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
Does anybody have any useful documents on using AC/tapping mode AFM to determine viscoelastic properties of materials?
I don’t have any papers on this, but believe it is a daunting task to try to extract viscoelasticity from Tapping or Phase. This is challenging because the phase signal is a non-linear function of set point but more so because the phase signal is the mix of multiple sample interaction contributions including viscoelasticity but also plasticity, elasticity, and adhesion.
I think the most direct approach would be to use Peak Force QNM Dissipation channel. This channel records the area between the approach and retract curves, at regular imaging speeds, for each pixel in the image (see figure).
Capturing this entire area gives you dissipation components from the both viscoelastic and the adhesive portions of the curve, however, you can normalize out the adhesive components by scanning the same area at “zero force” (this is something you can routinely do with Peak Force Tapping – contact me if you want more info). The difference between the two images will be representative of the viscoelastic properties (you will have to apply the model of your choice to convert the data from energy).
Steve
Hi Stephen,
How are you doing?
I have got a fast look on your attached picture and have to admit that I have quite some problems with it.
So first, wasn't the drawn curve rather obtained from a regular force distance acquisition and not from a PeakForce (= Pulsed Force Mode) acquisition, since in this case you would still see some remaining oscillations.
Then the stiffness is a value that is measured in N/m (and not in Pa) and that corresponds to the slope of the force versus deformation curve.
What can be determined by using a model like the DMT (Derjaguin, Muller, Toporov) model is a Youngs modulus (or reduced Youngs modulus in the case you ignore the Poisson ratio) which is not the same than the stiffness.
Finally, the equation that is written below "Stiffness: DMT model" is not really the equation corresponding to this model, but rather the one for tyhe Hertz model for the sphere.
One big question i'm asking myself about your Peak Force QNM implementation is the validity of the DMT model for most of the studied samples.
Do you have a way to check whether this model applies or not?
If it is not the case, which alternatives are you giving in order to more correctly analyze your curves?
I thank you very much in advance for your answer.
Best regards,
Philippe
I am well Philippe. Thanks for your note and good to hear from you again. I hope all is well with you.
Agreed, the figure is a schematic and as such makes it the trade off between detail and simplicity. To address your questions let me refer you to the Application Note on Peak Force QNM. Quantitative Mechanical Property Mapping at the Nanoscale with PeakForce QNM AN128.pdf
In particular note the section “Comparison with Force Volume and Pulse Force Mode” in the 3rd column of page 5, as well as the “Elastic Modulus” section on page 4, where in the 3rd column it is discussed “if the DMT model is not appropriate”.
Indeed the choice of models is complex; we offer one within the package – DMT, however, you are right that there are many options and considerations when choosing a model. We did not try to implement all possibilities in the first revision (but are interested in your feedback on which others to include in the future).
We do see the need for multiple model options, so instead we provide the ability to export the raw force curves from an area of the image, allowing the individual scientist to apply whatever model they see fit.
Best,Steve
Hi Steven,
I’m doing fine thank you and enjoy right now the pleasure of having become again father a couple of months ago.
As you probably know, I have already invested more than 10 years in the development of tools and methods in order to improve and make easier the analysis of force spectroscopy data.
Thus I get quite disturbed when I see mistakes like confusion between stiffness and Young’s modulus.
I get even more disturbed when such confusion are found in publication (Sudhir Husale, Henrik H. J. Persson, Ozgur Sahin, DNA nanomechanics allows direct digital detection of complementary DNA and microRNA targets, Nature, 462, 1075-1079, 2009).
Also, if I’m not wrong this publication is from the father of the Harmonix. Also if this paper is used as a reference paper for the Harmonix method, it is kind of scary.
Also concerning the Harmonix, it had been stated on its release (I don’t know if it is still the case) that this technique is quantitative, and I would really like to know how this is possible.
Indeed, it is already quite difficult to measure the spring constant of a regular cantilever which explains that there are a plenty of publications on this subject.
Additionally, it can be shown that for a regular cantilever there are already torsional contributions in the higher frequencies (higher than the fundamental) of the thermal vibrations of the cantilever.
For a Harmonix cantilever, the torsional spring constant will be a function of the torsion angle as well as the cantilever deflection.
In order to make quantitative measurements with the Harmonix, you need thus to be able to determine the torsional spring constant value of the cantilever.
How do you determine the torsional spring constant value relation with the torsion and deflection of the cantilever?
Which kind of function are you using for this determination?
How do you calibrate the function parameters?
Concerning the Peak Force QNM, I had a look to the document you referred me to.
I’m quite impressed about the value differences of the figures 5 with other techniques.
How have these values been determined?
Why does FV Imaging only have a lateral resolution of 100 nm?
Also what is the effect of the used filtering on the acquired force curves?
