Theory of elastic neutrino-electron scattering Oleksandr Tomalak 1,2,3,* and Richard J. Hill 1,2,† 1 Department of Physics and Astronomy, University of Kentucky, Lexington, Kentucky 40506, USA 2 Theoretical Physics Department, Fermi National Accelerator Laboratory, Batavia, Illinois 60510, USA 3 Institut für Kernphysik and PRISMA Cluster of Excellence, Johannes Gutenberg Universität, D-55099 Mainz, Germany (Received 19 July 2019; accepted 15 January 2020; published 21 February 2020) Theoretical predictions for elastic neutrino-electron scattering have no hadronic or nuclear uncertainties at leading order making this process an important tool for normalizing neutrino flux. However, the process is subject to large radiative corrections that differ according to experimental conditions. In this paper, we collect new and existing results for total and differential cross sections accompanied by radiation of one photon, νe → νeðγÞ. We perform calculations within the Fermi effective theory and provide analytic expressions for the electron energy spectrum and for the total electromagnetic energy spectrum as well as for double- and triple-differential cross sections with respect to electron energy, electron angle, photon energy, and photon angle. We discuss illustrative applications to accelerator-based neutrino experiments and provide the most precise up-to-date values of neutrino-electron scattering cross sections. We present an analysis of theoretical error, which is dominated by the ∼0.2%–0.4% uncertainty of the hadronic correction. We also discuss how searches for new physics can be affected by radiative corrections. DOI: 10.1103/PhysRevD.101.033006 I. INTRODUCTION In the Standard Model of particle physics, neutrinos are massless particles. However, experiments with solar [1–6], atmospheric [7,8], reactor [9–13], and accelerator [14–16] neutrinos 1 establish that neutrinos oscillate and have non- zero mass [17,18], thus providing a convincing example of physics beyond the Standard Model. Fundamental ques- tions about this definitive portal to new physics remain unanswered: What is the origin of neutrino mass? Are lepton number and CP symmetries violated? Do sterile neutrinos exist? What is the absolute scale and ordering of neutrino masses? New experiments aim to address these questions but rely on a precise description of neutrino interactions with the ordinary matter (electrons and nuclei) used to detect them. Interactions with atomic nuclei compose the bulk of neutrino scattering events at accelerator neutrino experi- ments. Although interactions with atomic electrons are rarer, they are nonetheless valuable. The neutrino-electron scattering process plays an important dual role: first, owing to a clean experimental signature and a small cross-section uncertainty, the process provides an incisive constraint on neutrino flux [19,20]; second, the bulk of next-to- leading order (NLO) radiative corrections can be evaluated analytically and thus serve as a prototype for the more complicated cases of neutrino-nucleon and neutrino- nucleus scattering. Radiative corrections to elastic neutrino-electron scat- tering of order α were calculated first in Ref. [21], where only soft-photon bremsstrahlung was considered. In Ref. [22], an analytical phase-space integration technique was developed to include hard-photon brems- strahlung, and the electron energy spectrum for neutrino- electron scattering accompanied by one radiated photon was obtained. The leading-order (LO) cross section in the low-energy limit of the Weinberg theory [23] was evaluated in Ref. [24]. References [25,26] presented the electron energy spectrum in the limit of small elec- tron mass accounting for corrections of order α and including other electroweak NLO radiative corrections. The electromagnetic energy spectrum was considered in Refs. [27,28]. Reference [29] reproduced results of Refs. [22,25] by numerically performing the phase-space integration, and accounted for the electron mass sup- pressed interference term; Ref. [29] also presented a numerical evaluation of the electromagnetic energy spec- trum. The hard-photon correction to the total elastic cross section was studied in Refs. [30,31]. Different aspects of * [email protected]† [email protected]Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. Funded by SCOAP 3 . 1 For the purposes of this paper, “accelerator” neutrinos have energy large compared to the electron mass. PHYSICAL REVIEW D 101, 033006 (2020) 2470-0010=2020=101(3)=033006(36) 033006-1 Published by the American Physical Society
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Theory of elastic neutrino-electron scattering
Oleksandr Tomalak 1,2,3,* and Richard J. Hill 1,2,†
1Department of Physics and Astronomy, University of Kentucky, Lexington, Kentucky 40506, USA2Theoretical Physics Department, Fermi National Accelerator Laboratory, Batavia, Illinois 60510, USA
3Institut für Kernphysik and PRISMA Cluster of Excellence, Johannes Gutenberg Universität,D-55099 Mainz, Germany
(Received 19 July 2019; accepted 15 January 2020; published 21 February 2020)
Theoretical predictions for elastic neutrino-electron scattering have no hadronic or nuclear uncertaintiesat leading order making this process an important tool for normalizing neutrino flux. However, the processis subject to large radiative corrections that differ according to experimental conditions. In this paper, wecollect new and existing results for total and differential cross sections accompanied by radiation of onephoton, νe → νeðγÞ. We perform calculations within the Fermi effective theory and provide analyticexpressions for the electron energy spectrum and for the total electromagnetic energy spectrum as well asfor double- and triple-differential cross sections with respect to electron energy, electron angle, photonenergy, and photon angle. We discuss illustrative applications to accelerator-based neutrino experimentsand provide the most precise up-to-date values of neutrino-electron scattering cross sections. We present ananalysis of theoretical error, which is dominated by the ∼0.2%–0.4% uncertainty of the hadroniccorrection. We also discuss how searches for new physics can be affected by radiative corrections.
DOI: 10.1103/PhysRevD.101.033006
I. INTRODUCTION
In the Standard Model of particle physics, neutrinos aremassless particles. However, experiments with solar [1–6],atmospheric [7,8], reactor [9–13], and accelerator [14–16]neutrinos1 establish that neutrinos oscillate and have non-zero mass [17,18], thus providing a convincing example ofphysics beyond the Standard Model. Fundamental ques-tions about this definitive portal to new physics remainunanswered: What is the origin of neutrino mass? Arelepton number and CP symmetries violated? Do sterileneutrinos exist? What is the absolute scale and ordering ofneutrino masses? New experiments aim to address thesequestions but rely on a precise description of neutrinointeractions with the ordinary matter (electrons and nuclei)used to detect them.Interactions with atomic nuclei compose the bulk of
neutrino scattering events at accelerator neutrino experi-ments. Although interactions with atomic electrons arerarer, they are nonetheless valuable. The neutrino-electron
scattering process plays an important dual role: first, owingto a clean experimental signature and a small cross-sectionuncertainty, the process provides an incisive constrainton neutrino flux [19,20]; second, the bulk of next-to-leading order (NLO) radiative corrections can be evaluatedanalytically and thus serve as a prototype for the morecomplicated cases of neutrino-nucleon and neutrino-nucleus scattering.Radiative corrections to elastic neutrino-electron scat-
tering of order α were calculated first in Ref. [21],where only soft-photon bremsstrahlung was considered.In Ref. [22], an analytical phase-space integrationtechnique was developed to include hard-photon brems-strahlung, and the electron energy spectrum for neutrino-electron scattering accompanied by one radiated photonwas obtained. The leading-order (LO) cross section inthe low-energy limit of the Weinberg theory [23] wasevaluated in Ref. [24]. References [25,26] presentedthe electron energy spectrum in the limit of small elec-tron mass accounting for corrections of order α andincluding other electroweak NLO radiative corrections.The electromagnetic energy spectrum was considered inRefs. [27,28]. Reference [29] reproduced results ofRefs. [22,25] by numerically performing the phase-spaceintegration, and accounted for the electron mass sup-pressed interference term; Ref. [29] also presented anumerical evaluation of the electromagnetic energy spec-trum. The hard-photon correction to the total elastic crosssection was studied in Refs. [30,31]. Different aspects of
Published by the American Physical Society under the terms ofthe Creative Commons Attribution 4.0 International license.Further distribution of this work must maintain attribution tothe author(s) and the published article’s title, journal citation,and DOI. Funded by SCOAP3.
1For the purposes of this paper, “accelerator” neutrinos haveenergy large compared to the electron mass.
PHYSICAL REVIEW D 101, 033006 (2020)
2470-0010=2020=101(3)=033006(36) 033006-1 Published by the American Physical Society
radiative corrections in elastic neutrino-electron scatteringwere also discussed in Refs. [27–44]. See Refs. [45,46] forrecent reviews.In this work, we analytically evaluate relevant dis-
tributions and spectra in elastic (anti)neutrino-electronscattering starting from four-fermion effective field theory(EFT). We take neutrino-lepton and neutrino-quark EFTcoefficients from Ref. [47] (with nf ¼ 4 active quarks atrenormalization scale μ ¼ 2 GeV) and calculate real andvirtual corrections in the MS renormalization schemewithin this theory. Exploiting the technique of Ref. [22],we evaluate the electron energy spectrum and present thiscalculation in a relatively compact form. We generalize thistechnique for the evaluation of the electromagnetic energyspectrum as well as triple- and double-differential crosssections. We discuss a new treatment of hadronic loopdiagrams; this contribution dominates the error budget forneutrino-electron scattering and impacts other neutralcurrent neutrino processes, such as coherent neutrino-nucleus scattering [48]. As illustrative applications usingaccelerator neutrino beams [16,49–51], we consider theimpact of radiative corrections on energy spectra andcompare observables employing electron energy vs totalelectromagnetic energy. For possible low-energy applica-tions, we provide results in analytic form keeping allcharged lepton mass terms. The complete mass dependencecould be useful in the analysis of future reactor and solarneutrino experiments [52–56]. We also discuss exampleswhere radiative corrections can impact searches for newphysics, including neutrino charge radius effects.The paper is organized as follows. Section II considers the
kinematics of neutrino-electron scattering and computes thetree-level scattering process including electroweak correc-tions to the low-energy four-fermion interaction. Section IIIcomputes virtual corrections to elastic scattering. Section IVrepresents the bulk of the paper and computes QEDcorrections involving real radiation. Section V presentsillustrative results for total cross sections and electron energyvs total electromagnetic energy spectra. Section VI presentsour conclusions and outlook. In the main text of the paper,we describe the general strategy of the computations andfocus on results in the limit of small electron mass (i.e.,neutrino beam energy much larger than electron mass).Appendixes provide general expressions retaining all elec-tron mass terms. Appendix A summarizes higher-orderperturbative QCD corrections to heavy-quark loops thatare discussed in Sec. III B. Appendix L displays flux-averaged spectra in experimental conditions of DUNE,MINERvA, NOvA, and T2K experiments.
II. NEUTRINO-ELECTRON SCATTERING
We begin in Sec. II A by reviewing the kinematics ofneutrino scattering on atomic electrons. Throughout thissection we consider general charged leptons l, but in thefollowing sections we specialize to the phenomenologically
most relevant case of the electron, l ¼ e. We introducethe relevant basis of four-fermion effective operators inSec. II B and discuss their coefficients in Sec. II C.
A. Kinematics for neutrino-electron scattering
Consider the scattering of neutrinos on atomic electrons.We neglect the atomic binding energy and momentumcompared to the energy and momentum transferred in thescattering process. Consequently, the initial electron istaken to be at rest in the laboratory frame, where thekinematics is given by pμ ¼ ðm; 0Þ (initial electron withp2 ¼ m2), p0μ ¼ ðE0; k − k0Þ (final charged lepton withp02 ¼ m02), kμ ¼ ðω; kÞ (initial neutrino), and k0μ ¼ðω0; k0Þ (final neutrino); see Fig. 1. The neutrino massscale is much lower than the electron mass and typicalneutrino beam energy, and we neglect the neutrino massmν
throughout. We will let qμ ¼ p0μ − pμ denote the momen-tum transfer to the charged lepton and writeme ¼ m for theelectron mass.Elastic scattering is described by two independent
kinematical variables. It is convenient to introduce theinvariant momentum transfer,
q2 ¼ ðp0 − pÞ2; ð1Þ
and the squared energy in the center-of-mass referenceframe,
s ¼ ðpþ kÞ2: ð2ÞNote that production of heavier charged leptons in neu-trino-electron scattering is possible when the neutrino beamenergy is high enough. Using s ¼ m2 þ 2mω ≥ m02 we seethat ω ≥ ðm2
μ −m2Þ=ð2mÞ ≈ 10.9 GeV to produce a muon(m0 ¼ mμ), while ω ≥ 3089 GeV for the production ofτ (m0 ¼ mτ).The neutrino scattering angle in the laboratory frame,Θν,
can be expressed in terms of the final neutrino energy ω0 as
cosΘν ¼ωω0 −mðω − ω0Þ − m2−m02
2
jkjjk0j
¼ 1þmω−mω0 −
m2 −m02
2ωω0 : ð3Þ
The final neutrino energy varies between backward andforward scattering in the range
FIG. 1. Neutrino-electron scattering kinematics.
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-2
mω
mþ 2ωþ m2 −m02
2ðmþ 2ωÞ ≤ ω0 ≤ ωþm2 −m02
2m; ð4Þ
corresponding to the charged lepton energy range
mþm02 −m2
2m≤ E0 ≤ mþ 2ω2
mþ 2ωþ m02 −m2
2ðmþ 2ωÞ : ð5Þ
The angle between recoil charged lepton direction and theneutrino beam direction, Θe, is given by
cosΘe ¼ωE0 −m2 −mðω − E0Þ þ m2−m02
2
ωjp0j ; ð6Þ
and scattering is possible only in the forward cone boundedby Θmax
when it varies between 0 and 1, i.e., the electron is scatteredalways into the forward hemisphere.
B. Effective neutrino-charged lepton operators
Neutrino-electron scattering is described by theexchange of weak vector bosons W and Z (with massesMW and MZ, respectively) in the Standard Model; cf.Fig. 2 for contributing Feynman diagrams. At energiesbelow the electroweak scale, the interactions of neutrinosand charged leptons are determined by an equivalenteffective Lagrangian [57–59]. Neglecting corrections sup-pressed by 1=M2
W, the effective Lagrangian consists ofmomentum-independent four-fermion operators.At tree level, the matching onto this effective Lagrangian
Leff is readily obtained,
Leff ¼ −g2
M2WðJWþÞμðJW−Þμ −
g2
2M2ZðJZÞμðJZÞμ; ð9Þ
where JμW− , JμWþ ¼ J†μW− , and JμZ are charged and neutralcurrents in the StandardModel Lagrangian coupling toWþ,W−, and Z, respectively, and g is the electroweak SUð2ÞLcoupling constant. Focusing on leptonic vs quark operators,we have
JμW− ¼ 1ffiffiffi2
pXl
l γμPLνl; ð10Þ
JμZ ¼ 1
cos θW
Xl
��−1
2þ sin2θW
�lγμPLl
þ sin2θWlγμPRlþ 1
2νlγ
μPLνl
�; ð11Þ
where PL ¼ ð1 − γ5Þ=2 and PR ¼ ð1þ γ5Þ=2 are projectionoperators onto left-handed and right-handed fermions andθW denotes the weak mixing angle satisfying MW=MZ ¼cos θW . After Fierz rearrangement of the charged currentcontribution, the result may be written as
Leff ¼ −Xl;l0
νlγμPLνll0γμðcνll
0L PL þ cRPRÞl0
− cXl≠l0
νl0γμPLνllγμPLl0; ð12Þ
with coefficients cνll0
L ; cR, and c,
cνll0
L ¼ 2ffiffiffi2
pGF
�sin2θW −
1
2þ δll0
�;
cR ¼ 2ffiffiffi2
pGFsin2θW; c ¼ 2
ffiffiffi2
pGF; ð13Þ
where we have introduced the Fermi constant GF ¼g2=ð4 ffiffiffi
2p
M2WÞ, and where the Kronecker symbol δll0
satisfies δll0 ¼ 1 for l ¼ l0 and δll0 ¼ 0 for l ≠ l0.Note that coefficients c and cR are the same for allcombinations of lepton flavors, while the coefficient
cνll0
L depends on whether the neutrino and charged leptonhave the same flavor.Neglecting the neutrino magnetic moment contribution
[60–66], the leading-order cross section of neutrino-leptonscattering can be expressed, in all possible cases, as[24,25,29,67–92]
dσνll0→νll0
LO
dω0 ¼ m4π
½ðcνll0L Þ2IL þ c2RIR þ cνll0
L cRILR�; ð14Þ
dσνll0→νll0
LO
dω0 ¼ m4π
½ðcνll0L Þ2IR þ c2RIL þ cνll0
L cRILR�; ð15Þ
FIG. 2. Leading-order contributions to neutrino-lepton scatter-ing in the Standard Model. The graph with the exchange of Zboson contributes to the neutrino and antineutrino scattering.l and l0 denote charged leptons of any flavor in this figure.
