Available online at www.sciencedirect.com
Nuclear Physics A 967 (2017) 192–199 www.elsevier.com/locate/nuclphysa
Heavy Flavor Production, Energy Loss and Flow — Experimental Overview Xin Dong MS70R0319, One Cyclotron Road, Berkeley, CA 94720, USA
Abstract In this proceedings, I highlight selected experimental results on heavy flavor quark production presented at the Quark Matter 2017 conference. I show experimental evidences that charm quarks interact strongly and flow with the QGP medium and that bottom quarks lose less energy compared to charm quarks. These high quality data will offer stringent constraints on theoretical model calculations and help precision determination of QGP medium transport parameters. In the end, I look forward to a more prospective future of the heavy flavor program with further improved detector equipments at both RHIC and LHC. Keywords: heavy quarks, energy loss, diffusion coefficient, nuclear modification factor, elliptic flow
1. Introduction Heavy flavor quarks (c, b), due to their large masses, are expected to have unique roles for studying QCD in both vacuum and medium. There have been extensive measurements of heavy quark production in elementary collisions that demonstrate their production is calculable in perturbative QCD. Heavy quark interaction with the hot Quark-Gluon Plasma (QGP) should shed light on the roles of radiative energy loss vs. elastic collisional energy loss in such a medium. In particular, one should expect the mass hierarchy for the parton energy loss in QCD medium: ΔEb < ΔEc < ΔEq < ΔEg . However, the hierarchy may not be directly revealed in the nuclear modification factor (RAA ) observable due to differences in initial parton spectra, fragmentation etc.. Model calculations show that although charm hadron RAA is similar to that of light hadrons, the RAA of bottom hadron production in heavy-ion collisions will be less suppressed compared to charm and other light hadrons, revealing the mass hierarchy of parton energy loss unambiguously, as seen in Fig. 1 (left) [1]. And such a difference in RAA will disappear at very high pT as the mass effect becomes less important or negligible. The heavy quark propagation inside the QGP medium can be treated as “Brownian” motion when the heavy quark mass is much larger than the medium temperature as well as the interaction strength. The heavy quark equation of motion can be described by a reliable stochastic Langevin simulation and characterized by one intrinsic medium transport parameter - the heavy quark diffusion coefficient. Here low pT measurements will be more relevant for the determination of this transport parameter. Figure 1 (right) summarizes the latest theory calculations for the charm quark spacial diffusion coefficient [2]. The goal of
http://dx.doi.org/10.1016/j.nuclphysa.2017.04.023 0375-9474/© 2017 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
X. Dong / Nuclear Physics A 967 (2017) 192–199
193
Ds × 2πT
studying heavy quark production in heavy-ion collisions is to reveal the parton energy loss mechanisms and in return to measure the QGP transport parameter, particularly its temperature dependence. 40 pQCD LO
30 F
T-Matrix V=
20 H
R
G
10
0 0.5
DQPM V=U T-Matrix
Lattice: Banerjee
MC@sHQ Lattice: Ding
1
1.5
2
T/Tc Fig. 1. (Left) CUJET pQCD predictions for different flavor hadrons in central Au+Au collisions at predictions on charm quark spacial diffusion coefficient as a function of temperature.
√
sNN = 200 GeV. (Right) Model
Heavy flavor hadrons decay shortly after they are produced (cτ ∼ 60 − 500μm). Precision measurements of heavy quark production in high multiplicity heavy-ion collisions will require separation of their decay vertices from primary collisions in order to reduce combinatorial background. Therefore, precision silicon pixel detectors are critical to achieve necessary pointing resolution in a wide kinematic region. All RHIC and LHC experiments are equipped by such detectors. Particularly, a high resolution pixel detector based on the MAPS technology was firstly applied to the STAR experiment. Its unique features - ultimate pitch size, thin material thickness are perfect fit for precision heavy flavor measurements over a broad momentum range in heavy-ion collisions. The track pointing resolution in Au+Au collisions has reached < 40μm at a transverse momentum of 1 GeV/c [3]. Heavy flavor hadrons are measured via either full reconstruction of their decays or inclusive/semiinclusive decay daughters. In the following part of this proceeding, I will show selected heavy flavor measurements at RHIC and LHC focusing on recent key achievements towards understanding the QGP properties at mid-rapidity.
