Pith. sign in

REVIEW 2 major objections 5 minor 33 references

New Weighted Sum of Gray Gases (WSGG) Models for Radiation Calculation in Carbon Capture Simulations: Evaluation and Different Implementation Techniques

T0 review · 2 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Switching coefficient interpolation of the classical air-fuel WSGGM from stepwise to piecewise linear roughly halves its deviation from a spectral-line reference in oxy-fuel dry-recycle radiation calculations.

desk verdict A careful, useful WSGG benchmark for oxy-fuel radiation that deserves a referee, provided the unpublished SLW reference gets disclosed or independently checked. read the letter →

arxiv 2411.18467 v1 pith:RVJOIMKP submitted 2024-11-27 physics.gen-ph

Osama A. Marzouk , E. David Huckaby This is my paper · ORCID
classification physics.gen-ph PACS 44.40.+a
keywords weightedsumofgraygasesWSGGMoxy-fuelcombustioncarboncaptureradiativeheattransferabsorptioncoefficientinterpolationdryrecyclespectralline-basedmodel
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper evaluates four recently developed weighted-sum-of-gray-gases models (WSGGMs) for oxy-fuel combustion and compares them to the classical air-fuel WSGGM that is built into common CFD codes, using a spectral-line-based WSGGM (SLW) as the reference. Its central finding is that in the dry-recycle oxy-fuel regime (10% H2O, 90% CO2), the classical model underpredicts the radiative heat flux and heat source, while the newer oxy-fuel models overpredict them. The paper also shows that the way the model coefficients are interpolated between tabulated H2O/CO2 ratios matters: switching from piecewise-constant to piecewise-linear interpolation raises the classical model's absorption coefficient at low H2O partial pressures and reduces its deviation from the SLW reference by about half in the two oxy-fuel test cases. A sympathetic reader would care because this suggests that a legacy CFD radiation model can be made markedly more accurate for carbon-capture simulations by a simple implementation change, without replacing the model itself.

What carries the argument

The central object is the weighted sum of gray gases model (WSGGM), which represents a real gas mixture as a small number of hypothetical gray gases: total emissivity is a temperature-weighted sum $\varepsilon = \sum_i a_i(T)(1-\exp[-k_i p L])$, with weights $a_i(T)$ as polynomials in $T$, and the mixture absorption coefficient is recovered by the Beer-Lambert relation $k = -\ln(1-\varepsilon)/L$. The paper's operational machinery is the coefficient-interpolation scheme. The tabulated coefficients are keyed to fixed H2O-to-CO2 ratios, expressed as $R = P_w/P_c$ or the bounded ratio $RR = P_w/(P_w+P_c)$; piecewise-constant interpolation picks the nearest tabulated set, while piecewise-linear interpolation linearly interpolates between adjacent tabulated sets (with extrapolation near the endpoints). At low but finite H2O partial pressures, this interpolation choice changes the absorption coefficient substantially, and that difference is what turns the classical model's underprediction into a much smaller underprediction.

What would settle it

Recompute the four box test cases with an independent, published line-by-line or multi-scale spectral solver using the same spatial and angular resolution as the WSGGM runs, and check the dry-recycle case: if the classical air-fuel model no longer underpredicts, or if linear interpolation no longer halves its error, the paper's central claim fails.

Watch

Extended reading notes

Core claim

The paper's claim is that the classical air-fuel WSGGM, when implemented with stepwise coefficient selection, systematically underpredicts radiative transfer in dry-recycle oxy-fuel conditions, and that this is a property of the interpolation step, not of the underlying gas model. In the 2m x 2m x 4m box test, the air-fuel model's RMS deviation from the SLW solution for the radiative flux was 5.13 kW/m2 with stepwise interpolation; with piecewise-linear interpolation this dropped to 2.58 kW/m2, and the mean error changed from -5.10 to -2.31 kW/m2, i.e., the underprediction shrank while remaining negative. The newer oxy-fuel models, in contrast, overpredict the flux and heat source in this regime, with the 4+1 version of one model closest to SLW in the dry-recycle case and another oxy-fuel model having the smallest overall deviation across all four test cases. The paper does not claim one oxy-fuel model is uniformly best; it claims the interpolation method materially changes the ranking and that piecewise-linear interpolation is a consistent improvement for the classical model.

