Strategies for electron pair reconstruction in CBM

The investigation of the inmedium properties of short lived neutral vector mesons (ρ, ω, φ) via detection of their decay in electron-positron pairs is one of the major issues of the CBM physics program. The strategies to reduce the combinatorial background in electron pair measurements with the CBM detector are discussed here. The simulations were performed for the collision system Au+Au at a beam energy of 25 AGeV. The fi nal state phase space distributions of hadrons and photons were generated using the UrQMD code. To study the most extream case the simulation was done for zero impact parameter. The phase space distribution of electrons and positrons from purely leptonic and semi-leptonic (i.e. Dalitz) decays of ρ, ω, φ were obtained using the PLUTO event generator. The ρ meson mass distribution was generated by including a Breit-Wigner shape around the pole mass, thermal phase space factors, and a factor 1/M to account for vector dominance in the decay into e e. All particles were fi nally propagated through the detector system using the GEANT3 package. The dominant background sources are random combinations of electrons and positrons from π-Dalitz decay and γ conversion. In a central Au + Au collision about 360 π mesons are produced, which decay into eeγ (Γ/Γtot = 0.012) and to 2γ (Γ/Γtot = 0.988). Hitherto existing electron pair spectrometers (DLS [1], HADES [2], CERES [3]) provide electron identifi cation in front of the magnetic fi eld, this permits the rejection of electron-positron pairs with small opening angle, which most probably originate from π-Dalitz decay and γ conversion. At CBM there is no electron ID in front of the Silicon Tracking Station and the magnetic fi eld. Hence, the rejection strategy has to rely to a large extend on the track topology of pairs in the compact Silicon Tracking Station. To increase the acceptance for low momentum particles the magnetic fi eld was reduced to Bmax = 0.33 T and shifted 20 cm downstream. A single 25 μm gold target was assumed. The fi rst STS station was not used in the analysis due to its limited acceptance for tracks with small laboratory polar angles. A track was considered fully reconstructed if tracking station 2 to 7 were traversed and the RICH and the 3 TRD detectors were hit. Perfect particle identifi cation was assumed for fully reconstructed tracks. Tracks were considered partially reconstructed if a minimum of 4 STS stations were traversed. For these tracks only the charge sign information and the Monte-Carlo momentum were used. The Monte-Carlo momenta where smeared assuming 1 % momentum resolution independent of particle momentum. The strategy of background rejection comprises three steps. First step is to identify and reject true pairs originating from conversion, second step is to remove single tracks where the true partner was not fully reconstructed, third step is to assign pairs with a characteristic pattern to π-Dalitz pairs. All cuts are summarized in Table 1.