One-Dimensional Nonlinear Quantum Walks

Abstract: We explore a continuous-time quantum walk starting at a single vertex on the discrete path and cycle with a cubic nonlinearity. Such nonlinearities arise in Bose-Einstein condensates described by the Gross-Pitaevskii equation or by nonlinear optical waveguide arrays. We analytically prove that the nonlinear quantum walk can be trapped to arbitrary fidelity depending on the coefficient of the nonlinear term. This contrasts with linear quantum walks, which are known for spreading quickly in one dimension. We propose that this trapping can be used for timing in quantum state transfer, where a qubit is held at a node until it is ready to be transferred, and it can also be held again at the receiving node. This scheme can also be interpreted as a form of quantum memory, with the trap and transfer corresponding to the storage and release of quantum information.

• The paper establishes analytic lower bounds for self-trapping, showing how localized quantum walkers depend on the nonlinearity coefficient and vertex degree.
• It demonstrates that tuning the cubic nonlinearity enables controlled localization and timed quantum state transfer in lattice systems.
• The study provides a framework for designing robust quantum memories and routers by dynamically modulating nonlinear interactions.

One-Dimensional Nonlinear Quantum Walks: Analytical Characterization and Applications

Introduction

The study of quantum walks (QWs), both in discrete and continuous time, has elucidated quantum-enhanced spreading and quantum search phenomena in one-dimensional (1D) lattice systems. While the traditional quantum walk formulation is linear, real physical systems like Bose-Einstein condensates (BECs) and nonlinear optical waveguide arrays are governed by the cubic nonlinear Schrödinger equation, also known as the Gross-Pitaevskii equation. These systems introduce cubic nonlinearities characterized by a real coefficient $g$, resulting in nonlinear quantum walks (NLQWs) whose dynamics diverge qualitatively from their linear counterparts. This work systematically analyzes the fundamental dynamical regime of self-trapping—a phenomenon where a walker remains localized indefinitely—on finite path and cycle graphs, providing both rigorous analytical lower bounds and proposals for protocols in quantum information applications (2605.20464).

Formalism and Self-Trapping Phenomenon

The dynamics of the one-dimensional nonlinear quantum walk are governed by the time-dependent equation:

$$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$

where $H$ is the standard adjacency Hamiltonian of the discrete path or cycle and $g$ quantifies the cubic nonlinearity. Initial conditions focus on a walker localized at site $v_r$.

While the linear quantum walk ($g=0$) is characterized by ballistic expansion, the inclusion of nonzero $g$ leads to the emergence of self-trapping. For sufficiently large $|g|$, the walker remains localized at its initial vertex with high probability; this nonlinear, interaction-induced localization is in stark contrast to any scenario permitted by purely linear unitary evolution.

Analytical Characterization of Self-Trapping

The paper presents rigorous analytic bounds for the onset and magnitude of self-trapping. The central technical result is expressing a lower bound for the time-averaged probability $p(t)$ at the initial site in terms of the nonlinear coefficient $g$ and local degree $$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$0:

$$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$1

This result yields the following implications:

• Self-trapping onset threshold: For a walker initially at a degree-1 (endpoint) site of the path, self-trapping is rigorously guaranteed for $$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$2, and for degree-2 vertices (interior or cycle) for $$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$3. These analytical thresholds are close to the numerically observed critical nonlinearity $$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$4 but are not tight, leaving a gap in the analytic characterization—a direction for future work.
• Guaranteed localization: For any $$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$5, one determines a nonzero lower bound $$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$6 such that $$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$7 for all $$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$8, where $$i \frac{d}{dt}|\psi(t)\rangle = (H - g|\psi(t)|^2) |\psi(t)\rangle,$$9 is a solution to an explicit algebraic equation dependent on $H$0 and $H$1. For example, choosing $H$2 (endpoint), the minimum probability at the initial site is at least 0.987 for all time.
• Inverse design: Given a required trapping probability (e.g., $H$3), one can determine the minimum $H$4 needed to guarantee this.

The analysis establishes that these localization bounds are independent of lattice size $H$5, depending solely on the local degree. These results link the controllable physical nonlinearity parameter $H$6 to observable dynamical behavior in quantum lattice systems.

Implications for Quantum Information Protocols

A central application proposed is in the controlled timing of quantum state transfer on spin chains or photonic lattices. Standard state transfer involves ballistic QW evolution from source to target. By tuning $H$7 to large values, a walker can be dynamically "trapped" at the source node, effectively "pausing" information transfer. Setting $H$8 releases the localization, permitting state transfer with known fidelity, and setting $H$9 at the target can trap the qubit again. This approach enables active quantum memory and synchronized release protocols using only nonlinear parameter modulation.

This strategy is directly implementable in physical systems where $g$0 is tunable, e.g., BECs via Feshbach resonances or nonlinear photonic lattices. The same physical degree of freedom embodies both the flying and storage qubit, simplifying hardware requirements.

Theoretical and Practical Outlook

This study provides a rare instance of strong analytic bounds for nonlinear quantum walk dynamics on generic finite graphs, a setting where most previous results were purely numerical or limited to special cases (dimers or trimers). The ability to deterministically control localization via nonlinear interactions raises fundamental questions on the design of robust quantum memories, routers, and circuit elements based on dynamical self-trapping. The non-tightness of the analytic bounds identifies a technically precise open question: to close the gap between rigorous lower and true numerical thresholds for the self-trapping transition.

From a broader nonlinear dynamics viewpoint, connecting self-trapping phenomena in quantum walks to corresponding results in condensed matter (e.g., discrete nonlinear Schrödinger solitons) and experimental results in nonlinear optics reinforces the universality and cross-platform relevance of these results.

Conclusion

The paper provides a comprehensive analytic treatment of continuous-time one-dimensional nonlinear quantum walks governed by the Gross-Pitaevskii equation. It rigorously establishes lower bounds for self-trapping at the initial site as a function of the cubic nonlinearity and applies these bounds to propose quantum information timing and memory protocols. While the derived thresholds are not tight compared to numerical observations, this work advances the analytic understanding of nonlinear quantum walk localization and sets a foundation for application-oriented quantum control schemes and refined theory development (2605.20464).