives 1, a different hexametallic tetracatecholate is isolated: solvothermal reaction of [V(acac)₃], dbcatH₂ and PhCO₂H (1 : 1 : 2) in EtOH at 150 °C for 12 h gives dark blue/black needles of [V₆O₂(dbcat)₄(acac)₂(PhCO₂)₄(EtO)₂] (2; Fig. 2) directly on slow cooling (62%).‡2 is centrosymmetric and contains three unique vanadium ions. The six vanadium ions form an edge-sharing bitetrahedral core. V1,1A,2A,3 are connected via a μ₄-oxide (O1). V3 is bridged to V1 and V1A by two benzoates; the coordination at V3 is completed by a chelating diketonate and a μ-alkoxide (O7A) bridging to V2A. V2 (and s.e.) is chelated by two catecholates which also bridge to V1 and V1A (via O3 and O5A, respectively); the coordination geometry at V2,2A is trigonal prismatic. BVS analysis unambiguously assigns V1 and V3 as V^III (sum 3.01 and 3.05, respectively; Table S1†) but V2 is ambiguous giving intermediate values between 3+ and 4+. This leaves two charge-balanced possibilities: [V^III₆O₂(dbcat)₂(dbsq)₂(acac)₂(PhCO₂)₄(EtO)₂] or [V^III₄V^IV₂O₂(dbcat)₄(acac)₂(PhCO₂)₄(EtO)₂]. The “dbcat” C–C(ind) distances are significantly different for the two unique ligands [C1–C2 1.343(14) and C15–C16 1.428(15) Å] which could indicate a difference in redox state [C–C(ind) is longer for semiquinones than catecholates, typically ca. 1.44 cf. 1.40 Å, respectively].¹⁰ From magnetic measurements, χT(T) plateaus above 100 K up to room temperature at a value of 4.1 emu K mol⁻¹ (Fig. S3†). This could be consistent with {V^III₄V^IV₂} (calculated 4.3 emu K mol⁻¹ for g = 1.9) or {V^III₆(dbsq)₂} if the semiquinone spins are strongly antiferromagnetically coupled to the coordinated metal ions (V2,2A) to give two effective s = ½ centres. This is the case in [V^III(sq)₃] species.¹

Fig. 2 Top: crystal structure of 2. Bottom: core with ^tBu, Ph and Me groups removed for clarity. Selected bond lengths (Å): V–O1 1.993(6)–2.047(7), V–O(R) 1.930(7) and 2.010(7), V–O(acac) 1.937(7) and 1.972(7), V–O(benzoate) 1.997(7)–2.033(7). V–O(cat) distances: V2–O3,O8 2.036(6) and 1.863(7); V2–O5A,O6 2.034(7) and 1.896(6). Internal dbcat distances: C15,16–O3,O8 1.368(12) and 1.354(12); C1,2–O5,O6 1.393(12) and 1.402(12); C1–C2 1.343(14); C15–C16 1.428(15).

If a lower quantity of carboxylic acid is used in this reaction, a much smaller species results: reaction of [V(acac)₃], dbcatH₂ and PhCO₂H (2 : 2 : 1) in EtOH at 150 °C followed by standing for several days gives blue/black crystals of [V₂(dbcat)₄(EtOH)₂]·EtOH (3; 40%; Fig. 3).‡ Although benzoate is not incorporated in the product, the reaction fails in the absence of PhCO₂H. 3 crystallises in P1̄: the two vanadium ions are each chelated by two catecholates, one of which also bridges (via O1,5) to the other metal ion. The coordination at each vanadium is completed by a terminal alcohol. BVS analysis gives both vanadium ions as V^IV (valence sums 3.82 and 3.75, Table S1†) and metric parameters of the ligands are typical of catecholates, giving a charge balanced formula of [V^IV₂(dbcat)₄(EtOH)₂]. This product is intriguing as it appears to be a four-electron reduced form of Pierpont's [V^V₂O₂(dbcat)₂(dbsq)₂] dimer where alcohols have replaced the terminal oxides.

Fig. 3 Crystal structure of 3. Selected bond lengths (Å): V–O(H)R 2.0815(13) and 2.0442(14) Å; V1,2–O1,5 1.9823(12)–2.0809(12); other V–O(cat) 1.8293(13)–1.9337(12); C–O range 1.353(2)–1.369(2) Å. C–C(ind) range 1.403(2)–1.408(2).

