Factorization dynamics between quantum Fisher information and quantum coherence

AI summary

Overview

Research area: Quantum metrology and quantum information theory — specifically the relationship between quantum Fisher information (QFI) and quantum coherence in pure states under unitary evolution.

Technical level: Intermediate. The core results are compact algebraic relations, but the paper assumes familiarity with density matrices, Hamiltonians, and the quantum Cramér–Rao bound. The experimental sections require some knowledge of linear-optics state preparation.

Scope: A theoretical and experimental study (arXiv:2609.36571v1, 29 September 2026, by Zhao, Yu, Li, Zhou and Zhang of Ningbo University and the National Key Laboratory of Scattering and Radiation, Beijing) deriving and verifying a "factorization law" linking the QFI of an evolved pure state to the coherence of its input state, plus a coherence-based lower bound on parameter-estimation variance.

What This Paper Is About

Quantum Fisher information sets the ultimate precision limit of parameter estimation, and quantum coherence is the resource that lets a state carry phase information at all — but the dynamical relationship between the two was not established. The authors ask how the coherence of an input pure state determines the QFI of the unitarily evolved output state, and whether the variance of an estimate can be bounded using only the input state's coherence when the final state is unavailable. They answer both questions with a factorization relation, then test it on photonic qubit and qutrit systems.

Key Contributions

1. A factorization law for qubits (Eq. 4): For a pure qubit state, the evolved-state QFI equals exactly 2 × (l₂-norm coherence of the input) × (maximal QFI), i.e. ℱ_θ(|ψ_θ⟩) = 2C_{l₂}(|ψ⟩)ℱ_{θ,max}.

2. A generalized inequality for qudits (Eq. 5): For pure states of dimension D > 2, the QFI is bounded above by 2C_{l₂}(|ψ⟩)ℱ_{θ,max}, valid when Σ_{i<j}|a_i|²|a_j|² < 0.25; the inequality becomes trivial when that sum is ≥ 0.25.

3. A final-state-independent variance bound (Eq. 6): Combining the above with the quantum Cramér–Rao bound gives Var(θ̂) ≥ 1/(nℱ_θ) ≥ 1/(2nC_{l₂}(|ψ⟩)ℱ_{θ,max}), so estimation precision can be lower-bounded from initial-state coherence alone.

4. Experimental verification in two and three dimensions: Photonic tests of Eq. 4 (qubit) and Eq. 5 (qutrit, including the equality condition), plus a demonstration in the 2D case that measured variance tracks 1/ℱ_θ₁(|ψ_θ₁⟩) under an optimal measurement.

Main Findings

• Zero coherence implies zero QFI: Directly from Eqs. 4 and 5, if the input state has C_{l₂}(|ψ⟩) = 0, then the output state necessarily has ℱ_θ(|ψ_θ⟩) = 0. An input must be coherent for its output to be usable for estimating an unknown parameter θ.
• Qubit case agrees with the equality: With Ĥ₁ = |0⟩⟨1| + |1⟩⟨0| (eigenvalues 1 and −1, so ℱ_{θ₁,max} = 4) and θ₁ = 24°, the measured left- and right-hand sides of Eq. 4 showed only a slight difference across α₁₉ ∈ [0°, 21°], with theory curves and experimental points closely matching. The stated source of error was suboptimal wave-plate calibration, giving a slight deviation of the prepared initial state and unitary from the intended ones.
• Qutrit case satisfies the inequality: With Ĥ₂ = (|0⟩⟨1| + |1⟩⟨0| + |1⟩⟨2| + |2⟩⟨1|)/√2, θ₂ = 29° and ℱ_{θ₂,max} = 4, both tested settings (α₁ = 0° and α₁ = 10°) satisfied Eq. 5, with experimental values reasonably compatible with the theoretical curve.
• The equality condition is confirmed: Eq. 5 becomes an equality when the initial state is a superposition of the eigenstates |0_eig⟩ and |2_eig⟩ of Ĥ₂, written |ψ′⟩ = sin2α|0_eig⟩ + cos2α|2_eig⟩; this was verified experimentally (Fig. 3D).
• Variance bound verified in 2D: Using a projection onto |0⟩⟨0| in the Pauli σ_z basis and maximum-likelihood estimation, measured Var(θ̂₁)×n fitted the theoretical curve well. When α₁₉ = 0° (initial state |ψ₀⟩ = |0⟩) this measurement choice achieves the optimal measurement.
• Sources of experimental imperfection identified: Imperfect preparation of pure states, imperfect horizontal/vertical polarization separation by the beam displacers, and deviation of the experimentally realized unitary operators from the theoretical ones. Error bars for 3σ deviations from a Poisson distribution were smaller than the plotted point size and were omitted.
• Not reported: The paper does not give a single aggregate fidelity or numerical error metric for the fits; agreement is described qualitatively ("good agreement," "relatively compatible"). Details of the mixed-state treatment, the explicit unitary matrices, and maximum-likelihood estimation procedure are deferred to the Supplementary Materials, whose text in the provided content is truncated.

