{"id":"ad90644e-7479-4990-8878-ee33e6dcfebb","arxiv_id":"1909.00298","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper reports measuring a π geometric phase for a two-level Longuet-Higgins Hamiltonian on IBM quantum hardware, but the circuit is a direct rotation rather than an adiabatic simulation.","lead":"This paper claims to observe the molecular Aharonov-Bohm geometric phase using a two-qubit IBM quantum computer. The experiment reports a π phase for a full 2π traversal, but the circuit implements a direct rotation rather than an adiabatic evolution, so the central claim is overstated.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central inference is circular: M(t) is defined as the eigenvector rotation matrix, so the measured phase is the programmed gate angle, not an independently observed Berry phase.","rationale":"The reader's weakest assumption correctly identifies the load-bearing flaw: M(t) in Eq. (16)-(18) is constructed as the eigenvector matrix of H(φ), and the paper then applies this matrix as a gate. There is no derivation showing that this matrix equals the actual evolution operator for an adiabatic traversal of nuclear coordinates, and Eq. (13) is only a formal expression with h(t') undefined. The measured phase is therefore the phase of the applied rotation, not an independent observation of a geometric phase. This is not a disagreement with consensus or a stylistic issue; it is an internal gap between the molecular Hamiltonian and the implemented circuit. The proposed test, a Trotterized adiabatic sweep with dynamical phase subtraction, would directly show whether the single-gate protocol isolates the geometric phase or merely reproduces the programmed angle. Because the central claim as stated is not supported by the evidence in the paper, the reject verdict is appropriate.","tokens_in":7018,"tokens_out":8758,"duration_ms":91410,"concrete_test":"On the same IBM backend, replace the single CU3(θ) gate in Fig. 2 with an N-step Trotterized implementation of U(T)=T exp(-i∫_0^T H(φ(t))dt) for H(φ)=K(cosφ σ_z+sinφ σ_x) with φ swept from 0 to 2π, and record the raw overlap phase ⟨ψ(0)|U(T)|ψ(0)⟩ as a function of N and T. If the raw phase differs from the programmed rotation angle and only equals π after explicitly subtracting the dynamical phase ∫E(t)dt, then the single-gate result was the gate angle, not an observed geometric phase.","verdict_should_be":"REJECT","load_bearing_attack":"The central claim requires that the measured overlap phase be the geometric phase accumulated during an adiabatic cyclic evolution under the molecular Hamiltonian. The paper never establishes this. Equation (13) writes M(t)=exp(∫h dt') but h(t') is never defined. Equations (16)-(18) then define M(t) as Σ|output⟩⟨input|, i.e. the unitary matrix that maps the φ=0 eigenstates onto the eigenstates of H(φ) from Eq. (5); this is precisely the rotation matrix whose angle is the programmed parameter. The circuit in Fig. 2 applies this matrix as a controlled gate, so ⟨ψ(0)|M(t_f)|ψ(0)⟩ is the overlap programmed into the gate, not a quantity emergent from Hamiltonian dynamics. The adiabatic theorem, the Berry connection A(R)=i⟨χ|∇_R χ⟩ from Eqs. (7)-(9), and the dynamical phase are never used to connect H(φ(t)) to M(t). Consequently, the experiment cannot distinguish a Berry phase from a preprogrammed rotation, and the claim to have observed the molecular Aharonov-Bohm geometric phase is unsupported. The paper also contains an algebraic error in Eq. (3), where the off-diagonal term is not squared in the secular energy, but the decisive gap is the unjustified identification of the eigenvector matrix with the evolution operator.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports an experiment on IBM quantum hardware and simulators aimed at observing the geometric (Berry) phase of the Longuet-Higgins molecular Aharonov-Bohm model. The authors define a two-qubit circuit in which a probe qubit controls the application of an operator M(t) to a system qubit, and the geometric phase is extracted from the expectation values of σx and σy on the probe. They find that the phase accumulates linearly with the rotation angle and reaches π for a 2π rotation, which they interpret as the molecular Aharonov-Bohm phase. The abstract and discussion claim an exact estimation of the geometric phase for this molecular system.","tokens_in":7340,"tokens_out":3837,"duration_ms":34389,"significance":"If the claim were correct, it would provide a simple quantum-computer demonstration of the