论文标题
通过磁铁矿中的能量调整电子兴奋的超快生成隐藏相
Ultrafast generation of hidden phases via energy-tuned electronic photoexcitation in magnetite
论文作者
论文摘要
储层计算是预测湍流的有力工具,其简单的架构具有处理大型系统的计算效率。然而,其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间,然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用,该系统模拟了湍流的若干特征。我们表明,使用径向基函数作为非线性投影器,即使只有部分观测并且不知道控制方程,也能稳健地捕捉复杂的系统非线性。最后,我们表明,当测量稀疏、不完整且带有噪声,甚至控制方程变得不准确时,我们的网络仍然可以产生相当准确的预测,从而为实际湍流系统的无模型预测铺平了道路。
Metal-insulator transitions (MIT) occurring in non-adiabatic conditions can evolve through high-energy intermediate states that are difficult to observe and control via static methods. By monitoring the out-of-equilibrium structural dynamics of a magnetite (Fe3O4) crystal via ultrafast electron diffraction, we show that MITs can evolve through different pathways by properly selecting the electronic excitation with light. Near-infrared (800 nm) photons inducing d-d electronic transitions is found to favor the destruction of the long-range zigzag network of the trimerons and to generate a phase separation between cubic-metallic and monoclinic-insulating regions. Instead, visible light (400 nm) further promotes the long-range order of the trimerons by stabilizing the charge density wave fluctuations through the excitation of the oxygen 2p to iron 3d charge transfer and, thus, fosters a reinforcement of the monoclinic insulating phase. Our experiments demonstrate that tailored light pulses can drive strongly correlated materials into different hidden phases, influencing the lifetime and emergent properties of the intermediate states.