云学院回顾 | 离子阱量子模拟器中熵的度量
3月22日，奥地利因斯布鲁克大学量子光学和光谱研究所Rainer Blatt实验室的Tiffany Brydges博士给我们带来了离子阱量子模拟器中熵的度量的报告。下面我们一起来简单回顾本期报告的主要内容：
报告中Tiffany Brydges博士首先强调了量子模拟以及基于量子模拟的量子计算在未来一定会在实际应用中发挥其重要作用，并且使用空气动力学研究中的风洞作为实例证实量子模拟器在量子系统尤其是量子纠缠相关研究中起到的重要作用。随后，Tiffany Brydges博士着重介绍了实验装置及钙离子势阱模拟器的工作原理，展示了非常精彩的激光调控钙离子串中单个钙离子从基态到激发态转变的动力学过程，之后Tiffany Brydges博士将理论与实际相结合，展示了这种钙离子势阱在表征量子纠缠系统中特征参数二阶Rényi熵的测量中的应用。如果当时没有时间参加报告的老师，可通过我们的回顾视频观看，详细了解Tiffany Brydges博士的精彩报告。
在Tiffany Brydges博士的实验中，使用单个钙离子在792纳米是否能够发出荧光来表征钙离子处在基态或者激发态。单个离子发出的荧光非常的弱，因此在他们的系统中使用了Andor的iXon Ultra EMCCD作为检测器收集探测。iXon Ultra EMCCD是Andor公司的高端成像设备，具备超高灵敏度的特性，具有单光子级别的灵敏度，非常适合探测量子光学研究中单个离子、单原子发出的荧光。
Tiffany Brydges博士的精彩报告引起了参会人员的踊跃提问。下面是Tiffany Brydges博士针对大家的问题给出的回答，请大家参考。如果还有其他的问题，欢迎大家留言给我们。
1. How long your system can hold more than 20 ions without mielting?
On average there's a melting event about every 30 seconds with 20 ions. We have an automatic 'refreeze' system, so that when a melting event is detected any measurement is paused and the ions are recrystallised. In this way it's actually very rare that we will 'lose' ions and have to reload.
Interestingly (and contrary to what we expected) it seems that the number of melting events we see for 50 ions is not much higher than for 20 ions - we're still trying to determine why this is.
2. To get a good ion chain stability, what have you done to your trap and vacuum system?
(i) The trap is the classic 'Innsbruck style' design with blade electrodes, and was installed around 2012 (I think), so it's getting quite old now and we would like to upgrade this in the future. The large ion-electrode distance you get from the Innsbruck design helps lead to low heating rates. This trap was designed and installed well before my time and so I am not familiar with the exact intricacies of why it is so successful, other than that it was machined such that the alignment precision was extremely high. There is a particularly useful thesis from our group which deals with the trap in detail:
"Digital quantum simulation, Schroedinger cat state spectroscopy and setting up a linear ion trap", Cornelius Hempel, 2014.
(ii) We actively stabilise the RF going to the trap using a home-built PID system.
(iii) Our vacuum system isn't very specialised; we use an ion pump from Agilent Technologies along with an NEG. Luckily we haven't needed to open the vacuum chamber since ~ 2012, with our vacuum currently down at around 10^-10.
3. What is you traps axial mode frequency and is it easy to cool it down?
For the experiments I discussed, the axial frequency was around 220kHz, but we use anywhere between 118kHz and 1.2MHz (depending on the length of the string). It is relatively easy to cool these modes (particularly using polarisation gradient cooling), however the heating rate along the axial direction is much higher than with the radial modes. We're still unsure as to exactly why our axial heating rate is so high.
The entangling gates that I discussed were performed on the radial modes which we can effectively cool down to the ground state using sideband-cooling.