Pulse shape analysis for the GERDA experiment to set a new limit on the half-life of 0νββ decay of ⁷⁶Ge

The GERDA experiment searches for neutrinoless double beta (0νββ) decay of ⁷⁶Ge using high purity germanium (HPGe) detectors operated in liquid argon (LAr). The aim is to explore half-lives of the order of 10²⁶ yr. Therefore, GERDA relies on improved active background reduction techniques such as pulse shape discrimination (PSD) in which the time structure of the germanium signals is analyzed to discriminate signal- from background-like events. Two types of HPGe detectors are operated: semi-coaxial detectors previously used in the Heidelberg-Moscow and IGEX experiments and new Broad Energy Germanium (BEGe) detectors which feature an improved energy resolution and enhanced PSD. In Phase I of the experiment, five enriched BEGe detectors were used for the first time in the search for 0νββ decay. A PSD based on a single parameter, the ratio of the maximum current amplitude over the energy A/E is applied. 83% of the background events in a 232 keV region around Q_ββ are rejected with a high signal efficiency of (92.1 ± 1.9) %. The achieved background index (BI) is (5.4^+4.1_-3.4) . 10^-3 (counts)/(keV.kg.yr). This is an improvement by a factor of 10 compared to previous germanium based 0νββ experiments. Phase II of the experiment includes a major upgrade: for further background rejection, the LAr cryostat is instrumented to detect argon scintillation light. Additional 25 BEGe detectors are installed. After PSD and LAr veto a BI of (0.7^+1.3_-0.5) . 10^-3 (counts)/(keV.kg.yr) is achieved. This is the best BI achieved in 0νββ experiments so far. A frequentist statistical analysis is performed on the combined data collected in GERDA Phase I and the first Phase II release. A new limit on the half-life of 0νββ decay of ⁷⁶Ge is set to T^0ν_1/2>5.3.10²⁵ yr at 90% C.L., with a median sensitivity of T^0ν_1/2>4.0.10²⁵ yr at 90% C.L.