Discovery of ZZ Cetis in detached white dwarf plus main-sequence binaries
S. Pyrzas, B. T. Gaensicke, J. J. Hermes, C. M. Copperwheat, A. Rebassa-Mansergas, V. S. Dhillon, S. P. Littlefair, T. R. Marsh, S. G. Parsons, C. D. J. Savoury, M. R. Schreiber, S. C. C. Barros, J. Bento, E. Breedt, P. Kerry
aa r X i v : . [ a s t r o - ph . S R ] N ov Mon. Not. R. Astron. Soc. , 1–8 (2014) Printed 3 September 2018 (MN L A TEX style file v2.2)
Discovery of ZZ Cetis in detached white dwarf plusmain-sequence binaries
S. Pyrzas ⋆ , B. T. G¨ansicke , J. J. Hermes , C. M. Copperwheat , A. Rebassa-Mansergas , V. S. Dhillon , S. P. Littlefair , T. R. Marsh , S. G. Parsons ,C. D. J. Savoury , M. R. Schreiber , S. C. C. Barros , J. Bento , E. Breedt and P. Kerry Instituto de Astronom´ıa, Universidad Cat´olica del Norte, Avenida Angamos 0610, Casilla 1280, Antofagasta, Chile Department of Physics, University of Warwick, Coventry, CV4 7AL, UK Astrophysics Research Institute, Liverpool John Moores University, Twelve Quays House, Birkenhead CH41 1LD, UK Kavli Institute for Astronomy and Astrophysics, Peking University, Beijing 100871, China Department of Physics and Astronomy, University of Sheffield, Sheffield S3 7RH, UK Departamento de F´ısica y Astronom´ıa, Universidad de Valpara´ıso, Avenida Gran Breta˜na 1111, Valpara´ıso, Chile Aix Marseille Universit´e, CNRS, LAM, UMR 7326, 13388, Marseille, France Department of Physics and Astronomy, Macquarie University, NSW 2109, Australia
Accepted . Received ; in original form
ABSTRACT
We present the first results of a dedicated search for pulsating white dwarfs (WDs) indetached white dwarf plus main-sequence binaries. Candidate systems were selectedfrom a catalogue of WD+MS binaries, based on the surface gravities and effectivetemperatures of the WDs. We observed a total of 26 systems using ULTRACAMmounted on ESO’s 3.5 m New Technology Telescope (NTT) at La Silla. Our photo-metric observations reveal pulsations in seven WDs of our sample, including the firstpulsating white dwarf with a main-sequence companion in a post common envelope bi-nary, SDSS J1136+0409. Asteroseismology of these new pulsating systems will providecrucial insight into how binary interactions, particularly the common envelope phase,affect the internal structure and evolution of WDs. In addition, our observations haverevealed the partially eclipsing nature of one of our targets, SDSS J1223-0056.
Key words: stars: white dwarfs - stars: low-mass - binaries: close - binaries: eclipsing- asteroseismology
The class of ZZ Ceti stars comprises single, hydrogen-atmosphere (DA), photometrically variable whitedwarfs (WDs) (see e.g. McGraw 1977; Clemens 1993;Mukadam et al. 2006) that exhibit non-radial g-modepulsations (Chanmugam 1972; Warner & Robinson 1972).ZZ Ceti stars are tightly grouped together in the
ZZ Cetiinstability strip (e.g. Bergeron et al. 1995; Koester & Allard2000; Mukadam et al. 2004a; Gianninas et al. 2006;Van Grootel et al. 2012), a well defined region in the effec-tive temperature ( T eff )-surface gravity (log g ) plane, between11,100 and 12,600 K for a log g = 8 WD (Gianninas et al.2011), with a dependence on log g (Giovannini et al. 1998).One of the fundamental questions regarding the instabilitystrip is its assumed purity (e.g. Gianninas et al. 2005, ⋆ E-mail: [email protected] and thus, studies of ZZ Cetis can be applied to theentire population of DA white dwarfs. The means towardsthis end is asteroseismology.Asteroseismology, the study of stellar pulsations, is apowerful tool to gather direct information about the inte-riors of stars. In the past few decades it has been success-fully adapted to WDs to constrain the hydrogen layer mass, An impure strip doesn’t necessarily imply the opposite. Impu-rities could arise if a third parameter is in effect, as is the casefor cataclysmic variables, explained later in this Section. In sucha case, the third parameter needs to be taken into account whendefining the strip.c (cid:13)
S.Pyrzas et al.
Table 1.
