ebook img

Dark Matter and Dark Energy via Non-Perturbative (Flavour) Vacua PDF

0.61 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 Dark Matter and Dark Energy via Non-Perturbative (Flavour) Vacua

Dark Matter and Dark Energy via Non-Perturbative (Flavour) Vacua Walter Tarantino∗ King’s College London, Department of Physics, Strand, London WC2R 2LS, UK. A non-perturbative field theoretical approach to flavour physics (Blasone-Vitiello formalism) has been shown to imply a highly non-trivial vacuum state. Although still far from representing a satisfactory framework for a coherent and complete characterization of flavour states, in recent years the formalism has received attention for its possible implications at cosmological scales. In a previouswork,weimplementedtheapproachonasimplesupersymmetricmodel(freeWess-Zumino), withflavourmixing,whichwasregardedasamodelforfreeneutrinosandsneutrinos. Theresulting effectivevacuum(calledflavourvacuum)wasfoundtobecharacterizedbyastrongSUSYbreaking. Inthispaperweexplorethephenomenologyofthemodelandwearguethattheflavourvacuumisa consistentsourceforbothDarkEnergy(thankstothebosonicsectorofthemodel)andDarkMatter (viathefermionicone). Quiteremarkably,besidestheparametersconnectedwithneutrinophysics, in this model no other parameters have been introduced, possibly leading to a predictive theory 2 of Dark Energy/Matter. Despite its oversimplification, such a toy model already seems capable 1 to shed some light on the observed energy hierarchy between neutrino physics, Dark Energy and 0 Dark Matter. Furthermore, we move a step forth in the construction of a more realistic theory, by 2 presentinganovelapproachforcalculatingrelevantquantitiesandhenceextendingsomeresultsto n interactive theories, in a completely non-perturbative way. a J PACSnumbers: 14.60.Pq,95.36.+x,95.35.+d 1 1 INTRODUCTION by the theory. Such a flavour vacuum (which can be ] regarded as a vacuum condensate) represents the physi- h p NeutrinoFlavourOscillationisnowadaysafairlywell- cal state with no (flavour) particles in it. Despite being - established fact, thanks to a wide range of experimen- merely an “empty” state, the flavour vacuum is charac- p tal evidences [59]. A simple quantum mechanical model terized by a rich structure revealed by the non-vanishing e h (based on the work of Pontecorvo, Maki, Nagawa, and expectation value of the stress-energy tensor and the re- [ Sakata [36, 48, 56, 57]) is commonly considered as suf- latedequationofstate. WithinBVformalismoneisable ficient for accounting for experimental data. How- to fully describe it at a non-perturbative level, and its 2 v ever, this hides non-trivial difficulties in the formula- features depend on the specific model considered (spin, 1 tion of flavour oscillations in a Quantum Field Theoret- interactions,numberofparticlescharacterizedbyflavour 6 ical (QFT) framework [3]. Flavour states indeed do not mixing, etc.). 4 represent correct asymptotic states (by definition, since In a series of papers it was suggested that the flavour 3 their oscillating behaviour), which are required in the vacuum might behave as a source of Dark Energy [6– 2. usual perturbative approach to QFT (in the Lehmann, 10, 13, 23–25, 27]. Recently, it has been shown that 1 Symanzik and Zimmermann scheme). in a simple supersymmetric (Wess-Zumino) model with 1 Morethanadecadeago,anon-perturbativeapproachfor flavourmixing,inwhichtwoMajoranafields,twoscalars 1 buildingflavourstateswassuggestedbyBlasone,Vitiello and two pseudo-scalars were present (a simple model for v: andcoworkers(BVformalism forflavourphysics)[12]. A neutrinos and sneutrinos), the flavour vacuum was actu- i first version was proposed in 1995 [12], but some incon- ally characterized by a strong Supersymmetry breaking X sistencies in the derivation of oscillation formulae were [28, 54]. In the present work we argue that this breaking ar noticed [17, 20, 31, 32] shortly after; a revisited version, istheoriginofaninterestingphenomenology,thatmight in which these discrepancies were clarified and removed, shed some lights both on the Dark Energy and the Dark was suggested and developed [15, 18, 19, 41, 42] later on Matter problem. More precisely, in the supersymmetric [3]. However, for some aspects, the formalism remains context of [28, 54], the flavour vacuum can be thought controversial and its physical relevance is still matter of as made of two different fluids that fill in all universe: a debate [16, 34, 35]. first one related to the bosonic sector of the model, and Theapproachcorrectlyreducestothecommonquantum a second induced by the fermionic one. The former is mechanicalapproachinthesmall(neutrinos)masslimit, characterized by negative pressure, equal in modulus to but leads to corrections to those formulas that are cur- its energy density (acting as a source of Dark Energy). rently beyond experimental sensitivity [3, 26]. However, The latter is characterized by zero pressure, giving rise perhaps the most interesting feature of BV formalism is to a source of Cold Dark Matter. the non-trivial vacuum (called flavour vacuum) implied Thefirstpartofthepaper(SectionI)willbededicated to review BV formalism, complementing the original lit- erature with a discussion on non-perturbative theories ∗ [email protected] and Fock spaces. 