International audienceExperiments and computations are performed for the CuMOCVD fromcopper(I) N,N′-di-isopropylacetamidinate [Cu(iPr–Me–amd)]2 or [Cu(amd)]2 where amd = CH(CH3)2NC(CH3)NCH(CH3)2. The a priori choice of this precursor is dictated mainly by its oxygen and halogen–free ligands allowing co-deposition with oxophilic elements such as Al and by its ability to provide conformal Cu films in atomic layer deposition processes. The nucleation delay and the deposition rate as a function of deposition temperature and the evolution of the deposition rate along the radius of the substrate holder are experimentally determined with depositions performed at 1333 Pa in a vertical, warm wall, MOCVD reactor. With the aim to propose a kinetic scenario for Cu deposition, based on recently published experimental results for the decomposition of [Cu(amd)]2, a predictive 3D model of the process is built, based on the mass, momentum, energy and species transport equations. In agreement with the previously mentioned experimental results, it is demonstrated that a single surface reaction is responsible for the deposition of Cu. Two surface kinetics expressions are implemented depending on the deposition regime; a simple Arrhenius type expression in the reaction limited regime and a Langmuir–Hinshelwood type expression prevailing in the transport limited regimewhich takes into account the inhibition effects. The two different kinetics designate a modification in the surface reaction mechanism. The results show good agreement between experiments and computations. Complementary computations are performed, in order to compare the deposition rates of the Cu films deposited via the [Cu(amd)]2 and the (hfac)Cu(VTMS) and Cu(hfac)2 so as to determine relative advantages and disadvantages of Cu MOCVD from [Cu(amd)]2