Abstarct:   This research focuses on the design and fabrication of betavoltaic nuclear batteries. In the design of these batteries, factors such as the radioactive source, semiconductor type and thickness, suitable electrode type, and the arrangement of semiconductor laxyers and the source (geometry) must be considered. Using Fermi's golden rule, the continuous beta spectrum of some widely used sources such as 63Ni, 90Sr, 35S, 14C, 147Pm and 90Sr/90Y were calculated. Also, the effect of using the average beta energy instead of its spectrum on the energy deposited in a betavoltaic nuclear battery and soft tissue was investigated. Ultimately, it was determined that using the average energy of the 63Ni, 90Sr, 35S, 14C and 147Pm beta sources instead of their continuous beta spectrum causes 63.16%, 26.77%, 55.53%, 53.87% and 54.37% relative difference in the amount of energy deposition in the semiconductor and 63.28%, 27.20%, 54.91%, 53.28%, and 54.35% in the amount of energy deposited in soft tissue, respectively. Therefore, using the average energy of a beta source instead of its continuous energy spectrum affects the results obtained from Monte Carlo simulations. The beta spectrum of the 90Sr/90Y source was used in the simulation stages. In order to determine the suitable semiconductor with optimal thickness, the effect of the thickness of aluminum nitride, tin dioxide, zinc oxide, silicon carbide, gallium nitride, boron nitride, and diamond P-N junctions on the short-circuit current, open-circuit voltage, and energy conversion efficiency of the battery was investigated, and the optimal condition was determined for each case. Furthermore, in this study, the effect of three geometric models, planar, double-plane, and cubic on the electrical properties of the studied batteries was investigated. Finally, it was determined that the zinc oxide betavoltaic nuclear battery, while being cheap, is more efficient than other investigated batteries. Therefore, zinc oxide was used as a semiconductor material in the stage of making the laboratory sample of the studied battery. In this research, in order to build a betavoltaic nuclear battery with favorable results, about 40 samples were made. In the construction of the first samples, the experiences of others were used to select the electrode material. However, in the continuation of the work, in order to determine the suitable electrodes experimentally, an experiment was designed which could be used to determine the suitable electrodes in each of the P-Si, N-Si/ZnO, P-Si/ZnO, and N-Si modes at different voltages. Finally, for two of the optimized samples, one of the Schottky diode type and the other of the P-N heterojunction type, simulations were performed and the theoretical and experimental data were compared with each other. In this work, the difference between the theoretical and experimental data was about 25% to 52%, which is a good agreement compared to the works done by others, which have a difference of about 74% to 93%.