Holography and the problem of parton energy loss in a quark-gluon - - PowerPoint PPT Presentation
Holography and the problem of parton energy loss in a quark-gluon plasma Daniel Pablos SEWM 18 Barcelona 25th May 2018 Absence of quasiparticles? Most satisfactory description of QGP involves an almost ideal liquid phase pp pPb PbPb
studies of QGP formation in small systems suggest common hydrodynamic origin for flow effects!
pp pPb PbPb
Most satisfactory description of QGP involves an almost ideal liquid phase
Weller & Romatschke - PLB ‘17
2
Daniel Pablos McGill / JETSCAPE
Small value of shear viscosity over entropy density ratio
η s ∼ 0.08
τqp ∼ 5η s 1 T ∼ 1 T
challenges quasiparticle description Predicted by Policastro, Son and Starinets (2001) for a large class of non-abelian gauge theories at strong coupling which have a gravity dual
pp pPb PbPb
Bernhard et al. - PRC ‘16 York & Moore - PRD ‘08
3
Daniel Pablos McGill / JETSCAPE
pp pPb PbPb
Chesler - PRL ‘15, JHEP ‘16
4
Daniel Pablos McGill / JETSCAPE
Hydrodynamics at work with large gradients at very early times Natural situation at strong coupling
R ∼ 1 T
Even for system sizes of order hydrodynamic expansion is well behaved Appealing picture of hydrodynamization for all system sizes within strong coupling
would like to have description of jet/QGP interaction in harmony with sQGP hypothesis traditional pQCD computations need to assume separation of scales
λD ⌧ λm.f.p.
g ⌧ 1
for ex: neglect correlations among scatterers
q ∼ √ ET
But the typical virtuality exchanged between the jet and the medium
β1 loop(q)
g ∼ 2
There is motivation to explore an alternative, non-perturbative, description of jet quenching that does not assume the presence of quasiparticles
5
Daniel Pablos McGill / JETSCAPE
quarks are dual to open strings attached to probe flavour branes having a plasma in the gauge theory represents a black hole in the bulk
J Friess, et al., PRD75 (2007)
and QCD have very different vacuums but share similarities bulk metric perturbations encode boundary stress energy variations
N = 4 SYM
T 6= 0
T > Tc
N = 4
and QCD
6
Daniel Pablos McGill / JETSCAPE
presence of string perturbs metric satisfies linearised Einstein’s equations near boundary expression
string sourced hydro (long wavelength) non-hydro (jet modes)
dressed quarks are open strings attached to a D7 flavour brane charged under U(1) gauge field sourcing baryon current at boundary depth of string endpoint determines localisation of excitation at boundary
7
Daniel Pablos McGill / JETSCAPE
Chesler et al. - PRD ‘09 Chesler & Rajagopal, JHEP ‘16 horizon boundary
8
Daniel Pablos McGill / JETSCAPE
Nambu-Goto action
null geodesics endpoint angle string profile
Chesler & Rajagopal - PRD ‘15, JHEP ‘16
Schwarzschild-AdS Solve E.O.M. by finding null geodesic profile: xgeo(t), ugeo(t) Find energy carried by each geodesic: (peaks at the endpoint)
Πτ
0(σ)
Construct the string energy-momentum tensor:
1. 2. 3.
with rate:
as the jet loses energy … it also gets wider Fractional energy loss
initial jet opening angle
most energy at endpoint: Bragg-like peak
9
Daniel Pablos McGill / JETSCAPE
Chesler & Rajagopal - PRD ‘15, JHEP ‘16
Unambiguous determination of boundary jet properties: Pieces of string falling below horizon energy-momentum into hydro modes xtherm = 1 2κ E1/3
in
T 4/3
Also, quite interestingly:
High energy jet starts with a high virtuality, much greater than medium scale Parton shower well approximated by vacuum-like splittings (late stages?) Use non-perturbative holographic prescription for partonic energy loss Energy flowing into hydro modes: Estimate the hadronic spectra coming from medium response (assume small perturbation, instantaneous hydrodynamization) Lost jet energy converted into soft particles at large angles (corr. bkgd.)