molecular Aharonov-Bohm phase and a scalable protocol for evaluating geometric phases in more complex molecular Hamiltonians. The paper has some positive features: it presents a concrete ancilla-based interferometric scheme for measuring an overlap phase, it implements the circuit on real IBM hardware, and it connects the problem to the Longuet-Higgins conical-intersection literature. However, the central physical claim is not supported by the presented evidence. The operator M(t) is defined by the eigenvector matrix of the Hamiltonian, and the circuit applies that matrix as a gate, so the measured phase is essentially the phase programmed into the gate. The connection between M(t) and the adiabatic evolution under the molecular Hamiltonian is never derived, and the Berry connection and adiabatic theorem play no role in the analysis. The paper therefore does not demonstrate an observation of a geometric phase.","major_comments":[{"comment":"The operator M(t) applied in the circuit is defined in Eqs. (16)-(18) as the matrix mapping the φ=0 eigenstates of the Hamiltonian in Eq. (5) to the eigenstates at arbitrary φ, namely a rotation by φ/2. No derivation is given that this matrix is the unitary evolution generated by the molecular Hamiltonian H(φ(t)) during an adiabatic traversal; Eq. (13) invokes an exponent of an integral of h(t') but h(t') is never identified. Consequently, the measured quantity arg⟨ψ(0)|M(tf)|ψ(0)⟩ is the argument of the overlap with the programmed rotation matrix, not a geometric phase accumulated through the adiabatic theorem, and the central claim in the Discussion is unsupported.","section":"II, Eqs. (13)-(18)"},{"comment":"The secular equation in Eq. (3) reads ϵ = ±[(αx+βy)^2 + by]^{1/2}; this is dimensionally inconsistent because the first term in the parentheses has dimensions of energy squared while the second has dimensions of energy. The standard double-cone expression requires (by)^2, so the algebra preceding the conical-intersection model contains an error. While this error does not directly enter the quantum circuit, it undermines the presentation of the model.","section":"Eq. (3)"},{"comment":"The results report a linear accumulation of phase with the programmed rotation angle, but no error bars, shot counts, or statistical analysis are given, and there is no comparison between the simulator results and the two hardware backends shown in Figs. 5-7. Because the angle θ of the U3 gate is set to the same value as φ/2 used to compute the expected phase, the agreement between the measured phase and the theoretical curve is a check of the gate calibration rather than an independent observation of a geometric phase.","section":"III, Figs. 4-7"},{"comment":"The protocol places the system in an eigenstate at φ=0 and then directly applies M(t) as a controlled gate, rather than evolving the system under H(φ(t)) through a closed path with a time-dependent parameter. The Berry connection A(R) in Eqs. (7)-(9) and the parallel-transport condition are never used to relate H(φ(t)) to M(t). Thus the experiment cannot discriminate a geometric phase from a preprogrammed unitary rotation, and the connection to the molecular Aharonov-Bohm effect is not established.","section":"II, protocol and circuit"}],"minor_comments":[{"comment":"The name 'Lounguet-Higgins' appears in the abstract and Introduction; it should be 'Longuet-Higgins'.","section":"Abstract and Introduction"},{"comment":"Reference 12 is incomplete and garbled, listing multiple author sequences without a coherent citation.","section":"References"},{"comment":"The notation |ψθ(0)⟩ and |ψθ(tf)⟩ is used in Eq. (10) before the state |ψθ(t)⟩ is defined; the subscripts and arguments should be specified.","section":"Eq. (10)"},{"comment":"The horizontal axis is labeled 'angle(theta)' and the vertical axis 'phase(theta)', while the text calls the quantity 'berry phase'; these labels should be made consistent.","section":"Fig. 4"},{"comment":"The paper does not clarify which results were obtained on the quantum simulator versus the real devices IBMqx4 and IBMqx2; the Discussion refers to the 'nominal circuit on the quantum simulator' without specifying the hardware provenance of each figure.","section":"Results and Discussion"}],"recommendation":"reject","confidential_remarks":"The central problem is structural: the measured phase is defined to equal the programmed gate angle, so no amount of additional data or error analysis would convert this into an observation of the molecular Berry phase. The manuscript could potentially serve as a pedagogical demonstration of an ancilla-based interferometric phase measurement, but as a research claim about the molecular Aharonov-Bohm effect it does not meet the standard for publication. I also note that the reference list contains several incomplete and malformed entries, which would need correction in any future resubmission."