{"id":"44ba4e4f-bb10-4b17-885d-083300174e3f","arxiv_id":"1908.07597","paper_version":6,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"The authors construct locally-acting mirror Hamiltonians using position-space field operators that require both positive and negative frequency photons.","lead":"This paper quantizes the electromagnetic field in position space by including both positive and negative frequency solutions, then constructs mirror interaction Hamiltonians that act only on incoming light. It offers a way to model semi-transparent mirrors and broadband light-matter interactions with local equations of motion.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The derivation of the scattering operator in Eq. (50) assumes time-ordering can be ignored; for non-delta mirror profiles H_I(t) does not commute with itself at different times, so Eq. (53) is not the actual S-matrix of H_mirr.","rationale":"The reader's weakest assumption identifies exactly the time-ordering problem in the step from Eq. (50) to Eq. (53), and my independent review reaches the same conclusion. This is the most load-bearing concern because the existence claim for the mirror Hamiltonian rests on the scattering operator derivation: without a valid derivation, the general construction in Eq. (44) is not shown to reproduce classical mirror scattering. The concern is not a refutation of the whole paper: the special profile in Section 4.4 uses commuting generators and is supported by an exact Ehrenfest calculation, so the paper contains a correct special-case construction. However, the text explicitly presents Eq. (53) as an approximation-free general result, and that overclaim propagates into the discussion around Eq. (56) and into the claimed necessity of the doubled Hilbert space. The unbounded-below dynamical Hamiltonian is a separate physical concern, but it is not the central logical gap. Since the reader already issued a conditional verdict and this review confirms the same specific gap without escalating it, the verdict should remain unchanged: conditional acceptance pending a corrected (time-ordered or restricted-to-commuting) derivation of the scattering operator.","tokens_in":16458,"tokens_out":22380,"duration_ms":189970,"concrete_test":"Compute the second-order Magnus term for a finite-width, non-delta mirror profile, e.g. Omega_xx' = eta exp[-(x^2 + x'^2)/(2L^2)] with L > 0. Evaluate (1/2) integral_{-inf}^{inf} dt integral_{-inf}^{t} dt' [H_I(t), H_I(t')] acting on a single incoming right-moving wave packet. If this operator is nonzero, then T exp(-i/hbar integral H_I dt) differs from the ordinary exponential in Eq. (50), and Eq. (53) is not the S-matrix of H_mirr. A nonzero second-order term would directly falsify the 'without approximations' claim for general mirror profiles.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In Section 4.3 the paper writes S_I = exp(-i/hbar * integral H_I(t) dt) in Eq. (50) and then claims, after deriving Eq. (53), that 'The above scattering operator has been derived without approximations.' For the unitary evolution generated by H_mirr, the correct object is the time-ordered exponential T exp(-i/hbar * integral H_I(t) dt). These two agree only if [H_I(t), H_I(t')] = 0 for all t and t'. This is not true for the general interaction in Eqs. (48)-(49). The interaction picture Hamiltonian contains operators A_1(x) A_-1^dagger(x') and their Hermitian conjugates; the commutator of such terms with their adjoints is proportional to delta functions of spatial separations and does not vanish when the supports of Omega_{x+ct, x'-ct} and Omega_{y+ct', y'-ct'} overlap. For a mirror of finite thickness, such overlaps occur, so time-ordering matters. The Magnus expansion would generate additional multi-mode terms beyond the single integral in Eq. (50), and these terms can modify the on-shell scattering phases and the conclusion that positive and negative frequencies never mix. The special profile Omega_xx' = Omega(x) delta(x+x') used in Section 4.4 is a commuting case, because the generators become independent beam-splitter generators at different positions; for that profile Eq. (50) is justified and the Ehrenfest derivation in Section 4.4 provides independent support. But Eq. (53) is presented as a general result, and it is used in the argument leading to Eq. (56) and in the discussion of why a positive-frequency-only Hamiltonian cannot specify the mirror position. Thus the central claim is not established for the general locally-acting mirror Hamiltonians advertised in the abstract and in Eq. (44); it is only established for the special commuting profile. This is a load-bearing gap in the proof, rather than a refutation of the special construction.