LHeC and Physics Beyond the Standard Model
Emmanuelle Perez
CERN
• Sensitivity to new physics in ep collisions at 1.4 TeV :
quark radius, electron-quark resonances,
eeqq contact interactions.
• LHeC w.r.t. the interpretation of LHC discoveries :
are there limitations due to our limited
knowledge of high x pdfs ?
• Crude studies so far, and only a few topics.
LHeC Steering Committee meeting, 27 Oct. 2007
1
3 June 2006
Beyond the Standard Model in ep collisions : introductory remarks
• Since ep is not an annihilation machine, the cross-section is “small”
for (heavy) pair-produced new particles.
 Investigate single particle production instead
 The sensitivity depends, in addition to the new particle mass, on
its coupling to SM particles.
 In case of a discovery: information not only on M but also on this
coupling.
Examples : Leptoquarks, Rp-violating SUSY, excited fermions, …
• HERA vs. Tevatron showed nice examples of the complementarity between
both facilities, in such model-driven BSM searches.
• Once LHC has discovered “something”, additional data from ep and/or ee
may be necessary to tell what the “something” is.
E. Perez
2
October 07
DIS at highest Q2 : towards quark substructure ?
Assign a finite size < r > to the
EW charge distributions :
d/dQ2 = SMvalue x f(Q2)
f(Q ) = 1 2
< r2 >
6
2
Q
Global fit of PDFs and < r > using d/dxdQ2
from LHeC simulation, 10 fb-1 per charge,
Q2 up to 500000 GeV2 :
< rq > < 8. 10-20 m
One order of mag. better than current bounds.
At LHC : quark substructure may be seen as a deviation in the dijet spectrum.
Such effects could also be due to e.g. a very heavy resonance.
Could we establish quark substructure with pp data only ?
E. Perez
3
October 07
Electron-quark resonances
Apparent symmetry between the lepton & quark sectors ?
Exact cancellation of QED triangular anomaly ?
• “Leptoquarks” (LQs) appear in many extensions of SM
• Scalar or Vector color triplet bosons
• Carry both L and B, frac. em. charge
ep
pp
difficult, large
backgrounds
eq
q
llqq
lqq
qq
NC DIS
CC DIS
Z/DY + jj
QCD
W + jj
W/Z + jj
QCD
 (unknown) coupling l-q-LQ
LQ decays into (lq) or (q) :
• ep : resonant peak, ang. distr.
• pp : high ET lljj events
Such eq resonances also predicted in Supersymmetry with “R-parity” violation:
e+
e+ and e- probe different
~
~
u Lj ( dRk*)
squarks & couplings
’1jk
+ p,
Example:
resonant
stop
production
in
e
Rp
- p.
resonant
sbottom
in
e
_
dk ( u j )
E. Perez
4
October 07
 = BR( LQ  eq )
• Tevatron probes large masses for large
 (LQ  eq) independently of .
But large  (LQ  q) is difficult.
 = BR (LQ  eq)
“Leptoquarks” : pp versus ep
• HERA better probes LQs with small 
provided that  not too small
• LHC vs LHeC :
MLQ (GeV)

A.F. Zarnecki
LHC could discover eq resonances
with a mass of up to 1.5 – 2 TeV
via pair production.
em
LHeC
E. Perez
1 TeV
LHC
pair
prod
MLQ (GeV)
Quantum numbers ? Might be
difficult to determine in this
mode.
5
October 07
Determination of LQ properties
pp, pair production
e+
_
q or q ?
• Fermion
number
• Chiral
couplings
E. Perez
F = -1
+
e,
e-
_
q or q ?
• Scalar
or
Vector
ep, resonant production
F = +1
_
_
qq  g  LQ LQ :
angular distributions
depend on the structure
of g-LQ-LQ. If coupling
similar to WW, vector LQs
would be produced unpolarised…
q
F=0 LQs : (e+) higher
F=2 LQs : (e- ) higher
(high x i.e. mostly q in initial state)
e
*
e
q
q
cos(*)
distribution
gives the
LQ spin.
