Dual-frequency resonance tracking in switching spectroscopy piezoresponse force microscopy for ferroelectric thin films

Switching spectroscopy piezoresponse force microscopy (SSPFM) is widely used to probe local polarization reversal in ferroelectric thin films, but resonance-enhanced implementations remain difficult to apply to large datasets because they typically rely on repeated frequency sweeps, leading to longer acquisition times, higher noise levels, and fitting-related instability. Here, we introduce an approach that combines SSPFM with real-time dual-frequency resonance tracking (DFRT), allowing the excitation frequency to remain locked to the instantaneous contact-resonance frequency throughout the spectroscopy sequence. This strategy extends the benefits of real-time resonance tracking, commonly exploited in continuous measurements, to pulsed switching spectroscopy, while avoiding repeated reconstruction of the full resonance peak at each voltage step. The method is evaluated on a Pb(Zr0.2Ti0.8)O3 thin film and compared with two conventional SSPFM approaches: sweep-based and fixed-frequency measurements. At short segment durations, sweep-based SSPFM becomes increasingly affected by phase noise, phase-inversion artifacts, and fitting errors, whereas SSPFM-DFRT preserves low noise and measurement stability. In addition, the conversion of raw data into hysteresis loops is substantially faster because no repeated resonance fitting is required. Compared with fixed-frequency SSPFM, it also compensates resonance shifts, which become critical during prolonged measurements such as point matrix acquisitions and would otherwise reduce measurement repeatability and robustness. Finally, SSPFM-DFRT is demonstrated in mapping mode on a 50 × 30 point matrix, where it enables robust extraction of switching parameters with high fit quality across the dataset. Overall, this approach provides a practical route to faster, more stable, and more robust resonance-enhanced SSPFM for spectroscopy and mapping, particularly when short measurement times and reliable large-scale measurements are required.