We carry out a phase-curve analysis of the KELT-9 system using photometric observations from NASA's Transiting Exoplanet Survey Satellite (TESS). The measured secondary eclipse depth and peak-to-peak atmospheric brightness modulation are-+ 650 15 14 and 566±16 ppm, respectively. The planet's brightness variation reaches maximum 31±5minutes before the midpoint of the secondary eclipse, indicating a 5°.2±0°.9 eastward shift in the dayside hot spot from the substellar point. We also detect stellar pulsations on KELT-9 with a period of 7.58695±0.00091hr. The dayside emission of KELT-9b in the TESSbandpass is consistent with a blackbody brightness temperature of 4600±100 K. The corresponding nightside brightness temperature is 3040±100 K, comparable to the dayside temperatures of the hottest known exoplanets. In addition, we detect a significant phase-curve signal at the first harmonic of the orbital frequency and a marginal signal at the second harmonic. While the amplitude of the first harmonic component is consistent with the predicted ellipsoidal distortion modulation assuming equilibrium tides, the phase of this photometric variation is shifted relative to the expectation. Placing KELT-9b in the context of other exoplanets with phase-curve observations, we find that the elevated nightside temperature and relatively low day-night temperature contrast agree with the predictions of atmospheric models that include H 2 dissociation and recombination. The nightside temperature of KELT-9b implies an atmospheric composition containing about 50% molecular and 50% atomic hydrogen at 0.1bar, a nightside emission spectrum that deviates significantly from a blackbody, and a 0.5-2.0 μm transmission spectrum that is featureless at low resolution.

KELT-9 b is an ultra-hot Jupiter transiting a rapidly rotating, oblate early-A-type star in a polar orbit. We model the effect of rapid stellar rotation on KELT-9 b's transit light curve using photometry from the Transiting Exoplanet Survey Satellite to constrain the planet's true spin-orbit angle and to explore how KELT-9 b may be influenced by stellar gravity darkening. We constrain the host star's equatorial radius to be 1.089±0.017 times as large as its polar radius and its local surface brightness to vary by ∼38% between its hot poles and cooler equator. We model the stellar oblateness and surface brightness gradient and find that it causes the transit light curve to lack the usual symmetry around the time of minimum light. We take advantage of the light-curve asymmetry to constrain KELT-9 b's true spin-orbit angle (- +  87 11 10), agreeing with Gaudi et al. that KELT-9 b is in a nearly polar orbit. We also apply a gravity-darkening correction to the spectral energy distribution model from Gaudi et al. and find that accounting for rapid rotation gives a better fit to available spectroscopy and yields a more reliable estimate for the star's polar effective temperature.

We constrain the true spin-orbit alignment of the KOI-89 system by numerically fitting the two Kepler photometric lightcurves produced by transiting planets KOI-89.01 and KOI-89.02. The two planets have periods of 84.69 days and 207.58 days, respectively. We find that the two bodies are low-density giant planets with radii 0.45 ± 0.03 R jup and 0.43 ± 0.05 R jup and spin-orbit misalignments 72 • ± 3 • and 73 •+11 −5 , respectively. Via dynamic stability tests we demonstrate the general trend of higher system stability with the two planets close to mutual alignment and estimate their coalignment angle to 20 • ± 20 • -i.e. the planets are misaligned with the star but may be aligned with each other. From these results, we limit KOI-89's misalignment mechanisms to star-disk-binary interactions, disk warping via planet-disk interactions, planet-planet scattering, Kozai resonance, or internal gravity waves.

We fit the Kepler photometric light curve of the KOI-368 system using an oblate, gravity-darkened stellar model in order to constrain its spin-orbit alignment. We find that the system is relatively well-aligned with a sky-projected spin-orbit alignment of λ = 10 • ± 2 • , a stellar obliquity of ψ = 3 • ± 7 • , and a true spinorbit alignment of ϕ = 11 • ± 3 • . Although our measurement differs significantly from zero, the low value for ϕ is consistent with spin-orbit alignment. We also measure various transit parameters of the KOI-368 system: R KOI−368 = 2.28 ± 0.02R , R p = 1.83 ± 0.02R jup , and i = 89.221 • ± 0.013 • . This work shows that our gravity-darkened model can constrain long-period, well-aligned planets and M-class stars orbiting fast-rotators, allowing for measurement of a new subcategory of transiting bodies.

We present an optical-to-infrared transmission spectrum of the inflated sub-Saturn KELT-11b measured with the Transiting Exoplanet Survey Satellite (TESS), the Hubble Space Telescope (HST) Wide Field Camera 3 G141 spectroscopic grism, and the Spitzer Space Telescope (Spitzer) at 3.6 μm, in addition to a Spitzer 4.5 μm secondary eclipse. The precise HST transmission spectrum notably reveals a low-amplitude water feature with an unusual shape. Based on free-retrieval analyses with varying molecular abundances, we find strong evidence for water absorption. Depending on model assumptions, we also find tentative evidence for other absorbers (HCN, TiO, and AlO). The retrieved water abundance is generally ≲0.1× solar (0.001–0.7× solar over a range of model assumptions), several orders of magnitude lower than expected from planet formation models based on the solar system metallicity trend. We also consider chemical-equilibrium and self-consistent 1D radiative-convective equilibrium model fits and find that they, too, prefer low metallicities ([M/H] ≲ −2, consistent with the free-retrieval results). However, all of the retrievals should be interpreted with some caution because they either require additional absorbers that are far out of chemical equilibrium to explain the shape of the spectrum or are simply poor fits to the data. Finally, we find that the Spitzer secondary eclipse is indicative of full heat redistribution from KELT-11b’s dayside to nightside, assuming a clear dayside. These potentially unusual results for KELT-11b’s composition are suggestive of new challenges on the horizon for atmosphere and formation models in the face of increasingly precise measurements of exoplanet spectra.