Avalanche photodiodes (APDs) are specialized semiconductor devices designed to provide internal gain through impact ionization, enabling high sensitivity in low-light conditions. Unlike standard photodiodes, which convert photons into electron-hole pairs without amplification, APDs leverage the avalanche effect to multiply photogenerated carriers, significantly enhancing their responsivity. This makes them indispensable in applications where detecting weak optical signals is critical, such as LiDAR, fiber-optic communications, and biomedical imaging.

The internal gain mechanism in APDs is rooted in impact ionization, a process where high-energy charge carriers collide with the semiconductor lattice, generating additional electron-hole pairs. When a reverse bias voltage close to the breakdown threshold is applied, the electric field accelerates photogenerated carriers to sufficient kinetic energies. These energetic carriers then ionize bound electrons upon collision, creating secondary carriers that further contribute to the current. The multiplication factor (M) quantifies the gain, defined as the ratio of the total output current to the primary photocurrent. For silicon-based APDs, M can exceed 100, while InGaAs APDs typically achieve gains between 10 and 40 due to material-dependent ionization coefficients.

However, the stochastic nature of impact ionization introduces noise, characterized by the excess noise factor (F). This parameter accounts for the statistical fluctuations in the multiplication process and is influenced by the ratio of electron and hole ionization coefficients (k). For materials where one type of carrier dominates impact ionization (e.g., electrons in silicon), F remains relatively low. The excess noise factor is empirically modeled by F = M (1 - (1 - k) ((M - 1)/M)^2). In silicon APDs, k values range from 0.001 to 0.1, resulting in F values near 2 at M = 10. In contrast, InGaAs APDs exhibit higher k (~0.3–0.5), leading to F > 5 at similar gains, which necessitates careful trade-offs between sensitivity and noise performance.

APD structures are engineered to optimize carrier multiplication while minimizing dark current and noise. Reach-through APDs employ a layered design where a low-field absorption region ensures efficient photon absorption, while a high-field multiplication region triggers impact ionization. This separation reduces the likelihood of tunneling-induced dark current. Another advanced configuration is the separate absorption and multiplication (SAM) structure, commonly used in InGaAs/InP APDs. Here, InGaAs absorbs infrared photons due to its narrow bandgap, while InP, with its wider bandgap, hosts the multiplication region. The heterojunction between these materials must mitigate charge trapping at the interface to prevent response delays or gain suppression.

Low-light applications heavily rely on APDs due to their superior signal-to-noise ratios. In fiber-optic communications, APDs enable long-haul data transmission by detecting attenuated optical signals at gigabits-per-second rates. For instance, 1550 nm wavelength systems use InGaAs APDs with gains of ~10 to achieve receiver sensitivities below -30 dBm. LiDAR systems, particularly those used in autonomous vehicles, exploit APDs for time-of-flight measurements, where their rapid response and internal gain enhance detection ranges even under ambient light interference. Biomedical imaging techniques like fluorescence lifetime microscopy also benefit from APDs, as their gain allows single-photon detection without external amplification circuits.

The performance of APDs is further influenced by temperature and bias voltage stability. Cooling the device reduces dark current, which is critical for minimizing noise in high-gain operation. For example, silicon APDs operating at -20°C exhibit dark currents below 1 nA, compared to microampere-level currents at room temperature. Bias voltage must be precisely controlled, as even minor fluctuations can drastically alter the gain or push the device into irreversible breakdown. Modern APDs integrate temperature compensation circuits and active bias regulation to maintain optimal performance.

Future advancements in APD technology focus on improving materials and architectures to reduce excess noise and extend spectral coverage. Superlattice and quantum-dot-based APDs aim to engineer ionization coefficients for lower F, while wide-bandgap semiconductors like GaN and SiC are being explored for ultraviolet and high-temperature applications. Despite these innovations, the fundamental trade-offs between gain, noise, and speed will continue to guide APD design for emerging optoelectronic systems.

In summary, avalanche photodiodes leverage impact ionization to achieve internal gain, making them vital for low-light detection. Their performance is governed by the excess noise factor and carefully tailored structures like reach-through and SAM designs. With applications spanning telecommunications, LiDAR, and biomedical imaging, APDs represent a critical intersection of semiconductor physics and practical optoelectronics. Ongoing research seeks to further refine their noise characteristics and expand their operational domains, ensuring their relevance in future photonic technologies.