The question of why some quasars have unusually strong nitrogen lines has been debated for decades. Some studies explained it by the high overall metallicity of the gas in the broad-line region (BLR), others by selective nitrogen enrichment. However, directly separating these two effects within the BLR was impossible: earlier metallicity estimates relied on nitrogen lines, and thus depended on the assumed mechanism of its production. The work sits within the line of research on the chemical evolution of quasars and galaxies, but stands out because for the first time it systematically compares nitrogen-based and nitrogen-free metallicity indicators for a large sample of N-loud quasars and a matched control group of ordinary quasars.
From the DESI DR1 catalog, 121 quasars with anomalously strong nitrogen lines (N-loud) were selected in the redshift range 2.13–3.90. For each one, three to five ordinary quasars were matched by redshift, continuum luminosity and black hole mass estimated from the C IV line; in addition, a control sample based on Mg II mass was built to test robustness. All spectra were processed uniformly: a power-law continuum was subtracted, lines were decomposed into components with a double power-law profile, and line fluxes were measured with a Monte Carlo method. Metallicity was estimated using CLOUDY photoionization models within the locally optimally emitting cloud (LOC) framework, using both nitrogen-based (N V/C IV, N III]/C III], N III]/O III]) and nitrogen-free ((Si IV+O IV])/C IV, Al III/C IV) line ratios.
Within the N-loud quasar sample, the metallicity from N V/C IV is roughly 3–50 solar, whereas from the nitrogen-free indicator (Si IV+O IV])/C IV it is mostly 1–20 solar, with the former systematically exceeding the latter by a factor of 2.5 on average. The two nitrogen-free indicators, (Si IV+O IV])/C IV and Al III/C IV, agree well with each other. Compared with the control sample, N-loud quasars have a median metallicity of about 5 solar versus 1.5 solar from (Si IV+O IV])/C IV, and about 12 solar versus 4 solar from N V/C IV. In the metallicity overlap region, the slope of the relation between the nitrogen-based and nitrogen-free estimates is significantly steeper for N-loud quasars (0.96 versus 0.56, p=0.011), which points to additional nitrogen enrichment that grows with overall metallicity. The line excesses persist across the entire range of black hole masses and Eddington ratios, and for N V/C IV there is a hint of growing excess at lower masses and higher accretion rates.
The work is the first to directly separate the contributions of overall metallicity and relative nitrogen excess in the BLR of quasars with anomalously strong nitrogen lines. Unlike neighboring studies, which either relied only on nitrogen lines or examined a narrow line region (NLR), here nitrogen-free indicators are applied to the very same BLR and compared with a carefully matched control group. This allows the claim that N-loud quasars are not just richer in metals but also possess an additional nitrogen excess that is not described by standard secondary nitrogen production. Such a result casts doubt on the universality of the N/O versus O/H relation and calls for a revision of chemical evolution models in galactic nuclei.
The topic can develop in several directions. First, enlarging the sample of N-loud quasars with reliable Mg II masses and a full set of ultraviolet lines will allow a stricter test of how the nitrogen excess depends on overall metallicity and accretion properties. Second, more flexible abundance modeling, where nitrogen varies independently of metallicity, will help quantify the excess and compare it with predictions of specific enrichment scenarios, such as a delayed stellar wind from asymptotic giant branch stars or reprocessing of material in the accretion disk. Third, comparison with recently discovered nitrogen-rich systems at high redshifts, including compact objects like "little red dots," may clarify whether there is a common mechanism for retaining chemically processed gas in dense nuclear environments.
The results will affect studies of the chemical evolution of galaxies and active galactic nuclei, as well as the interpretation of spectra of distant objects where nitrogen lines are used as a metallicity indicator.
Build a larger sample of N-loud quasars with Mg II masses and run modeling with independent variation of the nitrogen abundance in order to quantify the excess and test specific enrichment scenarios.
The work is directly tied to the unsolved problem of the origin of anomalously strong nitrogen lines in quasars: is it a consequence of high overall metallicity or of selective nitrogen enrichment? The result that the relative nitrogen abundance is partly independent of overall metallicity indicates the need to revise standard chemical evolution models in galactic nuclei and to search for additional nitrogen sources.
A metallicity of 50 solar means that in the broad-line region gas of some quasars the share of elements heavier than helium is 50 times higher than in the Sun. On top of that, nitrogen relative to oxygen is even more abundant than the standard secondary production model predicts.