Revisiting the case of R Monocerotis: Is CO removed at R < 20 au?

Context. To our knowledge, R Mon is the only B0 star in which a gaseous Keplerian disk has been detected. However, there is some controversy about the spectral type of R Mon. Some authors propose that it could be a later B8e star, where disks are more common. Aims. Our goal is to re-evaluate the R Mon spectral type and characterize its protoplanetary disk. Methods. The spectral type of R Mon has been re-evaluated using the available continuum data and UVES emission lines. We used a power-law disk model to fit previous 12CO 1 →0 and 2 →1 interferometric observations and the PACS CO data to investigate the disk structure. Interferometric detections of 13CO J = 1 →0, HCO+ 1 →0, and CN 1 →0 lines using the IRAM Plateau de Bure Interferometer (PdBI) are presented. The HCN 1 →0 line was not detected. Results. Our analysis confirms that R Mon is a B0 star. The disk model compatible with the 12CO 1 →0 and 2 →1 interferometric observations falls short of predicting the observed fluxes of the 14 < Ju < 31 PACS lines; this is consistent with the scenario in which some contribution to these lines is coming from a warm envelope and/or UV-illuminated outflow walls. More interestingly, the upper limits to the fluxes of the Ju > 31 CO lines suggest the existence of a region empty of CO at R ≲ 20 au in the protoplanetary disk. The intense emission of the HCO+ and CN lines shows the strong influence of UV photons on gas chemistry. Conclusions. The observations gathered in this paper are consistent with the presence of a transition disk with a cavity of Rin ≳ 20 au around R Mon. This size is similar to the photoevaporation radius that supports the interpretation that UV photoevaporation is main disk dispersal mechanism in massive stars [ABSTRACT FROM AUTHOR]