Atomic Force Microscopy (AFM) Explained: How It Works and Key Modes

Hooke's Law states that the restoring force exerted by a spring is directly proportional to its displacement from equilibrium (F = -kx), where the spring constant k determines how much the spring stretches or compresses under a given force; when a spring is cut in half, its spring constant doubles because the number of coils is halved, making it stiffer and causing it to show less deformation for the same applied force.

A cantilever is a beam of uniform cross-sectional area fixed horizontally at one end and loaded at the other end. For a light cantilever (neglecting beam weight) with a point load W at the free end, the deflection at the free end is given by δ = (W L³)/(3 Y I), where L is the length, Y is Young's modulus, and I is the moment of inertia. For a rectangular cross-section (breadth B, depth D), I = (B D³)/12, so δ = (4 W L³)/(Y B D³). For a circular cross-section (radius R), I = (π R⁴)/4, so δ = (4 W L³)/(3 π Y R⁴).

When a beam is supported at both ends and loaded in the middle, the upper surface experiences tension while the lower surface experiences compression. The neutral axis experiences neither. For a cantilever beam fixed at one end and loaded at the free end, the depression is: Δ = (Wl³)/(3EI). For a cantilever with load at the center: Δ = (Wl³)/(8EI). The moment of inertia (I) for rectangular cross-section is (bd³)/12, and for circular is (πr⁴)/4. Beam depression is proportional to load, cube of length, and inversely proportional to Young's modulus and moment of inertia.

Beam deflection is the deformation of structural members from their original position under load. Key influencing factors include load magnitude/type, span length, material properties (modulus of elasticity E), cross-sectional shape (moment of inertia I), and beam type. Standard formulas for common configurations are: simply supported with uniform load: Δ_max = 5WL⁴/(384EI); simply supported with concentrated load at midspan: Δ_max = PL³/(48EI); cantilever with concentrated load at free end: Δ_max = PL³/(3EI). The mathematical foundation uses the governing differential equation EI(d²y/dx²) = M(x), where integration yields slope and deflection equations with constants determined by boundary conditions: simply supported beams have zero deflection at both ends; cantilever beams have zero deflection and zero slope at the fixed end.

The force constant, commonly referred to as the spring constant in physics, is defined as the force per unit displacement, expressed by the formula k = F/X, where F is the applied force and X is the displacement. Its dimensional formula is M¹T⁻², derived from the dimensions of force (M¹L¹T⁻²) divided by displacement (L¹). The SI unit of the force constant is newtons per meter (Nm⁻¹). This concept is central to Hooke’s law, which states that the force required to extend or compress a spring is directly proportional to the distance of deformation, provided the deformation is small relative to the spring’s total possible deformation. Hooke’s law applies not only to springs but also to other elastic systems such as wind-induced deformation of tall buildings, plucked guitar strings, and inflated party balloons. Materials obeying this linear relationship are termed linear-elastic or Hookean. The law enables the derivation of material-specific relationships between strain and stress. For instance, a homogeneous rod with uniform cross-section behaves like a spring when stretched, with its stiffness k directly proportional to its cross-sectional area and inversely proportional to its length.

Piezoelectric elements form the foundation of Kleindig's nanomanipulation technology. These elements generate mechanical movement when voltage is applied, or produce electrical signals when mechanically stressed. The principle is demonstrated in electric lighters, where pressing a button applies pressure to a crystal, generating voltage that creates a spark. Conversely, applying voltage to the crystal can change its length or size on a nanometer scale with extreme precision.

Piezoelectric actuators offer significant advantages over electromagnetic actuators for precision positioning. While electromagnetic actuators have zero stiffness in open-loop operation, piezoelectric actuators provide finite stiffness because they are solid objects that physically deform. This makes them suitable for nanometer-level displacement transducers. By driving two piezoelectric elements at 90 degrees to each other, two-axis motion can be achieved: synchronous driving produces Y-axis motion, while opposite-phase driving produces X-axis motion. This configuration enables applications in scanning probe microscopy and precision metrology.

Moving a sensor across the nanoworld requires extraordinary precision. To illustrate this challenge, imagine trying to position Mount Everest within one centimeter of a plate covered in salt granules, using only a pencil tip. The STM achieves this by using three piezoelectric bars arranged in a tripod configuration. When voltage is applied, these bars elongate with extreme precision, allowing the probe tip to move along three axes with accuracy measured in fractions of an angstrom (one ten-billionth of a meter).

