Pulsational frequencies of the eclipsing delta-Scuti star HD 172189
J. E. S. Costa, E. Michel, J. Pena, O. Creevey, Z. P. Li, M. Chevreton, J. A. Belmonte, M. Alvarez, L. Fox Machado, L. Parrao, F. Perez Hernendez, A. Fernandez, J. R. Fremy, S. Pau, R. Alonso
aa r X i v : . [ a s t r o - ph ] J un Astronomy&Astrophysicsmanuscript no. costajes c (cid:13)
ESO 2018October 31, 2018
Pulsational frequencies of the eclipsing δ Scuti star HD 172189.Results of the STEPHI XIII campaign.
J. E. S. Costa , E. Michel , J. Pe˜na , O. Creevey , Z. P. Li , M. Chevreton , J. A. Belmonte , M. Alvarez , L. FoxMachado , L. Parrao , F. P´erez Hern´endez , A. Fern´andez , J. R. Fremy , S. Pau , and R. Alonso Observatoire de Paris, LESIA, FRE 2461, 92195 Meudon, France Instituto de Astronom´ıa - Universidad Nacional Aut´onoma de M´exico, Ap.P. 877, Ensenada, BC, Mexico Instituto de Astrof´ısica de Canarias, 38200 La Laguna, Tenerife, Spain National Astronomical Observatories, Chinese Academy of Sciences, Beijing 100012, ChinaReceived 00.00.0000 / Accepted 22.02.2007
ABSTRACT
Context.
The eclipsing δ Scuti star HD 172189 is a probable member of the open cluster IC 4756 and a promising candidate targetfor the CoRoT mission.
Aims.
The detection of pulsation modes is the first step in the asteroseismological study of the star. Further, the calculation of theorbital parameters of the binary system allows us to make a dynamical determination of the mass of the star, which works as animportant constraint to test and calibrate the asteroseismological models.
Methods.
We performed a detailed frequency analysis of 210 hours of photometric data of HD 172189 obtained from the STEPHIXIII campaign ⋆ . Results.
We have identified six pulsation frequencies with a confidence level of 99% and a seventh with a 65% confidence level,in the range between 100 − µ Hz. In addiction, three eclipses were observed during the campaign, allowing us to improve thedetermination of the orbital period of the system.
Key words.
Stars: oscillations (including pulsations) — (Stars: variables:) δ Sct — (Stars:) binaries: eclipsing
1. Introduction
HD 172189 (IC 4756 93 = SAO 123754) is a binary star ofvisual magnitude m V = .
85 and spectral type A2 ( α = h m . s , δ = + o ′ . ′′ ) in the galactic clusterIC 4756. Recently, Mart´ın-Ruiz et al. . (2005) (M-R, hereafter)showed that HD 172189 is an eclipsing binary star with an or-bital period of 5 .
71 days and with δ Scuti-type pulsations witha clear frequency of 19 . − (226 . µ Hz), and evidence ofmore modes in the range 208 − µ Hz.The δ Scuti variables are stars with masses from 1.5 to 2.5 M ⊙ , located at the intersection of the lower part of the classicalCepheid instability strip with the main sequence. They consistof main sequence objects, pre-main sequence ones, and also ob-jects that have evolved o ff the main sequence, in the hydrogenshell burning phase. Most of the δ Scuti stars are multiperiodic,with frequencies between 50 and 600 µ Hz. The pulsation modesare radial as well as non-radial due to κ -mechanism associatedwith the zone where He is partially ionized. Thus, δ Scuti starsprovide a good opportunity to apply asteroseismology and studykey mechanisms at work in the main sequence stage, amongwhich are transport of angular momentum, transport of chem-ical species at the edge of convective cores (overshooting), andlarge-scale circulations.A few weeks of observations are usually required to resolvetheir rich spectra, often showing close pairs of peaks with a
Send o ff print requests to : J.E.S. Costa ⋆ The HD 172189 reduced light curves are available in theCDS via anonymous ftp to cdsarc.u-strasbg.fr (130.79.128.5) or viahttp: // cdsweb.u-strasbg.fr / cgi-bin / qcat?J / A + A / . spacing of ∼ . µ Hz. Three week-long multisite observationalcampaigns are regularly organized within the STEPHI network(Michel et al. 1992; Michel et al. 2000), trying to minimize theproblem of missing data due to the day-night cycle, which in-duces strong side lobes of the spectral window in the Fourierspectrum.The frequency analysis of HD 172189 shows that it is a δ Scuti with at least six pulsation frequencies. Eclipses that oc-curred during the campaign were observed, allowing