Using LFM to Determine the Composition of Materials

Over the last few decades, the recording of nanoscale frictional forces has received the attention of many researchers. This is due to the rise in demand for high resolution mechanical characterization of numerous recently-designed materials. Lateral force microscopy (LFM) [1] is notably practical for detecting and locating heterogeneities in surface frictional properties, and an increasing amount of applications are constantly being discovered.

Examples include the delineation of surface coverages of deposited materials on thin films, recognizing different compounds that make up polymer blends and composites, and the mechanical testing of micro- and nano-electrochemical systems (MEMS/NEMS) [2-7]. NanoScientific’s research detailed below provides evidence for the practicality of LFM with Park atomic force microscopy (AFM) for the detection of surface compositional discrepancies between two samples.

Prior to this, to provide evidence that supports the theoretical principles of the experiment, researchers used a sample holding a polymer deposited onto a glass substrate and used LFM to show the distinct differences in frictional properties from the LFM measurements.

Then, researchers looked at graphene grown on Si using LFM and scanning thermal microscopy (SThM), in order to obtain information not only on the frictional properties of the sample but also the thermal properties.

Principal of LFM

The way that LFM works is closely related to that of Contact Mode, where the deflection of the cantilever in the vertical plane is recorded and exploited to produce surface topography. This is in contrast to LFM, which has the deflection of the cantilever in the horizontal plane. The lateral deflection of the cantilever is produced by the force met by the cantilever as it travels horizontally across the sample surface.

Aspects that decide the degree in which this deflection occurs are: the topography of the sample surface, the frictional coefficient, the cantilever lateral spring constant and the direction of the cantilever movement.

When carrying out LFM, the cantilever movement in both the horizontal and vertical direction is monitored by a position sensitive photodetector (PSPD) which is made up of four compartments (known as a quad-cell), and is illustrated in Figure 1.

In order to acquire topographical details of the surface, a ‘bi-cell’ signal in relation to the difference between the cells on top (A+C) and the bottom cells (B+D) recorded from the quadrant detector is needed (Equation 1), commonly referenced as an ‘A-B’ signal.

Topographic information = (A+C) – (B+D) (Equation 1)

In contrast, to obtain the surface frictional properties, the LFM signal is taken from the difference between the right cells (A+B) and the left cells (C+D) (Equation 2).

Frictional information = (A+B) – (C+D) (Equation 2)

Figure 1. Schematic illustration of laser position on PSPD in the operation of AFM (top) and LFM (bottom).

Figure 2. (a) Schematic illustration of the sample of interest, (b) Cantilever deflection during scanning, (c) Line scan of topographical signal; Line scan of LFM signal when scanning from left to right (d) and right to left (e).

To provide a more direct illustration of how LFM works, Figure 2 shows a diagram of the structure of cantilever deflection during a scan by cantilever of a surface with various frictional areas. In Figure 2(a) the surface shown includes a raised step in the middle and smooth areas on the outer edges of the surface, and an area of increased frictional coefficient situated in the flat part to the left hand side.

Figure 2(b) shows the deflection of the cantilever as it traverses across the surface and meets the topographical properties in addition to heterogeneous frictional areas. Figure 2(c) gives an illustrative line profile of the AFM topographical signal.

When only looking at the topographical information provided in the AFM signal, it is easy to draw the conclusion that the properties of the surface do not change between points one and two, however when also taking into account the LSM signal, a distinct difference between these two points is observed (Figure 2(d) and 2(e)).

As a result of a rise in corresponding friction when passing across this area, the cantilever will alter its angle to the right as it scans from the left-hand side to the right-hand side and produces a higher signal in the LFM measurements.

Then when the scan direction is backwards, the cantilever alters its positioning to the left and a reduction in the LFM signal is recorded. Reaching the full potential of LFM is dependent on how well it detects different parts contained in the sample, relying on frictional properties when the surface is reasonably flat, enabling the researcher to increase the amount of information known about the surface.

