4. Recent results of MPD feasibility study
During 2010-14 a comprehensive program of feasibility studies for the MPD detector
was carried out. The results of these studies have been presented at many international
conferences and published. Below, we summarize several selected results of this activity.
4.1 Study of electron-positron pairs in A+A collisions at NICA/MPD
A detailed description of the analysis procedure and results can be found in Ref.[1].
Experimental study of dileptons in heavy-ion collisions is a challenging task, because of
a huge combinatorial background of uncorrelated lepton pairs from the Dalitz decays of
0 and conversions. The main goal of this study was to investigateMPD detector
performance for low-mass dilepton measurements in terms of hadron suppression factor,
signal-to-background ratio and invariant mass resolution.We used in this study central (03 fm) Au-Au collisions at the center-of-mass energy of 7A GeV from the UrQMD
generator in combination with the cocktail of vector mesons from the Pluto event
generator.The used detector set-up includes the Time Projection Chamber (TPC), TimeOf-Flight system (TOF) and Electromagnetic Calorimeter (EMC) covering the
pseudorapidity range ||<1.2. The particles were transported through the detector bythe
Geant 3.21 code, the track reconstruction was based on the Kalman filtering technique.
All the reconstructed in the TPC tracks were then extrapolated to the TOF detector and
matched with TOF hits. The efficiency of track matching with TOFwas found to be of
about 90% at pT>0.4 GeV/c, and the transverse momentum resolution appeared to be
better than 3% for tracks with pT below 1 GeV/c. Electron identification was achieved by
using combined information about the specific energy loss dE/dx from TPC, time-offlight from TOF and E/p from EMC. The achieved overall hadron rejection factorwas
about 3200.The conversion pairs were rejected by a cut,making use of the fact that the
magnetic field is orthogonal to the dilepton’smomenta plane. This selection was
complemented by a cut on the radial position of the production point. In addition, in order
to suppress electrons (positrons) from the conversion further we applied a transverse
momentum cut: 0.2<pT<2 GeV/c.
Fig.1.(Left) Background-subtracted invariant mass distributions of electron-positron pairs
from central Au+Au collisions at MPD.(Right) Signal-to-Background (S/B) ratio from
heavy-ion experiments as a function of total charged multiplicity.
In Fig.1 (left panel) a background-subtracted invariant mass distributions of electronpositron pairs in the pseudorapidity window ||<1.0 is plotted. A rough estimate of
invariant mass resolution of the MPD setup was made by fitting the dileptons spectra at
the poles for vector mesons: RMS of 14 and 17 MeV was obtained for the omega- and
phi-meson pole, respectively. The overall (integrated over the invariant mass window of
0.2-1.5 GeV/c2) signal-to-background ratio was found to be close to 10%.The obtained
results for the signal-to-background ratio are shown in Fig.1 (right panel) along with the
published data from other experiments. The expected parameters of the MPD setup are
among the best over the world.
4.2 Evaluation of the MPD detector capabilities for study of the strangeness
production at NICA
A detailed description of the analysis procedure and main results can be found in Ref. [2],
where the MPD detector performance for measurements of K0-mesons, (anti) and
hypertritons are presented. The analysis wasbased on the detectors covering the midrapidity region (|| < 1.3): the main tracker Time Projection Chamber (TPC) and barrel
Time-Of-Flight system (TOF). The overall detector material budget is dominated by the
contribution from the TPC inner and outer cages (made of composite materials kevlar
andtedlar) resulting in the total material budget not exceeding 10% of the radiation length
in the phase-space region of interest. The event samples used for the present study were
produced with the UrQMD and DCM-QGSM (Dubna Cascade Model - Quark-Gluon
String Model) generators, the latter one was used to predict light (hyper)nuclei in central
Au+Au collisions. The volume of the event samples (0-3.0 fm central) at 5A and 9A GeV
ranged from 104 to 5.105 corresponding to about 30 seconds to 30 minutes of data taking
time with the NICA design luminosity.Produced by the event generators particles have
been transported through the detector using the GEANT3 transport package (describing
particle decays, secondary interactions, etc.). The decay properties of hypernuclei (modes
and branching ratios) were taken from experimental data. The track reconstruction
method is based on the Kalman filtering technique and the energy losses were accounted
for in the track fitting procedure.All the TPC reconstructed tracks where then
extrapolated to the TOFdetector and matched to the TOF hits. Particles within the
pseudorapidity range |<1.1 were identified using the combination of the time-of-flight
information from the TOF detector and the dE/dx signal from TPC. Fig.3 (left panel)
shows a typical dE/dx versus mass-squared distribution for tracks with momentum p=1.5
GeV/c. Selected hadron and lightnuclei candidates fall within the 3 ellipses around the
nominal positionfor a given particle type. In addition, the probability for a given particle
to belong toeach of the species can be calculated knowing the widths of the
corresponding distributions (along the dE/dx and M2 axes) and the difference from the
predictedposition for the specie. It wasfound that by requiring this probability to be
greater than 0.75 one can get theefficiency and contamination distributions shown on the
right panel of Fig. 3.
Fig. 3. Left panel: a typical dE/dx versus mass-squared distribution for tracks with p=1.5
GeV/c from central Au+Au (LAQGSM generator). Right panel: PID efficiency and
contamination of wrongly identified specie for positively charged particles.
