The characterization of the electrical properties of materials on the nanometer scale with atomic force microscopy (AFM) is a rapidly expanding field. As materials are customized for ever more specialized roles, having an instrument that can be configured to perform these measurements is becoming a necessity. Piezo response (PR) imaging on an atomic force microscope, pioneered by Gruveman et al. a number of years ago, is one such electrical characterization technique. AFMs configured in this mode are used to measure the mechanical response of a piezoelectric material as a function of lateral position when a time-varying voltage is applied across grains or domains in the material. The MFP-3D AFM is ideally suited for these measurements because of the flexibility of the controller and software. Specifically, the all-digital controller and the easy accessibility to change the controller’s crosspoint switch in the IGOR Pro software interface makes PR imaging an ideal application.
How Piezo Response Imaging Works
In the MFP-3D system, the cantilever is scanned over the surface in contact mode. An oscillating voltage is applied to the tip, preferably at a frequency considerably higher than the feedback loop of the AFM (in this example, a frequency of 200kHz). In general, a frequency near, but not exactly on, the in-contact resonance of the cantilever is used. This concept is illustrated in Figure 1. The blue material is polarized “up”. When the tip applies a local potential, the grain shrinks. The red material, polarized in the opposite direction will instead expand. By taking the oscillating AFM deflection and measuring the phase signal with a lockin amplifier, you can clearly differentiate different domains.
Figure 1. Oscillating polarization response microscopy, or piezoresponse (PR) imaging.
Figure 2 shows a 0.5 volt, 200kHz sinusoidal dither added to the tip potential. The cantilever deflection, as shown in Figure 3, contains two components - a DC component that corresponds to the topography of the sample and an AC component that contains information about the mechanical response of the piezo material. The DC component is used to operate the usual contact mode feedback loop and generates the “height”, topography, image shown on the left in the Figure 4. The response to the oscillating voltage of the tip is processed by the two phase digital lockin implemented in the MFP-3D Controller.
Figure 2. DSP: The output of the PSD can be processed to yield the DC deflection, amplitude and phase of an oscillating cantilever. First, the cantilever is excited with the sinusoidal output from a direct digital synthesizer (DDS). This device generates two sinusoidal waveforms 90° out of phase with each other, designated coswt and sinwt. Coswt is used to excite the cantilever with a “shake” piezo. This in turn excites the cantilever to oscillate, the motion of which is detected by the PSD. The analog to digital converter (ADC) converts this into a digital signal again. This digital signal is multiplied by both the coswt and sinwt component, then low-pass filtered to get the in-phase (i) and quadrature (q) components of the cantilever motion as described in the text. The low frequency deflection can also be extracted from the digitized PSD signal by simply low pass filtering. The same functionality can be obtained with analog electronics, though digital electronics have quite a few advantages.
Figure 3. Feedback Loop: Most AFM imaging modes involve a feedback loop that regulate the tip-sample distance. The signal processing block generates an “error” signal. In contact mode, this is cantilever deflection. Other non-feedback signals such as the piezo response amplitude, phase, or tip-sample currents can be passed on to the display system to form an image. The setpoint is subtracted from the error signal. It is then input into a feedback calculation. The output of this calculation is then used to control the tip-sample separation through a high voltage amplifier and a piezo that modulates the tip-sample separation. The job of the feedback calculation is to do this in such a manner as to keep the error signal at the setpoint value (that is, to zero the input into the feedback calculation). The output of the feedback calculation is then a representation of the sample topography, or height.
Figure 4. Topography (l) and piezo response (r) of the highly conductive ZnO sample at a bias of 1.5 volts, 2μm scan. Sample courtesy K. Krishnan Lab, University of Washington
Both signals are obtained from the deflection signal after it is digitized at 5MHz and processed with the Field Programmable Gate Array (FPGA) and DSP in the controller. An example image is shown in Figure 4. These images were made with a 1.5 volt bias applied to the tip. The topography of the thin film is shown on the left hand side and the ferroelectric domain structure, as revealed by the piezo response imaging is shown on the right hand side.
Easy Configuration of the Crosspoint Switch
The crosspoint switch on the MFP-3D Controller allows users a great deal of flexibility in making measurements. PR imaging is just one example of how the controller can be configured to perform a custom measurement. The versatility of the crosspoint switch allows any reasonable connection to be made through the IGOR Pro software interface. In this example, the direct digital synthesizer, (DDS), is mapped to ‘chip’, which is a physical connection to the cantilever holder as shown in Figure 5. For this experiment, a wire is sent from this connection to the sample so that a bias can be applied to the sample relative to the tip. The DDS is basically an AC voltage source and is normally mapped to the drive peizo when operating in AC mode. For PR imaging, it is simply rerouted it to the sample. Note that the ‘defl’ signal is mapped to the ‘fast in’ point in the crosspoint switch. This sends the deflection data to the fast, 5MHz, ADC, which is controlled by the FPGA as shown in Figure 6. A high speed lock-in amplifier in the FPGA is then used to extract the amplitude and phase data from the deflection, while the DC deflection is used to control the tip-sample force feedback loop.
Figure 5. The IGOR Pro software interface allows easy changes to the crosspoint switch.
Figure 6. Crosspoint panel settings for piezo response imaging. “PogoIn0” is connected to one of the auxiliary ADCs. In this case, we are also using the ORCA current imaging cantilever holder to monitor the current flowing through the sample.
Also note that the ‘PogoIn0’ signal is mapped to ‘InA’. ‘PogoIn0 is another physical connection to the cantilever holder (see also Figure 6). In this case, the cantilever holder is an ORCA model. This linear amplifier can monitor the current through the tip. This maps to ‘InA’, which is one of three user accessible ADCs (aside from the ones used by the microscope for normal operation). Other physical connections can be mapped either to outputs from the controller, or back to ADCs to which the user has full access. BNC connectors on the front of the controller allow the user to input their own signals and, again, map them to whatever line is needed with the crosspoint switch.
With the MFP-3D AFM’s powerful controller and flexible IGOR Pro software interface, PR imaging is easily accomplished. PR imaging is an extremely useful imaging technique that can be used to characterize electrical properties for a variety of piezoelectric and ferro materials that will lead the way to many new material advances for a variety of industries.