NuTeV – charged, neutral currents induced by neutrinosNew measurement of Weinberg angle Possible “New Physics” Beyond Standard Model?

Normal (“QCD”) Explanations of NuTeV Anomaly? radiative correctionsparton Charge Symmetry Violationstrange quark momentum asymmetry

nuclear, shadowing corrections

Implications of NuTeV Standard Model Test Beyond the Standard Model?? “QCD Effects”??

Tim Londergan Nuclear Theory Center, Indiana UniversityPAVI06 Milos, Greece May 16-20, 2006 Supported by NSF PHY- 0244822

In collaboration with Tony Thomas (Adelaide/JLab)

The NuTeV Experiment: charged, neutral currents from neutrino DIS

800 GeV p at FNAL produce pi, K from interactions in BeO target; Decay of charged pi, K produces neutrinos, antineutrinos; Almost pure muon neutrinos; (small νe contamination from Ke3 decay)

Only neutrinos penetrate shielding

Dipoles select sign of charged meson:• Determine nu/nubar type•

NuTeV: Rochester/Columbia/FNAL/Cincinnati/Kansas State/Northwestern/Oregon/Pittsburgh neutrino collaborationG. Zeller etal, PRL 88, 091802 (02); PR D65, 111103 (02)

Separate Neutral, Charged-Current Events

Charged current: Track through several platesLarge visible energy deposit

Neutral current: Short visible trackLarge missing energy

NuTeV event selection: • Large E in calorimeter • event vertex in fiducial volume

NuTeV Events: • 1.62 million • 351,000

NuTeV Detector: 18 m long, 690-ton steel scintillator; Steel plates interspersed with liq scintillator, drift chambers

The Paschos-Wolfenstein Ratio: Neutrino Total Cross Sections on Isoscalar Target:

Paschos & Wolfenstein ( PR D7, 91 (73)): Independent measurement of Weinberg angle, using ratio of total X-sections for neutrinos, antineutrinos on isoscalar target: PW ratio minimizes sensitivity to PDFs, higher-order corrections

The Paschos-Wolfenstein Ratio:

PW Ratio depends on the following assumptions: • Isoscalar target (N=Z)

• include only light (u, d) quarks

• neglect heavy quark masses

• assume isospin symmetry for PDFs

• no nuclear effects (parton shadowing, EMC, ….)

• no contributions outside Standard Model

NuTeV Determination of Weinberg Angle: • Construct ratios Individual ratios less dependent on overall normalization Very precise charged/neutral current ratios: Different cuts, acceptance: don’t construct PW ratio directly• : depends strongly on Weinberg angle

• : weak dependence on Weinberg angle

These ratios lead to a NuTeV value for the Weinberg angle:

The NuTeV result is ~ 3 σ above very precise value (from EW processes at LEP)

3σ below SM

agree with SM

NuTeV work at LO in QCD (with improvements) and find

s2w(NuTeV)=0.2276±0.0013stat ±0.0006syst ±0.0006th

-0.00003(Mt/GeV-175)+0.00032 lnMH/100GeV

where s2w=1-M2

w/M2z (on-shell)

Global fit: s2w= 0.2229 ±0.0004

~2.8σ discrepancyNuclear effects and ν oscillationexplanations very unlikely

NuTeV Result: 3σ Discrepancy from LEP value

Corresponds to 1.2% decrease in gL2;

This is large compared to precision of EW measurements!

SM

Low Q2 (~10-4 GeV2) atomic PV exp’t in Cs [PRL 82, 2484 (99)]Moller scatt exp’t (E158) at SLAC; Q2~ 0.026 GeV2 [PRL 95, 081601 (05)]

NuTeV exp’t, Q2 ~ 20 GeV2 [PRL 88, 091802 (02)]LEP, e+ e- at Z pole, Q2 = MZ

2

Qweak, pol e-p at JLab (proposed),

“Running” of Weinberg Angle Effective Weinberg angle depends on Q2

Warning: gauge and process dependent parameter

Qweak

E158: Q dependence of Weinberg angle confirmedat 6σ levelNuTeV: discrepancy of ~3σ with SM prediction [Czarnecki-Marciano, PRD53, 1066 (96)]

“New Physics” explanation for NuTeV?

The problem: extremely precise EW data confirms SM! • Mass, width of Z, W• X-sections, branching ratios at Z peak [LEP, SLD]• LR and FB asymmetries in e+e- scattering• new particles must satisfy all these constraints• EW constraints ~ 0.1% level [NuTeV ~ 1.2%] • new particles have heavy masses Q2 dependence must be slow• “new physics” hard to satisfy EW constraints!

