---
title: Cooling Storage Ring External-Target Experiment
url: https://www.emergentmind.com/topics/cooling-storage-ring-external-target-experiment
type: topic
---

# Cooling Storage Ring External-Target Experiment

Searching arXiv for recent CEE/CSR external-target papers to support the article.
to=arxiv_search.search 彩票总代  天天爱彩票怎么json
{"query":"Cooling Storage Ring External-target Experiment CEE HIRFL Zero Degree Calorimeter beam monitor T0 MWDC", "max_results": 10, "sort_by": "submittedDate"} 
to=arxiv_search.search  ปมถวายสัตย์ฯ ായിjson
{"query":"all:\"CSR external-target experiment\" OR all:\"Cooling Storage Ring External-target Experiment\"", "max_results": 10, "sort_by": "submittedDate"} 
The Cooling Storage Ring External-target Experiment, commonly abbreviated CEE, is a fixed-target heavy-ion experiment at the Heavy Ion Research Facility in Lanzhou that is designed to study dense nuclear matter and the QCD phase diagram in the low-temperature, high-baryon-density region. Across the subsystem studies currently available, CEE is described as using beams with ion species from H to U and energies from about 0.4 to 1.1 GeV/u, within a broader CSR environment that can deliver beams of all stable ion species up to 2.8 GeV/u for protons and 0.5 GeV/u for uranium. Its spectrometer combines a dipole-magnet tracking system, beam timing and beam-position instrumentation, time-of-flight detectors, forward drift chambers, and a zero-degree calorimeter, with a recurring design principle that forward detectors should support centrality and event-plane reconstruction while minimizing autocorrelations with midrapidity observables [2509.10800][2406.12878][2304.14411].

## 1. Scientific programme and accelerator setting

CEE is intended to explore the properties of strongly interacting matter in the low-temperature, high-baryon-density regime, with a particular emphasis on the nuclear equation of state and the density dependence of the symmetry energy above saturation density. In the beam-energy domain discussed for CEE, heavy-ion collisions are expected to generate baryon densities reaching 2–3 times nuclear saturation density and temperatures near 40 MeV, making observables such as directed flow, elliptic flow, and particle production sensitive to pressure gradients, mean fields, and transport dynamics [2503.22774].

The experiment is implemented in fixed-target mode at HIRFL-CSR. The published CEE programme explicitly highlights systems such as \(^{238}\mathrm{U}+^{238}\mathrm{U}\) at 500 MeV/u and \(^{12}\mathrm{C}+^{12}\mathrm{C}\) at 1.1 GeV/u, and more generally describes a research range from a few hundred MeV/u to 1 GeV/u. Several papers frame this physics case in terms of the QCD phase structure at high baryon density, whereas detector-development papers emphasize the need for precise tracking, timing, and forward calorimetry under high-multiplicity fixed-target conditions [2302.11759][2503.22774][2509.10800].

A related accelerator-development strand concerns CSRe stochastic cooling. In that context, the beam extracted from CSRm is injected into CSRe, and the secondary beam produced at the internal target area is characterized by large momentum spread and large transverse emittance. The stochastic cooling system was designed to provide rapid pre-cooling, because electron cooling alone would be too slow for initial conditions such as \(\delta p/p = \pm 0.5\%\) and large emittance. The underlying rate relation was given as
\[
\frac{1}{\tau} = \frac{W}{N}[{2\rm g}(1-{\tilde{M}^{-2})-{\rm g}^{2}({\rm M}+\frac{U}{\rm Z}^{2})],
\]
with the bandwidth
\[
W = \frac{1}{2\times{\rm T}_{pk}\times{\eta}\times\frac{\delta\rm p}{p},
\qquad
\eta = \left|\frac{1}{\gamma}_{t}^{2}-\frac{1}{\gamma}^{2}\right|.
\]
This ring-level infrastructure is not the detector spectrometer itself, but it is part of the broader external-target experimental environment and of the beam-quality constraints motivating the CEE programme [1402.3180].

## 2. Spectrometer architecture

CEE is described as a universal charged-particle spectrometer optimized for the intermediate-energy HIRFL-CSR programme. The published subsystem descriptions consistently list a dipole magnet, beam instrumentation upstream of the target, central and forward tracking, time-of-flight systems, and a forward zero-degree calorimeter. One detector paper specifies a uniform 0.5 T dipole magnet, while another refers to a superconducting dipole magnet; both place the central tracking and TOF subsystems inside the magnet acceptance and the ZDC downstream in the forward region [2503.22774][2302.11759].

