Measurement of Surface Photovoltage by Atomic Force Microscopy under Pulsed Illumination

Measuring the structure-function relation in photovoltaic materials has been a major drive for atomic force microscopy (AFM) and Kelvin-probe force microscopy (KPFM). The local surface photovoltage (SPV) is measured as the change in contact potential difference (CPD) between the tip and sample upon illumination. The quantities of interest that one will like to correlate with the structure are the decay times of SPV and/or its wavelength dependence. KPFM depends on the tip and sample potential; therefore, SPV is prone to tip changes, rendering an accurate measurement of SPV challenging. We present a measurement technique which allows us to directly measure the difference in the CPD between illuminated and dark states and, thus, SPV as well as the capacitance derivative by using pulsed illumination. The variation of the measured SPV can be minimized due to the time-domain measurement, allowing accurate measurements of the SPV. The increased accuracy enables the systematic comparison of SPV across different measurement setups and excitation conditions (e.g., wavelength dependence and decay time of SPV).

Figure 1

Simulated frequency-shift bias spectroscopy for a CDP shift of 200 mV. For illustration purposes, the capacitance gradient is kept constant. A pulsed illumination will modulate the frequency shift at any dc bias between the two curves.

Figure 2

(a) Block diagram of the TD KPFM measurement. The frequency-shift signal is integrated with two gated integrations synchronized with the pulsed illumination. Each output of the gated integration is recorded at each applied bias to record a full Kelvin parabola. (b) Measurement scheme for a gated integration in the time domain of the modulated frequency shift.

Figure 3

TD KPFM measurement of the SPV under pulsed illumination. A gated integration is used to acquire the bias spectroscopy for the dark and illuminated states simultaneously. A shift of $213\pm 4\mathrm{mV}$ in the CPD and 5% in $(d^{2}C)/(d{z}^2)$ is measured under illumination.

Figure 4

Histogram of 162 SPV measurements recorded successively over more than 18 h. The differential TD KPFM SPV measurement ($202\pm 22\mathrm{mV}$) has a much higher precision than the standard SPV measurement ($207\pm 312\mathrm{mV}$) with a 14-times-smaller standard deviation. The inset shows two measurements $a$ and $b$ at different times.