It has been shown within density-functional theory that in Mn12-acetate there are effects due to disorder by solvent molecules and a coupling between vibrational and electronic degrees of freedom. We calculate the in-plane principal axes of the second-order anisotropy caused by the second effect and compare them with those of the fourth-order anisotropy due to the first effect. We find that the two types of the principal axes are not commensurate with each other, which results in a complete quenching of the tunnel-splitting oscillation as a function of an applied transverse field.PACS numbers: 75.50. Xx, 71.15.Mb, 75.30.Gw, 75.30.Et The observation of resonant tunneling of magnetization in the single molecule magnet Mn 12 -acetate 1 (hereafter Mn 12 ) with a ground-state spin of S = 10 has led to many experimental and theoretical investigations. 2,3,4,5,6 A simple anisotropy Hamiltonian for S = 10 provides an excellent approximation to the physics occurring at low temperatures. Deriving the Hamiltonian from density-functional (DF) calculations is challenging but possible. 4,5,6 For a particular molecular geometry, a magnetic anisotropy tensor may be calculated considering spin-orbit coupling within a DF framework. In the principal-axes coordinates, the lowest-order spin Hamiltonian can be simplified to the following form:where D and E are the uniaxial and second-order transverse anisotropy parameters and S z is the easy-axis component of the spin operator S. The S 4 symmetry of the ideal Mn 12 molecule causes the value of E to vanish and the lowest-order transverse terms to be fourth order. Only transverse anisotropy terms are responsible for the resonant tunneling between energy levels that are almost degenerate. Magnetic tunneling measurements, however, showed that some of the resonant tunneling occurred at a level lower than fourth order. For the Mn 12 , the calculated D value is -0.556 K, 4 which agrees well with experiment. The calculated (measured) value of D, however, does not account for all the measured anisotropy barrier of 65 K. 3,8 One needs a longitudinal anisotropy of order higher than second-order. In this spirit a fourth-order spin-orbit-vibron (SOV) interaction was proposed by Pederson et al. 11 Ideally, this SOV interaction can only contribute to modifying the second-order barrier and to activating two fourth-order transverse terms in the spin Hamiltonian. At this level a strong SOV interaction can complicate the tunneling experiments in several different ways. The simplest complication is that there are two different longitudinal 4th-order terms which scale as S 2 S 2 z and S 4 z respectively. If the S 4 z term is dominant, then at any resonant field only a single pair of states are involved with tunneling. The fourth-order transverse terms result in departures of the period of the tunnel-splitting oscillation from ∆H x = 2 2E(|D| + E)/gµ B that has been derived by Garg. 12 Such departures were first identified by Wernsdorfer et al. 13 and later modified to include the 4th-order transverse...