Ω and ϕ production in Au+Au collisions at sNN = 11.5 and 7.7 GeV in a dynamical quark coalescence model

NUCLEAR PHYSICS AND INTERDISCIPLINARY RESEARCH

Ω and ϕ production in Au+Au collisions at $\sqrt{s_{_{N N}}}$ = 11.5 and 7.7 GeV in a dynamical quark coalescence model

Xiao-Hai Jin，

Jin-Hui Chen，

Yu-Gang Ma，

Song Zhang，

Chun-Jian Zhang，

Chen Zhong

Nuclear Science and Techniques

Vol.29, No.4

Article number 54

Published in print 01 Apr 2018

Available online 16 Mar 2018

DOI：10.1007/s41365-018-0393-1

40100

The Ω and ϕ production in relativistic heavy-ion collisions is studied in a dynamical quark coalescence model using the phase space information of strange quarks from a multiphase transport (AMPT) model. Enhanced local parton density fluctuation is implemented in the AMPT to simulate the QCD phase transition dynamics. By studying the transverse momentum pT spectra and the elliptic flow of the multi-strangeness particles, such as Ω and ϕ, and the Ω/ϕ ratio as a function of pT in the AMPT, we find that the new development improves the description of experimental data. The study motivates further experimental investigations of Ω and ϕ production in Phase-II of the Beam Energy Scan program at RHIC.

QCD phase transitionmulti-strangeness particleselliptic flowAMPT

1 Introduction

Searching for the QCD critical point and mapping the QCD phase diagram are major scientific goals of the Beam Energy Scan (BES) program in Relativistic Heavy-Ion Collisions (RHIC) [1]. Finite temperature lattice QCD calculations show that at high temperature and low baryon chemical potential (μ_B), a phase transition from hadron gas to the quark gluon plasma (QGP) will happen, and the phase transition is smooth and continuous [2]. At large μ_B region, the QCD-based models predict that the phase transition is of the first order and there should exist a QCD critical point as the endpoint of the first order phase boundary [3-5].

Heavy ion collisions at relativistic energy provide a unique opportunity to study the properties of QGP and the QCD phase diagram. Many exciting results in the field show rich structure of QCD matter, such as the non-monotonic behaviour of conserved quantity fluctuation and final state correlations [6-15], net-proton direct flow [16], the strangeness enhancement [17-20], the local baryon-strangeness correlation measurement from the hypernuclei system [21-23], novel quantum phenomena induced by strong magnetic fields in heavy-ion collisions [24], the breaking of the number of constituent quark scaling on identified particles [25-27], and the deviation of the Ω/ϕ ratio in BES in comparison with data at top RHIC energy.[28]

Multi-strange hadrons, such as Ω and ϕ, are clean probes to explore the QCD phase diagram [29, 30], as they are expected to have a relatively small hadronic interaction cross section and little contribution from resonance decay [31, 32]. Therefore, they can carry the information directly from the chemical freeze-out stage with little or no distortion due to hadronic rescattering. As a result, the production of Ω and ϕ particles offers an advantage in probing the transition from partonic to hadronic dynamics [29]. In a recent measurement of Ω and ϕ production at midrapidity from Au+Au collisions at $\sqrt{s_{_{N N}}}$ = 7.7, 11.5, 19.6, 27, and 39 GeV from the STAR experiment at RHIC, one finds that the Ω/ϕ ratios scaled by their number of constituent quark show a suppression of strange quark production in central collisions at 11.5 GeV compared to results at $\sqrt{s_{_{N N}}}$ ≥ 19.6 GeV, and the data of 19.6 GeV or above show little beam energy dependence, which may suggest strange quark equilibration is achieved in higher energy but not in lower energy [28]. We have carried out a detail study on the underlying dynamics of strange quark matter at the RHIC based on AMPT model previously, and found that the AMPT model presents the trend of Ω/ϕ vs p_T of the data, though under-predicts the Ω yield by a factor of 5 [30]. New development of the naive quark coalescence model in the AMPT model improves the description of baryon production in heavy-ion collisions, such as protons and Λ, but still significantly under-predicts the Ω and Ξ yield [33]. An alternative solution using a dynamical quark coalescence model to replace the naive quark coalescence model in the AMPT model predicted the Ω and ϕ yield in Au+Au collisions at $\sqrt{s_{_{N N}}}$ = 200 GeV reasonably well [34]. In this paper, we follow the procedure described in Ref. [34] and introduce the local parton density fluctuation effect in the AMPT model to study the Ω and ϕ production as a possible signal of the QCD phase transition. We noted that similar idea was discussed in Ref. [35].

