DOI 10.1393/ncc/i2019-19248-9 Colloquia: La Thuile 2019
Heavy ion and fixed-target physics at ATLAS, CMS and LHCb
G. Grazianion behalf of the ATLAS, CMS and LHCb Collaborations
INFN, Sezione di Firenze - Firenze, Italy
received 10 October 2019
Summary. — Collisions with ion and proton beams at the LHC provide a testbed
to study strongly interacting matter at the most extreme density and temperature presently achievable at accelerators. The ATLAS, CMS and LHCb experiments, though not designed specifically for heavy ion physics, are providing unique inputs to this rich phenomenology. Recent results from the three experiments are reviewed, covering in particular measurements with heavy flavour and electromagnetic probes, and the first results from fixed-target collisions obtained by LHCb.
1. – Introduction
Collision data using lead beams have been collected by all the four major LHC experi-ments, as summarized in fig. 1. Using their general-purpose detectors, fully instrumented at mid rapidities, ATLAS and CMS can reconstruct even the most central PbPb colli-sions. The ability to acquire data at the maximum collision rate provided by the machine allowed the two experiments to exploit the full potential of the LHC luminosity, collect-ing the largest samples of PbPb and pPb collisions. ATLAS and CMS are thus in best position to study rare processes in heavy ion collisions. Noticeable examples are the first observations of light-by-light scattering in PbPb, by ATLAS [1], and of top production in pPb, by CMS [2], both announced in 2017.
The kinematic reach of the four LHC experiments in pPb collisions is compared in fig. 2. LHCb can complement the other LHC detectors by providing measurements in its forward coverage (pseudorapidity η between 2 and 5). When the proton beam points toward the “forward” direction, values of Bjorken-x as low as 10−5 in the target Pb nucleon can be probed, where gluon saturation effects are expected to occur. In the inverse configuration, measurements are sensitive to the anti-shadowing region x∼ 0.1. On the other hand, the LHCb tracking detectors are saturated by the very high track density produced in central PbPb collisions in its forward acceptance, so the experiment is more suited for smaller collision systems and for the study of peripheral PbPb collisions. Another unique feature of LHCb is the possibility to inject gas targets (currently He, Ne and Ar) in the LHC beam pipe and study beam-gas collisions, for which the forward
c
CERN on behalf of the ATLAS, CMS and LHCb Collaborations
0 200 400 600 800 1000 1200 1400 1600 μb-1 [email protected] TeV 2011 0 20 40 60 80 100 120 140 160 180 nb-1 [email protected] TeV 2013 0 200 400 600 800 1000 1200 1400 1600 μb-1 [email protected] TeV 2015 0 20 40 60 80 100 120 140 160 180 nb-1 [email protected] TeV 2016 0 200 400 600 800 1000 1200 1400 1600 μb-1 [email protected] TeV 2018 arm) μ ALICE ( ATLAS CMS LHCb
Fig. 1. – Integrated luminosities collected by the four major LHC experiments during the pPb and PbPb runs.
geometry of the detector is particularly well suited. Several samples in this fixed-target configuration were collected for nucleon-nucleon centre-of-mass energies √sNN between 69 and 110 GeV, a relatively unexplored energy scale between the reaches of SPS and LHC experiments. In this configuration the detector covers mid and backward rapidities, and is sensitive to large x values in the target nucleon, where PDFs are affected by the anti-shadowing and EMC effects, and also by a possible intrinsic heavy quark content.
The three collaborations have a very active heavy ion program, with 33 papers submit-ted altogether since the previous La Thuile conference. In this contribution we focus on
Fig. 2. – Accessible kinematic range in terms of Bjorken-x and squared parton-parton invariant mass Q2 for the four LHC experiments in pPb collisions. The forward acceptances of the LHCb and ALICE muon arm detectors correspond to two disjointed regions for the two possible orientations of the proton and lead beams. The kinematic regions accessible in the fixed-target configuration at LHCb and in e-p collisions at HERA are also shown.
