News and posts

Time: Thursday, October 25th, 11:40
Speaker: Lucian PREJBEANU, INAC

Spin transfer torque MRAM (STT-MRAM) are considered as one of the most promising technology for non-volatile RAM due to their non-volatility, quasi-infinite endurance, speed and high-density. In standard STT-MRAM, particularly the in-plane magnetized ones, the switching dynamics of the storage layer magnetization is characterized by an incubation delay that can take up to a few nanoseconds. This is because the spin transfer torque is initially zero at the onset of the write current, since at equilibrium the storage layer magnetization and spin-current polarization are parallel or antiparallel. The magnetization reversal is actually triggered by thermal fluctuations which makes the switching stochastic. Consequently, it is necessary to increase the write pulse duration and/or its amplitude to reach sufficiently low write error rates (for instance lower than 10-11). This is detrimental for the realization of short access time memory such as SRAM-like used for Cache applications, which means that addressing fast switching memories, as SRAM replacement in cache or beyond, up to the processor, it is mandatory to deploy new ultrafast writing strategies. Therefore, we present in this study three MRAM writing strategies allowing to greatly improve the write speed down to the sub-nanosecond range.
First, we demonstrate that sub-ns switching with final state determined by the current polarity through the stack can be achieved in STT-MRAM cells comprising two spin-polarizing layers having orthogonal magnetic anisotropies. Two solutions were found and demonstrated : one consists in increasing the cell aspect ratio, the other consists in applying a static transverse field on the cells, the second one being preferred since it does not require to increase the footprint of the cell.
Second, we have shown that the write stochasticity can be almost completely suppressed and the writing speed greatly increased by inducing an oblique anisotropy (also called easy-cone anisotropy) in the storage layer [7]. This means that, at equilibrium, the storage layer magnetization, instead of being aligned along the normal to the plane of the layer, lies along any direction on a cone of axis normal to the plane. In contrast, the reference layer is designed to keep its perpendicular anisotropy. Thanks to this easy cone anisotropy, the storage and reference layer magnetizations always keep a relative angle so that upon write, the storage layer magnetization reversal can be triggered at the very onset of the write current pulse. This quadratic anisotropy itself results from spatial fluctuations of uniaxial anisotropy which can be induced during deposition and annealing of the magnetic tunnel junction stacks. Thanks to this easy cone anisotropy, the writing is much more reproducible and can be faster and/or realized at lower write voltage thereby reducing the write energy consumption).
Third, we recently demonstrated the first proof of concept of a perpendicular Spin orbit torque (SOT)-MRAM. SOT-RAM combining high speed, non-volatility and potentially infinite endurance while being compatible with technological nodes below 22nm and as such appears as a strong candidate for future non-volatile cache memory. In SOT-MRAM, whose write can be achieved with sub-ns pulses, we have shown that the inclusion of SOT-MRAM at the L1-instruction cache and L2-cache can reduce the energy consumption of processors by 60% while solving radiation induced soft errors of SRAM-only configuration.

Time: Thursday, October 25th, 11:00
Speaker: Wangxiang FENG, Beijing

Magneto-optical (MO) effects have been known for more than a century as they reflect the basic interactions between light and magnetism.  The origin of MO effects is normally believed by the simultaneous presence of band exchange splitting and spin-orbit coupling.  Using a tight-binding model and first-principles calculations, we show that topological MO effects, in analogy to the topological Hall effect, can arise in noncoplanar antiferromagnets caused entirely by scalar spin chirality instead of spin-orbit coupling.  The band exchange splitting is not indispensable to topological MO effects.  Moreover, the Kerr and Faraday rotation angles in two-dimensional or layered noncoplanar antiferromagnets are found to be quantized in the low-frequency limit, implying the appearance of quantum topological MO effects, which can be measured by time-domain THz spectroscopy.

Time: Thursday, October 25th, 10:10
Speaker: Kamil OLEJNIK, Prague

The electrical control of magnetic moments of antiferromagnets (AFMs) using Néel-ordered current induced spin-orbit fields [1] opened possibility to use AFMs for practical memory applications [2]. Compared to ferromagnets (FMs) the AFMs theoretically promise up to three orders of magnitude faster magnetization dynamics.
Experimental results investigating the functionality of AFM memory cells fabricated from epitaxial CuMnAs films will be presented. First we investigate the electrical writing with pulse lengths ranging from milliseconds to hundreds of picoseconds [3]. Since shorter pulses cannot be applied using standard electrical circuitry, we use THz radiation pulses to investigate the writing in THz range. With the THz radiation pulses we observe analogous memory functionality demonstrating that the AFM memory cells can be written using picosecond electrical pulses [4].