Load-bearing premise

The results stand on the accuracy of the unpublished SLW reference solution; if that spectral benchmark is biased or not converged, the underprediction/overprediction pattern and the model ranking could change.

Editorial extensions

If this is right

  • In dry-recycle oxy-fuel simulations that use the classical air-fuel WSGGM with stepwise coefficients, radiative heat flux and heat source will be underpredicted, biasing predicted temperatures high.
  • Switching to piecewise-linear interpolation of the classical model's coefficients cuts its deviation from the SLW reference by roughly half in the wet- and dry-recycle oxy-fuel cases.
  • The newer oxy-fuel models generally overpredict the flux and source in the dry-recycle case, so they carry the opposite bias.
  • Because the classical model's interpolation sensitivity is largest in the dry-recycle regime, legacy CFD users targeting oxy-fuel carbon capture can improve accuracy by changing only the coefficient lookup, not the radiative solver.
  • Stepwise interpolation introduces discontinuities in emissivity and absorption coefficient at interval boundaries; linear interpolation reduces both discontinuities and model spread.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: since the paper's profiles show larger model spread at 10 m pathlength than at 1 m, the interpolation effect is expected to be more pronounced in full-scale boiler furnaces than in lab-scale or pilot-scale boxes.
  • Editorial inference: the ranking among the new oxy-fuel models is provisional because the SLW benchmark is unpublished, uses a coarser spatial grid and different angular quadrature than the WSGGM runs, and was produced by the developer of one of the ranked models; an independent benchmark would be needed to confirm the numerical ordering.
  • Editorial inference: a natural extension is to test whether linear interpolation also improves the classical model's predictions of the radiative source term in non-homogeneous gas mixtures, where local H2O/CO2 ratios vary; the paper only reports homogeneous mixtures.
  • Editorial inference: for dry-recycle conditions, the paper's coefficient profiles suggest the classical model's low absorption coefficient, not its spectral assumptions, causes the underprediction; a targeted emissivity measurement or narrow-band calculation at 10% H2O / 90% CO2 could validate that mechanism.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. The paper compares four recently developed oxy-fuel weighted sum of gray gases models (WSGGMs) — Johansson 3+1 and 4+1, Krishnamoorthy 3+1, and Yin 4+1 — against the classical air-fuel Smith et al. model. It compares absorption coefficients and emissivities over a range of H2O/CO2 ratios, and then solves the gray radiative transfer equation in a 2 m × 2 m × 4 m box with a prescribed inhomogeneous temperature field for four gas compositions representative of air-firing, wet-recycle oxy-firing, and dry-recycle oxy-firing. For each model, both piecewise-constant and piecewise-linear interpolation of tabulated coefficients are tested. The finite-volume results are compared with an SLW reference solution. The central finding is that in the dry-recycle case (10% H2O, 90% CO2), the Smith et al. model underpredicts radiative flux and source, while the newer oxy-fuel models overpredict; piecewise-linear interpolation substantially improves the Smith et al. predictions, with more modest effects on the oxy-fuel models.

Significance. If the SLW benchmark is reliable, the paper has a practically useful, low-cost result for carbon capture simulations: the widely used Smith et al. model can be made markedly more accurate in oxy-fuel dry-recycle conditions simply by changing how its coefficients are interpolated, and the Krishnamoorthy model offers the lowest overall deviation with reduced computational cost. The paper also compiles useful implementation details in the appendices, and Figure 4 provides a resolution-independence check for the finite-volume solver. The main caveat is that the reference SLW solution is unpublished, produced by a proprietary code, and not convergence-verified in this manuscript, which weakens the quantitative error estimates and the model ranking.

