Near-Field Scanning Optical Microscopy (NSOM) - History and Applications

The confinement of light in nanometric domains was only a dream when I first considered that such confinement might be possible. The idea occurred to me in 1979, while I was at Cornell University in the department of Applied Physics, attending a seminar given by Professor Lance Taylor then of Harvard University. Professor Taylor was trying to determine the distance between two protein receptors confined to a cell membrane. The protein receptors were situated at a distance that was smaller than the resolution limit of the microscope. The resolution limit - which had been defined by Lord Rayleigh at the end of the 1800s - was approximately 0.25 microns or 250 nanometers in x and y, and about five times worse in the z direction. Essentially, Lance Taylor had reached the limit of optical resolution and consequently was stymied in his research efforts to resolve the two proteins on the cell membrane.

After this seminar I immediately drew a simple diagram that described an approach to optics that would allow one to break this limit in Physics. This diagram has been replicated many times throughout the world in the last thirty years. Since the time of Galileo the confinement and the placement of light was defined by three optical elements: Lenses, prisms and mirrors. The idea that is illustrated in the figure below uses a very small hole to pass light onto the surface of a sample.

If the sample is brought very close to this hole within the red/pink rectangle, it can capture the confined light of the hole before the light spreads out. This confined area of light is called the Near-field. With this innovation, two points that were closer than the Rayleigh limit and could not be resolved with a lens could now be interrogated point-by-point with the very fine, relative-motion of the sample or the hole. We named the resulting technique Near-Field Scanning Optical Microscopy, or NSOM.

In parallel with these developments, Dieter Pohl at IBM Zurich developed a similar idea, originally called Optical Stethoscopy and later called Scanning Near-Field Optical Microscopy, or SNOM. Truth be told, approximately fourteen years before our paper and Dieter Pohl's 1984 paper, Eric Ash at Imperial College in London pioneered using a small hole and microwave radiation that relied on the same near-field principal. However, it was far from obvious that this research could be extended to optics, since the metal film in which a small hole is fabricated has infinite conductivity in the microwave regime, rendering the metal film opaque to microwave radiation. In the visible regime however, metal films have finite conductivity and therefore are partially transparent. Thus, it was not evident that a hole of nanometric dimension would have enough contrast to be able to see light coming through such a hole.

It took us nearly three years to make the first small hole in a metal film, for the simple reason that the tools for fabricating nanometric dimension holes in metal films were only just being invented. I recall the first night that I saw red light from a Helium Neon Laser coming through the smallest holes that we were able to fabricate, which were 15 nanometers. Nonetheless, we were still far from the dream of the object of this experiment, which was to develop a light microscope based on light coming through this small hole. That took another twelve years of research and development which led to the establishment of a company that focused on providing the world with these instruments. The company is owned by the three academic institutions of which I had been a part during this period of invention, research and product development: Cornell University - where I had been a professor before I moved to Israel - Hebrew University and the Hadassah University Hospital.

Without boring the reader with the difficulties of the research and development process that led to the realization of this form of microscopy, suffice it to say that the most successful near-field optical probe was built on the technology that had been prevalent in electrophysiology for decades. In electrophysiology, glass elements could be tapered to ultra-small dimensions using heat, a pulling force and controlled tension. Realizing the difficulty of applying nanofabricatioin tools, that were still in development, for the creation of nanometric holes - and the difficulty of using these holes to track real, rough surfaces - we invented a technique in 1986 using glass-pulling technology together with specific metal-coating protocols to form sub-wavelength apertures at the tip of glass structures.

The year 1986 also heralded one of the most important developments of the twentieth century: Atomic Force Microscopy, developed by Gerd Bennig and Calvin Quate. Using this invention, we rapidly advanced our method of creating apertures, and pioneered a methodology of cantilevering glass structures to produce a force-sensing optical waveguide with a sub-wavelength aperture at its tip. This enabled bringing a sub-wavelength aperture to sub-nanometer distances from a surface that could be rough, without destroying the nanometric light probes. As a penultimate step, these instruments were finally fully integrated into standard far-field optical microscopy.

It has taken a considerable period of time for people to fully appreciate the importance of these developments. First, the difficulty in development has only eased in the last several years; and second, the mindset in optics was one that did not appreciate the importance of nanophotonics in the early years near-field optics. In 1984, there were only two papers on nanophotonic applications. Last year, nearly eight hundred papers were published in the area of nanophotonics.

Today, one field after another is appreciating the critical role that NSOM can play in fundamental and applied developments. For example, new development in plasmonics allow light propogation to extend from nanometric to micrometric dimensions. This can impact everything from optical computing to solar energy concentration to biological sensing. Where would the development of nanophotonic devices based on plasmonics be without the invention of near-field optics? Much of the experimental device fabrication in plasmonics requires near-field optics for understanding the distribution of light in nanometric dimensionalities.

Aside from the sensing applications that we have mentioned above, only recently has it been possible to extend the power of near-field optics to biological imaging of soft samples. This is a result of a breakthrough that we have achieved that allows tuning forks with near-field optical probes to be completely immersed in biological liquid. Now fundamental investigations of the properties of cellular membranes - where 85% of the body's reactions take place - are open to near-field optical investigations, such as the original problem of Lance Taylor.

In addition to the above, there are applications as well in the area of Fluorescence Correlation Spectroscopy, whose signal to noise depends upon the size of illumination and near-field optics allows zeptoliter or 10-21 liters illumination volumes. This is three orders of magnitude smaller than the smallest available illumination volumes that can be achieved with standard optics. There is even the possibility that such small illumination volumes of the optical near-field can be used to delineate one of the most difficult problems in science - the nanometric structure of solutions.

Finally, there are important developments in chemistry, which can profitably employ such small illumination volumes. For example, the illumination of a region of a photocatalytic surface, while monitoring its electrochemical properties can be achieved with an NSOM probe. Such research is extremely important for the future of solar energy conversion and photocatalysis in general.

These are only a few of the wide spectrum of areas that the invention of near-field optics - now in its rising phase - has the ability to advance forward.