Has the Peak Force QNM been validated on protein unfolding experiences like spectrin for example?
I wish you a nice day,
Philippe,
It sounds like you have some concerns about Ozgur’s paper, and I can only recommend you give him a call to discuss his work; he is very engaging and always excited to talk about his research. I should mention, Nature is renowned for its quality and thorough peer review.
The information in Figure 5 in the Peak Force Tapping application note is comes from variety of sources including publications, calculations, and market literature. The 100nm spec in FV assumes a typical interaction area from sample deformation at typical FV forces. Regarding filtering; in our implementation, the filter’s corner frequency is at 40kHz, where our probe oscillation frequency is around 2kHz. (or ~20 harmonics before the filter). In testing we varied the cutoff frequency to see if it had an effect on calculated values and found it did not.
I think using Peak Force Tapping on unfolding applications is a rich future application area and one that I am sure will be explored in short order as more scientists gain access to the technology.
Best,
Dear Steve,
It seems to me, but I may be wrong, that the fact that a paper has once been published in a high quality and thorough peer review journal like Nature with a stiffness in Pa, doesn’t mean that stiffness and Young’s modulus are physical quantities that should have the same meaning (and/or unit).
Also, you have completely missed the most important questions about the Harmonix, which are the ones concerning the spring constant calibration of the Harmonix cantilevers.
How are you making the Harmonix cantilever spring constant calibration with respect to the defection and torsion (or even other parameters that may affect its value) so that you are finally able to make quantitative measurements with this technique?
Could you please send me references of the papers that had been used in order to fill out the values of the table of the figure 5.
Obviously, the 100nm spec reported for FV apply only for some very particular case or type of experiences, and already don’t apply for (for example) protein unfolding adsorbed on a hard surface (glass, mica,...) applications.
Finally, it is indeed nice that more and more scientists are going towards the field of force spectroscopy. Nevertheless, wouldn’t the fact to use the Pulsed Force Mode be a cheaper alternative to the Peak Force Tapping technique (at least for most, if not to say all applications)?
The HarmoniX calibration procedure is outlined in detail in our manual. Send me you email/phone number offline (sminne@veeco.com) and I will set up a call to go through it, and your other questions with you.
Stephen,
Thank you for the reply.I've been reading up on other alternative afm-based methods to measure viscoelasticity.I'm interested in reading more on the Peakforce QNM.Please let me know if you can provide some documents about this functionality.I'd like to learn to how to use it.However, I'm also assuming that this function is an add-on to the conventional Multimode AFMs(the one i use is a Veeco Multimode V SPM with a Nanoscope V controller),which requires the purchase of a certain package(?).Thanks for the feedback.
Alperen Ketene
anketene@vt.edu
Alperen,
The application note is the best place to start: Quantitative Mechanical Property Mapping at the Nanoscale with PeakForce QNM AN128.pdf
The brochure provides a good overview: B073-RevA0-PeakForce_QNM-Brochure_HiRes.pdf
Dr. Chanmin Su has a paper accepted on the technique, but not yet published, so I can’t distribute, but can be discussed offline. I will ask one of our Sales Application Scientist to contact you about this as well as how to see it live and upgrade your system.
Best Regards,Steve
Steve,
Thanks for your comments, but t would be nice if that manual was made available to everyone. I also have very large skepticism in the ability of Harmonix to produce quantitative data. Fruthermore the chart comparing the different methods is not wholly accurate it shows force-volume being limited to a minimum peak force of <50nN, that is way off. In fluid pN forces are no problem with the appropriate cantilever and KPa elastic moduli, if not 100's of Pa can be measured.
-Mike
Sorry I was referring to the calibration procedure manual.
The calibration procedure is based on this paper by Ozgur:
O. Sahin, "Harnessing Bifurcations in Tapping-Mode Atomic Force Microscopy to Calibrate Time-Varying Tip-Sample Force Measurements", Rev. Sci. Inst. 78 (2007) 103707.
In addition there are a few tweaks to adjust for scaling factors that our software introduces.
While there are many potential artifacts that can occur in HarmoniX (as with any indentation based technique), it is hard to ignore the results that Ozgur obtained over a wide range of samples:
O. Sahin and N. Erina, "High-resolution and large dynamic range nanomechanical mapping in tapping-mode atomic force microscopy", Nanotechnology 19 (2008) 445717
That is not to say that HarmoniX gives results with uncertainties comparable to dedicated nanoindenters where the calibration is much more sophisticated and complete. HarmoniX is the first (and still the only to my knowledge) TappingMode technique that actually quantifies material properties such as modulus so that we can begin the discussion of artifacts and uncertainties.
On the other hand, PeakForce QNM usually provides better signal to noise and simpler operation and calibration than HarmoniX. If it is not necessary to operate in TappingMode, it is easier to start with PeakForce QNM.