Pablos et al. - JHEP ‘14, ‘16, ‘17
O(1)
Plasma-jet interaction dominated by temperature scale
free parameter
10
Daniel Pablos McGill / JETSCAPE
string acts as a perturbation in the large Nc limit agreement between hydrodynamics & wake of a quark in gauge/gravity duality energy-momentum conservation in the jet+plasma interplay wake hadron distribution estimate
small perturbation on top of hydro
no extra free parameter (within hybrid model)
11
Daniel Pablos McGill / JETSCAPE
Chesler & Yaffe - PRD ‘08
12
Daniel Pablos McGill / JETSCAPE
ATLAS-CONF-2017-009 JHEP 1704 (2017) 039
Use hadron and jet suppression data for central collisions at LHC to fit the free parameter
Hadron suppression Jet suppression
# inelastic collisions per heavy ion collision estimated through the Glauber model Perform a global fit and extract the best value of κSC
13
Daniel Pablos McGill / JETSCAPE
0.3 0.4 0.5 0.6 0.7 0.8 CMS Had 5.02 ATLAS Had 5.02 CMS Had 2.76 ATLAS Had 2.76 RHIC Had 0.20 CMS Jets R=0.2 2.76 CMS Jets R=0.3 2.76 CMS Jets R=0.4 2.76 ATLAS Jets R=0.4 5.02 ATLAS Jets R=0.4 2.76 κ 2 σ 1 σ Global Fit Hadrons 0-5% Jets 0-10%
* with LHC data only
In preparation
14
Daniel Pablos McGill / JETSCAPE
Consistent, but some tension between hadrons & jets preferred value
0.3 0.4 0.5 0.6 0.7 0.8 CMS Had 5.02 ATLAS Had 5.02 CMS Had 2.76 ATLAS Had 2.76 RHIC Had 0.20 CMS Jets R=0.2 2.76 CMS Jets R=0.3 2.76 CMS Jets R=0.4 2.76 ATLAS Jets R=0.4 5.02 ATLAS Jets R=0.4 2.76 κ 2 σ 1 σ Global Fit Hadrons 0-5% Jets 0-10%
* with LHC data only
In preparation
15
Daniel Pablos McGill / JETSCAPE
Consistent, but some tension between hadrons & jets preferred value
16
Daniel Pablos McGill / JETSCAPE
0.2 0.4 0.6 0.8 1 1.2 1.4 10 100 1000 RAA Hadron or Jet pT [GeV] Hadrons Jets R = 0.4
In preparation
17
Daniel Pablos McGill / JETSCAPE
0.2 0.4 0.6 0.8 1 1.2 1.4 10 100 1000 RAA Hadron or Jet pT [GeV] Hadrons Jets R = 0.4
In preparation
High z enhancement
ATLAS-CONF-2017-005
Count the average number of hadrons, per jet, with energy fraction z ATLAS
18
Daniel Pablos McGill / JETSCAPE
0.6 0.8 1 1.2 1.4 1.6 1.8 0.5 1 1.5 2 2.5 3 3.5 4 Jet FFs ratio ln(1/z) Actual jet FFs
PbPb jet FFs
0.2 0.4 0.6 0.8 1 1.2 1.4 10 100 1000 RAA Hadron or Jet pT [GeV] Hadrons Jets R = 0.4 Jets ⊗ FF actual
High z enhancement
In preparation
Count the average number of hadrons, per jet, with energy fraction z HYBRID MODEL
19
Daniel Pablos McGill / JETSCAPE
Vacuum jet FFs
0.2 0.4 0.6 0.8 1 1.2 1.4 10 100 1000 RAA Hadron or Jet pT [GeV] Hadrons Jets R = 0.4 Jets ⊗ FF actual Jets ⊗ FF vacuum
0.6 0.8 1 1.2 1.4 1.6 1.8 0.5 1 1.5 2 2.5 3 3.5 4 Jet FFs ratio ln(1/z) Actual jet FFs Vacuum jet FFs
