Open access peer-reviewed chapter

Non-equilibrium Equation of State in the Approximation of the Local Density Functional and Its Application to the Emission of High-Energy Particles in Collisions of Heavy Ions

By A.T. D’yachenko and I.A. Mitropolsky

Submitted: December 26th 2019Reviewed: March 24th 2020Published: February 3rd 2021

DOI: 10.5772/intechopen.92247

Downloaded: 39


The non-equilibrium equation of state is found in the approximation of the functional on the local density, and its application to the description of the emission of protons and pions in heavy ion collisions is considered. The non-equilibrium equation of state is studied in the context of the hydrodynamic approach. The compression stage, the expansion stage, and the freeze-out stage of the hot spot formed during the collisions of heavy ions are considered. The energy spectra of protons and subthreshold pions produced in collisions of heavy ions are calculated with inclusion of the nuclear viscosity effects and compared with experimental data for various combinations of colliding nuclei with energies of several tens of MeV per nucleon.


  • local density functional
  • hydrodynamics
  • non-equilibrium equation of state
  • heavy ions
  • hot spot
  • nuclear viscosity
  • protons
  • subthreshold pions

1. Introduction

The main object of studying heavy ion collisions is to study the equation of state (EOS) of nuclear matter. Along with molecular dynamics and the Vlasov dynamic equation, nuclear hydrodynamics is an effective method for describing the interaction of heavy ions with medium and intermediate energies (see, e.g., [1]). Typically, the equilibrium EOS is used [1]; it involves the local thermodynamic equilibrium in the system. A hybrid model was proposed for use at high energies in [2, 3]. It includes a fast non-equilibrium stage and the subsequent description of the dynamics of a nucleus-nucleus collision based on equilibrium relativistic hydrodynamics of an ideal fluid. We showed in our works [4, 5, 6, 7, 8, 9, 10, 11] that local thermodynamic equilibrium is not immediately established in the process of collisions of heavy ions, since the non-equilibrium component of the distribution function, which leads to the formation of a collisionless shock wave, is important at the compression stage.

The kinetic equation for finding the distribution function of nucleons is used in this paper. It is solved in conjunction with the equations of hydrodynamics, which are essentially local conservation laws of mass, momentum, and energy. As a result, the non-equilibrium equation of state is found in the approximation of the functional on the local density. Since the emitted secondary particles (nucleons, fragments, and pions) contain the basic information about the EOS, it is necessary to know the differential cross sections for the emission of these particles. The energy spectra of protons and subthreshold pions with allowance for nuclear viscosity are analyzed in this paper as a follow-up to our works [11, 12, 13] devoted to the energy spectra of protons and fragments in which viscosity was neglected.

By subthreshold production, we mean the generation of πmesons with energies lower than the threshold for the production of pions ENNin free nucleon-nucleon collisions. The absolute thresholds for pion production are ENN=2mπ+mπ22m290MeV in nucleon-nucleon collisions, ENAmπ140MeV in nucleon-nucleus collisions, and EBA=mπ2+2A+Bmπm2ABm20MeV in nucleus-nucleus collisions at A=B=12, where mπis the pion mass and mis the nucleon mass. This expression for the absolute threshold energy is obtained from a comparison of the relativistic invariants J=E2P2before and after the collision, neglecting the binding energy of pion (Eis the total energy; Pis the total momentum).

The pion production threshold during the collision of heavy ions decreases owing to collective effects and the internal motion of nucleons. These effects are naturally taken into account using the hydrodynamic approach, which explicitly includes the many-particle nature of colliding heavy ions. In the case of low energies, the hydrodynamics should be modified to take into account the non-equilibrium EOS, which describes the transition from the initial non-equilibrium state to the state of local thermodynamic equilibrium.

Such an approach to describing the temporal evolution of the resulting hot spot includes a compression stage and an expansion stage taking into account the nuclear viscosity that we found. The calculated energy spectra of protons and pions produced in nuclear collisions (both identical and different in mass) at an energy of 92 MeV per nucleon in the case of protons and 94 MeV per nucleon in the case of subthreshold pions are in agreement with the available experimental data [1, 14], respectively.

