Gallium arsenide (GaAs) stands as a cornerstone of modern semiconductor technology, prized for its superior electron mobility and direct bandgap. These properties enable GaAs to excel in high-frequency electronics, optoelectronics, and photovoltaic devices. However, the precise electrical behavior of GaAs—particularly its electrical conductivity—is exquisitely sensitive to the presence and concentration of impurity atoms within its crystal lattice. Understanding how impurity levels shape conductivity is essential for engineers and researchers who design devices such as monolithic microwave integrated circuits (MMICs), high-electron-mobility transistors (HEMTs), laser diodes, and space-grade solar cells. This article explores the fundamental relationships between dopants, contaminants, and the resulting charge carrier dynamics in GaAs, providing a comprehensive view of how impurity management drives performance optimization.

Gallium Arsenide Fundamentals

Gallium arsenide is a compound semiconductor formed from elements in Group III and Group V of the periodic table. Its zincblende crystal structure provides high electron mobility—roughly five to six times greater than that of silicon at room temperature—and a direct bandgap of approximately 1.42 eV at 300 K. The direct bandgap means that electron‑hole recombination can occur with high radiative efficiency, making GaAs the material of choice for light‑emitting diodes and laser diodes operating in the near‑infrared spectrum. Additionally, the material's wide bandgap and high breakdown field enable operation at elevated temperatures and frequencies, which is why GaAs dominates the market for power amplifiers in mobile phone base stations and radar systems.

In its pure, intrinsic form, GaAs has a relatively low conductivity because the intrinsic carrier concentration (electrons and holes generated by thermal excitation across the bandgap) is on the order of 106 cm−3 at room temperature. For most device applications, the conductivity must be controlled over many orders of magnitude, a feat accomplished entirely through the intentional introduction of impurities—a process called doping.

The Role of Impurities in Semiconductors

Impurities can be classified into two broad categories: intentional dopants that are deliberately added to tailor electrical properties, and unintentional contaminants that arise from growth precursors, crucibles, or process environments. Both types profoundly affect the concentration and mobility of charge carriers. In GaAs, even sub‑parts‑per‑million levels of certain impurities can change the conductivity from semi‑insulating to highly conductive, or vice versa. Understanding the interplay between impurity type, concentration, and distribution is the foundation of semiconductor device engineering.

Donor and Acceptor Impurities

Donor impurities are elements that have more valence electrons than the host atom they replace. For GaAs, typical donors include silicon (Si), tellurium (Te), tin (Sn), and sulfur (S). When a donor atom substitutes for gallium (e.g., silicon on a gallium site), its extra electron is loosely bound and can be thermally excited into the conduction band, leaving behind a positively charged ion. The resulting material is n‑type, with electrons as the majority carriers. Conversely, acceptor impurities have fewer valence electrons and naturally create holes in the valence band. Common acceptors in GaAs are zinc (Zn), carbon (C), beryllium (Be), and magnesium (Mg). When an acceptor substitutes for arsenic (e.g., carbon on an arsenic site), it accepts an electron, generating a mobile hole and making the material p‑type.

Compensation and Deep Levels

Not all impurities act as simple donors or acceptors. Some create deep levels—energy states near the middle of the bandgap—that can trap charge carriers and reduce conductivity. For example, oxygen contamination in GaAs forms a deep donor level that tends to compensate p‑type materials, degrading device performance. Additionally, a well‑known deep level in GaAs is the EL2 defect, which is a complex involving an arsenic antisite (AsGa) and related vacancies. EL2 is commonly used to create semi‑insulating GaAs substrates; its deep donor level pins the Fermi level near mid‑gap, yielding resistivities exceeding 107 Ω·cm.

Compensation occurs when donors and acceptors are simultaneously present. In highly compensated material, the net carrier concentration is the difference between donor and acceptor densities, but mobility suffers because ionized impurity scattering increases. Controlling compensation is critical for achieving reproducible conductivity in GaAs layers grown by molecular‑beam epitaxy (MBE) or metal‑organic chemical‑vapor deposition (MOCVD).

Impact on Electrical Conductivity

The electrical conductivity σ of a semiconductor is given by

σ = q (n μn + p μsub>p)

where q is the elementary charge, n and p are the electron and hole concentrations, and μn and μp are the respective mobilities. Impurity levels influence both the carrier concentrations and the mobilities. For an extrinsic, n‑type GaAs sample doped with donors of concentration ND and containing acceptors of concentration NA, the net electron concentration is approximately n = ND − NA (assuming full ionization at room temperature). Increasing donor density raises n, but also introduces more ionized impurity centers that scatter electrons, reducing mobility.