[GeV] μ μ m 2.6 2.8 3 3.2 3.4 3.6 3.8 4 4.2 Entries / (0.04 GeV) 3 10 4 10 ATLAS -1 = 5.02 TeV, 0.42 nb NN s Pb+Pb < 40 GeV T p 9 < | < 2 y 0 < | 20-50% Data Fit (nS) ψ Prompt (nS) ψ Non-prompt Bkg [ps] τ 4 − −2 0 2 4 6 8 10 Entries / (0.2 ps) 1 10 2 10 3 10 4 10 5 10 ATLAS -1 = 5.02 TeV, 0.42 nb NN s Pb+Pb < 40 GeV T p 9 < | < 2 y 0 < | 20-50% Data Fit (nS) ψ Prompt (nS) ψ Non-prompt Bkg 9 10 11 ] 2 c ) [GeV/ − μ + μ ( M 0 100 200 300 400 500 600 ) 2 c Candidates / (20 MeV/ LHCb sNN=8.16 TeV, pPb ) nS ( ϒ Background Total ) 2 (GeV/c (KK) φ ) μ μ ( ψ J/ m 5 5.2 5.4 5.6 5.8 6 ) 2 Events / (20 MeV/c 0 2 4 6 8 10 12 Data Fit Signal Background CMS s 0 B (PbPb 5.02 TeV) -1 b μ 351 /nDOF: 39/45 = 0.86 2 χ 3.4 ± Yield = 9.0 < 15 GeV/c T 7 < p |y| < 2.4 Cent. 0-100% 5000 5200 5400 5600 ] 2 c ) [MeV/ + π 0 D ( m 0 50 100 150 200 250 300 350 400 450 500 ) 2c Candidates / (10 MeV/ Data Fit Signal Partial Combinatorial + K 0 D → + B LHCb = 8.16 TeV NN s Pb p
Fig. 3. – Examples of reconstruction of heavy flavour states in the three experiments: mass and proper time for ψ() candidates in PbPb from ATLAS [3], mass of Y (nS) states in pPb from LHCb [4], signal for B0
s→ J/ψ φ in PbPb from CMS [5] and signal for B+→ D0π+in pPb from
LHCb [6].
recent results using probes where the three experiments can better complement ALICE in the study of nuclear matter: heavy flavours, electroweak processes and fixed-target collisions.
2. – Heavy flavours
Rich heavy flavour samples have been collected by the three experiments, thanks, particularly for ATLAS and CMS, to the large integrated luminosities and, particularly for LHCb, to the excellent vertexing and particle identification performance. In fig. 3, ex-amples are shown of the mass and proper time resolutions that can be achieved, enabling the clean separation of different quarkonia states and to disentangle promptly produced charm particles from those produced in b-hadron decays. This is a key feature to study the effect on charm production of hot nuclear matter produced in PbPb collisions, since
b-hadron decays are expected to occur outside the Quark-Gluon Plasma (QGP) formed
in such collisions. Beyond inclusive charm production from b decays, the first results with exclusive open-b decays in heavy ion collisions, also shown in fig. 3, have been re-cently released: the observation of B0
s → J/ψ φ decays in PbPb by CMS [5], and of four
exclusive decays of B0, B+ and Λ0b hadrons in pPb by LHCb [6].