Information on all our target WD+MS binaries. We provide names, coordinates, SDSS u, g, r magnitudes ∗ , T eff , log g and M WD values. The penultimate column gives the classification of each target (see text for details), while the last column gives the lengthof the observations.SDSS J RA [J2000] Dec [J2000] u g r T eff a [K] log g b M WD c [M ⊙ ] Class Obs. Length [hrs]0017-0024 00:17:26.63 -00:24:51.1 19.66 19.21 18.96 14594 ± . ± .
30 0 . ± .
18 WDMS 2.460021-1103 00:21:57.90 -11:03:31.6 19.22 18.57 18.10 11045 ±
164 8 . ± .
10 1 . ± .
05 WDMS 2.890052-0051 00:52:08.42 -00:51:34.6 19.04 18.30 17.72 12300 ±
427 8 . ± .
12 0 . ± .
08 WDMS 6.82 ±
461 7 . ± .
25 0 . ± .
11 WDMS 2.850124-0023 01:24:03.11 -00:23:01.1 20.29 18.88 17.75 11972 ± . ± .
44 0 . ± .
10 WDMS 2.020203+0040 02:03:51.28 +00:40:25.1 20.32 19.43 18.67 10794 ±
475 8 . ± .
36 0 . ± .
22 WDMS 2.930212+0018 02:12:39.45 +00:18:56.9 19.69 19.22 18.87 13852 ± . ± .
47 0 . ± .
27 WDMS 2.480218+0057 02:18:49.98 +00:57:39.2 20.23 19.67 19.15 12250 ± . ± .
59 0 . ± .
29 UNKN 2.630255-0044 02:55:09.29 -00:44:14.7 20.76 19.69 18.74 13182 ± . ± .
61 0 . ± .
27 WDMS 2.710327-0022 03:27:58.15 -00:22:15.4 20.47 19.48 18.63 12902 ±
874 7 . ± .
31 0 . ± .
17 WDMS 3.170328+0017 03:28:42.92 +00:17:49.7 19.06 18.03 16.16 ±
514 7 . ± .
15 0 . ± .
07 WDMS 2.390336-0047 03:36:56.12 -00:47:27.8 20.09 19.49 19.23 15246 ± . ± .
24 0 . ± .
13 WDMS 3.100345-0614 03:45:14.71 -06:14:21.2 20.16 19.31 18.51 13904 ± . ± .
27 0 . ± .
17 PCEB 3.610824+1723 ±
358 8 . ± .
18 0 . ± .
12 WDMS 2.521043+0603 ±
299 8 . ± .
20 0 . ± .
13 WDMS 2.731054+1008 10:54:36.18 +10:08:37.3 18.98 18.61 18.58 11433 ±
229 7 . ± .
13 0 . ± .
09 UNKN 1.871117-1255 11:17:10.54 -12:55:40.9 20.30 19.65 19.25 11302 ±
357 8 . ± .
24 0 . ± .
15 WDMS 1.181136+0409 11:36:55.17 +04:09:52.6 17.57 17.07 17.17 11699 ±
152 7 . ± .
08 0 . ± .
05 PCEB 1.051223-0056 12:23:39.61 -00:56:31.2 18.41 17.91 18.06 11565 ±
159 7 . ± .
11 0 . ± .
06 PCEB ± . ± .
73 1 . ± .
38 WDMS 1.621329+2557 13:29:33.67 +25:57:43.2 20.09 18.75 17.75 11565 ± . ± .
44 0 . ± .
26 UNKN 0.711453+0010 14:53:05.77 +00:10:48.2 19.71 18.90 18.21 11565 ±
233 8 . ± .
09 0 . ± .
05 WDMS 1.931520+0634 15:20:33.43 +06:34:42.9 19.47 18.91 18.52 11302 ±
234 8 . ± .
12 0 . ± .
09 WDMS 1.391615+2357 16:15:05.51 +23:57:46.3 19.02 18.23 17.66 12250 ±
595 8 . ± .
12 0 . ± .
08 UNKN 1.991652+1340 16:52:40.74 +13:40:15.0 20.95 19.98 19.11 11173 ±
101 8 . ± .
44 0 . ± .
25 UNKN 1.311724+0733 17:24:45.28 +07:33:24.7 19.51 18.86 18.35 13588 ±
527 8 . ± .
12 0 . ± .