104 *
p + p(p) → D0/D + X 102 D 0 (A
LICE
)
y=0
dσ/dp dy|cc (mb/(GeV/c))
2. Heavy Flavor Production in p + p and p/d+A Collisions
1
pp 7 TeV [x103]
D 0 (C
DF)
pp 1.96 TeV [x102]
T
10−2
D 0+
D* (ST
AR
10−4
)
pp 500 GeV [x10]
D0 +D * (ST
AR
)
pp 200 GeV
10−6 FONLL 10−8
0
5
10
15
20
Transverse Momentum p (GeV/c) T
√ Fig. 2. (Left) D0 -meson production cross section measured in p+p( p) ¯ collisions from sNN = 0.2-7 TeV compared to pQCD FONLL √ calculations (grey bands); (Right) B+ -meson production cross section measured in p+p collisions at sNN = 7 TeV)
X. Dong / Nuclear Physics A 967 (2017) 192–199 R pPb
194
ALICE
1.6
p-Pb, s NN=5.02 TeV Prompt D mesons, -0.96
1.4
0
+
*+
Average D , D , D 0
D
1.2 1 0.8 0.6 CGC (Fujii-Watanabe)
0.4
pQCD NLO (MNR) with CTEQ6M+EPS09 PDF Vitev et al.: power corr. + k broad + CNM Eloss
0.2
T
Kang et al.: incoherent multiple scattering
0 0
5
10
15
20
25
p (GeV/c ) T
√ Fig. 3. (Left) D0 -meson nuclear modification factor RpPb at sNN = 5.02 TeV compared to various model calculations; (Right) Forward-backward rapidity asymmetry of D0 -meson nuclear modification factor as a function of pT at 2.5< |y∗ | <4 from LHCb.
Figure 2 shows the measured inclusive D0 -meson (left) and B+ -meson (right) production cross sections in p + p collisions at mid-rapidity from various high energy experiments [4, 5, 6, 7, 8, 9, 10, 11]. These measurements over broad collisional energy and particle kinematic ranges can be well described by pQCD FONLL calculations [12]. Recent D-meson and B-meson cross section measurements at forward rapidities measured by LHCb also show good agreements with FONLL calculations [13, 14]. An interesting observation is that most experimental data fall at the upper bounds of FONLL calculation uncertainty bands while both experimental data and the FONLL upper bounds start to approach the FONLL central values at high pT . There is a vast amount of precision experimental data, collected in the past few years from RHIC and LHC, which allows to constrain the pQCD theory calculations [15]. Au+Au → D0 @ 200 GeV 0-10%
2
0
D 2014 0
D 2010/11
1.5
RAA
π 0-12% STAR 1
p+p uncert. 0.5
0
0
2
4
6
8
p (GeV/c) T
Fig. 4. RAA of inclusive D0 mesons compared to inclusive charged hadrons in central Au+Au collisions at √ (left) and central Pb+Pb at sNN = 5.02 TeV collisions by CMS (right).
√
sNN = 200 GeV by STAR
Figure 3 (left) shows the recent measurement of D0 -meson RpPb in minimum bias p+Pb collisions at sNN = 5.02 TeV [16]. The data shows the RpPb is close to unity in the measured kinematic region indicating no significant cold nuclear matter effect. Also shown in the figure are various model calculations with either different initial conditions (CGC or nuclear PDF) or including energy loss in cold nuclear matter etc. The data precision need to be further improved to differentiate between models. In the right panel, LHCb √ measured the forward-backward asymmetry of RpPb , RFB , as a function of pT in p+Pb collisions at sNN = 5.02 TeV [17]. The small but finite asymmetry can be well described by a pQCD calculation which includes the EPS09 nuclear PDFs.
√
X. Dong / Nuclear Physics A 967 (2017) 192–199
195
3. Heavy Flavor Production in A+A Collisions
0
0
0.1
10-40%
0.05
0 0
0.5
1
1.5
2
(m - m0) / nq (GeV/c 2)
2.5
STAR D STAR SUBATECH TAMU c-quark diff. TAMU no c-quark diff. Duke - 2πTDs=7 LBT PHSD 3D viscous hydro
2
0.15
STAR Au+Au @ 200 GeV
Anisotropy Parameter, v
D Ξ Λ KS
2
Anisotropy Parameter, v / nq
3.1. Charm Measurements Figure 4 shows the RAA of inclusive D0 mesons compared to inclusive charged hadrons in central Au+Au √ √ collisions at sNN = 200 GeV by STAR (left) [18, 19] and central Pb+Pb at sNN = 5.02 TeV collisions by √ CMS (right) [20]. Similar measurements have been done by the ALICE collaboration at both sNN = 2.76 and 5.02 TeV [21, 22]. At pT > 4 GeV/c, one can observe that D-meson production in central heavy-ion collisions is significantly suppressed, and the suppression level is nearly similar compared to that of light flavor hadrons at both RHIC and LHC. Combined with no suppression seen in p+Pb collisions, the large suppression observed in A+A collisions demonstrates that charm quarks lose significant amount of energy in the hot QGP medium.
