IPPP/23/54

Three-body Entanglement in Particle Decays

Kazuki Sakurai∗
Institute of Theoretical Physics, Faculty of Physics,
University of Warsaw, ul. Pasteura 5, PL-02-093 Warsaw, Poland

Michael Spannowsky†
Institute for Particle Physics Phenomenology, Department of Physics,
Durham University, Durham DH1 3LE, U.K.
Abstract
Quantum entanglement has long served as a foundational pillar in understanding quantum mechanics, with a predominant
focus on two-particle systems. We extend the study of entanglement into the realm of three-body decays, offering a more
intricate understanding of quantum correlations. We introduce a novel approach for three-particle systems by utilising the
arXiv:2310.01477v2 [quant-ph] 10 Apr 2024

principles of entanglement monotone concurrence and the monogamy property. Our findings highlight the potential of studying
deviations from the Standard Model and emphasise its significance in particle phenomenology. This work paves the way for
new insights into particle physics through multi-particle quantum entanglement, particularly in decays of heavy fermions and
hadrons.

I. INTRODUCTION By building on the entanglement monotone concur-

rence and the monogamy property, we propose an ap-
proach to extend entanglement to three particles, chart-
The study of quantum entanglement has been a corner-
ing a course for future explorations in the realm of multi-
stone of quantum mechanics, providing profound insights
particle quantum entanglement in particle phenomenol-
into the non-local correlations between quantum systems
ogy. Extending the concept of entanglement to three
[1, 2]. Historically, much of the focus has been on two-
particles will bolster its applicability to uncharted terri-
particle entanglement, particularly in the context of the
tories in particle phenomenology, e.g. the decay of heavy
Bell inequalities [3, 4]. However, as the field of quantum
fermions and hadrons1 This exploration into hadron de-
mechanics evolves, it becomes imperative to explore more
cays offers a fresh perspective and a deeper understand-
complex systems, specifically, the realm of multi-particle
ing of the intricate quantum interactions within these
entanglement.
systems.
Recent research has delved into the intricacies of two-
Additionally, the introduction of the three-particle en-
particle entanglement in the context of particle physics.
tanglement measure presents a new observable. This
Bi-partite systems have been explored for top-anti-top
novel measure holds significant promise in aiding the
quarks [5–10], the Higgs boson [11–13], gauge bosons [14–
search for unknown heavy resonances and potentially
17] and leptons [18], revealing that the quantum infor-
discovering new physics. The entanglement property
mation properties of their spin states at proton colliders
is based on the fundamental interaction of the pro-
are accessible in current data. Furthermore, some stud-
cess. Thus, for a comprehensive assessment of the ex-
ies emphasised the importance of quantum observables
pected three-particle entanglement in three-body decays,
in probing the underlying dynamics of quantum systems
we calculate its value for the effective Lorentz struc-
[19–21].
tures generated by (pseudo)scalars, (pseudo)vectors and
Given the advancements in studying two-particle en- (pseudo)tensors exchanges, respectively.
tanglement, reflected by the large number of entangle-
In the following discussion, we assume these spin states
ment measures for bipartite systems [22–26], extending
do not decohere by hadronisation or interactions with
this research to three-particle systems is a natural yet
the environment before the measurement. For this as-
surprisingly underexplored advancement. The study of
sumption to be warranted, the spin-decoherence time
three-particle entanglement presents a richer tapestry of
scale needs to be significantly longer than the lifetime
quantum correlations and offers the potential to uncover
of the decaying resonance. This is usually the case for
new insights into the fundamental nature of quantum me-
electroweak-scale and many hadronic resonances. For
chanics. Moreover, extending the Bell inequality tests to
example, the spin-decoherence time scale in top quark
three particles can provide a more robust framework for
decays is O(mt /Λ2QCD ), while its lifetime is many orders
testing the foundational principles of quantum mechanics
and exploring potential deviations from the predictions
of the Standard Model.
1 In many high-energy processes, three-body decays are favoured
phenomenologically over two-body decays, either because of
kinematic reasons or improved observability over backgrounds.
∗ Electronic address: kazuki.sakurai@fuw.edu.pl Amongst others, those include top decays, t → bf f¯′ , and the
† Electronic address: michael.spannowsky@durham.ac.uk Higgs boson decay, H → Zµ+ µ− .