2. Non-equilibrium equation of state in a local density approximation

If the energies of colliding heavy ions are less than 300 MeV per nucleon (pion production threshold in free nucleon-nucleon collisions), we use the kinetic equation to find the nucleon distribution function frpt(rx1x2x3is the spatial coordinate; pp1p2p3is the momentum; tis the time) [11, 12]:


where f0rptis a local equilibrium distribution function; τis the relaxation time; Wρ(Wρ=αρ+βργ) is a one-particle Skyrme-type self-consistent potential depending on the density ρ, where three parameters αα<0, ββ>0, and γγ>1are determined by setting the equilibrium density ρ0= 0.145 fm−3, binding energy Eb= −16 MeV, and compression modulus K = 210 MeV; and mis the nucleon mass.

Equation (1) with allowance for the hydrodynamic equations obtained from (1) by taking the corresponding moments with a weight of 1, p, and p22m[11, 12] describes the dynamics of nuclear collisions and forms the basis of our approach. The solution of Eq. (1) can be simplified if we work out the distribution function frptdetermining EOS in the form


where the distribution function f1rptis defined in momentum space as an axially symmetric Fermi ellipsoid, which is a convenient form for describing excitations in the Fermi liquid theory and is assumed to be blurred along the axis p1with the temperature T1and frozen in the transverse directions p2and p3. The function f0rptis represented in the momentum space by the equilibrium Fermi sphere blurred with temperature T; qis a relaxation factor (q=expt0tdt/τ), where t0rtis the start time of the relaxation process in the system; τis the relaxation time, which can be specified as in [15]. However, we define τmore traditionally as τ=λ/υT, where λis the mean free path of nucleons at a given nucleon density, which is assumed to be equal to the mean distance between nucleons, and υTis the mean speed of the thermal Fermi motion of nucleons. This expression for τin the energy range under consideration is close in magnitude to the value proposed in [15], but it turns out to depend on temperature and compression ratio and seems to us more realistic. All calculations are carried out precisely for such τ. The equation for finding the relaxation factor qrtis obtained by taking the moment for the kinetic equation with a weight of p12p22+p32/2that determines the degree of anisotropy of the distribution function frptin momentum space.

So, the initial moments d3p2πℏ3with weights 1, p, p22m, p12p22+p32/2from kinetic equation (1) provide the corresponding hydrodynamic equations [11, 12] for finding nucleon density ρrt, velocity field υrt, internal energy density e=ε+I+eint, and pressure tensor Pij=Pkinij+Pintδij(the repeated indices imply the summation, δijis the Kronecker symbol). The terms of interaction for energy density eintand pressure Pintare, respectively,


The kinetic terms are.




which corresponds to diagonal tensor of pressure Pkinij=0ij, and heat terms I and I1are associated with temperatures T and T1, respectively. Since we assume that the integrals of motion (density ρ, momentum density mρυ, and energy density (ε+I=ε1+2ε2+I1)) are conserved during relaxation, hydrodynamic equations have no right-hand sides.

To find density ρ, velocity field υrelaxation factor q, and temperatures T and T1, we thus have the closed system of equations that considers expressions for terms of interaction and kinetic terms. These equations allow us to find distribution function frptin form (2). Relaxation factor q0q1allows us to describe the dynamics of the Fermi surface variation from the state with q=1, where function frptin the momentum space is maximally anisotropic, to the state with q=0, where it is completely isotropic.

3. Hydrodynamic stage

We simplify the description of the time evolution of colliding nuclei distinguishing the compression stage, the expansion stage, and the freeze-out stage of the resulting hot spot. We reduce the interaction between two nuclei to the interaction between their overlapping regions. This can be interpreted as a hot spot formation process. In this case, we take into account the conservation laws. Shock waves with changing front diverging in opposite directions are formed at the stage of compression during the interaction between overlapping regions of colliding nuclei [5, 6, 7, 8, 9].

In the process of compression, when the shock wave reaches the boundaries of the hot spot, the density reaches its maximum value. The dependence of the maximum compression ratio ρ/ρ0at the shockwave front (solid line) on the collision energy of nuclei E0is shown in Figure 1. It hardly depends on the composition of colliding nuclei, since we consider the interaction of the same overlapping regions in the system of equal speeds of the colliding nuclei. The dependence of ρ/ρ0on the energy E0for the distribution function corresponding to the equilibrium EOS with q=0is shown by a dashed line, and such a dependence for a completely non-equilibrium EOS with q=1is shown by a dash-dotted line.

Figure 1.

The dependence of the maximum compression ratio ρ/ρ0 on the collision energy E0 achieved during the interaction of the overlapping regions of colliding nuclei for the case of the relaxation factor q calculated by us (solid line), for the case where the factor q=0 (dashed line) and for the case where q=1 (dash-dotted line). All dependences correspond to the value of the compression modulus K=210 MeV.

