Table of Contents
Fetching ...

Quantum-Limited Acoustoelectric Amplification in a Piezoelectric-2DEG Heterostructure

Eric Chatterjee, Daniel Soh, Matt Eichenfield

TL;DR

This work develops a quantum-mechanical description of acoustoelectric amplification in a 2DEG on a piezoelectric substrate, showing how a dc drift creates population inversion that enables stimulated phonon emission and gain for propagating acoustic waves. By deriving the 2DEG first-order acoustic susceptibility via Lindhard theory with Mermin corrections and analyzing it across high-mobility/low-drift, low-mobility, and high-drift regimes, the paper connects quantum results to classical predictions in appropriate limits. The gain per unit length and the corresponding quantum noise are computed, and a full quantum model incorporating pump depletion yields a clamping condition that sets a maximum phonon intensity. These results inform the design of quantum acoustic devices, including phononic lasers and phase-insensitive quantum phononic amplifiers, with potential for on-chip, ultra-coherent phonon sources and integrated nonlinear phononics.

Abstract

We provide a quantum mechanical description of phonon amplification in a heterostructure consisting of a two-dimensional electron gas (2DEG) stacked on top of a piezoelectric material. An applied drift voltage effectively creates a population inversion in the momentum states of the 2DEG electrons, giving rise to spontaneous emission of phonons. Once an acoustic wave is launched, the pumped electrons release phonons via stimulated emission, returning to depleted ground states before being pumped back to the excited states. We show that whereas efficient amplification using a 1D electron gas requires the acoustic wavelength to roughly equal the average electron-electron spacing, a 2DEG enables efficient amplification for any wavelength greater than the average electron-electron spacing. We derive the imaginary and real parts of the 2DEG first-order acoustic susceptibility as functions of electronic drift velocity in specific limits and derive the gain per unit length for the signal and the quantum noise, with the gain matching the classical result in the short-electronic-lifetime (low-mobility) regime. Moreover, we analyze the gain clamping due to pump depletion and calculate the maximum achievable intensity. Our results provide a framework for designing novel acoustic devices including a quantum phononic laser and phase-insensitive quantum phononic amplifiers.

Quantum-Limited Acoustoelectric Amplification in a Piezoelectric-2DEG Heterostructure

TL;DR

This work develops a quantum-mechanical description of acoustoelectric amplification in a 2DEG on a piezoelectric substrate, showing how a dc drift creates population inversion that enables stimulated phonon emission and gain for propagating acoustic waves. By deriving the 2DEG first-order acoustic susceptibility via Lindhard theory with Mermin corrections and analyzing it across high-mobility/low-drift, low-mobility, and high-drift regimes, the paper connects quantum results to classical predictions in appropriate limits. The gain per unit length and the corresponding quantum noise are computed, and a full quantum model incorporating pump depletion yields a clamping condition that sets a maximum phonon intensity. These results inform the design of quantum acoustic devices, including phononic lasers and phase-insensitive quantum phononic amplifiers, with potential for on-chip, ultra-coherent phonon sources and integrated nonlinear phononics.

Abstract

We provide a quantum mechanical description of phonon amplification in a heterostructure consisting of a two-dimensional electron gas (2DEG) stacked on top of a piezoelectric material. An applied drift voltage effectively creates a population inversion in the momentum states of the 2DEG electrons, giving rise to spontaneous emission of phonons. Once an acoustic wave is launched, the pumped electrons release phonons via stimulated emission, returning to depleted ground states before being pumped back to the excited states. We show that whereas efficient amplification using a 1D electron gas requires the acoustic wavelength to roughly equal the average electron-electron spacing, a 2DEG enables efficient amplification for any wavelength greater than the average electron-electron spacing. We derive the imaginary and real parts of the 2DEG first-order acoustic susceptibility as functions of electronic drift velocity in specific limits and derive the gain per unit length for the signal and the quantum noise, with the gain matching the classical result in the short-electronic-lifetime (low-mobility) regime. Moreover, we analyze the gain clamping due to pump depletion and calculate the maximum achievable intensity. Our results provide a framework for designing novel acoustic devices including a quantum phononic laser and phase-insensitive quantum phononic amplifiers.

Paper Structure

This paper contains 19 sections, 125 equations, 8 figures.

Figures (8)

  • Figure 1: Diagram of acoustoelectric amplifier. Note that the external dc electric field $\bm{E_d}$ provides the energy for amplifying the traveling acoustic wave.
  • Figure 2: Shift in semiconductor electronic spectrum due to drift electric field for the 1D case. Note that the range of occupied states shifts by the drift wavevector $k_d$, raising (lowering) the highest occupied energy level for carriers traveling along (against) the dc electric field by $\Delta E$ in the limit $k_d \ll k_F$.
  • Figure 3: Phase-space diagram of the regions $S$ (a), and $S'$ (b), depicted as the green-shaded crescent-shaped areas, representing the electrons that can emit (absorb) phonons of wavevector $\bm{q} = q\hat{x}$ due to the corresponding final states being unoccupied.
  • Figure 4: Phase-space diagram (a) of the acoustoelectric amplification process in the high-mobility/low-drift-velocity limit given an acoustic field propagating in the $+\hat{x}$-direction with wavevector $q$, along with dimensions of each emissive (red) region (b).
  • Figure 5: Numerical results for the real (a) imaginary (b) parts of $\chi^{(1)}(-\omega_0)$ as functions of the drift velocity $v_d$, given a mobility (in units of $\textrm{m}^2/(\textrm{V} \cdot \textrm{s})$) of 100 (solid, blue), or 1000 (solid, green), along with the analytical results (dotted, red) for the high-mobility/low-drift-velocity limit $\gamma/q,v_d \ll v_F$. We assume a phonon angular frequency of $\omega_0 = 2\pi \times 10^9 \textrm{ s}^{-1}$, speed of sound $v_s = 4 \times 10^3$ m/s, a carrier density $n = 2 \times 10^{15} \textrm{ m}^{-2}$, a carrier effective mass $m = 0.067 m_0$, and a 2DEG thickness $t_\mathrm{2DEG} = 2 \times 10^{-8} \textrm{ m}$.
  • ...and 3 more figures