Ionization fronts mark the transition from neutral to ionized gas and are found, for example, at the edges of H II regions, around photoionized clumps and in stellar wind bubbles. Ionization front (IF) structure and propagation were first systematically studied in the classic papers of Kahn (1954), Axford (1961) and Goldsworthy (1961). The standard classification into R-type (`rarefied') and D-type (`dense') IFs was introduced by Kahn (1954). Yorke (1986) has reviewed work on H II region evolution and dynamics up to that date. In such studies the neutral gas density is assumed to be either uniform or to vary smoothly.
However, practically all diffuse astronomical sources are clumpy; this includes Wolf-Rayet and planetary nebulae, molecular clouds and active galactic nuclei. Hartquist & Dyson (1993) and Dyson et al. (1997) have reviewed the response of clumpy sources to mass, momentum and energy input. In many sources, energy input by ionizing photons is important. Clumps photoionized by an external radiation field lose their heated surfaces by simple expansion, provided that their surface pressure exceeds that of their surroundings (Dyson 1968, 1994; Kahn 1969; Bertoldi & McKee 1990). Mass injection from photoionized clumps has been suggested as the source of the ionized material in ultracompact H II regions (Dyson 1994; Dyson et al. 1995; Redman et al. 1996; Williams et al. 1996; Lizano & Canto 1995; Lizano et al. 1996). The photoevaporation of neutral material is graphically shown in the HST images of the elephant-trunk structures in M16 (Hester et al. 1996) and the cometary knots in the Helix nebula (O'Dell & Handron 1996).
The evaporation rates and lifetimes of photoionized clumps are central to the clumpy UCHII R models. The studies above deal with the photoionization of non-magnetized, isothermal, self-gravitating clumps. All these assumptions are questionable. The statistical studies of Myers et al. (1995) and Myers & Khersonsky (1995) show that the ratio between magnetic and thermal pressure may be of the order ten or so in diffuse neutral clouds and up to an order of magnitude higher than that in clumps like those identified by Williams et al (1995) from CO maps of the Rosette Molecular Cloud. The ratios of the magnetic pressure to thermal pressure in some dense cores associated with low mass star formation are each only of the order of a few, but, after mass loss from a high mass star begins in a region, such cores can be compressed, and this might lead to increases of the ratios. Magnetic fields have been included in the study of shock fronts (see e.g. Draine & McKee 1993) but the effect on the different types of ionization front has not been described.
Ideally, the effects of turbulence in the ambient medium on the propagation of an ionization front should be considered. We intend to address this point in future papers. However, there are situations in which the ratios of the magnetic pressure to thermal pressure and of the magnetic pressure to the turbulent pressure could be high; for instance, dense cores that have been compressed by hot shocked stellar winds in regions of high mass star formation may not be very turbulent. Even if the turbulent pressure is comparable to the magnetic pressure, so long as the H ii region is small compared to the wavelength at which the turbulent power is concentrated, a change of reference frame to one comoving with the ambient medium immediately ahead of the front maintains the applicability of our analysis. From Spitzer (1978) one finds that a D-type ionization front forms around a B1 star's H ii region in a clump like those found by Williams et al. (1995) when the H ii region's radius is only , which is almost certainly small compared to the aforementioned wavelength.
In Sect. 2 we describe the modifications to the jump conditions across and the propagation of IFs where there is a magnetic field in the plane of the front. We will address the more general and complex case of IF propagation at an arbitrary angle to the magnetic field in a later paper. In Sect. 3 we briefly comment on these modifications and note some of the applications of these ideas in relation to UCHII R and other diffuse sources.
© European Southern Observatory (ESO) 1998
Online publication: March 3, 1998