Near-field scanning optical microscopy (NSOM) originally was used simply to generate high-resolution images of samples. Now, Raman and infrared spectroscopy are being combined with NSOM to generate images with chemically specific information, filling a gap left by scanning probe microscopies that can reveal shape and structure on the nanometer scale but don't provide information about chemical composition.

The near field--defined as a distance to the sample much shorter than the wavelength of light--provides much better spatial resolution than conventional optical methods, which are limited by diffraction to approximately half the wavelength of light.


	REMOVING COUPLING Strong topographic coupling has a large effect on the IR absorption collected by a near-field apertureless microscope of polystyrene-poly(dimethylsiloxane) (PDMS) block copolymer surface. The tall rings in the upper panel show strong absorption. In the lower panel, near-field IR data were collected using an alternative scanning technique that reduces topographic coupling, and the true domains of concentrated PDMS absorbing the IR emerge as the regions colored purple.
	COURTESY OF GILBERT WALKER AND BORIS AKHREMITCHEV

The ability to generate chemical information with such fine spatial resolution is becoming increasingly important as the "critical dimensions of emerging technologies shrink," said Stephan J. Stranick, a chemist at the National Institute of Standards & Technology, at a symposium about near-field spectroscopy sponsored by the Division of Analytical Chemistry at the American Chemical Society national meeting in San Diego earlier this month. For example, molecular electronics and single-site catalysts are at the forefront of their respective areas, Stranick pointed out. Understanding these technologies will require spectroscopic techniques that provide chemically specific information on the nanometer scale.

Near-field spectroscopy has many advantages for analyzing the chemical composition of nanostructures. "You can use this in a natural environment. You can use this on your tabletop," Gilbert C. Walker, a chemistry professor at the University of Pittsburgh, said. "There are other methods of analyzing chemical composition at less than 100-nm length scales, but they all involve using a vacuum chamber. Many samples don't look the same when they're in vacuum. That's why these near-field techniques, both the Raman and the infrared, are really promising. You can examine heterogeneous systems under ambient conditions."

THE SCANNING PROBE techniques on which near-field spectroscopy is based automatically provide topographic information. "Because there's a feedback mechanism keeping the probe close to the surface, you're going to learn about the topography," Walker explained. "It's something that comes for free."

Near-field spectroscopy can be divided into two broad categories: aperture and apertureless. In aperture near-field spectroscopy, light is delivered to the sample through an opening much smaller than the wavelength of light. The size of the aperture and its distance from the sample surface dictate the spatial resolution. In the apertureless techniques, a sharp metallic probe is used to scatter electromagnetic radiation. The apertureless techniques provide better spatial resolution than the aperture techniques.

At the symposium, Yasushi Inouye, an assistant professor of applied physics at Osaka University and a member of Satoshi Kawata's group, described the apertureless probe that his group uses for near-field Raman studies. A silver-coated atomic force microscope (AFM) cantilever is held in contact with a dielectric substrate on which the sample is placed. A laser beam illuminates the metal tip through the substrate. Inouye uses a combination of illumination methods to depress the background. The electric field--and consequently the Raman signal--is enhanced by a factor of 80 only in a spot that corresponds to the size of the tip radius. A metal tip and p-polarized light are required for such a large electric field enhancement, Inouye said.

Inouye and coworkers obtained AFM and Raman images of a mixture of the dyes rhodamine 6G and crystal violet. The regions containing the different dyes were identified using unique peaks that were well separated from peaks from the other dye. They also used the near-field probe to obtain images of mouse myocardial cells.

David N. Batchelder, a professor of physics at the University of Leeds in England, uses both aperture and apertureless probes for near-field Raman spectroscopy. He monitored stress in silicon with an aperture probe. A stress profile was generated by converting the shift in Raman frequency to a measure of stress.

In one case where Batchelder used apertureless probes for near-field Raman spectroscopy, a C₆₀ monolayer was placed on a mica substrate. The tip was brought near the sample, either touching it or in close proximity. The laser illuminated a 1-µm spot, but the area of interest was only 10 nm. "You need an enhancement or you won't see the spot you're interested in," Batchelder said. When the tip was in contact with the sample surface, the signal was much greater than when the tip was pulled away.

Batchelder is studying polydiacetylene (PDA) solutions with apertureless near-field Raman spectroscopy. PDA, a model for conjugated polymer systems, undergoes a structural change below 60 šC; it appears yellow when hot and red when cool. In the yellow solution, PDA is a random coil. Batchelder wanted to find out if the PDA is random or ordered in the red solution. The Raman frequencies and polarization from PDA particles attached to a mica substrate suggest that the polymer chains are highly ordered and probably chain-folded.

