{"id":"302a1375-11c4-44e2-afa8-a69f6d507879","arxiv_id":"1908.05226","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper presents two formal path-integral methods and reproduces known propagators for a charged harmonic oscillator in electric and magnetic fields, but contains sign errors in the electric-field terms and omits the general anisotropic case.","lead":"A single-author physics paper derives the propagator for a charged anisotropic harmonic oscillator in crossed electric and magnetic fields using two path-integral tricks: complex Fourier series with zeta-function regularization, and a rotating frame that cancels the magnetic force. The general anisotropic result is promised but only limiting cases are worked out, and one energy-spectrum formula has a sign error.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The anisotropic main result Eq. (65) follows from an invalid rotating-frame step: the Larmor rotation does not diagonalize an anisotropic potential, so the effective frequencies in Eq. (64) are not the true normal-mode frequencies of the charged anisotropic oscillator.","rationale":"The paper's central claim is the path integral of a charged anisotropic harmonic oscillator in crossed electric and magnetic fields. The complex-Fourier method in Sec. III is only presented for the isotropic case, so the anisotropic claim rests entirely on the rotating-frame derivation in Sec. IV B. That derivation is invalid: the rotation to the Larmor frame cancels the magnetic term but does not preserve the anisotropic harmonic potential, which acquires time-dependent coefficients proportional to cos(2 omega_L t) and sin(2 omega_L t). These terms are missing from Eq. (63), and without them the effective frequencies in Eq. (64) are not justified. The exact normal-mode frequencies of a two-dimensional anisotropic oscillator in a magnetic field obey a quartic characteristic equation, and the paper's frequencies sqrt(omega_L^2+omega_x^2), sqrt(omega_L^2+omega_y^2) are not roots except in the isotropic limit. The omega_y=0 limit gives a sharp physical contradiction: the system has a continuous spectrum, while Eq. (65) predicts discrete y-oscillator levels. The reader's weakest assumption focused on zeta-regularization of the infinite product in Eq. (26); that is a meaningful formal concern, but it is not the load-bearing failure. The defect in the anisotropic rotating-frame argument is concrete, demonstrable, and invalidates the main promised result. Because the anisotropic generalization is the point of the title and the paper cannot be saved by typographical corrections alone, the current manuscript should be rejected; a corrected version would need a genuine treatment of the coupled anisotropic system, for example via the exact normal-mode diagonalization.","tokens_in":13666,"tokens_out":24437,"duration_ms":245231,"concrete_test":"Set E=0, omega_x=omega, omega_y=0 in Eq. (55) and compute the trace of the propagator in Eq. (65). Since the Hamiltonian is independent of y, the canonical momentum p_y is conserved and the spectrum is continuous, consisting of free motion in y plus one oscillator with frequency sqrt(omega^2+4 omega_L^2). Eq. (65) gives a product of two discrete oscillators with frequencies sqrt(omega^2+omega_L^2) and omega_L, which predicts discrete energy levels that do not exist. A direct check is to solve the classical equations of motion from Eq. (55) for omega_y=0: the nonzero characteristic frequency is sqrt(omega^2+4 omega_L^2), not sqrt(omega^2+omega_L^2). Alternatively, substitute the rotation (62) into the anisotropic potential and verify that Eq. (63) is obtained only when omega_x=omega_y.","verdict_should_be":"REJECT","load_bearing_attack":"The anisotropic result Eqs. (63)-(65) is not derived and is false for omega_x != omega_y. The rotating-frame step (62) is harmless only for a rotation-invariant potential. For the anisotropic potential in Eq. (55), substituting bar x = tilde x cos(omega_L t) + tilde y sin(omega_L t), bar y = -tilde x sin(omega_L t) + tilde y cos(omega_L t) gives -(m/2)(omega_x^2 bar x^2 + omega_y^2 bar y^2) = -(m/4)[(omega_x^2+omega_y^2)(tilde x^2+tilde y^2) + (omega_x^2-omega_y^2)((tilde x^2-tilde y^2) cos(2 omega_L