4. GALACTIC H II REGIONS

Recent reviews on abundances of galactic H II regions are those by Peimbert (1993) and Wilson & Rood (1994).

Photoionization models with uniform chemical composition predict t² values around 0.01 (e.g. Garnett 1992; Gruenwald & Viegas 1992). Baldwin et al. (1991) produced a photoionization model for the Orion nebula, in which grains and gas are well mixed, that included photoelectric heating and cooling of the gas by grain ionization and grain collisions respectively; for this model Peimbert et al. (1993a) find that t² = 0.004.

In Table 1 I present observational t² values for galactic H II regions of the solar vicinity based on three methods: a) the combination of T_e(Bac / H) and T_e (4363/5007) (see equations 2.3 and 2.4), b) the condition that N(C⁺⁺ recombination) / N(C⁺⁺, collisions) is equal to one; c) the condition that N(O⁺⁺ recombination) / N(O⁺⁺, collisions) is equal to one. In addition to the results presented in Table 1, by combining the observations at high radiofrequencies (H56, H66) and the [O III] observations at 88.35 µm with T_e(4363/5007), the following values of t² are obtained for the Orion nebula: 0.037, 0.011 and 0.035 (Berulis, Smirnov & Sorochenko 1975; Wilson, Bieging & Wilson 1979; Moorwood et al. 1978; Torres-Peimbert, Peimbert & Daltabuit 1980).

Walter & Dufour (1994) from longslit spectra of the Orion nebula find a decrease in T_e(Bac) moving to the west of ¹ Ori C, from 8400 K at a distance of 40" to a low of 2800 K at a distance of 220"; this decrease implies an increase in t² with distance from ¹ Ori C reaching values considerably higher than any reported to date.

From the previous discussion it follows that typical t² values in galactic H II regions of the solar vicinity are in the 0.03 to 0.05 range, values considerably higher than those predicted by photoionized chemical homogeneous models with constant density.

As for PN there are three possible explanations for the large t² values: a) that the T_e(4363/5007) temperatures have been overestimated due to clumps and regions with N_e ~ 10⁶ cm^-3, b) that the H II regions are chemically inhomogeneous and c) that there is deposition of kinetic energy by shocks and by subsonic turbulence produced by mass loss from stars inside the H II regions.

The Orion nebula does show regions of N_e ~ 10⁶ cm^-3 (Bautista, Pradhan & Osterbrock 1994; Bautista & Pradhan 1994) but due to their high densities, these regions are of low degree of ionization, probably in the partially ionized zone that separates the H II region from the H I region, and do not affect the distribution of T_e(4363/5007). Also the determination of N_e(3727/7325) indicates the presence of density fluctuations but the fraction of the emission due to regions with N_e 10⁵ cm^-3 is negligible (e.g. Peimbert & Torres-Peimbert 1977).

Cunha & Lambert (1992) and Cunha (1993) have divided the B stars of the Orion association into four subgroups according to age. They have found that the younger the subgroup the higher the O/H ratio and that the spread in the O/H ratio in one age group reaches 0.5 dex. Inhomogeneities of O/H in the 0.3 dex to 0.5 dex range in a given nebula can produce t² values in the 0.03 to 0.04 range.

There is a large body of evidence for the presence of high velocity components and shock waves inside shell H II regions (e.g. Dufour 1989 and references therein) and inside typical H II regions (e.g. Castañeda 1988; Clayton 1990; Massey & Meaburn 1993). These high velocity components are due to stellar winds from massive stars embedded in the nebulae.

Temperature fluctuations also play a paramount role in the determination of Y / Z. M17 is the H II region of the solar vicinity with the best He/H determination (Peimbert et al. 1992). This determination combined with a t² = 0.04 and the pregalactic helium abundance, Y_p, of 0.230 derived from O-poor extragalactic H II regions (e.g. Pagel et al. 1992, Pagel 1995) yields Y / Z = 2.5 ± 0.5. From M17, Y_p = 0.230 and t² 0.00 a Y / Z = 4.59 is derived.

The large Y / Z and the small O/H values for t² = 0.00 found for O-poor H II regions led Maeder (1992, 1993) to suggest that there is a mass, m(BH), above which massive stars would end their evolution without enriching the ISM at the time of the SN explosion. Carigi (1994) and Peimbert, Sarmiento & Colín (1994b), based on the stellar evolution models by Maeder (1992) find that it is possible to explain the Y / Z value of 2.5 ± 0.5 (t² = 0.04) for the solar vicinity without the need to suppress the Z enrichment of massive stars at the end of their evolution. Alternatively for t² = 0.00, the mechanism suggested by Maeder becomes necessary to explain the Y / Z ratio but would enter in contradiction with observational constraints such as C / O.