Context. The detection and characterization of exoplanet atmospheres is currently one of the main drivers pushing the development of new observing facilities. In this context, high-resolution spectrographs are proving their potential and showing that high-resolution spectroscopy will be paramount in this field. Aims. We aim to make use of ESPRESSO high-resolution spectra, which cover two transits of HD 209458b, to probe the broadband transmission optical spectrum of the planet. Methods. We applied the chromatic Rossiter–McLaughin method to derive the transmission spectrum of HD 209458b. We compared the results with previous HST observations and with synthetic spectra. Results. We recover a transmission spectrum of HD 209458b similar to the one obtained with HST data. The models suggest that the observed signal can be explained by only Na, only TiO, or both Na and TiO, even though none is fully capable of explaining our observed transmission spectrum. Extra absorbers may be needed to explain the full dataset, though modeling approximations and observational errors can also be responsible for the observed mismatch. Conclusions. Using the chromatic Rossiter–McLaughlin technique, ESPRESSO is able to provide broadband transmission spectra of exoplanets from the ground, in conjunction with space-based facilities, opening good perspectives for similar studies of other planets.

Atmospheric studies at high spectral resolution have shown the presence of molecules, neutral and ionised metals, and hydrogen in the transmission spectrum of ultra-hot Jupiters, and have started to probe the dynamics of their atmospheres. We analyse the transmission spectrum of MASCARA-1b, one of the densest ultra-hot Jupiters orbiting a bright (V=8.3) star. We focus on the Ca ii H&K, Na i doublet, Li i, Hα, and K i D1 spectral lines and on the cross-correlated Fe i, Fe ii, Ca i, Y i, V i, V ii, CaH, and TiO lines. For those species that are not present in the stellar spectrum, no detections are reported, but we are able to measure upper limits with an excellent precision (∼ 10 ppm for particular species) thanks to the signal-to-noise ratio (S/N) achieved with Echelle SPectrograph for Rocky Exoplanets and Stable Spectroscopic Observations (ESPRESSO) observations. For those species that are present in the stellar spectrum and whose planet-occulted spectral lines induce spurious features in the planetary transmission spectrum, an accurate modelling of the Rossiter-McLaughlin effect (RM) and centre-to-limb variations (CLV) is necessary to recover possible atmospheric signals.In the case of MASCARA-1b, this is difficult due to the overlap between the radial velocities of the stellar surface regions occulted by MASCARA-1b and the orbital track along which the planet atmospheric signal is expected to be found. To try to disentangle a possible planetary signal, we compare our results with models of the RM and CLV effects, and estimate the uncertainties of our models depending on the different system parameters. Unfortunately, more precise measurements of the spin-orbit angle are necessary to better constrain the planet-occulted track and correct for the transit effects in the transmission spectrum with enough precision to be able to detect or discard possible planetary absorptions. Finally, we discuss the possibility that non-detections are related to the low absorption expected for a high surface gravity planet such as MASCARA-1b. Other techniques such as emission spectroscopy may be more useful for exploring their atmospheric composition. Key words. planetary systems -planets and satellites: individual: MASCARA-1 b -planets and satellites: atmospheres -methods: observational -techniques: spectroscopic * Based on guaranteed time observations (GTOs) collected at the European Southern Observatory (ESO) under ESO programme 1102.C-0744 by the ESPRESSO Consortium.

<div>High resolution spectra is the standard source to derive precise radial velocities. With the development of state-of-art instrumentation, like ESPRESSO, we are able to have higher resolutions and long term stability.</div> <div>During a transit the star light from a observer point of view is filtered in the day-night terminator of the exoplanet. As the light transverses it, diverse chemical species, dust, hazes and clouds let an imprint in the host star light. This allows to detect exoplanet atmospheres using transmission spectroscopy. However, a chromatic approach of the Rossiter-McLaughlin (RM) effect also permits to measure the planet radius wavelength dependence as a composition proxy.</div> <div>The Rossiter-McLaughlin effect is an phenomenon in radial velocities. When a star rotates, there is a velocity distribution created by the half of the surface that is moving towards the observer, presenting a intrinsic blueshift, compensated by the redshifted portion rotating away. A transiting planet produces an asymmetric distortion of the stellar line profiles of an observer giving origin to a RM profile.</div> <div>We present CaRM (Chromatic Rossiter-McLaughlin), a code developed to retrieve the transmission spectrum of an exoplanet using HARPS and ESPRESSO data. It employs a Markov chain Monte Carlo algorithm to fit two distinct RM models, with a linear or quadratic limb-darkening law, to Cross Correlated spectra. The orbital parameters retrieved as result, e.g. spin-orbit misalignment, are of foremost importance when constraining the models of planetary formation and evolution.</div> <div>A chromatic approach to the RM has been used in HD 189733b HARPS transit data, which is reproduced by this code with a good degree of agreement to previous results. We expect now to apply CaRM to ESPRESSO data (ESO-VLT), taking this method to a new level.</div>