major comments (2)
  1. [4.2, Tables 3-6] The central quantitative claims rest on the SLW reference solution, but that solution is unpublished, was generated by Dr. Krishnamoorthy with a proprietary discrete-ordinates code, and is not accompanied by any convergence study in this manuscript. The SLW uses a 26×19×19 spatial grid with T4 quadrature, while the WSGGM solutions use 41×41×80 with 7×7 angular divisions; Figure 4 shows resolution insensitivity only for the J31s WSGGM solution, not for the SLW. Since the mean and RMS deviations in Tables 3-6 and the under/overprediction findings are all differences with respect to this benchmark, a biased or under-resolved SLW solution could change the ranking and the claimed improvement from piecewise-linear interpolation. Please provide the SLW flux/source profiles, a convergence study for the SLW discretization, or an independent comparison with a published spectral benchmark for the same configuration.
  2. [5, with Section 4.2] The model ranked best overall (Krishnamoorthy et al.) was authored by the same researcher who generated the unpublished SLW reference solution. This is not a derivational circularity, but it is a real independence concern for the benchmark: the error estimates in Tables 3-6 and the conclusions in Section 5 could be affected by unrecognized choices in the SLW setup. The manuscript should either add an independent benchmark (e.g., SLW results from another code or published SNB/SLW solutions) or explicitly document measures taken to avoid bias in generating the reference solution.
minor comments (5)
  1. [3.4] The list of Smith et al. tabulated sets is numbered 1, 2, 4, and 5; an item 3 is missing, making the enumeration incomplete.
  2. [Throughout] There are repeated typographical errors in the header and body, e.g., "Nation al Combustion Meeting", "Combu stion Institut e", and "WSSGM" instead of "WSGGM".
  3. [Acknowledgments] The sentence "The thank Dr. Krishnamoorthy for the use of his SLW results" should be "The authors thank Dr. Krishnamoorthy for the use of his SLW results."
  4. [4.2] The y-axis label in Figure 5 appears as a corrupted symbol in the manuscript; the figure should be regenerated with a clear label for the radiative heat flux magnitude.
  5. [2.3] The description of piecewise-linear interpolation states that extrapolation is used near RR=0 and RR=1, but it does not specify how this is done for models other than Smith et al.; a short general rule or reference to the appendix for each model would improve reproducibility.

Circularity Check

1 steps flagged · score 4.0 of 10

Benchmark provenance rather than derivational circularity: the SLW reference is unpublished and supplied by the same researcher whose WSGGM is ranked best, making the central ranking load-bearing on a self-referential source.

  1. other [Section 4.2 (Figure 5), Concluding Remarks, Acknowledgments]
    "All plots also contain results using the SLW model. These are unpublished results generated by Dr. Gautham Krishnamoorthy. ... Overall, the WSGGM developed by Krishnamoorthy et al. had the smallest deviation from the SLW results of all the models considering all 4 test cases."

    The paper's central accuracy claims and model ranking are defined as deviations from SLW reference fluxes (Tables 3-6). Those SLW results are not an external, published benchmark: they are unpublished computations supplied by Dr. Krishnamoorthy, who is also the lead author of the Krishnamoorthy et al. WSGGM that the paper concludes has the smallest overall deviation. The present author Huckaby is a co-author of that same WSGGM (reference [7]). The paper checks spatial/angular insensitivity only for its own finite-volume solutions (Figure 4), not for the SLW runs (26x19x19 grid, T4 quadrature), and provides no raw SLW profiles or convergence study.

full rationale

The core derivation chain—WSGGM coefficients to emissivity to absorption coefficient to finite-volume RTE solution—is not circular: the five WSGGMs are externally published (Smith, Johansson, Krishnamoorthy, Yin), their coefficients are taken as prescribed inputs, and the two interpolation variants are applied and compared independently. The under/overprediction findings and the interpolation improvements are genuine numerical outcomes, not fitted parameters or quantities defined in terms of themselves. No equation in the paper is constructed so that X derives from Y while Y is defined by X. The only circularity concern is the benchmark: the SLW reference is unpublished, supplied by Dr. Krishnamoorthy, whose own WSGGM is ranked best, and the present author is a co-author of that model. The paper does not demonstrate convergence or provide raw data for the SLW, so the ranking is not independent of the top-ranked model's origin. This does not reduce the derivation to an identity or a fit, but it does make the headline comparison load-bearing on a self-referential evidentiary source. For that reason the score is 4: the central claim has independent technical content, yet its supporting benchmark is not independent of the model it favors.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No new free parameters or entities are introduced. The central comparison relies on published WSGGM coefficients, a specified synthetic temperature field, and an unpublished SLW reference. The main ledger item is the reliance on the proprietary SLW benchmark, which is treated as ground truth.