t) + 2 tilde x tilde y sin(2 omega_L t))], with explicit time dependence that is dropped in Eq. (63). Hence the effective frequencies in Eq. (64) do not follow. Independently, the exact normal-mode frequencies of the quadratic Lagrangian in Eq. (55) solve the characteristic equation (omega_x^2-Omega^2)(omega_y^2-Omega^2)-4 omega_L^2 Omega^2=0. The claimed sqrt(omega_L^2+omega_x^2) and sqrt(omega_L^2+omega_y^2) satisfy this only in the isotropic case. A limiting case makes the failure concrete: for omega_y=0 there is no confinement in y, the system has a continuous spectrum, and the nonzero oscillator frequency is sqrt(omega_x^2+4 omega_L^2); Eq. (65) instead predicts discrete oscillators with frequencies sqrt(omega_x^2+omega_L^2) and omega_L. The title's promised anisotropic propagator is therefore not established; the zeta-regularization question in Eq. (26) is secondary to this.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes two calculational methods for the Feynman path integral of a charged anisotropic harmonic oscillator in crossed electric and magnetic fields. The first method expands quantum fluctuations in complex Fourier series and uses Riemann zeta-function regularization to evaluate divergent products; it produces the fluctuation factor F(T) = m omega_eff/(2 pi i hbar sin omega_eff T) for the isotropic magnetic case, Eq. (42). The second method transforms to a uniformly rotating frame, exploiting cancellation of Lorentz and Coriolis forces, and is used to derive the isotropic action Eq. (52) and, in Sec. IV B, a claimed full anisotropic propagator Eq. (65) with effective frequencies sqrt(omega_L^2+omega_x^2) and sqrt(omega_L^2+omega_y^2). The paper also gives an isotropic energy spectrum, Eq. (76), including the electric-field Stark shift.","tokens_in":14029,"tokens_out":15075,"duration_ms":131222,"significance":"If the anisotropic result were correct, the paper would provide a compact and formally simple derivation of the propagator for a charged anisotropic oscillator in crossed fields, which is a useful quantum-mechanics benchmark. The paper is self-contained, uses no fitted parameters, and correctly reproduces several known special cases: the free particle, the harmonic oscillator, Landau levels in the zero-oscillator limit, and the isotropic magnetic-oscillator propagator. The zeta-regularization calculation is heuristic but transparent, and the rotating-frame idea is elegant in the isotropic case. However, the central advertised anisotropic result is invalid, and several formulas in the isotropic electric-field case are internally inconsistent. The positive value of the paper is therefore confined to the isotropic special cases, which are mostly known results presented in a new way.","major_comments":[{"comment":"The rotating-frame step used to obtain the anisotropic propagator is invalid. Substituting the rotation (bar x, bar y)^T = R(t)(tilde x, tilde y)^T with R(t)=[[cos omega_L t, sin omega_L t],[-sin omega_L t, cos omega_L t]] into the potential in Eq. (55) gives -(m/2)(omega_x^2 bar x^2 + omega_y^2 bar y^2) = -(m/4)[(omega_x^2+omega_y^2)(tilde x^2+tilde y^2) + (omega_x^2-omega_y^2)((tilde x^2-tilde y^2) cos 2 omega_L t + 2 tilde x tilde y sin 2 omega_L t)], which is explicitly time-dependent and non-diagonal whenever omega_x != omega_y. The Lagrangian in the rotating frame is therefore not the static anisotropic oscillator of Eq. (63), and the effective frequencies in Eq. (64) are not the normal-mode frequencies. The exact normal-mode frequencies of the quadratic Lagrangian in Eq. (55) solve (omega_x^2 - Omega^2)(omega_y^2 - Omega^2) - 4 omega_L^2 Omega^2 = 0; for omega_y = 0 this yields Omega^2 = omega_x^2 + 4 omega_L^2 together with a zero mode, whereas Eq. (65) predicts discrete frequencies sqrt(omega_x^2+omega_L^2) and omega_L. The claimed anisotropic propagator is therefore not established and, in the form stated, is false.","section":"§IV B, Eqs. (63)–(65)"},{"comment":"For E = 0, Eq. (70) does not reduce to the earlier result Eq. (52). Using A1,...,A4 from Eq. (71), the E = 0 limit of Eq. (70) contains the term -2(x_a y_b - x_b y_a) sin omega_L T, whereas Eq. (52) contains -2(y_a x_b - x_a y_b) sin omega_L T = +2(x_a y_b - x_b