{"id":"49af4e62-7ca4-4ca0-a0bd-93c0bf0188bd","arxiv_id":"2506.22029","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Collision-induced absorption spectra of noble gas pairs are insensitive to the ab initio method and basis set, and helium-neon shows a predicted double-peak structure from a short-range dipole sign change.","lead":"This paper tests how much the quality of quantum chemistry calculations matters for predicting collision-induced absorption spectra of noble gas pairs. It finds that even rough calculations give spectra within about 10 percent at room temperature, and predicts a new double-peak feature for helium-neon.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Classical F(R) integrand in Eq. (4) bounds only integrated intensity; the paper uses it to claim vdW dipoles 'never contribute' despite excluding bound-bound transitions, so the frequency-resolved 'never' claim is unsupported.","rationale":"The reader's weakest assumption is the same one I would flag: the classical F(R) criterion is used to draw a frequency-resolved and bound-state-inclusive conclusion. The paper has genuine strengths: the basis-set and coupled-cluster convergence trends are systematic, the Ar-He and Ar-Ne comparisons with experiment are positive controls, and the double-exponential dipole analysis is a useful constructive explanation of the Ne-He feature. The problem is that the paper states 'never' and 'complete insensitivity' from a calculation that (i) uses a classical integrated-intensity weight, (ii) explicitly leaves out bound-bound transitions, and (iii) does not directly test truncating the dipole at 8 a0 in a full quantum line-shape calculation. The proposed truncation and bound-state check would settle whether this concern actually lands. If the check passes, the central qualitative claims survive; if it fails, the conclusions need to be weakened to 'insensitive for the integrated and free-free/bound-free spectrum' rather than 'never.' I therefore keep the reader's CONDITIONAL verdict unchanged.","tokens_in":13245,"tokens_out":8933,"duration_ms":109062,"concrete_test":"Recompute the Ne-He and Ar-Ne absorption coefficient at 77 K (and 10 K if feasible) including all bound-bound, bound-free, and free-free transitions using the published CCSD(T)/CBS surfaces, and repeat the calculation with the dipole moment set to zero for R > 8 a0. If the low-frequency spectrum (0-100 cm-1) or the peak intensity changes by more than about 5%, the claim that van der Waals induced dipoles 'never contribute' is falsified; if the spectra are unchanged, the concern is settled.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the inference, made in Sec. IV and restated in the Abstract and Conclusions, that because the classical phase-space integrand F(R) = 4πR² exp(-V/kBT) D²(R) in Eq. (4) is negligible for R > 8 a0, the long-range van der Waals induced dipole 'never contributes' to the absorption spectrum. Eq. (4) is a classical sum rule for the frequency-integrated intensity; it says nothing directly about how that intensity is distributed in frequency. A small D(R) in the long-range tail can produce a narrow, low-frequency feature, especially from bound-bound or quasi-bound transitions, whose peak is noticeable even though its integrated area is small. The manuscript explicitly omits bound-bound transitions from the line-shape calculations (Sec. VI) and dismisses the poorly converged near-threshold bound states as 'not expected to contribute significantly because they are long-ranged' (Sec. II), which is asserted rather than demonstrated. Consequently the central 'never contribute' conclusion, and the accompanying 'complete insensitivity' framing, exceeds what the calculation actually shows. The 10% insensitivity estimate is also based on integrated-intensity or peak comparisons for the free-free/bound-free spectrum, so it does not by itself cover the full quantum spectrum.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports a systematic study of how the accuracy of collision-induced absorption (CIA) spectra of the noble-gas pairs Ne–He, Ar–He, and Ar–Ne depends on the quality of the underlying ab initio potential energy and induced dipole surfaces. Spectra are computed at the CCSD, CCSD(T), and estimated FCI levels with basis sets up to CBS, and the sensitivity of the spectra to the PES and IDS is quantified. The authors find that the spectra are rather insensitive to the ab initio level, with worst-case integrated-intensity differences around 10% at room temperature and smaller at high temperature, and they report a previously unnoticed double-peak structure in the Ne–He spectrum. They also argue, based on a classical phase-space integrand, that long-range van der Waals induced dipoles never contribute to the spectra. The calculated Ar–He and Ar–Ne spectra agree with experiment within estimated error bars, while the Ne–He 77 K measurement disagrees by about a factor of