0.2
Au+Au sNN = 200 GeV 0-80%
0.1
0 0
1
2
3
4
5
6
7
p (GeV/c) T
T
0.2
Pb+Pb 30-50%
ALICE 2.76 TeV KS
2
Anisotropy Parameter, v / nq
Fig. 5. (Left) v2 as a function of transverse kinetic energy mT − m0 normalized by the number-of-constituent-quarks (nq ) for D0 √ compared to other identified particles in 10–40% centrality Au+Au collisions at sNN = 200 GeV; (Right) v2 as a function of pT in √ 0–80% centrality Au+Au collisions at sNN = 200 GeV compared with model calculations.
0.15
p+p Λ+Λ
0.1 0.05 0
D @5.02TeV CMS
0
D0 @5.02TeV ALICE
0
0.5
1
1.5
2
(m - m0) / nq (GeV/c 2)
2.5
T
Fig. 6. (Left) v2 as a function of transverse kinetic energy mT − m0 normalized by the number-of-constituent-quarks (nq ) for D0 compared to other identified particles in 30–50% centrality Pb+Pb collisions; (Right) v2 as a function of pT in 30–50% centrality √ Pb+Pb collisions at sNN = 5.02 TeV compared with model calculations.
Figure 5 (left) shows the latest D0 meson v2 measurement compared to other identified particles (KS0 , Λ √ and Ξ− ) in 10–40% centrality Au+Au collisions at sNN = 200 GeV [3]. The data are plotted as a function of transverse kinetic energy mT − m0 , where mT = p2T + m20 and both axises are normalized by the numberof-constituent-quarks (nq ). One can observe that the D0 mesons v2 is significant and follows the same trend as light flavor hadrons. This suggests that charm quarks should have gain sufficient interactions in the QGP medium to reach such a significant flow. In the right panel, the measured D0 meson v2 in 0–80%
196
X. Dong / Nuclear Physics A 967 (2017) 192–199
centrality Au+Au collisions is compared to various theory model calculations. A 3D viscous event-byevent hydrodynamic simulation with η/s = 0.12 using the AMPT initial condition and tuned to describe v2 for light hadrons, predicts D0 v2 that is consistent with the data for pT < 4 GeV/c [23]. This suggests that charm quarks have achieved thermal equilibrium in these collisions. These models that can describe both the RAA and v2 data include the temperature–dependent charm diffusion coefficient 2πT D s in the range of ∼2–12 consistent with the predictions by lattice QCD calculations [24, 25]. Figure 6 (left) shows the D-meson v2 measured at LHC and in a similar way compared to other identified particles in 30–50% centrality Pb+Pb collisions [20, 22]. Note that the D-meson v2 is in a different collision energy. Nevertheless, one can observe the similar universal trend for all particles measured here including D-mesons. There might be a small deviation from a universal trend at 1 < (mT − m0 )/nq < 2 GeV/c2 for the D-meson data at LHC, but the deviation shows also in light flavor hadrons. And more precise measurements at low pT will be needed to confirm the hydrodynamic mass effect. In the right panel, the CMS data points are compared to various model calculations. The data especially at intermediate to high pT now have reached great precision, providing strong constraints on different models. The next important task would be a joint effort between experimentalists and theorists to firstly pin down all the non-trivial differences in different model implementations and then to use the data to constrain precisely the temperature dependent of heavy quark diffusion coefficient 2πT D s .
Fig. 7. D+s /D0 ratios as a function of pT measured by STAR (left) and ALICE (right) in mid-central heavy-ion collisions compared to p + p references as well as model predictions.