2 The ampt model

The AMPT model is a hybrid model developed to describe heavy-ion collisions at relativistic energies [36]. It has two versions: default AMPT and string melting AMPT. In our study, we adopt the string melting version to study the QCD fluctuation effect. In the string melting version, all excited strings are converted into partons [36], it has a clear advantage over the default version in describing the flow of harmonic and pion interferometry [33, 37- 39]. The AMPT model with the string melting scenario consists of four components. The initial stage is described by a heavy-ion jet interaction generator (HIJING) [40, 41], which is designed to simulate multiple jets and particle production in heavy ion collisions. Zhang’s parton cascade (ZPC) [42] is used to describe scattering among partons. It only included two-body elastic interaction until now. The hadronization of parton is based on a naive quark coalescence model from coordinate space distribution [36]. The scattering of resulting hadrons is described by a relativistic transport (ART) model [43, 44].

In the transport approach, interactions among partons are described by their scattering cross sections, which in the AMPT model are given by[42]

\frac{d σ_{p}}{d t} ≃ \frac{9 π α_{s}^{2}}{2 (t - μ^{2})^{2}},

(1)

σ_{p} ≃ \frac{9 π α_{s}^{2}}{2 μ^{2}} \frac{1}{1 + μ^{2} / s} .

(2)

where α_s is the strong coupling constant with a typical value of 0.47 in AMPT, μ is the screening mass which depends on the medium effect [42], and s and t are the usual Mandelstam variables. By applying a large parton scattering cross sections of 6-10 mb, the AMPT model can reproduce the centrality and transverse momentum dependence of hadron elliptic flow [45]. Recent development on the input parameters of AMPT model seems to obtain a smaller value, and more close to the value used in pQCD calculation [38]. For our current study, we follow the instruction from Ref. [38] and take the parton scattering cross section of 3 mb to simulate parton interaction.

In order to understand the QCD phase diagram on a wide range of collision energies, we add local parton density fluctuation in AMPT. Viewed from the thermodynamics, a critical point is a point at which a single thermodynamic state bifurcates into two macroscopically distinct states. This bifurcation may lead to long-range thermal fluctuations. To model this effect, a large density fluctuation is introduced in the end of parton scattering in ZPC. Specifically, we assume that the partonic matter prior to the QCD phase transition consists of clusters of various sizes. We redistribute partons produced in the AMPT model to a few clusters but keep their momenta unchanged. In our algorithm, the center of a cluster is determined by the maximum number density of parton distributions in that event. Cluster positions are selected based on the freeze out positions of partons, and then far away partons in the transverse plane are moved close to the nearest cluster. All partons then coalescence into hadrons. To demonstrate this effect, we only allow in the present study the formation of four quark clusters before hadronization. An example of the parton spatial distribution is shown in panel (b) of Fig. 1. For comparison, the distribution from the AMPT model is shown in panel (a). The subsequent distribution of hadrons formed immediately after hadronization from partons are shown in panel (c) and (d), respetively. From Fig. 1, it is seen that the enhanced local parton density fluctuation has a clear effect on the spatial distributions of hadronic matter in AMPT. The fluctuation effect (panel (d)) is stronger than the original AMPT model (panel (c)). In this paper, we perform our calculations with string melting AMPT in version 2.24 for Au+Au collision.

