When scanning a 1-micron area at 0.5 Hz using the Dimension Icon in QNM mode, I observed wave-like artifacts in the topography channel. Even after attempting various order flattening methods, the artifacts persisted. However, reducing the scan rate to below 0.3 Hz eliminated these artifacts. What might be the cause of these wave-like artifacts in AFM topography
99.9% sure that you have a very severe tip artifact, quasi all patterns are identical. The true sample shape could be very small and nicely spherical particles, but your tip is a ca. 100 nm long "L".
As for the "waves", they are very likely oscillations, I would guess electronics (power supply frequency), but the mechanics (fast scanning) can sometimes enhance them, so no surprise you get rid of them at slow scanning.
I am quite sure that a fresh tip (see above) will solve the problem!
Alexander M Bittner, thank you for your response. Does a higher scan rate result in a shorter tip-sample interaction time, leading to increased detection of oscillation signals? Alternatively, could there be another reason? And do these waves affect the measured mechanical properties?
There can be too many reasons for the waves or rather oscillations. Basically, mechanics and electronics - it is a feedback loop, so principally sensitive to any interference. I can only guess: The power supply might run at50 Hz or 60 Hz, and at this frequency (or at a multiple) your signal (height) has a perturbation. When scanning more slowly, it is still there, but such that it is not well visible (i.e. the angle of the "waves" changes). Or indeed, the system has a little bit more time to adapt to the height change from pixel to pixel - in this case, I would expect the oscillations especially at the side wall of a high surface feature.
Dear Tseng,
I concur with Prof. Bittner's observation about the likelihood of tip artifacts, given the patterns and angles. As for the wavy features, troubleshooting is multifaceted.
I'm curious about the real-time scan graph; perhaps you could share this data for further insight.
A few considerations:
My initial inclination is that the gains might indeed be too high causing the piezo oscillation. Nevertheless, a brief sample description along with the file metadata would facilitate a better assessment of the primary parameters. Try exploring the aspects above though.
Additionally, employing FFT filtering or a low-pass filter could enhance the image quality by selectively removing these repetitive wave patterns.
Good luck,
RC
Alexander M Bittner Thanks for your response again. My power supply runs at 60 Hz. The files below contain other mechanical properties channels, and there are no wave-like features in the images. Therefore, the oscillations might be caused by electronics or other reasons that do not affect the mechanical properties?
Rafael N.P. Colombo Thanks for your suggestion.
Since I'm using Peak Force QNM mode, there's only feedback gain to adjust. Currently, my feedback gain is set at 15 with a peak force setpoint of 30 nN. I'll experiment with different scan angles to see if the wave angles change.
Since my experiments are conducted in a Faraday cage, the environmental factors may not be the issue. However, the wave-like artifacts remain consistent across almost all samples and tips.
1. Analysis: Counting the oscillations is probably the best way to analyze the problem. I predict you will find about 120 in your image (careful if you run an overscan). The scan took 1 second (2 seconds for trace and retrace), and 120 Hz is twice that of your power line? Then it is likely electronics, but, as mentioned, his is usually resolved with a new tip.
2. Resolving the problem: You need more images with other tips and/or with a fresh sample anyway, to reproduce your results, so I am sure this is the best way to go. I would forget analysis and take new scans instead.
If I were You I woluld check cables. If connections are ok, be sure the AFM's table is well grounded. Next, try to move out (or turn off) any electric devices, which can produce electromagnetic field. If nothing helps, use probes with other frequencies. It looks as echo from elecric power source.... let us know how you solve problem, please.
- Badges
- Science topic