{"id":"a8b71939-3fe5-4349-9138-4a67dca097a5","arxiv_id":"2509.07684","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"low","formal_verification":"none","parameter_count":3,"one_line_summary":"For Sr II, Plane-Wave Born and Distorted Waves approximations reproduce R-matrix effective collision strengths within an order of magnitude (DW within a factor of 2.5), offering cheap atomic data for kilonova nebular modeling.","lead":"This paper tests two fast, approximate methods for calculating how electrons excite the strontium ion Sr II, comparing them to a slower, more accurate reference method. The authors find the fast methods give the right order of magnitude for these collision rates, which could help model kilonova spectra when accurate data for heavy elements is unavailable.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Factor-2.5 benchmark uses an R-matrix reference whose target levels are 5–8% too high; at 1000 K those threshold errors suppress the reference effective collision strengths, so the central accuracy claim is not yet established.","rationale":"The paper is a useful and honest benchmark: it presents new radiative rates, clear level comparisons, and a direct table of effective collision strengths against R-matrix data. The reader's conditional verdict is appropriate. However, the most load-bearing weakness is not primarily the transferability from Sr II to lanthanides/actinides, but the energy consistency of the benchmark itself. The paper's own Table 1 shows that the AS target energies underlying the reference R-matrix calculation are 5–8% high for the low-lying 4d and 5p levels. Because the effective collision strength at T=1000 K is exponentially sensitive to threshold energy (Eq. 2), a threshold shift of ~0.1 eV can change Υ by factors of several. Thus the claimed 'within a factor of 2.5' agreement may be a consequence of comparing accurate-threshold approximate calculations to an inaccurate-threshold R-matrix reference. This is concrete, internal to the paper, and testable. The transferability concern raised by the reader is real but secondary; it can only be evaluated after the Sr II benchmark itself is placed on an energy-consistent footing. I therefore keep the conditional verdict, with the condition expanded to require an energy-consistent R-matrix comparison.","tokens_in":6945,"tokens_out":10799,"duration_ms":108119,"concrete_test":"Obtain the R-matrix collision strengths of [18] as a function of incident energy, shift the energy axis so that each excitation threshold matches the NIST/HFR-fitted levels from Table 1, re-evaluate Eq. (2) at T=1000 K, and recompute the Table 3 ratios. If any AS-DW ratio then exceeds 2.5, or if HFR-PWB ratios worsen substantially, the benchmark must be redone with an energy-consistent R-matrix calculation before the central factor-2.5 claim is accepted.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim—AS-DW within ~2.5 of CC/R-matrix and HFR-PWB within an order of magnitude—rests entirely on Table 3. However, the R-matrix reference [18] is built on AS target energies that the paper itself finds to be 5.29% off on average (Table 1); the lowest excited levels are 7.8–7.9% too high (e.g., 4d 2D5/2 at 16001 cm^-1 vs NIST 14836 cm^-1). Effective collision strengths are Maxwellian averages over incident energy (Eq. 2). At T=1000 K, kT ≈ 0.086 eV, while the 4d excitation thresholds are ~1.8 eV. A threshold placed 0.10–0.14 eV too high removes precisely the highest-weight part of the integral, lowering the R-matrix Υ by factors of order 2–5 relative to an energy-consistent target. The favorable ratios in Table 3 (e.g., AS-DW 1–4 is 2.16× the reference) may therefore be partly an artifact of comparing against a reference with shifted thresholds. The paper reports improved level energies but never propagates this known discrepancy into the benchmark. Until this is quantified, the conclusion that these approximations can provide reliable large-scale data is conditional even for Sr II, before any extrapolation to lanthanides and actinides.