The PID efficiency is defined as the ratio of those particles that are correctlytagged to
the total number of generated particles. The contamination is determined as the number
of incorrectly tagged particles divided by the number of correctly tagged particles. As
seen from Fig. 3, the overall PID efficiency for p, d and 3He is close to 100%, while due
to a partial overlap of the distributions for pionsand kaons the efficiency of  drops down
to ~0.8 at p=2.5 GeV/c. The contamination of wrongly identified pions(basically from,
e, and K) does not exceed 10%. For other species the observed contribution from the misidentified particles is negligible.
Fig. 4. Invariant mass distributions for p and  (left panel) and  pairs (right panel).
In Fig. 4 the MPD detector performance for reconstruction of L and K0 are demonstrated.
The overall reconstruction efficiency and significance are comparable to those from the
STAR and ALICE experiments. Hypertritons were reconstructed using their decay modes
into charged tracks, the event topology is shown in Fig. 5 (left panel). The procedure of
reconstruction and applied cuts are described in [2]. Fig. 5 (right panel) shows invariant
mass distribution of 3He and p- candidates, the overall efficiency and significance are
also presented.
Fig. 5. (Left) The decay topology of a hypertrion. (Right) Invariant mass distribution of
3
He and  candidates.
Based on the results of this study, a rough estimates of our expectations for the yields of
(anti)andH3 ate tabulated in Table 1.
Table 1 Expected particle yields at MPD for 10 weeks of NICA running time
4.3 Study of the MPD detector performance for measurement of
(multi)strange hyperon production in central Au+Au collisions.
The goal of this study was to evaluate the performance of the MPD detector for
reconstruction of cascade and omega (anti)hyperons and the details of the analysis
procedure can be found elsewhere [3].The event samples used for the present study were
produced with the UrQMD generator at the center of mass energy of 9A GeV. Particle
transport procedure as well as the reconstruction and particle identification algorithms
was the same as described in the previous section. We reconstruct (multi)strange
hyperons by combining charged tracksreconstructed in the TPC, first to select a V0candidate (a characteristic topology of the two opposite charge daughter tracks) and then
to match it with one of the secondary pion or kaon candidate. To guarantee that track
combinations are associated with real decays we applied several selection criteria. To
ensure that the charged tracks are secondary ones, distinct cuts were applied on the
minimum value of the impact parameters to the primary vertex. In addition, a pair of
tracks was rejected if the distance of closest approach (DCA) in space between the two
opposite charged tracks was larger than a given value. Once the secondary vertex
position was defined, only those falling within a fiducial region starting from a given
distance from the main vertex were kept. Finally, the invariant mass was calculated under
the proper hypothesis.The exact values of selection cuts were found by performing a
multidimensionalscan over the whole set of selection criteria with a requirement to
maximize the invariant mass peak significance. The combinatorial background was
derived using the “event-mixing” method, i.e. we combined each reconstructed hyperon candidate from an event with all kaon candidates from other 100 events.
Fig. 3. Reconstructed invariant mass of  and - candidates (left panel) as well as of
and K- candidates (right panel)
In Figure 3 are plotted invariant mass distributions for the reconstructed  and -, Kcandidates. The estimated efficiencies, Signal-to-Background (S/B) ratio and significance
are also shown. Based on the results of this study and model predictions, we have
estimated the expectedyields of particle species under interest for 10 weeks of data taking
(see Table 2).
Table 2. Expected particle yields in central Au+Au collisions for 10 weeks of running
time at NICA/MPD
4.4 Analysis of particle direct and elliptic flow in Au+Au collisions with the
MPD detector
The analysis procedure (more details about it can be found in [4] )was carried out using
the MPDroot software. A total of 3 .105 Au+Au collisions with impact parameter in the
range of 0-9 fm at center-of-mass energy of 11A GeV were generated with the UrQMD
model. The produced particles were transported and reconstructed through the MPD
reconstruction chain using only TPC and TOF detectors. A track selection criteria of
minimum 10 TPC points was required. For this study the Bayesian approach based on
energy loss dE/dx from TPC and TOF information was used: the overall PID efficiency
above 90% was achieved for protons and pions. Reconstruction of -hyperons was
performed using the decay mode to proton and pion. The secondary vertex reconstruction
utilizes a Kalman filtering approach and a set of topological cuts was applied to the
daughter particle candidates (see section 4.2 for details).For the flow analysis the wellestablished event-plane method was used. The event-plane angle was reconstructed by all
identified charged TPC tracks using two sub-events in negative and positive rapidity:
where the sum is on every selected track in the sub-event A or B, wiare weights, n denotes
the order of harmonic,  is the track azimuth angle and is the reconstructed event-plane
angle.
Following the determination of the event-plane angle and the corresponding resolution,
particles are correlated to the event-plane, however, they need to be corrected by the
event-plane resolution to obtain the final results.The  elliptic flow is evaluated in the
invariant mass peak region and in the sidebands region for each pT bin, and the
background contribution is removed from the signal by:
where2S is the "pure"-hyperon flow signal, v2S+B is the flow signal as measured in the
mass peak region and 2B is the flow signal contribution by the background.
Fig. 5 depicts the corrected differential elliptic flow for (anti)protons, charged pions, and
-hyperons versus transverse momentum of the particle.
Fig. 5 Elliptic flow coefficients for (anti)protons, charged pions, and -hyperons as a
function of transverse momentum.
Literature
1.
2.
3.
4.
V. Vasendina et al. Phys. Part.Nucl. Lett. 2013, V. 10, pp. 769-777.
V. Vasendina et al. Phys. Part. Nucl. Lett. 2015, V. 12, pp. 100-112.
M. Ilieva et al. accepted for publication in Phys. Part. Nucl. Lett.
N. Geraksiev for the MPD Collaboration, PoS (Baldin ISHEPP XXII) 131.