Christopher R. K.
Glasson
a,
Jack K.
Clegg
bc,
John C.
McMurtrie
d,
George V.
Meehan
*a,
Leonard F.
Lindoy
*ab,
Cherie A.
Motti
e,
Boujemaa
Moubaraki
f,
Keith S.
Murray
f and
John D.
Cashion
g
aSchool of Pharmacy and Molecular Sciences, James Cook University, Townsville, Qld 4814, Australia. E-mail: George.Meehan@jcu.edu.au
bSchool of Chemistry, F11, The University of Sydney, NSW 2006, Australia. Fax: +61 293513329; E-mail: lindoy@chem.usyd.edu.au
cDepartment of Chemistry, The University of Cambridge, Lensfield Rd, Cambridge, UK CB2 1EW
dChemistry, Queensland University of Technology, GPO Box 2434, Brisbane, 4001, Australia
eThe Australian Institute of Marine Science, Townsville, 4810, Qld, Australia
fSchool of Chemistry, Monash University, PO Box 23, Clayton, Vic. 3800, Australia
gSchool of Physics, Monash University, PO Box 23, Clayton, Vic. 3800, Australia
First published on 25th November 2010
A unique example of incorporation of a tetrahalometalate anion in a small supramolecular cage is described in which a tetrahedral cage of type [Fe4L6]8+ selectively encapsulates a [FeIIICl4]− anion over a [FeIICl4]2− anion in its central cavity to yield a discrete, mixed oxidation state, Fe(II)/Fe(III) supramolecular assembly. This unusual outcome has been achieved using two alternative synthetic strategies.
Studies within our group have also seen the development of both neutral12,19 and positively charged12M4L6 hosts. For example, the [Fe4L6]8+ cation (1) was generated from the interaction of FeII with 5,5′′′-dimethyl-2,2′:5′,5′′:2′′,2′′′-quaterpyridine (L) (Fig. 1). This system was shown to encapsulate both BF4− and PF6− guest species, with uptake selectivity for PF6− over BF4− being evident.
Fig. 1 Schematic representation of the assembly of the tetrahedral [Fe4L6]8+host (1) incorporating a guest anion. |
We now report the encapsulation of a [FeIIICl4]− anion by 1, which to the best of our knowledge represents the first reported example of such inclusion of a tetrahalometallate anion in a small supramolecular cage. This outcome has been achieved using two (alternative) synthetic strategies.
In our previous report12 we described the formation of [Fe4L6 ⊃ BF4](BF4)7 from the reaction of Fe(BF4)2·6H2O with L in acetonitrile and showed that the BF4− anion spontaneously occupies the central cavity of the tetrahedral cage. A related synthesis was undertaken as part of the present study in which a solution of L was heated with FeCl2·5H2O in acetonitrile at reflux over 24 h to yield a fine suspension of a purple precipitate. On removal of the acetonitrile the resulting solid was dissolved in water and excess KPF6 was added to yield a deep red precipitate which was then purified by chromatography on Sephadex LH-20 using acetonitrile as eluent. Even though the 1H NMR spectrum of the purified material showed significant line broadening, it was evident from the spectrum that the product was symmetrical. Microanalysis of the red product was in accordance with the stoichiometry Fe4L6·(FeCl4)·(PF6)7·CH3OH.
Crystals of X-ray quality of the above complex were obtained by diffusion of methanol into an acetonitrile solution of this product. The crystal structure determination showed that the product was a unique solvated tetranuclear complex of type [FeII4L6 ⊃ FeIIICl4](PF6)7 (Fig. 2), in which a tetrahedral [FeIIICl4]− anion (see later) occupies the central cavity (see Section S1 in the ESI†). The above product crystallised in the cubic space groupP3n with each FeII centre existing on a 3-fold special position. The asymmetric unit contains 1/12th of the complex; the crystal corresponds to a racemic mixture in which the four chiral pseudo-octahedral FeII centres at the apices either have all Δ or all Λ configurations. The internuclear distance between apical FeII centres is 9.43 Å, equating to a cavity volume of ∼100 Å3.
