Scanning Tunneling Microscope

The scanning tunneling microscope (STM) was the first of several "proximal probes" that in the past decade have revolutionized our ability to explore, and manipulate, solid surfaces on the size scale of atoms. At its heart, the STM is little more than a pointed electrode scanned over a conducting surface (or "specimen") of interest, via electronic control of a piezo-electric crystal's shape. Gerd Binnig and Heinrich Rohrer of IBM's Zurich Research Center were awarded the 1986 Nobel Prize in Physics for discovering the STM. Excitement about this instrument, and its "atomic force microscope" cousin, remains high. One reason is that scanning probe microscopes continue to open new doors, while serving as our eyes and hands in the exploration of solid surfaces on submicron size scales. In fact, the 1996 Nobel Prize in Chemistry was awarded to Richard E. Smalley of Rice University (a scanning probe microscope enthusiast) for his role in the joint discovery of fullerines. Named after geodesic dome inventor R. Buckminster Fuller, fullerines are spherical carbon molecules whose cousin (the carbon nanotube or "Bucky tube") promises to make scanning tunneling microscopes even more powerful (in a microscopic way) in days ahead.

What are scanning tunneling microscopes doing that is new? To begin with, we now understand more of the detail seen in images. For example, a bump is not always a bump. Atomic oxygen chemisorbed on a metal surface appears as a depression in STM images, even if the atom is positioned above the metal surface layer! The reason is that the tip needs to press closer to the specimen to get the same current, if an oxygen atom is in the way. Adsorbed molecules whose STM images we understand better now include carbon monoxide, and quite a few loopy and straight hydrocarbons (like benzene and ethylene). Also, the "electronic structure" that complicates STM images often comes from electrons "hanging out" on the surface. For example, M. F. Crommie, C. P. Lutz, and D. M. Eigler working with the IBM Almaden Research Center Visualization Lab have found ripples near step edges on a copper surface, which come from such electrons. They have even gone a step further, and positioned individual Iron atoms to build "electron corrals" of various shapes on copper metal (e.g. Fig. 1). The result: Some of the most striking pictures to date of observed structures predicted only by the laws of quantum mechanics!

With the development of techniques for studying devices "cleaved" open to cross-section viewing, compound semiconductors (like gallium arsenide) have revealed some of their electronic secrets including the location of single dopant atoms beneath their surface. Elemental semiconductors, like the silicon pre-eminent in today's computers, still play "hard to get" because of surface defects introduced when they are broken open for cross-section study. The STM has also revealed in recent years how metal atoms of type A behave on metal surfaces of type B. For example, we're now certain that some metal atoms (like silver) on a surface (like platinum) avoid the differing atoms of the substrate, while other atoms (like gold on certain silver surfaces) form unexpected alloys or layered structures. In both cases, the behavior of atoms on a two-dimensional surface differs from the behavior of the same atoms when they find themselves contained within a three-dimensional solid!

Scanning probe microscopes are increasingly able to do more than report on one property of the specimen as a function of position (like the piezo-height at which current between STM tip and specimen is held at some set-point value). Thus modern "atomic force" microscopes can map lateral force and conductivity along with height, and image-pairs from an air-based STM scanning to and fro can be used to map friction coefficient along with height. As these instruments provide more robust ways for "getting small" and checking things out, vizualization facilities are improving rapidly as well. Already, topographic maps may be colored using non-topographic information like data on coefficient of friction. This allows ray-tracing programs to seriously put our 3D pattern recognition abilities to work in the nano-world (e.g. Fig. 2), and allows software like that in virtual reality markup language (VRML) browsers to offer human-viewpoint exploration & travel between nano-locations, even for participants connected far and wide across the world-wide web.

Lastly, of course, scanning tunneling microscopes allow us not just to look, but to touch. As Ralph Merkle of Xerox points out, the properties of a material depend on how its atoms are arranged. Rearrange the atoms in coal and you get diamonds. Rearrange the atoms of soil, water & air, and you have grass. Rearrange the atoms in grass, and you have a cheeseburger. The goal, of course, is not to build cheeseburgers one atom at a time! We are merely learning to explore material properties atom by atom. Eventually, however, we might learn to build tiny assemblers that, like mitochondria inside our cells today, aid the reshaping of matter by replicating molecular structures according to design specification.