1
of magnitude shorter, i.e. 1/Γt with Γt ≃ 1.4 GeV.
II. DEFINITION OF ENTANGLEMENT
Entanglement can be quantified by a class of non-
negative functions called entanglement monotones [26,
27], whose values do not increase under local opera-
tions and classical communication (LOCC). A particu-
larly convenient entanglement monotone is concurrence
[25, 28]. For a mixed state ρ of two qubits the concur-
rence is defined as
C[ρ] = max(0, η1 − η2 − η3 − η4 ) ∈ [0, 1], (1)
where ηi (ηi p
> ηj for i < j) are the eigenvalues of the
√ √
matrix R ≡ ρρ̃ ρ with ρ̃ ≡ (σy ⊗ σy )ρ∗ (σy ⊗ σy ).
For separable states C = 0, while C = 1 for maxi-
mally entangled states. For a pure state of two qubits,
|ψ⟩ ∈ HA ⊗ HB , the concurrence can be computed more
straightforwardly as
q
C[|ψ⟩] = 2(1 − Trρ2B ) , (2)
where ρB is the reduced density operator of subsys-
tem B obtained by tracing over subsystem A: ρB ≡
TrA (|ψ⟩⟨ψ|).
For a three-qubit state, |Ψ⟩ ∈ Hi ⊗ Hj ⊗ Hk , one can
consider two types of entanglement. One is an entangle-
ment between two individual particles, say between i and
j. This entanglement can be computed by first tracing
out subsystem k and use formula (1):
Cij = C[ρij ], ρij = Trk (|Ψ⟩⟨Ψ|) . (3)
Another type is an entanglement between one particle
and the rest of the system, known as one-to-other bipar-
tite entanglement. The concurrence between i and the
composite subsystem (kj) can be computed using Eq.
(2):
q a positive x-component. The opening angles 1 → 2 and
Ci(kj) = 2(1 − Trρ2kj ), ρkj = Tri (|Ψ⟩⟨Ψ|) . (4)
Here we used “Qubit Power” of the Schmidt theorem [29]
(see also e.g. [30]) and applied the two-qubit formula Eq.
(2) to a three-qubit state |Ψ⟩.
The entanglement between i and subsystem (kj) can-
not be freely shared between i-j and i-k. Namely, there
is a trade-off between i’s entanglements with j and k.
This property, called monogamy, is one of the most fun-
damental traits of entanglement and formulated by the