Play with lepton beam
polarisation.
?
6
October 07
ep : golden machine to study LQ properties
F = 0 or 2 ?
Spin ?
Chiral couplings ?
Couples to  ?
Compare rates in e-p and e+p
Angular distributions
Play with polarisation of lepton beam
Easy to see since good S/B in j channel
F=0
F=2
Classification in the table below relies on minimal assumptions.
ep observables would allow to disentangle most of the possibilities (having
a polarised p beam would complete the picture).
E. Perez
7
October 07
“LQ spectroscopy” : LHeC versus single LQ production at LHC
Single LQ production at LHC brings information in principle. But cross-section
e.g. at 1 TeV is a factor ~ 400 lower than at LHeC.
Idem for the simultaneous determination
of coupling  at e-q-LQ and the quark flavor q.
If LHC observes a LQ-like resonance,
M < 1 TeV, with indications (single prod)
that  not too small, LHeC would solve the
possibly remaining ambiguities.
E. Perez
8
Asymmetry
Example : Fermion Number F.
Look at signal separately when resonance is formed
by (e+ + jet) and (e- + jet).
Sign of the asymmetry gives F, but could
be statistically limited at LHC. Easier in ep !
q
e+
F=0
g
•

q
e-
LHeC: 10 fb-1 per charge
 = 0.1
LHC: 100 fb-1
07
MLQOctober
(GeV)
Beyond LQs… To be further studied :
 SUSY with R-parity conservation
e
 in pb, e- p
q
~
e
0
~
q
Pair production via
t-channel exchange of
a neutralino.
Cross-section sizeable when
M < 1 TeV i.e. if squarks
are “light”, could observe
selectrons up to ~ 500 GeV.
 in pb, e+ p
 Single production of excited fermions
E. Perez
9
- Could extend a bit over the
LHC slepton sensitivity
- Possible information on
couplings by playing with
e+ / e- / L / R
October 07
 Further models …
Esp. models which predict new effects
- “easy” to see in ep because of low, well-understood backgrounds
- more difficult to establish in pp because of large bckgs
Example from HERA experience:
Anomalous top production via
BSM coupling tu.
Single top analysis at Tevatron is
difficult (large W+jets bckgd).
E. Perez
10
October 07
p structure & interpretation of LHC discoveries
The interpretation of discoveries in AA at Alice may require direct
measurements on pdfs in A – not covered here.
Here, focus on ATLAS & CMS discoveries : highest masses  highest x.
Constraints on d and g at high x still limited :
E. Perez
11
October 07
_
Current high x uncertainties and NP processes at LHC : qq processes
Example: new W ’, resonant slepton production in RpV SUSY
(reach for a W’ with
SM like couplings)
d
_
u
CMS
Physics
TDR
Vol II
• W’
RpV SUSY : reach would depend on the
strength of the coupling ’ .
s
_
With sea quarks involved, uncertainties
c
large already well below the kinematical limit.
40% uncertainty
on part. lum. For
a 6 TeV W ’.
 g(W ’) ?
• ~
’
Would make the measurement of the coupling difficult.
E. Perez
12
October 07
Drell-Yan mass spectrum
4 TeV
8 TeV
(shown uncertainties:
from CTEQ 61 sets)
CMS Physics TDR Vol II
Example : new Z’ boson, KK gravitons
in Randall-Sundrum models etc.. Signal = a mass peak.
Partonic luminosities can be “normalised” to the side-bands data if enough stat.
But close to the discovery limit, couplings of a Z’ boson may not be
measured accurately.
