High-[N/O] Stars: Tracers in Cluster & Galaxy Evolution

• High-[N/O] stars are stellar populations with nitrogen-to-oxygen ratios that greatly exceed typical levels, marking rapid, localized chemical enrichment.
• They are identified via high-resolution spectroscopy and key emission lines (e.g., [N II], [O II]), providing precise chemical and dynamical diagnostics.
• Their presence in globular clusters, the Milky Way halo, and high-redshift galaxies offers critical insights into early cluster formation and galaxy evolution.

High-[N/O] stars are stellar populations, or individual stars, that exhibit anomalously high nitrogen-to-oxygen abundance ratios relative to expectations for their metallicity and evolutionary context. These high [N/O] signatures are fundamental tracers of nucleosynthetic processes, star cluster formation physics, and galactic assembly, and are now recognized as key chemical imprints both in the local universe (globular clusters, the Milky Way halo, field stars) and in the high-redshift universe (as revealed by recent JWST spectroscopy). Their paper provides unique constraints on the sites and mechanisms of rapid nitrogen enrichment, the dynamics of early star formation in dense stellar systems, and the chemical evolution of galaxies across cosmic time.

1. Definition, Measurement, and Occurrence

High-[N/O] stars are defined by nitrogen-to-oxygen number ratios (${\rm [N/O]}$) that significantly exceed the “primary” plateau value found in low-metallicity star-forming regions, often reaching or surpassing the solar $\log$(N/O) $\approx$ –0.86 even at subsolar oxygen abundance. Empirically, the threshold for “high” is context dependent, but abundances such as $\log({\rm N/O}) > -0.3$ at $12+\log({\rm O/H})<8.0$ are well established as anomalous in both local and high-$z$ environments (Charbonnel et al., 2023, Marques-Chaves et al., 2023, Ji et al., 18 May 2025).

Abundances are determined via high-resolution spectroscopy of individual stars (assessing CN, NH, and atomic N lines), or, in the ISM of distant galaxies, via strong emission lines such as [N II], [O II], and recombination lines (requiring accurate electron density and ionization correction factor estimation) (Zhang et al., 7 Feb 2025). Notably, stars showing [N/Fe] $\gtrsim$ +0.5 or [N/O] $>$ 0 in the MW halo, or systems with $\log$(N/O) of order $-$0.2 at $12+\log$(O/H) $<$ 7.8 in high-$z$ galaxies, belong to this class.

High-[N/O] stars are observed:

• In second-generation stars of globular clusters (GCs), often accompanied by Na-Al enrichment and O, Mg depletion (0902.1773, Gratton et al., 2014, Kim et al., 2023, Belokurov et al., 2023, Kane et al., 4 Sep 2025).
• As rare field halo stars (“N-rich” at low [Fe/H]) likely originating from disrupted or dissolved GCs (Kim et al., 2023, Belokurov et al., 2023, Kane et al., 4 Sep 2025).
• In resolved high-surface-brightness clusters in local galaxies and compact, strongly star-forming high-redshift galaxies (so-called N/O–enhanced galaxies or “NOEGs”) identified by JWST (Charbonnel et al., 2023, Marques-Chaves et al., 2023, Ji et al., 18 May 2025).
• As products of binary mass transfer from AGB stars in “NEMP” field stars (Simpson et al., 2019).

2. Physical Origins: Enrichment and Nucleosynthesis

The high [N/O] signature requires local or rapid nitrogen enrichment, incomparable with the “primary” plateau from normal core-collapse supernovae. The main identified mechanisms are:

