COSMO (COSmological Monopole Observer) is a ground-based cryogenic Martin Puplett Interferomenter which aims at measure the monopole spectral distortions of the Cosmic Microwave Background (CMB) by comparing the sky signal with a reference blackbody. Isotropic CMB spectral distortions can be produced by processes related to energy injections/extraction or photon production/destruction, therefore holding information about such processes in the thermal history of the Universe, opening a new window not only to the early-Universe but also to phenomena from the world of particle physics: particle decays, dark matter annihilation, reionization, structure formation and so on. To date, no isotropic spectral distortions have been detected and only upper limits have been set on their amplitude to one part in 100000. Nevertheless, since many of these processes are included in the standard cosmological model, the related spectral distortions are guaranteed to exist. CMB spectral distortions represent an independent and complementary source of information about processes in the Universe with respect to temperature and polarization anisotropies, which have already provided invaluable information about the Universe. The observations will be performed in the 150 and 220 GHz atmospheric windows and will take place in Dome-C, Antarctica, the best site on Earth in terms of atmospheric stability. The instrument is fully funded and the cryostat is under construction. The blackbody cavity design is currently under study by means of HFSS simulations, which have rarely been used for the study of such optical elements. A deep performance analysis, followed by dedicated measurements, is required to design and characterize a reference calibrator which needs to be as close as possible to a blackbody, otherwise distortions due to the calibrator itself would be produced.
Currently there is no calibrator available performing as a perfect blackbody ,and only cavities approximating a blackbody have been developed. The approximation is mainly driven by the non-zero reflectance of the inner surface. In order to detect CMB spectral distortions, a calibrator with an emissivity close to unity within the level of the distortions is required, otherwise spectral distortions due to a non-perfect blackbody calibrator will be produced, and this systematic error will dominate the measurements. While for the y-distortions the necessary level is almost reached by instruments as ARCADE-II, for the smaller distortions the requirement is much more demanding, and dedicated simulations that give reliable predictions of the performance are necessary, given also the evident difficulty in measuring the properties of a calibrator so close to a real blackbody. The most common method to measure the properties of a cavity consists of measuring its reflectance, which in turn allows to estimate the emissivity of the cavity. This kind of measurements requires an almost perfect control over undesired reflections from any optical element included in the setup. In addition, a very powerful input source is necessary, such that the reflected component, reduced by many orders of magnitudes by the cavity, is still detectable. Alongside all these difficulties, HFSS simulations provide a verypowerful tool for the cavity performance prediction, allowing to optimize the cavity design by fine-tuning all its features until the desired limit is reached. This approach has been rarely used for the study of blackbody cavities, being exploited only for the low-frequency case of the ARCADE-II calibrator.