Scanning Probe Microscopy (SPM)

Humans wish to explore the physical structure of ever smaller objects, as well as the very fine features of large objects.  To this end, we invented microscopes to see better.  However, there are fundamental limitations to the resolution of a microscope.  Features smaller than half the wavelength (λ/2) of the light being used cannot be resolved.  The wavelength can be shortened by placing the object to be observed within a material other than air or vacuum to buy a little more resolution, but it is not as much as can be desired.  Using shorter wavelength light can increase resolution, but this light has more energy, eventually resulting in destruction of the sample and requiring different optics.  Games can be played with uncertainty to increase resolution in one dimension while losing it in another using squeezed light.  Eventually, the call for greater resolution leaves us blind and we must follow the example of the blind man and reach out and touch the object in order to see it.

Scanning probe microscopy (SPM) is the blanket term for technologies that take some sort of "finger" into close proximity of an object to be studied.  This "finger" is kept at a set distance from the object using some sort of feedback and moved with respect to the object in order to explore its shape.  The surface is mapped out, usually in a grid, in order to form the complete picture.  Two main forms of SPM are scanning tunneling microscopy (STM) and atomic force microscopy (AFM).

Schematic of a scanning tunneling microscope.  Probability of tunneling (P) is dependent on the distance between the surface and the tip (d) such that it falls off exponentially.

The first, STM, uses a metal tip into close proximity of a conductive sample as the "finger", or tip, inside a vacuum.  The tip can be as simple as a wire sliced so that it comes to a sharp point.  When this is brought close to the sample, but not touching, the vacuum between the tip and the sample acts as an insulator preventing current from flowing.  However, there is a finite probability of electrons tunneling across the gap resulting in a small current.  The probability is sensitively dependent on the distance.  To map out the surface, the surface is moved with respect to the tip and a feedback loop is used to maintain the tip at a constant distance over the surface.  The vertical motion required to keep the current constant is recorded and used for the image.  This is used to achieve atomic resolution and even observe subatomic structure such as orbitals.  (This later is, of course, completely impossible.)

Variations on this theme are also used.  Most simply, the sample can be scanned at a constant height while the changing current is recorded to produce the picture.  One I have found particularly interesting is inelastic electron tunneling spectroscopy wherein electrons of very carefully controlled velocity (via controlling the energy) are used to reveal vibrational modes of chemisorbed species on a metallic surface.

Schematic of an atomic force microscope.

The second, AFM, is more generalized as it does not require any specialized materials.  The tip is often a microfabicated cantilever of some insulator which is placed close to or in contact with the surface.  Placed in contact, the surface is moved around while the deflection of the cantilever is measured or the motion required to keep the deflection constant is measured, similarly to the two basic modes of using STM.  The cantilever may also be operated in tapping mode, where the cantilever still comes into contact with the surface, but is not dragged along it.  Even this can be destructive to the surface, so it may also be operated in non-contact mode.  For this, the cantilever is vibrated at a resonant frequency and the amplitude of the vibration is monitored.  When it comes close to the surface, the surface dampens the vibrations even without any contact, providing a measure.

Measurement of the deflection of the cantilever was initially done using an STM above the cantilever reasoning that this would be very sensitive to the position.  This turned out to be surprisingly difficult and now a laser is usually reflected off the cantilever onto a photosensitive diode or array.  Atomic resolution has been a much harder fight, but now can be achieved.  The variations of AFM can seem endless.  Changing the material of the tip or adding material to the tip changes the interaction between the tip and the surface.  For instance, magnetic material may be added to the tip to image magnetic domains in a sample.  Another variation places chemicals on the probe and measures chemical interactions.