Ionized Impurity Scattering

As the impurity concentration exceeds about 1017 cm−3 in GaAs, the dominant scattering mechanism shifts from lattice vibrations (phonon scattering) to ionized impurity scattering. Coulomb interactions between charge carriers and ionized dopants increase the scattering rate, causing mobility to drop sharply. This effect is particularly pronounced for low‑temperature applications, where phonon scattering is suppressed. For example, at 77 K, mobility in GaAs can exceed 105 cm2/V·s at low doping, but it plummets to a few thousand at doping levels above 1018 cm−3.

Neutral Impurity Scattering

When dopants are not fully ionized—which can occur at low temperatures or for deep levels—the neutral impurity atoms also scatter carriers, albeit with a cross‑section that is generally smaller than for ionized impurities. In heavily doped GaAs, both ionized and neutral scattering must be considered to accurately model conductivity. Additionally, scattering from alloy disorder (in ternary compounds like AlGaAs) and from interface roughness in heterostructures can further reduce mobility.

Optimal Doping Strategies for Device Performance

Choosing the right impurity level is a balancing act that depends on the specific device application. Too few impurities result in high resistivity and low current drive; too many cause mobility collapse, parasitic series resistance, and increased leakage. Engineers employ several techniques to achieve the desired conductivity profile.

Uniform versus Modulation Doping

In traditional field‑effect transistors, a uniformly doped channel is used. For a GaAs MESFET, typical channel doping levels range from 1×1017 to 5×1017 cm−3 to balance sheet carrier density and mobility. However, for high‑speed HEMTs, a technique called modulation doping (or delta doping) is employed. A thin, heavily doped donor layer (e.g., Si delta‑doping sheet density of 5×1012 cm−2) is placed in a wider‑bandgap barrier material (such as AlGaAs) adjacent to an undoped GaAs channel. The electrons transfer into the channel, where they experience greatly reduced ionized impurity scattering. This yields very high carrier mobilities—over 106 cm2/V·s at cryogenic temperatures—enabling ultra‑low‑noise amplifiers for satellite communications.

Doping for Optoelectronic Devices

In laser diodes and LEDs, the active region typically consists of undoped or lightly doped GaAs quantum wells to sustain high radiative efficiency. The surrounding cladding layers are doped heavily (n‑type and p‑type, e.g., 1×1018 cm−3) to provide efficient carrier injection while minimizing series resistance. P‑type doping in GaAs is often achieved with beryllium or carbon, because carbon has a low diffusion coefficient and can be introduced at high concentrations without degrading the sharp interfaces required for quantum wells.

Semi‑Insulating Substrates

For microwave devices, a semi‑insulating GaAs substrate is essential to reduce parasitic capacitance and cross‑talk. Semi‑insulating GaAs is produced by intentionally incorporating deep‑level impurities or defects that compensate residual shallow donors. The EL2 defect, an intrinsic deep donor with a concentration around 1–2×1016 cm−3, is the most common compensation mechanism. Alternatively, doping with chromium (a deep acceptor) can produce semi‑insulating behavior, though chromium contamination is often undesirable due to its detrimental effect on device reliability. The key is to achieve a net carrier concentration below 106 cm−3, corresponding to a resistivity greater than 107 Ω·cm.

Unintentional Impurities and Their Consequences

Even when clean growth conditions are maintained, unintentional impurities can sneak into GaAs crystals during growth or processing. The most common contaminants include carbon, oxygen, silicon, and hydrogen. Their sources vary: carbon originates from metal‑organic precursors (e.g., trimethylgallium) in MOCVD; oxygen can come from oxidizing agents or residual moisture; silicon may be released from quartz reactor components. Each contaminant interacts differently with the GaAs lattice.

Carbon in GaAs

Carbon typically incorporates on arsenic sites, acting as a shallow acceptor with an ionization energy of about 26 meV. In MOCVD‑grown GaAs, carbon is the dominant residual acceptor. Concentrations can range from 1014 to 1016 cm−3 depending on the growth conditions (e.g., V/III ratio, temperature). If not carefully controlled, carbon compensation can ruin the intentional doping profile, especially in thin channel layers. Precise tuning of the growth parameters—such as lowering the growth temperature or increasing the arsenic flux—can reduce carbon incorporation to acceptable levels.

Oxygen and Deep‑Level Traps

Oxygen forms a deep donor level at about Ec − 0.79 eV in GaAs. Even trace amounts (1014 cm−3 or less) can trap carriers and reduce the lifetime of injected carriers. This is especially harmful for solar cells and photodiodes, where minority‑carrier lifetime directly affects quantum efficiency. In semi‑insulating materials, oxygen can also influence the compensation ratio. Gettering techniques, such as using low‑temperature buffer layers or incorporating rare‑earth getters, help mitigate oxygen contamination.