2 In the second part of the paper (Section II) we shall that connects flavour fields ν , ν with massive ones ν , A B 1 explore in details the phenomenology of the model stud- ν . Such a relation is connected with the linearization of 2 ied in [28, 54]. We shall clarify why the flavour vacuum the following Lagrangian for free spin-1/2 fields is a good Dark Energy and Dark Matter candidate, with emphasis on this latter. More importantly, we will ex- L=iν¯ (x)∂/ν (x)+iν¯ (x)∂/ν (x)+ A A B B plain how to relate all parameters of the model to ob- −m ν¯ (x)ν (x)−m ν¯ (x)ν (x)+ servational data. This is a quite important aspect of A A A B B B −m (ν¯ (x)ν (x)+ν¯ (x)ν (x)) (4) the approach: the model introduces very few parameters AB A B B A whichareallrelatedtoneutrinophysics. Hopefully,more which becomes realisticmodelswillnotrelyonuncontrolledfreeparam- eters, leading to a truly falsifiable theory for both Dark Energy and Dark Matter. A first encouraging result in L=iν¯1(x)∂/ν1(x)+iν¯2(x)∂/ν2(x)+ this direction comes already from the simple model here −m ν¯ (x)ν (x)−m ν¯ (x)ν (x) (5) 1 1 1 2 2 2 studied: this is indeed fairly consistent with a choice of the parameters modeled on real world data, as we shall when see in the relevant section. A first step towards more realistic models will be mA =m1cos2θ+m2sin2θ moved in the last part of this work (Section III). We will m =m sin2θ+m cos2θ B 1 2 present a novel method for calculating relevant quanti- m =(m −m )sinθcosθ. (6) AB 2 1 ties,specificallythoughtforanalyzingthefeaturesofthe flavour vacuum, which might enable us to study inter- However,inthefield-theoreticalframeworkthedecom- active theories completely at a non-perturbative level. position of the fields (3) into ladder operators associated In particular, we shall show how, under reasonable as- withflavourparticlestatesishighlynon-trivial[3]. Ithas sumptions, the method can discriminate which interac- been shown, indeed, that states defined as the relativis- tions preserve the behaviour of the condensate as Dark ticequivalentof(1),forwhich|ν (cid:105)belongstothemass- 1,2 Energy/Matter source. m irreduciblerepresentationofthePoincar´egroup,are 1,2 not eigenstates of the flavour charge operators [11, 14], which for the theory (4) read [11] I. BV FORMALISM (cid:90) Q (t)= d(cid:126)xν†(x)ν (x), Neutrino oscillations can be described in a non- A A A relativisticquantummechanicalframeworkbyconstruct- (cid:90) Q (t)= d(cid:126)xν†(x)ν (x). (7) ingparticlestates,labelledbyaflavour number,thatare B B B not eigenstates of the Hamiltonian. In its simplest for- mulationfortwodistinctflavours,thePontecorvomodel, BVformalismcompensateforthis[12,39,42],bydefin- flavour states are constructed as follow [4]: ing appropriate flavour eigenstates via the action of a certain operator G on massive states: θ |ν (cid:105)=cosθ|ν (cid:105)+sinθ|ν (cid:105) A 1 2 |νB(cid:105)=−sinθ|ν1(cid:105)+cosθ|ν2(cid:105) (1) |(cid:126)k ,f ;(cid:126)k ,f ;(cid:126)k ,f ;...(cid:105)≡ 1 1 2 2 3 3 where |ν1(cid:105) and |ν2(cid:105) are massive eigenstate of the free G†θ|(cid:126)k1,m(1);(cid:126)k2,m(2);(cid:126)k3,m(3);...(cid:105) (8) Hamiltonian (particles with well defined mass m , with i i=1,2), and from which where |(cid:126)k ,f ;(cid:126)k ,f ;(cid:126)k ,f ;...(cid:105) denotes a state with 1 1 2 2 3 3 different flavour particles described by their mo- ℘A→B =|(cid:104)νB|νA(t)(cid:105)|2 =|(cid:104)νB|e−iHt|νA(cid:105)|2 menta (cid:126)k and their flavours f = A,B, whereas (cid:18) (cid:19) i i =sin2θsin2 ω1(k)t−ω2(k)t (2) |(cid:126)k1,m(1);(cid:126)k2,m(2);(cid:126)k3,m(3);...(cid:105) denotes a state with dif- 2 ferent massive particles described by their momenta (cid:126)k i (cid:113) and their masses m(i) = m1,m2, defined from the lin- with ωi(k) ≡ (cid:126)k2+m2i, describing the non-vanishing earized theory (5). The operator Gθ is defined by the probability of a flavoured particle with momentum (cid:126)k to equations: be created with a certain flavour (A) and be detected ν (x)=G−1ν (x)G later on with a different flavour (B). A θ 1 θ Theformofthetransformationbetweenflavouredand νB(x)=G−θ1ν2(x)Gθ (9) massive particles (1) is reflected in the relativistic field formalism by the relation and its explicit form depends on the specific theory con- sidered. For the theory (4), G is written as [12] θ ν (x)=ν (x)cosθ+ν (x)sinθ A 1 2 νB(x)=−ν1(x)sinθ+ν2(x)cosθ (3) Gθ(t)=eθ2(cid:82)d(cid:126)x(ν1†(x)ν1(x)−ν1†(x)ν2(x)). (10) 3 AmongallflavourstatesdefinedintheBVapproach,the framework,ascatteringprocessisrepresentedbyagraph one called flavour vacuum and defined by with a certain number of external legs (incoming and outgoing particles). The total number of internal lines |0(cid:105)f ≡G†θ|0(cid:105) (11) is connected to the precision of the approximation used in the perturbative expansion: the higher the order of plays a special role, since it represents the physical vac- the perturbation, the higher the number of vertices, and uum. In this context, by physical vacuum we mean the thereforethehigherthenumberofinternallinesinvolved. state that represents the physical empty state, i.e. with Eachlinecanbena¨ıvelyinterpretedasasinglefreeparti- no particle in it. Since only particles with well defined clestate,whichisemittedinthestartingvertexandthen flavour, rather than mass (i.e. Hamiltonian eigenstates), absorbed in the ending one. At each order in perturba- can be created/detected, one expects the physical vac- tion theory, a finite number of free particle states enters uumtoberepresentedbythestatethatcountsnoflavour in the description of the scattering process. However, particles in it. It can be shown that this is state is |0(cid:105) , f under the assumption that the perturbative series con- rather than |0(cid:105) [12].1 verges, its limit would be described by an infinite num- Furthermore, it has been proven that all flavour beroflines/one-free-particle-states. Inbra-ketformalism states|(cid:126)k1,f1;(cid:126)k2,f2;(cid:126)k3,f3;...(cid:105)areorthogonal toeachmas- suchalimitstatewouldthereforeberepresentedbyavec- sive state |(cid:126)k ,m ;(cid:126)k ,m ;(cid:126)k ,m ;...(cid:105), and therefore tor of H, the space of all physical states, but not of F , 1 (1) 2 (2) 3 (3) 0 (cid:104)0|0(cid:105) = 0 follows as a particular case. This result en- the space of states with finite number of free particles. f ables us to talk of a Fock space for flavour states, in With this example we want to remark the non-trivial opposition of the usual Fock space, whose basis is given difference existing from a free theory and an interactive by {|(cid:126)k ,m ;(cid:126)k ,m ;(cid:126)k ,m ;...