Introduction to Scanning Tunneling Microscopes

Nanotechnology is a relatively recent topic. In part this is because the tools required to interact with matter on such a small scale have been difficult to construct, and have only been available since 1981. Light-based microscopes have been in use for the last 400 years, but their resolution is limited to about 2000 nanometers (nm). Scanning Electron Microscopes (SEM) have been in use since their invention in the 1930s. Current SEM resolution limit is several nanometers. In 1981, Gerd Binnig and Heinrich Rohrer invented the Scanning Tunneling Microscope (STM). STM resolution has been improving since that time and now most scanning tunneling microscopes have resolutions of 0.1 nm or better.

A scanning tunneling microscope really only requires a metal tip and a metal sample. A portion of an STM is shown in the picture below. This picture was taken with a scanning electron microscope and color was added to make the relevant portions stand out. The metal tip is colored yellow and the sample is colored reddish brown. The left pane of the picture shows the tip relatively high above the sample and is magnified 350 times. The right pane of the picture is magnified 1000 times and shows the tip about 1,000 nm above a bump on the sample. The bump is about 18,000 nm wide.


As the tip is brought closer to the sample, electrons from the tip can tunnel through the air gap and reappear on the sample. The rate at which the electrons appear on the sample is dependent on the separation of the tip and sample. Tunneling does not happen very frequently if the separation distance is much greater than 1 nm. As the distance decreases below 1 nm, the tunneling rate (and the tunneling current) increase exponentially. The tip is usually attached to a piezo motor to allow its position to be changed.

Tunneling can be used to map the surface height of the sample in two different ways. In “Constant Height” mode, the tip is scanned across the sample at a constant height relative to the lab frame. The path of the tip is shown in the picture below as the red line. As separation between the tip and sample changes, so will the tunnel current. In this mode, the magnitude of the tunnel current is plotted as a function of horizontal position to give a contrast map. This mode is only useful if the sample is very flat so the tip will remain close enough to tunnel but not so close that it crashes into the sample. In our example picture below, the red line intersects a particle on the right side so the tip will crash into it if this path is followed.

In “Constant Current” mode, a feedback system moves the tip up and down by applying voltage to the piezo motor as the tip scans across the sample such that the tunnel current is maintained at a constant magnitude. This path is represented in the picture by the white line. In this mode, the output of the feedback system is plotted as a function of horizontal position to give a height map. This mode is the one that is used most of the time. The resolution in either mode is very good, but since the tip must physically pass over each point of the image, the process is very slow.

Building an STM requires a mechanical subsystem, an electrical subsystem, and the computer subsystem. The mechanical subsystem consists of a method of holding the tip and sample in close proximity while allowing the tip motion in three dimensions. The electrical and computer subsystems work together to control the tip motion and collect the data. In general, a more complex computer subsystem means the electrical subsystem can be less complicated.