ebook img

Lorentz invariance violation in top-down scenarios of ultrahigh energy cosmic ray creation PDF

0.16 MB·English
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Lorentz invariance violation in top-down scenarios of ultrahigh energy cosmic ray creation

FERMILAB-Pub-03/183-A Lorentz invariance violation in top-down scenarios of ultrahigh energy cosmic ray creation James R. Chisholm and Edward W. Kolb Fermilab Astrophysics, Fermi National Accelerator Laboratory, Batavia, Illinois 60510-0500, and Enrico Fermi Institute, University of Chicago, Illinois 60637 (Dated: February 1, 2008) The violation of Lorentz invariance (LI) has been invoked in a number of ways to explain issues dealingwithultrahighenergycosmicray(UHECR)productionandpropagation. Thesetreatments, however, have mostly been limited to examples in the proton-neutron system and photon-electron system. InthispaperweshowhowabroaderviolationofLorentzinvariancewouldallowforaseries ofpreviouslyforbiddendecaystooccur,andhowthatcouldleadtoUHECRprimariesbeingheavy baryonicstates or Higgs bosons. 4 PACSnumbers: 11.30.Cp,96.40.De,98.70.Sa 0 0 2 I. INTRODUCTION produce superheavy UHECR primaries. n a J UHECRs are an enduring mystery. Without the in- 8 troduction of new particles or interactions, evading the II. BREAKING THE LAW 2 Greisen–Zatsepin–Kuz’min (GZK) cutoff [1, 2] requires unidentified nearbysources. Evenwithoutthe GZKcut- A. Modified Dispersion Relations 2 off,the“bottom-up”approachfacesthechallengeoffind- v ing in nature anacceleratorcapable of energiesin excess There are two approaches to breaking LI commonly 8 of 1020eV. used in the literature. The first [16, 20, 21, 25, 26] is 8 2 “Top-down”scenariosassumethattheUHECRsresult genericallyto modify the dispersionrelationsfor aparti- 6 from the fragmentation of a ultra-high energy hadronic cle of mass m and 3-momentum p= p. | | 0 jet produced by cosmic strings [3] or by the decay of The second method, which we will not utilize here, is 3 a supermassive particle [4]. In the supermassive par- instead to write down a particle Lagrangian [27] which 0 ticle (wimpzilla) scenario, the UHECRs are of galactic includes Lorentz and CPT violating terms. A mapping / h origin, resulting from the decay of relic supermassive of the Lagrangian terms to a dispersion relation is non- p (M > 1013GeV) particles. Wimpzillas can be produced trivial(see [23]for a simple case); however,to first order p- copio∼uslyintheearlyuniverse[5],thussolvingtheenergy the changes to the photon and electron propagators in- e problem. Since they would cluster in the dark matter duce shifts to cγ and ce [17]. Similarly, it was shown by h halo of our galaxy [6, 7, 8], they also solve the distance [28] that loop quantum gravity effects produce modified v: problem. Detailed analysis of these decays, however, dispersionrelationssimilartothoseconsideredelsewhere i show that at high energy top-down scenarios produce and here. X more photons than protons [9, 10, 11] in the UHECR The simplest way to break LI, as shown in [17], is to r spectrum seen at Earth. write down a dispersion relation for a particle species i a as The top-down prediction of photon preponderance in UHECRsistheonemajorprobleminanotherwisesimple E2 =p2c2+m2c4. (1) explanation. Results from the Fly’s Eye [12], Haverah i i i i i Park [13], and AGASA [14] cosmic ray experiments all This is changing the “speed of light” or Maximum At- indicate that at energies above