Material modifications are often required when implementing graphene in devices for nanoelectronics or energy conversion. These modifications generally take the form of adsorption or covalent binding, however, local surface inhomogeneities of pristine graphene (for example, wrinkles) may impact their uniformity.
Accurate nanoscale characterization of the graphene topography is essential, and this must be performed in combination with an assessment of the material’s functional properties.
Atomic force microscopy (AFM) is ideal for this task, combining real-space topography imaging with accurate detection of functional surface properties; for example, the adhesion, electronic potential, and modulus. AFM facilitates a holistic approach to the characterization of graphene and other 2D materials at the nanoscale.
Unique properties emerge when materials are reduced to two dimensions, and as such, graphene exhibits a range of exceptional physical characteristics. These include excellent charge carrier dynamics, high mechanical strength and high thermal conductivity.
There are a number of potential future applications of 2D materials such as graphene—these range from optoelectronics to flexible electronics and electrochemical energy storage. The light weight and low dimensionality of 2D materials have also invited the attention of nanoelectronics researchers looking into the ongoing downscaling of electronic devices.1
Industrial applications of graphene in electronic devices necessitate the use of large wafer-scale graphene films, prompting researchers to focus on improving monolayer growth procedures. Chemical vapor deposition (CVD) on catalytic copper (Cu) has proven to be the most explored route throughout this research.
Graphene growth on Cu substrates requires a subsequent transfer onto an insulating substrate, however, and this process can damage the monolayer, introducing contaminations. A process for direct growth on insulating substrates is, therefore a vital step towards the development of future graphene applications.2
This article outlines the study of wafer-scale graphene grown on LED grade c-plane sapphire within an AIXTRON CCS R&D reactor.
AFM’s versatility makes it uniquely suitable for investigations into the morphology and functional properties of CVD-grown graphene on insulating sapphire. Sideband Kelvin Probe Force Microscopy (KPFM) on a Park Systems NX20 AFM was therefore employed in the characterization of graphene on sapphire.
A distinct contrast was observed in the surface potential between graphene and the underlying sapphire surface. Variations of surface potential over graphene wrinkles and step edges were also noted.
Figure 1 displays a 3D overlay of topography and surface potential, clearly showing the correlation of topographic features and the KPFM surface potential. This overlay confirms a significantly lower surface potential on graphene wrinkles and around sapphire steps when compared to that of the sapphire terraces. Surface potential distribution also corresponds to mechanical features resolved using Park’s PinPoint nanomechanical mode.
This correlation between the KPFM and nanomechanical signals suggests a potential connection between graphene’s electronic and mechanical properties, demonstrating AFM’s potential as a holistic characterization technique that is ideal for 2D materials.
Figure 1. 3D overlay of the graphene topography, which displayed wrinkles and underlying sapphire steps as indicated, with the surface potential imaged via sideband KPFM. Image Credit: Park Systems Europe
Surface Potential Imaging via Sideband KPFM
KPFM is an AFM technique that can capture surface topography and surface potential simultaneously. When a conductive cantilever and conductive or semiconductive sample are electrically connected, their Fermi levels will align, giving rise to a contact potential difference (CPD) that corresponds to the material’s work functions. This CPD between tip and sample will also introduce an electrostatic force.
The conductive cantilever scans the surface in KPFM, while simultaneously applying an AC voltage in order to detect local changes in this electrostatic force. These changes may be caused by local variations of the CPD, determined by the surface potential distribution.3 A DC bias counteracts the CPD at each point of the scan in order to measure the surface potential locally.
Based on this applied DC bias, the sample’s surface potential distribution is reconstructed in the KPFM signal. The electrostatic force detection method will determine the resolution and accuracy of the surface potential in KPFM.
Figure 2. Schematic frequency spectra of the cantilever oscillation for off-resonance KPFM in a) and sideband KPFM in b). Image Credit: Park Systems Europe
The off-resonance KPFM method involves the AC voltage modulating electrostatic force at a frequency far from the resonance of the cantilever. This method is commonly used for topography imaging. The schematic frequency spectrum of cantilever oscillations displayed in Figure 2a visualizes the separation of mechanical and electrical excitation at frequencies f0 and fAC, respectively.
Electrostatic force is detected via the oscillation amplitude at the AC frequency. The application of a DC bias matching the potential difference between tip and sample the amplitude at the AC frequency allows the electrostatic force to be nullified.
The dependence of the KPFM signal on the long-ranged force lowers the sensitivity of the measurement, however. Non-local interactions between sample and cantilever superimpose on the local signal, and this, in turn, hinders the accuracy and spatial resolution of surface potential when this is measured via off-resonance KPFM.3
Park Systems has integrated a user-friendly sideband KPFM mode into their NX research AFMs in order to enhance the accuracy and resolution of the KPFM signal. In sideband KPFM, a low-frequency AC voltage (2-5 kHz) is applied to the tip in order to modulate the electrostatic force gradient.
Modulating the force gradient introduces frequency sidebands on the left and right of the cantilever resonance. This is shown in Figure 2b in the schematic cantilever frequency spectrum. Much like off-resonance KPFM, the feedback of sideband KPFM nullifies the amplitude of these sidebands through the application of a DC bias matching the potential difference at each individual measurement position.
