We present the application of local ultrahigh pressures (> 10 GPa) with diamond Atomic Force Microscopy (AFM) tips for a controlled tuning of graphene electronics. It enables the creation of very stable p-doped regions with nanometer precision. This methodology exploits further the application of AFM and opens the door to tunning the electronics of other 2D materials with a simple experimental set up.
The Potential of AFM to Reach Ultrahigh Pressures
Most of the knowledge acquired by the human being has been gained through studying the nature at or near one atmosphere, which is the pressure at the Earth’s surface. Nevertheless, much of the matter in the Universe exists under much more pressurized conditions, as for example deep inside planets and stars. The most common device employed to reach ultrahigh pressures is the diamond anvil cell , which is typically used to synthesize materials and phases not observed under normal ambient conditions. It compresses a submillimeter-sized sample between two opposing diamond polished tips, being able to reach pressures typically up to around 100-200 GPa and up to 750 GPa in special set ups.
An AFM with a diamond tip can be seen as a nanotechnology version of the diamond anvil cell. Thanks to the final radius of the AFM tips (tipically tens of nanometers or below), ultrahigh presssures can be achieved with relative low forces. Using commercially available diamond AFM probes, we can estimate that pressures around 300 GPa could be reached. This will enable ultrahigh pressure studies with a widespread tool such as an AFM.
Graphene unique combination of extraordinary strength, stiffness and high flexibility have generated great expectations in applications such as flexible electronics, electromechanical devices, protective coatings or as a reinforcement agent. These outstanding mechanical properties make graphene an ideal material to withstand ultrahigh pressure. By applying ultrahigh pressures with diamond AFM tips we have locally tuned the electronic properties of graphene on silicon dioxide substrates with nanometer precision, creating selective p-doped areas .
A direct application is the fabrication of graphene regions with low electrical contact resistance, highly relevant for graphene electronics.
Graphene Modifications with Ultrahigh Pressures
Graphene flakes were deposited by microexfoliation on 300 nm SiO2/Si substrates. The modification of areas with ultrahigh pressures was done according to the following procedure : firstly the flake was imaged in non-contact dynamic mode. Then the AFM tip was brought into contact and the applied load was selected according to the desired pressure, scanning an area of interest under these conditions. Finally, the tip was brought back to non-contact dynamic mode and the flake was imaged again. Figure 1 depicts the procedure of a graphene area modified with ultrahigh pressure (fig. 1 a-c) and a topography image of areas in a same flake modified with different pressures (fig. 1d). These pressure-induced modifications are irreversible and stable over time.
Tunable Doping with Pressure
The degree of coupling between the graphene layer and the underlying substrate is an important factor on the electronic properties of graphene [3-5]. To asses the effect of the modifications at different pressures on the electronic properties, a battery of techniques were employed: Raman spectroscopy, Kelvin Probe Force Microscopy (KPFM) and Scanning X-ray Photoelectron Microscopy (SPEM). These measurements gave evidence of a tunable p-doping effect with increasing pressure , being proportional to the graphene-SiO2 distance variation within the modified areas. Figure 2a shows a comparison of the depth of the different modified areas with the KPFM and Raman measurements. A clear correlation can be observed between the depth (from AFM), the Fermi level shift (from KPFM) and the intensity ratio of the 2D and G peaks (from Raman) of the modified areas with the applied pressure: the higher the pressure, i) the deeper the area, which means that the adsorption strenght can be precisely controlled; ii) the higher the Fermi level shift down with respect to the Dirac cone vertex and iii) the lower the I(2D)/I(G) ratio. Both ii) and iii) indicate a fine tunable pressure-dependent p-doping effect due to a charge transfer [6-7]. These results are in very good agreement with the SPEM observations (fig. 2b), where the C 1s peaks from modified areas present a rigid shift to lower binding energies, reflecting a shift of the Fermi level towards the valence band, as expected for a p-doping effect.
Origin and Implications of the Permanent Changes in Height and Doping Level of the Modified Graphene Areas
Several effects could be behind the observed p-doping, such as charge accumulation, electrochemical strain effects or flexoelectricity. However, the timescales when they occur or the response of graphene and silicon dioxide to electric fields allow discarding them as the main cause of the observed effects.
The formation of very few strong covalent bonds between the graphene and the substrate would be enough to explain with a single hypothesis both the irreversible approach of graphene towards the substrate and the observed tunable doping. To verify this idea, firstly we analyzed the line shape of the C 1s core levels shown in fig. 2b. A first visual analysis of the C 1s line shape does not show significant changes. However, while the C 1s line shape of the peaks for pristine graphene can be fitted using a single component, it is needed an extra small second component to explain the line shape for graphene modified at higher pressures . This second component is compatible with carbon atoms in sp3 hybridization, in agreement with the covalent bonding hypothesis.