us to refinethe period of the binary system.The star HD 172189 was found to be a ”likely member” ofthe open cluster IC 4756 by Kop ff (1943). Herzog et al. (1975)updated the results of a previous work of Seggewiss (1968), con-cluded that the probability of the membership to the cluster forHD 172189, based on its proper motion alone, is 92%. Missana& Missana (1995) re-studied the cluster IC 4756, estimating theprobability of membership of each stars and arrived to a similarresult for the case of HD 172189: 95% of probability of being amember of the cluster.The interest of pulsating stars in cluster and / or in well char-acterized binaries for asteroseismology studies has been stressedfor long and by several authors. As mentioned in a recent reviewby Aerts (2006) (see also Aerts and Harmanec 2004), beside theperspective to study processes specific to binarity, like the influ-ence of tidal forces on angular momentum transport or eventualmass transfer, the possibility to determine precisely global pa-rameters (like individual masses, radii, chemical composition orage) allows to reduce significantly the parameters space to ex-plore in the necessary stellar modeling work. The seismic inter- Costa J.E.S. et al.: HD 172189 — Results of the STEPHI XIII campaign. pretation is thus expected to become more constraining in termsof diagnostics on physical processes.In this spirit, Brown et al. (1994) applied singular value de-convolution technique to study the relative impact of precisionon eigenfrequencies and global parameters in the illustrativecase of two solar-like pulsators in a visual binary. They con-sidered the optimal case of the close Alpha Cen binary. Theirconclusions necessarily remained theoretical, since no oscilla-tions were detected at that time neither in Alpha Cen A nor inAlpha Cen B. But this seminal work found a great developmentin Miglio and Montalb´an (2005) where an optimization usingLevenberg-Marquardt algorithm is applied to Alpha Cen A + B,considering all available observables, including oscillation fre-quencies now detected in both objects. This work confirms thatsatisfying all the observables with present physical assumptionsis becoming more and more challenging and ineluctably leads toquestion the description of key-physical processes like mixinglength theory versus full spectrum theory. Miglio and Montalb´anhowever conclude that the present seismic data do not allow togo further and discriminate between di ff erent physical optionsfor the equation of state or element di ff usion descriptions.In the specific case of δ Scuti stars, Creevey et al. (2006)proposed to develop the approach of Brown et al. (1994), in thespecific case of eclipsing binaries featuring a δ Scuti. Their workis however, so far limited only to non seismic observables.A few δ Scuti pulsators in eclipsing binaries have alreadybeen intensively searched for oscillations. For RZ Cas andAB Cas (Rodr´ıguez et al. 2004a; Rodr´ıguez et al. 2004b), theseobservations only brought one oscillation frequency, and theydid not lead to a seismic interpretation so far. RS Cha is anotherexample of eclipsing binary featuring pre-main sequence ob-jects, one of them showing δ Scuti type oscillations (Alecian etal. 2006). Recently, new observations have been organized suc-cessfully, showing apparent multiperiodicity of both component(Bohm et al. In prep.).The fact that HD 172189 is a δ Scuti star and at the sametime a binary system and a member of the open cluster IC 4756makes it a very interesting target for the CoRoT mission.
2. The observations
The STEPHI campaigns include three observatories around theworld: Observatorio del Teide (Iza˜na, Tenerife Island, Spain),Observatorio Astron´omico Nacional de San Pedro M´artir (BajaCalifornia, Mexico) and the Xinglong Station of the BeijingObservatory (China). The observations were carried out for 21days over the period 7-28 June 2004. The B9 star HD 172248(IC 4756 117 = SAO 123762) was used as the primary compari-son star (comp1) with a brightness of m V = .
91. Two secondarycomparison stars (comp2) were observed: IC 4756 142 at theMexican observatory and IC 4756 115 at the Spanish observa-tory. The first one is an A3 star with m V = .