Flat FFs ratio PbPb jet FFs Jet substructure is important for jet quenching phenomenology
In preparation
20
Daniel Pablos McGill / JETSCAPE
Wider, more active jets lose more energy than narrower, hard fragmenting ones Steeply falling jet spectrum bias inclusive jet sample to narrower ones, explains high z enhancement High pT hadrons belong to such subsample
and so
AA
AA >
Effect seen in the literature, for different models,
Milhano & Zapp - EPJ ‘16 Brewer et al. - JHEP ‘18 Pablos et al. - JHEP ‘17
21
Daniel Pablos McGill / JETSCAPE
0.2 0.4 0.6 0.8 1 1.2 10 100 RAA Jet pT [GeV] R=0.1 R=0.2 R=0.3 R=0.4 R=0.5 R=0.6 R=0.7 R=0.8
Wider jets lose more energy decreases with increasing jet radius
Rjet
AA
WITHOUT MEDIUM RESPONSE
Pablos et al. - JHEP ‘16
22
Daniel Pablos McGill / JETSCAPE
Opening the jet radius allows the recovery of some of the lost energy WITH MEDIUM RESPONSE
0.2 0.4 0.6 0.8 1 1.2 10 100 RAA Jet pT [GeV] R=0.1 R=0.2 R=0.3 R=0.4 R=0.5 R=0.6 R=0.7 R=0.8
Hint of ordering reversal around R ∼ 0.7 Characteristic behaviour of strong coupling: efficient energy transfer into hydro modes Upcoming precise data from CMS Stay tuned!
Pablos et al. - JHEP ‘16
23
0.05 0.1 0.15 0.2 0.25 5 10 15 20 25 0 − 10% R = 0.4 60 < PT, ch jet < 80 GeV Event Fraction Mch jet (GeV) Back No Back pp reference (PYTHIA) ALICE Data R=0.4 0.05 0.1 0.15 0.2 0.25 5 10 15 20 25 0 − 10% R = 0.4 80 < PT, ch jet < 100 GeV Event Fraction Mch jet (GeV) Back No Back pp reference (PYTHIA) ALICE Data R=0.4 0.05 0.1 0.15 0.2 0.25 5 10 15 20 25 0 − 10% R = 0.4 100 < PT, ch jet < 120 GeV Event Fraction Mch jet (GeV) Back No Back pp reference (PYTHIA) ALICE Data R=0.4
quenching back-reaction
cancellation between two effects
M = q E2 − p2
T − p2 z
jet narrowing reduces jet mass soft particles at edges rapidly increase mass
In preparation
Daniel Pablos McGill / JETSCAPE
24
Grooming techniques:
Measured anti-kT jet Soft Drop condition: Analytically well understood observable: strongly relates to QCD splitting function Provides momentum balance between the two groomed subjets
pT 1 pT 2
Remove large angle & soft:
zcut = 0.1
β = 0
Taken from M. Verweij’s slides @ MIT HI workshop ‘16 Larkoski et al. - JHEP ‘14, PRD ‘15
Daniel Pablos McGill / JETSCAPE
25
Count the number of times that the same jet satisfies the Soft Drop condition
0.1 0.2 0.3 0.4 0.5 0.6 1 2 3 4 5 6 7 R = 0.4 80 < P ch
T,jet < 120 GeV
1/Njets dN/dnSD nSD PYTHIA Hybrid Model
No enhancement in the # splittings passing Soft Drop in medium Suppression of wide structures tends to slightly reduce nSD
In preparation
(ALICE) in QM ‘18
Daniel Pablos McGill / JETSCAPE
26
Wide structure suppression modifies probability
Data not unfolded:
necessarily cancel in the ratio!
