Hostname: page-component-7b7567c8c-tnzcf Total loading time: 0 Render date: 2026-11-13T20:53:29.903Z Has data issue: false hasContentIssue false
  • English
  • Français

Can Wolf-Rayet stars be the missing ingredient to explain high-z He II ionizing radiation?

Published online by Cambridge University Press:  09 November 2026

G. González-ToràOpen the ORCID record for G. González-Torà[Opens in a new window],
A. A. C. Sander,
E. Egorova,
O. Egorov,
M. Bernini-Peron,
J. JosiekOpen the ORCID record for J. Josiek[Opens in a new window],
K. Kreckel,
R. R. LefeverOpen the ORCID record for R. R. Lefever[Opens in a new window],
V. RamachandranOpen the ORCID record for V. Ramachandran[Opens in a new window] and
E. C. Schösser
G. González-Torà*
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 Heidelberg
A. A. C. Sander
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 HeidelbergUniversität Heidelberg, Interdiszipliäres Zentrum für Wissenschaftliches Rechnen, 69120 Heidelberg, Germany
E. Egorova
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 Heidelberg
O. Egorov
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 Heidelberg
M. Bernini-Peron
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 Heidelberg
J. Josiek
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 Heidelberg
K. Kreckel
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 Heidelberg
R. R. Lefever
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 Heidelberg
V. Ramachandran
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 Heidelberg
E. C. Schösser
Affiliation:
Zentrum für Astronomie der Universität Heidelberg, Astronomisches Rechen-Institut, Mönchhofstr. 12-14, 69120 Heidelberg
*
email: gemma.gonzalez-tora@uni-heidelberg.de
Rights & Permissions [Opens in a new window]

Abstract

Classical Wolf-Rayet (WR) stars are hot, massive stars with depleted hydrogen. At low metallicities (Z), WN3-type WR stars have relatively thin winds and are major sources of ionizing flux. The detection of high-ionization emission lines in high-redshift (z) galaxies as well as nearby low-Z dwarf galaxies raises questions about the origin of He II ionizing radiation and its role in galaxy evolution, as stellar population models fail to reproduce the required fluxes. Low-Z WN3 stars may provide the missing contribution but are easily hidden in integrated light. Using the Local Volume Mapper, we compare resolved optical spectra of SMC WN3 stars with integrated regions, focusing on the broad He II $\lambda4686\,\unicode{x00C5}$ line. We find stellar emission diluted within nebular regions, becoming undetectable when integrating over areas larger than 24 pc. Nonetheless, these stars emit enough ionizing photons to explain observed He II nebular emission, being strong candidates for the He II ionizing sources in low-Z and high-z galaxies.

Keywords

stars: Wolf-Rayetstars: mass-lossstars: massiveGalaxies: ISMGalaxies: stellar contentISM: HII regions

Information

Type
Poster Paper
Information
Proceedings of the International Astronomical Union , Volume 21 , Symposium S402: Massive Stars Across Redshifts in the Era of JWST and Large Surveys , January 2025, pp. 274 - 277
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - NCCreative Common License - SA
This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike licence (https://creativecommons.org/licenses/by-nc-sa/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the same Creative Commons licence is used to distribute the re-used or adapted article and the original article is properly cited. The written permission of Cambridge University Press or the rights holder(s) must be obtained prior to any commercial use.
Copyright
© The Author(s), 2026. Published by Cambridge University Press on behalf of The International Astronomical Union

1. Introduction

Nebular He II recombination emission lines are present in both high-z (e.g., Topping et al. Reference Topping2025) and nearby low-metallicity (Z) galaxies (e.g., Kehrig et al. Reference Kehrig2018; Wofford et al. Reference Wofford2021), implying the existence of sources that can produce a significant fraction of high energy photons ( lamda less than 228 $\lambda\lt228\,$ Å) to ionize He Superscript plus $^{+}$ in their surrounding interstellar medium. Wolf-Rayet stars are hot ( upper T Subscript normal e normal f normal f Baseline greater than 30 $T_{\mathrm{eff}}\gt30$ kK) massive stars with very dense winds that present broad emission lines in their spectrum. Among them, a subclass of WR stars showing strong helium and nitrogen emission lines are defined as WN stars. Particularly, WN3 stars are WN stars showing emission from high ionization species (e.g., He II, N V), with many of them showing relatively thin winds among the WR regime. Spectral modeling has revealed them as important sources of He Superscript plus $^{+}$ ionizing flux (e.g., Hainich et al. Reference Hainich2015; Sander Reference Sander2022), which makes them particularly interesting as the WR subtype occurrence shifts to earlier types at lower Z (e.g., Crowther Reference Crowther2007).

