Picosecond magnetization dynamics of nanomagnets: time resolved XMCD and XPEEM

Nanomagnets are the building blocks for future data storage technology and spintronic devices that have applications in telecommunications technology and quantum computation. By far the most successful spintronic devices today are the spin and tunnel valve sensors used in hard disk recording heads. The valve consists of two ferromagnetic layers, the fixed layer and the free layer , separated by either a non-magnetic metal or an ultrathin oxide tunnel barrier. When a magnetic field is applied to the device, the free layer magnetization reorients from anti-parallel to parallel to the fixed layer magnetization, giving rise to a magnetoresistance (MR) that provides the signal from the sensor. Arrays of valves with bi-stable magnetic states provide the basis for a magnetic random access memory (MRAM). It was recently observed that spin-valves of nano-scale lateral size also exhibit spin-transfer-torque (STT) effects when current is passed perpendicular to the plane of the device. Due to conservation of angular momentum, injection of spin-polarized current into a ferromagnetic layer can produce a torque upon the local magnetization that generates a response similar to that induced by a pulsed magnetic field. Magnetic switching due to STT is proposed for 2nd generation MRAM and is thought to be essential for scaling to smaller cell sizes. STT may also be used to maintain a steady precession of the free layer in a valve leading to microwave generation. The spin-valve already lies well within the realm of nanotechnology with layer thicknesses of just a few nanometers and lateral sizes as small as 50 nm in research level recording heads. However, smaller feature sizes and hence increased storage capacity are of limited use without a commensurate reduction in the time taken to read and write each bit of information. The dynamic response of these nanoscale valve devices is the subject of this proposal. In Exeter we have ten years experience of using femtosecond lasers to stimulate and detect sub-nanosecond magnetic processes. We have developed a time resolved scanning Kerr microscope (TRSKM) that allows the response of a sample to a pulsed magnetic field to be imaged stroboscopically. Recently, in collaboration with Hitachi Global Storage Technologies, we have studied the picosecond dynamics of arrays of nano-scale elements of free layer material. We have inferred that, for element sizes less than 200 nm, the response of the magnetization to a pulsed field is spatially non-uniform and dominated by spin wave modes localised at the element's edges. This non-uniformity may result in a degradation of the signal to noise ratio in future recording sensors. Indeed we believe that STT devices will be similarly blighted. We now propose to use time resolved X-ray measurement techniques to obtain unique information about the picosecond magnetization dynamics within nanoscale magnets and specifically spin-valves. We will use time resolved x-ray magnetic circular dichroism (TRXMCD) to make element specific measurements of precessional dynamics within spin-valve structures, while time resolved x-ray photoemission electron microscopy (TRXPEEM) will be used to obtain time resolved images of the magnetic state. We will hence confirm that edge modes dominate the magnetic response of elements within the deep nanoscale regime, and test strategies for the selective excitation of a more spatially uniform response. The development of time resolved magnetic measurements at UK synchrotron sources will open this rather new field to a large community of UK magnetism researchers, while the project will benefit from our longstanding experience of time resolved measurement techniques. Finally, we believe that this project presents an excellent training opportunity for a PhD student who will be integral to the development of measurement techniques that are likely to find much wider application in years to come.