© 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Adv. Mater. 2010, 22, 4275–4279 4275 www.advmat.de www.MaterialsViews.com COMMUNICATION wileyonlinelibrary.com By Thomas McGlone, Carsten Streb, De-Liang Long, and Leroy Cronin* Assembly of Pure Silver-Tungsten-Oxide Frameworks from Nanostructured Solution Processable Clusters and Their Evolution into Materials with a Metallic Component [∗] T. McGlone, Dr. C. Streb, Dr. D.-L. Long, Prof. L. Cronin WestCHEM, Department of Chemistry The University of Glasgow G12 8QQ (UK) http://www.croninlab.com; E-mail: L.Cronin@chem.gla.ac.uk DOI: 10.1002/adma.201001398 Polyoxometalates (POMs) are anionic metal oxide clusters con- structed from early transition metals in their highest oxida- tion states, most commonly Mo VI , W VI and V V . [1] They can be described as an array of MO x polyhedral units (M = W, Mo, V, Nb or Ta; x = 4–7) linked via edge, corner and occasionally face sharing modes and are assembled from the hydrolytic aggre- gation of mononuclear oxometalates. [2,3] With an unrivalled range of tuneable properties, POMs exhibit a diverse range of applications in many areas of chemistry including catalysis, [4,5] medicine, [6] magnetism, [7–9] materials [10,11] and surface studies. [12] Despite the vast potential of anionic POM clusters as precursors to form novel materials and their application in materials science, the ability to exploit the molecular structure of the clusters in the assembly of framework materials, and to control their reactivity, has been greatly limited. This is because of the seemingly infinite numbers of non specific ionic assem- blies that can be assembled by the combination of the anionic POM clusters with the charge balancing cations. Any new break- through in this area requires the ability to exploit the bottom up assembly of nanostructured clusters in solution, with a well defined linking strategy that gives rise to the overall material. Taking all of these factors into account, it is clear that the traditional routes to POM formation involved an element of serendipity, due to a poor understanding of reaction mecha- nisms. Given these issues it is not surprising that solution processable POM precursors, although potentially transforma- tive, have not been routinely used in the assembly of novel ‘pure’ inorganic solid-state frameworks. In an effort to combat these complex issues, we recently developed a synthetic strategy to design novel clusters, by cation exchange reactions, which have the role of preventing the formation of thermodynami- cally favourable, highly symmetrical species in solution. [13,14] This approach was shown to be general when we applied it to tungstate-based cluster systems; by using protonated triethanolamine (TEAH) we were able to isolate the largest isopolyoxotungstate to date, [H 12 W 36 O 120 ] 12 - which has been described as an inorganic crown ether [15,16] and very recently has been incorporated into a family of hybrid inorganic- organic frameworks exhibiting a new type of supramolecular host-guest chemistry. [17,18] Of importance to this work, we have used this approach to assemble the cylinder-shaped cluster, [H 4 W VI 19 O 62 ] 6 - labelled {W 19 } here, which can be assembled and isolated as the organic-cation TEAH salt. [19] Indeed, the presence of the organic templating TEAH cations are crucial in directing the assembly of the {W 19 } archetype, see Figure 1, since in the absence of TEAH (e.g. with inorganic cations) only the well known [W 10 O 32 ] 4 - unit [20] can be obtained. The {W 19 } cluster is structurally highly unusual: the outer cluster shell resembles the traditional Dawson-type {W 18 } cage, however instead of the traditional set of two heteroatoms an additional W atom is located in the internal cluster cavity. Hence the cluster was, and still remains, the only example of an isopolyanion yielding a Dawson-type heteropolyanion framework. The system cocrystallises as a series of isomers with respect to the central W coordination environment and rotation of the trimeric caps, and these can only be separated by precipitation with n-tetrapropylammonium (TPA) and frac- tional recrystallisation from acetonitrile. The two most abun- dant isomers are (TPA) 6 [ α-H 4 W 19 O 62 ] 6CH 3 CN, { α-W 19 } and (TPA) 6 [ γ ∗ -H 4 W 19 O 62 ] 3CH 3 CN, { γ ∗ -W 19 }, which are illustrated in Figure 1. The main differentiating feature of these cluster isomers is the coordination mode of the central W unit; in the α-isomer, the W is coordinated by six oxo ligands in a trigonal prismatic fashion, whereas in the γ ∗ -isomer the W is located in an octahedral environment. Consequently, the cluster shell of { α-W 19 } features an eclipsed arrangement of the trimeric top and bottom caps, whereas in the { γ ∗ -W 19 } the caps are arranged in a staggered fashion, see Figure 1. Given the fact we can use cation-exchange-reactions to con- trol the assembly of the nanoscale cluster architectures, we Figure 1. Representation of the {W 19 } isomers with the central tung- sten coordination environment highlighted as green transparent poly- hedra. Left: the { α-W 19 } isomer contains a trigonal moiety and right: the { γ ∗ -W 19 } isomer contains an octahedral environment in the centre. W: blue spheres, O: red spheres.