Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2609.36431

Transverse momentum as the counter-diabatic generator in bent waveguide couplers

Yu-kai Lee, Shuo-Yen Tseng

quant-phphysics.opticsSubmitted Sep 29, 2026• Announced Sep 30, 2026• Revised Oct 16, 2026 (v2)
v1 · Sep 29, 2026v2 · Oct 16, 2026
arXivHTMLPDF

Abstract

Bending the axis of a mode-evolution coupler suppresses the nonadiabatic coupling between its supermodes, and bent couplers of this kind were recently shown to realize the counter-diabatic (CD) protocol. Here we identify the operator responsible. Rigid lateral displacement of a waveguide structure is generated by the transverse momentum $\hat p_x$, whose diagonal matrix elements vanish in a real supermode basis and whose off-diagonal element is purely imaginary; in a two-mode system such an operator is proportional to $\sy$. The CD term of a two-waveguide coupler is itself proportional to $\sy$, while the detuning and the coupling, the two parameters set by the waveguide widths and spacing, lie in the $σ_z$--$σ_x$ plane and cannot produce it. The transverse momentum is therefore the CD generator of the bent coupler, by necessity rather than by design. Equating the term it supplies to the nonadiabatic coupling gives the axis slope in closed form as a ratio of two matrix elements of the unperturbed supermodes at a single cross section, which a commutator identity recasts in coupled-mode variables as $\dot{\xo}=\dotθ/(γ\dbeta^{2})$. The expression agrees with the numerical offset search to $0.3\%$ and requires no search. Beam propagation simulations confirm the design and compare it, for the first time, with the CD protocol realized by unitary transformation in a straight coupler built from the same reference structure. The two devices reach $0.994$ and $1.000$ supermode fidelity where the untreated coupler reaches $0.886$, and agree in bandwidth and fabrication tolerance, while their internal trajectories differ by exactly the frame transformation that relates them.

AI Summary

The paper asks which operator makes axis bending implement counter-diabatic (CD) driving in a two-waveguide mode-evolution coupler, and whether it yields a direct bending prescription. It treats scalar paraxial propagation as Schrödinger evolution and projects onto two real instantaneous supermodes. In this basis, transverse momentum has zero diagonal and purely imaginary off-diagonal matrix elements, so the comoving-frame term $-\dot{x}_0\hat p_x$ is proportional to $\sigma_y$, the component required for CD driving but unavailable through detuning and coupling alone. Cancelling the nonadiabatic coupling gives $\dot{x}_0=\langle\psi_+|\partial_z\psi_-\rangle/\langle\psi_+|\partial_x\psi_-\rangle$. A commutator identity converts this into coupled-mode variables. The expression uses unshifted local modes without scanning displaced geometries; it is related to an existing curvature-based prescription by a frame transformation. Numerical offset searches agree pointwise within $0.3\%$. Scalar beam-propagation simulations compare bent and straight, gauge-transformed CD devices derived from the same reference coupler, with endpoint-flat tapers. At $1.3$ mm length, output supermode fidelities are $0.994$ and $1.000$, respectively, versus $0.886$ untreated. Both exceed $0.99$ over more than $170$ nm of wavelength and show comparable fabrication-error windows. Their different internal supermode populations are interpreted through the frame transformations. The derivation is exact only within the paraxial two-mode model. Momentum also couples guided modes to radiation, which the reduction discards; simulated radiated fractions remain below $0.6\%$. Scalar paraxial propagation additionally omits obliquity corrections of relative order $\dot{x}_0^2$. Stronger bends and high-index-contrast platforms require vectorial treatment including radiation modes.

Little sign of AI-generated textIncremental

Error found: The eigenvalue-splitting convention changes between the adiabatic Hamiltonian and the commutator section, reversing the sign of the stated coupled-mode bending law if the earlier convention is retained. The abstract's exact trajectory-equivalence claim requires the ideal two-mode approximation, whereas the propagation simulations and acknowledged radiation coupling exhibit residual differences. Facet-matched transformations at the design parameters do not imply identical wavelength and fabrication-error responses for fixed devices, and the reported sweeps demonstrate only comparable robustness.

Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.

