{"id":"3c313d57-5123-4e2f-8217-b9dd6de7a8b2","arxiv_id":"1908.03722","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The recommended 225Ra electric dipole moment enhancement factors are -6.29(1) x 10^-17 |e| cm/(|e| fm^3) for the Schiff moment and -12.66(14) x 10^-20 <sigma_N> |e| cm for the T-PT interaction, with alpha_d = 244(13) ea0^3.","lead":"This paper computes how strongly radium-225 amplifies tiny time-reversal-violating effects in its electric dipole moment using a refined coupled-cluster method. The numbers it recommends will help future experiments use radium-225 to search for physics beyond the Standard Model.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Recommended EDM values hinge on post hoc selection of the d-perturbation variants; smooth convergence is not an accuracy benchmark, and the values disagree with earlier RCCSD(T) results.","rationale":"The reader's verdict is CONDITIONAL, and this stress-test confirms that conditionality rather than moving it. The paper contains a substantial many-body calculation with detailed term-by-term tables, and the RNCC formalism is standard and clearly presented. The dipole-polarizability recommendation, 244(13) ea0^3, is supported by agreement with several independent RCCSD(T) calculations and is therefore not the main concern. The load-bearing weakness is the unsupported selection of the d-perturbation variants for the EDM recommendations. The manuscript's own text provides the key evidence: it calls the w-perturbation expression 'more appropriate' in Sec. III, yet abandons those results because of iterative convergence behavior; it acknowledges that triples may change the results; and the abstract's averaging description is arithmetically inconsistent with the body. The recommended error bars exclude known truncation effects and the spread of the discarded variants. A triples-level calculation of the d-variants would directly test whether the recommended values are stable or are artifacts of the SD truncation. Until such a test is performed, the recommended EDM enhancement factors should be treated as conditional.","tokens_in":18053,"tokens_out":4229,"duration_ms":47914,"concrete_test":"Compute the T-PT and NSM EDM enhancement factors in 225Ra at the RCCSD(T) level (or with a perturbative triples correction) using both the w- and d-perturbation variants, with the same Gaussian basis and V_N potential. If the RCCSD(T) d-values remain within the quoted error bars of -12.66(14) and -6.29(1), the post hoc selection is vindicated. If they shift toward the earlier RCCSD(T) values near -10 and -6.8, the recommended values and their uncertainties are not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the recommended enhancement factors for 225Ra: -12.66(14) x 10^-20 for T-PT and -6.29(1) x 10^-17 for NSM. These are, in fact, averages of only the dipole-perturbation variants RCCSDd and RNCCSDd, not of all four variants as the abstract states. The paper's justification in Sec. IV is that the d-perturbed amplitudes 'converge smoothly' whereas the w-perturbed amplitudes do not, and on this basis the d results are 'more reliable.' This is a numerical-stability criterion, not an accuracy criterion; two methods sharing the same perturbation operator D can agree because they share the same approximation, not because they are correct. The internal contradiction is sharpened by Sec. III, which explicitly states that the w-perturbation expression, Eq. (6), 'should be treated as being more appropriate than Eq. (8)' (the d-perturbation recasting). Moreover, the recommended T-PT value differs by about 26% from the earlier RCCSD(T) result of Singh and Sahoo (-10.01 in Table I), and the w-variants are the ones consistent with previous RCC calculations. The quoted uncertainties, ±0.14 and ±0.01, reflect only the spread between the two d-variants and do not cover triples, Breit/QED, basis incompleteness, or the full four-way spread (which is roughly 3.4 for T-PT and 0.87 for NSM). The paper itself admits triples 'may lead to a much better agreement,' explicitly acknowledging that the SD-level results may be incomplete. For the recommendation to hold, the d-variants must be independently validated as accurate; that evidence is absent.