Complexes 1–3 are all highly air sensitive. This is not surprising given the low valency; Pierpont's dimer itself is formed on aerial oxidation of [V^III(dbsq)₃] and the dimer reacts further with O₂ to ultimately give V₂O₅ with liberation of 3,5-^tBu₂-benzoquinone³ [dioxygenation catalysis is observed in the presence of excess dbcatH₂].⁴ For 1 and 3 we have managed to isolate the products of reaction with atmospheric O₂.

1 rapidly oxidizes to a dark blue material on exposure to air, both in the solid state and in solution. This blue material is difficult to crystallize, but on occasion we have isolated poor quality crystals from THF solution, revealing these to be [V₆O₃(“dbcat”)₄(OMe)₈] (4; Fig. 4).‡ Mass spectra confirm 4 as the species in solution. Better quality crystals of the same species can be isolated from reaction of [V(acac)₃], dbcatH₂ and Mg(OH)₂ in MeOH at 150 °C, crystallising in P1̄ as [V₆O₃(“dbcat”)₄(OMe)₈]·1.4MeOH·H₂O with two half-molecules per asymmetric unit. There is one major difference from the structure of 1: the apical vanadium ions now have terminal oxides [V1–O6 1.598(4) Å, υ_V=O 986 cm⁻¹], replacing the coordinated MeOH in 1. The redox assignment for 4 is not clear-cut. The apical vanadium ions (V1,1A) have clearly oxidised and BVS assigns them as V^IV (valence sum +4.10, Table S1†) with V2,2A (+3.22) and V3,3A (+3.10) as V^III. This would require two “dbcat” ligands to be in the semiquinone state to charge balance; [V^III₄V^IV₂O₃(dbcat)₂(dbsq)₂(OMe)₈]. However, there are no significant differences in the C–O(cat) [average 1.359 cf. 1.360 Å] or C–C(ind) [1.388(8) cf. 1.397(8) Å] lengths for the two crystallographically unique ligands in a molecule to support a localised mixed valence description for the ligands. This does not preclude a delocalised description, and this could be favoured by the planar {V₄(“dbcat”)₄} structure. Such delocalisation could also include the metal ions [{V^III(dbsq)}²⁺ ↔ {V^IV(dbcat)}²⁺] which could be the cause of the slightly high BVS sum for V2,2A.

Fig. 4 Structure of one of the unique molecules in the crystal structure of 4. Selected bond lengths (Å): V–O1 2.0848(12)–2.4650(12), V1–O6 1.598(4), V–O(Me) 1.974(4)–1.997(4), V–O(cat) 1.858(4)–2.012(4), C–O(cat) 1.346(7)–1.372(6), C–C(ind) 1.388(4) and 1.397(4).

The magnetic properties of 4 do not clarify this issue, but do confirm that V1,1A are V^IV rather than V^V (Fig. S4†). Monitoring the aerial oxidation of 1 to 4 by UV-Vis (CH₂Cl₂; Fig. S5†)¹² shows the collapse of a single peak at 468 nm (ε = 9 × 10³ M⁻¹ cm⁻¹) with growth of broad bands at 578 and 734 nm (16 × 10³ M⁻¹ cm⁻¹). Broad bands between 700–800 nm are a common feature of many dbsq⁻ complexes,¹³ but are not observed in e.g. [V^IV(dbcat)₃]²⁻.¹ This supports the oxidation of some of the ligands to dbsq⁻.

The primary oxidation product of the dimetallic 3 can also be isolated: when the filtrate from the synthesis of 3 is left to evaporate slowly in air dark crystals of [V₂O₂(“dbcat”)₄]·3EtOH (5) form (14%, based on initial reagent amounts in the synthesis of 3).‡ The two coordinated alcohols of 3 have been replaced by terminal oxides (O5) (Fig. 5). Although we have been unable to obtain satisfactory elemental analyses for 5, its formula is unambiguous from the X-ray structure. 5 is an isomer of Pierpont's dimer, the only structural difference being the location of the oxo-group relative to the bridging ligand, being trans to O3 in 5 and cis in Pierpont's complex. Again, assignment of redox states in the oxidation product is not straightforward. BVS analysis gives the metal ions as V^V (valence sum 5.02, Table S1†). Charge balance then requires [V^V₂O₂(dbcat)₂(dbsq)₂], which is equivalent to Pierpont's dimer.³ However, there is no significant difference between the metric parameters of the chelating and bridging “dbcat” ligands in the structure of 5, all being in the range expected for catecholate. To date we have been unable to obtain reliable magnetic data for 5 due to magnetic impurities (presumably vanadium oxides).