Fig. 2 (a) Space filling representation of the crystal structure of [Fe4L6 ⊃ FeCl4]7+ viewed down the C3-axis of the ΔΔΔΔ enantiomer and (b) schematic representation of the encapsulated tetrahedral [FeCl4]− anion. |
The encapsulated [FeCl4]− anion has perfect tetrahedral symmetry with Cl–Fe–Cl bond angles of 109.5° and Fe–Cl bond lengths of 2.1994(17) Å that lie within the expected range for a [FeIIICl4]− anion.21 For comparison, the corresponding average for [FeIICl4]2− is ∼2.32 Å (although some overlap of the respective FeII and FeIII ranges is evident in the crystallographic literature).22,23 With respect to the above, it is noted that several examples of the formation of FeII metal complexes incorporating associated [FeIIICl4]− counterions derived from FeCl2 precursor solutions in the presence of air under a variety of conditions have been documented previously.24
Clearly the PF6− anion, present in solutions of the initial product during its purification, does not successfully compete with [FeCl4]− for encapsulation. Interestingly, as observed in our previous study, the tetrahedral host [Fe4L6]8+ does, however, exhibit clear encapsulation selectivity for PF6− over BF4−.12 This selectivity for [FeCl4]−, taken together with the X-ray structural results, strongly suggest that the size, shape (and symmetry) of the [FeCl4]− guest is quite complementary to the cavity of 1. Nevertheless, it is noted that the encapsulation of the similarly sized [ZnCl4]2− was not observed when [Fe4L6]([ZnCl4])4 was prepared previously in the presence of this anion.12 Both this result and our current observation are consistent with the observations of Raymond et al.,25 that the larger enthalpy of desolvation associated with doubly charged species tends to make their encapsulation significantly less favourable.
The high resolution mass spectrum of the present product revealed +2 to +7 ions corresponding to the successive losses of PF6− from a formula corresponding to [Fe4L6 ⊃ FeCl4](PF6)7 in agreement with our observation that the encapsulated anionic guest species is indeed [FeIIICl4]− (Fig. 3a; see Section S2 in the ESI†). However, as an aside it is noted that the mass spectrum also contained two additional ion cluster series: a series of ions corresponding to successive losses of PF6− from a parent species of formula [Fe4L6](PF6)8 and a further series consistent with the successive losses of PF6− from a parent species of formula [Fe4L6 ⊃ FeIICl4](PF6)6; the observed and calculated isotopic distributions of the +3 ions from each series are illustrated in Fig. 3 (b)–(d) The former of these latter two series may be rationalised by the loss of the [FeCl4]− guest in the electrospray while the latter presumably reflects the occurrence of a redox reaction26 under the ionisation conditions present in the positive ion ESI measurement; consistent with this is the observation that the peak ratio for the peaks arising from [Fe4L6 ⊃ FeIICl4](PF6)6 and [Fe4L6 ⊃ FeIIICl4](PF6)7 is varied on changing the applied potential of the capillary exit in the electrospray process.
Fig. 3 (a) The mass spectrum of [Fe4L6 ⊃ FeIIICl4](PF6)7 illustrating the +2 to +4 ion clusters; (b), (c) and (d) illustrate the observed (top) and calculated (bottom) isotopic distributions consistent with the respective ions:{[Fe4L6 ⊃ FeIICl4](PF6)3}3+, {[Fe4L6](PF6)5}3+ and {[Fe4L6 ⊃ FeIIICl4](PF6)4}3+. |
A bulk electrolysis (oxidation) experiment involving [FeII4L6 ⊃ FeIIICl4](PF6)7 in acetonitrile was also undertaken. In an initial study, the cyclic voltammogram showed a single broad wave (E1/2 = 1.08 V; ΔEp = 111 mV; 4 e−) under the conditions employed (see Section S3 in the ESI†). Based on this result, an oxidation potential of 1.4 V was chosen for the bulk electrolysis experiment. During the progress of the electrolysis the characteristic deep red colour of the tris-bipyridyl FeIIchromophore was discharged, with the final colour of the electrolysis solution being light green. The experimentally measured charge (Q) for the oxidation of [FeII4L6 ⊃ FeIIICl4](PF6)7 was 1.692 C (101%) versus the theoretical value for a 4e− (4FeII → 4FeIII) process of 1.673 C. Thus, the results from the bulk electrolysis investigation are in full agreement with the mixed FeII/FeIII formulation for the [FeII4L6 ⊃ FeIIICl4]7+ cation discussed above.
