New techniques for nanokelvin condensed matter physics

对越来越冷的温度的探索带来了许多了不起的发现。最难液化的气体是氦，但最终在20世纪初实现了液化。这一突破导致了超流体（一种没有摩擦的液体，就像超导体中的电流没有阻力一样流动）的有趣现象的观察，如今用氦制冷在技术上具有重要的基础意义，例如对牛津仪器这样的大型国际公司。在20世纪末，新技术被开发出来，使用激光冷却原子（尽管这似乎违反直觉）和磁阱，将原子限制在一个非常好的真空区域。这项新技术允许在低温下制备碱金属原子的稀蒸气。令人惊讶的是，这些金属原子不会聚集在一起（形成分子），蒸发使云进一步冷却到几十纳开尔文的极低温度。正如著名物理学家爱因斯坦和玻色所预言的那样，这使得在弱相互作用的稀释气体中首次实验观察到玻色-爱因斯坦凝聚（BEC）。这一极其有趣的量子现象与之前对超流氦的研究有关，但这种液体比超冷原子气体要复杂得多。使用最先进的天文学和显微镜相机可以拍摄到冷原子的非常详细的图像，这种直接看到量子流体中发生的事情的能力使得理解这些系统取得了非常迅速的进展，并提供了丰富的新知识。这在BEC的第一个实验中就很明显，从普通气体到量子态的相变被观察到密度和形状的巨大变化。我们已经建造了一种新的装置，其中原子在磁场中移动时所经历的势能景观是通过应用射频辐射以精确可控的方式定制的。（施加的辐射改变了原子在空间中特定位置的量子态，从而改变了它们感受到的电势。）这已被证明对二维系统和创建有趣的几何形状（如环形陷阱）特别有用（环路周围有量子相干性）。这种方法的一个很大的优点是，与用激光捕获原子相比，电势非常光滑，没有缺陷，那里会出现干涉条纹。我们将这种新方法与时间平均相结合，使陷阱的潜力范围更大，寿命更长。我们将继续开发和测试捕获原子的新方案，例如创建双环（原子在两个同心圆上）和在相互作用有共振增强的磁场中捕获原子（法诺-费什巴赫共振）。我们将把这项技术应用于强相关系统的直接量子模拟，并探索物波干涉测量法在精密测量设备上的应用。将原子气体冷却到纳米开尔文温度的这一进展现在使我们能够实现理查德·费曼（Richard Feynman）的说法：“因此，我相信，使用合适的量子机器，你可以模拟任何量子系统，包括物理世界。”这在量子信息处理的背景下经常被引用，但它更直接地适用于可控量子系统的创建，我们设计量子哈密顿量，使其看起来与感兴趣的物理系统相同。这是我们直接量子模拟工作的基本原理。在这种情况下，量子力学被用来设计量子机器，即一种可以控制地创造多体量子态的装置，就像汽车力学被用于制造车辆一样。

The quest for colder and colder temperatures has led to many remarkable discoveries. The most difficult gas to liquify is helium but this was finally achieved at the beginning of the 20th century. That breakthrough led the observation of the intriguing phenomena of superfluidity (a liquid that flows without friction just like the current in a superconductor flows without resistance) and nowadays refrigeration with helium is of fundamental technological importance, e.g. for large international companies such as Oxford Instruments. At end of the the 20th century new techniques were developed that used laser light to cool atoms (although this may seem counterintuitive) and magnetic traps that confine the atoms in a region of very good vacuum. This new technology allowed dilute vapours of alkali metal atoms to be prepared a low temperatures. Amazingly this atoms of metal do no clump together (to form molecules) and evaporation cools the cloud further to extremely low temperatures of tens of nanokelvin. This allowed the first experimental observation of Bose-Einstein condensation (BEC) in a weakly interacting dilute gas, as predicted by the famous physicists Einstein and Bose. This extremely interesting quantum phenomenon has links with previous work on superfluid helium but the liquid is more complicated to understand than the ultra-cold atomic gases. Wonderfully detailed images of the cold atoms can be taken using state-of-the-art cameras developed for astronomy and microscopy and this ability to see directly what is going on in quantum fluids has allowed very rapid progress in understanding these systems and provided a wealth of new knowledge. This was evident in the very first experiment on BEC where the phase transition from an ordinary gas to the quantum regime was observed as a dramatic change in the density and shape.We have built a novel apparatus in which in the the potential energy landscape that atoms experience as they move through the magnetic field is tailored in a precisely controllable way by the application of radio-frequency radiation. (The applied radiation changes the quantum state of the atoms at particular positions in space hence changing the potential they feel.) This has proven to be particularly useful for two-dimensional systems and for creating interesting geometries such as ring traps (with quantum coherence around the loop). A great advantage of this approach is that the potential is very smooth and free from defects, as compared to trapping atoms with laser light where interference fringes arise. We have combined this new approach with time-averaging to allow an even greater range of potentials and long lifetimes in the traps. We shall continue to develop and test new schemes for trapping atoms, such as the create of double rings (atoms on two concentric circles) and trapping atoms at magnetic fields where there is resonant enhancement of the interactions (Fano-Feshbach resonance). We shall apply this technology to the direct quantum simulation of strongly correlated systems, and explore applications such a matter-wave interferometry for precision measuring devices. This progress in cooling atomic gases to nanokelvin temperatures now allows us to fulfill the statement of Richard Feynman,``I therefore believe it's true that with a suitable class of quantum machines you could imitate any quantum system, including the physical world''. This is often quoted in the context of quantum information processing but it applies more directly to the creation of controllable quantum systems in which we engineer the quantum Hamiltonian so that it looks the same as that of the physical system of interest. This is the underlying principle of our work on Direct Quantum Simulation. In this context quantum mechanics is used to design a quantum machine, i.e. an apparatus to controllably create many-body quantum states, in the same way that automotive mechanics is used in the creation of vehicles.