08 WDMS 1.85 ∗ These magnitudes are on the SDSS 2.5 m photometric system. Note, however, that ULTRACAM filters are primed,i.e. u ′ , g ′ , r ′ , i ′ , as they are closer to the USNO 40-in system (Fukugita et al. 1996; Smith et al. 2002). a,b,c Values obtained using the updated spectral decomposition/fitting technique described in detail in Rebassa-Mansergas et al. (2012). Total time over two observing nights ; This is an i magnitude, and the system was observed in the i ′ -band ; Exposure time for u ′ -band was 40 sec ; Exposure time for all bands was 10 sec ; Originally mis-classified as a WDMS, see Sec. 4.1 ; Total time the depth of the degenerate core boundary, the overall stel-lar mass and temperature, the mass of any other chemi-cally stratified layers, and the rotation rates of these stel-lar remnants (Winget & Kepler 2008; Fontaine & Brassard2008; Althaus et al. 2010).As the end-point of all low-mass stars, WDs are also fre-quently found in binary configurations, such as cataclysmicvariables (CVs), close interacting binaries in which the WDprimary accretes from a low-mass companion. With the dis-covery of ZZ Ceti-type pulsations in the WD primary of theCV GW Lib (Warner & van Zyl 1998; van Zyl et al. 2000,2004), the analytic power of asteroseismology has becomeavailable to determine accurate stellar parameters for theWDs in these binaries, which is otherwise very difficult,if not impossible, to achieve, as the light from the WD iscontaminated with emission from the accretion disc. Aster-oseismology also offers the opportunity to study how theaccretion of mass and angular momentum onto the WDsin CVs can affect the WD structure, shedding light on thepotential single-degenerate scenario for SN Type-Ia progen-itors. Most of the (non-magnetic) WDs in CVs were not ex-pected to be pulsating, as accretion heats them to effectivetemperatures greater than 12000 K (Townsley & G¨ansicke2009). The work of Townsley et al. (2004) and Arras et al.(2006) suggested that the atmospheric composition of ac-creting WDs, especially the He abundance, the accretion of heavier elements and the rapid rotation of the WD primariesin CVs could significantly affect the pulsation modes. Thereare currently 16 known pulsating WDs in CVs. Most of themshow a single pulsation mode, and thus it has not been possi-ble to derive stellar parameters through asteroseismology forany of them. Their corresponding instability strip seems tobe wider than the one for non-interacting ZZ Cetis, rangingfrom 10500 K . T eff . > T eff andlog g .The occurrence of pulsating WDs in CVs raises thequestion of the existence of pulsating WDs in the pro-genitors of CVs and, generally, in non-interacting binaries,such as detached white dwarf plus main-sequence binaries(WD+MS). In recent years, owing mainly to the Sloan Dig-ital Sky Survey (SDSS, York et al. 2000), WD+MS binarieshave been discovered in large numbers (e.g. Silvestri et al.2005; Heller et al. 2009; Rebassa-Mansergas et al. 2010,2012; Liu et al. 2012; Rebassa-Mansergas et al. 2013), offer-ing the exciting possibility for ensemble asteroseismology c (cid:13) , 1–8 ulsating WDs in WDMS binaries studies of a large and homogeneous sample of WDs in bina-ries. About two-thirds of the SDSS WD+MS binaries arewide enough that the WD progenitors have evolved as if theywere single stars (e.g. Willems & Kolb 2004; Schreiber et al.2010; Nebot G´omez-Mor´an et al. 2011). Thus, pulsatingWDs in such systems should exhibit properties very similarto the single ZZ Cetis. This hypothesis remains effectivelyuntested by observations; to our knowledge, there is onlyone confirmed ZZ Ceti in a wide common proper motion bi-nary, G117-B15A (e.g. Kepler et al. 1991).The rest of the WD+MS binaries are expected to bepost-common envelope binaries (PCEBs), i.e. close binariesthat have formed through common envelope (CE) evolution(e.g. Webbink 2008, for a review). The CE phase is believedto be the main mechanism for the formation of low mass,He-core WDs (e.g Paczynski 1976; Iben & Tutukov 1986).While the presence of such WDs in PCEBs hasbeen established observationally, in both double-degenerate(Marsh et al. 1995; Marsh 1995; Steinfadt et al. 2010;Parsons et al. 2011; and see also Nelemans & Tout 2005and references therein) and single-degenerate (i.e. with aMS companion) configurations (Marsh & Duck 1996; Bruch1999; Pyrzas et al. 2012; Parsons et al. 2012 and see alsoSchreiber & G¨ansicke 2003 and references therein), pulsat-ing He-core WDs have resisted detection (Steinfadt et al.2012) and have been discovered only very recently, in double-degenerate configurations (extremely low mass WDs (ELM),Hermes et al. 2012, 2013b,a). Previous to our work, no pul-sating WDs in single-degenerate PCEBs were known.The study of pulsating WDs in PCEBs can providecrucial insight into whether, and if so how, the internalstructure of WDs, and by extension the pulsation proper-ties, are affected by the common envelope phase. Whitedwarfs in single-degenerate PCEBs can be found mainlyin two “flavours”, those with M WD < .
