New Pulsating Stars Detected in EA-type Eclipsing-binary Systems Based on TESS Data

Binary systems are the most reliable objects to determine absolute stellar parameters (mass, radius, luminosity, etc.) with, based on photometric and spectroscopic observations, which is very important in establishing stellar evolutionary models. Generally, a star's internal structure cannot be observed directly, but it can be interpreted by asteroseismology for pulsating stars. Therefore, binary systems containing at least one pulsating component have greatly attracted the interest of researchers. In addition to the physical parameters obtained by the analysis method of binary systems, the interior structure of the pulsating component can also be analyzed by asteroseismology, which is extremely useful for us in understanding stellar structure, stellar evolution, and the tidal interactions on stellar pulsations (e.g., Fuller 2017; Fuller et al. 2020; Guo 2021).

In the early 1970s, some researchers (Tempesti 1971; Broglia & Marin 1974; McInally & Austin 1977) observed the existence of eclipsing binaries with a pulsating component. However, very few such objects were discovered before the observations made by space telescopes such as CoRot and Kepler over the last 20 years (e.g., Damiani et al. 2010; Southworth et al. 2011; Maceroni et al. 2013; Gaulme & Guzik 2019) and other telescopes on the ground (e.g., Liakos et al. 2012). So far, the number of such targets has reached into the hundreds or more (e.g., Zhou 2010; Liakos & Niarchos 2017, etc.), among which a large number of them (more than 200) are in the δ Sct component in the classical semidetached Algols (e.g., Soydugan et al. 2006a; Zhou 2010; Liakos et al. 2012; Liakos & Niarchos 2017; Kahraman Aliçavuş et al. 2017). This kind of star is defined as an oscillating eclipsing binary of the Algol type (oEA) by Mkrtichian et al. (2002). Of course, the EA-type eclipsing binaries can be semidetached or detached systems. And some pulsating components in EA-type eclipsing binaries are not the primary star or not in semidetached systems, which is different from the original definition of oEA systems.

Generally, the mass transfer from the secondary to primary components (e.g., Guo et al. 2017; Miszuda et al. 2021) makes the primary component of oEA stars appear inside the bottom of the classical Cepheid instability strip and show δ Sct–like oscillations with a frequency range of 3–80 cycles day⁻¹. Thus, the pulsation of the primary component should be associated with the existence of the companion star. Soydugan et al. (2006b) proposed, for the first time, that there is a correlation between the orbital (P_orb) and dominant pulsation (P_pul) periods for oEA stars, while Soydugan et al. (2006a) published a catalog including 25 oEA stars. Liakos et al. (2012) published a catalog including 74 oEA stars and updated the correlation between P_orb and P_pul. Zhang et al. (2013) made a first theoretical explanation for this relationship. Liakos & Niarchos (2017) published a catalog including 199 oEA stars.

In 2018 April, the Transiting Exoplanet Survey Satellite (TESS; Ricker et al. 2015) was launched by NASA as an exoplanet survey mission. It monitored bright stars with a 2 minute short cadence and provided full-frame images every 30 minutes. The primary TESS mission is to search for planets transiting bright and nearby stars over all of the sky. Meanwhile, its high-precision photometric data also provides unprecedented opportunities for the study of binaries and variable stars.

Due to the observational characteristics of the pulsating stars in eclipsing binaries, the high-precision and continuous time-series photometry data of space telescopes have incomparable advantages for the discovery and research of oEA stars. These observational features reveal that the low-amplitude pulsations of the components are contained in the light curve of binary stars, their orbital periods are usually greater than or close to 1 day, and so on. In this paper, we report the detection of new pulsating stars in TESS EA-type eclipsing binaries and make a preliminary analysis of the pulsation properties of these objects.

In the catalog of the International Variable Star Index (VSX; Watson et al. 2006), 95,914 EA-type eclipsing-binary systems were listed by 2022 January 15. We searched these targets in the TESS catalog of 2 minutes short cadence from 1–45 sectors based on the criterion Dist < 5'', where Dist is the distance (in arcseconds) between the two positions determined by the coordinates given in TESS and VSX. It was found that 1626 EA-type eclipsing-binary systems were observed by TESS. We also carried on to cross-match these samples with the catalog of 4584 eclipsing binaries observed by TESS (Prša et al. 2022), and a total of 889 objects are listed in this catalog. The information of the first 50 EA-type binary eclipsing systems are listed in Table 1 and a total of 1626 objects are available in a machine-readable format. In this table, these objects are listed in the order of increasing TESS ID, and the parameters in Columns 2–7 are from the VSX catalog.

