Michael Abdul-Masih

Context. Accurate modeling of stellar spectra is essential for deriving the physical properties of stars. However, traditional model atmospheres often oversimplify key phenomena – such as deformations due to rotation or multiplicity – that break spherical symmetry. The Spectroscopic PAtch Model for Massive Stars ( SPAMMS ) addresses these limitations by explicitly accounting for surface distortions, resulting in more realistic spectra for deformed stars. Yet, SPAMMS has been limited by the spectral types covered by its model atmosphere grids. Aims. We aim to extend the parameter space of the stellar atmosphere grids available to the SPAMMS framework. By expanding the coverage of effective temperatures, surface gravities, and metallicities, we enable the synthesis of spectra across a broader range of spectral types. Methods. We computed specific intensities, I ( λ, µ ), for 101 emergent angles using PRISMAS (Pipeline of Radiative Intensity Synthesis for Meshed Atmospheric Surfaces). Built upon SYNPLE , PRISMAS employs precomputed local thermodynamic equilibrium (LTE) and non-LTE model atmospheres, including two ATLAS9-Kurucz grids and the TLUSTY-based OSTAR2002 and BSTAR2006 models. Results. The computed intensity grids span effective temperatures from 3500 to 55 000 K and surface gravities between 0.0 and 5.0 dex – covering the entire range of spectral types from O- to K-type stars. These grids also incorporate multiple metallicities (0−30 Z _⊙ ) and microturbulent velocities (1, 3, 5, 10km s ⁻¹ ), enabling detailed modeling of stars across diverse evolutionary stages. The spectral coverage extends from 3000 to 9000 Å, spanning both the ultraviolet and optical regimes with a sampling of ∆ λ = 0.01 A. As a proof of concept, we adopted the new intensity grids to model (i) a rapidly rotating B-type star, and (ii) an eclipsing Algol-type binary using SPAMMS . Conclusions. The implementation of LTE-Kurucz and NLTE-TLUSTY model atmospheres significantly expands the parameter space accessible to SPAMMS . The code can now generate realistic synthetic spectra for a wide range of stellar types and geometries, including nonspherical and multiple systems. Therefore, by explicitly modeling three-dimensional surface effects, SPAMMS reproduces key observational signatures across a broad range of stellar masses and effective temperatures – from O- to K-type stars.

Aims. K 1-6 has long been classified as a planetary nebula (PN) hosting a binary central star; however, it has remained poorly studied due to its faintness. The central star exhibits pronounced photometric variability whose origin is still unclear. We present a comprehensive characterisation of the K 1-6 system, including the physical properties of its stellar components and the nature of the surrounding nebulosity. Methods. We conducted a multi-wavelength analysis combining optical and UV spectroscopy obtained with the Gran Telescopio Canarias, the Telescopio Nazionale Galileo, the Nordic Optical Telescope, and the Hubble Space Telescope. We also present long-term multi-band ground- and space-based photometry, including high-cadence data from the Transiting Exoplanet Survey Satellite, narrowband imaging, and the latest astrometric constraints from Gaia . Results. Our results show that the nebula is not a remnant PN, but instead consists of interstellar medium photoionised by a hot white dwarf, which is relatively evolved. It has a cooling age of 1–2 Myr, implying that any original PN has long since dissipated. We further find that the central object is a hierarchical triple system, comprising an inner binary with an orbital period likely of the order of thousands of days and a distant tertiary companion orbiting on a timescale of tens of thousands of years. The optically dominant cool component of the inner binary is an inflated K-type star displaying extreme magnetic activity, including large-amplitude variability and flaring. Its properties resemble those of BY Dra-type binaries and Abell 35-type systems, and are difficult to reconcile with single-star evolution, pointing instead to a history of binary interaction.

Context . The importance of massive stars cannot be overstated: they are powerful probes of the early universe, they play a vital role in the chemical and mechanical evolution of their host environments, and their end products allow us to study the most extreme physics in the universe. Obtaining accurate stellar and surface parameters for large samples of massive stars is vital to our understanding of how they evolve and how their births, lives, and deaths affect their surroundings. With the large volume of data expected from the next generation of spectroscopic surveys, it is likely that the computational cost of our current analysis methods (especially the calculation of model stellar atmospheres and synthetic spectra) will prove to be the most important bottleneck impeding our progress. Aims . In order to overcome these limitations and dramatically decrease computing times, we aim to develop a robust emulator for the FASTWIND radiative transfer and spectral synthesis code. Additionally, we aim to explore alternative fitting methods that have not been feasible up to this point due to computational costs. Methods . We calculated a large set (∼50 000 models in total) of FASTWIND synthetic spectra of OB-type stars, varying temperature, surface gravity, helium mass fractions, and CNOSi abundances, and we trained a collection of neural networks to emulate these models with separate networks for each line. We also developed the open-source python package SpecFANN (Spectral Fitting via Artificial Neural Networks), which provides users with a suite of fitting methods that can be used with these or other user-generated neural networks. Results . The majority of the trained neural networks reach average accuracies of better than ∼0.01–0.1% for photospheric lines and better than ∼0.1–1% for wind lines. We find that SpecFANN is able to obtain robust and accurate stellar parameters that are consistent with the literature for a sample of 52 early-type stars. Using SpecFANN we find that we can achieve the same fit in ∼1/360 000 of the time when compared to alternative techniques that rely on “on-the-fly” FASTWIND computations. Conclusions . We have demonstrated that neural networks offer a viable path forward to address the computational limitations of our current atmosphere analysis and stellar parameter determination methods for hot stars.