45 M ⊙ and a He-core and those with M WD > . ⊙ and a C/O core (e.g.Pyrzas et al. 2009; Rebassa-Mansergas et al. 2011); the jux-taposition of the two classes can provide fertile ground forstudies on the potential effects of mass loss on the WD struc-ture and pulsations. Furthermore, these systems can pop-ulate the gap in WD mass between the normal ZZ Cetis(with M WD = 0 . ⊙ and above) and the ELMs (with M WD < .
25 M ⊙ ), leading to a complete census of pulsat-ing WDs.In this paper we present the first results from a dedi-cated search for pulsating WDs in WD+MS binaries. Thestructure is as follows: our target selection, photometric ob-servations and period analysis techniques are described inSec. 2 and 3 respectively, while Section 4 presents our results.We discuss our findings in Sec. 5 and conclude in Sec. 6 witha summary and suggestions for future work. Targets were selected from the white dwarf plus main-sequence binaries catalogue by Rebassa-Mansergas et al.(2012). Systems were chosen on the basis of their log g and T eff values, as determined from their SDSS spectra, us-ing the spectral decomposition/fitting technique describedin detail in Rebassa-Mansergas et al. (2007) (and see also Rebassa-Mansergas et al. 2012). Table 1 lists basic informa-tion on all our targets and the observations. Based on avail-able radial velocity (RV) information, either from the origi-nal SDSS spectroscopy or from follow-up observations, these26 targets can be divided into three groups:(i) PCEB: systems for which spectroscopy reveals radialvelocity variations (close binaries)(ii) WDMS: systems for which spectroscopy reveals noradial velocity variations (wide binary candidates)(iii) UNKN: systems for which no or insufficient spec-troscopy is availableWhile the detection of short period RV variations un-ambiguously identifies the PCEBs among our targets, theabsence of RV variations is not a guarantee that a systemis a wide binary. Insufficient spectral resolution, low orbitalinclination or an unlucky orbital phase sampling can preventthe identification of a PCEB (see e.g. Schreiber et al. 2010,for a discussion). An example is SDSS J1223-0056, whoseavailable spectroscopy did not show any RV variations andthe system was classified as a WDMS. However, our subse-quent photometric observations revealed the system to beeclipsing (see Sec. 4.1); the orbital period was determined tobe P orb = 2 . Photometric light curves of our targets were obtained dur-ing two observing runs in December 2010 and May 2011.We used the high-speed camera ULTRACAM (Dhillon et al.2007) mounted as a visitor instrument on ESO’s 3.5 m NewTechnology Telescope (NTT) at La Silla Observatory, Chile.Each target was observed in full-frame mode, for a contin-uous block of time, but varying in length depending on theschedule. The exposure time was 20 s, with a dead-time be-tween exposures of ∼
25 ms. ULTRACAM is a triple-beamcamera, so data were obtained simultaneously in the Sloan u ′ , g ′ and r ′ bands. In one case an i ′ filter was used insteadof r ′ for scheduling reasons. All of the data were reducedwith aperture photometry using the ULTRACAM pipelinesoftware, with debiassing, flat-fielding and sky backgroundsubtraction performed in the standard way. The fluxes of thetargets were determined using a variable aperture, wherebythe radius of the aperture is scaled according to the fullwidth at half-maximum of the stellar profile. Variations intransparency were accounted for by dividing each light curveby the light curve of a nearby comparison star. The stabilityof these comparison stars was checked against other stars inthe field, and no variations were seen. For the period analysis and the detection of pulsations,each light curve was first converted into fractional inten-sity (dividing by the mean and subtracting 1). Subsequently, c (cid:13) , 1–8 S.Pyrzas et al.
Table 2.
Summary of the pulsation properties of the seven ZZ Ceti white dwarfs. We provide the frequency, the corresponding periodand the amplitude of each detection for all three ULTRACAM arms, as well as the corresponding detection thresholds. A dash indicatesno detection.System Frequency [ µ Hz] Period [s] Amplitude [mma ] Det. Thresh. [mma]SDSS J u ′ g ′ r ′ u ′ g ′ r ′ u ′ g ′ r ′ u ′ g ′ r ′ - 928 . ± .
73 932 . ± .
84 - 1077 . ± . . ± . . ± .
21 - - 1093 . ± . . ± .
30 1131 . ± .
15 1134 . ± .
55 884 . ± . . ± . . ± . . ± .
87 1583 . ± .
51* 1581 . ± .
21 633 . ± . . ± . . ± . . ± .
49 1714 . ± .
21 1704 . ± .
06 582 . ± . . ± . . ± . . ± .
45 1960 . ± .
44 1955 . ± .
10 512 . ± . . ± . . ± . . ± .
47 2723 . ± . a - 366 . ± . . ± . . ± .
