Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userSpin-Singlet and Spin-Triplet Josephson Junctions for Cryogenic Memory
104
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Due to the ever increasing power and cooling requirements of large-scale computing and data facilities, there is a worldwide search for low-power alternatives to CMOS. One approach under consideration is superconducting computing based on single-flux-quantum logic. Unfortunately, there is not yet a low-power, high-density superconducting memory technology that is fully compatible with superconducting logic. We are working toward developing cryogenic memory based on Josephson junctions that contain two or more ferromagnetic (F) layers. Such junctions have been demonstrated to be programmable by changing the relative direction of the F layer magnetizations. There are at least two different types of such junctions -- those that carry the innate spin-singlet supercurrent associated with the conventional superconducting electrodes, and those that convert spin-singlet to spin-triplet supercurrent in the middle of the device. In this paper we compare the performance and requirements of the two kinds of junctions. Whereas the spin-singlet junctions need only two ferromagnetic layers to function, the spin-triplet junctions require at least three. In the devices demonstrated to date, the spin-singlet junctions have considerably larger critical current densities than the spin-triplet devices. On the other hand, the spin-triplet devices have less stringent constraints on the thicknesses of the F layers, which might be beneficial in large-scale manufacturing.
rate research
Read More
Josephson junctions containing two ferromagnetic layers are being considered for use in cryogenic memory. Our group recently demonstrated that the ground-state phase difference across such a junction with carefully chosen layer thicknesses could be controllably toggled between zero and $pi$ by switching the relative magnetization directions of the two layers between the antiparallel and parallel configurations. However, several technological issues must be addressed before those junctions can be used in a large-scale memory. Many of these issues can be more easily studied in single junctions, rather than in the Superconducting QUantum Interference Device (SQUID) used for the phase-sensitive measurements. In this work, we report a comprehensive study of spin-valve junctions containing a Ni layer with a fixed thickness of 2.0 nm, and a NiFe layer of thickness varying between 1.1 and 1.8 nm in steps of 0.1 nm. We extract the field shift of the Fraunhofer patterns and the critical currents of the junctions in the parallel and antiparallel magnetic states, as well as the switching fields of both magnetic layers. We also report a partial study of similar junctions containing a slightly thinner Ni layer of 1.6 nm and the same range of NiFe thicknesses. These results represent the first step toward mapping out a ``phase diagram for phase-controllable spin-valve Josephson junctions as a function of the two magnetic layer thicknesses.
We study the Andreev bound states in a Josephson junction between a singlet and a triplet superconductors. Because of the mismatch in the spin symmetries of pairing, the energies of the spin up and down quasiparticles are generally different. This results in imbalance of spin populations and net spin accumulation at the junction in equilibrium. This effect can be detected using probes of local magnetic field, such as the scanning SQUID, Hall, and Kerr probes. It may help to identify potential triplet pairing in $rm(TMTSF)_2X$, $rm Sr_2RuO_4$, and oxypnictides.
The increased capabilities of coupling more and more materials through functional interfaces are paving the way to a series of exciting experiments and extremely advanced devices. Here we focus on the capability of magnetically inhomogeneous superconductor/ferromagnet (S/F) interfaces to generate spin-polarized triplet pairs. We use the power of the Josephson effect for a quantitatively accurate proof of the coexistence and tunability of singlet and triplet transport in ferromagnetic spin filter junctions. We build on previous achievements and find unique correspondence between neat experimental benchmarks in the temperature behavior of the critical current and theoretical modeling based on microscopic calculations. This turns to be a unique opportunity to model disorder and spin-mixing effects in a Josephson junction (JJ) to enlarge the space of parameters, which regulate the phenomenology of the Josephson effect and could be applied to a variety of novel types of JJs.
We demonstrate a Josephson junction with a weak link containing two ferromagnets, with perpendicular magnetic anisotropy and independent switching fields in which the critical current can be set by the mutual orientation of the two layers. Such pseudospin-valve Josephson junctions are a candidate cryogenic memory in an all superconducting computational scheme. Here, we use Pt/Co/Pt/CoB/Pt as the weak link of the junction with $d_text{Co} = 0.6$ nm, $d_text{CoB} = 0.3$ nm, and $d_text{Pt} = 5$ nm and obtain a $60%$ change in the critical current for the two magnetization configurations of the pseudospin-valve. Ferromagnets with perpendicular magnetic anisotropy have advantages over magnetization in-plane systems which have been exclusively considered to this point, as in principle the magnetization and magnetic switching of layers in the junction should not affect the in-plane magnetic flux.
Josephson junctions containing ferromagnetic materials have attracted intense interest both because of their unusual physical properties and because they have potential application for cryogenic memory. There are two ways to store information in such a junction: either in the amplitude of the critical current or in the ground-state phase difference across the junction; the latter is the topic of this paper. We have recently demonstrated two different ways to achieve phase control in such junctions: the first uses junctions containing two magnetic layers in a pseudo spin valve configuration, while the second uses junctions containing three magnetic layers with non-collinear magnetizations. The demonstration devices, however, have not yet been optimized for use in a large-scale cryogenic memory array. In this paper we outline some of the issues that must be considered to perform such an optimization, and we provide a speculative phase-diagram for the nickel-permalloy spin-valve system showing which combinations of ferromagnetic layer thicknesses should produce useful devices.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