assumptions (4)
  • domain assumption Gray-gas approximation: the total radiative intensity is described by a single wavelength-independent absorption coefficient (Eq. 4).
    Invoked in Section 2.2 to justify solving one RTE instead of many; the authors acknowledge the assumption is unrealistic for gases but argue it is acceptable for industrial wall-flux problems.
  • ad hoc to paper SLW reference solution accuracy: the unpublished SLW results by Krishnamoorthy are treated as the benchmark.
    Used in Section 4.2 and Tables 3-6 as the reference for all error estimates; the SLW code is proprietary and the reference results are not included in the paper.
  • standard math WSGGM representation: total emissivity is a weighted sum of gray gases with temperature-polynomial weights (Eq. 5), with tabulated coefficients taken from the original model papers.
    Assumed throughout; this is the standard WSGG formalism from Hottel and the cited model papers.
  • domain assumption The four gas mixtures in Table 1, particularly the dry-recycle case (10% H2O, 90% CO2), are representative of oxy-fuel combustion with carbon capture.
    The box test configurations are synthetic; the authors note the limited scope of the test but generalize to boiler simulations in Section 4.1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of New Weighted Sum of Gray Gases (WSGG) Models for Radiation Calculation in Carbon Capture Simulations: Evaluation and Different Implementation Techniques." pith.science (2026). https://pith.science/paper/RVJOIMKP

@misc{pith2026241118467,
  author       = {Pith},
  title        = {Pith review of: New Weighted Sum of Gray Gases (WSGG) Models for Radiation Calculation in Carbon Capture Simulations: Evaluation and Different Implementation Techniques},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RVJOIMKP}},
  note         = {Machine review of arXiv:2411.18467}
}
read the original abstract

We apply several weighted sum of gray gases models (WSGGMs) to calculate the radiative absorption coefficient for gas mixtures containing H2O and CO2. Our main objectives are to analyze and compare four WSGGMs which have been recently developed for oxy-fuel combustion. The models are compared with the widely-used air-fuel WSGGM of Smith et al. In addition to direct comparison of the absorption coefficients, we compare finite-volume solutions of the radiative equation of transfer in a 2m x 2m x 4m box with a specified inhomogeneous temperature field and a homogeneous mixture of the H2O, CO2 and N2. Calculations using a spectral line-based WSGGM (SLW) are used as a reference solution to estimate the accuracy. For each WSGGM, we apply two interpolation methods for determining the model coefficients at arbitrary H2O-to-CO2 ratios. For wet-recycle oxy-fuel combustion, we found that the deviation of the air-fuel WSGGM was not significantly larger than several of the newer models. However, with dry recycle (90 vol%-CO2) the air-fuel WSGGM model underpredicts the radiative flux and radiative heat source in contrast to the other models which overpredict these fields. Piecewise linear interpolation consistently improves the predictions of the air-fuel WSGGM, but only has a modest effect on the predictions of the oxy-fuel WSGGM's.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages

  1. [1]

    Andersson, Characteri zation of Oxy-fuel Flames - Their Composition, Temperature and Radiat ion, Ph.D

    K. Andersson, Characteri zation of Oxy-fuel Flames - Their Composition, Temperature and Radiat ion, Ph.D . Dissertation, Chalmers University of Technology, Sweden , 2007

    work page 2007

  2. [2]

    Chui , M

    E. Chui , M. A. Douglas, Y. Tan, Fuel 82 (2003) 1201- 1210

    work page 2003

  3. [3]

    S. P. Khare , T . F. Wall, A. Z. Farida , Y. Liu , B. Moghtaderi , R. P. Gupta , Fuel 87 (2008) 1042- 1049

    work page 2008

  4. [4]

    P. Heil, D. Toporov , H. Stad ler, S. Tschunko, M. Forster , R. Kneer , Fuel 88 (2009) 1269-1274

    work page 2009

  5. [5]

    T. F. Wall, Proceedings of the Combustion Institut e 31 (2007) 3 1-47

    work page 2007

  6. [6]

    Johansson, K

    R. Johansson, K. Andersson, B. Leckner, H. Thunman , International Journal of Heat and Mass Transf er 53 (2010) 220-230

    work page 2010

  7. [7]

    Krishnamoorthy , M

    G. Krishnamoorthy , M. Sarni, S. Orsino, A. P erera, M. Shahnam , E. D. Huckaby, International Journal of Computational Fluid Dynam ics 24 (2010) 69- 82

    work page 2010

  8. [8]

    C. Yin, L. C. R. Johansen, L. A. Rosendah l, S. K. Krer , Energy e3 Fuels 24 (2010) 6275- 6282

    work page 2010

Show all 33 references
  1. [9]

    T. F. Smit h, Z . F. Shen, J. N. Friedman, Journ al of Heat Transfer 104 (1982) 602- 608

    1982

  2. [10]