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper benchmarks two computationally inexpensive approximations—HFR with Plane-Wave Born (HFR-PWB) and AUTOSTRUCTURE with Distorted Waves (AS-DW)—for computing electron-impact excitation effective collision strengths in Sr II. The benchmark case uses the five lowest levels and ten transitions, comparing against published CC/R-matrix calculations [18] at T=1000 K (Table 3). The paper reports that AS-DW agrees with the R-matrix reference within a factor of about 2.5 and that HFR-PWB generally reproduces the correct order of magnitude, with one isolated outlier. It also presents new forbidden M1/E2 radiative rates for the 4d-5s transitions. The stated motivation is to validate cheap methods for large-scale production of collisional data for heavy elements needed in nebular-phase kilonova modeling.","tokens_in":7339,"tokens_out":8888,"duration_ms":102371,"significance":"If the reported accuracy were robust, the paper would provide a practical pathway for filling a genuine gap in atomic data for non-LTE kilonova spectral modeling. The HFR+Fit target energies (0.02% agreement with NIST, Table 1) and the radiative rates (within 4% for HFR+Fit, Table 2) are useful results in their own right. The comparison with R-matrix data is honest in that no parameter is fitted to the reference collision strengths, and the paper explicitly flags the HFR-PWB outlier. However, the central quantitative benchmark is weakened by an energy-threshold inconsistency in the reference data and by the narrow scope of the extrapolation, as detailed below.","major_comments":[{"comment":"The R-matrix reference [18] and the present approximate calculations use different target energy thresholds, and this difference is large enough to affect the T=1000 K benchmark. Table 1 shows that the AS energies used in [18] for the 4d levels are 15686 and 16001 cm^-1 versus NIST values of 14556 and 14836 cm^-1, i.e. offsets of 1130 and 1165 cm^-1. At T=1000 K, kT is about 695 cm^-1, so the Boltzmann factor in Eq. (2) differs by roughly exp(1.6)~5 for the 1-2 and 1-3 excitations. The reference effective collision strengths are therefore suppressed relative to an energy-consistent target, and the ratios in Table 3 mix target-structure errors with the errors of the PWB/DW continuum approximations. For example, AS-DW is 2.16 times the reference for 1-4, but if the reference threshold is corrected upward by the exp(-deltaE/kT) factor the ratio changes substantially. The claim that AS-DW ag","section":"Table 1, Eq. (2), Table 3"},{"comment":"The extrapolation from Sr II to lanthanides and actinides is not supported by the evidence in the manuscript. The benchmark covers only five levels of one alkali-like ion with a single valence electron. Open f-shell lanthanides and actinides have much higher level densities, stronger channel coupling, and more prominent resonance effects, all of which are neglected in PWB and DW. The sentence in the Conclusions stating that the benchmark 'demonstrates that HFR-PWB and AS-DW approximations can provide reliable large-scale data sets ... for systematic applications to lanthanides, actinides' overstates what a single sample case can show. Either add a representative heavy open-f-shell test case or restrict the conclusion to Sr II and clearly state that transferability to complex ions remains to be validated.","section":"Section 6, Conclusions"},{"comment":"The HFR-PWB outlier 2-3 (4d 2D3/2 - 4d 2D5/2) is a factor of about 143 below the R-matrix value (9.50E-02 versus 1.36E+01). It is not merely a numerical curiosity: this is a forbidden transition within the ground configuration and is likely to be relevant for nebular-phase level populations. The text repeatedly summarizes the HFR-PWB accuracy as 'correct order of magnitude' and calls the failure 'isolated,' but with only ten transitions tested, one failure is a 10% failure rate on the most physically important class of transitions. The conclusion that HFR-PWB can provide reliable large-scale data should be qualified, and the manuscript should either explain the origin of this discrepancy or soften the recommendation.","section":"Table 3, Section 5"}],"minor_comments":[{"comment":"The integration variable in the definition of the effective collision strength is written as epsilon_j. The standard definition integrates over the incident electron energy; please clarify the notation and identify which energy the subscript j refers to.","section":"Eq. (2)"},{"comment":"Please state explicitly whether the collision calculations use the HFR or HFR+Fit target energies, and whether the AS calculations include the TEC. Table 1 gives both HFR and HFR+Fit, so the reader cannot tell which target is used in Table 3 without inferring from context.","section":"Section 5"},{"comment":"There is a typo: 'metatable' should be 'metastable', and 'in comparison the those' should be 'in comparison with those'.","section":"Table 2 caption"},{"comment":"The statement that 'results from [18] show a difference of 5.29% with NIST values on average' is not directly supported by the five