Our present accomplishments are more modest. The ability to see surface atoms was demonstrated with the first STM. The ability to move xenon atoms around on nickel at liquid nitrogen temperature (e.g. to spell IBM) was demonstrated by Scientists at IBM Almaden in 1989. In recent years, now, selective bond-breaking has been demonstrated by T. C. Shen et al. at the University of Illinois, and by Ph. Avouris et al. at IBM Watson Research Center. In both cases the work has been done on silicon surfaces, and involves the selective breaking of either Si-Si bonds, Si-H Bonds, or Si-O-O bonds. The actual breaking has been done either by electrical pulses from the STM tip, or by vibrational excitation. This work is exciting because it involves construction at the level of single atoms: the ultimate frontier for lithographic miniaturization.

The mechanism used at IBM Watson for modifying silicon surfaces an atom at a time works at room temperature, unlike the earlier IBM Almaden work with xenon atoms on nickel. However, the electrical pulses used would likely cause problems if delicate molecules were to be repositioned. This brings us back to recent work by researchers at IBM's Zurich Research Center, the French National Center for Scientific Research (CNRS), and Cambridge University. They have now succeeded in positioning individual molecules at room temperature by purely mechanical means. They "hand-picked" an organic molecule with 173 atoms, including a porphyrin core, for the experiment. Choosing 6 such molecules from a set randomly positioned on a copper surface, they pushed each individually into position to form a ring which would not normally be found in nature (Fig. 3). Other molecules in the vicinity were not disturbed.

Thus the scanning tunneling microscope, and its other scanned probe cousins, are serving as the vehicle for an increasingly wider range of explorations which take place in worlds on the size scale of atoms and molecules. The interface to these microscopes has also become less abstract, allowing the explorers to apply pattern-recognition skills developed in the macro world toward the accomplishment of their objectives. Tools developed for "pretend" virtual reality environments are being put to use in the exploration of real places on the nano-scale. New molecular allies, like carbon nano-tubes, are being discovered and put to use as well. In the near future, you will likely be able to bring a "specimen" to your favorite scanned probe microscope lab, put on a head-set designed for virtual reality, and find yourself standing among friends on a landscape of molecules resting on your specimen's surface. Every few seconds, the surface you stand on will ripple a bit, as the scanning probe tip passes by.

Bibliography. S. Chaing (editor), Force and tunneling microscopy, Chemical Reviews 97 (4), 1015-1230 (1997); M. F. Crommie, C. P. Lutz, D. M. Eigler, Confinement of electrons to quantum corrals on a metal surface, Science 262, 218-220 (1993); M. F. Crommie, C. P. Lutz, D. M. Eigler and E. J. Heller, Waves on a metal surface and quantum corrals, Surface Review and Letters 2 (1), 127-137 (1995); T. C. Shen, C. Wang, G. C. Abeln, J. R. Tucker, J. W. Lyding, Ph. Avouris and R. E. Walkup, Atomic-scale desorption through electronic and vibrational excitation mechanisms, Science 268, 1590-1592 (1995); Ph. Avouris, Manipulation of matter at the atomic and molecular levels, Accounts of Chemical Research 28, 95-101 (1996); R. Martel, Ph. Avouris and I.-W. Lyo, Molecularly adsorbed oxygen species on Si(111)-(7x7): STM-induced dissociative attachment studies, Science 272, 385-388 (1996); T. A. Jung, R. R. Schlittler, J. K. Gimzewski, H. Tang, and C. Joachim, Controlled room-temperature positioning of individual molecules: Molecular flexure and motion, Science 271, 181-184 (1996); H. Dai, J. H. Hafner, A. G. Rinzler, D. T. Colbert, and R. E. Smalley, Nanotubes as nanoprobes in scanning probe microscopy, Nature 384, 147-151 (1996).