From the structural point of view, A. Hoffman et al suggest  that the transformation of diamond under ion beam irradiation involves a stage of amorphous sp3 phase formation, at a critical dose, followed, upon further implantation, by a sp3 to sp2 bond transition. Below the critical dose, which corresponds to a critical density of defects, thermal annealing can nearly restore the diamond lattice, and above it, graphitization of the damage can occur . Because of annealing effects (defects diffusion and crystal reconstruction), this critical dose (and thus the defect density) depends on the implantation temperature, for the same ion and energy [5,27]. Indeed, if, for instance, the vacancies and the interstitials are spatially distributed in the same manner (as in a ``cold'' implantation process), they may annihilate under annealing process, leading to diamond reconstruction. Furthermore, for an implantation temperature greater than 800 K (at which vacancies are thought to become mobile), the formation of stable defects is inhibited by diffusive process . Thus, even high dose implantion in diamond at high temperature would never lead to amorphization.
Another feature of the damage induced in diamond crystal is the change in electrical resistivity as a function of the implantation dose [5,28,26]. A structure rich in sp2 bonds should be electrically conductive due to the delocalized electron available. Since the defect density increases with the number of bombarding atoms, and sp3 to sp2 bond transition may occur during the implantation, the resistivity of diamond roughly decreases with increasing dose. It was found that for low dose implantation, the logarithm of the resistivity changes linearly with T-1/4 (where T is the measurement temperature), as expected for variable range hopping conduction , the sp2 bonds acting as the hopping centers . At higher dose, these damage centers can overlap, leading to a metallic conduction [5,28].
It has been shown that implantation in diamond of atoms like xenon or krypton at high temperature yields n-type conductivity . Since these atom are inert and do not interact electrically with the diamond matrix, the effect of their implantation is only to create damage. The n-type conductivity is therefore due to the defects formed, and should also appears when doping by ion implantation is carried out. In that case, the substantial conductivity induced by the defects may also screen the conductivity due to the dopant atoms. Thus, control experiments must be done in parallel to the doping itself, in which a damage region is created by inert atoms implantation, with the same damage profile and subjected to the same annealing process. In this way, the conductivity measured can be attributed either to the defects induced by implantation, or to the dopant itself.
The damage created can also give rise to energy levels in the band gap, which can compensate the doping levels and modify the electrical properties of the sample. For example, boron doped p-type diamond can be partially compensated by donor levels induced by defects, to a percentage of 5 % . Moreover, these defects may interact with the dopant itself to form electrically inactive complexes. The passivation of the impurities can thus affect the doping control. Finally, scattering centers created during the ion implantation and related to the damage, reduce the carrier mobility. Consequently, the essential condition to use diamond as an efficient semiconductor is to reduce as far as possible radiation damage and to drive dopant atoms to suitable sites.
As explained above, self-interstitial diffusion occurs at about 300 K while vacancies are immobile until about 800 K . Based on these experimental facts, Prins  suggested a scheme called cold-implantation-rapid annealing (CIRA) for optimizing the properties of the doped diamond. Accordingly, the dopant implantation is carried out at low temperature ( 77 K) to freeze in the damage cascade, and thus to minimize the number of defects created. The sample is then rapidly heated to a temperature larger than 800 K, at which the interstitials can move and recombine with vacancies before any further diffusion out of the implanted layer takes place. Finally, the system is annealed at high temperature ( 1700 K) to reduce as far as possible radiation damage and to drive dopant atoms to suitable sites, making them electrically active centers.
To increase the probability for the dopant to be located in a region rich in vacancies, it was suggested  to broaden the implanted layer by multiple energy implantation. This process should raise the rate of dopant-vacancy recombination in the case of substitutional dopants and improve the properties of the doped sample. Prawer et. al.  proposed to implant the dopant at a very high energy of few Mev to form a deeply buried impurity layer. The defects could be more easly annealed since the surrounding diamond matrix applies a hugh pressure on the damaged region, especially at large depths. This could facilitate annealing, and thus increase the carrier concentration and their mobility, and reduce the compensation ratio.