values in Table 1, which give an average absolute deviation of about 4.2%. Please specify the set of levels over which the 5.29% average is computed.","section":"Section 3, Table 1"},{"comment":"The empirical correction factor (X+3)/(1+X) is introduced without discussion of its range of validity or the data from which it was derived. A sentence with a reference or a qualitative explanation would help the reader assess whether its use is appropriate for low-temperature effective collision strengths.","section":"Eq. (3)"},{"comment":"No uncertainties or convergence checks are reported for the collision strengths in Table 3. Since the central claim is a factor-level accuracy statement, the paper should at least discuss numerical uncertainties of the present calculations and of the R-matrix reference.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and addresses a real need in kilonova atomic data. The authors are transparent about the limitations of the approximate methods. The main issue is technical: the benchmark reference has threshold energies that differ from NIST by ~0.1 eV, which at 1000 K can easily shift effective collision strengths by factors of several. This is a fixable issue in the sense that the paper can re-analyze with corrected threshold factors or add a second benchmark. The extrapolation to heavy open-f-shell ions is a separate concern that should be addressed by more cautious wording or an additional test case. I see no indication of circular fitting to the reference data."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Plain English: this is a useful benchmark paper with a notable unaddressed problem. The authors compute new effective collision strengths for Sr II using HFR-PWB and AS-DW, compare against R-matrix data from another group, and provide corrected forbidden-line radiative rates. The radiative rates are a genuine improvement (HFR+Fit within 4% of NIST), and the benchmark clearly shows both approximations beat the Axelrod formula for forbidden transitions, which is the practical takeaway for kilonova modelers.\n\nThe central accuracy claim—AS-DW within a factor of 2.5, HFR-PWB within an order of magnitude—is presented as if Table 3 settles it. It doesn't. The R-matrix reference [18] uses AS target energies that are 5.29% off on average; the 4d 2D levels are about 8% (roughly 0.12–0.14 eV) too high. At T=1000 K, kT is 0.086 eV, and the effective collision strength is a Maxwellian average whose low-energy region is exponentially sensitive to where the threshold sits. A threshold that high removes the most heavily weighted part of the integral, so the reference Υ are likely suppressed relative to an energy-consistent calculation. The authors report their own energy levels are far closer to NIST, but they never use that information to correct or even quantify the effect on the benchmark. Until that is done, the factor-2.5 statement is conditional even for this one ion.\n\nThe other soft spot is scope: five levels of one alkali-like ion says little about lanthanides and actinides with open f shells, dense level mixing, and strong resonances. The extrapolation in the Conclusions is more hopeful than supported. Neither issue is fatal; both are addressable with a threshold-adjusted comparison and a second test case.\n\nWho this is for: atomic data producers and people building nebular-phase kilonova models. It deserves peer review—the data table and the radiative rates are useful—but the referee should send it back for a quantitative treatment of the reference-energy problem. I'd take the numbers with a grain of salt until then.","headline":"Useful benchmark of cheap collisional approximations for Sr II, but the R-matrix reference's 5–8% level-energy error is never propagated into the central accuracy claim, so the factor-2.5 headline is not yet established.","tokens_in":7802,"tokens_out":2634,"would_cite":false,"duration_ms":29126,"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 benchmarks two low-cost approximations for electron-impact excitation in Sr II against close-coupling R-matrix data, finding they reproduce the correct order of magnitude of effective collision strengths, with distorted waves with","keywords":["kilonova","electron-impact excitation","collision strengths","Sr II","plane-wave Born","distorted waves","forbidden transitions","nebular phase"],"falsifier":"Compute HFR-PWB and AS-DW effective collision strengths for a lanthanide ion such as Ce II or Sm II for its five lowest levels at 1000 K and compare with an R-matrix close-coupling calculation; if a substantial