By detecting the short-range force gradient rather than the force, long-range crosstalk is reduced and both the lateral resolution and local potential sensitivity are considerably enhanced.3
Figure 3. Topography and surface potential captured via sideband KPFM on CVD-grown graphene on a sapphire substrate. The line profiles of the topography in green and the surface potential in red show a correlation of the two signals with a distinct potential contrast between underlying sapphire steps and terraces as well as graphene wrinkles. Image Credit: Park Systems Europe
A distinct potential contrast that directly correlated with the topography of the sample (Figure 3) was resolved. This was achieved by measuring sideband KPFM on the CVD-grown graphene film on a sapphire substrate. The topography signal displayed underlying sapphire terraces and varying step heights of up to 3 nm. Graphene wrinkles with heights ranging between 0.5 and 2 nm were also noted.
Each sapphire terrace also included finer substructures with elevations of 0.6 nm in proximity to the step edges. These were visible in the enlarged image detail (Figure 3).
Simultaneously recorded surface potential showed a distinct contrast between two discrete states. A low potential state was observed at the position of the elevations close to the sapphire steps and on the graphene wrinkles, while a high potential state was observed on the bulk film on underlying sapphire terraces.
The wrinkle captured in the line profile of Figure 3 led to the detection of a potential contrast of around 0.5 V, regarding the bulk film present on the sapphire terraces.4 The potential on the elevations near the sapphire steps, however, differed by approximately 0.7 V from the potential of the bulk film on the terraces.
Correlation of Surface Potential and Nanomechanics
The sample’s adhesion force and modulus were imaged at the same position using Park System’s PinPoint nanomechanical mode. This was done in order to further investigate the graphene wrinkles, the elevations close to the sapphire steps, and their surface potential contrast.
Figure 4. Schematic diagram of Park Systems’ PinPoint nanomechanical mode. In this force spectroscopy technique, the tip approaches the sample and retracts at each pixel before moving to the next pixel, as indicated by positions 1, 2, 3 and 4. The resulting force curves and their automated analysis allow real-time visualization of nanomechanical properties including adhesion force, modulus, stiffness, and deformation. Image Credit: Park Systems Europe
PinPoint provides precise quantitative nanomechanical images through the use of fast force spectroscopy mapping (Figure 4). The cantilever approaches and retracts at each individual pixel of the whole scan area, simultaneously acquiring nanomechanical information and 3D topography of the sample surface.
The XY scanner stops at each pixel, allowing high-speed force-distance curves to be acquired with well-defined control of contact force and contact time between the tip and the sample. Automated analysis of each force curve facilitates real-time visualization of sample modulus, deformation and adhesion force at the same time as topography imaging is displayed.
Figure 5. Adhesion force and modulus acquired on graphene on a sapphire substrate via Park Systems’ PinPoint nanomechanical mode and the corresponding surface potential imaged via sideband KPFM at the same measurement area. The white box highlights the same sapphire terrace featuring a higher adhesion force, deformation and surface potential compared to surrounding sapphire steps. Image Credit: Park Systems Europe
Adhesion force and modulus information acquired via PinPoint’s nanomechanical mode reveals a contrast between the graphene wrinkles and the underlying sapphire terraces and steps (Figure 5). Elevations in proximity to the sapphire steps also featured a decreased surface potential, therefore exhibiting a lower adhesion force coupled with a higher modulus.
This lower adhesion and higher modulus indicate that the graphene film may become harder at these points. On the sapphire terraces, however, adhesion force increases while modulus decreases. This correlation of nanomechanical properties with surface potential and sample topography suggests a potential connection between the material’s electronic and mechanical properties.
Summary
The range of AFM techniques (such as sideband KPFM and PinPoint nanomechanical mode) offered via Park Systems’ research AFMs enables the holistic, in-depth characterization of 2D materials. This was demonstrated here via an investigation into wafer-scale CVD-grown graphene on sapphire that had been produced in an AIXTRON CCS R&D reactor.
The measurements illustrated a distinctive correlation between the sample’s surface potential and the adhesion force and modulus, therefore suggesting a link between the mechanical and electronic properties of graphene. Through proper characterization of the highly localized mechanical and electronic properties of 2D materials, it will be possible to effectively tailor the development of future nanoelectronics applications.
References
- S. J. Kim, K. Choi, B. Lee, Y. Kim and B. H. Hong, Annu. Rev. Mater. Res., 2015, 45, 63–84.
- N. Mishra, S. Forti, F. Fabbri, L. Martini, C. McAleese, B. R. Conran, P. R. Whelan, A. Shivayogimath, B. S. Jessen and L. Buß, Small, 2019, 15, 1904906.
- U. Zerweck, C. Loppacher, T. Otto, S. Grafström and L. M. Eng, Phys. Rev. B, 2005, 71, 125424.
- S. Samaddar, J. Coraux, S. C. Martin, B. Grévin, H. Courtois and C. B. Winkelmann, Nanoscale, 2016, 8, 15162–15166.
Acknowledgments
Produced from materials originally authored by Ilka M. Hermes,1 Simonas Krotkus,2 Ben Conran,3 Clifford McAleese,3 Xiaochen Wang,3 Oliver Whear,3 Michael Heuken2.
1 Park Systems Europe GmbH, Mannheim, Germany
2 AIXTRON SE, Herzogenrath, Germany
3 AIXTRON Ltd, Cambridge, United Kingdom
This information has been sourced, reviewed and adapted from materials provided by Park Systems Europe.
For more information on this source, please visit Park Systems Europe.