Nuclear effects affecting heavy quark production in AA (PbPb) collisions are quanti-fied by the nuclear modification factor RAA= YieldAA/(Ncoll×Yieldpp) where production yields are compared to those measured in p p collisions at the same energy, scaled by the number of colliding nucleons Ncoll. Data could shed light on quarkonia sequential
suppression from colour screening, a classical signature for QGP formation [7], but many other effects can play a role, notably parton energy loss and statistical regeneration of
[GeV] T p 6 7 8 10 20 30 40 100 AA R 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 ATLAS -1 = 5.02 TeV, 0.42 nb NN s
Charged particles, 0-5%, 2.76 TeV , 0-10% ψ Non-prompt J/ , 0-10% ψ Prompt J/ part N 0 50 100 150 200 250 300 350 400 PbPb ψ (2S)/J/ ψρ 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 ATLAS = 5.02 TeV, 0.42 nb-1 NN s Pb+Pb, -1 = 5.02 TeV, 25 pb s , pp < 40 GeV, |y| < 2 T 9 < p double ratio ψ (2S) to J/ ψ Prompt part N 0 50 100 150 200 250 300 350 400 PbPb ψ (2S)/J/ ψρ 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 ATLAS -1 = 5.02 TeV, 0.42 nb NN s Pb+Pb, -1 = 5.02 TeV, 25 pb s , pp < 40 GeV, |y| < 2 T 9 < p double ratio ψ (2S) to J/ ψ Non-prompt
Fig. 4. – On the left plot, the RAA values as a function of pT are compared for prompt
and non-prompt J/ψ and charged particles for the most central collisions. The double ratio
RAA(ψ(2S))/RAA(J/ψ ) is shown as a function of particle multiplicity in the middle plot for
prompt particles, and in the right plot for non-prompt ones [3].
quarkonia. Cold nuclear matter (CNM) effects like modification of nuclear PDFs, co-herent energy loss and interaction with comoving particles are also relevant and can be probed in smaller collision systems as pPb.
ATLAS released measurements of RAA for prompt and non-prompt J/ψ and ψ(2S)
mesons from the 2015 PbPb data [3]. Results indicate a sizeable J/ψ suppression, with a strong transverse momentum (pT) dependence, similar for prompt, non-prompt J/ψ
and light charged particles (see fig. 4). This could indicate a common dominating effect from parton energy loss in hot matter. However, prompt ψ(2S) are observed to be more suppressed than J/ψ , which is not the case for non-prompt particles, as expected from colour screening, with a reduced effect for more central collisions, possibly because of the increasing importance of statistical regeneration. However, there is some tension with earlier CMS results [8]. The larger data samples collected in 2018 by both experiments should help clarifying the situation.
CMS released RAAmeasurements also for bottomonia states [9], where a spectacular
suppression of Υ (2S) and Υ (3S) states with respect to Υ (1S) is observed, as shown in fig. 5. Results could be predicted by a model [10], tuned on charged-hadron measured
(GeV) -μ + μ m 8 9 10 11 12 13 14 Events / (0.1 GeV) 0 1 2 3 4 5 6 7 8 9 3 10 × < 30 GeV μ μ T p | < 2.4 μ μ |y > 4 GeV μ T p | < 2.4 μ η | Centrality 0-100% PbPb Data Total fit Background scaled AA R (5.02 TeV) -1 b μ PbPb 368 CMS part N 0 50 100 150 200 250 300 350 400 AA R 0 0.2 0.4 0.6 0.8 1 1.2 (1S) ϒ (2S) ϒ (3S) 68% CL ϒ (3S) 95% CL ϒ < 30 GeV T p |y| < 2.4 /s=1 η π 4 /s=2 η π 4 /s=3 η π 4 Krouppa, Strickland (5.02 TeV) -1 , pp 28.0 pb -1 b μ PbPb 368/464 CMS Cent. 0-100%
Fig. 5. – On the left plot, mass distribution for Υ(nS) candidates in PbPb collisions from CMS, compared to the normalized spectrum in p p collisions. The resulting RAA as a function of
CM y -3 -2 -1 0 1 2 3 pPb R 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 < 10 GeV/c T 6.5 < p [EPJC 77, 269 (2017)] ψ Prompt J/ < 10 GeV/c T 6.5 < p 5.02 TeV -1 , pp 28.0 pb -1 pPb 34.6 nb CMS (2S) ψ Prompt 4 − −2 0 2 4 y* 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 ) S (1 ϒ )/ S (3 ϒ pp )/ p Pb|Pb p( ℜ =8.16 TeV NN s c <25 GeV/ T p LHCb LHCb comovers
Fig. 6. – Examples for nuclear modification factor of quarkonia states in pPb collisions. On the left plot, comparison of RpPb for J/ψ and ψ(2S) by CMS [11]. On the right plot, double ratio RpPb(Y (3S))/RpPb(Y (1S)) measured by LHCb [4].