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things about this paper. First, the circuit is simple and the data show a clean linear phase accumulation ending at pi for a 2pi rotation. Second, the central claim that this observes the molecular Aharonov-Bohm geometric phase does not survive reading the methods. The measured quantity is essentially the angle programmed into the rotation gate, so the experiment cannot distinguish a Berry phase from a preprogrammed unitary.\n\nWhat is genuinely here: the authors take the Longuet-Higgins two-level Hamiltonian, map its eigenstates onto qubit rotations, and run a textbook interferometric phase measurement on IBM hardware. The one-step circuit in Fig. 2 is transparent, and reproducing the expected pi phase with a simulator and real hardware is a reasonable pedagogical exercise. If reframed as a demonstration of how a Berry phase would appear in a qubit rotation, the paper has some value for students.\n\nThe load-bearing problem is in Eqs. (13)-(18). The operator M(t) is introduced as an exponential of an integral involving h(t'), but h is never defined. Then M is explicitly constructed as the sum of |output><input| mappings between eigenstates, which gives the rotation matrix whose angle is exactly the parameter phi/2. The gate in the circuit is set to that same angle. So the overlap <psi(0)|M(tf)|psi(0)> is the phase written into the gate by construction. The adiabatic theorem, the Berry connection, and the dynamical phase are never used to connect the Hamiltonian H(phi) to M(t). This is circular as an observation of a geometric phase, and it is the central unsupported step. This is not a minor gap; it is the claim.\n\nThere are smaller issues. Equation (3) appears dimensionally off: the off-diagonal term should be squared under the square root, likely a typo. There are no error bars, no statistical analysis, and no code or raw data. The reference list has some garbled entries, though not enough to matter.\n\nWho gets value from this: instructors looking for a simple IBM Q exercise, and students learning why one must be careful about what a circuit actually simulates. It is not a research result on molecular Aharonov-Bohm systems. I would not cite it as an observation, but I might point students to it as an example of how easy it is to program a phase and then measure it back.\n\nIf I were the editor, I would send it to a referee rather than desk-reject, because the flaw is instructive and the paper makes a concrete, checkable claim. I would expect the referee to reject the central claim but to allow that the protocol could be salvaged if reframed as a simulation exercise.","headline":"A clean but circular educational demonstration: the measured phase is the gate angle programmed into the circuit, not an independently observed molecular Berry phase.","tokens_in":7822,"tokens_out":1668,"would_cite":false,"duration_ms":19887,"reading_group":"maybe","serious_thinker":"unclear","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper reports exact estimation of the molecular Aharonov-Bohm geometric phase on a quantum computer: a full 2π parameter sweep produces the expected π phase.","keywords":["geometric phase","Berry phase","molecular Aharonov-Bohm effect","conical intersection","Born-Oppenheimer approximation","two-qubit circuit","Jahn-Teller effect","quantum simulation"],"falsifier":"Replace the hand-built matrix M(t) with a time-ordered evolution generated by the adiabatic Hamiltonian H(t) = K(I + cos φ(t) σz + sin φ(t) σx) integrated over the same closed loop, and compare the measured overlap phase with the rotation-gate result; if the two disagree, the reported phase is the gate's rotation angle rather than a dynamically accumulated geometric phase.","tokens_in":6880,"feed_emoji":"⚛️","tokens_out":10502,"duration_ms":89138,"temperature":0.7,"pith_summary":"The paper reports an exact estimation of the molecular Aharonov-Bohm geometric phase on a quantum computer. In the molecular version of the Aharonov-Bohm effect, a closed loop of nuclear coordinates around a conical intersection changes the sign of the electronic wavefunction, a global π phase that acts like