All scanning probe techniques such as STM and AFM depend on precise spatial control of a probe in all three orthogonal directions*. This is normally achieved using piezo-electric materials. When a voltage is applied across the piezo material it causes the material to expand or contract (depending which way round the voltage is applied). The effect is small, and typically several hundreds of volts are required to cause a change of a few microns in the material dimensions. This effect can be used as the basis of a mechanical actuator, the movement of which can be controlled down to the sub-angstrom level (i.e. on the atomic scale).

The most common way to form a piezo-based scanner is to fashion the piezo material into a hollow tube. Pairs of electrodes (on the inner and outer walls) are placed on either side of the tube. When suitable voltage differences are applied to these electrodes one side of the tube expands and the other side contracts. This results in a bending of the tube, hence if one end is fixed the other end moves, resulting in the scanning motion. Two sets of electrodes 90 degrees apart allow motion in the x-y plane*. A further pair of electrodes extending around the entire circumference of the tube cause an entire section of the tube to expand or contract, resulting in the free end of the tube moving parallel to the tube axis (=the z axis). The combination of all three sets of electrodes allows movement of the free end of the tube to be controlled very precisely in all three axes.

There are two ways of exploiting this effect. A probe can be mounted on the moving end of the piezo tube and scanned over the sample in question. This can present difficulties when using probes that require optical detection of the parameter being measured, for example the bending of a cantilever, as the probe is a moving target. A simpler method is to mount the sample on the scanner tube and fix the probe to a stationary stage. This makes the transfer of information from the probe easier to implement, and also facilitates the design of rigid systems (making higher movement resolution possible). The main drawback is that only small samples can be accommodated, as the piezo scanner cannot cope with the inertia of heavier samples. Hence when examining large (typically greater than 1x1 cm) samples a scanned-probe design is indicated.

Given the limited range of the piezo scanning system, most SPM systems also incorporate coarse positioning stages to position the probe over the area of interest of a sample. These positioning stages normally use screws to move the sample and/or the scanning system, and are actuated either by hand or by computer-controlled stepper motors.

There are two main disadvantages with the use of piezo actuators. The breakdown voltage* of the material limits the total range of movement, typically to tens of microns at most. Also, the response of the material to voltage changes is hysteretic*. If the piezo scanner is used in an open-loop system* some characterisation of the piezo response is required so that the movement of the scanner can be controlled accurately. The alternative is to use position sensors in the scanner, allowing you to use a closed-loop control system for greater positioning accuracy. The main downside of closed-loop systems is the increased complexity, and therefore cost.