PeakForce QNM and HarmoniX

Reply

Mike Maybrun replied on Wed, Feb 24 2010 1:03 PM

rated by 0 users

Suggested by Bede Pittenger

From draft version of PeakForce QNM application note:

Comparison with Tapping & HarmoniX

As mentioned earlier, TappingMode Phase Imaging is another popular method for material property mapping.  TappingMode uses cantilevers that are relatively stiff and the cantilever is vibrated near the fundamental resonant frequency of the cantilever.  The system adjusts the Z position to keep the amplitude of oscillation at the drive frequency constant.  The tapping amplitude can change as a result of both long range and short range interactions.  As a result, the height data represents the SPM response to a combination of interactions.  Peak Force Tapping, on the other hand, only responds to repulsive force, or short range interaction. The long range interactions (adhesive and electrostatic forces) are not useful for height control.  Short range interactions are the key to high resolution imaging. By consistently controlling the short range interaction forces, Peak Force Tapping enables image quality control with fewer artifacts linked to complication of the tip surface interactions and cantilever dynamics. For example, TappingMode needs to operate with a relatively stiff cantilever and sufficiently high amplitude to avoid tip entrapment on surface due to capillary forces. Once such entrapment occurs feedback will fail as a result of the lost amplitude. Peak Force Tapping control operates independent of the capillary force. Feedback only responds to the repulsive force and moves the tip away from the surface once the peak force reaches the setpoint. The benefit of such control is not only the consistent tip/surface interaction mechanism but also that it enables tapping imaging with far broader range of cantilevers, including very soft levers which were only used for contact mode. This broadened cantilever selection allows much wider range of interaction force control, from piconewtons to millinewtons, thereby enabling a very wide range of materials to be studied.

The phase of the TappingMode cantilever vibration relative to the drive is a useful indication of different mechanical properties.  Unfortunately, the phase signal is a mixture of material properties, depending both on dissipative and conservative forces.[ref Sahin, Cleveland?]  Many efforts have been made to extract quantitative data from phase signal.[ref Tamayo, Garcia, …]  However since elasticity, hardness, adhesion and energy dissipation all contributes to phase shift. Solving one equation with multiple unknown variables can hardly be quantitative.  Additionally, the phase signal depends on imaging parameters such as drive amplitude, drive frequency, and set-point.  This makes it difficult and sometimes impossible to interpret the source of the contrast and the user can only conclude that there are differences in the sample without further knowledge of contributing physical factors.   Fig2 demonstrates this difficulty with phase imaging and compares it to PeakForce QNM for a multilayer polymer sample.  It is often assumed that the phase contrast is primarily caused by variations in sample modulus.  Comparing Fig2(c) and (e), it is clear that this is not true in this case.  By tapping harder (reducing the amplitude setpoint) one would expect to deform the sample more, increasing the contribution of the modulus to the phase, but in this case the contrast does not change significantly (Fig2(f)).  The PeakForce QNM data (Fig2(b)) shows that the phase signal is dominated by the adhesion independent of tapping setpoint for this tip-sample interaction.  It is easy to see that one must be very careful in interpreting phase results even for qualitative use.  The PeakForce QNM Modulus channel, on the other hand, has unambiguous contrast that can be quantified (inset section in Fig2(c)) – the narrow strips have a modulus of about 300MPa, while the wide ones have a modulus of about 100MPa.

HarmoniX microscopy is a TappingMode technique that uses many harmonics of the tapping drive signal to reconstruct the force curves that occur during tapping.  From these force curves the material properties can be measured independently.  The harmonic signals are excited at each tap in the TappingMode oscillation period.  They are detected by measuring the torsional motion of the special HarmoniX cantilever during TappingMode imaging (the TappingMode feedback loop does not change – it still keeps the flexural amplitude constant).  The torsional bending of the cantilever acts as the high bandwidth force sensor and allows force separation plots to be extracted during tapping, and these are analyzed in real-time in analogous to PeakForce QNM.  HarmoniX is a very powerful mode and it has the benefit of working with TappingMode feedback and Phase Imaging for easy comparison with previous results.   On the other hand, HarmoniX mode shares all difficulties of tapping mode in imaging control. For example, HarmoniX can be challenging to use, especially in fluids where the cantilever Q is much lower.  Additionally, interpretation can be complex if the torsional sensor doesn’t have enough bandwidth, if perturbations to the flexural motion become significant, or if cantilever overtones coincide with integer multiples of the drive frequency.

There are two other TappingMode derived modes that have recently gained popularity: TappingMode while observing a separate harmonic of the tapping drive,[ref Ozgur?] and Dual AC mode.[ref Garcia, Cleveland?]  Single Harmonic imaging depends on either special cantilevers or a lucky coincidence of an overtone with a harmonic, while Dual AC adds a second frequency (usually at an overtone) to the vibration driving the cantilever.  Both of these techniques provide contrast that is analogous to Phase contrast in that it does not fully separate the mechanical properties and therefore cannot be quantified. 

A common characteristic of all these tapping based nanomechanical mapping technologies is that they are all based on higher frequency components. Theoretically, in order to accurately reconstruct the real-time tip surface interaction an infinite number of frequency components is needed. HarmoniX can detect 15 to 20 harmonic components, rendering a good approximation and effective for many materials in reconstructing tip surface interaction and deriving quantitative data. Fewer frequency components will further limit the ability of a technique to deriving quantitative data.

In all TappingMode techniques, the feedback uses the near-resonance amplitude as a control parameter. In a normal tapping control, the peak interaction force varies from fraction of a nanonewton to tens of nanonewtons, depending on the operating amplitude, cantilever spring constant, and set point.  Such interaction force is well controlled when cantilever oscillation is in steady-state.  However, when tip is scanning on a sample surface, especially a rough surface, the amplitude error occurring at the sharp edges can correspond to interaction force one order of magnitude higher than that of steady-state.  Amplitude error incurred force is the leading cause of tip damage. Such damage occurs because the feedback is not directly controlling interaction force.   In contrast, Peak Force Tapping directly controls the peak force on the sample.  This protects the tip and sample while maintaining excellent surface tracking.