Related papers

Counterdiabatic mode-evolution based coupled-waveguide devices

Shuo‐Yen Tseng

The goal in designing mode-evolution based devices is to realise short and high-fidelity components. The counterdiabatic protocol in coherent quantum state control can be used to cancel unwanted coupling between adiabatic modes in mode evolution but is not directly realisable in the coupled-waveguide system. By finding alternative coupled-mode equations that links to the same interaction picture dynamical equation as the counterdiabatic protocol via unitary transformations, we have derived a universal formalism for the design of short and high-fidelity mode-evolution based coupled-waveguide devices. Starting from a traditional adiabatic device design, the counterdiabatic protocol leads to a high-fidelity device, with its evolution following the adiabatic modes exactly even when the adiabatic condition is violated. Tolerance analysis shows that the countera-diabatic devices combine the advantages of adiabatic and resonant devices. The formalism is used to design asymmetric waveguide couplers.

Optics Express•Sep 3, 2013•35 citations

Inertial Counterdiabatic Driving

Julián Ferreiro-Vélez, F. A. Cárdenas-López, F. Motzoi +1 more

We introduce an inertial counterdiabatic (CD) protocol that reformulates shortcut-to-adiabaticity design as a moving-frame diagonalization problem. The method employs the Magnus expansion to construct a variational propagator generated by an accessible CD ansatz. In the new rotating frame, the controls are optimized to suppress the off-diagonal Hamiltonian terms, and the CD terms are defined to counteract the corresponding inertial contribution. The propagator generated with the Magnus expansion naturally exploits additional control directions generated through its nested commutator, while the sequential design of the propagator introduces temporal nonlocality into the optimization. We benchmark the method on a two-qubit system and on the one-dimensional quantum Ising chain, where it consistently outperforms variational CD protocols based on action minimization. The largest improvements are observed near gap-closing regions, demonstrating the advantage of combining temporal and algebraic nonlocality for suppressing diabatic excitations.

quant-pharXiv:2610.08235•Oct 7, 2026•0 citations

Nonadiabatic couplings and gauge-theoretical structure of curved quantum waveguides

Jan Stockhofe, Peter Schmelcher

We investigate the quantum mechanics of a single particle constrained to move along an arbitrary smooth reference curve by a confinement that is allowed to vary along the waveguide. The Schr\"odinger equation is evaluated in the adapted coordinate frame and a transverse-mode decomposition is performed, taking into account both curvature and torsion effects and the possibility of a cross-section potential that changes along the curve in an arbitrary way. We discuss the adiabatic structure of the problem, and examine nonadiabatic couplings that arise due to the curved geometry, the varying transverse profile, and their interplay. The exact multimode matrix Hamiltonian is taken as the natural starting point for few-mode approximations. Such approximate equations are provided, and it is worked out how these recover known results for twisting waveguides and can be applied to other types of waveguide designs. The quantum waveguide Hamiltonian is recast into a form that clearly illustrates how it generalizes the Born-Oppenheimer Hamiltonian encountered in molecular physics. In analogy to the latter, we explore the local gauge structure inherent to the quantum waveguide problem and suggest the usefulness of diabatic states, giving an explicit construction of the adiabatic-to-diabatic basis transformation.

arXiv:1311.6925•Physical Review A•Mar 24, 2014•20 citations

Landau–Zener dynamics in a curved optical directional coupler

Stefano Longhi

A simple and experimentally accessible optical realization of Landau–Zener dynamics of two-level quantum systems with a finite coupling duration is proposed using a waveguide-based directional coupler with a cubically bent axis.

Journal of Optics B Quantum and Semiclassical Optics•May 16, 2005•28 citations

Generalized directional coupling for high-precision manipulation of the optical phase for classical and quantum light

