In this work, the crystal structure of epitaxially grown semiconductor nanowires has been analysed using electron microscopy and to some extent X-ray diffractometry. The goal of the EU project which this work was a part of was to build multi-junction solar cells with nanowires as the main building blocks. Higher efficiency solar cells today tend to need rare elements and be very costly. Using nanowires should reduce the amount of material used and also reduce the cost. The geometry of nanowires allow epitaxially grown materials with a greater level of lattice mismatch, without introducing large amounts of defects due to strain. This increases the selection of possible materials to build the photo diodes. However, the strain will not be lower in the region of the junction and can change material properties, such as the bandgap. The substrate the wires are grown on was used as back contact of the solar cells. The interface between this layer and each nanowire is of course also a very sensitive area of the device. In order to examine these two interfaces a sample preparation method was further developed and adapted these types of samples. With this sample preparation method it was also possible to examine the very same nanowire sample using both X-ray diffraction and transmission electron microscopy. The examined structures were probed for their relative tilt to the substrate. Nanobeam electron diffraction was used in order to probe the local crystal structure of a nanowire, especially across the junction. The strain across the junction measured by the difference in crystal lattice distance extends over a larger volume than the gradual change in composition found for the same sample. Measurements of other nanowires showed a continuously changing tilt along the wires and a local distortion of the crystal structure at the junction. This thesis also comments on some unusual properties and _ndings of the examined nanowires: Some nanowires sported a droplet-like protrusion of the catalyst gold particle reaching into the solid center of the nanowire. This feature can be discussed in terms of nanowire growth process and might influence device performance. Several samples of nanowires grown in the [1 1 1]-direction were found to host unusual twin-defects in the (1 1 1)-plane. The advent of [1 1 1] rotational twins are common during the growth of nanowires in the zincblende structure, however, twins in the diagonal [1 1 1] direction are not. Energy dispersive X-ray spectroscopy showed that these defects influence the local composition of the wire. A thick contamination layer on top of the InP substrate of a sample with GaP nanowires was discovered and reported. This clearly showed the necessity of proper sample analysis post growth and device preparation. The results of this work hopefully contributes to the understanding of the properties and growth mechanics of the nanowires. The methods developed should help improve the nanowire and device manufacturing through better characterization steps and ultimately lead to more ecient solar cells.