Fig. 1. Quantum memory based on a Nd
3+:YVO
4 crystal
[35]. The single photons are emitted from a quantum dot on another optical table.
基于钕离子掺杂钒酸钇晶体的量子存储(单光子来源于另一个光学平台的量子点)
[35] Fig. 2. EIT quantum memory based on cold atoms
[40]. The rubidium atomic ensemble in the first cigar-shaped magneto-optical trap (MOT1) is used to generate photon pairs. The rubidium atomic ensemble in MOT2 acts as a quantum memory, and is used to store the anti-Stokes photons from MOT1.
基于冷原子EIT的量子存储(第一个磁光阱(MOT1)囚禁的雪茄型铷原子系综用来产生双光子对, 第二个磁光阱(MOT2)囚禁的铷原子系综作为量子存储器, 用来存储来自于MOT1的anti-Stokes光子)
[40] Fig. 3. GEM schematic
[45]: (a) A three-level system; (b) an ensemble of atoms with linearly varying frequency shift in the
z direction; (c) a pulse of light is stored in the frequency-shifted ensemble; (d) after reversal of the frequency gradient at time
, a photon echo emerges at time 2
; (e) the optical layout; (f) the applied magnetic field,
Bz.
磁场GEM的原理图
[45] (a) 三能级系统; (b) 沿
z方向线性频移的原子系综; (c) 光脉冲将要存入频移了的原子系综; (d) 在
时刻, 频率梯度反转, 在2
时刻出现光子回波; (e) 主要光路; (f) 施加的梯度磁场
Bz Fig. 4. Far off-resonance Raman memory: (a) Principle of experiment
[48]; (b)
-type energy level; the strong control light (blue line) induces a virtual energy level (black dashed line), and couples (retrieve) the signal photons (read lines) into (from) the caesium atomic ensemble
[48]; (c) broadband single-photon-level memory in a hollow-core photonic crystal fibre
[49].
远失谐拉曼存储 (a) 实验原理
[48]; (b) Λ型能级结构, 强的控制光(蓝线)激发出虚能级(黑色虚线), 并将信号光子(红线)耦合进铯原子系综或者将存储的信号光子读取出来
[48]; (c) 在空心光子晶体光纤中实现单光子量级的宽带光存储
[49] Fig. 5. Quantum memory protocol based on off-resonant cascaded absorption (ORCA)
[27]: (a) The weak input signal pulse (blue line) and strong control pulse (orange line) are counter-propagating; (b) the relevant caesium atomic levels, where the storage state is 6D
5/2; (c) the experimental setup.
基于非共振梯形吸收的量子存储
[27] (a) 信号光子(蓝线)和控制光(橙线)反向传播; (b) 具体采用的铯原子能级, 其中6D
5/2是存储态; (c) 实验装置图
Fig. 6. Milestone works of quantum memory towards broadband and quantum regime in atomic ensemble
[51]. The quantum memory experiments in cold atoms are shown in black diamond. The quantum memory experiments in room-temperature atoms are shown in red diamond. ORCA: Quantum memory based on off-resonant cascaded absorption. FORD: Quantum memory based on far off-resonance DLCZ protocol.
在实现宽带量子存储的历程中, 基于原子系综的量子存储的代表性工作
[51], 其中冷原子实验用黑色方块表示; 热原子实验用红色方块表示; ORCA表示梯形量子存储; FORD表示远失谐DLCZ量子存储
Fig. 7. Experimental principle
[51]. (a) Experimental setup. The caesium cell is packed in a three-layer magnetic shielding and is heated up to 61.3 ℃. WP, Wollaston prism; QWP, quarter-wave plate; HWP, half-wave plate; PBS, polarization beam splitter. (b) The write process of FORD quantum memory. (c) The read process.
FORD存储方案的原理图
[51] (a) 实验装置图, 其中铯池置于3层磁屏蔽筒内并被加热到61.3 ℃, WP代表沃拉斯顿棱镜; QWP代表四分之一波片; HWP代表二分之一波片; PBS为偏振分束器; (b) FORD存储的写过程; (c) 读过程
Fig. 8. Generation of entanglement based on DLCZ protocol. (a) Initially, the atoms are prepared in state
. Then a write light interacts with atoms and generates a Stokes photon with a probability of a few percent. (b) The photons detected by the detector 1 (denoted by D1) and detector 2 (denoted by D2) may come from either the cell A or the cell B. If the two detectors (D1 and D2) detect only one photon, and one cannot distinguish whether the photon is from the cell A or cell B, then the entanglement between the cell A and cell B is established. BS: beam spliter with a splitting ratio of 50 : 50. (c) Young’s double slit experiment. We can not distinguish which slit the photon passes through.
