No abstract

Magnets are widely used in a large number of applications, and their market is larger than that of semiconductors. Information storage is certainly one of the most important uses of magnets, and the lower limit to the size of the memory elements is provided by the superparamagnetic size, below which information cannot be permanently stored because the magnetization freely fluctuates. This occurs at room temperature for particles in the range of 10–100 nm, owing to the nature of the material. However, even smaller particles can in principle be used either by working at lower temperatures or by taking advantage of the onset of quantum size effects, which can make nanomagnets candidates for the construction of quantum computers.

Various present and future specialized applications of magnets require monodisperse, small magnetic particles, and the discovery of molecules that can function as nanoscale magnets was an important development in this regard. These molecules act as single-domain magnetic particles that, below their blocking temperature, exhibit magnetization hysteresis, a classical property of macroscopic magnets. Such 'single-molecule magnets' (SMMs) straddle the interface between classical and quantum mechanical behaviour because they also display quantum tunnelling of magnetization and quantum phase interference. Quantum tunnelling of magnetization can be advantageous for some potential applications of SMMs, for example, in providing the quantum superposition of states required for quantum computing. However, it is a disadvantage in other applications, such as information storage, where it would lead to information loss. Thus it is important to both understand and control the quantum properties of SMMs. Here we report a supramolecular SMM dimer in which antiferromagnetic coupling between the two components results in quantum behaviour different from that of the individual SMMs. Our experimental observations and theoretical analysis suggest a means of tuning the quantum tunnelling of magnetization in SMMs. This system may also prove useful for studying quantum tunnelling of relevance to mesoscopic antiferromagnets.

No abstract

The preparation, characterization, and X-ray structure are reported for the single-molecule magnet (PPh4)[Mn12O12(O2CPh)16(H2O)4]·8(CH2Cl2) (2). Complex 2 crystallizes in the triclinic space group P1̄, which at 213 K has a = 17.2329(2), b = 17.8347(2), c = 26.8052(2) Å, α = 90.515(2), β = 94.242(2), γ = 101.437(2)°, and Z = 2. The salt consists of PPh4 + cations and [Mn12O12(O2CPh)16(H2O)4]- anions. The (Mn12O12)15+ core of the anion is formed by an external ring of eight Mn atoms bridged by μ3−O2- ions to an internal tetrahedron of four Mn atoms. Because of disorder in both phenyl rings and solvate molecules, it was difficult to use bond valence sum values to determine definitively the oxidation state of each Mn atom. There is a Mn4O4 cubane unit in the internal part of the molecule and these Mn atoms are all MnIV ions. For the eight “external” Mn atoms the bond valence sum values did not define well their oxidation states. For these eight Mn atoms, it was not possible to determine whether a trapped-valence MnIIMnIII 7 or an electronically delocalized description is appropriate. High-frequency EPR (HFEPR) data were collected for the previously structurally characterized MnIV 4MnIII 7MnII valence-trapped salt (PPh4)[Mn12O12(O2CEt)16(H2O)4] (1) at 328.2 and 437.69 GHz. In the high magnetic field the crystallites orient and the HFEPR spectra are pseudo−single-crystal like, not powder patterns. The spectral features are attributed to the fine structure expected for a S = 19/2 complex experiencing axial zero-field splitting D Ŝ z 2, where D = −0.62 cm-1. The sign of D was definitively determined by the temperature dependence of the spectrum. Complex 2 exhibits one out-of-phase ac magnetic susceptibility (χ‘ ‘M) signal in the 3−6 K range. The temperature of the χ‘ ‘M peak is frequency dependent, as expected for a single-molecule magnet. The rate at which the direction of magnetization reverses from “up” to “down” was evaluated from χ‘ ‘M data collected at various frequencies (1−1512 Hz) of oscillation of the ac magnetic field. This gives magnetization relaxation rates in the 2.86−4.51 K range for complex 2 and in the 3.2−7.2 K range for complex 1. Rates were also determined in the 1.80−2.50 K range for complex 1 via magnetization decay experiments. In this latter case, the polycrystalline sample is magnetically saturated in a large dc field. After the magnetic field is rapidly decreased to zero, the decay of the magnetization to zero is monitored. The rates evaluated by both the frequency dependence of the out-of-phase ac signal and dc relaxation decay experiments for complex 1 fit on an Arrhenius plot to give an activation energy of U eff = 57 K and a preexponential rate of 1/τ0 = 7.2 × 107 s-1. From the HFEPR data, complex 1 has a S = 19/2 ground state with D = −0.62 cm-1. This gives a potential-energy barrier of U = 79 K for the double-well potential-energy diagram. The value of U eff is less than the barrier height U, because when individual [Mn12 -] anions convert from spin “up” to spin “down”, they can not only...