two.","tokens_in":13511,"tokens_out":3360,"duration_ms":38976,"significance":"If the central claim is correct, the paper provides an important practical message: quantitatively predictive CIA spectra for noble-gas mixtures can be obtained from relatively inexpensive electronic-structure data, and effort invested in very accurate long-range surfaces may be less important than usually assumed. The systematic convergence study is a useful contribution, the calculations are carried out without fitted parameters in the main spectral predictions, and the agreement with experiment for Ar–He and Ar–Ne lends credibility to the computational approach. However, the paper's strongest claim—that the absorption spectrum is 'never' sensitive to van der Waals distances—is not fully supported by the evidence presented, and the error-bar estimate is based on integrated intensities rather than on frequency-resolved spectral features.","major_comments":[{"comment":"The claim that long-range van der Waals induced dipoles 'never contribute' rests on the classical phase-space integrand F(R) of Eq. (4), which bounds only the frequency-integrated intensity. That integrand does not directly constrain how the intensity is distributed in frequency: a small but long-ranged dipole tail can produce a narrow, low-frequency feature whose peak is noticeable even though its integrated area is small. Because the line-shape calculations explicitly exclude bound-bound transitions (Section VI) and the near-threshold bound states are dismissed as 'not expected to contribute significantly' without a demonstrated calculation (Section II), the 'never' conclusion is stronger than the evidence supports. I recommend either computing the spectra with the IDS truncated or replaced by its long-range asymptotic form beyond 8 a0 and comparing the full frequency-resolved spectra, or restricting the claim to the integrated intensity of the free-free and bound-free contributions.","section":"Section IV, Eq. (4), Fig. 6; Abstract; Conclusions"},{"comment":"The approximately 10% insensitivity estimate is based on the relative difference in integrated intensities, which is then applied as an overall multiplicative scaling of the spectrum. This implicitly assumes that all uncertainty manifests as an overall intensity scale factor. It does not bound frequency-dependent shape errors, including the position and depth of the Ne–He dip that is a central new prediction. Since the paper's headline quantitative claim is about spectral accuracy, the error analysis should also report a frequency-resolved metric, such as the maximum relative difference per frequency bin between spectra computed with different PES/IDS levels.","section":"Section IV, paragraph beginning 'We then estimate a theoretical error bar'"},{"comment":"The Ne–He 77 K spectrum disagrees with the experimental measurement by roughly a factor of two, outside the paper's own conservative error bars. The manuscript suggests that the experimental data may be unreliable, but it does not quantitatively rule out missing physics in the calculation. In particular, Ne–He has one bound state, and bound-bound contributions are excluded from the line-shape calculation; at 77 K these could contribute at low frequencies. A quantitative estimate of the bound-bound contribution, or a discussion of why it is negligible, is needed before the conclusion that the spectrum is insensitive to ab initio data can be considered fully supported for this system.","section":"Section VI, Fig. 8(e)"}],"minor_comments":[{"comment":"The title has a missing space: 'insensitivity toab initio data' should be 'insensitivity to ab initio data'. In addition, the Table III and Table IV captions contain the typo 'basis stes' for 'basis sets'.","section":"Title and captions"},{"comment":"The text states that 'the change in IDS caused a difference of around 10% at the frequency with maximum intensity,' while Table I reports relative differences in integrated intensity. Please clarify which metric is used in each comparison so the reader can properly interpret the convergence tables.","section":"Section IV, text near Fig. 2 and Table I"},{"comment":"The notation for the soft-sphere approximation uses |kBT, 0> and |kBT + ħω, 0> as if kBT were a wavefunction label. It would be clearer to define the collision energy explicitly, e.g., E = kBT, and write |E, 0> and |E + ħω, 0>.","section":"Eq. (5)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a solid convergence study with a potentially useful practical message, but the 'never contributes' claim and the 10% accuracy statement are currently overgeneralized relative to the evidence. The authors should be able to address the major comments with additional calculations or a more careful wording of the claims. The paper fits the scope of the journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know before you read this one. First, the core quantitative result—CIA spectra for Ne–He, Ar–He, and Ar–Ne are insensitive to the ab initio method and basis set at the 10% level near room temperature, with better convergence at higher T—is new and, as far as I can tell, well supported by the convergence tests. Second, the paper's grander phrasing ('complete insensitivity,' 'never contribute') overstates what the calculation actually shows. The authors know their line shapes; they need to rein in the conclusions.