Figure 1:

(Color online) The spatial distributions of partons and hadrons in the transverse plane from an AMPT event in Au+Au collisions at

\sqrt{s_{_{N N}}}

= 11.5 GeV. Panel (a) and (b) are distributions of partons from AMPT with a string melting scenario and AMPT with enhanced local parton density fluctuation scenario, respectively. Panel (c) and (d) represent distributions of hadrons accordingly.

3 The dynamical quark coalescence model of Ω and ϕ

A dynamical quark coalescence model has been used to study the production of Ω and ϕ [34]. In this model, the probability for producing a hadron from partons is given by the overlap of parton phase space distributions with the parton Wigner function inside the hadron [46]. The multiplicity of a M-parton hadron in an heavy-ion collision is given by

\begin{array}{l} N_{M} = G \int d r_{i_{1}} d q_{i_{1}} \dots d r_{i_{M - 1}} d q_{i_{M - 1}} \\ \times 〈 \sum_{i_{1} > i_{2} > \dots > i_{M}} ρ_{i}^{W} (r_{i_{1}}, q_{i_{1}} \dots r_{i_{M - 1}}, q_{i_{M - 1}}) 〉 . \end{array}

(3)

In Eq. (3), ⟨⟩ represents the event averaging; $r_{i_{1}}$ ,, $r_{i_{M - 1}}$ and $q_{i_{1}}$ ,, $q_{i_{M - 1}}$ are the M-1 relative coordinates and momenta in the M-parton rest frame; $ρ_{i}^{W}$ is the Wigner phase space function inside the hadron, and G is the statistical factor for the M partons.

To determine the quark Wigner phase space functions inside Ω and ϕ, we need quark wave functions. The quark wave functions can be taken as a spherical harmonic oscillator as done in early work [34]. For the ϕ particle, it can be expressed as

ψ (r_{1}, r_{2})=1/(π σ_{ϕ}^{2})^{3/4} exp [- r^{2} /(2 σ_{ϕ}^{2})],

(4)

where r = r₁ - r₂ is the relative coordinate and σϕ is a size parameter of ϕ. Its normalized wave function leads to a root-mean-square (RMS) radius of R_ϕ = (3/8)^1/2σ_ϕ. The quark Wigner function in the ϕ particle can be expressed as

ρ_{ϕ}^{W} (r,k)=8exp (- \frac{r^{2}}{σ_{ϕ}^{2}} - σ_{ϕ}^{2} k^{2}),

(5)

where k = (k₁-k₂)/2 is the relative momentum between s and $\bar{s}$ .

Similarly, for Ω^- and ${\bar{Ω}}^{+}$ particles, their wave function can be described by the following equation

ψ (r_{1}, r_{2}, r_{3})=(3 π^{2} σ_{Ω}^{4})^{- 3/4} exp (- \frac{ρ^{2} + λ^{2}}{2 σ_{Ω}^{2}}) .

(6)

The quark Wigner phase-space function inside the Ω^- and ${\bar{Ω}}^{+}$ baryon can be expressed as

\begin{array}{l} ρ_{Ω}^{W} (ρ, λ, k_{ρ}, k_{λ}) = \\ 64 exp (- \frac{ρ^{2} + λ^{2}}{σ_{Ω}^{2}}) exp [- (k_{ρ}^{2} + k_{λ}^{2}) σ_{Ω}^{2}], \end{array}

(7)

where ρ and λ are relative coordinates of quark, kρ and k_λ are relative momenta, and σΩ is a size parameter that is related to the RMS radius, R_Ω.

The two parameters σϕ and σΩ in the quark Wigner phase space functions inside the ϕ meson and Ω baryon are related to their RMS radii. We take the same values of RMS, R_ϕ = 0.65 fm and R_Ω = 1.2 fm, as used in Ref. [34]. One should note that the quark coalescence model violates energy conservation. The effect is much more ambiguous at low p_T. In the present study, the coalescence model is considered as a perturbative approach, valid only if the numbers of partons coalesced into hadrons is small compared with the total numbers of partons in the system [34]. This condition is satisfied for Ω and ϕ produced in relativistic heavy-ion collisions because their numbers are indeed significantly smaller than that of kaons.