    Goutiere, F

    V. Goutiere, F. Liu, A. Charette, Journal of Quantitative Spectroscopy e3 Radiative Transf er64 (2000) 299-326

    2000

  3. [11]

    ANSYS FLUENT 12.0, User's Guide, 2009, Canonsburg, PA, USA

    2009

  4. [12]

    Besset te , T

    D. Besset te , T . Marchal, A. Verma, XXI International Congress on Glass (ICC 2007), 2007

    2007

  5. [13]

    Porter, F

    R. Porter, F . Liu, M. Pomkashanian, A. William s, D. Smith, Journal of Quantitat ive Spectroscopy e3 Radiative Transfer 111 (2010) 2084-2094

    2010

  6. [14]

    Liu , Journal of Heat Transfer 121 (1999) 200-203

    F. Liu , Journal of Heat Transfer 121 (1999) 200-203

    1999

  7. [15]

    Siegel, J

    R. Siegel, J. R. Howell, Thermal Radiation Heat Transfer , 4th Edition, Tay lor & Francis , USA, 2002

    2002

  8. [16]

    M. F. Modest , Radiative Heat Transfer, 2nd Edit ion, Academ ic Press, USA, 2003

    2003

  9. [17]

    H. C. Hottel , in: W. H. McAdams (Ed.) , Heat Transmiss ion, 3rd Edition , McGraw -Hill, New York, 1954, Ch. 4, pp. 55- 125

    1954

  10. [18]

    H. C. Hottel Journal of the Institute of Fuel 34 (1961) 220- 234

    1961

  11. [19]

    A. F. Sarofim , Radiant Heat Transmission in Enclosures, Sc.D. Dissertatio n, Massachuset ts Institute of Tech- nology, Cambr idge, USA, 1962

    1962

  12. [20]

    H. C. Hotte l, A. F. Sarofim, Radiative Transfer, McGraw -Hill , New York , USA, 1967

    1967

  13. [21]

    Lallemant , A

    N. Lallemant , A. Sayrer, R. Weber, Progress in Energy and Combustion Science 22 (1996) 543- 574

    1996

  14. [22]

    D. C. Haworth, Progress in Energy and Combustion Science 36 (2010) 168-259

    2010

  15. [23]

    H. C. Hottel, J. J. Noble, A. F. Sarofim , G. D. Silcox, P. C. Wankat, K. S. Knaebel, in: D. W. Green, R. H. Perry (Eds.), Perry's Chemic al Engineers ' Handbook, 8th Edition, McGraw -Hill , New York, 2007, Ch. 5, pp. 1- 83

    2007

  16. [24]

    National Combustion Meeting, March 2011 Paper OT08

    EM2C Lab: Laboratoire d'Energ etique Moleculaire et Macrosco pique, Comb ustio n du CNR et Ecole Cent rale Paris Trans latio n: Molecular and Macroscopic Molecular Energetics Labora tory of CNRS [French nation al center for scienti fic research ] and Ecole Centra le Paris [Par...

    2011

  17. [25]

    W. L. Grosshandler , RADCAL: a Narrow-Band Model for Radi ation Calculat ions in a Combusti on Environment , NIST Technical Note 1402, 1993

    1993

  18. [26]

    D. K. Edward s, W. A. Menard , Applied Optics 3 (1964) 621- 625

    1964

  19. [27]

    D. K. Edwards, A. Balakrishn an, International Journal of Heat and Mass Transf er 16 (1973) 25-40

    1973

  20. [28]

    D. K. Edward s Advances in Heat Transfer 12 (1976) 115- 193

    1976

  21. [29]

    A. T . Modak, Journal of Quant ita tive Spectroscopy and Radiative Transf er 21 (1978) 131- 142

    1978

  22. [30]

    M. K. Denison, B. W. Webb , Journal of Quan titative Spectroscopy & Radiative Transf er 50 (1993) 499-510

    1993

  23. [31]

    M. K. Denison, B. W. Webb, International Journal of Heat and Mass Transfer 38 (1995) 1813- 1821

    1995

  24. [32]

    M. K. Denison, B. W. Webb, Journal of Heat Transfer 117 (1995) 788- 792

    1995

  25. [33]

    C. P. Thurgood , A. Pollard , A. B. Becker, Journal of Heat Transfer 117 (1995) 1068-1070. 14

    1995

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.