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a position-space quantization of the one-dimensional electromagnetic field using both positive- and negative-frequency photon modes, introduces truly-local bosonic annihilation operators A_{sλ}(x), and uses them to construct a Hermitian, time-independent 'mirror Hamiltonian' H_mirr = H_dyn + H_int (Eq. 41). The authors claim that this Hamiltonian acts only on incoming wave packets and that the resulting scattering operator S_I (Eq. 53) maps incoming modes to outgoing modes, preserving frequency and polarization, with complete reflection when Ξ_k = π/2. They also provide an exactly solvable example in Section 4.4 with Ω_{xx'} = Ω(x) δ(x+x'), where Ehrenfest equations give a closed rotation of incoming and outgoing amplitudes.","tokens_in":16814,"tokens_out":14422,"duration_ms":286851,"significance":"If the general construction were valid, it would provide a local Hamiltonian model of a two-sided semi-transparent mirror and avoid the nonlocality of standard positive-frequency-only input-output treatments. The paper's strengths are its explicit construction of local bosonic position-space operators, the clean separation of the dynamical Hamiltonian from the positive energy observable, and the exactly solvable delta-profile example in Section 4.4, which is self-consistent and supported by a direct Ehrenfest calculation. The main limitation is that the general scattering-operator derivation omits time-ordering, so Eq. (53) is not established as the actual S-matrix of H_mirr for general Ω_{xx'}. The central idea is interesting and the special case works, but the general claim needs repair.","major_comments":[{"comment":"The interaction-picture evolution generated by H_mirr is the time-ordered exponential U_I(∞,-∞) = T exp(-i/ℏ ∫_{-∞}^{∞} dt H_I(t)), not the ordinary exponential in Eq. (50). The latter agrees with the former only if [H_I(t), H_I(t')] = 0 for all t and t'. For the general profile in Eqs. (48)-(49) this commutativity is not established and is generally false: commutators of the beam-splitter terms A_1(x) A_{-1}^†(x') with their Hermitian conjugates contain δ-function factors such as δ(x-y) and δ(x'-y'), and these give a nonzero integrated commutator when the supports of Ω_{x+ct,x'-ct} and Ω_{y+ct',y'-ct'} overlap, as they do for any mirror of finite thickness. Hence Eq. (53) is not the exact S-matrix of the Hamiltonian dynamics, and the sentence after Eq. (53) stating that it 'has been derived without approximations' is not justified in the general case. A Magnus expansion would produce higher-order multi-mode terms that can modify the k-dependent phases and can mix the frequency sectors; only the special profile Ω_{xx'} = Ω(x) δ(x+x') of Section 4.4, where the generators at different positions commute, avoids this problem. Since Eq. (53) is the basis for the no-frequency-mixing conclusion and for the comparison in Eq. (56), this is a load-bearing issue.","section":"§4.3, Eqs. (50)–(53)"},{"comment":"The free dynamical Hamiltonian H_dyn in Eq. (23) has eigenvalues ℏck with k ranging over all real values and is therefore unbounded from below. The paper explicitly distinguishes H_dyn from the positive-definite energy observable H_eng in Eq. (17), but H_dyn remains the generator of time translations in Eq. (24) and is part of H_mirr in Eq. (41). Consequently the vacuum |0⟩ is not the ground state, and the model can in principle release arbitrarily large amounts of energy in transitions to deeply negative-frequency states. The manuscript should either show that the interaction-picture computation and the scattering operator are nevertheless well-defined and physically meaningful, or state clearly that the construction is a formal effective model whose stability is not addressed.","section":"§3, Eqs. (23) and (17); §4.1"}],"minor_comments":[{"comment":"In the paragraph defining incoming and outgoing wave packets, the second occurrence of 'x>0 and s=-1' for outgoing packets should presumably be 'x>0 and s=1'; as written, both outgoing cases have s=-1.","section":"§4.2"},{"comment":"The word 'locally-acing' should be 'locally-acting'.","section":"§3.2"},{"comment":"According to Eq. (54), Ξ_k is defined as i times a real integral, so for real symmetric Ω_{xx'} it is purely imaginary. The condition for complete reflection should be Ξ_k = iπ/2 (or equivalently -iΞ_k = π/2), not Ξ_k = π/2 as stated in