1.Key Laboratory of Artificial Structures and Quantum Control (Ministry of Education), Shenyang National Laboratory for Materials Science, School of Physics and Astronomy, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China 2.Tsung-Dao Lee Institute, Shanghai 200240, China
Fund Project:Project supported by the National Basic Research Program of China (Grant Nos. 2016YFA0301003, 2016YFA0300403), the National Natural Science Foundation of China (Grant Nos. 11521404, 11634009, U1632102, 11504230, 11674222, 11574202, 11674226, 11574201, U1632272, 11655002), and the Strategic Priority Research Program of Chinese Academy of Sciences, China (Grant No. XDB28000000).
Received Date:12 September 2018
Accepted Date:11 October 2018
Available Online:01 July 2019
Published Online:05 July 2019
Abstract:The search for new states that exhibit topological order is currently a very active and exciting area of research. Like a topological insulator, superconducting order can also exhibit topological order, which is different from that of a conventional superconductor. This superconductor is so-called " topological superconductor”, which has a full pairing gap in the bulk and gapless surface state. Majorana Fermions obey non-Abelian fractional statistics, and have been proposed to construct topological qubits, so there is a great prospect of scientific research and application in topological quantum computing. It is very interesting that Majorana Fermions are predicted to exist in topological superconductors. However, natural topological superconductor is very rare. Inspired by the realization of topological insulators, theoretical physicists have proposed that via the fabrication of the s-wave superconductor/topological insulator heterostructure, Majorana Fermions may exist in the superconducting topological insulator induced by proximate effect. Due to various kinds of topological insulators and conventional s-wave superconductors, heterostructures constructed by this method can greatly increase the variety of artificial topological superconductors. In this paper we review the experimental progress in the heterostructure composed of the Bi2Te3-type topological insulator and the conventional s-wave superconductor NbSe2. Using molecular beam epitaxy, atomically flat topological insulator film can be fabricated at the top of superconductor substrate. The spatial distribution of Majorana Fermions on the surface of topological insulator can be directly observed by in situ scanning tunneling microscopy/spectroscopy. In the center of a magnetic vortex, Majorana Fermions will appear as the Majorana zero mode, a zero-energy peak inside the superconducting gap. Although the energy gap between low energy quasiparticle excitation and the Majorana zero mode is very small, the evidences such as zero bias conductance anomaly, Y-shape splitting of zero-bias conductance, spin-selective Andreev reflection are self-consistent and reveal that the Majorana zero mode exists in the center of a magnetic vortex. These experiments have led to a new insight into superconductivity. It may open a door to probing the novel physics of Majorana fermions. Keywords:proximity effects/ superconducting films and low-dimensional structures/ vortex phases/ tunneling phenomena
全文HTML
--> --> --> -->
2.1.Bi2Se3/NbSe2异质结
由于拓扑绝缘体和超导体材料之间的晶格失配、界面反应、再加上拓扑绝缘体的热稳定性不高, 实验制备拓扑绝缘体/超导体异质结是非常有挑战的事情. 然而, 这些问题可以用van der Waals外延生长的方法解决, 这种生长方法所用的材料都是具有van der Waals间隙的层状结构[42]. 拓扑绝缘体Bi2Se3每一层由Se–Bi–Se–Bi–Se五原子层(QL)构成[43], 超导体NbSe2每一层由Se–Nb–Se三原子层(TL)构成[44], 层于层之间都具有van der Waals间隙. Wang等[30]利用分子束外延生长方法, 将拓扑绝缘体Bi2Se3薄膜长在解理的超导体NbSe2衬底上, 并用低温扫描隧道显微镜(STM)和扫描隧道显微谱(STS)对其进行了研究. 图1(a)显示了大范围原子级平整的覆盖度为2 QL厚的Bi2Se3薄膜的STM图像. 图中大部分地方厚度为2 QL, 有些地方的厚度为1 QL和3 QL. 图1(b)是图1(a)中红色虚线处Bi2Se3/NbSe2异质结侧视结构示意图. 图1(c)为原子分辨的STM图, 图1(c)显示NbSe2衬底表面有3 × 3的重构, 这是电荷密度波引起的图案. 而Bi2Se3薄膜表面为Se原子终止的六角晶格结构(图1(d)), 周期为0.41 nm, 这说明薄膜的表面没有重构, 是平整的Bi2Se3 (111)-(1 × 1)表面. 所以, van der Waals外延生长方法可以得到原子级锐利的拓扑绝缘体/超导体异质结界面. 图 1 (a)在NbSe2衬底上生长的Bi2Se3薄膜的形貌; (b)Bi2Se3/NbSe2异质结示意图; (c)NbSe2衬底表面的原子分辨STM图; (d) Bi2Se3薄膜表面的原子分辨STM图[30] Figure1. (a) Morphology of Bi2Se3 thin films grown on NbSe2 substrate; (b) schematic diagram of the Bi2Se3/NbSe2 heterostructure; (c) atomically resolved STM image of the NbSe2 substrate; (d) atomically resolved STM image of the Bi2Se3 film[30].
