Probing Topcolor-Assisted from Top-Charm Associated Production at LHC

Junjie Caoa,b, Zhaohua Xiong c, Jin Min Yang b a CCAST (World Laboratory), P.O.Box 8730, Beijing 100080, China b Institute of Theoretical Physics, Academia Sinica, Beijing 100080, China c Graduate School of Science, Hiroshima University, Hiroshima 937-6256, Japan (April 2, 2003)

tc bb + +√2KUR∗ KDLc¯RbLπt + h.c. , (1) We propose to probe the topcolor-assisted technicolor i (TC2) model from the top-charm associated productions at where v 174 GeV, F 50 GeV is the top-pion decay the LHC, which are highly suppressed in the Standard Model. t constant' and the factor' v2 F 2/v reflects the effect Due to the flavor-changing couplings of the with − t the scalars (top-pions and top-Higgs) in TC2 model, the of the mixing between thep top-pions and the would-be top-charm associated productions can occur via both the s- goldstone bosons [8]. KUL, KDL and KUR are the rota- channel and t-channel parton processes by exchanging a scalar tion matrices that transform the weak eigenstates of left- field at the LHC. We examined these processes through Monte handed up-type, down-type and right-handed up-type Carlo simulation and found that they can reach the observ- quarks to the mass eigenstates, respectively. As pointed able level at the LHC in quite a large part of the parameter tc out in [6], the transition between tR and cR, KUR,canbe space of the TC2 model. naturally around 10% 30% without conflict with low energy experimental data∼ and Ks can be parameterized 14.65.Ha, 12.60.Fr, 12.60.Jv as follows:

Ktt = Kbb =1,Ktt =1 , The fancy idea of technicolor (TC) provides a possi- UL DL DR − ble mechanism of breaking the electroweak symmetry dy- tc tt 2 tu 2 2 KUR = 1 KUR KUR 2  , (2) namically. However, it is hard for technicolor to generate q − − ≤ p − the masses, especially the heavy top quark mass. with  representing the fraction of top quark mass gen- As a realistic TC model, the topcolor-assisted technicolor erated from TC interactions. Throughout this paper, we (TC2) model [1] combines technicolor with topcolor, with fix Ktc = √2 2 and treat  as an input parameter UR − the former mainly responsible for electroweak symmetry in the range 0 0.1. TC2 model also predicts a CP-even ∼ breaking and the latter for generating a major part of scalar ht called top-Higgs [7]. Its couplings to quarks are top quark mass. This model is one of the promising can- similar to that of the neutral top-pion except that the didates of new physics, awaiting to be tested at the up- neutral top-pion is CP-odd.

coming CERN (LHC).

g

Ø Ø As shown in numerous previous studies, the top quark b

processes are sensitive to new physics [2]. In some new Ø

h

Ø

¦

Ø



Ø

¼ 

physics models like supersymmetry, there may emerge Ø Ø

some new production and decay mechanisms for the top

c

c

 g

quark at hadron colliders [3–5]. The study of these new b

aµ ´bµ mechanisms will help to reveal the new physics effects. ´ It is noticeable that the TC2 model may have richer FIG. 1. Feynman diagrams contributing to tc¯ associated top-quark phenomenology than other new physics mod- production in TC2 model. els since it treats the top quark differently from other quarks to single out top quark for condensation [1]. In fact, the top-color interaction is flavor non-universal and The above large Yukawa couplings will induce the top- consequently may induce new large flavor-changing (FC) charm associated production pp tc¯ + X through both 0 → interactions [6,7]. One kind of such FC interactions occur the s-channel gg πt ,ht tc¯, as shown in Fig.1(a), → → + between quarks and the top-pion, the pseudo-goldstone and the t-channel bb tc¯ by exchanging a π ,asshown → t boson predicted by TC2 model with mass of a few hun- in Fig.1(b). Compared to the t¯b productions, which is dred GeV, which are given by [1,6] also sensitive to TC2 [6], the tc¯ productions are highly suppressed in the SM and thus its observation would be m v2 F 2 a robust evidence of new physics. t t iKtt Ktt t¯ t π0 = − UR UL∗ L R t The tc¯ production through the s-channel gg h tc¯ L √2Ft p v t  has been briefly studied in the literature [7].→ But→ a +√2Ktt Kbb t¯ b π+ + iKtc Ktt t¯ c π0 UR∗ DL R L t UR UL∗ L R t detailed Monte Carlo study of its observability, with