| Subsystem | Salient configuration | Primary role |
|---|---|---|
| Beam monitor | Upstream gaseous monitor with two orthogonal micro-TPCs | Per-particle beam tracking and primary-vertex constraint |
| TPC | Double-volume or two-half-TPC configuration | Midrapidity tracking and \(dE/dx\) |
| MWDC | Three forward chambers; prototype uses 6 X/U/V sense layers | Forward tracking |
| TOF | T0, iTOF, eTOF | Start time and PID |
| ZDC | 24 sectors \(\times\) 8 rings, downstream at zero degrees | Centrality and event-plane determination |

The TPC acceptance and the forward tracking system are complementary. CEE Fast Simulation studies quoted for flow analyses state that the experiment covers laboratory polar angles approximately from \(10^\circ\) to \(120^\circ\), corresponding roughly to proton rapidities \(-0.7\) to \(+1\) in the center-of-mass frame for \(\sqrt{s_{NN}} \approx 2.1\) GeV. The two-half TPC design generates azimuthal efficiency modulations at \(\phi \approx 90^\circ\) and \(270^\circ\), whereas the ZDC occupies the forward plane with pseudorapidity coverage \(\eta \approx 1.8\)–4.8 [2503.22774][2302.11759].

The ZDC geometry is central to several CEE analyses. It is installed downstream of the other subsystems at \(Z = 295\)–299 cm and consists of a wheel with inner radius \(R = 5\) cm and outer radius \(R = 100\) cm, segmented into 24 azimuthal sectors and 8 concentric rings. Another subsystem paper describes the same detector as symmetrical and fan-shaped, with a central beam hole and a total of 384 readout channels because each module is coupled to a PMT providing two charge signals via two dynodes [2302.11759][2304.14411].

## 3. Tracking, timing, and beam instrumentation

Upstream beam instrumentation is a distinctive element of CEE. A gaseous beam monitor based on GEM amplification and Topmetal-CEE pixel sensors is being developed to track each beam particle just upstream of the target. The prototype is a \(120 \text{ mm} \times 120 \text{ mm} \times 212 \text{ mm}\) gas vessel containing two micro-TPCs whose drift fields are orthogonal to each other. It operates in Ar(70%) + CO\(_2\)(30%) at room temperature and near local atmospheric pressure, with a drift field of 300 V/cm, \(V_{\mathrm{GEM}} = 350\) V, and an induction field of 1000 V/cm. In heavy-ion beam tests with Kr ions at \(\sim 320\) MeV/u, the per-row spatial resolution was \(47.8 \pm 1.0~\mu\mathrm{m}\) using the center of geometry and \(43.1 \pm 0.9~\mu\mathrm{m}\) using the center of gravity; in laser timing studies, the drift-time resolution increased from \(\sim 9\) ns at \(L = 1.7\) cm to \(\sim 13\) ns at \(L = 4.3\) cm, and the drift velocity was measured as \(0.728 \pm 0.003\) cm/\(\mu\)s [2509.10800].

The beam monitor’s readout architecture is also unusually explicit in the published record. Each micro-TPC uses four Topmetal-CEE chips, each chip measuring \(19 \text{ mm} \times 4 \text{ mm}\) and containing a single row of 180 pixels with a \(100~\mu\mathrm{m}\) pitch along the measured coordinate. Each pixel includes a charge-sensitive amplifier, discriminator, and 8-bit time-over-threshold logic with 25 ns binning, and the data-driven readout runs at 40 MPixels/s. A second-generation Topmetal-CEE chip reduced the minimum operating threshold from \(\sim 20\mathrm{k}\,e^{-}\) to \(\sim 5\mathrm{k}\,e^{-}\) and the shaping time from \(\sim 1~\mu\)s to \(\sim 0.5~\mu\)s, while the temporal noise remained \(\sim 350\,e^{-}\) and the input dynamic range remained \(>100\mathrm{k}\,e^{-}\) [2509.10800].

Forward charged-particle tracking is provided by MWDCs. A half-size prototype tested in 350 MeV/u Kr+Fe reactions consisted of 6 sense layers with X, U, and V wire orientations at \(0^\circ\), \(30^\circ\), and \(-30^\circ\), respectively, and a sensitive area of \(76 \text{ cm} \times 76 \text{ cm}\). Operated with Ar/CO\(_2\) (80/20) at slightly above atmospheric pressure and with 1500 V high voltage on the anode wires, the efficiency for each layer was beyond 95%, and the tracking residual for 6-layer tracks was \(301 \pm 2~\mu\mathrm{m}\). The calibrated drift velocity at the center of the cell was \(v_d \approx 4.5\) cm/\(\mu\)s [2406.12878].

The start-time system has been studied in two published detector lines. One paper reports an MRPC-based T0 detector surrounding the target at about 10 cm radius, with an intrinsic single-module time resolution of \(\sim 40\)–50 ps and a T0 resolution of \(\sim 60\) ps within one group using two tracks, while a two-group comparison yielded \(\sim 100\) ps because of vertex smearing [1909.11407]. A later T0 paper describes a thin EJ-200 plastic-scintillator detector with 32 Hamamatsu SiPMs, installed approximately 30 cm upstream of the fixed target, and reports beam-test timing better than 30 ps after T–TOT correction and iterative weighted averaging, satisfying the stated requirement that the T0 timing jitter be no more than about 30 ps for an overall TOF resolution better than 50 ps [2304.02944].

