Gate-tunable black phosphorus spin valve with nanosecond spin lifetimes

Abstract

Two-dimensional materials offer new opportunities for both fundamental science and technological applications, by exploiting the electron's spin 1 . While graphene is very promising for spin communication due to its extraordinary electron mobility, the lack of a band gap restricts its prospects for semiconducting spin devices such as spin diodes and bipolar spin transistors 2 . The recent emergence of 2D semiconductors could help overcome this basic challenge. In this letter we report the first important step towards making 2D semiconductor spin devices. We have fabricated a spin valve based on ultra-thin (~ 5 nm) semiconducting black phosphorus (bP), and established fundamental spin properties of this spin channel material which supports all electrical spin injection, transport, precession and detection up to room temperature (RT). Inserting a few layers of boron nitride between the ferromagnetic electrodes and bP alleviates the notorious conductivity mismatch problem and allows efficient electrical spin injection into an n-type bP. In the non-local spin valve geometry we measure Hanle spin precession and observe spin relaxation times as high as 4 ns, with spin relaxation lengths exceeding 6 µm. Our experimental results are in a very good agreement with first-principles calculations and demonstrate that Elliott-Yafet spin relaxation mechanism is dominant. We also demonstrate that spin transport in ultra-thin bP depends strongly on the charge carrier concentration, and can be manipulated by the electric field effect. 2 Electron spin is an important degree of freedom which can complement or even replace charge in information storage and logic devices 3 . For spin-based electronics, it is essential to have materials with long spin relaxation times at RT 1 . With respect to the material selection, semiconductors in particular offer new opportunities that are unfeasible in metal-based spintronics devices. These include doping by the electric field effect and gate controlled amplification/switching actions 4 . The boom of semiconductor spintronics started with the demonstration of the electrical spin injection into GaAs by R. Fiederling et al., also later by H.Ohno et al. 5,6 . I. Appelbaum and his colleagues further fostered this by adding silicon into the spintronics materials family 7 . More recently 2D materials such as graphene have captured the interest of engineers and scientists on the grounds of their high electronic mobility and the simultaneous ability to tune their charge carrier concentrations by the electric field effect 8 .Graphene has already been investigated very extensively in the spintronics community since the first unequivocal demonstration of RT spin injection by N. Tombros et al., 1,9 . While the first devices showed spin lifetimes of only 0.1 ns 9 , the new generation of graphene devices that have surfaces within the last year show remarkable spin lifetimes of ~ 10 ns, making graphene suitable for spin communication channels 10 .Conversely, the zero-band gap nature of graphene does not al...

Measuring anisotropic spin relaxation in graphene

“…This tool can also be used to probe the spin relaxation in similar 2D materials that have recently started to attract interest like black phosphorus [42]. The data collected from the xHanle experiment pointed to a nontrivial magnetization of the contacts which is in line with the other experiments we performed.…”

Spin-orbit coupling in elemental two-dimensional materials

“…26 The proportionality factor, b 2 , is the Elliott-Yafet spin mixing (or spin admixture) parameter. It has been extensively studied for bulk materials and thin films of heavy elements [32][33][34][35][36][37][38] , but the knowledge about b 2 in atomically thin 2D systems is very limited 39,40 .…”

Stabilizing ultra-thin black phosphorus with in-situ-grown 1 nm-Al2O3 barrier

“…12,[22][23][24][25][26][27][28][29][30][31] Some attempts to reduce the thickness of the capping A 2 O 3 have also been reported. 32 Thin capping may simultaneously passivate and serve as a tunnel barrier (which may even lead to improved electrical contacts 33,34 ) representing a double advantage over thick capping in which electrical contacts must be patterned before passivation (ex-situ). 20 Finally, this leads us to an important limitation in many of the reported procedures, which lies in their ex-situ character: the time exposure to air, even if very short, of the samples during the fabrication process 23 is critical for the degradation of the thinnest few-layered BP flakes.…”