the text. In the Section 4.4 example the real rotation angle is ∫Ω dt, so the Ξ_k of Eq. (54) differs from the angle appearing in Eq. (62) by a factor i; the notation should be made consistent.","section":"§4.3, Eqs. (54) and (62)"},{"comment":"The comparison of ∫ dt Ω_k with Ξ_k assumes a time-independent Ω_k in Eq. (55). The text should clarify why a time-dependent coupling Ω_k(t) with finite integral is not allowed, since such a coupling could satisfy Eq. (56).","section":"§4.3, Eq. (56)"}],"recommendation":"major_revision","confidential_remarks":"This manuscript deliberately departs from the standard Hilbert space by including negative-frequency photons and separating the dynamical Hamiltonian from the energy observable. The time-ordering issue in §4.3 is the central technical problem; if the authors restrict the general claim to commuting profiles or provide a controlled Magnus expansion, the paper could be publishable. The companion reference [28] appears to carry much of the justification for the position-space construction, and the present paper should be self-contained enough for the claims it makes."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper constructs locally-acting mirror Hamiltonians from position-space bosonic operators with negative-frequency modes, and the central idea is worth taking seriously. What is new is the explicit Hamiltonian construction and the derivation of a scattering operator that maps incoming to outgoing modes. The Section 4.4 example with Ω_{xx'}=Ω(x)δ(x+x') is clean and self-consistent: the interaction-picture Hamiltonian is built from mutually commuting generators at all times, so the time-ordering issue disappears, and the Ehrenfest equations close exactly, giving a proper beam-splitter dynamics. That is a genuine, reproducible result.\n\nThe soft spot is the general derivation in Section 4.3. Eq. (50) exponentiates the integral of H_I(t) without a time-ordered product, and the paper claims this is exact. For general Ω_{xx'}, H_I(t) does not commute with itself at different times—commutators like [A_1(x)A†_{-1}(x'), A_{-1}(y')A†_1(y)] produce delta-function terms that do not vanish. So exp(-i∫H_I dt) is not the exact S-matrix for a general mirror, and Eq. (53) is not established in that generality. This matters because Eq. (53) is used to argue that positive and negative frequencies never mix and to motivate the Hilbert-space doubling. That argument is rigorous for the delta profile but remains an assumption for the general locally-acting Hamiltonians advertised in the abstract. Also, H_dyn is unbounded below and no normal-ordering or renormalization discussion is given; that is a minor worry for a toy model, but it should be stated.\n\nThe paper is aimed at quantum optics people working with broadband pulses and local light-matter coupling in one dimension. They will get a useful technique and a good discussion of why positive-frequency-only effective Hamiltonians cannot specify the mirror position. I agree with the reader's conditional verdict: the gap is specific and addressable, not a refutation of the main idea.\n\nSend this to peer review. The Section 4.4 construction is a real result, and the general claim can be fixed by either restricting it to the commuting profile or presenting the time-ordered version and stating what is actually approximate.","headline":"A real, useful construction of position-space mirror Hamiltonians with a clean special case, but the general scattering-operator claim in Eq. (53) goes beyond what is proved; the Section 4.4 example is solid.","tokens_in":17400,"tokens_out":6517,"would_cite":true,"duration_ms":500442,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A Hermitian, time-independent mirror Hamiltonian built from position-space photon operators acts only on incoming wave packets and reproduces classical mirror-image scattering for semi-transparent mirrors.","keywords":["position-space quantisation","negative-frequency photons","mirror Hamiltonian","semi-transparent mirror","scattering operator","bosonic commutator relations","wave packets","quantum optics"],"falsifier":"Take two incoming wave packets with distinct frequencies $k_1$ and $k_2$ and calculate the exact time-ordered evolution generated by $H_I(t)$ from Eq. (52); if the result differs from the paper's $S_I$ acting on the same two-photon state, the 'no approximations' step from Eq. (50) to Eq. (53) is not valid beyond single-excitation inputs. A simpler check is to evaluate $[H_I(t),H_I(t')]$ for a generic profile $\\Omega_{xx'}$: whenever it is