spectra and flow, which attributes the effect to colour screening, though large CNM effects are not excluded in other models. Final-state effects in cold matter are indeed evident from the results in pPb collisions illustrated in fig. 6. Recent results from CMS [11] clearly show a larger suppression of ψ(2S) with respect to J/ψ , confirming similar findings by the other LHC experiments. LHCb showed evidence for the same effect in Υ (3S) with respect to Υ (1S) [4], in nice agreement with predictions based on quarkonia dissociation due to the interaction with comoving final-state particles [12] (“comovers” in the plot).
The flow of heavy flavour particles is another key observable. ATLAS could measure the elliptic flow v2 separately for prompt and non-prompt J/ψ particles [13]. Results,
shown in fig. 7, suggest a higher flow at low pTfor prompt J/ψ , as expected from
statisti-cal regeneration, and a universal behavior at high pT, as observed for RAA. Surprisingly,
a sizeable elliptic flow of heavy hadrons is observed by CMS also in high-multiplicity
pPb collisions, where QGP was not expected to be formed. The latest results with
J/ψ mesons [14] are compatible with what observed previously with D0 mesons [15],
excluding the hypothesis that the flow could be originated only by light particles in the hadronization stage and adding new inputs to the puzzle of collectivity in small systems (see e.g. [16]). 10 15 20 25 30 [GeV] T p 0 0.05 0.1 0.15 0.2 2 v ATLAS -1 = 5.02 TeV, 0.42 nb NN s Pb+Pb, < 2, 0 - 60% y , ψ Prompt J/ < 2, 0 - 60% y , ψ Non-Prompt J/ (GeV) T p 0 2 4 6 8 sub 2 v 0.0 0.1 0.2 CMS pPb 8.16TeV < 250 offline trk N ≤ 185 ψ Prompt J/ 0 Prompt D S 0 K Λ
Fig. 7. – On the left plot, J/ψ elliptic flow in PbPb collisions, measured by ATLAS [13]. In the right plot, the same measurement in pPb by CMS, compared with similar measurements for D0 mesons and strange hadrons [14].
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 γ J x 0.2 0.4 0.6 0.81 1.2 1.4 1.6 1.82 2.2 )γ J x /d N )(dγ N (1/ 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 γ J x 0.2 0.4 0.6 0.81 1.2 1.4 1.6 1.82 2.2 )γ J x /d N )(dγ N (1/ 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 γ J x 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8xJγ 50-80% 30-50% 20-30% 10-20% 0-10% ATLAS -1 5.02 TeV, 25 pb pp -1 Pb+Pb 5.02 TeV, 0.49 nb = 100-158 GeV γ T p
(same each panel)
pp
Pb+Pb
Fig. 8. – Distribution of xJ γ ≡ pjetT /p γ
T in γ-jet events with a high-pT photon, measured by
ATLAS in p p collisions and PbPb collisions of different centrality at√sNN= 5TeV [19].
3. – Electromagnetic probes
Electroweak boson production (γ and W/Z decaying leptonically) provides a clean probe of the initial state, since final states are not interacting strongly with the medium. CMS exploited its large sample of 8 TeV pPb data to reconstruct 180k W → μν can-didates, obtaining precise constraints on nuclear PDFs [17]. ATLAS released new W/Z results in 5 TeV p p data [18], where production is measured with accuracy close to 2%, providing an accurate reference for similar studies in heavy ion data.
Studies of jet-quenching also take profit of associated production of high-pTphotons,
which can tag the initial pTof the recoiling jet. As shown in fig. 8, the symmetry in the
transverse plane observed by ATLAS [19] for γ-jet events produced in p p or peripheral PbPb collisions is strongly modified in central collisions. The amount of jet quenching and its fluctuations can thus be quantified.