E. Perez
13
October 07
DY mass spectrum and “Contact Interactions”
NP in Drell-Yan spectrum might not manifest itself as a mass peak…
e.g. large extra-dimensions, interference with very heavy boson etc…
Effective “contact-term” Lagrangian :
d/ds = SMvalue + …  s /2 + … ( s / 2) 2
LHeC
HERA
VV model
VV model
LHeC sensitivity
(10 fb-1 e- & e+) :
25 – 45 TeV
depending on the model
Similar to the
expected sensitivity
at LHC.
 / 2
E. Perez
 / 2
14
(GeV-2)
October 07
eeqq Contact interactions in DY and DIS
• LL model,  = 30 TeV, sign = -1 :
effect in DY can be “absorbed” in pdf unc.
• In some cases, may be difficult to
determine the sign of the interference
of the new amplitude with SM.
LHeC, e- p, LL
• At LHeC, sign of the interference
can be determined by
looking at the asym. between
/SM in e- and e+.
Polarisation can further help
disentangle various models.
E. Perez
15
October 07
High x gluon and dijets at LHC
Some NP models predict deviations in dijet mass spectrum at high mass.
Example : some extra-dimension models.
Mc = 2 TeV
Mc = 4 TeV
S. Ferrag,
hep-ph/0407303
Mjj (GeV)
Mjj (GeV)
Due to pdf uncertainties, sensitivity to compactification scales reduced
from 6 TeV to 2 TeV in this example.
[ sensitivity has to be checked taking into account jet energy scale uncert… ]
E. Perez
16
October 07
Conclusions
For “new physics” phenomena “coupling” directly electrons and quarks
(e.g. leptoquarks, eeqq contact interactions) : LHeC has a sensitivity
similar to that of LHC.
The further study, in ep, of such phenomena would bring important
insights : leptoquark quantum numbers, structure of the “eeqq” new
interaction. These studies may be difficult, if possible at all, in pp.
LHC sensitivity to new (directly produced) particles not much limited
by our pdf knowledge. “Contact-interactions” deviations may be more
demanding.
However, the interpretation of discoveries at LHC may require a better
knowledge of the high x pdfs : e.g. determination of the couplings of a
W ’ or Z ’ if “at the edge” .
E. Perez
17
October 07
Work ahead … (not an exhaustive list)
- Determination of LQ quantum numbers at LHC : real MC analysis with realistic
simulation & backgrounds.
- Contact interactions : systematic analysis of
- how pp data could discriminate between various models.
- the complementarity between pp, ep, ee (cf A. Zarnecki, Tevatron/LEP/Hera)
- Assess the limitations due to our poor knowledge of high-x gluon in
searches with jets.
- Further study of the LHeC potential especially in Supersymmetry.
If slepton + squark accessible at LHeC, what additional information do we
learn compared to LHC ?
Need to involve people familiar with analysis in Atlas / CMS.
Establish contacts on BSM at LHeC with theorists / phenomenologists.
E. Perez
18
October 07
Backups
E. Perez
19
October 07
 (pb)
Single LQ production at LHC
Single LQ production also possible at
the LHC.
g
ep
q
pp
LQ
•

e-
g
q
•

q
LQ
e-
  ee followed by eq -> LQ not
considered yet. Not expected to change
much the results shown here (Tevatron).
200 GeV
MLQ (GeV)
1200 GeV Smaller x-section than at LHeC.
And large background from Z + 1 jet.
Can be used in principle to determine the LQ properties in pp.
E. Perez
20
October 07
Current high x uncertainties and NP processes at
LHC : quark-quark processes
Example: squark production
d
~
g
d
~
d
~
d
(shown uncertainties:
from CTEQ 61 sets)
(pdf) on the relevant partonic luminosities instead of that on the  of a given BSM process.
But better if couple
to u quarks as well…
Estimated
LHC
sensitivity
(SUSY)
For a process involving high x d quarks, pdf uncertainty ~ 20% at the
corner of the LHC phase space.
Could be ~ 50% with extended sensitivity (e.g. LHC upgrade)
E. Perez
21
October 07