a vector potential on the nuclear motion. The authors implement a two-qubit interferometric protocol: a system qubit carries the molecular eigenstate, a probe qubit controls the application of the rotation matrix built from the instantaneous eigenstates, and the accumulated overlap phase is recovered from the probe's σx and σy expectation values. The measured phase grows linearly from 0 to π as the angle parameter runs from 0 to 2π, which the paper takes as confirmation that the protocol estimates the molecular geometric phase exactly.","feed_headline":"Two-qubit circuit reads the molecular Berry phase of π","feed_subtitle":"A probe qubit maps the conical-intersection sign change to a linear phase from 0 to 180 degrees.","key_machinery":"The load-bearing object is the 2×2 operation matrix M(t), built from the instantaneous eigenstates of the molecular Hamiltonian and applied as a controlled gate on the system qubit; the paper writes this matrix as $e^{{∫ h dt'}}$ without identifying the generator h. Starting from the φ=0 eigenstates |0⟩ and |1⟩, the rotation by φ/2 maps each basis state to the corresponding eigenstate at angle φ, so the inner product between the initial state and the rotated state carries the phase accumulated over the parameter interval. The probe qubit turns that inner product into an observable: tracing out the system leaves the probe in a state whose σx and σy expectation values are the real and imaginary parts of the overlap, and the geometric phase is their argument. The circuit therefore works as a direct phase meter requiring only single-qubit measurements on the probe, not full tomography of the system.","core_discovery":"The central claim is that the π Berry phase of the molecular Aharonov-Bohm effect can be measured directly and exactly with a state-independent two-qubit circuit. For the two-level conical-intersection Hamiltonian whose eigenstates are [cos(φ/2), sin(φ/2)]^T and [-sin(φ/2), cos(φ/2)]^T, the operation matrix M(t) formed from these eigenstates is the rotation matrix [[cos(φ/2), -sin(φ/2)], [sin(φ/2), cos(φ/2)]]. The circuit prepares the φ=0 eigenstate, applies M(t) conditionally on the probe qubit, and reads the geometric phase as arg(⟨σx⟩ + i⟨σy⟩) from the probe's reduced state. At φ=2π the rotation becomes -I, the overlap is -1, and the phase is π; intermediate sweeps yield the reported linear phase accumulation, plotted against the rotation angle. The authors present this agreement as exact estimation of the geometric phase and propose the protocol as a first-principles route for more complex molecular Hamiltonians.","pith_inferences":["A direct dynamical test would replace the hand-built matrix M(t) with a time-ordered evolution generated by the time-dependent molecular Hamiltonian and check that the readout still equals the rotation-matrix phase; that would separate a programmed rotation phase from a genuinely accumulated geometric phase.","The probe-qubit overlap readout is a pure-state version of the mixed-state Uhlmann phase protocol, so the same circuit could be adapted to measure molecular geometric phases at finite temperature or under decoherence.","Because the measured phase is the argument of an overlap, the protocol is effectively a universal geometric-phase meter for two-level systems; any parameter loop, not only a conical intersection, could be characterized with the same basic circuit.","The linear phase plot implies the circuit could be used as a calibration device for quantum gates: deviations from linearity in the phase-versus-angle curve would expose rotation-angle errors in the applied single-qubit gate."],"forward_implications":["A complete 2π cycle of the molecular parameter returns the eigenstates with a sign change, so the circuit confirms the Born-Oppenheimer wavefunction's multivaluedness that the molecular Aharonov-Bohm vector potential is designed to remove.","Because the phase is read from Pauli expectation values on a single probe qubit, the protocol does not require state tomography and can be run on two-qubit processors as they exist today.","For Hamiltonians with m-fold degeneracy, the same construction generalizes to an m×m secular problem, and the paper argues that quantum eigensolvers can supply the eigenstates when analytic ones are unavailable.","The reported linear phase-versus-angle dependence provides a calibration curve: any two-level adiabatic loop whose eigenstates wind by half the loop angle should show the same linear readout."],"supporting_citations":[{"why":"Establishes the