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript applies the relativistic normal coupled-cluster (RNCC) method at the singles-and-doubles level to compute the electric dipole moment enhancement factors of 225Ra due to the nuclear Schiff moment (NSM) and the tensor-pseudotensor (T-PT) electron-nucleus interaction, and also the static dipole polarizability alpha_d. The authors develop both weak-interaction-perturbed (w) and dipole-perturbed (d) variants within RCCSD and RNCCSD, solve the corresponding amplitude equations, and tabulate individual correlation contributions. They recommend final enhancement factors -6.29(1) x 10^-17 |e| cm/(|e| fm^3) and -12.66(14) x 10^-20 <sigma_N> |e| cm, obtained by averaging the RCCSDd and RNCCSDd results, and an alpha_d of 244(13) ea0^3 obtained by averaging the RCCSD and RNCCSD alpha_d values. The paper also compares all results with previous RCC, RPA, and CI+MBPT calculations.","tokens_in":18368,"tokens_out":6471,"duration_ms":65292,"significance":"If the recommended EDM enhancement factors are reliable, they would serve as important benchmarks for extracting CP-violating couplings from the ongoing 225Ra EDM experiment. The RNCC treatment is methodologically interesting because it avoids the non-terminating series and normalization ambiguity of the RCC approach, and the alpha_d value is consistent with several earlier high-level calculations. The paper provides a useful comparison of term-by-term contributions between RCC and RNCC. The main weakness is that the central recommendation rests on a post hoc choice of the d-perturbation results whose justification contradicts the paper's own statement that the w-perturbation expression is more appropriate, and the quoted uncertainties do not reflect the full spread of the computed values.","major_comments":[{"comment":"The abstract states that the recommended EDM values are an average of the results from two variants each of both the RCC and RNCC methods, but Table I shows that the T-PT and NSM recommendations are exactly the averages of RCCSDd and RNCCSDd only: (-12.519-12.803)/2 = -12.661 and (-6.284-6.295)/2 = -6.2895. The w-variant results (-9.774 and -13.148 for T-PT; -6.183 and -7.053 for NSM) are excluded without being reflected in the stated central values. This discrepancy between the abstract and the actual procedure must be corrected and the basis for excluding the w-variants stated explicitly.","section":"Abstract and Table I"},{"comment":"The paper says in Sec. III that Eq. (6) 'should be treated as being more appropriate than Eq. (8)' and that Eq. (8) is 'just a mathematical recast.' However, the recommended EDM values are based on the d-perturbation variants that implement the recast expression (Eq. (8) or its analogue). No argument is given for why numerical convergence should override the formal preference for Eq. (6), so the selection of the d-variants for the final recommendation appears post hoc relative to the stated theoretical preference.","section":"Sec. III (Eqs. (6) and (8)) and Sec. IV"},{"comment":"The justification for preferring the RCCSDd and RNCCSDd results is that the first-order perturbed amplitudes due to the dipole operator 'converge smoothly' whereas the w amplitudes are 'unusually large in the first few iterations.' This is a statement about the numerical stability of the Jacobi iteration, not about the physical accuracy of the truncated many-body expansion. Two methods that share the same perturbation operator D can agree because they share the same systematic error, and agreement alone does not establish correctness; no independent accuracy benchmark is provided for the EDM enhancement factors.","section":"Sec. IV, convergence discussion"},{"comment":"The quoted uncertainties, +/-0.14 for T-PT and +/-0.01 for NSM, are simply half the difference between the RCCSDd and RNCCSDd results. They do not include the spread among all four variants (about 3.4 for T-PT and 0.87 for NSM), the estimated effect of triples (which the text says 'may lead to a much better agreement' in the alpha_d analysis), or the Breit/QED contributions that are neglected based on Ref. [21]. The recommended error bars therefore understate the theoretical uncertainty and should be recomputed to reflect all identified error sources.","section":"Sec. IV, uncertainty estimates"},{"comment":"The recommended T-PT central value, -12.66, differs from the earlier RCCSD(T) result -10.01 of Ref. [21] by about 26%, which is far outside the quoted uncertainty. Since the earlier result includes partial triples, the paper needs to explain why the SD-level d-variant values should replace it, or demonstrate that the triples contribution is already captured differently. Without such an explanation, the recommendation conflicts with the highest-order previous calculation listed in the paper's own comparison table.","section":"Table I and Sec. IV"}],"minor_comments":[{"comment":"The phrase 'It to be noted that' should be 'It is to be noted that.'","section":"Sec. III, after Eq. (8)"},{"comment":"The tables use a dagger on the Lambda operator (e.g., Lambda^(0)_1^dagger D T^(1)_1) although the RNCC expectation values in Eqs. (35)-(36) do