Fig. 5 Crystal structure of 5. Selected bond lengths (Å): V1–O5 1.591(3). Bridging ligand: C1–O1 1.375(4), C2–O2 1.353(5), C1–C2 1.389(5). Terminal ligand: C15–O3 1.372(4), C16–O4 1.357(4), C15–C16 1.399(5).

Conclusions

This work shows that high-nuclearity, low-valent vanadium catecholates are readily accessible via solvothermal routes. Given the redox active nature of metal and ligand this chemistry is rich and very sensitive to, for example, subtle changes in stoichiometry. The highly reducing nature of the complexes may lead to useful reactivity, and we are currently attempting to extend the chemistry to V^II with a view to exploring Shilov's observations on nitrogen fixation. 1 and 3 (the complexes for which we have isolated the primary oxidation products) both undergo four-electron oxidation and addition of two oxygen atoms in air, to 4 and 5, respectively. While it seems likely that this process involves oxidation of both metal ion and catecholate, unambiguous assignment of the charge distribution in the oxidised species requires further detailed physical studies—these will be reported later.

Acknowledgements

We thank the EPSRC and the EC (“MagMaNet” NoE, and a Marie Curie Host Fellowship for Young Researchers award to E. S.) for funding and Daresbury Laboratories for access to synchrotron X-ray facilities.

Notes and references

1. (a) R. M. Buchanan, H. H. Downs, W. B. Shorthill, C. G. Pierpont, S. L. Kessel and D. N. Hendrickson, J. Am. Chem. Soc., 1978, 100, 4318; (b) S. R. Cooper, Y.-B. Koh and K. N. Raymond, J. Am. Chem. Soc., 1982, 104, 5092; (c) P. J. Bosserman and D. T. Sawyer, Inorg. Chem., 1982, 21, 1545; (d) M. E. Cass, N. R. Gordon and C. G. Pierpont, Inorg. Chem., 1986, 25, 3962.
2. See for example: (a) A. R. Bulls, C. G. Pippin, F. E. Hahn and K. N. Raymond, J. Am. Chem. Soc., 1990, 112, 2627; (b) C. L. Simpson and C. G. Pierpont, Inorg. Chem., 1992, 31, 4308.
3. M. E. Cass, D. L. Greene, R. M. Buchanon and C. G. Pierpont, J. Am. Chem. Soc., 1983, 105, 2680.
4. C.-X. Yin and R. G. Finke, J. Am. Chem. Soc., 2005, 127, 13988; C.-X. Yin, Y. Sasaki and R. G. Finke, Inorg. Chem., 2005, 44, 8521.
5. (a) S. A. Isalva, L. A. Nikonova and A. E. Shilov, Nouv. J. Chim., 1981, 5, 21; (b) A. F. Shestakov and A. E. Shilov, Kinet. Catal., 2001, 42, 653.
6. (a) S. Lee, K. Nakanishi, M. Y. Chiang, R. B. Frankel and K. Spartalian, J. Chem. Soc., Chem. Commun., 1988, 785; (b) M. Mazzanti, C. Floriani, A. Chiesi-Villa and C. Guastini, J. Chem. Soc., Dalton Trans., 1989, 1793; (c) N. P. Luneva, S. A. Mironova, A. E. Shilov, M. Y. Antipin and Y. T. Struchkov, Angew. Chem., Int. Ed. Engl., 1993, 32, 1178.
7. See for example: M. I. Khan and J. Zubieta, Prog. Inorg. Chem., 1995, 43, 1.
8. R. H. Laye, M. Murrie, S. Ochsenbein, A. R. Bell, S. J. Teat, J. Raftery, H. U. Güdel and E. J. L. McInnes, Chem.–Eur. J., 2003, 9, 6215; R. H. Laye, Q. Wei, P. V. Mason, M. Shanmugan, S. J. Teat, E. K. Brechin, D. Collison and E. J. L. McInnes, J. Am. Chem. Soc., 2006, 128, 9020; I. S. Tidmarsh, R. H. Laye, P. R. Brearley, M. Shanmugam, E. C. Sañudo, L. Sorace, A. Caneschi and E. J. L. McInnes, Chem.–Eur. J., 2007, 13, 6329; R. Shaw, R. H. Laye, L. F. Jones, D. M. Low, C. Talbot-Eeckelaers, Q. Wei, C. J. Milios, S. Teat, M. Helliwell, J. Raftery, M. Evangelisti, M. Affronte, D. Collison, E. K. Brechin and E. J. L. McInnes, Inorg. Chem., 2007, 46, 4968; S. Khanra, M. Kloth, H. Mansaray, C. A. Muryn, F. Tuna, E. C. Sañudo, M. Helliwell, E. J. L. McInnes and R. E. P. Winpenny, Angew. Chem., Int. Ed., 2007, 46, 5568.
9. R. H. Laye and E. J. L. McInnes, Eur. J. Inorg. Chem., 2004, 2811.
10. C. G. Pierpont and R. M. Buchanan, Coord. Chem. Rev., 1981, 38, 45; C. G. Pierpont and C. W. Lange, Prog. Inorg. Chem., 1994, 41, 331; C. G. Pierpont, Coord. Chem. Rev., 2001, 219–221, 415; P. Chaudhuri, C. N. Verani, E. Bill, E. Bothe, T. Weyermuller and K. Wieghardt, J. Am. Chem. Soc., 2001, 123, 2213.
11. C. Aronica, G. Chastanet, E. Zueva, S. A. Borshch, J. M. Clemente-Juan and D. Luneau, J. Am. Chem. Soc., 2008, 130, 2365; L. J. Batchelor, R. Shaw, S. J. Markey, M. Helliwell and E. J. L. McInnes, Chem.–Eur. J., 2010, 16, 5554.
12. We have also monitored the cyclic voltammetry response of 1 as a function of time, see Fig S6†.
13. See for example: M. W. Lynch, D. N. Hendrickson, B. J. Fitzgerald and C. G. Pierpont, J. Am. Chem. Soc., 1984, 106, 2041.

