Probing new physics with underground accelerators and radioactive sources

(1)

Citation for this paper:

Izaguirre, E., Krnjaic, G. & Pospelov, M. (2015). Probing new physics with

underground accelerators and radioactive sources. Physics Letters B, 740, 61-65.

http://dx.doi.org/10.1016/j.physletb.2014.11.037

UVicSPACE: Research & Learning Repository

_____________________________________________________________

Faculty of Science

Faculty Publications

_____________________________________________________________

Probing new physics with underground accelerators and radioactive sources

Eder Izaguirre, Gordan Krnjaic, Maxim Pospelov

2015

©2014 The Authors. Published by Elsevier B.V. This is an open access article under

the CC BY license (http://creativecommons.org/licenses/by/3.0/).

This article was originally published at:

(2)

Contents lists available atScienceDirect

Physics

Letters

B

www.elsevier.com/locate/physletb

Probing

new

physics

with

underground

accelerators

and

radioactive

sources

Eder Izaguirre

a

,

Gordan Krnjaic

a

,

∗

,

Maxim Pospelov

a

,

b

a_Perimeter_Institute_for_Theoretical_Physics,_Waterloo,_Ontario,_Canada

b_Department_of_Physics_and_Astronomy,_University_of_Victoria,_Victoria,_British_Columbia,_Canada

a

r

t

i

c

l

e

i

n

f

o

a

b

s

t

r

a

c

t

Articlehistory:

Received3September2014

Receivedinrevisedform18November2014 Accepted18November2014

Availableonline21November2014 Editor:M.Trodden

New light, weakly coupledparticlescan be eﬃciently produced atexisting and future high-intensity accelerators and radioactivesources indeepunderground laboratories.Onceproduced, theseparticles canscatterordecayinlargeneutrinodetectors(e.g. Super-KandBorexino)housedinthesamefacilities. Wediscusstheproductionofweaklycoupledscalarsφ vianuclearde-excitationofanexcitedelement intothegroundstateintwoviableconcretereactions:thedecayofthe0+excitedstateof16Opopulated viaa(p,

α

) reactiononﬂuorineandfromradioactive144_Ce_decay_where_the_scalar_is_produced_in_the

de-excitationof144_Nd∗_,_which_occurs_along_the_decay_chain._Subsequent_scattering_on_electrons,_e_(φ,

_γ

₎_e,

yields amono-energetic signal that is observable inneutrino detectors. We show that thisproposed experimentalsetup cancovernewterritoryformasses250 keV≤mφ≤2meandcouplingstoprotonsand

electrons,10−11_≤_g

egp≤10−7.Thisparameterspaceismotivatedbyexplanationsofthe“protoncharge

radius puzzle”,thus thisstrategyadds a viablenew physics component to the neutrinoand nuclear astrophysicsprogramsatundergroundfacilities.

1. Introduction

Inrecentyears,therehasemergedauniversalappreciationfor new light, weakly-coupled degrees of freedom as generic possi-bilities for New Physics (NP) beyond Standard Model (SM). Con-siderableeffort in“intensityfrontier”experimentsisnowdevoted to NPsearches [1]. In this paperwe argue that there is a pow-erfulnew possibilityforprobing thesestates by combininglarge underground neutrino-detectors with either high luminosity un-dergroundacceleratorsorradioactivesources.

Underground laboratories, typically located a few km under-ground, are shielded from most environmental backgrounds and areidealvenuesforstudyingrareprocessessuch aslow-rate nu-clearreactions andsolar neutrinos. Thus far, thesephysics goals have been achievedwith very differentinstruments: nuclear re-actionsrelevantforastrophysicsinvolvelow-energy,high-intensity protonorionbeamscollidingwithﬁxedtargets(suchastheLUNA experimentatGranSasso),whilesolarneutrinosaredetectedwith largevolume ultra-cleanliquid scintillatoror waterCerenkov de-tectors(SNO,SNO

+

,Borexino,Super-K,etc.).

*

Correspondingauthor.

E-mailaddress:gkrnjaic@perimeterinstitute.ca(G. Krnjaic).