not involve a Hermitian conjugate of Lambda; please clarify the notation for the bra-side de-excitation operator or use a consistent convention.","section":"Tables II-IV"},{"comment":"The text 'we find that the they are very similar' contains a typo and should read 'we find that they are very similar.'","section":"Sec. IV, first paragraph"},{"comment":"The figures are labeled with Roman numerals (Fig. I, II, III); while not incorrect, Arabic numerals would be more consistent with standard journal style.","section":"Figure captions"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the scope of the journal and the RNCC machinery itself appears sound. The main concern is the post hoc selection of the d-perturbation variants for the final EDM recommendation, which contradicts the paper's own formal preference for the w-perturbation expression and yields error bars that exclude the w-variant spread. However, the authors have all the data needed to re-derive a more defensible recommendation (e.g., reporting a range that covers all four variants, or providing a rigorous justification for excluding the w-variants), and the alpha_d benchmark is a useful contribution. I therefore recommend major revision rather than rejection. The abstract's wording will need to be corrected regardless of the outcome."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know before anything else. This is the first RNCC calculation of the 225Ra EDM enhancement factors, for a system with a live experimental program, and the alpha_d value, 244(13) ea0^3, overlaps the earlier accurate calculations (241-251 ea0^3), a decent sanity check. But the recommended EDM numbers come from a post hoc choice of the dipole-perturbed variants, and the abstract describes the averaging in a way the body does not support.\n\nWhat is genuinely good: the machinery is standard, parameter-free, and the term-by-term tables are detailed enough to audit where the numbers come from. The RNCC vs RCC comparison for this atom is new, and the two-perturbation-route (w vs d) diagnostic is a sensible cross-check even if the conclusion is debatable.\n\nThe soft spot is real and is the same one the referee report will land on. The abstract says the recommended values are an average of two variants each of both methods, but the body averages only RCCSDd and RNCCSDd. The w-perturbed values disagree: for T-PT, RCCSDw gives -9.77, RNCCSDw gives -13.15, and the earlier RCCSD(T) result from the same group is -10.01. The defense is that the d amplitudes converge smoothly while the w amplitudes are slow and large in early iterations. Smooth convergence is a stability criterion, not an accuracy criterion. Two calculations that share the same D operator can agree because they share the same flaw.\n\nThe paper knows the tension. Sec. III says the w-route expression, Eq. (6), should be treated as more appropriate than the d-route recasting, and later it admits triples may be needed for much better agreement. Then Sec. IV recommends the d-route values, with error bars (+/-0.14, +/-0.01) that cover only the two d-variants, not the four-way spread (about 3.4 for T-PT, 0.87 for NSM). That is not a demonstrated numerical error, but the recommendation procedure is not justified as written.\n\nI would not desk-reject. The alpha_d analysis is a legitimate calibration check, the RNCC application is novel for this atom, and the tables are useful. A serious referee could get it into publishable shape by requiring an abstract that describes the actual averaging, an honest quote of the four-way spread, a defense of why the d-variants beat the w-route when the paper itself says the w-route is formally more appropriate, and less finality for the SD-level numbers. No code or data is provided, so the numerics can't be independently checked, which is common in this subfield but worth noting. The citation pattern is unremarkable; self-citations are to method papers and prior RCC work.\n\nFor a reading group on atomic EDM calculations, this is a useful case study in how method uncertainty gets converted into recommended values. Send it to peer review, but expect heavy revision.","headline":"First RNCC calculation for 225Ra is a serious cross-check with a solid alpha_d sanity check, but the recommended EDM values rest on a post hoc choice of the d-perturbed variants and the abstract oversells the averaging.","tokens_in":18950,"tokens_out":7930,"would_cite":true,"duration_ms":70354,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["31.15.bw","32.10.Dk","11.30.Er"],"model":"deepseek-v4-flash","headline":"Relativistic normal coupled-cluster pins down the 225Ra EDM enhancement factors.","keywords":["electric dipole moment","radium-225","Schiff