Polarity of Light Contributes to Switching of Magnetic Memory Bits
by Tim Palucka
Materials Research Society | Published: 18 March 2014
As the magnetic bits for storing data in memory get smaller and smaller, it is becoming more difficult for applied magnetic fields by themselves to switch a bit from a digital 0 to a digital 1 state in a stable manner. Tiny magnetic bits can be randomly switched by thermal fluctuations, wiping out recorded data. By adding a laser to the magnetic writing process to briefly heat the bit and apply a magnetic field before cooling it down again, greater stability against these thermal fluctuations can be gained. Now, researchers have demonstrated that there's more in the laser beam than just heat-the helicity, or polarization, of the light can contribute to the switching of a bit's magnetic state. What's more, this effect can be seen in rare-earth-free Co-Ir synthetic ferrimagnets (SFIs), possibly eliminating the need for expensive rare earth elements altogether.
"The hard drive industry is running into this problem that you can make something you can write but it won't store the data, or you can make something that is able to store the data but you can't write it," says Eric Fullerton of the University of California San Diego. "It's known as the 'superparamagnetic effect.' This has been a huge problem. People have been working for the last 20 years to address this problem."
Fullerton's and Fainman's groups at UCSD, in cooperation with investigators from the Université de Lorrraine and the University of Kaiserslautern in Germany, experimented with 400 to 500 hundred different materials, including rare earth-transition metal (RE-TM) alloys, multilayers, and heterostructures, in addition to the rare-earth-free SFIs. They observed the phenomenon of all-optical helicity-dependent switching (AO-HDS)-magnetic switching in the absence of an applied magnetic field-in all four of these categories.
Since the discovery of AO-HDS in Gd-Fe-Co, an RE-TM material, in 2007, which Fullerton calls "a really amazing discovery," most of the work in the field has been on Gd-Fe-Co. "So the question was always 'is AO-HDS very specific to this material, or is this a more general phenomena, and can you engineer materials that would show it more broadly?'" Fullerton says.
As reported in a recent issue of Nature Materials , the researchers took a comprehensive approach to the investigation, working with (1) amorphous RE-TM alloys, (2) RE-TM multilayers, (3) RE-TM multilayer heterostructures, and (4) SFI heterostructures to investigate both AO-HDS and thermal demagnetization phenomena.
In the first category, they experimented with four amorphous, ferrimagnetic RE-TM alloys, with RE = Gd, Tb, Dy, and Ho, and TM = Fe, Co or Fe-Co alloys. When these materials were placed between 4-nm-thick layers of Ta and irradiated with a laser beam having right circularly polarized, left circularly polarized, and linearly polarized light, AO-HDS was seen in samples with RE concentrations around 25%. That is, the magnetization state was seen to switch based only on the thermal energy and polarity of the laser beam. In category 2, alternating RE/TM layers, such as Gd/Co, Tb/Co, and Ho/CoFe, with a total thickness of 25 nm between the Ta films, also showed AO-HDS over a narrow range of compositions. To add further complexity, in category 3 exchanged-coupled stacks composed of two types of multilayers, such as [Tb(0.5 nm)/Co (0.45 nm)] N and [Tb(0.35 nm)/Co(0.7 nm)] 25- N , where N is the number of repeats in the 25-nm thick overall system, were formed. In this case, AO-HDS was seen in cases where N = 25 to 15, with thermal degradation occurring for N < 15.
The non-rare-earth SFIs were made by of two TM layers antiferromagnetically coupled through 0.4 nm Ir interlayers. SFIs with a magnetization compensation temperature ( TMcomp ) greater than room temperature exhibited AO-HDS behavior. Indeed, magnetization compensation appears to be essential for AO-HDS.
The researchers concluded that for AO-HDS to be seen, the materials must (1) have structures that exhibit perpendicular magnetic anisotropy, (2) have two distinct magnetic sublattices that are antiferromagnetically coupled, and (3) have magnetic sublattices with two different temperature dependencies such that compensation of the average magnetization is present near or above room temperature.
Fullerton was careful to note that the micron-scale laser spots and the femto-second duration of the laser bursts used in these experiments must be significantly improved for any possible practical application of the findings of this research. They are now focusing on pushing the technology into the nanoscale and improving the time scale, as well as investigating other materials.
In addition to expanding the range of materials that are subject to AO-HDS, this new research, allows for the development of new theories to explain AO-HDS. "Up until now, theories were very specific to the physics of rare earth ions or atoms," Fullerton says. "If you can show it occurring without a rare earth, then if you have a theory that explicitly depends on the properties of rare earths, it makes you tend to re-think it."
As a start toward a better theoretical understanding, Fullerton and his colleagues are revisiting earlier models based on the transfer of both angular momentum and heat from the light beam to the sample. The heat increases the temperature of the sample so that only a minimal amount of angular momentum from the polarization of the laser beam is needed to induce magnetization switching.
"The hard drive industry is working hard to incorporate nanolasers into their drives to heat up the media and record information," says Fullerton, who worked for ten years in the industry before joining academia. "Now if you could directly record the data using just the laser, or the laser could provide an additional component besides just heating the magnetic bit, this might greatly move this technology along. But there are many, many barriers to overcome, and we've just shown we're one step-a little bit-closer, I would say."
Read the abstract in Nature Materials here.