Coffman-Kundu-Wootters (CKW) monogamy inequality
[31, 32]:
2 2 2
Ci(kj) ≥ Cij + Cik . (5)
For multipartite systems, one can define a so-called
genuine multipartite entanglement (GME) [30, 33, 34]. A
good GME measure should (1) vanish for all product and
2
z
<latexit sha1_base64="iJGb+vz25DO3nUJ04ncih8QKTI4=">AAAB6HicdVBNS8NAEN3Ur1q/qh69LBbBU0jEtPZW9OKxBfsBbSib7aRdu9mE3Y1QS3+BFw+KePUnefPfuG0jqOiDgcd7M8zMCxLOlHacDyu3srq2vpHfLGxt7+zuFfcPWipOJYUmjXksOwFRwJmApmaaQyeRQKKAQzsYX8399h1IxWJxoycJ+BEZChYySrSRGvf9YsmxPcetelW8JJXzjJQ97NrOAiWUod4vvvcGMU0jEJpyolTXdRLtT4nUjHKYFXqpgoTQMRlC11BBIlD+dHHoDJ8YZYDDWJoSGi/U7xNTEik1iQLTGRE9Ur+9ufiX1011eOFPmUhSDYIuF4UpxzrG86/xgEmgmk8MIVQycyumIyIJ1Sabggnh61P8P2md2W7Z9hrnpdplFkceHaFjdIpcVEE1dI3qqIkoAvSAntCzdWs9Wi/W67I1Z2Uzh+gHrLdPYbeNWQ==</latexit>
1
<latexit sha1_base64="xTnVU9LIIvMUIutHcVm1RgxoeJo=">AAAB6HicbVDLSgNBEOz1GeMr6tHLYBA8hV3xdQx68ZiAeUCyhNlJbzJmdnaZmRXCki/w4kERr36SN//GSbIHTSxoKKq66e4KEsG1cd1vZ2V1bX1js7BV3N7Z3dsvHRw2dZwqhg0Wi1i1A6pRcIkNw43AdqKQRoHAVjC6m/qtJ1Sax/LBjBP0IzqQPOSMGivVvV6p7FbcGcgy8XJShhy1Xumr249ZGqE0TFCtO56bGD+jynAmcFLsphoTykZ0gB1LJY1Q+9ns0Ak5tUqfhLGyJQ2Zqb8nMhppPY4C2xlRM9SL3lT8z+ukJrzxMy6T1KBk80VhKoiJyfRr0ucKmRFjSyhT3N5K2JAqyozNpmhD8BZfXibN84p3VbmsX5Srt3kcBTiGEzgDD66hCvdQgwYwQHiGV3hzHp0X5935mLeuOPnMEfyB8/kDfd2Mvw==</latexit>
<latexit sha1_base64="RIYmO+ELdl7IduoDTJXNRkVfGAc=">AAAB8HicdVDLSgMxFL1TX7W+qi7dBIvgqmSK1nZXdOOygn1IO5RMJtOGJjNDkhFK6Ve4caGIWz/HnX9jpq2gogcCh3POJfcePxFcG4w/nNzK6tr6Rn6zsLW9s7tX3D9o6zhVlLVoLGLV9YlmgkesZbgRrJsoRqQvWMcfX2V+554pzePo1kwS5kkyjHjIKTFWuusLGw3IwB0US7iMLapVlBG3hl1L6vVapVJH7tzCuARLNAfF934Q01SyyFBBtO65ODHelCjDqWCzQj/VLCF0TIasZ2lEJNPedL7wDJ1YJUBhrOyLDJqr3yemRGo9kb5NSmJG+reXiX95vdSENW/KoyQ1LKKLj8JUIBOj7HoUcMWoERNLCFXc7oroiChCje2oYEv4uhT9T9qVslstn9+clRqXyzrycATHcAouXEADrqEJLaAg4QGe4NlRzqPz4rwuojlnOXMIP+C8fQLBVJBo</latexit>
1
y
