Quantum oscillations, terahertz cyclotron resonance and two-carrier Hall analysis of polarization-induced GaN/AlN two-dimensional hole gases report subband masses and lifetimes that disagree. We solve the six-band $k·p$ problem self-consistently with Poisson's equation at the measured density, with a finite AlN barrier. The computed heavy-hole mass, 1.92 to $1.99m_0$, agrees with the measured $1.92\pm 0.16m_0$; the cyclotron-resonance value differs because that resonance is overdamped. The two measured subband densities share one Fermi level only if the light-hole band is about one and a half times heavier than calculated, the discrepancy that both techniques report in the light-hole mass. Only $A_6$, the least well determined band parameter, moves the light subband alone. Rescaled to the measured light-hole density, it leaves the heavy holes unchanged, agrees with the light-hole cyclotron-resonance mass of a second sample and gives a Lifshitz-Kosevich mass that rises with field; as it lies near the limit of a bounded Hamiltonian, we treat it as an effective parameter. For scattering, a coupled two-subband Boltzmann equation excludes any single mechanism. The heavy-hole quantum mobility, inferred from the onset of its oscillations, is overestimated, because the heavy holes dominate the resistance; a Dingle fit of the published oscillations and their amplitude ratio both give 95 instead of 167 to $200{\mathrm{cm}}^2\mathrm{V}{}^{-1}{\mathrm{s}}^{-1}$. The four mobilities then fix the long-range disorder spectrum, $q^{-3.2}$ to $q^{-3.6}$, between in-plane and threading charged lines, and the two-carrier Hall analysis is not the cause of the disagreement.

Close

PrevNext

View figure in article

• Electronic structure
• Spin-orbit coupling
• Quantum wells
• Two-dimensional electron gas
• Wide band gap systems
• Boltzmann theory
• Effective mass theory