Inthispaperweoutlineanovelexperimentalstrategyinwhich light,“invisible”states

φ

areproducedinundergroundaccelerators orradioactivematerialswithO

(

MeV

)

energyrelease,andobserved innearby neutrinodetectors inthesame facilities asdepictedin Fig. 1:

X∗

→

X

+ φ,

production at “LUNA” or “SOX” (1)

e

+ φ →

e

+

γ

,

detection at “Borexino”

.

(2) Here X∗ is anexcited state of element X ,accessed via anuclear reactioninitiatedby anundergroundaccelerator(“LUNA”)orby a radioactive material (“SOX”).1 Inthe “LUNA”-type setup a proton beam collides against a ﬁxed target, emitting a new light parti-cle that travelsunimpeded through the rock and scattersinside a “Borexino”-typedetector.Alternatively, inthe “SOX”production scenario,designedtostudyneutrinooscillationsatshortbaselines, aradioactivematerialplacednearaneutrinodetectorgivesriseto the reactionin Eq.(1)asan intermediate step ofthe radioactive material’sdecaychain.

We study one particularly well-motivated NP scenario witha

MeV scalarparticle,veryweakly O

(

10−4

₎

_coupled_to_nucleons

1 _Our_idea_is_very_generic,_not_speciﬁc_to_any_single_experiment_or_location,_which

iswhyquotationmarksareused. http://dx.doi.org/10.1016/j.physletb.2014.11.037

(3)

62 E. Izaguirre et al. / Physics Letters B 740 (2015) 61–65

Fig. 1. Schematicﬁgureofφproductionina“LUNA”-typeundergroundaccelerator viap+19_F_{→ (}16_O∗_→16_O_{+ φ)}₊_α_or_a_{“SOX”-type}_radioactive_source_via144_Ce_→

144_Pr

(ν¯e)→Nd∗→Nd+ φ.Subsequentdetectionat“Borexino”proceedsviaφe→ eγscalarconversion.

andelectrons.Thisrangeofmassesandcouplingsisnotexcluded by astrophysical or laboratory bounds, and is motivated by the persistentprotoncharge-radiusanomaly.Twoconcrete,viable pos-sibilitiesforproducinglightscalarsareconsidered:

•

For the LUNA-type setup, we show that such light particles can be eﬃciently produced by populating the ﬁrst excited 6.05 MeV0+stateof16Oin

(

p

,

α

)

reactionsonﬂuorine.

•

FortheSOX-type setupwe ﬁndsimilarly powerfulsensitivity from the 144_Ce–144_Pr

₍

_ν

_¯

e

)

radioactive source, which can pro-duceascalarwith2.19or1.49 MeVenergiesfromthe144Nd∗ de-excitationthatoccursalongthedecaychain.

The subsequent detection of a mono-energetic release in a Borexino-type detector with6.05, 2.19, or 1.49 MeV will be free from substantial environmental backgrounds. The strategy pro-posedinthisLetter iscapableofadvancingthesensitivitytosuch statesby manyordersofmagnitude, completelycovering the pa-rameterspacerelevantfortherp puzzle.

2. Scalarparticlesbelow1MeV

Newparticles in theMeV andsub-MeVmass rangeare moti-vatedbytherecent7

σ

discrepancybetweenthestandard determi-nationsoftheprotonchargeradius,rp,basedone–p interactions [2], and the recent, most precise determination of rp from the Lamb shift in muonic Hydrogen [3,4]. One possible explanation forthisanomalyisanewforcebetweentheelectron(muon)and proton[5–7]mediatedbya

∼

100 fmrangeforce(scalar- or vector-mediated)that shiftsthebinding energies ofHydrogenicsystems andskewsthedeterminationofrp.Motivatedbythisanomaly,we considerasimplemodelwithonelightscalar

φ

thatinteractswith protonsandleptons,

Lφ

=

1 2

(∂

μ

φ)