 

[1] J. Železný et al. Phys. Rev. Lett. 113 (2014).
[2] P. Wadley et al., Science 351, 587–590 (2016).
[3] K. Olejník et al., Nat. Commun. 8, 15434 (2017).
[4] K. Olejník et al., Sci. Adv. 4, eaar3566 (2018).

Time: Thursday, October 25th, 9:00
Speaker: Davide BOSSINI, Dortmund

Magnetism in solid state materials is one of the most widely investigated phenomena in condensed matter physics. The conventional description of a magnetic material is formulated in the framework of thermodynamics, since it relies on the concept of equilibrium. While this approach is effective for the ground state properties, its application to the dynamical regime is limited to to spin dynamics with characteristic timescales in which the adiabatic approximation can still be invoked. The technical progresses of pulsed laser sources have provided the possibility to generate intense laser pulses with duration in the 10-100 femtosecond range. Such laser pulses are among the shortest stimuli in contemporary solid state physics. They provide the groundbreaking possibility to drive and detect spin dynamics in magnetic materials in real-time experiments, whose time-resolution is comparable to or even shorter than the two main magnetic interactions, i.e. the spin-orbit coupling and the exchange interaction. Note that aside from the clear academic interest, investigating the optical control of spins on ever-shorter timescales may have relevant implications for possible future developments of the magnetic recording industry. In this talk I will present the basic concepts, methods and goals of the field called “ultrafast magnetism”[1]. In particular, I aim at demonstrating the potentiality and wide applicability of the optical methods, by describing the major breakthroughs reported in this research area. Spectacular phenomena have already been observed, such as the ultrafast demagnetisation[2], the picosecond-deterministic reversal of the magnetisation[3], the coherent control collective spin excitations[4] and even the photo-induced magnetic phase transitions on the picosecond timescale[5]. In the last part of my talk, I plan to discuss the most recent trends[6,7] and some possible future directions.

[1] A. Kirilyuk et al. Rev. Mod. Phys. 82, 2731 (2010)
[2] E. Beaurepaire et al. PRL 76, 4250 (1996)
[3] C. Stanciu et al. PRL 99, 047601 (2007)
[4] A.V. Kimel et al Nature 435, 655 (2005)
[5] D. Afanasiev et al. PRL 116, 097401 (2016)
[6] D. Bossini et al. Nat. Comm. 7, 10645 (2016)
[7] D. Bossini et a. Nat. Phys. 14, 370 (2018)

Time: Wednesday, October 24th, 17:50
Speaker: Jeffrey BOKOR, Berkeley

Single‐shot helicity independent all optical switching (AOS) of GdFeCo ferrimagnetic alloys has received considerable attention since it was first reported in 2011 [1] and has been understood as arising from an ultrafast thermal effect [2]. We have demonstrated that ultrafast heating of GdFeCo may be achieved by electrical charge current pulses from an electronic circuit [3]. We observe deterministic, repeatable, and reversible ultrafast switching of the magnetization of GdFeCo with a single sub‐10 picosecond electrical current pulse. The magnetization reverses in ~10 ps, which is more than one order of magnitude faster than any other electrically controlled magnetic switching, and demonstrates a fundamentally new switching mechanism without the need for spin polarized currents or spin transfer/orbit torques. The original observation of all‐optical switching was limited to GdFeCo [1, 2]. In 2014 it was discovered that all‐optical switching could be obtained in a range of ferromagnetic films [3]. However, it has now been understood that, in the ferromagnets that have been tested to date, all‐optical switching is a multi‐shot process [5]. This means that for full magnetization reversal to occur, multiple laser pulses are needed, which limits the operation speed. We have experimentally demonstrated a new family of ferromagnets grown on GdFeCo which present single‐shot all optical switching [6]. These coupled bilayer stacks may offer distinct advantages of ferromagnets over pure ferrimagnets for applications, including ultrafast electric current switching of ferromagnets, thereby enabling a high magnetoresistance electrical readout. We are also studying scaling of the required switching current with size of the magnetic element down to 50 nm dots and smaller. Results on switching of such nanoscale magnetic dots in GdFeCo as well as a number of new ferrimagnetic materials will be presented. Our long‐term goal is to realize fully integrated devices suitable for on‐chip magnetic memory (and perhaps even logic) that can be switched with current delivered by an on‐chip CMOS drive circuit (no laser involved!) with switching speed in the range of picoseconds, i.e. up to two orders of magnitude faster than present spin‐torque based spintronics.

