Nanoscience Research: Identifying Nanomagnetic Alignments
The summer before my senior year of high school, I interned for a program called SHARP, which placed me inside the Nanoelectronics and Nanostructures Group at UC Berkeley.
There I met Dave and Pat, two phenomenal grad student mentors whom I assisted with their research. I worked on two projects, one with nanomagnets, and the other with carbon nanotubes.
Nanomagnets are <100 nanometers in size and have just one domain, meaning that all their magnetic moments are pointing in the same direction. They’re very useful because their magnetic fields can easily be aligned in a particular direction. Imagine a wire of nanomagnets where the individual fields are pointing either up or down, essentially replicating binary. Thus, they can potentially be used in computers to replace silicon-based semiconductors.
Dave’s Project
My grad mentor Dave fabricated permalloy nanomagnets of ~100nm in lines of 20 or so and imaged them using a magnetic force microscope (MFM). Lined up this way, the magnetic fields were made to point up or down by simply triggering the first magnet in the series. A domino-like effect took place and unaligned nanomagnets became fixed, pointing in alternating directions. The end result was a series of magnetic domains in an up-down-up-down pattern.
My goal was to see if there were any errors in the domino effect; that is, if there were any sequences in the chains that were up-up or down-down.
I used MATLAB to process the images from the MFM and detect anomalous patterns, i.e. anything that was not a nicely alternating sequence. The problem was, there was so much noise in the image that simply looking at the color intensities across the image wouldn’t cut it — there was too much ambiguity across pixels.
We had to minimize the noise to get a consistently accurate reading of the nanomagnets. So, with Dave’s help, I used fast Fourier transforms to isolate the color profile going across the image.
It worked beautifully, and actually helped us identify a trend — the errors were occurring in the second half of the chain. We suspected that there were potentially some environmental/thermal perturbations destabilizing the domino flow. Perhaps a more tightly controlled environment or different spacing would help improve this process.
All in all, an incredible learning experience; a great first foray into lab research that taught me how to ask questions and come up with even more questions. I learned some MATLAB, helped fabricate nanomagnets and image them at Lawrence Berkeley National Labs (in this great building called the Foundry that’s shaped like a cantilever and hangs off the edge of a hill), and ate plenty of delicious potato puffs from Gregoire. Totally worth it.
