SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 322, 242-255 (1997)

Previous Section Next Section Title Page Table of Contents

1. Introduction

  Cataclysmic variables (CVs) are binary systems in which matter flows from a Roche-lobe filling dwarf secondary star onto a white dwarf primary. The CV AE Aqr contains one of the most rapidly spinning white dwarfs known ([FORMULA] sec) and is a member of the Intermediate Polar (IP) sub- class of CVs. (For reference we give the most pertinent system parameters of AE Aqr in Table 1).


[TABLE]

Table 1. Some of the most important system parameters of AE Aqr (most of which are taken from Welsh et al. 1994).


The magnetic field strength of the white dwarf in IP's is not sufficient to synchronise the spin period of the white dwarf with the binary orbital period, but prevents or disrupts the formation of an accretion disk which is usually present in CVs (see Patterson 1994 for a recent review). In Polars (or AM Her stars), the white dwarf has a sufficiently strong magnetic field to prevent the formation of an accretion disk and the accretion flow impacts directly onto the white dwarf (see Cropper 1990 for a review). However, a direct measurement of the magnetic field strength of AE Aqr has yet to be made. The most common method of determining this parameter in Magnetic CVs is from polarisation measurements. Measurements by Stockman et al. (1992) gave an upper limit to the circular polarisation of 0.06% indicating [FORMULA] 500 T. Other more indirect estimates of the field strength include B =1-10 T and 10-100 T (Lamb & Patterson 1983 and Lamb 1988 respectively) from the spin-down and pulsing behaviour of the white dwarf.

Although the magnetic field strength of AE Aqr is uncertain, there is evidence to suggest that there is no accretion disk at the present epoch, and that very little of the material leaving the secondary star is actually accreted onto the surface of the white dwarf. The observed emission lines are not double peaked, and a Doppler tomograph (Wynn et al. 1996) shows no signature of the expected Keplerian flow. This would suggest that the accretion is largely stream-like, or possibly blob-like. Wynn et al. (1995) simulate the accretion flow in AE Aqr and find that the vast bulk of material is ejected from the binary system altogether. This is also consistent with optical observations by Mouchet et al. (1995) which show no evidence for the presence of an accretion disc.

AE Aqr is observed to emit at radio wavelengths with a quiescent and a flaring component. The generation mechanism probably is synchrotron emission from relativistic electrons gyrating in magnetic fields. The integrated quiescent radio power at a distance of 84 pc (Bailey 1981) is [FORMULA]  W, that in flares [FORMULA]  W. The X-ray emission is [FORMULA] W ([FORMULA]) implying an accretion rate of at most [FORMULA] kg s [FORMULA] [FORMULA] /yr (if all luminosity is from accretion, Eracleous et al. 1991, Reinsch et al. 1995). This emission is smaller by more than an order of magnitude than the spin-down luminosity of the white dwarf ([FORMULA] W; De Jager 1994). The spin-down energy is much larger than the (total observed) quiescent accretion luminosity plus the energy released in typical UV flares (Horne & Eracleous 1995). There is therefore a reservoir of energy which might be tapped to accelerate to high energy the particles thought to be the source of quiescent and flaring radio emission.

Bastian et al. (1988) and Abada-Simon et al. (1993) reported radio flares on timescales ranging from a few minutes to several hours. They interpreted these flares as the superposition of synchotron-emitting plasma clouds. Observations of AE Aqr were extended down to sub-mm wavelengths by Abada-Simon et al. (1995a). Typically the radio power is [FORMULA] W during one hour for a big flare, and a total radio energy of a few times [FORMULA] Joules ([FORMULA] J for the largest flare observed). The optical and radio flare events are not correlated and further, unusual dips were seen in the radio (Abada-Simon et al. 1995b). From VLBI a source expansion up to [FORMULA] (a is the binary separation) has been inferred at a speed 0.01 c (A.E. Neill 1995 private communication, De Jager & Meintjes 1993). Although Bastian et al. (1988) proposed a model in which energetic particles were the direct cause of these flares, they did not attempt to explain what the source of these energetic particles was. Under the synchrotron hypothesis a total energy of [FORMULA] J is required in [FORMULA] energetic electrons with Lorentz factors of [FORMULA] distributed over a source of extent [FORMULA] m with an ambient magnetic field of 0.005 T. In the present work we attempt to explain the origin of these fast particles. We will address the following questions:

  • What is the power in MHD disturbances? (§ 2)
  • Is magnetic pumping a suitable candidate for the particle acceleration required to power the synchrotron emission? (§ 3,4,5,6)
  • What causes the radio emission to be produced in flares rather than more steadily? (§ 6)
  • What other acceleration processes are relevant to AE Aqr? (§ 7)
  • Why do radio outbursts such as observed on AE Aqr occur only in 4 CVs and in only a small fraction of X-ray binaries (Nelson & Spencer 1988, Hjellming & Penninx 1991, Kuijpers 1989)? (§ 7)

In Sect. 2 we first consider how effectively magnetic disturbances are created in the magnetised white dwarf companion system. In Sect. 3 we work out which adiabatic invariants exist in AE Aqr and discuss their relevance to particle acceleration. In Sect. 4 we specifically study acceleration by magnetic pumping, and in Sect. 5 the limiting particle energies when account is taken of various losses. In Sect. 6 we model the radio emission. In Sect. 7 we discuss the initial particle source, other potentially significant acceleration processes, and the relation of AE Aqr to other magnetised binaries. Finally Sect. 8 sums up the results of the present paper.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: June 30, 1998
helpdesk.link@springer.de