Exoplanet atmospheres
Directly imaged giants and young brown dwarfs are still warm from formation, so their thermal emission is bright enough for detailed spectroscopy. That advantage turned high-contrast targets into laboratories for atmospheric physics: molecular inventories, temperature structure, clouds, and — more recently — isotopes (Konopacky et al. 2013; Snellen et al. 2014; review by Snellen 2025).
The broader motivation is formation. Snow lines in discs rearrange the volatile budget available for accretion (Öberg et al. 2011), and elemental ratios such as C/O have been proposed as fingerprints of that history (Madhusudhan 2019). Isotopes offer a complementary handle. Zhang et al. (2021) reported the first $^{13}$CO detection in an exoplanet atmosphere (TYC 8998-760-1 b), with a $^{12}\mathrm{C}/^{13}\mathrm{C}$ ratio well below the local ISM, interpreted as ice-rich accretion beyond the CO snow line. Follow-up work with JWST has since opened oxygen isotopes as well (e.g. VHS 1256 b). The open question is no longer whether isotopes are measurable, but what a growing sample actually says — and how often host–companion comparisons are possible.
Where I contribute
I lead retrievals that turn that programme into a comparative census within the ESO SupJup Survey and complementary JWST observations.
For β Pic b, eleven nights of VLT/CRIRES+ K-band spectroscopy yield $^{12}\mathrm{C}/^{13}\mathrm{C}=58^{+18}_{-15}$, with $T_{\mathrm{eff}}\approx1630\,\mathrm{K}$, near-solar metallicity, and solar-like C/O (A&A 2026). That places a classic imaged planet — interior to the CO snow line — on the same isotopic footing as the early Zhang et al. result, but consistent with the present-day ISM rather than strongly $^{13}$C-enriched. Multi-epoch analysis matters here: night-to-night scatter has to enter the error budget, or the ratio looks more precise than it is.
GQ Lup goes a step further by measuring both the companion and the star. GQ Lup B and its K7 host share indistinguishable carbon isotope ratios ($53^{+7}_{-6}$ and $51^{+10}_{-8}$), favouring a shared natal reservoir over a chemically distinct core-accretion pathway (SupJup IV). Host–companion pairs like this are still rare, yet they are exactly what the formation debate needs.
Young free-floating brown dwarfs fill out the sample. With CRIRES+, TWA 28 and J0856 give ISM-like ratios and the first HF detection in a brown dwarf (SupJup II). With JWST/NIRSpec, TWA 27A and TWA 28 require joint atmosphere–disc models: warm continuum and hot optically thin CO sit on top of rich photospheres, while carbon and oxygen isotopes are constrained from the CO fundamental (A&A 2025). And for the close-in T dwarf HD 19467 B, residual speckles can masquerade as chemistry unless they are modelled jointly with the atmosphere — otherwise abundance and isotope posteriors look tight for the wrong reasons (A&A 2026).
In short: the field showed that isotopes are accessible; my papers ask how they behave across ages, separations, and observing modes — and what systematics still hide in the data.