We downloaded the simple aperture photometry data of these 1626 EA-type binary systems from the Mikulski Space Telescope Archive (MAST) database, and used the steps described by Shi et al. (2021a, 2021b) to process the original light curves. Then we examined the light curve of each target with the same criteria and searched the published literature. The criteria include an amplitude larger than 50 μmag and an obvious regular variation in the light curves outside eclipse. A total of 57 new eclipsing-binary systems with possible pulsating components were detected. Their light curves are displayed in Figures 1 and 2, where TJD equals BJD minus 2,457,000.0. As shown in these two figures, there are similar pulsating variations in the light curves of each target. Meanwhile, some targets have obvious pulsating variations, but others only can be seen in an enlarged view. The information of these objects with the pulsating components are listed in Tables 2 and 3, while these parameters of Columns 2–4 are from the VSX catalog.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Some EA-type binary systems may have a late-type component, which is likely to be accompanied by magnetic activity, such as KIC 06669809, KIC 10581918, KIC 10619109, KIC 11175495 (Liakos 2017), KIC 6048106 (Lee 2016), and KIC 06852488 (Shi et al. 2021a). In this case, spots may cover the surface of the late-type component and the light curves of the binary system show the O'Connell effect, i.e., the phenomenon that the magnitude of the light maximum near the 0.25 phase is different from that near the 0.75 phase (e.g., Milone 1969; Liu & Yang 1999; Qian et al. 2014). The magnetic activity cycle of stars generally is over years or decades, such as the average period of sunspot activity being 11 yr. Compared with the orbital period of less than 30 days and an observation span of tens of days, the spot modulations are shown as harmonic frequency peaks of the orbital frequency in the Fourier spectrum.

We notice that the light curve of TIC158794976 outside eclipse is continuously changing, which may be a β Lyrae type (EB-type) or W UMa type (EW-type) eclipsing-binary system, i.e., a target misclassified as EA-type eclipsing-binary system. At the same time, we also note that the light curves of TIC299160301 and TIC322208686 are different from other targets. After removing the data during eclipse, their curves are shown in Figure 3, which are similar to typical multiperiod pulsating stars and may be γ Dor in an EA-type eclipsing-binary system.

Zoom In Zoom Out Reset image size

PERIOD04 Lenz & Breger (2005) is a software package based on classical Fourier analysis, which will be used to analyze the orbital frequency and the dominant pulsation frequency of the light curves of 57 new pulsating stars in EA-type eclipsing binaries. The errors of frequency and amplitude are calculated following Montgomery & Odonoghue (1999).

We used the PERIOD04 software to analyze the orbital frequency from the light curves of these 57 new targets, and calculate the corresponding orbital periods. These new orbital periods are listed in Column 6 of Tables 2 and 3, where the errors of the orbital periods are derived from the errors of their orbital frequencies using the error-transfer formula. Here we did not obtain a reliable orbital period for TIC29052041, TIC200135791, TIC233829984, and TIC391643922 from the light curves with a short time span, because the orbital periods of these four targets are more than 10 days. The new orbital periods of three targets are different from that given by VSX: the new orbital periods of TIC56127811 and TIC191221080 are half of that given by VSX, and the new orbital period of TIC266920154 is twice as long as that given by VSX. As can be seen in Tables 2 and 3, the orbital periods of these 57 targets are distributed in the range of 0.4–27 days.