87 1044 . ± .
78* 1049 . ± .
40 962 . ± . . ± . . ± . . ± . a . ± .
41 1483 . ± .
14 685 . ± . . ± . . ± . . ± .
56 2510 . ± .
69 - 398 . ± . . ± . . ± .
34 1012 . ± .
91 1031 . ± .
18 989 . ± . . ± . . ± . . ± .
58 1238 . ± .
30 1190 . ± .
92 848 . ± . . ± . . ± . . ± .
82 - b . ± .
33 761 . ± . . ± . . ± .
35 1603 . ± .
75* 1626 . ± .
69 629 . ± . . ± . . ± . . ± .
83 1967 . ± .
11 - 513 . ± . . ± . . ± .
71 1567 . ± .
69 - 634 . ± . . ± . . ± .
20* 3082 . ± .
70 - 324 . ± . . ± . . ± .
97 - - 292 . ± . . ± . † . ± . a - 164 . ± . . ± . . ± .
51* 1248 . ± . a - 835 . ± .
80 801 . ± .
10 - 21.9 13.1 77.2 10.9 13.61136+0409 - 3616 . ± .
31* 3656 . ± .
36 - 276 . ± . . ± . . ± .
91 - - 182 . ± . mma: milli-modulation amplitude, a 0.1% relative amplitude change ; Two observations - solution for each independent observation* Pre-whitened signal ; a Only > σ detection ; b Barely 1 σ detection ; † First harmonic of the 3080.21 signal
Figure 1.
The results of our campaign. The 26 systems plotted on the T eff - log g plane. Pulsating systems are shown with green, filledsymbols; non-pulsating systems with red, open symbols (colour version available only online). Circles denote PCEB, squares WDMS andtriangles UNKN binaries, following Table 1. The dashed lines are the boundaries of the ZZ Ceti instability strip found by Gianninas et al.(2011). c (cid:13) , 1–8 ulsating WDs in WDMS binaries Figure 2.
The seven pulsating WDs: g ′ -band light curves (left panels) and the corresponding amplitude spectra (right panels) for eachsystem, identified in the right panels. Tickmarks indicate the significant detections for each system, as listed in Table 2. we calculated amplitude spectra using the TSA package(Schwarzenberg-Czerny 1993) as implemented in
MIDAS .In order to judge the significance of a signal, we calcu-lated a detection threshold in the following fashion (see alsoGreiss et al. 2014): for each light curve we created artificialdata sets using a shuffling technique (see e.g. Kepler 1993;Schreiber 2007, for discussions), where each intensity point f i is randomly re-assigned to a time point t j with i = j ; allintensity and time points are used. This shuffling destroysany coherent signal, but retains the time- and overall noiseproperties of a light curve. We then calculated the ampli-tude spectra of the shuffled light curves and recorded thevalue of highest amplitude. Using the results from 10,000shufflings, we calculated the amplitudes corresponding tothe 68.3, 95.5 and 99.7 per cent confidence levels (1-, 2- and3- σ respectively) and set the 3 σ amplitude as our detectionthreshold. Signals with amplitudes above this threshold wereconsidered significant. Finally, the frequencies and errors ofsignificant detections were determined using the bootstrapmethod (Press 2002).When strong signals are present, using the maximumamplitude to determine the threshold typically leads to over-estimated threshold values. Therefore, our analysis was car-ried out as follows: for each light curve, we used our shuffling technique to calculate the detection threshold and to ensurethat the strongest signal present in the power spectrum isindeed a 3- σ detection. Subsequently, we prewhitened thissignal, and subjected the light curve to a second shuffling, inorder to calculate the revised detection threshold and lookfor additional significant signals. We did not proceed withfurther iterations, as we have only one relatively short lightcurve per target. Our photometric observations and subsequent period anal-ysis revealed the pulsating nature of seven white dwarfs inour sample, while the remaining 19 were found to be non-variable at our detection threshold. Our results are illus-trated in Figure 1.Table 2 lists the seven systems with > σ signal detec-tions, along with the corresponding frequencies, periods, am-plitudes and detection thresholds in each filter, while Fig. 2shows the g ′ -band light curves and the corresponding am-plitude spectra of all seven pulsating systems.In Table 3 we list all the non-pulsating systems alongwith the corresponding detection thresholds in each of the c (cid:13) , 1–8 S.Pyrzas et al.
Figure 3.
Sample g ′ -band light curves (left panels) and their corresponding amplitude spectra (right panels) of two of our non-pulsatingWDs, SDSS J1520+0634 (top panels) and SDSS J1615+2357 (bottom panels). Red lines indicate the 3 σ detection threshold. Table 3.