2

₋

1 2m 2 φ

φ

2

+ (

gpp p

¯

+

geee

¯

+

gμ

μμ

¯

)φ,

(3) anddeﬁne

2

_{≡ (}

_g

egp

)/

e2. We assume mass-weighted couplings to leptons, ge

∝ (

me

/

mμ

)

gμ, and no couplings to neutrons. The apparentcorrectionstothecharge radiusoftheprotoninregular andmuonichydrogenare[5–7]

r2_p

_eH

= −

6

2 m2_φ

;

r 2 p

μH

= −

6

2

₍

_g μ

/

ge

)

m2_φ f

(

amφ

)

(4)

where a

≡ (

α

mμmp

)

−1

(

mμ

+

mp

)

is the

μ

H Bohr radius and

f

(

x

)

=

x4

₍

₁

₊

_x

₎

−4_. _Equating

_r2

p

|

μH

−

r2p

|

eH tothe current

dis-Fig. 2. Sensitivity projections for various experimentalsetups interms of 2₌

gpge/e2 _and _m

φ, which parametrize the NP explanation ofthe rp anomalyin

Eq. (4); the blue band is the parameter space that resolves the puzzle. The “LUNA/Borexino”curve assumesa400 keVprotonbeamwith1025 _POT_incident

on a C3F8 targettoinduce p+19F→ (16O∗→16O+ φ)+αreactions 100 m

awayfromBorexinoandyield10signalevents(>3σ)abovebackgrounds[9].The Borexino3 MeVandSuper-K 3 MeVlinesassumethesamesetupwitha3 MeV p-accelerator10 mawayfromeachdetector.TheSuper-K projectionshows100 sig-nalevents(>3σ)abovebackgroundsat 6.05 MeV[10].TheSOXlinesassumea radioactive144_Ce–144_Pr_source_{7.15 m}_away_from_Borexino_with₅₀_and₁₆₅_events

(>3σ)abovebackgroundsfor2.19and1.49 MeVlinesrespectively.Shadedingray areconstraintsfromsolarproduction[9],LSNDelectron–neutrinoscattering[11], andstellarcooling[12],forwhichweassumege= (me/mp)gp.(Forinterpretation ofthereferencestocolorinthisﬁgurelegend,thereaderisreferredtotheweb versionofthisarticle.)

crepancy of

−

0

.

063

±

0

.

009 fm2 [4], one obtains a relation be-tween mφ and

.Thus, formφ

=

0

.

5 MeV,the anomaly suggests

2

₁

_.

₃

_×

₁₀−8_. _For_m

φ

>

2me,the

φ

→

e+e− process ishighly constrained by searches for light Higgs bosons [1], so we con-siderthemφ

<

2me region,whichisrelativelyunconstrained.Since

ge

gp, the

φ

−

e coupling is suppressed relative to that of a massivephoton-likeparticle,soprecisionmeasurementsof

α

and

(

g

−

2

)

e donotconstrainthisscenario.

We wouldliketo emphasizethat currently,thereare nogood modelsofnewphysicscapableofﬁtting

rp discrepancyandnot sufferingfromadditionalﬁne-tuningissues,especiallyifonetries to ﬁndasatisfactory descriptionforsuch modelsatorabove the electroweak scale.Thus, models withvery light vector mediators haveto beconstructedto avoidcouplings withneutrinos [7],but thesecannot avoidthetuningofthemuon g

−

2 andtheatomic parity violation constraints [8]. In that sense, a sub-MeV scalar may be presentingthe leastamount oftuning [5].Still, the van-ishing couplingtoneutrons(constrainedinneutronscattering ex-periments tobe below10−4 _level),_is _challenging_to _achieve:_the onlypossibilityathandseemstobeaﬁne-tuningof

φ

uu and

¯

φ ¯

dd

operatorsatthequarklevel.Thisinturn,wouldcorrespondto tun-ing ofdimensionﬁveoperators, when

φ

qq are

¯

generalizedtothe full SM gauge invariance. To summarizethis discussion, we take model(3)asaphenomenologicalmodel,capableofresolving

rp discrepancy,butnotfreeofﬁne-tuningissues.

