Amorphous carbon films have widespread applications in many domains such as protective coatings, optoelectronics or biomedicals owing to their high chemical stability, mechanical properties, tribological performance and resistance to corrosion. The addition of nitrogen into the amorphous carbon matrix multiplies the possible applications since it significantly reduces the stress, improves the thermal and chemical stability, enhances the elastic properties providing hardness and flexibility and modifies the electro-optical properties. However, for the development of industrially relevant materials and their optimisation, there is still a huge need of obtaining a complete understanding of the detailed bonding configuration of C and N atoms in the a-CN structure since their physical properties are strongly correlated to their local structure. The huge complexity in the different hybridisation states in a-CN makes it difficult to work out the actual local structure using a single characterisation technique (Fig. 1). Nevertheless, this information is essential for tailoring of their properties as required in the proposed applications, such as magnetic storage disks, head readers, field emission displays, sensors and protective coatings for biomedical devices.In addition to the research and applications of disordered carbon nitride films, it must be remembered that the research on carbon nitride compounds started y15 years ago with the theoretical prediction of a material that could be harder than diamond, b-C 3 N 4 with a similar crystalline structure to the well known Si 3 N 4 . Figure 2 shows the bonding configuration expected for the super hard phase. Based upon the semi-empirical equation deduced by Cohen, 1 which relates high bulk modulus with short bond lengths and low ionicity, Sung and Sung 2 realised that crystalline phases of covalently bonded carbon and nitrogen materials might be harder than diamond. Liu and Cohen 3 then used pseudopotential calculations to determine the possible crystalline structures and found two super hard covalent bonded phases (a and b) that were predicted to be metastable under ambient conditions. Following this work, a great number of groups have attempted to synthesise C 3 N 4 using a variety of techniques, all of them suitable for the formation of metastable phases, such as thin film deposition or high pressure-high temperature systems.Other hard crystalline structures have also been proposed. Teter and Hamley 4 studied five different structures with the same stoichiometry. Using ab initio molecular dynamics, they calculated the energy-volume (EV) curves for several possible phases of C 3 N 4 and also the elastic constants for the cubic phases. In particular, the pressure, enthalpy and bulk modulus at zero pressure were evaluated. More importantly, Teter also estimated the shear moduli since the critical structural parameter describing the hardness is the shear and not the bulk modulus. Among all the calculated shear moduli of the C 3 N 4 phases, none of them was higher than the ...