around 1019 to 1020eV, tainable Velocity (MAV) for each particle to something protons are more abundant than photons in UHECRs. slightly different than c.1 The MAVs for different par- While photons are disfavored, it is not possible to be ticle species are assumed a priori to be different in that sure that the primary is indeed a proton. and similar treatments [26]. The idea of violating LI (see Ref. [15] for a broad Insuchacase,itispossible forpreviouslyunallowable overview) has recently been studied in the context of reactionstooccur,suchasp ne+ν [16,17],γ e+e− e UHECRs for proton decay [16, 17], atmospheric shower [25, 26] and γ νν¯ [25].→A conclusion reach→ed, us- development[18],andmodificationstothecutoffsofpro- ing the first reac→tion, is that neutrons may make up the ton and photon spectra due to cosmic background fields dominantbaryoniccomponentofthe UHECRprimaries. [19, 20, 21, 22, 23, 24]. As shown in Ref. [25], LI vi- olation can also lead to vacuum photon decay, and in general the decay of particles to more massive species. We exploit this fact to show how particles produced in 1 Althoughfactorsofcareexplicitlyincludedinequations,westill top-down decays can undergo an “inverse cascade” to takeittobedimensionlesswithcγ =c=1. 2 This, strictly, would be true if one was only tuning the B. Example: Photon Decay proton and neutron MAVs and leaving all others equiv- alent (= c). On the other hand, keeping every MAV as In this section we consider the tree-levelphoton decay individually tunable gives an overwhelming number of process γ e+e−, which is kinematically forbidden for free parameters. c = c. Fo→llowing Ref. [17], we define the parameter δ e γe In order to examine the consequences of varying all as3 particle MAVs using the fewest number of free parame- ters, we extend the method used in Ref. [21] and write δ =c2 c2. (9) γe − e the dispersion relation as Using astrophysical constraints [29, 30], δ can be lim- γe E2 = p2c2+m2c4+p2c2f(p/Mc) ited to the range 10−16 > δγe > 10−17. For δγe > 0, − +m2c4g(p/Mc)+mc2pch(p/Mc). (2) photon decay occurs above a photon energy threshold given by Here, f(x),g(x), and h(x) are dimensionless universal 2m c2 functions having the property f(0) = g(0) = h(0) = 0, E = e ec. (10) th δ so that as p/M 0 the normal dispersion relations are γe → recovered. Here M is the mass scale that determines the ThisdecayrateofwascompputedinColeman&Glashow relative degree of Lorentz violation; for our purposes we (1997) as set it at the Planck mass M 1019 GeV. Expanding ≈ these functions in Taylor series about x=0 and keeping 1 δ 3/2 terms of O(p2) and smaller gives Γ= 2α cγ2e E 1−(Eth/E)2 . (11) (cid:18) e (cid:19) h i E2 = m2c4+mc2pc 1g′(0)m Fordecayproductenergyaboveabout2 1020eV,pho- 2 M tons will outnumber protons, so we set E×to this value (cid:18) (cid:19) th 1 m 1 m2 and suggestively rewrite the decay rate above threshold +p2c2 1+ 2h′(0)M + 6g′′(0)M2 . (3) as: (cid:18) (cid:19) Γ E δ The g′(0) term can be neglected for two reasons: at (1 km)−1 γe . (12) ¯hc ≈ 2 1020 eV 3 10−29 p mc it would be experimentally detected as devia- (cid:18) × (cid:19)(cid:18) × (cid:19) ≪ tions in the non-relativistic kinetic energy; for p mc ThiswouldruleoutallbutaterrestrialsourceofUHE ≫ it is negligible to the quadratic p term. Note that to photons. The value δ = 3 10−29 is used is a lower γe this (lowest) order, f(x) has no effect.2 For this section, limit; if δ would be any lo×wer the energy threshold γe we examine the cases of massless (photon) and massive wouldbecometoohigh. Thismeansthatonlyδ (3 γe (electron) particles: 10−29,10−16) will correct for the photon overabun∈danc×e in top-down scenarios. E2 = m2c4+p2c2 1+ 1h′(0)m + 1g′′(0)m2 (4) Now we see what is required to have δγe in the neces- e e 2 M 6 M2 saryrangeto allowforthis photondecay. UsingEqs.