Figure 7 shows the recent measurements of D+s /D0 ratios as a function of pT measured by STAR (left) [26] and ALICE (right) [22] in mid-central heavy-ion collisions compared to p + p references as well as model predictions. The PYTHIA and world average values for D+s /D0 in elementary collisions are √ about 0.15. The ALICE measurement in p + p collisions at s = 7 TeV is slightly higher, but consistent within their uncertainties. However, in mid-central heavy-ion collisions, both STAR and ALICE observe that the D+s /D0 ratios are significantly higher than the expected p + p reference in the measured pT region. This is qualitatively consistent with the expectation from the coalescence hadronization coupled with strangeness enhancement in heavy-ion collisions. However, the ratio observed in data is larger than the model calculation from TAMU with these ingredients implemented as shown in the left plot for RHIC [27]. Λ+c baryon has an extreme short life time and has never been reconstructed in heavy-ion collisions. √ Figure 8 (left) shows the first Λ+c signal reconstructed in 10–60% Au+Au collisions at sNN = 200 GeV by 0 STAR [28]. The right figure shows the measured Λc /D ratio compared to baryon-to-meson ratios measured for light flavor hadrons and model predictions. The Λ+c /D0 ratio measured in mid-central Au+Au collisions is significantly larger than the PYTHIA expectation or the world average fragmentation ratio from elementary collisions. Also shown in the figure are two model calculations with coalescence hadronization for charm quarks. In Ko’s model, thermalized charm quark spectrum is used for recombination and the predicted Λ+c /D0 ratio is comparable to the measurement despite the calculation is for 0–5% central collisions. One needs low pT measurement to further differentiate between three-quark vs. quark-di-quark recombina-
Baryon/Meson Ratios
X. Dong / Nuclear Physics A 967 (2017) 192–199
197
(p+p ) (π++π-)
Au+Au @ 200GeV
(Λ+Λ) 2K 0S 0-5%
(D 0+D 0)
Ko: di-quark
0-12%
1
(Λc+Λc)
STAR Preliminary
0-5%
10-60%
Ko: three-quark 60-80%
60-80% Greco (Λ c /D)
SHM
PYTHIA
10−1
0
1
2
3
4
5
60
1
2
3
4
5
60
1
2
3
4
5
6
Transverse Momentum (GeV/c) Fig. 8. (Left) First reconstructed Λc signal in heavy-ion collisions from STAR; (Right) Baryon-to-meson ratios for light flavor hadrons, strangeness hadrons and the new charm hadrons measured by STAR. The Λc /D0 data is compared to various model calculations in the right panel.
tion scenarios. In Greco’s model, charm quarks diffuse in the QGP medium and then recombine with light quarks to form charm hadrons. The calculated ratio is w.r.t the total charm mesons that include D+ and D s . One may expect a factor of 1.5 (p + p baseline) or larger (if D s is enhanced) increase to compare to the measured Λ+c /D0 ratio. To summarize the charm measurements, it is observed at both RHIC and LHC that D-meson RAA is significantly suppressed at pT > 4 GeV/c, similar as other light hadrons. D-meson v2 follows the same universal trend as other light hadrons from low to high pT too. Furthermore D s /D0 and Λc /D0 ratios are significantly enhanced in mid-central heavy-ion collisions compared to the p + p references. These results show clear evidences that charm quarks lose a significant amount of energy in the medium and they flow with the QGP medium. This indicates that charm quarks may have reached thermal equilibrium in the heavy-ion collisions at RHIC and LHC. 3.2. Bottom Measurements RAA
CMS 1.4 1.2 1
sNN = 2.76 TeV Open beauty: nonprompt J/ ψ 6.5 < p < 30 GeV/c, |y| < 1.2 T
Open charm: prompt D (ALICE) 8 < p < 16 GeV/c, |y| < 0.5 T
0.8 0.6 0.4 0.2 0 0
50 100 150 200 250 300 350 400
Npart
Fig. 9. (Left) Di-jet asymmetry parameter (x j ) difference between Pb+Pb and p + p collisions as a function of centrality for inclusive jets and b-jets measured by CMS. (Middle) RAA of non-prompt J/ψ and fully reconstructed B+ -meson in Pb+Pb collisions compared to inclusive charged hadrons and D0 -mesons. (Right) RAA of non-prompt Jψ (CMS) and D0 (ALICE) integrated at high pT region as a function of centrality.
