In semiconductor physics, perfection is the exception, and understanding why is essential. Defects aren't just flaws; they're the foundation of how we engineer semiconductor devices. Every time you encounter a question about doping, carrier mobility, device reliability, or material strength, you're really being asked about defects and their consequences. The physics of defects explains everything from why silicon can conduct electricity to why transistors eventually fail.
You need to connect defect types to their physical mechanisms and device-level impacts. Can a vacancy enhance diffusion? How does a dislocation affect mechanical properties? Why do deep-level traps kill device performance while shallow dopants improve it? Don't just memorize names. Know what category each defect belongs to, what physical principle it demonstrates, and how it changes material behavior.
Point Defects: The Zero-Dimensional Disruptors
Point defects are localized atomic-scale disruptions: single atoms missing, misplaced, or swapped. Despite their tiny size, they dominate diffusion rates, carrier concentrations, and doping behavior. These are the defects you'll encounter most in semiconductor device questions.
Vacancies
Missing lattice atoms: the simplest point defect, created when an atom leaves its regular site
Accelerate diffusion by providing empty sites for neighboring atoms to jump into, following DโeโEaโ/kT
Affect carrier lifetime and mechanical properties; vacancy concentration increases exponentially with temperature because it takes thermal energy to form them
Interstitials
Extra atoms squeezed between lattice sites, creating local strain fields in the crystal
Smaller atoms preferred (hydrogen, carbon, oxygen) because they fit more easily into interstitial spaces
Alter conductivity by introducing scattering centers or acting as unintentional dopants
Substitutional Impurities
Foreign atoms replacing host atoms: the basis of intentional semiconductor doping
Modify band structure by adding donor or acceptor energy levels within the bandgap
Size matching matters: atoms with similar radii to the host substitute more readily (Hume-Rothery rules)
Compare: Interstitials vs. Substitutional impurities: both introduce foreign atoms, but interstitials squeeze between lattice sites while substitutionals replace host atoms. When asked about doping mechanisms, substitutional impurities are your go-to example.
Frenkel Defects
A Frenkel defect is a vacancy-interstitial pair: an atom leaves its lattice site and moves to a nearby interstitial position. This means the total number of atoms in the crystal stays the same, so density is preserved and stoichiometry is maintained.
Common in ionic crystals with large size differences between cations and anions (like AgCl, where the small Ag+ ion can easily fit into interstitial sites)
Increase ionic conductivity through the vacancy mechanism, since the newly created vacancy allows neighboring ions to hop
Schottky Defects
A Schottky defect is a pair of vacancies: one cation and one anion are both missing, which maintains charge neutrality in ionic compounds.
Reduces material density because atoms are removed entirely from the crystal (they migrate to the surface)
Temperature-dependent concentration following nsโโeโEsโ/2kT where Esโ is the formation energy of the defect pair
Compare: Frenkel vs. Schottky defects: both occur in ionic crystals, but Frenkel moves an atom to an interstitial (constant density) while Schottky removes atom pairs entirely (reduced density). Which type dominates depends on ion size ratios: if the cation is much smaller than the anion, Frenkel defects are favored because the cation fits into interstitial sites. If cation and anion are similar in size, Schottky defects tend to dominate.
Line Defects: One-Dimensional Disruptions
Dislocations are linear defects where atomic planes are displaced relative to each other. They're the key to understanding plastic deformation. Without dislocations, materials would be either perfectly brittle or impossibly strong.
Edge Dislocations
Extra half-plane of atoms: imagine inserting a partial sheet of paper into a stack of sheets
Burgers vector perpendicular to the dislocation line, defining the magnitude and direction of lattice distortion
Enable slip at stresses far below theoretical crystal strength; their movement explains why metals are ductile
Screw Dislocations
Helical atomic arrangement: atoms spiral around the dislocation line like a parking garage ramp
Burgers vector parallel to the dislocation line
More mobile than edge dislocations because the parallel Burgers vector allows them to cross-slip onto different crystallographic planes, while edge dislocations are confined to a single slip plane
Critical for crystal growth: they provide self-perpetuating steps where new atoms can attach
Compare: Edge vs. Screw dislocations: both enable plastic deformation, but edge dislocations have Burgers vectors perpendicular to the line while screw dislocations have Burgers vectors parallel to it. This geometric difference is why screw dislocations can cross-slip and edge dislocations generally cannot.
Planar Defects: Two-Dimensional Boundaries
Planar defects are interfaces where the regular crystal structure changes abruptly. They control grain structure, affect carrier scattering, and determine how materials respond to stress.
Grain Boundaries
Interfaces between misoriented crystal regions: polycrystalline materials contain many grains at different angles
Impede dislocation motion through the Hall-Petch relationship: ฯyโ=ฯ0โ+k/dโ, where d is the grain diameter. Smaller grains mean more boundaries, which means higher yield strength.