In ultra-peripheral collisions (UPC) involving lead beams, the large flux of quasi-real photons can be used to probe the nuclear matter by observing exclusive photoproduction of hadronic states. Perturbative QCD calculations of heavy flavour production are sensi-tive to nuclear PDFs for low x values (down to 10−5in the forward region). CMS recently observed the first Υ signal in pPb UPC at 5 TeV [20], while LHCb released a preliminary measurement of J/ψ production in its first small sample of PbPb collisions [21]. These results are limited by statistics but indicate very good prospects with the largest samples already collected.
ATLAS demonstrated that dimuon photoproduction can be observed also on top of hadronic collisions [22]. By detecting an effect on the dimuon acoplanarity in the most
0 0.005 0.01 0.015 α 0 200 400 600 α d s N d s N 1 0 - 10 % 0 0.005 0.01 0.015 α 10 - 20 % 0 0.005 0.01 0.015 α 20 - 40 % ATLAS = 5.02 TeV NN s -1 Pb+Pb, 0.49 nb 0 0.005 0.01 0.015 α 40 - 80 % Pb+Pb data > 80% data data overlay STARlight +
Fig. 9. – Distribution of the acoplanarity α≡ 1−|φ+−φπ −| for low-pTdimuons produced in PbPb
collisions of different centrality [22], compared to the distribution in peripheral collisions, which is well described by the STARlight photoproduction model.
pNe pHe pAr pAr PbAr pHe pHe pNe pNe PbNe ] 22 protons (Pb) on target [10 2 − 10 1 − 10 1 10 2 10 2500 GeV 4000 GeV 6500 GeV Beam Energy 2015 | 2016 | 2017 | 2018
Fig. 10. – Summary of collected fixed-target samples. The data size is given in terms of delivered protons (ions) on target (POT). For the nominal target pressure of 2× 10−7 mbar, 1022POT correspond to an integrated luminosity of about 5 nb−1 per meter of gas.
central PbPb collisions, as illustrated in fig. 9, they identified a novel clean probe for the hot nuclear medium.
CMS confirmed the mentioned discovery of light-by-light scattering in PbPb UPC collisions, observing 11.1±1.1 signal events with an expected background of 4.0±1.2 [23]. The result is used to set competitive limits on the coupling of a hypothetical axion-like particle with mass above 5 GeV.
4. – Fixed-target collisions
Figure 10 summarises the fixed-target samples acquired by LHCb during the LHC Run 2. One of the main goals of these special runs is to study charm production in different collision systems at the energy scale around 100 GeV/c. The potential of this program was demonstrated by measuring D0 and J/ψ production in two of the first available
samples: pHe collisions at √sNN = 86.6 GeV and pAr collisions at 110 GeV [24]. As shown in fig. 11, the differential rapidity distributions are found to be in good agreement with models based on collinear factorisation and a sizeable intrinsic charm content is not favoured by these data. More precise measurements and additional final states are expected in the near future from the larger samples already collected.
Data with a helium target are also very interesting to cosmic ray physics, as they reproduce collisions in space between primary cosmic rays and the interstellar medium.
y* 2 − −1 0 b/nucleon]μ [ y* /d 0 D σ d 0 10 20 30 40 50 60 70 80 He p = 86.6 GeV NN s LHCb LHCb data 1.44 CT14NLO+nCTEQ15 × He p 1.44 CT14NLO × pp 20 40 60 80 100 ] c [GeV/ p 5 10 15 20 25 30 35 /GeV]c bμ [ p /dσ d DATA EPOS LHC EPOS 1.99 QGSJETII-04 QGSJETII-04m HIJING 1.38 PYTHIA 6.4 LHCb X p → pHe = 110 GeV NN s c < 2.8 GeV/ T p 1.2 <
Fig. 11. – Examples of production studies in fixed-target configuration. On the left plot, dif-ferential cross-section in the centre-of-mass rapidity y∗ for D0 production in pHe collisions at 86.6 GeV [24]. On the right plot, differential cross-section in momentum for p production in pHe collisions at 110 GeV for pTabove 1.2 GeV, compared to predictions from several models [25].
