Forums
Bruker Media
Community Media
Bruker AFM Probes
SPM Digest
Application Notes
NanoTheater
Website
中文
Brochures & Datasheets
Publications
Probes Catalog
Events
Manuals & Documentation
Presentations
Guide to AFM Modes
News
Journal Club
Webinars & Video
Nanovations
Other
Please I would like know how to overcome strong Z piezo drift when working with magnetic samples which are very rough.
I'd like to look at magnetic domains on a sample with a roughness ~300nm. When lifting the probe up 800 nm the Z piezo gets very unstable. Even on this large scale of lift it is still possible to observe some features due to the sample topography.
Is there a way to settle the parameters to improve the results?
Thanks
Luci
Hi Luci,
800 nm Lift height is i) usually way to high to actually get any MFM and ii) a big step for the z-piezo that can cause creep. My suggestion would be to lift only <100nm and use the closed-loop z feature if your AFM happens to have one. The LiftMode operation will take care to preserve the tip sample even on rough samples.
Stefan
This sounds pretty difficult. Due to many problems, one of which could be drift in z, I think using lift mode on a sample with 300nm features will be very difficult. If you think about the way lift mode works, it's based on the idea that the magnetic fields will be propagating (more or less) vertically from the surface. If your sample has features of ca. 300nm, that unless you lift the tip a *long* way (as you have tried), then some magnetic fields will come from the side towards the tip. I have never tried to lift the tip that far, but I would think that if you are 800nm from the surface, your sensitivity to magnetic fields should be VERY small. Furthermore, there is a lot of inaccuracy in lift mode (due to , for example, drift in x, y, and z, which will make it hard to compensate for sample topography on such rough surfaces. I have a paper which discusses some of the many issues in lift mode MFM, you can find it here:
http://www.fc.up.pt/pessoas/peter.eaton/abstracts/Neves_etal_nanotechnology_2010_abstract.html
As Stephan said, if you have closed loop scanning it could certainly help with the drift.
Good luck!Pete.
Peter,
There is actually very little inaccuracy in the LiftMode procedure if you have a good AFM. You mention drift as a major source. Drift should be adressed by design, mechanical and electronic and should therefore be quite low in a modern AFM like e.g. the ICON.
The 300nm features that are mentioned will not prevent you from performing MFM measurements. The main point being that it would be better if your sample had less topgraphy but sometimes you have to deal with the sample in front of you so to speak. Depending on the nature of the features there can be of course some edge artifacts, but that is almost true for a lot of AFM techniques. The main thing is that you actually do not pull back 300nm or even more, a height at which one would not get a useful signal, but stay much closer to the surface.
Stephan,
Well, perhaps you are right, I have no experience with the ICON...but I have tried to do multiple lift mode images over a single nanoparticle with a (nonlinearised) multimode, and I can assure you that it's challenging! (but possible!)
One of the main issues with lift mode for MFM is that the "magnetic" and "topographic" measurements are not made a the same time. This leads to some uncertainty in what you are measuring. Of course, under ideal conditions, and with an instrument that actually moves the probe to the same place each time, you cna be more sure that the two measurements are more or less in the same place. And of course there are different ways to measure magnetic properties...
AFAIK, all lift mode implementations actually use a constant voltage ramp in the slow scan axis, so that the "normal" and "interleaved" lines are always lightly offset from each other. Is that right, or is it more sophisticated?Pete.
Hi Pete,
Not all LiftMode implementations work the way you describe it. That particular implementation that you are mentioning was developed to operate a software linearized AFM in the best possible way. By using a low noise closed-loop (and I mean so low noise that you never have to turn it off!) one can actually perform the Lift scan on the backward trace. This avoids the issue of the slight offset amoung other things. Check out a DImension Edge or an Innova when you have a chance.
Dear Stefan and Pete,
thank you for your suggestions on the MFM experiment.
We have a NSIIIA equipped with a quadrex. Unfortunately, we do not have a closed loop scanner. We are trying to look at material remaining from biological tissues. We hope to have some iron inside this sample and that the MFM could detect it. The iron remaining appears to form deposit that are rough (the height of the crystal is about 500-600 nm) and thus we have to apply very high lift height in the range of 500-600 nm.
Do you think that smaller lift height would work on such sample?
Keeping the lift height constant and varying the drive phase we observed a contrast inversion on the frequency image, do you know what is going on?
Luci and Thiago
I guess you mean that you are trying to identify magnetic fields associated with this iron, because of course there are much better ways than MFM to detect iron! I am also assuming that you have already checked that the material is magnetic with magnetometry.
If you have such large crystals of magnetic material, and it's actually magnetised, then in principle, it should be pretty easy to detect the magnetic fields by MFM.
Firstly, don't vary the drive phase..this *will* lead to phase contrast inversion ,i am not surprised this happens.Keep the drive phase (and frequency) fixed, and monitor changes in the phase image during scanning.
Secondly, as Stephan said, you MUST use lift heights below 100nm for any MFM. I would start with 50nm lift, and use fairly slow scanning (maybe 0.5-1 Hz for a 2micron scan), and importantly feedback parameters as high as you can get without introducing noise. Use a low resolution scan (128-256 pixels) to begin with. These parameters will help the lift mode to keep the tip OFF the surface during the "interleave" scan line.
Just because your crystal is 500nm high does NOT mean you need lift heights of 500nm! The further you get away from the sample, the harder to detect the magnetic fields it is. NOTE phase signal in lift mode is very small, for example, you might be used to seeing contrast with a phase scale of 30 degrees in tapping mode. In lift mode, magnetic signals will be invisible at this scale. Reduce the z scale to 1-5 degrees, and look for contrast. If you see very BRIGHT spots, it could be the tip touching the sample in "lift". So increase lift height a bit, or scan with higher gains. On the other hand, if you see no contrast, decrease lift height. Do not forget to magnetise the tip before you start!
Regards,
Pete.