Van der Waals heterostructures are comprised of stacked atomically thin two-dimensional crystals and serve as novel materials providing unprecedented properties. However, the random natures in positions and shapes of exfoliated two-dimensional crystals have required the repetitive manual tasks of optical microscopy-based searching and mechanical transferring, thereby severely limiting the complexity of heterostructures. To solve the problem, here we develop a robotic system that searches exfoliated two-dimensional crystals and assembles them into superlattices inside the glovebox. The system can autonomously detect 400 monolayer graphene flakes per hour with a small error rate (<7%) and stack four cycles of the designated two-dimensional crystals per hour with few minutes of human intervention for each stack cycle. The system enabled fabrication of the superlattice consisting of 29 alternating layers of the graphene and the hexagonal boron nitride. This capacity provides a scalable approach for prototyping a variety of van der Waals superlattices.

We investigates exciton-exciton annihilation (EEA) in tungsten disulfide (WS 2) monolayers encapsulated by hexagonal boron nitride (hBN). It is revealed that decay signals observed by time-resolved photoluminescence (PL) are not strongly dependent on the exciton densities of hBN-encapsulated WS 2 monolayers (WS 2 /hBN). In contrast, the sample without the bottom hBN layer (WS 2 /SiO 2) exhibits a drastic decrease of decay time with increasing exciton density due to the appearance of a rapid PL decay component, signifying nonradiative EEA-mediated recombination. Furthermore, the EEA rate constant of WS 2 /hBN was determined as (6.3 ± 1.7) × 10-3 cm 2 s-1 , being about two orders of magnitude smaller than that of WS 2 /SiO 2. Thus, the observed EEA rate reduction played a key role in enhancing luminescence intensity at high exciton densities in the WS 2 monolayer.

The spin-orbit interaction (SOI) of a two-dimensional hole gas in the inversion symmetric semiconductor Ge is studied in a strained-Ge=SiGe quantum well structure. We observe weak antilocalization (WAL) in the magnetoconductivity measurement, revealing that the WAL feature can be fully described by the k-cubic Rashba SOI theory. Furthermore, we demonstrate electric field control of the Rashba SOI. Our findings reveal that the heavy hole (HH) in strained Ge is a purely cubic Rashba system, which is consistent with the spin angular momentum m j ¼ AE3=2 nature of the HH wave function. DOI: 10.1103/PhysRevLett.113.086601 PACS numbers: 72.25.Dc, 73.20.Fz, 73.21.-b The spin-orbit interaction (SOI) in a two-dimensional system is a subject of considerable interest because the SOI induces spin splitting at a zero magnetic field, which is important in both fundamental research and electronic device applications [1]. Recent developments of SOI-induced phenomena in the solid state demonstrate many possibilities utilizing spin current and the emergence of new physics such as the spin interferometer [2,3], persistent spin helix [4,5], spin Hall effect [6][7][8], and quantum spin Hall effect [9,10]. Up to now, there have been two well-known SOIs existing in solids: the Dresselhaus SOI [11] due to bulk inversion asymmetry (BIA) in the crystal structure and the Rashba SOI [12,13] due to spatial inversion asymmetry (SIA).In low-dimensional systems, the Rashba SOI becomes more important because it is stronger at the heterointerface and can be controlled by an external electric field. Many of the pioneering studies on the SOI-induced phenomena mentioned above were performed in two-dimensional electron systems, where the Rashba SOI is described by the k-linear Rashba term. In the Hamiltonian, the k-linear Rashba term can be written aswhere σ AE ¼ 1=2ðσ x AE iσ y Þ denote combinations of Pauli spin matrices, k AE ¼ k x AE ik y , and k x , k y are the components of the in-plane wave vector k ∥ . The effective magnetic field Ω 1 ðk ∥ Þ acting on the transport carrier due to the k-linear Rashba term is illustrated in Fig. 1(a).Recently, a higher-order contribution of the Rashba SOI, the so-called k 3 (k-cubic) Rashba SOI, has received more attention [14,15]. The Hamiltonian for the k-cubic Rashba SOI is expressed asand the effective magnetic field Ω 3 ðk ∥ Þ in k space is illustrated in Fig. 1(b) [15]. There is a significant difference in the effective field symmetry between the k-linear and the k-cubic Rashba SOI with one and three rotations in k space, respectively. The k 3 symmetry of the SOI is an interesting subject because it influences all of the SOI-induced phenomena as opposed to the k-linear Rashba term. For example, in case of the spin Hall effect, the k-cubic Rashba term is predicted to give rise to a larger spin Hall conductivity [17][18][19].

