REVIEW 3 major objections 4 minor 40 references
Oxygen Isotope Ratios in Hydrogen-Deficient Carbon Stars: A Correlation with Effective Temperature and Implications for White Dwarf Merger Outcomes
T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read This paper claims that hydrogen-deficient carbon stars come in two cleanly separated subclasses: dustless ones with 16O/18O below 1 and dusty RCB stars with ratios above 4, plus a newly seen trend of lower ratios in hotter stars that…
desk verdict Solid, reproducible isotope dichotomy; the Teff-trend headline is confounded with class and needs tempering. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the oxygen isotope ratio 16O/18O, extracted by fitting synthetic spectra to the relative strengths of 12C16O versus 12C18O vibrational bandheads and lines in the K band (2.25–2.48 μm, R≈75,000). The measurements sit on a grid of hydrogen-deficient model atmospheres, with a semi-automated chi-squared fitting routine that treats the isotope ratio as a free parameter. A secondary mechanism is the correlation between nitrogen and oxygen-18 abundances, which the paper attributes to alpha-capture nucleosynthesis during partial helium burning; because both isotopes are measured on the same spectra, the correlation is relatively insensitive to model uncertainties.
What would settle it
Measure the oxygen isotope ratio for an HdC star with an independently determined effective temperature in the 5250–5500 K gap between the two classes (for example, using a Gaia parallax and SED fit), or check whether the adopted temperatures in Table 3 are biased by comparing with such independent values. A dLHdC star cooler than 5250 K with 16O/18O > 1, or an RCB star hotter than 5500 K with 16O/18O < 4, would directly contradict the dichotomy and the temperature trend.
Extended reading notes
Core claim
On the paper's own terms, the discovery is that RCB and dLHdC stars—two spectroscopically similar subclasses of hydrogen-deficient carbon stars—are cleanly separated by their oxygen isotope ratios when measured at high resolution: all six dLHdC stars have 16O/18O < 1, while all six RCB stars have 16O/18O > 4. The paper further claims a first-ever monotonic trend of decreasing 16O/18O with increasing effective temperature across the whole HdC class, a trend it says is consistent in direction with theoretical models of white-dwarf merger remnants, though those models overpredict the dLHdC values by two orders of magnitude. It also reports a linear correlation between nitrogen and oxygen-18 abundances, interpreted as a fixed ~8% conversion of 14N to 18O via the alpha-capture chain 14N(α,γ)18F(β+)18O. The authors take these results to affirm that differences in the total mass and mass ratio of the progenitor white-dwarf binary determine whether a merger becomes an RCB or a dLHdC star.
Load-bearing premise
The claimed trend with effective temperature rests on adopted temperature values, and the two star classes do not overlap in temperature (RCB stars are all cooler than 5250 K, dLHdC stars all hotter than 5500 K), so the 'trend with temperature' could amount to nothing more than the already known class difference.
Editorial extensions
If this is right
- The RCB versus dLHdC dichotomy in oxygen isotope ratios (16O/18O < 1 versus > 4) is confirmed with the largest high-resolution sample to date and extends to newly discovered dLHdC stars.
- The trend of lower 16O/18O at higher effective temperature, if real, gives an observational diagnostic connecting a star's surface composition to the mass ratio and total mass of its white-dwarf progenitor binary.
- Existing white-dwarf merger models reproduce the direction of the trend but not its magnitude for dLHdC stars, indicating that additional physics (e.g., helium-shell burning temperature, hydrogen content, or convection) must be added.
- The measured dust-dilution correction implies RCB oxygen isotope ratios could be underestimated by roughly a factor of two, further widening the gap between the two classes.
- The roughly 8% fixed conversion of nitrogen to oxygen-18 via alpha capture gives a nucleosynthetic benchmark that merger models will need to reproduce.
Reading between the lines
- If the temperature trend is set by the mass ratio of the merging white dwarfs, a single measured 16O/18O ratio plus temperature could one day be inverted to estimate the unseen binary parameters of an individual HdC star.
- Because the two classes occupy non-overlapping temperature ranges, the new 'trend with temperature' is observationally indistinguishable from the old class dichotomy with current data; a future sample spanning the 5250–5500 K gap would settle whether the relation is continuous.
- The model overprediction by two orders of magnitude for dLHdC stars is a sharper test than the RCB comparison; finding the physics that lowers the ratio (e.g., lower helium-burning temperatures) may also predict other observable signatures, such as the C/O ratio or luminosity distribution.