---

Footnotes

† Electronic supplementary information (ESI) available: BVS results for 1–5; polyhedral view of 1; χ(T), χT(T) for 1, 2 {also M(H)} and 4; UV/Vis for 1 → 4; cyclic voltammogram for 1; cif files for 1–5. CCDC reference numbers 762899–762903. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c0sc00142b

‡ Synthesis: all manipulations were conducted under anaerobic conditions. The reactions were conducted under solvothermal conditions in a sealed Teflon-lined autoclave, heated to 150 °C for 12 h, then cooled to room temperature at 0.05 °C min⁻¹. 

1: [V(acac)₃] (0.25 g) and dbcatH₂ (0.166 g) in MeOH (7.5 ml) were treated as above. Brown plates of [V₆O(dbcat)₄(MeO)₈(MeOH)₂)]·4MeOH formed directly on cooling (61%). Elemental analysis calcd. (%) for C₇₀H₁₂₈O₂₃V₆: C 51.16, H 7.85, V 18.58; found: C 51.68, H 7.42, V 18.27. Selected IR data (KBr; ν/cm⁻¹: 1465, 1243, 988. UV-Vis (CH₂Cl₂): 468 nm (ε = 8981 M⁻¹ cm⁻¹). 

4: brown crystals of 1 oxidise on exposure to air to give a blue/black powder of [V₆O₃(“dbcat”)₄(MeO)₈] which can occasionally be recrystallised from THF. MALDI-MS (dithranol matrix in CH₂Cl₂): m/z 1483 (M⁺). Elemental analysis calcd. (%) for C₆₄H₁₀₄O₁₉V₆: C 51.80, H 7.07, V 20.61; found: C 49.75, H 7.14, V 20.30. Selected IR data (KBr; ν/cm⁻¹: 1465, 1240, 986. UV-Vis [CH₂Cl₂; ν/nm (ε/M⁻¹ cm⁻¹)] 578 (15946), 734 (15554). Crystals of 4·1.4MeOH·H₂O can be obtained from solvothermal reaction of [V(acac)₃] (0.25 g), dbcatH₂ (0.159 g) and Mg(OH)₂ (0.042 g, 0.718 mmol) in EtOH (low yield). 

