Hierarchical Structures DOI: 10.1002/ange.200500208 Hierarchical Assembly of {Fe 13 } Oxygen-Bridged Clusters into a Close-Packed Superstructure** Muralee Murugesu, Rodolphe ClØrac, Wolfgang Wernsdorfer, ChristopherE. Anson, and AnnieK. Powell* Dedicated to Professor Heinrich Vahrenkamp on the occasion of his 65th birthday Hierarchical materials can be regarded as systems in which small units are incorporated into larger superstructures, thereby investing the resulting construct with a hierarchy of properties that operate on different scales. Such systems become of particular interest when nanometer scales are involved, as these represent a situation in which nanoparticles are organized into a specific array. The organization of nanoparticles is important in terms of creating devices but poses significant difficulties if the particles have to be physically arranged, for example, by using electron-beam lithography. Therefore, the idea of the combination of such particles along chemical principles is attractive as it leaves the chemical-bonding interactions to do the work. Herein, we describe the realization of this idea in a material obtained from aqueous solution and composed of a close-packed superstructure of connected iron( iii ) nanomagnets, which themselves display close-packed core architectures. Our conceptual hierarchy begins at the atomic level with Fe III ions in water in the form of the hexaaqua ion [Fe- (H 2 O) 6 ] 3+ . Under normal conditions, a variety of hydrolysis reactions would lead to the precipitation of an amorphous hydroxide, which would then age into well-defined mineral phases, such as the oxyhydroxide and goethite phases, and finally transform into the most thermodynamically stable oxide haematite. [1] We have observed that supply of tripodal chelating ligands of the general form N(RCOOH) 2 R’ (in which R’ can be any organic residue) to such solutions can halt this process through the stabilization of captured intermediate phases composed of close-packed cores, which are portions of the brucite structure (exemplified by Mg(OH) 2 ) encased in a shell of the ligand units. [2,3] It is worth looking at this hydrolysis process in more detail: The brucite structure is an AB 2 lattice composed of close-packed hydroxide (B) layers that are arranged as double strips and M II ions (A) that sit in the octahedral holes between these strips so that each hydroxide unit bridges three metal ions (Figure 1). In the hydrolysis process, we can imagine the starting point (first generation) to be a hexaaqua metal ion which will link to six further metal ions (second generation) on production of the hexahydroxo metal ion, and the process will continue through production of the hydroxide ions and coordination to a further six metal centers (third generation), then twelve (fourth generation), and so on, as can be seen in Figure 1. Aggregates that contain portions of such core structures have been structurally characterized for a variety of metal ions. [2–15] Clearly, an infinite brucite structure is not possible for the examples of M III ions on the basis of charge imbalance, which could be stabilized by the shell of ligands coordinated to the outermost metal ions of the hydroxide lattice in the observed aggregates. A further pertinent point is that removal of half the protons from the double strips of hydroxide moieties in brucite leads to a layer of the a-oxyhydroxide structure, which is exemplified for the Fe III center by goethite, a-FeO(OH), itself the precursor to the most thermodynami- cally stable phase haematite. Thus, we can think of such aggregates as metastable intermediates that are trapped through the process of crystallization. These aggregates prove to be magnetically interesting for the Fe III center because, although the overall coupling is antiferromagnetic, the pairwise interactions are unequal in magnitude over the whole molecule, thus leading to residual ground-state spins of up to 33/2 and a display of hysteresis of the magnetiza- tion. [3,16–18] Thus, a secondary level of organization both in terms of the structure and properties of the compound is demonstrated. In the specific case we describe herein, the tripodal proligand nitrilotripropionic acid (H 3 ntp = N(CH 2 CH 2 COOH) 3 ) has been used to successfully trap {Fe 13 } aggregates (Figure 2), in which the production of the Figure 1. The AB 2 lattice as seen in the brucite Mg(OH) 2 structure. Green: metal atoms; red: oxygen atoms. [*] Dr. M. Murugesu, Dr. C. E. Anson, Prof. A. K. Powell Institut für Anorganische Chemie der Universität Karlsruhe Engesserstrasse Geb. 30.45 76128 Karlsruhe (Germany) Fax:(+ 49)721-608-8142 E-mail: powell@chemie.uni-karlsruhe.de Dr. R. ClØrac Centre de Recherche Paul Pascal CNRS-UPR 8641, 115 Avenue Dr. A. Schweitzer 33600 Pessac (France) Dr.W.Wernsdorfer Laboratoire Louis NeØl, CNRS 25 Avenue Des Martyrs BP166 38042 Grenoble Cedex 9 (France) [**] This work was supported by the DFG through the Schwerpunkt Programm SPP 1137 on “Molecular Magnetism”, the Center for FunctionalNanostructures(CFN),theConseilRØgionald’Aquitaine, Bordeaux I University, and the CNRS. Zuschriften 6836 # 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. 2005, 117, 6836–6840