Modeling integrated frequency shifters and beam splitters

Photonic quantum computing is a strong contender in the race to fault-tolerance. Recent proposals using qubits encoded in frequency modes promise a large reduction in hardware footprint, and have garnered much attention. In this encoding, linear optics, i.e., beam splitters and phase shifters, is necessarily not energy-conserving, and is costly to implement. In this work, we present designs of frequency-mode beam splitters based on modulated arrays of coupled resonators. We develop a methodology to construct their effective transfer matrices based on the SLH formalism for quantum input-output networks. Our methodology is flexible and highly composable, allowing us to define $N$-mode beam splitters either natively based on arrays of $N$-resonators of arbitrary connectivity or as networks of interconnected $l$-mode beam splitters, with $l<N$. We apply our methodology to analyze a two-resonator device, a frequency-domain phase shifter and a Mach-Zehnder interferometer obtained from composing these devices, a four-resonator device, and present a formal no-go theorem on the possibility of natively generating certain $N$-mode frequency-domain beam splitters with arrays of $N$-resonators.

💡 Research Summary

This paper addresses a critical challenge for frequency‑encoded photonic quantum computing: the implementation of linear optical components—beam splitters and phase shifters—that necessarily exchange energy with the optical field. The authors propose a family of devices built from arrays of electro‑optically modulated coupled ring resonators and develop a rigorous theoretical framework based on the SLH (Scattering‑Lindblad‑Hamiltonian) formalism for quantum input‑output networks (QIONs).

First, the SLH description of each resonator network is mapped to an ABCD state‑space representation, allowing the construction of a frequency‑domain transfer function Ξ(ω)=C(iωI−A)^{-1}B+D. Because the resonators are driven by a time‑dependent modulation, the A matrix is initially time‑varying, precluding a straightforward Fourier analysis. The authors resolve this by applying two successive rotating‑wave approximations (RWAs). The first RWA aligns the modulation frequency ωd with the splitting Δ12 of the two normal modes of the unmodulated two‑resonator system, eliminating fast oscillating terms. The second RWA exploits the regime ΓL≪Δ12, where the coupling to the waveguide is weak compared with the mode splitting, allowing each normal mode to be treated as coupled to an independent input port. In this regime the system becomes effectively time‑invariant, and a closed‑form transfer matrix can be derived.

Using this methodology, the paper analyses several concrete devices.

1. Two‑resonator beam splitter (single‑waveguide) – With identical resonators (ωa1=ωa2=ω0) and strong inter‑resonator coupling u, the normal modes are symmetric and antisymmetric combinations separated by Δ12=2u. By setting ωd≈Δ12 and choosing the modulation amplitude ε and phase ϕ, the device implements an arbitrary 2‑mode frequency beam splitter. The derived transfer matrix reproduces the experimentally observed splitting ratios reported by Hu et al.

2. Two‑resonator Mach‑Zehnder interferometer (dual‑waveguide) – Adding a second waveguide on the right side creates two independent output ports. By cascading two of the above beam splitters with a relative phase shifter (realized by a static detuning of one resonator), the authors construct a frequency‑domain Mach‑Zehnder interferometer. The overall transfer matrix is the product of the constituent matrices, demonstrating full composability.

3. Four‑resonator 4‑mode beam splitter – Extending the architecture to four resonators permits native implementation of a 4‑mode unitary transformation. Various coupling topologies (linear chain, ring, cross‑linked) are explored. Numerical optimization of the modulation phases and amplitudes shows that certain topologies achieve near‑unitary mixing across all four frequency channels while maintaining low insertion loss. Sensitivity analysis reveals that performance is limited primarily by resonator quality factor (Q) and the precision of the modulation waveform.

4. No‑go theorem for N‑mode native beam splitters – The authors generalize the SLH/ABCD approach to an arbitrary number N of resonators. By examining the rank and symmetry constraints imposed by the linearized Hamiltonian and the coupling matrix, they prove that for N>4 there exist families of unitary N‑mode beam splitters that cannot be realized with any static inter‑resonator connectivity combined with a single-tone modulation. The theorem formalizes an intuitive limitation: the number of independent control parameters (modulation amplitudes, phases, and coupling strengths) grows linearly with N, whereas a generic N‑mode unitary requires O(N^2) parameters. Consequently, only a restricted subset of unitaries is accessible natively; the rest must be synthesized from cascaded lower‑mode blocks.

The paper concludes by emphasizing the practical relevance of the derived transfer matrices: they enable rapid circuit‑level simulation of large frequency‑domain photonic networks without resorting to full time‑domain master‑equation integration. The authors also discuss experimental considerations—high‑speed electro‑optic modulators, low‑loss waveguide‑resonator coupling, and thermal stability—and outline future directions such as multi‑tone modulation schemes, error‑corrected frequency‑bin operations, and integration with superconducting microwave‑to‑optical transducers.

Overall, the work provides a solid theoretical foundation for designing scalable, low‑loss frequency‑mode linear optics, bridging the gap between abstract quantum‑information protocols and realistic integrated photonic hardware.