Magnetic excitations of small magnetic clusters and single atoms

非弹性扫描隧道光谱（ISTS）已被证明是一个很好的工具，用于研究纳米结构中的磁激发，允许确定激发光谱[1，2，3]和激发态的寿命[4，5]。在微观上，自旋-轨道相互作用和晶体场的相互作用导致局部磁晶各向异性能（MAE），即取决于净磁矩方向的能量。这提升了原子或团簇的磁性多重态的简并性，并且可以在自旋激发实验中进行探测。在团簇中，也可以激发非共线的自旋态，其中单个原子的磁矩相对于彼此转动，并增加交换能的代价。激发态的寿命主要由电子弛豫过程决定，其中激发态的能量被给予衬底的传导电子[4，5，6]。此外，MAE雅阁Stenvens算子诱导磁性多重态内的混合，从而实现磁化的隧穿。对于磁性过渡元素，可以观察到每个原子10 meV数量级的大各向异性[7，2，8]。由于原子3d态与衬底的有效耦合，无论是金属[2，6]还是薄绝缘层[5]，激发态和基态的寿命都在10−14和10−7s之间，并且无法实现稳定的磁性状态。在这个提议中，我们将研究4f金属原子，其中磁矩强烈地位于4f轨道中，使得与周围状态的杂化显着减少。通过这种方式，可以在更长的时间尺度上实现稳定的自旋，从而可以研究它们的量子性质并执行简单的量子计算操作。

Inelastic scanning tunneling spectroscopy (ISTS) has proven to be an excellent tool for the investigation of magnetic excitations in nanostructures allowing both to determine the excitation spectrum [1, 2, 3] and the life times of the excited states [4, 5]. Microscopically, the interplay of spin-orbit interaction and the crystal field causes a local magnetocrystalline anisotropy energy (MAE), i.e. an energy that depends on the direction of the net magnetic moment. This lifts the degeneracy of the magnetic multiplet of the atom or cluster and can be probed in a spin excitation experiment. In clusters, also non-collinear spin states can be excited, in which the moments of individual atoms are turned with respect to each other with the additional cost of the exchange energy. The life time of the excited state is mainly determined by an electronic relaxation process, in which the energy of the excited state is given to a conduction electron of the substrate [4, 5, 6]. Further the MAE induces mixing within the magnetic multiplet in accord to the Stenvens operators enabling tunneling of the magnetization. For magnetic transition elements, large anisotropies of the order of 10 meV per atom could be observed [7, 2, 8]. Due to the effective coupling of the atomic 3d states to the substrate, be it metallic [2, 6] or a thin insulating layer [5], the life times of the excited and ground states are between 10−14 and 10−7s, and no stable magnetic states could be achieved. In this proposal, we will work on 4f metallic atoms in which the magnetic moment is strongly located in the 4f orbitals such that the hybridization to the surrounding states is significantly reduced. This way, stable spins can be achieved for longer time scales that allow to investigate their quantum nature and to perform simple quantum computational operations.