To analyze the dominant pulsation frequency, we excluded the data during eclipse, such as the light curve of TIC139701741 in Figure 4. In this figure, the black open circles and the red solid circles represent the light curve during eclipse and that outside eclipse, respectively. We used the PERIOD04 software to extract and prewhiten the frequency from the light curve outside eclipse, until we obtained a frequency that is not the harmonic of the orbital frequency and can basically fit the most dominant pulsation in the curve. Fourier spectra for the light curve outside eclipse of TIC139701741 are shown in the top panel of Figure 5, while the Fourier spectra after prewhitening all the orbital-harmonic peaks are shown in the bottom panel of Figure 5. In this figure, the red dotted lines represent the harmonics of the orbital frequency, the most prominent pulsation frequency is 16.91 cycles day⁻¹, and the second most prominent frequency of 1.52 cycles day⁻¹ is the fourth-order harmonic of the orbital frequency. The most prominent pulsation frequency has side-lobe frequency peaks that are separated by the orbital frequency, and they can be prewhitened with the prominent pulsation frequency.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The dominant pulsation period and amplitude are listed in Columns 7 and 8 of Tables 2 and 3, where the errors are calculated as described above. As can be seen in those two tables, these 57 targets pulsate in a period range of 0.02–5 days and an amplitude range of 0.09–21 mmag. We could not obtain a pulsation frequency for TIC455509774, which may be due to the aperiodic variation of the light curve. Meanwhile, we also find that the two eclipses of TIC455509774 are not separated by exactly 0.5 in phase, which mean it may has an eccentric orbit.

Except for TIC455509774, the positions of 56 new pulsating stars in EA-type eclipsing binaries on the P_pul–P_orb relation diagram for oEA stars are shown in Figure 6 as red solid circles, where the size of the circles indicates their pulsation amplitude. The orbital periods of TIC29052041, TIC200135791, TIC233829984, and TIC391643922 are from VSX. In this figure, 43 normal objects show a consistent distribution trend with the theoretical relation (black solid line) derived by Zhang et al. (2013), where those objects (green open circles) published by Liakos et al. (2012) are also displayed. As also can be seen in this figure, 13 objects are distinguished from the P_orb–P_pul relation. Among these 13 objects, there are 10 targets pulsating with a frequency of less than 3 cycles day⁻¹, and 3 targets pulsating in the frequency range of δ Sct but deviating from the trend of the P_pul–P_orb relation.

Zoom In Zoom Out Reset image size

Among 95,914 EA-type binaries listed in the catalog of VSX, 1626 EA-type binaries are observed by TESS in a 2 minute short cadence from 1–45 sectors. A total of 57 new pulsating stars are detected from 1626 EA-type binaries, and we analyze their new orbital periods and the dominant pulsation periods. Among the 57 targets, 43 samples show a consistent distribution trend with the P_pul–P_orb relation of oEA stars, which means that they should be oEA stars. Meanwhile, 10 targets pulsate with a frequency of less than 3 cycles day⁻¹, which means that they are likely to be γ Dor, such as TIC299160301 and TIC322208686, or other variable stars in EA-type eclipsing-binary systems. Three targets (TIC191221080, TIC202490797, and TIC80042405) pulsate in the frequency range of δ Sct but deviate from the trend of the P_pul–P_orb relation, which may mean that they are a δ Sct star in eclipsing-binary systems, but maybe with a different formation mechanism from oEA stars.

The light curve of TIC455509774 shows an eccentric eclipse and similar pulsating variation, but could not obtain a pulsation frequency, which may be due to an aperiodic pulsating variation caused by stellar activities, such as spots or chemical-concentration spots. The light curve of TIC158794976 shows continuous variation outside eclipse, which may be an EB- or EW-type eclipsing-binary system. It is an important object for studying the influences of tidal forces on stellar pulsations. It resembles HL Dra (Shi et al. 2021b) where both the primary and secondary components have a high filling factor.

These new objects are very interesting sources for further investigations of binary formation and evolutions and the influences of tidal forces and mass transfer on stellar pulsations. However, many characteristics of these objects are still unclear, and require more spectral or photometric observations in the future.

This work is partly supported by Chinese Natural Science Foundation (Nos. 11933008 and 12103084). The TESS data presented in this paper were obtained from the Mikulski Archive for Space Telescopes (MAST) at the Space Telescope Science Institute (STScI). STScI is operated by the Association of Universities for Research in Astronomy, Inc. Support to MAST for these data is provided by the NASA Office of Space Science. Funding for the TESS mission is provided by the NASA Explorer Program. We are grateful to the anonymous referee for valuable comments and suggestions which have improved the manuscript greatly.