The [Fe IV] discrepancy:: Constraining the iron abundances in nebulae

We study the current discrepancy between the model- predicted and measured concentrations of Fe++ and Fe+3 in ionized nebulae. We calculate a set of photoionization models, updated with the atomic data relevant to the problem, and compare their results with those derived for the available nebulae where both [ Fe III] and [ Fe IV] lines have been measured. Our new model results are closer to the measured values than the results of previous calculations, but a discrepancy remains. This discrepancy translates into an uncertainty in the derived Fe abundances of a factor of up to similar to 4. We explore the possible causes of this discrepancy and find that errors in the Fe atomic data may be the most likely explanation. The discrepancy can be fully accounted for by any of the following changes: ( 1) an increase by a factor of similar to 10 in the recombination rate ( radiative plus dielectronic, or charge transfer) for Fe+3, ( 2) an increase by a factor of 2 - 3 in the effective collision strengths for Fe++, or ( 3) a decrease by a factor of 2 - 3 in the effective collision strengths for Fe+3. We derive the Fe abundances implied by these three explanations and use the results to constrain the degree of depletion of Fe in our sample nebulae. The Galactic H II regions and planetary nebulae are found to have high depletion factors, with less than 5% of their Fe atoms in the gas phase. The extragalactic H II regions ( LMC 30 Doradus, SMC N88A, and SBS 0335 - 052) have somewhat lower depletions. The metal- deficient blue compact galaxy SBS 0335 - 052 could have from 13% to 40% of Fe in the gas phase. The depletions derived for the different objects define a trend of increasing depletion at higher metallicities.