Three Populations of IPs

It now seems likely that there are three populations of IPs, with the discovery of LAMOST J024048.51+195226.9 as a twin of AE Aqr. These are propeller systems, where the white dwarf spin is so fast that some of the matter is flung off, but it is generally accepted that at least some of the matter is being accreted by the white dwarf. This is in contrast to AR Sco, whih is believed to be driven by the spin-down power of the white dwarf, as in radio pulsars (powered by the spin-down of neutron stars), with no evidence that accretion is currently taking place.

It appears that the rest of IPs (non-propeller, or at least presumed non-propeller, systems) can be divided to regular IPs, with a typical BAT-band X-ray luminosity of over 10³³ erg/s, and low-luminosity IPs (LLIPs) with a typical BAT-band luminosity of 10³¹ erg/s (Pretorius & Mukai 2014). Pretorius & Mukai only identified 3 LLIPs in the BAT sample (EX Hya, DO Dra, and V1025 Cen), there appear to be quite a few more that are not detected by BAT, despite many of them being quite nearby. This is consistent with their low luminosity in the BAT band, which also means that the BAT-band luminosity may be more bimodal than the <10 keV or bolometric X-ray luminosity. This may be due to lower shock temperature (due to tall shocks) as well as lower accretion rate.

In turn, the lower accretion rate in the low luminosity IPs can usually be explained as due to evolutionary effects: most are short period IPs below the 2-3 hr preiod gap of CVs, where the secular average of mass transfer rate is a few times 10^-11 solar masses per year up to 10^-10 solar masses per year (see, e.g., Figure 11 of Knigge et al. 2011). Therefore, short orbital period IPs are expected to have lower than typical X-ray luminosity. That is, any IPs below the period gap are candidate LLIPs. There is one problematic case: given the short orbital period, IGR J18173-2509 ought to be an LLIP, but it is in fact extremely luminous (probably above 10³⁴ ergs s^-1 in the BAT band), which is a mystery. On the other side, DO Dra is an LLIP even though its orbital period is above the period gap.

Also, there are two types of X-ray spectra among IPs: while most IPs are characterized by heavy, complex absorption, whose change is the primary cause of X-ray spin modulation, little intrinsic absorption is apparent in some IPs. In the Chandra HETG spectra of CVs analyzed by Mukai et al. 2003, EX Hya was classifed as having a "cooling flow" type spectrum, along with dwarf novae U Gem and SS Cyg, and the old nova V603 Aql. In contrast, normal IPs AO Psc, V1223 Sgr, and GK Per all had a very hard continuum and the low energy lines were better explained as due to photoionization. Of course, X-ray spectra of these (and other traditional) IPs make it clear that their central engine is also cooling flow-like, and the hard continuum is due to the complex absorber. Partial-covering absorber models are inadequate to describe the true complex absorption in the pre-shock flow of magnetic CVs, which requires something like the pwab model of Done & Magdziarz 1998 (see also Islam & Mukai 2021), resulting in the appearance of a hard power-law like continuum.

While the X-ray spin modulation in most IPs is energy dependent, it is not as steep as that of a simple photoelectric absorber (Norton & Watson 1989). However, the spin modulation of EX Hya is better modeled as due to self-occultation of a tall shock (Allan et al. 1998). V1025 Cen (Hellier, Beardmore & Buckley 1998) and HT Cam (de Martino et al. 2005) appear similar to EX Hya in that their average X-ray spectra do not demand the presence of a complex absorber, and that their spin modulation can be explained readily as normalization, rather than absorption, changes.

The complex absorbers in IPs, and the resulting energy-dependent modulation, are boty caused by the intrinsic absorption in the pre-shock flow. It therefore makes sense that these features are apparent in high accretion rate IPs but not in low accretion rate IPs. Not only does the maximum N_H lower in proportion to the accretion rate, but the shocks are tall in LLIPs, which allows X-rays to escape from the sides of the post-shock region, rather that through the top, thereby escaping any interaction with the pre-shock flow.

Because LLIPs likely have tall shocks, the measured shock temperature does not reflect the gravitational potential of the white darf at its surface. Finally, the tall shocks result in both poles being observable over a wide range of viewing angles, which is a likely mechanism for spin modulation in LLIps (Mukai 1999).

Away from the X-rays, the presence of outbursts in DO Dra and EX Hya, for example, is likely linked with their low luminosity. Also, the FUV spectra of both these system are dominated by the white dwarf photosphere (Froning et al. 2012), which is another clue to the low luminosity.

Please send your comments, suggestions etc. to Koji.Mukai@nasa.gov and/or Koji.Mukai@umbc.edu

National Aeronautics and Space Administration

Goddard Space Flight Center

IPs

Full Catalog

Related Systems

Three Populations of IPs