Scatter charge carriers and provide fast diffusion paths, affecting both electrical and chemical behavior
Stacking Faults
Errors in layer stacking sequence: in FCC metals, the normal ABCABC pattern might become ABCBCABC
Lower energy than grain boundaries but still affect slip behavior and mechanical properties
Bounded by partial dislocations: the fault region exists between two Shockley partial dislocations
Twin Boundaries
Mirror-symmetric crystal regions: atoms on opposite sides are crystallographic reflections of each other
Enhance ductility and toughness by providing additional deformation mechanisms beyond ordinary slip
Form during deformation or annealing: mechanical twins and annealing twins have different origins but the same mirror-plane geometry
Compare: Grain boundaries vs. Twin boundaries: both separate crystal regions, but grain boundaries are disordered high-energy interfaces while twin boundaries are ordered, low-energy mirror planes. Twin boundaries generally have less impact on electrical properties because their atomic arrangement is more regular.
Bulk Defects: Three-Dimensional Imperfections
Volume defects are macroscopic features that affect overall material integrity. These become critical in device reliability and failure analysis.
Voids
Empty cavities within the material, ranging from atomic-scale clusters to visible pores
Act as stress concentrators: crack initiation often begins at voids, with local stress amplified by a concentration factor Ktโ such that ฯlocalโ=ฯappliedโรKtโ
Form during processing (trapped gas, vacancy coalescence) or during device operation (electromigration in metal interconnects, where current pushes atoms and leaves voids behind)
Precipitates
Clusters of secondary-phase atoms that form when solubility limits are exceeded, typically during cooling
Affect electrical properties by scattering carriers or introducing interface states at the precipitate-matrix boundary
Compare: Voids vs. Precipitates: both are bulk defects, but voids weaken materials by concentrating stress while precipitates can strengthen materials through dislocation pinning. Voids tend to form from vacancy coalescence or electromigration; precipitates form when impurity concentrations exceed solid solubility during thermal processing.
Electronic Defects: Carrier Traps and Dopants
These defects directly control electrical behavior by introducing energy states within the bandgap. This is where defect physics meets device physics.
Deep-Level Traps
Energy levels far from band edges, typically more than a few kT from the conduction or valence band
Capture and hold carriers, acting as recombination centers that reduce carrier lifetime via ฯโ1/Ntโ, where Ntโ is the trap density
Degrade device performance: they reduce LED efficiency, lower solar cell output, and hurt transistor reliability. Common deep-level impurities in silicon include iron and gold.
Shallow-Level Dopants
Energy levels close to band edges, within roughly kT of the conduction or valence band at room temperature
Ionize easily at room temperature, so nearly every dopant atom contributes a free carrier
Control carrier concentration: donors (P, As in Si) provide electrons; acceptors (B, Ga) provide holes
Foundation of all semiconductor devices: without shallow dopants, there are no p-n junctions and no transistors
Dangling Bonds
Dangling bonds are unpaired electrons that occur wherever the crystal periodicity terminates, such as at surfaces, grain boundaries, or in amorphous material. They create midgap states that trap carriers and pin the Fermi level, which is bad for device performance.
Passivated by hydrogen: forming SiโH bonds eliminates most dangling bond states, which is why hydrogen passivation is a standard step in amorphous silicon solar cell fabrication
Surface States
Electronic states localized at material surfaces, arising from broken translational symmetry at the crystal edge
Cause Fermi level pinning, which can dominate metal-semiconductor contact behavior and make it difficult to control Schottky barrier heights
Density depends on surface preparation: clean, oxidized, and contaminated surfaces behave very differently. High-quality thermal SiO2โ on silicon dramatically reduces interface state density.
Compare: Deep-level traps vs. Shallow-level dopants: both introduce bandgap states, but shallow dopants ionize completely and enhance conductivity while deep-level traps capture carriers and reduce device performance. This distinction is fundamental to semiconductor engineering.
Both Frenkel and Schottky defects occur in ionic crystals. What physical difference determines which type dominates, and how does each affect material density?
Compare edge and screw dislocations: how do their Burgers vectors differ relative to the dislocation line, and why does this geometric difference make screw dislocations more mobile?
If asked to explain why polycrystalline silicon has lower carrier mobility than single-crystal silicon, which defect types would you discuss and why?
A semiconductor device shows reduced efficiency due to carrier recombination. Would you suspect deep-level traps or shallow-level dopants? Explain the energy-level reasoning.
Contrast how voids and precipitates affect mechanical properties. Under what processing conditions might each form during semiconductor fabrication?
Key Semiconductor Defects to Know for Semiconductor Physics