nonzero, time-ordering cannot be ignored.","tokens_in":16247,"feed_emoji":"🪞","tokens_out":12058,"duration_ms":97130,"temperature":0.7,"pith_summary":"Most descriptions of quantised light decompose the field into monochromatic waves that are spread over all space, which makes local optical elements like mirrors awkward to model. This paper develops a position-space quantisation of the one-dimensional electromagnetic field that includes both positive- and negative-frequency solutions of Maxwell's equations, and uses the resulting truly-local bosonic field operators to construct Hermitian, time-independent mirror Hamiltonians. The paper claims these Hamiltonians act only on wave packets approaching a semi-transparent mirror, turn them into outgoing mirror-image wave packets, and preserve frequency and polarisation. If this works, classical mirror scattering becomes a first-principles unitary quantum evolution that can describe light arriving from both sides of the interface without the nonlocal artefacts of standard treatments.","feed_headline":"Negative-frequency photons enable local mirror Hamiltonians","feed_subtitle":"Position-space quantisation makes semi-transparent mirror scattering a local, unitary process that preserves frequency and polarisation.","key_machinery":"The central object is the family of truly-local bosonic field operators $A_{s\\lambda}(x)$, defined through the Fourier kernel $f(k)=1/\\sqrt{2\\pi}\\,e^{i\\,\\mathrm{sgn}(k)\\varphi}$, which makes single-excitation states at different positions pairwise orthogonal and gives the operators delta-function commutation relations. These operators are the building blocks of the mirror interaction Hamiltonian $H_{\\rm int}=\\sum_{\\lambda=\\pm}\\int\\!\\!\\int dx\\,dx'\\,i\\hbar\\Omega_{xx'}\\,[A_{1\\lambda}^{(S)}(x)A_{-1\\lambda}^{(S)\\dagger}(x')-\\mathrm{H.c.}]$, whose real coupling profile $\\Omega_{xx'}$ is nonzero only near the mirror interface. The argument is carried by the scattering operator $S_I$ obtained from this Hamiltonian: because the interaction-picture Hamiltonian only contains combinations that conserve $|k|$ after the time integration, the resulting exponential acts as a frequency-preserving unitary that swaps left- and right-moving excitations. The phase $\\Xi_k$ controls the reflection amplitude and recovers the classical transmission and reflection ratios of a semi-transparent mirror.","core_discovery":"The central claim is the existence of a Hermitian, time-independent mirror Hamiltonian $H_{\\rm mirr}=H_{\\rm dyn}+H_{\\rm int}$ for light in one dimension that acts locally and only on incoming wave packets. The construction quantises both the positive- and the negative-frequency solutions of Maxwell's equations, effectively doubling the Hilbert space, and defines annihilation operators $A_{s\\lambda}(x)$ for truly-localised field excitations with the bosonic commutator relations $[A_{s\\lambda}(x),A_{s'\\lambda'}^{\\dagger}(x')]=\\delta_{ss'}\\delta_{\\lambda\\lambda'}\\delta(x-x')$. The interaction Hamiltonian $H_{\\rm int}$ couples a right-moving excitation at position $x$ to a left-moving excitation at $-x$ near the mirror plane, which is the quantum version of the classical mirror-image method. Working in the interaction picture, the paper derives the scattering operator $S_I=\\exp\\bigl(-(i/\\hbar)\\int_{-\\infty}^{\\infty}dt\\,H_I(t)\\bigr)$ and simplifies it to $S_I=\\exp\\!\\bigl(-i\\sum_{\\lambda}\\int_{-\\infty}^{\\infty}dk\\,[\\Xi_k\\,a_{1\\lambda}(k)\\,a_{-1\\lambda}^{\\dagger}(k)+\\mathrm{H.c.}]\\bigr)$, with $\\Xi_k=(i/c)\\int\\!\\!\\int dx\\,dx'\\,\\Omega_{xx'}e^{ik(x+x')}$. This operator maps incoming to outgoing modes without changing frequency or polarisation, couples positive frequencies only to positive and negative only to negative, and for $|\\Xi_k|=\\pi/2$ produces complete reflection.","pith_inferences":["A rigorous treatment of the time-ordering in Eq. (50) would likely introduce small corrections for multi-photon inputs; testing two-photon scattering through the same mirror would expose them.","The same doubled-Hilbert-space construction could yield local Hamiltonians for other optical elements, such as beam splitters with frequency-dependent reflectivity or dispersive media, by choosing $f(k)$ and $\\Omega_{xx'}$ appropriately.","If the scattering operator is truly unitary on the full Hilbert space, it implies an exact mapping between mirror scattering and a beamsplitter transformation on the doubled Hilbert space, which could be exploited in cascaded linear-optics networks.","The position-space operators