fraction of transitions deviate by more than an order of magnitude from the R-matrix values, the paper's central claim of reliable large-scale applicability to heavy elements would be falsified for those elements.","tokens_in":6858,"feed_emoji":"💥","tokens_out":2733,"duration_ms":32102,"temperature":0.7,"pith_summary":"The paper tests whether two computationally cheap approximations, the plane-wave Born (PWB) method implemented in the Hartree-Fock code and the distorted-waves (DW) method implemented in AUTOSTRUCTURE, can replace much costlier close-coupling R-matrix calculations for electron-impact excitation in Sr II. Benchmarked against published R-matrix effective collision strengths for the five lowest energy levels, both approximations give the right order of magnitude at the kilonova-relevant temperature of 1000 K, and the DW results agree within a factor of about 2.5. This matters because nebular-phase kilonova modeling needs collisional atomic data for dozens of heavy elements, for which complete R-matrix data are not available. The paper argues that these fast methods can supply large-scale datasets that are far better than the crude empirical formulas currently used, especially for forbidden transitions.","feed_headline":"Cheap collision codes hit factor 2.5 of kilonova data","feed_subtitle":"Sr II benchmark shows plane-wave Born and distorted waves can supply heavy-element atomic rates for nebular-phase spectra.","key_machinery":"The central objects are the collision strength and its Maxwellian-averaged effective collision strength, computed with two continuum-electron models: the Plane-Wave Born approximation, which neglects the interaction between the continuum electron and the target potential (with an empirical correction in the Cowan HFR code), and the Distorted-Waves approximation, which includes a static scattering potential but neglects channel coupling and resonances. These approximate rates are benchmarked against CC/R-matrix results at 1000 K for the five lowest levels of Sr II, the comparison being the mechanism that supports the claim.","core_discovery":"Using Sr II as a sample case, the authors compute effective electron-impact collision strengths for the ten transitions among the five lowest energy levels, at temperatures below 10,000 K, with two approximate treatments of the continuum electron: plane-wave Born in the pseudo-relativistic Hartree-Fock method, and distorted waves in AUTOSTRUCTURE. Comparing to the CC/R-matrix calculations of Mulholland et al. [18], they find that the DW results agree within a factor of 2.5 or less, and even the PWB results capture the correct order of magnitude in all but one isolated transition. They also compute radiative rates for the two forbidden M1/E2 transitions from the 4d metastable levels, obtainin","pith_inferences":["The benchmark result may not transfer directly to lanthanides and actinides, whose open f-shells produce dense level structures and strong resonances; the paper provides no test in such ions, so the extrapolation is an unproven but testable expectation.","The isolated PWB failure for the 2–3 forbidden transition suggests that for some low-energy transitions the plane-wave treatment may be unreliable even in simple ions, so a safety margin or a hybrid approach (PWB for allowed lines, DW for forbidden lines) might be prudent.","The same benchmarking protocol could be applied to a heavier ion with many low-lying levels, such as Ce II or Sm II, to map where the factor-of-2.5 agreement breaks down.","The effective collision strength comparisons at 1000 K are a single temperature point; extending the benchmark over a wider temperature range would clarify whether the approximations degrade at higher energies where resonances differ."],"forward_implications":["Large-scale collisional data for a wide range of heavy ions can be produced at modest computational cost, replacing empirical formulas that can be off by several orders of magnitude for forbidden transitions.","The existing HFR atomic targets for elements from Z=20 to Z=103 can be directly reused to compute PWB collision strengths without constructing new models from scratch.","Nebular-phase non-LTE spectral models of kilonovae can incorporate these approximate collision rates, potentially improving line identifications and abundance estimates.","For applications where a factor of 2.5 in collision strength is acceptable, AS-DW offers a reliable compromise between accuracy and completeness.","The accurate HFR radiative rates for the Sr II forbidden lines support their use in kilonova spectral