y_a) sin omega_L T. The sign of the sin omega_L T cross term is opposite, so the claimed general classical action for the isotropic case is internally inconsistent.","section":"§IV B, Eq. (70)"},{"comment":"The self-propagating kernel quoted in Eq. (73) is not consistent with the coincident-endpoint limit of Eqs. (70)–(71). Setting x_b = x_a = x and y_b = y_a = y in the previously derived classical action gives an exponent proportional to (cos omega_eff T + cos omega_L T)/sin omega_eff T acting on a^2, together with additional q^2 E^2 terms in the bracket, whereas Eq. (73) uses (cos omega_eff T - cos omega_L T)/sin omega_eff T and drops those q^2 E^2 terms. In addition, Eq. (73) omits the prefactor m omega_eff/(2 pi i hbar sin omega_eff T) that is needed to evaluate the trace in Eq. (72). Since the energy spectrum in Sec. V is derived from Eq. (73), the derivation of Eq. (76) rests on a formula that does not follow from the preceding classical action.","section":"§IV B, Eq. (73) and §V"},{"comment":"The zeta-regularization step is the main formal premise of the first method. In Eq. (26) the k-integral is interchanged with an infinite product over Fourier modes, and the divergent product over n is subsequently evaluated using zeta(0) = -1/2 and zeta'(0) = -1/2 log(2 pi). These manipulations are taken as a definition of the path-integral normalization rather than derived from a time-sliced midpoint discretization, which is precisely the magnetic-field ambiguity the Introduction promises to circumvent. The final isotropic result Eq. (42) agrees with known results, which is reassuring, but the method as stated is not shown to be equivalent to the standard discretized construction. This concern is secondary to the anisotropic error, but it should be addressed explicitly if the paper is revised.","section":"§III B, Eq. (26)"}],"minor_comments":[{"comment":"The potential term in the shifted Lagrangian is written as -m omega_y^2 bar x^2 / 2 in the displayed equation; it should be -m omega_y^2 bar y^2 / 2.","section":"§IV B, Eq. (58)"},{"comment":"The notation bar S_t is introduced but not defined; it should denote the action of the oscillator part bar L_t. The split of the action into C and bar S_t would benefit from an explicit statement of the integration limits.","section":"§IV B, Eq. (59)"},{"comment":"The integral measure is written as D[gamma(t)] in a context where the variable shift has been performed in Cartesian variables; the notation should be D[bar x] D[bar y] for consistency with Eq. (61).","section":"§IV B, Eq. (62)"},{"comment":"The expression for tilde S_c^{tilde y} is abbreviated as (x ⇔ y). This is not a self-contained formula and should be written explicitly.","section":"§IV B, Eq. (66)"},{"comment":"The comparison with Ref. 4 states that a numerical check shows the expressions are identical, but no numerical data or explicit algebraic identification is provided. Either give the comparison or remove the claim.","section":"§IV B, after Eq. (68)"},{"comment":"The derivation of zeta'(0) via the Wallis product is labeled as heuristic, which is appropriate, but the chain of equalities leading from Eq. (C16) to Eq. (C18) is difficult to follow and would benefit from an explicit statement that divergent series are being manipulated formally.","section":"Appendix C, Eq. (C20)"}],"recommendation":"reject","confidential_remarks":"The isotropic portions of the paper contain useful and correct special cases, and the zeta-regularization method is pedagogically interesting. However, the paper's advertised central result, the anisotropic propagator of Sec. IV B, is derived through an invalid rotating-frame diagonalization and is false as stated. This is a load-bearing error that cannot be repaired by small local corrections; a substantial rewrite, including a change of the title and abstract, would be needed to restrict the claims to the isotropic case. I therefore cannot recommend publication in the present form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the bottom line: the paper has a genuinely useful formal trick for one class of problems, but the anisotropic result advertised in the title is not derived and is, as far as I can tell, wrong.