Models of eclipsing binaries require the assignment of appropriate emergent intensities to the surface elements of the binary components. For distance-dependent modeling of flux-calibrated light curves, this necessitates an approximation of the absolute normal intensities of both components of the binary, as well as how their brightness varies across the stellar disks (limb darkening). Such surface intensities are often inferred from other physical properties of the synthetic binary (effective temperature, surface gravity, etc.) through the use of model atmospheres, which in turn are generally suited to a particular range of stellar types or parameters. Here, we present the major developments included in the PHOEBE 2.5 release (publicly available from http://phoebe-project.org ), which improve the fidelity of model binaries comprising hot, compact stars. These developments include the incorporation of model atmospheres produced using the Tübingen Model-Atmosphere Package and Montréal/Tremblay codes (complementing the already incorporated PHOENIX and Castelli & Kurucz models, primarily suited to main-sequence stars and low-temperature giants). Similarly, PHOEBE v2.5 now allows for blending/extrapolation of model atmospheres, meaning one can continue to make use of model atmospheres in cases when a small number of surface elements have parameters outside the model-atmosphere grid. As an added value product, we also present tables of limb-darkening coefficients derived from the newly incorporated model atmospheres, such that they can be used as inputs in other binary modeling codes.

Binary interactions during the asymptotic giant branch phase can lead to the formation of chemically peculiar stars such as Ba-, CH-, and CEMP-s stars with overabundances of s-process elements as well as carbon. Only a few hundred of these stars have been subject to detailed chemical and/or dynamical studies, typically in non-homogeneous samples. We compile a sample of s-process-polluted candidates using the fourth data release of the large spectroscopic survey GALAH. systematic We also compare the chemical and orbital properties of these candidates with those of confirmed s-process-polluted stars to gain stronger evidence of their nature. GALAH DR4 uses neural networks and automatic spectral analysis methods as well as data of a lower spectral resolution than is normally used to characterise objects. . We compared our stellar parameters and individual abundances with those of the survey and used this validation to define the thresholds that a star in GALAH DR4 must pass to be flagged as a these We therefore built a validation sample of s-rich stars before querying the survey, for which we obtained VLT/UVES and Mercator/HERMES high-resolution (>80,000) spectra and analysed them with traditional methods good s-process-rich candidate. Based on our comparisons with the high-resolution complementary spectra, we defined thresholds on five quantities: rm s/Fe > 0.40 dex, rm Y/Fe > 0.22 dex, rm Zr/Fe > 0.26 dex, rm Ba/Fe > 0.19 dex, and rm La/Fe > 0.26 dex. We identified stars in GALAH DR4 that are good candidates to be s-process-polluted stars, covering a broad parameter space . They share many similarities with , which strengthens our confidence in . When we cross-matched the sample with the literature, we found that only 7% of the candidates have measured orbital periods and eccentricities. This limits a full comparison with confirmed 1059 (4000 K < < 0.5 dex) < 6000 K, 0.5 dex < łogg < 5.0 dex, -2.0 dex < Fe/H the samples of confirmed s-rich stars, especially their ratios of heavy over light s-elements the purity of the sample Ba and related star samples for now. However, their binary fraction is, as expected, higher than the one we found for the full GALAH DR4 catalogue. Our sample of candidates is almost five times larger than the number of currently confirmed polluted stars. This and the fact that it has been homogeneously treated by the GALAH survey opens very interesting avenues to compare nucleosynthesis models as well as binary evolution models.

Context . Despite the importance of magnetic fields in massive stars, their origin is widely debated and still not well understood. Aims. With the mounting evidence for the importance of studying magnetic fields in interacting massive binary and multiple systems, it appears necessary to investigate the presence of magnetic fields in semi-detached systems with ongoing mass transfer, and in contact systems where mass is actively being exchanged. Methods . We present an analysis of 53 high-resolution HARPSpol spectropolarimetric observations of a sample of 14 massive binary and multiple systems using the least-squares deconvolution technique. The majority of the studied systems are classified as semi-detached or contact binaries. Results . Definite detections of the presence of a magnetic field are achieved in all studied systems apart from the rather faint system SV Cen, for which only a marginal detection was obtained. The fact that the presence of magnetic fields is detected in all but one of the studied systems strongly suggests that interaction between the system components plays a definite role in the generation of magnetic fields in massive stars. The measured mean longitudinal magnetic field strength for all targets is of the order of a few hundred Gauss to a few kiloGauss. The strongest longitudinal magnetic fields of 4–5 kG are discovered in the massive O-type triple system MY Ser in both components of the contact binary. kiloGauss-order magnetic fields are also detected in two other systems, V1294 Sco and V606 Cen. It is possible that there is an implication of some system characteristics, such as multiplicity, the mass ratio between the components, and a large fillout factor, on the measured magnetic field strength. Our results for the magnetic field measurements in interacting binaries present the first assessment of the occurrence rate of magnetic fields in a representative sample of such systems.