 

[1] I. Radu, et al., Nature 472, 205 (2011).
[2] T. A. Ostler, et al., Nat Commun 3, 666 (2012).
[3] Y. Yang, et al., Science Advances 3, E1603117 (2017).
[4] C. H. Lambert, et al., Science 345, 1337 (2014).
[5] M. S. El Hadri, Phys. Rev. B 94, 064412 (2016).
[6] J. Gorchon, et al., Appl. Phys. Lett. 111 (2017).

Time: Wednesday, October 24th, 17:10
Speaker: Martijn HECK, Aarhus

Photonics has been the technology of choice for long-distance telecommunications for many decades. The key enablers here are the vast bandwidth of optical fibers and their low loss of, currently, below 0.2 dB/km. The loss of an optical fiber or waveguide is independent of the bandwidth of the optical signal. This is unlike electrical cables, like copper, where the losses tend to increase rapidly with increasing bandwidth, to multiple dBs per meter at 25‑GHz bandwidths. This means that there is a clear trade-off for interconnects: the higher the bandwidth, the shorter the length becomes where optical communications is the preferable solution.
Following this trend, in high-end datacenters, vertical-cavity-laser-based multimode fiber interconnects connect the servers, with silicon-photonic-based single-mode fiber interconnects as the highest bandwidth solutions, operating at 100/200 Gbps [1]. For on-board interconnects, i.e., between processors, memory and the edge of the server, new standards are currently being drawn up for so-called optical engines, making this a near-term reality [2]. And finally, pioneering work has shown that the optics can even be brought onto the processor, into an existing CMOS process, for the highest-bandwidth communications [3]. This technology is currently being commercialized [4].
Looking even further, optical networks-on-chip are being considered, albeit mostly theoretical. Initial design studies have shown the validity of replacing copper interconnects on the processor with an optical network, e.g., to connect cores in a multi-core processor [5,6]. Such approaches are based on energy-efficient silicon photonics, promising attojoule-per-bit communications [7].
Leveraging these technology developments and trends, in the H2020-funded project SPICE, we intend to push the boundaries even further, by trying to address single memory elements optically, e.g., to write magnetic tunnel junctions that have magneto-optic layers [8]. This eventually requires ultra-dense and ultra-low power networks, for overall memory energy consumption in the sub-20-fJ/bit range. I will present the opportunities for this technology.

This project has received funding from the European Union’s Horizon 2020 research
and innovation programme under grant agreement No 713481.

1. Mekis et al., "A grating-coupler-enabled CMOS photonics platform." IEEE Journal of Selected Topics in Quantum Electronics 17, no. 3 (2011): 597-608.
2. https://onboardoptics.org/, checked d.d. 11-10-2018
3. Sun et al., "Single-chip microprocessor that communicates directly using light." Nature 528, no. 7583 (2015): 534.
4. https://ayarlabs.com/, checked d.d. 11-10-2018
5. Batten et al., "Building many-core processor-to-DRAM networks with monolithic CMOS silicon photonics." IEEE Micro 29, no. 4 (2009).
6. J. R. Heck, and J. E. Bowers, "Energy efficient and energy proportional optical interconnects for multi-core processors: Driving the need for on-chip sources." IEEE Journal of Selected Topics in Quantum Electronics 20, no. 4 (2014): 332-343.
7. A. B. Miller, "Attojoule optoelectronics for low-energy information processing and communications." Journal of Lightwave Technology 35, no. 3 (2017): 346-396.
8. http://spice-fetopen.eu/, checked d.d. 11-10-2018

Time: Wednesday, October 24th, 16:30
Speaker: Dries VAN THOURHOUT, Ghent

Silicon Photonics uses existing CMOS technology to realise complex photonic integrated chips that allow to route, control, switch and filter light on a chip.  Silicon waveguides strongly confine light allowing to make these chips very compact and enhance light-matter interaction.  In this talk I will introduce the status of the field and the operation of most relevant basic devices (sources, modulators, detectors, passive guiding and routing of signals…).  Then I will show how these building blocks can be combined to build more complex chips that can route and switch optical pulses, as would be needed for optically switching magnetic tunnel junctions.