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-3
dσνll0→νl0l
LO
dω0
����l≠l0
¼ m4π
c2IL; ð16Þ
dσνll→νl0l0
LO
dω0
����l≠l0
¼ m4π
c2IR; ð17Þ
with kinematical factors:
IL ¼ ðk · pÞðk0 · p0Þm2ω2
¼ 1þm2 −m02
2mω→ 1; ð18Þ
IR ¼ ðk · p0Þðk0 · pÞm2ω2
¼ ω02
ω2
�1þm02 −m2
2mω0
�→
ω02
ω2; ð19Þ
ILR ¼ −mm0ðk · k0Þ
m2ω2¼ −
m0
ω
�1 −
ω0
ωþm2 −m02
2mω
�
→ −mω
�1 −
ω0
ω
�; ð20Þ
where the limit of elastic process, i.e.,m0 ¼ m, is presentedin the last step. The neutrino-energy spectra in Eqs. (14)–(17) are equivalent to the recoil electron energy spectra dueto energy conservation: mþ ω ¼ E0 þ ω0. In particular,dσ=dE0 ¼ dσ=dω0. We later apply this observation tocompute differential cross sections with respect to totalelectromagnetic energy in the presence of radiative cor-rections. To study the angular spectrum, the differentialcross section can be obtained by exploiting
dE0 ¼ 4mω2ðmþ ωÞ2 cosΘed cosΘe
½ðmþ ωÞ2 − ω2cos2Θe�2: ð21Þ
We observe that the contribution from the interference termILR is suppressed by the charged lepton mass. The neutrinoand antineutrino scattering are related by the substitution
IL ↔ IR (k ↔ k0) or equivalently cνll0
L ↔ cR.Note that νll → νll and νll → νll cross sections
involving one flavor seem to be not positive definite forenergies comparable with the charged lepton mass due tothe helicity-flip interference term cνllL cR. However, thecross section is always positive in the physical region ofscattering mω=ðmþ 2ωÞ < ω0 < ω and can vanish only inthe case of forward recoil electrons with maximum energyE0 ¼ mþ 2ω2=ðmþ 2ωÞ [93–96] in the scattering of anelectron antineutrino of energy ω:
ω ¼�cνllL
cR− 1
�m2: ð22Þ
We discuss the impact of radiative corrections on thecancellation (22) in Sec. V D.
C. Effective neutrino-lepton and neutrino-quarkinteractions beyond leading order
Higher-order electroweak and QCD contributions modifycouplings in the effective Lagrangian of Eq. (12). Theevaluation of virtual NLO corrections to elastic neutrino-charged lepton scattering also involves interaction withquarks and gluons; see Secs. III B and III C. The relevantneutral current part of the effective neutrino-quarkLagrangian is
Lqeff ¼ −
Xl;q
νlγμPLνlqγμðcqLPL þ cqRPRÞq; ð23Þ
with (neutrino flavor independent) left- and right-handedcouplings cqL and cqR, respectively. At tree level,
cqL ¼ 2ffiffiffi2
pGFðT3
q −Qqsin2θWÞ; cqR ¼ −2ffiffiffi2
pGFQqsin2θW;
ð24Þ
where T3q denotes the quark isospin (þ1=2 for q ¼ u; c,
−1=2 for q ¼ d; s) and Qq its electric charge in units of thepositron charge (þ2=3 for q ¼ u; c, −1=3 for q ¼ d; s).For numerical analysis, we employ low-energy effectivecouplings from Ref. [47]. For definiteness, we take inputsin four-flavor QCD (nf ¼ 4) at renormalization scaleμ ¼ 2 GeV in the MS scheme and do not distinguishbetween couplings to u (d) and c (s) quarks.2
The effective Lagrangians of Eqs. (12) and (23), and thecorresponding charged current quark operators [47], deter-mine neutrino scattering rates at GeV energy scales, up tocorrections suppressed by powers of electroweak scaleparticle masses. Electroweak scale physics is encoded in
TABLE I. Effective couplings (in units 10−5 GeV−2) in the Fermi theory of neutrino-fermion scattering with four quark flavors at thescale μ ¼ 2 GeV. The error due to the uncertainty of Standard Model parameters is added in quadrature to a perturbative error ofmatching.
2In Ref. [47], one-loop matching to the Standard Model isperformed at the electroweak scale accounting for the leadingQCD corrections with one exchanged gluon inside quark loopsand neglecting masses of all fermions except the top quarkcompared to the electroweak scale. The matching is accompaniedby renormalization group evolution to GeV scales to resum largeelectroweak logarithms in the effective couplings. The relation ofthe couplings in Table I to various definitions of GF and sin2 θW isdiscussed in Ref. [47].
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-4
the values of the operator coefficients, summarized inTable I. Real photon radiation and virtual correctionsinvolving the photon and other light particles must stillbe calculated within the effective theory.
III. VIRTUAL QED CORRECTIONS
In this section, we present virtual corrections, consider-ing QED vertex corrections involving virtual photons inSec. III A and closed fermion loop contributions fromleptons and heavy quarks in Sec. III B. We estimate thecorrection coming from light-quark loops in Sec. III C.
A. QED vertex correction
We consider one-loop virtual corrections in elastic(anti)neutrino-electron scattering νle → νle (νle → νle).Within the Standard Model, the vertex correction isgiven by the diagrams in Fig. 3, while only the singlediagram in Fig. 4 contributes in the effective theory. Theusual field renormalization factors must be applied toexternal legs.First, we evaluate the one-loop vertex correction to the
matrix element of left-handed (L) and right-handed (R)charged lepton currents JL;Rμ ¼ eðp0ÞγμPL;ReðpÞ fromEq. (12). We perform the integration in d ¼ 4 − 2ε dimen-sions of spacetime to regularize the ultraviolet divergence,
where =k≡ kμγμ for any four-vector k, ξγ is the photon gauge parameter, and a is an arbitrary constant associated with thephoton mass regulator. The small photon mass λ is introduced to regulate infrared (IR) divergences. The corresponding fieldrenormalization factor of external charged leptons is
Zl ¼ 1 −α
4π
ξγε−
α
4π
�ln
μ2
m2þ 2 ln
λ2
m2þ 4
�þ α
4πð1 − ξγÞ
�lnμ2
λ2þ 1þ aξγ ln aξγ
1 − aξγ
�: ð26Þ
Neglecting Lorentz structures whose contractions withthe neutrino current vanish at mν ¼ 0, the resulting cor-rection can be expressed as3
ðZl − 1ÞJL;Rμ þ δJL;Rμ ¼ α
πðf1JL;Rμ þ f2jL;Rμ Þ; ð29Þ
in terms of form factors f1 and f2, and the additionalcurrents jLμ and jRμ :
jLμ ¼ 1
2eðp0Þ
�γμγ5 þ
iσμνqν
2m
�eðpÞ; ð30Þ
jRμ ¼ 1
2eðp0Þ
�−γμγ5 þ
iσμνqν
2m
�eðpÞ: ð31Þ
Here σμν ¼ i2½γμ; γν�.
Using Eqs. (25) and (26), the UV finite and gauge-independent virtual correction is given in Eq. (29) by one-loop QED form factors [97,98]:
f1ðβÞ ¼ −1
2β
�β −
1
2ln1þ β
1 − β
�ln
λ2
m2
þ 1
β
�3þ ρ
8ln1þ β
1 − β−1
8ln1þ β
1 − βln
�21þ ρ
ρ
��
−1
2β
�Li2
β − 1þ ρ
2β− Li2
β þ 1 − ρ
2β
�− 1; ð32Þ
FIG. 3. Virtual corrections to elastic neutrino-electron scatter-ing in the Standard Model corresponding to the vertex correctionin effective theory.
FIG. 4. QED vertex correction to elastic neutrino-electronscattering in effective theory.
3Note that the vertex correction can be expressed as amodification of vector and axial currents:
eðp0ÞγμeðpÞ → eðp0ÞγμeðpÞ þα
πeðp0Þ
�f1γμ þ f2
iσμνqν
2m
�eðpÞ;
ð27Þ
eðp0Þγμγ5eðpÞ → eðp0Þγμγ5eðpÞ þα
πðf1 − f2Þeðp0Þγμγ5eðpÞ:
ð28Þ
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-5
f2ðβÞ ¼ρ
4βln1þ β
1 − β; ð33Þ
which are expressed in terms of the recoil electron velocityβ and the parameter ρ:
β ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffi1 −
m2
E02
r; ρ ¼
ffiffiffiffiffiffiffiffiffiffiffiffiffi1 − β2
q¼ m
E0 : ð34Þ
The vertex correction (29) to the unpolarized crosssection can be expressed as a sum of factorizable andnonfactorizable terms:
dσνle→νlev ¼ α
πδvdσ
νle→νleLO þ dσνle→νle
v;NF : ð35Þ
The factorizable correction is given by
δv ¼ 2f1: ð36Þ
The nonfactorizable term dσνle→νlev;NF is obtained by modi-
fying kinematical factors Ii in Eqs. (14) and (15) as Ii →Ii þ α
π f2δvIi where
δvIL ¼ δvIR ¼ 1
2ILR −
ω0
ω; ð37Þ
δvILR ¼ 2
�IL þ IR −
ω0
ω
�− ILR: ð38Þ
The resulting vertex correction to the unpolarized crosssection of Eq. (35) is in agreement with Refs. [29,37]. Inthe limit of a massless electron, the Pauli form factorvanishes, f2ðβÞ → 0, and the correction becomes exactlyfactorizable.
B. Closed fermion loops: Leptons and heavy quarks
In addition to the corrections involving virtual photons inSec. III A, we must account for the corrections with aclosed fermion loop of Fig. 5. These corrections correspondto the diagram of penguin type and the effects of γ-Zmixing in the StandardModel; cf. Fig. 6. They represent theEFT determination of the kinematical dependence ofelectroweak corrections; cf. Refs. [25,27].
In this section, we consider the loop contribution from anarbitrary fermion with mass mf and charge Qf (in units ofthe positive positron charge) and effective left- and right-handed couplings cfL and cfR, respectively, as in Eqs. (12)and (23). Note that the coupling cfL for charged leptons(f ¼ l) depends on the neutrino flavor. This perturbativetreatment applies to loops involving charged leptons orheavy quarks (mf ≫ ΛQCD). Light quarks require a non-perturbative treatment, as discussed in Sec. III C below.Starting from the nf ¼ 4 flavor theory discussed inSec. II C, we treat the charm quark as heavy and the up,down, and strange quarks as light.The correction can be expressed as a modification
of electron left- and right-handed currents, cL;RJL;Rμ →
cL;RJL;Rμ þ cfL;RδJ
L;Rμ ,
δJL;Rμ ¼ Qfe2eðp0ÞγλeðpÞ−gλρq2
×Z
iddLð2πÞd
Tr½γρð=LþmfÞγμPL;Rð=L − =qþmfÞ�ðL2 −m2
fÞððL − qÞ2 −m2fÞ
;
ð39Þ
and does not depend on the photon gauge. Corrections toeither left- or right-handed currents are vectorlike and maybe written
δJLμ ¼ δJRμ ¼ Qfα
2πΠðq2; mfÞðJLμ þ JRμ Þ: ð40Þ
At renormalization scale μ in the MS scheme, the formfactor Π is
and corresponds to vacuum polarization in QED [99–103].
FIG. 5. Long-range dynamics in elastic neutrino-electron scat-tering in the effective theory. Loops with all interacting fields inthe theory are summed up.
FIG. 6. Standard Model diagrams giving rise to long-rangedynamics in EFT: γ-Z mixing and penguin-type diagram.
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-6
The resulting “dynamical” correction to the unpolarizedcross section, dσνle→νle
dyn , can be expressed in the followingform:
dσνle→νledyn ¼ α
π
Xf≠uds
QfΠðq2; mfÞdσνle→νledyn;f þ dσνle→νle
dyn;uds :
ð42Þ
The contribution from three light flavors dσνle→νledyn;uds is
discussed below in Sec. III C. The reduced cross sectiondσνle→νle
dyn;f is obtained by replacing νle couplings inEqs. (14) and (15) as
ðcνll0L Þ2 → cνll0
L ðcfL þ cfRÞ; ð43Þ
ðcRÞ2 → cRðcfL þ cfRÞ; ð44Þ
cνll0
L cR →1
2ðcνll0L þ cRÞðcfL þ cfRÞ: ð45Þ
The sum in Eq. (42) extends over all charged leptons(e, μ, τ) and heavy quarks (c) in the theory (a factor Nc ¼ 3is obtained in the sum over colors for heavy quarks). Wealso include QCD corrections due to exchanged gluonsinside the quark loop; see Refs. [104–107] and Appendix Afor exact expressions.The momentum transfer in elastic neutrino-electron
scattering is suppressed by the electron mass,
0 ≤ −q2 < 2mω: ð46Þ
For neutrino beam energies smaller than 10 GeV, thisimplies jq2j≲ 0.01 GeV2. Consequently, the contributionof loops with heavy quarks can be well approximatedsubstituting Πðq2; mfÞ → Πð0; mfÞ.
