Scanning Tunneling Microscope

1. Introduction.- 

The scanning tunneling microscope was invented in 1981 by Binnig and Rohrer. Wikipedia (STM) has an excellent article on the microscope. The microscope scans a metallic tip over a small area of a solid surface where the dimensions can laterally vary between a few Ångstroems to tens of nanometers. The tip is maintain at distances generally below the nanometer from the surface. When a small bias (below a few volts) is applied between tip and sample, a current is established. As the tip scans the surface, a system of piezoelectrics controlled by some electronic equipment keeps the current constant. The maps of tip position constitute a constant-current image of the scanned surface.

A constant current image of a chain of 9 Mn atoms and a single Fe atom is shown in the figure. Since the image is basically given by the electronic features of the very similar Mn and Fe atoms, they cannot be distinguished.

Figure: Constant current STM image of Mn9Fe on a Cu (100) surface with a single Cu2N layer.
Image taken from our work on spin chains on surfaces.



The working principle is that electrons spill out of surfaces. This is a purely quantum effect. The wave function of the electron has an exponentially decaying tail in the vacuum side of the surface. Typically the decaying length scales with the inverse of the square root of the work function of the surface. In the limit of an infinite work function, the electrons would be strictly confined inside the surface without spill out. This is another way of saying that in the limit of an infinite potential well representing the surface, the wave function of the electron is totally confined to the interior of the well.

It is the finiteness of the confining potential that allows the electron to "exponentially" explore the nearest region of the surface in the vacuum side. Typical decay lengths of metallic systems are in the Ångstroems range. When two metals approach at a distance of a few Ångstroems, their electronic wavefunctions overlap. An electron of one of the metals can be found in the other metal without any energy transfer.

We have just described the effect tunnel where electrons from occupied states of one metal flow into unoccupied states of the other metal. The resonance condition when occupied states are at the same energy as the unoccupied states of the other metal is achieved through the bias difference. Even in the presence of this bias, the electrons of one electrode do not have enough energy to overcome the vacuum barrier between the two metals. This is the reason behind the name of the tunnel effect.

In our group we have long experience both in experiments using STM and in theory modelling STM constant-current images. Using density functional theory, it is possible to produce images that are virtually indistinguishable from the experimental images.
Figure: Molecular structure using fullerene molecules and the simulated/experimental images taken from one of our articles.