Measurements of the electron-positron continuum in ALICE

The status of the analysis of electron-positron pairs measured by ALICE in pp collisions at $\sqrt{s} = 7$ TeV and central Pb-Pb collisions at $\sqrt{s_\mathrm{NN}}=2.76$ TeV is presented. Key questions and the main challenges of the analysis are discussed on the basis of first raw invariant mass spectra for both collision systems.


Analysis of the pp Data
For the pp data, 300M minimum-bias events have been available for analysis. Figure 1 shows the TPC signal distribution for the tracks selected for the analysis. Besides passing basic quality criteria, a TPC signal for each track within −1.5 < σ e < 3 of the expected mean electron signal for a given momentum was required.
For tracks with p T > 0.4 GeV/c, the tracks are additionally constrained, by requiring a TOF signal within 3σ of the expected electron signal at the corresponding momentum. As shown in the right panel of Figure 1, this leads to a negligible contamination from other particles, most of all charged pions, over the full momentum range used for this analysis.
To determine he invariant mass distribution of the dielectron signal all opposite-sign pair combinations of electron and positron candidates are determined in each event. The combinatorial background a e-mail: cbaumann@ikf.uni-frankfurt.de   TeV. The like-sign distribution has been corrected to account for the difference in acceptance between like-sign and unlike-sign pairs. is calculated from same-event like-sign pairs: mixed ) accounts for the acceptance difference between like-sign and unlike-sign pairs, which is modeled using a mixed-events technique ( Figure 2).
To reduce the combinatorial background, pairs with opening angles θ < 0.035 rad are removed from the sample, as such pairs are predominantly originating from photon conversions. The signal-tobackground ratio for this measurement varies between 10 −1 − 1 as shown in Figure 3.    This value is comparable to results obtained by ALICE in the dimuon channel at forward rapidities. They are complementary to the dielectron measurements, details are given in [10].

Resonance Workshop at UT Austin
Subtracting the like-sign distribution from the unlike-sign yield gives the raw electron-positron yield shown in Figure 4. Within the statistical uncertainties, the yield is significant up to the J/Ψ mass, the peaks of the ω and φ mesons are clearly distinguishable. The p T dependence of the resonance yields will be compared in the future to the results obtained in the µ and K decay channels.
Currently, the acceptance and efficiency corrections are being determined and a hadronic cocktail calculation that also accounts for the contribution from open-charm decays is under development. With these, it will be possible to compare the spectrum to theoretical expectations assuming vacuum decay of the hadrons and to measure or set limits on the charm cross-section.

Analysis of the Pb-Pb Data
In the 2011 beamtime ALICE has recorded enhanced samples of central and semi-central events in addition to minimum-bias data for Pb-Pb collisions at √ s NN = 2.76 TeV. In total, 8M minimum bias, 27M central and 32M semi-central events are available for analysis. While the general analysis method is identical to the approach used for pp collisions, stricter cuts for the track selection have been chosen to account for the increase of the combinatorial background due to the higher event multiplicities: a signal within 3σ of the expected electron signal in both TPC and TOF for p T > 0.4 GeV/c is required for electron candidates.
A main focus of this analysis is the search for a thermal contribution to the photon spectrum, measuring internal conversions in the electron-positron channel. The kinematical region relevant for such pairs is characterized by p T,ee > 1.0 GeV/c and m ee < 0.5 GeV/c 2 . The virtual photon yield from internal conversions can be gauged by fitting the measured distribution with a function of the form f total (m ee ) = r · f hadronic (m ee ) + (1 − r) f photon (m ee ), where f hadronic describes the shape of the hadronic contributions to the spectrum and f photon the shape of the photonic contributions, where both are independently normalized. r can then be interpreted as the fraction of photons coming from internal conversions in the inclusive photon spectrum. Such a fit is typically made at masses around 200-300 MeV/c 2 , where the spectrum is no longer dominated by the contribution from π 0 -Dalitz decays. The raw yield of such pairs was extracted in several p T,ee intervals following the method described above. Figure 5 shows the unlike-sign and like-sign distributions for two p T intervals as an example. The signal-to-background ratio shows, as can be expected, a dependence on the p T region. A typical value for m ee ≈ 300 MeV/c 2 is S /B = 10 −2 ( Figure 6).
The background subtracted dielectron spectrum is shown in Figure 7. At low invariant masses electron-positron pair from conversions of real photons contribute significantly to the spectrum. Different approaches to identify and reject these with good efficiency are currently under evaluation.
To allow quantitative conclusions on the dielectron production, these raw spectra need to be corrected for acceptance and efficiency. Furthermore, a hadronic cocktail calculation including contributions from open charm and bottom decays is being developed.

Summary
We have presented first results on the measurement of electron-positron pairs in both pp and Pb-Pb collisons measured with the ALICE experiment. Making use of its unique PID capabilities at low Resonance Workshop at UT Austin   momenta, a raw electron-positron pair yield has been extracted in both collision systems. Using the pp results as a base-line, the search for a possible thermal contribution to the electron-positron spectrum at low invariant masses in the Pb-Pb data is a main focus of the ongoing analysis.