0.5 1 1.5 2 0.1 0.2 0.3 0.4 0.5 0 − 5% R = 0.4, 80 < P ch
T,jet < 120 GeV
1/Njets dN/dzg (PbPb/pp) zg all ∆R ∆R < 0.1 ∆R > 0.2
∆R ≡ R12
angular cuts Shape of the distribution not modified because our model assumes vacuum-like shower
In preparation
Daniel Pablos McGill / JETSCAPE
27
1 2 3 4 5 ln(1/∆R) −10 −8 −6 −4 −2 ln(z ∆R) −0.1 −0.08 −0.06 −0.04 −0.02 0.02 0.04 0.06 0.08 0.1
In preparation
Suppression Enhancement
Fill a 2D density map by using both momentum balance (zg) and angular separation (ΔR) PbPb − pp A suppression of large angle splittings and enhancement of collinear splittings is observed - consistent with observation in zg measurement
Daniel Pablos McGill / JETSCAPE
28
energy loss at strong coupling is a necessary tool to assess the nature of QGP dynamics degree of hydrodynamization of lost energy can be tested with current observables much progress has been made in developing models that can be compared to data further effort is needed on bringing holographic models to a next level of sophistication
by taking the core ingredients from pQCD that apply in HIC due to scale separation need a systematic confrontation with data within a common framework
v1.0 has been now released!
https://github.com/JETSCAPE
Modular simulator of all aspects of heavy ion collisions Energy loss modules: MATTER, LBT, MARTINI, AdS/CFT Will soon feature concurrent Jet+Hydro evolution!
Daniel Pablos McGill / JETSCAPE
29
Can we constrain QGP transport coefficients by confronting to data the effect of medium response to the presence of the jet? (or, inversely, can we constrain e-loss models based on their consistency with extracted transport coefficients from minimum bias soft physics?) Hydro can describe flow in small systems, but there is absence of jet quenching… Strong coupling e-loss has a strong path length dependence, which means very small quenching in pPb, but… how do we explain high pT v2? Strong coupling prescription determines the rate of energy transfer from hard (and hardish) modes into hydro modes If jet hydrodynamization is related to hydrodynamic behaviour appearance itself… can we use holography insights to develop a model of mini-jet quenching for the bulk?
Daniel Pablos McGill / JETSCAPE
Daniel Pablos McGill / JETSCAPE
Core features of the model have been validated by e.g. photon-jet observables predictions No strong evidence so far of hard point-like scatterers Talk by R. Bi on Wed
31
Daniel Pablos McGill / JETSCAPE
J
A
0.1 0.2 0.3 0.4
[GeV] 〉
|| T
p 〈
20 40
PbPb - pp R = 0.3, 0-10%
J
A
0.1 0.2 0.3 0.4
20 40
∆
〉
|| T
p 〈 0.5-1.0 1.0-2.0 2.0-4.0 4.0-8.0 8.0-300.0
PbPb - pp R = 0.3, 10-30%
Daniel Pablos McGill / JETSCAPE
40 30 20 10 10 0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 Quenching Only h/ pk
T i
∆ 40 30 20 10 10 0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 PbPb R=0.3, 0-30% Quenching + Medium Response h/ pk
T i
∆ 8.0-300.0 4.0-8.0 2.0-4.0 1.0-2.0 0.5-1.0 h/ pk
T i∆
‘missing-pt’ observables
40 20 20 40 0.1 0.2 0.3 0.4 PbPb-pp R=0.3, 0-10% 0.1 0.2 0.3 0.4 PbPb-pp R=0.3, 10-30% h/ pk
T i [GeV]
AJ AJ 8.0-300.0 4.0-8.0 2.0-4.0 1.0-2.0 0.5-1.0 h/ pk
T iΣ
CMS
CMS
leading jet subleading jet
energy is recovered at large angles in the form of soft particles data suggests that implementation of back-reaction might mistreat semi-hard particles
32
J
A
0.1 0.2 0.3 0.4
[GeV] 〉
|| T
p 〈
20 40
PbPb - pp R = 0.3, 0-10%
J
A
0.1 0.2 0.3 0.4
20 40
∆
〉
|| T
p 〈 0.5-1.0 1.0-2.0 2.0-4.0 4.0-8.0 8.0-300.0
PbPb - pp R = 0.3, 10-30%
Daniel Pablos McGill / JETSCAPE
40 30 20 10 10 0.1 0.3 0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 PbPb R=0.3, 0-30% Quenching + Medium Response h/ pk
T i
∆ 8.0-300.0 4.0-8.0 2.0-4.0 1.0-2.0 0.5-1.0 h/ pk
T i∆
‘missing-pt’ observables
