The enrichment of the interstellar medium (ISM) in galaxies is a complex interplay between the star-formation history, supernovae-driven Galactic outflows, and infall of metal-poor gas. Chemical abundances are usually described by HII region abundances corresponding to the youngest star-forming regions. The enrichment of the large reservoir of neutral gas in galaxies, reservoir for future star formation either in situ or through infalling gas, remains largely unknown.

Previous generation FUV spectrographs, in particular IUE, FUSE but also HST/STIS and COS, suggested that the neutral gas in blue compact dwarf galaxies is less enriched with heavy elements as compared to the ionized gas in HII regions (e.g., Aloisi et al. 2003; Lecavelier de Etangs et al. 2004; Lebouteiller et al. 2004, 2009, 2013). This was found by observing far-UV absorption lines of resonant transitions toward young star clusters (Kunth et al. 1994). These results also hinted at the existence of a threshold metallicity at redshift 0, possibly set by the IGM enrichment.

A firm conclusion could never be reached because of the complex arrangement of multiple lines of sight intersecting multiple absorption components with different properties (e.g., turbulent velocities, column densities, radial velocity), with non-linear effects and resolution effects (in particular hidden saturated components).

LUVOIR could provide abundance measurements at exquisite sensitivity toward individual stars in nearby galaxies, enabling tomograpic abundance mapping of the neutral gas envelope, which, together with ionized gas abundance mapping with MUSE would provide for the first time a complete 3D abundance distribution. LUVOIR would then allow constraining local enrichment of heavy elements released in young star-forming regions (e.g., wround Wolf-Rayet stars), examining whether star formation is biased toward the more metal-rich regions, measure abundances in photodissociation regions near molecular clouds, measure the metallicity threshold in a wide redshift range, but also constraining abundances near active galactic nuclei.

in addition to abundances, LUVOIR would also be able to measure the cooling rate of the neutral gas via the fine-structure absorption lines of neutral carbon and oxygen. This provides interesting constraints for simulations in order to constrain the galaxy chemical evolution models.