Direct search for light dark matter with quantum detectors

Due to observations of celestial phenomena, it is evident that only about 37% of the matter in our universe is visible. The majority of matter appears to be dark or does not interact with light and has not been directly observed till now. However, the analysis and simulations of observations strongly indicate that the milky way and our solar system is completely populated with dark matter particles. For this reason, the detection and characterization of such dark matter candidates on earths surface is feasible and presents one of the major challenges of modern physics. Most former experiments concentrated on the search for dark matter particles with heavy masses. To detect those particles, the interaction processes between dark matter with e.g. up to some 1000 kg of liquefied noble gas was studied. But those experiments could just exclude different mass ranges and many underlying models of theoretical physics have been disproved by particle accelerator experiments in recent years. Thus, the search for light dark matter particles became more attractive nowadays. The lighter the particles become, the more of them have to exist. As a consequence, light dark matter experiments need a significant smaller amount of sensitive mass of some kg to detect a reasonable amount of interactions or exclude a further parameter space. For this reason, we propose to use highly specified, low noise silicon detectors to search for light dark matter candidates. The interaction of a dark matter particle with the silicon is expected to excite only some electrons, which are weakly bound to the silicon atoms. In order to measure this tiny interaction, those detectors are required to be sensitive to signals as small as a single electron. The development and operation of these very complex devices require extensive studies and preparation. For instance, background events caused by impurities or thermally exited electrons have to be identified and avoided or reduced in upgraded sensors in order to lay the base for further progress in the direct search for dark matter.

Despite the constant progress of science, the exact composition of our universe remains a mystery. By observing large celestial structures such as galaxies or clusters of galaxies and modeling their formation, we are able to conclude on the amount of matter present. However, up to thirty percent of this matter is not visible. To investigate the properties and nature of this dark matter, it is necessary to directly detect it with extremely sensitive experiments. Since the search for heavy dark matter particles has not yielded any clear results in recent decades, current experiments are increasingly focusing on possible dark matter particles with a comparable low mass. This approach has two crucial consequences. Firstly, the particles are no longer heavy enough to generate a measurable signal through interaction with massive atomic nuclei, so that the scattering process on light electrons is now being investigated. Secondly, an increased frequency of such processes is expected, as the density of dark matter is known and lighter particles occur correspondingly more often. Due to both aspects, the use of semiconductor detectors, which are precise enough to measure signals of individual electrons but can only be built up in comparatively small masses, is of increasing interest. RNDR-DEPFET detectors are a promising technology for this purpose. The high accuracy of these detectors is achieved by repeatedly sampling a similar signal, in a way that the noise of the overall measurement is reduced with each additional sampling. The sensors are manufactured in a scientific environment by the Semiconductor Laboratory of the Max Planck Society and assembled to detectors within this project and tested for their suitability for detecting dark matter. After we demonstrated the ability of these detectors to measure single electrons with a detector of several thousand pixels on an experimental setup, we were able to develop an analysis software with which this measurement can be carried out with an accuracy in the range of a 10,000th of a second. This high time resolution significantly reduces the probability of two single electrons being collected within one sampling. Since the noise of a measurement increases with the amount of sensitive sampled mass, but the signal is always constant, the sensitivity of the measurement or the signal-to-noise ratio is improved by using many small pixels in a similar manner. The project enabled us to significantly increase the sensitivity of these experiments for the future search for dark matter.