Resonant magnetization tunneling inMn12acetate: The absence of inhomogeneous hyperfine broadening

Abstract

We present the results of a detailed study of the thermally-assisted-resonant-tunneling relaxation rate of Mn12 acetate as a function of an external, longitudinal magnetic field and find that the data can be fit extremely well to a Lorentzian function. No hint of inhomogeneous broadening is found, even though some is expected from the Mn nuclear hyperfine interaction. This inconsistency implies that the tunneling mechanism cannot be described simply in terms of a random hyperfine field.

Detailed single-crystal EPR line shape measurements for the single-molecule magnetsFe8BrandMnHill1, Maccagnano2, Park3 et al. 2002Phys. Rev. B1188771View full textAdd to dashboardCite

“…It has also been shown that long-range strains produced by dislocations result in broad distributions of relaxation times, something which is also observed experimentally in Mn 12 −ac [40]. Indeed, assuming realistic concentrations of dislocations, Garanin and Chudnovsky [40] have been able to account for the experimentally observed tunneling in Mn 12 −ac [49], and predict a distribution in the uniaxial spin-spin coupling parameter σ D ≈ 0.027D, which is of the same order as we have found from linewidth analyses (σ D ≈ 0.01 − 0.02D). It should be pointed out, however, that any disorder amongst the ligand molecules, as well as any loss of solvent molecules, may be expected to produce many of the same effects as those proposed by Garanin and Chudnovsky [40], i.e.…”

“…Several EPR (and INS) experiments have indicated the presence of significant fourth order single-ion transverse anisotropies for both Mn 12 − ac [16,26,32,35] and Fe 8 Br [33,35]. Nevertheless, fourth order anisotropy cannot fully explain the observed spacing in magnetic field of magnetization steps observed in low temperature hysteresis experiments [49], and there remains considerable disagreement as to the magnitude of the fourth order terms [16,26,32,33,34,35,51]. Consequently, higher resolution EPR measurements are neces-sary in order to refine estimates of the Hamiltonian parameters, and to determine whether EPR lineshapes contain additional information concerning the MQT phenomenon.…”

Single crystal EPR determination of the spin Hamiltonian parameters for Fe8 molecular clusters

“…1 suggest that the spin levels are inhomogeneously broadened, since a mixture of Gaussian and Lorentzian lineshapes provide the best fits to the data. A significant contribution to this broadening is likely due to a distribution in D and/or g values (strain), in additon to possible random dipolar or hyperfine fields; we note, a spread in D and/or g would not affect the zero field quantum tunneling resonance which has been shown to be Lorentzian in the Mn 12 system by Friedman et al [25]. To verify this, we plot the M S dependence of the linewidths in Fig.…”

Dislocation-induced spin tunneling inMn12acetate

“…Thus the transverse anisotropy and the transverse field generated by dislocations [see, e.g., Eq. (27)] are typically small in comparison with the uniaxial term in the Hamiltonian of the ideal crystal. In this situation one can obtain the tunneling splittings ∆ mm ′ of resonant pairs of levels in the lowest-order of the perturbation theory.…”

“…6 There exist two macroscopic experimental approaches to the study of spin tunneling in Mn 12 (see, e.g., Refs. 6,7,8,25,26,27,28,29,10,9,30,11,12,31,32 and references therein). Both clearly demonstrate the effect of resonant spin tunneling.…”