BiFeO₃ is one of the most promising functional materials on perovskite basis. It is well known for its multiferroic properties, which it even maintains at temperatures far above room temperature, in the form of coupling between its ferroelectric polarization and antiferromagnetic ordering. With the successful demonstration of a direct switching of the antiferromagnetic domains via an electric field, research interest in BiFeO₃ has enormously increased in the past years because it offers intriguing application prospects. Besides the magnetoelectric properties, BiFeO₃ is also piezoelectric, shows a photovoltaic effect, and is highly birefringent. Additionally, it does not use a toxic and environmentally problematic component like Pb-based perovskites. Possible applications include smaller, faster, and more energy-efficient digital storage since changing magnetic orientation via the application of an electric field and the magnetoelectric coupling needs much less energy than to have electric current create a magnetic field to rewrite magnetic domains. With an increasing demand for digital data storage, this could be valuable to reduce environmental impact by reduced energy consumption. Other interesting applications include spin valves, spintronic devices, sensors, and optoelectronic devices. Doping and strain engineering of BiFeO₃ can be used to tune and change the material properties for applications. For example, the antiferromagnetic behavior can be changed to a ferromagnetic one. To understand effects in the material at an atomic scale, advanced transmission electron microscopy is one of the best methods since it delivers not just structural but also elemental and chemical information. This is combined in this work with density functional theory calculations, which showed substantial synergy effects. Scanning electron microscopy, X-Ray diffraction, X-ray photoelectron spectroscopy, and atomic force microscopy were used for additional characterization of the BiFeO₃ thin films. This dissertation focuses on thin films with Ca as a dopant or co-dopant. Specifically, it contains 3 studies focused on different aspects: • A new finding of segregation of Ca dopant toward the in-plane compressively strained interface between the Ca and Mn co-doped bismuth ferrite film and a strontium titanate substrate. The Ca segregation triggers atomic and electronic structure changes at the interface. The strain at the interface is reduced according to the Ca concentration gradient. Variations in the interplanar spacing and oxygen vacancies are introduced. The observed segregation behavior is confirmed with density functional theory calculations. • A study on the interaction of oxygen vacancies and ferroelectric domain walls on the case of a Ca doped BiFeO₃ film. The results revealed that the oxygen vacancies agglomerate in plates which simultaneously represent negatively charged domain walls in a tail-to-tail configuration. The plates appear without any periodicity between each other in out-of-plane as well as in-plane direction. Within the plates, the oxygen vacancies form 1D channels in pseudocubic [010] direction with one site containing lots of vacancies and the adjacent ones on both sites few. Interestingly, no variety between the characteristics of out-of-plane and in-plane plates could be found. Charged defects such as oxygen vacancies are known for their application-decisive pinning effect on domain walls, which on the one hand leads to fatigue mechanisms but on the other hand also counteracts retention failure. Charged defects also strongly influence domain wall conductivity. • The discovery of a significantly higher Ca solubility in BiFeO₃ than in the secondary Bi₂O₃ phase. The solubility behavior is confirmed and expanded on with density functional theory calculations. Bi₂O₃ can be used to evoke the super-tetragonal phase in BiFeO