4 Transverse momentum spectra of Ω and ϕ

Using the parton phase space information and the dynamical quark coalescence model, we first study the effect of local parton density fluctuation on the transverse momentum p_T spectra of Ω and ϕ.

Figure 2 presents the transverse momentum distributions of Ω and ϕ at midrapidity (|y| ≤ 1) in Au+Au collisions at $\sqrt{s_{_{N N}}}$ = 11.5 and 7.7 GeV. The local parton density fluctuation causes the trend to increase the production rate of Ω and ϕ in comparison with the two AMPT calculations shown in Fig. 2. The fluctuation scenario describes the experimental data better than the original AMPT case, both with dynamical quark coalescence. We note that the calculations at p_T > 1 GeV/c are below the data, particularly for the Ω. Our result is similar to that shown in Ref. [34], which may be attributed to the deficiency of the AMPT model in treating baryon production. Development on this direction is on the way and better description on proton and Λ productions have been achieved [33]. Nevertheless, one should note that the AMPT model with naive quark coalescence only predicts 20% of Ω yield [30]. The new calculation certainly improves the description of Ω and ϕ production.

Figure 2:

(Color online) The transverse momentum spectra of Ω and ϕ at midrapidity in Au+Au collisions at

\sqrt{s_{_{N N}}}

= 11.5 and 7.7 GeV. The pink circle points are results with enhanced local parton density fluctuation scenarios, while the red square points are results from AMPT with the dynamical quark coalescence model. Black triangle points are experimental data [28].

We further study the ratio of Ω/ϕ as a function of p_T, which is believed to be sensitive to local strange quark fluctuation. The results are presented in Fig. 3. The $N (Ω^{-} + {\bar{Ω}}^{+}) / 2 N (ϕ)$ ratio from string melting AMPT increases slowly as p_T increases. It is close to the experiment data at p_T< 1.0 GeV/c and underestimates the ratio at high p_T significantly. For the result from AMPT with enhanced local parton density fluctuation scenario, the ratio is larger and describes the experimental data well, especially for p_T> 1.5 GeV/c. It seems that the density fluctuation scenario allows a larger possibility for strange quarks or antiquarks to overlap in the coordinate space, as shown in Fig. 1. The production rate on Ω and ϕ thus enhances via quark coalescence. The effect is stronger on Ω than ϕ from the Ω/ϕ ratio analysis.

Figure 3:

(Color online) The ratio of Ω to ϕ in Au+Au collisions at

\sqrt{s_{_{N N}}}

=11.5 and 7.7 GeV. Pink circle points are calculations from AMPT with enhanced local parton density fluctuation while red square points are results from AMPT, both with the dynamical quark coalescence model. The original AMPT with the naive quark coalescence model only predicts 20% of the experimental data [30].

5 Anisotropic flows of Ω and ϕ

In this section, we evaluate the effect of variations in local parton density spacial distributions and discuss its consequence on elliptic flow. The collectivity in high-energy heavy-ion collisions can be measured through final particle azimuthal anisotropy [47]. The anisotropy coefficients are generally obtained from Fourier expansion of final particle azimuthal distortion [48]. i.e.,

E \frac{d^{3} N}{d^{3} p} = \frac{1}{2 π} \frac{d^{2} N}{p_{T} d p_{T} d y} (1 + \sum_{i = 1}^{N} 2 v_{n} \cos [n (φ - ψ_{RP})]) .

(8)

where E is the energy of the final particle, p_T is the transverse momentum, y is the rapidity, φ is the azimuthal angle of the particle and ψ_RP is the reaction plane angle. The Fourier coefficients, vn(n=1,2,...), can be described by the following equation

v_{n} = 〈 \cos (n [φ - ψ_{RP}]) 〉 .