The astrophysical and ﬁxed-target constraints depend on the cross section for e

φ

→

e

γ

conversion, which formφ

me with astationaryelectrontargetis

d

σ

dE

=

π

(

ge

/

e

)

2

α

2

(

E

−

me

)

meQ4

(

Q

−

E

+

me

)

2

E

Q2

−

E Q

−

2meQ

−

2m2e

+

me

3Q2

+

3Q me

+

2m2e

,

(5)

where E isthe electron recoilenergyand Q is the

φ

energy. At

(4)

σ

eφ

π

(

ge

/

e

)

2

α

2 2meQ

=

13 mb

×

5 MeV Q

×

ge e

2

,

(6)

which determines the in-medium

φ

-absorption probability. Ab-sorption competes with the

φ

→

γ γ

decay, proceeding through loopsoffermions f withthewidthgivenbyastandardformula,

Γ (φ

→

γ γ

)

=

α

2_m3 φ 512

π

3

f gf mf NcQ2fA1/2

(

τ

f

)

2

,

(7) whereQf isthefermioncharge,

τ

f

≡

m2φ

/

4m2f,and

A1/2

(

τ

)

=

2

τ

−2

τ

+ (

τ

−

1

)

arcsin

√

τ

.

(8)

An approximate proportionality to particle masses ensures that couplingstoneutrinosarenegligible.

Processes(5),(7)deﬁnethegrossfeaturesof

φ

-phenomenology incosmologicalandastrophysicalsettings.Theensuingconstraints aresummarizedasfollows:

•

Energy loss in stars via e

γ

→

e

φ

(red giants, white dwarfs, etc.)isexponentiallysuppressedformφ

>

Tstar.Wecalculatea boundofmφ

250 keV,fortheﬁducialrangeofcouplings.

•

The decay of

φ

in theearly Universe at T

∼

mφ resultsin a

negativeshiftofthe“effectivenumberofneutrinos.”Formφ

>

250 keV the shiftis moderate, Neff

∼ −

0

.

5 [13], and can be easily compensated by the positive contributions fromother lightparticles(e.g.sterileneutrinos).

•

SN physics: Low masses and sizable couplings, ge,p

∼

10−4, ensures the

φ

are trappedduring theexplosions,andneither takeenergyfromtheexplosivezonesnordegradetheneutrino energiesonaccountofgν

=

0.

•

Emission of

φ

in solar nuclear reactions can be constrained using the Borexino search forsolar axions [9],and disfavors some fraction of the parameter space with

2 _in _between 10−12and10−10,asshowninthiswork.

Inaddition toastrophysical constraints,boundson

from di-rectsearchesofverylightscalarstypicallyprobe

2

₁₀−7_._When combined,existingconstraintsleaveanunexploredpartofthe pa-rameterspaceforthescalarmodel,250 keV

mφ

<

2me,10−10

2

₁₀−7_,_and_the

_r

p-motivatedrangefallsinthemiddleofthis allowedterritory.TheexistingconstraintsaresummarizedinFig. 2.

3. Productionofscalarsinnuclearreactions

Searches of light scalar particles in nuclear reactions, such as3_H

₍

_p

_,

_γ

₎

4_{He and} 19_F

₍

_p

_,

_α

₎

16_O∗ _have_been _successfully imple-mented[14,15]onthesurface,wherethemainbackgroundcomes fromcosmic events.Forsub-MeVmassesof

φ

,thelatterreaction isespeciallyadvantageousas

φ

isproducedinthede-excitationof the0+state:

16_O∗

₍

₆

_.

₀₅

₎

_→

16_O

_{+ φ,}

₍₉₎

withenergyrelease Q

=

6

.