NuTeV

“Designer Particles” I: delicately adjust to fit all existing data

• oblique corrections [high mass scale, couples only to vector bosons]: parameters constrained by EW data – can’t fit NuTeV

• extra Z’ (unmixed) – possible; runs into trouble with latest muon g-2

“Designer Particles” II: More Attempts to fit NuTeV

• minimal SUSY loops – No – most have wrong sign; others violate existing constraints

• Leptoquarks (bosons that couple to leptons & quarks): models with very carefully tuned mass splittings still possible – could be tested at Tevatron, LHC [Davidson, Gambino etal, J High E Phys 2, 37 (2002)]

MSSM, light sleptinos, gauginos Leptoquark (solid); extra gauge bosons (red)

• EM radiative corrections, process specific to NuTeV, are large– Bremsstrahlung from final state lepton in CC

significant correction.• Not present in NC (promotes CC events to

higher y so they pass energy cut) R , R, sin2W} ≈

{+.0074,+.0109,-.0030}

– New calculation of this effect (Diener, Dittmaier, Hollik PR D69, 073005 (04) )

• not yet implemented in analysis• Corrections likely larger than NuTeV estimate

EW Radiative Corrections

D. Yu. Bardin and V. A. Dokuchaeva, JINR-E2-86-260, (1986)

Isospin Violating Corrections to PW Ratio: Changes in PW ratio from isospin violating PDFs:

PW Correction valence parton charge symmetry violation (CSV)parton charge symmetry:

CS: (rotation of 180o about “2” axis in isospin space)

We know the origins of parton CSV: o quark mass difference:

o Electromagnetic contributions: most important EM effect:

n-p mass difference

CSV Contribution to NuTeV Result: Sather: Analytic Quark Model Approximation for Valence Parton CSV.

PW Ratio CSV Corr’n using Sather:Rodionov: Sather: CTEQ4LQ:

40% decreasein anomaly!

NuTeV: Don’t evaluate PW Ratio! CSV Contrib’n to NuTeV result:

Calculate parton CSV at low (quark model) momentum scale Evolve up to Q2 of NuTeV exp’t (20 GeV2) Evaluate with NuTeV functional

30% decrease in anomaly

Leads to analytic results(model-dependent)

Phenomenological Parton CSV PDFsMRST PDFs from global fits include CSV for 1st time:Martin, Roberts, Stirling, Thorne [Eur Phys J C35, 325 (04)]:

Choose restricted form for parton CSV:

Best fit: κ = -0.2, large uncertainty ! Best fit remarkably similar to quark model CSV calculations

MRST (2004) ADEL (1994)

[f(x): 0 integral; matches to valence PDFs at small, large x]

90% conf limit (κ)

“QED Splitting”: a New Source of Isospin Violation

MRST, Eur.Phys.J. 39, 155 (05); Glueck, Jimenez-Delgado, Reya, PRL95, 022002 (05)

“QED evolution”, quark radiates photonEvolve in Q2

• qualitatively similar to quark model CSV • QED varied while quarks “frozen”• contributes even if mu = md and Mn= Mp

• add to quark model CSV term • increase CSV ~ factor 2• MRST incorporate QED splitting with PDFs in global fit to high energy data

Phenomenology: MRST global fit limits valence CSV -0.8 ≤ κ ≤ +0.65 κ = -0.6 remove NuTeV anomaly! κ = +0.6 anomaly twice as large!

Theoretical estimates: quark model remove ~1/3 anomaly “QED splitting” remove ~1/3 at current limits, CSV could produce observable effects

(~ 5-6%) in certain reactions

Summary, CSV Effects on Weinberg angle

MRST global fit to valence CSV

90% confidence limits:

Strange Quark Contributions to PW Ratio: Contribution from strange quarks:

Strange quark normalization: constrained(no net strangeness in nucleon)

If s quarks carry more momentum than sbar decrease anomaly

Determination of s, sbar quark PDFs: Opposite sign dimuons from neutrinos

• CCFR: charge of faster muon determines neutrino or antineutrino; • most precise way to determine s, sbar PDFs CCFR, NuTeV

s-sbar momentum asymmetry

Analyses of s quark momentum asymmetry • NuTeV: analyzed s, sbar for small x: • Best fit: • Do not enforce normalization condition• NLO treatment of dimuon prod’n

• But, from normalization: • If s < sbar for small x, requires s > sbar for large x

CTEQ: [Kretzer etal, PRL 93, 041802 (04), Olness etal, Eur Phys J C40, 145 (05)]

• Global analysis of parton PDFs CTEQ6• Includes CCFR, NuTeV dimuon data • (includes expt’l cuts on dimuons)• Extract “best fit” for s, sbar dist’ns [enforce s normalization cond’n]

Catani et al. hep-ph/0404240; 0406338• in NNLO, can generate nonzero momentum asymmetry• find small mom asymmetry < 0 from 3-loop contrib’ns

positive [S-]

Results: CTEQ Global fit

vs. Bjorken x

• CTEQ: S- > 0, strange asymmetry decreases NuTeV anomaly;• dimuon data: most sensitive for s PDFs • CTEQ: s contrib’n removes 0 25% of anomaly

μ±

other

-.001 < [S-] < +.004

NuTeV Analysis of s, sbar Quark Dist’n:

Re-analyzed dimuon data: preliminary analysis [S-] ~ 0 appears negative strangeness not conserved? NLO treatment of dimuon prod’n

NuTeV group: now using CTEQ PDFs (u,d) very similar analysis method acceptance corrections? fragmentation functions? charm mass (CTEQ uses HERA data)?