51 while the latteris a star of spectral type G0 with m V = . + comparison 1 + comparison 2), and the fourth channel is devotedto measuring the sky level.A total of 210 hours of photometric data were obtained dur-ing the 21 days of the campaign. The integration time was 1 sec-ond. This corresponds to a duty cycle of 40% which is typicalfor the STEPHI campaigns.A log of the campaign is given in Table 1. The digits in thedata set names (in the first column) indicate the date of the startof the observations and the two last letters indicate the site of Hour
E1E2 E3E407080910111213141516171819202122232425262728 XL ZN SP 6 4 −6 −8 −10 12 10 8 2 0 −2 −4 D a t e ( J un ) U T Primary eclipse
Fig. 1.
Coverage diagram of the STEPHI XIII campaign. Eachrectangular strip represents a night of observation. The labelsXL, ZN and SP indicate the three observatories. The total num-ber of hours of photometric data is 210 with an e ff ective cover-age of 40%. The times of minimum of the primary eclipses areindicated by E1, E2, E3 and E4.observation: XL = Xinglong, ZN = Obs. del Teide (Iza˜na), andSP = San Pedro M´artir. Hereafter, we will use these three ab-breviations to represent the three sites. Column 2 gives the UTdate at the start of the observations and columns 3 and 4 give theHJD time (Heliocentric Julian Date) at the start and end of theobservations. The last two columns show the number of usefulmeasurements and the observation length (in hours). The cov-erage diagram of the campaign is shown in Fig. 1. The lettersin the top of the figure indicate the positions of the observationsfrom the three sites. The rectangular strips represent the duration(in hours) of the observations in relation to 0h UT of the date inthe vertical axis. The diagram was made this way to be consis-tent with Table 1. Note that the observations from each site arealigned inside the same approximate range of time. The times ofminimum of the primary eclipses discussed in Sec. 5 are indi-cated in the figure.
3. Data reduction
The data reduction followed similar steps as reported in pre-vious STEPHI campaigns (see, for instance, Hernandez et al.1998). We start with the inspection of the light curve and, whennecessary, elimination of bad points and anomalous parts of thedataset (including the parts where the eclipses are occurring).In the second step we proceed to the calibration using measure-ments of the sky background simultaneously taken by all the fourchannels, usually taken at the beginning and at the end of theobservations. The third step is the subtraction of the sky back-ground. The measurements of the sky background are subtractedfrom the light curve of each channel. However, due to technicalproblems, in this campaign the photometer sky channel couldnot be used for the sky monitoring on some nights. The fourth osta J.E.S. et al.: HD 172189 — Results of the STEPHI XIII campaign. 3Data set Date T begin T end Number Durationname 245 0000 . +
245 0000 . + of (hours)(HJD) (HJD) pointsIC0607ZN 07-Jun-2004 3164.51865 3164.70145 18111 5.05IC0607SP 07-Jun-2004 3164.78305 3164.94251 11327 3.61IC0608ZN 08-Jun-2004 3165.44773 3165.70545 10067 2.97IC0608SP 08-Jun-2004 3165.74305 3165.97506 19082 5.37IC0609XL 09-Jun-2004 3166.19328 3166.30368 9499 2.65IC0609ZN 09-Jun-2004 3166.46177 3166.70003 20586 5.72IC0609SP 09-Jun-2004 3166.72632 3166.96818 20407 5.67IC0610XL 10-Jun-2004 3167.07915 3167.30217 19267 5.35IC0610ZN 10-Jun-2004 3167.43952 3167.69994 22485 6.25IC0610SP 10-Jun-2004 3167.72274 3167.97030 16098 4.50IC0611ZN 11-Jun-2004 3168.44450 3168.69808 21910 6.09IC0611SP 11-Jun-2004 3168.69851 3168.97172 23592 6.56IC0612XL 12-Jun-2004 3169.08350 3169.29475 16178 5.07IC0612ZN 12-Jun-2004 3169.43251 3169.69744 22884 6.36IC0612SP 12-Jun-2004 3169.69386 3169.97170 23435 6.52IC0613ZN 13-Jun-2004 3170.43495 3170.70100 22986 6.39IC0613SP 13-Jun-2004 3170.69817 3170.97177 23591 6.57IC0614ZN 14-Jun-2004 3171.42569 3171.69950 23605 6.57IC0614SP 14-Jun-2004 3171.74160 3171.96188 18238 5.07IC0615ZN 15-Jun-2004 3172.41819 3172.69828 24191 6.72IC0615SP 