To Make a Fine Point, New NIST Instrument Analyzes Microscope Tips

A new research facility at the National Institute of Standards and Technology accepts tips six at a time, in fact. But the tips are not of the monetary variety. Rather, they are the super-fine probes of the scanning microscopes used by researchers to glimpse atoms and molecules and, increasingly, by companies to inspect products with features and tolerances fast approaching molecular scales.

NIST has built a customized, two-in-one instrument to help it solve key issues that limit the measurement capabilities of scanning probe microscopes and, as a result, the instruments' usefulness in industrial and research applications.

SPMs include scanning tunneling microscopes, atomic force microscopes and a growing repertoire of related instruments with sharp tips, often needle-like in shape. They are best known for their ability to generate three-dimensional images of specimens in atomic detail, but depending on the design, SPMs also are used to analyze a wide range of physical and chemical properties, from magnetism to temperature to hardness.

Microelectronics firms and others that make products with ultraminiature, high-precision components are pressing SPMs into practical service. Worldwide, sales of the instruments are reported to be growing by more than 15 percent annually and may approach $170 million this year. Yet, unaccounted for and often large variations in the shape of probe tips and the inability to determine accurately tip dimensions undermine the instruments' usefulness in high-precision measurement, inspection and other quality-control applications.

As a practical consequence, measurements made with one probe tip may differ significantly from those made with another tip used on the same microscope. This poses a serious problem when trying to inspect, for example, electronic devices whose specified dimensions are equivalent to a string of a few atoms.

Obstacles to improved SPM accuracy, explains NIST physicist Rick Silver, co-designer of the new instrument, include measurement uncertainties stemming from uncontrolled variations in the geometry of fabricated tips and from dynamic shape and dimensional changes that occur during scanning. Interactions between tips and samples further complicate matters, he says.

In SPM-obtained images, tip and sample appear to merge, distorting measurements. Such distortions can be corrected, but only to the degree that geometry and dimensions of the tip are known. Today, this key information either is lacking, or it is acquired with great difficulty.

NIST aims to resolve this critical void through a combination of measurement-focused experiments and computer modeling efforts. The centerpiece of the research program is an imposing, NIST-built apparatus that features a scanning tunneling microscope, or STM, and a field-ion microscope, or FIM. Both microscopes can "see" individual atoms, but, for FIMs, this level of scrutiny is limited to needle-like specimens with a radius measuring between 1 nanometer (billionth of a meter) and about 100 nanometers.

This is not a serious limitation for the task at hand, since most SPM tips fall within this range. For probes with blunter tips, however, characterization will be done with a scanning electron microscope.

The two microscopes function much like a tag team: Probe tips are characterized with the FIM and then transferred in ultrahigh vacuum to the neighboring but physically isolated ultrahigh vacuum STM. With the STM, researchers will "test drive" the probes on silicon samples with well-controlled, atomically ordered surfaces, featuring, for example, a series of nanometer-scale steps or terraces. Like all STMs, this instrument produces profiles of surfaces by sensing a tunneling electrical current that flows between atoms on the probe tip and those on the sample surface. Maintaining the probe at a constant height relative to the surface maintains the current. Successively scanning the tip over the surface yields a three-dimensional image with atomic resolution.

NIST's combination instrument can accommodate up to six probes, which are shuttled between microscopes on a rotating carousel. In addition, the apparatus features a separate ultrahigh vacuum chamber, where tip and sample surfaces can be heat treated.

Tips also can be "sculpted" with the FIM. By varying the electric field in the chamber, researchers can evaporate clusters and even planes of atoms from probe tips. Silver says that while this capability cannot be controlled precisely, it will permit modifications useful to the research effort.

The tandem arrangement of microscopes will enable measurements of probe tip geometries and dimensions to be correlated with tip performance in STMs and other SPMs. One goal of the NIST effort is to develop methods and tools that enable direct characterization of SPM tips, which would eliminate much of the measurement uncertainty resulting from unknown variations in tip geometry. Initially, work will focus on tungsten and platinum-iridium tips.

"Just as a good mechanic would not grab a screw driver without inspecting the size and type of the blade," Silver explains, "SPM users require basic information on tip geometry so that they can confidently select tips that are most appropriate for particular jobs."

Another goal is to prepare the equivalent of a recipe book that prescribes standard methods for making SPM tips with reproducible geometries, as well as for modifying tips with etching and field evaporation techniques. The NIST team also intends to identify cleaning and transporting methods that minimize the risk of unwanted alterations of probes.

Characterization artifacts are another anticipated outcome. Samples with atomically ordered surfaces and nanometer-sized surface features would enable SPM users to calibrate their tips and verify their performance before applying them.

Information gained through this effort also will be used to refine and extend NIST's computer modeling work on SPM tips and on measurement-distorting interactions between tip and sample. This work already has yielded a "blind reconstruction method" that aids in determining the real dimensions of features in an SPM-obtained image. With a model developed by NIST physicist John Villarrubia, an SPM user can estimate the maximum possible dimensions of a probe tip, thereby reducing measurement uncertainty.

Plans call for including the NIST tip characterization software in the agency's Guide to Available Mathematical Software. GAMS is an on-line repository of publicly available software for mathematical modeling and statistical analysis. It can be found on the World Wide Web at http://gams.nist.gov/gams/.

As a non-regulatory agency of the Commerce Department's Technology Administration, NIST promotes U.S. economic growth by working with industry to develop and apply technology, measurements and standards.