An alternative synthetic strategy involving a microwave driven procedure analogous to that described previously by us12 for obtaining [Fe4L6]8+ free of its included polyatomic anion was repeated (but with methanol substituted for water as the reaction solvent) and the product from this reaction then precipitated by the addition of an ether solution of H[FeCl4]27,28 to the reaction solution. Microanalysis of the resulting red product was consistent with the formula [Fe4L6 ⊃ FeCl4](FeCl4)7·2CH3OH·CH3CN. X-ray quality crystals of this material were obtained by diffusion of Et2O into an acetonitrile solution of the above product. The product crystallises in the cubic space groupF23 and the structure once again revealed that the product was the M4L6 complex of type [Fe4L6 ⊃ FeCl4](FeCl4)7 with a tetrahedral [FeCl4]− anion occupying the central cavity. Importantly, the encapsulated [FeCl4]− anion again had perfect tetrahedral symmetry with Cl–Fe–Cl bond angles of 109.5° and Fe–Cl bond lengths of 2.2059(8) Å, which are almost identical to those in the ‘external' [FeCl4]− counterions whose mean Fe–Cl bond length is 2.20 Å. As anticipated, this X-ray result is in complete accordance with the formulation of the corresponding PF6− salt as [Fe4L6 ⊃ FeIIICl4](PF6)7 (for which the mean of the Fe–Cl bond lengths, at 2.1997(17) Å, is identical within experimental error to both the above values).
[Fe4L6 ⊃ FeIIICl4](FeCl4)7·2CH3OH·CH3CN is water soluble and the addition of KPF6 to an aqueous solution of this product led to the isolation of the corresponding PF6− salt whose microanalysis, mass spectrum and unit cell analysis were in agreement with the results obtained for the earlier-prepared PF6− salt discussed above.
Finally, in view of the unusual mixed FeII/FeIII oxidation states present in the above unique inclusion products, we embarked on variable temperature investigations (4.2 to 300 K) of the magnetic susceptibility behaviour of both [Fe4L6 ⊃ FeIIICl4](PF6)·CH3OH and [Fe4L6 ⊃ FeIIICl4](FeCl4)·2CH3OH·CH3CN, together with Mössbauer spectral measurements obtained at 78 K. In both cases the results accord well with the oxidation states assigned for the iron centres in each of the above compounds (full details of these studies are presented in Section S5 of the ESI†).
In this report we describe the synthesis and characterisation of a new cationic inclusion complex of type [Fe4L6 ⊃ FeCl4]7+. This system is especially noteworthy on at least three counts. First, the product is a rare example of a supramolecular mixed-valent Fe(II)/Fe(III) inclusion assembly. Secondly, it also appears unprecedented in that it selectively extracts a [FeIIICl4]− anion from a mix of FeII and/or FeIII chloro species undoubtedly present in the respective reaction solutions under the conditions employed. Finally, to the best of our knowledge it represents a unique example of the incorporation of a tetrahalometallate anion in a small supramolecular cage.
We (GVM, LFL, KSM) thank the Australian Research Council for support. JKC acknowledges a Marie Curie International Incoming Fellowship within the 7th European Community Framework Programme.
Footnote |
† Electronic supplementary information (ESI) available: The ESI contains experimental details of the synthesis of the inclusion products together with their X-ray structure determinations. Further details of the mass spectral measurements, electrochemistry, magnetic susceptibility and Mössbauer studies are also presented. CCDC 783500 ([Fe4L6 ⊃ FeCl4](FeCl4)7·3MeOH·3MeCN·9H2O)) and 783501 ([Fe4L6 ⊃ FeCl4](PF6)7·12H2O) contain the supplementary crystallographic data for this paper. These can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.]. CCDC reference numbers 783500, 783501. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c0sc00523a |
This journal is © The Royal Society of Chemistry 2011 |