CMS collaboration pioneered the measurement of heavy flavor tagged jet in heavy-ion collisions. The early measurement on inclusive b-jet RAA is comparable to inclusive jet RAA at pT > 80 GeV/c [29]. Figure 9 (left) shows a new CMS measurement on di-jet asymmetry difference between Pb+Pb and p + p collisions for inclusive jets and b-tagged jets [30]. The fraction x j is defined as the pT ratio of the sub-leading jet w.r.t. the leading jet - x j = pT,2 /pT,1 . The results show that there is no significant difference between inclusive di-jets and b-tagged di-jets in Pb+Pb collisions in the jet pT regions indicated in the figure. Moving towards
X. Dong / Nuclear Physics A 967 (2017) 192–199
198
RAA
lower pT region, at this conference, CMS reported the first fully reconstructed B+ -meson RAA measurement √ in 0-100% centrality Pb+Pb collisions at sNN = 5.02 TeV [31]. In the pT region between 8–40 GeV/c, the B+ -meson RAA is comparable to that of light hadrons and D0 -meson. ATLAS also reported the measurement of non-prompt J/ψ RAA which also levels around 0.35–0.4 that is similar to that of B+ mesons in the pT of 9–40 GeV/c [32]. In short, at pT > 10 GeV/c, the LHC measurements for B-mesons, b-jet, and non-prompt J/ψ show no difference from those of light hadrons or D-mesons beyond the current uncertainties. When moving further down in pT , in a recently submitted paper [33] and also reported in this conference, the CMS collaboration measured the non-prompt J/ψ RAA and the result shows differences w.r.t. the light √ hadron and D0 -meson RAA in the region of 4–10 GeV/c in Pb+Pb collisions at sNN = 5.02 TeV, as seen in Fig. 9 middle figure [31]. The right panel of Fig. 9 shows a direct comparison of the integrated RAA in the high pT region for non-prompt J/ψ (CMS) and prompt D0 (ALICE) as a function of collision centrality √ at sNN = 2.76 TeV [33]. The pT regions are chosen so that parent B-meson pT for the non-prompt J/ψ is similar to that of prompt D-meson. The figure clearly demonstrates the difference in RAA between B mesons and D mesons in this kinematic region, suggesting the mass hierarchy of parton energy loss - a feature predicted by pQCD calculations [1]. A recent ALICE paper on the charm/bottom separated single electron RAA also shows the RAA of electrons from bottom decays has a higher value than that of electrons from charm+bottom together [34]. According to the pQCD expectations, such a difference due to the mass effect will gradually disappear with increasing pT and this is consistent with what was observed in high pT B-mesons, b-jet, and non-prompt J/ψ RAA measurements. +Au Au+Au 0-80% 0%
sNN = 200 Ge GeV
STAR Preliminary
B→J/ ψ (0-80%/FONLL) inclusive D 0 DUKE : B→J/ ψ
Au Au+Au 0-10% %
STAR Preliminary
GeV sNN = 200 GeV
B→D 0 (0-20%/FONLL) inclusive D 0 DUKE : B→D 0 DUKE : prompt D 0
Au Au+Au 0-80% %
STAR Preliminary
sNN = 200 G GeV eV
B→e D →e DUKE : B→e DUKE : D →e
DUKE: Phys. Rev. C 92, 024907
1
0.2
p+p uncertainty 0 FONLL B →D uncertainty
FONLL B →J/ ψ uncertainty
1
2
3
4
5
6
7
8
1
2
3
B+D →e
RAA
4
5
6
7
Transverse Momentum (GeV/ c)
8
1
uncertainty
2
3
4
5
6
7
8
Fig. 10. (Left) RAA of charm and bottom separated single electrons in central Au+Au collisions by PHENIX. (Right) RAA of nonprompt J/ψ, non-prompt D0 and charm/bottom separated electrons in Au+Au collisions by STAR.
