Event Selection
The first step of the procedure is the event selection. For the aim of this work minimum bias events will be analyzed. In the ALICE context, “Minimum Bias” event means an event with at least an hit revealed in the SPD or in the V0 detector, i.e. a charged particle in the 8 pseudo-rapidity units covered by these detectors. In order to have an efficient and user-transparent event selection the AliRoot framework provides a class (called AliPhysicsSelection) that automatically selects the requested events just before the starting of the real physical analysis, avoiding the inclusion, for instance, of non physical data taking runs (i.e. calibration runs) or events without the interaction trigger, i.e. trigger on bunch crossings, or events flagged as beam-gas by either V0A or V0C detectors. This class can be applied both to the real and to the simulated data (even if the latter are by definition good physics events) in order to treat both of them with homogeneity. This aspect is very important in particular when we will use the number of events for the normalizations of the data. Moreover, another event selection looking at the reconstructed primary vertex is performed. Only events with a good primary vertex (i.e. a vertex reconstructed with the information from the complete tracks reconstructed with “ITS + TPC” or at least a vertex identified with the data from the SPD detector) will be analyzed.
Σ(1385) Candidates Selection
The Σ(1385) has two main decay channels and they are strong channels. This has two consequences: first of all it will not be possible to separate the Σ∗ decay vertex from the primary interaction vertex, and secondly it will be difficult to separate the resonance peak from the background because it will be quite broad (Γ > 35 MeV). That means that no strong topological selection will be possible and the Σ∗ signal can be extracted only through an invariant mass analysis of Λ and π decay particles, i.e. using only one of the possible Σ∗ decay channels since it has the highest branching ratio and since, in ALICE, we are not able to identify with an acceptable efficiency the Σ’s (needed for the second Σ∗ decay channel). The first step is therefore to identify the Λ’s and π’s candidates. The Λ’s, identified by the ALICE reconstruction program are combined with all the remaining pions to get the Σ∗. During this phase some selection criteria were applied to select them. This criteria are described in the next paragraph.
Λ’s and π’s selection criteria
The ALICE automatic general reconstruction analysis provides the identification of all the Λ but for the purpose of this analysis, it is mandatory to select, among them, those which pass a specific set of quality selection criteria. These criteria are choosen as a compromize between the requirement to have both a sample as clean as possible and a good Signal over Background ratio. The same philosophy is applied to the pions selection criteria. Since the Σ∗ decay vertex is not separable from the primary vertex, the Λ’s coming from the Σ∗’s decay can be considered as primaries, and we try to select (and so reject most of the secondaries) them applying a standard ALICE pt dependent cut in the Distance of Closest Approach (DCA) between the Λ’s and the primary vertex. This cut takes into account the resolution in the impact parameter estimation for different particle types. Then a check is performed in order to exclude the so called “Like-Sign” Λ’s, i.e. misreconstructed Λ’s whose daughter assigned particles have the same charge sign. Finally in order to try to improve the signal over noise ratio and the signal purity, a set of kinematical and quality cuts were applied and studied. The quality cuts, applied to Λ’s and π’s involves the following quantities:
- for the Λ’s:
- TPC refitting of both the positive and negative daughter track
- number of TPC clusters per track for the positive and negative Λ’s daughter
- TPC χ2divided by the number of TPC clusters per track both for the positive and negative Λ’s daughter
- DCA between Λ and primary vertex
- Λ daughter’s cosine of the pointing angle
- Λ mass window
- for the π’s
- TPC refitting of both the positive and negative daughter track
- number of TPC clusters per track for the positive and negative Λ’s daughter
- TPC χ2divided by the number of TPC clusters per track both for the positive and negative Λ’s daughter
- DCA between π and primary vertex
- for both the Λ’s and the π’s:
The residual Λ’s and π’s are then paired to obtain the invariant mass histogram. In order to evaluate the background to techniques has been implemented, the "Side-Bands Technique"and the "Event Mixing Technique".
Side Bands Technique
The “Side-Bands” technique consists in a fit of the invariant mass distribution after excluding the region where the signal is supposed to appear (“exclusion region”). The “exclusion region” was chosen as centered at the Σ∗ mass published in the PDG booklet (1385 MeV) and with a width equals to 3×Σ∗ width (35 MeV). The chosen fit function is a polynomial function containing both Taylor and Laurent terms in order to be able to follow the distribution profile both at low and high invariant masses. The fitting function is then extrapolated also inside the “exclusion region”.
Event Mixing Technique
The “Event Mixing” technique consists in a mix of the Λ’s from one event and the π’s from another event in order to obtain a fully uncorrelated background. This technique foresee to be applied with particular care in the events selection. Infact in order to proper reproduce the required background it is mandatory to mix similar events in terms at least of particle multiplicity and vertex positions. For this reason a proper selection of the events to be mixed is needed before the applying of the procedure.
