Héctor here, your AFM expert at Nanosurf calling out for people to share their Friday afternoon experiments. Today I test if I can see how magnetic fields dampen AFM probe oscillations.You will learn:
I couldn't complete my plan for this week, and I thought it might be useful to share why. Afterall, what distinguishes fridayAFM from everything else is that we talk about things that are not yet finished, and by doing so, we bring them closer to completion (see the posts about measuring oxide thickness for instance, FridayAFM - Coherer effect, it had some open questions that were answered in another post, FridayAFM - You only measure twice).
So, this is the background:
Standard MFM is a 2-pass technique where you oscillate the probe at its resonant frequency. The first pass close to the surface to image the topography, the second pass, at a higher distance using the pre-recorded topography to image the magnetic field gradient (either looking at a phase or frequency shift in the deflection).
Can we do MFM in a different way? More precisely, can we obtain the same or similar information about the magnetic stray field by analyzing existing data from Wavemode?
A few weeks ago I published FridayAFM - Ringing microscopy where I analyzed the deflection curve in WaveMode to extract the dissipation energy (or at least to get a signal proportional to it). At the time I present you with this image I reproduce here below.
The figure shows the different parts of the deflection curve vs time as the cantilever performs WaveMode. From part number 5 we extract the dissipation which I discussed in the previous post. But what about part 7?
Part 7, the ringing decay is affected quite a lot by the damping present on the medium. For instance, if the same curve is taken in liquid, the damping is so high that the ringing decreases within one oscillation or less. In vacuum it stays for a long time and by the time the probe touches the surface again, it might still be ringing.
So what?
So what if the probe has magnetic coating and it is immersed on the stray field of a magnetic sample? Think of a pendulum immersed in oil. The oscillation of the pendulum decreases exponentially due to the oil, and the frequency of the oscillations depend on the length of the pendulum and the gravity. Come back to the AFM probe and you have air damping the oscillation, the cantilever acting as the pendulum, and the pull (or push) from the magnetic field gradient as equivalent to gravity. It is the Cavendish experiment to measure gravity at nanoscale but to measure magnetic field gradient. It is what I call ringing MFM (maybe there is a name out here for it already, but I couldn't find it, so I claim this one). Do not confuse it with a mode where you might excite the probe in between taps of the Wavemode, this is not it, this is purely letting the probe oscillate and performing data analysis.
Does it work?
The answer to that is not that easy.
First I need a nice sample which shows some strong stray field and large domains. For MFM, my test sample it has been for a long time floppy disks (or hard drives sometimes if I'm checking how much lateral resolution I can squeeze out of the probe). But I have done a lot of floppy disks before in fridayAFM, how to make this one somehow new?
I managed to get hold of a Nintendo floppy disk! This is not something you will see everyday. In principle they where only distributed in Japan for the Famicon disk system as a cheap alternative to store games on ICs. Their capacity: 112 kB.
How they look under MFM?
They look like reeeeeeaaaally large bits.
But it is ok, I wanted large domains with strong magnetic field.
So, next step, perform some Wavemode and capture the force curves vs distance including the ringing.
Something like these curves:
Or if you prefer vs distance:
Some Python code and some fitting afterwards....
We got the ringing part and a fitting to a decaying sinusoidal curve.
So, how the ringing frequency looks like for each pixel, do we obtain the MFM signal?
First, let me show you how it should look like using standard 2-pass MFM:
Then this is what I obtained looking at the frequency of the decay ringing:
Did I made a mistake?
No mistake in the analysis, the signal is not there. So what went wrong?
I didn't realize it, but the amplitude of the probe oscillation was too small (if you check on the graph above, the z distance between the minimum, when the probe detaches from the surface, and the maximum distance from the sample is less than 10 nm) and it never went out of the Lennard-Jones potential and onto the regime where only the magnetic interaction is present. In other words, the ringing at that distance so close to the surface is dominated by the adhesion, not the magnetic signal.
Should try again with a cleaned sample (to reduce the adhesion), and a much larger WaveMode amplitude to guarantee that at some point the ringing is happening far away from the surface.
Let's recap. On paper, Ringing MFM was a nice idea. In real life, if the adhesion is too large and the amplitude of WaveMode too small, the ringing will not happen at the distance from the sample where the magnetic signal dominates over other interactions. So... it is worth trying again with another sample with stronger stray field, less adhesion (i.e. cleaned), and with as much amplitude as I can get on WaveMode. (In other words, to be continued...)
I hope you find this useful, entertaining, and try it yourselves. Please let me know if you use some of this, and as usual, if you have suggestions or requests, don't hesitate to contact me.
Extra:
Did you remember that I keep a data base of magnetic media imaged with MFM? So... How these disks density compares to similar devices from the time? Well, it turns out that although these disks appeared in 1986, they are a slight modification of the Mitsumi's "Quick Disk" 3-inch floppies initially released in 1984. Even that, was low density for the time! Explaining why it was considered a cheap alternative.