15-Jun-2004 3172.69699 3172.96155 22845 6.35IC0616ZN 16-Jun-2004 3173.41763 3173.69907 24256 6.75IC0616SP 16-Jun-2004 3173.69528 3173.95722 22574 6.29IC0617ZN 17-Jun-2004 3174.41541 3174.69871 24476 6.80IC0617SP 17-Jun-2004 3174.68134 3174.96363 24302 6.77IC0618ZN 18-Jun-2004 3175.41786 3175.69765 24172 6.72IC0618SP 18-Jun-2004 3175.71757 3175.96352 20072 5.58IC0619ZN 19-Jun-2004 3176.42122 3176.69892 23990 6.67IC0619SP 19-Jun-2004 3176.68105 3176.96442 24959 6.96IC0620SP 20-Jun-2004 3177.67549 3177.96542 24959 6.96IC0621SP 21-Jun-2004 3178.68041 3178.96473 22922 6.40IC0622SP 22-Jun-2004 3179.67059 3179.96465 24506 6.84IC06S3SP 23-Jun-2004 3180.68252 3180.96416 24180 6.76IC0624SP 24-Jun-2004 3181.67754 3181.95713 23220 6.46IC0625SP 25-Jun-2004 3182.67586 3182.96803 25104 7.01IC0626SP 26-Jun-2004 3183.68674 3183.70567 23307 6.52IC0627SP 27-Jun-2004 3184.67712 3184.89731 21940 6.14IC0628SP 28-Jun-2004 3185.67427 3185.96532 14789 4.34 Table 1.
Log of the observations of the STEPHI XIII campaign. The letters XL, ZN and SP in the data set names indicate the siteof observation.step is the division of the light curve of the variable star (var)by the primary comparison star light curve, var / comp1, and bythe secondary comparison star light curve, var / comp2. The pur-pose of this step is to minimize the e ff ect of changes in the skytransparency on long time scales.In order to remove low-frequency trends which a ff ect the de-tection of pulsation modes, the next step is to fit and subtract apolynomial of low-order (order ≤
2) from the light curve. Thelast step is to divide the whole light curve by the average valueand subtract 1. The resulting light curve is the relative variationin the magnitude of the star in relation to its mean magnitude.Most of the light curves show maximum variations around 0 . ∼
5% relative to the mean inten-sity. Finally, all times are converted to heliocentric Julian dates(HJD).
4. Frequency analysis
In Fig. 2 we show the periodograms of the light curves ofvar / comp1 and var / comp2 for San Pedro M´artir and Tenerifedata sets for the range of 0 − µ Hz with amplitudes given in ppt (part per thousand). The respective spectral windows areshown on the right side.To our surprise, the periodogram of the light curve of theTenerife data shows the presence of a series of at least 12 har-monics of the frequency of 1 cd − f d = . µ Hz). This isnot the case for the San Pedro M´artir data, although the reduc-tion procedure is the same for the two data sets. We discardedthe possibility of these harmonics being an artifact of an elec-tronic problem in one of the channels because they are present inthe periodograms of the light curves of var / comp1, var / comp2and comp2 / comp1. The harmonics, perhaps, are resulting of anoccasional change in the sky transparency at Iza˜na during thecampaign.In the periodogram of the SP light curve var / comp1, we canclearly see peaks of pulsation modes with high amplitudes andfrequencies within the range 150 − µ Hz, completely sepa-rated from the bump of peaks of low frequencies ( f < µ Hz).The same peaks appear in the periodogram of the light curveof var / comp2 of the same site, proving that they are related tothe target star and not to the comparison star. The di ff erence be-tween the relative heights of the bumps in the two cases can Costa J.E.S. et al.: HD 172189 — Results of the STEPHI XIII campaign.
Frequency ( µ Hz)
024 Tenerife −− var/comp20 50 100 150 200 250 300 350024 A m p li t ude s ( pp t ) Tenerife −− var/comp10 50 100 150 200 250 300 350024 San Pedro Martir −− var/comp20 50 100 150 200 250 300 350024
Periodograms of HD 172189 −− STEPHI Jun/2004
San Pedro Martir −− var/comp1 −50 0 50Frequency ( µ Hz) −50 0 50−50 0 50−50 0 50
Spectral Windows
Fig. 2.
Periodograms of the one-site light curves of var / comp1and var / comp2 for the San Pedro M´artir and Tenerife data. In theright side are shown the respective one-site spectral windows. µ Hz) −6−5−4−3−2−10123456 A m p li t ude ( pp t ) HD 172189 − STEPHI Jun/2004 − Tenerife Data
Fig. 3.