1 the consideration of the SM backgrouds, is still lacking. s-andt-channel productions, one sees that for a com- Given the great importance of the LHC phenomenology, mon scalar mass the s-channel rate is higher than the such a study of observability is absolutely necessary. Fur- t-channel rate only in the range from mt to 400 GeV. thermore, the tc¯ production via the t-channel bb tc¯, From this figure one can also see that the cross section which is more important when the top-pion is heavy,→ has of pp π0 tc¯ is at least a couple of times larger than → t → not been studied before. In this work we will give a that of pp h tc¯ for m 0 = m . t πt ht comparative study of both processes. We will not only → → study the production rates, but also perform a Monte Carlo simulation to show explicitly the observability at - the LHC. Our result shows that both the s-channel and LHC: pp→tc+X t-channel processes can reach the observable level at the TC2 Model LHC in quite a large part of the parameter space. There- ε = 0.03 fore, looking for the top-charm associated productions at 10 4 the LHC will serve as an important probe for the TC2 model. The spin- and color-averaged amplitudes of the s- channel and t-channel processes are given by (fb) 3

σ 10

2 2 2 2 π0 1 m v F αs bb- t 2 = t t (2 2)(1 )2 → s 2 −2 tc- |M | 64  2Ft v π  − − gg 2 2 2 → gg π 0 sˆ mt (ˆs mt ) 2 c , (3) 10 → t → 1 2 2 − 2 2 tc- m (ˆs m )2 + m Γ h → ×  t  π0 π0 0 t πt tc- − 2 2 2 2 1 m v F αs ht 2 = t t (2 2)(1 )2 s 2 −2 |M | 64  2Ft v π  − − 100 250 400 550 700 850 1000 2 m (GeV) sˆ m2(ˆs m2) scalar c t t , (4) 2 2 2 2− 2 2 × mt  (ˆs m ) + m Γ − ht ht ht FIG. 2. Cross sections of top-charm associated production 2 2 2 2 π+ m v F as a function of corresponding scalar mass for different parton t 2 = t − t (2 2)(1 )2 t 2 2 processes at the LHC. |M | 2Ft v  − − tˆ(tˆ m2) − t , (5) ˆ 2 2 ×(t m + ) Under the assumption that the top quark decays via − π the normal weak interactions to Wb, the final state of tc¯ where α = g2/(4π)withg denoting strong coupling s s s production is Wbc¯. We look for events with the leptonic constant,s ˆ and tˆ are patron level Mandelstam variables. decay of the W , W `ν¯ (` = e or µ) and thus the signa- 0 Γπ (Γht ) is the width of neutral top-pion (top-higgs) → t ture of tc¯ production is an energetic charged lepton, one which can be calculated by considering all its decay b-quark jet, one light c-quark jet, plus missing E from modes. c (R) are loop functions defined by c (R)= T 1,2 1 the neutrino. The potential SM backgrounds are the sin- 1 ln (1 Rx(1 x)) 4 dx − − and c (R)= 2+(1 )c (R). Note 0 x 2 R 1 gle top productions, top pair (tt¯) productions, and Wb¯b, − π0 − R t ht there is no interference between Ms and Ms due to Wcc¯, Wcj and Wjj productions. These backgrounds 0 different CP property of πt and ht. In our numerical cal- have been studied extensively in [11]. In order to use the culation, we use the CTEQ5L parton distribution func- background results in [11], in our analysis we applied the tions [9] with Q = √s/ˆ 2. same selection cuts as in [11]. In Fig.2 we plotted the two-body tc¯ production cross First, we assumed silicon vertex tagging of the b-quark section versus corresponding scalar mass for both the s- jet with 60% efficiency and the probability of 0.5% (15%) channel and t-channel processes. The charge conjugate for a light quark (c-quark) jet to be mis-identified as a production channel, i.e., the tc¯ production, is also in- b-jet, which can reduce the background Wjj efficiently. cluded in our analysis throughout this paper. We required the reconstructed top quark mass M(bW ) We see from Fig.2 that for the t-channel process the to lie within the mass range M(bW ) mt < 20 GeV, production rate drops monotonously with the increase which can reduce all the non-top| backgrounds− | efficiently. of the charged top-pion mass. For the s-channel pro- To simulate the detector acceptance, we made a se- cesses the production rates are maximum when the neu- ries of basic cuts on the transverse momentum (pT ), the tral scalars lie in the range of m + m m 0 2m . pseudo-rapidity (η), and the separation in the azimuthal t c . πt ,ht . t The reason is that in this range, tc¯ is the dominant de- angle-pseudo-rapidity plane ( ∆R = (∆φ)2 +(∆η)2 ) cay mode of the neutral scalars. Comparing the rates of between a jet and a lepton or betweenp two jets. These

