Eclipse time variations & the continued search for companions to short-period eclipsing binary systems

2020 December 4

George Faillace, David Pulley, John Mallett, Americo Watkins, Ian Sharp, Xinyu Mai

Introduction

From observations of eclipse time variations (ETVs), many claims have been made of the detection of circumbinary objects orbiting subdwarf B (sdB) binary systems, members of the HW Vir family, and binary systems where the primary has evolved into a white dwarf. Typically, these systems have a very hot primary component with temperatures in excess of 30,000K, and a secondary M dwarf or brown dwarf companion with temperatures of 3,500K or less. The separation between the two components is usually less than one solar radius, causing the secondary to be heavily irradiated by the primary star. This gives rise to significant amounts of reflected energy from the secondary.

An HW Vir-type system’s compact structure, short periods and large temperature differences between the two components give rise to short and well-defined primary eclipses, allowing times of minima to be determined with high precision. These systems have undergone a post-common envelope binary (PCEB) evolution, as described in the Appendix, with many showing apparent periodic variations in their eclipse timings. An overview is provided by Zorotovic & Schreiber (2013) and Lohr et al. (2014).^1,2

In this paper we consider three PCEB systems – HS 0705+6700, NSVS 14256825, and NN Ser – as well as one evolving into a W UMa-type of system: NSVS 01286630. Claims have been made for the presence of circumbinary objects, with calculated parameters, orbiting all four systems. Of all the putative objects discovered through ETVs, those around NN Ser were thought to be amongst the most compelling because of the high quality of the data and because its main-sequence companion is a late-M star, which restricts the possibility of other causes for period changes, for example magnetic coupling. With our new observations we investigated the periodic variation in the position of the barycentres of these four systems, to see if they fit with previous predictions of circumbinary orbits. While the NN Ser and NSVS 14256825 systems’ planets are listed in various international databases, e.g. NASA Exoplanet Archive, neither of the other two systems have this level of recognition, calling into question their proposed circumbinary hypotheses.

While the more exciting explanation of attributing these period variations to the presence of planets or brown dwarfs orbiting the systems has been a popular consideration, other factors could also explain their cyclical behaviour (see last paragraph of the ‘Observing method & data reduction’ section which follows). Other claims of circumbinary objects are for close eclipsing binary systems of the W UMa-type or binary systems evolving to a W UMa once their Roche lobes are filled, such as NSVS 01286630, the fourth system described herein.

We discuss the possibility that these other factors could explain the cyclical behaviour of these four systems, attributed in the literature to circumbinary objects. We note that two of the systems (HS 0705+6700 and NSVS 14256825) have been included in our recent study of seven short-period eclipsing binaries (Pulley et al., 2018).³ As will be seen, our recent observations, which cover two additional seasons, support the conclusion of further deviations from previous predictions. In this paper we present the ETVs exhibited by four eclipsing binary systems at somewhat different stages of binary evolution, and we analyse the results in the context of circumbinary planet hypotheses. The analysis is preceded by a brief historical review of the hypotheses presented by earlier observers. The four systems studied are listed with their parameters in Tables 1 & 2.

Observing method & data reduction

In Table 3 we list the telescopes used to obtain the new data and in Table 4 we report 108 new observations of the four eclipsing binary systems between 2017 May and 2019 September. The effects of atmospheric extinction were minimised by making all observations at altitudes greater than 40°. All images were calibrated using dark, flat, and bias frames, then analysed with Maxim DL or Astroart.^4,5 The source flux was determined with aperture photometry using a constant aperture for all images, and the radius scaled according to the full width at half maximum (FWHM). Variations in observing conditions were accounted for by determining the flux relative to an ensemble of comparison stars in the field of view. The apparent magnitude of the target was derived from those of the comparison stars and its average magnitude calculated by the software.

This was done as follows. Firstly, the same comparison stars were used for each image. Using the average derived magnitude of the target from each comparison star and the standard deviation of the average, the final value for the target was obtained for each frame. The magnitudes of the comparison stars were chosen, appropriate to the filter being used. The comparison stars’ catalogue magnitudes for the various filters were taken from the American Association of Variable Star Observers (AAVSO) Photometric All Sky Survey (APASS) catalogue.⁶ These were similar to the target magnitudes and, whenever possible, had similar colour indices to the target stars. Because the APASS catalogue does not include the R pass band, in the few cases where observations were taken with the R filter a conversion formula recommended by the AAVSO was used to transform the catalogue’s Sloan r′ magnitudes to the corresponding R magnitudes. Whether observations were performed with or without filters, check stars were used to ensure that there was no variability in the reference star selected.

All of our new mid-image timings used in this analysis were first converted to Barycentric Julian Date Dynamical Time (BJD_TDB) using the Ohio State University or the Astropy time utilities.^7,8 Computer clocks were synchronised with external atomic clocks during the imaging process.

(Login or click above to view the full illustrated article in PDF format)