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
Dear experienced AFM users!,
I am basically trying to understand why, when I try to calibrate the deflection sensitivity during a force curve analysis (by pushing a DNP standard contact probe into a very clean glass surface) I get different values literally every 10 seconds... 54.8 nm/V.... 56.4 nm/V... 53.3 nm/V...
There is no drift in the values (over an hour), they seemingly randomly vary. The system temperature is constant (looking at short scale differences anyways). No aircon passing by. No significant noise. I leave the equipment on prior to all this for 1 hr. I literally take a force curve, press update sensitivity it gives one number. 10 seconds later i repeat the process having done nothing but wait and bam, a different number, sometimes up to 10-20% change, which is clearly enormous. The vertical, horizontal and sum signals stay within a well defined range, and each well within 2V where I expect signals to be nice and linear. Erm, what else... Ah - the setup is a MultiMode AFM using 6.14r1 software and a Nanoscope IV controller. I've tried reseating the probe, repeated realignments (SUM always between 4.5 and 6 ish). WHY DOES THE SENSITIVITY VARY SO MUCH?!?!?!? What is a typical variation waveform to waveform?
Help would be greatly appreciated!
Kind regards
James Karamath
P.S. if I felt that the probe tip cleanness had something to do with it, how does one clean one?!
Dear Mr. Karamath,
The reproducibility of the deflection sensitivity should lie within 5% according to what I've been told by Bruker employees. This is approximately the case for the values you mention here. However, "10-20%" is clearly too much. If you use Peak Force QNM the measured modulus depends on the deflection sensitivity by the power of three, meaning that a 10%-different deflection sensitivity results in a 33%-different modulus. However, if you measure several values and then use the mean of them, it shouldn't cause high errors. Additionally, it might be wise to capture the deflection sensitivity from captured ramp curves rather than from real-time ramps as the real-time fit uses only two points in the force-distance curves to calculate the slope. You can obtain higher precision offline. But don't expect too much; the slopes will not be perfectly linear, so you will have some variation in the deflection sensitivity even if you use only one single curve to measure it but not exactly the same fit limits.
Best regards, Dietmar Haba