Qualitative disagreement between NuTeV/CTEQ, using same data, techniques to extract s, sbar ??

Nuclear Effects ?? NuTeV: All data neutrino DIS on ironNuclear modification of PDFs: Shadowing EMC Effect Fermi motion ….Miller/Thomas Int J Mod Phys A20, 95 (05): Shadowing effect for NuTeV??

charged leptons: Vector meson dominance

CC events:

NC events:

Miller/Thomas: shadowing very different from

Zo: very small coupling to NC shadowing ~ 0

Different shadowing for account for anomaly??

NuTeV: NO ! Shadowing low Q2, data much higher Q2

value very close to SM value (should be quite different in MT scenario)

Nuclear effects in reactions: Hirai, Kumano, Nagai, PR D71, 113007 (05)

Shadowing: increase NuTeVanomaly??

Nuclear Effects (part II)

Brodsky, Schmidt, Yang (PR D70, 116003 (04)) Shadowing/Antishadowing ν-A processes

• include Pomeron, Odderon, Reggeon effects• obtain both shadowing, antishadowing effects • both constructive, destructive interference

• processes modified in different ways

Predict greater effect on

Remove ~ 20% of Weinberg angle anomaly Partial contribution along with CSV, s quarks ??

than

Conclusions: NuTeV: Measured CC, NC X-sections for on Fe

Large (~ 3σ), surprising discrepancy for

“New Physics” – difficult to fit LEP results, NuTeV “designer particles” unlikely very delicate “QCD Corrections” to NuTeV measurement??

• radiative corr’ns new calc’n, not in analyses• parton CSV ~ could remove effect [MRST]• strange quark asymmetry ~ 1-2 σ [CTEQ]• nuclear effects ~ 20%, Brodsky etal

NuTeV: no!!

CSV, strangeness at present, most plausible explanations for NuTeV anomaly

small additive contrib’ns from CSV, strangeness, nuclear

effects ??

(model dependence)

NuTeV – charged, neutral currents induced by neutrinosNew measurement of Weinberg angle Possible “New Physics” Beyond Standard Model?

Normal (“QCD”) Explanations of NuTeV Anomaly? radiative correctionsparton Charge Symmetry Violationstrange quark momentum asymmetry

nuclear, shadowing corrections

Implications of NuTeV Standard Model Test Beyond the Standard Model?? “QCD Effects”??

Tim Londergan Nuclear Theory Center, Indiana UniversityPAVI06 Milos, Greece May 16-20, 2006 Supported by NSF PHY- 0244822

In collaboration with Tony Thomas (Adelaide/JLab)

NuTeV/LEP/ PV MØller/ Qweak

E158 at SLACfirst measurementof PV in MØller sc.

huge luminosityhigh polarization

(~80%)

Suppressed sensitive to sin2θw

Large radiative corrections, ≈-40%Czarnecki-Marciano,Denner-Pozzorini,Petriello,Ferroglia et al

Very low Q2 atomic PV exp’t in CsJLab: Qweak e-p PV exp’t

Warning: gauge and process dependent parameterAt tree level, APV≈280 10-9

Qweak !!

“QCD Corrections” to the NuTeV Result:

( “Something Old”) Isoscalar Effect (N ≠Z for Fe)

Depends only on 2nd moment of light quark valence PDFs (momentum carried by up, down valence quarks) Isoscalar corrrection to PW ratio:

NuTeV Isoscalar correction:

• NuTeV exp’t doesn’t evaluate PW ratio • Isoscalar correction from Monte Carlo simulation of exp’t• Although NuTeV significant difference from PW corr’n, under control

At low Q2, generate from quark models:

residual (ud) diquark

Under CSV n p,

residual (ud) diquark

[only n-p mass difference]

Qualitative Features of Parton Distributions

[True for all intermediate states!]

“Majority” valence quark: “Minority” valence quark

residual (uu) diquark

residual (dd) diquark

[diquark mass difference]

Calculating CSV Corrections to PW Ratio: o Construct models for valence parton distributions o Test how these models behave under CSV transformation:

o Predict sign, magnitude of CSV effects in NuTeV exp’t

CSV Corrections to PW Ratio:

CSV correction to PW ratio independent of Q2

Both numerator, denominator involve 2nd moment of valence dist’ns numerator, denominator evolve identically with Q2

under LO evolution,

Evaluate valence PDFs at low Q2

o Dominated by configurations with 3 valence quarks o Easy to calculate, study behavior

Quantitative Estimates of Parton CSVQuark-model formula for parton PDFs:

Large x: dominated by X=diquark statesSather: study variation with nucleon, quark mass: analytic expression

small shift in PDFs, due to quark, nucleon mass. For large x, Sather predicts d > u in both cases. (From normalization, d < u at small x)

X = qq; 3q+qbar; 4q+2 qbar. ….

[work at low Q2 where 3-quark states dominate]