NEAR-FIELD RAMAN, or "nano-Raman," spectroscopy might appear to be simply a higher resolution version of "micro-Raman." However, the results that Eric Ayars, a physics professor at Walla Walla College in Washington, described at the symposium indicate otherwise. Extra peaks show up in the near-field Raman spectrum that aren't in the far-field spectrum, he found. Those peaks are caused by alternate polarization states and a "gradient-field Raman effect," Ayars explained.

A mathematical model called the Bethe-Bouwkamp model demonstrates that the electric field of light coming through a subwavelength aperture has both a strong gradient and a z-polarized component, Ayars said. The z-polarized light allows peaks that would appear in the Raman spectrum of other crystal orientations to become visible, whereas the gradient causes IR-active peaks to become Raman-active.

"The normal derivation of Raman spectroscopy assumes that your electric field is constant over the length scale of the vibration, which is a good assumption in any normal situation," Ayars told C&EN. "Because of this very strong gradient, you can't assume that the electric field is constant. It's only because of this very strong rapidly changing field near the tip that you have this effect." The gradient field effect is 1,000-fold stronger near metal than in a vacuum. The gradient field effect may also explain some of the Raman modes that are observed in surface-enhanced Raman spectroscopy, or SERS, Ayars believes.

IR spectroscopy is also being done in the near field. Walker and Stranick both described applications of near-field IR spectroscopy. In the far field, the diffraction limit for IR imaging is nominally 3 µm, but such high resolution is difficult to achieve, Stranick said. The achievable spatial resolution is closer to 20 µm, he asserted. Some of the challenges to using IR in the near field are the need for IR-transparent fibers, for rapid data acquisition, and for bright, tunable sources.

Stranick used IR near-field spectroscopy to characterize and model polymeric systems such as TiO₂ particles in acrylic melamine and a polyethylacrylate-polystyrene blend. In a transmission near-field measurement, the thickness of the film can set a limit on the achievable spatial resolution, he said. If the sample is thicker than the spatial extent of the near field (approximately an aperture diameter deep), diffraction effects will begin to degrade the resolution. However, the resolution is still better than is achievable in conventional IR microscopy.

Walker uses an oscillating tungsten-coated probe for apertureless near-field IR spectroscopy of nanostructured surfaces. When a block copolymer is cast in air, the rate of solvent evaporation and humidity influences the formation of surface features. For example, in a block copolymer of polystyrene-poly(dimethylsiloxane), changes in hydration cause changes in the surface topography and presumably the chemical composition.

"Topographic variations of samples in near-field experiments contribute to signals whose origins are not simply based on chemical concentration, but rather on the topographic structure of whatever you're studying," Walker told C&EN. "Anytime you collect a near-field signal on anything that has any roughness, you're always going to get some contribution to the signal as a consequence of that roughness."

IF THE TOPOGRAPHIC contribution to the signal can be understood through modeling, the effects can be minimized, although not completely eliminated, Walker said. At the symposium, he presented two models--one that was analytical and a second that was numerical and allowed for surface inhomogeneity and mapped topographic coupling. One problem is what are known as "multiple scattering effects," Walker said. The light gets trapped between the tip and the surface and "rattles back and forth before it escapes," Walker explained. Such light interacts with the sample multiple times, increasing the observed signal. With the numerical model that Walker presented, these multiple scattering effects can be easily identified, he said.

One way to deal with the topographic coupling in the experiment is to maintain a constant distance between the tip and the underlying substrate (not the tip and the surface). "That is a first-order way of minimizing the topographic coupling," Walker said. "The topographic contribution is reduced enough that the main thing you see is chemical composition variation. However, there is still a contribution from topography that you simply can't entirely escape."

Near-field vibrational spectroscopy is currently being used in only a few laboratories, Stranick told C&EN. However, he thinks that will change. "The instrumentation is becoming sophisticated enough and the measurements robust enough that nonspecialists are going to be able to put it to use. With the recent commercialization of this technology [for Raman], we will see rapid growth in the number of users similar to the accelerated growth of AFM use after its commercialization."

SMALL SCALE Near-field probes allow samples to be chemically analyzed with high spatial resolution.
COURTESY OF STEPHAN STRANICK

[Previous Story] [Next Story]

Top



Home \| Table of Contents \| News of the Week \| Cover Story
Business \| Government & Policy \| Science/Technology

Chemical & Engineering News Copyright © 2001 American Chemical Society - All Right Reserved 1155 16th Street NW • Washington DC 20036 • (202) 872-4600 • (800) 227-5558