Context. Cygnus OB2, located within the Cygnus X complex – one of the most active star-forming regions of the Galaxy – hosts hundreds of O- and B-type stars at different evolutionary stages. This rich association offers a unique opportunity to study the evolution and dynamic interactions of massive stars. However, despite extensive studies, a notable absence of a fast-rotating group ( v sin i > 200 km s ⁻¹ ) among the O-type population of Cygnus OB2 challenges current models of massive star evolution. Aims. Stellar rotation strongly impacts spectral line shapes of O-type stars, with high rotational velocities potentially leading to misclassifications. This study investigates whether some stars in Cygnus OB2, classified at low spectral resolution as B0, are actually rapidly rotating late-O types. Such cases could explain the observed lack of fast rotators in Cygnus OB2. Methods. Considering the effects of rotation, we reclassified the known B0 population in Cygnus OB2, using the MGB tool and both the new and pre-existing optical spectroscopy. Finally, we computed the projected rotational velocities using iacob-broad . Results. We find that approximately 19% of the initial B0 population in Cygnus OB2 are, in fact, late-O types. Further analysis shows that only six stars in the entire dataset have projected rotational velocities above 200 km s ⁻¹ , with just one new O-type star exceeding this threshold. Conclusions. In our study of Cygnus OB2, we continue to find a notable lack of fast rotators among its O-type population. We propose a combination of three factors as the most likely explanation: (i) the young age of Cygnus OB2 may imply that fast rotators have not been produced yet due to binary interactions; (ii) fast rotators may have been dynamically ejected from the core as runaway stars; and (iii) local star formation conditions may hinder binary formation (reducing spin-up interactions) or result in slower rotational velocities at birth.

Although triple systems are common, their orbital dynamics and stellar evolution remain poorly understood. We investigated the V1371 Tau system using TESS photometry, multiepoch spectroscopy, and recent interferometric data, confirming it as a rare triple system consisting of a eclipsing binary orbited by a classical Be star, with a spectral classification of (B1V + B0V) + B0Ve. The eclipsing binary exhibits an orbital period of ≈34 days, and the Be star orbits the inner pair on a timescale of a few years. Weak H α emission lines suggest the presence of a Keplerian disk with variability on a timescale of months around the Be star, and nearly constant V / R ratio no detectable asymmetry variations. Besides the eclipses, frequencies at 0.24 and 0.26 cycles day ^–1 dominate the photometric variability. Higher-frequency signals are present, which appear associated with nonradial pulsations. The eclipsing pair ( i ≈ 90 ^∘ ) shows projected rotational velocities of 160 and 200 km s ⁻¹ . The Be star’s measured v sin i ≈ 250 km s ⁻¹ implies a critical rotation fraction between 0.44 and 0.76 for plausible inclinations, significantly faster than the eclipsing components. The shallower eclipses in the KELT data compared to TESS suggest a variation in orbital inclination, possibly induced by Kozai–Lidov cycles from the outer Be star. The evolution analysis suggests that all components are massive main-sequence stars, with the secondary star in the eclipsing binary being overluminous. This study emphasizes the complexity of triple systems with Be stars and provides a basis for future research on their formation, evolution, and dynamics.

Despite their importance, the dynamical properties of massive stars remain poorly understood. Rotation is a key driver of internal mixing and angular momentum transport, significantly influencing massive star evolution. However, constraining rotation from spectroscopy is challenging, as spectral lines often exhibit excess broadening beyond rotation. The origin of this additional broadening, typically attributed to large-scale velocity fields and commonly referred to as macroturbulence, remains uncertain and unconstrained. Here, we present a combined analysis of TESS photometry and rapid time-series spectroscopy obtained using the high-resolution ESPRESSO instrument at the Very Large Telescope of the European Southern Observatory for the rapidly rotating and pulsating massive star ζ Oph. Leveraging excellent temporal coverage, our analysis demonstrates that pulsation-induced variability leads to peak-to-peak scatter as large as 88 km s ⁻¹ in the observed macroturbulent velocity time series. We also demonstrate that time-averaged macroturbulent velocities are spectral-line-specific and can exceed 100 km s ⁻¹ . Furthermore, the macroturbulent velocities from shorter integration times are systematically lower than those derived from stacked spectra that mimic the longer exposure times typically needed for fainter stars. These results highlight the role of pulsations in driving variable macroturbulence in massive stars, while also pointing to a potential bias in spectroscopic estimates of macroturbulence for fainter massive stars.