While resolving individual stars in high-z galaxies is unfeasible, we have a nearby low-Z environment with a resolvable WN3 stellar population, the Small Magellanic Cloud (SMC, upper Z almost equals 0.2 upper Z Subscript circled dot $Z \approx 0.2\,Z_\odot$ , Bouret et al. Reference Bouret2003; Ramachandran et al. Reference Ramachandran2019). Hosting 12 known, resolvable WR stars, we compare the intrinsic stellar properties and spectra of 6 WN3 in the SMC (AB 1, 7, 9, 10, 11 and 12) with unresolved integral field unit (IFU) spectra from the SDSS V Local Volume Mapper survey (LVM, Drory et al. Reference Drory2024), focusing on the 4686 modifying above upper A with ring $\unicode{x00C5}$ He II line, the strongest optical emission line for WN stars (see also González-Torà et al. Reference González-Torà2025).

2. Atmosphere analyses and IFU observations

All the individual WN3 targets have optical (3700 - 6830 modifying above upper A with ring $\unicode{x00C5}$ ) data from Foellmi et al. (Reference Foellmi, Moffat and Guerrero2003), and have been previously analyzed with the PoWR atmosphere code (Gräfener et al. Reference Gräfener, Koesterke and Hamann2002; Sander et al. Reference Sander2015) in Hainich et al. (Reference Hainich2015) or Shenar et al. (Reference Shenar2016). Despite having a WR classification, only the models for the WN3 star in AB 7 filter yields a wind optical depth larger than unity, namely tau Subscript upper F Baseline left parenthesis upper R Subscript sonic Baseline right parenthesis equals 2.1 $\tau_F(R_{\text{sonic}}) = 2.1$ . All single WN3 in the SMC analyzed in Hainich et al. (Reference Hainich2015) show tau Subscript upper F Baseline left parenthesis upper R Subscript sonic Baseline right parenthesis less than 1 $\tau_F(R_{\text{sonic}}) \lt 1$ , spanning between 0.29 $0.29$ (AB 1) and 0.8 $0.8$ (AB 10). This means that the winds of the single SMC WN3 stars are not globally optically thick. The transition in optical thickness further does not align with transparency in He II ionizing flux, as outlined by AB 7 and the theoretical study by Sander et al. (Reference Sander2023). From the PoWR atmosphere models (taken from Hainich et al. Reference Hainich2015; Shenar et al. Reference Shenar2016), we can compute the emerging ionizing fluxes Q Subscript normal e normal d normal g normal e $_{\mathrm{edge}}$ for an ion with a certain ionization edge frequency nu Subscript normal e normal d normal g normal e $\nu_{\mathrm{edge}}$ via

(1) upper Q Subscript normal e normal d normal g normal e Baseline colon equals 4 pi upper R Subscript asterisk Superscript 2 Baseline integral Subscript nu Subscript normal e normal d normal g normal e Baseline Superscript normal infinity Baseline StartFraction upper F Subscript nu Baseline Over h nu EndFraction normal d nu comma \begin{equation} Q_{\mathrm{edge}}:= 4 \pi R_\ast^2 \int_{\nu_{\mathrm{edge}}}^{\infty} \frac{F_{\nu}}{h\nu} \,\mathrm{d}\nu,\end{equation}

with upper F Subscript nu $F_{\nu}$ denoting the stellar flux and upper R Subscript asterisk $R_\ast$ the stellar radius.

