Physical Review Letters

In 1982 Gerd Binnig and colleagues successfully demonstrated that the surface topography of a material could be mapped with atomic resolution. Using their newly designed STM, the group imaged both gold and CaIrSn₄ surfaces. In STM, a conducting tip is moved back and forth over a surface. The tip is close enough to the surface that electrons are able to tunnel across the vacuum gap between the surface and tip, creating a current. By adjusting the height of the tip so that this current stays constant, a 3D height map of the surface can be generated. One year after this demonstration, Jerry Tersoff and Donald Hamann published a theory for STM, which they verified using data from the experiments of Binnig et al.

Binnig and Heinrich Rohrer went on to win the Nobel Prize in 1986 for their design of the scanning tunneling microscope.

One of the main advantages of STM is that it allows direct real-space imaging of surfaces. One of the first surface science questions that STM helped to solve was the unit cell structure of silicon (111), an important model system and a material that is today used in everything from mobile phones to solar cells. The images obtained by Binnig et al. for Si(111) reveal a rhombohedral unit cell with the atoms arranged in a sixfold rotationally symmetric pattern. As well as mapping out atomically flat surfaces, the technique can also be used to capture nanosized clusters and features that are many atoms thick, as demonstrated, for example, by Olga Dulub et al. STM is not confined to simply mapping geometric surface structures; the images can also contain information about electronic effects. Ulrike Diebold and colleagues, for example, studied oxygen vacancies that develop in titanium dioxide when the material is heated. They found that their STM images were dominated by electronic structure effects rather than by physical topography.

STM is typically performed under ultrahigh vacuum. In 2001, Lars Österlund and colleagues demonstrated a modification to the instrument that enabled it to operate at pressures of up to 1 bar. This update allowed a gas to be controllably introduced into the STM so that surfaces could be imaged while undergoing a chemical reaction at technologically relevant conditions. Österland et al. monitored a copper surface reacting with hydrogen gas. One year later, Bas Hendriksen and Joost Frenken used this new high-pressure STM to study the oxidation of platinum, and they observed new reaction mechanisms not seen at low pressures.

While STM was designed as an atomic resolution imaging technique, it is now employed for a whole host of applications, ranging from assembling individual atoms on a surface into specific patterns to measuring the lifetimes of surface electrons and imaging the electronic orbitals of individual molecules.

If the vertical gap between the STM tip and an atom is decreased, the tip-atom interaction increases. Tune this interaction carefully and the tip will drag an atom along as it is moves along the surface. A group at IBM first demonstrated single-atom positioning via STM by spelling out the company’s initials in xenon atoms adsorbed onto a nickel surface. Ludwig Bartels and colleagues later showed that atoms can also be pushed, slid, and forced to hop (skip over absorption sites) along a surface.

In 1986, Binnig, Quate, and Gerber combined the principles of STM with a stylus profilometer to create the AFM. Rather than the constant current used in STM, in AFM a constant force is maintained between the tip and the object being imaged as the tip is scanned over the object’s surface. While STM can only be used to image conductive samples, AFM does not require a current flow between the tip and the sample and can map a surface regardless of its conductivity. AFM has become a standard laboratory tool that is widely used to image not only inorganic materials but also biological samples such as individual proteins and DNA.

AFM has been widely studied from a theoretical point of view. The first theory papers looked at issues relating to the components used to make the AFM and how the AFM was operated. For example, Farid Abraham and colleagues studied how the AFM tip shape can alter the image that is produced. They found that the tip shape significantly influences the resolution of the image and that differently shaped tips can lead to slightly different outlines being obtained for the same object. If the force applied to the surface by the tip is too high then the tip can also change the morphology of the surface being imaged. Abraham et al. made predictions on how to avoid this.

An AFM can be operated in many different modes. One such mode is known as frequency modulation. In this mode, the cantilever, on which the tip is attached, is made to oscillate via an actuator. The change in the resonant frequency is monitored and used to reconstruct an image. Franz Giessibl theoretically determined how the frequency shift of the cantilever relates to the force between the tip and the sample. It turns out that the geometry of the tip also plays an important role here.

While the AFM was initially used to image hard surfaces in air or under vacuum, it can also be operated in liquids. Albrecht Weisenhorn and colleagues experimentally studied how the interaction between the tip and a sample surface changes depending on what liquid the sample is submersed in. They found that ethanol works well for imaging macromolecules—such as polymers, DNA, and proteins— adsorbed on a substrate, while pure water should be avoided.

In addition, more recent updates to the technique enabled imaging of soft materials. R. Mahaffy and colleagues modified the tip geometry by attaching a polystyrene sphere to its end. This made it possible to probe more delicate and thinner samples and to measure the local frequency-dependent viscoelastic behavior of polymer gels and individual fibroblasts.

The forces involved in AFM are not only useful for imaging. They can be used to detect small modifications in the surface energy of a material, to study the mechanical properties of monolayers, and to manipulate single atoms.

Nancy Burnham and colleagues measured the attractive and adhesive forces between an AFM tip and various surfaces, some of them covered with monolayer films. They found that the tip-sample force increased as the surface energy of the sample increased. They were also able to clearly detect small changes in the surface energy that resulted from modifying the surface group of the monolayer film.

Noriarki Oyabu et al. used an AFM to lift a single atom up off a surface and then replace it, demonstrating reproducible vertical manipulation. Unlike similar atom manipulations performed using STM, the AFM method of Oyabu et al. did not require a voltage to be applied between the tip and the atom. Instead, the tip was lowered so that it just pushes down on a single atom. The pressure this applies is enough for the atom to bind to the tip via a short-range chemical interaction. The atom can then be lifted away. To replace the atom, the tip is lowered, and the atom is pushed back into place.