Procedure of AFM Imaging

Two samples were selected and imaged in ambient conditions with a Park NX10 AFM. Sample 1 consists of a polymer deposited on a glass substrate. Sample 2 contains graphene grown on Si. Images for Sample 1 were acquired in LFM Mode using a scan rate of 1.0 Hz. Images for Sample 2 were acquired in both SThM Mode and LFM Mode with a scan rate of 0.6 Hz. In LFM imaging, a NSC36-C cantilever (nominal spring constant k = 1 N/m) was used. In SThM imaging, a NanoThermal-10 (nominal spring constant k = 0.25 N/m) was used.

Results and Discussion

Polymer on Glass

To demonstrate the performance of the LFM mode by Park AFM, Sample 1 was first imaged as a proof-of-principle experiment, with the topography image, LFM images and the representative line profiles shown in Figure 3. Images were acquired at a pixel size of 256 × 256 and a scan size of 20 µm × 20 µm.

From the AFM topography image (Figure 3a), circular features were observed with diameters ranging from 1 to 2 µm and heights of 20 to 200 nm within the sample. In Figure 3a, red arrows labeled as 1, 2, and 3 were added to three representative circular features for better illustration. In terms of the feature dimension, diameters of 1.72 µm, 1.41 µm, and 1.33 µm were measured for features 1, 2, and 3.

The heights of each of the three features were 50.1 nm, 158.2 nm and 42.1 nm, respectively. The distribution of such features is relatively homogeneous throughout the sample. In addition to the appearance of the circular features, slight height variations were also observed within the area under investigation, i.e., ∆Z = 29.2 nm between the two triangular cursors as indicated in Figure 3a.

LFM forward (Figure 3b) and backward (Figure 3c) images were captured simultaneously along with the topography image, through which the frictional characteristic of the sample can be obtained. The LFM images revealed the compositional heterogeneity within the sample, and two domains with distinct frictional coefficients were observed.

To give a more direct comparison of the signal, line profiles along the solid lines drawn in the topography image (red line in Figure 3a), the LFM forward image (green line in Figure 3b), and the LFM backward image (blue line in Figure 3d) were generated with the XEI software and the three traces are displayed in Figure 3d.

From the line profiles obtained from the LFM forward scan (green line in Figure 3d) and backward scan (blue line in Figure 3d), a qualitative idea can be gained regarding the frictional characteristic of the sample. In summary, a downward shift in the LFM signal was observed during the forward scan (green line in Figure 3d), indicating that the movement of the cantilever was hindered by the underlying substrate due to frictional force.

The cantilever was dragged by the surface and eventually a backward torsion occurred as it scanned from left to right, which was then observed as a negative shift in the lateral deflection signal. Oppositely, the LFM signal shifted upwardly during the backward scan (blue line in Figure 3d), which, again, is a result of the cantilever being dragged by the surface because of the larger frictional interaction between the cantilever and the surface. As a result, it can be concluded that the frictional coefficient for the central area is higher as compared to the surrounding areas.

Figure 3. (a) Topography image, red arrows were added to better illustrate representative circular features;
(b) LFM forward scan image;
(c) LFM backward scan image;
(d) line profiles plotted along the red line seen in 3a, green line seen in 3b and blue line seen in 3c, respectively.

Another interesting finding is, although drastic changes were seen in both the green trace (LFM forward) and the blue trace (LFM backward), only minor height variations were observed in the red trace (topography). Results here showcased the strength of LFM to identify different components within a sample based on the frictional properties even when no significant difference can be seen from the topographical data.

Graphene on Si

Upon demonstration of the LFM operation with Sample 1, the LFM measurements were then repeated on Sample 2 to examine the difference in frictional characteristics between graphene and Si. Also, SThM imaging was conducted to study the thermal properties of the two materials. Images were acquired with a pixel size of 256 × 256 and a scan size of 15 µm × 15 µm.

In Figure 4a, topography of Sample 2 is shown. The boundary between graphene and Si, as indicated by the white dashed line, is discernable. A representative line profile was plotted along the red line drawn in Figure 4a and is shown in Figure 4d (red trace), a ∆Z of ~5 nm was measured between graphene and Si.