05 MeV.IntheSM,thesingle-

γ

decay ofthisstate isnot possibledueto angularmomentum conserva-tion, and the main de-excitation process is 16O∗

→

16O

+

e+e−

withthe long lifetime 96

±

7 ps [16]; thus, the relative branch-ing to new physics can be greatly enhanced. Following [17] for

mφ

Q ,theNPbranchingratio

Γ

φ

/Γ

e+e− is

B

rφ

=

8

π

(

gp

/

e

)

2Q5

α

b

(

s

)(

Q

−

2me

)

3

(

Q

+

2me

)

2

4

×

103

gp e

2

,

(10)

wheres

= (

Q

−

2me

)/(

Q

+

2me

)

andb

(

s

)

≈

0

.

92 isdeﬁnedin[17]. The excited state 16O∗ can be eﬃcientlyproduced in

∼

100 keV– MeV p accelerators.

To estimate the

φ

yield from p

+

19_F

_→

16_O∗

₍

₆

_.

₀₅

₎

₊

_α

_, _we

modelthecrosssectionbelow3 MeVusing[18,19]andextrapolate to the Coulomb-suppressed region. Speciﬁcally, we take

σ

(

E

)

σ

0f

(

E

)

,with

σ

0

=

18 mbn andmodeltheCoulombrepulsionwith

f

(

E

<

E0

)

=

E0 E exp

Eg

/

E0

−

Eg

/

E

,

(11)

intheE

<

E0

≡

1

.

5 MeV range.HereEg

=

2

(

π α

ZF

)

2

μ

=

45

.

5 MeV istheGamowenergyand

μ

istheproton–ﬂuorinereducedmass,

E is the c.o.m. energy, and normalization ensures continuity at

f

(

E0

)

=

1,whererepulsioncanbeneglected.

ThesignalyieldforaprotonbeamofenergyEp (i.e. the prob-abilityto producea quantum of

φ

per each injectedproton) and targetmaterialofﬂuorine number-densitynF is

Nφ

(

Ep

)

=

B

rφ

×

nF Ep

0 dE

σ

p

(

E

)

|

dE

/

dx

|

.

(12)

|

dE

/

dx

|

depends on the material that includes ﬂuorine, and is readily available in[20]. For example,for the C3F8 material, the probabilityofproducingone

φ

perinjectedprotonisNφ(3 MeV

)

∼

3

×

10−2

(

gp

/

e

)

2.

Theangulardistributionofemerging

φ

isfullyisotropicas nu-clearrecoilvelocitiesarenegligible,andtheﬂuxatthepositionof thedetectorisgivenby

Φφ

=

Nφ(Ep

)

× (

dNp

/

dt

)/

4

π

L2.Insidethe detector,theemitted

φ

scatteroffelectronsthroughe

φ

→

e

γ

with crosssectionsgivenby(5).Thus,theonlyremainingfree parame-ters(distanceL,numberofacceleratedprotonsperseconddNp

/

dt, theirenergyEp aswellasthenumberofelectronsinthedetector volume)arelocation,source,anddetector-speciﬁc.2

4. Productionoflightstatesinradioactivedecays

An alternative realistic mechanism forproducing light weakly coupled particles is using the high-intensity radiative sources placed near a neutrino detector. In particular, we focus on the speciﬁc radioactive source 144_Ce–144_Pr

₍

_ν

_¯

e

)

motivatedby the SOX proposal by the Borexino Collaboration. The production of the scalar inthis reaction proceedsvia 144Ce

→ β ¯

ν

+

144_{Pr followed} by144Pr

→ β ¯

ν

+ (

144_Nd∗

_→

144_Nd

_{+ φ)}

_._Once_produced,_the_scalar canbedetectedataneutrinodetector.

5. Possibleacceleratorrealizations

Alltheingredients forasuccessfulrealizationofouridea cur-rentlyexistattheundergroundLaboratoriNazionalidelGranSasso (LNGS) inItaly, home ofboth theLUNAacceleratorandBorexino detector.Inaddition,thereareseveralotherfacilitiesofinterest in-cludingSNOLABinCanadaandtheKamiokaObservatoryinJapan. Both SNO

+

and Super-Kdetectors inthese laboratoriescould be sensitive to new sub-MeV states if a proton accelerator were to beplaced intheir vicinity.Furthermore,theSanfordUnderground ResearchFacility(SURF)hascurrentplanstohosttheDualIon Ac-celerators for Nuclear Astrophysics (DIANA), which are expected

2 _Depending_on_the_UV_model_assumptions_that_yield_the_effective_theory_in₍₃₎_,

theφ→γ γdecaymaydominatethesignalyieldinsidethedetector.However,this ishighlymodel-dependent,soweconservativelyrestrictourfocustothe model-independentscatteringsignalthatdependsonlyonthe couplings ge,p intheIR effectivetheory.

