Office Location & Contact
Micromagnetics Research Lab
Ferromagnetic materials (like iron) can have a spontaneous magnetization even in the absence of any externally applied magnetic field. This property is useful for a wide variety of applications, but it is particularly useful in the area of data storage because the magnetization has two values for any given field. (This behavior is demonstrated in the hysteresis loop shown to the left).
Since the first magnetic hard drives were built the areal density of magnetic bits has been doubling at a rate of about every other year. This is accomplished primarily by reducing the amount of magnetic material necessary to store information. This process can not continue indefinitely, therefore other data storage techniques are necessary.
Nanowires made out of ferromagnetic materials are a potential solution. The wire geometry is advantageous because a wire is not necessarily confined to the surface of a disk; spatial constraints are no longer an issue.
Additionally, we have learned that it is possible to move information rapidly along the nanowire with weak magnetic fields; leading to devices that could operate significantly faster than currently available.
Here in the Physics Department at Marquette University my undergraduate students and I use computer simulation to model and understand high speed magnetization dynamics and to develop techniques to read and write information quickly and reliable. Computer simulation is useful because we need to work on a scale that has a spatial resolution of a few nanometers (1/1000 of the width of a hair) and on a time scale of 100's of picoseconds (10-12 s). Experimentally it is possible to explore the behavior with comparable spatial resolution, or with comparable time resolution, but techniques which have access to both simultaneously are lacking. Computer simulation has proven to be a valuable tool for both understanding high speed magnetization dynamics and for predicting behavior in new systems. We use the computer predictively and develop the results to gain new understanding of the important processes involved.
Below is a description of some of the current areas of active research going on in my group. A list of publications follows, a majority of my undergraduate students are co-authors on these papers as they are in fact generating the bulk of the data.
Domain Wall Control: It is necessary to inject, move, store, and potentially reposition a domain wall in a nanowire quickly and reliably. This is our definition of control. In a planar nanowire, such as the ones we study, the domain wall moves quickest when driven by a weak magnetic field. We were the first to demonstrate an injection pad that allows for low field injection of not just one, but multiple domain walls (see ref. 8) into a single nanowire. Once the domain wall is in the wire, it is necessary to hold it there and the most common technique is via a notch inscribed in the wire. It is possible to select which notch to position a moving domain wall at with the application of a magnetic field component applied perpendicular to the wire. Ref. 5 below has a nice description and visualization of the control process.
Nanowire Arrays: The majority of nanowire research has been conducted on individual nanowires, whereas devices are going to consist of many wires, each containing many domain walls. The ability to individually select, and controllably move one domain wall in such a system is critical. Recently (see ref. 1) we have demonstrated a technique to create a localized magnetic field that can be applied to select and move a predetermined domain wall. Improving the reliability of this process is a very active component of our current research efforts. A recently discovered feature actually increases the selectivity and viability of domain wall control, a discovery we are very excited about. More to come as we finish this development.
Domain Wall Interactions: In both single wires and in nanowire arrays, the domain walls interact with each other. Knowledge of the interaction strength is important for understanding how closely the wires and the domain walls can be packed.
Defects: All wires contain defects and it turns out that this can be a good thing. A current student, with support from the Wisconsin Space Grant Consortium (WSGC) has shown that lots of microscopic defects actually improve the speed of a moving domain wall especially when driven by strong magnetic fields where the motion is typically quite slow. We have worked to develop a simplified model of the interaction of a moving domain wall with the defects to explain the result. A lot of fundamental physics is applied to most easily and accurately describe the counter-intuitive behavior.
Publications at Marquette: denotes MU undergraduate student