2: [V(acac)₃] (0.25 g), dbcatH₂ (0.159 g) and PhCO₂H (0.175 g) in EtOH (7.5 ml) were treated as above. Black needles of [V₆O₂(dbcat)₄(acac)₂(PhCO₂)₄(EtO)₂] form directly on cooling, with more isolated on concentration of the filtrate (62%). Elemental analysis calcd. (%) for V₆C₉₈H₁₃₄O₂₄: C 58.80, H 6.75, V 15.27; found: C 58.68, H 6.48, V 14.89. 

3: [V(acac)₃] (0.25 g), dbcatH₂ (0.318 g), PhCO₂H (0.044 g) and EtOH (7.5 ml) were treated as above. Blue/black crystals of [V₂(dbcat)₄(EtOH)₂]·EtOH formed from the resultant dark blue/black solution after several days (40%). Elemental analysis calcd. (%) for C₆₂H₉₈O₁₁V₂: C 66.41, H 8.81, V 9.08; found: C 66.20, H 8.44, V 9.47. 

5: the filtrate from the above reaction was left to evaporate in air, forming blue/black crystals of [V₂O₂(“dbcat”)₄]·3EtOH after several hours (14%). Elemental analysis calcd. (%) for C₆₂H₉₅O₁₃V₂: C 64.57, H 8.56, V 8.83; found: C 61.23, H 7.26, V 13.23. We were unable to obtain a good analysis for 5. 

Crystal data: 

1: C₇₀H₁₂₈O₂₃V₆, M = 1643.36, triclinic, a = 11.232(2), b = 13.693(3), c = 14.106(3) Å, α = 96.27(3), β = 91.40(3), γ = 90.45(3)°, U = 2155.9(7) ų, T = 150 K, space group P1̄, Z = 1, μ = 0.688 mm⁻¹, 27728 reflections measured, 7070 unique (R_int = 0.0717). Final R₁ = 0.0550, GOF = 1.033, wR₂ [I > 2σ(I)] = 0.1518. 

4: C₆₇H₁₂₀O₂₄V₆, M = 1545.99, triclinic, a = 13.520(3), b = 18.220(4), c = 19.290(4) Å, α = 89.59(3), β = 74.53(3), γ = 82.69(3)°, U = 4540.6(7) ų, T = 100 K, space group P1̄, Z = 2, μ = 0.648 mm⁻¹, 33491 reflections measured, 15418 unique (R_int = 0.0572). Final R₁ = 0.0724, GOF = 1.032, wR₂ [I > 2σ(I)] = 0.2100. 

2: C₉₈H₁₂₄O₂₄V₆, M = 1991.61, monoclinic, a = 11.486(4), b = 28.417(11), c = 14.966(5) Å, β = 100.286(9)°, U = 4806(3) ų, T = 100 K, space group P2₁/n, Z = 2, μ = 0.631 mm⁻¹, 8118 reflections measured, 4254 unique (R_int = 0.1527). Final R₁ = 0.0778, GOF = 0.836, wR₂ [I > 2σ(I)] = 0.1614. 

3: C₆₂H₉₈O₁₁V₂, M = 1121.28, triclinic, a = 11.2421(8), b = 14.0373(10), c = 21.6847(16) Å, α = 105.9570(16), β = 93.7700(10), γ = 95.8850(10)°, U = 3257.0(4) ų, T = 150 K, space group P1̄, Z = 2, μ = 0.34 mm⁻¹, 57516 reflections measured, 15525 unique (R_int = 0.0628). Final R₁ = 0.0554, GOF = 1.069, wR₂ [I > 2σ(I)] = 0.1512. 

5: C₆₂H₉₈O₁₃V₂, M = 1153.28, monoclinic, a = 13.7603(12), b = 11.8807(10), c = 20.3294(18) Å, β = 95.249(2)°, U = 3309.6(5) ų, T = 100 K, space group P2₁/c, Z = 2, μ = 0.338 mm⁻¹, 12901 reflections measured, 4046 unique (R_int = 0.0572). Final R₁ = 0.0613, GOF = 0.985, wR₂ [I > 2σ(I)] = 0.1712. 

---