Nuclear power plants are expected to play an important role in the worldwide electricity production in the coming decades, since they provide an economically attractive, reliable and low-carbon source of electricity with plenty of resources available for at least the coming hundreds of years. However, the design of nuclear reactors can be improved significantly in terms of safety, by designing reactors with fully passive safety systems, and sustainability, by making more efficient use of natural resources in so-called breeder reactors, and by reducing the radiotoxicity and storage time of the waste produced by the reactor. The application of a thorium breeder fuel cycle within a Pebble Bed High Temperature Reactor could provide such improvements. The core of a Pebble Bed Reactor (PBR) consists of a helium cooled random stacking of graphite spheres, which contain many small coated fuel particles retaining radioactive fission products for temperatures below 1600 degrees Celsius. In case of a Depressurized Loss of Forced Cooling (DLOFC), the most serious accident that can occur in a PBR, decay heat can be removed from the core by passive means without the maximum fuel temperature exceeding 1600 degrees Celsius. This is due to its relatively small core radius and low power density. Thorium fuel cycles offer several interesting advantages. Thorium is three to four times more abundant in the earth's crust than uranium. In a closed breeder cycle, thorium can significantly reduce the radiotoxicity and storage time of nuclear waste, and all natural resources mined, i.e. Th-232 or U-235 and U-238, can be used for electricity production instead of roughly 1% of the natural uranium resources consumed by current light water reactors. Furthermore, the Th/U-233 fuel cycle has favourable nuclear properties, i.e. the neutron reproduction factor of U-233 and the relatively high thermal neutron capture cross section of Th-232, for use in thermal breeder reactors, like a PBR. This thesis work investigates whether it is possible to achieve a thorium breeder fuel cycle in a pebble bed reactor within a practical operating regime and without compromising passive safety. Furthermore, the reactor should also achieve a net U-233 production within a limited time frame. Neutronics studies of the fuel design show that the conversion of thorium into U-233 can be maximized for thorium pebbles with a large heavy metal loading of 30 g, a conservative maximum from a fuel fabrication perspective, and a standard fuel kernel radius of 0.025 cm, preferably irradiated at low specific power to improve k-inf. Reprocessing or adding moderator pebbles is required to raise the k-inf of these pebbles above unity. Cylindrical cores, consisting of a central driver zone surrounded by a breeder zone and reflector regions, were investigated during equilibrium core design studies. 30 g thorium pebbles are inserted in the breeder zone, while a higher carbon to heavy metal ratio is used in the driver zone. Results obtained by an equilibrium core calculation scheme show that the uranium content from both the discharged driver and uranium pebbles has to be reprocessed after their final passage to achieve breeding. As a next step, coupled neutronic and thermal-hydraulic design studies using a coupled DALTON/THERMIX code scheme were performed to investigate whether it is possible to combine passive safety and breeding, within a practical operating regime, inside a thorium PBR. Additionally, the equilibrium core calculation scheme was extended to include the spectral influence of surrounding zones (driver, breeder and reflector) into the fuel depletion calculations. High conversion ratios (CR > 0.96) and passive safety can be combined in a thorium PBR within a practical operating regime. Increasing the U-233 content of the fresh driver pebbles (18 w%), breeding (CR=1.0135) can already be achieved for a 220 cm core and 80 cm driver zone radius, but the temperature feedback is too weak to compensate the reactivity insertion due to decay of Xe-135 during a DLOFC without scram. With a lower U-233 content per driver pebble (10 w%), breeding (CR=1.0036) and passive safety can be combined for a 300 cm core and 100 cm driver zone radius operating at a power of 100 MWth, but this requires a doubling of the pebble handling speed and a high fuel pebble reprocessing rate, which may present a challenge from an engineering perspective. The maximum fuel temperature during a DLOFC without scram was 1481 degrees Celsius, still quite a bit below the TRISO failure temperature. The maximum reactivity insertion due to water ingress is also limited (+1497 pcm). For this 100 MWth passively safe thorium breeder PBR design, the total control rod worth is far insufficient in the radial reflector to achieve cold reactor shutdown, requiring a worth of over 15,000 pcm. 3D heterogeneous KENO calculations show that 20 control rods, positioned just outside the driver zone, can provide sufficient reactivity worth. Furthermore, the insertion of a neutron absorber gas, like BF3, can be considered as an additional emergency shutdown system. Finally, the running-in phase of the passively safe thorium breeder PBR was analysed using a simplified core depletion model, which solves the depletion equations with an axial fuel movement term for the most relevant actinides and a lumped fission product pair. Enriched uranium (U-235/U-238) is used during the first 1300 days of reactor operation and U-233/Th fuel is used afterwards. By clever adjustment of the enrichment or U-233 weight fraction of the feed driver fuel over time, the thorium PBR can start to breed U-233 within 7 years after starting reactor operation. Furthermore, a basic safety analysis for the various stages of the running-in phase indicates that passive safety is also ensured during any stage of the running-in phase. This thesis work demonstrates that it is possible to achieve breeding within a passively safe thorium pebble bed reactor after a limited time (< 7 years), though the margins in terms of breeding, passive safety and practical constraints are rather small. Therefore, the passively safe thorium breeder PBR design presented requires a doubling of the fuel pebble handling speed, compared to the HTR-PM, and a high fuel pebble reprocessing rate, while also some other technical difficulties still have to be addressed, such as the large pumping power requirement and the modelling of the conus region and defueling chutes. However, these issues are no fundamental limitation for the application of a thorium breeder cycle within a passively safe PBR, which could offer a huge improvement in terms of safety and sustainability of nuclear power for future generations.