Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

Manuscript Title: THERMINATOR: THERMal heavy-IoN generATOR
Authors: Adam Kisiel, Tomasz Taluc, Wojciech Broniowski, Wojciech Florkowski
Program title: THERMINATOR
Catalogue identifier: ADXL_v1_0
Distribution format: tar.gz
Journal reference: Comput. Phys. Commun. 174(2006)669
Programming language: c++.
Computer: PC, Pentium III, IV, or Athlon, 512 MB RAM; NOT HARDWARE DEPENDENT; (any computer with the c++ compiler and the ROOT environment [1]).
Operating system: Linux: Mandrake 9.0, Debian 3.0, SuSE 9.0, Red Hat FEDORA 3, etc., Windows XP with Cygwin ver.1.5.13-1 and gcc ver. 3.3.3 (cygwin special); NOT SYSTEM DEPENDENT.
RAM: 50 Mbytes
Keywords: Monte Carlo generator, relativistic heavy-ion collisions, thermal models, particle production, transverse-momentum spectra.
PACS: 25.75.-q, 25.75.Dw, 25.75.Gz, 25.75.Ld.
Classification: 11.2.

External routines: ROOT ver. 4.02.00.

Nature of problem:
Statistical models have proved to be very useful in the description of soft physics in relativistic heavy-ion collisions [2]. In particular, with a few physical input parameters, such as the temperature, chemical potentials, and velocity of the collective flow, the models reproduce the observed particle abundances [3-15], the transverse-momentum spectra [16], balance functions [17,18], or the elliptic flow [19,20] in both non-strange and strange sectors. The key element of the approach is the inclusion of the complete list of hadronic resonances, which at the rather high temperature at freeze-out, ~ 165 MeV, contribute very significantly to the observed quantities. Their two- and three-body decays, taken from the tables, proceed in cascades, ultimately producing the stable particles observed in detectors. At the moment there exist several codes to compute the abundances of particles (the publicly available programs for this purpose are SHARE [21] and THERMUS [22], which is a rather simple task, since the abundances are insensitive to the geometry of the fireball and its expansion. On the other hand, the calculation of the transverse-momentum spectra of particles is much more complicated due to the sensitivity to these phenomena. THERMINATOR deals with this problem, offering the full information on the space-time positions and momenta of the produced particles. As a result, the program allows for the very efficient computation of the transverse-momentum spectra of identified particles and the examination of the implications of the assumed expansion model. THERMINATOR allows easily for the departure from symmetries typically assumed in other approaches. This opens the possibility to study the dependence of physical quantities on rapidity and the azimuthal angle. The contribution of the resonances to various observables may be traced conveniently, and their role in the statistical approach may be verified. As a Monte Carlo event generator written in the object-oriented c++ language in the ROOT [1] environment, THERMINATOR can be straightforwardly interfaced to the standard software routinely used in the data analysis for relativistic heavy-ion colliders, such as SPS, RHIC, and, in the future, LHC. In this way the inclusion of experimental acceptance, kinematic cuts, or interfacing with other programs poses no difficulty.

Solution method:
THERMINATOR uses the particle data tables [23] in the universal input form used by the SHARE [21] package. The user decides on the thermal parameters and the preferred expansion model. The optimum thermal parameters may be taken, e.g., as those obtained with the help of SHARE [21] or THERMUS [22]. At the moment there are two different expansion models implemented in the code: the model of Ref. [16] based on the so-called Buda-Lund [24] parameterization, and the Blast-Wave model [25,26]. The positions and velocities of the particles are randomly generated on the hypersurface according to the statistical (Bose-Einstein of Fermi-Dirac) distribution factors. All particles, stable and unstable, are included. The particles move along classical trajectories from their initial positions, with velocities composed of the thermal motion and the collective expansion of the system. Stable particles just stream freely, while the resonances decay after some (randomly generated) time, which is controled by the particle's lifetime. The decays are two-body or three-body, and their implementation involves simple kinematic formulas. The decays can proceed in cascades, down to the stage where only stable particles are present. All particles have tags indicating their parent. The secondary rescatterings are not considered in this approach. Full history of the event is stored in an output file, allowing for a detailed examination of the space-time evolutions and the calculation of the transverse-momentum spectra.