Methodology in Plain English

The authors first do the algebra: write the QFI of a pure state under a θ-independent Hamiltonian as four times the variance of the Hamiltonian, express the same state's coherence as the sum of squared off-diagonal density-matrix elements (the l₂-norm), and show the two quantities differ by a factor of 2 times the maximal QFI. That gives an exact equality in two dimensions and an upper bound in higher dimensions, which they then convert into a variance bound via the quantum Cramér–Rao bound.

To test it, they generate heralded single photons from a 405-nm continuous-wave diode laser pumping a periodically poled potassium titanyl phosphate (PPKTP) crystal via collinear type II spontaneous parametric down-conversion. Detecting the idler photon heralds the signal photon. Half-wave plates, beam displacers, polarizing beam splitters and quarter-wave plates encode the chosen input state and implement the unitary evolution; a beam splitter taps off part of the light so the input state's coherence can be measured directly, while the transmitted part evolves and is measured (via σ_x) to reconstruct the final-state density matrix and hence the QFI. The qutrit unitary is decomposed as U_θ₂ = U₁₂U₀₂U₀₁, each U_ij acting on a 2D subspace of the qutrit with the complementary subspace unchanged, and implemented with wave plates H3–H12 and Q1–Q4. Fixed parameters (θ₁ = 24°, θ₂ = 29°) were used while scanning the input-state angle. To reduce drift, coincidence events were recorded in 1-s windows and the initial-state angle was adjusted every 180 data points.

Why This Matters

Impact on research: The work converts a qualitative intuition — "coherence is needed for metrology" — into a quantitative dynamical law, and it shifts the practical question of precision from the final state back to the input state. It also extends the statement beyond qubits to general qudits, unlike earlier qubit-only treatments, and connects to existing work on the role of coherence in metrology tasks such as the phase discrimination game and parameter estimation.

Real-world applications (as framed by the authors):

• Quantum sensing — predicting achievable sensitivity from how the probe is prepared.
• Quantum metrology and phase estimation, including scenarios with state-preparation error, loss and decoherence.
• Quantum key distribution and quantum secure direct communication, where coherence is a required resource.
• Quantum computing, where coherence underpins the advantage over classical information processing.

Industry relevance: For anyone building quantum sensors or optical metrology hardware, the bound in Eq. 6 allows a precision guarantee to be certified from the source state alone — useful when the output state after a lossy or inaccessible channel cannot be fully characterized.

Future Directions

• Extending the experimental test to dimensions D > 3, which the authors explicitly note will be more challenging and is left as future work.
• The paper states that a detailed discussion of the factorization law for mixed states is provided in the Supplementary Materials, leaving mixed-state dynamics as an open avenue for experimental confirmation.
• Comparing this framework against non-Hermitian quantum metrology approaches, which the authors note have achieved or even surpassed the Heisenberg limit under certain conditions.
• Optimizing the input state's coherence and the channel properties jointly to minimize estimation variance, which the authors propose as a route to higher precision.
• Several related works (references 56–58 in the paper) are flagged as warranting further exploration by the community.

Target Audience

Researchers and graduate students in quantum metrology, quantum sensing and quantum information theory; experimentalists working with photonic qubit/qutrit platforms; and engineers developing precision-measurement hardware who need a resource-theoretic handle on how probe-state preparation caps achievable sensitivity. Readers without a background in density-matrix formalism or the quantum Cramér–Rao bound will need to consult the cited literature (references 7–14) first.

Authors’ abstract

Quantum Fisher information (QFI) quantifies the sensitivity of a quantum state to a parameter change and plays a key role in quantum metrology. Meanwhile, quantum coherence is a crucial resource for quantum information processing. However, despite extensive studies on various aspects of quantum metrology, the interplay between the dynamics of the QFI and quantum coherence remains unexplored. Here, we explore the factorization relationship between the QFI and quantum coherence, both theoretically and experimentally. We prepare pure states in qubit and qutrit systems and apply unitary evolution to investigate this relationship. Our results confirm the factorization law, offering strong evidence for this relationship. Furthermore, we establish a lower bound on the variance of the estimated parameter that does not require the final state information. This approach reveals a deep connection between the QFI and quantum coherence, providing insights into quantum parameter estimation. These findings have potential applications in quantum sensing and quantum metrology.