Time: Wednesday, October 24th, 15:40
Speaker: Mo LI, Minnesota

All-optical switching (AOS) has been extensively explored for its prospects of enabling ultrafast magnetic recording and operation of spintronic devices. Nevertheless, there has not been any demonstration of AOS in realistic spintronic devices, such as magnetic tunnel junctions (MTJs). Since previous studies of AOS have only been performed on a single magnetic layer, how additional magnetic layers affect the AOS phenomenon remains unknown. In this work, we develop a perpendicularly magnetized MTJ (p-MTJ) employing GdFeCo as the free layer. We demonstrate, for the first time, all-optical switching of an MTJ, without using any external magnetic field, but rather with single sub-picosecond infrared laser pulses. The switching is read out electrically through measuring the tunneling magnetoresistance (TMR). We further demonstrate MHz repetition rate of switching GdFeCo film, which is limited by our instruments, but the fundamental upper limit of this rate should be higher than tens of GHz as has been revealed by previous time-resolved studies28. The demonstrated picosecond switching time of an MTJ by AOS is two orders of magnitude faster than that of any other switching methods.

Time: Wednesday, October 24th, 14:30
Speaker: Bernard DIENY, INAC

Heat assisted magnetic recording (HAMR)  uses temporary near field heating of the media during write to increase hard disk drive storage density. By using a plasmonic antenna embedded in the write head, an extremely high thermal gradient iscreated in the recording media (up to 10K/nm). State of the art HAMR media consists of grains of L1₀-FePt exhibiting high perpendicular anisotropy separated by 1 to 2 nm thick
carbon segregant. Next to the plasmonic antenna, the difference of temperature between two nanosized FePt grains in the media can reach 80K across the 2 nm thick grain boundary. This represents a gigantic local thermal gradient of 40K/nm across a carbon tunnel barrier. In the field of spincaloritronics, much weaker thermal gradients of ~1K/nm were shown to cause a thermal spin-transfer torque capable of inducing magnetization switching in magnetic tunnel junctions. Considering on one hand that two neighboring grains separated by an insulating grain boundary in a HAMR media can be viewed as a magnetic tunnel junction, and on the other hand, that the thermal gradients in HAMR are one to two orders of magnitude larger than those used in conventional spincaloritronic experiments, one may expect a strong impact from these thermal spin-transfer torques on magnetization switching dynamics in HAMR recording. This issue has been totally overlooked in earlier development of HAMR technology. It is shown that the thermal in-plane torque can have a detrimental impact on the recording performances by favoring antiparallel magnetic alignment between neighboring grains during the media cooling. Implications on media design are discussed in order to overcome the influence of these thermal torques. Suggestions of spincaloritronic experiments taking advantage of these huge thermal gradients produced by plasmonic antenna are also given.

Time: Wednesday, October 24th, 12:20
Speaker: Enrique DEL BARCO, Florida

Emerging phenomena, such as the spin-Hall effect (SHE), spin pumping, and spin-transfer torque (STT), allow for interconversion between charge and spin currents and the generation of magnetization dynamics that could potentially lead to faster, denser, and more energy efficient, non-volatile memory and logic devices. Present STT-based devices rely on ferromagnetic (FM) materials as their active constituents. However, the flexibility offered by the intrinsic net magnetization and anisotropy for detecting and manipulating the magnetic state of ferromagnets also translates into limitations in terms of density (e.g., because neighboring units can couple through stray fields) and speed (frequencies are limited to the GHz range). A new direction in the field of spintronics is to employ antiferromagnetic (AF) materials, particularly antiferromagnetic insulators. In contrast to ferromagnets, where magnetic anisotropy dominates spin dynamics, in antiferromagnets spin dynamics are governed by the interatomic exchange interaction energies which are orders of magnitude larger than the magnetic anisotropy energy, leading to the potential for ultrafast information processing and communication in the THz frequency range. We will present studies of spin pumping at Manganese Difluoride(MnF₂) / platinum (Pt) interfaces at temperatures below the MnF₂Néel temperature (T_N =  67.34K). In particular, measurements of the inverse spin Hall effect (ISHE) voltage arising from the interconversion of the dynamically injected spin currents into Pt will be reported. We observe a clear electrostatic potential signal coinciding with the MnF²spin flip-flop transition (H_SF= 9T). The signal reverses switching the polarity of the magnetic field, and displays a marked dependence on the power of the microwave stimuli, as expected from the ISHE.