progress in jet physics at colliders from the lhc to eic · 2018-07-23 · progress in jet physics...

Progress in jet physics at colliders from the LHC to EIC

Xiaohui Liu

Probing Quark-Gluon Matter with Jets @ BNL 2018

1

2

Status of arts

Available NNLO predictions up to 2-jets

• Rapid progress in jet physics

• NNLO

3

Status of arts


• NNLO Fixed order

• Improved predictive power

• Potentially useful for studying jet quenching with V + 1j

• Good for pT, HT spectrum … May not be sufficient when sensitive to high multiplicities

Radja, XL, Petriello


• Boost in the jet substructure

• Sensitive to multiple emissions

• Resummation

4

Status of arts

- Jet shape Li, Li, Yuan, Chien, Vitev

- Jet mass/angularity Larkoski, Schwartz, Kang, Lee, Ringer…

- Jet radius Dasgupta, Salam, Soyez, Kang, Ringer, Vitev, Liu, Moch…

- Jet grooming/pruning/trimming

- NGLs

- …

Larkoski, Marzani, Soyez, Thaler …

Dasgupta, Salam, BMS, Caron-Huot … Becher, Neubert, Shao, Larkoski, Moult, Neill …

- Jet shape

- Jet mass/angularity

- Jet radius

- Jet grooming/pruning/trimming

- NGLs

- …

• Single inclusive jet production

• small R + threshold

• Jet-lepton pT imbalance

• small R + small pT

5

This talk

6

Single inclusive jet production

Liu, Felix, Moch

• Benchmark process at the LHC

7

Inclusive jet production

• Still large discrepancies and uncertainties for different PDF fittings in large x region.

• Can be constrained by near future inclusive jet pT analysis.

• alphas extraction

• PDF fitting

• Current Status of arts

8


• Theory is systematically higher than the data for all rap bins.

• Also seen in other LHC inclusive jet analyses, with machine energy ranging from 2.76TeV to 13TeV

• NLO known for decades: NLOJET++ (Nagy, 02), (M)EKS (Ellis,

Kunszt, Soper, 92) …

CMS, 1406.0324


9


• R-dependent discrepancies between the theory and the data



CMS, 1609.05383


• NNLO with leading color

approximation

10

scale = individual jet pT scale = leading jet pT

Currie, Glover, Pires, 17

• NNLO theory seems to describe the data with a specific scale choice, (still systematically higher than the data).

• Slightly different scale choices lead to different conclusions.





• NNLO with leading color approximation

11


“While no cuts were applied to all jet datasets included in NNPDF3.1, for the 2011 ATLAS 7 TeV dataset a good agreement between data and theory was obtained when fitting only the central rapidity bin, |yjet| < 0.5. Concurrently, it was found that achieving a good description of the ATLAS 2011 7 TeV data set would be impossible, if all five rapidity bins were included simultaneously and if all cross-correlations among rapidity bins were taken into account accordingly.”

Nocera and Ubiali, 1709.09690


• persistent tension between NLO/NNLO theory and the LHC

data

12


Large higher order corrections?

• Possible large corrections

• small R

• threshold

13


X

m=0,k=1

↵ns

"ln2n�m�k z

z

#

+

lnm R

luminosity

energy

machine energy

seems away from the threshold

but enhanced by the steep falling of the luminosity

z measures the invariant mass outside the signal jet, characterize the distance to the threshold

• Possible large corrections

• small R

• threshold

14


X

m=0,k=1

↵ns

"ln2n�m�k z

z

#

+

lnm R

800 1000 1200 1400

0.4

0.6

0.8

1.0

1.2

pT @GeVDKR

K0.40.8 ¥ K0.21.2 ¥ K0.6

NLO singular /NLO full

singular terms make up the dominant bulk of the NLO result.

Or, threshold gives a very good approximation of the full calculation

• Factorization and resummation

15


X

m=0,k=1

↵ns

"ln2n�m�k z

z

#

+

lnm R

derived within SCET + Becher, Neubert, Rothen, Shao …

lnR = lnpTR

pT= ln

pTR

µ+ ln

µ

pT

• Factorization and resummation

16


LO

NLO

NNLO

... ... ... ... ... ...

