エキゾティック原子の量子気体の理論

在这一年中，我改进了描述超冷偶极费米气体动力学的数值方法，并获得了关于这种系统中的爱因斯坦-德-哈斯效应的令人信服的新结果。这是必要的，因为我得到的第一个结果被意想不到的人为因素所缺陷，因此我仔细地确定了数值误差的所有来源，并制定了克服它们的措施。利用改进后的数值设置，我研究了具有两个自旋分量的二维偶极费米气体的动力学，重点研究了所谓的爱因斯坦-德哈斯（EdH）效应，即磁化向轨道角动量的转移。我的结果表明，在费米气体中，EdH效应与一个额外的过程一起出现，一个扭曲运动，其中两个自旋分量以相反的方向旋转，其单个角动量比整个系统大得多。使用能量泛函，我发现这是费米表面变形的结果，由各向异性偶极子-偶极子相互作用引起，因此在BEC中不存在。这种扭转运动的一个主要结果是，EdH效应可能更容易在费米气体中进行实验检测。这种扭转运动的第二个特征是它对s波散射长度或外部磁场变化的敏感性，这可能导致相对旋转的逆转。这可能对超冷偶极气体（如镝）的实验研究非常有用，在这些实验中，s波散射长度的值到目前为止仍然未知。

During this year, I improved the numerical methods for describing the dynamics of ultracold dipolar Fermi gases and obtained compelling new results regarding the Einstein-de Haas effect in such systems. This was necessary because the first results I had obtained turned out to be flawed by unexpected artifacts, therefore I meticulously identified all sources of numerical errors and developed measures to overcome them. With the resulting improved numerical setup, I studied the dynamics of two-dimensional dipolar Fermi gases with two spin components, focussing on the so-called Einstein-de Haas (EdH) Effect, the transfer of magnetization into orbital angular momentum. My results show that in a Fermi gas, the EdH effect appears together with an additional process, a twisting motion, where both spin components rotate in opposite directions with considerably larger individual angular momenta than the full system. Using energy functionals I found out that this is the result of Fermi surface deformation, induced by the anisotropic dipole-dipole interactions and therefore absent in a BEC. A main consequence of this twisting motion is that the EdH effect might be easier to detect experimentally in a Fermi gas. A second feature of this twisting motion is its sensitivity to changes in the s-wave scattering length or an external magnetic field, which can lead to a reversal of the relative rotation. This could prove very useful for experimental research of ultracold dipolar gases such as Dysprosium, where the values of s-wave scattering lengths remain unknown so far.