moment","tensor-pseudotensor interaction","relativistic normal coupled-cluster","relativistic coupled-cluster","dipole polarizability","CP violation"],"falsifier":"A calculation with full connected triple excitations in both RCC and RNCC, or a direct experimental measurement of the 225Ra dipole polarizability, would test the recommendation: if the dipole-perturbed EDM factors shift by more than their quoted 1% and 1.1% uncertainties, or if $\\alpha_d$ falls outside $244\\pm13\\,e a_0^3$, the smooth-convergence criterion for reliability is called into question.","tokens_in":17817,"feed_emoji":"⚛️","tokens_out":9421,"duration_ms":82788,"temperature":0.7,"pith_summary":"This paper tries to settle the atomic enhancement factors that will convert a future electric dipole moment (EDM) measurement in 225Ra into bounds on CP-violating physics. Earlier relativistic coupled-cluster (RCC) results for these factors disagreed with other many-body methods by about 40%. Because 225Ra has an exceptionally large octupole deformation, its EDM is enhanced by about three orders of magnitude relative to 129Xe and 199Hg, so accurate atomic factors are especially valuable. To resolve the discrepancy, the authors apply the relativistic normal coupled-cluster (RNCC) method, whose wave-function normalization is unity by construction and whose EDM expression terminates naturally. From the dipole-perturbed RCCSD and RNCCSD calculations they recommend enhancement factors of $-6.29(1)\\times10^{-17}\\,|e|\\,\\mathrm{cm}/(|e|\\,\\mathrm{fm}^3)$ for the nuclear Schiff moment and $-12.66(14)\\times10^{-20}\\,\\langle\\sigma_N\\rangle|e|\\,\\mathrm{cm}$ for the tensor-pseudotensor electron-nucleus interaction, together with a static dipole polarizability of $244(13)\\,e a_0^3$.","feed_headline":"225Ra EDM enhancement factors pinned to 1% accuracy","feed_subtitle":"Coupled-cluster cross-check gives the atomic factors needed to turn a radium EDM measurement into new-physics bounds.","key_machinery":"The central object is the relativistic normal coupled-cluster (RNCC) wave function, which uses a bi-orthogonal bra state $\\langle\\tilde{\\Psi}_0| = \\langle\\Phi_0|(1+\\Lambda)e^{-T}$ so that the normalization $\\langle\\tilde{\\Psi}_0|\\Psi_0\\rangle = 1$ is exact by construction. This removes the approximate cancellation of disconnected normalization terms that plagues truncated RCC theory and makes the EDM expression terminate naturally while satisfying the Hellmann-Feynman theorem. The paper also uses two mathematically equivalent perturbation routes, perturbing the wave function with the P,T-odd Hamiltonian (the 'w' variants) or with the electric dipole operator (the 'd' variants); the d variants converge smoothly and give consistent RCC and RNCC results, which is why the recommended values are based on them.","core_discovery":"The paper's central claim is that the relativistic normal coupled-cluster (RNCC) method removes the normalization ambiguity of truncated relativistic coupled-cluster (RCC) theory, and that when the EDM is computed by perturbing with the dipole operator rather than with the P,T-odd Hamiltonian, the RCCSD and RNCCSD results agree closely. On this basis the paper recommends $d_a^{\\mathrm{NSM}} = -6.29(1)\\times10^{-17}|e|\\,\\mathrm{cm}/(|e|\\,\\mathrm{fm}^3)$ and $d_a^{\\mathrm{T-PT}} = -12.66(14)\\times10^{-20}\\langle\\sigma_N\\rangle|e|\\,\\mathrm{cm}$ for 225Ra, with a static dipole polarizability $\\alpha_d = 244(13)\\,e a_0^3$ used as a corroborating benchmark.","pith_inferences":["A sharper test of the recommended values would be to compute the same EDM enhancement factors using full triple excitations in both RCC and RNCC; the paper itself notes that triples may improve the agreement, but does not include them.","The difference between the w and d routes could be used as a diagnostic for basis-set completeness: in the complete-basis, full-correlation limit the two routes must give identical values, so their residual disagreement measures the combined truncation error.","The same RCCSDd/RNCCSDd agreement criterion could be applied to other octupole-deformed nuclei, such as 225Ac or 229Pa, where method spreads of similar size may affect EDM interpretation.","The averaging choice matters: using all four variants (w and d for both methods) would change the recommended T-PT factor by about 5% relative to the d-only average, so the stated uncertainty may under-represent the method spread."],"forward_implications":["A 225Ra EDM measurement with limit $|d_a| \\le L$ can be converted into independent bounds on the nuclear Schiff moment and the T-PT coupling by dividing $L$ by $6.29(1)\\times10^{-17}$ and $12.66(14)\\times10^{-20}$, respectively.","The recommended $\\alpha_d = 244(13)\\,e a_0^3$ gives a target that future many-body calculations should reproduce, and because it agrees with earlier RCC values it supports the reliability of the d-variant EDM results.","The close agreement between RCCSDd and RNCCSDd suggests that the dipole-perturbation route is the numerically safer way to compute EDMs in truncated coupled-cluster theories, even when the weak-interaction-perturbed route is the formal definition.","The large spread among the w variants implies that previous RCC-only EDM calculations that relied on the weak-perturbation route may have uncertainties of tens of percent, and should be re-examined with the d route.","Because the RNCC expression terminates naturally, the method can be extended to higher excitations (e.g., triples) without the disconnected-diagram cancellation ambiguity that grows with truncation level."],"supporting_citations":[{"why":"Supplies the earlier RCC values for 225Ra EDM and alpha_d that the paper extends and compares with.","marker":"[21]"},{"why":"Introduces the RNCC method and demonstrates its agreement with RCC for Hg and Xe.","marker":"[25]"},{"why":"Establishes the normal coupled-cluster formalism with the bi-orthogonal bra state used here.","marker":"[29]"},{"why":"Provides the Hellmann-Feynman and termination properties of the RNCC energy and property expressions.","marker":"[27]"},{"why":"Defines the nuclear Schiff moment interaction Hamiltonian and earlier estimates of its atomic enhancement.","marker":"[2]"},{"why":"Defines the T-PT electron-nucleus interaction and gives earlier RCC results for the enhancement factor.","marker":"[20]"},{"why":"Provides a CCSD(T) alpha_d value for Ra used as a comparison benchmark.","marker":"[34]"},{"why":"Provides a perturbed RCC alpha_d value with Dirac-Coulomb-Breit Hamiltonian used as another benchmark.","marker":"[37]"}],"fun_headline_variants":["Radium-225 EDM factors settled by normal coupled-cluster","RNCC resolves radium-225 EDM discrepancies","Radium EDM factors now consistent across methods","225Ra EDM: normal coupled-cluster removes ambiguity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The recommended values assume that the dipole-perturbed (d) variants are the reliable ones because their first-order wave functions converge smoothly, and that the spread between the RCCSDd and RNCCSDd results is a fair estimate of the remaining uncertainty.","fun_headline_variants_meta":{"raw":{"variants":["Radium-225 EDM factors settled by normal coupled-cluster","RNCC resolves radium-225 EDM discrepancies","Radium EDM factors now consistent across methods","225Ra EDM: normal coupled-cluster removes ambiguity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000581,"raw_usage":{"total_tokens":2812,"prompt_tokens":1096,"completion_tokens":1716,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":712,"completion_tokens_details":{"reasoning_tokens":1650}},"tokens_in":712,"tokens_out":1716,"duration_ms":12830,"temperature":1.0,"reasoning_tokens":1650,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:04:40.873759+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A calculation with full connected triple excitations in both RCC and RNCC, or a direct experimental measurement of the 225Ra dipole polarizability, would test the recommendation: if the dipole-perturbed EDM factors shift by more than their quoted 1% and 1.1% uncertainties, or if $\\alpha_d$ falls outside $244\\pm13\\,e a_0^3$, the smooth-convergence criterion for reliability is called into question.","supporting_citations":[{"cited_title":"Spevak, N","cited_arxiv_id":null,"evidence_quote":"Supplies the earlier RCC values for 225Ra EDM and alpha_d that the paper extends and compares with."},{"cited_title":"Shavitt and R","cited_arxiv_id":null,"evidence_quote":"Establishes the normal coupled-cluster formalism with the bi-orthogonal bra state used here."},{"cited_title":"Borschevsky et al [36] performed calculations to obtain αd using an uncontracted universal basis set in the rel- ativistic CCSD(T) method (RCCSD(T) method)","cited_arxiv_id":null,"evidence_quote":"Defines the nuclear Schiff moment interaction Hamiltonian and earlier estimates of its atomic enhancement."},{"cited_title":"Auerbach, V","cited_arxiv_id":null,"evidence_quote":"Defines the T-PT electron-nucleus interaction and gives earlier RCC results for the enhancement factor."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides a CCSD(T) alpha_d value for Ra used as a comparison benchmark."},{"cited_title":"Ban, J .Dobaczewski, J","cited_arxiv_id":null,"evidence_quote":"Provides a perturbed RCC alpha_d value with Dirac-Coulomb-Breit Hamiltonian used as another benchmark."}],"review_version":1}