At RHIC, both PHENIX and STAR have reported the latest measurements on the bottom production in this conference [35, 36]. Figure 10 (left) shows the PHENIX RAA measurement of electrons from charm and bottom decays in central Au+Au collisions. The RAA of electrons from bottom decays is clearly larger than that of electrons from charm decays in 3–5 GeV/c indicating less suppression of bottom quark in the hot medium in these collisions. STAR reported recent measurements of bottom production via multi-channels: non-prompt J/ψ, non-prompt D0 and charm/bottom separated electrons in Au+Au collisions, shown in the right of Fig. 10. The RAA of non-prompt J/ψ and non-prompt D0 are calculated using the FONLL value as the p+ p reference, and they show a suppression at pT > 5 GeV/c indicating bottom quarks suffer energy loss in the medium created in these collisions. In the third panel, the bottom-electron RAA is clearly above the charm-electron RAA . Considering the uncertainties between the two are fully anti-correlated, the significance of the difference is about 2σ. Bottom production measurements are more challenging due to their low production cross section. Observations at RHIC and LHC have shown evidences that bottom quarks lose less energy compared to charm and light flavor quarks in the medium created in heavy-ion collisions. At LHC, the difference in RAA between bottom hadrons or b-jets and charm/light flavor hadrons/jets disappear at high pT . These results are consistent with the mass hierarchy feature for parton energy loss inside the QGP medium. 4. Summary and Future Outlook There have been great achievements in heavy flavor measurements in the past few years with new instrumentation and large datasets collected at RHIC and LHC. At this conference, we have seen clear evidences
X. Dong / Nuclear Physics A 967 (2017) 192–199
199
that charm quarks flow the same as other light hadrons and strong suppression in RAA , which indicates charm quarks may be thermalized in the QGP medium at top RHIC and LHC energies. We also see evidences of less energy loss for bottom quarks than charm or light quarks, consistent with the suppression mass hierarchy of parton energy loss. These precision measurements will provide strong constraints on theory model calculations. I look forward to joint efforts between experimentalists and theorists for precision extraction of QGP medium transport parameter, e.g. the temperature dependence of heavy quark diffusion coefficient. The next phase of the heavy quark program will be focusing on precision open bottom measurements and heavy quark correlations. We have observed the evidences of mass hierarchy of parton energy loss. A detailed investigation on open bottom production in heavy-ion collisions will be necessary to evaluate quantitatively the roles between radiative energy loss vs. collisional energy loss. Open bottom production will also offer the cleanest way to measure the heavy quark diffusion coefficient due to its much larger quark mass compared to the charm quark. Total bottom yield will further help a precision interpretation of Upsilon results measured in heavy-ion collisions. This requires precision measurements down to low pT . The ALICE ITS-upgrade at LHC [37] as well as the sPHENIX MVTX [38] utilizing the next generation fast MAPS sensor are planned to explore these measurements at LHC and RHIC. The upgraded detectors will also enable many other heavy flavor observables, e.g. high statistics charmed baryon production, heavy flavor correlations etc. I look forward to heavy flavor programs at RHIC and LHC in the coming decade. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37]
A. Buzzatti, M. Gyulassy, Phys. Rev. Lett. 108 (2012) 022301. Y. Akiba, et al., arXiv:1502.02730. L. Adamczyk, et al., arXiv:1701.06060. L. Adamczyk, et al., Phys. Rev. D86 (2012) 072013. D. Acosta, et al., Phys. Rev. Lett. 91 (2003) 241804. B. Abelev, et al., JHEP 01 (2012) 128. S. Acharya, et al., arXiv:1702.00766. V. Khachatryan, et al., Phys. Rev. Lett. 106 (2011) 112001. G. Aad, et al., JHEP 10 (2013) 042. V. Khachatryan, et al., arXiv:1609.00873. C. Aidala, et al., arXiv:1701.01342. M. Cacciari, P. Nason, R. Vogt, Phys. Rev. Lett. 95 (2005) 122001. R. Aaij, et al., JHEP 03 (2016) 159. R. Aaij, et al., JHEP 08 (2013) 117. R. E. Nelson, R. Vogt, A. D. Frawley, Phys. Rev. C87 (1) (2013) 014908. B. B. Abelev, et al., Phys. Rev. Lett. 113 (23) (2014) 232301. P. Robbe, these proceedings. L. Adamczyk, et al., Phys. Rev. Lett. 113 (14) (2014) 142301. G. Xie, Nucl. Phys. A956 (2016) 473–476. J. Sun, these proceedings. J. Adam, et al., JHEP 03 (2016) 081. A. Barbano, these proceedings. L. G. Pang, Y. Hatta, X. N. Wang, B. W. Xiao, Phys. Rev. D91 (2015) 074027. D. Banerjee, S. Datta, R. Gavai, P. Majumdar, Phys. Rev. D85 (2012) 014510. H. T. Ding, F. Karsch, S. Mukherjee, Int. J. Mod. Phys. E24 (10) (2015) 1530007. L. Zhou, these proceedings. M. He, R. J. Fries, R. Rapp, Phys. Rev. Lett. 110 (11) (2013) 112301. G. Xie, these proceedings. S. Chatrchyan, et al., Phys. Rev. Lett. 113 (2014) 132301. K. Jung, these proceedings. T.-W. Wang, these proceedings. J. Lopez, these proceedings. V. Khachatryan, et al., arXiv:1610.00613. J. Adam, et al., arXiv:1609.03898. K. Nagashima, these proceedings. K. Oh, these proceedings. ALICE ITS Technical Design Proposal. URL http://cds.cern.ch/record/1625842/ [38] sPHENIX MVTX pre-proposal. URL https://www.phenix.bnl.gov/WWW/publish/mxliu/sPHENIX/MVTX/sPHENIX-MVTX-Preproposal-022017-final.pdf