Periodograms of the Tenerife data obtained with polyno-mial fitting of order 2 (top) and order 3 (bottom).be explained by the di ff erence in spectral type: the variable starHD 172189 is an A2 star and the first comparison star (comp1)is of spectral class B9, while the second one (comp2) is a G0star.The presence of the harmonics complicates the identifica-tion of the peaks of pulsation frequencies in the spectrum. Wetried a new modification in the data reduction procedure: we fit-ted and subtracted a polynomial of order three (instead of ordertwo) from the Tenerife data from each night. This procedure wase ff ective in eliminating all the peaks with high amplitude in thelow-frequency region of the periodograms (0-100 µ Hz) and theharmonics of the frequency of the day, without any apparent ef-fect over the peaks with higher amplitude in the pulsation rangeof frequencies (100 − µ Hz) as can be seen in Fig. 3.In Fig. 4 we show the spectral window (top) and the pe-riodograms of the whole light curve (ZN + SP + XL) ofvar / comp1 (middle) and var / comp2 (bottom), where we cansee the presence of pulsation modes in both cases. The centralpeak of the spectral window was arbitrarily positioned to coin-cide with the position of the highest peak in the periodogram.The FWHM of the central peak is 0 . µ Hz and the two nearest
Frequency ( µ Hz) A m p li t ude ( pp t ) Average in amplitude99% confidence level40% confidence level
Original f f f f f f f Fig. 5.
Prewhitening process in HD 172189 (var / comp1). Ineach panel the peaks above the 99% confidence level (dashedline) are selected and removed from the original light curve, to-gether with all previous selected frequencies, and a new peri-odogram is obtained. The selected peaks are indicated by f , f ,etc. The lower level of 40% is indicated by the dotted curve andthe solid curve represents the average amplitude.sidelobes have ∼
40% of its amplitude. All subsequent sidelobeshave heights below ∼
4% of the amplitude of the central peak.The pulsation frequencies appear in both periodograms, in therange 150 − µ Hz.To find pulsation frequencies we used the usual iterative ap-proach: starting with an empty list of candidate frequencies andthe periodogram of the original light curve (var / comp1); (1) in-side the region of interest in the amplitude spectrum we identifythe peaks with the highest confidence levels (taking care to dis-card aliases). If there is no peak with a significant probabilitythe algorithm stops. (2) Put the selected frequencies in the listof candidate frequencies; (3) using a non-linear method, fit si-nusoidals with all frequencies of the list of frequencies to theoriginal light curve. The fitting refines the values of the initialfrequencies and calculates the amplitudes and phases, as well asthe respective uncertainties. (4) The fitted sinusoidals are sub-tracted from the original light curve and the periodogram of theresidual light curve is calculated, and return to step (1) to searchfor additional possible pulsation frequencies.Before applying the algorithm on the light curve var / comp2,we used a high-pass filter to remove signal at low frequencies (¡100 mu Hz). As in previous STEPHI articles (e.g. ´Alvarez et al.1998 Fox-Machado et al. 2002), the confidence levels are cal-culated with Fisher’s test as prescribed by Nowroozi (1967). Wesearch for oscillation peaks in the range 100 − µ Hz, whichin Nowroozi’s description corresponds to around m = . q ¯ A f , which is fitted by the exponential ¯ A f = c exp( c f ), where c and c are constants and f is the frequency.The result of the prewhitening process in HD 172189 isshown in Fig. 5. The dashed curve in each graph indicates the99% confidence level, adopted as a detection limit while the dot-ted curve indicates the position of the 40% confidence level. Inthe periodogram of the original light curve, we found three peaks osta J.E.S. et al.: HD 172189 — Results of the STEPHI XIII campaign. 5 µ Hz) A m p li t ude s ( pp t ) HD 172189 − STEPHI Jun/2006
Spectral windowvar/comp1var/comp2
Fig. 4.
Spectral window (top) and periodograms of the whole light curves of HD 172189 of var / comp1 (middle) and var / comp2(bottom). f i Frequency Amplitude Confidence( µ Hz) (ppt) Level f . ± .
003 2 . ± .
02 99% f . ± .
006 0 . ± .
02 99% f . ± .
006 0 . ± .