(9)

Similarly, the calculation of harmonic flow, vn, can be relative to the participant plane angle, ψn{PP} [47, 48]. For the study of the local parton density fluctuation effect on anisotropy flow, the participant plane method may be more straight forward [49]. The participant plane angle can be defined by the following equation

ψ_{n} {P P} = \frac{1}{n} [\arctan \frac{〈 r^{2} \sin (n φ) 〉}{〈 r^{2} \cos (n φ) 〉} + π],

(10)

where n is nth-order participant plane, r and φ are the coordinates position and azimuthal angle of partons, and ⟨...⟩ represents parton number density weighting.

Figure 4 shows the v₂ of Ω and ϕ in Au+Au collisions at $\sqrt{s_{_{N N}}}$ = 11.5 and 7.7 GeV. Our results from AMPT with string melting versions and the enhanced local parton density fluctuation scenario are consistent with the experimental data, considering the large statistical uncertainty. Comparing the two sets of AMPT model results, the difference on v₂ is small. It may be because the v₂ is mainly developed in the parton cascade stage in the AMPT model [36]. Another reason may be related to the spacial anisotropy.

Figure 4:

(Color online) The elliptic flow of Ω (upper panel) and ϕ (bottom panel) as a function of p_T in Au+Au collisions at

\sqrt{s_{_{N N}}}

= 11.5 and 7.7 GeV. Red square points are results from AMPT with dynamical quark coalescence, and pink circle points represent AMPT with enhanced local parton density fluctuation scenario. Experimental data are plotted for comparison [50].

The spacial anisotropy is quantified by the participant eccentricity coefficients ε_part[51, 52]:

ε_{2} (part) = \frac{\sqrt{{(σ_{y}^{2} - σ_{x}^{2})}^{2} + 4 σ_{x y}^{2}}}{σ_{y}^{2} + σ_{x}^{2}},

(11)

where

σ_{x}^{2} = {x^{2}} - {x}^{2}

(12)

σ_{y}^{2} = {y^{2}} - {y}^{2}

(13)

σ_{x y} = {x y} - {x} {y},

(14)

and {…} denotes the average over all participants in one event. The results from the AMPT model prior to quark hadronization with different sets of configuration are shown in Fig.5. It is found that the local parton density fluctuation does not have a large impact on eccentricity.

Figure 5:

(Color online) Results of participant eccentricity from the AMPT with and without the enhanced local parton density fluctuation scenario. The upper panel presents the ε₂ distribution and the bottom panel shows the absolute difference |Δε₂| distribution.

6 Summary

In summary, we have studied the local parton density fluctuation effect on Ω and ϕ production in heavy ion collisions based on the parton phase space information from the AMPT model. We have calculated he transverse momentum spectra, the particle ratio, and the elliptic flow of Ω and ϕ in Au+Au collisions at $\sqrt{s_{_{N N}}}$ = 11.5 and 7.7 GeV. For the production of Ω and ϕ, it is found that the local parton density fluctuation increases the yield more than normal AMPT calculation. Our new results improves the description of experimental data. The increase is more visible in the Ω/ϕ ratio vs. p_T. For the v₂ of Ω and ϕ, the results are almost the same between the two scenarios of our calculations may be due to small difference of spatial eccentricity between the two scenarios. Our study implies that the Ω/ϕ ratio could provide valuable insight into possible strong density fluctuations pertaining to the search of critical point or first order phase transition in the future Phase II of the BES experiment at RHIC. We note that there are many other observations that may be sensitive to the strong local parton density fluctuation, such as the high moment analysis of conserve quantum number[9]. The study is on-going and not ready to be published yet.

References

[1]

M. M. Aggarwal et al., arXiv:1007.2613

[2]

Y. Aoki, G. Endrodi, Z. Fodor, et al.,

The order of the quantum chromodynamics transition predicted by the standard model of particle physics

. Nature 443, 675 (2006). doi: 10.1038/nature05120