Scanning Kelvin probe microscopy (KPFM) is a technique used for determining work function as well as the local electrical potential distribution of numerous materials that have nanoscale features.
This technique is mostly used for analyzing the electronic characteristics of semiconductor surfaces and nanostructures. Some of the measurements that can be made using this method with nanoscale lateral resolution include potential profiling of semiconductors composed of positively and negatively charged substances. The application of this method to expose the differences in charge distribution occurring on the nanoscale can help in interpreting and enhancing the performance of CMOS semiconductor devices to a greater extent [1–2]. Moreover, KPFM has been helpful in defining and establishing the quantitative physical data such as the charge position and total charge of polymer materials [3, 4].
Other work function or surface potential measuring tools, like photoelectron spectroscopy (PES), scanning electron microscopy (SEM), and electron-beam-induced current (EBIC), were introduced. Yet, some of these methods are only relevant for inorganic-type semiconductor samples — some samples do not offer a high spatial resolution, while others are destructive and need high vacuum [5]. KPFM is an air-ambient, compatible and nondestructive method when compared to other methods, and is clearly one of the most user-friendly electrical failure analysis and nanometrology tools to be currently available in the market.
When compared to other measurement methods, traditional KPFM technique, also called Amplitude Modulation AM-KPFM, has already contributed immensely in preserving device reliability and inspection of sophisticated materials. However, there is more scope for improving the AM-KPFM’s signal-to-noise detection ability so that even the features that have less optimal surface potential strength can be identified and mapped out with higher accuracy and lateral resolution.
For these reasons, Park Systems recently devised a new method known as Frequency Modulation FM-KPFM with atomic force microscope (AFM). Useful for electrical characterization, this method enables measurements with better sensitivity when compared to the AM-KPFM technique. In this analysis, a polymer material was determined to compare the performance of the AM-KPFM method with that of FM-KPFM. The results obtained in the experiment demonstrate that FM-KPFM is considerably more sensitive when compared to the AM-KPFM method in determining work function as well as the surface potential distribution of varied materials.
Experiment
A Park NX10 AFM was used to examine a polymer patterned array material deposited on a silicon substrate. FM-KPFM and AM-KPFM measurements were acquired by performing two separate scans. The same tip and scan parameters were used for acquiring images. For the experiment, a conductive Mikromasch NSC36Cr-Au cantilever (nominal spring constant k = 1 N/m with resonant frequency f = 90 kHz) was employed.
In KPFM mode, two interaction forces exist between the sample and the AC biased tip — the Van der Waals force and the electrostatic force. The former force is harnessed to produce the surface topography of the sample, whereas the latter force existing between the sample and the tip produces data for the electrical properties of the sample. The cantilever deflection signal, thus obtained, includes both sets of data; therefore a technique that can fully isolate these signals is important for effective imaging.
Lock-in amplifiers within the Park NX10 are integrated into its electronics and are used for isolating the signals. This makes it possible to acquire both EFM and topography data at the same time. The system uses two amplifiers called Lock-in 1 and Lock-in 2. The former acquires the topography data by examining the tip motion induced by the Van der Waals interaction, whereas the latter acquires electrical property data by inspecting the frequency of the applied AC voltage signal to the tip that, consecutively, produces an electrostatic force interaction with the sample. Care was taken to make sure that the frequency of the applied AC voltage signal is sufficiently smaller (5–20 kHz) than the cantilever oscillation frequency (70–330 kHz) so that the two signals do not obstruct one another.
In FM-KPFM setup [6], the NCM phase signal of the Lock-in 1 amplifier is conveyed to the Lock-in 2 amplifier to act as a source for EFM operation by joining a BNC cable between “Aux1 in” and “Aux2 out,” as depicted in Figure 1.
.jpg)
Figure 1. Diagram of FM-KPFM
A separate DC bias was also applied to the cantilever and regulated to produce a feedback loop that would cancel out the electrical oscillation between the sample and the tip induced by applying an AC bias to the cantilever. Therefore, the value of this offsetting DC bias that annuls the electrical oscillation caused by AC bias is regarded as a measure of surface potential.
Results and Discussion
In this experiment, the resultant topography information demonstrates that the polymer patterned array with square-like features was effectively deposited on the silicon substrate but does not show any considerable information in relation to its surface potential. On the contrary, the surface potential data obtained in the FM-KPFM and the AM-KPFM techniques reveals the surface potential structure but does not show any major information in relation to the sample’s physical structure.