- The nitrogen–oxygen-18 correlation, if it holds in a larger sample, could serve as a chemical clock: the fraction of nitrogen converted could constrain the duration and temperature of the partial helium-burning phase in a merger remnant.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper presents high-resolution (R≈75000) K-band spectra of six RCB and six dLHdC stars and derives oxygen isotope ratios and CNO abundances using a semi-automated spectral-fitting routine with MARCS model atmospheres. The authors report a clear dichotomy: all dLHdC stars have 16O/18O<1 while all RCB stars have 16O/18O>4. They further claim, for the first time, a trend of decreasing 16O/18O with increasing effective temperature, which they compare to WD merger models, and a linear correlation between 18O and 14N abundances suggesting a fixed conversion fraction via alpha-capture.
Significance. If the dichotomy holds, this is a valuable observational result: it triples the sample of dLHdC stars with high-resolution isotope ratios, validates the method against two previously measured stars, and strengthens the case that RCB and dLHdC stars form from distinct WD merger configurations. The paper also ships public spectra and fitting software, and includes a careful discussion of dust-dilution systematics. These are concrete strengths. However, the paper's headline new claim—a continuous Teff–16O/18O trend—is not established by the current sample because stellar class and Teff are perfectly confounded, and the within-class trends are weak or absent. The central dichotomy is robust; the trend claim needs reframing or additional analysis.
major comments (3)
- [Section 4, Figure 5, Table 4] The claimed trend of decreasing 16O/18O with increasing Teff is confounded with class membership. All six RCB stars have Teff between 4500 and 5250 K, while all six dLHdC stars have Teff between 5500 and 6250 K, so Teff and class are perfectly separated. Within each class the trend is not monotonic: for dLHdCs, B42 at 5500 K has 0.32, HD 137613 at 5500 K has 0.52, A223 at 6250 K has 0.60, and B566 at 5750 K has 0.78; for RCBs, the two coolest stars have enormous asymmetric errors. The apparent continuous relation is therefore driven almost entirely by the inter-class gap. The authors acknowledge this in the Conclusion ('this has a potential to bias our results'), but the abstract and Section 4 present the trend as a new finding without the same caveat. Because this is the paper's novel claim, it must be reanalyzed or substantially reframed, for example by fitting the trend separately within each class, regressing out class membership, or explicitly stating that the data only support a class difference rather than a continuous temperature dependence.
- [Table 4 and Section 3.5.2] The cool end of the claimed trend rests on two RCB stars whose 16O/18O values are effectively unconstrained. AO Her is measured at 69 (+228/−35) and WISE J1942+ at 93 (+500/−55), with the upper bound for WISE J1942+ explicitly reported as a limit because the chi-square curve never rises above Δχ2=1. These two points dominate the visual steepening at low Teff in Figure 5, yet their ratios are consistent with values as low as ~34 and ~38, respectively. The text and figure present the best-fit values without adequate emphasis on how weakly these anchor the trend. This should be addressed quantitatively, e.g., by showing the trend with these points removed or with their asymmetric uncertainties propagated into the fitted slope.
- [Section 4, Figure 5 (right)] The claimed linear correlation between A(18O/Fe) and A(14N/Fe) is also potentially a class-separation artifact. The dLHdC stars occupy a high-N/high-O locus (A(N)≈8.7–9.5, A(O)≈8.0–9.0) while the RCB stars occupy a low-N/low-O locus (A(N)≈6.6–7.8, A(O)≈7.2–8.0). A linear fit across the combined sample therefore does not demonstrate that a fixed fraction of 14N is converted to 18O within each class; it may simply trace the same RCB/dLHdC dichotomy. The paper should show the fit within each class or explicitly demonstrate that the slope and intercept are not driven by the class grouping.
minor comments (4)
- [Abstract / Section 5] The abstract states 'we find a trend' while the Conclusion says 'we find a possible correlation'; these should be harmonized to avoid overstating the result.
- [Section 3.5.3, Table 5] The Teff sensitivity check is presented for only one RCB star (ASAS-RCB-21); given that Teff is a fixed adopted parameter for all stars, a similar test for a dLHdC star would strengthen confidence that the derived isotope ratios are not strongly Teff-dependent.