(6) (cid:18) (cid:19) E2 = p2c2. (5) and (9), we can write γ γ 1 m 1 m2 We can now define δ = c2h′(0) c2g′′(0) . (13) γe −2 M − 6 M2 1 m 1 m2 c2 =c2 1+ h′(0) + g′′(0) (6) Since the functions g(x) and h(x) are a priori ar- e 2 M 6 M2 (cid:18) (cid:19) bitrary, so are the values and signs of their deriva- tives. It is argued in Ref. [21] that too strongly vary- ing functions would be unphysical, and that the deriva- m2 =m2 1+ 1h′(0)m + 1g′′(0)m2 −2 (7) tives should be of order unity.4 Adopting this, we can e 2 M 6 M2 ignore the g′′(0) term in the above equation. In this (cid:18) (cid:19) case, in order to have photon decay we need to require to then write the dispersion relationin the familiar form of Eqn. 1 as Ee2 =m2ec4e+p2ec2e. (8) 3 Thisisidentical toξ−η forn=2inthenotationofRef.[29]. 4 Notethatifthisholdsalsoforf(x),weerredinnotincludingthe f′(0)terminEq.(3),asthatdominatesovertheh′(0)termwhen p>∼mc. Wewillnotapologizemuchforthis,aswearelookingat the“lowestorder”changeinphenomenology (thechange inthe 2 Issuesofphotondispersioninthevacuum, asdiscussedinRefs. “speedsoflight”)andnot“nextorder”changes(suchasvacuum [20,29],willnotbeaddressedhere. photon dispersion). 3 h′(0) negative. Then δ mc2/2M 3 10−23 and exactly zero momenta [17]. Since the c are the max- γe X ∼ ∼ × { } E = 2m c2c 2M/mc2 = 2√2m Mc2 2 1017eV. imum attainable velocies for the decay products, that aTlhtlhoiwsvaabllueereaonfegδeγp.e is(logarithmically)einthe≈mid×dle ofour imIptliiesstthhiastlaλst<pmoiinnt{cwXe}f.ocus on: min cX in our case { } Returning to our original impetus of addressing the corresponds to max mX . In the ultrarelativistic limit, { } feasibilityof“TopDown”scenarios,weseethatrequiring EA pAcA and so on, and the particle with the lowest ≈ h′(0) 1 is sufficient to eliminate any photons among speed of light will carry most of the momentum (in the the U∼HE−CR primaries. This model, however, eliminates minimum energy case). Further, in this threshold case, any protons as well, which we will now address. λ is also the velocity of the incident particle [17]. Since λ c intheregimeweareconsidering,thatmeansthat A ≈ the decaycanonlyoccurif c >min c . FromEq.14, A X { } C. Making the impossible possible that implies that mA <max mX . { } Inotherwords,aboveacertainthreshold,adecaymay Unlike previousinvestigations[17,29] wherethe MAV only be allowed if at least one decay product is more for each particle could be chosen independently, this is massivethanthedecayingparticle. Thiscanhaveserious not so in our case. Since Eq. 2 holds for all types of consequences for the particle content of UHECRs, as we particles, it is straightforwardto see other effects of this will now address. levelofLorentzinvarianceviolation. UsingEq.6,wecan in general write for two particles A and X III. THE INVERSE CASCADE 1 m m δ = c2 c2 = h′(0)c2 A− X AX A− X 2 M In this model, for each individual reaction there is a (mX −mA)c2, (14) thresholdenergyET abovewhichparticlescanonlydecay ≈ 2M ifatleastoneofthedecayproductsismoremassive. For the decay A B + ..., where B is the most massive where we have