The EL2 Defect

As mentioned earlier, the EL2 defect is ubiquitous in GaAs grown from a melt (LEC, VGF). Its concentration depends on the stoichiometry: arsenic‑rich conditions promote EL2 formation. While EL2 is useful for semi‑insulating substrates, it is detrimental in active layers because it causes carrier trapping and low‑frequency noise. In epitaxially grown layers, EL2 levels are typically much lower than in bulk substrates, but out‑diffusion from the substrate can contaminate the epitaxial film. Buffer layers and careful thermal management are used to suppress this effect.

Measurement and Characterization of Impurity Levels

Determining the exact impurity concentration and its impact on conductivity requires a suite of complementary characterization techniques.

Hall Effect Measurements

The van der Pauw Hall method is the most common technique for measuring sheet resistivity, carrier concentration, and Hall mobility. By measuring the Hall voltage in a magnetic field, one can extract the net carrier density n or p and the Hall mobility μH. Temperature‑dependent Hall measurements (e.g., from 77 K to 300 K) can separate the contributions of different impurity species, as the carrier freeze‑out behavior reveals activation energies of shallow and deep levels.

Secondary‑Ion Mass Spectrometry (SIMS)

SIMS provides direct chemical detection of impurity atoms with detection limits in the parts‑per‑billion range for many elements. Depth profiling of dopants like silicon, carbon, and oxygen is routine in epitaxial layer analysis. Correlating SIMS profiles with electrical data (e.g., from capacitance‑voltage profiling) enables the calibration of dopant activation efficiency.

Capacitance‑Voltage (C‑V) Profiling

C‑V measurements on Schottky diodes yield the net ionized impurity concentration as a function of depth. This technique is especially useful for detecting compensation and for verifying the steepness of doping profiles in δ‑doped layers. Combined with Hall data, C‑V can reveal whether all dopants are electrically active or if some are compensated by deep levels.

Photoluminescence and Deep‑Level Transient Spectroscopy (DLTS)

Low‑temperature photoluminescence (PL) spectra often show sharp excitonic peaks from bound carriers, which can be linked to specific impurities. For example, a peak at 1.5128 eV is associated with carbon acceptors. DLTS is the gold standard for characterizing deep levels, providing trap concentration, energy level, and capture cross‑section. These techniques are invaluable for identifying unintentional contaminants that degrade conductivity.

Future Directions and Research Frontiers

As device dimensions shrink and operating frequencies push into the terahertz range, controlling impurity levels with atomic precision becomes ever more important. Researchers are exploring several avenues to further optimize the conductivity of GaAs:

  • Ultra‑high‑purity GaAs: Advances in growth techniques, such as solid‑source MBE and high‑purity precursor synthesis, are pushing residual impurity levels below 1012 cm−3 in lightly doped layers, enabling record electron mobilities exceeding 3×105 cm2/V·s at room temperature.
  • Novel dopant species: Tellurium and tin are being revisited as donors because they can achieve higher doping densities than silicon without forming compensating defects. Similarly, carbon is being refined as a p‑type dopant in MBE, where it can be introduced using a heated graphite filament.
  • Digital doping and nanostructures: Atomic‑layer doping (e.g., δ‑doping with sub‑monolayer control) and the use of quantum dots and nanowires allow impurity placement with nanometer precision, minimizing scattering and maximizing charge‑carrier density.
  • In‑situ monitoring: Reflection high‑energy electron diffraction (RHEED), spectroscopic ellipsometry, and mass spectrometry are being integrated into growth chambers to provide real‑time feedback on impurity incorporation, allowing closed‑loop process control.

These innovations promise to extend the performance envelope of GaAs‑based devices, enabling next‑generation communication systems, high‑efficiency photovoltaics, and quantum‑computing heterostructures.

Conclusion

The electrical conductivity of gallium arsenide is governed by a delicate interplay between intentional doping, unintentional contamination, and crystal perfection. Donor and acceptor impurities determine the carrier concentration, while scattering from ionized and neutral impurities limits mobility. Achieving the desired conductivity for a particular device—whether it be a high‑speed HEMT, a semi‑insulating substrate, or a high‑efficiency solar cell—requires precise control over both the type and concentration of impurities. By combining advanced growth techniques with thorough characterization, engineers can harness the full potential of GaAs. Ongoing research continues to refine these methods, promising ever‑higher performance in the electronic and optoelectronic devices that underpin modern technology.

For further reading: Explore the basic properties of gallium arsenide, the physics of doping in semiconductors, and the details of the EL2 defect. A comprehensive review of impurity scattering in GaAs can be found in this recent review article.