2 cuts are chosen to be Fig.2. Outside the region the signal is observable only for enough large  value. But given 0.001  0.1, ` j miss ≤ ≤ pT ,pT ,pT 20 GeV , (6) the signal is observable for m 0 550 GeV. In case of ≥ πt . η` 2.5, ηj 4, ηb 2 , (7) nonobservation, the 2σ lower limit on mπ0 is about 600 | |≤ | |≤ | |≤ t ∆R , ∆R 0.7 . (8) GeV. jj j` ≥ -1 Further simulation of detector effects is made by assum- 10 ing a Gaussian smearing of the energy of the final state 5 σ σ σ σ particles, given by ∆E/E = 30%/√E 1% for leptons 5 3 2 ⊕ and ∆E/E = 80%/√E 5% for hadrons, where indi- 3 σ cates that the energy dependent⊕ and independent⊕ terms are added in quadrature and E is in GeV. 2 σ Under the above cuts the total cross section of back- -2 grounds is 16839 fb [11], of which 7159 fb is from single ε 10 Observable top productions, 2770 fb from tt¯ productions, and 5070 fb, 1460 fb, 230 fb and 150 fb from Wcj, Wjj, Wb¯b, Wcc¯ productions, respectively. A few remarks are due regarding our analysis: (1) The main kinematic difference between the s-and LHC: pp→tc+X- t-channel processes is that for the former the signal has a s-channel (TC2) peak in the top-charm invariant mass distribution and -3 10 - this feature might be used to effectively reduce back- 200 300 400 500 600 1000 ground [7]. In our analysis, we ignored this difference mπ0 (GeV) and adopt the same strategy to probe the top-charm sig- t FIG. 3. Observable region in the plane (m 0 ,)forthe nal. The reason is twofold. One is that without any πt production pp tc¯ + X through s-channel parton process information of m 0 and m ,wedonotknowwherethe π ht 0 → 1 gg π tc¯ at the LHC with L = 100 fb− . peak lies at. The other is, due to strong interaction of → t → the neutral scalars with quarks, their widths are of sev-

0 eral hundred GeV for mπ ,ht > 2mt and as a result, the -1 peak is highly smeared. 10 (2) The two jets in the signal are bc¯ (or ¯bc). Since the c-quark jet could be mis-identified as a b-jet with a probability of 15%, the efficiency of tagging one b-jet from bc¯ (or ¯bc) should be slightly higher than 60%. In our σ σ σ 5 3 2 analysis we conservatively assumed the tagging efficiency Observable of 60%. -2 (3) For the same reason stated above, we could pos- ε 10 sibly require to tag two b-jets for the signal. Compared with tagging only one b-jet, this will further reduce the signal rate by a factor of 15% while suppress the Wjj and Wcj backgrounds by a factor of 0.5%. However, the large backgrounds of top quark productions cannot be LHC: pp→tc+X- relatively suppressed because they contain two b-jets in t-channel (TC2) their final states. As a result, the total background is re- -3 10 duced only by a factor of 37%. So this strategy of tagging 200 300 400 500 600 1000 two b-jetsdoesnobetter. mπ+ (GeV) The observability of the tc¯ productions through gg t 0 ¯ → FIG. 4. Same as Fif.3, but for the production pp tc¯+ X πt tc¯ and bb tc¯ are shown in Figs. 3 and 4 for ¯ → → → 1 through t-channel bb tc¯. the LHC with L = 100 fb− . Throughout our analysis → we restrict the value of the parameter  in the range of The t-channel process exhibits a different feature from 0.001 0.1. the s-channel process, which can be inferred from the be- We∼ see from Figs. 3 and 4 that for both processes havior of its two-body tc¯ production rate shown in Fig.2. the observable parameter region is quite large. For the The observable region of the parameter space shrinks s-channel process the region mt +mc mπ0 2mt is ob- . t . monotonously with the increase of the charged top-pion servable for any  value in the range of 0.001 0.1. The mass. But for a relatively high  value in the range of reason for this is already elucidated in the discussions∼ for