\n\nThe complex Fourier series with zeta-regularization applied to the magnetic-field path integral is a nice variation. For the isotropic charged oscillator in a constant magnetic field, the method reproduces the known propagator and fluctuation determinant, and the demonstration that the real sine basis fails is instructive. The energies in Eq. (76) are the textbook Landau-oscillator spectrum. That part is solid.\n\nThe problem is Section IVB. The author tries to get the anisotropic propagator by shifting away the electric field and then rotating with the Larmor frequency. For an isotropic potential this works fine: the Larmor rotation removes the magnetic term and leaves a static isotropic oscillator at frequency sqrt(ω_L^2+ω^2). For an anisotropic potential the same rotation leaves explicit 2ω_L t time dependence in the potential. The author drops it and writes down a product of two 1D propagators with frequencies sqrt(ω_L^2+ω_x^2) and sqrt(ω_L^2+ω_y^2). But those are not the normal-mode frequencies of the charged anisotropic oscillator; the actual characteristic equation is (ω_x^2−Ω^2)(ω_y^2−Ω^2)−4ω_L^2 Ω^2=0, which the claimed frequencies do not satisfy except in the isotropic limit. So the main result of the paper is not established.\n\nThere are also smaller issues that would need fixing: Eq. (70) has a sign inconsistency with the E=0 limit in Eq. (52), and Eq. (73) drops a term quadratic in E. The zeta-regularization steps in Sec. III are formal (interchange of product and integral), but they give the right isotropic answers and are the kind of thing one can accept as a calculational device.\n\nVerdict: this paper should be sent to a referee, not desk-rejected, because the isotropic methodology is worth attention and the alleged anisotropic extension might be repairable. But in the current form it is not publishable. The author needs to either derive the anisotropic propagator correctly (using the true normal modes) or restrict the claim to the isotropic case. If the latter, the title and much of Sec. IVB should be revised accordingly.","headline":"The complex Fourier method is a neat formal trick that works for the isotropic charged oscillator, but the anisotropic propagator advertised in the title is based on an invalid Larmor rotation and is not correct as derived.","tokens_in":14579,"tokens_out":5862,"would_cite":false,"duration_ms":50497,"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":"The paper shows that complex Fourier series with zeta regularization and a rotating-frame transformation both evaluate the charged anisotropic harmonic oscillator path integral in crossed electric and magnetic fields.","keywords":["Feynman path integral","zeta-function regularization","anisotropic harmonic oscillator","crossed electric and magnetic fields","Larmor frequency","rotating frame","Landau levels","fluctuation integral"],"falsifier":"Take the charged isotropic oscillator in a uniform magnetic field, compute its propagator numerically with the standard time-sliced definition in which the vector potential is evaluated at the midpoint of each segment, extrapolate the result to zero slice width, and compare the fluctuation factor with $F(T)=m\\omega_{\\rm eff}/(2\\pi i\\hbar\\sin(\\omega_{\\rm eff}T))$; a mismatch in normalization or phase would show that the zeta-regularized definition is not the midpoint-discretized path integral.","tokens_in":13390,"feed_emoji":"⚛️","tokens_out":11385,"duration_ms":96242,"temperature":0.7,"pith_summary":"The paper shows that two-dimensional Gaussian path integrals with velocity-dependent magnetic terms and a static electric field can be evaluated formally without the usual obstacles: an infinite normalization constant and the need to discretize the vector potential at segment midpoints. The first method expands quantum fluctuations in a complex Fourier series and regulates the resulting divergent products with the Riemann zeta function, obtaining the fluctuation integral $F(T)=m\\omega_{\\rm eff}/(2\\pi i\\hbar\\sin\\omega_{\\rm eff} T)$ for a charged oscillator in a magnetic field. The second method rotates the coordinate frame at the Larmor frequency $\\omega_L=qB/2m$, so the Lorentz force cancels the Coriolis force and the Lagrangian