The relaxation factor at the energy region of E0<100MeV per nucleon is maximal (q=1), and it decreases with increasing energy, leading to a greater isotropy of the distribution function. We calculated the dependence of the maximum compression ratio on energy for E0>100MeV per nucleon. It is found in between the extreme cases with q=0and q=1. At E0<100MeV per nucleon, the dependence ρ/ρ0on energy coincides with the dash-dotted curve corresponding to the case with q=1(i.e., completely non-equilibrium EOS) and is located above the dashed curve corresponding to the case with q=0belonging to traditional hydrodynamics and the onset of local thermodynamic equilibrium.

A compressed and heated hot spot (a ball with radius R) expands when the shock wave reaches the boundaries of the system. The hot spot expands in accordance with the equations of hydrodynamics for radial motion of nucleon density ρrt, velocity υrt, energy density ert, and pressure Prt, following from (1) [11, 12]:


The heat flux for a local equilibrium distribution function is Q=0. Here, the internal energy density is e=ekin+eintand pressure is P=Pkin+Pint, where ekinand Pkinare the kinetic terms, and the interaction terms eintand Pintare.


The velocity field is found from Eq. (3) in the approximation of a homogeneous but time-dependent density of hot spot ρrt=ρt:


where Rtis the radius of the hot spot; R1tis the radius of the velocity field kink determined from the solution of equations; and Ṙtand Ṙ1tare the derivatives in time (speed), which are also found from the Eqs. A system of ordinary integro-differential equations is obtained after integrating Eqs. (4) and (5) over the hotspot volume. It is solved numerically.

However, the deviation of the distribution function frptfrom the local equilibrium function f0rptis not taken into account in these equations. Expressing frptfrom the right side of Eq. (1) through its left side, we find


where Iis the thermal term depending on the temperature T. When obtaining (9), we substituted f0ρrtUrptTrtinto the left part of Eq. (1) instead of frpt, taking into account Eqs. (3)(5), where U=pmυ22m. In this case, the hot spot was averaged over the volume to derive Eq. (1), and at the expansion stage, the density ρrt, the temperature Trt, and the thermal term Irtwere considered to be homogeneous functions of time tand independent of the radius r. Substituting expression (9) into the equations of hydrodynamics [11, 12], we find the corrections to kinetic terms of the energy density ekinand pressure Pkin:


where e0,kin=eF+Iand P0,kin=23e0,kinare the equilibrium kinetic parts of the energy density and pressure density, eF=3102m32π2ρ2/3ρ, and η=43P0,kin+56eFτis the viscosity coefficient. The following correction terms turn out to be an order of magnitude smaller and they are not taken into account. The heat flux is Q=0. The corrections to kinetic terms significantly affect the hotspot expansion and slow it down, because the Reynolds number is not large Re=mρυlη1for the viscosity coefficient ηfound by us (formula (10)) in the energy range under consideration of E0≈ 100 MeV per nucleon with a characteristic nuclear size of l ≈ 3 fm. In our case, the temperature is T20 MeV; P0,kinρT; τ31023s; the viscosity coefficient is η41010kg m−1 s−1. It coincides in the order of magnitude with the gas estimate ηmT/σ[16] if we take σ40 mb for the elementary cross section. Moreover, ηs>>4π, where sis the entropy density (sρ). That is, in our case, the ratio ηsis more than an order of magnitude higher than the limiting value of 4π[17] (achievable, e.g., in the state of a quark-gluon plasma). Thus, the viscosity coefficient is quite large in the energy range under consideration. This reduces the expansion speed of the hot spot and increases its temperature. Secondary particles (nucleons, fragments, and pions) form and freeze out when the expanding nuclear system reaches a critical density (freezing density) ρdetermined from the condition dPintdρ=ρdWdρ=0.

