Perovskite solar cells are an emerging technology that is evolving rapidly, with power conversion efficiency values that compete with traditional materials such as silicon. In this type of solar cells the photons are absorbed in the perovskite and the charges are extracted using transport materials. By sandwiching the perovskite between a material with and excess of negative (N) or positive (P) charge, one can fabricate a P-i-N or N-i-P structure depending on the deposition order of the materials. Perovskite solar cells have several advantages, mainly the possibility of being very thin thanks to the high absorption coefficient of the perovskite and the wide range of deposition techniques, compatible with industry. From all the deposition techniques, vacuum sublimation stands out due to several advantageous characteristics. This method consists in subliming in high vacuum the required precursor materials for a certain composition, depositing the resulting perovskite onto a substrate. Sublimed perovskite solar cells generally rely on doped organic layers for transporting the electrons and holes photogenerated, which might be chemically unstable and absorb some part of the light, reducing the current generated. This thesis aims to improve vacuum deposited perovskite solar cells, enhancing the stability and performance of the devices. To fulfill this objective, we explore the substitution of the weakest part of the device, the doped transport layers, by implementing the use of transition metal oxides. This type of metal oxides stands out due to the wide range of work functions available and their self-doping, which improves the charge transport thanks to their natural crystal defects. The most interesting metal oxides for the N side are TiO2 and SnO2 due to their adequate energy levels and their proper charge transport. For the P side we chose MoO3 thanks to the possibility of being sublimed and its good hole injection behavior. For these reasons, the thesis is structured as follows: -Chapter 3: Substitution of doped C60 in N-i-P vacuum deposited perovskite solar cells by a TiO2 nanoparticle dispersion, leading to devices with higher efficiencies and better reproducibility. -Chapter 4: Substitution of doped TaTm in P-i-N vacuum deposited perovskite solar cells by MoO3 and then the implementation of this layer in the N-i-P configuration, leading to a design with metal oxides on both contacts. -Chapter 5: Characterization of the P-i-N architecture with MoO3 from chapter 4 under space conditions, which was proven to be very stable and opened the possibility of using perovskite solar cells in high altitude conditions. The substitution of the doped transport layers led to devices with higher power conversion efficiencies (more than 20%, among the highest values for vacuum deposited CH3NH3PbI3 solar cells to date) and more robustness, passing stability tests under space conditions. The work developed in this thesis has opened an interesting field for vacuum deposited perovskite solar cells and changed the main fabrication routes in our laboratoryPerovskite solar cells are an emerging technology that is evolving rapidly, with power conversion efficiency values that compete with traditional materials such as silicon. In this type of solar cells the photons are absorbed in the perovskite and the charges are extracted using transport materials. By sandwiching the perovskite between a material with and excess of negative (N) or positive (P) charge, one can fabricate a P-i-N or N-i-P structure depending on the deposition order of the materials. Perovskite solar cells have several advantages, mainly the possibility of being very thin thanks to the high absorption coefficient of the perovskite and the wide range of deposition techniques, compatible with industry. From all the deposition techniques, vacuum sublimation stands out due to several advantageous characteristics. This method consists in subliming in high vacuum the required precursor materials for a certain composition, depositing the resulting perovskite onto a substrate. Sublimed perovskite solar cells generally rely on doped organic layers for transporting the electrons and holes photogenerated, which might be chemically unstable and absorb some part of the light, reducing the current generated. This thesis aims to improve vacuum deposited perovskite solar cells, enhancing the stability and performance of the devices. To fulfill this objective, we explore the substitution of the weakest part of the device, the doped transport layers, by implementing the use of transition metal oxides. This type of metal oxides stands out due to the wide range of work functions available and their self-doping, which improves the charge transport thanks to their natural crystal defects. The most interesting metal oxides for the N side are TiO2 and SnO2 due to their adequate energy levels and their proper charge transport. For the P side we chose MoO3 thanks to the possibility of being sublimed and its good hole injection behavior. For these reasons, the thesis is structured as follows: -Chapter 3: Substitution of doped C60 in N-i-P vacuum deposited perovskite solar cells by a TiO2 nanoparticle dispersion, leading to devices with higher efficiencies and better reproducibility. -Chapter 4: Substitution of doped TaTm in P-i-N vacuum deposited perovskite solar cells by MoO3 and then the implementation of this layer in the N-i-P configuration, leading to a design with metal oxides on both contacts. -Chapter 5: Characterization of the P-i-N architecture with MoO3 from chapter 4 under space conditions, which was proven to be very stable and opened the possibility of using perovskite solar cells in high altitude conditions. The substitution of the doped transport layers led to devices with higher power conversion efficiencies (more than 20%, among the highest values for vacuum deposited CH3NH3PbI3 solar cells to date) and more robustness, passing stability tests under space conditions. The work developed in this thesis has opened an interesting field for vacuum deposited perovskite solar cells and changed the main fabrication routes in our laboratory