The piezoelectric effect is a bidirectional phenomenon where applying mechanical force to a polarized piezoelectric ceramic body generates an electrical charge (direct effect), while applying electrical voltage causes the body to expand or contract (inverse effect); manufacturing these transducers involves creating custom ceramic powders, pressing them into desired shapes, drilling precise holes using diamond tip tools, metallizing electrodes, and assembling components in clean rooms with high-precision mounting into metal housings for both actuator and sensor applications.

Achieving precise movement at the nanometer scale requires specialized piezoelectric motors that leverage the piezoelectric effect, where certain crystals expand or contract in response to electrical voltage, enabling both extremely high precision (down to single-digit nanometers) and high speeds (up to 1 m/s) without the backlash and thermal issues common in traditional mechanical systems like screws and gears.

This section examines why conventional microscopy cannot achieve nanoscale imaging. Optical microscopy resolution is fundamentally limited by diffraction, with resolving power d ≈ λ/(2n·sinα) depending on wavelength, refractive index, and numerical aperture. Common objectives achieve ~150 nm resolution for 400 nm light, insufficient for nanoscale imaging. Electron microscopy achieves atomic resolution but requires vacuum-fixed samples, limiting biological applications to cryo-EM or dried specimens. Scanning Tunneling Microscopy, developed by Binnig and Rohrer (Nobel Prize 1986), exploits quantum tunneling where electrons exhibit wave-like properties extending beyond classical barriers. Tunneling current requires small tip-sample distances and bias voltage, with current depending exponentially on separation. Feedback loops maintain constant distance using piezoelectric elements that deform precisely (0.1 nm/volt) in response to force sensor signals, enabling quantitative force measurements down to 10 piconewtons.

Electron microscopy achieves dramatically higher resolution than optical microscopy due to the much shorter wavelength of electrons (10 picometers) compared to visible light (550 nanometers). This gives electron microscopes a diffraction-limited resolution of approximately 5 picometers—about 10,000 times better than optical microscopes limited to ~250 nanometers. Unlike traditional projection microscopy that forms images simultaneously using lenses, scanning microscopy builds images sequentially by raster-scanning a probe over the sample and collecting data point-by-point. This scanning approach enables higher resolution because the image is constructed from localized interactions rather than relying on optical projection. In scanning microscopy, magnification and resolution are independent parameters: magnification is controlled by scan pattern size, while resolution depends on probe characteristics.

Optical microscope resolution is fundamentally limited by light wavelength, refractive index, and numerical aperture. Resolution (d) depends on wavelength (λ) and refractive index (n) - smaller wavelengths and higher refractive indices improve resolution. Immersion oils (n≈1.5) improve resolution over air (n≈1.0), but visible light cannot be reduced below 400nm without entering ultraviolet range. Electron microscopes overcome this limit by using electrons with controllable wavelengths via kinetic energy adjustment. Transmission electron microscopes (TEM) direct electrons through thin samples for internal structure imaging. Scanning electron microscopes (SEM) scan electron beams across surfaces to detect secondary electrons for topographical imaging. Atomic force microscopy (AFM) uses a fine tip that interacts with surface features, detecting tip movement to create atomic-scale topographical maps.

The theoretical resolution limit of optical microscopes, defined by Abbe's equation, was reached at the beginning of the 20th century and cannot be overcome through traditional lens design or electronic improvements; however, super-resolution microscopy techniques developed around 2014 (which earned a Nobel Prize) can bypass this limit using fluorescent markers and image processing, though these advanced methods are not applicable to hobby microscopy.

Resolution in imaging refers to the clarity and detail level, measured in pixels per inch. In microscopy, resolution is associated with the richness of observable details. The limit of resolution is the minimum distance between two points that can still be distinguished as separate entities. The human eye has a resolution limit of approximately 0.2 mm, while optical microscopes improve this by about 1000 times, allowing observation down to 0.2 micrometers. The formula for resolution limit is: Limit = k × (λ / NA), where k ≈ 0.61, λ is the wavelength of light, and NA is the numerical aperture. In microscope systems, objective lenses determine resolution while ocular lenses only provide magnification. The relationship between resolution limit and detail richness is inverse: smaller limits mean greater observable detail.