4. Double differential cross sections of the emission of protons and pions: comparison with the experimental data

Protons and pions are emitted when the nuclear system reaches a critical density. The cross section of the emission of protons (pions) is found from the condition that the number of particles fd3pand the value d3p/Eof are relativistic invariants [18, 19]. As a result, the inclusive double differential cross section of reaction A + B → p(π) + X is


where bis an impact parameter and the distribution function of protons (pions) has the form


Here Eand pare the total energy and momentum of the proton (pion), respectively; E=p2+mpπ2; Ωis the solid angle; Sis the spin; υrtand Trtare the velocity field and temperature at the time of freeze-out (they are solutions of the equations of hydrodynamics); γ=1/1υ/c2is the Lorentz factor; μis the chemical potential (for pions μ=0, because the number of pions is not specified). The factor Gb=σtb/σgbintroduced in (12) takes into account the difference between the total cross section and the geometric cross section, where σtbis defined as the cross section of the formation of a hot spot for a given impact parameter bfrom two overlapping regions in colliding nuclei, and σgbis equal to the geometric cross section of these overlapping regions. Here, the total cross section is always greater than geometric one, as in the case of the fusion of two nuclei comparable in size. In addition, the function frptincluded in Eq. (12) was modified in comparison with Eq. (13) according to relation (2): the sign “+” refers to protons, and the sign “−” refers to pions. Expressions (12) and (13) refer to protons (pions) emitted from a hot spot as a result of the interaction of the overlapping regions of colliding nuclei. In addition to this contribution, we took into account the contribution from the emission of protons (pions) as a result of the fusion of non-overlapping regions of colliding nuclei. The calculated double differential cross sections of proton emission (energy spectra) were compared with similar calculations obtained by solving the Vlasov-Uling-Uhlenbeck (VUU) kinetic equation [1] and with available experimental data. Our calculations corresponded to the equation of state with selected compression modulus equal to K = 210 MeV, i.e., with the same which was taken for the best description of the experiment in the calculations that we performed in [8, 9] at energies of 250 and 400 MeV per nucleon for colliding Ne and U nuclei.

We present the proton spectra in the 40Ar + 40Ca p+Xreaction at the angles of 30° (1), 50° (2), 70° (3), and 90° (4) for the energy of projectile nuclei of 40Ar of 92 MeV per nucleon (Figure 2). In Figure 2, the solid curves correspond to our calculation, the histograms correspond to the calculations performed by the method of solving the VUU equation [1], and the dots are the experimental data from [1].

Figure 2.

Spectra of protons emitted in the reaction 40Ar + 40Ca with the energy of 40Ar ions of 92 MeV per nucleon at angles of 30° (1), 50° (2), 70° (3), and 90° (4). The solid lines are the results of calculations according to this model with the calculated q corresponding to K=210 MeV; the histograms are the results of calculations obtained from the solution of the VUU kinetic equation (1); the dots are the experimental data from [1].

As can be seen, our calculation (this is not the Monte Carlo method and not histograms) is in good agreement with the experimental data. This is especially true for small angles of emission of protons (30°, 50°, and 70°). Our approach has an advantage over the more detailed method of solving the VUU equation [1], since the solid curves (but not histograms) are the result of the calculation. Note here that simple cascade models, as mentioned in [1], cannot describe these experimental data at all.

We compared our data with the available experimental data on the emission of pions. Figure 3 illustrates the comparison of our calculated (solid lines) and experimental [14] (dots) double differential cross sections for the reactions of π+-meson production when 16O ions collide with 27Al nuclei (curve 1), 58Ni nuclei (curve 2), and 197Au nuclei (curve 3) at energies of 16O ions of E0=94 MeV per nucleon at an angle of 90°. It can be seen that the calculation is in good agreement with the experiment for chosen parameters of the nuclear interaction and taking into account the viscosity of the medium ηthat is found by us and proportional to the relaxation time τwithin the experimental errors. In this case, the effect of viscosity on the calculated cross section of emitted pions is stronger for more asymmetric combinations of colliding nuclei, when the contribution of the emission of pions from the hot spot prevails. Thus, inclusive pion spectra in asymmetric nuclear collisions can be used to measure the viscosity of a nuclear medium.

Figure 3.

The calculated (solid lines) and experimental (dots) [14] inclusive double differential cross sections of the emission of mesons at the observation angle of 90° in the reactions 16O + 27Al (1), 16O + 58Ni (2), and 16O + 197Au (3) with energy of 16О ions of E0=94 MeV per nucleon.

Figure 4 illustrates the comparison of the calculations (solid lines) with the experimental data [14] (dots) for the reaction 16O + 27Al → π++ X at energy of 16O ions of 94 MeV per nucleon at pion emission angles of 70° (curve 1), 90° (curve 2), and 120° (curve 3). The calculation is in agreement with the experimental data if its parameters are constant. In all the illustrations under consideration, the agreement of calculation with the experiment was achieved without introducing fitting parameters and is more successful than our previous works [11, 19, 20].