3 0.001 0.1, the signal is observable up to mπ+ 800 LHC can either observe the productions or set stringent ∼ t ' GeV. This value is much larger than the constraint from bounds on the parameters of the TC2 model. Rb [12], i.e. m + 250 GeV. In case of nonobservation, The work of Z.X. was supported by the postdoctor πt & the 2σ lower limit on m + can reach 1 TeV. followship of Japan Society for the Promotion of Science. πt We did not show the observable region in the plane (ht, ) for the s-channel process gg ht tc¯ because it is similar to Fig.3. The only difference→ → is if  0.1, ≤ the signal is observable at 3σ level for ht . 480 GeV and unobservable at 2σ level for ht & 530 GeV. In our analysis of neutral scalar production, we did not [1] C. T. Hill, Phys. Lett. B 345, 483 (1995). K. Lane and ¯ 0 choose tt as the signal of the new physics for mπ ,mht > E. Eichten, Phys. Lett. B 352, 382 (1995); K. Lane and 2mt. The reason is the effect of these scalars on the cross E. Eichten, Phys. Lett. B 433, 96 (1998). section of tt¯ is small, resulting in a contribution of the [2] See, e.g., C. T. Hill and S. J. Parke, Phys. Rev. D 2 3 order of σ(tt¯)TC2/σ(tt¯)SM =10− 10− at the LHC. 49, 4454 (1994); K. Whisnant, et al., Phys. Rev. D In some extensions of the SM, such∼ as the generic two- 56, 467 (1997); K. Hikasa, et al., Phys. Rev. D 58, Higgs doublet models with tree level FC scalar interac- 114003 (1998). For most recent reviews, see, e.g., C. tions [13], both s-andt-channel top-charm associated T. Hill and E. Simmons, hep-ph/0203079; C.-P. Yuan, productions can also occur via exchanging neutral and hep-ph/0203088; E. Simmons, hep-ph/0211335; S. Wil- lenbrock, hep-ph/0211067. charged Higgs bosons, respectively. However, although [3] See, e.g., A. Datta, et al., Phys. Rev. D 56, 3107 (1997); the corresponding cross sections are large, they are at R. J. Oakes, et al., Phys. Rev. D 57, 534 (1998); K.- least one order of magnitude smaller than those of TC2 I. Hikasa, J. M. Yang, B.-L. Young, Phys. Rev. D 60, for a common mass of Higgs bosons and top-pions [6]. 114041 (1999); P. Chiappetta, et al., Phys. Rev. D 61, The reason is that TC2 model predicts a large Yukawa 115008 (2000); C. Balazs, H.-J. He, C.-P. Yuan, Phys. coupling mt 2.5 and a possible large FC coefficient Rev. D 60, 114001 (1999). 2Ft tc ' [4] See, e.g., M. Hosch, et al., Phys. Rev. D 58, 034002 KUR (see Eq.(1)). The interactions between the scalars and quarks can (1998); S. Mrenna and C.P. Yuan, Phys. Lett. B 367, also induce large neutral flavor changing gauge interac- 188 (1996). tion, such as the coupling tcg¯ [14]. Since it is not easy [5] C. S. Li, R. J. Oakes and J. M. Yang, Phys. Rev. D 49, 293 (1994); J. L. Lopez, D. V. Nanopoulos and R. Ran- to observe top rare decay t cg at hadron collider [15], → garajan, Phys. Rev. D 56, 3100 (1997); G. M. de Divitiis, such a coupling may be well tested via single top pro- R. Petronzio and L. Silvestrini, Nucl. Phys. B 504,45 duction, cg t, at