C. Light-quark contribution
At small q2, QCD perturbation theory cannot be appliedto evaluate the light-quark contribution in Fig. 5. Weinstead evaluate this contribution by relating it to measuredexperimental quantities.For GeV energy neutrino beams, momenta in the range
(46) are small compared to hadronic mass scales, and wethus evaluate the relevant hadronic tensor at q2 ¼ 0.Neglecting NLO electroweak corrections to the quarkcoefficients of Eq. (23), the light-quark contribution inEq. (42) may be written as
dσνle→νledyn;uds ¼ α
πðΠð3Þ
3γ ð0Þ − 2sin2θWΠð3Þγγ ð0ÞÞdσνle→νle
dyn;uds : ð47Þ
The reduced cross section dσνle→νledyn;uds is obtained replacing
νle couplings in Eqs. (14) and (15) as
ðcνll0L Þ2 → 2ffiffiffi2
pGFc
νll0L ; c2R → 2
ffiffiffi2
pGFcR;
cνll0
L cR →ffiffiffi2
pGFðcνll
0L þ cRÞ: ð48Þ
The quantity Πγγ is defined by the vacuum correlationfunction,
ðqμqν − q2gμνÞΠγγðq2Þ
¼ 4iπ2Z
ddxeiq·xh0jTfJμγ ðxÞJνγð0Þgj0i; ð49Þ
where Jμγ ¼ Pq Qqqγμq is the quark electromagnetic
current. Similarly, Π3γ is given by
ðqμqν − q2gμνÞΠ3γðq2Þ
¼ 4iπ2Z
ddxeiq·xh0jTfJμ3ðxÞJνγð0Þgj0i; ð50Þ
where Jμ3 ¼P
q T3qqγμq is (the third component of) the
quark isospin current. The current-current correlation
functions Πð3Þij ð0Þ are evaluated at q2 ¼ 0 for nf ¼ 3
flavors, in the MS scheme.Unlike the light-quark contribution to the photon propa-
gator, involving only Πγγ, the correction to neutral currentneutrino-electron scattering involves also Π3γ and cannotbe directly related to the total hadron production crosssection in eþe− collisions. However, an approximate rela-
tion between Πð3Þγγ and Πð3Þ
3γ holds in the limit of SUð3Þfflavor symmetry for three light quarks [108,109]. Ingeneral, the flavor sums read
Πð3Þγγ ¼
Xi;j
QiQjΠij ¼ 4
9Πuu þ 1
9Πdd þ 1
9Πss −
4
9Πud
−4
9Πus þ 2
9Πds; ð51Þ
Πð3Þ3γ ¼
Xi;j
T3i QjΠij ¼ 1
2
�2
3Πuu þ 1
3Πdd þ 1
3Πss − Πud
− Πus þ 2
3Πds
�: ð52Þ
SUð3Þf symmetry implies Πuu ¼ Πdd ¼ Πss and Πud ¼Πus ¼ Πds, and consequently, the simple relation [108]
Πð3Þ3γ ð0Þ ≈ Πð3Þ
γγ ð0Þ. This allows us to express the entirelight-quark contribution to the unpolarized cross section
dσνle→νleuds in terms of the single observable Πð3Þ
γγ ð0Þ.For numerical evaluation, we use the dispersive analysis
of eþe− cross-section data and measurements of hadronic τdecays combined with a perturbative treatment of the high-energy contribution in Refs. [110–112],
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-7
Πð3Þγγ ð0Þjμ¼2GeV ¼ 3.597ð21Þ: ð53Þ
To estimate uncertainty due to the SUð3Þf symmetryapproximation, we may consider an alternative SUð2Þfansatz that sets Πuu ¼ Πdd, Πss ¼ 0 and neglects discon-nected, Okubo, Zweig and Iizuka (OZI)-suppressed termsΠud ¼ Πus ¼ Πds ¼ 0. The flavor sums (51) and (52) then
yield Πð3Þ3γ ¼ 9Πð3Þ
γγ =10, only a 10% correction to theSUð3Þf symmetry limit. In the final error budget, weconsider a more conservative 20% uncertainty on thisrelation,
Πð3Þ3γ ð0Þ ¼ ð1� 0.2ÞΠð3Þ
γγ ð0Þ: ð54Þ
Renormalization scale dependence of the light-quark con-tribution (47) is perturbatively calculable. For μ ≠ 2 GeV,the additional correction corresponds with 3Πð0; mf ¼2 GeVÞ of Eq. (41) for each quark (accounting forNc ¼ 3 quark colors).The replacement Πðq2Þ → Πð0Þ introduces an error of
relative order mω=m2ρ ≲ 10−3 for ω≲ GeV, where we use
mρ ¼ 770 MeV as a typical hadronic scale. This regimeincludes neutrinos of energy up to the TeV range pro-duced at modern high-energy accelerators, and the uncer-tainty is contained in the error budgets (53) and (54).At much higher neutrino energies where q2 correctionsare appreciable but still in the nonperturbative domain,the same SUð3Þf approximation [at momentum transferq2 ≠ 0 in Eq. (54)] can be used to describe the light-quarkcontribution.4
IV. REAL PHOTON EMISSION
Let us consider one-photon bremsstrahlung. Section IVA provides basic expressions for this process. We thenstudy relevant differential observables accounting for bothsoft and hard photons. We start with the electron energy,electron angle, and photon energy triple-differential crosssection in Sec. IV B. Integrating over one energy variable,we obtain double-differential distributions in Secs. IV Cand IV D. The double-differential cross section with respectto two energy variables is described in Sec. IV E. Weprovide the distribution with respect to the photon energyand photon angle in Sec. IV F. Integrating it over the photonangle, we provide the photon energy spectrum in Sec. IVG.Finally, we discuss the real soft-photon correction to elasticneutrino-electron scattering and present electron andelectromagnetic energy spectra in Secs. IV H and IV I,respectively. We also provide the absolute scattering crosssection in Sec. IV J. Throughout this section, we present allexpressions in the limit of small electron mass and provideexpressions for general mass in the Appendix. For the
energy spectra in Secs. IV H and IV I, we provide a generaldiscussion of momentum regions at arbitrary mass, butpresent the massless limit and relegate details to theAppendix.
A. Radiation of one photon
The one-photon bremsstrahlung amplitude T1γ (cf.Fig. 7) contains terms corresponding to radiation fromthe initial electron T1γ
i and from the final electron T1γf ,
T1γ ¼ T1γi þ T1γ
f : ð55Þ
The amplitude T1γi is obtained from the tree-level amplitude
with the substitution
eðpÞ → eε�ρ=p − =kγ þm
ðp − kγÞ2 −m2γρeðpÞ; ð56Þ
where kγ is a photon momentum and ε�ρ is the photon
polarization vector. The amplitude T1γf is obtained from the
tree-level amplitude with the substitution
eðp0Þ → eε�ρeðp0Þγρ =p0 þ =kγ þm0
ðp0 þ kγÞ2 −m02 : ð57Þ
Evaluating the spin-averaged squared matrix element,Pspin jT1γj2, we obtain for the bremsstrahlung cross sec-
tions:
dσνle→νleγLO ¼ α
4π
mω
π3½ðcνleL Þ2IL þ c2RIR þ cνleL cRI
LR�; ð58Þ
dσνle→νleγLO ¼ α
4π
mω
π3½ðcνleL Þ2IR þ c2RIL þ cνleL cRI
LR�; ð59Þ
where terms Ii contain the phase-space integration
Ii ¼Z
Ri
m2ω2δ4ðkþ p − kγ − k0 − p0Þ d
3kγ2kγ
d3k0
2ω0d3p0
2E0 ;
ð60Þ
and kinematical factors Ri are expressed in terms of particlemomenta as
FIG. 7. One-photon bremsstrahlung in elastic neutrino-electronscattering.
4See Ref. [113] for a discussion of Πγγðq2Þ − Πγγð0Þ.
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-8
RL ¼ −IL�
pμ
ðp · kγÞ−
p0μ
ðp0 · kγÞ�2
m2ω2 þ ðk · p0Þðk0 · p0Þðkγ · p0Þ −
ðk · pÞðk0 · pÞðkγ · pÞ
þ ðk · pÞðk0 · p0Þðkγ · p0Þ −
ðk · pÞðk0 · p0Þðkγ · pÞ
þ ðk0 · p0Þðk · kγÞðkγ · pÞ
�1þ m2
ðkγ · pÞ−
ðp · p0Þðkγ · p0Þ
�þ ðk · pÞðk0 · kγÞ
ðkγ · p0Þ�1 −
m02
ðkγ · p0Þ þðp · p0Þðkγ · pÞ
�; ð61Þ
RR ¼ −IR�
pμ
ðp · kγÞ−
p0μ
ðp0 · kγÞ�2
m2ω2 þ ðk · p0Þðk0 · p0Þðkγ · p0Þ −
ðk · pÞðk0 · pÞðkγ · pÞ
þ ðk0 · pÞðk · p0Þðkγ · p0Þ −
ðk0 · pÞðk · p0Þðkγ · pÞ
þ ðk · p0Þðk0 · kγÞðkγ · pÞ
�1þ m2
ðkγ · pÞ−
ðp · p0Þðkγ · p0Þ
�þ ðk0 · pÞðk · kγÞ
ðkγ · p0Þ�1 −
m02
ðkγ · p0Þ þðp · p0Þðkγ · pÞ
�; ð62Þ
RLR ¼ −ILR
�pμ
ðp · kγÞ−
p0μ
ðp0 · kγÞ�2
m2ω2 −2mm0ðk · kγÞðk0 · kγÞ
ðp · kγÞðp0 · kγÞ: ð63Þ
Kinematical factors IL; IR; ILR are given in terms of momentum invariants in Eqs. (18)–(20) and are evaluated in thekinematics of 2 → 3 scattering. Neutrino and antineutrino scattering are related by the substitution RL ↔ RR (equivalently,k ↔ k0). The IR-divergent parts of RL and RR correspond to integrals R and R in Ref. [25], respectively.
B. Triple-differential distribution
We evaluate the bremsstrahlung cross section using the integration technique of Ref. [22] and provide expressions for thetriple-differential cross section with respect to electron angle, electron energy, and photon energy keeping all electron massterms in Appendix B. In the limit of small electron mass,5 the result can be approximated by the following substitutions inEqs. (58) and (59),6
where ω0 ¼ ω − kγ − E0 and the variable z ≤ 1 is intro-duced to emphasize the forward direction of the relativisticelectron:
1 − cos θe ≡mωð1 − zÞ: ð68Þ
Note the difference between the electron scattering angle inthe elastic process [Θe of Eq. (6)] and in the scatteringprocess with radiation (θe). At m → 0, the physical regionof kinematical variables is given by
0 ≤ E0 ≤ ω; 2 −ω
E0 ≤ z ≤ 1; 0 ≤ kγ ≤ ω − E0:
ð69ÞIn the vicinity of the elastic peak,
z → Z ¼ 1 −ω0
ω − ω0 ; ð70Þ
5In the following, we denote the limit of small electron masscompared to all other relevant energy scales as ω ≫ m.
6Note that suppressed terms in the lepton mass expansion of ILand IR contribute to the cross section at the same order as ILR. For aconsistent power counting, one has either to neglect the inter-ference term completely or to expand IL and IR further.
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-9
the cross section of Eqs. (64)–(66) diverges. The smallmass approximation in Eqs. (64)–(66) is valid only awayfrom this region:
jz − Zj ≫ mE0
k2γðE0 þ kγÞ2
ω0
ω − ω0 : ð71Þ
For a correct description in the elastic peak region, and toobtain distributions (such as energy spectra) that involveintegration through this region, expressions with an elec-tron mass of Appendix B must be used.
C. Double-differential distribution in electronenergy and electron angle
Integrating the triple-differential distribution over thephoton energy kγ, we obtain the double-differential crosssection with respect to the recoil electron energy andelectron angle. We provide the double-differential distri-bution in electron energy and electron angle keeping allelectron mass terms in Appendix C. In the limit of smallelectron mass, the cross section is given by the followingsubstitutions in Eqs. (58) and (59)6:
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-10
The variable z ≤ 1 is introduced to emphasize the forwarddirection of the relativistic electron,
1 − cos θe ≡ mE0 ð1 − zÞ: ð74Þ
At m → 0, the physical region of kinematical variables isgiven by
m ≤ E0 ≤ ω;E0
ω≤ z ≤ 1: ð75Þ
D. Double-differential distribution in electromagneticenergy and electron angle
To obtain the distribution with respect to the electromag-netic energy and electron angle, we use the neutrino energyω0instead of kγ in the triple-differential cross section, changethe integration order, and integrate first over the electronenergy. The final neutrino energy determines the total electro-magnetic energy EEM: EEM ¼ E0 þ kγ ¼ mþ ω − ω0 andcan be used to obtain EEM distributions since dEEM ¼ −dω0.In the limit of small electron mass, the neutrino energy
and electron angle distribution is given by the followingsubstitutions in Eqs. (58) and (59)6:
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-11
This approximation is valid only away from the elastic peak [cf. Eq. (70)] when
jz − Zj ≫ ω0
ω − ω0 : ð78Þ
At m → 0, the physical region is given by
0 ≤ ω0 ≤ ω; 1 −ω
m≤ z ≤ 1: ð79Þ
We discuss the double-differential distribution in electromagnetic energy and electron angle keeping all electron massterms in Appendix D.
E. Double-differential distribution in photon energy and electron energy
To obtain the distribution with respect to photon energy and electron energy, we can change the integration order andintegrate the triple-differential cross section first over the electron scattering angle. In the limit of small electron mass, theleading terms of the photon energy and electron energy distribution are given by the following substitutions in Eqs. (58) and(59)6:
IL⟶ω≫m
�−29E02 þ 8E0kγðω0ω − 3Þ þ k2γðω02
ω2 − 6Þ12E2
EMþ 1
2
�1þ E02
E2EM
�ln2E0EEM
mkγ
�Dγ; ð80Þ
IR⟶ω≫m
�−29E02 þ 8E0kγðωω0 − 3Þ þ k2γðω2
ω02 − 6Þ12E2
EMþ 1
2
�1þ E02
E2EM
�ln2E0EEM
mkγ
�ω02
ω2Dγ; ð81Þ
ILR⟶ω≫m
�E02ð4 E2
EMωω0 − 1Þ − E0kγðωω0 − 3Þðω0
ω − 3Þ þ 3k2γ2E2
EM−�E0EEM
ωω0 þ k2γE2EM
�ln2E0EEM
mkγ
�mEEM
ω0
ωDγ; ð82Þ
valid in the physical region, 0 ≤ E0 þ kγ ≤ ω, with the phase-space factor Dγ,
Dγ ¼ π2dkγkγ
dE0
ω: ð83Þ
We discuss the double-differential distribution in photon energy and electron energy keeping all electron mass terms inAppendix E.
F. Double-differential distribution in photon energy and photon angle
Besides the electron angle, the photon scattering angle θγ can be measured in principle. We consider the distribution withrespect to the photon energy and the photon angle in the following. We present the double-differential distribution in photonenergy and photon angle keeping all electron mass terms in Appendix F.In the limit of small electron mass, the cross section is given by the following substitutions in Eqs. (58) and (59)6:
where the variable z ≤ 1 is introduced to emphasize theforward direction of the photon,
1 − cos θγ ≡mωð1 − zÞ: ð86Þ
The photon angle with respect to the neutrino beamdirection is bounded as
cos θγ ≥ 1 −mkγ
�1 −
kγω
�; ð87Þ
while the physical region for the photon energy is0 ≤ kγ ≤ ω.
