We present a hybrid ab initio molecular dynamics scheme that includes both DFT and Hartree−Fock-based extended Lagrangian and converged post-Hartree−Fock Born−Oppenheimer components, combined within the framework of a molecular fragmentation-based electronic structure. The specific fragmentation algorithm used here is derived from ONIOM but includes multiple, overlapping "model" systems. The interaction between the various overlapping model systems is approximated by invoking the principle of inclusion−exclusion at the chosen higher level of theory and within a "real" calculation performed at the chosen lower level of theory. Furthermore, here, the lower level electronic structure of the full system is propagated through an extended Lagrangian formalism, whereas the fragments, treated using post-Hartree−Fock electronic structure theories, are computed using the normal converged Born−Oppenheimer treatment. This conservative dynamical approach largely reduces the computational cost to approximate on-the-fly dynamics using post-Hartree−Fock electronic structure techniques and can be very beneficial for large systems where SCF convergence may be challenging and time consuming. Benchmarks are provided for medium-sized protonated water clusters, H 9 O 4 + and H 13 O 6 + , and polypeptide fragments, including a proline tripeptide fragment, and alanine decamer. Structural features are in excellent agreement between the hybrid approach using an MP2:B3LYP fragment-based electronic structure and BOMD using MP2 for the full system. Vibrational properties derived from dynamical correlation functions do show a small redshift for the extended Lagrangian treatments, especially at higher frequencies. Strategies are discussed to improve this redshift. The computational methodology works in parallel using both MPI and OpenMP and shows good scaling with the processor number. The timing benchmarks are provided for the alanine decamer. A powerful feature of the computational implementation is the fact that it is completely decoupled from the electronic structure package being employed and thus allows for an integrated approach that may include several different packages. These computational aspects will be further probed in future publications.

A pseudotetrahedral cobalt(II) complex with a positive axial zero-field splitting parameter of D = 12.7 cm(-1), as determined by high-field EPR spectroscopy, is shown to exhibit slow magnetic relaxation under an applied dc field.

This paper reports the experimental and theoretical investigations of two trigonal bipyramidal Ni(II) complexes, [Ni(Me(6)tren)Cl](ClO(4)) (1) and [Ni(Me(6)tren)Br](Br) (2). High-field, high-frequency electron paramagnetic resonance spectroscopy performed on a single crystal of 1 shows a giant uniaxial magnetic anisotropy with an experimental D(expt) value (energy difference between the M(s) = ± 1 and M(s) = 0 components of the ground spin state S = 1) estimated to be between -120 and -180 cm(-1). The theoretical study shows that, for an ideally trigonal Ni(II) complex, the orbital degeneracy leads to a first-order spin-orbit coupling that results in a splitting of the M(s) = ± 1 and M(s) = 0 components of approximately -600 cm(-1). Despite the Jahn-Teller distortion that removes the ground term degeneracy and reduces the effects of the first-order spin-orbit interaction, the D value remains very large. A good agreement between theoretical and experimental results (theoretical D(theor) between -100 and -200 cm(-1)) is obtained.

This perspectives article takes a broad view of the current understanding of magnetic bistability and magnetic quantum tunneling in single-molecule magnets (SMMs), focusing on three families of relatively simple, low-nuclearity transition metal clusters: spin S = 4 II 4 Ni , III 3 Mn (S = 2 and 6) and III 6 Mn (S = 4 and 12). The Mn III complexes are related by the fact that they contain triangular III 3 Mn units in which the exchange may be switched from antiferromagnetic to ferromagnetic without significantly altering the coordination around the Mn III centers, thereby leaving the single-ion physics more-or-less unaltered. This allows for a detailed and systematic study of the way in which the individual-ion anisotropies project onto the molecular spin ground state in otherwise identical low-and high-spin molecules, thus providing unique insights into the key factors that control the quantum dynamics of SMMs, namely: (i) the height of the kinetic barrier to magnetization relaxation; and (ii) the transverse interactions that cause tunneling through this barrier. Numerical calculations are supported by an unprecedented experimental data set (17 different compounds), including very detailed spectroscopic information obtained from high-frequency electron paramagnetic resonance and low-temperature hysteresis measurements. Comparisons are made between the giant spin and multi-spin phenomenologies. The giant spin approach assumes the ground state spin, S, to be exact, enabling implementation of simple anisotropy projection techniques. This methodology provides a basic understanding of the concept of anisotropy dilution whereby the cluster anisotropy decreases as the total spin increases, resulting in a barrier that depends weakly on S. This partly explains why the record barrier for a SMM (86 K for Mn 6 ) has barely increased in the 15 years since the first studies of Mn 12 -acetate, and why the tiny Mn 3 molecule can have a barrier approaching 60% of this record. Ultimately, the giant spin approach fails to capture all of the key physics, although it works remarkably well for the purely ferromagnetic cases. Nevertheless, diagonalization of the multi-spin Hamiltonian matrix is necessary in order to fully capture the interplay between exchange and local anisotropy, and the resultant spin-state mixing which ultimately gives rise to the tunneling matrix elements in the high symmetry SMMs (ferromagnetic Mn 3 and Ni 4 ). The simplicity (low-nuclearity, high-symmetry, weak disorder, etc..) of the molecules highlighted in this study proves to be of crucial importance. Not only that, these simple molecules may be considered among the best SMMs: Mn 6 possesses the record anisotropy barrier, and Mn 3 is the first SMM to exhibit quantum tunneling selection rules that reflect the intrinsic symmetry of the molecule.