Graphene-based vertical field effect transistors have attracted considerable attention in the light of realizing high-speed switching devices; however, the functionality of such devices has been limited by either their small ON-OFF current ratios or ON current densities. We fabricate a graphene/MoS 2 /metal vertical heterostructure by using mechanical exfoliation and dry transfer of graphene and MoS 2 layers. The van der Waals interface between graphene and MoS 2 exhibits a Schottky barrier, thus enabling the possibility of well-defined current rectification. The height of the Schottky barrier can be strongly modulated by an external gate electric field owing to the small density of states of graphene. We obtain large current modulation exceeding 10 5 simultaneously with a large current density of ~10 4 A/cm 2 , thereby demonstrating the superior performance of the exfoliated-graphene/MoS 2 /metal vertical field effect transistor.a)

We conducted local anodic oxidation (LAO) lithography in single-layer, bilayer, and multilayer graphene using tapping-mode atomic force microscope. The width of insulating oxidized area depends systematically on the number of graphene layers. An 800-nm-wide bar-shaped device fabricated in single-layer graphene exhibits the half-integer quantum Hall effect. We also fabricated a 55-nm-wide graphene nanoribbon (GNR). The conductance of the GNR at the charge neutrality point was suppressed at low temperature, which suggests the opening of an energy gap due to lateral confinement of charge carriers. These results show that LAO lithography is an effective technique for the fabrication of graphene nanodevices. PACS numbers:Graphene, a single atomic layer of graphite, has a unique band structure [1,2] and exceptionally high carrier mobility [3]. Therefore, it has been used to develop carbon-based electronic devices [4]. To date, graphene nanodevices such as quantum dots [5,6], AharonovBohm rings [7], and nanoribbons [8,9] have been fabricated by conventional electron-beam lithography combined with plasma etching. However, plasma etching introduces defects in graphene [3,7,10], which causes localization of charge carriers [7,10]. Further, this technique cannot be used to control the edge structure of graphene, which is expected to have significant effects on its electronic properties [8]. Therefore, in order to fabricate high-quality devices, we need a new lithography technique that will allow us to perform high-resolution patterning without damaging the graphene layer.Local anodic oxidation (LAO) lithography using atomic force microscope (AFM) is a promising technique for the fabrication of graphene nanodevices. This is because LAO lithography has been successfully used for fabricating nanodevices based on semiconductors [11,12,13,14,15]. The confinement of charge carriers obtained by LAO is highly specular [16] than that obtained by plasma etching [17], i.e. the charge carriers conserve their momentum along the normal to the confinement. Recently, Weng et al. produced insulating trenches in graphene flakes with a thickness of 1-2 atomic layers using tapping-mode AFM [18]. However, in their experiment, the number of graphene layers was not determined, though the LAO conditions such as the width of oxidized area are expected to be totally different for single-layer, bilayer, and multilayer graphene. Geisbers et al. used contact-mode AFM to produce an insulating trench on single-layer graphene [19]. However, * Electronic address: msatoru@iis.u-tokyo.ac.jp † Electronic address: tmachida@iis.u-tokyo.ac.jp the contact-mode AFM cantilever can damage the fabricated device [20]. Further, the transport phenomena of Dirac fermions were not demonstrated clearly in either experiment above [18,19]. In this letter, we describe LAO lithography experiments that were conducted in single-layer, bilayer, and multilayer graphene using tapping-mode AFM. We show that the width of oxidized area depends on the number of graphene layers. We hav...

Coherent control of local nuclear spins in a solid-state device is demonstrated. By unequally populating spin-resolved quantum-Hall edge channels, nuclear spins in a limited region along the edge channels are strongly polarized via the hyperfine interaction. Pulsed rf magnetic fields, generated by a built-in micrometal strip, cause the nuclear-spin state to evolve coherently. The nuclear-spin state reached during the pulse duration is finally read out via the edge-channel conductance, which shows Rabi oscillation.

We have fabricated a lateral double barrier magnetic tunnel junction (MTJ) which consists of a single self-assembled InAs quantum dot (QD) with ferromagnetic Co leads. The MTJ shows clear hysteretic tunnel magnetoresistance (TMR) effect, which is evidence for spin transport through a single semiconductor QD. The TMR ratio and the curve shapes are varied by changing the gate voltage. PACS numbers:The research field of semiconductor-based spin electronics (spintronics) has opened up a new technology for spin manipulation by means other than magnetic field.[1, 2] For developing semiconductor nanospintronic applications and discovering novel physical phenomena, one is extremely interested in technological possibilities for spin injection into a single semiconductor quantum dot (QD) which behaves as an artificial atom.[3] To date, many theoretical studies of spin transport through a single nonmagnetic island with ferromagnetic leads have been reported, [4,5,6,7,8,9] and spin accumulation in the island was predicted in their reports. Very recently, for metallic systems, spin injection into a single nonmagnetic nanoparticle was achieved,[10] which indicates the occurrence of spin accumulation. For an individual carbon nanotube (CNT) with ferromagnetic leads, the spin transport [11,12] and its gate-control [13,14,15,16] have also been demonstrated, showing possible spintronic applications using CNTs. However, no experimental work on spin-dependent transport through a single semiconductor QD has been reported yet.Recently, Jung et al. [17] succeeded in transport measurements for a single self-assembled InAs QD in contact with nonmagnetic leads and clearly observed shell structures due to an artificial atomic nature. Replacing the nonmagnetic leads with ferromagnetic ones, we

We demonstrate electrical spin injection from a ferromagnet to a bilayer graphene