The domain structure shape visualized in the sample’s KPFM data was found to be comparable to the physical structure seen in the topography data, which was a patterned array containing square dots. Here, Park Systems’ Park XEI software was used to examine the data obtained in this experiment.
The AM-KPFM (right), FM-KPFM (center), and topography (left) images of the polymer patterned array sample are shown in Figure 2. The topography data can be obtained at the same time as the KPFM data. The topography image shown in Figure 2 was obtained simultaneously with the AM-KPFM image.
.jpg)
Figure 2. 10 x 10 μm image of polymer patterned array. Topography image (left), FM-KPFM image (center), and AM-KPFM image (right).
A distinct image of a well-defined lattice structure is shown by this topography. The acquired signals are mapped into a color table by the Park XEI software. The square-like features in the topography image seem to have a brighter color and indicate the greater height areas, whereas the flat surface with darker color indicates the lower height areas.
The quantified peak to valley in the topography image was about 150 nm. By contrast, both FM-KPFM and AM-KPFM data demonstrates a patterned array that has an irregularity on the surface (indicated by red arrows). The square-like features in both KPFM images seem to have a darker color and indicate the regions with comparatively lower surface potential, whereas the flat surface with brighter color indicates the regions with higher surface potential.
The surface potential results obtained from FM-KPFM and AM-KPFM can be compared to easily establish the fact that the FM-KPFM method has better sensitivity in identifying surface potential variation when compared to the AM-KPFM technique. The FM-KPFM technique used in this experiment offers a higher resolution image that displays sharper edges of the square features when compared to the AM-KPFM method. FM-KPFM was also able to identify weak potential in the uneven surface, whereas AM-KPFM failed to detect any potential difference.
Furthermore, the line profile produced by the XEI software in Figure 3 (bottom) supplies the potential data of the polymer patterned array.
.jpg)
Figure 3. FM-KPFM image (top-left), AM-KPFM image (top-right), and corresponding line profile (bottom) acquired from the polymer sample. FM-KPFM line profile (red line, y-axis on left) and AM-KPFM line profile (green line, y-axis on right).
For the FM-KPFM method, the line profile is denoted in red color, while for the AM-KPFM, the line profile is indicated in green color. In addition, the line profile of the FM-KPFM image eventually demonstrates that there is a small reduction in potential (encircled in red) in the uneven surface when compared to the areas adjacent to it. The line profile of the AM-KPFM method, on the other hand, fails to reveal any changes of potential in that specific surface. Comparison of the line profile of both methods shows that the FM-KPFM method can be said to be relatively more sensitive than the traditional AM-KPFM method.
Conclusion
The Park NX10 AFM utilizing AM-KPFM and FM-KPFM imaging enabled effective characterization of the polymer patterned array. The topography data showed that the surface of the sample contains a patterned array with square-like features. Both FM-KPFM and AM-KPFM images showed some amount of unevenness in the patterned array. The FM-KPFM method demonstrates that the surface pattern’s irregularity is a surface that has weak potential. The AM-KPFM method, by contrast, was not sufficiently sensitive to identify the weak potential in the same area.
Considering the above outcomes, FM-KPFM can be concluded to be an order of magnitude more sensitive than the AM-KPFM method. It is also useful for detecting differences in surface potential. The technique’s increased sensitivity will also increase the likelihood of identifying defects with better accuracy and resolution in various sophisticated materials, including semiconductor devices.
References and Further Reading
- LanFei (2018) FUNDAMENTALS OF KELVIN PROBE FORCE MICROSCOPY AND ITS APPLICATIONS IN THE CHARACTERIZATION OF SOLAR CELLS. Doctoral Dissertation, University of Pittsburgh.
- J. Pineda, et al., Electrical Characterization of Semiconductor Device Using SCM and KPFM Imaging.
- J. Gonzalez, et al., Charge distribution from KPFM images, PCCP, Issue 40, 2017.
- 4. M. Ortuño, et al., Conducting polymers as electron glasses: surface charge domains and slow relaxation, Scientific Reports volume 6, Article number: 21647 (2016).
- W. Melitz, et al., Kelvin probe force microscopy and its application, Surface Science Reports 66 (2011) 1–27.
- Charles Kim, et al., How to Measure FM-KPFM, Park Systems White Paper, To be published (2018).
This information has been sourced, reviewed and adapted from materials provided by Park Systems.
For more information on this source, please visit Park Systems.