The infrared S(1) line of molecular hydrogen has been searched for in twenty-two planetary nebulae using the imaging mode of the Anglo-Australian Telescope. The line was detected and mapped in eleven objects. It has been demonstrated that all those with strong excited molecular hydrogen belong to a subclass of the Type I planetary nebulae, morphologically consisting of an equatorial toroid with faint bipolar extensions. Furthermore, nearly all planetaries with these characteristics have strong molecular hydrogen. The molecular line ratios in the 2.0 to 2.5 micron window are consistent with shock excitation. The observations suggest that the morphology of these planetaries has been controlled by a fast stellar wind interacting with a disc of gas concentrated in the equatorial plane.

We report the first detection of OH emission at 1667 MHz from a planetary nebula, IC 4997 (Figure 1). OH emission in satellite line of 1612 MHz was detected already from planetary nebulae not only Vy 2-2 (Davis and Seaquist 1979), but also two IRAS objects which were distinguished from OH/IR stars as planetary nebulae (Pottasch et al 1987). OH observations of IC 4997 have been carried out with the large radio telescope at Nançay, France. Highly resolved optical emission lines were obtained at the Okayama Astrophysical Observatory, Japan.

OH/IR stars are now generally accepted to be progenitors of planetary nebulae. We have carried out a project to try to find objects in the transition phase between OH/IR stars and PN. Transition objects should be characterized by having both continuum radio emission and a 1612 MHz OH maser line. The continuuum emission would indicate that the central star has become hot enough to ionize part of the envelope, while the OH maser indicates that a large part of the envelope is still neutral. Only one such object was known, namely Vy 2-2.

New collision strengths, from a 19-state quantum calculation for He I, are used to derive revised He/H ratios in planetary nebulae (PN). Empirical formulae are given, for the correction of He I recombination line fluxes for collisional effects, and for the calculation of the population of metastable helium (He I 23S) in gaseous nebulae. The revised He abundances for PN, for four samples of published line fluxes, show a mean ratio He/H = 0.100 ±0.007 if nebulae with neutral He and Type I PN are excluded. The mean reduction due to collisional effects is only 10% for Galactic PN. It is shown that the hypothesis, that He/H should be independent of nebular temperature and density, is better satisfied when collisional effects are allowed for. The new He abundances indicate that there is very little He enrichment in Galactic PN of Types II, III, and IV, and that the enhancement of Type I PN in He over H II regions is reduced from earlier values by one third.

From the study of the λλ5876, 7065 and 10830 He I line intensities in NGC 6572, NGC 6803, NGC 7009, NGC 7027, NGC 7662 and IC 418, it is found that the I(10830)/I(5876) ratio is weaker than expected. By considering estimates of the optical depth at λ10830 due to dust absorption and by determining the optical depth at λ10830 due to atomic absorption, it is argued that dust absorption of λ(10830) photons is not the cause for the low I(10830)/I(5876) ratios. By assuming that the 23S He0 state is depopulated only by radiative transitions to the 11S state and by triplet-singlet exchange collisions, it is found that its population is about a factor of two smaller than expected. This result is in agreement with a previous study of the λ3889, 4472, 5876, 6678 and 7065 line intensities in a group of thirteen Type I planetary nebulae. One of the main implications of the underpopulation of the 23S level is that the collisional effects in the N(He)/N(H) abundance ratios of planetary nebulae and 0-poor extragalactic H II regions are smaller than previously thought.

Recent work has emphasized the determination of the pregalactic helium abundance by mass Yp and the slope ΔY/ΔZ based on the chemical composition of both galactic and extragalactic H II regions (Pagel, 1987; Pagel et al., 1986).

