Mitch Jacoby
C&EN Chicago

Ask a dozen surface scientists to  identify key developments in in strumentation that are responsible for catapulting nanotechnology to the front lines of physical science research. Nearly all of them will point to the advent of scanning probe microscopy.

Despite being newcomers to the field of instrumental analysis, scanning tunneling microscopy (STM), atomic force microscopy (AFM), and other scanning probe techniques derived from those two main ones, have become popular--even indispensable--in many labs because of the volume of nanometer-scale information these techniques provide.

Compared with other instruments that open a window to a world of molecule-sized spaces, scanning probe microscopes are relatively simple, inexpensive, and easy to operate. These devices, known as proximal probes because they rely on a probe tip in close proximity to a specimen, can provide data in the form of topographic relief images that can often be interpreted in a straightforward, intuitive way. And especially appealing about the proximal probes is their multipurpose nature that offers not only a view of individual atoms and molecules but also ways to pick them up, move them around, and position them at will.

"What I want to talk to you about is the problem of manipulating and controlling things on a small scale," said Richard P. Feynman at an American Physical Society meeting in 1959. Speaking at California Institute of Technology, where he was then a physics professor, Feynman proposed radical ideas about miniaturizing printed matter, circuits, and machines. "There's no question that there is enough room on the head of a pin to put all of the Encyclopaedia Britannica," he said. Underscoring his belief that such a feat is physically possible while trying to motivate researchers to think small, Feynman offered $1,000 prizes to the first people to meet certain goals in shrinking books and electric motors.

"I'm not inventing antigravity, which is possible someday only if the laws [of physics] are not what we think," Feynman insisted. "I am telling what could be done if the laws are what we think; we are not doing it simply because we haven't yet gotten around to it."

Forty-one years later we're getting around to it. A key step in that direction was taken in the early 1980s when Gerd K. Binnig and Heinrich Rohrer , staff scientists at IBM's Zürich Research Laboratory , invented the scanning tunneling microscope. To make headway into a realm of molecule-sized devices, it would be necessary to survey the landscape at that tiny scale. And Binning and Rohrer's microscope offered a new way to do just that.

STM reveals nanoscale

Unlike conventional microscopes that provide direct images of an object under investigation, STM generates surface contour maps in which atomic-scale features can be resolved by carefully scanning an ultrasharp metal tip across a specimen surface while measuring electrical currents. Just a few years after debuting their new microscope, the IBM researchers jointly won a Nobel Prize in Physics for the design of their analytical tool.

STM hinges on the fact that, when a metal probe tip is brought to within approximately 1 nm of a conducting surface and a small voltage is applied to the tip, an electric current can be caused to flow between tip and sample. (The direction of flow depends on applied voltage.) The term "tunneling" is used because of the quantum mechanical effect of the same name that explains how it's possible for electrons with very little energy to pass through the gap between the two surfaces. The electrons make the trip by tunneling through the huge potential barrier.

[Photo by Mitch Jacoby]

The tunneling current is strongly dependent on the distance between the sample and the tip. Withdrawing the tip by just 1 can cause the current to decrease by a factor of 10. By monitoring the tunneling current with a feedback circuit while the sample is being scanned, the researcher can continuously adjust the probe to maintain a fixed height above the surface. The subtle up-and-down and lateral scanning motions of the tip are controlled by piezoelectric elements and recorded by a computer. The three-dimensional information can be displayed as an elevation map.

An important limitation of scanning tunneling microscopes is that they can be used only with specimens that conduct electric current. In contrast, atomic force microscopes, which were invented by Binnig and coworkers in the mid-1980s, can be used to study conducting and nonconducting materials.

Atomic force microscopy

Similar to STM in many ways, AFM is based on scanning a flexible, force-sensing cantilever across a specimen. Attractive and repulsive forces acting on the tiny diving-board-like arm cause deflections that can be measured with laser methods. The newer proximal probe can be used in a number of modes of operation. Included are a contact mode, in which the tip touches the specimen surface and senses internuclear repulsive forces between nuclei in the tip and sample, and a noncontact mode that exploits electrostatic or van der Waals forces. As with STM, a feedback circuit can be used to adjust the tip-to-sample distance to maintain constant force. The tip motions can be recorded and converted into relief maps.

	Researchers at the Canadian National Research Council show that minimal pretreatment of a silicon surface with an STM tip is all it takes to coax a flow of styrene molecules (orange protrusions) to line up in an orderly fashion.

Armed with new tools to study structures measuring just a few billionths of an inch, researchers quickly recognized the proximal probes' ability to move individual atoms and molecules--not just image them. A particularly well-known example came in 1990 when physicists Donald M. Eigler and Erhard K. Schweizer at IBM's Almaden Research Center in San Jose, Calif., announced that they had drawn the letters "IBM" on a cold nickel crystal by carefully positioning a handful of xenon atoms one at a time.