From the LFM image in Figure 4b, the two materials were clearly differentiated. From the line profile (green trace in Figure 4d) of the LFM signal, a larger frictional coefficient was observed for Si compared to that of graphene, as evident by the downward shift in the LFM signal over Si as compared to that over graphene.

According to previous literature, the nominal friction coefficient for graphene is 0.03,8 whereas the nominal friction coefficient for Si is 0.2.9 Our LFM results are consistent with previous literature.

Furthermore, to gain insights about the thermal properties of the two materials, SThM was performed, with the resultant SThM error image shown in Figure 4c and a representative line profile (blue trace) shown in Figure 4d. A higher SThM error was observed over Si than graphene, which indicates a higher thermal conductivity of graphene as compared to that of Si.

Figure 4. (a) Topography image; (b) LFM image and (c) SThM image of Sample 2 (graphene on Si); (d) Line profiles plotted along red line seen in 4a, green line seen in 4b, and blue line seen in 4c.

Conclusion

Here NanoScientific demonstrates the use of LFM to differentiate surface compositional variations based on the relative differences in frictional properties. Implemented with Contact Mode AFM imaging, this technique enables nanoscale characterization of frictional domains within a sample.

The strength of LFM is first shown in Sample 1, where the two different materials (i.e., polymer and glass) within the sample were not easily distinguishable from the topography image. However, from the LFM images, the two domains were clearly separated by their difference in frictional coefficient.

Then both LFM and SThM were applied to examine the frictional as well as thermal properties of Sample 2, which is graphene grown on Si. Qualitative results show that graphene has a lower frictional coefficient and a higher thermal conductivity as compared to that of Si.

In summary, LFM has already found a wide range of applications in nanoscale frictional measurements and will continue to facilitate the development of many exciting technologies.

Acknowledgments

Wenqing Shi, Gerald Pascual, Byong Kim, and Keibock Lee, Park Systems Inc., Santa Clara, CA USA

References and Further Reading

1. Mate, C. M., McClelland, G. M., Erlandsson, R., & Chiang, S., Atomic-scale friction of a tungsten tip on a graphite surface. Phys. Rev. Lett., 1987, 59, 1942.
2. Sheiko, S. S. Imaging of polymers using scanning force microscopy: from superstructures to individual molecules. New Developments in Polymer Analytics II. Springer Berlin Heidelberg, 2000. 61-174.
3. Perry, S. S. Scanning probe microscopy measurements of friction. MRS bulletin, 2004, 29, 478-483.
4. Perry, S. S., &Tysoe, W. T. Frontiers of fundamental tribological research. Tribology Letters, 2005, 19, 151-161.
5. Munz, M., Schulz, E., & Sturm, H., Use of scanning force microscopy studies with combined friction, stiffness and thermal diffusivity contrasts for microscopic characterization of automotive brake pads. Surface and interface analysis, 2002, 33, 100-107.
6. Raczkowska, J., Montenegro, R., Budkowski, A., Landfester, K., Bernasik, A., Rysz, J., &Czuba, P., Structure evolution in layers of polymer blend nanoparticles. Langmuir, 2007, 23, 7235-7240.
7. Sun, S., Chong, K. S., Leggett, G. J., Photopatterning of self-assembled monolayers at 244 nm and applications to the fabrication of functional microstructures and nanostructures. Nanotechnology, 2005, 16, 1798.
8. Shin, Y. J., Stromberg, R., Nay, R., Huang, H., Wee, A. T., Yang, H., & Bhatia, C. S., Frictional characteristics of exfoliated and epitaxial graphene. Carbon, 2011, 49, 4070-4073.
9. Deng, K., Ko, W. H., A study of static friction between silicon and silicon compounds. Journal of Micromechanics and Microengineering, 1992, 2, 14.

This information has been sourced, reviewed and adapted from materials provided by Park Systems Inc.

For more information on this source, please visit Park Systems Inc.