02 99% f . ± .
008 0 . ± .
02 99% f . ± .
010 0 . ± .
02 99% f . ± .
010 0 . ± .
02 99% f . ± .
012 0 . ± .
02 65%
Table 2.
Detected pulsation frequencies in HD 172189 with con-fidence levels ≥ f , f , and f . In the two successive prewhitened periodogramswe found three more frequencies above this level, f , f and f .After the prewhitening of the six frequencies, only one peak wasfound with a confidence level upper the 40% probability limit,with a 65% confidence level. The six detected pulsation frequen-cies and the seventh frequency with lower probability are givenin Table 2.We search for the seven detected peaks in the periodogramof the light curve of var / comp2 and using a least square fittingwe estimated the probability of each peak not being due to noisein the periodogram of this light curve: 99 .
9% for f ; 92% for f ;96% for f ; 80% for f ; 30% for f ; 45% for f ; and 35% for f .As expected, the probabilities found here are lower than thoseones found for the case of var / comp1 because the light curve ofvar / comp2 is more noisy. Even so, the four highest peaks areclearly confirmed in this test.M-R found two pulsation modes in HD 172189: 226 . µ Hzand 218 . µ Hz. The first mode is in agreement with f , but thesecond one does not agree with our frequencies. However, weidentify the peak at 218 . µ Hz as an alias of f .
5. Eclipses in HD 172189
The binarity of HD 172189 was discovered from analysis of thedata of a campaign carried out in 1997 for detecting γ Doraduspulsating stars in the cluster IC 4756 (Mart´ın 2000; Mart´ın2003). M-R analyzed Str¨omgren uvby − H β observations ob-tained in di ff erent campaigns during 1997, 2003, and 2004, showed the presence of three light minima due to eclipses, butonly one of the eclipses was followed past the minimum inbrightness with a depth of ∼ .
12 mag. From the data analysis,they found a period of P = . T min I = . + . · E (in HJD), where E is the integer number of primary eclipsessince the epoch 245 2914 .
644 HJD. The results indicate a sys-tem with an orbital eccentricity of e ≃ .
24, a longitude of theperiastron of ω ≃ o and an inclination of i ≃ o . The analysisalso suggests that the two stars have di ff erent radii ( r / r ≃ . T e ff ;1 ≃ . T e ff ;2 .According to the ephemeris, there were four primary eclipsesduring the present campaign, but, as shown in Fig. 1, only theeclipses E E E IC0608ZN , IC0619SP and
IC0625SP , respectively (cf. Tab 1).Fig. 6 shows the positions of the eclipses in the light curveof HD 172189. The seven detected pulsation frequencies weresubtracted and points collected within 60 s were binned. Thethree primary eclipses are clearly visible.Fig. 7 shows the light curves during the eclipse. In the firsttwo panels we see the light decreasing, passing through a mini-mum in brightness and then increasing. The primary eclipse E .
520 (HJD). Unfortunately, the ob-servations were interrupted several times during the night of thiseclipse and the reduction of the data presented some di ffi culties.A better scenario is shown in the second graph of Fig. 7 forthe primary eclipse E
3, where the light curve changes slowlyand the minimum is well defined. Fitting a polynomial curveof degree 3 to the eclipse part of the light curve, we obtain T min I = . ff ers by ∼ σ fromthe predicted value using the ephemeris from M-R, indicatingthat the uncertainty in the orbital period is underestimated. Usingthis new minimum and the previous ephemeris, we recalculatedthe orbital period to obtain P = . E T min I = . + . · E (in HJD). (1)Table 3 shows the ephemeris dates for the four eclipses. Thecalculated date for the eclipse E . Costa J.E.S. et al.: HD 172189 — Results of the STEPHI XIII campaign.
Date (HJD−2450000.) −0.10−0.050.000.050.100.15 ∆ m a g HD 172189 − STEPHI Jun/2004E1 E3 E4
Fig. 6.
Light curve of HD 172189. The measurements werebinned to 60 seconds. Three primary eclipses, E E E −0.050.000.050.10 ∆ m a g Eclipse E1
Dates (HJD−3453000)Eclipse E3
Eclipse E4
Fig. 7.
The three primary eclipses ( E E E
4) observed dur-ing the campaign. The points were binned and the pulsation fre-quencies were subtracted.
Eclipse Date (HJD)E1 245 3165 . . . . Table 3.