Thanks to SDSS V (Kollmeier et al. Reference Kollmeier2025), the availability of the LVM wide field IFU data now enables us to cross-check the spectra from the individual WN3 targets as well as their inferred quantities with nebular diagnostics. LVM covers a field of view of 0.5 degrees with 1801 hexagonally packed fibers of 35.3” apertures, with a spectral coverage 3600-9800 modifying above upper A with ring $\unicode{x00C5}$ , and a spectral resolution of R tilde $\sim$ 4000. We have studied the integrated spectra from the fiber apertures around the SMC WN3 stars with angular radii of 20”, 40”, 80”, 160”, and 300”, corresponding to a diameter of 12, 24, 48, 96 and 181 pc (see left panel in Figure 1). From the nebular emission, we estimated Q Subscript normal upper H normal e Sub Subscript normal upper I normal upper I $_\mathrm{He_{II}}$ using the formula from Kehrig et al. (Reference Kehrig2015):

(2) upper Q left parenthesis HeII right parenthesis equals upper L Subscript HeII Baseline divided by left bracket j left parenthesis lamda Baseline 4686 right parenthesis divided by alpha Subscript upper B Baseline left parenthesis HeII right parenthesis right bracket almost equals upper L Subscript HeII Baseline divided by 3.66 times 10 Superscript negative 13 Baseline comma \begin{equation} \text{Q}(\text{HeII})=L_{\text{HeII}}/[j(\lambda 4686)/\alpha_{\text{B}}(\text{HeII})]\approx L_{\text{HeII}}/3.66\times10^{-13} ,\end{equation}

where upper L Subscript HeII $L_{\text{HeII}}$ is the line luminosity obtained from the LVM spectra, j left parenthesis lamda Baseline 4686 right parenthesis $j(\lambda 4686)$ is the photon energy at 4686 modifying above upper A with ring $\,\unicode{x00C5}\,$ , and alpha Subscript upper B Baseline left parenthesis HeII right parenthesis $\alpha_{\text{B}}(\text{HeII})$ the recombination rate coefficient assuming case B recombination and upper T Subscript e Baseline tilde 2 times 10 Superscript 4 $T_{\text{e}}\sim2\times10^{4}$ K (Osterbrock & Ferland Reference Osterbrock and Ferland2006).

Figure 1.

Left: AB7 fiber apertures studied from the LVM. Center: Normalized 4686 modifying above upper A with ring $\unicode{x00C5}$ flux of the star and the different fiber apertures of the LVM. Right: Intrinsic flux of the WN3h component in the AB 7 binary.

Three panels show data related to the star AB7 and its ionizing flux.

3. Results and conclusions

In all cases, the stellar emission is strongly diluted in the LVM data. An example is shown in the right panel of Figure 1, were we plot the normalized spectra around He II 4686 modifying above upper A with ring $4686\,\unicode{x00C5}$ for AB7 and the corresponding LVM fiber apertures. While still present in the smallest 20” aperture, the stellar component completely vanishes at 40”, corresponding to a diameter of 24 pc. In González-Torà et al. (Reference González-Torà2025) we show that the stellar contribution for the rest of the targets is always completely diluted at an aperture of only 24 pc. These results imply that WN3 stars at higher distances in unresolved environments can be hard or even impossible to detect.

However, their non-detection does by no means imply that such stars could not contribute significant numbers of ionizing photons. Using again the LVM data, we can compare how much of the intrinsic Q Subscript normal upper H normal e Sub Subscript normal upper I normal upper I $_\mathrm{He_{II}}$ is visible in the observed nebular He II emission lines. Figure 2 shows the ratio of ionizing fluxes Q Subscript normal upper H normal e Sub Subscript normal upper I normal upper I Subscript comma normal upper L normal upper V normal upper M Baseline divided by normal upper Q Subscript normal upper H normal e Sub Subscript normal upper I normal upper I Subscript comma normal s normal t normal a normal r $_\mathrm{He_{II},LVM}/\mathrm{Q}_\mathrm{He_{II},star}$ with respect to the diameter of the LVM integrated fiber in pc. We see that for all cases – except AB 7 with d greater than $\gt$ 48 pc – the star alone produces more than enough photons to produce the observed emission in largest aperture (i.e., Q Subscript normal upper H normal e Sub Subscript normal upper I normal upper I Subscript comma normal upper L normal upper V normal upper M Baseline divided by normal upper Q Subscript normal upper H normal e Sub Subscript normal upper I normal upper I Subscript comma normal s normal t normal a normal r Baseline less than 1 $_\mathrm{He_{II},LVM}/\mathrm{Q}_\mathrm{He_{II},star}\lt1$ ). For AB7 at d greater than $\gt$ 48 pc, there is a blending with other nearby sources of He II ionizing flux, possibly from within NGC 395.

