High-nuclearity manganese carboxylate clusters have been attracting intense interest during the last several years from scientists in various disciplines. This is because of a combination of factors, not least their aesthetic qualities and their unusual magnetic properties. [1,2] The latter arise from their large, and often abnormally large, ground-state spin values, which, in combination with significant easy-axis-type magnetoanisotropy, leads to Mn clusters exhibiting the new magnetic phenomenon of single-molecule magnetism. [2] This property is the ability of discrete molecules to exhibit the superparamagnet-like property of slow magnetization relaxation and thus to behave as magnets below their blocking temperature by exhibiting magnetization versus field hysteresis. [3,4] The first single-molecule magnet (SMM) was [Mn 12 O 12 (O 2 CMe) 16 (H 2 O) 4 ] (1), which possesses an S 10 ground state, [3,5] and a number of other Mn x SMMs have since been discovered. [1, 2, 6±9] As part of a continuing effort to prepare new clusters with large S values that might be SMMs, we have been exploring new reactions of 1, or its derivatives, such as [Mn 12 O 12 (O 2 CCH 2 tBu) 16 (H 2 O) 4 ] (2), which can readily be prepared from 1 by a ligand-substitution procedure. [10] A solution of 2 in CH 2 Cl 2 was treated with an equal volume of MeOH, and the solution concentrated by slow evaporation over several days. After a brown solid was removed by filtration, black crystals of [Mn 21 O 24 (OMe) 8 (O 2 CCH 2 tBu) 16 -(H 2 O) 10 ] (3) formed over few days. The yield is very low ($ 3 %) but the reaction has been reproduced several times. The structure of 3 (Figure 1, top) [11] consists of an Mn core that is approximately planar and is ligated on the periphery by 16 m-O 2 CCH 2 tBu groups and 10 H 2 O molecules. The complex is trapped valence, the Mn III ions being the outer Mn6 ± Mn11 atoms and their symmetry-related partners; the complex has crystallographic C i symmetry. The Mn III ions were identified by their metric parameters and Jahn ± Teller distortions. As expected, the Jahn ± Teller elongated Mn III ÀO bonds (2.135(9) ± 2.323(8) ä) are significantly longer than the other Mn III ÀO bonds (1.858(7) ± 1.981(8) ä). The Mn IV ÀO bonds are in a narrower range (1.838(7) ± 1.956(8) ä). The 21 Mn ions are not all in the same plane: the nine Mn IV ions (Mn1 ± Mn5) and two Mn III ions (Mn8) are co-planar, but the two Mn 5 Figure 1. Top: ORTEP plot of the molecular structure of complex 3 (the thermal ellipsoids are set at 50 % probability). Bottom: side view showing Mn 4 (red), Mn 3 (green), and O (yellow) atoms to emphasize the planar central Mn 11 unit.

No abstract

The reactions of [Mn12O12(O2CEt)16(H2O)4] with phenylphosphinic acid (PhHPO2H) in MeCN and MeCN/CH2Cl2 have led to isolation of [Mn22O12(O2CEt)22(O3PPh)8(H2O)8] (2) and [Mn22O12(O2CEt)20(O3PPh)8(O2PPhH)2(H2O)8]n (3), respectively, both containing PhPO3(2-) groups from in situ oxidation of PhHPO(2)(-). Complex 2 is molecular and consists of two Mn9 subunits linked by four additional Mn atoms. Complex 3 contains almost identical Mn22 units as 2, but they are linked into a one-dimensional chain structure. The Mn22 unit in both compounds is mixed-valence Mn(III)18, Mn(II)4. Solid-state, variable-temperature dc magnetic susceptibility and magnetization measurements were performed on vacuum-dried samples of 2 and 3, indicating dominant antiferromagnetic interactions. A good fit of low-temperature magnetization data for 2 could not be obtained because of problems associated with low-lying excited states, as expected for a high nuclearity complex containing Mn(II) atoms. An approximate fit using only data collected in small applied fields indicated an S = 7 or 8 ground state for 2. Solid-state ac susceptibility data established that the true ground state of 2 is S = 7 and that the connected Mn22 units of 3 are ferromagnetically coupled. Both 2 and 3 displayed weak out-of-phase ac signals indicative of slow magnetization relaxation. Single-crystal magnetization versus applied dc field scans exhibited hysteresis loops for both compounds, establishing them as new single-molecule and single-chain magnets, respectively. Complex 2 also showed steps in its hysteresis loops characteristic of quantum tunneling of magnetization, the highest nuclearity molecule to show such QTM steps. Arrhenius plots constructed from dc magnetization versus time decay plots gave effective barriers to magnetization relaxation (U(eff)) of 6 and 11 cm(-1) for 2 and 3, respectively.