René Heilmann, Robert Keil, Stefan Nolte +1 more

Summary form only given. The manipulation of the phase of classical and quantum light is a major key to quantum photonics [1]. Integrated optics has several advantages over bulk optical systems, most notably in robustness, handling and size. Writing three-dimensional waveguides circuits into transparent materials using an ultrafast laser is thereby a new emerging technology [2; 3] with possible applications in quantum communication [4], and quantum computing [5]. In our work, we experimentally demonstrate the high-precision manipulation of the optical phase of classical and quantum light, using a generalized directional coupler.The behavior of light in directional coupler consisting of two waveguides can be described by the Hamiltonian H=T+V1+V2, where T is the kinetic energy and Vi is the potential energy of waveguide i = 1 and i = 2. The wavefunction of the system is constructed as a superposition of the single waveguide modes Ψ=a|1+b|2, and the Schrödinger equation of the coupler reads as HΨ=ksmΨ, where ksm denotes the propagation constant of the supermode. The propagation constants of the supermodes can be calculated straightforwardly as ksm1,2=k0±κ(1∓ σ)/(1-σ2) , where k0 is the propagation constant of the individual waveguides, κ=2|V2|1=1|V1|2 represents the coupling constant, and σ=1|=2|1is the direct overlap of the individual waveguide modes. For σ → 0 the well known distribution of ksm1,2 is preserved. As ksm = (ksm1+ksm2)/2>k0, in a generalized directional coupler a phase delay Δφ is introduced that readswith lC = π/κ as the coupling length. The phase delay depends only on σ, which itself is a function of the spacing between the waveguides. As the spacing can be controlled in a very accurate manner, this approach allows the highly precise manipulation of the optical phase along a directional coupler. In order to measure this effect, we set up a tuned Mach-Zehnder (MZ) interferometer with identical lengths of the interferometer arms. A third waveguide was applied to one of the arms, forming an additional directional coupler in which the light completely couples forth and back with no intensity losses in the relevant interferometer arm (Fig. 1(a)). However, as the change in the propagation constant in this arm results in a phase delay, at the output of the MZ the ratio of intensities (for classical light) and the count rates (for single photons) changes. To obtain an all-reflecting coupler (ARC) for different coupling constants, the respective length of a coupler has to be adjusted and the associated phase delays are shown in Fig. 1(b). Due to decreasing σ for larger distances between the waveguides, the accumulated phase difference shrinks, even if the length of the coupler increases. Normalizing the phase difference by length allows the direct calculation of the change of the propagation constant caused by the ARC (Fig. 1(c)).In conclusion, we demonstrated a highly precise method for the manipulation of the optical phase in integrated waveguide structures by employing generalized directional coupling. The propagation constants of the supermodes in the structures are controlled by the spacing between the waveguides in the ARC. Our approach holds for classical as well as quantum light, and has the potential to find its use in a multitude of integrated-optical devices.

Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC)•May 1, 2013•0 citations

Engineered photon-pair generation by four-wave mixing in asymmetric coupled waveguides

Robert J. A. Francis-Jones, Wright, Thomas A., Andriy Gorbach +1 more

Third-order nonlinear processes require phase matching between the interacting fields to achieve high efficiencies. Typically in guided-wave $χ^{(3)}$ platforms this is achieved by engineering the dispersion of the modes through the transverse profile of the device. However, this limits the flexibility of the phase matching that can be achieved. Instead, we analyze four-wave mixing in a pair of asymmetric waveguides and show that phasematching may be achieved in any $χ^{(3)}$ waveguide by coupling of a nondegenerate pump from an adjacent waveguide. We demonstrate the additional flexibility that this approach yields in the case of photon-pair generation by spontaneous FWM, where the supermode dispersion may be modified to produce pure heralded single photons -- a critical capability required for example by silicon platforms for chip-scale quantum photonics.

