The reduction of energetic disorder in the absorber layers has played a major role in the progress in organic solar cell development towards 19% efficiency.[1, 2] Hence, accurate characterization of this energetic disorder is crucial on the search for high-efficiency organic solar cells. The quantities typically used as a measure of disorder are the Urbach energy E_U, which is the slope of an exponential band tail,[3] or the width of a Gaussian distribution.[4] Both models are fit to data on the density of subgap states that can be obtained by various methods. These characterization techniques can be classified by the way the energy axis is probed which can be optical or thermal excitation, a variation of applied voltage or even time- or frequency-dependent scans. In our literature research, we have observed that methods based on optical excitation like Fourier-transform photocurrent spectroscopy (FTPS) and photothermal deflection spectroscopy (PDS) typically yield low Urbach energies whereas for voltage-dependent methods such as charge extraction and capacitance measurements, values of up to 3kT are as frequently reported as lower ones. This discrepancy points towards a systematic difference between the methods and highlights the need for a study of the potentials and limitations of characterization techniques of energetic disorder in organic solar cells.

For this purpose, we determine the Urbach energy with PDS, FTPS and capacitance measurements. Indeed, we also observe much higher values of E_U from the capacitance than from the optical measurements. With the help of drift-diffusion simulations, we demonstrate that the different ways of varying the energy axis of the density of states can also lead to different energy ranges of the subgap density of states being reflected in the Urbach energy. So, a more shallow trap distribution towards midgap will cause a higher Urbach energy in the voltage-dependent measurements than in the optical measurements that probe the steep band tail near the band edge. We further show that factors that generally deteriorate the solar cell performance like high injection, high recombination or low mobilities can lead to an overestimation of the Urbach energy in the capacitance measurements. So overall, we demonstrate the potential of combining different methods to characterize energetic disorder on a wider energy range but also to avoid misinterpretation of data due to a method’s limitations.