used h′(0) 1. → ∼− decay product, mB >mA, and m... mB the threshold The consequencesof this for particle decaysare as fol- ≪ energy goes like lows,as firstnoted by [17], and we reproducetheir argu- ment here. m2 m2 Consider the decay A X , where X is a col- E c3 B − A. (19) lection of massive particle→s. {We}want to {find}the min- T ∼ s c2A−c2B imum possible energy E where this can occur. We min can always lower final state energy by removing trans- Using equation 14, this reduces to verse components of momenta, so we’re working in one dimension only. We want to minimize E c2 M(m +m ) h′(0) −1/2. (20) T A B ∼ 1 (cid:18) (cid:19) E( p )= m2 c4 +p2 c2 (15) p { X} X X X X Due tothe largevalueofthePlanckmass,this thresh- XX q old is around the same order of magnitude for a wide subject to the constraint range of reactions involving particles of mass much less than M. G( p )= p p =0. (16) ConsideranUHECRofmassm0producedwithenergy { X} XX X!− A 0E0ca≫n bEeTa. dIefcnaoyrmpraoldlyuc(tatofenaermgoierseEma≪ssivEeTp),arptaicrlteic1le, sTthaent{s.pXU}sinagrethvearmiaebtlheos,daonfdLapnAg,rEanAge>mEulmtiipnliearrse,cEonis- tlihgehntehreprero(dEuc≫tsEthTa)t, 0wicllanredsuecltayasinwtoell1;.thWeesiugbnsotraenttihael minimized when fractionoftheincidentenergywillbecarriedbythemost massive decay product. ∂E ∂G =λ , (17) Now, we have a particle of energy E1 < E0 and mass ∂pX ∂pX m1 > m0. If it can, it will also decay into a particle which becomes 2 with m2 > m1 and energy E2 < E1. As the decays continue into more and more massive final states, the p c2 p c2 final energy continues to decrease, until at some point λ= X X = X X . (18) EX m2Xc4X +p2Xc2X wmeNre>achma0.paSritniccelewNe awriethnoenwerbgeyloEwNth<reEshToladn,dthmeaNss Thusallproductsmovewpiththesamevelocityλ. Note particle can decay as usual into N-1 and so on into less that λ < c for all X. For the example we consider, in massive particles. X the event of massless particles among the decay prod- So, to summarize, this inverse cascade decay process ucts, they would minimize final state energy by carrying occurs in two stages. 4 1. Decay up: Decays into more and more massive, thanthetopquark. Ifitislighterthanthetop,theHiggs yet less energetic particles, when E >E , and is unstable to H tt¯. T → 2. Decay down: Regular decay schemes once E < E . T C. Tree-level Strong IV. PANDORA’S BOX (NEWLY ALLOWED Since the bare QCD vertex preserves quark flavorand REACTIONS) charge,notmuchhappensinthebarequarksector,apart fromthereversingsofstrongdecayssuchasN ∆πand → π ρπ. Justasforthemasslessphoton,freegluondecay We saw in Section II that at sufficiently high energies → is allowed via g qq¯. particles can only decay if one of the decay products is → moremassive. InSectionIIIwesawhowthiscanproduce an “inverse cascade” to more and more massive particle D. One-loop processes species. In this section we address the specifics of how this cascade occurs, by examining which previously for- One could formulate many more decays at the 1-loop bidden decays can now occur. All of the following are assumedto be occuringwellabovethreshold,andwelist level, but the rates for these for them are suppressed them in several classes. compared to the tree level decays. For this reason, we do not consider them further here, except to note two interesting decays of the photon: γ ν ν¯ (QED and weak process). Note that since A. Tree-level QED → l l neutrinos are so much lighter than their heavy partners