## 4. Forward calorimetry, centrality, and event-plane reconstruction

The ZDC is one of the defining detectors of CEE. It is tasked with measuring forward spectator fragments and thereby providing centrality and reaction-plane information. At intermediate energies, semi-central and peripheral collisions produce sizable spectator fragments at small angles relative to the beam, so the forward energy and hit topology in the ZDC become highly informative. In the geometric picture used by the centrality paper, the centrality percentile is
\[
c(b) = \frac{\int_0^b P(b')\,db'}{\int_0^{b_{\max}} P(b')\,db'},
\qquad P(b)\propto b,
\]
and the approximate ZDC–spectator relation is summarized as \(E_{\mathrm{ZDC}} \approx \alpha N_{\mathrm{spec}}\), with the important caveat that the beam hole and finite segmentation weaken the monotonicity of simple one-dimensional observables [2304.14411].

A recurrent misconception is that total ZDC energy or the number of fired channels should be sufficient as a centrality estimator. The CEE simulation study explicitly argues otherwise: because of the beam hole and acceptance effects, neither “total ZDC energy” nor “number of fired channels” alone yields a strong monotonic mapping to \(b\). The adopted solution is a multivariate classifier based on Extreme Gradient Boosting. Using IQMD \(^{238}\mathrm{U}+^{238}\mathrm{U}\) collisions at 500 MeV/u, transported through GEANT4, the model is trained on three impact-parameter classes—central \(0 \le b \le 3\) fm, semi-central \(3 < b \le 7\) fm, and peripheral \(7 < b \le 10\) fm—using full-ZDC fired-channel counts, full-ZDC deposited energy, and ring-resolved deposited energies. The resulting three-class classifier achieved an average test-set AUC of \(\approx 0.96\); at 90% purity, the efficiencies were 67% for central, 66% for semi-central, and 97% for peripheral events, with only minor sensitivity to variations in scintillator thickness, hit efficiency, Gaussian energy smearing, and IQMD de-excitation [2304.14411].

Event-plane reconstruction with the ZDC is based on the first-harmonic flow vector built from forward hits:
\[
\vec{Q}_{1}=
\left(
\begin{array}{c}
\sum_i w_i \sin(\phi_i) \\
\sum_i w_i \cos(\phi_i)
\end{array}
\right),
\qquad
\Psi_{1}=\tan^{-1}\left(
\frac{\sum_i w_i \sin(\phi_i)}
{\sum_i w_i \cos(\phi_i)}
\right).
\]
The deposited energy \(\Delta E\) is used as the base weight, but the field-induced left–right acceptance asymmetry requires an additional position weight \(P(x,y,\Delta E)\), followed by a Fourier-shift flattening procedure applied in centrality bins. In IQMD \(U+U\) at 500 MeV/u with GEANT4 detector simulation, the first-order event-plane resolution from the full ZDC reaches \(\approx 90\%\) in mid-central collisions \((4 < b < 7~\mathrm{fm})\) after these corrections [2302.11759].

The event-plane methodology has also been validated in a separate JAM + CEE Fast Simulation study of proton directed flow. There, MWDC-based event planes and ZDC-based event planes are reconstructed with re-centering, flattening, two-subevent resolution estimates, and explicit non-flow mitigation. In that study, the first-order resolution peaks at mid-central collisions \((5 < b < 7~\mathrm{fm})\), reaching about 90% for MWDC and 70% for ZDC. After correction, the midrapidity proton directed-flow slopes are reported as
\(0.631 \pm 0.002\) with respect to \(\Psi_{\mathrm{RP}}\),
\(0.629 \pm 0.003\) with respect to the MWDC event plane,
and \(0.634 \pm 0.012\) with respect to the ZDC event plane [2503.22774].

## 5. Triggering, simulation frameworks, and analysis methodology

CEE analyses are strongly simulation-driven at the current stage, and the published work uses several distinct frameworks. IQMD plus GEANT4 is used for ZDC centrality and event-plane performance studies, JAM plus CEE Fast Simulation is used for directed-flow validation, UrQMD 3.4 plus GEANT4 is used in T0 timing studies, and FLUKA with the PRECISION default option set is used for beam-monitor radiation simulations [2304.14411][2503.22774][1909.11407][2509.11148].

