Understanding atomic force microscopy

The development of active-cantilever technology spans three decades of pioneering research, from the first piezoresistive cantilever to the fully integrated systems produced by nano analytik today. Over this period, the technology has evolved into increasingly compact, self-sensing, and high-performance AFM systems.

1993

C.F. Quate

First piezoresistive cantilever

Introduced for first time the piezoresistive single resistor read-out scheme.

1994

I.W. Rangelow

Integration of Wheatstone bridge sensing

Introduced for first time the Wheatstone Bridge onto SPM-Cantilever.

1998

I.W. Rangelow

First fully active cantilever

Introduced for first time an integrated thermomechanical actuator onto the cantilever system.

Today

nano analytik Active Cantilevers

Ongoing Cantilever Innovation

Continuous development of integrated self-sensing and self-actuating cantilevers for higher sensitivity, faster imaging, and scalable AFM platforms.

Piezoresistive readout on an active AFM cantilever works by exploiting the change in electrical resistance that occurs in a doped semiconductor when it is strained. An AFM cantilever is made from a silicon-based material (often doped). A thin resistive element (a piezoresistor) is embedded in or patterned onto the cantilever, typically near the region where deflection is largest. When the cantilever deflects (due to tip-sample interaction forces), the bending creates mechanical strain in the cantilever material. This strain changes the crystal lattice spacing in the doped silicon, which alters the charge carrier mobility and density in the piezoresistive region. The change in resistance ΔR is proportional to the original resistance R and to the fractional strain, roughly ΔR/R ∝ ε, where ε is the normal strain in the piezoresistor. The proportionality depends on the dopant type, crystallographic orientation, and whether the current runs along or across certain crystal directions. A constant current or constant voltage is applied to the piezoresistive element. As the cantilever bends, its resistance changes, causing a measurable change in voltage drop (or current) across the element. This voltage (or current) signal is proportional to the cantilever deflection. Because the resistance change is typically small, the readout often uses a Wheatstone bridge to maximize sensitivity and common-mode rejection, followed by amplification, filtering, and demodulation if needed (especially in dynamic modes). Actuation and coupling (for active cantilevers): In actively driven cantilevers, the cantilever is excited (e.g., electrothermal). The deflection due to interaction with the sample modulates the piezoresistive element, providing a direct electrical signal that tracks deflection. Because the readout is integrated on the cantilever, no optical path is required. In this manner a piezoresistive readout enables compact, integrated sensors and simple optical-free operation, but it typically trades some sensitivity and noise performance against optical readouts. Thermal and electrical noise, temperature drift, and material anisotropy must be managed through design (bridge ratio, shielding, and temperature compensation). Summarizing it we can say that deflection-induced strain in the cantilever changes the resistance of an embedded piezoresistor; this resistance change, measured via a bridge and amplified electronics, provides a direct electrical readout of the cantilever’s motion.

Electrothermal actuation uses local heating to cause thermal expansion, bending the cantilever to drive motion. AFM cantilevers are usually silicon with an integrated heater on the cantilever consisting metal, oxide and polymer layers. An electrical current through the heater raises its temperature. The heated region expands. If heating is asymmetric (one side hotter than the other), it creates bending. Electrical power P supplied to the heater is P = I^2 R = V^2 / R, where I is current, V is voltage, and R is the heater resistance. Temperature rise ΔT is approximately proportional to the absorbed power minus losses (ΔT ∝ P, with the proportionality depending on thermal resistance R th and heat sinking). Thermal expansion ΔL on the heated side is proportional to ΔT (ΔL ∝ ΔT). The cantilever deflection δ is driven by the differential expansion between the heated region and the rest of the structure, so δ ≈ k · ΔT ≈ k' · P, where k and k' collect geometry, material properties, and thermal pathways. In practice, the relationship is roughly linear for modest temperature rises within a given design, but it can exhibit nonlinearity and hysteresis due to material limits, thermal lag, and drift. Calibrations that map heater power to stable deflection are often performed for accurate control. The asymmetric expansion yields a mechanical deflection that can be controlled by adjusting the heating power. Turning off the current cools the cantilever and it returns toward its original shape. This actuation is suitable from single Hz to MHz frequencies (up to ~2000 kHz depending on design); not as fast as optical or piezoelectric actuators for high-speed AFM. No external optical drive needed, compact on-chip integration, potential for high actuation speed with proper geometry.