G. Photon energy spectrum
Integrating the double-differential distribution in photonand electron energies over the electron energy, or thedouble-differential distribution in the photon energy and
photon scattering angle over the angle, we obtain thephoton energy spectrum. We present the photon energyspectrum keeping all electron mass terms in Appendix G.The leading terms in the electron mass expansion are givenby the following substitutions in Eqs. (58) and (59):
IL⟶ω≫m
π2
ωgL
�kγω
�dkγ; ð88Þ
IR⟶ω≫m
π2
ωgR
�kγω
�dkγ; ð89Þ
ILR⟶ω≫m
π2
ω
mωgLR
�kγω
�dkγ; ð90Þ
with functions gLðxÞ, gRðxÞ, and gLRðxÞ derived first in thepresent paper6,
gLðxÞ ¼ð1 − xÞðx2 − 20x − 53Þ
12x−�3þ 1
x
�ln x −
x2 þ x − 2
2xln2ωð1 − xÞ
mþ ln
2ω
mln xþ π2
6− Li2x; ð91Þ
gRðxÞ ¼ −ð1 − xÞð37x2 þ 223xþ 73Þ
36x−�1
3xþ 9þ 5x
2
�ln xþ ð1 − xÞðx2 þ 4xþ 1Þ
3xln2ωð1 − xÞ
m
þ�ln2ω
mln xþ π2
6− Li2x
�ð1þ xÞ; ð92Þ
gLRðxÞ ¼ð1 − xÞð11 − 13xÞ
4xþ 1 − 2x
2xln x −
ð1 − xÞ2x
ln2ωð1 − xÞ
m: ð93Þ
The integral of the photon energy spectrum obtainedfrom Eqs. (88)–(90) is infrared divergent if extended toarbitrary small photon energy. The total NLO cross sectionis obtained by implementing an infrared regulator andincluding the (separately infrared divergent) virtual cor-rection from Sec. III.
H. Electron energy spectrum
All of our following calculations for neutrino andantineutrino scattering contain the same IR contribution
arising from the soft-photon phase space, when the elasticprocess (without radiation) and scattering with bremsstrah-lung are experimentally indistinguishable. The soft-photoncontribution has to be accounted for in differential crosssections with respect to one kinematical variable (except forthe photon energy spectrum of Sec. IVG, where one simplyevaluates the spectrum above a chosen minimum photonenergy). The amplitude T1γ
soft for the radiation of one softphoton with energy kγ ≤ ε, where ε ≪ m;ω denotes acutoff regulator, can be expressed in factorizable form as
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-13
T1γsoft ¼
�ðε� · p0Þðkγ · p0Þ −
ðε� · pÞðkγ · pÞ
�eT; ð94Þ
where T corresponds to the amplitude without radiation.The corresponding contribution dσνle→νleγ
soft to the brems-strahlung spectrum is given by
dσνle→νleγsoft ¼ α
πδsdσ
νle→νleLO ; ð95Þ
with the soft correction factor δs [21,25,29,37],
δs ¼1
β
�Li2
1 − β
1þ β−π2
6
�−2
β
�β −
1
2ln1þ β
1 − β
�ln2ε
λ
þ 1
2βln1þ β
1 − β
�1þ ln
ρð1þ βÞ4β2
�þ 1: ð96Þ
The velocity β of Eq. (34) (and ρ ¼ffiffiffiffiffiffiffiffiffiffiffiffiffi1 − β2
p),
β ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffi1 −
m2
E2
r; ð97Þ
now describes either electron or electromagnetic energyspectra and E stands for the corresponding energy, i.e.,E ¼ E0 or E ¼ EEM. Note the exact cancellation of the IRdivergence in the sum of the vertex correction and the soft-photon emission; i.e., δs þ δv does not depend on thefictitious photon mass λ [114–117]. The correction ofEq. (96) comes entirely from the first (factorizable) termsin Eqs. (61)–(63) and still contains an unphysical depend-ence on the photon energy cutoff ε.For further evaluation of the electron angle distributions,
we introduce the four-vector l [22],
l ¼ kþ p − p0 ¼ ðl0; fÞ; ð98Þ
with the laboratory frame values,
l0 ¼ mþ ω − E0; ð99Þ
f2 ¼ jfj2 ¼ ω2 þ β2E02 − 2ωβE0 cos θe: ð100Þ
Besides the soft-photon correction, the first factorizableterms in Eqs. (61)–(63) contribute from the region kγ ≥ ε. Itis convenient to split this contribution into two parts. Thereare no restrictions on the phase-space integration in regionI: l2 ¼ l20 − f2 ≥ 2εðl0 þ fÞ. In region II: l2 ≤ 2εðl0 þ fÞ,which includes the region of scattering with elastic kin-ematics, the phase space of the final photon is bounded by
cos γ ≥1
f
�l0 −
l2
2ε
�; ð101Þ
where γ is the angle between f and kγ . The bremsstrahlungcontribution from region I, dσνle→νleγ
I , cancels the ln ε
divergence of the soft-photon correction. It may be writtenas the sum of factorizable and nonfactorizable corrections,
dσνle→νleγI ¼ α
πδIdσ
νle→νleLO þ dσνle→νleγ
I;NF : ð102Þ
The factorizable correction δI is obtained from the first,factorizable, terms in Eqs. (61)–(63), evaluating kinemati-cal factors IL; IR; ILR in the kinematics of the elastic 2 → 2
process,
δI ¼2
β
�β −
1
2ln1þ β
1 − β
�ln
2ð1þ βÞεβmð1þ cos δ0Þ
; ð103Þ
where the angle δ0 is given by
cos δ0 ¼ω2 − β2E02 − l20
2βE0l0: ð104Þ
The nonfactorizable part dσνle→νleγI;NF is discussed below.
The bremsstrahlung contribution from region II can beexpressed in factorizable form
dσνle→νleγII ¼ α
πδIIdσ
νle→νleLO ; ð105Þ
where
δII ¼1
β
��1
2þ ln
ρð1þ cos δ0Þ4β
�ln1 − β
1þ β− Li2
1 − β
1þ β
− Li2cos δ0 − 1
cos δ0 þ 1þ Li2
�cos δ0 − 1
cos δ0 þ 1
1þ β
1 − β
�þ π2
6
�
þ ln1 − β cos δ0
ρ− 1: ð106Þ
Consequently, the complete electron energy spectrum isgiven by
dσνle→νleγLO þ dσνle→νle
NLO
¼h1þ α
πðδv þ δs þ δI þ δIIÞ
idσνle→νle
LO
þ dσνle→νlev þ dσνle→νle
dyn þ dσνle→νleγNF ð107Þ
and does not depend on the unphysical parameters ε and λ.We remark that although individual corrections containdouble logarithms, i.e.,
δv ∼β→1
−1
8ln2ð1 − βÞ; δs ∼
β→1−1
4ln2ð1 − βÞ;
δII ∼β→1
3
8ln2ð1 − βÞ; ð108Þ
the complete cross-section correction is free from suchSudakov double logarithms [118,119]. In Appendix H,
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-14
we obtain the remaining nonfactorizable piece dσνle→νleγNF
from the region of hard photons (kγ ≥ ε), which containsdσνle→νleγ
I;NF as well as the contribution beyond the firstfactorizable terms in Eqs. (61)–(63), integrating the elec-tron angle and electron energy distribution over the variablef (equivalent to the electron scattering angle θe), andretaining all electron mass terms.The resulting correction to the electron energy spectrum
reproduces the result of Ref. [25] in the limit m → 0;E0=ω ¼ const. Besides the closed fermion loop contribu-tion of Secs. III B and III C, it is represented by thefollowing substitutions in Eqs. (58) and (59):
IL⟶ω≫m
π2
ωf−
�E0
ω
�dE0; ð109Þ
IR⟶ω≫m
π2
ω
�1 −
E0
ω
�2
fþ
�E0
ω
�dE0; ð110Þ
ILR⟶ω≫m−π2
ω
mω
E0
ωf−þ
�E0
ω
�dE0; ð111Þ
with functions f−ðxÞ, fþðxÞ [25], and f−þðxÞ derived first inthe present paper6,
f−ðxÞ ¼ −2
3ln2ω
mþ�ln1 − xffiffiffi
xp þ x
2þ 1
4
�ln2ω
m−1
2
�Li2ðxÞ −
π2
6
�þ x2
24−11x12
−47
36
−1
2ln2
1 − xx
−�x2þ 23
12
�lnð1 − xÞ þ x ln x; ð112Þ
ð1 − xÞ2fþðxÞ ¼ −2
3ð1 − xÞ2 ln 2ω
mþ�x − 1
2þ ð1 − xÞ2 lnð1 − xÞ
�ln2ω
m−ð1 − xÞ2
2ln1 − xx2
lnð1 − xÞ
þ�ð1 − xÞx − 1
2
��Li2ðxÞ þ ln
2ωxm
ln x −π2
6
�þ�x2 þ x
2−3
4
�ln x
−31 − 49x
72ð1 − xÞ þ 1 − x
3
�5x −
7
2
�lnð1 − xÞ; ð113Þ
−xf−þðxÞ ¼ 2þ 2 ln xþ�x − ln x −
1
2
�ln2ωxm
þ�3
2xþ 1
2− x ln
2ωxm
�ln1 − xx
þ 1
2xln2ð1 − xÞ
þ ðx − 1Þ�Li2ðxÞ −
π2
6þ 5
4
�: ð114Þ
We observe that in exactly forward kinematics at electronthreshold, when E0 ¼ m, the energy spectrum is given bythe nonfactorizable contribution from the electromagneticvertex and closed fermion loops,
dσνle→νleγLO þ dσνle→νle
NLO ⟶E0→m
dσνle→νleNLO → dσνle→νle
LO
þ dσνle→νlev þ dσνle→νle
dyn ; ð115Þwith f2ð0Þ ¼ 1=2 in Eqs. (35), (37), (38) and Πð0; mfÞ,Πð3Þ
γγ ð0Þ, Πð3Þ3γ ð0Þ of Eqs. (42), (47). This equation provides a
universal limit for electron energy and electromagneticenergy spectra.The electron energy spectrum has the following
logarithmically divergent behavior near its end pointE0 ≤ E0
0 ¼ mþ 2ω2
mþ2ω :
dσνle→νleγLO þ dσνle→νle
NLO
dσνle→νleLO
≈−α
π
2
β
�β−
1
2ln1þ β
1− β
�lnE00 −E0
m;
ð116Þas determined by infrared logarithms in Eqs. (36) and (96).
I. Electromagnetic energy spectrum
We evaluate the bremsstrahlung cross section withrespect to the sum of electron and photon energies con-sidering the final neutrino energy spectrum instead of theelectron energy spectrum [22]; see Sec. IV D for explan-ations. For the neutrino scattering angle distributions, weintroduce the four-vector l,
l ¼ kþ p − k0 ¼ ðl0; fÞ; ð117Þwith the laboratory frame values,
l0 ¼ EEM; ð118Þ
f2 ¼ j fj2 ¼ ω2 þ ω02 − 2ωω0 cos θν: ð119Þ
Note the difference between the neutrino scattering angle inthe elastic process [Θν of Eq. (3)] and in the scattering withradiation (θν).Below the end point of maximal electron energy,
EEM ≤ E00 ¼ mþ 2ω2
mþ2ω, we can use the same integration
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-15
technique as in Ref. [22]. Above the end point, the photonenergy is bounded from below kγ ≥ EEM − E0
0, and there isno corresponding elastic process as well as no contributionfrom the soft region. We consider these two regionsseparately in the following.
1. Below electron end point: EEM ≤ E00 =m+ 2ω2
m+ 2ω
The contribution from the soft-photon region kγ ≤ ε isgiven by Eqs. (95) and (96). We split the integration regionwith kγ ≥ ε for factorizable terms in Eqs. (61)–(63) intotwo regions similar to Sec. IV H. In region I: l2 −m2 ¼l20 − f2 −m2 ≥ 2εðl0 þ fÞ, there are no restrictions on thephase space. In region II: l2 −m2 ≤ 2εðl0 þ fÞ, the phasespace of the final neutrino is restricted to
cos γ ≥1
f
�l0 −
l2 −m2
2ε
�; ð120Þ
where γ is the angle between f and kγ. The correction factorfrom region II, δII [cf. Eq. (105)], is given by
δII ¼ −1
β
�β −
1
2ln1þ β
1 − β
�ln1þ β
1 − β: ð121Þ
Here β is expressed in terms of electromagnetic energy as inEq. (97). As for the electron energy spectrum, the brems-strahlung contribution from region I may be written as thesum of factorizable and nonfactorizable corrections; cf.Eq. (102). The factorizable correction δI is obtained fromthe first factorizable terms in Eqs. (61)–(63), evaluatingkinematical factors IL, IR, ILR in the kinematics of the elastic2 → 2 process,
δI ¼2
β
�β −
1
2ln1þ β
1 − β
�ln
ε
m: ð122Þ
In Appendix I we evaluate the remaining nonfactorizablepiece dσνle→νleγ
NF of the electromagnetic energy spectrumbelow the electron end point, performing straightforwardintegrations and keeping all electron mass terms. Itaccounts for the region of hard photons (kγ ≥ ε) andcontains dσνle→νleγ
I;NF as well as the contribution beyondthe first factorizable terms in Eqs. (61)–(63).The resulting correction to the electromagnetic energy
spectrum reproduces the result of Refs. [27,28] in the limitm → 0, EEM=ω ¼ const. Besides the closed fermion loopcontribution of Secs. III B and III C, it is represented by thefollowing substitutions in Eqs. (58) and (59):
IL⟶ω≫m
π2
ωfL
�EEM
ω
�dEEM; ð123Þ
IR⟶ω≫m
π2
ω
�1 −
EEM
ω
�2
fR
�EEM
ω
�dEEM; ð124Þ
ILR⟶ω≫m−π2
ω
mω
EEM
ωfLR
�EEM
ω
�dEEM; ð125Þ
with functions fLðxÞ; fRðxÞ [27,28], and fLRðxÞ derived firstin the present work6,
fLðxÞ ¼3x2 − 30xþ 23
72−2
3ln2ωxm
−π2
6; ð126Þ
fRðxÞ ¼−4x2 − 16xþ 23
72ð1 − xÞ2 −2
3ln2ωxm
−π2
6; ð127Þ
fLRðxÞ ¼x2 þ 3x − 3
4x2−3
2ln2ωxm
−π2
6: ð128Þ
In exactly forward kinematics at electromagnetic energythreshold when EEM ¼ m, the electromagnetic energyspectrum coincides with the electron energy spectrum;see Eq. (115).Just below electron end point (EEM < E0
0 ¼ mþ2ω2
mþ2ω ≈ ω), the electromagnetic energy spectrum, besidesthe closed fermion loop contribution, is given by thefollowing substitutions in the nonfactorizable correction6:
IL⟶ω≫m
−π2
3
�ln4ω2
m2þ π2
2þ 1
6
�dEEM
ω; ð129Þ
IR⟶ω≫m
π2
24
dEEM
ω; ð130Þ
ILR⟶ω≫m
π2
4
mω
�3 ln
4ω2
m2þ 2π2
3− 1
�dEEM
ω: ð131Þ
Equations (129) and (130) are in agreement with the similarlimit taken from the result of Refs. [27,28].