Measurements of the [O III] 52, 88 μm and [N III] 57 μm fine-structure emission lines have been obtained for nine planetary nebulae, using the facility far-infrared array spectrometer on NASA's Kuiper Airborne Observatory. The N++/O++ ratios determined from these observations range by more than an order of magnitude among the sample. Using recent improved values for the atomic parameters, we find that the N++/O++ ratios agree fairly well with values of N+/O+ determined from optical lines in the same objects. The highest N++/O++ values, found for the extreme “Type I” nebulae NGC 2440 and NGC 6302, are approximately unity. These results imply that the synthesis and mixing of nitrogen must be extremely efficient in the progenitor stars of some planetary nebulae, and that these nebulae are significant sources of nitrogen to the interstellar medium. The local electron densities derived from the intensity ratios of the two [O III] lines are generally lower than values in the literature determined from small-beam optical observations of other ions, such as [O II]. This effect can be understood in terms of the presence of clumpy structure in the nebula, since the far-infrared lines have fairly low critical densities for collisional de-excitation and therefore are preferentially emitted from low-density gas.

Spectral (λλ 4100 - 7300 A) and photoelectric observations of DDDM 1 (61 + 41°1; δ1950 = +38°48′) were made in 1985–86 on Crimean Station of Sternberg State Institute. Photographic absolute H3 flux is found to be F(Hβ) = (3.0±0.57) 10−12 erg cm−2 s−1, angular radius φ = 1″.0±0″.5. Interstellar reddening E(B-V) for the object is found to be less than 0″.02. Integrated flux from central star and gaseous nebula was measured in UBV-filters V = 14m.71, B-V = 0m.12, U-B = −0m.9.

The question of whether M1-78 is a PN or a compact H II region is discussed. We have obtained new high resolution radio continuum maps, optical spectra and CO maps. Arguments for it being a PN include spectral information, far infrared continuum emission, and radio morphology. It is the strongest CO emitting PN known. Its abundances are peculiar: high helium and very low oxygen and nitrogen abundances. If it is a PN it must be within 4 kpc, but 21-cm absorption measurements indicate that it may be further away.

Recent studies by the Peimberts have noted an anticorrelation between 0 and N abundances in the Type I He- and N-rich PN, such that N+) is approximately constant. We report observations of the spectra and composition of several “planetary nebula-like” shells surrounding more luminous population I O- and WR-stars, which indicate that this 0-N anticorrelation extends upwards in the HR diagram to among the most luminous stars known, with O/H values in the shells ranging down to −2 dex below Solar. We report optical and UV spectrophotometry of the shell nebulae NGC 2359, NGC 6164-5, NGC 6888, NGC 7635, AG Carinae, and the condensations around Eta Carinae, which generally support this anticorrelation trend. We also discuss variations in other elements such as He, C, S, and Ar in these shell nebulae, and compare the compositional variations to the expectations from stellar evolution and nucleosynthesis models of intermediate and massive stars.

High-resolution IUE images have provided line profiles of emission cut by interstellar absorption in the magnesium II lines λλ2795, 2802 A in nine objects. The underlying emission has been reconstructed in seven of these by fitting Gaussian profiles to the remaining line wings. Estimates of the ratio of intrinsic to observed emission ranging from 1.5 to 23 have been obtained. Photo-ionization models of IC 418, IC 4997 and NGC 2440 have been used to obtain Mg/H ratios from the corrected Mg II line strengths. The logarithmic magnesium abundances obtained were 7.6±0.12, 6.96±0.3 and 6.5±0.3 respectively. These results indicate that depletion of Mg into grains is less than previously thought and reduce the observed gradient in the magnesium abundance. The Mg depletion may be related to the C/O ratio within the nebula; carbon-rich planetary nebulae may have less magnesium depletion than those with C/0 < 1.

The abundances of C, N and O in planetary nebulae must correspond to the evolutionary status of their progenitor red giant stars. The best spectral features for abundance determination of these elements are the recombination lines, which depend weakly on the variations of Te and ne. The abundance ratio of the ions A+ and H+ can be given by [1–3].

Data from a survey of 51 planetary nebulae (PN) observed with the image tube scanner on Shane Telescope at Lick Observatory are combined with those for a comparable number of objects previously reported. For nearly all of the PN included in the later program, it was possible to obtain adequately accurate plasma diagnostics and line intensities to derive ionic concentrations for He, N, O, Ne, S, Cl, and Ar. To get ionization correction factors we calculated theoretical nebular models to fit the excitation level and the intensities of individual important lines. Final model parameters include the stellar radius, emergent flux, Fv(*), from Husfeld et al. (1984, Astron. Astrophys., 134, 139), nebular size, the optical depth at the hydrogenic Lyman limit, and chemical abundances. Many PN do not appear to be optically very thick in the Lyman continuum.