analysis and in atomic databases."],"supporting_citations":[{"why":"Supplies the CC/R-matrix reference effective collision strengths against which the PWB and DW results are benchmarked.","marker":"[18]"},{"why":"Provides the HFR method and the plane-wave Born approximation with its empirical correction as implemented in Cowan's code.","marker":"[15]"},{"why":"Provides the AUTOSTRUCTURE code and the distorted-waves method used for the second set of collision strengths.","marker":"[16]"},{"why":"Defines the close-coupling R-matrix method as the standard that the approximate methods are compared against.","marker":"[17]"},{"why":"Establishes the existing HFR targets for elements Z=20 to Z=103, enabling large-scale PWB calculations.","marker":"[11]"},{"why":"Documents that the Axelrod formula underestimates collision strengths, motivating the need for better approximations.","marker":"[28]"},{"why":"Provides the van Regemorter formula used as a comparison for allowed transitions.","marker":"[26]"},{"why":"Provides the Axelrod formula used as a comparison for forbidden transitions.","marker":"[27]"}],"fun_headline_variants":["Sr II benchmark: DW and PWB collision strengths within 2.5x","Approximate electron-impact codes pass Sr II kilonova test","Kilonova atomic rates: cheap methods within factor 2.5","PWB and DW rates benchmarked for Sr II, kilonova modeling","Sr II electron excitation: approximations match R-matrix closely"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The conclusion that these cheap approximations work well enough is drawn from one alkali-like ion with a single valence electron and five low-lying levels, and is assumed to extend to lanthanides and actinides with open f-shell structures, dense level mixing, and strong resonances.","fun_headline_variants_meta":{"raw":{"variants":["Sr II benchmark: DW and PWB collision strengths within 2.5x","Approximate electron-impact codes pass Sr II kilonova test","Kilonova atomic rates: cheap methods within factor 2.5","PWB and DW rates benchmarked for Sr II, kilonova modeling","Sr II electron excitation: approximations match R-matrix closely"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000694,"raw_usage":{"total_tokens":2998,"prompt_tokens":786,"completion_tokens":2212,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":530,"completion_tokens_details":{"reasoning_tokens":2118}},"tokens_in":530,"tokens_out":2212,"duration_ms":15277,"temperature":1.0,"reasoning_tokens":2118,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T21:49:45.861508+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute HFR-PWB and AS-DW effective collision strengths for a lanthanide ion such as Ce II or Sm II for its five lowest levels at 1000 K and compare with an R-matrix close-coupling calculation; if a substantial fraction of transitions deviate by more than an order of magnitude from the R-matrix values, the paper's central claim of reliable large-scale applicability to heavy elements would be falsified for those elements.","supporting_citations":[{"cited_title":"We find that our AS transition rates underestimate the NIST values by about 10%, while the results from [18] overestimate them by 10%","cited_arxiv_id":null,"evidence_quote":"Supplies the CC/R-matrix reference effective collision strengths against which the PWB and DW results are benchmarked."},{"cited_title":"Pognan, A","cited_arxiv_id":null,"evidence_quote":"Provides the HFR method and the plane-wave Born approximation with its empirical correction as implemented in Cowan's code."},{"cited_title":"Badnell, CPC182, 1528 (2011)","cited_arxiv_id":null,"evidence_quote":"Provides the AUTOSTRUCTURE code and the distorted-waves method used for the second set of collision strengths."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the close-coupling R-matrix method as the standard that the approximate methods are compared against."},{"cited_title":"Deprince, G","cited_arxiv_id":null,"evidence_quote":"Establishes the existing HFR targets for elements Z=20 to Z=103, enabling large-scale PWB calculations."},{"cited_title":"Pognan, J","cited_arxiv_id":null,"evidence_quote":"Documents that the Axelrod formula underestimates collision strengths, motivating the need for better approximations."},{"cited_title":"Letchumanan, M","cited_arxiv_id":null,"evidence_quote":"Provides the van Regemorter formula used as a comparison for allowed transitions."},{"cited_title":"Bi ´emont, J","cited_arxiv_id":null,"evidence_quote":"Provides the Axelrod formula used as a comparison for forbidden transitions."}],"review_version":1}