2. Above electron end point: EEM > E00 =m+ 2ω2
m+ 2ω
Above the electron end point energy, the correspondingelastic process is kinematically forbidden. For ω ≫ m, thisregion is relatively small but finite,
EEM − E00 ≤
1
1þ m2ω
m2<
m2: ð132Þ
Since the photon energy is bounded from below in thisregion, kγ > EEM − E0
0, the calculation does not require IRregularization. We present the electromagnetic energyspectrum above the electron end point keeping all electronmass terms in Appendix J.The electromagnetic energy spectrum has the following
logarithmically divergent behavior just above the electronend point EEM > E0
0 ¼ mþ 2ω2
mþ2ω:
dσνle→νleγLO
dσνle→νleLO
≈α
π
2
β
�β −
1
2ln1þ β
1 − β
�lnEEM − E0
0
m: ð133Þ
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-16
J. Absolute cross section
The resulting total cross-section correction, besidesclosed fermion loop contributions, in the ultrarelativisticlimit is given by the following substitutions in Eqs. (58) and(59) for IL, IR [25], and ILR derived first in the present paper6:
IL⟶ω≫m
π2
24
�19 − 4π2 − 16 ln
2ω
m
�; ð134Þ
IR⟶ω≫m
π2
72
�19 − 4π2 − 16 ln
2ω
m
�þ π2
3; ð135Þ
ILR⟶ω≫m−π2
24
mω
�15 − 2π2 − 36 ln
2ω
m
�: ð136Þ
Factors IL and IR of Eqs. (134) and (135) can be obtainedintegrating Eqs. (126) and (127) or Eqs. (112) and (113)over the energy variable. To evaluate the factor ILR, one hasto regulate the logarithmic mass singularity properly or takethe limit from the general expression of Appendix K. Notethe absence of double logarithms in the resulting cross-section correction in Eqs. (126)–(128) and (134)–(136),although individual corrections contain them; cf. Eq. (108).Note also that the total elastic cross section at leading orderis given by the following substitutions in Eqs. (14)and (15):
Zdω0IL⟶
ω≫mω;
Zdω0IR⟶
ω≫m
ω
3;
Zdω0ILR⟶ω≫m
−m2:
ð137ÞResults for the absolute cross section including the electronmass dependence are presented in Appendix K.
V. ILLUSTRATIVE RESULTS
Our results may be used to compute absolute anddifferential cross sections for neutrino-electron scatteringover a broad range of energies and experimental setups. Wefocus on the application to flux normalization at accel-erator-based neutrino experiments in Secs. VA through VC and discuss radiative corrections in the context of newphysics searches in Sec. V D.
A. Total cross section: Energy dependenceand error analysis
The total cross sections for νμe; νee; νμe, and νeescattering are shown in Fig. 8. For ω ≫ m, cross sectionsgrow approximately linearly with neutrino beam energy. Asa benchmark point, we determine at ω ¼ 1 GeV
FIG. 8. Total cross section in the (anti)neutrino-electron scattering processes νμe → νμeðXγÞ, νee → νeeðXγÞ, νμe → νμeðXγÞ, andνee → νeeðXγÞ as a function of (anti)neutrino beam energy ω.
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-17
The cross section is evaluated using four-flavor QCD,with running QED and QCD couplings αðμÞ and αsðμÞevaluated using two and five loop running, respectively,with αð2 GeVÞ ¼ 1=133.309 and αsð2 GeVÞ ¼ 0.3068.The uncertainties in Eq. (138) are from the following: (i)
the hadronic parameter Πð3Þ3γ ð0Þ=Πð3Þ
γγ ð0Þ in Eq. (54) and
from Πð3Þγγ ð0Þ in Eq. (53)7; (ii) from uncertainties in the
four-fermion operator coefficients cνll0
L , cR in Table I; and(iii) from higher-order perturbative corrections, estimatedby varying renormalization scale μ20=2 < μ2 < 2μ20, whereμ0 ¼ 2 GeV. For simplicity, we evaluate the light-quarkcontribution of Eq. (47) neglecting NLO electroweakcorrections and renormalization group corrections to thefour-fermion operator coefficients, taking for definitenessGF ¼ 1.166378 7 × 10−5 GeV−2 and sin2 θW ¼ 0.23112in Eqs. (47) and (48); it is straightforward to includethese corrections, whose impact is given by the fewpermille shift in the coefficients [47], times the ∼1%fractional contribution of light quarks to the crosssection. The charm-quark contribution in Eq. (42) isevaluated including the OðαsÞ and Oðα2sÞ correctionsfrom Appendix A and using the MS mass mcð2GeVÞ¼1.096GeV [corresponding to mcðmcÞ ¼ 1.28ð2Þ GeV[120] ]. The fractional uncertainty coming from the charmquark mass error is ≈1–2 × 10−5 and is not displayed inEq. (138), nor is the uncertainty of a similar magnitudecoming from higher orders in GF expansion. The e-, μ-,and τ-lepton contributions in Eq. (42) are evaluated usinglepton pole masses and the complete kinematic depend-ence of Πðq2; mlÞ in Eq. (41).8
For ω ≫ m, the relative cross-section error is approx-imately constant, independent of neutrino energy. Relativeuncertainties on total cross sections from different sourcesare summarized in Table II. The dominant uncertainty fromthe light-quark contribution in differential and absolutecross sections can be expressed as9
δ
�dσνle→νle
uds
dE0
�≈ η
GFmffiffiffi2
pπ
α
πΠð3Þ
γγ ð0Þ
×
����cνleL IL þ cRIR þ cνleL þ cR2
ILR
����; ð140Þ
δσνle→νleuds ≈ η
GFmωffiffiffi2
pπ
α
πΠð3Þ
γγ ð0Þ�2ωcνleL
mþ 2ω
þ�1 −
m3
ðmþ 2ωÞ3�cR3−mωðcνleL þ cRÞðmþ 2ωÞ2
�;
ð141Þ
with the relative uncertainty η ¼ ðΠð3Þ3γ ð0Þ=Πð3Þ
γγ ð0Þ − 1.0Þ ≈0.2 and the substitution cνleL ↔ cR in the case of antineu-trino scattering.To illustrate the impact of radiative corrections on the
total cross section, Eq. (138) may be compared to theleading-order result of our calculation at scale μ ¼ 2 GeVand ω ¼ 1 GeV:
Radiative corrections change the total cross section by1.7%. We turn now to a discussion of the energy depend-ence of the radiative corrections.
B. Electron and total electromagnetic energy spectra
Figures 9 and 10 display the typical size of the radiativecorrections to energy spectra with respect to the finalelectron energy (E ¼ E0) and with respect to the totalelectromagnetic energy (i.e., the electron energy plusphoton energy, E ¼ E0 þ kγ). We consider muon typeneutrinos and antineutrinos, the primary component inthe accelerator neutrino beam. In these figures, we showthe quantity δ representing the radiative correction normal-ized to the leading-order elastic cross section:
TABLE II. Relative errors of the total neutrino-electron scatter-ing cross section.
7The error of Πð3Þγγ ð0Þ in Eq. (53) contributes �0.00006.
8One can safely evaluate a τ-lepton contribution considering Πð0; mτÞ since jq2j ≪ m2τ .
9It can be seen [cf. Eq. (140)] that the muon antineutrino-electron scattering cross section is free from hadronic uncertainty, and alsoeffective coupling uncertainty induced by cR, at the particular recoil antineutrino energy ω:
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-18
δ ¼ dσνle→νleγLO þ dσνle→νle
NLO − dσνle→νleLO
dσνle→νleLO
: ð143Þ
The correction to the electromagnetic energy spectrum isrelatively flat over a wide energy, whereas the correction tothe electron energy spectrum is logarithmically divergentbelow the electron end point; cf. Eq. (116). The logarithmicdivergence of the electromagnetic energy spectrum abovethe electron end point [cf. Eq. (133)] is not seen in Fig. 9due to the small size of the region in Sec. IV I 2 comparedto the scale of the figure. Both corrections start from thelimit of Eq. (115) at E ¼ m. Note that the correction δdepends on the renormalization scale μ since the numeratordoes not contain the leading-order elastic process, ratherjust the virtual correction to it, leaving the scale dependenceof the closed fermion loops (Secs. III B and III C) withoutcancellations. The large renormalization scale dependencein Figs. 9 and 10 illustrates the cancellations occurringbetween LO and NLO in arriving at the total cross sectionin Eq. (138). Other uncertainties are not shown in thefigure.
C. Electron angular spectrum
In this section, we consider the angular smearing ofdifferential cross sections. It can be presented as a functionof the variable X,
X ¼ 2m
�1 −
Eω
�; ð144Þ
which becomes X ≈ E0θ2e for (anti)neutrinos of high energyin the case of the electron energy spectrum. We present theresulting NLO spectrum in Figs. 11 and 12 for two (anti)neutrino beam energies: ω ¼ 1 GeV and 10 GeV. Althoughthe electromagnetic and electron energy spectra integrate tothe same total cross section, shape effects induced byradiative corrections can potentially impact the calibrationof neutrino flux. For example, experimental cuts requiringa minimum observed energy will result in different num-bers of accepted events depending on which distribution(electromagnetic or electron energy) is chosen. In a practicalanalysis, neither the electron spectrum nor the electromag-netic spectrum will perfectly represent the experimental
e, = 1 GeV
e, = 10 GeV
EM, = 1 GeV
EM, = 10 GeV
e e( ), %
5
0
E/0 0.2 0.4 0.6 0.8 1.0
FIG. 9. Radiative corrections to the neutrino-electron scatteringprocess νμe → νμeðXγÞ for two neutrino beam energies ω ¼ 1,10 GeV. The quantity δ is defined in Eq. (143) and stronglydepends on the MS scale μ. Three curves for μ ¼ μ0=
ffiffiffi2
p, μ ¼ μ0,
and μ ¼ ffiffiffi2
pμ0 with μ0 ¼ 2 GeV are presented. The solid and
dash-dotted curves correspond with electron spectrum, i.e.,E ¼ E0, dashed curves with electromagnetic spectrum, i.e.,E ¼ E0 þ kγ . Uncertainties are not shown on this plot with ascale-dependent quantity. Lower curves correspond to a largervalue of μ.
e, = 1 GeV
e, = 10 GeV
EM, = 1 GeV
EM, = 10 GeV
e e( )
, %
5
0
E/0 0.2 0.4 0.6 0.8 1.0
FIG. 10. Same as Fig. 9 for antineutrino-electron scatteringprocess νμe → νμeðXγÞ. Uncertainties are not shown on this plotwith a scale-dependent quantity. Lower curves correspond to alarge value of μ for E=ω≲ 0.07–0.1 and to a smaller valueof μ above.
e, = 1 GeV
e, = 10 GeV
EM, = 1 GeV
EM, = 10 GeV
e e( )
1042
d/ d
E, c
m2 G
eV-1
1.0
1.5
2.0
2.5
103 X, GeV0 0.2 0.4 0.6 0.8 1.0
FIG. 11. Energy spectrum in the neutrino-electron scatteringνμe → νμeðγÞ, plotted as a function of X ¼ 2mð1 − E=ωÞ fortwo neutrino beam energies ω ¼ 1, 10 GeV. The solid anddash-dotted curves correspond with the electron spectrum, i.e.,E ¼ E0, dashed curves with the electromagnetic spectrum, i.e.,E ¼ E0 þ kγ .
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-19
conditions, and the more general distributions presentedelsewhere in this paper can be used.Results comparing E0 and EEM distributions after aver-
aging over typical experimental flux profiles are collectedin Appendix L.
D. New physics considerations
In this section, we consider the impact of radiativecorrections on the dynamical zero (22) and isolate thedependence of the neutrino-electron scattering cross sec-tion on effective neutrino charge radii. Both effects arepresent in the Standard Model but may also be used tosearch for or constrain new physics.Recall that an amplitude cancellation causes the tree
level electron energy spectrum to vanish at the end point ofthe maximal electron energy, E0
0 ¼ mþ 2ω2
mþ2ω, when elec-tron antineutrinos of a particular energy ω scatter onelectrons; cf. Eq. (22). This feature could have implicationsfor novel neutrino oscillation experiments (see, e.g.,Refs. [94,95]), and it is thus interesting to determine theimpact of radiative corrections on the cancellation. Toinvestigate this question, it is convenient to represent thecross section near the end point as the factorized product ofsoft and hard functions [121], dσ ∼ SH. The soft functionaccounts for infrared divergences and real photon emission.Using the explicit forms for virtual corrections fromSec. III, the hard function through first order in α takesthe form
H ∝�ω
ω− 1
��ω
ω− 1þOðαÞ
�: ð145Þ
For ω=ω − 1 ¼ OðαÞ, the cross section including radiativecorrections is suppressed by Oðα2Þ. The electromagneticenergy spectrum is equal to the electron energy spectrum attree level and vanishes at the same kinematic point.However, radiative corrections now receive a contribution
from “hard” real photon emission, and the electromagneticspectrum in the vicinity of ω ¼ ω and EEM ¼ E0
0 isnonvanishing at first order in α. For general ω ≠ ω, theelectromagnetic energy spectrum vanishes at the end pointEEM ¼ mþ ω and is discontinuous at EEM ¼ E0
0; at ω ¼ ωthe discontinuity is replaced by a kink.Neutrino charge radii [122–125] may be systematically
defined and computed with low-energy effective fieldtheory [47], where new physics contributions are repre-sented as10
δcνleL ¼ δcνleR ¼ e2
6δr2νl : ð146Þ
The impact on neutrino-electron scattering is given by
δ
�dσνle→νle
dE
�¼ mα
3δr2νl
����cνleL IL þ cRIR þ cνleL þ cR2
ILR
����;ð147Þ
δσνle→νle ¼ mωα
3δr2νl
�2ωcνleL
mþ 2ωþ�1 −
m3
ðmþ 2ωÞ3�cR3
−mωðcνleL þ cRÞðmþ 2ωÞ2
�; ð148Þ
with the substitution cνleL ↔ cR in the case of antineutrinoscattering.
VI. CONCLUSIONS AND OUTLOOK
In this work, we have presented analytical results forelastic (anti)neutrino-electron scattering starting from four-fermion effective field theory. Total cross sections, theelectron and electromagnetic energy spectra, as well asdouble- and triple-differential cross sections were presentedin a relatively compact form. Our results can be applied toimprove constraints of neutrino flux measurements viaelastic neutrino-electron scattering. All expressions wereobtained for finite electron mass and can also be used inlow-energy applications such as oscillation measurementswith solar and reactor (anti)neutrinos.Next-to-leading order corrections with bremsstrahlung
of one photon are typically of order few percent and dependon the experimental setup. For instance, as discussed inSec. V C, electron and electromagnetic energy spectradiffer significantly. Although these two spectra integrateto the same total cross section, kinematical cuts can alter
e, = 1 GeV
e, = 10 GeV
EM, = 1 GeV
EM, = 10 GeV
e e( )
1042
d/ d
E, c
m2 G
eV-1
1.0
1.5
2.0
2.5
103 X, GeV0 0.2 0.4 0.6 0.8 1.0
FIG. 12. Same as Fig. 11 for antineutrino-electron scatteringprocess νμe → νμeðXγÞ.