becomes that of an effective harmonic oscillator. In the anisotropic case with crossed fields the full propagator follows as a product of two effective one-dimensional oscillators with frequencies $\\sqrt{\\omega_L^2+\\omega_x^2}$ and $\\sqrt{\\omega_L^2+\\omega_y^2}$. The two routes are shorter and more direct than the standard time-slicing computation, and they agree with known limiting results such as Landau levels.","feed_headline":"Zeta regularization and frame rotation solve magnetic path integrals","feed_subtitle":"Two methods yield the full propagator for a charged anisotropic oscillator without infinite normalization or midpoint discretization.","key_machinery":"The first carrier is the complex Fourier expansion of the fluctuation loop, $\\delta\\gamma(t)=\\sum_{n=-\\infty}^{\\infty} c_n e^{i2\\pi n t/T}$, which diagonalizes the action into a quadratic form in the Fourier coefficients; the boundary conditions are imposed by a delta function $\\delta(\\sum_n c_n)$. Divergent infinite products are then evaluated formally with zeta-regularization, using $\\zeta(0)=-1/2$ and $\\zeta'(0)=-\\frac12\\log(2\\pi)$, together with Euler's sine product. The second carrier is the transformation to a uniformly rotating frame at the Larmor frequency $\\omega_L=qB/2m$, the frequency at which the Lorentz force and the Coriolis force have the same form and cancel; this leaves an anisotropic harmonic oscillator with effective frequencies $\\omega_x^{\\rm eff}$ and $\\omega_y^{\\rm eff}$. Both methods assume the functional measure is invariant under the relevant translations and rotations.","core_discovery":"The central claim is that the charged anisotropic oscillator propagator is, up to a phase coming from completing the square with the electric field, exactly the product of two independent anisotropic-oscillator propagators with effective frequencies $\\omega_x^{\\rm eff}=\\sqrt{\\omega_L^2+\\omega_x^2}$ and $\\omega_y^{\\rm eff}=\\sqrt{\\omega_L^2+\\omega_y^2}$ after a frame rotation at the Larmor frequency and a spatial shift by the electric field. In the isotropic case the fluctuation factor reduces to $F(T)=m\\omega_{\\rm eff}/(2\\pi i\\hbar\\sin(\\omega_{\\rm eff} T))$, a symmetric expression that the paper states does not occur in the same form in standard references. The paper claims that both derivations are fully formal yet avoid introducing the infinite normalization constant and avoid the ambiguities of discretizing a magnetic vector potential; the zeta-regularized complex Fourier expansion and the rotation and translation invariance of the measure replace those steps.","pith_inferences":["If the zeta-regularized complex Fourier procedure is viewed as a definition of the path integral, the same regularization could be tested on non-quadratic perturbations such as a quartic anharmonic term, where the action no longer diagonalizes and the choice of regularization may change the result; the paper does not address that regime.","The rotating-frame argument is not tied to a uniform magnetic field in an essential way, so it may generalize to inhomogeneous fields or rotating traps where a local Larmor frequency appears, although the paper only treats the uniform-field case.","A direct comparison with a finite-dimensional time-sliced definition of the same propagator, with the vector potential evaluated at midpoints, would identify precisely which formal steps are equivalent to the standard definition and which are new conventions; that comparison is not made in the paper."],"forward_implications":["For a charged isotropic oscillator in a magnetic field, the fluctuation integral is $F(T)=m\\sqrt{\\omega^2+\\omega_L^2}/(2\\pi i\\hbar\\sin(T\\sqrt{\\omega^2+\\omega_L^2}))$, so no separate normalization constant is needed in path-integral evaluations.","For an anisotropic potential, the full propagator factorizes into $x$- and $y$-effective oscillators with frequencies $\\sqrt{\\omega_L^2+\\omega_x^2}$ and $\\sqrt{\\omega_L^2+\\omega_y^2}$, which directly gives the classical action from the known anisotropic oscillator action.","The energy spectrum in the isotropic case is $E(n,m)=q^2 E^2/(2m\\omega^2)+\\hbar(\\sqrt{\\omega^2+\\omega_L^2}+\\omega_L)(n+\\tfrac12)+\\hbar(\\sqrt{\\omega^2+\\omega_L^2}-\\omega_L)(m+\\tfrac12)$, which