electromagnetic Aharonov-Bohm effect that the molecular phase is meant to parallel.","marker":"1"},{"why":"Supplies the Berry-phase formalism and the adiabatic-phase context the protocol claims to measure.","marker":"2"},{"why":"Provides the molecular geometric-phase treatment and vector-potential gauge structure used in the derivation.","marker":"4"},{"why":"Gives the two-state conical-intersection Hamiltonian whose eigenstates are the input-output states of the circuit.","marker":"6"},{"why":"Defines geometric phase through parallel transport, the criterion invoked for the overlap phase.","marker":"8"},{"why":"Contributes the probe-qubit scheme for extracting the overlap phase from Pauli expectation values.","marker":"12"},{"why":"Demonstrates Berry-phase measurement in a superconducting qubit, the experimental template the circuit adapts.","marker":"13"},{"why":"Supplies the quantum eigensolver subroutine that would make the protocol applicable when eigenstates are not known analytically.","marker":"19"}],"fun_headline_variants":["IBM quantum chip pins down molecular Berry phase","Exact π phase measured on two-qubit IBM circuit","Quantum circuit nails molecular Aharonov-Bohm phase","Probe qubit maps conical-intersection sign flip","IBM quantum computer reads π geometric phase"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The protocol rests on identifying the eigenvector matrix M(t) of the instantaneous molecular Hamiltonian with the actual adiabatic evolution operator, so the phase read from the circuit is assumed to be the geometric phase accumulated by a nuclear traversal rather than merely the phase of the rotation that was programmed into the gate.","fun_headline_variants_meta":{"raw":{"variants":["IBM quantum chip pins down molecular Berry phase","Exact π phase measured on two-qubit IBM circuit","Quantum circuit nails molecular Aharonov-Bohm phase","Probe qubit maps conical-intersection sign flip","IBM quantum computer reads π geometric phase"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000396,"raw_usage":{"total_tokens":2084,"prompt_tokens":962,"completion_tokens":1122,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":578,"completion_tokens_details":{"reasoning_tokens":1049}},"tokens_in":578,"tokens_out":1122,"duration_ms":28061,"temperature":1.0,"reasoning_tokens":1049,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:56:23.041583+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Replace the hand-built matrix M(t) with a time-ordered evolution generated by the adiabatic Hamiltonian H(t) = K(I + cos φ(t) σz + sin φ(t) σx) integrated over the same closed loop, and compare the measured overlap phase with the rotation-gate result; if the two disagree, the reported phase is the gate's rotation angle rather than a dynamically accumulated geometric phase.","supporting_citations":[{"cited_title":"Aharonov, D","cited_arxiv_id":null,"evidence_quote":"Establishes the electromagnetic Aharonov-Bohm effect that the molecular phase is meant to parallel."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Berry-phase formalism and the adiabatic-phase context the protocol claims to measure."},{"cited_title":"Allen Mead title The geometric phase in Molecular Systems journal Review Modern Physics volume 64 , pages 51 ( year 1992 )","cited_arxiv_id":null,"evidence_quote":"Provides the molecular geometric-phase treatment and vector-potential gauge structure used in the derivation."},{"cited_title":"Hertzburg and H.C","cited_arxiv_id":null,"evidence_quote":"Gives the two-state conical-intersection Hamiltonian whose eigenstates are the input-output states of the circuit."},{"cited_title":"Anandan and Y.Aharonov title Geometric Quantum Phase and Angles journal Physics Review D , volume 38 , pages 1863 ( year 1988 )","cited_arxiv_id":null,"evidence_quote":"Defines geometric phase through parallel transport, the criterion invoked for the overlap phase."},{"cited_title":"Viyuela,, A","cited_arxiv_id":null,"evidence_quote":"Contributes the probe-qubit scheme for extracting the overlap phase from Pauli expectation values."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Demonstrates Berry-phase measurement in a superconducting qubit, the experimental template the circuit adapts."},{"cited_title":"Reinforcement learning for semi-autonomous approximate quantum eigensolver","cited_arxiv_id":"1906.06702","evidence_quote":"Supplies the quantum eigensolver subroutine that would make the protocol applicable when eigenstates are not known analytically."}],"review_version":1}