ap-6-11.dvi Acta Polytechnica Vol. 51 No. 6/2011 Cataclysmic Variables — X-rays and Optical Activity in V1223Sgr and V709Cas R. Gális, L. Hric, E. Kundra, F. Münz Abstract Intermediatepolars are amajor fraction of all cataclysmic variables detectedby INTEGRAL inhardX-ray. These objects have recently been proposed to be the dominant X-ray source population detected near the Galactic centre, and they also contribute significantly to X-ray diffuse Galactic ridge emission. Nevertheless, only 25% of all known intermediate polars have been detected in hard X-ray. This fact can be related to the activity state of these close interacting binaries. A multi-frequency (from optical to X-ray) investigation of intermediate polars is essential for understanding the physical mechanisms responsible for the observed activity of these objects. Keywords: Cataclysmic variables, optical and X-ray variability, mass transfer and accretion. 1 Introduction Cataclysmic variables (CVs)manifest strong activity in a whole spectrum from radio up to γ-rays. CVs are close binary systems consisting of a hot white dwarf (WD) and a red main-sequence star of spec- tral type M or K, which fills the volume of its inner Roche lobe and transfers matter to the vicinity of the WD [1]. The mass transfer between components causes the observed activity of theCVs, which varies from relatively small light variations (flickering) to enormous photometric changes (outbursts of novae) on time scales that cover awide range from very fast variability of fractions of seconds (flickering) to long- term variations of several years or decades (activity cycles). According to strength of the WD magnetic field, the transferredmatter creates anaccretiondisk (clas- sical CVs) or follows magnetic lines and falls to the surface of the WD (magnetic CVs). Magnetic CVs are a small sub-set of all catalogued CV systems, and fall into two categories: polars and intermedi- ate polars (IPs). In IPs, the WD magnetic field (106 − 107 G) is not strong enough to disrupt the disc entirely (as in the case of polars) and it simply truncates the inner part of the disc [2]. An accretion flow is channelled down towards the magnetic poles and onto the WD surface. Increased interest in CVs was aroused after the discovery of the very hard X-ray emission (up to 80 keV) of many IPs. When the transferred ma- terial impacts the WD atmosphere, a strong shock will form above its surface [3]. The temperature in the post-shock region (PSR) can be very high and the plasma is cooled mainly via optically thin bremsstrahlung radiation in the hardX-ray band [4]. Aswe showed in our previous analysis [5], the broad- band spectra (3–100keV) of the studied IPs can be well fitted by a thermal bremsstrahlung model with PSR temperature kT ≈ (20–25)keV. Reflection of thebremsstrahlungphotons at anoptically thick cold medium can also contribute to the hard X-ray spec- trum [6]. In the case of V2400 Oph, a significant emission excess detected at ≈ 26keV can be caused by the reflection of X-rays from the surface of the WD [5]. In total, 32 CVs (and 2 symbiotic systems) were detected in the data of INTEGRAL [7–9,5]. This is more than was expected, and it represents ≈ 5% of all detections of this space observatory. The sample of IPs detected in the (20–40)keV energy band has 23 members, representingmore than 70% of all CVs detected by INTEGRAL [5]. IPs are the most lumi- nous and the hardest X-ray sources among accreting WDs. In hardX-rays, these objects seem to be more luminous (up to a factor of 10) than polars [10]. In strongly magnetized (B ≥ 107G) polar systems, cy- clotron radiation is an important coolingmechanism, which suppresses high temperature bremsstrahlung emission, whilst it should be negligible for IPs. This could explain why most of the CVs observed in the hard X-ray band are IPs. Short-term soft X-raymodulations have been ob- served in the orbital period, in the spin period of the WD, or a beat between the two. No significant modulation has been found so far in the 20–30keV light curves [11]. Most CVs seem to have persis- tent soft gamma rayfluxes. Nevertheless, the sample detected by INTEGRAL represents only 25% of all knownIPs [5]. Some IPsarenotdetectable even ifwe have significant exposure time (more than 4Msec) for these sources. This fact can be related to the activ- 13 Acta Polytechnica Vol. 51 No. 6/2011 Fig. 1: Left panel: INTEGRAL/IBIS flux curves of V1223Sgr in the corresponding energy bands. The arrows represent 3σ upper limits. Right panel: INTEGRAL/OMC light curve of V1223Sgr in the optical spectral band. The crosses represent average magnitude values in the corresponding seasons ity state of these interacting binaries. To understand this relation, it is necessary to study the correlation between the activity state and the X-ray emission of IPs. 2 Observations, analysis and results We used all publicly available observational data from INTEGRAL/JEM-X and INTEGRAL/IBIS to study possible variability of the selected IPs in the hard X-ray and soft γ-ray spectral bands. In addi- tion, observations from INTEGRAL/OMCdatawere used to look for long-termvariability of these objects in optical band. The observational data used in our analysiswasprocessedby INTEGRAL’sOfflineStan- dardAnalysis PackageOSA7. The libraries grid.py and ms_suite.py developed by FM were also used for the data processing and preparation of the final mosaics. 