arXiv:1809.10494•Sep 27, 2018•4 citations
Bending the axis of a mode-evolution coupler suppresses the nonadiabatic coupling between its supermodes, and bent couplers of this kind were recently shown to realize the counter-diabatic (CD) protocol. Here we identify the operator responsible. Rigid lateral displacement of a waveguide structure is generated by the transverse momentum p^x\hat p_xp^​x​, whose diagonal matrix elements vanish in a real supermode basis and whose off-diagonal element is purely imaginary; in a two-mode system such an operator is proportional to \sy\sy\sy. The CD term of a two-waveguide coupler is itself proportional to \sy\sy\sy, while the detuning and the coupling, the two parameters set by the waveguide widths and spacing, lie in the σzσ_zσz​--σxσ_xσx​ plane and cannot produce it. The transverse momentum is therefore the CD generator of the bent coupler, by necessity rather than by design. Equating the term it supplies to the nonadiabatic coupling gives the axis slope in closed form as a ratio of two matrix elements of the unperturbed supermodes at a single cross section, which a commutator identity recasts in coupled-mode variables as \xo˙=θ˙/(γ\dbeta2)\dot{\xo}=\dotθ/(γ\dbeta^{2})\xo˙​=θ˙/(γ\dbeta2). The expression agrees with the numerical offset search to 0.3%0.3\%0.3% and requires no search. Beam propagation simulations confirm the design and compare it, for the first time, with the CD protocol realized by unitary transformation in a straight coupler built from the same reference structure. The two devices reach 0.9940.9940.994 and 1.0001.0001.000 supermode fidelity where the untreated coupler reaches 0.8860.8860.886, and agree in bandwidth and fabrication tolerance, while their internal trajectories differ by exactly the frame transformation that relates them.
The paper asks which operator makes axis bending implement counter-diabatic (CD) driving in a two-waveguide mode-evolution coupler, and whether it yields a direct bending prescription. It treats scalar paraxial propagation as Schrödinger evolution and projects onto two real instantaneous supermodes. In this basis, transverse momentum has zero diagonal and purely imaginary off-diagonal matrix elements, so the comoving-frame term −x˙0p^x-\dot{x}_0\hat p_x−x˙0​p^​x​ is proportional to σy\sigma_yσy​, the component required for CD driving but unavailable through detuning and coupling alone.
Cancelling the nonadiabatic coupling gives x˙0=⟨ψ+∣∂zψ−⟩/⟨ψ+∣∂xψ−⟩\dot{x}_0=\langle\psi_+|\partial_z\psi_-\rangle/\langle\psi_+|\partial_x\psi_-\ranglex˙0​=⟨ψ+​∣∂z​ψ−​⟩/⟨ψ+​∣∂x​ψ−​⟩. A commutator identity converts this into coupled-mode variables. The expression uses unshifted local modes without scanning displaced geometries; it is related to an existing curvature-based prescription by a frame transformation. Numerical offset searches agree pointwise within 0.3%0.3\%0.3%.
Scalar beam-propagation simulations compare bent and straight, gauge-transformed CD devices derived from the same reference coupler, with endpoint-flat tapers. At 1.31.31.3 mm length, output supermode fidelities are 0.9940.9940.994 and 1.0001.0001.000, respectively, versus 0.8860.8860.886 untreated. Both exceed 0.990.990.99 over more than 170170170 nm of wavelength and show comparable fabrication-error windows. Their different internal supermode populations are interpreted through the frame transformations.
The derivation is exact only within the paraxial two-mode model. Momentum also couples guided modes to radiation, which the reduction discards; simulated radiated fractions remain below 0.6%0.6\%0.6%. Scalar paraxial propagation additionally omits obliquity corrections of relative order x˙02\dot{x}_0^2x˙02​. Stronger bends and high-index-contrast platforms require vectorial treatment including radiation modes.
The goal in designing mode-evolution based devices is to realise short and high-fidelity components. The counterdiabatic protocol in coherent quantum state control can be used to cancel unwanted coupling between adiabatic modes in mode evolution but is not directly realisable in the coupled-waveguide system. By finding alternative coupled-mode equations that links to the same interaction picture dynamical equation as the counterdiabatic protocol via unitary transformations, we have derived a universal formalism for the design of short and high-fidelity mode-evolution based coupled-waveguide devices. Starting from a traditional adiabatic device design, the counterdiabatic protocol leads to a high-fidelity device, with its evolution following the adiabatic modes exactly even when the adiabatic condition is violated. Tolerance analysis shows that the countera-diabatic devices combine the advantages of adiabatic and resonant devices. The formalism is used to design asymmetric waveguide couplers.
We introduce an inertial counterdiabatic (CD) protocol that reformulates shortcut-to-adiabaticity design as a moving-frame diagonalization problem. The method employs the Magnus expansion to construct a variational propagator generated by an accessible CD ansatz. In the new rotating frame, the controls are optimized to suppress the off-diagonal Hamiltonian terms, and the CD terms are defined to counteract the corresponding inertial contribution. The propagator generated with the Magnus expansion naturally exploits additional control directions generated through its nested commutator, while the sequential design of the propagator introduces temporal nonlocality into the optimization. We benchmark the method on a two-qubit system and on the one-dimensional quantum Ising chain, where it consistently outperforms variational CD protocols based on action minimization. The largest improvements are observed near gap-closing regions, demonstrating the advantage of combining temporal and algebraic nonlocality for suppressing diabatic excitations.