基于DLCZ方案建立纠缠的原理 (a) 一开始, 原子被制备在初态
上, 然后写光与原子相互作用, 并以百分之几的概率产生Stokes光子; (b) 探测器1 (记为D1)和探测器2 (记为D2)探测到的光子有可能是来自于原子池A也有可能来自于原子池B, 在不能分辨Stokes光子是来源于原子池A还是B的前提下, 如果D1和D2两个探测器只有一个探测到光子且只探测到一个光子, 则原子池A和B之间存在纠缠; BS,光束分束器, 这里用的分束比是50 : 50; (c) 杨氏双缝干涉, 我们不能确定光子会从哪个狭缝通过
Fig. 9. Entanglement swapping based on DLCZ protocol. The retrieval process. The read light retrieves the storage state
out as an anti-Stokes photon. (b) Entanglement swapping. Initially, cell A and B are entangled, cell C and D are entangled. Under the influence of read light, both the cell B and C will emit anti-Stokes photons with a certain probability. If the two detectors (D1 and D2) detect only one photon, and one can not distinguish whether the photon is from cell B or cell C, then the cell A and D are entangled. By analogy, one can establish an entanglement between two atomic ensembles separated by great distance.
基于DLCZ 方案的纠缠交换 (a) 读取过程, 读光将存储态
读出为anti-Stokes光子; (b) 纠缠交换, 一开始, 原子池A和B存在纠缠, 原子池C和D存在纠缠, 原子池B和C在读光作用下有一定概率产生anti-Stokes光子, 在不能分辨光子是来源于原子池B还是C的前提下, 如果 D1和D2两个探测器只有一个探测到光子且只探测到一个光子, 则原子池A和D之间会产生纠缠; 以此类推, 便可在距离很远的两个原子系综之间建立纠缠
Fig. 10. Multiphoton synchronization based on DLCZ protocol. N cells interacting with write light can stochastically generate Stokes photons (green circles) and collective excitations. Repeatedly write the cell until a Stokes photon is generated. When each of the cells successfully stores a collective excitation, turn on the read light and retrieve all of the collective excitations out as N synchronous anti-Stokes photons (blue circles).
基于DLCZ方案的多光子同步, 其中在写光作用下, N个原子池随机产生Stokes光子(绿色圆)和与Stokes光子对应的集体激发态; 对每个原子池反复进行写操作, 直到产生Stokes光子为止; 当所有原子池都成功存储了集体激发态, 用读光将所有原子池内的集体激发态同时读取, 以产生N个时间上同步的anti-Stokes光子(蓝色圆)
| 具有代表性的工作 | 存储方案 | 存储器温度 | 互关联函数g(2) | 带宽 | 时间带宽积 | 1 | Phys. Rev. Lett.110 083601 (2013)
| EIT | 300 μK | ≤2 | <5 MHz | 74 | 2 | Nature438 837 (2005)
| EIT | 303—320 K | 2—3 | ~1 MHz | ~1 | 3 | Nature438 833 (2005)
| EIT | ~100 μK | 8.5 | 12 MHz | 120 | 4 | Nat. Photon.5 628 (2011)
| EIT | ~100 μK | 10 | 5.5 MHz | 13 | 5 | Phys. Rev. A75 040101 (2007)
| DLCZ | 333 K | 1.3 | 1 MHz | NA | 6 | Nat. Phys.5 95 (2009)
| DLCZ | 100 μK | 37 | <10 MHz | <10000 | 7 | Opt. Lett.37 142 (2012)
| DLCZ | 310 K | 4 | 1 MHz | 5 | 8 | Nat. Photon.10 381 (2016)
| DLCZ | ~100 μK | ~37 | <10 MHz | <2200000 | 9 | Nature461 241 (2009)
| GEM | 300K | ≤2 | 1 MHz | NA | 10 | Nat. Commun. 174 (2011)
| GEM | 351 K | ≤2 | ~1 MHz | ≤10 | 11 | Optica3 100 (2016)
| GEM | 100 μK | ≤2 | <10 MHz | 84 | 12 | Nat. Photon.4 218 (2010)
| Far off-resonance Raman | 335.5 K | ≤2 | 1.5 GHz | 18 | 13 | Phys. Rev. Lett.107 053603 (2011)
| Far off-resonance Raman | 335.5 K | ≤2 | 1.5 GHz | 2250 | 14 | Phys. Rev. Lett.116 090501 (2016)
| Far off-resonance Raman | 343 K | ≤2 | 1 GHz | 95 | 15 | Nat. Photon. 9 332 (2015)
| Raman memory | ~100 μK | 13.6 | 140 MHz | 200 | 16 | Nature432 482 (2004)
| Off-resonant Faraday interaction | 300 K | ≤2 | NA | NA | 17 | Phys. Rev. A97 042316 (2018)
| Off-resonant cascaded absorption (ORCA) | 364 K | 120 | 1 GHz | 5 | 18 | Commun. Phys.1 55 (2018)
| Far off-resonance DLCZ (FORD) | 334 K | 28 | 537 MHz | 700 |
|
Table 1. Milestone works on quantum memory in atomic ensemble and key figures of merit[51].
各种基于原子系综的具有代表性的量子存储器及其重要参数[51]