• Asymptotic Giant Branch (AGB) stars: Intermediate-mass AGBs (3–6 M$_\odot$) undergo third dredge-up and hot-bottom burning, producing N-rich, C-depleted, sometimes s-process-enriched ejecta (0902.1773, Gratton et al., 2014). This scenario is strongly supported by correlations of C+N+O, Na, Al, Zr, and La abundances among stars in GCs.
• Supermassive and Massive Stars: Models predict that runaway stellar mergers in dense proto-cluster environments can form supermassive stars (SMS; $>10^3$ M$_\odot$) that efficiently process C and O into N via the CNO cycle at high temperature. Their winds contaminate the intra-cluster medium, generating the extreme [N/O] enhancements observed in both GCs and certain high-$z$ “N emitters,” such as GN-z11 and CEERS-1019 (Charbonnel et al., 2023, Marques-Chaves et al., 2023, Nandal et al., 5 Feb 2024). This “conveyer-belt” model can deliver sustained N-rich enrichment without overproducing C or Ne.
• Wolf–Rayet (WR) Stars: At both low and moderate metallicity, the winds of massive WN (nitrogen sequence) WR stars can inject large amounts of nitrogen before supernovae have delivered their oxygen yield. If star formation timescales are matched to the WR lifetime and core-collapse SNe are either suppressed or delayed (e.g., via direct collapse to BH), N/O can be enhanced for short intervals (Fukushima et al., 16 Apr 2024, Zhang et al., 7 Feb 2025).
• Binary Mass Transfer: In rare cases, especially for NEMP field stars ([N/Fe] $>2$), mass transfer from an intermediate-mass AGB companion is implicated, but this is rare in GCs because of the dynamical disruption of binaries (Simpson et al., 2019).
• Differential Galactic Winds and Star Formation Physics: Galactic winds that preferentially expel O-rich SN ejecta but not N-rich AGB or WR wind products artificially raise the N/O ratio in the residual ISM if a strong, short burst of star formation is combined with such selective winds (Vincenzo et al., 2016, Rizzuti et al., 6 Dec 2024).

3. Cluster Formation, Multiple Populations, and Environmental Constraints

The cluster environment is fundamental in producing high-[N/O] stars:

• High [N/O] is most readily attained in massive, dense clusters ($M_\star \gtrsim 10^6$ M$_\odot$, $\Sigma_\star \gtrsim 10^{2.5}$ M$_\odot$ pc$^{-2}$), where the deep potential is able to retain stellar ejecta, and the star formation timescale is fine-tuned (typically $3–10$ Myr) relative to the WR and AGB evolutionary timescales (Kim et al., 2023, Belokurov et al., 2023, Fukushima et al., 16 Apr 2024, Ji et al., 18 May 2025).
• The presence of multiple populations (e.g., “2P” stars) with enhanced N, Na, Al, and depleted O, Mg (the light element anti-correlation) is a defining property of GC evolution, resulting from sequential star formation out of gas polluted by previous generations (0902.1773, Gratton et al., 2014, Marques-Chaves et al., 2023).
• This environment-dependent formation is further supported by the observation that NOEGs are only found in extremely dense, compact star-forming regions—suggesting a close analogy to proto-globular cluster conditions at high redshift (Ji et al., 18 May 2025).

4. Chemo-dynamical Tagging and the Connection to Galactic Halo Assembly

High-[N/O] field stars can be linked to their GC origin through combined chemical and dynamical tagging:

• Stars exhibiting high [N/O], [Al/Fe], Na/Fe are tagged as GC debris even after escape or GC dissolution (Belokurov et al., 2023, Kane et al., 4 Sep 2025).
• Chemo-dynamical association is quantified by comparing probability densities in phase-space (e.g., energy, angular momentum, metallicity space) between field stars and surviving GCs. The most massive, low-energy (inner Galaxy) GCs show the highest degree of association (Kane et al., 4 Sep 2025).
• However, the association signal is diluted when accounting for the effects of dynamical friction, bar-induced orbital evolution, and the fact that even “chemically normal” field stars in the inner halo may be former cluster members. Precise star-to-cluster tracing becomes unreliable, but the population-level evidence is robust for a substantial GC contribution to the halo’s high-[N/O] stars (Kane et al., 4 Sep 2025, Belokurov et al., 2023).
• An increasing fraction of high-[N/O] stars at lower metallicities ($\mathrm{[Fe/H]} < -2$) is consistent with early, large-scale GC dissolution and in situ cluster-dominated assembly during the Milky Way’s Aurora phase (Belokurov et al., 2023).