\n\nThe study does several things well. The electronic structure calculations are careful: counterpoise correction, midbond functions, CBS extrapolation, and an FCI estimate by continued-fraction extrapolation. The spectral calculations use a standard, converged quantum treatment, and the Ar–He and Ar–Ne spectra agree with experiment within the conservatively estimated error bars. That is a genuine positive control. The Ne–He double-peak is a real prediction from the ab initio dipole, and the hard-sphere/soft-sphere Fourier analysis explaining it is a nice piece of physical insight. No fitted parameters enter the main spectral predictions, which is worth emphasizing.\n\nThe soft spots are in the interpretation, not the numerics. The 'long-range van der Waals dipoles never contribute' claim rests on the classical integrand F(R) in Eq. (4), a sum rule for integrated intensity. That tells you where the integrated intensity comes from, but not how it is distributed in frequency. A small long-range dipole tail can produce narrow low-frequency features, especially from bound–bound transitions, and the paper explicitly excludes bound–bound contributions from the line shapes. At 77 K, where bound states are more relevant, the conclusion is not safe without a direct truncation test or an estimate of the bound–bound spectrum. The paper's own Fig. 7 shows bound-free contributions becoming significant at 10 K, so the low-T regime is not fully covered. 'Never' should be softened to 'negligible for the free-free/bound-free continuum in the temperature range studied.'\n\nAlso, 'complete insensitivity' is too strong: the numbers themselves show up to 43% uncertainty at 77 K in the worst case. That is good predictive power, not insensitivity. The Ne–He double-peak prediction conflicts with the only available low-T measurement, which is about twice as intense as the calculation; the authors note the data stop at the predicted minimum, but the disagreement remains unresolved. A 295 K measurement would be the clean test.\n\nMinor point: the double-exponential model is fitted to the same dipole that produces the spectrum, so it is illustrative rather than an independent validation. The citation pattern is fine—relevant prior work on CIA and long-range dipoles is cited, and the literature claiming sensitivity is explicitly engaged.\n\nWho this is for: molecular spectroscopists and anyone building CIA opacity databases. The sensitivity analysis is worth having, and the paper deserves a serious referee. It needs revision to align the claims with the evidence, but the core contribution is solid.","headline":"A genuinely useful sensitivity analysis of CIA spectra to ab initio quality, but the 'never contribute' long-range claim and the 'complete insensitivity' framing outrun the evidence.","tokens_in":14019,"tokens_out":2142,"would_cite":true,"duration_ms":23194,"reading_group":"yes","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 claims that collision-induced absorption spectra of noble-gas pairs are largely insensitive to the quality of ab initio potential and dipole surfaces, with even the cheapest calculations accurate to about 10% at room temperature…","keywords":["collision-induced absorption","noble gas dimers","induced dipole surface","ab initio sensitivity","van der Waals interactions","spectral line shapes","He-Ne double-peak structure","high-temperature opacities"],"falsifier":"Recompute the full quantum spectrum including bound-bound and bound-free transitions and set $D(R)=0$ for $R>8\\,a_0$; if the spectrum changes measurably, then long-range van der Waals dipoles do contribute, falsifying the central claim. A complementary experimental check is to look for the predicted dip in the room-temperature He–Ne spectrum between 200 and 600 cm−1.","tokens_in":13045,"feed_emoji":"⚛️","tokens_out":8207,"duration_ms":79954,"temperature":0.7,"pith_summary":"This paper investigates whether the accuracy of collision-induced absorption (CIA) spectra of noble-gas pairs is limited by the quality of the underlying ab initio potential and dipole surfaces. For He–Ne, Ar–He, and Ar–Ne, the authors compute spectra from several levels of electronic-structure theory and find that the spectra change by only about 10% at room temperature when the cheapest surfaces are used, and by less at high temperature. The reason is that the classical integrand $F(R)=4\\pi R^2\\exp[-V(R)/k_\\mathrm{B}T]D^2(R)$ is sharply peaked at short distances, so the spectrum never probes the long-range van der Waals region beyond about 8 $a_0$. The paper also reports that the He–Ne spectrum has a