The goal of the study is the investigation of the build-up of copper clusters in reactors pressure vessel (RPV) steels. The study comprises tests on RPV surveillance samples provided by a nuclear power plant, and on RPV reference samples. The reference steels, named Japanese Reference Quality (JRQ), were provided by the International Atomic Energy Agency (IAEA). Moreover the study was completed with the investigation of some unirradiated binary alloys. To better understand the mechanism of cluster build-up in a simplified system, tests were performed on the binary alloy Fe-Cu containing 1.3 at % Cu. The samples were annealed at 775 K for different times. They were analysed using x-ray absorption fine structure (XAFS) spectroscopy at the Cu K-edge, x-ray diffraction (XRD) and transmission electron microscopy (TEM). The results show that Cu cluster formation occurs even with short annealing times. These clusters with a diameter of a few nm have been observed to switch easily from bcc Fe-like to fcc Cu-like structure. While a short annealing time of 2.5 h at 775 K does practically not change the good dilution of Cu in the bcc Fe matrix, an annealing treatment for 312 h leads to large fcc Cu precipitates. A linear combination analysis suggests that in the sample annealed for 8 h, Cu clusters are formed mostly in the same structure as the bcc matrix. A coexistence of bcc and fcc clusters is obtained for 115 h of annealing. TEM indicates the presence of precipitates with a diameter up to 60 nm for an annealing time of 312 h, and XRD provides complementary data about the clusters size distributions in differently annealed samples. The reactor pressure vessel (RPV) reference steel from IAEA submitted to neutron irradiation in a research reactor and subsequently thermally annealed and re-irradiated was investigated by XAFS spectroscopy. It could be expected that Cu and Ni atoms form nano-clusters. In the unirradiated sample and in the irradiated sample no significant clustering is detected. In all irradiated and subsequently annealed samples an increase of Cu and Ni atom densities are recorded around the x-ray absorbing atoms. Furthermore, the density of Cu and Ni atoms determined in the first and second shell around the absorber is found to be affected by irradiation and the annealing treatment. The comparison of the XAFS data at the Cu and Ni K-edges shows that these elements are in a bcc structure like the Fe host lattice. However, the local irradiation damage reduces the number of next neighbour atoms around copper; while the annealing reduces the number of vacancies. The number of Cu and Ni atoms around (Cu or Ni) absorbers then increases. The IAEA samples are very well characterised due to earlier analysis by atom probe tomography (APT) at the Oak-Ridge National Lab (ORNL) and mechanical tests, at the Paul Scherrer Institut. The possibility to compare XAFS and APT provided an invaluable opportunity to understand the differences and similarities among the two techniques. XAFS and APT are both very sensitive to nanostructures and both are element selective techniques. XAFS is highly sensible to the average environment of the absorber element, within < 0.5 nm, so it delivers information about the short range order. The results show a general agreement and the differences were considered within the experimental error. Finally, surveillance samples from a Swiss nuclear power plant (KKG Kernkraftwerk Gösgen) were analysed by XAFS. The number of Cu and Ni atoms determined in the first and second shell around the absorbers is affected by irradiation and temperature. The comparison of the EXAFS data at the Cu and Ni K-edges shows that these elements are in a crystallographic structure similar to bcc Fe. There are indications that the formation of Cu and Ni clusters differs significantly; irradiation damage reduces the next neighbour number for both elements. Clusters were detected only around copper atoms: the number of Cu atoms in the first two shells was found to increase with the irradiation time. With EXAFS it could be verified that neutron irradiation in a power reactor leads to small changes of the local atomic environment around the investigated Cu and Ni atoms. Moreover, most of the local structural changes are obtained for the sample with the longest irradiation time. This lets assume that the reduction of the number of next neighbours in the bcc structure around Cu and Ni in the steel matrix is steadily increasing with the years of irradiation in the reactor. However it can be expected that the elevated temperature helps refilling the irradiation induced vacancies by a permanent annealing process. The detailed EXAFS analysis has demonstrated that the number of Cu atoms in the local vicinity of copper is enriched during irradiation in the power plant, while the local atomic environment around nickel seems to be less affected. Cu and Ni atoms behave similarly in the IAEA samples, but they show differences in the power reactor samples. The reasons for the different behaviour reside in the different irradiation processes: IAEA samples were strongly irradiated and then annealed at high temperature, while the KKG samples were irradiated at a higher temperature than the IAEA samples without any post irradiation annealing. Obviously neutron irradiation forms many vacancies and vacancies clusters, which are easily filled by solute atoms. In case of irradiation in the power reactor, there are fewer vacancies and the temperatures involved are lower than the annealing temperature of the IAEA reference samples; the atom cluster formation of Cu is slowed down and Ni is even less affected.