5. High-[N/O] Phenomenology in Distant Galaxies and the Cosmic Context

JWST spectroscopy has uncovered a significant population of high-redshift galaxies with supersolar N/O at low O/H—the NOEGs— offering direct analogs to the N-rich conditions of early GC formation:

• Objects such as GN-z11 and CEERS-1019 ($z\sim 8.7$–10.6) exhibit log(N/O) $> -0.2$ at 12+log(O/H) $\sim$ 7.7, almost an order of magnitude higher than ordinary galaxies at similar metallicity (Charbonnel et al., 2023, Marques-Chaves et al., 2023, Nandal et al., 5 Feb 2024).
• These systems are extremely compact and have high stellar mass surface densities, aligning with expectations for proto-cluster formation in the early universe (Ji et al., 18 May 2025).
• The high frequency of NOEGs at $z\gtrsim 4$ is matched by elevated N/O in the MW’s Aurora-phase field stars and GCs, suggesting that cluster-dominated star formation and rapid in situ chemical enrichment were ubiquitous at early times (Charbonnel et al., 2023, Marques-Chaves et al., 2023, Belokurov et al., 2023, Ji et al., 18 May 2025).
• Comparative studies reveal chemical tracks (N/O vs O/H) of NOEGs that closely match those expected from both self-enrichment in proto-cluster environments and the observed MW GC populations (Ji et al., 18 May 2025).

6. Alternative Explanations, Uncertainties, and Theoretical Considerations

Several theoretical models and enrichment channels have been proposed and critically evaluated:

• While very massive Pop III stars (2000–9000 M$_\odot$) can, under certain circumstances, produce the extreme [N/O] found in some high-$z$ galaxies (Nandal et al., 5 Feb 2024), chemical evolution models demonstrate that short, intense bursts of star formation coupled with differential, oxygen-selective galactic winds are sufficient to reproduce observed N/O without invoking exotic nucleosynthetic sources (Rizzuti et al., 6 Dec 2024, Vincenzo et al., 2016).
• The presence of direct-collapse WR stars is shown to temporarily elevate N/O in young galaxies, especially where star formation timescales and gas densities allow for the retention of N-rich winds and suppression of oxygen yield by failed supernovae (Zhang et al., 7 Feb 2025, Fukushima et al., 16 Apr 2024).
• Binary mass transfer accounts for a small fraction of N-rich field stars (notably NEMP stars with [N/Fe] $>2$), but this process is inefficient in cluster environments (Simpson et al., 2019).
• Theoretical and observational evidence point to the necessity of considering cluster mass, compactness, feedback, and the detailed timing of enrichment and star formation. Only massive, dense clusters with star formation episodes lasting several Myr after the onset of WR winds, and prior to widespread SN onset, can produce large populations of high-[N/O] stars via self-enrichment (Fukushima et al., 16 Apr 2024).
• Rotational mixing in single massive stars increases N/O modestly. However, the most extreme N-rich stars (e.g., ON stars, second-generation GC stars) demand more efficient or multiple channels (binary evolution, enhanced mixing, or formation in dense cluster environments) (Martins et al., 2015, Martins et al., 2011).

7. Implications for Galaxy Evolution and Future Studies

The ubiquity of high-[N/O] stars in GCs, MW field stars, and high-$z$ NOEGs has far-reaching implications:

• The fraction of high-[N/O] stars is a sensitive tracer of the transition from cluster-dominated, bursty star formation (Aurora phase) to extended, quiescent disk formation in the MW and other galaxies (Belokurov et al., 2023).
• The strong environmental dependence of high N/O signals indicates that clustered self-enrichment and rapid recycling of massive star ejecta in compact starbursts were commonplace in the early universe and largely absent in later disk-dominated phases (Ji et al., 18 May 2025, Belokurov et al., 2023).
• Measuring the occurrence, chemical properties, internal dynamics, and environmental context of high-[N/O] stars and galaxies with next-generation spectroscopic surveys will directly test scenarios for globular cluster, nuclear cluster, and early galaxy formation, and illuminate the interplay of feedback, cluster physics, and ISM mixing in regulating rapid chemical evolution.

In summary, high-[N/O] stars represent robust signatures of clustered, rapid chemical self-enrichment in regions of intense, compact star formation. Their chemical, dynamical, and spatial distribution in both the Milky Way and the high-redshift universe encode key information about massive cluster formation, the early assembly of galaxies, and the transformation from the primordial halo to the modern disk. Their paper interlinks stellar evolution, cluster dynamics, nucleosynthetic physics, and cosmological galaxy formation (0902.1773, Gratton et al., 2014, Charbonnel et al., 2023, Kim et al., 2023, Belokurov et al., 2023, Marques-Chaves et al., 2023, Nandal et al., 5 Feb 2024, Fukushima et al., 16 Apr 2024, Rizzuti et al., 6 Dec 2024, Zhang et al., 7 Feb 2025, Ji et al., 18 May 2025, Kane et al., 4 Sep 2025).