10In terms of weak scale matching coefficients, this corre-sponds to a contribution to the neutrino-photon coupling inRef. [47], δcνlγ ¼ ðe2=6Þδr2νl . The “charge radius” as a low-energy observable quantity is unambiguously defined in terms offour-Fermi coefficients in Ref. [47]. For a diagrammatic formu-lation of neutrino charge radii in the Standard Model seeRef. [126] and references therein.
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-20
inferred flux constraints if radiative corrections are notmatched correctly to experimental conditions. Future pre-cise measurements of the electron angular spectrum inneutrino-electron scattering can provide energy-dependentneutrino flux constraints. Our results provide a completedescription of the kinematic dependence of radiativecorrections needed to control uncertainties in neutrinoenergy reconstruction. We have discussed the impact ofradiative corrections on cross sections and energy distri-butions in searches for physics beyond the Standard Modelin Sec. V D.We provided a complete error budget for neutrino-
electron scattering observables. The light-quark contribu-tion to the radiative correction is the dominant source ofuncertainty. We have expressed this contribution in termsof well-defined Standard Model observables, independentof “constituent quark” models used in previous treatments,and determined the relevant hadronic parameter, denoted
Πð3Þ3γ ð0Þ, using SUð3Þf symmetry to relate it to the exper-
imentally constrained parameter Πð3Þγγ ð0Þ. To further pin
down the uncertainty of this light-quark contribution, onecan evaluate a closed fermion loop contribution within thedispersion relation approach decomposing eþe− cross-section data and measurements of hadronic τ decays intoflavor components [108,109,112,127–129] or perform acalculation in lattice QCD [130,131].We note that due to the restrictive kinematics of neutrino-
electron scattering (jq2j < 2mω for the elastic process) thelight-quark contribution enters as a single constant, repre-senting the q2 → 0 limit of the relevant hadronic tensor.This single constant will also impact (and may be con-strained by) other low q2 processes such as coherentneutrino-nucleus scattering.Besides its phenomenological relevance, the neutrino-
electron scattering process provides an analytically calculableprototype for the more complicated case of neutrino-nucleusscattering [132]. In general, radiative corrections can bedecomposed (“factorized”) into soft and hard functions usingeffective field theory [121].11 The soft functions depend onexperimental configuration but are independent of hadronicphysics and describe universal large logarithms that are
present in general kinematics. The hard functions are inde-pendent of experimental configuration and describe hadronicphysics. In neutrino-electron scattering the analogous hardfunctions are perturbatively calculable, whereas in neutrino-nucleus scattering they must be parametrized and experimen-tally constrained.
ACKNOWLEDGMENTS
WethankK.McFarland for useful discussions.O. T. thanksMatthias Heller for useful discussions regarding radiativecorrections in QED. O. T. acknowledges the Fermilab theorygroup for warm hospitality and support. The work of O. T. issupported by the Visiting Scholars Award Program of theUniversities Research Association. The work is supported bythe U.S. Department of Energy, Office of Science, Office ofHigh Energy Physics, under Award No. DE-SC0019095 andby the Deutsche Forschungsgemeinschaft DFG through theCollaborative Research Center [The Low-Energy Frontier ofthe Standard Model (SFB 1044)]. Fermilab is operated byFermiResearchAlliance,LLCunderContractNo.DE-AC02-07CH11359 with the United States Department of Energy.FEYNCALC [133,134], LOOPTOOLS [135], JAXODRAW [136],Mathematica [137], andDATAGRAPHwere extremelyuseful inthis work.
APPENDIX A: QCD CORRECTION TO QEDVACUUM POLARIZATION
For quark loop contributions in Sec. III B, we includethe leading QCD correction due to one exchanged gluoninside the quark loop. This correction modifies the formfactor Π in Eq. (41) as Π → Πþ ΠQCD with ΠQCD fromRefs. [104–107]12:
ΠQCD ¼ αs3π
�ln
μ2
m2f
− 4ζð3Þ þ 55
12þ 4m2
f
q2V1
�q2
4m2f
��;
ðA1Þwhere αs is a strong coupling constant, ζðsÞ denotes theRiemann zeta functions, and the function VðrÞ is given by(for spacelike momentum transfer, r < 0)
VðrÞ ¼ffiffiffiffiffiffiffiffiffiffiffi1 −
1
r
r �8
3
�rþ 1
2
��Li2ðr2−Þ − Li2ðr4−Þ þ ln
−64ð1 − rÞ2rr3þ
ln rþ
�− 2
�rþ 3
2
�ln rþ
�
þ 4
�r −
1
4r
��2Li3ðr2−Þ − Li3ðr4−Þ þ
8
3ðLi2ðr2−Þ − Li2ðr4−ÞÞ ln rþ
�þ 13
6þ ζð3Þ
r
þ 16
3
�r −
1
4r
�ln8ð1 − rÞ ffiffiffiffiffiffi
−rp
r3þln2rþ − 8
�r −
1
6−
7
48r
�ln2rþ; ðA2Þ
12Note that the color factor applies as NcðΠþ ΠQCDÞ.11An application of this formalism to the discussion of the dynamical zero in νee scattering was described in Sec. (V D).
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-21
with notations r� ¼ ffiffiffiffiffiffiffiffiffiffi1 − r
p � r. As discussed at the end ofSec. III B, the relevant limit for neutrino-electron scatteringis q2 → −0, corresponding with
ΠQCDjq2→−0 ¼αs3π
�ln
μ2
m2f
þ 15
4
�: ðA3Þ
For practical evaluation of a c-quark contribution, wetake the well-convergent expression in terms of MS quarkmass mc from Refs. [110,138–140],
Π ¼ 1
3ln
μ2
m2cþ αs3π
�− ln
μ2
m2cþ 13
12
�
þ α2s3π2
�655
144ζð3Þ − 3847
864−5
6ln
μ2
m2c−11
8ln2
μ2
m2c
þ nf
�361
1296−
1
18ln
μ2
m2cþ 1
12ln2
μ2
m2c
��; ðA4Þ
where nf ¼ 4 denotes the number of active quarks. Thecorrection of order α2s in Eq. (A4) does not change ourresults within significant digits.
APPENDIX B: TRIPLE-DIFFERENTIALDISTRIBUTION
We evaluate the bremsstrahlung cross section followingRef. [22]. For the electron angle distributions, we introducethe four-vector l,
l ¼ kþ p − p0 ¼ ðl0; fÞ; ðB1Þ
with the laboratory frame values,
l0 ¼ mþ ω − E0; ðB2Þ
f2 ¼ jfj2 ¼ ω2 þ β2E02 − 2ωβE0 cos θe: ðB3Þ
Note the difference between the electron scattering angle inthe elastic process [Θe of Eq. (6)] and in the scattering withradiation (θe).The triple-differential cross section with respect to the
electron angle, electron energy, and photon energy is givenby the following substitutions in Eqs. (58) and (59):
The physical region of variables corresponding to theradiation of hard photons with energy kγ ≥ ε (ε ≪ m, ω),is the following (see Sec. IV H for a description of hard-and soft-photon regions):
mþ 2ε2
m − 2ε≤ E0 ≤ mþ 2ðω − εÞ2
mþ 2ðω − εÞ ; ðB10Þ
jω − jp0jj ≤ f ≤ l0 − 2ε; ðB11Þ
l0 − f2
≤ kγ ≤l0 þ f2
: ðB12Þ
We keep the exact dependence on the unphysicalparameter ε which is important in the evaluation of theelectron energy spectrum in Sec. IV H. Our integrationregion in Eqs. (B10)–(B12) corresponds to region I inSec. IV H.
APPENDIX C: DOUBLE-DIFFERENTIALDISTRIBUTION IN ELECTRON ENERGY AND
ELECTRON ANGLE
Integrating Eqs. (B4)–(B6) over the photon energy kγ,we obtain the double-differential cross section with respectto the recoil electron energy and angle. The result is givenby the following substitutions in Eqs. (58) and (59):
Ii →π2mω3
�ai þ bi ln
1þ β
1 − βþ ci ln
l0 þ fl0 − f
þ di lnl0 − βf cos δ − ffiffiffi
gp
l0 − βf cos δþ ffiffiffig
p�dfdE0 −
π2
ω
2
β
�β −
1
2ln1þ β
1 − β
�Iidf2
l2dE0; ðC1Þ
with g ¼ ðf cos δ − βl0Þ2 þ ρ2f2 sin2 δ and the angle δ between vectors l and p0,
cos δ ¼ ω2 − β2E02 − f2
2βE0f: ðC2Þ
Kinematical factors IL; IR; ILR in Eq. (C1) correspond to the 2 → 2 process.The coefficients in integrals Ii are given by
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-23
cR ¼ −l2ðl2 − 4mðl0 − m
2ÞÞ
8m4−l0ðl0 −mÞ
m2;
dR ¼ βfl2ðð2l0ρ −mð−2ρ2 þ ρþ 1ÞÞðβl0 − f cos δÞ − 12βρl2Þ
4g3=2m2−ρ2fl0ðfðf − βl0 cos δÞ þ 1
2β2l2Þ
g3=2m
−ρfðl2ðl2 − 4l0mþ 2m2Þ þ 8l0m2ðω − 2E0ÞÞ
8ffiffiffig
pm4
þ 3β2ρ3f3l2sin2δðl2 þ 4E0ðl0 − βf cos δÞÞ16g5=2m2
;
aLR ¼ βfðβl0 − f cos δÞg
−β cos δ
ρ;
bLR ¼ −ρfl2
2βm3;
cLR ¼ βl0 cos δ2ρf
−l2 −mðmþ E0Þ
2m2;
dLR ¼ ρfððmþ 2E0Þm − l2Þ2m2 ffiffiffi
gp −
ρ2f2ðf − βl0 cos δÞ2g3=2
:
APPENDIX D: DOUBLE-DIFFERENTIALDISTRIBUTION IN ELECTROMAGNETIC
ENERGY AND ELECTRON ANGLE
To obtain neutrino energy (equivalently, electromagneticenergy) and electron angle distribution, Eqs. (B4)–(B6) canbe integrated over the electron energy, exploiting theenergy conservation: kγ ¼ mþ ω − ω0 − E0. The integra-tion measure of Eq. (B9) is replaced as
Dm ¼ π2
m2ω2df
dkγkγ
dE0
ω→
π2
m2ω2d cos θe
dω0
kγ
βE0dE0
f:
ðD1ÞThe physical integration region is contained in
0 ≤ ω0 ≤ ω; ðD2Þ
0 ≤ cos θe ≤ 1; ðD3Þ
m ≤ E0 ≤ mðmþ ωÞ2 þ ω2cos2θeðmþ ωÞ2 − ω2cos2θe
; ðD4Þ
which is actually larger than the physical region. Theextraneous regions I and II are above the electron end point(EEM ≥ E0
The presentation here in terms of a larger region (D2)–(D4) and subtractions (D5)–(D10) is designed as a simpledescription of the actual physical region. In practice, onemay perform the integration over this larger region and usesubtractions above the electron end point EEM ≥ E0
0 ¼mþ 2ω2
mþ2ω; or one may break up the integration region(D2)–(D4) and integrate once only over the physical region.