the LHC. This process was exten- → (1997); G. Eilam, et al., Phys. Lett. B 510, 227 (2001); sively studied in effective Lagrangian approach [16] with K.J. Abraham, et al., Phys. Lett. B 514, 72 (2001); Phys. 1 the conclusion that for LHC with 100 fb− the discov- Rev. D 63, 034011 (2001); T. Han and J. Hewett, Phys. ery limit on the effective interaction coefficient κc/Λis Rev. D 60, 074015 (1999); F. del Aguila, J. A. Aguilar- 1 0.0048 TeV− at 3σ level, or alternatively, new physics Saavedra, R. Miquel, Phys. Rev. Lett. 82, 1628 (1999); ¯ contribution to the cross section pp t∗ bleve must J.L. Diaz-Cruz, H.-J. He, C.-P. Yuan, Phys. Lett. B 530, be larger than 40 fb. We examined this→ process→ in TC2 179 (2002). [6] H.-J. He, C.P. Yuan, Phys. Rev. Lett. 83, 28 (1999). model and found that for  0.1 only when mh 400 ≤ t ≤ [7] G. Burdman, Phys. Rev. Lett. 83, 2888 (1999). GeV and m + 0 350 GeV can such single top events π ,π [8] G. Burdman and D. Kominis, Phys. Lett. B 403, 101 be observable. Obviously,≤ the observable region is smaller (1997). than that from top-charm associated production. [9] H. L. Lai, et al. (CTEQ collaboration), hep-ph/9903282. TC2 model also predicts some new top-quark decay [10] E. Malkawi and T. Tait, Phys. Rev. D 54, 5758 (1996). + modes, such as t b¯bc via exchanging a π and t [11] T. Stelzer, Z. Sullivan, S. Willenbrock, Phys. Rev. D , → t → 58 cW W via an intermediate ht. Among these new decay 094021 (1998). modes, t b¯bc is dominant for a not heavy charged top- [12] C. X. Yue et al., Phys. Rev. D 62, 055005 (2000). → pion. For example, for  =0.1andmπ+ = 250 GeV, [13] See, e.g., D. Atwood et al, Phys. Rev. D 54, 3296 (1996). ¯ 3 the ratio Γ(t bbc)/Γ(t bW ) can reach 9 10− , [14] X.L. Wang et al., Phys. Rev. D 50, 5781 (1994); J. Phys. which maybe not→ accessible→ at the LHC due to× the SM G20, L91 (1994); Commun. Theor. Phys. 24, 359 (1995); backgrounds [17]. For the decay mode t cW W [18], it G. R. Lu, et al., Phys. Rev. D 57, 1755 (1998). → [15] T. Han et al., Phys. Lett. B 385, 311 (1996). drop quickly with the increase of mht and for mht > 250 7 [16] M. Hosch, K. Whisnant and B.-L. Young, Phys. Rev. D GeV the ratio Γ(t WW)/Γ(t bW )isof10− .Asa → → 56, 5725 (1997). result, such a decay mode cannot put severe bounds on [17] J. A. Aguilar-Saavedra and G. C. Branco, Phys. Lett. B TC2 parameter. 495, 347 (2000). So we conclude the top-charm associated productions [18] C. X. Yue et al, Phys. Lett. B 508, 290 (2001). at the LHC are a powerful probe for the TC2 model. The

4 -1 10

5 σ σ σ σ 5 3 2

3 σ

2 σ

-2 ε 10 Observable

LHC: pp→tc+X- s-channel (TC2)

-3 10 - 200 300 400 500 600 1000

mπ0 (GeV) t -1 10

σ σ σ 5 3 2 Observable

-2 ε 10

LHC: pp→tc+X- t-channel (TC2)

-3 10 200 300 400 500 600 1000

mπ+ (GeV) t

g

Ø Ø

b

Ø

h

Ø

¦

Ø



Ø

¼



Ø

Ø

c

c



g

b

´aµ ´bµ LHC: pp→tc+X- TC2 Model ε = 0.03

10 4

(fb) 3

σ 10

bb- → tc- gg → gg π 0 2 10 → t → h tc- t → tc-

100 250 400 550 700 850 1000

mscalar (GeV)