Additional comments:
The ongoing analysis of the SPS and the RHIC data as well as the future heavy-ion program at LHC will certainly benefit from THERMINATOR as a tool for generating events in a simple statistical model. The Monte Carlo code written in c++ and using the standard ROOT [1] environment can be easily adapted to purposes directly linked to experimental data analysis. The space-time tracking capability will allow, in the framework of the statistical approach, a better understanding of the physics of relativistic heavy-ion collisions. THERMINATOR calculates the particle spectra and other observables related to the space-time evolution of the system. It provides a c++ framework which may be easily developed for detailed analysis of more involved observables such as, e.g., correlation functions or HBT radii.

Running time:
The generation of 500 events from scratch takes about 1 hour 15 minutes on a PC with Athlon-Barthon 2.5 GHz under Red Hat Fedora 3. Each subsequent 500 events take about 1 hour. To store 500 events about 1.1 GB disk storage is needed, depending on the kinematic range. After converting the output to the ROOT TTree format, 900 MB may be freed.

References:
[1] R. Brun and F. Rademakers, Nucl. Instrum. Meth. A389 (1997) 81, (http://root.cern.ch).
[2] P. Braun-Munzinger, K. Redlich and J. Stachel, (2003), nuclth/ 0304013.
[3] P. Koch and J. Rafelski, South Afr. J. Phys. 9 (1986) 8.
[4] J. Cleymans and H. Satz, Z. Phys. C57 (1993) 135, hep-ph/9207204.
[5] J. Sollfrank et al., Z. Phys. C61 (1994) 659.
[6] P. Braun-Munzinger et al., Phys. Lett. B344 (1995) 43, nucl-th/9410026.
[7] P. Braun-Munzinger et al., Phys. Lett. B365 (1996) 1, nucl-th/9508020.
[8] J. Cleymans et al., Z. Phys. C74 (1997) 319, nucl-th/9603004.
[9] F. Becattini, J. Phys. G23 (1997) 1933, hep-ph/9708248.
[10] G.D. Yen and M.I. Gorenstein, Phys. Rev. C59 (1999) 2788, nuclth/ 9808012.
[11] P. Braun-Munzinger, I. Heppe and J. Stachel, Phys. Lett. B465 (1999) 15, nucl-th/9903010.
[12] J. Cleymans and K. Redlich, Phys. Rev. C60 (1999) 054908, nuclth/ 9903063.
[13] F. Becattini et al., Phys. Rev. C64 (2001) 024901, hep-ph/0002267.
[14] P. Braun-Munzinger et al., Phys. Lett. B518 (2001) 41, hep-ph/0105229.
[15] W. Florkowski, W. Broniowski and M. Michalec, Acta Phys. Polon. B33 (2002) 761, nucl-th/0106009. 33
[16] W. Broniowski and W. Florkowski, Phys. Rev. Lett. 87 (2001) 272302, nucl-th/0106050.
[17] W. Florkowski, W. Broniowski and P. Bozek, J. Phys. G30 (2004) S1321, nucl-th/0403038.
[18] P. Bozek, W. Broniowski and W. Florkowski, Acta Phys. Hung. A22 (2005) 149, nucl-th/0310062.
[19] W. Broniowski, A. Baran and W. Florkowski, AIP Conf. Proc. 660 (2003) 185, nucl-th/0212053.
[20] W. Florkowski, W. Broniowski and A. Baran, (2004), nucl-th/0412077.
[21] G. Torrieri et al., (2004), nucl-th/0404083.
[22] S. Wheaton and J. Cleymans, (2004), hep-ph/0407174.
[23] Particle Data Group, K. Hagiwara et al., Phys. Rev. D66 (2002) 010001.
[24] T. Csorgo and B. Lorstad, Phys. Rev. C54 (1996) 1390, hep-ph/9509213.
[25] E. Schnedermann, J. Sollfrank and U.W. Heinz, Phys. Rev. C48 (1993) 2462, nucl-th/9307020.
[26] F. Retiere and M.A. Lisa, Phys. Rev. C70 (2004) 044907, nuclth/ 0312024.