(5)

64 E. Izaguirre et al. / Physics Letters B 740 (2015) 61–65

to deliver 10–100 mA 3 MeV proton beams. SURF is also home totheLargeUndergroundXenon(LUX) experiment,whichdespite its smallervolumecompared toBorexino andSuper-Kamiokande, couldalsobesensitivetonewsub-MeVstates.

TheLUNAaccelerator[21]candelivermAcurrentsofMeVscale proton energies [22]. Our main results and the plot with sensi-tivityprojectionsassume a target whichis notcurrentlyused by the LUNA experiment (e.g. C3F8), but can easily be installed. In Fig. 2weshowarealistic scenarioassumingtheexisting400 keV accelerator L

=

100 m away inthe canonical LUNAscenario. We also show projections for an upgraded 3 MeV beam [23] 10 m awayfromtheBorexinodetectorintheGranSassoservicetunnel. Forall our acceleratorprojections we optimistically assume 1025 protons-on-target (POT), achievable witha 50 mA beamrunning foroneyear.Veryimportantly,at6.05 MeVenergyBorexinois al-mostbackground-freeandhasgoodenergyresolution,sothateven ahandful ofevents

(

∼

10

)

wouldshow asigniﬁcantexcessinthe correspondingenergybin,andconstituteadiscovery.

Onepracticallimitationofthisproposalcouldbearequirement ofnotincreasingtheneutronbackgroundinLNGS.Inourexample, the main source of neutrons is

α

nuclei produced in each reac-tionstep,whichyieldneutronsinsecondarycollisionswithtarget nuclei. Using[24],we estimate theneutron yield from19F

(

α

,

n

)

23_Na _in _our _setup _to _be

_∼

_O

₍

_{few Hz}

₎

_. _Such _low _rates _are irrele-vantatLNGS, whichcanaccommodate 103 _Hz,_but_might_matter ifalternateproductionmethodsareemployed,thusrequiringextra shielding.

The Super-Kamiokande (Super-K) detector [25] in Kamioka, Japan,containsa50,000-tonwaterˇCerenkovdetector.InFig. 2we showtheexpected

sensitivityofahigh-intensity3 MeV proton source,assuminga C3F8 target 10 mawayfromthedetector. De-spiteapenalty duetoarelatively highthresholdfortheelectron energyinSuper-K, onecan see an incrediblystrong potential for thereachtonewphysics.

6. Possibleradioactivesourcerealizations

Forscalar productionvia radioactivedecays, one possibilityis phase B of the SOX proposalby the Borexino Collaboration [26], which intendsto deploy a

∼

2 PBq sourceof 144Ce–144Pr7.15 m fromtheBorexinocenter.Roughly2%of144Ce decaysare accom-paniedby the

γ

-radiation fromthedecay ofthemetastable Nd∗ daughternucleidescribedabove.The1.49and2.19 MeVtransition energies are well above the Borexino threshold, so this method covers the full mass range of interest, generating

∼

1013

(

gp

/

e

)

2

φ

-particles per second.Giventheplanned exposures[26],we es-timate the Borexino reach in this case, and add corresponding sensitivitylinesonFig. 2.