Figure 4.

The calculated (solid curves) and experimental (dots) [14] inclusive double differential cross sections of the emission of mesons in the reaction 16O + 27Al with energy of 16O ions of 94 MeV per nucleon at the observation angles of 70° (curve 1), 90° (curve 2), and 120° (curve 3).

5. Conclusions

Thus, the idea of using the hydrodynamic approach with a non-equilibrium equation of state in describing collisions of heavy ions is further developed in this work. The non-equilibrium equation of state is found in the approximation of the functional on the local density. The differential cross sections of the emission of protons and the production of subthreshold pions in heavy ion collisions are uniformly described with the same fixed parameters of the equation of state and in the same approach as in the previous papers [11, 12, 13], which describe the differential cross sections for the formation of protons and light fragments. It is shown that the non-equilibrium equation of state included in the hydrodynamic equations allows us to describe the experimental energy spectra of protons produced in collisions of heavy ions with intermediate energies better than the equation of state corresponding to traditional hydrodynamics, which initially implies the local thermodynamic equilibrium.

This simplified hydrodynamic approach including a description of the stages of compression, expansion, and freeze-out of a substance during heavy ion collisions turned out to be no worse than a more detailed approach based on the Monte Carlo solution of the Vlasov-Uling-Uhlenbeck kinetic equation.

In comparison with previous works, the inclusion of the effects of nuclear viscosity, which we found in the relaxation approximation for the kinetic equation, is new. This did not add new parameters in describing the temporal evolution of nuclear collisions. The relaxation time τ, which determines the nuclear viscosity coefficient η, turned out to be close to the value found on the basis of the behavior of nuclear Fermi liquid [15] and is not a fitting parameter. When describing the emission of protons and fragments, the inclusion of the viscosity of the medium is not so significant, and the pions are very sensitive to the viscosity.

The highlighting of proton (pion) emission after the temporal evolution of the resulting hot spot and the contribution to the particle emission cross sections during the fusion of “spectators” (non-overlapping regions of colliding nuclei) were significant in calculating the cross sections. This made it possible to describe the differential cross sections of the emission of protons (pions) for collisions of nuclei in various combinations. Highlighting this feature of our approach can be useful in comparison with other ways of pion production in heavy ion collisions, for example [21, 22], based on the solution of the Vlasov-Uling-Uhlenbek equation. These works include a range of higher energies of colliding heavy ions (more than 300 MeV per nucleon) and the production of pions by means of Δ-isobar production. We included this channel at low subthreshold energies, not limited to the production of thermal pions. However, this channel appears on the higher energy tails of the energy spectra of pions [23].

Studies of the formation of protons, fragments, and subthreshold production of pions may be of interest for the development of a scientific program planned with radioactive beams in Dubna using the COMBAS facility [24], which is designed to study nuclear collisions in the energy range of 20–100 MeV per nucleon.

© 2020 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution 3.0 License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite and reference

Link to this chapter Copy to clipboard

Cite this chapter Copy to clipboard

A.T. D’yachenko and I.A. Mitropolsky (February 3rd 2021). Non-equilibrium Equation of State in the Approximation of the Local Density Functional and Its Application to the Emission of High-Energy Particles in Collisions of Heavy Ions, Density Functional Theory Calculations, Sergio Ricardo De Lazaro, Luis Henrique Da Silveira Lacerda and Renan Augusto Pontes Ribeiro, IntechOpen, DOI: 10.5772/intechopen.92247. Available from:

chapter statistics

39total chapter downloads

More statistics for editors and authors

Login to your personal dashboard for more detailed statistics on your publications.

Access personal reporting

Related Content

This Book

Next chapter

Introductory Chapter: A Brief Mention for High-Pressure in Oxides from DFT Simulations

By Sérgio Ricardo de Lázaro

Related Book

First chapter

Thermodynamic Properties of Nano-Silver and Alloy Particles

By Wangyu Hu, Shifang Xiao, Huiqiu Deng, Wenhua Luo and Lei Deng

We are IntechOpen, the world's leading publisher of Open Access books. Built by scientists, for scientists. Our readership spans scientists, professors, researchers, librarians, and students, as well as business professionals. We share our knowledge and peer-reveiwed research papers with libraries, scientific and engineering societies, and also work with corporate R&D departments and government entities.

More About Us