AFM can be used to image atoms on the surfaces of both insulators and conductors. The basic process involves scanning a sharp metal tip across the sample to produce images based on the balance of tiny forces between the tip and the sample. To obtain really high-resolution images, however, the microscope tip needs to be moved to within 1 nm of the sample surface but at this tiny range the AFM tip and sample can interact via weak electrostatic attractions, or Van der Waals forces.

A team led by Peter Liljeroth of the Aalto University School of Science in Finland and Daniel Vanmaekelbergh of Utrecht University in the Netherlands has shown how the atomic contrast in AFM images of graphene depends on how reactive the atoms on the AFM tip are. The team compared the images obtained using two different tip apexes. In the first experiment, a reactive tip terminated by iridium atoms was used. In the second experiment, the tip was chemically passivated by controllably picking up a single carbon monoxide molecule.

With the reactive tip, the researchers found that both attractive and repulsive forces between the atoms on the AFM tip apex and on the graphene samples produced the atomic contrast observed. Conversely, the image obtained using the nonreactive tip depended only on repulsive interactions. The results from the experiments could help clarify many existing, and conflicting, previous reports where researchers had assigned features in AFM images of graphene to either the atoms or the hollow sites in the samples without really knowing which was which, says the team.

Controlling tip apex atoms

“The key to our experiments was to control the type of atoms at the AFM tip apex, so that we could tune its chemical reactivity,” explained Liljeroth. “Such an approach allows us to select the type of interaction (chemical bonding or Pauli repulsion) between the AFM and the graphene, which in turn determines the atomic-scale contrast in the final AFM images.”

The images were obtained using a low-temperature (5K) ultrahigh vacuum scanning tunnelling/atomic force microscope based on a “quartz tuning fork” (or Qplus) force sensor. The conditions employed mean that both STM and AFM can be performed on the same atomic region in the sample. “An added benefit of the Qplus force sensor is that it allows for non-contact AFM experiments in which the tip oscillates only a little – by less than 1 angstrom in fact – which boosts the sensitivity of the instruments to short range forces,” said Liljeroth.


The technique can be used to resolve the structure of graphene nanostructures and edges with greater precision, he adds, something that is difficult to do without the detailed understanding of the contrast formation mechanisms at play in AFM imaging. “Such precision is not possible using an STM alone because of the contrast originating from variations in the local density of states that does not correctly reflect true atomic positions. It is also very hard to achieve using an AFM if you have a reactive tip apex.”

There is nothing to prevent researchers using the new AFM imaging technique in conjunction with STM, which, for its part, can be used to analyse the electronic properties of a sample, he says. “On the contrary, a quantitative link between atomic and electronic structures is crucial for testing advanced theoretical ideas on how to use graphene nanostructures for next-generation electronics,” he told

The team is now busy studying graphene nanoribbons using its technique. “We have quite a few questions to answer – for example, about the magnetism of certain kinds of graphene edges and how it depends on the atomic-scale details of these structures.”

The current work is detailed in ACS Nano.