2.1 Intermediate polar V1223Sgr IP V1223Sgr is a bright X-ray source (4U 1849-31) with possibleX-rayflare activity. A short-termburst has also been detected from this system in the opti- cal [12]. These outbursts areprobably a result of disk instabilities or an increase inmass transfer, but there is no correlation between optical andX-ray burst ac- tivity. Moreover, episodes of deep low state (decrease by several magnitudes) of V1223Sgr in the optical band have also been detected [13]. The observationalmaterial for IP V1223Sgr con- sisted of 132 (INTEGRAL/JEM-X) and 1375 (IN- TEGRAL/IBIS) individual pointings obtained in the course of almost three years (MJD 52710.38– 53809.25). The overall INTEGRAL/IBIS mosaics (total exposure time1405.5ksec) showedthatV1223 Sgr is detectableup to80keVwith ≈ 5σ significance. The broad-band (3–100)keV spectrumwas verywell fitted by a bremsstrahlung model with temperature kT =23.7+1.4−1.3keV [5]. During the monitored period, the mean fluxes of this object were (103.0 ± 4.0) × 10−12 ergcm−2 s−1, (46.4±1.4)×10−12 ergcm−2 s−1, (15.1±1.5)×10−12 ergcm−2 s−1 and (12.3±2.0)×10−12 ergcm−2 s−1 in (15–25)keV, (25–40)keV, (40–60)keV and (60–80)keV band, re- spectively. The inspection of the data showed that the ob- servations were obtained in the course of 7 separate seasons. As the next stepwe split the data according these seasons to investigate long-term X-ray/γ-ray variability. The flux curves are displayed in Figure 1 (left panel). It is clear that during the monitored period the fluxes of V1223Sgr were long-term vari- able (especially in the softerbands),witha significant drop around MJD ≈ 53650. The optical light curve of V1223Sgr (INTE- GRAL/OMC) is shown in Figure 1 (right panel). We can see that optical brightness of this sourcewas long-term variable, too. Moreover, these light vari- ations are strongly correlated with the changes in the (15–25)keV, (25–40)keV and (40–60)keV spec- tral bandswith correlation coefficients 0.81, 0.82 and 0.89, respectively. Our detailed period analysis of INTEGRAL/ OMC data did not yield any significant period, only partial detection of the orbital period Porb = 3.37hrs [14]. This was probably caused by complex intrinsic variability (flickering) of V1223Sgr, possi- ble period variations or a the drastic change in the brightness of this object during monitored period. However, a particular analysis is not possible due to inappropriate time distribution of the data. 14 Acta Polytechnica Vol. 51 No. 6/2011 Fig. 2: Left panel: INTEGRAL/JEM-X flux curves of V 1223 Sgr in the corresponding energy bands. The arrows represent 3σ upper limits. Right panel: INTEGRAL/IBIS phase diagram of V1223Sgr in (15–25)keV band folded with orbital period (3.37hrs) Fig. 3: INTEGRAL/IBIS light curves of V709Cas in the corresponding energy bands. The arrows represent 3σ upper limits We also prepared overall mosaics using all avail- able data from INTEGRAL/JEM- X. The total exposure time was 80.6 ksec. The medial fluxes of V1223Sgr during the monitored period were (2.8 ± 0.4) × 10−12 ergcm−2 s−1, (3.7 ± 0.3) × 10−12 ergcm−2 s−1, (2.6±0.4)×10−12 ergcm−2 s−1 and (4.9 ± 1.1) × 10−12 ergcm−2 s−1 in (3–6)keV, (6–10)keV, (10–15)keV and (15–25)keV band, re- spectively. Wealsospit thedata to7 seasonsto inves- tigate the long-term X-ray variability of V1223Sgr. The flux curves are displayed in Figure 2 (left panel). As we can see the corresponding errors are too large for subsequentanalysis andthereforewecanconclude that the INTEGRAL/JEM-X fluxes of V1223Sgr were persistent within their errors in the monitored period. Typically, soft X-ray modulations were observed in the orbital period, in spin period ofWD, or a beat between the two in IPs. However, the IPs are close binary systems with orbital periods in the order of hours and these objects, are not detectable on these time scales by INTEGRAL/IBIS. As the next step, we attempted to investigate the possible short-term variability of V1223Sgr in INTEGRAL/IBIS data. We prepared a uniquemethod of folding the particu- larphase intervalon thebasis ofproper time intervals from the individual science windows. Our method applied Good Time Intervals (GTIs) according to the (orbital or other) phase bin and created phase resolved mosaics (assuming sufficient exposure) of a periodic source. A phase diagram of the fluxes V1223Sgr in the (15–25)keV band folded with the orbital period and constructed using the data from time interval MJD (52917.17–52926.84) is shown in Figure 2 (right panel). 