- [Section 1] The text refers to Karambelkar et al. (2021) and Tisserand et al. (2020) for the sources WISE-ToI-222 and WISE-ToI-223; the correspondence between those names and WISE J1818+/WISE J1820+ is only given in a table note and would be clearer in the main text.
- [Throughout] There are several typographic and spacing issues, e.g., 'idlsoftwarespextool' and 'and V arun Bhalerao' in the author list; these should be corrected in the final version.
Circularity Check
No significant circularity: the isotope ratios and abundances are measured from spectra and validated against independent data; the Teff trend is a caveated confound, not a construction.
full rationale
This is fundamentally an observational measurement paper. The 16O/18O ratios and CNO abundances are obtained by chi-square fitting of synthetic MARCS/Turbospectrum spectra to individual 12C16O, 12C18O, CN, and C2 features (Section 3.4), not by applying the theoretical models being tested. The fitting procedure is externally anchored: the ratio measurements for HD 137613 and HD 182040 are consistent with the independent high-resolution results of García-Hernández et al. (2009). The claimed Teff trend is a post-fit correlation between measured ratios and Teff values adopted from empirical color-Teff calibrations and SED fits (Table 3); it is not constructed from the isotope fits. The main statistical caveat, that the RCB and dLHdC samples have disjoint Teff ranges, is acknowledged in the conclusion ('all RCB stars in our sample are cooler than 5250 K, while the dLHdC stars are warmer than 5500 K, and this has a potential to bias our results'). That is a confound in interpretation, not a definitional circularity. The comparison to Crawford et al. (2024) involves coauthors and is self-citational, but it is not load-bearing: the models are independent simulations and are explicitly found to overpredict dLHdC ratios by two orders of magnitude, so the observational result does not reduce to the models. The 8% 14N-to-18O conversion fraction is simply the intercept of the empirical linear fit and is not presented as an independent prediction. No step in the derivation chain equates an output to an input by construction; the only circularity-adjacent feature is the minor self-citation in the interpretive model comparison.
Assumptions & free parameters
free parameters (7)
- 16O/18O ratio (per star) =
0.32 to 93, Table 4
- A(C), A(N), A(O) (per star) =
Table 4 values
- Microturbulence xi (per star) =
6.0-7.0 km/s, Table 3
- Macroturbulence v_mac (RCB stars) =
6-8 km/s, Section 3.4
- Dust shell fraction f_shell =
0.2 (test value)
- Effective temperature Teff (per star) =
4500-6250 K, Table 3
- log g =
1.0 dex (fixed)
assumptions (7)
- domain assumption MARCS hydrogen-deficient model atmospheres with C/He=0.01, A(O)=8.8, and solar metallicity are appropriate for all HdC stars in the sample
- domain assumption LTE spectrum synthesis is adequate for deriving molecular abundances
- domain assumption Adopted effective temperatures from Crawford et al. (2023) color-temperature calibration and SED literature are accurate to the level needed for the Teff trend
- domain assumption HdC stars originate from double-degenerate CO-He white dwarf mergers
- domain assumption Crawford et al. (2024) theoretical WD merger models are predictive enough to compare observed isotope ratio trends
- domain assumption Atomic and molecular line lists used (CO, CN, C2, VALD) are accurate in the K band
- domain assumption The absence of 13C16O bandheads means 12C/13C is high and isotope fits are not contaminated by 13C lines