dip or double-peak structure that follows from a zero crossing of the induced dipole at $R\\approx 4.2\\,a_0$, and that simple single-exponential dipole models cannot produce this feature. If correct, this means quantitative CIA modeling for these systems does not require expensive high-level electronic-structure data.","feed_headline":"Cheap ab initio data keeps CIA spectra good to 10%","feed_subtitle":"Even low-level potential and dipole surfaces give ~10% collision-induced absorption at room temperature, and better when hot.","key_machinery":"The machinery is the induced dipole surface $D(R)$ together with the classical phase-space integrand $F(R)=4\\pi R^2\\exp[-V(R)/k_\\mathrm{B}T]D^2(R)$ (Eq. 4), which identifies the internuclear distances that actually contribute to the integrated absorption at a given temperature. The line shape itself is computed from quantum scattering wavefunctions (Eqs. 1–3), and an approximate hard-sphere/Fourier model (Eq. 7) is used to attribute spectral features to the shape of the dipole function. The zero crossing of the He–Ne dipole at $R\\approx 4.2\\,a_0$ plays the pivotal role: because the integrand samples this region at room temperature, the Fourier transform of the dipole produces a dip and double-peak structure that a single exponential cannot capture.","core_discovery":"The central claim is that for the three noble-gas heterodimers, CIA spectra are largely insensitive to the chosen ab initio method and basis set: replacing the recommended CCSD(T)/CBS surfaces with the much cheaper CCSD(T)/AVTZ surfaces changes the integrated intensity by up to about 10% at 295 K, with the dipole surface responsible for most of that variation, and the differences shrink as temperature rises to 2000 K. The paper explains this through the classical phase-space integrand, which shows that the integrated intensity is dominated by a narrow window of internuclear separations — a few bohr wide, around 3–5 $a_0$ — set by the trade-off between the decay of the squared dipole and the Boltzmann suppression of the repulsive wall. Consequently, the long-range region, where the induced dipole is governed by van der Waals interactions at $R\\gtrsim 8\\,a_0$, never contributes to the spectra at any temperature studied. For He–Ne, the dipole function changes sign near $R=4.2\\,a_0$, producing a previously unreported double-peak absorption feature that is reproduced by a double-exponential dipole model but not by a single-exponential one; a hard-sphere/Fourier analysis shows the feature is a direct consequence of the dipole's non-monotonic shape. This is taken as evidence that long-range van der Waals data are irrelevant for these spectra, and that short-range interactions are the controlling factor for astronomical CIA modeling.","pith_inferences":["The 'never contribute' conclusion may not carry over to other collisional pairs (e.g., H2–H2 or H2–rare-gas systems) where lighter masses and deeper potentials let bound states sample much larger distances; a similar F(R) diagnostic should be checked case by case.","The insensitivity to the long-range dipole suggests that the main source of error in CIA theory is the functional form of the short-range dipole, so experimental tests of predicted features (like the He–Ne dip) would provide a sharper benchmark than global intensity comparisons.","The same F(R) analysis could be used in reverse: given a target temperature and desired accuracy, one could determine the smallest ab initio grid and the largest R that needs to be computed, saving cost for high-throughput opacity databases.","If the double-peak in He–Ne is confirmed, it would indicate that current spectral databases based on older single-exponential models may have systematic shape errors in the 200–600 cm−1 region for light gas pairs."],"forward_implications":["CIA opacities for He–Ne, Ar–He, and Ar–Ne can be computed with low-level ab initio surfaces and still be roughly 10% accurate at room temperature, and better at higher T, so astronomical models do not have to wait for CBS-quality surfaces.","The long-range van der Waals region of the induced dipole can be excluded from future electronic-structure calculations for these systems, since it never contributes to the spectrum.","Short-range dipole models need to be flexible enough to capture zero crossings and other non-monotonic features; single-exponential models will fail for systems like He–Ne.","As temperature rises, the spectra probe shorter internuclear distances on the repulsive wall, so high-temperature CIA modeling depends on accurate short-range repulsion.","For Ar–He and Ar–Ne, the computed spectra match experiment within the estimated uncertainty, so these calculations can directly supply absorption coefficients for atmosphere models over an expanded temperature range."],"supporting_citations":[{"why":"Supplies the CIA spectral-density formulas (Eqs. 1–2) and the classical phase-space integrated-intensity relation (Eq. 4) used