Nuclear fusion in a tokamak is a promising source of energy. However, a question arises: which plasma configuration is most likely to produce net energy? In order to contribute to answering this, during this PhD, we will study the impact of magnetic geometry (comparison between positive and negative triangularity) on the collisional and turbulent transport of tungsten (W). The performance of a tokamak strongly depends on the energy confinement it can achieve. The latter degrades significantly due to turbulent transport and radiation (primarily from W). On ITER, the tolerated amount of W in the core of the plasma is about 0.3 micrograms. Experiments have shown that the plasma geometry with negative triangularity (NT) is beneficial for confinement as it significantly reduces turbulent transport. With this geometry, it is possible to reach confinement levels similar to those of the ITER configuration (H-mode in positive triangularity), without the need for a minimum power threshold and without the associated plasma edge relaxations. However, questions remain: what level of W transport is found in NT compared to a positive geometry? What level of radiation can be predicted in future NT reactors? To contribute to answering these questions, during this PhD, we will evaluate the role of triangularity on impurity transport in different scenarios in WEST. The first phase of the work is experimental. Subsequently, the modeling of impurity transport will be carried out using collisional and turbulent models. Collaboration is planned with international plasma experts in NT configurations, with UCSD (United States) and EPFL (Switzerland).