The nuclei collision research programme at the LHC accelerator gives a unique possibility for the investigation of the properties of the strong-interacting matter of so high energy density as expected sufficient for deconfinement of hadronic matter and creation of quark-gluon plasma (QGP), wherein collective effects result in screening of colour interaction between partons. Strong-interacting objects as heavy quarkonium (Upsilon-family) and hard partonic jets, are produced at the initial phase of the nuclei collision and propagate through the dense matter carring information about its state. The actual task is simulating the processes of generation and detection of such objects in application to the CMS apparatus.

1. Dimuon resonance reconstruction in heavy ions collisions.

The muon tracks recognition is necessary to observe the Upsilon-suppression effect, the latter being one of the most important signatures of QGP. This effect is of special interest in heavy nuclei interactions. In particular, the heavy quark bound states Upsilon(2S) and Upsilon(3S) are predicted to be suppressed by colour screening strongly relatively to Upsilon(1S), what can be examined with different ion species. However in heavy ion case we get the serious problem of the tracker detector occupancy. It may happen in central Pb-Pb collisions (the multiplicity is estimated up to 8000 charged and 4000 neutral particles per rapidity unit) that only these detectors at the distance more than 80 cm, and the pixel detector, would have a reasonable occupancy (20-25%). Thus one needs to have an algorithm, which could work effectively and fast under maximum occupancy using only the part of tracker system.

The algorithm of dimuon reconstruction in the case of heavy ions with CMS apparatus has been developed [1]. It includes a program package, which is able to trace tracks through the apparatus, get the responses of detectors, and reconstruct the resonances up to multiplicites expected for central Pb-Pb collisions. The main goal was to find criteria, which would allow getting the maximum number of dimuons from Upsilon decay as well as providing the maximum suppression of uncorrelated muon pair background from pion/kaon decays, which gives the response in muon stations.

The efficiency of reconstruction depends on the multiplicity: from 66% -- for the maximum multiplicity corresponding to central Pb-Pb interactions, up to 90% -- for multiplicity reduced by a factor 3. The signal/background ratio in Upsilon-mass region is 6 times higher than we had without the algorithm using and almost do not depend on the multiplicity.

2. Jet production in heavy ion collisions.

High-transverse momentum parton pair (dijet) from a single hard scattering is produced at the initial stage of the collision process ( < 0.01 fm/c). Jet then propagates through the QGP formed due to mini-jet production on larger time scale ( ~0.1 fm/c) and interacts strongly with the comoving constituents of the medium. The consequence of this got to be the energy losses of a hard partonic jets: radiative losses due to gluon "bremsstrahlung" induced by multiple scattering and collisional losses due to the final state interactions (elastic rescatterings) of jet partons with the medium constituents [2]. Since the jet rescattering intensity is a strongly increasing function of temperature, the formation of super dense and hot partonic matter in heavy ion collisions at initial temperature up to ~1 GeV should result in significantly larger jet energy losses as compared to the case of hadronic gas at temperature ~0.2 GeV or cold nuclear matter.

To study energy losses of a hard parton in QGP ("jet quenching") three different processes are considered: QCD jet pair, Z+jet and gamma+jet production. However utilization of hard jet characteristics to investigate QGP in heavy ion collisions going to be difficult because of "false" jets background -- statistical fluctuations of the transverse energy flux arising from a huge multiplicity of "thermal" secondary particles in an event. Predictions of different models give from 3000 to 8000 charged particles per rapidity unit in a central Pb-Pb collision at LHC energies. All cells of the hadronic calorimeter are filled completely in this situation (the cell size in eta-phi space is ~ 0.1 X 0.1 in central rapidity region < 1.5), and the response of the calorimeter to the transverse energy flux of many "thermal" particles hit the cell can imitate a signal from single high-p_T particle. Under this condition the reconstruction of "true" QCD jets from hard parton-parton scattering is a vital question for the CMS heavy ion physics programme. A number of attempts were dedicated to jet finding algorithms optimization in heavy ion collisions in CMS conditions[3,4].

The threshold transverse energy of hard jet recognition in central Pb-Pb collisions has been calculated to be E_T = 100 GeV with the recognition efficiency close to 100% and low background from "thermal" particles ( < 10%). In order to avoid the increasing of energy threshold for jet recognition due to energy losses of jet partons in the dense matter, the optimization of jet finding algorithm has been made using a different intrinsic structure of "false" and QCD-jets, in particular, different dependence of the mean transverse momentum of jet particles on the total jet energy [5]. Expected statistics should be sufficient to study high-E_T (> 100 GeV) jet production for different values of the jet transverse energy and centrality of the collisions.

3. The transverse energy measurement and impact parameter estimation.

In order to estimate parameters of the dense matter created in heavy ion collision one have to know the "centrality" of interaction. The measurement of the total transverse energy of secondary particles in the CMS calorimeters gives a tool for estimation of the nuclei collision impact parameter.

The huge multiplicity of particles, produced in heavy ion interactions, requires significant computing time to simulate the response of calorimeters. Therefore a software package for the fast simulation of hadronic calorimeter response based on the hadron shower profile parameterization has been developed and used with the CMS simulation tool cmsim008 [6].

In particular, the Ca-Ca nuclei interactions at energy 7.3 GeV per nucleon pair were simulated with FRITIOF 7.02 generator. The simulation of transverse energy measurement in these events has shown that the energy cut permits the selection of the events with impact parameter less than radius of the Ca nucleus. Average impact parameter = 2 fm with the 8% of events being survived.


[1] M.Bedjidian, O.L.Kodolova. "Dimuon reconstruction in heavy ion collisions." CERN CMS Note/97-095

[2] I.P. Lokhtin, A.M. Snigirev. "Probing the space-time evolution of hot parton matter by hard jets" Zeit. f. Phys. C73 (1997) 315

[3] R.Kvatadze, R.Shanidze. "The jet recognition in heavy ion collisions with CMS". CERN CMS TN/94-270

[4] N.A.Kruglov et al. "Jet finding algorithms in heavy ion collisions in CMS." CERN CMS TN/96-084

[5] N.A. Kruglov, I.P. Lokhtin, L.I. Sarycheva, A.M. Snigirev. "On the problem of hard jet recognition in heavy ion collisions." Zeit. f.Phys. C76 (1997) 99

[6] N.A.Kruglov. "Simulation of the measurement of transverse energy in heavy ion collisions in CMS". Talk given in CMS Heavy Ion Workshop (Lyon, June 10-11, 1998). CMS Document 96-112.