The cosmic microwave background (CMB) has been a treasure trove for cosmology. CMB experiments such as Planck have put stringent constraints on the now standard model for Cosmology: ΛCDM. Over the next decade, current and planned CMB experiments are expected to exhaust nearly all primary CMB information. To reach beyond the ΛCDM model of the universe, requires to observe beyond the CMB primary modes. Rayleigh scattering of the CMB is one source of additional cosmological information. It is caused by the additional scattering of CMB photons by neutral species formed during recombination and exhibits a strong and unique frequency scaling (∝ ν4). In this thesis, we explore the detectablity and the cosmological information carried by this small distortion of the CMB signal. In particular, we present detectability forecasts for the next generation of CMB surveys. We also show that more futuristic experiments, such as PICO will be able to use Rayleigh scattering to improve constraints on Neff and Σmν. Even more futuristic experiments will be able to use the Rayleigh scattering signal to improve measurements of primordial Non-Gausianities. We will provide heuristic arguments for these improvements. Detections of the Rayleigh scattering signal and secondary anisotropies in general are made difficult by the presence of astrophysical foregrounds and instrument systematics which hinder the signals of interest. In the second part of this thesis, we study the impact of these noise sources in the context of two future ground-based CMB surveys: the Simons Observatory (SO) and the CCAT-prime collaboration. In particular, we will present an implementation of a novel component separation method developed for SO. This semi parametric method has been designed to mitigate the limitations of the currently used foreground cleaning techniques. Secondly, we will present how instrumental systematics may impact the detectability of Rayleigh scattering with CCAT-prime.