ZTF J021804.16+071152.93: a dead cataclysmic variable and potential solution to the missing period bouncers

Abstract

It is predicted that half or more of all cataclysmic variables (CVs) should have evolved past the period minimum and now exist as so-called "period bouncers" where a white dwarf should be accreting from a Roche-lobe filling substellar companion. However, this prediction stands in stark contrast to observations, where only a few per cent of CVs are found in this evolutionary phase. A potential solution to this discrepancy is that a magnetic field emerges from within the white dwarf after the system has reached the period minimum. The transfer of angular momentum from the spin of the white dwarf into the orbit then pushes the two stars apart, detaching them for potentially billions of years. Here we present the discovery of ZTF J021804.16+071152.93, a detached $0.69\pm0.01 M_{\odot}$, 19 MG magnetic white dwarf plus $37\pm5 M_\mathrm{Jup}$ brown dwarf binary with an orbital period of 1.7 hours. The kinematics of the system indicate that it is a high probability member of the galactic thick disk. However, this strongly disagrees with the much younger age of the system obtained from the white dwarf parameters, implying that the system may have been accreting in the past. This system is therefore consistent with having detached as a result of the emergence of the magnetic field of the white dwarf when the system was still mass transferring, and may represent the ultimate fate for many (perhaps even most) CVs.

Figures (8)

• Figure 1: Left: Gaia DR3 colour-magnitude diagram of the white dwarf cooling track. Red points show confirmed close, detached white dwarf plus brown dwarf binaries Farihi04Farihi17Maxted06Casewell12Casewell18Casewell20Beuermann13Parsons17bParsons25Roestel21, ZTF J0218+0711 is shown as a blue star, demonstrating that the white dwarf is considerably cooler than in other known systems and shows no clear optical excess from the companion to the white dwarf. Right: phase-folded ZTF light curves of ZTF J0218+0711 in the $g$ band (top) and $r$ band (folded on a period of 1.7 hours). The light curve is consistent with no variability outside of the deep eclipse of the white dwarf.
• Figure 2: Circular polarisation (V/I) spectra of 7 observations of ZTF J0218+0711 showing variability, which is caused by the rotation of the white dwarf. The top panel shows all spectra overlaid over the full wavelength range, with the black box marking the area of clearest variability. The bottom-left panel shows the polarised spectra corresponding to the box in the top panel in phase with the orbital ephemeris. The bottom-right panel shows the mean of the absolute polarisation values in phase with the orbital ephemeris. The data appears periodic on a timescale consistent with the orbital period, however, a clear period cannot be determined given the limited baseline of the observations.
• Figure 3: Normalised, averaged X-shooter spectrum of ZTF J0218+0711 binned by a factor of 5 (black line). Overplotted are the Zeeman split components as a function of the magnetic field strength for the H$\beta$ line (orange) and H$\gamma$ (green). The dips in the spectrum agree well with a field strength of around 19 MG.
• Figure 4: Trailed spectra of the H$\alpha$ emission line originating from the companion of the white dwarf in ZTF J0218+0711. Left: original data, where there is a time jump between spectrum 6 and 7, middle: our best fit model to the data, right: the residuals of the fit. Colours are normalised flux density, as shown by the colourbars.
• Figure 5: GTC+HiPERCAM eclipse light curves of ZTF J0218+0711 (top to bottom: $u_s$, $g_s$, $r_s$, $i_s$ and $z_s$ bands) along with the best fit models (black lines). Residuals to the fits are shown beneath each panel (in standard deviations).