Cite this review
Pith. "Pith review of Oxygen Isotope Ratios in Hydrogen-Deficient Carbon Stars: A Correlation with Effective Temperature and Implications for White Dwarf Merger Outcomes." pith.science (2026). https://pith.science/paper/YNEHJOYQ
@misc{pith2026241203664,
author = {Pith},
title = {Pith review of: Oxygen Isotope Ratios in Hydrogen-Deficient Carbon Stars: A Correlation with Effective Temperature and Implications for White Dwarf Merger Outcomes},
year = {2026},
howpublished = {\url{https://pith.science/paper/YNEHJOYQ}},
note = {Machine review of arXiv:2412.03664}
}
abstract
Hydrogen-deficient Carbon (HdC) stars are a class of supergiants with anomalous chemical compositions, suggesting that they are remnants of CO-He white dwarf (WD) mergers. This class comprises two spectroscopically similar subclasses - dusty R Coronae Borealis (RCB) and dustless Hydrogen-deficient Carbon (dLHdC) stars. Both subclasses have a stark overabundance of $^{18}\textrm{O}$ in their atmospheres, but spectroscopic differences between them remain poorly studied. We present high-resolution ($R \approx 75000$) K-band spectra of six RCB and six dLHdC stars, including four newly discovered dLHdC stars, making this the largest sample to date. We develop a semi-automated fitting routine to measure $^{16}\textrm{O}/^{18}\textrm{O}$ ratios for this sample, tripling the number of dLHdC stars with oxygen isotope ratios measured from high resolution spectra. All six dLHdC stars have $^{16}\textrm{O}/^{18}\textrm{O}<1$, while the RCB stars have $^{16}\textrm{O}/^{18}\textrm{O}>4$. Additionally, for the first time, we find a trend of decreasing $^{16}\textrm{O}/^{18}\textrm{O}$ ratios with increasing effective temperature for HdC stars, consistent with predictions of theoretical WD merger models. However, we note that current models overpredict the low $^{16}\textrm{O}/^{18}\textrm{O}$ ratios of dLHdC stars by two orders of magnitude. We also measure abundances of C, N, O, Fe, S, Si, Mg, Na, and Ca for these stars. We observe a correlation between the abundances of $^{14}\textrm{N}$ and $^{18}\textrm{O}$ in our sample, suggesting that a fixed fraction of the $^{14}\textrm{N}$ is converted to $^{18}\textrm{O}$ in these stars via $\alpha$-capture. Our results affirm the emerging picture that the mass ratio/total mass of the WD binary determine whether an RCB or dLHdC is formed post-merger.
Figures
Figures from the paper (2 more)
Reference graph
Works this paper leans on
- [1]
-
[2]
Asplund, M., Gustafsson, B., Lambert, D. L., & Rao, N. K. 2000, A&A, 353, 287
work page 2000
-
[3]
A., Eriksson, K., Gustafsson, B., & Nordlund, A
Bell, R. A., Eriksson, K., Gustafsson, B., & Nordlund, A. 1976, A&AS, 23, 37
work page 1976
- [4]
-
[5]
Clayton, G. C. 1996, PASP, 108, 225 —. 2012, JAA VSO, 40, 539
work page 1996
-
[6]
2007, ApJ, 662, 1220 Oxygen Isotope Ratios in HdC Stars15
Asplund, M. 2007, ApJ, 662, 1220 Oxygen Isotope Ratios in HdC Stars15
work page 2007
-
[7]
Clayton, G. C., Herwig, F., Geballe, T. R., et al. 2005, ApJ, 623, L141
work page 2005
-
[8]
Crawford, C. L., Nikultsev, N., Clayton, G. C., et al. 2024, MNRAS, 534, 1018
work page 2024
Show all 40 references
-
[9]
L., Tisserand, P., Clayton, G
Crawford, C. L., Tisserand, P., Clayton, G. C., et al. 2023, MNRAS, 521, 1674
2023
-
[10]
C., Vacca, W
Cushing, M. C., Vacca, W. D., & Rayner, J. T. 2004, PASP, 116, 362
2004
-
[11]
Catchpole, R. M. 1997, MNRAS, 285, 317 Garc ´ ıa-Hern´ andez, D. A., Hinkle, K. H., Lambert, D. L., &
1997
-
[12]
2009, ApJ, 696, 1733 Garc ´ ıa-Hern´ andez, D