reduces to the Landau levels $\\hbar\\omega_c(n+\\tfrac12)$ in the zero-potential limit.","The complex Fourier series method extends the Fourier-expansion technique to path integrals with velocity-dependent potentials, where the real sine-series expansion fails because the magnetic cross term $\\langle\\delta x|\\dot{\\delta y}\\rangle$ vanishes identically."],"supporting_citations":[{"why":"supplies the standard path-integral framework and the harmonic-oscillator propagator that the paper uses as its basic check.","marker":"[8]"},{"why":"sets out the time-sliced propagator with a magnetic field and the midpoint rule that the new methods aim to circumvent.","marker":"[21]"},{"why":"gives the exact propagator for the anisotropic oscillator in a magnetic field used by the paper to confirm its formula by numerical comparison.","marker":"[4]"},{"why":"provides the complex Fourier-series treatment of closed curves that motivates expanding the quantum fluctuations as loops.","marker":"[12]"},{"why":"supplies the trigonometric series and infinite-product identities used in evaluating the zeta-regularized mode sums.","marker":"[10]"}],"fun_headline_variants":["Larmor rotation cancels magnetic force in path integral","Two clean ways to do magnetic path integrals without infinities","Zeta regularization tames anisotropic oscillator path integral","Frame rotation turns magnetic path integral into two oscillators","Charged anisotropic HO propagator via zeta and rotating frame"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument stands on the assumption that the formal way of summing divergent infinite products used here is a valid definition of the path integral, matching the standard definition that breaks time into small slices and evaluates the magnetic interaction at the middle of each slice; if the two definitions disagreed, the normalization and the propagator would change.","fun_headline_variants_meta":{"raw":{"variants":["Larmor rotation cancels magnetic force in path integral","Two clean ways to do magnetic path integrals without infinities","Zeta regularization tames anisotropic oscillator path integral","Frame rotation turns magnetic path integral into two oscillators","Charged anisotropic HO propagator via zeta and rotating frame"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000596,"raw_usage":{"total_tokens":2767,"prompt_tokens":899,"completion_tokens":1868,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":515,"completion_tokens_details":{"reasoning_tokens":1788}},"tokens_in":515,"tokens_out":1868,"duration_ms":14513,"temperature":1.0,"reasoning_tokens":1788,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:36:03.530470+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the charged isotropic oscillator in a uniform magnetic field, compute its propagator numerically with the standard time-sliced definition in which the vector potential is evaluated at the midpoint of each segment, extrapolate the result to zero slice width, and compare the fluctuation factor with $F(T)=m\\omega_{\\rm eff}/(2\\pi i\\hbar\\sin(\\omega_{\\rm eff}T))$; a mismatch in normalization or phase would show that the zeta-regularized definition is not the midpoint-discretized path integral.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies the standard path-integral framework and the harmonic-oscillator propagator that the paper uses as its basic check."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"sets out the time-sliced propagator with a magnetic field and the midpoint rule that the new methods aim to circumvent."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"gives the exact propagator for the anisotropic oscillator in a magnetic field used by the paper to confirm its formula by numerical comparison."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"provides the complex Fourier-series treatment of closed curves that motivates expanding the quantum fluctuations as loops."},{"cited_title":"Gradshteyn and I.M","cited_arxiv_id":null,"evidence_quote":"supplies the trigonometric series and infinite-product identities used in evaluating the zeta-regularized mode sums."}],"review_version":1}