40 20 20 40 0.1 0.2 0.3 0.4 PbPb-pp R=0.3, 0-10% 0.1 0.2 0.3 0.4 PbPb-pp R=0.3, 10-30% h/ pk
T i [GeV]
AJ AJ 8.0-300.0 4.0-8.0 2.0-4.0 1.0-2.0 0.5-1.0 h/ pk
T iΣ
CMS
CMS
leading jet subleading jet
energy is recovered at large angles in the form of soft particles data suggests that implementation of back-reaction might mistreat semi-hard particles
CMS
33
Weak coupling: interplay between antenna angle, formation time and emission wavelength medium interactions can destroy antenna color correlations radiation from the global charge only if system not resolved by QGP Strong coupling: quark-gluon system emulated by string with kink stopping distance modulated by angular separation between endpoint & kink In Hybrid Model: unresolved dipoles lose energy as a single effective excitation two partons are resolved if their separation is greater than resolution length
Casalderrey & Ficnar - arXiv:1512.00371 Hulcher et al. - JHEP ‘18 Mehtar-Tani et al. - PLB ‘12
needs further study!
Casalderrey & Iancu - JHEP ‘11 Casalderrey et al. - PLB ‘13
Daniel Pablos McGill / JETSCAPE
34
Daniel Pablos McGill / JETSCAPE
0.3 0.4 0.5 0.6 0.7 0.8 CMS Had 5.02 ATLAS Had 5.02 CMS Had 2.76 ATLAS Had 2.76 RHIC Had 0.20 CMS Jets R=0.2 2.76 CMS Jets R=0.3 2.76 CMS Jets R=0.4 2.76 ATLAS Jets R=0.4 5.02 ATLAS Jets R=0.4 2.76 κ 2 σ 1 σ Global Fit
With increasing , hadrons & jets preferred value is more similar…
Lres = 2/πT
Lres
*
Hadrons 0-5% Jets 0-10%
* with LHC data only
In preparation
35
36
Daniel Pablos McGill / JETSCAPE
κ ∈ {0.420, 0.445} (global fit)
0.2 0.4 0.6 0.8 1 1.2 10 100 RAA Hadron or Jet PT (GeV) √s = 2.76 ATeV
√s = 5.02 ATeV
37
Daniel Pablos McGill / JETSCAPE
0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0.01 0.1 1 125 < pjet
T
< 160 GeV |y| < 2.1, R = 0.4 PbPb/pp z ATLAS Prelim. Data Lres = 0 Lres = 2/πT
(global fits)
38
Daniel Pablos McGill / JETSCAPE
all ΔR
Not unfolded data: need to smear theory results
39
Daniel Pablos McGill / JETSCAPE
0.1 0.2 0.3 0.4 0.5 zg 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Rg 5 10 15 20 25 30 35 40
Correlation between momentum balance & subjet angular separation Absolute Normalisation pp (PYTHIA)
40
Daniel Pablos McGill / JETSCAPE
Correlation between momentum balance & subjet angular separation Separately normalised for each zg
0.1 0.2 0.3 0.4 0.5 zg 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Rg 20 40 60 80 100 120
pp (PYTHIA)
41
Daniel Pablos McGill / JETSCAPE
2 4 6 8 10 12 14 16 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 R = 0.4 80 < P ch
T,jet < 120 GeV
1/NRg dN/dRg Rg nSD=1 nSD=2 nSD=3 nSD=4 nSD=5 nSD=6
Strong correlation between subjet angular separation and number of Soft Drop splitting Suppression of wide structures Reduction of nSD pp (PYTHIA)
42
Daniel Pablos McGill / JETSCAPE
Rough independence of momentum balance w.r.t. nSD Suppression of wide structures Not necessarily modifies zg
1 2 3 4 5 6 7 8 0.1 0.2 0.3 0.4 0.5 R = 0.4 80 < P ch
T,jet < 120 GeV
1/Nzg dN/dzg zg nSD=1 nSD=2 nSD=3 nSD=4 nSD=5 nSD=6
pp (PYTHIA)
Daniel Pablos McGill / JETSCAPE smallest angular separation between two jets that the medium can resolve? assign a transverse structure to the string such that a quark-gluon system is emulated holographic description of 3-jet events study the stopping distances as a function of
different scaling than pQCD in a dense plasma
θpQCD
res
∝ E−3/4
Casalderrey & Ficnar - arXiv:1512.00371
43
competing effects: each individual jet widens, while wider jets lose more energy the string is treated as a model for the jet as a whole consider ensemble of jets by choosing initial distributions of energy & angle from pQCD
effect also observed in pQCD
measures jet angle in pQCD for the same jet suppression different final angle dist.