and the quarks,the threshold for photon to neutrino de- Here the bare QED vertex γ Q+Q− where Q is cay could be very low depending on how light the neu- → any charged particle is allowed as a decay. This also trinos are. includesfinalstateswithextraphotonsandboundstates γ Zγ (weak process). This photon Z-Cerenkov de- of Q+Q−, like γ π0γ. → cay occurs due to a W-loop. → B. Tree-level weak E. Spin up It is in the weak sector that most of the interesting For bound states (and now restricting ourselves to decays in the inverse cascade occur, due to the flavor- bound states of three valence quarks) there exist higher changing W vertex. mass resonances that only differ in the mass and total In this Lorentz-violating scheme, weak decays (such angular momentum (e.g., the ∆ and higher resonances as that of the neutron) can happen in reverse, allowing for the protonand neutron). Since the state with higher proton decay via p ne+ν and flavor changing decays angularmomentum is more energetic,it is more massive such as n Λπ0. → and thus processes like N N∗π, N N∗∗π and so → → ∗ → Not only that, decays to weak bosons become permis- on can act to “spin up” the particle to more and more sible, such as ν l±W∓ and l± ν W±. Included massive states. l l in this are decays→of leptons to hea→vier leptons due to Thus,eventhealltopquarkbaryontttisn’tcompletely virtual W’s. In the quark sector,this allows decays such stable, but can decay via spin up. as p nW+ and so on. Generally, the flavor changing weak→decay q q′W will convert all quarks to t quarks. → V. ESTIMATES OF DECAY RATES In the absence of a fourth generation, a bare t quark is stable. Since the Z is massive, decays such as q,l,ν Forthe photondecayγ e+e−, the decayrateΓwell Z q,l,ν (“Z-Cerenkov”)become permissib{le as wl}ell.→ above threshold was → l {For th}e bosons; W+ t¯b is allowed, and Z-Cerenkov 1 → Γ = αδ E. (21) is possible for the W as well (this is a bare weak vertex, γ→ee 2 γe as is W WZZ and W WZγ). The Z is unstable to the de→cay Z tt¯. → Weadaptthisformofthedecayratetootherprocesses → to make estimates, and argue for it as follows. At ultra- NowconsidertheHiggs: indeedwemightbeabovethe relativistic energies,the only energy scale is the incident energy of weak symmetry breaking. Including a single energy E, so Γ E. It is proportional also to the ef- massive Higgs H would allow; f,W,Z f,W,Z H ∝ { } → { } fective coupling constant involved in the decay, and to (“Higgs-Cerenkov” for fermions and weak bosons) and kinematic factors. We schematically write this as W,Z W,Z HH. In the absence of new physics { } → { } beyondtheStandardModel,theH isstableifitisheavier Γ (coupling) (kinematics) E. (22) ∼ × × 5 Forthephotondecay,thecouplingisαandthekinematic Second, and more interesting, if the initial UHE pri- factor is δ . We generalize these factors as follows. mary is of high enough energy, the decay up sequence γe We (first) consider decays of the form A B + C may occur and leave particles that can no longer decay → where m <m , m m . At threshold and beyond, into moremassiveparticles,but arestillabovethreshold A B B C ≫ most of the momentum of the final state is carried by for decay into less massive particles. These Most Mas- the particle with the lowest speed of light, thus highest sive Particles (MMPs) would then be the most energetic mass (particle B). The relevant kinematical factor will componentoftheUHECRpopulation. Whatexactlythe then be δ . This will