Measurement protocol guidelines for the J-V characteristics in perovskite solar cells
Dynamic electric model (DEM) simulations, at a bias scan rate of 20 mV/s (solid lines) and 500 mV/s (dashed lines), under bias pre-poling conditions (P0).
Measured dynamic J-V characteristics, at a bias scan rate of 20 mV/s (black symbols) and 500 mV/s (red symbols), under bias pre-poling conditions (Vpol).
Forward-reverse (F-R) and reverse-forward (R-F) scans present different types of hysteresis, normal or inverted, depending on the initial polarization P0 (Vpol).
Forward scan (F): the bias is varied from V=0 to V=Voc.
Reverse scan (R): the bias is varied from V=Voc to V=0.
Measured stationary J-V characteristics (insets), obtained by stabilizing Voc (R-F) and Isc (F-R), at a bias scan rate of 5 mV/s.
For a bias scan rate of 20 mV/s, the hysteresis is still visible.
Measurement protocol guidelines for obtaining the stationary characteristics:
Bias scan rate: should be as small as possible; the measurement time should not however interfere with degradation timescale.
Bias scan range: should be as close as possible to (0,Voc) range.
The open circuit bias, Voc, should be stabilized in advance.
For R-F scans start the measurement from stabilized Voc. Alternatively, for F-R scans start the measurement from stabilized Isc.
Measurement protocol guidelines for a meaningful characterization of the hysteresis:
Determine the stabilized Voc.
Pre-pole the sample at fixed bias Vpol, for a time tpol.
Perform the J-V scans, R-F or F-R, in (0,Voc) range.
A large scan rate combined with (unintentional) bias pre-poling can artificially enhance the PCE.
The J-V characteristics will be far away from the stationary case.
A large scan rate can misleadingly render a hysteresis-free behavior.
Exploring only a narrow interval of bias scan rates can render the hysteresis either increasing or decreasing in magnitude.
The hysteresis magnitude is maximized for intermediate bias scan rates. Very fast or very slow scan rates will render a vanishing hysteresis.
Bias scan range:
A bias scan range exceeding (0,Voc) will introduce additional poling compared the J-V characteristics measured within this interval.
A fixed bias scan range can lead to misinterpretations of dynamical effects when comparing samples with different Voc's.
In this context, a degradation affected Voc can pose additional problems.
Independent forward (F) and reverse (R) scans:
Independent forward and reverse bias scans can lead to significant errors in evaluating the hysteresis:
the initial poling of the individual scans is hard to control; the time between separate F and R scans may lead to uncotrollable depolarization.
Consecutive F-R or R-F scans are recommended, i.e. the second scan is conducted in the opposite sweep direction and begins immediately after
the first scan is complete.
G. A. Nemnes, Cristina Besleaga, A. G. Tomulescu, Alexandra Palici, L. Pintilie, A. Manolescu and Ioana Pintilie,
"How measurement protocols influence the dynamic J-V characteristics of perovskite solar cells: theory and experiment", Solar Energy 173, 976 (2018) arXiv:1803.00285