For flow analyses, the event-plane procedure follows the standard chain of \(Q_n\)-vector construction, re-centering, flattening, subevent resolution determination, and resolution correction,
\[
v_n = \frac{\langle \cos[n(\phi-\Psi_n)] \rangle}{R_n}.
\]
The directed-flow study emphasizes two technical points. First, self-correlations are removed in MWDC-based event-plane analyses by excluding the particle of interest from the event-plane construction and correcting momentum-conservation effects when MWDC/TPC matching is available. Second, when correlating TPC protons with the ZDC event plane, the proposed mitigation against autocorrelation is to restrict the particle of interest to backward rapidity, \(-0.5 < y < 0\), while the broader optimized region for proton \(v_1\) is \(-0.5 < y < 0.5\) and \(0.2 < p_T < 0.7\) GeV/\(c\) [2503.22774].

The trigger system for the external target experiment at CSR was designed as a three-level master–slave hierarchy integrated into PXI 6U crates distributed over distances up to about 100 m. Level-1 preprocessing resides in front-end measurement modules, Level-2 sub-trigger formation is handled by Slave Trigger Modules, and Level-3 global decision logic resides in a Master Trigger Module. The MTM aggregates data from up to 16 STMs using fiber links implemented with Xilinx 7-series GTP transceivers and 8b/10b coding at 800 Mbps, synchronized to a global 40 MHz clock and aligned with the K28.5 comma sequence. Laboratory validation reported zero errors over \(6.912\times10^{13}\) transmitted bits in each of four directions, corresponding to a measured upper bound \(\mathrm{BER} < 1\times10^{-13}\) [1512.05337].

The global trigger combines a programmable multiplicity condition with a programmable topological coincidence among detector groups. This architecture matters for CEE because the detector systems are geographically distributed and because the experiment relies on flexible online configuration for different physics programmes. The same technical emphasis on modularity also appears at the detector level, where subsystem papers repeatedly describe calibration workflows based on T–TOT corrections, iterative drift-time calibrations, position-weight maps, and detector-response variations rather than on idealized responses alone [1512.05337][2304.02944][2406.12878].

## 6. Performance envelope, limitations, and development trajectory

The published subsystem record gives a relatively coherent performance envelope for CEE. The beam monitor prototype exceeds its design goals of spatial resolution \(< 50~\mu\mathrm{m}\) and time resolution \(< 1~\mu\)s, reaching better than \(50~\mu\mathrm{m}\) spatial resolution and better than 15 ns timing resolution. The MWDC prototype reaches per-layer efficiency beyond 95% and a tracking residual of \(301 \pm 2~\mu\mathrm{m}\). The T0 detector studies report either intrinsic MRPC timing of \(\sim 40\)–50 ps with a \(\sim 60\) ps group-level T0 or a scintillator–SiPM start detector with timing better than 30 ps. The ZDC studies show both high event-plane resolution in simulation and robust ZDC-only centrality classification using multivariate methods [2509.10800][2406.12878][1909.11407][2304.02944][2304.14411].

Several limitations are stated explicitly. Much of the current centrality and event-plane work is simulation-based, so real-data domain shift, pile-up, channel nonuniformities, and calibration transfer remain open issues. The centrality classifier currently uses only three classes, and the paper notes that future work may incorporate regression tasks and explore other ML algorithms. The beam-monitor paper does not quantify aggregate throughput, sustainable hit rates, or latency for the multi-chip system, and the radiation-environment paper excludes front-end cards, supports, and cables from the FLUKA geometry [2304.14411][2509.10800][2509.11148].

Radiation tolerance is a particularly concrete development issue for the upstream beam monitor. For a six-month total beam time at 1 MHz, the FLUKA study reports a maximum integrated TID of \(\approx 10.6\) kGy and a maximum integrated \(1~\mathrm{MeV}\) neutron-equivalent fluence of \(\approx 2.2\times10^{11}~\mathrm{cm}^{-2}\) at the chip planes for the 500 MeV/u U benchmark. The same study concludes that thinner field-cage windows and the use of two GEM layers are the most effective geometry modifications for lowering chip-plane TID and NIEL, whereas internal lead shielding is not cost-effective for those damage channels [2509.11148].

The broader trajectory is therefore one of detector commissioning, calibration refinement, and progressive replacement of idealized assumptions by experimentally constrained ones. Prototype studies already exist for the T0 detector, beam monitor, MWDC, ZDC analysis chain, and distributed trigger. The literature also points to next steps: second-generation Topmetal-CEE chips and double-stage GEMs are under development, thinner beam-monitor windows are favored by radiation simulations, and centrality/event-plane studies anticipate retraining and validation for other systems and beam energies. This suggests a maturing fixed-target heavy-ion programme whose forward detectors, timing systems, and beam instrumentation are being optimized jointly rather than as isolated subsystems [2509.10800][2509.11148][2304.14411].

Source: https://www.emergentmind.com/topics/cooling-storage-ring-external-target-experiment