Compound Semiconductors

Introduction

Elemental silicon is not the only semiconductor platform. Combining elements can produce direct optical gaps, high electron mobility and layer structures tailored to light emission, detection or high-frequency electronics. Group 13–15 compounds such as GaAs and InP are called III–V semiconductors in the traditional group notation; group 12–16 compounds such as CdTe and ZnSe are called II–VI. Alloying can tune properties continuously in principle, but chemical composition, crystal strain and defects must be controlled together.

Core explanation

Compound semiconductors have periodic bonding networks with contributions from covalency and ionicity that vary by composition. Their valence and conduction bands reflect orbital energies, bonding strength and crystal symmetry. GaAs has a direct fundamental gap in its usual zinc-blende structure; silicon's familiar phase has an indirect gap. This makes GaAs attractive for efficient near-band-edge emission. Other III–V and II–VI compounds span different gaps and lattice parameters. The family labels describe periodic-table origins, not a universal electrical charge on each atom in the crystal. MIT's compound-semiconductor course begins with these families, their band structures and gap-versus-lattice-parameter trends.

In an alloy such as Al xGa (1−x)As, the fraction x changes the average chemical environment and often changes the band gap. A simple interpolation E g(x) ≈ (1−x)E g(A) + xE g(B) − bx(1−x) introduces a bowing parameter b to represent nonlinearity, for a specified phase and temperature. The expression is empirical and not a guarantee that an alloy is miscible at every composition or retains the same direct-gap character. Band extrema can cross as composition changes, producing a direct-to-indirect transition. Consequently an optical-device designer checks both gap energy and its k-space character.

Layers with different compositions form heterojunctions. A wider-gap layer can confine carriers in a narrower-gap active region, helping LED and laser operation; carefully designed band offsets matter. Lattice mismatch creates elastic strain, and beyond a thickness or strain condition the material may relax by forming dislocations. Those defects can aid nonradiative recombination and undermine the very optical benefit sought from the alloy. An ideal band diagram must therefore be paired with a growth and defect analysis. Some systems deliberately use strained layers below a critical thickness to tune bands without creating excessive dislocations.

Material choice includes toxicity, abundance, substrate compatibility and process stability. Cadmium-containing semiconductors can be excellent optical absorbers, but containment and lifecycle management matter. III–V growth can be expensive compared with bulk silicon, yet their high optical performance may justify use in space photovoltaics, lasers or high-speed components. The US Department of Energy's III–V photovoltaic program discusses multijunction III–V materials and their spectral advantages.

Step-by-step reasoning

1. Identify the elements and semiconductor family without assuming fully ionic bonding. 2. Determine gap energy and direct or indirect character for the composition and crystal phase. 3. Consider alloy bowing rather than linear interpolation by default. 4. Compare lattice parameters and band offsets when stacking different materials. 5. Check defects, growth method, stability and environmental constraints for the application.

Visual explanation

Plot band gap against lattice parameter for several compounds and draw an alloy trajectory between two endpoints. A second sketch shows a narrow-gap active layer between wider-gap layers, with conduction and valence band offsets forming carrier confinement. Mark dislocations at a heavily mismatched interface to show why a favourable band diagram alone is insufficient.

Real-world analogy

Mixing two ingredients can tune a recipe's flavour, but the result need not be an exact average: interactions create nonlinear effects. Stacking two baked layers also requires them to expand similarly, or cracks form. Alloy bowing and lattice mismatch are not literal cooking processes, but the analogy reminds us that composition and structural compatibility are distinct design axes.

Real-world example

A multijunction solar cell can place a larger-gap III–V absorber above a smaller-gap absorber. High-energy photons are captured first, while lower-energy photons pass to lower layers. The alloy compositions are tuned for spectral coverage, but the stack must maintain suitable lattice quality, contacts and current matching. The physics of tandem collection is the same spectral tradeoff discussed for single-junction cells, now addressed with multiple materials.

Why?

Why can changing alloy fraction alter optical colour? Replacing atoms changes orbital energies, bond lengths and the periodic potential, shifting band edges. An emitted or absorbed photon probes an allowed electronic transition between those bands. The colour is therefore an electronic-structure consequence of composition, not a simple paint mixture of the endpoint solids.

Common misconception

“Every III–V compound has a direct gap” is false; gap character varies with material and composition. “II–VI means ions have exactly +2 and −2 formal charges everywhere” overstates a family name and ignores covalent bonding. “Lattice match guarantees a good device” ignores impurities, interface traps and growth defects.

Worked example

Suppose a hypothetical direct-gap alloy has endpoint gaps 1.4 eV and 2.2 eV and bowing parameter b = 0.4 eV. At x = 0.50, the estimate is E g ≈ 0.5(1.4) + 0.5(2.2) − 0.4(0.5)(0.5) = 1.70 eV . Pure linear interpolation would give 1.80 eV. This is an illustrative use of the model, not a measured gap for a named alloy; one would also verify that the gap remains direct at x = 0.50.

Quick check

1. Why can an alloy's gap differ from a straight-line average of its endpoint gaps? Answer: Different local bonding and electronic interactions can produce nonlinear composition dependence, represented by a bowing term.

Exam focus

Name the periodic-table families accurately and state whether the specific material's gap is direct or indirect. Use a bowing parameter only when supplied or justified. In heterostructure answers, mention lattice strain and dislocation risk as well as band offsets. Separate a material's electronic promise from practical growth and safety considerations.

Advanced insight

Band offsets between two semiconductors depend on interface chemistry and strain, not only on the difference between their bulk gaps. Quantum wells in heterostructures create discrete subbands whose transition energies can be tuned by layer thickness. Alloy disorder can broaden optical spectra and scatter carriers, so composition tuning can trade spectral flexibility for mobility or linewidth.

Summary

III–V and II–VI compounds expand the range of semiconductor band structures. Alloying and layering tune optical transitions and carrier confinement, while bowing, lattice mismatch, defects and stability constrain real devices. A successful design treats chemical composition and crystal quality together.

Practice questions

1. Classify GaAs and CdTe by traditional semiconductor family. Answer: GaAs is III–V; CdTe is II–VI. 2. Why is a direct gap useful for an LED active region? Answer: Band-edge radiative recombination can occur without a phonon supplying a large crystal-momentum change. 3. What is a possible consequence of severe lattice mismatch in a thick grown layer? Answer: Strain relaxation can create dislocations that scatter carriers or promote nonradiative recombination. 4. Does x = 0.5 in a binary alloy always imply the midpoint gap? Answer: No. Bowing and possible changes of phase or band extrema can make the dependence nonlinear.