[en] Evidence for physics beyond the well-established standard model of particle physics is found in the sector of neutrino physics, in particular in neutrino oscillations, and in experimental hints requiring the presence of Dark Matter. Neutrino oscillations demand the neutrinos to be massive and at least four additional parameters, three mixing angles and one phase, are introduced. A non-vanishing value for the third mixing angle, θ₁₃, has only recently been found, amongst others by the reactor antineutrino disappearance experiment Double Chooz. This experiment detects anti ν_e's by means of the Inverse Beta Decay (IBD), which has a clear signature that can very effectively be discriminated from most of the background. However, some background still survives the selection cuts applied to the data, partly induced by radioactivity. In order to determine the amount of radioimpurities in the detector, germanium spectroscopy measurements and neutron activation analyses have been carried out for various parts of the Double Chooz far detector. A dedicated Monte-Carlo simulation was performed to obtain the singles event rate induced by the identified radioimpurities in the fiducial volume of Double Chooz. In the present thesis, parts from the outer detector systems, as well as components of the inner detector liquids were measured. In sum, a singles rate of less than 0.35 Hz above the antineutrino detection threshold of 0.7 MeV has been found. This is by far below the design goal of Double Chooz of ∝ 20 Hz. The analysis of bismuth-polonium (BiPo) coincidences in the first Double Chooz data allows to directly determine the number of decays from the U- and the Th-decay chain in the active detector parts. Assuming radioactive equilibrium, concentrations of (1.71±0.08).10^-14(g)/(g) for uranium and (8.16±0.49).10^-14(g)/(g) for thorium have been found, which are also well below the design goal of Double Chooz (2.10^-13(g)/(g)). Both gamma spectroscopy measurements and the BiPo analysis show the high level of radiopurity reached in Double Chooz. In addition, with the BiPo analysis the α-quenching factors for the Target and the GammaCatcher liquids have been determined, respectively, to 9.94±0.04 and 13.69±0.02 at 7.7 MeV, and 9.05±0.01 and 14.3±0.1 at 8.8 MeV. The former values show a good agreement with the values obtained in a dedicated laboratory measurement. The time stability of the peak position of the ²¹⁴Po α-peak could be proven, too, showing a stable detector performance at low visible energies. The direct search for Dark Matter can, amongst others, be performed with liquid rare gas detectors, which make use of the scintillation light. However, a good background discrimination is needed. Studies on the wavelength- and time-resolved scintillation properties of liquid argon have therefore been carried out with high resolution and best statistics. The results obtained for different ion beams show that particle discrimination is not feasible in any realistic experiment by means of the wavelength-resolved scintillation light only, but the time structure of the emitted light provides a good handle to distinguish between different incident particles. For heavy ions (sulfur) a ratio of the fast to the slow scintillation component of (1.6 ± 0.6) is found, while lighter particles (protons) exhibit a ratio of (0.25 ± 0.05). The outcome of the present studies shows that this ratio can also be used in wavelength-integrating measurements which have a comparable detection efficiency for wavelengths below and above ∝170 nm. The present results demonstrate that for a number of 90 detected photons the singlet-to-triplet distributions obtained for sulfur ions and protons as exciting particles cease to overlap. In a Dark Matter experiment, if all photons produced can be detected, this corresponds to a discrimination threshold of only 2.25 keV.