APPENDIX E: DOUBLE-DIFFERENTIALDISTRIBUTION IN PHOTON ENERGY
AND ELECTRON ENERGY
To obtain the distribution with respect to the photonenergy and electron energy, Eqs. (B4)–(B6) can be inte-grated first over the variable f after the change of theintegration order. The kinematical region of electron energyis bounded as
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-24
m ≤ E0 ≤ mþ 2ω2
mþ 2ω: ðE1Þ
The physical region of f for different values of kγ is thengiven by
0 ≤ kγ ≤l0 − jω − jp0jj
2; l0 − 2kγ ≤ f ≤ l0; ðE2Þ
l0 − jω− jp0jj2
≤ kγ ≤l0 þ jω− jp0j
2; jω− jp0jj ≤ f ≤ l0;
ðE3Þ
l0 þ jω − jp0jj2
≤ kγ ≤ l0; −l0 þ 2kγ ≤ f ≤ l0: ðE4Þ
APPENDIX F: DOUBLE-DIFFERENTIALDISTRIBUTION IN PHOTON ENERGY AND
PHOTON ANGLE
We evaluate the bremsstrahlung cross section withrespect to the photon energy and photon angle consideringthe final photon energy spectrum instead of the electron
spectrum [22]; see Sec. IV D for explanations. For thephoton scattering angle (with respect to the neutrino beamdirection) distributions, we introduce the four-vector l,
l ¼ kþ p − kγ ¼ ðl0; fÞ; ðF1Þ
with the laboratory frame values,
l0 ¼ mþ ω − kγ; ðF2Þ
f2 ¼ j fj2 ¼ ω2 þ k2γ − 2ωkγ cos θγ; ðF3Þ
where θγ denotes the photon scattering angle.The photon energy spectrum accounting for electron
mass terms is given by the following substitutions inEqs. (58) and (59):
Ii →π2
2mω2
�aiðl2 −m2Þ þ bi ln
m2
l2
�fdf
ðl2 − sÞ2 dkγ; ðF4Þ
with s ¼ m2 þ 2mω and coefficients ai and bi in Eq. (F4):
The nonfactorizable contribution to the electron energy spectrum dσνle→νleγNF from Eq. (107) is given by the following
substitutions in Eqs. (58) and (59):
Ii →π2
ω3
�zi þ yi ln
2ωm
−1þ ρ1þβ ð1þ 2ω
m Þþ xi ln
2l0m
1þ 2ωm − 1þβ
ρ
þ ri ln1 − 1þβ
ρ1þβ1−β −
1þβρ ð1þ 2ω
m Þ
�dE0
þ π2
ω3
�qi ln
1þ β
1 − βþ vi
�Li2
1þ β
ρ− Li2
�1þ 2ω
m
�þ Li2
ð1þ 2ωm Þρ
1þ β−π2
6
��dE0: ðH1Þ
Exact expressions for coefficients zi, yi, xi, ri, qi, and vi in Eq. (H1) are given by
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-27
vL ¼ 1
2
�m2
2þ 2mωþ ω2
�; vR ¼ 1
2
�l20 þ
β2 þ ρ
ρ2m2
�; vLR ¼ 1
2mð2l0 −mÞ;
xL ¼ −2
15
ω5
m3þ 1
3
ω3
mþ�1þ 3β2
3ρ3−4β4 − 11β2 þ 7
3ρ4
�ω2 þ
�2
ρ3−β4 − β2 þ 2
ρ4
�mω
þ�−7β4 þ 14β2 − 22
15ρ4þ 15β4 − 25β2 þ 22
15ρ5
�m2;
xR ¼ −l20ð35l0m2 − 10l20mþ 2l30 − 30m3Þ
15m3;
xLR ¼ 3l0m2 − 3l20m − 2l30 þ 3m2ω
3m;
yL ¼ 1
2ωðω −mÞ;
yR ¼−ω4 − 2ð5 − 1
ρÞmω3 þ 12β2þ11ρ−16ρ2
m2ω2 þ 6β2þ9ρ−10ρ2
m3ωþ β2þ2ρ−2ρ2
m4
ðmþ 2ωÞ2 ;
yLR ¼ mE0�1 −
ðmþ 2ωÞ2 −mω
E0ðmþ 2ωÞ�;
rL ¼�−2þ β
3
ρ
ð1þ βÞ2 þ1
6
4þ β
1þ β
�ω2 þ
�β − ρ2
ρð1þ βÞ þ1
2
�1þ 1
ð1þ βÞ2��
mω
þ�−ð17β2 þ 36β þ 22Þρ
30ð1þ βÞ3 þ 14β2 þ 43β þ 44
60ð1þ βÞ2�m2;
rR ¼�−2þ β
3
ρ
ð1þ βÞ2 þ1
6
4þ β
1þ β
�ω02 þ
�β2 − 5β þ 1
3ρð1þ βÞ þ 1
6
7β2 þ 8β − 2
ð1þ βÞ2�mω0
þ�−23β3 þ 14β2 þ 41β − 2
30ρð1þ βÞ2 þ −28βρ2 þ 43β2 þ 2
30ρ2ð1þ βÞ�m2;
rLR ¼ 1
6
�14þ 5β þ 2
2β2 − 4β − 7
ρ
�m2
1þ βþ�1þ 1 − 2ρ
1þ β
�mω;
qL ¼�1
2β
ρ
1þ β−1þ β
2β
�ω2 þ β
2ρmωþ 1 − ρ
2βm2;
qR ¼�1
2β
ρ
1þ β−1þ β
2β
�ω02 þ
�2 −
1
1þ β−2 − β
2ρ
�mω0
þ�4β3 þ β2 − 4β þ 2
4βρ2þ −β3 þ 2β2 þ β − 1
2βρð1þ βÞ�m2;
qLR ¼ ð1 − βÞω2 − 2ρmωþ ð1þ β2Þm2
β
l0 − ω
mþ βmE0;
zL ¼ zω4
ω4 þ zω3
L mω3 þ zω2
L m2ω2 þ zωLm3ωþ z0Lm
4
m2;
zR ¼ 2zω4
ω5 þ zω4
R mω4 þ zω3
R m2ω3 þ zω2
R m3ω2 þ zωRm4ωþ z0Rm
5
m2ðmþ 2ωÞ ;
zLR ¼ 2l0 þ 9m6
�l0 −
ρω
1þ β
�;
zω4 ¼ 1
15−
1
15
ρ
1þ β; z0L ¼ 25β2 − 49
60ρ3
�1 −
1
ρ
�−
8β2
15ρ2;
zω3
L ¼ 3 − β
30ρ−
3þ 2β
30ð1þ βÞ ; zω2
L ¼ 7β2 þ 8β − 23
30ð1þ βÞρ −15β2 þ 6β − 23
30ρ2;
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-28
zωL ¼ −20β3 þ 51β2 þ 38β − 105
60ρ3−55β3 þ 54β2 − 82β − 105
60ρ2ð1þ βÞ ;
zω4
R ¼ −8
15ρþ 1
15
8 − β
1þ β; zω
3
R ¼ 113β2 − 2β − 133
30ð1þ βÞρ −143β2 − 34β − 133
30ρ2;
zω2
R ¼ −339β3 − 805β2 − 353β þ 851
60ρ3þ −760β3 − 825β2 þ 778β þ 851
60ρ2ð1þ βÞ ;
zωR ¼ βðð433 − 45βÞβ þ 44Þ − 439
30ρ3þ βðβð9βð33β þ 3Þ − 730Þ − 29Þ þ 439
30ρ4;
z0R ¼ 270β2 − 269
60ρ3þ 309β4 − 839β2 þ 538
120ρ4;
where l0 ¼ mþ ω − E0 and ω0 ¼ l0. Our result agreesnumerically with Refs. [29,38]. Integrated over the electronenergy, it agrees with the total cross section of Appendix K.
APPENDIX I: ELECTROMAGNETIC ENERGYSPECTRUM BELOW ELECTRON END POINT
For the remaining nonfactorizable contribution to theelectromagnetic energy spectrum dσνle→νleγ
NF , it is conven-ient to express the result as
dσνle→νleγNF ¼ α
πδγdσ
νle→νleLO þ ðdσνle→νleγ
NF Þ0; ðI1Þ
where in the first term the cross section of the elasticprocess is expressed as a function of the final state neutrinoenergy, and
δγ ¼1
2βln1 − β
1þ β
�1þ ln
ρ17=2
4β4ð1 − βÞ9=2�− 1 − 2 ln
1 − ρ
ρ
−1
β
�Li2
−ρ3
ð1þ βÞ3 þ1
2Li2
1 − β
1þ β− Li2
ρ
1þ βþ π2
6
�:
ðI2Þ
As for the electron energy spectrum, individual correctionscontain double logarithms,
δv ∼β→1
−1
8ln2ð1 − βÞ; δs ∼
β→1−1
4ln2ð1 − βÞ;
δII ∼β→1
1
2ln2ð1 − βÞ; δγ ∼
β→1−1
8ln2ð1 − βÞ; ðI3Þ
but the complete electromagnetic energy spectrum is freefrom Sudakov double logarithms [118]. The residual non-factorizable piece of the bremsstrahlung contribution,ðdσνle→νleγ
NF Þ0 is given by the following substitutions inEqs. (58) and (59):
Ii →π2
ω3
�ai þ bi ln
1þ β
1 − βþ ci ln
2 − ρ
1 − β
�dω0; ðI4Þ
where coefficients ai; bi, and ci can be expressed in termsof the initial and final neutrino energies, ω and ω0,respectively, in the following form:
The interference part of the energy spectrum isdetermined by
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-29
aLR ¼�−
ρ
2 − ρ
2ωω0
m2−
1
3ρþ 4
�ω2ILR;
bLR ¼�1þ β
2β
2ωω0
m2−
1
3β2þ 1
3ð1 − βÞ þ7
6−1þ β
ρ
�1
3β2þ 7
6βþ 1
6
��ω2ILR;
cLR ¼�−2ωω0
m2þ β4 − 5β2 þ 2
3β2ρ2þ 2ð1þ 4β2Þ
3β2ρ
�ω2ILR;
where ILR is given by Eq. (20).Our result agrees with the numerical evaluation in Ref. [29].
APPENDIX J: ELECTROMAGNETIC ENERGY SPECTRUM ABOVE ELECTRON END POINT
The electromagnetic energy spectrum above the electron end point can be conveniently expressed as a sum of thefactorizable and nonfactorizable corrections,
dσνle→νleγ ¼ α
πδγdσ
νle→νleLO þ ðdσνle→νleγÞ0: ðJ1Þ
The factorizable part is given by
δγ ¼1
β
�−π2
3þ 7
8ln2
1þ β
1 − βþ 2 ln
�1þ 2ω
m
�ln1þ β
1 − β−3
2ln1þ β
1 − βln2 − ρ
1 − βþ 2Li2
ρ
1þ β
þ ln2 − ρð1þ 2ω
m Þρð1þ 2ω
m Þln
�1þ β
1 − β
1þ β − ρð1þ 2ωm Þ
−1þ β þ ρð1þ 2ωm Þ
�− Li2
ρð1þ 2ωm Þ
1þ β− Li2
2 − ρð1þ 2ωm Þ
1þ β
þ Li22 − ρ
1þ βþℜ
�Li2
ρð1þ 2ωm Þ
1 − βþ Li2
2 − ρð1þ 2ωm Þ
1 − β− Li2
2 − ρ
1 − β
��þ 2 ln
�2 − ρð1þ 2ω
m Þ1 − ρ
2ωω0 þmðω0 − ωÞ−m2
�;
ðJ2Þ
where the elastic cross section dσνle→νleLO is expressed in terms of ω0. The nonfactorizable part is given by the following
where EEM ¼ mþ ω − ω0, and as explained in Sec. II B dσ=dE0 ¼ dσ=dω0. Our result agrees with a numerical evaluation ofRef. [29]. The total cross section from both regions of Secs. IV I 1 and IV I 2 is in agreement with Ref. [31]. Correctingobvious typos, the function IRL and only the function IL of Eq. (J3) with the interchange IL ↔ IR are in agreement withRef. [31]. For all other kinematical factors of Secs. IV I 1 and IV I 2, we find nontrivial discrepancies with Ref. [31].
APPENDIX K: ABSOLUTE CROSS SECTION
The total cross-section correction including both real and virtual contributions, besides the closed fermion loop correctionof Secs. III B and III C, is given by the following substitutions in Eqs. (58) and (59) [31]:
The reduced cross section arising from the quark loopcontributions σνle→νle
dyn;q is obtained replacing νle couplingsin Eqs. (14) and (15) as
Zdω0ðcνll0L Þ2IL → 2
ffiffiffi2
pGFc
νll0L ωð1 − RÞ;
Zdω0c2RIR → 2
ffiffiffi2
pGFcR
ωð1 − R3Þ3
;
Zdω0cνll
0L cRILR → −
ffiffiffi2
pGFðcνll
0L þ cRÞ
ωR2
r: ðK13Þ
APPENDIX L: AVERAGED OVER FLUXNEUTRINO CROSS SECTIONS
In the following, we average the energy spectrum withanticipated flux profiles of the DUNE Near Detector
DUNE
e,
e,
e, e
e, e
e
EM,
EM,
EM, e
EM, e
EM
1042
d/ d
E, c
m2 G
eV-1
0
1.0
2.0
0.5
1.5
2.5
E, GeV1 2 3 4 5 6 7 8 9 10
FIG. 13. Electron (e) and electromagnetic (EM) energy spectra inelastic neutrino-electron scattering for the neutrino beam mode ofDUNE experiments. The electron energy spectrum is above at lowenergy. Electron and muon (anti)neutrino contributions are shown.
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
033006-32
[141,142] at Fermilab. In Figs. 13 and 14, we show theresulting electron and electromagnetic energy spectra forneutrino and antineutrino beam modes.The corresponding figures for MINERvA [19,20,143–
145], NOvA [146], and T2K [147,148] experiments areshown in Figs. 15–20. The difference between the electronand electromagnetic energy spectra slightly washes outafter averaging over the typical neutrino flux. It is larger atlow energies, where it can reach an effect of the relativeorder 1%–3%, and smaller at higher energies reflecting thedependence in Fig. 11.
DUNE
e,
e,
e, e
e, e
e
EM,
EM,
EM, e
EM, e
EM
1042
d/ d
E, c
m2 G
eV-1
0
1.0
2.0
0.5
1.5
2.5
E, GeV1 2 3 4 5 6 7 8 9 10
FIG. 14. Same as Fig. 13 for the antineutrino beam mode.
MINERvA
e,
e,
e, e
e, e
e
EM,
EM,
EM, e
EM, e
EM
1042
d/ d
E, c
m2 G
eV-1
0
1.0
2.0
0.5
1.5
2.5
E, GeV2 4 6 8 10
FIG. 15. Same as Fig. 13 for the MINERvA experiment.
MINERvA
e,
e,
e, e
e, e
e
EM,
EM,
EM, e
EM, e
EM
1042
d/ d
E, c
m2 G
eV-1
0
1.0
2.0
0.5
1.5
2.5
E, GeV2 4 6 8 10
FIG. 16. Same as Fig. 14 for the MINERvA experiment.
NOvA
e,
e,
e, e
e, e
e
EM,
EM,
EM, e
EM, e
EM
1042
d/ d
E, c
m2 G
eV-1
0
1.0
2.0
0.5
1.5
2.5
E, GeV1 2 3 4 5 6 7 8 9 10
FIG. 17. Same as Fig. 13 for the NOvA experiment.
NOvA
e,
e,
e, e
e, e
e
EM,
EM,
EM, e
EM, e
EM
1042
d/ d
E, c
m2 G
eV-1
0
1.0
2.0
0.5
1.5
2.5
E, GeV1 2 3 4 5 6 7 8 9 10
FIG. 18. Same as Fig. 14 for the NOvA experiment.
T2K
e,
e,
e, e
e, e
e
EM,
EM,
EM, e
EM, e
EM
1042
d/ d
E, c
m2 G
eV-1
0
1.0
2.0
0.5
1.5
2.5
E, GeV0.5 1.0 1.5 2.0 2.5 3.0
FIG. 19. Same as Fig. 13 for the T2K experiment.
T2K
e,
e,
e, e
e, e
e
EM,
EM,
EM, e
EM, e
EM
1042
d/ d
E, c
m2 G
eV-1
0
1.0
2.0
0.5
1.5
2.5
E, GeV0.5 1.0 1.5 2.0 2.5 3.0
FIG. 20. Same as Fig. 14 for the T2K experiment.
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
033006-33
[1] B. T. Cleveland, T. Daily, R. Davis, Jr., J. R. Distel, K.Lande, C. K. Lee, P. S. Wildenhain, and J. Ullman,Astrophys. J. 496, 505 (1998).
[2] W. Hampel et al. (GALLEX Collaboration), Phys. Lett. B447, 127 (1999).
[3] Q. R. Ahmad et al. (SNO Collaboration), Phys. Rev. Lett.89, 011301 (2002).
[4] J. N. Abdurashitov et al. (SAGE Collaboration), Zh. Eksp.Teor. Fiz. 122, 211 (2002) [J. Exp. Theor. Phys. 95, 181(2002)].
[5] S. Fukuda et al. (Super-Kamiokande Collaboration), Phys.Rev. Lett. 86, 5651 (2001).
[6] S. N. Ahmed et al. (SNO Collaboration), Phys. Rev. Lett.92, 181301 (2004).
[7] Y. Fukuda et al. (Super-Kamiokande Collaboration), Phys.Rev. Lett. 81, 1562 (1998).
[8] Y. Ashie et al. (Super-Kamiokande Collaboration), Phys.Rev. Lett. 93, 101801 (2004).
[9] K. Eguchi et al. (KamLAND Collaboration), Phys. Rev.Lett. 90, 021802 (2003).
[10] T. Araki et al. (KamLAND Collaboration), Phys. Rev. Lett.94, 081801 (2005).
[11] Y. Abe et al. (Double Chooz Collaboration), Phys. Rev. D86, 052008 (2012).
[12] J. K. Ahn et al. (RENO Collaboration), Phys. Rev. Lett.108, 191802 (2012).
[13] F. P. An et al. (Daya Bay Collaboration), Chin. Phys. C 37,011001 (2013).
[14] M. H. Ahn et al. (K2K Collaboration), Phys. Rev. Lett. 90,041801 (2003).
[15] D. G. Michael et al. (MINOS Collaboration), Phys. Rev.Lett. 97, 191801 (2006).
[16] K. Abe et al. (T2K Collaboration), Phys. Rev. D 88,032002 (2013).
[19] J. Park et al. (MINERvA Collaboration), Phys. Rev. D 93,112007 (2016).