resemble temporal-mode descriptions of light, so the construction may connect to ultrafast quantum optics experiments with broadband photonic wave packets."],"forward_implications":["Light scattering by a two-sided semi-transparent mirror can be modelled by a Hermitian, time-independent Hamiltonian that acts only on incoming wave packets.","The scattering operator preserves the frequency and polarisation of incoming photons, so positive-frequency photons remain positive and negative-frequency ones remain negative.","For coupling strength $|\\Xi_k|=\\pi/2$, the model gives complete conversion of incoming into outgoing wave packets, i.e., perfect reflection.","The Hamiltonian reproduces the classical mirror-image dynamics, including shape preservation of reflected wave packets, and can describe light approaching the mirror from both sides without unphysical interference."],"supporting_citations":[{"why":"Supplies the classical mirror-image scattering dynamics that the constructed Hamiltonians are designed to reproduce.","marker":"[22]"},{"why":"Provides the position-space quantisation formalism, including the generalised Fourier transform and the choice of $f(k)$, on which the local field operators are built.","marker":"[28]"},{"why":"Gives the quantum mirror-image detector method whose Hilbert-space doubling motivates including negative-frequency photons.","marker":"[19]"},{"why":"Extends the mirror-image method to asymmetric mirrors and supports the claim that the doubled Hilbert space is needed to specify the mirror position.","marker":"[20]"},{"why":"Serves as the standard momentum-space description of the quantised field that must be recovered as a special case.","marker":"[26]"}],"fun_headline_variants":["Negative frequencies make mirror interactions local","Position-space quantisation yields local mirror Hamiltonians","Local mirror Hamiltonians via negative-frequency photons","Mirror Hamiltonians act locally with negative modes","Negative-frequency modes enable local mirror scattering"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The derivation of the scattering operator treats the time-ordered exponential as an ordinary exponential of the integrated Hamiltonian, which requires the interaction-picture Hamiltonian to commute with itself at different times; the paper does not prove this commutativity, and for the multimode operators it is generally false.","fun_headline_variants_meta":{"raw":{"variants":["Negative frequencies make mirror interactions local","Position-space quantisation yields local mirror Hamiltonians","Local mirror Hamiltonians via negative-frequency photons","Mirror Hamiltonians act locally with negative modes","Negative-frequency modes enable local mirror scattering"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.0009,"raw_usage":{"total_tokens":3894,"prompt_tokens":986,"completion_tokens":2908,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":602,"completion_tokens_details":{"reasoning_tokens":2853}},"tokens_in":602,"tokens_out":2908,"duration_ms":22834,"temperature":1.0,"reasoning_tokens":2853,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:02:23.412448+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take two incoming wave packets with distinct frequencies $k_1$ and $k_2$ and calculate the exact time-ordered evolution generated by $H_I(t)$ from Eq. (52); if the result differs from the paper's $S_I$ acting on the same two-photon state, the 'no approximations' step from Eq. (50) to Eq. (53) is not valid beyond single-excitation inputs. A simpler check is to evaluate $[H_I(t),H_I(t')]$ for a generic profile $\\Omega_{xx'}$: whenever it is nonzero, time-ordering cannot be ignored.","supporting_citations":[{"cited_title":"Local photons","cited_arxiv_id":"2104.04499","evidence_quote":"Provides the position-space quantisation formalism, including the generalised Fourier transform and the choice of $f(k)$, on which the local field operators are built."},{"cited_title":"Furtak-Wells, L","cited_arxiv_id":null,"evidence_quote":"Gives the quantum mirror-image detector method whose Hilbert-space doubling motivates including negative-frequency photons."},{"cited_title":"Dawson, N","cited_arxiv_id":null,"evidence_quote":"Extends the mirror-image method to asymmetric mirrors and supports the claim that the doubled Hilbert space is needed to specify the mirror position."},{"cited_title":"Bennett, T","cited_arxiv_id":null,"evidence_quote":"Serves as the standard momentum-space description of the quantised field that must be recovered as a special case."}],"review_version":1}