TSYM = b TQCD
44
Daniel Pablos McGill / JETSCAPE
Rajagopal et al. - PRL ‘16 Milhano & Zapp - EPJ ‘16
use pp jet shapes to determine angle distribution nuclear jet shape modification captures core dynamics - lacks contribution from medium response
a ∈ {1.8, 2.5}
as the string nullifies, different initial choices tend to converge
45
Daniel Pablos McGill / JETSCAPE
determine string energy density by considering different initial profiles evolved within full string dynamics
Brewer et al. - JHEP ‘18
Daniel Pablos McGill / JETSCAPE the heavy quark transfers energy to the plasma via frictional drag external force required to keep quark’s velocity constant need to consider the fluctuations within the thermal bath for a sensible phenomenology
zh zq
bdy
(coloured noise) (validity regime)
l i m i t
v a l i d i t y ?
∝ M
Herzog et al., Gubser, Casalderrey&Teaney ‘06 Casalderrey-Solana & Teaney ‘06, Gubser ‘07 Horowitz ‘15
46
Daniel Pablos McGill / JETSCAPE compute distance square travelled by the endpoint as it falls in holographic direction
faster than speed limit wobbly limp noodle interpolation between heavy quark and light quark results
(2 − 4)π √ λT 3
2π √ λT 3 v
a(t)
speed at which endpoint falls ˆ qLRW ' 7.5 p λT 3
ˆ qGubser = 2π √ λT 3 v √γ
light quark heavy quark
v = 0
(late times)
Moerman & Horowitz ‘16
47
Daniel Pablos McGill / JETSCAPE light quark endpoint can fall unimpeded towards the black brane external boosted U(1) fields semiclassical string description Arnold & Vaman ‘11 Chesler et al. ‘09
κsc ∝ λ0
robust result at strong coupling zq → 0
48
Daniel Pablos McGill / JETSCAPE from Gubser's ‘14 talk in ECT*
finite endpoint momentum strings: energy loss = energy flux from endpoint to bulk of string including higher derivative corrections (Gauss-Bonnet), but need to rescale LHC temperature string endpoint starts close to the horizon, then stops at boundary and falls back again
Ficnar et al. ‘14
49
Daniel Pablos McGill / JETSCAPE
0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0.05 0.1 0.15 0.2 0.25 0.3 0-10% Centrality 100 < P jet
T
< 300 GeV 0.3 < |η| < 2, r < 0.3 P parton
T
> 1 GeV PbPb/pp r K=100 K=40 K=20 K=0 CMS Data 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0.2 0.4 0.6 0.8 1 0-10% Centrality 5 < P hadron
T
< 10 GeV PbPb/pp r K=100 K=40 K=20 K=0
Inclusive jets - all tracks Subleading jets - semi-hard tracks strong quenching suppresses the effect of broadening early wide fragments quenched late narrow fragments survive selection bias towards narrower jets, merely a jet axis deflection kinematical limits chosen such that:
deviations from such Gaussian broadening hard momentum transfers from QGP quasiparticles
50