[20] E. Valencia et al. (MINERvACollaboration), Phys. Rev. D100, 092001 (2019).
[21] T. D. Lee and A. Sirlin, Rev. Mod. Phys. 36, 666 (1964).[22] M. Ram, Phys. Rev. 155, 1539 (1967).[23] S. Weinberg, Phys. Rev. Lett. 19, 1264 (1967).[24] G. ’t Hooft, Phys. Lett. 37B, 195 (1971).[25] S. Sarantakos, A. Sirlin, and W. J. Marciano, Nucl. Phys.
B217, 84 (1983).[26] J. N. Bahcall, M. Kamionkowski, and A. Sirlin, Phys. Rev.
D 51, 6146 (1995).[27] D. Yu. Bardin and V. A. Dokuchaeva, Nucl. Phys. B246,
221 (1984).[28] D. Yu. Bardin and V. A. Dokuchaeva, Yad. Fiz. 43, 1513
(1986) [Sov. J. Nucl. Phys. 43, 975 (1986)].[29] M. Passera, Phys. Rev. D 64, 113002 (2001).[30] M. Green, J. Phys. G 7, 1169 (1981).[31] D. Yu. Bardin and V. A. Dokuchaeva, Yad. Fiz. 39, 888
(1984) [Sov. J. Nucl. Phys. 39, 563 (1984)].[32] P. Salomonson and Y. Ueda, Phys. Rev. D 11, 2606 (1975).
[33] E. D. Zhizhin, R. V. Konoplich, and Yu. P. Nikitin, Izv.Vuz. Fiz. 1975, 82 (1975).
[34] N. Byers, R. Ruckl, and A. Yano, Physica (Amsterdam)96A, 163 (1979).
[35] M. Green and M. J. G. Veltman, Nucl. Phys. B169, 137(1980); B175, 547(E) (1980).
[36] W. J. Marciano and A. Sirlin, Phys. Rev. D 22, 2695(1980); 31, 213(E) (1985).
[37] K.-i. Aoki, Z. Hioki, R. Kawabe, M. Konuma, and T. Muta,Prog. Theor. Phys. 65, 1001 (1981).
[38] K.-i. Aoki and Z. Hioki, Prog. Theor. Phys. 66, 2234(1981).
[39] Z. Hioki, Prog. Theor. Phys. 67, 1165 (1982).[40] A. M. Mourao, L. Bento, and B. K. Kerimov, Phys. Lett. B
237, 469 (1990).[41] A. Weber and L. M. Sehgal, Nucl. Phys. B359, 262 (1991).[42] F. Buccella, C. Gualdi, G. Miele, and P. Santorelli, Nuovo
Cimento B 107, 1343 (1992).[43] J. Bernabeu, S. M. Bilenky, F. J. Botella, and J. Segura,
Nucl. Phys. B426, 434 (1994).[44] E. Akhmedov, G. Arcadi, M. Lindner, and S. Vogl, J. High
Energy Phys. 10 (2018) 045.[45] W. J. Marciano and Z. Parsa, J. Phys. G 29, 2629 (2003).[46] A. Sirlin and A. Ferroglia, Rev. Mod. Phys. 85, 263 (2013).[47] R. J. Hill and O. Tomalak, arXiv:1911.01493.[48] D. Akimov et al. (COHERENT Collaboration), Science
357, 1123 (2017).[49] D. S. Ayres et al. (NOvA Collaboration), Fermi National
[50] R. Acciarri et al. (DUNE Collaboration), arXiv:1512.06148.
[51] K. Abe et al., arXiv:1109.3262.[52] K. Abe et al. (Super-Kamiokande Collaboration), Phys.
Rev. D 83, 052010 (2011).[53] M. Wurm et al. (LENA Collaboration), Astropart. Phys.
35, 685 (2012).[54] Y.-F. Li, J. Cao, Y. Wang, and L. Zhan, Phys. Rev. D 88,
013008 (2013).[55] C. Giunti and A. Studenikin, Rev. Mod. Phys. 87, 531
(2015).[56] C. Giunti, K. A. Kouzakov, Y.-F. Li, A. V. Lokhov, A. I.
Studenikin, and S. Zhou, Ann. Phys. (Berlin) 528, 198(2016).
[57] E. Fermi, Z. Phys. 88, 161 (1934).[58] R. P. Feynman and M. Gell-Mann, Phys. Rev. 109, 193
(1958).[59] E. C. G. Sudarshan and R. E. Marshak, Phys. Rev. 109,
1860 (1958).[60] M. E. Nahmias, Math. Proc. Cambridge Philos. Soc. 31, 99
(1935).[61] H. R. Crane, Rev. Mod. Phys. 20, 278 (1948).[62] J. H. Barrett, Phys. Rev. 79, 907 (1950).[63] J. L. Kulp and L. E. Tyron, Rev. Sci. Instrum. 23, 296
(1952).[64] F. G. Houtermans and W. Thirring, Helv. Phys. Acta 27, 81
(1954).[65] C. L. Cowan, F. Reines, and F. B. Harrison, Phys. Rev. 96,
1294 (1954).
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)
[66] J. Bernstein, M. Ruderman, and G. Feinberg, Phys. Rev.132, 1227 (1963).
[67] F. J. Hasert et al., Phys. Lett. 46B, 121 (1973).[68] F. Reines, H. S. Gurr, and H.W. Sobel, Phys. Rev. Lett. 37,
315 (1976).[69] L. Wolfenstein, Phys. Rev. D 17, 2369 (1978).[70] L. Wolfenstein, Phys. Rev. D 20, 2634 (1979).[71] L. B. Okun, Leptons and Quarks (North-Holland,
Amsterdam, Netherlands, 1982).[72] R. C. Allen et al., Phys. Rev. Lett. 55, 2401 (1985).[73] S. P. Mikheyev and A. Yu. Smirnov, Sov. J. Nucl. Phys. 42,
913 (1985).[74] H. A. Bethe, Phys. Rev. Lett. 56, 1305 (1986).[75] S. J. Parke, Phys. Rev. Lett. 57, 1275 (1986).[76] S. P. Rosen and J. M. Gelb, Phys. Rev. D 34, 969
(1986).[77] J. Dorenbosch et al. (CHARM Collaboration), Z. Phys. C
41, 567 (1989); 51, 142(E) (1991).[78] R. C. Allen et al., Phys. Rev. Lett. 64, 1330 (1990).[79] L. A. Ahrens et al., Phys. Rev. D 41, 3297 (1990).[80] G. S. Vidyakin, V. N. Vyrodov, I. I. Gurevich, Yu. V.
Kozlov,V. P.Martemyanov, S. V. Sukhotin,V. G.Tarasenkov,E. V. Turbin, and S. K. Khakhimov, Pis’ma Zh. Eksp. Teor.Fiz. 55, 212 (1992) [JETP Lett. 55, 206 (1992)].
[81] R. C. Allen et al., Phys. Rev. D 47, 11 (1993).[82] A. I. Derbin, A. V. Chernyi, L. A. Popeko, V. N. Muratova,
G. A. Shishkina, and S. I. Bakhlanov, Pis’ma Zh. Eksp.Teor. Fiz. 57, 755 (1993) [JETP Lett. 57, 768 (1993)].
[83] J. Horejsi, Introduction to Electroweak Unification: Stan-dard Model from Tree Unitarity (World Scientific Publish-ing, Singapore, 1993).
[84] P. Vilain et al. (CHARM-II Collaboration), Phys. Lett. B335, 246 (1994).
[85] L. B. Auerbach et al. (LSND Collaboration), Phys. Rev. D63, 112001 (2001).
[86] D.W. Liu et al. (Super-Kamiokande Collaboration), Phys.Rev. Lett. 93, 021802 (2004).
[87] Z. Daraktchieva et al. (MUNU Collaboration), Phys. Lett.B 615, 153 (2005).
[88] A. G. Beda, V. B. Brudanin, E. V. Demidova, V. G. Egorov,M. G. Gavrilov, M. V. Shirchenko, A. S. Starostin, and T.Vylov, Phys. At. Nucl. 70, 1873 (2007).
[89] C. Arpesella et al. (Borexino Collaboration), Phys. Rev.Lett. 101, 091302 (2008).
[90] M. Deniz et al. (TEXONO Collaboration), Phys. Rev. D81, 072001 (2010).
[91] A. G. Beda, V. B. Brudanin, V. G. Egorov, D. V. Medvedev,V. S. Pogosov, M. V. Shirchenko, and A. S. Starostin, Adv.High Energy Phys. 2012, 1 (2012).
[92] J. A. Formaggio and G. P. Zeller, Rev. Mod. Phys. 84, 1307(2012).
[93] J. Segura, J. Bernabeu, F. J. Botella, and J. Penarrocha,Phys. Rev. D 49, 1633 (1994).
[94] J. Segura, J. Bernabeu, F. J. Botella, and J. A. Penarrocha,Phys. Lett. B 335, 93 (1994).
[95] J. Bernabeu and S. Palomares-Ruiz, J. High Energy Phys.02 (2004) 068.
[96] J. Bernabeu, J. Papavassiliou, and M. Passera, Phys. Lett.B 613, 162 (2005).
[97] J. S. Schwinger, Phys. Rev. 76, 790 (1949).
[98] R. Barbieri, J. A. Mignaco, and E. Remiddi, NuovoCimento A 11, 824 (1972).
[99] W. Pauli and M. E. Rose, Phys. Rev. 49, 462 (1936).[100] R. P. Feynman, Phys. Rev. 76, 769 (1949).[101] Y. S. Tsai, Phys. Rev. 120, 269 (1960).[102] M. Vanderhaeghen, J. M. Friedrich, D. Lhuillier, D.
Marchand, L. Van Hoorebeke, and J. Van de Wiele, Phys.Rev. C 62, 025501 (2000).
[103] M. Heller, O. Tomalak, M. Vanderhaeghen, and S. Wu,Phys. Rev. D 100, 076013 (2019).
[104] A. Djouadi and C. Verzegnassi, Phys. Lett. B 195, 265(1987).
[105] A. Djouadi, Nuovo Cimento A 100, 357 (1988).[106] B. A. Kniehl, Nucl. Phys. B347, 86 (1990).[107] S. Fanchiotti, B. A. Kniehl, and A. Sirlin, Phys. Rev. D 48,
307 (1993).[108] F. Jegerlehner, Z. Phys. C 32, 195 (1986).[109] F. Jegerlehner, Nuovo Cimento C 034S1, 31 (2011).[110] J. Erler, Phys. Rev. D 59, 054008 (1999).[111] J. Erler and M. J. Ramsey-Musolf, Phys. Rev. D 72,
073003 (2005).[112] J. Erler and R. Ferro-Hernòndez, J. High Energy Phys. 03
(2018) 196.[113] S. Abreu et al., in Theory report on the 11th FCC-ee
workshop, edited by A. Blondel, J. Gluza, S. Jadach, P.Janot, and T. Riemann (2019) [arXiv:1905.05078].
[114] F. Bloch and A. Nordsieck, Phys. Rev. 52, 54 (1937).[115] N. Nakanishi, Prog. Theor. Phys. 19, 159 (1958).[116] T. Kinoshita, J. Math. Phys. (N.Y.) 3, 650 (1962).[117] T. D. Lee and M. Nauenberg, Phys. Rev. 133, B1549
(1964).[118] V. V. Sudakov, Zh. Eksp. Teor. Fiz. 30, 87 (1956) [Sov.
Phys. JETP 3, 65 (1956)].[119] D. R. Yennie, S. C. Frautschi, and H. Suura, Ann. Phys.
(N.Y.) 13, 379 (1961).[120] M. Tanabashi et al. (Particle Data Group), Phys. Rev. D 98,
030001 (2018).[121] R. J. Hill, Phys. Rev. D 95, 013001 (2017).[122] J. Bernstein and T. D. Lee, Phys. Rev. Lett. 11, 512
(1963).[123] W. A. Bardeen, R. Gastmans, and B. E. Lautrup, Nucl.
Phys. B46, 319 (1972).[124] S. Y. Lee, Phys. Rev. D 6, 1701 (1972).[125] A. D. Dolgov and Ya. B. Zeldovich, Rev. Mod. Phys. 53, 1
(1981).[126] J. Bernabeu, J. Papavassiliou, and D. Binosi, Nucl. Phys.
B716, 352 (2005).[127] W. Wetzel, Z. Phys. C 11, 117 (1981).[128] N. A. Papadopoulos, J. A. Penarrocha, F. Scheck, and K.
Schilcher, Nucl. Phys. B258, 1 (1985).[129] N. A. Papadopoulos, J. A. Penarrocha, F. Scheck, and K.
Schilcher, Phys. Lett. 149B, 213 (1984).[130] F. Burger, K. Jansen, M. Petschlies, and G. Pientka, J. High
Energy Phys. 11 (2015) 215.[131] M. Ce, A. Gerardin, K. Ottnad, and H. B. Meyer, Proc. Sci.
LATTICE2018 (2018) 137.[132] R. J. Hill, K. S. McFarland, and O. Tomalak (to be
published).[133] R. Mertig, M. Bohm, and A. Denner, Comput. Phys.
Commun. 64, 345 (1991).
THEORY OF ELASTIC NEUTRINO-ELECTRON SCATTERING PHYS. REV. D 101, 033006 (2020)
[134] V. Shtabovenko, R. Mertig, and F. Orellana, Comput. Phys.Commun. 207, 432 (2016).
[135] T. Hahn and M. Perez-Victoria, Comput. Phys. Commun.118, 153 (1999).
[136] D. Binosi and L. Theussl, Comput. Phys. Commun. 161,76 (2004).
[137] W. R. Inc., Mathematica, Version 11.0.1.0 (WolframResearch, Inc., Champaign, IL, 2016).
[138] K. G. Chetyrkin, J. H. Kuhn, and A. Kwiatkowski, Phys.Rep. 277, 189 (1996).
[139] K. G. Chetyrkin, J. H. Kuhn, and M. Steinhauser, Nucl.Phys. B482, 213 (1996).
[140] K. G. Chetyrkin, B. A. Kniehl, and M. Steinhauser, Nucl.Phys. B510, 61 (1998).
[141] T. Alion et al. (DUNE Collaboration), arXiv:1606.09550.[142] Dune fluxes, http://home.fnal.gov/~ljf26/DUNEFluxes/.[143] J. Devan et al. (MINERvA Collaboration), Phys. Rev. D
94, 112007 (2016).[144] L. Aliaga et al. (MINERvA Collaboration), Phys. Rev. D
94, 092005 (2016); 95, 039903(A) (2017).[145] L. A. Soplin, Neutrino flux prediction for the NuMI beam-
line, Ph.D. thesis, The College of William and Mary, 2016.[146] Nova fluxes, http://nova-docdb.fnal.gov/cgi-bin/Show
Document?docid=25266.[147] K. Abe et al. (T2K Collaboration), Phys. Rev. D 87,
012001 (2013); 87, 019902(A) (2013).[148] K. Abe et al. (T2K Collaboration), Phys. Rev. D 91,
072010 (2015).
OLEKSANDR TOMALAK and RICHARD J. HILL PHYS. REV. D 101, 033006 (2020)