continue to be true when final MMPsaredependsonthemodelofparticlephysicsused. AB states with more than 2 particles are allowed, as long as Of particles knownto existknow, the MMPs would con- B continues to be the most massive particle. sist of top quarks and excited bound states of top (tt¯ The coupling factor is determined by looking at the mesons and ttt baryons) quarks. If the Higgs is more tree level Feynman diagrams and counting vertices in- massive than the top, then the MMPs would consist of volved in the reaction. single Higgs bosons. Including Supersymmetry and its Consider then the decay of the proton; in addition to expected spectrumof more massivesuperpartners,if the the inverse of the neutron decay p ne+ν (“leptonic” decay rate is still rapid enough the MMP would be the e decay), the decay p nW+ (“bo→sonic” decay) is also most massive superparticle (MMSP). → allowed. Of course, in the first outcome, our proposed solution For the bosonic decay; E √m M 2.8 1019 of using LI violation to solve the photon abundance is- T W ∼ ∼ × eV, and as most of the energy goes into the W, δ sues in top-down models is not obviously successful. For pW m /2M 4 10−18. Thevertexisasingleweakvertex∼; bothoutcomes,infact,onewouldhavetore-examinethe W with coup∼ling×α =α/sin2θ 0.0316. Thus, at E = decayprocessstartingwiththeinitialdecayofthesuper- W W ∼ E ,Γ 4 eV. massive relic (which is assumed to be sub-threshold, so T For∼the leptonic decay; E M(2m ) 4.5 1018 it decays “normally”), and follow the decay products to T p ∼ ∼ × eV, and as most of the energy goes into the n, δ see what results. pn (m m )/2M 5 10−23. pThis reaction has tw∼o Simulations of MMP-induced shower development n p weak−vertices wit∼h co×upling α2 10−3. Thus, at would have to be performed to see if they are consis- E =E ,Γ 2.2 10−7 eV. W ∼ tent with actual events. It was shown in Ref. [32] that T ∼ × Notethatthedistancetravelledbeforedecayis 1/Γ, stableUHEprimariesofmassgreaterthanabout50GeV and givenh¯c=6.57 10−24 eV pc, the 1/Γ for th∼ese re- shouldbe distinguishable fromprotonprimariesby their actions is incredibly×small: 10−24 pc for the bosonic and atmospheric shower profiles. As the lightest conceivable 10−17 pc for the leptonic. In the context of a UHECR, MMP (a tt¯ meson) would be much heavier than this assuming comparable rates (within a few orders of mag- bound, the result of [32] might be used to exclude the nitude)tothe above,this inversecascadehappensessen- MMP as the primary for UHE showers. However, a key tially immediately, so that by the time any decay prod- assumptionisthattheprimaryisstable. ForaMMPjust uctsreachtheEarth,theyhavegonethroughbothstages above the threshold energy, it may take only a few colli- of the cascade and are now sub-threshold particles. sionsbeforetheMMPenergyisloweredbelowthreshold, where it will decay “normally” (e.g., tt¯ W+W−) into → a shower of particles, perhaps mimicking the shower of VI. CONSEQUENCES FOR UHECR a nucleus UHECR. A MMP with incident energyfarther COMPOSITION above threshold would then penetrate deeper. In any event, further investigation into MMP UHECR shower Given the rapid rate of the inverse cascade process, profiles is warranted. there can be two outcomes for the UHECR population Also of concern is that, while energy and momentum at earth. are conserved in a certain preferred frame, they are nec- First,ifboththeDecayUpandDecayDownsequences essarily not conserved in all other frames. As