2 thoughts on “How to Perform PUND Measurements — A Beginner’s Lab Manual”

1. 

   Payal

   December 19, 2025 at 12:39 PM

   Hello Ma’am. I have been trying P-E loop measurement on my sample (thin films), but the loops i have obtained (majorly the loop is broken) shows exceptionally high polarization values (2000-2500 microC/cm2). I have done PUND measurement the remnant polarization I have obtained is around 3-4 microC/cm2. I am trying to optimize the sample to get a proper loop with reasonable polarization values. But so far I am not able to obtain a proper data. Any suggestion from you will be highly appreciable.
   I am using Radiant technologies, i have prepared samples with spin coating with thickness of 40-50 nm on ITO substrate, I am using Gallium indium eutectic as my bottom and top electrodes.

   Reply
   1. 

      Dr. Rolly Verma

      December 19, 2025 at 5:03 PM

      Hello Payal,
      Thank you for your detailed and thoughtful question. This is a very common—and important—issue in thin-film ferroelectric measurements, and you are absolutely right to cross-validate P–E loop data with PUND analysis.
      Based on your description, the exceptionally high polarization values obtained from the P–E loops (~2000–2500 µC/cm²) are not physically realistic for ferroelectric thin films. Such values strongly indicate that the measured loop is dominated by non-switching contributions, including leakage current, capacitive charging, and electrode-related effects. In contrast, the remnant polarization obtained from your PUND measurement (~3–4 µC/cm²) is far more reasonable and should be considered a much closer representation of the intrinsic ferroelectric response of your film.
      A few important points to consider while optimizing your measurements:
      1. In very thin films (≈40–50 nm), distorted or broken P–E loops are commonly observed due to a combination of high leakage current, imperfect electrode interfaces, and non-uniform electric fields.
      2. The use of liquid gallium–indium eutectic as both top and bottom electrodes can further complicate the measurement. It has various limitations at the nm scale. Liquid electrodes often introduce unstable contact areas, additional series resistance, and parasitic capacitance. These effects can severely distort P–E loops, even though PUND measurements may still appear relatively cleaner.
      3. For ferroelectric films in the 40–50 nm thickness range, solid electrodes are strongly recommended if your goal is to extract meaningful polarization values. Liquid Ga–In can be useful as a preliminary or temporary contact, but it should not be relied upon for final conclusions. However, proceed carefully, just remember that ITO is not a perfect metal, so interface effects must be interpreted carefully, not ignored.
      4. At this stage, the PUND result should be treated as the more reliable indicator of ferroelectric switching. A low PUND polarization combined with an inflated P–E loop almost always signifies that the loop is leakage-dominated rather than polarization-dominated.
      5. Before attempting further loop optimization, focus on reducing leakage. This may involve lowering the applied electric field, optimizing the measurement frequency, or improving film quality through suitable annealing (if compatible with your material system).
      • During optimization, it is crucial to apply the electric field gradually rather than sweeping directly to a high field. Increasing the field step-by-step—from a low value toward saturation or breakdown—helps clearly identify the onset of switching, leakage, or dielectric failure, and avoids artificial loop distortion caused by sudden high-field stressing.
      • Measurements should also be performed at a sufficiently low frequency appropriate for your material system. If the frequency is too high, leakage and capacitive currents can dominate the response, especially in ultrathin films, leading to exaggerated polarization values. Joint optimization of frequency and field amplitude is therefore essential.
      • Finally, it is important to remember that for many thin-film ferroelectrics—particularly at nm-scale thicknesses—a perfectly “textbook-like” P–E loop is not always achievable. In such cases, PUND analysis becomes essential and is often preferred in research publications for reporting reliable remnant polarization values.
      You are approaching this problem in exactly the right way by comparing P–E and PUND results. I encourage you to continue focusing on leakage suppression and interface optimization, and to rely more strongly on PUND data for intrinsic polarization assessment.
      I hope this clarifies your observations. Please feel free to share further results if you would like to discuss them.
      Best regards,
      Dr. Rolly Verma
      Advance Materials Lab

      Reply