Sequential Strangeness Freeze-out

Rene Bellwied1,*
1
University of Houston, Physics Department, 617 SR1 Bldg., Houston, TX 77204, USA

Abstract. I will describe the latest results from lattice QCD pertaining to
a potential flavour hierarchy in the hadronic freeze-out from the QCD
crossover region. I will compare these results to a variety of improved
hadronic resonance gas calculations and to experimental data of
fluctuations of net-charge, net-proton and net-kaon multiplicity
distributions, which serve as a proxy for the susceptibilities of conserved
quantum numbers on the lattice. I will conclude that there is intriguing
evidence for a flavour dependent freeze-out, and I will suggest expansions
to the experimental program at RHIC and the LHC that could potentially
demonstrate the impact of a flavour separation during hadronization.

1 Introduction
The detailed determination of a pseudo-critical temperature based on continuum
extrapolations of the temperature dependence of the chiral susceptibility on the lattice, in
comparison to calculations of the chemical freeze-out temperature using particle yields at
RHIC and the LHC, seems to indicate that hadronization and freeze-out coincide near the
phase boundary in the QCD phase diagram. The question arises whether this transition from
quark to hadron degrees of freedom occurs at the same temperature for all particle species
and/or quark flavours. The application of statistical hadronization models is successful in
describing hadronic particle yields over many orders of magnitude. From the abundant pion
yields to the rare alpha particle, the thermally equilibrated system can be broadly defined
with two common freeze-out parameters, namely the chemical freeze-out temperature and
the baryo-chemical potential. These calculations were applied over a wide range of
collision energies from the SPS to the LHC, and system sizes from pPb to PbPb. However,
recent high resolution measurements of the particle yields in ALICE at the LHC and STAR
at RHIC, as well as the net-particle fluctuations in STAR, seem to indicate that there might
be evidence for a sub-structure in the common freeze-out picture. The early LHC
measurements attributed the tension in a common fit to a proton anomaly, because, at the
time, only the proton yields seemed to deviate from the anticipated particle yields. As of
late, though, i.e. with the Run-2 data from ALICE, also the multi-strange baryons show a
significant deviation from the common temperature fit. These results can be related to
attempts to deduce the freeze-out temperatures of particular conserved quantum numbers
independently from fluctuations of net-particle distributions and lattice QCD.

*
Corresponding author: [email protected]
2 Susceptibilities from lattice QCD

Susceptibilities are defined as the derivatives of the pressure with respect to the chemical
potential. Continuum extrapolated susceptibility calculations of single flavour quantum
numbers showed that there is a difference between flavours in the crossover region [1], see
Fig.1(left). Fig.1(right) shows the flavour specific susceptibility ratio 4/2 [2], which was
suggested as a specific observable to deduce chemical freeze-out temperatures directly [3],
from a comparison of experimental data to first principle calculations. The lattice data
themselves show a peak at different temperatures and their agreement with Hadron
Resonance Gas (HRG) model calculations begins to deviate at these temperatures as well.
This is, without a direct comparison to experimental data, not yet proof of a flavour
hierarchy in the crossover region, but it is suggestive of different freeze-out temperatures
for light and strange flavour particles.

Fig. 1. (left): Continuum extrapolated lattice QCD results for 2u and 2s [1], (right): Continuum
extrapolated lattice QCD results for 4/2 for light and strange quarks in comparison to HRG model
calculations [2].

3 Comparison to the Hadron Resonance Gas Model

The relevance of the hadronic spectrum included in the HRG calculations was
highlighted early on by Bazavov et al. [4] in order to explain a potential deviation from the
lattice curves at different temperatures for different flavours. It was shown that the
inclusion of certain states significantly improves the agreement between HRG and lattice
QCD for certain susceptibilities in the QCD crossover region. The initial study simply
expanded the list of available states by expectations from the non-relativistic Quark Model
[5]. More refined Quark Model variations, which include the possibility of quark-quark
interactions in the hadron, significantly reduce the number of expected states, while at the
same time the number of experimentally verified states in the listing of the particle data
group, e.g. PDG-2016 [6], continues to increase. A detailed study of the HRG approach
including the latest findings, was presented at this conference and in a recent publication [7,
8] and indicated that a compromise between too many and too few excited states in the high
mass hadronic spectrum can be found, based on all states listed in PDG-2016. With this
input the HRG calculation can describe most features of a multitude of susceptibility ratios
calculated with lattice QCD up to the pseudo-critical temperature(s). As an example, Fig.2
shows the two most relevant susceptibility ratios that can be measured to determine flavour
dependencies, namely the (s/B)LO and the 4/2 for strange quarks [8]. It is evident that the
so-called PDG-2016+ list gives the best compromise between old PDG and QM listings.

Fig. 2. Continuum extrapolated lattice QCD results for (s/B)LO (left) and for 4s/2s (right) in
comparison to HRG model calculations with varying number of resonant states based on PDG-2012,
PDG-2016+ (i.e. incl. one star states), and non-relativistic Quark Model predictions [8].