such, due occur, then the consituent particles could, depending on to the motion of the earth and the solar system, small theexactdecaychainsandnatureofinitialUHEprimary, departures from energy conservation might be observed consist of protons, electrons, photons, and their antipar- as well in the UHECR atmospheric showers. ticles. Each particle spectrum would exhibit a cutoff at ThisworkwassupportedinpartbytheDepartmentof theirprospectivethresholdenergies,whichareallrelated Energy and by NASA (NAG5-10842) and by NSF grant by the h′(0) parameter. PHY 00-79251. [1] K.Greisen, Phys. Rev.Lett. 16, 148 (1966). [3] C.T.Hill,D.N.Schramm,andT.P.Walker,Phys.Rev. [2] G. T. Zatsepsin and V. A. Kuz’min, JETP Lett. 4, 78 D, 36, 1007 (1987); P. Bhattacharjee, C. T. Hill, and D. (1966). N. Schramm, Phys.Rev.Lett. 69, 567 (1992). 6 [4] V. Berezinsky, M. Kachelrieß, A. Vilenkin, Phys. Rev. [16] S. L. Dubovsky, and P. G. Tinyakov, Astropart. Phys. Lett. 79, 4302 (1997); V. A. Kuzmin, V. A. Rubakov, 18, 89 (2002). Phys. Atom. Nucl. 61, 1028 (1998); K. Benakli, J. El- [17] S.Coleman,andS.L.Glashow,Phys.Rev.D59,116008 lis, D. V. Nanopoulos, Phys. Rev. D 59, 047301 (1999); (1999). V. Berezinsky, P. Blasi, A. Vilenkin, Phys. Rev. D 58, [18] H.VankovandT.Stanev,Phys.Lett.B538,251(2002). 103515 (1998). [19] H.Sato,andT.Tati,Prog.Theor.Phys.47,1788(1972). [5] D. J. H. Chung, E. W. Kolb, A. Riotto, Phys. Rev. D [20] T. Kifune, Ap.J. Lett. 518, L21 (1999). 59, 023501 (1999); Phys. Rev. Lett. 81, 4048 (1998); [21] R. Aloisio, P. Blasi, P. L. Ghia, and A. F. Grillo, Phys. Phys. Rev. D 60, 063504 (1999); in Dark Matter in As- Rev. D 62, 053010 (2000). trophysics and Particle Physics 1998, edited by H. V. [22] O. Bertolami, Nuc.Phys. Proc. Supp.88, 49 (2000). Klapdor-Kleingrothaus and L. Baudis, IoP Publishing, [23] O. Bertolami, and C. S. Carvalho, Phys. Rev. D 61, Bristol 1999, pp.592–611; D.J. H.Chung, P.Crotty, E. 103002 (2000). W.Kolb, A.Riotto, Phys.Rev.D 64, 043503 (2001); V. [24] J.Alfaro,andG.Palma,Phys.Rev.D67,083003(2003). Kuzmin, I. Tkachev, JETP Lett. 68, 271, (1998), Phys. [25] S. Coleman, and S. L. Glashow, Phys. Lett. B 405, 249 Rev.D 59, 123006 (1999), Phys. Rep.320, 199 (1999). (1997). [6] M.Birkel,andS.Sarkar,Astropart.Phys.9,297(1998). [26] T. Jacobson, S. Liberati, and D. Mattingly, Phys. Rev. [7] P.Blasi,andR.K.Sheth,Phys.Lett.B486,233(2000). D 66, 081302 (2002). [8] P. Blasi, R. Dick, and E. W. Kolb, Astropart. Phys. 18, [27] D. Colladay, and V. A. Kostelecky´, Phys. Rev. D 58, 57 (2002). 116002 (1998). [9] S.Sarkar and R.Toldr`a, Nuc. Phys.B 621, 495 (2002). [28] J.Alfaro,andG.Palma,Phys.Rev.D65,103516(2002). [10] V. Berezinsky, P. Blasi, and A. Vilenkin, Phys. Rev. D [29] T. Jacobson, S. Liberati, and D. Mattingly, Phys. Rev. 58, 103515 (1998). D 67, 124011 (2003). [11] C. Barbot and M. Drees, Phys. Lett. B 533, 107 (2002). [30] F. W. Stecker and S. L. Glashow, Astropart. Phys. 16, [12] D. J. Bird et. al., Ap. J. 441, 144 (1995); F. Halzen et 97 (2001). al.,Astropart. Phys. 3, 151 (1995). [31] V. A. Kostelecky´ and A. G. M. Pickering, Phys. Rev. [13] M. Aveet al.,Phys. Rev.Lett 85, 2244 (2000). Lett. 91, 031801 (2003). [14] K.Shinozaki et al.,Ap. J. Lett. 571, L117 (2002). [32] I. F. M. Albuquerque, G. R. Farrar, and E. W. Kolb, [15] V. A. Kostelecky´, ed., CPT and Lorentz Symmetry, Phys. Rev.D 59, 015021 (1998). (World Scientific, Singapore, 1999).

See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.