NkLO ...

Fixed Order

Resum

mation

1

↵sL2 ↵sL ↵s

↵2sL

4 ↵2sL

3 ↵2sL

2 ↵2sL ↵2

s

↵ksL

2k ↵ksL

2k�1 ↵ksL

2k�2 ↵ksL

2k�3 ↵ksL

2k�4

LL NLL NNLL

� = �NLO � �sin. + �NLL

• Phenomenology

• compare with 2.76 TeV LHC data

17


160 180 200 220 240 260 280 3000.00.20.40.60.81.01.21.4

pT @GeVD

sêsNLO

R = 0.2

160 180 200 220 240 260 280 3000.00.20.40.60.81.01.21.4

pT @GeVD

R = 0.3

160 180 200 220 240 260 280 3000.00.20.40.60.81.01.21.4

pT @GeVD

R = 0.4

• NLO consistently higher than the data for all choices of radii.

• NLL + NLO agrees very well with the data

with NLO CT10 PDF sets NP effects included

XL, Moch, Ringer, 2017

NLO using MEKS, Gao, et. al. 2012

• Phenomenology

• compare with 7 TeV LHC data

18


102 1030.4

0.6

0.8

1.0

1.2

1.4

1.6

pT @GeVD

sêsNLL+NLO

0.0 b »y» <0.5102 103

0.4

0.6

0.8

1.0

1.2

1.4

1.6

pT @GeVD

sêsNLL+NLO

0.5 b »y» <1.0

102 1030.4

0.6

0.8

1.0

1.2

1.4

1.6

pT @GeVD

sêsNLO

0.0 b »y» <0.5102 103

0.4

0.6

0.8

1.0

1.2

1.4

1.6

pT @GeVD

sêsNLO

0.5 b »y» <1.0

• NLO is higher than the data while NLL + NLO agrees well

with NLO CT10 PDF sets NP effects included

similar story for higher

rapidity bins

• Phenomenology

• Impact of PDFs

19


• PDF uncertainty is small

• Considering the impact of PDFs does not change the conclusion: resummation improves the theory prediction quite a lot.

102 1030.4

0.6

0.8

1.0

1.2

1.4

1.6

pT @GeVD

sêsCT10

NLO+NLL, 0.0 b »y» < 0.5 û 7TeV

CT14DATA

ABMP16NNPDF

102 1030.4

0.6

0.8

1.0

1.2

1.4

1.6

pT @GeVDsêsCT10

NLO, 0.0 b »y» < 0.5 û 7TeV

ABMP16NNPDF

CT14DATA

• Phenomenology

• compare with 7 TeV LHC data

20


102 1030.0

0.5

1.0

1.5

2.0

pT @GeVD

DR=sHRLêsHR=0.5L

NLO+NLL, 0.0 b »y» <0.5 û 7TeV

D0.2D0.5

D0.7 ¥NPD0.9

DATA D0.7

102 1030.0

0.5

1.0

1.5

2.0

pT @GeVDDR=sHRLêsHR=0.5L

NLO, 0.0 b »y» <0.5 û 7TeV

D0.2D0.5

D0.7 ¥NPD0.9

DATA D0.7

• Take the ratio to reduce the PDF impacts

• NLO does not describe the data while NLO + NLL does a much better job

21

Jet-lepton pT imbalance

Liu, Felix, Yuan in preparation

• Good for spin asymmetries and eA collisions

22


Requires TMD resummation and the small jet radius resummation

COM or laboratory frame; analogy to pp collisions

• Factorization

23


Rapidity regulator by Chiu, Jain, Neill, Rothstein

• Factorization

24


All components have been calculated to NLO

• Each function depends only on one single scale.

• Rapidity regulator only in the beam and the global soft function.

• Can check to see all poles cancel.

• Stay tuned for numerics.

• Framework allows to do small R resummation joint with threshold and small pT.

• Can achieve high precision in the future, eg. NNLO + NNLL.

• Applied to the inclusive jet production at the LHC.

• Tensions reduced when joint resummation is considered.

25

Conclusions

progress in jet physics at colliders from the lhc to eic · 2018-07-23 · progress in jet physics...

Documents