A more provocative extension of the standard HRG approach was also presented at this
conference and in a recent publication, expanding on the question whether a non-interacting
resonance gas is indeed the most realistic proxy for the hadronic interaction strength near
the phase transition [9]. As an alternative the authors proposed to either use a
parametrization of the low energy van-der-Waals interactions or an excluded volume to
describe the attractive and repulsive hadron-hadron interactions. These changes to the HRG
approach seemingly extend the agreement between HRG and lattice QCD to higher
temperatures, but a.) the agreement now reaches beyond the pseudo-critical temperature
and b.) the rather unconstrained parameter base for the interactions and the excluded
volume allows for rather large variations in the fit. Nevertheless the approach is intriguing
and, in the future, could also take into account potential flavour dependent differences in
the parameters.

4 Fits based on experimental studies of fluctuations and yields

Recent results published by the STAR collaboration [10] regarding the evolution of the
chemical freeze-out temperature for the various energies of the RHIC beam energy scan
showed that although a common freeze-out temperature can be found through fits to the
measured yields of all particles, this temperature will be about 15-20 MeV lower, if only
light flavour particles (and kaons) are included in the fit (see Fig. 35 in [10]). The kaon
yield is included, but was shown to be rather insensitive to the freeze-out temperature [11].
Independently the work by Chatterjee et al. [12] showed that the fits to ALICE yields
significantly improve if two separate chemical freeze-out temperatures are assumed for
light and strange particles. This requirement for two temperatures seemingly disappears
when the yields in small systems (pp and pPb collisions) are fitted [13].
Regarding the use of net-particle fluctuations rather than the yields, our studies that
tried to determine net-electric charge and net-baryon number fluctuations, by using the net-
charged particle and net-proton distributions as a proxy, found a chemical freeze-out
temperature for these particles, based on a HRG fit, that was consistently 15-20 MeV below
the expected value from a common fit to all measured particle yields, see Fig.3(left) [14].
The resulting temperatures are in agreement with lattice QCD calculations based on the
related susceptibility ratios [15,16]. One should note that all HRG fits are applied to the
2/1 ratios for a particular particle species. We found that the error bars for the higher
moments are still too large and that the higher moments might potentially have
contributions from critical fluctuations at the RHIC beam energies, which will negatively
impact the chemical freeze-out fits [14]. In our conclusions, we pointed out that both, the
net-charged particles and the net-protons, are dominated by light quark particles (pions and
protons, respectively), and that a fit to a strangeness proxy, i.e. the fluctuations of net-
kaons, might shed light on the question whether the chemical freeze-out could indeed be
flavour dependent. In Fig.3(right) I am presenting preliminary results of a HRG fit to the
recently published net-kaon data from STAR [17]. The HRG model is the same used in
Fig.3(left), in fact the baryo-chemical potential was adopted from the fit to net-protons and
net-charges and only the temperature was left as a free parameters. As one can see the net-
kaons tend towards higher freeze-out temperatures than the other net-particle distributions.
This might be surprising because as mentioned before, the kaon yields are rather insensitive
to the temperature in a statistical hadronization fit, but our group also showed that the
higher order moments of the net-kaon distributions are more sensitive to the freeze-out
parameters than the yields themselves [18]. In Fig.3(right) we also show that the final result
is rather insensitive to the number of hadronic states included in the HRG calculation.
Lattice calculations assuming a Boltzmann approach to the partial pressure in order to
isolate the contribution of the kaons to the strangeness susceptibilities confirm this result
[19].

Fig. 3. (left): Results from a combined HRG fit to the 2/1 measurements for net-charges and
net-protons from STAR (blue points) [14]. The extracted freeze-out parameters are compared to the
curve by Cleymans et al. [20], which tried to parametrize all freeze-out results from statistical
hadronization models to yields from SIS, SPS, RHIC and LHC. (right): Preliminary results from
fitting the 2/1 measurements for net-kaons from STAR with the same model used in Fig.3(left). The
B was fixed by the net-p,Q results. The figure also shows a comparison between results using
different PDG lists in the HRG calculation.

