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
Hello all,
I'm using a Dimension 3100 in taping mode with a common silicon probe. On my samples I can observe mode III cracks which produce a step. So basically, I'm imaging steps.
My issue is that if I change the scan size from 512nm to 1024nm, the step height is doubled. And if I increase it to 2µm, it's again doubled, as if there were a linear correlation between the scan size and the height of the step. Changing the speed of scan wouldn't change anything (I kept the absolute speed constant for the different scan size also). So I cannot evaluate the real step height.
I tried on other steps, it's the same. I get 2/4/8 nm for small ones or 10/20/40nm for big ones.
Have you already heard of this artifact? How can I correct it?
Thanks for your help
H-Y
Sure. I can give an example.
I made a 500nm*500nm scan of my crack. Without changing any other parameter, I change the scan size to 600nm. The scanned area is centered on the very same point. Then I change again to 700nm, and for some reason the scanned area is several tens of nanometers higher up in the Y-axis. As the crack height gets bigger for bigger "Ys", I had the feeling that there were a artificial correlation between the scan size and the Z height.
My mistake was to consider only my sample. I learned that checking regularly the tip and the scanner on calibration samples is important before making any judgment on "weird" results.
Dear H-Y,
That sounds weird to me. I have some rough ideas what might cause effects like that, but I'd like to see your images before. Could you please post them so that we know better what you're talking about?
Regards, Dietmar
---
For posting images here:Go to the top right of the window and click on your profile name. On the right side of the opening window you can find the title "My files". As you click "View All Files" you get to a window where you can upload data. I recommend to upload them in the form of ZIP-files. Once uploaded, you can add the files to your postings by clicking "Insert Media" here in the posting dialogue.
Hello Dietmar,
I'm sorry but I cannot use the AFM before some weeks and I didn't save any line profiles :s .
I had the same problem by turning 90° the scan on other steps.
Also the same height profile is showed in both directions, there is no visible hysteresis.
What I did to characterize it was to look at only line profiles, disable the slow scan axis and change different parameters (scan size, speed, rotation angle).
Once I can have access to the AFM again I could run the same thing on calibration grids, but I don't think it'll change anything as I think my steps do exist (one can see them by SEM imaging with secondary electron mode). But if you have already ideas, feel free to share them :) .
Thank you.
ps: my AFM expert, who's now on vacations, told me the AFM was qualibrated just one month ago.
Hello, H-Y,
Haven't you got any images offline? Just to give us a clue of what you're talking about? I'm afraid I can't tell anything without having seen the images.
Hello,
I exported one of the images of my scan in ASCII. This is it imported on Matlab, the blue line corresponds to the line profile located on the top right hand corner. The scan size is here 3840nm, but if I zoom in the step height changes. It's not a very good example because I have this deep indent in the middle so I certainly tuned the feedback to respond correctly to such a big change of topology (it's a wafer so the RMS is below 1nm). And the direct scan is not showed (I saved with reverse). I hope this can help. Otherwise I'll have to wait for him to come back.
All units are in nm by the way.
Thanks.
H-Y.
Ok, I made an artificial line profile with matlab. The noise is of course not so high and the tip angle (or the step angle) is perhaps not so perfect, but it represents what I experienced.
It's still hard to imagine what your problem is. Haven't you got the original data (file name ends with ".000" or so) that we could open in NanoscopeAnalysis? From what I know so far, I cannot really help you. You seem to have produced some weird kind of artifact, just I cannot tell you yet how to avoid it.
Dietmar
I found out my problem. by increasing slowing the scan size with the slow direction (Y-axis) disabled, I noticed that the Y-position changes, hence a different height is measured. On a calibration grid I was obtaining always the right height for any scan size and any speed. So problem solved.
Thank you anyways.
Dear hyamanieu,
Glad to read you managed to solve your problem. However, could you explain the solution you found once again, just in case somebody else has the same problem and finds this discussion looking for the answer?