Detection thresholds for the 19 white dwarfs where nopulsations were detected.System Threshold [mma]SDSS J u ′ g ′ r ′ three filters, while Fig. 3 shows two sample g ′ -band lightcurves and their respective amplitude spectra.The periods observed in our systems are compara-ble to those observed in typical pulsating WDs (e.g.Mukadam et al. 2006). All our systems are detached, sowe do not expect to see any accretion-related variability,such as flickering and QPO’s. In detached systems, photo-metric variability could be caused by ellipsoidal modulationor irradiation/reflection effects (see e.g. Pyrzas et al. 2009;Parsons et al. 2010). However, these forms of variability aremodulated on longer timescales (effectively on the orbitalperiod, with two maxima and one maximum per orbit re-spectively) than the periods observed in our sample, andhave a distinctive qualitative imprint on the light curves,very different to the observed variability. Henceforth, we will assume that each > σ detection corresponds to a pulsa- Figure 4.
Phase-folded g ′ -band lightcurve of SDSS J1223-0056,using the orbital ephemeris Eq. 1. Inset panel: a zoom-in versionaround the eclipse itself. tion mode, although repeated detections of each pulsationwould be required for unambiguous confirmation. Our photometric observations revealed that the white dwarfin SDSS J1223-0056 undergoes partial eclipses. Measuringthe times of mid-eclipse from four observed eclipses, we de-termine the orbital ephemeris of the system to beBMJD (TDB) = 55706 . E ∗ . P orb = 2 . g ′ -band lightcurve of the new eclipsing system.SDSS J1223-0056 has also been detected as an eclipsing sys-tem in data from the Catalina Real-time Transient Survey(see Parsons et al. 2013). Inspection of Tab. 2 and 3 reveals that in the majority ofcases, our u ′ -band detection threshold is much higher thanthe g ′ - and r ′ -band, since the quality of the data in thisband is compromised by the faintness of the targets and thelack of good comparison stars in the fields. As a result, insome of the pulsating systems there is no u ′ -band detectionof the pulsations, although the respective g ′ - and r ′ -banddetections are unambiguous. c (cid:13) , 1–8 ulsating WDs in WDMS binaries Our survey for pulsations generally has a relatively lowsensitivity; the median value for WDs not observed to varyis 6.3 mma in the g ′ -band, only slightly below the medianamplitude of 8.8 mma for detected pulsations in a sample of35 ZZ Cetis discovered from SDSS (Mukadam et al. 2004b).Pulsation signals could also be diluted if the M-dwarf com-panion contributes a significant amount of flux. Thus, it ispossible that we might have missed some pulsators amongthese seemingly non-variable systems.For our seven pulsating systems, the frequencies of thedetected signals are consistent with those of g-mode pulsa-tions observed in single WDs (e.g. Mukadam et al. 2006).The multicolour amplitudes of the pulsations (particularlyin the cases where u ′ -band pulsations are also detected) areindicative of ℓ = 1 , T eff and log g of these WDs at face value, two of our pulsating systems(SDSS J0111+0009 and J0203+0040) lie nominally outsidethe Gianninas et al. (2011) instability strip, while six sys-tems with no detections lie inside it. However, it is impor-tant to bear in mind that the parameters determined fromthe spectral decomposition (Rebassa-Mansergas et al. 2010,2012) have rather large statistical errors (evident in the er-rors bars of Fig. 1) and can be subject to systematic uncer-tainties (see Parsons et al. 2013). In WD+MS binaries, thesubtraction of the M-dwarf component from the spectrumdirectly affects the disambiguation between the “hot” and“cold” solutions of the subsequent WD fit. It is possible thatfor some of our systems we have selected the “wrong” solu-tion. Also, log g values are overestimated for those systemswith T eff < g values. However, since these 3D correctionshave not yet been implemented in the context of an empir-ical ZZ Ceti instability strip, we have not adopted them inour analysis. We have carried out the first dedicated survey to iden-tify pulsating white dwarfs in detached WD+MS binaries.Among a sample of 26 such systems, selected based on the T eff and log g values of their WDs obtained from spectralfitting, we have idenified seven new pulsating white dwarfs.One of these WDs is found in SDSS J1136+0409, a confirmedsingle-degenerate post-common-envelope binary, which con-stitutes the first detection of such kind.For the immediate future, work needs to be carried outon multiple fronts: (i) high signal-to-noise spectroscopy ofall our targets with wide wavelentgh coverage (across theBalmer jump) in order to pinpoint their exact location onthe T eff - log g plane, (ii) higher signal-to-noise photometrictime series of the seemingly non-variable systems and (iii) follow-up photometric observations of all pulsating WDs inorder to identify candidates for precision asteroseismology. ACKNOWLEDGEMENTS
We thank the referee Anjum Mukadam for a construc-tive report. Based on observations made with ESO tele-scopes at the La Silla Paranal Observatory under pro-gramme ID 086.D-0555 and 087.D-0557. The research lead-ing to this results has received funding from the Euro-pean Research Council under the European Union’s Sev-enth Framework Programme (FP/2007-2013)/ERC GrantAgreement n.320964 (WDTracer). SP gratefully acknowl-edges support from an ALMA-CONICYT grant (31110019).ARM acknowledges financial support from the Postdoc-toral Science Foundation of China (grant 2013M530470)and from the Research Fund for International Young Scien-tists by the National Natural Science Foundation of China(grant 11350110496). VSD, SPL and ULTRACAM are sup-ported by the STFC. TRM and EB acknowledge the sup-port of STFC under grant ST/L000733/1. SGP acknowl-edges support from FONDECYT in the form of grant num-ber 3140585. MRS acknowledges support from FONDECYT(1141269).
