PRL Top Ten: #4
Atomic Force Microscope
(G. Binnig, C. Quate, and Ch. Gerber, Phys. Rev. Lett. 56 (1986), 930), 3469 citations
This is the seventh in a series of articles by James Riordon. The first article appeared in the November 2002 issue.
Nineteen eighty-six was a busy year for Gerd Binnig: the Zurich-based IBM physicist was blessed with birth of a son, jointly won Nobel Prize in physics with Heinrich Rohrer for the invention of the scanning tunneling microscope (the prize was also was shared with Ernst Ruska for his work on electron optics), and published a highly cited PRL succinctly entitled Atomic Force Microscope. "Surprising, isn't it," is Binnig's understated reply when asked to reflect on the events that surrounded the development of one of the most versatile analytical tools to appear on the research scene in a century. "It was a very exciting year for me working in this wonderful group in Stanford."
Binnig clearly recalls the moment that the inspiration for the AFM came to him. "After the STM was working I was torturing my brain for many years how to get atomic resolution also on insulators." (STMs are limited to imaging conducting samples.) "I tried so many ideas, none of them was promising. One day I was lying on the couch and suddenly saw a drawing in the statistically structured ceiling. It was the drawing of an AFM with a tip mounted on a cantilever," says Binnig, "I talked with Cal [Quate] and Christoph [Gerber] about it. It appeared that nobody had asked the question before whether one could measure the force between two single atoms. What should the instrument look like; how should it function? Christoph then built the first AFM at IBM. He was mainly working at IBM, I mainly in Stanford."
The AFM that Binnig envisioned is a conceptually simple device; a sharp probe of silicon, carbon, or some other material is mounted on a cantilever spring and dragged across the surface of a sample. In one of its most common operating modes, a feedback system adjusts the distance between the sample and the probe tip to maintain a constant deflection of the cantilever as it traverses the sample, and the structure of the surface contour is deduced by monitoring signals in the feedback loop. The interaction between the probe and sample may be mediated by various forces?electric, magnetic, van Der Waals?depending on the sample material and the specific scanning mode selected. Unlike the STM, which monitors fluctuations in currents flowing between the probe tip and a conductive sample, the AFM can provide images of a broad spectrum of both conducting and insulating materials.
"Basically," says Cal Quate, leader of the group at Stanford where Binnig began developing the AFM during his sabbatical from IBM, "it's a phonograph that's scaled down to look at atoms." In fact, Christoph Gerber scavenged a portion of the first AFM from a commercially-produced phonograph. "The first cantilever spring was a gold foil with a glued-on tip," explains Gerber, "and that was a diamond from an old record player needle, which I went down to Palo Alto and bought." Gerber crushed the diamond, and selected one of the sharpest fragments to serve as the their first probe tip.
Although the AFM is now widely recognized as an atomic resolution microscope, it was not immediately clear that the AFM would be capable of detecting the detail that Binnig, Gerber and Quate first proposed. "The fact that 'atomic' was in the title implied that we would see atoms," says Quate.
Initially, however, the PRL reviewers dwelt on the lower resolution scans reported in the letter, and were doubtful of the ultimate performance that the authors predicted for AFMs. "We argued with the reviewers," laughs Quate, "it took us a long time to convince them that it should be published."
Of course, the reviewers eventually conceded the point, but in retrospect their skepticism was at least partly justified by challenges the authors faced in the years following publication of their notable PRL. Early AFM images of atomic-scale structure, it turned out, proved to be deceptive. "I was always wondering why the surfaces investigated by AFM looked atomically very ordered, but very disordered when studied by STM," recalls Binnig. "The explanation is that many tip atoms of the AFM image the atomic structure of the surface. The result is an overlay of many atomic images and defects average out, but the periodicity remains. it took seven years to get [true] atomic resolution. Today this is under control."
For the most part, the authors note, AFMs are rarely pushed to their ultimate resolution. Much of their popularity as analytical tools, and therefore many of the letter's citations, result from the incredible versatility of the microscope in a wide variety of fields. "It is obvious that investigating all kinds of materials on the atomic scale, or close to that, makes a big difference," says Binnig, "Seeing is believing. Understanding the structure of matter on that scale, and being able to manipulate it by the AFM or other means in a controlled, 'seeing' way opens a new world. That was clear, and the STM already worked, being the biggest step in this direction. The AFM, however, broadened this in several ways. A large community can operate the AFM on a wide range of samples with the option, besides measuring tunnel currents and doing tunneling spectroscopy, or doing force spectroscopy."
"We developed the AFM solely to get atomic resolution on nonconductive surfaces," adds Gerber, "That was the idea, and that took many years [following publication of the letter]. In the meantime many researchers picked up the simplicity of the device and developed the AFM into this versatile instrument that we have today. It's incredible what you can do with it." In recent years, AFMs have been improved and modified to the point that they can probe soft, biological samples as well as the rugged crystals that were the focus of early studies. Other versions detect chemical properties, respond to magnetic fields, or measure frictional forces on a minute scale, to name a few of the seemingly endless AFM variations.
"When you look at instrumentation," Gerber adds, "it's stuff like deep space, like the Hubble and all that, which contribute to the understanding of our universe. What we've done with the STM and AFM helped to open up the nanoworld for visualization. Those are the two great things, in my opinion, to have happened with regard to instrumentation in the last twenty years."
All three researchers remain active in AFM research and development. Although he is now retired from IBM, Gerber continues to explore biological AFM applications at the university of Basel. Quate, who holds the Leland T. Edwards chair in the Stanford department of engineering, is working on AFM arrays with carbon nanotube tips. Binnig, like Quate, spends some of his time working on AFM arrays, including a massively parallel array of thousands of probes known as project Millipede. In addition, Binnig, currently ponders complexity informatics and leads a Munich research group attempting to model human perception.
©1995 - 2015, AMERICAN PHYSICAL SOCIETY
APS encourages the redistribution of the materials included in this newspaper provided that attribution to the source is noted and the materials are not truncated or changed.
Associate Editor: Jennifer Ouellette