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
Hi,
I am curious as to where on the cantilever the thermal method gives the spring constant to.
Many methods will specify the spring constant at the end of the cantilever, in which case the value must be corrected back to the tip; but I am unsure of whether this is the case when using the thermal method.
Any advice would be greatly appreciated
Cheers,
Ash
Ash,
There are many good references on calibration, here is one from Ben Ohler: AN094-RevA0-Practical_Advice_on_the_Determination_of_Cantilever_Spring_Constants-AppNote.pdf
In general, the thermal tune calibration method takes into consideration the tip location because the method uses the measured deflection-sensitivity. The measured deflection-sensitivity takes into consideration the real location of the tip because it is calculated from a force curve, which is generated by pressing the sample against the tip (not the end of the cantilever).
There are many additional small corrections you can chase including spot size and location, and tilt angle and tip/cantilever length ratio (see Hutter). In general, the thermal tune method is generally well accepted.
Steve
Hi Ash,
Unfortunately the thermal tune function doesn't have separate parameters for the "prefactor" (i.e. 0.971 or 0.965) and the X (Chi) squared deflection sensitivity correction. There is only a single correction factor that is treated like Chi (i.e. squared in the denominator). In other words, the equation it uses is like Eqn 16 in the app note except without the 0.971 factor.
The workaround for this is to lump together both corrections into an "effective Chi" factor. So for a rectangular cantilever, it would be SQRT(1.09^2/0.971)=1.106. This is mentioned in Step 5 of the thermal tune instructions on page 10 of the App Note. These instructions are better than the ones in the manual. We should really update the function to support entry of both types of correction factors or base it on a simple rectangular vs. v-shaped shape selection. In the meantime, this correction accomplishes the same result.
Regarding the sampling rate, the NanoScope V manual is incorrect. That section must have been copied from an older manual that assumed our NS3a or NSIV controllers. Those old controllers were limited to 64 kHz sampling rate. You are quite correct that the NSV can sample much faster, either 500 kHz with the "low speed" ADCs or 50 MHz with the "high speed" ADCs. There is a "Thermal Tune Range" parameter in that view where you select either 1-100 kHz or 5-2000 kHz. These ranges use the low speed and high speed ADCs, respectively. You simply pick whichever one is large enough to cover the cantilever resonance.
The NanoScope PI controller (used only on the now obsolete Icon PI, not the current Icon or Icon PT) had a slightly different architecture. The thermal tune range on it was limited to 1-100 kHz..
I hope this clarifies these issues. Sorry for the errors in the documentation. Thanks for participating on the Nanoscale World site.
Regards,
-Ben
Thankyou for your help Stephen,
I am reasonably familiar with the thermal tune method, however that aspect of it makes a lot more sense now.
I have two follow-up questions if you're able to help:
Are the correction factors of 0.971 and 0.817 described in the app note by Ben Ohler applied automatically by the nanoscope 8 software, or must they be introduced after? I understand that an 8% offset is applied to the measured deflection sensitivity, however the nanoscope manual seems to describe this as a correction to something other than the factors described in the app note.
In addition, the nanoscope V manual also says that with the standard controller, thermal tune samples the deflection at 64kHz, limiting it to calibrating cantilevers with resonant frequencies lower than 32kHz.
Is this only the case with the Nanoscope V-PI controller, or also with the Nanoscope V?
I have been assuming (quite possibly naively) that due to the controller being able to sample at effectively 25MHz, the deflection sampling rate would be much higher than 64kHz.
Again, thankyou for your help; this forum has been an excellent resource for SPM information.
Regards