REFERENCES
Althaus, L. G., C´orsico, A. H., Isern, J., Garc´ıa-Berro, E.,2010, A&AR, 18, 471Arras, P., Townsley, D. M., Bildsten, L., 2006, ApJ Lett.,643, L119Bergeron, P., Wesemael, F., Lamontagne, R., Fontaine, G.,Saffer, R. A., Allard, N. F., 1995, ApJ, 449, 258Bergeron, P., Fontaine, G., Bill`eres, M., Boudreault, S.,Green, E. M., 2004, ApJ, 600, 404Bruch, A., 1999, AJ, 117, 3031Castanheira, B. G., Kepler, S. O., Kleinman, S. J., Nitta,A., Fraga, L., 2010, MNRAS, 405, 2561Castanheira, B. G., et al., 2007, A&A, 462, 989Chanmugam, G., 1972, Nat, 236, 83Clemens, J. C., 1993, Baltic Astronomy, 2, 407Dhillon, V. S., et al., 2007, MNRAS, 378, 825Fontaine, G., Brassard, P., 2008, PASP, 120, 1043Fukugita, M., Ichikawa, T., Gunn, J. E., Doi, M., Shi-masaku, K., Schneider, D. P., 1996, AJ, 111, 1748Gianninas, A., Bergeron, P., Fontaine, G., 2005, ApJ, 631,1100Gianninas, A., Bergeron, P., Fontaine, G., 2006, AJ, 132,831Gianninas, A., Bergeron, P., Ruiz, M. T., 2011, ApJ, 743,138Giovannini, O., Kepler, S. O., Kanaan, A., Wood, A.,Claver, C. F., Koester, D., 1998, BaltA, 7, 131Greiss, S., G¨ansicke, B. T., Hermes, J. J., Steeghs, D.,Koester, D., Ramsay, G., Barclay, T., Townsley, D. M.,2014, MNRAS, 438, 3086Heller, R., Homeier, D., Dreizler, S., Østensen, R., 2009,A&A, 496, 191Hermes, J. J., Montgomery, M. H., Winget, D. E., Brown,W. R., Kilic, M., Kenyon, S. J., 2012, ApJ Lett., 750, L28 c (cid:13) , 1–8 S.Pyrzas et al.
Hermes, J. J., et al., 2013a, MNRAS, 436, 3573Hermes, J. J., et al., 2013b, ApJ, 765, 102Iben, Jr., I., Tutukov, A. V., 1986, ApJ, 311, 742Kepler, S. O., 1993, BaltA, 2, 515Kepler, S. O., et al., 1991, ApJ Lett., 378, L45Koester, D., Allard, N. F., 2000, BaltA, 9, 119Liu, C., Li, L., Zhang, F., Zhang, Y., Jiang, D., Liu, J.,2012, MNRAS, 424, 1841Marsh, T. R., 1995, MNRAS, 275, L1Marsh, T. R., Duck, S. R., 1996, MNRAS, 278, 565Marsh, T. R., Dhillon, V. S., Duck, S. R., 1995, MNRAS,275, 828McGraw, J. T., 1977, The ZZ Ceti stars: A new class ofpulsating white dwarfs, Ph.D. thesis, Texas Univ., Austin.Mukadam, A. S., Winget, D. E., von Hippel, T., Mont-gomery, M. H., Kepler, S. O., Costa, A. F. M., 2004a,ApJ, 612, 1052Mukadam, A. S., Montgomery, M. H., Winget, D. E., Ke-pler, S. O., Clemens, J. C., 2006, ApJ, 640, 956Mukadam, A. S., et al., 2004b, ApJ, 607, 982Nebot G´omez-Mor´an, A., et al., 2011, A&A, 536, A43Nelemans, G., Tout, C. A., 2005, MNRAS, 356, 753Paczynski, B., 1976, in P. Eggleton, S. Mitton, & J. Whe-lan, ed., Structure and Evolution of Close Binary Systems,vol. 73 of
