2019ApJ...873...32M

A carbon-to-oxygen ratio (C/O) of around unity is believed to act as a natural separator of water- and methane-dominated spectra when characterizing exoplanet atmospheres. In this paper, we quantify the C/O ratios at which this separation occurs by calculating a large self-consistent grid of cloud-free atmospheric models in chemical equilibrium using the latest version of petitCODE. Our study covers a broad range of parameter space: 400 K < T_eff < 2600 K, 2.0 < log(g) < 5.0, -1.0 < [Fe/H] < 2.0, 0.25 < C/O < 1.25, and stellar types from M to F. We make the synthetic transmission and emission spectra, as well as the temperature structures, publicly available. We find that the transition C/O ratio depends on many parameters, such as effective temperature, surface gravity, metallicity, and spectral type of the host star, and could have values less than, equal to, or higher than unity. By mapping all of the transition C/O ratios, we propose a "four-class" classification scheme for irradiated planets in this temperature range. We find a parameter space where methane always remains the cause of dominant spectral features. Detection of CH₄ in this region, or the lack of it, provides a diagnostic tool to identify the prevalence of cloud formation and nonequilibrium chemistry. As another diagnostic tool, we construct synthetic Spitzer Infrared Array Camera color diagrams showing two distinguishable populations of planets. Since most of the exoplanet atmospheres appear cloudy when studied in transmission, we regard this study as a starting point of how such a C/O-sensitive observation-based classification scheme should be constructed. This preparatory work will have to be refined by future cloudy and nonequilibrium modeling to further investigate the existence and exact location of the classes, as well as the color-diagram analysis.