Know your full potential: Quantitative Kelvin probe force microscopy on nanoscale electrical devices

Abstract

In this study we investigate the influence of the operation method in Kelvin probe force microscopy (KPFM) on the measured potential distribution. KPFM is widely used to map the nanoscale potential distribution in operating devices, e.g., in thin film transistors or on cross sections of functional solar cells. Quantitative surface potential measurements are crucial for understanding the operation principles of functional nanostructures in these electronic devices. Nevertheless, KPFM is prone to certain imaging artifacts, such as crosstalk from topography or stray electric fields. Here, we compare different amplitude modulation (AM) and frequency modulation (FM) KPFM methods on a reference structure consisting of an interdigitated electrode array. This structure mimics the sample geometry in device measurements, e.g., on thin film transistors or on solar cell cross sections. In particular, we investigate how quantitative different KPFM methods can measure a predefined externally applied voltage difference between the electrodes. We found that generally, FM-KPFM methods provide more quantitative results that are less affected by the presence of stray electric fields compared to AM-KPFM methods.

Keywords: AM lift mode; AM off resonance; AM second eigenmode; AM-KPFM; FM-KPFM; cross section; crosstalk; field effect transistor; frequency modulation heterodyne; frequency modulation sideband; quantitative Kelvin probe force microscopy; solar cells.

Figure 1

CPD line profiles of two KPFM experiments on the same cross section of a mesoscopic perovskite solar cell under short circuit conditions with and without illumination, visualized by the red and blue line, respectively. The cell consisted of a fluorine-doped tin oxide (FTO) electrode, a compact TiO₂ electron extraction layer and a mesoscopic TiO₂ layer (meso) filled with the perovskite light-absorber methylammonium lead iodide (MAPI). The mesoscopic layer was followed by a compact MAPI capping layer, the hole transport material spiro-OMETAD and a gold electrode. Prior to the measurement the cross section of the solar cell was polished with a focused ion beam (FIB) to minimize topographic crosstalk. The CPD line profiles in a) were extracted from double side band frequency modulation KPFM (FM sideband) scans in single pass with VAC of 3 V [7]. The CPD line profiles in b) were extracted from amplitude modulation KPFM (AM lift mode) scans in lift mode with a tip–sample distance of 10 nm, an oscillation amplitude of ≈80 nm and a tip voltage U_AC of 1 V. Each line profile is an average of three adjacent scan lines.

Figure 2

Overview of excitation and detection frequencies for KPFM methods used in this work. The lower part shows the transfer function of the cantilever, amplitude plotted vs frequency. The upper part shows excitation (arrow upwards) and detection (arrow downwards) for the corresponding methods with the respective frequencies. Red color is used for the topography signal and blue for the electrical excitation and detection. The color code in the upper part corresponds to plots in the results. Representation inspired by [26].

Figure 3

Sketch of the setup as well as marking scheme of electrodes on the edge. U_ext represents the electrical excitation applied to the cantilever, U_pot the potential applied to the electrodes.

Figure 4

Comparison of the deviation of the measured potential difference from the applied potential plotted against the applied potential for AM (warm colors) and FM (cool colors), black represents an ideal measurement. Inset in the lower right visualizes the fraction of the potential captured by the respective method. Legend in the upper left shows the offset in brackets. Data shown captured in the middle of the electrode structure without an additional electrostatic force.

Figure 5

Comparison of the deviation of the measured potential difference from the applied potential plotted against the applied potential for AM (warm colors) and FM (cool colors), black represents an ideal measurement. Insert in the lower right visualizes the fraction of the potential captured by the respective method. Legend in the upper left shows the offset in brackets. Data shown captured on the electrodes 1/2 (left) and 3/4 (right) of the electrode structure without an additional electrostatic force.

Figure 6

Comparison of the deviation of the measured potential difference from the applied potential plotted against the applied potential for AM (warm colors) and FM (cool colors), black represents an ideal measurement. Insert in the lower right visualizes the fraction of the potential captured by the respective method. Legend in the upper left shows the offset in brackets. Data shown captured on the electrodes 1/2 (left) and 3/4 (right) of the electrode structure with an additional electrostatic force.

Figure 7

Comparison of the absolute measured potential plotted against the applied potential for AM (warm colors) and FM (cool colors). Legend in the upper left shows the measured CPD at 0 V applied voltage in brackets. Data shown captured in the middle of the electrode structure (left) and on the outer most electrodes (1/2) (right) without an additional electrostatic force.

Figure 8

Comparison of the absolute measured potential plotted against the applied potential for AM(warm colors) and FM(cool colors). Legend in the upper left shows the measured CPD at 0 V applied voltage. Data shown captured on the outer most electrodes (1/2) of the electrode structure with an additional electrostatic force.