Temperature Dependence of Semiconductor Bandgaps: Varshni and Bose-Einstein Models

Introduction to Bandgap Temperature Dependence

The temperature dependence of a semiconductor’s bandgap is a fundamental property with significant implications for electronic and optoelectronic device performance. This phenomenon, characterized by a narrowing bandgap with increasing temperature, is primarily governed by two physical mechanisms: lattice thermal expansion and electron-phonon coupling. Understanding this behavior is crucial for the design and optimization of devices such as lasers, photodetectors, and solar cells.

Primary Models for Bandgap Narrowing

Two principal models are employed to describe the temperature-induced reduction in bandgap energy: the empirical Varshni model and the more fundamental Bose-Einstein model.

The Varshni Model

Proposed by Y.P. Varshni, this model provides a widely used empirical formula:

E_g(T) = E_g(0) – (αT²)/(T + β)

• E_g(0): Bandgap energy at absolute zero temperature.
• α: A parameter related to the strength of the electron-phonon interaction.
• β: A parameter often associated with the Debye temperature or an average phonon energy.

The model effectively combines the contributions from lattice expansion and electron-phonon coupling into a single expression. It offers good accuracy for many common semiconductors, including silicon and germanium.

The Bose-Einstein Model

This model is derived from a theoretical framework incorporating Bose-Einstein statistics for phonons:

E_g(T) = E_g(0) – a_B / [exp(Θ_B/T) – 1]

• a_B: Represents the electron-phonon coupling strength.
• Θ_B: A characteristic phonon temperature.

This formulation provides a more physically grounded description, particularly for materials with strong electron-phonon interactions, such as polar semiconductors like gallium arsenide (GaAs).

Underlying Physical Mechanisms

The temperature dependence arises from two distinct effects:

• Lattice Thermal Expansion: Heating the crystal increases interatomic distances, which reduces orbital overlap and consequently decreases the bandgap. For instance, silicon’s lattice constant increases by approximately 2.6 × 10^-6 K^-1.
• Electron-Phonon Coupling: Phonons introduce dynamic perturbations to the crystal potential, leading to bandgap renormalization. This effect is dominant in polar materials where interactions with longitudinal optical phonons are significant.

Comparison and Application

While the Varshni model is valued for its simplicity and broad applicability, the Bose-Einstein model offers deeper physical insight, especially for specific material classes. The choice of model depends on the required accuracy and the specific semiconductor material under investigation.