The functionalization of carbon nanotubes (CNTs) is an ongoing and actively researched topic in the scientific community. The modification of tubes through functionalities leads to new physical properties of the tubes and opens the way to real nanotechnological applications. Functionalization methods can be classified into three different classes - endohedral, covalent, and non-covalent approaches. All three methods come with specific advantages and challenges, as I will discuss in the introductory chapters of this thesis. In the first part of this thesis, I will introduce a new covalent functionalization routine that we have developed. It is based on a nitrine based [2+1] cycloaddition reaction and for the very first time in literature, it maintains the extended π-network and preserves the outstanding optoelectronic properties of carbon nanotubes, even at an high degree of functionalization. This new method we developed offers a robust way to attach functional moieties on the tubes and creates a new toolbox for advanced tailoring of the nanotubes properties. In the second part of this work I will describe how non-covalent functionalization methods can be used to attach moieties on the CNTs sidewalls and I will compare their outcome with the results from our new established [2+1] cycloaddition. I will discuss in detail the influence of three different type of functional moieties immobilized to the tubes’ sidewalls: The dipole switch spiropyran, the molecular dye perylene, and gold nanoparticles. The comparative study between covalent and non-covalent functionalization methods shows that functional moieties can be used to strongly influence the optical property of tubes. The study furthermore shows that the tubes’ physical properties are also highly sensitive towards the attachment routine used for immobilization of the functional group. The same moiety yield different effects depending upon how it has been attached onto the tubes.