通过测量表面的微分电导谱(dI/dV)可以获得表面局域态密度(LDOS)的信息. 图2是在NbSe2衬底上生长的厚度分别为3 QL和6 QL的Bi2Se3薄膜表面测得的dI/dV谱. 在4.2 K温度下测得的dI/dV谱显示LDOS在费米能级处有明显的下降, 而在0.4 K温度下可以看到在 ± 1 meV处有非常强的相干峰. 这些都是非常明显的超导能隙的特征. 这些实验结果说明, 在NbSe2衬底上生长的Bi2Se3薄膜在低温下确实变成了超导体, 而且随着磁场或者温度的增加, 零偏压电导(ZBC)增强, 能隙两边相干峰的强度减弱直到消失. 所以通过加磁场和变温测量进一步证明了Bi2Se3薄膜的确发生了超导转变. 另外, 在0.4 K温度下, Bi2Se3薄膜厚度在1 QL—7 QL的范围内都观测到了超导能隙. 图 2 在Bi2Se3/NbSe2上探测的超导能隙[30] (a) 4.2 K和 (b) 0.4 K温度下3 QL厚的Bi2Se3薄膜的dI/dV谱; (c) 4.2 K和 (d) 0.4 K温度下6 QL厚的Bi2Se3薄膜的dI/dV谱 Figure2. Superconducting energy gap detected in Bi2Se3 thin films grown on NbSe2 substrate[30]: dI/dV spectra measured on 3 QL Bi2Se3 films at (a) 4.2 K and (b) 0.4 K; dI/dV spectra measured on 6 QL Bi2Se3 films at (c) 4.2 K and (d) 0.4 K.
在NbSe2衬底上生长的Bi2Te3薄膜, 当厚度达到11 QL时, 在0.4 K下都能观测到非常明显的超导能隙(图5(a)). 图5(b)给出了NbSe2衬底表面以及2 QL和3 QL厚的Bi2Te3薄膜的dI/dV谱线. 直到厚度为2 QL时, STS谱线在零偏压附近都是平的, 微分电导数值为零, STS谱线可以被BCS型谱函数拟合得非常好(图5(b)). 从1 QL到11 QL厚度上测得的dI/dV谱线都用s波BCS型曲线拟合, 数据整理在图5(c)中. 超导能隙随着薄膜厚度增加呈指数衰减, 这与超导近邻效应诱导的超导能隙随距离的衰减趋势在定性上来讲是一致的. 另外, 当薄膜厚度大于2 QL时, STS谱线在零偏压附近不是平的, 而且微分电导数值不为零. STS谱线不能被BCS型谱函数完全拟合(图5(b)). 这可能是由于随着薄膜厚度不断增加, 在达到3 QL厚时, 拓扑表面态在薄膜中出现. 拓扑表面态的出现对dI/dV谱线产生了重要影响, 使之偏离s波BCS谱型[49,50]. 为了证实这一点, 将3 QL厚的Bi2Se3薄膜在0.4 K测得的dI/dV谱线也与标准的BCS隧穿谱进行了比较(图5(c) 插图), 可以看出它比3 QL厚的Bi2Te3薄膜的拟合符合得更好. 因此可以做出判断, 拓扑表面态在超导近邻效应下可以发生超导转变, 它会使超导能隙的BCS谱型发生偏离. 换言之, Bi2Te3/NbSe2异质结是人工构造的拓扑超导体[31]. 图 5 在Bi2Te3/NbSe2上探测的超导能隙[31] (a)各种厚度的Bi2Te3薄膜上测得的超导能隙; (b)在NbSe2衬底, 2 QL以及3 QL Bi2Te3/NbSe2上测得的超导能隙; (c)超导能隙随厚度的变化, 插图为3 QL Bi2Se3/NbSe2的超导能隙. 这些dI/dV谱都是在0.4 K温度下测量的 Figure5. Superconducting energy gap observed on Bi2Te3/NbSe2[31]: (a) A series of dI/dV spectra taken on different thicknesses of Bi2Te3 thin films at 0.4 K; (b) dI/dV spectra taken on pristine NbSe2, 2 QL, and 3 QL Bi2Te3/NbSe2; (c) thickness dependence of the superconducting energy gap; Inset is the dI/dV spectra measured at 0.4 K on 3 QL Bi2Se3/NbSe2.
最近, Xu等[51]利用极低温高分辨的ARPES在7 QL Bi2Se3/NbSe2的能带结构中直接观测到了拓扑表面态的超导能隙. 图6(a)展示了4 QL Bi2Se3/NbSe2的能带分布. 位于波矢k1的表面电子态和位于波矢k2的体电子态随温度的变化分别显示在图6(b)和图6(c)中. 在低温下超导信号的特征非常明显, 温度升高到7 K以上时, 相干峰和超导能隙都消失. 对于7 QL Bi2Se3/NbSe2, Dirac点清晰可见, 而且表面态是拓扑非平庸的(图6(d)). 位于波矢k1的拓扑表面电子态(图6(e))和位于波矢k2的体电子态(图6(f))的超导信号的特征也很明显. 这些数据非常强有力地说明了拓扑超导体/NbSe2异质结是研究超导近邻效应诱导的新奇物理现象的理想平台. 图 6 (a) 12 K时测得的4 QL Bi2Se3/NbSe2的能带结构, 入射光子能量为18 eV; 4 QL 厚的Bi2Se3/NbSe2在 (b) k1和 (c) k2处的ARPES谱随温度的变化关系; (d) 12 K时测得的7 QL Bi2Se3/NbSe2的能带结构, 入射光子能量为18 eV; 7 QL 厚的Bi2Se3/NbSe2在 (e) k1和(f) k2处的ARPES谱随温度的变化关系[51] Figure6. (a) Band structure of a 4 QL Bi2Se3/NbSe2 measured at 12 K using an incident photon energy of 18 eV; Temperature dependence of ARPES spectra at (b) k1 and (c) k2 indicated in Fig. (a); (d) Band structure of a 7 QL Bi2Se3/NbSe2 measured at 12 K using an incident photon energy of 18 eV; Temperature dependence of ARPES spectra at (e) k1 and (f) k2 indicated in Fig. (d)[51].