IAU Symposium , p. 75Parsons, S. G., Marsh, T. R., Copperwheat, C. M., Dhillon,V. S., Littlefair, S. P., G¨ansicke, B. T., Hickman, R., 2010,MNRAS, 402, 2591Parsons, S. G., Marsh, T. R., G¨ansicke, B. T., Drake, A. J.,Koester, D., 2011, ApJ Lett., 735, L30Parsons, S. G., et al., 2012, MNRAS, 420, 3281Parsons, S. G., et al., 2013, MNRAS, 429, 256Press, W. H., 2002, Numerical recipes in C++ : the art ofscientific computing, Cambridge Univ. Press, CambridgePyrzas, S., et al., 2009, MNRAS, 394, 978Pyrzas, S., et al., 2012, MNRAS, 419, 817Rebassa-Mansergas, A., G¨ansicke, B. T., Rodr´ıguez-Gil, P.,Schreiber, M. R., Koester, D., 2007, MNRAS, 382, 1377Rebassa-Mansergas, A., G¨ansicke, B. T., Schreiber, M. R.,Koester, D., Rodr´ıguez-Gil, P., 2010, MNRAS, 402, 620Rebassa-Mansergas, A., Nebot G´omez-Mor´an, A.,Schreiber, M. R., Girven, J., G¨ansicke, B. T., 2011,MNRAS, 413, 1121Rebassa-Mansergas, A., Nebot G´omez-Mor´an, A.,Schreiber, M. R., G¨ansicke, B. T., Schwope, A., Gallardo,J., Koester, D., 2012, MNRAS, 419, 806Rebassa-Mansergas, A., Agurto-Gangas, C., Schreiber,M. R., G¨ansicke, B. T., Koester, D., 2013, MNRAS, 433,3398Robinson, E. L., et al., 1995, ApJ, 438, 908Schreiber, M. R., 2007, A&A, 466, 1025Schreiber, M. R., G¨ansicke, B. T., 2003, A&A, 406, 305Schreiber, M. R., et al., 2010, A&A, 513, L7Schwarzenberg-Czerny, A., 1993, in P. Grosbol & R. deRuijsscher, ed., 5. ESO/ST-ECF Data Analysis Work-shop, vol. 47 of
ESO Conference and Workshop Proceed-ings , p. 149Silvestri, N. M., Hawley, S. L., Oswalt, T. D., 2005, AJ,129, 2428Smith, J. A., et al., 2002, AJ, 123, 2121Steinfadt, J. D. R., Kaplan, D. L., Shporer, A., Bildsten,L., Howell, S. B., 2010, ApJ Lett., 716, L146 Steinfadt, J. D. R., Bildsten, L., Kaplan, D. L., Fulton,B. J., Howell, S. B., Marsh, T. R., Ofek, E. O., Shporer,A., 2012, PASP, 124, 1Szkody, P., et al., 2010, ApJ, 710, 64Townsley, D. M., G¨ansicke, B. T., 2009, ApJ, 693, 1007Townsley, D. M., Arras, P., Bildsten, L., 2004, ApJ Lett.,608, L105Tremblay, P.-E., Ludwig, H.-G., Steffen, M., Freytag, B.,2013, A&A, 559, A104Van Grootel, V., Dupret, M.-A., Fontaine, G., Brassard,P., Grigahc`ene, A., Quirion, P.-O., 2012, A&A, 539, A87van Zyl, L., Warner, B., O’Donoghue, D., Sullivan, D.,Pritchard, J., Kemp, J., 2000, BaltA, 9, 231van Zyl, L., et al., 2004, 350, 307Warner, B., Robinson, E. L., 1972, Nat, 239, 2Warner, B., van Zyl, L., 1998, in F.-L. Deubner,J. Christensen-Dalsgaard, & D. Kurtz, ed., New Eyes toSee Inside the Sun and Stars, vol. 185 of
IAU Symposium ,p. 321Webbink, R. F., 2008, in E. F. Milone, D. A. Leahy, &D. W. Hobill, ed., Astrophysics and Space Science Library,vol. 352 of
Astrophysics and Space Science Library , p. 233Willems, B., Kolb, U., 2004, A&A, 419, 1057Winget, D., Kepler, S. O., 2008, ARA&A, 46, 157York, D. G., et al., 2000, AJ, 120, 1579This paper has been typeset from a TEX/ L A TEX file preparedby the author. c (cid:13)000