Primary eclipses of HD 172189.depth for the primary eclipses E E ∆ mag = . ∆ mag = . ∆ mag = .
12 for the eclipse observed by M-R.The light curves show that the duration of each primaryeclipse is ∼ seven hours and the half-depths occur around onehour before and after the instant of minimum. From the orbitalelements calculated by M-R, we calculated a phase of ∼ . ∼ .
75 days ( ∼
18 hours) after each primary eclipse and havethe same duration. We searched for minima in the light curvearound these dates, but as was remarked by M-R in their analy- sis, no minimum with an unequivocal and statistically significantdepth was found.
6. Conclusions
The STEPHI XIII campaign carried out in June 2004, allowed astudy of the pulsating behavior of the δ Scuti star HD 172189.The Fourier analysis of the light curve from three sites showsthe presence of six pulsation frequencies with a confidence levelof 99%. An additional frequency with a lower probability of65% was also found. The observation of three eclipses duringthe campaign — two of them including the minimum in inten-sity — allows us to refine the previous value for orbital periodof the system, obtaining 5.70165(8) days.With the detection of six pulsation frequency with a highconfidence level, our present campaign confirms the interest ofHD 172189 as eclipsing binary featuring a δ Scuti pulsator forseismology. Besides the number of modes detected, the inter-est of HD 172189 Compared with other objects like RZ Cas(Rodriguez et al. 2004a) or AB Cas (Rodriguez et al. 2004b) isalso associated with its orbital period. Both, AB Cas and RZ Cas,reveal a short orbital period ( ∼ ff ect of tidal forces on the star shape and structure, po-tential e ff ects on oscillations. In fact, both these objects are con-sidered Algol-type binaries, which supposes that mass transferoccurred between companions and changed their evolution com-pared with single stars. With a 5.7d orbital period, HD 172189is expected to be more representative of a ”normal” single star,both for its pulsational behavior and for its evolution history.Dedicated spectroscopic and photometric observations areunder way to characterize precisely the orbital parameters andradii of HD 172189 members. These results when available,completed by our present oscillation frequencies analysis willallow to start the modeling and seismic analysis of this objecttaking advantage of these numerous constraints. In addition tothis, within a few years now, it is planed to observe HD 172189with CoRoT for five months. This is expected to allow detec-tion of many more oscillation modes by pushing down the noiselevel by a factor 500 in the power spectrum (see Michel et al.2006). Beside this, the fact that HD 172189 belongs to the clus-ter IC 4756, even if this one is not so well studied yet, o ff ers aninteresting potential source of complementary information forfurther developments. Acknowledgements.
This work received financial support from the SpanishDGES (ESP2004-03855-C03-03), the Chinese National Natural ScienceFoundation under grant number 10573023 and 10433010, and the Brazilianagency - Conselho Nacional de Desenvolvimento Cient´ıfico e Tecnol´ogico,CNPq. The 1.5 m Carlos S´anchez Telescope is operated on the island of Tenerifeby the Instituto de Astrof´ısica de Canarias at the Spanish Observatorio del Teide.
References
Alecian, E., Goupil, M.-J., Lebreton, Y., et al. 2006, in Proc. IAU Symp. 240- Binary Stars as Critical Tools and Tests in Contemporary Astrophysics,2006, Prague, Czech Republic, p.176´Alvarez, M., Hern´andez, M. M., Michel, E., et al. 1998. A&A, 340, 149Aerts, C. and Harmanec, P. 2004, in ASP Conf. Ser. 318: Spectroscopically andSpatially Resolving the Components of the Close Binary Stars, 2004, eds.Hilditch, R. W., Hensberge, H. and Pavlovski, K., p. 325Aerts, C. 2006, in Proc. IAU Symp. 240 - Binary Stars as Critical Tools and Testsin Contemporary Astrophysics, 2006, Prague, Czech Republic, p.40Brown, T.M., Christensen-Dalsgaard, J., Weibel-Mihalas, B., Gilliland, R. L.1994, ApJ, 427, 1013Creevey, O. L., Brown, T. M., Jim´nez-Reyes, S., Belmonte, J. A. 2006, in ASPConference Ser. 349: Astrophysics of Variable Stars, eds. C. Sterken and C.Aerts, p.387 osta J.E.S. et al.: HD 172189 — Results of the STEPHI XIII campaign. 7