Laser Induced Breakdown Spectroscopy (LIBS) is an analytical technique used to classify and potentially quantify elements in complex hosts (or matrices). Vacuum Ultraviolet Laser Induced Breakdown Spectroscopy (VUV LIBS) offers measurable improvements over traditional visible region LIBS, due to the presence of strong resonance and well spectrally isolated transitions at these shorter wavelengths. This extends the ability of LIBS to discriminate between emission lines from different elements, particularly light elements such as carbon, nitrogen, oxygen, sulphur, etc. In addition, as with visible LIBS, the precision obtainable can be limited by the continuum emission at the early stage of the plasma lifetime and greater performance can be achieved by moving to a time resolved detection system, reducing the contribution from the continuum. Hence, a (single channel) time resolved VUV LIBS system has been designed by replacing a back illuminated CCD with a combination of a plastic scintillator and a photomultiplier tube on an existing LIBS system. In this research, the performance of time integrated and time resolved VUV LIBS for the determination of the carbon content in steel samples is compared. The results demonstrated an improved limit of detection (LOD) for time resolved VUV LIBS over time integrated VUV LIBS for traditional point plasmas. In addition, VUV emission from plasma plumes created at a line (laser) focus was also studied for the same steel samples and the limit of detection was optimised and compared the limit for conventional point plasmas. The results demonstrate that the line focused plasma source in multichannel time integrated space resolved (TISR) mode and the single channel time resolved VUV mode yield comparable LoD values of 50 and 56 ppm respectively. It also showed that both were significantly better than TISR VUV LIBS with point plasma VUV sources which exhibited a poorer LoD of 316 ppm. In addition, two ultraviolet-visible (UV-Vis) LIBS studies were carried out in University of Padova. One of these focused on exploring the potential of conventional LIBS for depth profiling of thin films. The other focused on the classification and sorting of scrap metal samples using multivariate analysis, in particular using principal component analysis (PCA).