With eyebrows raised and curiosities piqued, scientists at many institutions began to explore the possibility of using scanning probe methods to form and investigate more complex structures such as tiny circuit elements and custom-made molecules. Dazzling demonstrations have been reported many times.

Earlier this year, for example, Eigler and coworkers at IBM formed an ellipse out of a few dozen cobalt atoms on a copper surface. Using the ring as a "quantum corral" to reflect copper surface electrons into predictable wave patterns, the group showed that when a single cobalt atom was placed at one of the focus points of the ellipse, a phantom atom appeared at the other focus point even though that position was vacant [Nature, 403, 512 (2000)].

After arranging a few dozen cobalt atoms as an ellipse on a copper surface and imaging the assembly (purple ring on orange, bottom), IBM researchers use a complimentary imaging technique to show that when a single cobalt atom is placed at one of the focus points of the ellipse, certain electronic properties (such as spin, indicated by spheres with arrows) are detected in the vicinity of the focus-point atom (large peak, top curve). A lower-intensity projection of those properties appears as a phantom atom at the other ellipse, even though that position is vacant.

The arrangement of atoms causes surface electronic states surrounding the focus-point cobalt atom to be projected to the other side of the ellipse. The IBM group notes that this finding, dubbed the quantum-mirage effect, may be useful for wireless transport of information across nanometer dimensions.

Wilson Ho showed that a probe tip can be used to deliver carbon monoxide molecules one at a time to an iron atom, forming Fe(CO), then Fe(CO)₂. In that experiment, Ho, who is now professor of physics and chemistry at the University of California, Irvine, used a scanning tunneling microscope to form chemical bonds, to image reactants and products, and to measure single-molecule vibrational spectra ( C&EN, Nov. 29, 1999, page 9 ).

At the University of North Carolina, Chapel Hill, chemistry professor John J. Boland uses scanning probe methods to follow chemical vapor deposition dynamics on semiconductor surfaces. Recently, Boland imaged unsaturated silicon valences (dangling bonds) as they wandered about on a silicon surface and showed that, eventually, dangling bonds tend to pair up with other dangling bonds--a key requirement for material growth processes.

Success in controlling and imaging tiny bits of matter with STM and AFM have spurred development of other scanning probe procedures. Chemical force microscopy, for example, is a variation of AFM in which a probe tip is capped with carboxylic acid or other chemical functionalities and used to map out surface functional groups. And magnetic resonance force microscopy--a technique developed at IBM and elsewhere--may one day be able to combine the chemical specificity of magnetic resonance spectroscopy with the atomic resolution of AFM.

Color mapping in STM images aids the eye in following atomic-scale landscapes: periodically spaced rows of atoms on those (110) face of a nickel crystals (top) assembled with a probe tip on a copper surface and (bottom) a molecule consisting of eight cesium atom and eight iodine atoms. ([Images courtesy of IBM]

At Northwestern University, Evanston, Ill., chemistry professor Chad A. Mirkin has devised a dip-pen procedure in which an atomic force microscope tip conveys a molecular "ink" to a substrate. Able to draw intricate patterns with nanometer dimensions, the technique has been extended recently to accommodate multiple inks and allow several writing heads to be operated simultaneously. And some manufacturers are investigating ways of using thousands of probe tips in parallel in data-storage devices that promise storage densities of 10¹² bits per sq inch.

A key shortcoming of scanning probe procedures is the slow, serial method by which they operate. This characteristic has limited their use mainly to laboratory applications involving atom-at-a-time maneuvers. Older techniques have also made their mark on science and technology at the small scale, and new instrumental methods continue to be developed. Transmission electron microscopy, for example, has been used for decades to examine tiny structures and currently can resolve features as small as 1 to 2 .

And photolithography, the workhorse of the microelectronics world, is used regularly to pattern semiconductors with features in the range of hundreds of nanometers. Other lithography methods, such as those based on electron beams or extreme ultraviolet light, offer higher resolution and are being investigated in some laboratories [Nature, 406, 1027 (2000)]. Optical microscopy can also play a part in this area--as demonstrated by researchers at Los Alamos National Laboratory who have developed a video procedure for monitoring colloid deposition and other nanoparticle processes in real time. But for stimulating the imagination via hands-on nanoscale manipulation, many scientists say STM and AFM are second to none.

"It's absolutely clear that the invention of the scanning tunneling microscope enabled this great emergence of interest in nanotechnology," asserts Robert A. Wolkow, principal research officer at the Canadian National Research Council's Steacie Institute for Molecular Sciences , Ottawa. "It has had an enormous direct practical impact but also an inspirational effect. People are inspired by this new opportunity to understand and control matter on the atomic scale. It has changed our way of thinking."

Top