2005MNRAS.362..424W

We present deep optical spectra of 23 galactic planetary nebulae, which are analysed in conjunction with archival infrared and ultraviolet spectra. We derive nebular electron temperatures based on standard collisionally excited line (CEL) diagnostics as well as the hydrogen Balmer jump and find that, as expected, the Balmer jump almost always yields a lower temperature than the [Oiii] nebular-to-auroral line ratio. We also make use of the weak temperature dependence of helium and Oii recombination line ratios to further investigate the temperature structure of the sample nebulae. We find that, in almost every case, the derived temperatures follow the relation, which is the relation predicted by two-component nebular models in which one component is cold and hydrogen-deficient. T_e(Oii) may be as low as a few hundred Kelvin, in line with the low temperatures found for the hydrogen-deficient knots of Abell 30 by Wesson, Liu and Barlow.

Elemental abundances are derived for the sample nebulae from both CELs and optical recombination lines (ORLs). ORL abundances are higher than CEL abundances in every case, by factors ranging from 1.5 to 12. Five objects with O²⁺ abundance discrepancy factors greater than 5 are found. DdDm 1 and Vy 2-2 are both found to have a very large abundance discrepancy factor of 11.8.

We consider the possible explanations for the observed discrepancies. From the observed differences between T_e(Oiii) and T_e(BJ), we find that temperature fluctuations cannot resolve the abundance discrepancies in 22 of the 23 sample nebulae, implying some additional mechanism for enhancing ORL emission. In the one ambiguous case, the good agreement between abundances derived from temperature-insensitive infrared lines and temperature-sensitive optical lines also points away from temperature fluctuations being present. The observed recombination line temperatures, the large abundance discrepancies and the generally good agreement between infrared and optical CEL abundances all suggest instead the existence of a cold hydrogen-deficient component within the `normal' nebular gas. The origin of this component is as yet unknown.