(cid:126)k ,m (cid:105)|∀n∈N}. one: wecanexpressinteractiveprocessesintermsoffree 1 (1) 2 (2) 3 (3) n (n) Theorthogonalityofthetwospacesisnotverysurpris- states (which have no direct physical meaning or inter- ing if one regards BV formalism as a non-perturbative pretation,beingjustabasisinwhichwechoosetoexpress approach to the interactive theory defined by (4). our process) but only in a weakly-interactive/perturbed In the Second-quantization framework, the Hilbert framework. A full non-perturbative treatment for inter- space representing physical states is defined by vectors active particle states requires subspaces of H, that are in the number occupation representation: assuming that orthogonal to the Fock space of free states F [38]. 0 a single-particle state can be classified by a discrete set Coming back to our case, a Lagrangian with flavour of states labeled by the index i=1,2,3,..., a vector rep- mixing, such as (4), can be regarded as an interactive resenting a many-particle state can be identified by the theory, thanks to its non-diagonal terms. Flavour par- number ni of particles occupying the i-th state and it is ticle states defined `a la BV form a Fock space Ff that denotedwith|n1,n2,n3,...(cid:105);theHilbertspaceofphysical is therefore orthogonal to F0. In other words, we could states H is therefore defined as the vector space gener- express flavour states in a perturbative way by means ated by the basis {|n1,n2,n3,...(cid:105)}. For bosons ni ∈ N0, of F0; however, BV formalism enables us to construct whereas for fermions ni = 0,1. It can be shown that flavour states in a completely non-perturbative manner, in both cases the set {|n1,n2,n3,...(cid:105)} is uncountable and andthereforeitrequiresstatesthatarepartofHbutnot therefore H is non-separable [60]. In particle physics a of F . 0 separable subset of H, a Fock space that we will denote Different Fock spaces are ordinarily used in QFT on with F , is usually considered [58]. F carries an irre- 0 0 curved backgrounds and other contexts [5, 22]. In the ducible representation of the Poincar´e group and par- former, for instance, one identifies Fock spaces for physi- ticle states belonging to it have well-defined mass and cal states in flat regions. However, these Fock spaces do spin. Such a subset is spanned by the countable basis of not coincide (they are different/orthogonal subset of H) all states with an arbitrary, yet finite in total, number of if the regions are not connected and curved regions exist freeparticles. Althoughthisbasisdoesnotfullydescribe inbetween. Asaconsequence,thevacuumdefinedbyan H (for instance the vector |1,1,1,...(cid:105) which counts an in- observer in a certain region is not necessarily described finite number of particles is not included), it is sufficient by the state with no particles by an observer in another for accounting for scattering processes at a perturbative region. The particle creation phenomenon may occur: level. In the usual perturbation theory all interactive a state that is empty for an observer can actually con- processes are indeed approximated by means of a super- tain particles according with a different observer. This position of a finite number of free particle states. This is mechanism characterizes both of the two main results of quite clear in the functional formalism, when Feynman QFT in curved spacetime: the Unruh effect [61] and the diagrams are considered. In a simplified picture of this Hawking radiation [40]. Inaformalanalogy,BVformalismintroducesaground state,calledflavourvacuum,whichisnotastrivialasthe groundstateforthefreetheory. Asalreadysaid, sinceit 1 In short: flavour states can be built by means of specific cre- ation/annihilationoperatorsforflavourparticles;itcanbeshown is the state in which no flavour particles are present, it thattheonlystatethatisannihilatedbyallannihilationopera- correctly represents the physical vacuum. Even though torsistheflavourvacuumdefinedvia(11). it is empty, it is characterized by a non-zero expectation 4 value of the stress-energy tensor (cid:104)0|T |0(cid:105) , whose ef- condensate, might provide a suitable description of the f µν f fects must be gravitationally testable. This is true as low energy limit of the model. long as we fix as zero-point of our theory the usual vac- In [54] we presented the behaviour of the flavour vac- uum |0(cid:105) for the free theory and belonging to F , or, uum,inasimplesupersymmetrictheory. Themodelthat 0 in other words, we consider the usual normal ordering wasconsideredinvolvestwofreerealscalarsS (x),S (x) A B (cid:104)0|:T :|0(cid:105) ≡ (cid:104)0|T |0(cid:105) −(cid:104)0|T |0(cid:105), which is valid with mixing, two free real pseudo-scalars P (x), P (x) f µν f f µν f µν A B inperturbationtheoryaswellasinthisnon-perturbative withmixingandtwofreeMajoranaspinorsψ (x),ψ (x) A B approach. Onecommonlyreferstotheflavourvacuumas with mixing: a condensate for the following reason: once expressed in (cid:88) (cid:104) terms of particles with well defined mass (eigenstates of L= ∂ S (x)∂µS (x)+∂ P (x)∂µP (x)+ µ ι ι µ ι ι the Hamiltonian), the flavour vacuum contains an non- ι=A,B vanishing number of those particles, per unit of volume. +iψ¯(x)∂/ψ (x)(cid:105)− (cid:88) (cid:104)m2 S2(x)+m2 S2(x)+ In our example, they are characterized by the following ι ικ ι ικ ι distribution over the momentum space [26] ι,κ=A,B (cid:105) +m ψ¯(x) ψ (x) (13) sin2θω (k)ω (k)−m m −k2 ικ ι κ (cid:104)0|n((cid:126)k)|0(cid:105) = 1 2 1 2 (12) f f 4π3 ω (k)ω (k) with m =m . Terms involving products of fields of 1 2 AB BA differentflavoursdisappearwhenoneexpressesthemodel with n((cid:126)k) ≡ (cid:80) (ar†((cid:126)k)ar((cid:126)k) + ar†((cid:126)k)ar((cid:126)k)), k ≡ |(cid:126)k|, in terms of new fields, obtained by appropriate rotations r 1 1 2 2 andar(†) representingladderoperatorsforparticleswith of the previous ones: i well-defined mass. However, since the physical degrees φ (x)=cosθφ (x)+sinθφ (x) A 1 2 of freedom of the theory are flavour particles (the only φ (x)=−sinθφ (x)+cosθφ (x) (14) kind of particle that can be produced and detected), the B 1 2 interpretation as a gas or collection of particles remains with φ=S,P,ψ, leading to at a mere mathematical level, the flavour vacuum be- ing absolutely “empty” from a physical point of view (in L= (cid:88) (cid:104)∂ S (x)∂µS (x)−m2S2(x)+ µ i i i i the sense that no flavoured particles are present in it), i=1,2 and only characterized by a non vanishing stress-energy +∂ P (x)∂µP (x)−m2P2(x)+ tensor expectation value which is detectable via gravita- µ i i i i (cid:105) tional effects. +ψ¯(x)(i∂/−m )ψ (x) . (15) i i i The features of the flavour vacuum depend on the model considered and a preliminary investigation on a From this latter it is possible to build the usual Fock simple supersymmetric model [54] showed that it might space for massive particles, previously denoted as F , 0 behave very differently, according with the spin of the which has as ground state the “massive” vacuum |0(cid:105). particles involved. The flavour vacuum is hence defined as |0(cid:105) ≡eθ(cid:82)d(cid:126)x(X12(x)−X21(x))|0(cid:105) (16) f II. PHENOMENOLOGY OF A SUSY FLAVOUR with VACUUM 1 X (x)≡ ψ†(x)ψ (x)+iS˙ (x)S (x)+iP˙ (x)P (x). A. Free WZ `a la BV 12 2 1 2 2 1 2 1 (17) Its features have been explored via its stress-energy ten- Our interest in BV formalism was firstly motivated sor expectation value, being by physics beyond the Standard Model. The Wess- Zumino model here discussed has been considered in (cid:88)(cid:104) T (x)= 2∂ S (x)∂ S (x)+2∂ P (x)∂ P (x)+ [54] after two works in which the flavour vacuum has µν (µ i ν) i (µ i ν) i i=1,2 been regarded as an effective vacuum arising in a string- (cid:105) theoretical framework [51, 53]. Indeed, a specific model +iψ¯(x)γ ∂ ψ (x) −η L. (18) i (µ ν) i µν from the braneworld scenario, called D-particle foam model [29, 30, 49, 50, 52], seems to explain neutrino It has been shown that the flavour vacuum behaves as a flavour oscillations in terms of flavour oscillation of fun- perfect relativistic fluid, i.e. damental strings, in presence of a “cloud” (or foam) of (cid:104)0|T |0(cid:105) =diag{ρ,P,P,P} (19) point-like topological defects in the bulk space. In the f µν f spirit of weak coupling string theory, the interaction be- with tween the foam and strings/branes in the theory can be ρ≡ (cid:104)0|T (x)|0(cid:105) = regardedas“vacuumdefects”fromthepointofviewofa f 00 f macroscopical observer. Therefore it has been suggested (m −m )2 (cid:90) K (cid:18) 1 1 (cid:19) (20) =sin2θ 1 2 dk k2 + that BV formalism, together with its “flavour vacuum” 2π2 ω (k) ω (k) 0 1 2 5 P≡ (cid:104)0|T (x)|0(cid:105) = f jj f cannotexplaintheobservedstructuresatgalacticscales, m2−m2 (cid:90) K (cid:18) 1 1 (cid:19) (21) thereforeDarkMatterisexpectedtobemadeoutoffairly =−sin2θ 1 2 dk k2 − 2π2 ω (k) ω (k) massiveand“cold”(non-relativistic)particles. Big-Bang 0 2 1 nucleosynthesislimitsontheaveragebaryoniccontentof ρ representing its energy density, P its pressure and K a the Universe exclude that (the majority of) Dark Mat- momentum cutoff (cf [28, 54]).2 terismadeoutofordinarybaryonicmatter(i.e. atoms). In particular, disentangling the contribution of the Furthermore,although“dark”,inthesensethatdoesnot bosonic sector from the fermionic one, one finds emit nor absorb light (i.e. electromagnetically neutral), Dark Matter might couple to ordinary matter in other (cid:90) K k2 ways (besides gravity); however, arguments on its den- ρb =sin2θ dkπ2 (ω1(k)+ω2(k))× sity and thermal production at early times imply that 0 such a coupling must be weak. (cid:32) (cid:33) (ω (k)−ω (k))2 × 1 2 (22) Both astrophysics and particle physics have been 2ω (k)ω (k) 1 2 proposing suitable candidates for Dark Matter through thelastthreedecades, givingrisetoanenormouswealth of choice. However, because of the absence of direct de- (cid:90) K k2 tectionsandthelackofpredictionsbytheoreticalmodels, ρ =sin2θ dk (ω (k)+ω (k))× f π2 1 2 plaguedbyanundesirableabundanceoffreeparameters, 0 (cid:18)ω (k)ω (k)−m m −k2(cid:19) the nature of Dark Matter remains elusive. × 1 2 1 2 (23) The fermionic sector of the flavour vacuum in the ω (k)ω (k) 1 2 model here presented clearly fulfills basic requests for a Dark Matter candidate: it contributes to the energy P =−ρ , P =0 (24) content of the universe; it is “dark” (i.e. it is an electro- b b f magnetically neutral object, since (s)neutrino fields do in which the standard normal order has being adopted. not couple with the electromagnetic field); furthermore, it does not interact with any other of the SM particles (excluding gravitational effects), being the empty state B. Flavour vacuum as a source of Dark Matter forthe(s)neutrinosector; unlikeitsbosoniccounterpart, it is purely pressureless.3 An important result emerges from the above outlined A possible concern about its uniform distribution in analysis: the equation of states w ≡P/ρ for the bosonic space, incontrastwiththeobserveddistributionofDark and the fermionic sectors are different, w = −1 and b Matter which is usually gathered in clusters around and w = 0 holding. The emphasis on the novelty of this f inside galaxies, can be easily dispelled by recalling that SUSY breaking mechanism has already being remarked we are actually modeling a simple “empty” universe. If [54]. We are now aimed to explore the interesting phe- a non-uniform matter distribution is considered in our nomenology connected with such a result. Our simple toy universe in addition to the flavour vacuum, it would model implies a physical vacuum that is a combination start to interact gravitationally with our vacuum con- oftwofluidswhichbehavequitedifferently: bothperme- densate. Thanks to initial irregularities in the matter atetheemptyspaceuniformlyandstatically,butonehas distribution, we expect them to form clusters via gravi- a cosmological-constant-like behaviour (w = −1), while tational instability (gravity tends to enhance irregulari- theotherbehavesasdust(w =0). Theroleoftheflavour ties, pulling matter towards denser regions [47]), as the vacuum as source of Dark Energy (which now is played system evolves with time. It is known that such an ef- onlybythebosonicsectorofthetheory)hasbeenexten- fect, on the other hand, does not necessarily occurs for sively discussed in literature [6–10, 13, 23–25, 27]. Here Dark Energy-like fluids [1], as the bosonic component of we present a new feature of the flavour vacuum: its con- the flavour vacuum, which can persist in their state of tribution to Dark Matter. spatial uniformity even in presence of clustered matter. Dark Matter is the name given to unknown sources Theevolutionofourflavourvacuum,consideringbothits of gravitational effects, whose presence, primarily within bosonic and fermionic components, in presence of other andaroundgalaxies,hasbeenestablishedbymanyastro- matterandgravitationalinteraction,representsnecessar- physical data [2, 43]. Numerical simulations of structure ily an object of future studies. formation have shown that “hot” (relativistic) particles 2 FromtheperspectiveofconsideringBVformalismasaneffective formalismforphysicsbeyondtheStandardModel[51,54],such 3 This is certainly true for the free above mentioned model; the acutoffmustbeinterpretedastheenergyscaleuptowhichthe possibility of extending this result to interactive models will be formalismprovidestheframeworkforagoodeffectivetheory. discussedlateron. 6 C. Testability gles, and we can constrain parameters induced by SUSY breaking, the cutoff is the only parameter left to deter- mine. An interesting aspect of the model concerns the in- As our toy universe expands, we assume that the two terplay between the two fluids. Supersymmetry imposes fluidsobeyEinsteinequationsandthereforetheyscaleas that the energy density of the bosonic component is tied up to the energy density of the fermionic component; ρ (t) a(t)3 =ρ a3 in a more realistic theory, therefore, one should be able f f 0 (25) to reproduce the current experimental value of the ratio ρb(t)=ρb. between the Dark Energy density and the Dark Mat- This means that today their value is ter energy density (∼ 2.8), in the optimist belief that the flavour vacuum is the only responsible for both of ρ (t )=ρ a3 them. The role of a curved background in the formu- f now f 0 (26) lation of the theory might be crucial, since the energy ρb(tnow)=ρb. density of a dust-like fluid gets diluted by the expansion of the universe, whereas such an effect does not occur respectively, being a(tnow) = 1 by convention. Those for a cosmological-constant type, and therefore the ratio two quantities depends on the following parameters: between those two quantities changes dramatically with (s)neutrino masses, mixing angles, cutoff energy, scale time. factor at t0. Provided with these expressions, we can then test our model in two ways. Within a momentum cutoff regularization framework, as the one here presented, the two energy densities de- 1.If observational data enable us to constrain pend on such a cutoff, which is the same for both quan- (s)neutrino masses, mixing angles, Dark Matter and tities. The ratio between them can in general be cutoff Dark Energy densities, from (26) we can derive the dependent, as it actually is in the case here presented. otherparametersleft: thecutoffenergyandthescale On one side, one might hope that in a more realistic factor a0. Well equipped with all the parameters of theory (on a curved background, for instance) the ratio thetheory,wewillthenbeabletocheckifthemodel might be cutoff independent. On the other hand, one isinreasonableagreementwithotherstandardmod- could consider the opposite situation, in which the ratio els. Forexample,ifthescalefactora0 fittingalldata varies with the cutoff, as highly desirable: if the ratio corresponds to a time in the future (for a0 >1), the is fixed from the cutoff, the same value for the cutoff model has to be rejected, or corrected at least. would also fix the value of the energy. This implies that 2.On the other hand, theoretical reasons might sug- once the cutoff is decided on the basis of experimental gest specific values for the cutoff (if for instance the data on the ratio, the model gives a precise prediction flavourvacuumrisesinthelowenergylimitofanun- fortheabsolutevaluesoftheenergydensities, whichcan derlyingtheory,and/orthescalefactora0,beingthe be compared with their observational estimates. temperature of the universe inversely proportional In order to illustrate these ideas we will present a con- to its scale factor. In this case, we might be able crete example. Let us assume that our supersymmetric to make a prediction on the value of the Dark Mat- modeliseffectiveuptotheenergyscaleK (whichcomes terandDarkEnergydensity,viaformulae(26),that from deeper theories, as, for instance, in [54] ). In the mightbecomparedwithobservationalestimates. On standard Big-Bang picture, this means that when the thisbasisourmodelisthereforeacceptedorrefused. universe cools down to that energy, the flavour vacuum starts to be the effective description of the vacuum state of the (unknown) underlying theory. We call t the time D. A Preliminary Test 0 correspondingwiththistransitionanda thecorrespond- 0 ing scale factor. The simple toy model discussed in Section IIA is not Inourtoyuniverse,weassumethatatt ,inabsenceof realisticenoughtoholdthecomparisonwithdataalready 0 anyothersourcesofenergyormatter,theenergy/matter available: only two generations of neutrinos have been contentofourtoyuniverseisonlyduetotheflavourvac- considered, neither matter or interactions are present, uum. Moreover, we assume that it can be describe, at SUSY is unbroken, there is no prescription for the cutoff a classic level (i.e. on sufficiently large scales), in terms K.4 However, some preliminary tests can be performed. of two fluids: a first one, due to the bosonic component As just explained, in a realistic theory with three gen- of the flavour vacuum and described by ρ and w =−1, erations instead of two, it is possible to constrain the b and a second one, due to its fermionic component and parameter space of mixing angles and masses thanks to described by ρ and w = 0. We will regard the bosonic observational data. In absence of such a theory, we will f component as the only source of Dark Energy and the fermionicastheonlysourceofDarkMatter. Bothρ and b ρ are function of the following parameters: (s)neutrino f masses, mixing angles, and the cutoff K. If we know 4 The assumption of treating neutrinos as Majorana particles (from observations) the neutrino masses and mixing an- mightalsobequestioned. 7 limit our selves to check if our simpler model admits a (26) hold). We will further assume that some radiation choiceofparametersthatgivesrisetophysically“plausi- and matter are present in our toy universe, whose den- ble” estimations for Dark Energy and Dark Matter den- sity is at least one order lower than the flavour vacuum sities. Moreprecisely,weshallcheckthecompatibilityof density. Their presence justifies the notion of “tempera- our model with the relation ture” and defines the profile of the time-evolution of our toy universe. Since we assume the neutrino sector not ρΛ ∼ρDM ∼(∆m2ij)2, (27) being coupled with any other fields, the flavour vacuum and the matter/radiation content of the universe inter- with ρΛ/DM the Dark Energy/Matter density today and act only gravitationally. As mentioned, we expect this ∆mij the difference of the squared masses of neutrinos5, interaction to lead the fermionic flavour vacuum to clus- which for our model becomes ter together with ordinary matter, leaving the bosonic flavourvacuumhomogeneouslydistributed. Theseeffects ρb ∼ρf(tnow)∼(∆m2)2, (28) are reasonably expected as long as gravitational effects are relevant only on cosmological scales, at which the intheassumptionthatallDarkEnergyandDarkMatter flavour vacuum is well approximated by a classical fluid. ofourtoyuniverseisduetotheflavourvacuum. Inother As a consequence of a possible non-uniform distribution words, is there a sensible region of the parameter space inspaceofthefermionicfluid,weshallconsiderthevalue that gives rise to (28)? of (23) not as a local attribute but as a global, or “aver- Once provided with a realistic theory, the reasoning aged” over sufficiently large scales, property. goestheotherwayaround: giventhespaceofparameters In order to check the compatibility of our model with constrainedbyobservationaldata, does(28)hold? How- (28),westartbydefiningthequantityξ ≡(1−m2/m2)2, 2 1 ever, the analysis on which we are embarking is neither withm thesmallerofthetwomasses(m <m ). Since 2 2 1 irrelevantnornegligible: previousanalyseshardlyconcil- 0<ξ <1, we can expand (22) and (23) in series around iate the very different scales of energies entering into the ξ =0 and get to problem, such as the momentum cutoff, which presum- ably is greater then the TeV scale, and neutrino mass sin2θ ρ = m4f(K/m )ξ+O(ξ3/2) differences (cf for instance [24]). The aim of this section b π2 1 1 is therefore to show that even in our simple toy model, sin2θ non-perturbative formulae describing the features of the ≈ π2 (∆m2)2f(K/m1) flavourvacuumcanaccommodateverydifferentscalesin sin2θ anaturalway,givingrisetophysicallysensiblevaluesfor ρ = m4g(K/m )ξ+O(ξ3/2) f π2 1 1 Dark Energy and Dark Matter densities. sin2θ Recapitulating, in the following we shall assume that ≈ (∆m2)2g(K/m ) (29) thephysicalvacuumiseffectivelydescribedbytheflavour π2 1 vacuum defined by (16) for energies lower than K; over with largedistancescales,suchaflavourvacuumbehavesasa classicalfluids, obeyingEinsteinequations(i.e. (25)and (cid:90) K/m1 x2 f(K/m )= dx (30) 1 4(1+x2)3/2 0 (cid:90) K/m1 x4 g(K/m )= dx (31) 5 The energy scale of Dark Energy is far away from all natural 1 0 4(1+x2)3/2 scalesprovidedbytheSMviaparticlemasses. Onlyonefunda- mental scale is known to be comparable with the Dark Energy and m4ξ = (∆m2)2 ≡ (m2 −m2)2. Relations (29) are 1 1 2 one: thescaleofneutrinophysics. Boundaryontotalmassesof good approximations of the exact values, as long as the neutrinos show that they are much lighter then all other parti- two masses are very similar m ∼ m . All divergen- cles: astrophysicaldataindicatethatΣmν <058eV,with95% ciesconnectedwithourproblem1areinc2ludedinfunction ofconfidence[44](thesumrunsoverallpossiblespecies-possi- blymorethenthree-thatwherepresentintheearlyUniverse). f(K/m1) and g(K/m1), for their argument running to Moreover, direct observations on solar and atmospheric neutri- infinity. Inthefollowinganalysisaphysicalcutoffofmo- (nboesinshgomw2th−atm∆2m≡212∆≈m28·,1c0f−[54]eVan2danredfe∆remnc22e3s≈th2e.r5e·in1)0.−T3heVes2e menta (K, rescaled by the neutrino mass m1) will be i j ij considered, in the belief that flavour physics `a la BV massscales10−1÷10−2eV havetobecomparedwiththescale must be regarded as an effective description at low en- 10−3eV, that one obtains from ρΛ = 3×10−11eV4. This “co- ergy scales of a deeper theory [51]. Clearly other renor- incidence” gave rise to many works, besides the ones connected with BV formalism, aimed to provide a theoretical explanation malization tools might be required if this assumption is forit(see[37]andreferencestherein). dropped, for instance, in a pure self consistent quantum The more famous coincidence problem regarding Dark Energy field theoretical approach. Despite this choice, it is re- concerns the similar density of Dark Energy and Dark Matter markable, however, that the relation ρ ∝(∆m2)2 has (ρΛ ≈ 2.8ρDM) as measured today, which requires a notable b/f fine-tuning of initial conditions considering their very different been derived entirely analytically. evolutionintime. Inthefollowing,wewouldliketoshowthatinacutoff- Thesetwo“coincidences”arecombinedtogetherinformula(27). regularization scheme the two functions in (30) can give 8 rise to physically sensible values, i. e. the function sin2 Θ f(K/m1) remains relatively small even when the cutoff 1.0 isveryhigh(givingrisetothehierarchy: highcutoff/low DarkEnergydensity),whereasg(K/m )canbeconsider- 1 0.8 (cid:72) (cid:76) ablygreater then f(K/m )for thesamechoice ofcutoff, 1 motivating the observed discrepancy between the Dark 0.6 Energy and Dark Matter densities at early times. We should stress once more that a strict comparison with available experimental data would be possible once a 0.4 more realistic model will be given. We will now focus on the former of (29). Being SUSY 0.2 LHC Planck unbrokeninourtoymodel,neutrinosandsneutrinoshave K thesamemasses. Sothem andm appearingin(29)are the masses of two neutrino1s, even t2hough ρ encodes the 105 1010 1015 1020 1025 1030 m1 b contribution of the bosonic sector of the theory, which in a realistic case would be affected by the breaking of FIG. 1. The parameter space (K/m1,sin2θ) is plotted. In SUSY via effective masses greater than m . Recalling blue,theacceptanceregionforthecondition(32)(boundaries i oftheregioncorrespondtolhs=0.9andlhs=1.1). Physical theobservedrelation(27), wewondernowifisthereany mixinganglesforthreegenerationofneutrinosare: sin2θ = region of the parameter space (θ,K) that might generate 13 0.000+0.028, sin2θ = 0.30+0.04, sin2θ = 0.50+1.4 (within a similar situation (i.e. ρ ∼ (∆m2)2) in our model. 12 −0.05 23 −1.2 b 2σ) [33]. These values are also plotted in the graph (red Quite interestingly, the condition regions). DottedlinesrepresentsLHCenergyscale(∼1TeV) andthePlanckenergyscale(∼1028eV)fixingm =10−2eV. sin2θ 1 f(K/m )∼1 (32) π2 1 is satisfied by a region of the plane (θ,K), which is not We already know that a must be extremely small, but 0 that far from the expected value for a realistic theory. what is the ratio between g(K/m ) and f(K/m ) for 1 1 As shown in Figure 1, if the cutoff of the theory lies large K/m (K/m > 1014)? As shown in Figure 2, 1 1 somewhere between the TeV (1012 eV) scale and the equation (35) is indeed satisfied for large values of K Plankscale(1028 eV),thevalueofsin2θ mustbearound and very small values of a , as one would expect from a 0 the 0.5÷1 region. Moreover, a complete overlap with realistic theory. If, for instance, the cutoff is set equal one real mixing angle is obtained in a region with a to the Planck scale, relation (35) is satisfied for a = 0 very high cutoff, close to the Planck scale. It should be 10−20, which sets the transition phase (when the flavour emphasized that the existence itself of such a region is vacuum became effective) well far in the past, when our highlynon-trivial, sinceitappearsfromthecombination toy universe was 1020 times smaller than “now”. Once of parameters spanning a wide range of energies, being more, parameters characterized by very different values ∆m2 ∼10−4eV2 and K >1012 eV. (a ∼ 10−20, K/m ∼ 1030) combine together to give 0 1 Focusing now on the latter of (29), we shall proceed rise to a third scale, which has a physical significance withasimilaranalysisinordertotestthehypothesisthat (nowadays Dark Matter density). thefermionicsectorofthemodelprovidesasensibleDark Despite the unrealistic nature of our toy model, these Matter candidate. As explained in the previous section, first quantitative tests go in the right direction and cer- under the assumption that the flavour vacuum behaves tainlymotivatethestudyofmorerealisticmodels,which as a perfect classical fluid on large scales, the fermionic hopefully will share with our toy model all the good fea- contribution would get diluted with time as the universe tures here discussed. expands. Its value today would then be ρ (t )=ρ a3 (33) f now f 0 III. TOWARDS INTERACTIVE FLAVOUR with a the scale factor corresponding to the time at VACUA 0 which the model became effective. In order to reproduce the relation (∆m2)∼ρ (t ) we expect f now A. A new method of calculation: Free WZ Revised sin2θ a30 π2 g(K/m1)∼1 (34) Following the standard literature, the results in [54] discussed so far have been derived by the following ap- that can be obtained by combining (33), (28) and (29). proach: BecauseoftheconstrainontheDarkEnergydensity(32) 1.stress-energy tensor has been written in terms of the above condition becomes massive fields; g(K/m ) 2.massive fields have been decomposed in terms of a3 1 ∼1. (35) 0f(K/m ) massive ladder operators; 1 9 a Recalling the discussion in Section IIA, in the study of 0 1020 the free WZ model we can consider the bosonic and the fermionic component separately, by evaluating relevant quantities in two separated contexts (a bosonic theory 1010 and a fermionic one) and eventually combining together theresults. Furthermore,thepseudoscalarandthescalar EXPANDINGUNIVERSE field are indistinguishable for our purposes, therefore we 1 CONTRACTINGUNIVERSE are allowed to consider just the scalar field, keeping in mind to sum its contribution to the relevant quantities 10(cid:45)10 Planck twice. In the real scalar case, we have LHC K 10(cid:45)21005 1010 1015 1020 1025 1030 m1 T0b0(x)=(cid:88)(cid:18)πi2(x)+(cid:16)∇(cid:126)φi(x)(cid:17)2+m2iφ2i(x)(cid:19) (36) i FIG.2. Theparameterspace(K/m1,a0)isplotted. Inblue, with π ≡φ˙ , the conjugate momentum of φ . Since the acceptance region for the condition (32) (boundaries of i i i the region correspond to lhs = 10 and lhs = 0.1). The red G (t)=eiθ(cid:82)d(cid:126)x(π2(x)φ1(x)−π1(x)φ2(x)) (37) line depicts the scale factor of the real universe as a function θ of its average temperature, fixing m = 10−2eV. In a more 1 from which realistictheory,theacceptanceregion(whichmaydifferfrom the one here presented) is further reduced by comparison of G (t)φ (x)G†(t)=φ (x)cosθ−φ (x)sinθ the first of (29) with Dark Energy data, which gives a con- θ 1 θ 1 2 (38) straintonpossiblemomentumcutoffs;themodelisruledout G (t)φ (x)G†(t)=φ (x)sinθ+φ (x)cosθ θ 2 θ 1 2 if the resulting region lies below the red line, since it would predict an amount of Dark Matter greater than what is ob- and served; the model is consistent with data on Dark Matter if the region overlaps or lies above the red line; in this latter Gθ(t)π1(x)G†θ(t)=π1(x)cosθ−π2(x)sinθ (39) casetheflavourvacuumcontributesonlytoafraction oftotal G (t)π (x)G†(t)=π (x)sinθ+π (x)cosθ Dark Matter density. θ 2 θ 1 2 via the Baker-Campbell-Hausdorff formula 3.massiveladderoperatorshavebeenwritteninterms 1 1 eYXe−Y =X+[Y,X]+ [Y,[Y,X]]+ [Y,[Y,[Y,X]]]+... of flavour ladder operators; 2 3! 4.hence,stress-energytensorhasbeenwritteninterms (40) of flavour ladder operators; we can write 5.flavour-vev of the stress-energy tensor has been re-     ducedbyactingwiththeflavourladderoperatorson (cid:88) (cid:88) the flavour vacuum. Gθ(t) πi2(x)G†θ(t)= πi2(x), (41) i=1,2 i=1,2 However, such a long procedure can be avoided and the sameexactresultmaybeobtainedinamuchshorterway, following the steps:   1.flavour vacuum is written as |0(cid:105)f = G†θ|0(cid:105), the Gθ(t)(cid:88) (cid:16)∇(cid:126)φi(x)(cid:17)2G†θ(t)= flavour-vev of T becoming a vev of O ≡G T G†; i=1,2 µν θ µν θ 2.the stress-energy operator is transformed under the   action of the Gθ(cid:4)G†θ, using (9); the transformed op- (cid:88) (cid:16)∇(cid:126)φi(x)(cid:17)2 (42) erator O will be expressed in terms of the flavour i=1,2 fields, rather then the massive ones; 3.(3) is used to write the flavour fields in terms of the and massive fields; O is again expressed in terms of the massive fields, but this time the operator G , and (cid:104)0|G (t)(m2φ2(x)+m2φ2(x))G†(t)|0(cid:105)= θ θ 1 1 2 2 θ its complicated exponential structure, is not present (cid:104)0|(m2φ2(x)+m2φ2(x))|0(cid:105)+ any more; 1 1 2 2 4.a vev of O, expressed as a simple combination of sin2θ(m21−m22)(cid:104)0|(φ22(x)−φ21(x))|0(cid:105). (43) massive fields, is left, which can be reduced by de- It follows that composingthemassivefieldsintomassiveladderop- erators. (cid:104)0|T (x)|0(cid:105) =(cid:104)0|T (x)|0(cid:105)+ f 00 f 00 As a neat example of the above procedure, we will +sin2θ(m2−m2)(cid:104)0|(φ2(x)−φ2(x))|0(cid:105) (44) derive the results of Section IIA via this new method. 1 2 2 1 10 and therefore its (anti-)commutation rules, regardless of the tensorial or spinorial structure of the field itself. Furthermore, ex- ρ = (cid:104)0|:T (x):|0(cid:105) = tending these results to more then two flavours is rather b f 00 f straightforward. =sin2θ(m2−m2)(cid:104)0|(φ2(x)−φ2(x))|0(cid:105) (45) 1 2 2 1 Finally, the method enables us to distinguish among all the terms of the stress-energy tensors the ones that Equivalently for T (x) we have jj reallycontributetothefinalresult. Thismightbehelpful in understanding the behaviour of the flavour vacuum in P = (cid:104)0|:T (x):|0(cid:105) =−sin2θ(m2−m2)× b f jj f 1 2 morerealistictheories,asweshallseeintheforthcoming ×(cid:104)0|(φ22(x)−φ21(x))|0(cid:105)=−ρb. (46) sections.6 Oncethefieldsin(45)and(46)aredecomposedinterms of the ladder operators and the quantum algebra is sim- B. Self-interactive Bosons plified, expressions (22) and (23) are correctly repro- duced. Anexampleoftheapplicationsjustdiscussedisoffered In the fermionic case, a similar procedure leads to by a λφ4 model. The theory ρf =f (cid:104)0|:T0f0(x):|0(cid:105)f =sin2θ(m1−m2)× L= (cid:88) (cid:0)∂µφi∂µφi−m2iφ2i −λφ4i(cid:1) (53) ×(cid:104)0|(cid:0)ψ¯2(x)ψ2(x)−ψ¯1(x)ψ1(x)(cid:1)|0(cid:105) (47) i=1,2 and can be regarded as derived from a model with flavour mixing: P = (cid:104)0|:Tf(x):|0(cid:105) =0. (48) f f jj f L=∂ φ ∂µφ +∂ φ ∂µφ −m2φ2 −m2φ2+ µ A A µ B B A A B B By comparing (47) and (45), the analogy between the (cid:88) fermionic and the bosonic condensate that earlier was −m2ABφAφB − gικλρφιφκφλφρ (54) hidden in formulae (22) and (23) is now more evident. ι,κ,λ,ρ=A,B Again,formula(23)iscorrectlyreproduced,oncetheop- with the usual rotation eratorial structure of the fields is simplified with respect to (cid:104)0|(cid:4)|0(cid:105). The expression (47) dispels any doubts con- φ =cosθφ −sinθφ A 1 2 cerning formula (23) and its possible dependency on the (55) φ =sinθφ +cosθφ specific form of the gamma matrices and spinors used to B 1 2 achievetheresultsof[54],being(47)independentofsuch and a specific choice of the coupling constants g(cid:4). Since a choice [55]. theexpressionofG intermsofthefieldscanbededuced θ Furthermore, Supersymmetry enables us to rewrite from this result in terms of the bosonic fields only. For the massive vacuum |0(cid:105) we know that G†φ G =cosθφ −sinθφ θ 1 θ 1 2 (56) (cid:104)0|Tµν(x)|0(cid:105)=0 (49) G†θφ2Gθ =sinθφ1+cosθφ2 which leads to justusingcommutationrelationsbetweenfieldsandcon- jugate momenta, which are not modified by the form of (cid:104)0|ψ¯(x)ψ (x)|0(cid:105)=−4m (cid:104)0|φ2(x)|0(cid:105) (50) the Lagrangian [55], expression i i i i and hence G =eiθ(cid:82)d(cid:126)x(φ˙2φ1−φ˙1φ2) (57) θ ρ =2sin2θ(m −m )2(cid:104)0|(cid:0)φ2(x)+φ2(x)(cid:1)|0(cid:105) (51) WZ 1 2 1 2 P =−2sin2θ(m2−m2)(cid:104)0|(φ2(x)−φ2(x))|0(cid:105) (52) 6 In addition, we would like to stress that the method does not WZ 1 2 2 1 requireanexplicitdecompositionoftheflavourfieldsintermsof flavourladderoperators. Suchadecompositionhasbeenobject in accordance with (20) and (21). ofadebateinliterature,raisedbytheauthorsof[31]. Although The procedure exemplified in the previous section can the problem was exhaustively discussed in [20] and [32], not all beeasilyimplementedforotherfields: onemightwantto the community was convinced by the arguments presented [34]. considerDiracortwocomponentWeylspinorsaswellas Without entering into the details of the dispute, here we would complex scalar fields, getting to analogous results. For like to suggest that a different point of view on the formalism, suchastheoneofferedbyformulae(47)and(45),whereanob- mere speculative reasons, applications to vector fields servablequantityconcerningaflavourstatehasbeencalculated or even more complex objects might be thought: the withouttheexplicituseofthecontroversialdecomposition,might methodinvolvesamanipulationofthestress-energyten- helpinadeeperunderstandingoftheproblemandtheformalism sor, with the use of equation of motion of the field and itself.

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.