2.2 Intermediate polar V709Cas This X-ray source was recognized as an IP after it had been detected in the ROSAT All Sky Sur- vey as RXJ0028.8+5917 and was identified with a 14th magnitude blue star, V709Cas. The broad- band spectrum (3–100)keV of this object was fit- ted by the bremsstrahlung model with temperature kT =29.6±2.5keV, covering factor CF =0.31±0.04, column density nH = (58 ± 15) × 1022 cm−2 and iron line energy 6.5 ± 0.2keV [15]. The analysis showed that accounting for Compton scattering does not significantly change the obtainedmass of theWD (MWD ≈ 0.9M�) in the case of V709Cas [15]. The broad-band (3–100)keV spectrum from INTEGRAL was well fitted by a thermal bremsstrahlung model with post-shock temperature kT =24.4+1.5−1.4keV [5]. Our analysis of all available observational data for V709Cas showed that this source is detectable up to 100keV. The hard X-ray/soft γ-ray fluxes are not persistent, and the flux curves indicate that the brightness of this IP increased by a factor of ≈ 2 from MJD 52700 to MJD 53700 in (15–25)keV en- ergy band (Figure 3). 15 Acta Polytechnica Vol. 51 No. 6/2011 3 Conclusions Weanalysedall availableobservationaldata from IN- TEGRAL for IP V1223Sgr andV709Cas. Our anal- ysis of the data from INTEGRAL/IBIS showed that the fluxes of these objects are long-term variable, mainly in the (15–25)keV and (25–40)keV bands. MoreoverthishardX-ray/softγ-rayvariability is cor- relatedwith the changes in the optical spectral band in the case of V1223Sgr. Our analysis revealed a deep flux drop around MJD ≈ 53650 observed in both the X-ray band and the optical band for this intermediate polar. A significantpart of the optical emission from IPs is produced by a hot spot, where the matter from a donor star interacts with the outer rim of the accre- tion disk. X-ray emission is produced by the inter- action of the accreting matter with the WD surface. The emission inboth the optical and theX-raybands is therefore related to the mass transfer, and the ob- served variations are therefore probably caused by changes in the mass accretion rate. We are preparing a photometric campaign to ob- tain long-term homogeneous observations (to cover whole activity cycles) as well as sets of observations with high time resolution (to cover the orbital cycles indetail) of selectedCVs,mainlyasa followupto the INTEGRAL observations. Simultaneous analysis of multi-frequency observation (from optical to X-ray) enables a complex study of the physical mechanism related to the mass transfer in these interacting bi- naries. Acknowledgement The International Gamma-Ray Astrophysics Obser- vatory (INTEGRAL) is a European Space Agency mission with instruments and a science data cen- tre funded by the ESA member states (especially by the PI countries: Denmark, France, Germany, Italy, Spain, Switzerland), theCzechRepublic andPoland, and with the participation of Russia and the USA. This study was supported by a project of the Slovak Academy of Sciences, VEGA Grant No. 2/0078/10. References [1] Warner, B.: Cataclysmic variable stars. Cam- bridge : Cambridge University Press, 1995. [2] Patterson, J.: Publ. Astron. Soc. Pacific, 1994, 106, 209. [3] Aizu, K.: Prog. Theor. Phys. 1973, 49, 1184. [4] King,A.R., Lasota, J.P.: Mon.Not.R.Astron. Soc. 1979, 188, 653. [5] Gális, R., Eckert, D., Paltani, S., Münz, F., Kocka, M.: Baltic Astronomy, 2009, 18, 321. [6] van Teeseling, A., Kaastra, J. S., Heise, J.: As- tron. Astrophys. 1996, 312, 186. [7] Bird, A. J., Malizia, A., Bazzano, A., Bar- low, E. J., et al.: Astrophys. J. Suppl. Ser. 2007, 170, 175. [8] Kniazev, A., Revnivtsev, M., Sazonov, S., Bu- renin, R., Tekola, A.: The Astronomer’s Tele- gram, 2008, 1488. [9] Masetti, N., Parisi, P., Palazzi, E., Landi, R., et al.: The Astronomer’s Telegram, 2008, 1620. [10] Chanmugam, G., Ray, A., Singh, K. P.: Astro- phys. J. 1991, 375, 600. [11] Barlow,E. J., Knigge, C., Bird, A. J., Dean, A., et al.: Mon. Not. R. Astron. Soc. 2006, 372, 224. [12] van Amerongen, S., van Paradijs, J.: Astron. Astrophys. 1989, 219, 195. [13] Garnavich, P., Szkody, P.: Publ. Astron. Soc. Pacific, 1988, 100, 1522. [14] Jablonski, F., Steiner, J. E.: Astrophys. J. 1987, 323, 672. [15] Suleimanov, V., Poutanen, J., Falanga, M., Werner, K.: Astron. Astrophys. 2008, 491, 525. R. Gális Department of Theoretical Physics andAstrophysics Institute of Physics Faculty of Sciences P. J. Šafárik University Park Angelinum 9, 04001 Košice, Slovakia L. Hric E. Kundra Astronomical Institute of the Slovak Academy of Sciences 05960 Tatranská Lomnica, Slovakia F. Münz Faculty of Science Masaryk University Kotlářská 2, 611 37 Brno, Czech Republic 16