07 August 2007

Y.A.B.A.F.M.I. (Yet another brilliant AFM innovation)

Atomic force microscopy is a leading tool for nanoscale studies of surfaces and objects as small as molecules and even atoms. With nanoscale features commonplace in semiconductor, data storage and life-sciences applications, AFMs are important for both research and industrial uses. These sophisticated instruments build a picture of the nano-world in much the same way that a blind man with a white cane does, using an atomic-sharp tip on a tiny spring cantilever whose motion is observed by sensitive instrumentation. But, as Nanowerk describes, researchers at Harvard and Stanford have literally put a new twist on the conventional way of doing things:

"In order to create a high speed and sensitive nanomechanical measurement tool, we have started from the most commonly used AFM technique called the tapping mode" explains [Harvard's Ozgur] Sahin. "The primary advantage of this technique is that it protects the tip and the sample during the imaging process and minimizes the interaction forces.
"For our goal of performing mechanical measurements, tapping mode also provides a unique opportunity because the sharp tip is moving back and forth against the surface and feels the variation of force during the interaction. If one can detect those forces varying with tip sample distance, one can perform a clear and detailed mechanical analysis."
Unfortunately, there are major difficulties in measuring the forces between the tip and the sample. These forces change at a rate much faster than the vibration of the cantilever, therefore the force sensing cantilever cannot respond to them. Indeed, there is a wealth of publications in the literature working on the non-linear dynamics of tapping cantilevers that seek indirect ways to measure these forces.
Hmph. I got bit by the non-linear dynamics they're talking about in a customer's advanced AFM application just ten days ago. Not being an atomic force microscopist, at first I had no idea what I was looking at and thought our instrumentation had gone bonkers. Nanowerk and Sahin continue:

"In a way, our work stands on the 'shoulders of these giants', because they have reached a very good understanding of the complicated cantilever dynamics in AFMs" says Sahin. "Nevertheless, we have taken a different approach by engineering the force sensing cantilever to measure the interaction forces directly."
The AFM cantilever has many vibration modes. Each one of these modes can act as an independent force sensor. The rapidly changing forces demand a fast (high resonance frequency) mode to be used. The problem with high resonance frequency modes is that they are stiff and do not bend easily to give a good signal.
"What we have noticed is that torsional vibration modes allow good signal levels and they have high enough resonance frequencies" says Sahin. "Unfortunately, tip sample forces do not excite torsional oscillations because the conventional cantilevers have their tips on the center line. Therefore, we designed cantilevers that have their tips off-centered. When this cantilever hits the surface, tip-sample forces generate a torque that bends the cantilever torsionally. Torsional vibrations can be detected in a commercial AFM system simultaneously with the vertical vibrations." When this cantilever is operated in conventional tapping-mode – touching the surface ever so lightly some 50,000 times per second (50 kHz) – the torsional vibrations can be simultaneously detected and translated into a time-varying tip-sample force waveform which contains detailed information about the mechanical properties of the sample.
"In principle, the speed of these measurements is limited by the oscillation frequency of the cantilever" says Sahin. "At the moment we are not fully benefiting from the speed enhancement, however, it is still more than a factor of thousand times faster than conventional mechanical measurements, yet it is much gentler to the sample.
"Improved speed enables mapping mechanical properties across a surface with nanometer resolution. I believe that in the near future we will see mechanical measurements performed within a microsecond. This will open up a new window to study time dependent phenomena at the nanoscale, such as protein folding and chemical reactions in general."
This is looking like another fundamental advance in a field that's littered with them.

No comments: