Table Of Content

EPJWebofConferenceswillbesetbythepublisher DOI:willbesetbythepublisher c Ownedbytheauthors,publishedbyEDPSciences,2015 (cid:13) Nucleosynthesis constraints on the faint vector portal AnthonyFradette1,MaximPospelov1,2,JosefPradler3,a,andAdamRitz1 1DepartmentofPhysicsandAstronomy,UniversityofVictoria,Victoria,BCV8P5C2,Canada 2PerimeterInstituteforTheoreticalPhysics,Waterloo,ONN2J2W9,Canada 3InstituteofHighEnergyPhysics,AustrianAcademyofSciences,A-1050Vienna,Austria 5 1 0 Abstract. New Abelian U(1) gauge bosons V can couple to the Standard Model through mixing of the ′ µ 2 associated field strength tensor V with the one from hypercharge, FY . Here we consider early Universe µν µν n sensitivitytothisvectorportalandshowthattheeffectivemixingparameterwiththephoton,κ,isbeingprobed a forvectormassesintheGeVballparkdowntovalues10 10 .κ.10 14wherenoterrestrialprobesexist. The − − J ensuingconstraintsarebasedonadetailedcalculationofthevectorrelicabundanceandanin-depthanalysisof 2 relevantnucleosynthesisprocesses. ] h p 1 Introduction the following we will concentrate on a Stückelberg ori- - p gin of m that allows to maintain gauge invariance in V e The origins of our Universe may well be rooted in infla- U(1) withoutcomplicatingthephenomenologybyhidden ′ h tion or alternative cataclysmic scenarios that regard the Higgsparticlesh;seee.g.[4]forthephenomenologyand [ ′ veryearliestmomentsofexistence. However,despitethe [5]forcosmologicalconstraintsonthelatterscenario. 1 impressive success of observational cosmology over the The SM decay modes of V are well known. When v past decades, the earliest true direct window into the be- hadronicdecaysarekinematicallyaccessible,onecanuse 9 ginningsremainobservationsoflightelementabundances. experimentaldataontheR-ratiotoinfercouplingstopho- 5 Theyconcerntheepochofprimoridalnucleartransforma- 4 tonsinthetime-likedirection,andhencetodeterminethe tions at cosmic times t & 1s. The overall concordance 0 decayrateΓ andallbranchingratios.Belowthedi-muon V 0 of the Big Bang nucleosynthesis (BBN) predictionswith threshold and for m > 1MeV the vector V decays to V . theobservationallyinferredprimordialvaluesisoneofthe 1 electron-positron pairs only, thereby setting its principal 0 mostimpressivesuccessesofmoderndaycosmologyand lifetime, 5 particle physics. Today, BBN is used as a toolboxto put 1 models of new physics to a stringent test [1], whenever 3 1GeV 10 12 2 v: they predictsome interferencewith the the standardpro- τV ≃ κ2αm =270s× m κ− ! , (2) i cessesintheobservablesectoratt&1s. V V X Undertheassumptionofacanonicalsequenceofcos- whereαiselectromagneticfinestructureconstant. Inthe r mologicalevents,theUniverseemergedfrominflationand a followingthecosmologicalconsequencesofU(1) vectors ′ baryogenesismuchpriortoBBN.Suchsequencethenal- with masses in the MeV-GeV range, and lifetimes long lowsonetoputstringentconstraintsonveryweaklyinter- enoughforthedecayproductstodirectlyinfluenceprimor- actingsectorsofnewphysicsbeyondtheStandardModel dial nucleosynthesis are explored. These vectors have a (SM). The kinetic mixing of a new U(1) vector V with ′ µ parametricallysmallcouplingto theelectromagneticcur- hypercharge FY Vµν is of particular interest as the mix- µν rent,andthusanextremelysmallproductioncrosssection ingwiththephotonleadstonumerousexperimentalcon- for e+e Vγ, σ κ2πα2/s 10 54cm2 wherewe − prod − sequences and much attention was devoted to this vector → ∼ ∼ took √s = 200 MeV and κ = 10 12 from above. Such − portalinrecentyears[2].Belowtheelectroweakscale,the smallcouplingsrenderthesevectorstatescompletelyun- couplingofV totheSMisessentiallygivenbyitsmixing detectableinterrestrialparticlephysicsexperiments. withthephoton[3], Despitethetinyproductioncrosssection,anycharged κ SM state that is populated in appreciable number in the = F Vµν =eκV Jµ . (1) LV −2 µν µ em early Universe at temperature T mV may yet emit V. ∼ With the above ballpark numbers in (2), parametric es- With κandm beingtheonlyfreeparameters,themodel V timates suggest that an amount of MeV/baryon may be provides a simple, and technically natural prototype sce- storedinV-particles.FollowedbylatedecaysbacktoSM narioforalight,weaklyinteractingnewparticlesector.In states, visible energy is therefore being injected into the ae-mail:[email protected] primordialplasma atlevelsthatare probedbyBBN. The EPJWebofConferences abundancecanbeperformedexplicitly, ¯ l κ Y(e) = 9 m3VΓV→ee¯. (5) V 4π(Hs) Aµ Vµ T=mV While further leptonic production channels are easy l time to be included into (5), hadronicproductionchannelsre- quire assumptions about the primordial hadron gas and Figure1. Darkphotoninversedecaysaretheleadingcontribu- the strength of interaction with photons, denoted by α . eff tion to sub-Hubble production rates inthe calculation of the V At temperatures above the QCD confinement scale T c ∼ relicdensity. 200MeVlightquarksaredeconfinedandindividualquark contributionscanbeaddedtoY inastraightforwardman- V ner. BelowT onemayuseafreegasofmesonsasanap- c proximation to the hadronic (non-baryonic)particle con- early Universe is therefore likely to be the only “labora- tent in the early Universe. The production via inverse tsomryal”lewrhaererebseuincghpvreorbyedda.rkHeprheowtoensrewpiotrhtoκn∼th1e0d−e1t2aialnedd tchheanrgbeedinpciolundeadnduskinaognadseccaalayrsQ{πE+Dπ−m,oKd+eKlw−}ith→effVectciavne analysisperformedin[7]wherealsoCMBlimitsoneven couplingstrengthslikeαπeffπ(mV)=κ2αππ(√s=mV)where laterdecayswereconsidered.Previouspartialdiscussions αππ is extracted from BaBar cross section measurements ofcosmologicalsignaturesofdecayingdarkphotonsmay ofe+e− γ∗ π+π−(γ)[11],andsimilarlyforcharged → → alsobefoundin[5,6]. kaons[12]. Inthefollowingweassumenootherlightstatesχthat Finally,thereisapossibilityofresonantproductionof are charged under U(1). Therefore, there are no decays V by virtue of the thermalbath. Such in-mediumeffects ′ V χχ¯ thatpotentiallydrainvisible modesandthereby maybecastintoaneffectivemixingangle, → ameliorate the derived limits from BBN. Using some re- κ κ = , (6) centinsightaboutthein-mediumproductionofdarkvec- T,L 1 Π /m2 tors[8,9](seealso[10])wefirstdiscusstheproductionof | − T,L V| darkvectorsinthenextsectionandexploreconstraintson with Π being the transverse (T) and longitudinal (L) T,L V-decaysintoSMinSec.3beforeconcludingwithSec.4. photonpolarizationfunctionsin the primordial, isotropic plasma.TheexpressionsforΠ cane.g.befoundin[13]; T,L the longitudinal polarization function [8] used here is, 2 Abundance prior to decay Πhere = m2/(ω2 m2)ΠRef.[13] and ω is the (dark) pho- L V − V L tonenergy.Equation(6)informsusabouttheconditionof Therelicabundanceofweaklycoupleddarkphotonsprior resonantdarkphotonproduction, totheirdecayisobtainedthroughacalculationoftheleak- age from the observable sector to the hidden sector with ReΠ (ω,T )=m2. (7) T,L r,T,L V sub-Hubblerates, Γ /H 1. This“freeze-in”process prod ≪ is dominatedby inversedecaysof V, throughthe coales- TheconditiondependsontemperatureT asΠ arepro- T,L cence of e , µ ..., and, similarly, throughhadroniccon- portional to the plasma frequency, ω (T). Most impor- ± ± P tributions;seetheillustrationinFig.1. tantly,theresonancetemperatureT (ω)asafunctionof r,T,L The Boltzmannequationfor the totalnumberdensity frequencyωisparametricallylargerthanm withamini- V ofV takestheform mumfrequencyatwhichtheresonancecanhappen, n˙ +3Hn =C. (3) 3 1/2 V V T =m 8m . (8) r,min V"2πα# ≃ V The righthandside is the collision integraland provided thatVneverreachesthermalequilibrium,itgivesthenum- Thus resonances occur at parametrically larger tempera- ber ofV states emittedperunitvolumeandunittime. In tures (by α−1/2) than mV, for which H(T) is significantly theMaxwell-BoltzmannapproximationfortheSMdistri- largerthanatT mV atwhichtheV freeze-inproduction ≃ butionfunctions,andin thelimitthatonlyelectronscoa- has its biggest contribution. Therefore, resonant contri- lesce, the integrationcan be done analytically [5, 7] (see butions to YV do not alter the picture drastically though also[6]) numericallytheymayconstituteasmuchas30%. Afterproduction,themomentumofV redhiftsquickly 3 sothatatthetimeofdecaytheenergyofV istogoodap- C Γ m2TK (m /T). (4) ≃ 2π2 V→ee¯ V 1 V proximationgivenbytherestmass, EV = mV. Thedecay deposits this energy into leptons, hadrons, and hadronic Here, Γ = κ2αm /3 is the decay width of V to elec- V ee¯ V resonances. The energy prior to decay that is stored per trons, up→to corrections(m /m )2 and K is the modified e V 1 baryonisthereforegivenby Besselfunctionofthesecondkind.Intermsofthenumber s of V particlesnormalizedto entropydensity, YV = nV/s, Ep.b. =mVYV 0 , (9) the cosmic time integral in (3) for the final “freeze-in” nb,0 DarkMatter,HadronPhysicsandFusionPhysics where n /s = 0.9 10 10 is the baryon-to-entropyra- 101 .. b,0 0 × − p(p¯) tio today. Equipped with Ep.b. as a function of mV and n(n¯) V →hadr. κ followingthe detailed calculationsof the V “freeze in” π+(π−) abundance in [7], we may now explore its consequences K+(K−) ) forBBN. ay KL c e d c 1 3 V-decaysduring BBN ni ro hEemi/mV d a Primordial nucleosynthesis predictions are affected for h ( darkphotondecayswithcosmiclifetimet & 1sorlarger. s/ e Ensuingconstraintsarethengovernedbyacombinationof ticl r lifetime and abundance, both being complementary with pa 10−1 respect to the vector mass: τ (Y ) decreases(increases) V V withlargerm . Fromthisoneexpectsconstraintsaslocal- V ized islands in those parameters where the epoch of pri- mordialnucleosynthesisexhibitsitsgreatestsensitivity. 10 mV (GeV) 3.1 Majoreffectsandtreatment Figure2. Pythiasimulationontheaveragenumberofparticles TheeffectsonBBNareunderstoodbyconsideringelectro- perV decayandontheinjectedelectromagneticenergyafterall magnetic and hadronicenergy injection separately. Prior particles havedecayed or annihilated toelectrons and photons. to decay, the V abundancerelativeto baryonsis substan- Aboutonethirdoftheenergyiscarriedawaybyneutrinos. Nar- tial, nV/nb . 108 for τV < 1s, and the decays of V in- rowhadronicresonancesareneglected. jectelectrons,muons,andmesonsinnumberslargerthan baryons. Darkphotondecayswith m 2m = 279MeVre- V π ≤ ± sultexclusivelyininjectionofelectromagneticenergy,be- frommuondecayconstituteonlyminorcorrectionstothe causeV e+e ,µ+µ aretheonlykinematicallyaccessi- − − → photon-inducedprocesseslistedabove[5]. blemodes. Muonstypicallydecaybeforeinteracting,and For vector masses abovethe di-pion threshold, m > electron-positronpairsarequicklythermalizedbyinterac- V 2m ,hadronicmodesareaccessibleinthedecayofVand tionswithbackgroundphotons.Theresultingelectromag- π± theeffectsonBBNaremoreintricate. Inthehadronicde- neticcascadewithspectrum f (E )entailsalargenumber γ γ cay ofV onlyπ , K , and K , with lifetimesτ 10 8s, ofnon-thermalphotonsthatmaythenspalllightelements. ± ± L ∼ − and(anti-)nucleonshaveachancetoundergoastrongin- Importantly,thespectrumhasarelativelysharpcut-off teractionreactionbeforedecayingbythemselves. for energiesabove the e pair-creationthreshold, E ± pair m2/(22T). Photons with E > E are being dissipate≃d Before deuterium formation at T 100keV, only e γ pair chargeexchangereactionsonnucleons,≃suchasπ + p before they interact with nuclei, and to good approxima- − → π0+n,arepossible.Theychangethen/pratioandthereby tion f (E ) = 0for E > E . Photonswith E < E , γ γ γ pair γ pair mostprominentlytheprimordial4Hevalue.Afterthedeu- however, undergoslower degradationprocesses and may terium bottleneck—once light elements have formed— interactwiththelightelementsbeforebeingthermalized. charge exchange creates “extra neutrons” on top of the EquatingE againstthephoto-destructionthresholds(in pair residual and declining neutron abundance. In addition, brackets below) yields the temperature and thereby the absorption with subsequent destruction of light elements cosmictimet ofbiggestimpactforaspallationchannel: ph suchasπ +4He T +nisnowoperative.Spallationof − → 2 104s, 7Be+γ 3He+4He (1.59MeV), 4Hemayalsohaveasecondaryconsequence: theproduc- × → tph  5 104s, D+γ n+p (2.22MeV), tion of mass-3 elements with non-thermalkinetic energy ≃ 4××106s, 4He+→γ→3He/T+n/p (20MeV), may induce reactions of the sort T + 4Hebg → 6Li+ n. The spallation rate of species N with numberdensity In the numericalanalysis, these processesaswell as sec- ondarypopulationsofπ fromkaondecays,andhyperon ± n isgivenby N producing channels from reactions of kaons on nucleons Emax and nuclei are being accounted for. Furthermore, in our Γph(T)=2nN dEγ fγ(Eγ)σγ+N X(Eγ), (10) analysis, we restrict ourselves to reactions at threshold, ZEthr → withchargedpionsandkaonsbeingthermalizedbeforere- whereσ (E )isthephoto-dissociationcrosssection actingonlightelements;suchapproximationgenerallyre- γ+N X γ for γ + N → X with threshold E . The factor of two sultsinmoreconservativeconstraints.Adetailedquantita- thr → accountsfor the two independentcascadesthatformin a tivediscussionofincompletestoppingcanbefoundin[5]. back-to-back decay of V at rest, each with a maximum Finally, baryonpairsare producedin the V-decayfor energyof E = max E ,E /2 . All spallation reac- m & 2GeV. Final state nucleons n¯ and p¯ will prefer- max pair inj V tions listed in [14] arentaken into aoccount in the numer- entially annihilate on protons with an annihilation cross ical analysis. We note in passing that neutrino injection section σ v m 2. Theinjectionofnn¯ thenresultsin h ann i ∼ −π± EPJWebofConferences agreement with [17] when using a baryon asymmetry of 1 e± π± K± ηb =6.2 10−10andaneutronlifetimeofτn =885.7s. Brem × ng 3.2 Lightelementobservations hi c bran 10−1 µ± BofBlNighsetnesleitmiveitnytiasbautntadiannecdebsyanthdetohbeisreervsatitmioantaeldinefrerorrrebnacre. e tiv Here we briefly discuss those observations that form the c p(p¯) e basisofourobtainedregionsofinterest. ff e d Themostabundantelementafterhydrogenishelium. e pt 10−2 n(n¯) ItsmassfractionYp isinferredfromextragalacticHIIre- o d gions,andvaluesintherange a 0.24 Y 0.26 (11) p ≤ ≤ 10−3 havebeenreportedovertheyears.Owingtopotentialsys- 0.1 1 tematicuncertainties[18,19]weadopt(11)asthecosmo- mV (GeV) logicallyviablerange. Amongrecentdevelopments,theprecisiondetermina- Figure3. Finalstatebranchingratiosoflong-livedmesonsand tionofD/Hfromhighredshiftquasarabsorptionsystems otherrelevantdecayproducts.BaBarmeasurementsofthee ± → standsout[20,21]. Errorbarshavereducedbyafactorof π ande K crosssectionsuptom = 1.8GeVarestiched ± ± → ± V fiveincomparisontopreviouslyavailabledeterminations. together with a Pythia simulation starting at m 2.5GeV; V ≥ Theweightedmeannowreads[21], a branching to K was neglected and the fraction of m that L V is ultimately being converted to electromagnetic energy is la- D/H=(2.53 0.04) 10 5. (12) beledBr . − em ± × D astration on dustgrainsis, however,a potentialsource of systematic uncertainty,andvaluesas high as 4 10 5 − × havealsobeenreported[22,23]. Inlightofthis,weadopt onenetp nconversionwithassociatedenergyinjection → anupperlimitof, of m +m . Annihilation on neutronswith similar cross p n sectionisalsopossibleand pp¯ injectionresultsinonenet D/H<3 10 5. (13) − n p conversion. Assuming equal cross sections, the × → relativeefficiencyis p/(n+p)andn/(n+p),respectively. as well. Finally, producingtoo little D/H yields a robust At threshold, the rate for neutroninjection can be in- limitbecausenoknownastrophysicalsourcesofthisfrag- ferred from a measurementof the e+e nn¯ cross sec- ilelightelementexist.Wethereforeeitherusethenominal − tion, σ 1nb [15]. Given a tota→l hadronic cross lower2σ-limitfrom(12)orrequire(robustly), e+e nn¯ −→ ∼ section σ 50nb at this energy, the branching e+e−→had ∼ 3He/D<1 (14) fractiontoaneutron-antineutronpairis 2%. Awayfrom ∼ the di-nucleon threshold, with multi-pion(kaon) produc- instead. Thelattervalueisderivedformsolarsystemob- tion anddecaysto hyperonsandbaryonicresonancesbe- servations[24]. ing prevalent, V-decays may be simulated using Pythia. Finally, and with much smaller abundance, the pri- The ultimate yield of π ,K , K , and nucleons prior to ± ± L mordial value of 7Li/H [25], is lower than the lithium their decay is shown in Fig. 2; dots depict the average yield from standard BBN by a factor of 3-5, 7Li/H = electromagnetic energy that is injected after all particles (5.24+0.71) 10 10[17].Weconsiderlithiumbeingincon- have decayedto electronsandphotons; e+ have been an- 0.67 × − corda−ncewithobservationsifBBNpredictionsyield nihilated on e . The rest of the decay-energy is carried − away by neutrinos. At lower energies, decay events are 10 10 <7Li/H<2.5 10 10. (15) − − eventuallydominatedbytwobodydecays. Abovethedi- × pion(di-kaon)threshold,we use BaBar measurementsof Whilenewphysicsmaybeattheheartofthelithiumprob- thee± π± ande± K± crosssectionuntilanreported lem, we caution thatastrophysicaldepletionmechanisms → → energyof √s=mV =1.8GeV. Relevantultimatebranch- may also play their part in solution to this long-standing ing ratios are shown in Fig. 3; the effects of KL are, for puzzle;see[26]forarecentreview. simplicity,neglectedinourBBNanaslysis. A moredetaileddiscussionalongwitha listofallin- 3.3 Results cludedreactionscanbefoundintheoriginalpaper[7]as wellasintheprecedingwork[5]. Numericalresultswere Our results from V-decays and their effect on BBN are obtained by usage of a Boltzmann code that is based on presentedinthem ,κparameterspaceinFig.4. Contours V Ref. [16], with significant improvements and updates as ofconstantlifetime,τ andfreeze-inabundancen /n are V V b detailed in [5]. Standard BBN yields are found to be in shownbythediagonalsolidanddottedlines,respectively. DarkMatter,HadronPhysicsandFusionPhysics 1100−−1100 4He followedby7Li+p→4He+4He.WithareducedCoulomb 2 II 3He/D barrier, 7Li is more susceptible to proton burning in the 1100−−1111 6 7LDi//HH secondstepandthedeclining7Litrendisdepictedbythe 6Li/H dashed lines in Fig. 4. Most of the extra neutrons, how- logτV ever,endupbeingcapturedbyprotonsandtheassociated 1100−−1122 5 Ia logYV D/Hconstraint(13)isgivenbytheorangeregion. III 3 κκ 1100−−1133 Ib 4 Conclusions 8 The kinetic mixing of a new U(1) gaugegroup with the Ic ′ 1100−−1144 0 Standard Model U(1) factors of hyperchargeand, below the electroweak scale, of electromagnetism is one of the fewportalstothehiddensectorwithrenormalizablecou- 1100−−1155 plings. The associated gauge boson V is often called a 11 110011 110022 110033 110044 mmVV((MMeeVV)) “darkphoton”andinthismanuscriptwehavereportedthe cosmologicallimitsfromBBN astheyhavebeenderived Figure4. Vector mass mV and kineticmixing parameter κ pa- in[7]. BBNsensitivityreachesphotonkineticmixingpa- rameter space with BBN sensitivity. Diagonal gray contours rametersofκ 10 14for1MeV m .10GeV,unchal- − V depict τ (solid) or n /n prior to decay (dotted). Shaded re- ∼ ≤ V V b lengedfromterrestrialdarkphotonsearches,see, e.g.the gionsareexcludedbyobservationsaslabeled.Thesolid(orange) worksandpresentations[27–33] closedlineisa2σconstraintfromunderproduction ofD/Hde- The presented limits are based on a thermal abun- rived from (12). Dashed black linesshow decreasing levels of dance of V and a standard cosmological history of 7Li/H, 4 10 10 and 3 10 10, fromouter cirlcestotheinner − − ones,resp×ectively. Along×thedottedline6Li/H=10 12signify- the Universe—i.e.uneventfuluntilV-decay—withreheat − ingantwoordersmagnitudeenhanced6Liyield,thatis,however, temperaturesinexcessofmV. Additionalcontributionsto notyetconstrainedbyobservations. theV-abundancesuchasfromaninitialV-condensateaf- terinflationmayonlystrengthenthederivedbounds. The latter source of primordial V-particles is particularly in- teresting in the context of smaller V-masses. Below the di-electronthreshold(notconsideredinthiswork),V has TheregionslabeledI-IIIareinconflictwithobservations anaturallylonglifetimewithV 3γbeingtheonlyde- asdetailedintheprevioussection. → cay mode. ThereforeV can evenbe a darkmatter candi- InregionsI,Vdecaystoe+e resultinelectromagnetic − date [6, 34] and ensuing constraints on the photoelectric energyinjection.RegionIa(τ 105s)ismarkedbyade- V ∼ absorption of V on atoms in dark matter detectors have struction of 7Be and D. The 7Li/H abundanceis reduced startedtoreceiveattentiononlyveryrecently[34–37]. to4 10 10 and3 10 10 fromtheoutsidetotheinside, − − × × respectively. However, cosmologically favored smaller 7Li/H abundances are challenged by3He/D < 1 (pink Acknowledgements shaded region). Using (12), lower 7Li/H values are ex- The speaker thanks S. Eidelman for the invitation to the cludedbythenominal2σlowerlimitonD/Hasdepicted workshop. JPissupportedbytheNewFrontiersprogram bythesolidclosedline. RegionIbisadditionallymarked bytheAustrianAcademyofSciences. by spallation of 7Li and 7Be from non-thermal photons. Thisresultsindirectproductionof6Li/H>10 12—values − yet too low for being observationally constrained at the References moment. Finally, Region Ic with τ 107s is marked V ∼ [1] M.PospelovandJ.Pradler,Ann.Rev.Nucl.Part.Sci. by 4He dissociationandnetcreationof3He/D rulingout 60,539(2010)[arXiv:1011.1054[hep-ph]]. thisparameterregion. Secondaryproductionof6Liisnot [2] R. Essig, J. A. Jaros, W. Wester, P. H. Adrian, efficientenoughtoyieldanadditionallimit. S. Andreas, T. Averett, O. Baker and B. Batell et al., In region II, τ < 100s and V decays before the V arXiv:1311.0029[hep-ph]. end of the D-bottleneck (T 100keV). Injection of pi- ∼ [3] B.Holdom,Phys.Lett.B166,196(1986). ons, kaons, and nucleons, results in anomalous n p ↔ [4] B.Batell, M.PospelovandA.Ritz, Phys.Rev.D79, inter-conversion. The consequence is an elevated n/p- ratioandthereforeenhancedD and4Heyields. Thelow- 115008(2009)[arXiv:0903.0363[hep-ph]]. lifetime/high-abundanceregionII is correspondinglydis- [5] M.PospelovandJ.Pradler,Phys.Rev.D82,103514 favoredbyY 0.26andD/H 3 10 5. (2010)[arXiv:1006.4172[hep-ph]]. p − ≤ ≤ × Finally,regionIIIismarkedbytheproductionof“ex- [6] J. Redondo and M. Postma, JCAP 0902, 005 (2009) tra neutrons”att 103s fromV nn¯ and fromcharge [arXiv:0811.0326[hep-ph]]. ∼ → exchange of π on protons, π p nπ0 or π p nγ. [7] A.Fradette,M.Pospelov,J.PradlerandA.Ritz,Phys. − − − → → Inaddition,hyperonproductionby“s-quark”exchangeof Rev.D90,035022(2014)[arXiv:1407.0993[hep-ph]]. K on protonsmay also result in extra neutrons. With it [8] H.An,M.PospelovandJ.Pradler,Phys.Lett.B725, − comesapaththatmaydepletelithium,7Be+n 7Li+p, 190(2013)[arXiv:1302.3884[hep-ph]]. → EPJWebofConferences [9] H. An, M. Pospelov and J. Pradler, Phys. Rev. Lett. [24] Prantzos, N., Vangioni-Flam, E. and Cassé, M. 111,041302(2013)[arXiv:1304.3461[hep-ph]]. 1993, Origin and evolution of the elements. Proceed- [10] J. Redondo and G. Raffelt, JCAP 1308, 034 (2013) ings., Cambridge University Press, Cambridge (UK), [arXiv:1305.2920[hep-ph]]. 1993 [11] J.P.Leesetal.[BaBarCollaboration],Phys.Rev.D [25] F. Spite and M. Spite, Astron. Astrophys. 115, 357 86,032013(2012)[arXiv:1205.2228[hep-ex]]. (1982). [12] J.P.Leesetal.[BaBarCollaboration],Phys.Rev.D [26] B.D.Fields,Ann.Rev.Nucl.Part.Sci.61,47(2011) 88,no.3,032013(2013)[arXiv:1306.3600[hep-ex]]. [arXiv:1203.3551[astro-ph.CO]]. [13] E. Braaten and D. Segel, Phys. Rev. D 48, 1478 [27] J.Beacham,EPJWebConf.73,07011(2014). (1993)[hep-ph/9302213]. [28] A.Celentano[theHPSCollaboration],J.Phys.Conf. [14] R.H.Cyburt,J.R.Ellis,B.D.FieldsandK.A.Olive, Ser.556,no.1,012064(2014). Phys.Rev.D67,103521(2003)[astro-ph/0211258]. [29] M. Battaglieri et al. [BDX Collaboration], [15] A. Antonelli, R. Baldini, P. Benasi, M. Bertani, arXiv:1406.3028[physics.ins-det]. M.E.Biagini,V.Bidoli,C.BiniandT.Bressanietal., [30] F. Curciarello [KLOE/KLOE-2 Collaboration], EPJ Nucl.Phys.B517,3(1998). WebConf.72,00004(2014). [16] L.Kawano,FERMILAB-PUB-92-004-A. [31] H. Merkel [A1 Collaboration], EPJ Web Conf. 81, [17] R. H. Cyburt, B. D. Fields and K. A. Olive, JCAP 01020(2014). 0811,012(2008)[arXiv:0808.2818[astro-ph]]. [32] M. Raggi and V. Kozhuharov, Adv. High En- [18] Y.I.IzotovandT.X.Thuan,Astrophys.J.710,L67 ergy Phys. 2014, 959802 (2014) [arXiv:1403.3041 (2010)[arXiv:1001.4440[astro-ph.CO]]. [physics.ins-det]]. [19] E. Aver, K. A. Olive, R. L. Porter and E. D. Skill- [33] R.Essig,P.Schuster,N.ToroandB.Wojtsekhowski, man,JCAP1311,017(2013)[arXiv:1309.0047[astro- JHEP1102,009(2011)[arXiv:1001.2557[hep-ph]]. ph.CO]]. [34] M.Pospelov,A.RitzandM.B.Voloshin,Phys.Rev. [20] M. Pettini and R. Cooke, Mon. Not. Roy. As- D78,115012(2008)[arXiv:0807.3279[hep-ph]]. tron. Soc. 425, 2477 (2012) [arXiv:1205.3785 [astro- [35] K.Arisaka,P.Beltrame,C.Ghag,J.Kaidi,K.Lung, ph.CO]]. A. Lyashenko, R. D. Peccei and P. Smith et al., As- [21] R.Cooke,M.Pettini,R.A.Jorgenson,M.T.Murphy tropart. Phys. 44, 59 (2013) [arXiv:1209.3810 [astro- andC.C.Steidel,arXiv:1308.3240[astro-ph.CO]. ph.CO]]. [22] S. Burles and D. Tytler, Astrophys. J. 507, 732 [36] K. Abe et al. [XMASS Collaboration], Phys. Rev. (1998)[astro-ph/9712109]. Lett. 113, 121301 (2014) [arXiv:1406.0502 [astro- [23] D. Kirkman, D. Tytler, N. Suzuki, J. M. O’Meara ph.CO]]. and D. Lubin, Astrophys. J. Suppl. 149, 1 (2003) [37] H. An, M. Pospelov, J. Pradler and A. Ritz, [astro-ph/0302006]. arXiv:1412.8378[hep-ph].