High spectral resolution spectroscopy has proved to be very useful for the advancement of chemical abundances studies in photoionized nebulae, such as H II regions and planetary nebulae (PNe). Classical analyses make use of the intensity of bright collisionally excited lines (CELs), which have a strong dependence on the electron temperature and density. By using high resolution spectrophotometric data, our group has led the determination of chemical abundances of some heavy element ions, mainly O ++ , O + and C ++ from faint recombination lines (RLs), allowing us to deblend them from other nearby emission lines or sky features. The importance of these lines is that their emissivity depends weakly on the temperature and density structure of the gas. The unresolved issue in this field is that recombination lines of heavy element ions give abundances that are about 2-3 times higher than those derived from CELs −in H II regions− for the same ion, and can even be a factor of 70 times higher in some PNe. This uncertainty puts into doubt the validity of face values of metallicity that we use as representative not only for ionized nebulae in the Local Universe, but also for star-forming dwarf and spiral galaxies at different redshifts. Additionally, high-resolution data can allow us to detect and deblend faint lines of neutron capture element ions in PNe. This information would introduce further restrictions to evolution models of AGBs and would help to quantify the chemical enrichment in s-elements produced by low and intermediate mass stars. The availability of an echelle spectrograph at the E-ELT will be of paramount interest to: (a) extend the studies of heavyelement recombination lines to low metallicity objects, (b) to extend abundance determinations of s-elements to planetary nebulae in the extragalactic domain and to bright Galactic and extragalactic H II regions. . However, high spectral resolution spectroscopy permits to obtain proper measurements of very faint lines of much interest for abundance studies due to their closeness to other lines or telluric emission features, specially in the reddest part of the op-⋆ Corresponding author. e-mail: cel@iac.es tical range or the near infrared (NIR). In most PNe, high spectral resolution is necessary in order to resolve their usually complex velocity structure, where velocity splitting of the order of few to several tens of km s −1 is the common rule. Since the velocity structure of H II regions is normally much simpler, spectral resolutions of about R∼10,000 are necessary to resolve their line widths and thus optimize the contrast of the faint lines with respect to the underlying continuum. The continuum may be very intense in the case of the spectra of extragalactic H II regions, where most of the contribution comes from the stellar underlying population we are including in our observing area and may even contain absorption features. Moreover, there are also important bright collisionally excited emission lines that need the use of high spectral resolution −R∼3000 ...