The Atomic Force Microscope (AFM) a.k.a. Scanning Force Microscope (SFM) or Scanning Probe Microscope (SPM), has been around for almost 15 years. The microscope was an offshoot of the Scanning Tunneling Microscope (STM) and designed to measure the topography of a nonconductive sample. The AFM has undergone several enhancements over the years, allowing it to measure the local resistivity, temperature, elasticity, tribology, as well as allowing studies beyond the limitations of conventional optics.
The information presented below pertains to the simplest use of the AFM - obtaining an image of the sample. The general principles apply to most every other application.
EXECUTIVE SUMMARY : The AFM works much the same way a phonograph or profilometer works only on a much, much smaller scale: a very sharp tip is dragged across a sample surface and the change in the vertical position (denoted the "z" axis) reflects the topography of the surface. By collecting the height data for a succession of lines it is possible to form a three dimensional map of the suface features.
The Tip : The first detail we'll discuss is the AFM tip or the part that makes contact with the sample. This represents one of the most critical features of the AFM, and advances are always being made to create a better tip (we'll soon see what makes a really good tip). The tip is generally made of silicon or silicon nitride using the same technology that makes the integrated chips in your computer. The tips come in various shapes - the one shown is square pyramidal in shape roughly two microns wide. The end of this tip is often slightly rounded (unlike the atomically sharp edges you see in your textbooks), being around ten nanometers (0.00001 millimeters, or about 100 atoms) in radius.
There are several requirements for the tip - it must be sharp and thin to get into all the nooks and crannies of the sample but it must be very durable to survive the forces and it shouldn't bend under normal loads. Research is still being done to improve the tip, but the most promising outcome will probably have long, thin crystals or fibers (such as buckytubes) attached or an integrated part of these.
The Cantilever : The AFM tip is held at the end of a thin, flexible beam, or "cantilever". This cantilever is made just as the tip was, but its shape is usually triangular ("V" shaped) or long and rectangular (an "I" beam). These are roughly 100 microns long (which is 0.1 millimeters, about the width of a hair) and only a few microns thick. This makes them very flexible but strong enough to securely hold the tips on their end.
The Scanner : Now let's discuss how the sample is "scanned" in an AFM. There are two ways of moving the sample under the AFM tip - in one case you can move the sample and keep the tip in place, or, alternatively, you could move the tip over the stationary sample. In the first case, the small sample being studied is placed on a very special scanner (shown in blue). This scanner is made of piezoelectric crystals, that neat stuff which replaced flint in cigarette lighters and makes the spark used to start a barbeque. This crystal creates a voltage if pressure is applied, or in reverse, can create a pressure by expanding or contracting if a voltage is applied. Using the contraction and expansion of the crystal, the configuration in a scanner allows for the controlled movement on the order of a fraction of a nanometer. Such precise manipulation of the sample could not be possible using traditional mechanical methods with gears and pistons.
Alternatively, the tip could be placed on the piezo scanner and brought in contact over the sample. This is a far better configuration if the sample is large or immersed in a liquid.
The Height Transducer : With the tip making contact with the sample we now ask the question "How do we monitor the height changes of the tip as it passes over the sample?" These vertical movements could be extremely small - sometimes less than a fraction of a nanometer when imaging atomic structures.
For most AFM, the height of the tip (and therefore sample) is usually monitored using a laser beam which reflects off the backside of the cantilever (a gold coating on the cantilever makes it behave like a mirror). This reflected beam hits a multi-segment photodiode which can detect the movement of the beam, and therefore the movement of the cantilever and tip. The vertical movement of the tip is thereby measured as a voltage change.
Gain Settings : Now that the tip is in contact with the sample and we have the means of moving the tip or sample, we have a question as to how control the tip and its response. The AFM has to be instructed what to do when the tip moves up and down. There are two means of doing this.
No Gain : The first way to respond to height changes in the tip is no response at all. In this mode of operation, called "no gain" or "varying force/constant height", the tip height is allowed to freely fluctuate. As the tip moves up or down, the cantilever bends, and since the cantilever is a spring this changes the force exerted on the surface. The laser reflecting off the cantilever changes a great deal, and thus the voltage readings from the photodiode vary. If the voltage to a given height of sample has been calibrated, the electrical signals can be converted to height readings.
This method of operation is best suited for very flat surfaces, such as atomic structures, since there is an instantaneous response of the photodiode voltage to the surface topography and the tip movement is not too great meaning the force doesn't change much.
High Gain : The second mode of operation is called "high gain" or "constant force/varying height". As the tip moves up or down, the reflected laser position begins to change but the AFM will attempt to restore the beam's position. Depending on whether the sample or tip sit on the piezo scanner, the AFM will do this by moving the sample underneath the tip or changing the position of the tip and cantilever. If the AFM can respond well enough to the height variations, the cantilever's curvature will remain vitually unchanged nad thus the force exerted by the tip on the sample is constant. The height changes of the scanner reflect the topography of the sample.
This method is best for almost every sample, but is difficult to perfect - set too high, the AFM becomes sensitive to random noise which can make the system unstable, and too low makes the AFM sluggish giving poor images and slightly varying forces.
What these two different modes of operation show is that the parameter controlled or monitored by the AFM is the point on the cantilever that reflects the laser - in the no gain mode this point mimics the surface, while in high gain this point is kept stationary. In both these modes, the actual contact with the sample is at the end of the tip, and this can be several microns from where the laser is reflected. This translation wouldn't be important if it remained constant, but the size of the tip and sample means the contact point will vary. This, as will be shown later, creates the image distortion called a "convolution' of the tip and sample geometries.
With everything properly set, the sample or tip is moved in a straight line (denoted the "x" axis) with the tip's height being recorded at regular intervals. Once the scanline is complete, the sample is moved slightly in the "y" direction and the scan is repeated. When several of these lines are viewed, one gets an image of the surface topography. One other "feature" used on the surface rendering is the false coloring whereby the "higher" data points are in a light shade while the "lower" ones are darker. This coloring is like what you see on regular topographic maps, where the mountains are white (what else) and the valleys are dark green.
* See my CMS Bulletin or Chem13 News reprints.
Last revised on 01-12-2000 by Peter Markiewicz