Predicted third-order sweet spots for phi-junction Josephson parametric amplifiers

Hybrid superconductor-semiconductor nanowire Josephson junctions exhibit skewed and phi-shifted current phase relations when an in-plane magnetic field is applied along the weak link’s spin-orbit effective field direction. These junctions can have an asymmetric Josephson potential with odd-order nonlinearities. A dominant third-order nonlinearity can be achieved by tuning the magnetic field to a sweet spot. Sweet spots persist when higher order Josephson harmonics are included. This makes it possible to have a single Josephson junction dipole element with three-wave mixing capability, which is favorable for pump-efficient amplification. Electrostatic gate tunability of the semiconductor weak link can make it operable within an extended range of working frequencies, and the inclusion of micromagnets can facilitate near-zero magnetic field operation.

💡 Research Summary

The paper proposes a novel single‑junction Josephson parametric amplifier (JPA) based on a hybrid superconductor‑semiconductor nanowire that operates in a three‑wave‑mixing regime. The key idea is to exploit the so‑called φ₀‑junction effect, which arises when an in‑plane magnetic field is applied parallel to the effective spin‑orbit field of the nanowire. Under these conditions the current‑phase relation (CPR) becomes asymmetric and acquires a global phase shift ϕ₀, i.e. I(ϕ)=I₁ sin(ϕ+ϕ₀)+I₂ sin(2ϕ+2ϕ₀+δ₁₂). Both the amplitudes I₁, I₂ and the phase offsets ϕ₀, δ₁₂ depend on the magnetic field B (ϕ₀≈a B, δ₁₂≈c B, I₁≈α₁(1−B²), I₂≈α₂(1−B²) for B≪B_c≈100 mT).

Expanding the Josephson potential U(ϕ)=−∫I(ϕ)dϕ around its minimum yields a Taylor series U(ϕ̃)=c₂ϕ̃²+c₃ϕ̃³+c₄ϕ̃⁴+…, where c₃ and c₄ are the third‑ and fourth‑order nonlinear coefficients that determine three‑wave mixing strength and Kerr (self‑phase) nonlinearity, respectively. The authors search for “sweet spots” where |c₄|≈0 (Kerr‑free) while |c₃| is maximized. Using a hybrid genetic algorithm to vary the magnetic field B and the phenomenological parameters a and c, they identify multiple sweet‑spot regions at B≈±0.26 B_c and ±0.39 B_c. The global phase shift a does not affect c₃ or c₄, whereas the relative phase c controls the location of the sweet spots.

Importantly, the authors verify that the sweet‑spot condition persists when higher harmonics (e.g., a third harmonic I₃) are added to the CPR, indicating robustness against realistic multi‑mode CPRs.

For device implementation, a single φ₀‑junction is embedded in a microwave resonator. The junction is realized with a Sn‑InSb or Sn‑InAs nanowire partially covered by a thin superconducting shell; a break in the shell defines the weak link. Gate voltage V_g tunes the carrier density, allowing the critical current (≈100–500 nA) and thus the Josephson energy (E_J≈50–250 GHz) to be varied, which in turn tunes the resonant frequency over several gigahertz.

To achieve the required magnetic field without disturbing the surrounding circuitry, the authors propose placing a micromagnet (e.g., CoFe) near the junction. Micromagnets can generate local fields of ±100 mT, while the bulk superconducting structures remain shielded, enabling near‑zero‑field operation.

Using realistic circuit parameters (L_J≈3.2 nH, L≈0.4 nH, ω₀≈20 GHz) and the optimal magnetic field, the calculated three‑wave coupling is g₃/2π≈32 MHz and the Kerr term g₄ is essentially zero. Within the rotating‑wave approximation the Hamiltonian reads
Ĥ/ħ = ω_r a†a + g₃(a+a†)³ + g₄(a+a†)⁴,
with ω_r≈3.9 GHz. Input‑output theory predicts a gain of 20 dB over a 40 MHz bandwidth for a coupling rate κ/2π≈0.4 GHz and modest signal/idler amplitudes (α_s≈α_i≈0.01). The 1‑dB compression point P₋1 dB is comparable to state‑of‑the‑art SNAIL‑based JPAs, indicating that the single‑junction design does not sacrifice dynamic range.

In summary, the work demonstrates that a φ₀‑junction can provide dominant third‑order nonlinearity while suppressing the fourth‑order Kerr term, enabling efficient three‑wave mixing with a single Josephson element. The approach offers several practical advantages: reduced circuit complexity and footprint (no need for arrays of SNAILs), gate‑tunable operating frequency, and the possibility of magnetic‑field‑free operation via integrated micromagnets. The authors outline a clear path toward experimental realization, including device fabrication, micromagnet design, and integration with magnetic‑compatible superconducting materials. This platform could become a valuable building block for quantum‑limited amplification, frequency conversion, and other microwave quantum optics applications in superconducting quantum processors.