7. Existingconstraints

Whilemanyofthepastbeam-dumpexperimentscanbe sensi-tiveto sub-MeVparticles,weconcentrate ontheone thatisable toconstraintheproductof gpge,namelytheLSNDexperimentat LosAlamos.Itsmeasurementoftheelasticelectron–neutrinocross section[11]isalsosensitivetolightscalarsthatinducee

γ

events dueto scattering on electrons. This analysishas previously been usedto constrain newvector particles producedin

π

0 _decays_to dark sector states [27,28]. In our scenario, a scalar

φ

cannot be producedfrompseudoscalar

π

0_decays._Instead,_the_dominant pro-cessis

π

− absorption via

π

−p

→

n

φ

.The analogousSM process

π

−p

→

n

γ

hasbranchingratio

∼

35%[29],soweapproximatethe

φ

branchingas

∼

2

_×

_35%._Taking_the

_π

−_production_rate_at_LSND

toberoughly10%ofthe

π

+productionimplies

∼

1022

_π

−_for_the

exposure in [11].Assuming isotropic

φ

emission andthe scatter-ing crosssectioninEq.(5)with Q

→

mp

+

mπ−

−

mn

mπ ,and implementingthecutsfromthisanalysis,weobtainaroughlyﬂat bound

2

₁₀−8 _for_m

φ

<

MeV as showninFig. 2.This

sensitiv-ityexceedseven theboundsfrom

(

g

−

2

)

e from[30],whichonly imply

2

₁₀−7 _over _this_mass _range, _assuming _mass _weighted couplings gp

= (

mp

/

me

)

ge;forge

=

gp,theboundsfrom

(

g

−

2

)

e arecomparabletothosesetbyLSND.

In the 100 keV–MeV mass window

φ

’s cannot be produced thermallyinthesolarinterior,butcanbe producedinnuclear re-actions. A particularly relevantprocess is p

+

d

→

3He

+ φ

(that accompanies thed

(

p

,

γ

)

3He reaction occurringforevery individ-ual pp eventofenergygeneration). If

φ

is suﬃcientlylonglived, andnotabsorbedinthesolarinterior,it willreachtheEarthand deposit5.5 MeVofenergyinBorexino.Theabsenceofsuchevents [9]setsanimportantconstraintonourmodel.

The solarﬂux of5.5 MeV

φ

particles at Borexino is approxi-matedusingthe pp-neutrinoﬂuxvia

Φ

φ,solar

2PescPsurv

Φ

ppν

,

(13)

where

Φ

ppν

=

6

.

0

×

1010cm−2s−1 [9].The probability of

escap-ingthesunis Pesc

=

exp

(

−

R _{dr n}

σ

eφ),theprobabilitythatthe scalar doesnot decaybetweenthe Sun andthe Earth is Psurv

=

exp

(

−

/

φ

)

, where

φ

=

Q c

/

mφ

Γ (φ

→

γ γ

)

is theboosted

de-cay length, and

is the Earth–Sun distance. The Borexino rate is

˙

Nφe

= Φ

φ,solarnB

σ

eφVB (14)

where n_,B are mean-solar andBorexino e− densities, VB is the Borexinovolume,andthecrosssectionoffelectronsisgivenin(6). ThecurrentlimitsonthisprocessareO

(

5

)

events[9]andthe con-straint is depicted by the oval region in Fig. 2. For

2

₁₀−10_, scattering off electrons prevents

φ

from leavingthe Sun andfor

2

_×

₁₀−12_the_production_and_scattering_are_insuﬃcient_to_yield anappreciablesignalatBorexino.

The constraints from thermal energy loss in red giants and white dwarfs follow the standard considerations. Calculating the thermal energy loss

∝

g2

eexp

(

−

mφ/Tstar

)

and reinterpreting the axionconstraintsfrom[12],weexcludethemφ

250 keV

param-eterspaceforall

ofinterest.

To conclude, inthispaper we haveproposed a novel strategy to hunt forsub-MeV particles produced in underground acceler-ators andradioactivesources located 10–100 m away from large undergroundneutrinodetectors.Thisexperimentalprogramoffers unprecedented sensitivity to a variety of NP scenarios including thosethatresolvetherppuzzle.

Acknowledgements

We thank Drs.A.Arvanitaki, J.Beacom, andI.Yavinfor help-fulconversations.ThePerimeterInstituteforTheoreticalPhysicsis supportedby theGovernmentofCanadathroughIndustry Canada andbytheProvinceofOntario.