A large sample of forbidden lines -[O II], [S II], [Ar IV], and [N I]- have been analyzed to obtain electron densities for 134 planetary nebulae.

We have obtained long-slit low resolution spectra (7.5 A resolution) of a sample of 13 low surface brightness planetary nebulae using the 2.5-m Isaac Newton Telescope (La Palma) with the Image Photon Counting System (IPCS) covering a spectral range from 3300 A to 7300 A. From those spectra we calculated the ionic and total abundances of 0, N, Ne and Ar. Variations in the ionization structure between the inner and the outer part are found in some nebulae although the total abundances appear not to change significantly along the nebulae.

When there are changes in gas density within a nebula, various methods of determining the electron density Ne can give different results. Irrespective of differences in ionization structure, there will be deviations in derived values of Ne due to the physics of populating the energy levels. To focus on Ne variations, the electron temperature is held constant. For two cases presented, the values of Ne inferred range over a factor of ten from nine species (line pairs); in order of increasing Ne, they are N+(122/204 μm), O++(52/88 μm), S+(6716/6731 A), S++(18.7/33.5 μm), 0+(3726/3729 A), Ne++(15/36 μm), Ar+3 (4711/4740 A), Ar++(8.99/21.8 μm), and C++(1906/1909 A). This is basically a progression from lower to higher critical densities, Nc, for the lines involved, although other factors are involved. The above order can change somewhat for different mixes of densities.

A revision is made of the classification scheme of planetary nebulae proposed by Peimbert (1978), taking into account the observed heavy-element abundances and radial abundance gradients (Faúndez-Abans and Maciel, 1986; 1987a). A subdivision is proposed of type II PN into the subtypes a and b, according to their nitrogen abundances relative to hydrogen (Faundez-Abans and Maciel, 1987b). Type IIa PN show a higher enrichment in nitrogen than type IIb, which present a nitrogen abundance close to the value of population I objects and a radial N/H gradient as well.

Long-slit IPCS 2-dimensional spectra in radial directions across the inner and outer shells and across some of the brightest cometary globules are presented. The spectra show the ionization structure of the nebula quantitatively, with He II λ4686 A strong in the inner regions and [N II] λ6584 A prominent in the outer shell, almost due east of the central star. The “[Ne III] anomaly”, previously reported for NGC 6720 and 7293 by Hawley and Miller, is clearly seen. It, together with the [O I] λ6300 A flux, provide constraints on our photo-ionization models, as both depend on the concentration of neutral H in the background gas.

The physical characteristics of NGC7027 and the nature of the observations available make it an exceptional object to check the ability of steady photoionization modelling to correctly predict the emission spectrum of nebulae. After a decade of atomic physics and observation improvements, a complete updating of the analysis performed by Pequignot et al (1978) (Astron. Astrophys. 63, 313) is in order.

A model potential method (Caves and Dalgarno, 1972, J. Quant. Spect. Rad. Transf., 12, 1539) was used to calculate accurate non-hydrogenic radiative recombination rates and transition probabilities of singly excited states of CI and N II. The results can be used to determine the excitation mechanism of emission lines and to estimate N III concentrations in nebulae with CI and N II emission lines. In most nebulae, observed permitted lines of N II are produced by radiative recombination, but sometimes stronger recombination lines are missing in their spectra. The [CI] lines observed in NGC 7027 cannot be explained by simple radiative and dielectronic recombination. The low [CI] λλ9850 + 23/ λ8727 value may indicate that the emission is produced in high density (NeZ 105 cm−3) condensations where partial collisional deexcitation of metastable levels, takes place. N III concentrations were determined using published data of NGC 3242, NGC 3918, and NGC 6572. The procedure outlined by Wilkes et al. (1981, M.N.R.A.S., 197, 1) to determine N abundances from (N+ + N++)/He+ ratios does not always give consistent results with UV or [N II] data. The problem may be due to errors in the calculation of transition probabilities involving the doubly excited levels 2s2p33P0 and 3D0 of N II that affect the branching and effective recombination rate of the multiplet N II λ5680.