This is further supported by Density Functional Theory (DFT) calculations. We carried out DFT-based simulations considering graphene on different SiO2 surface terminations and we found a pressure threshold of the order of 10-20 GPa for the covalent bonding to occur , in good agreement with the experimental results. The DFT calculations pointed out as well the need of chemical bonding to obtain a p-doping effect as observed experimentally, with a remarkable low fraction of bonds being enough for both this charge transfer and to keep graphene modified areas close to the silicon dioxide substrate.
We can envisage some direct applications of these results. Thanks to the increase of the coupling strength between the graphene layer and the underlying substrate, this procedure can be used to improve the sealing of graphene blisters, achieving significant drops of the leak rates . This approach could have relevant implications in graphene membrane applications.
The change in the doping level implies that the modified areas present lower electrical contact resistances than pristine graphene, as verified by conductive AFM (C-AFM) (fig. 3). Modified areas present higher currents for the same applied voltages than pristine graphene, indicative of a decrease of the contact resistance between the metal tip and the modified areas . This result might allow a sizable power consumption reduction of future graphene-based electronic devices.
The modification of graphene under ultrahigh pressures with a diamond AFM tip might have a direct impact in different fields, such as graphene electronics or sensors. These results suggest local application of ultrahigh pressures with AFM as a very powerful tool to tune 2D materials and heterostructures properties. Furthermore, thanks to the remarkable mechanical properties of graphene, it can provide a platform to carry out chemical reactions at ultrahigh pressures on trapped self-assembled monolayers of molecules between graphene layers and a substrate.
Financial support from the Spanish Ministry of Economy and Competitiveness, through the “María de Maeztu” Programme for Units of Excellence in R&D (MDM-2014-0377).
Pablo Ares1 and Julio Gómez-Herrero2
1School of Physics and Astronomy & National Graphene Institute, University of Manchester, Manchester, UK
2Departamento de Física de la Materia Condensada & Condensed Matter Physics Center (IFIMAC), Universidad Autónoma de Madrid, Madrid, Spain
Dr. Pablo Ares
School of Physics and Astronomy
National Graphene Institute
University of Manchester
More information on Atomic Force
Read more about graphene imaging
 Aiyasami Jayaraman: Diamond Anvil Cell and High-Pressure Physical Investigations, Reviews of Modern Physics, 55, 65-108 (1983) doi: 10.1103/RevModPhys.55.65
 Pablo Ares, Michele Pisarra, Pilar Segovia, Cristina Díaz, Fernando Martín, Enrique G. Michel, Félix Zamora, Cristina Gómez-Navarro, and Julio Gómez-Herrero: Tunable Graphene Electronics with Local Ultrahigh Pressure, Advanced Functional Materials, 1806715 (2019) doi: 10.1002/adfm.201806715
 Jimmy Nicolle, Denis Machon, Philippe Poncharal, Olivier Pierre-Louis, and Alfonso San-Miguel: Pressure-Mediated Doping in Graphene, Nano Letters, 11, 3564-68 (2011) doi: 10.1021/nl201243c
 Sunmin Ryu, Li Liu, Stephane Berciaud, Young-Jun Yu, Haitao Liu, Philip Kim, George W. Flynn, and Louis E. Brus: Atmospheric Oxygen Binding and Hole Doping in Deformed Graphene on a SiO2 Substrate, Nano Letters, 10, 4944-51 (2010) doi: 10.1021/nl1029607
 Matthew Yankowitz, Jeil Jung, Evan Laksono, Nicolas Leconte, Bheema L. Chittari, Kenji Watanabe, Takashi Taniguchi, Shaffique Adam, David Graf, and Cory R. Dean: Dynamic band-structure tuning of graphene moiré superlattices with pressure, Nature, 557, 404-8 (2018) doi: 10.1038/s41586-018-0107-1
 Anindya Das, Simone Pisana, Biswanath Chakraborty, Stefano Piscanec, Srijan K. Saha, Umesh V. Waghmare, Konstantin S. Novoselov, Hulikal R. Krishnamurthy, Andre K. Geim, Andrea C. Ferrari, and Ajay K. Sood: Monitoring dopants by Raman scattering in an electrochemically top-gated graphene transistor, Nature Nanotechnoly, 3, 210-5 (2008) doi: 10.1038/nnano.2008.67
 Young-Jun Yu, Yue Zhao, Sunmin Ryu, Louis E. Brus, Kwang S. Kim, and Philip Kim: Tuning the Graphene Work Function by Electric Field Effect, Nano Letters, 9, 3430-4 (2009) doi: 10.1021/nl901572a
 Yolanda Manzanares-Negro, Pablo Ares, Miriam Jaafar, Guillermo Lopez-Polin, Cristina Gomez-Navarro, and Julio Gomez-Herrero, arXiv:1809.03786 (2018)