Search for the QCD Critical Point in High Energy Nuclear Collisions: A Status Report Yu Zhang, Zhaohui Wang, Xiaofeng Luo, Nu Xu Abstract We review recent results of net-proton multiplicity fluctuations from STAR experiment, aiming to locate the QCD critical point in high-energy nuclear collisions at RHIC. We show net-proton number cumulant and proton number factorial cumulant ratios up to fourth order using experimental data from RHIC BES-II Au+Au collisions in collider mode and fixed-target mode. The comparison is made between experimental data and non-critical model calculations from Lattice QCD, HRG, hydrodynamic simulations and transport model UrQMD. In addition, we discuss initial volume fluctuation effect, which plays significant role in fixed-target energies. Finally, an outlook on experimental research on the QCD critical point in future experiments will be presented. 1 Introduction One of main goals in heavy-ion collisions is to explore the phase structure of strongly interacting matter. In the conjectured QCD phase diagram, plotted in terms of temperature \rm T and baryon chemical potential \mu_{\rm B} , lattice QCD calculations from first principles predict a smooth crossover transition between the quark–gluon plasma (QGP) and hadronic phase at vanishing \mu_{\rm B} 15 . However, at large \mu_{\rm B} , lattice QCD encounters the fermion sign problem and cannot provide definitive conclusions about the nature of the QCD phase transition. Figure 1: Theoretical predictions on the QCD critical point from various approaches. Many theoretical calculations predict that, at high baryon chemical potential, the transition between the hadronic phase and the QGP phase becomes first-order, terminating at a second-order phase transition point commonly referred to as the QCD critical point. On the experimental side, extensive measurements of various observables have been carried out in an effort to locate this critical point. Among these, higher-order cumulants of conserved quantities, such as net baryon number, net electric charge, and net strangeness, have been shown to be particularly sensitive probes of critical fluctuations 59 ; 16 ; 60 ; 24 ; 47 ; 53 ; 23 . A key signature of the QCD critical point is the non-monotonic energy dependence of the fourth-order cumulant ratio ( \kappa\sigma^{2} ) 59 . Such measurements have been performed by the STAR Collaboration at the Relativistic Heavy Ion Collider (RHIC) 8 ; 13 ; 10 ; 9 ; 2 ; 3 ; 7 ; 1 ; 4 . While fluctuations of conserved quantities are sensitive to critical phenomena, their interpretation in heavy-ion collisions requires careful consideration of finite-size effects, critical slowing down, and non-equilibrium dynamics such as baryon transport. These effects are not merely additive to equilibrium baselines but can fundamentally alter fluctuation observables. Therefore, deviations from model calculations should be interpreted with caution, and alternative approaches incorporating finite-size scaling 30 ; 57 ; 58 and dynamical modeling are also essential in the search for the QCD critical point. Recently, theoretical studies 32 ; 44 ; 25 ; 36 ; 66 ; 17 ; 28 ; 62 ; 42 ; 55 employing diverse approaches have converged on a critical point located in the high, \mu_{\rm B} region (see Fig. 1 ), a domain accessible to the RHIC Beam Energy Scan program 11 ; 27 and the STAR fixed-target experiment 50 . This review is organized as follows. In Sec. 2 , we briefly introduce the RHIC-STAR experiment and details of higher-order cumulant measurements thereon. Recent progress on (net-)proton multiplicity fluctuation analyses from STAR is presented in Sec. 3 , and we conclude with a summary in Sec. 4 . 2 Experiment and Data Analysis During the Beam Energy Scan (BES) program at RHIC from 2010 to 2021, the STAR detector collected a set of gold on gold nucleus collision data across a center-of-mass energy range of \sqrt{s_{\rm NN}}=3-200 GeV, operating at collider mode and fixed-target mode (FXT). The Au+Au collision data span a broad range of \mu_{\rm B} \approx 25-750 MeV at chemical freeze-out 11 . The measurements emphasized in this review utilize data from the second phase of the BES program (BES-II), which includes collider-mode data at \sqrt{s_{\rm NN}}=7.7 , 9.2, 11.5, 14.6, 17.3, 19.6, and 27 GeV, as well as fixed-target data at \sqrt{s_{\rm NN}}=3.0 , 3.2, 3.5, and 3.9 GeV. The observables used to characterize (net-)proton multiplicity fluctuations are n -th order cumulants ( C_{n} ) and factorial cumulants ( \kappa_{n} ) in Eq. ( 1 ), where N denotes (net-)proton number. Proton number is used as a proxy for baryon number as electrically neutral particles, mainly neutrons in baryon fluctuation analysis, cannot be detected by STAR detector. We present the direct measurements of net-proton fluctuations in this review. As claimed in Refs. 41 ; 40 , over a wide energy range, isospins of nucleons are randomized and uncorrelated in the final state due to reactions of nucleons with pions in the hadronic stage, which leads to a factorization of proton and neutron probability distributions. In this case, net-baryon cumulants can be obtained in terms of net-proton cumulants. The deviation of N from its mean value \langle N\rangle is defined as \delta N=N-\langle N\rangle . Ratios of cumulants ( C_{n} ) or factorial cumulants ( \kappa_{n} ) of different orders are constructed in order to cancel volume dependence and to enable direct comparison with susceptibility ratios from theoretical calculations 34 ; 35 ; 38 ; 33 . \displaystyle\begin{aligned} C_{1}&=\langle N\rangle,&\kappa_{1}&=C_{1},\\ C_{2}&=\langle(\delta N)^{2}\rangle,&\kappa_{2}&=-C_{1}+C_{2},\\ C_{3}&=\langle(\delta N)^{3}\rangle,&\kappa_{3}&=2C_{1}-3C_{2}+C_{3},\\ C_{4}&=\langle(\delta N)^{4}\rangle-3\langle(\delta N)^{2}\rangle^{2},&\kappa_{4}&=-6C_{1}+11C_{2}-6C_{3}+C_{4}.