Eriksson, K. 2009, ApJ, 696, 1733 Garc ´ ıa-Hern´ andez, D. A., Lambert, D. L., Kameswara Rao, N., Hinkle, K. H., & Eriksson, K. 2010, ApJ, 714, 144 Garc ´ ıa-Hern´ andez, D. A., Rao, N. K., Lambert, D. L., et al. 2023, ApJ, 948, 15
2009
-
[13]
M., Magg, E., Plez, B., et al
Gerber, J. M., Magg, E., Plez, B., et al. 2023, A&A, 669, A43
2023
-
[14]
1994, ApJS, 95, 535
Goorvitch, D. 1994, ApJS, 95, 535
1994
-
[15]
A., Eriksson, K., & Nordlund, A
Gustafsson, B., Bell, R. A., Eriksson, K., & Nordlund, A. 1975, A&A, 42, 407
1975
-
[16]
2008, A&A, 486, 951
Gustafsson, B., Edvardsson, B., Eriksson, K., et al. 2008, A&A, 486, 951
2008
-
[17]
P., Pandey, G., Kamath, D., et al
Hema, B. P., Pandey, G., Kamath, D., et al. 2017, PASP, 129, 104202
2017
-
[18]
P., Pandey, G., & Lambert, D
Hema, B. P., Pandey, G., & Lambert, D. L. 2012, ApJ, 747, 102
2012
-
[19]
S., Karakas, A
Jeffery, C. S., Karakas, A. I., & Saio, H. 2011, MNRAS, 414, 3599 Kameswara Rao, N., Reddy, B. E., & Lambert, D. L. 2004, MNRAS, 355, 855 Kameswara Rao, N., Lambert, D. L., Adams, M. T., et al. 1999, MNRAS, 310, 717
2011
-
[20]
M., Tisserand, P., et al
Karambelkar, V., Kasliwal, M. M., Tisserand, P., et al. 2022, A&A, 667, A84
2022
-
[21]
R., Kasliwal, M
Karambelkar, V. R., Kasliwal, M. M., Tisserand, P., et al. 2021, ApJ, 910, 132 —. 2024, PASP, 136, 084201
2021
-
[22]
Kurucz, R. L. 2010, Robert L. Kurucz on-line database of observed and predicted atomic transitions
2010
-
[23]
L., & Rao, N
Lambert, D. L., & Rao, N. K. 1994, Journal of Astrophysics and Astronomy, 15, 47
1994
-
[24]
1984, A&A, 134, 189
Magain, P. 1984, A&A, 134, 189
1984
-
[25]
2022, A&A, 661, A140
Magg, E., Bergemann, M., Serenelli, A., et al. 2022, A&A, 661, A140
2022
-
[26]
M., et al
Mainzer, A., Bauer, J., Cutri, R. M., et al. 2014, ApJ, 792, 30
2014
-
[27]
2021, ApJ, 911, 103
Munson, B., Chatzopoulos, E., Frank, J., et al. 2021, ApJ, 911, 103
2021
-
[28]
L., Jeffery, C
Pandey, G., Lambert, D. L., Jeffery, C. S., & Rao, N. K. 2006, ApJ, 638, 454
2006
-
[29]
2008, Physica Scripta Volume T, 133, 014003 —
Plez, B. 2008, Physica Scripta Volume T, 133, 014003 —. 2012, Turbospectrum: Code for spectral synthesis, Astrophysics Source Code Library, record ascl:1205.004
2008
-
[30]
2022, PASP, 134, 015002
Rayner, J., Tokunaga, A., Jaffe, D., et al. 2022, PASP, 134, 015002
2022
-
[31]
L., et al
Ryabchikova, T., Piskunov, N., Kurucz, R. L., et al. 2015, Phys. Scr, 90, 054005
2015
-
[32]
2023, MNRAS, 525, 3718
Storm, N., & Bergemann, M. 2023, MNRAS, 525, 3718
2023
-
[33]
N., Zhang, J., et al
Tennyson, J., Yurchenko, S. N., Zhang, J., et al. 2024, Journal of Quantitative Spectroscopy and Radiative Transfer, 326, 109083
2024
-
[34]
2012, A&A, 539, A51
Tisserand, P. 2012, A&A, 539, A51
2012
-
[35]
L., Soon, J., et al
Tisserand, P., Crawford, C. L., Soon, J., et al. 2024, A&A, 684, A131
2024
-
[36]
C., Bessell, M
Tisserand, P., Clayton, G. C., Bessell, M. S., et al. 2020, A&A, 635, A14
2020
-
[37]
L., Clayton, G
Tisserand, P., Crawford, C. L., Clayton, G. C., et al. 2022, A&A, 667, A83
2022
-
[38]
D., Cushing, M
Vacca, W. D., Cushing, M. C., & Rayner, J. T. 2003, PASP, 115, 389
2003
-
[39]
N., Szab´ o, I., Pyatenko, E., & Tennyson, J
Yurchenko, S. N., Szab´ o, I., Pyatenko, E., & Tennyson, J. 2018, MNRAS, 480, 3397
2018
-
[40]
A., et al
Zhang, J., Wang, X., Mazzali, P. A., et al. 2014, ApJ, 797, 5 16Mehla et al. 8.APPENDIX Wavelength (˚A) Lower Energy Level (eV) loggfExcitation Potential (eV) logγBroadening (s −1) 22906.458 9.1718 -0.279 2.5 0.0 3.98×10 8 23062.949 10.3851 -1.842 2.5 3.0 3.98×10 8 23091.400 1...
2014
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.