In most condensed matter systems we think of electrons as delocalized particles that roam about the internal landscape of a material. This is especially true when the number of electrons in a crystal is less than the number of lattice sites, in which case most materials are metallic. But what if Coulomb repulsion between electrons is the dominant energy in a material? In this case you might expect electrons to freeze in place, like water turning into ice, since it costs too much potential energy for them to move around. Such behavior is, in fact, a 90-year-old prediction of quantum mechanics, but achieving such “electron-ice” (also called a “Wigner crystal”) is surprisingly difficult in practice. So far Wigner crystals have only been seen in a few experimental systems. The first experimental platform for studying Wigner crystals involved floating electrons on the surface of liquid helium back in the 1970s, and the next utilized electrons trapped at buried semiconductor interfaces in the 1980s and 1990s. These systems, however, are not compatible with high-resolution microscopy techniques, and so for almost 90 years it was impossible to glimpse the inner structure of theoretically predicted electron-ice. This situation has changed recently due to the development of new 2D materials only a few atoms thick. The electrons contained in these materials are very close to the surface and can, in principle, be imaged using scanned probe techniques such as scanning tunneling microscopy (STM). Here I will describe our recent efforts to image Wigner crystals in single-layer and bilayer sandwiches of 2D semiconductors using STM techniques. Normally, the large metal tip of an STM disturbs the electrons inside a fragile Wigner crystal due to electrostatic interactions, and makes imaging the Wigner crystal very difficult. I will discuss how we overcame this challenge using a new “sensor layer” geometry to perform the first imaging of Wigner crystals for systems having dimension greater than one.