throughout.","marker":"[43]"},{"why":"Provides the classical statistical integrated intensity that the paper uses to validate the quantum spectra and define the error estimates.","marker":"[46]"},{"why":"Supplies the renormalized Numerov propagation method used to compute the scattering wavefunctions that enter the line-shape calculation.","marker":"[44]"},{"why":"RKHS interpolation/extrapolation turns the ab initio grid into the dense potential and dipole surfaces used by the CIA calculations.","marker":"[37]"},{"why":"Furnishes the basis-set extrapolation formula used to obtain the CBS limits of the interaction energies.","marker":"[36]"},{"why":"Counterpoise correction method used for computing interaction energies, affecting the PES quality.","marker":"[35]"},{"why":"Provides reference ab initio potential/dipole data and bound-state counts against which the paper's surfaces are checked.","marker":"[38]"},{"why":"Earlier computation of bound states for the three pairs, used to discuss near-threshold states excluded from the line-shape calculation.","marker":"[39]"},{"why":"Experimental Ne–He CIA spectrum at 77 K used as the comparison that lies outside the calculated uncertainty and that motivates the double-peak prediction.","marker":"[49]"}],"fun_headline_variants":["CIA spectra shrug off ab initio quality: 10% at room temp","Short-range dipoles drive noble-gas CIA; long-range never matter","He-Ne reveals double-peak CIA feature, defies single-exponential models","Low-cost ab initio gives 10% CIA accuracy, improves with temperature"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim that long-range van der Waals dipoles 'never contribute' rests on the assumption that the classical phase-space integrand $F(R)$ faithfully represents the internuclear distances sampled by the quantum spectrum, including the bound states and near-threshold states that are excluded from the line-shape calculations; if those quantum states sample distances beyond 8 $a_0$, the conclusion would fail.","fun_headline_variants_meta":{"raw":{"variants":["CIA spectra shrug off ab initio quality: 10% at room temp","Short-range dipoles drive noble-gas CIA; long-range never matter","He-Ne reveals double-peak CIA feature, defies single-exponential models","Low-cost ab initio gives 10% CIA accuracy, improves with temperature"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00027,"raw_usage":{"total_tokens":1657,"prompt_tokens":1010,"completion_tokens":647,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":626,"completion_tokens_details":{"reasoning_tokens":564}},"tokens_in":626,"tokens_out":647,"duration_ms":6718,"temperature":1.0,"reasoning_tokens":564,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T22:13:19.548964+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the full quantum spectrum including bound-bound and bound-free transitions and set $D(R)=0$ for $R>8\\,a_0$; if the spectrum changes measurably, then long-range van der Waals dipoles do contribute, falsifying the central claim. A complementary experimental check is to look for the predicted dip in the room-temperature He–Ne spectrum between 200 and 600 cm−1.","supporting_citations":[{"cited_title":"Collision-induced absorption in gases,","cited_arxiv_id":null,"evidence_quote":"Supplies the CIA spectral-density formulas (Eqs. 1–2) and the classical phase-space integrated-intensity relation (Eq. 4) used throughout."},{"cited_title":"Buryak, S","cited_arxiv_id":null,"evidence_quote":"Provides the classical statistical integrated intensity that the paper uses to validate the quantum spectra and define the error estimates."},{"cited_title":"Karman, A","cited_arxiv_id":null,"evidence_quote":"Supplies the renormalized Numerov propagation method used to compute the scattering wavefunctions that enter the line-shape calculation."},{"cited_title":"Ho and H","cited_arxiv_id":null,"evidence_quote":"RKHS interpolation/extrapolation turns the ab initio grid into the dense potential and dipole surfaces used by the CIA calculations."},{"cited_title":"Halkier, T","cited_arxiv_id":null,"evidence_quote":"Furnishes the basis-set extrapolation formula used to obtain the CBS limits of the interaction energies."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Counterpoise correction method used for computing interaction energies, affecting the PES quality."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides reference ab initio potential/dipole data and bound-state counts against which the paper's surfaces are checked."},{"cited_title":"L´ opez Cacheiro, B","cited_arxiv_id":null,"evidence_quote":"Earlier computation of bound states for the three pairs, used to discuss near-threshold states excluded from the line-shape calculation."},{"cited_title":"Marteau, J","cited_arxiv_id":null,"evidence_quote":"Experimental Ne–He CIA spectrum at 77 K used as the comparison that lies outside the calculated uncertainty and that motivates the double-peak prediction."}],"review_version":1}