\end{aligned} (1) The event-by-event (net-)proton multiplicity distributions are constructed by identifying (anti-)protons using STAR time projection chamber (TPC) 14 and time-of-flight (TOF) 43 detectors. Combining trajectory information from the TPC with flight-time measurements from the TOF enables proton identification with a purity exceeding 99%. Collision centrality is determined indirectly using the multiplicity distribution of charged hadrons excluding (anti-)protons. These reference multiplicity distributions are then compared with Monte Carlo Glauber simulations 51 and classified into centrality bins such as 0–5%, 5–10%, …, 70–80%. Thanks to the upgrade of inner time projection chamber (iTPC) 29 to STAR TPC, the reference multiplicity distributions in BES-II( \sqrt{s_{\rm NN}}=7.7- 19.6 GeV) is measured over an extended pseudo-rapidity acceptance |\eta|<1.6 , compared to |\eta|<1.0 in BES-I. Two other major upgrades, namely end-cap time of flight (eTOF) 54 and event plane detector (EPD) 12 have further enhanced STAR’s capabilities. Particularly, the eTOF detector upgrade plays a crucial role in particle identification for the STAR fixed-target experiment. Cumulants of (net-)proton multiplicity distributions are calculated in fine reference multiplicity intervals and then averaged over broader centrality classes (e.g. 0-5%) following the centrality bin width correction procedure shown in Ref. 46 . Analytical corrections for detector efficiency loss 52 ; 45 are applied under the assumption of a binomial detector response 10 ; 48 . Additional experimental factors, such as track selection criteria, are considered in systematical uncertainty evaluation. 3 Recent results 3.1 (Net-)proton multiplicity fluctuations from STAR Figure 2 shows collision energy dependence of net-proton cumulant ratios C_{2}/\langle p+\bar{p}\rangle , C_{3}/C_{1} , and C_{4}/C_{2} and proton factorial cumulant ratios \kappa_{2}/\kappa_{1} , \kappa_{3}/\kappa_{1} , and \kappa_{4}/\kappa_{1} in Au+Au collisions from BES-II ( \sqrt{s_{\rm NN}}=7.7-27 GeV) 6 and BES-I ( \sqrt{s_{\rm NN}}=39-200 GeV) 9 ; 2 at RHIC. Measurements are performed in 0-5% most central (red filled squares) and 70-80% peripheral collisions (black open diamonds) with the rapidity window |y|<0.5 in the center-of-mass frame and transverse momentum range 0.440 GeV, the decreasing trends are reproduced by model calculations including Lattice QCD, HRG and hadronic transport model UrQMD. While when it goes below around 11 GeV, all ratios show clear increase trends especially for C_{2}/\langle p+\bar{p}\rangle and C_{3}/C_{1} , which could not be reproduced by theoretical calculations. To quantify the deviations of net-proton C_{4}/C_{2} of data from non-critical baselines, significance (shown in Fig. 3 ) are calculated by (data - reference) divided by total uncertainties. The non-critical references include UrQMD, HRG CE, Hydro EV, and data from 70-80% peripheral collisions. From Fig. 3 , a maximum deviation at around 20 GeV is observed from all references with significance levels of 2 to 5 \sigma , which is a characteristic feature of proposed signature of the QCD critical point predicted in Ref. 60 . At lower collision energy, on the other hand, changes are also see in proton factorial cumulant ratios, see panels (d), (e), and (f) of Fig. 2 . The values of both \kappa_{2}/\kappa_{1} and C_{2}/\langle p+\bar{p}\rangle are monotonically decreasing as collision energy decreases from 200 GeV to about 10 GeV primarily due to the baryon number conservation in the central Au+Au collisions. Below 10 GeV, however, the decreasing trend turns to increase in the lower energy collisions, i.e. , high baryon density region. The change in coverture in the energy dependence is also seen in peripheral collisions, open diamonds in Fig. 2 , although it appears not as strong as in the central collisions. While these decreasing energy dependence are reproduced by most of the model calculations without criticality, these calculations all failed to reproduce the increase in the ratios at lower energies. The change in the energy dependence may indicate the nature of interactions, namely, strongly repulsive interactions at the high energy end while attractive interactions become necessary at the high baryon density region 31 . Note that in case of Van der Waals interaction, both attractive and repulsive interactions are required in order to form the critical point. Figure 4: Collision energy dependence of ratios for the cumulants of net-proton (top panels) and the factorial cumulant of protons (lower panels) in Au+Au collisions from STAR at RHIC. The red squares in the energy range 3<\sqrt{s_{\rm NN}}<3.9 GeV are the new results shown at Quark Matter 2025. Recent measurements on STAR fixed-target datasets in Fig. 4 are shown at Quark Matter 2025 conference 64 , which extend the energy dependence of proton cumulant and factorial cumulant ratios down to 3 GeV. In Fig. 4 the new measurements in 0-5% most central Au+Au collisions at \sqrt{s_{\rm NN}}=3.0-3.9 GeV are shown with red squares while results from collider data are shown with blue circles. Those measurements are performed with the rapidity window -0.5