The recent measurements in space of antiprotons with momentum above 10 GeV, no-tably by AMS-02 [26], are sensitive to a possible dark matter contribution but their interpretation is limited by the uncertainty on the production cross-sections for antipro-tons in such collisions (see e.g. [27]). The first measurement of antiproton production in
pHe collisions, accounting for about 40% of the expected p flux, was performed in the p
momentum range 12-110 GeV at√sNN= 110 GeV. Results [25], also shown in fig. 11, are substantially more precise than the spread among the predictions from the commonly used models of hadronic interactions.
5. – Conclusions and Prospects
Dramatic improvements in the understanding of the phenomenology of ultra-relativis-tic heavy ion collision are being provided by the LHC experiments. Unique contributions come from the ATLAS, CMS and LHCb experiments, including many other results, notably on jet quenching and flow of light particles, that could not be covered here. This progress motivates a development of the heavy ion program in the coming runs of the LHC, exploiting the possibilities offered by the HL-LHC [28]. The fixed-target physics program at LHCb is also being pushed forward. An upgraded gas target is foreseen for the LHC Run 3, allowing measurements at higher luminosity and using more gas species, opening a wide range of novel physics opportunities [29].
REFERENCES
[1] Aaboud M. et al., Nat. Phys., 13 (2017) 852.
[2] Sirunyan A. M. et al., Phys. Rev. Lett., 119 (2017) 242001. [3] Aaboud M. et al., Eur. Phys. J. C, 78 (2018) 762.
[4] Aaij R. et al., JHEP, 11 (2018) 194.
[5] Sirunyan A. M. et al., Phys. Lett. B, 796 (2019) 168, arXiv:1810.03022. [6] Aaij R. et al., Phys. Rev. D, 99 (2019) 052011.
[7] Matsui T. and Satz H., Phys. Lett. B, 178 (1986) 416. [8] Sirunyan A. M. et al., Phys. Rev. Lett., 118 (2017) 162301. [9] Sirunyan A. M. et al., Phys. Lett. B, 790 (2019) 270. [10] Krouppa B. and Strickland M., Universe, 2 (2016) 16. [11] Sirunyan A. M. et al., Phys. Lett. B, 790 (2019) 509. [12] Ferreiro E. G., Phys. Lett. B, 749 (2015) 98. [13] Aaboud M. et al., Eur. Phys. J. C, 78 (2018) 784. [14] Sirunyan A. M. et al., Phys. Lett. B, 791 (2019) 172. [15] Sirunyan A. M. et al., Phys. Rev. Lett., 121 (2018) 082301.
[16] Dusling K., Li W. and Schenke B., Int. J. Mod. Phys. E, 25 (2016) 1630002. [17] Sirunyan A. M. et al., Phys. Lett. B, 800 (2020) 135048, arXiv:1905.01486. [18] Aaboud M. et al., Eur. Phys. J. C, 79 (2019) 128, 79 (2019) 374(E). [19] Aaboud M. et al., Phys. Lett. B, 789 (2019) 167.
[20] Sirunyan A. M. et al., Eur. Phys. J. C, 79 (2019) 277.
[21] LHCb Collaboration, Tech. Rep. LHCb-CONF-2018-003 (2018). [22] Aaboud M. et al., Phys. Rev. Lett., 121 (2018) 212301.
[23] Sirunyan A. M. et al., Phys. Lett. B, 797 (2019) 134826, arXiv:1810.04602. [24] Aaij R. et al., Phys. Rev. Lett., 122 (2019) 132002.
[25] Aaij R. et al., Phys. Rev. Lett., 121 (2019) 222001. [26] Aguilar M. et al., Phys. Rev. Lett., 117 (2016) 091103. [27] Giesen G. et al., JCAP, 09 (2015) 023.
[28] Citron Z. et al., Tech. Rep. CERN-LPCC-2018-07 (2018) arXiv:1812.06772. [29] Graziani G. et al., Tech. Rep. CERN-LHCb-PUB-2018-015 (2018).