Figure 2.

Ratio of Q Subscript normal upper H normal e normal upper I normal upper I $_{\mathrm{HeII}}$ for the different stars and the corresponding LVM integrated region with respect to the diameter of the LVM aperture. The Q Subscript normal upper H normal e normal upper I normal upper I comma normal s normal t normal a normal r $_{\mathrm{HeII, star}}$ ratio is obtained by the PoWR models, while the Q Subscript normal upper H normal e normal upper I normal upper I comma normal upper L normal upper V normal upper M $_{\mathrm{HeII, LVM}}$ is calculated using Eq. 2, using only for the narrow nebular component at each aperture.

Line graph showing the ratio of Q HeII for different stars and corresponding LVM integrated regions with respect to the diameter of the LVM aperture.

Our results demonstrate that WN3 stars with comparably thin winds and thus weak emission lines are easy to hide, despite producing enough photons to reproduce the observed nebular He II emission. Current evolution models cannot reproduce these observed WR stage, in particular at SMC metallicities and below, thereby causing a major deficiency in current population synthesis. Solving the origin puzzle of the WN3 stars and other related objects such as WN2 stars (Sander et al. Reference Sander2025) is urgently needed and will likely have major impacts on stellar population predictions and cosmological simulations of galaxy evolution.

References

Bouret, J. C., et al. 2003, ApJ, 595, 1182 CrossRefGoogle Scholar
Crowther, P. A. 2007, ARA&A, 45, 177 Google Scholar
Drory, N., et al. 2024, AJ, 168, 198 CrossRefGoogle Scholar
Foellmi, C., Moffat, A. F. J., & Guerrero, M. A. 2003, MNRAS, 338, 360 CrossRefGoogle Scholar
González-Torà, G., et al. 2025, A&A, 703, L11 Google Scholar
Gräfener, G., Koesterke, L., & Hamann, W. R. 2002, A&A, 387, 244 Google Scholar
Hainich, R., et al. 2015, A&A, 581, A21 Google Scholar
Kehrig, C., et al. 2018, MNRAS, 480, 1081 Google Scholar
Kehrig, C., et al. 2015, ApJ, 801, L28 CrossRefGoogle Scholar
Kollmeier, J. A., et al. 2025, arXiv e-prints, arXiv:2507.06989 Google Scholar
Osterbrock, D. E., & Ferland, G. J. 2006, Astrophysics of gaseous nebulae and active galactic nucleiGoogle Scholar
Ramachandran, V., et al. 2019, A&A, 625, A104 Google Scholar
Sander, A., et al. 2015, A&A, 577, A13 Google Scholar
Sander, A. A. C. 2022, arXiv e-prints, arXiv:2211.05424 Google Scholar
Sander, A. A. C., et al. 2023, A&A, 670, A83 Google Scholar
Sander, A. A. C., et al. 2025, arXiv e-prints, arXiv:2508.18410 Google Scholar
Shenar, T., et al. 2016, A&A, 591, A22 Google Scholar
Topping, M. W., et al. 2025, ApJ, 980, 225 CrossRefGoogle Scholar
Wofford, A., et al. 2021, MNRAS, 500, 2908 CrossRefGoogle Scholar
Figure 0

Figure 1.Left: AB7 fiber apertures studied from the LVM. Center: Normalized 4686 Å$\unicode{x00C5}$ flux of the star and the different fiber apertures of the LVM. Right: Intrinsic flux of the WN3h component in the AB 7 binary.

Figure 1

Figure 2.Ratio of QHeII$_{\mathrm{HeII}}$ for the different stars and the corresponding LVM integrated region with respect to the diameter of the LVM aperture. The QHeII,star$_{\mathrm{HeII, star}}$ ratio is obtained by the PoWR models, while the QHeII,LVM$_{\mathrm{HeII, LVM}}$ is calculated using Eq. 2, using only for the narrow nebular component at each aperture.