We investigate the quantum mechanics of a single particle constrained to move along an arbitrary smooth reference curve by a confinement that is allowed to vary along the waveguide. The Schr\"odinger equation is evaluated in the adapted coordinate frame and a transverse-mode decomposition is performed, taking into account both curvature and torsion effects and the possibility of a cross-section potential that changes along the curve in an arbitrary way. We discuss the adiabatic structure of the problem, and examine nonadiabatic couplings that arise due to the curved geometry, the varying transverse profile, and their interplay. The exact multimode matrix Hamiltonian is taken as the natural starting point for few-mode approximations. Such approximate equations are provided, and it is worked out how these recover known results for twisting waveguides and can be applied to other types of waveguide designs. The quantum waveguide Hamiltonian is recast into a form that clearly illustrates how it generalizes the Born-Oppenheimer Hamiltonian encountered in molecular physics. In analogy to the latter, we explore the local gauge structure inherent to the quantum waveguide problem and suggest the usefulness of diabatic states, giving an explicit construction of the adiabatic-to-diabatic basis transformation.
Summary form only given. The manipulation of the phase of classical and quantum light is a major key to quantum photonics [1]. Integrated optics has several advantages over bulk optical systems, most notably in robustness, handling and size. Writing three-dimensional waveguides circuits into transparent materials using an ultrafast laser is thereby a new emerging technology [2; 3] with possible applications in quantum communication [4], and quantum computing [5]. In our work, we experimentally demonstrate the high-precision manipulation of the optical phase of classical and quantum light, using a generalized directional coupler.The behavior of light in directional coupler consisting of two waveguides can be described by the Hamiltonian H=T+V1+V2, where T is the kinetic energy and Vi is the potential energy of waveguide i = 1 and i = 2. The wavefunction of the system is constructed as a superposition of the single waveguide modes Ψ=a|1+b|2, and the Schrödinger equation of the coupler reads as HΨ=ksmΨ, where ksm denotes the propagation constant of the supermode. The propagation constants of the supermodes can be calculated straightforwardly as ksm1,2=k0±κ(1∓ σ)/(1-σ2) , where k0 is the propagation constant of the individual waveguides, κ=2|V2|1=1|V1|2 represents the coupling constant, and σ=1|=2|1is the direct overlap of the individual waveguide modes. For σ → 0 the well known distribution of ksm1,2 is preserved. As ksm = (ksm1+ksm2)/2>k0, in a generalized directional coupler a phase delay Δφ is introduced that readswith lC = π/κ as the coupling length. The phase delay depends only on σ, which itself is a function of the spacing between the waveguides. As the spacing can be controlled in a very accurate manner, this approach allows the highly precise manipulation of the optical phase along a directional coupler. In order to measure this effect, we set up a tuned Mach-Zehnder (MZ) interferometer with identical lengths of the interferometer arms. A third waveguide was applied to one of the arms, forming an additional directional coupler in which the light completely couples forth and back with no intensity losses in the relevant interferometer arm (Fig. 1(a)). However, as the change in the propagation constant in this arm results in a phase delay, at the output of the MZ the ratio of intensities (for classical light) and the count rates (for single photons) changes. To obtain an all-reflecting coupler (ARC) for different coupling constants, the respective length of a coupler has to be adjusted and the associated phase delays are shown in Fig. 1(b). Due to decreasing σ for larger distances between the waveguides, the accumulated phase difference shrinks, even if the length of the coupler increases. Normalizing the phase difference by length allows the direct calculation of the change of the propagation constant caused by the ARC (Fig. 1(c)).In conclusion, we demonstrated a highly precise method for the manipulation of the optical phase in integrated waveguide structures by employing generalized directional coupling. The propagation constants of the supermodes in the structures are controlled by the spacing between the waveguides in the ARC. Our approach holds for classical as well as quantum light, and has the potential to find its use in a multitude of integrated-optical devices.
Third-order nonlinear processes require phase matching between the interacting fields to achieve high efficiencies. Typically in guided-wave χ(3)χ^{(3)}χ(3) platforms this is achieved by engineering the dispersion of the modes through the transverse profile of the device. However, this limits the flexibility of the phase matching that can be achieved. Instead, we analyze four-wave mixing in a pair of asymmetric waveguides and show that phasematching may be achieved in any χ(3)χ^{(3)}χ(3) waveguide by coupling of a nondegenerate pump from an adjacent waveguide. We demonstrate the additional flexibility that this approach yields in the case of photon-pair generation by spontaneous FWM, where the supermode dispersion may be modified to produce pure heralded single photons -- a critical capability required for example by silicon platforms for chip-scale quantum photonics.