Investigating the electron temperature of in planetary nebulae using the DESIRED database.

Determining the chemical composition of the interstellar medium relies heavily on the analysis of emission lines from ionised nebulae. While most studies focus on low- to mid-ionisation species, the physical conditions in the innermost, high-ionisation regions of planetary nebulae (PNe) remain poorly constrained, particularly regarding the role of additional heating mechanisms. We investigated the behaviour of the electron temperature derived from lines, ( ), in a sample of PNe to characterise the thermal structure of high-excitation gas and compare it with the predictions of standard photoionisation models. Using the DEep Spectra of ionised REgions Data Base Extended (DESIRED-E), we selected a sample of 57 PNe for which ( ), ( ), and ( ) could be determined simultaneously and homogeneously. We performed a detailed comparison between these observational diagnostics and a suite of over 160,000 photoionisation models from the Mexican Million Models Database (3MdB). We find that the observed ( ) values are systematically higher than those predicted by pure photoionisation models for a given ( ). Namely, approximately 31% of the PNe sample exhibits a ) more than 2σ higher than photoionisation model predictions. This discrepancy persists regardless of the specific set of auroral lines used for the diagnostic or the choice of atomic data (transition probabilities and collision strengths) adopted in calculations. Other high-ionisation diagnostics, compiled from the DESIRED database or from the literature, however, do not show such behaviour, though the statistics for these are much more limited. The fact that models succeed for the highest-ionisation species but fail specifically for suggests that the discrepancy is not due to a global inner-nebula heating mechanism. Instead, it indicates a localised physical effect or limitation in the current understanding of the ionisation stratification and atomic physics specific to the argon ion stages. This ` anomaly' must be resolved to ensure the reliability of abundance determinations in high-excitation nebulae.