3 Results and discussion

For three objects narrow band images have been obtained, the remaining images are reproductions from ESO/SERC films. In all images north is up and east to the left. A summary of the identified emission lines and some derived properties, like electron density, extinction coefficient, radial velocity is given in Table 2; for NeVe3-1 values for both the eastern crescent and the inner part are listed. The excitation class is derived from the ratio of the [OIII] lines at 4959 and 5007 Å and H$_{\beta}$, follwing Aller's (1956) scheme.

 

Table 2:   Line fluxes for the new PNe

 

Figure 1:   Spectrum and [NII]-image for object KeWe2, PN G228.5-11.4

 

Figure 2:   Spectrum and ESO-R-image for KeWe3, PN G238.4-01.8

 

Figure 3:   Spectrum and ESO-R-image for KeWe4, PN G257.8-05.4

 

Figure 4:   Spectrum of eastern crescent and ESO-R-image for NeVe3-1, PN G275.9-01.0

 

Figure 5:   Spectrum and (H$_{\alpha}$ + [NII])-image for NeVe3-2, PN G326.4+07.0

 

Figure 6:   Spectrum and ESO-R-image for KeWe5, PN G348.9+04.6

 

Figure 7:   (H$_{\alpha}$ + [NII]) and [OIII] images for KeWe5, PN G348.9+04.6

KeWe2 appeared as an elliptical ring of nebulosity on the ESO/SERC. In contrast to most elliptical PNe it shows brightness enhancements at the ends of the major axis. The new narrow-band image ([NII]) reveals faint outer extensions that indicate that this object is not really an elliptical but rather a bipolar PN with a pronounced equatorial density enhancement and very faint lobes; on the survey only the ring forming the waist of the butterfly according to the classification schemes of Balick (1987) and Schwarz et al. (1993) can be seen. The equatorial ring is seen almost face on. The spectrum shows a low excitation nebula with strong [NII] emission. H$_{\beta}$ and [OIII] are very faint but from comparison with H$_{\alpha}$ it is obvious that reddening must be rather small, while the ratio of H$_{\beta}$ and [OIII] indicates an excitation class < 4.

KeWe3 is a rather large (2 arcmin) nebula consisting of two crescent shaped arcs forming a highly incomplete shell. The eastern crescent is brighter on the ESO/SERC. A possible central star is identified in Fig.2b. The spectrum is remarkable showing strong [NII] emission but very weak lines of H$_{\alpha}$ and [OIII]. The strange nitrogen ring nebula PL 1547.3-5612 has a very similar spectrum (Ruiz 1983). Due to the lack of useable lines no reliable information on reddening or density of the object could be derived. Still the distinct morphology and the extreme H$_{\alpha}$ to [NII] ratio makes this object a likely candidate for an evolved PN in an advanced stage of interaction with the ISM. This is corroborated by the fact that no complete shell is visible, compared with A34 and A61 whose structure is still intact, see Tweedy & Kwitter (1994).

From a morphological point of view KeWe4 is a typical old PN - interestingly enough it is the only such archetypical object in our sample - sporting a disklike structure of homogeneous but extremely low surface brightness and a central star (CS). Similarily the physical properties derived from the spectrum are quite ordinary for an evolved PN. A comparatively small reddening of c = 0.7 is obtained from the Balmer decrement, while the excitation class from H$_{\beta}$ and [OIII] is relatively high, $\rm EC \approx 6 {-} 7$. The sulfur doublet gives an electron density of $\rm < 100\; cm^{-3}$ making this a normal old PN at a statistical distance of 3.5kpc (diameter $\approx$ 0.7pc) that would certainly have been discovered and identified much earlier had it not been for its extremely low surface brightness.

NeVe3-1 looks somewhat similar to KeWe5 but the brightness distribution is even more lopsided with the E crescent being much brighter. In the central region that appears empty on the ESO/SERC the spectrum is of poor S/N and therefore it is not shown here, the only useful parameters derived are a density of $\approx$ 450 from the [SII] lines and a H$_{\alpha}$ to [NII] ratio of 0.7. The spectrum of the two crescents are very interesting: both show a nebula of medium excitation (EC 4-5) with strong [NII] lines and considerable reddening, c = 1.1. Comparison of the two spectra reveals a number of important differences, in particular some observed line ratios are not constant. While the ratio of H$_{\alpha}$/[OIII] is essentially constant the ratios of H$_{\alpha}$/[NII] and [NII]/[OIII] are not. [NII] is much stronger in the East, the brighter crescent. The electron density in both crescents is decidedly lower than in the central region (450); an upper limit of <100 is found for the fainter arc, compared to a value of $\approx$150 in the brighter one. These observational facts: onesided brightness enhancement, enhancement of density and lower ionisation stages are exactly what is predicted for an evolved PN interacting with the ISM (Soker et al. 1991). NeVe3-1 might be an excellent object to study these processes in more detail.

NeVe3-2 is a roundish nebula with a bright central object. The spectrum is dominated by the Balmer lines and [OIII] emission, HeI is present, while the low ionisation stages like [NII] and [OI] are very weak. In fact almost all [NII] seems to be concentrated in a small knot east of the CS, as seen on the raw data, that has a density of 1000 cm^-3. The reddening is determined as c = 0.8 from the Balmer decrement giving a diameter of 0.25pc at a statistical distance of 1.8kpc.

At first sight KeWe5 looks like an ordinary PN on the ESO/SERC showing a circular shape and a CS. The spectrum reveals a nebula of medium excitation with strong [NII] emission that must be in a highly evolved state since the [SII] doublet is in its low density limit, indicating a density < 100 cm^-3. At a distance of some 15 arcseconds a faint nebular emission was identified in the spectrum; despite its poor S/N ratio it is obvious that the ratio of H$_{\alpha}$ to [NII] is higher in this patch ($\approx$ 0.8) than in the nebula itself (0.4); the location of this faint nebular patch is marked in the (H$_{\alpha}$ + [NII])-image by two arrows. Note that this patch cannot be explained by diffuse galactic emission which has been found in [SII] by Kingsburgh & English (1992) around a number of objects. A statistical distance of 8kpc is found but this can only be considered a very rough estimate due to the low H$_{\beta}$ flux observed. The narrow band images taken in H$_{\alpha}$ and [OIII] reveal a much more complicated structure than the ESO/SERC. In the (H$_{\alpha}$ + [NII]) image the PN looks like a broken elliptical ring with a very pronounced brightness enhancement in the North. In [OIII] in contrast the object takes on an oddly deformed shape again showing a strong brightness enhancement but the position does not coincide with the brightness maximum in H$_{\alpha}$. Concentrated asymmetric brightness enhancements and disrupted structure are suggestive of an interaction with the interstellar medium (ISM). This could also account for the strong [NII] emission, see Borkowski et al. (1990) and Tweedy & Kwitter (1994) for details. The low density observed is also consistent with this notion.

KeWe2, KeWe3, NeVe3-1 and KeWe5 all show very high [NII]/H$_{\alpha}$ and [SII]/H$_{\alpha}$ratios. In fact the observed [NII]/H$_{\alpha}$ ratios are among the highest ever found in PNe (e.g. Guerrero et al. 1995). Unfortunately, due to the poor signal to noise ratio of the spectra, a proper analysis of the abundances of the ionized gas could not be carried out. However we have placed those values in a diagnostic diagram (Canto 1981), and they are located very close to the position of the supernova remants. This indicates that a strong contribution of the excitation in these nebulae is due to shock.