One should note that the usage of net-kaons as a proxy for net-strangeness is less justified
than the proxies for net-charge and net-baryon number [21]. Certainly the inclusion of
fluctuation data for strange baryons is very important in order to gauge the relative
contributions of flavour and baryon number to the final result. Studies of these fluctuations
are underway in STAR and ALICE.
3 Conclusions and Outlook
There is intriguing evidence that the flavour composition of the produced hadrons might
play a role in their freeze-out parameters and thus in their hadronization dynamics.
Particularly it seems that the quark mass plays a significant role in calculating the transition
of flavour specific susceptibilities on the lattice as long as the mass is not negligible
compared to the temperature of the equilibrated system. Experimentally this could mean
that heavier quark particles prefer to freeze-out at a higher temperature. I have presented
evidence to that effect based on yield and fluctuation measurements from STAR and
ALICE. Studies of the baryon mass evolution in the crossover region, based on PNJL [22]
and lattice calculations [23], also point at finite quark mass, and thus flavour, dependencies.
Certainly more evidence is needed, in particular from strange baryon fluctuation
measurements, but in theory this flavour dependence should also be measurable in the
charm sector as long as the charm quarks thermalize with the system.
A direct impact of a higher freeze-out temperature would be an enhancement of strange
particles relative to the yield obtained at a common lower freeze-out temperature. Indirect
evidence can be found in the ALICE data based on the latest yields in central PbPb
collisions and on the comparison of PbPb yields to pp yields for strange particles. The latter
point is often attributed to canonical suppression in the small system [24], but a comparison
between preliminary RHIC-BES data [25] and ALICE data [26] leads us to believe that the
energy dependence of the canonical suppression makes this effect almost negligible at LHC
energies, see Fig.4. Based on the data it seems that the strange anti-baryon yields follow the
same trend from sqrt(s) = 62 GeV on up. In addition, HRG fits, assuming a thermal system
is produced in pp collisions, show that the temperature in the small systems is about 15
MeV lower than the strange particle freeze-out temperature in heavy ion collisions [12],
which could explain at least part of the strangeness enhancement. Further studies are
needed, though.

Fig. 4. Anti-baryon over p- production as a function of collision energy and charged particle density
(for LHC: <dNch/d> in ||<0.5, for RHIC: <dNch/dy> in |y|<0.1) based on STAR [25] and ALICE
[26] data.

Regarding a more speculative aspect of particle production, this enhancement in strange

quarks could lead to strangeness clustering, which might manifest itself in strange multi-
quark states. The discovery of charmed tetra-and penta-quark states by LHCb certainly has
triggered renewed interest in exotica searches in the strangeness sector.
In terms of more dynamic quantities, a higher freeze-out temperature could potentially lead
to a shortened partonic phase for strange baryons. This could reduce dynamic quantities,
such as RAA and v2, as long as a significant contribution to the flow or the suppression is
generated close to pseudo-critical temperature.
Achnowledgements
I would like to thank my collaborators for various contributions to this work: Paolo Alba,
Livio Bianchi, Szabolcs Borsanyi, Zoltan Fodor, Jana Guenther, Anders, Knospe, Valentina
Mantovani-Sarti, Jackie Noronha-Hostler, Paolo Parotto, Attila Pasztor, and Claudia Ratti. I
also thank Kryzstof Redlich and Frithjof Karsch for some inspiring discussions. This work
was supported by the U.S. Department of Energy under DE-FG02-07ER41521.

References
1. C. Ratti et al., Phys. Rev. D 85, 014004 (2012)
2. R. Bellwied et al., Phys. Rev. Lett. 111, 202302 (2013)
3. F. Karsch, Central Eur. J. Phys. 10, 1234 (2012)
4. A. Bazavov et al, Phys. Rev. Lett. 113, 072001 (2014)
5. S. Capstick and N. Isgur, Phys. Rev. D 34, 2809 (1986)
6. C. Patrignani et al. (Particle Data Group), Chin. Phys. C 40, 10001 (2016)
7. C. Ratti, contribution to this conference (2017)
8. P. Alba et al., Phys. Rev. D 96, 034517 (2017)
9. V. Vovchenko et al., Phys. Rev. Lett. 118, 182301 (2017)
10. L. Adamczyk et al., Phys. Rev. C 96, 044904 (2017)
11. D. Magestro, J. Phys. G 28, 1745 (2002)
12. S. Chatterjee, A.K. Dash and B. Mohanty, J. Phys. G 44, 105106 (2017)
13. S. Chatterjee, contribution to this conference (2017)
14. P. Alba et al., Phys. Lett. B 738, 305 (2014)
15. S. Borsanyi et al., Phys. Rev. Lett. 113, 052301 (2014)
16. F. Karsch, Nucl. Phys. A 956, 352 (2016)
17. L. Adamczyk et al. (STAR Collaboration), arXiv:1709.00773
18. P. Alba et al., Phys. Rev. C 92, 064910 (2015)
19. J. Noronha-Hostler et al., arXiv :1607.02527
20. J. Cleymans, H. Oschler, K. Redlich, S. Wheaton, Phys. Rev. C 73, 034905 (2006)
21. C. Zhou, J. Xu, X. Luo, F. Liu, Phys. Rev. C 96, 014909 (2017)
22. J.M. Torres-Rincon, contribution to this conference (2017)
23. G. Aarts, contribution to this conference (2017) and arXiv:1710.08294
24. K. Redlich and A. Tounsi, Eur. Phys. J. C 24, 589 (2002)
25. X. Zhu for the STAR Collaboration, PoS (CPOD 2014), 008 (2015)
26. J. Adam et al. (ALICE Collaboration), Nature Phys. 13, 535 (2017)