AFM Lab Aplication note P01. AD8429 Piezoresponse Force Microscopy Amplifier

Transcription

1 AD8429 Piezoresponse Force Microscopy Amplifier - New standard for PFM measurements - State of the art signal amplifier - Designed and built in AFM Lab - Compatible with PFM,EFM,MFM Based in the Analog Devices AD8429 IC Operational Amplifier, we have developed a new low signal amplifier which overpasses the signal-to-noise ratio of AFM integrated amplifier. This new device is ready to be used, and will be our standard for PFM measurements in The amplifier exhibits the best signal-to-noise ratio, 1nV/sqrt(Hz) over a wide bandwidth of 1,2MHz. The amplifier is designed to be couple AC or DC, user selectable. With this amplifier we can read low deflection signals which were buried in noise with conventional AFM amplifier. Figure 1. Amplitude vs Frequency over a GND signal, comparison between AFM Amplifier (black) and AD8429 (red) line. Less deviation means less noise.

2 Figure 2 depicts the test of the amplifier with our standard Periodically Poled Lithium Niobate (PPLN) sample. The sample was scanned in the exact same area, at the same moment, comparison the signal coming from the AFM built-in amplifier and the new AD8429. Both Amplitude and Phase images of AD8429 amplifiers are cleaner, with less noise than the built-in AFM amplifier. Figure 2. PFM Amplitude (a) and PFM Phase (c) of a PPLN sample measured with the AFM built in amplifier. PFM Amplitude (b) and PFM Phase (d) of a PPLN sample measured with the new AD8429 Amplifier. AC Voltage was reduced in order to decrease the signal amplitude. With the amplifier, domains are seen cleaner, as denoted by histograms, less voltage is needed to measure the same phase difference

3 The new amplifier is capable of providing better phase difference between domains in high noise environments. In Figure 3 we compare the phase difference between the ideal case- theoretical case with no noise present-, the new AD8429 amplifier and the AFM amplifier. The Phase difference between domains were acquired from Fig. 3c and Fig. 3d, calculating the phase different from the area average. As the AC voltage is decreased, we lose PFM signal as the signal has more noise coming from the background. With the new amplifier, the background noise, which is also amplified, is reduced meaning that we can read domains phase difference closer to the ideal case. Figure 3. Domains Phase Difference vs AV voltage used. As the VAC signal is reduced, the PFM signal is also reduced, and thus the phase see by the Lock-in amplifier is the noise phase signal, which has a Gaussian distribution centered in a constant phase value. With the new amplifier, we are closer to the ideal case than before.

4 The measure D33 constants and provide accurate values, resonance curve must be cleaner. With the new amplifier, we can get cleaner resonance curves, and moreover, we can read PFM signals were we could not read them before. In Figure 4a we acquired the resonance curve, in the same spot, with the same tip with the new amplifier and the AFM amplifier. The resonance curve is cleaner, and sharper, meaning that it will be easier to acquired D33 constants. Moreover, in extremely noisy environments, we can get D33 measurements, as denoted by figure 4b where the amplifier is unable to acquire the PFM signal, while the AD8429 Amplifier can see it. Figure 4. Amplitude vs Frequency of a PPLN sample, resonance curves acquired with 2 VAC (a) and 0.2 VAC (b). Signal is cleaner in (a) for the new amplifier, while in (b) there is only signal for the new equipment.

5 Figure 5. Schematic diagram of the proposed amplifier. There two AD8429 configured with 100xgain, that can be concatenated. Figure 6. PCB board completed with populated elements, ready to be used.

6 Figure 7. Aluminum box with the amplifier inside, acting as a faraday box. Figure 8. Connection outside Aluminum box allowing easy access to signals.