References

[1]R.Essig,J.A.Jaros,W.Wester,P.H.Adrian,S.Andreas,etal.,arXiv:1311.0029, 2013.

[2]P.J.Mohr, B.N. Taylor,D.B. Newell, Rev.Mod. Phys. 84 (2012) 1527, arXiv: 1203.5425.

[3]R.Pohl,A.Antognini,F.Nez,F.D.Amaro,F.Biraben,etal.,Nature466(2010) 213.

[4]A.Antognini,F.Nez,K.Schuhmann,F.D.Amaro,FrancoisBiraben,etal.,Science 339(2013)417.

(6)

[5]D.Tucker-Smith,I.Yavin,Phys.Rev.D83(2011)101702,arXiv:1011.4922. [6]V.Barger,C.-W.Chiang,W.-Y.Keung,D.Marfatia,Phys.Rev.Lett.106(2011)

153001,arXiv:1011.3519.

[7]B.Batell,D.McKeen,M.Pospelov,Phys.Rev.Lett.107(2011)011803,arXiv: 1103.0721.

[8]S.G.Karshenboim,D.McKeen,M.Pospelov,arXiv:1401.6154,2014.

[9]G.Bellini,etal.,BorexinoCollaboration,Phys.Rev.D85(2012)092003,arXiv: 1203.6258.

[10]J. Hosaka, et al., Super-Kamiokande Collaboration, Phys. Rev. D 73 (2006) 112001,arXiv:hep-ex/0508053.

[11]L.Auerbach,etal.,LSNDCollaboration,Phys.Rev.D63(2001)112001,arXiv: hep-ex/0101039.

[12]G.Raffelt,A.Weiss,Phys.Rev.D51(1995)1495,arXiv:hep-ph/9410205. [13]K.M.Nollett,G.Steigman,Phys.Rev.D89(2014)083508,arXiv:1312.5725. [14]S.Freedman,J.Napolitano,J.Camp,M.Kroupa,Phys.Rev.Lett.52(1984)240. [15]D.Kohler,B.Watson,J.Becker,Phys.Rev.Lett.33(1974)1628.

[16]D.Tilley,H.Weller,C.Cheves,Nucl.Phys.A564(1993)1. [17]K.A.Snover,A.E.Hurd,Phys.Rev.C67(2003)055801. [18]P.Cuzzocrea,etal.,Lett.NuovoCimento2 (28)(1980)515.

[19]C.Angulo,M.Arnould,M.Rayet,P.Descouvemont,D.Baye,etal.,Nucl.Phys.A 656(1999)3.

[20] M.Z.M.J.Berger,J.S.Coursey,J.Chang,http://physics.nist.gov/PhysRefData/Star/ Text/PSTAR.html,2009.

[21]C.Broggini,D.Bemmerer,A.Guglielmetti,R.Menegazzo,Annu.Rev.Nucl.Part. Sci.60(2010)53,arXiv:1010.4165.

[22] D.Montanari,BorexinoCollaboration,2009.

[23]A.Guglielmetti,LUNACollaboration,EPJWebConf.66(2014)07007. [24]P.R.Wrean,R.W.Kavanagh,Phys.Rev.C62(2000)055805.

[25]Y. Fukuda, et al., Super-KamiokandeCollaboration, Nucl. Instrum. Methods Phys.Res.,Sect.A,Accel.Spectrom.Detect.Assoc.Equip.501(2003)418. [26]G.Bellini,etal.,BorexinoCollaboration,J.HighEnergyPhys.1308(2013)038,

arXiv:1304.7721.

[27]B.Batell,M.Pospelov,A.Ritz,Phys.Rev.D80(2009)095024,arXiv:0906.5614. [28]P. deNiverville,M. Pospelov,A.Ritz,Phys. Rev.D84(2011) 075020,arXiv:

1107.4580.

[29]N.Samios,Phys.Rev.Lett.4(1960)470.

[30]G. Giudice, P. Paradisi,M. Passera, J. High Energy Phys. 1211 (2012) 113, arXiv:1208.6583.