<latexit sha1_base64="wfUzHSsXyUCCV6so0j1lLayMk+g=">AAAB6HicdVDLSsNAFL2pr1pfVZduBovgKiTStHZXdOOyBfuANpTJdNKOnTyYmQgh9AvcuFDErZ/kzr9x2kZQ0QMXDufcy733eDFnUlnWh1FYW9/Y3Cpul3Z29/YPyodHXRklgtAOiXgk+h6WlLOQdhRTnPZjQXHgcdrzZtcLv3dPhWRReKvSmLoBnoTMZwQrLbXTUblimY5lN5wGWpF6NSc1B9mmtUQFcrRG5ffhOCJJQENFOJZyYFuxcjMsFCOczkvDRNIYkxme0IGmIQ6odLPloXN0ppUx8iOhK1RoqX6fyHAgZRp4ujPAaip/ewvxL2+QKP/SzVgYJ4qGZLXITzhSEVp8jcZMUKJ4qgkmgulbEZligYnS2ZR0CF+fov9J98K0a6bTrlaaV3kcRTiBUzgHG+rQhBtoQQcIUHiAJ3g27oxH48V4XbUWjHzmGH7AePsEYDONWA==</latexit>
<latexit sha1_base64="Pc8UEa6S1d9p3YyyclCZj1pSbSs=">AAAB/nicdVBNS8NAEN3Ur1q/quLJy2IRKkhJU23rTfTiUcF+QFPKZrtpl242YXcilFDwr3jxoIhXf4c3/42btoKKPhh4vDfDzDwvElyDbX9YmYXFpeWV7GpubX1jcyu/vdPUYawoa9BQhKrtEc0El6wBHARrR4qRwBOs5Y0uU791x5TmobyFccS6ARlI7nNKwEi9/F7iej6Wk6ILQwbkGLvRkB/18gW7dFavOqcOtku2XXMq1ZQ4tROngstGSVFAc1z38u9uP6RxwCRQQbTulO0IuglRwKlgk5wbaxYROiID1jFUkoDpbjI9f4IPjdLHfqhMScBT9ftEQgKtx4FnOgMCQ/3bS8W/vE4Mfr2bcBnFwCSdLfJjgSHEaRa4zxWjIMaGEKq4uRXTIVGEgkksZ0L4+hT/T5pOqVwtnd6cFM4v5nFk0T46QEVURjV0jq7QNWogihL0gJ7Qs3VvPVov1uusNWPNZ3bRD1hvn12xlSE=</latexit>
n(✓, )
✓3
<latexit sha1_base64="ZR0wIEJritTYgHi4ZDwUNaQRXZU=">AAAB73icbVDLSgNBEOz1GeMr6tHLYBA8hV3fx6AXjxHMA5IlzE5mkyGzs+tMrxCW/IQXD4p49Xe8+TdOkj1oYkFDUdVNd1eQSGHQdb+dpeWV1bX1wkZxc2t7Z7e0t98wcaoZr7NYxroVUMOlULyOAiVvJZrTKJC8GQxvJ37ziWsjYvWAo4T7Ee0rEQpG0UqtDg440u5Zt1R2K+4UZJF4OSlDjlq39NXpxSyNuEImqTFtz03Qz6hGwSQfFzup4QllQ9rnbUsVjbjxs+m9Y3JslR4JY21LIZmqvycyGhkzigLbGVEcmHlvIv7ntVMMr/1MqCRFrthsUZhKgjGZPE96QnOGcmQJZVrYWwkbUE0Z2oiKNgRv/uVF0jiteJeVi/vzcvUmj6MAh3AEJ+DBFVThDmpQBwYSnuEV3pxH58V5dz5mrUtOPnMAf+B8/gDTr4/Y</latexit>
✓2
<latexit sha1_base64="wRfLP6hExabdJ1/4gfGhpJDZYXY=">AAAB73icbVDLSgNBEOz1GeMr6tHLYBA8hd3g6xj04jGCeUCyhNnJbDJkdnad6RXCkp/w4kERr/6ON//GSbIHTSxoKKq66e4KEikMuu63s7K6tr6xWdgqbu/s7u2XDg6bJk414w0Wy1i3A2q4FIo3UKDk7URzGgWSt4LR7dRvPXFtRKwecJxwP6IDJULBKFqp3cUhR9qr9kplt+LOQJaJl5My5Kj3Sl/dfszSiCtkkhrT8dwE/YxqFEzySbGbGp5QNqID3rFU0YgbP5vdOyGnVumTMNa2FJKZ+nsio5Ex4yiwnRHFoVn0puJ/XifF8NrPhEpS5IrNF4WpJBiT6fOkLzRnKMeWUKaFvZWwIdWUoY2oaEPwFl9eJs1qxbusXNyfl2s3eRwFOIYTOAMPrqAGd1CHBjCQ8Ayv8OY8Oi/Ou/Mxb11x8pkj+APn8wfSK4/X</latexit>
x
<latexit sha1_base64="LYvhi36knHjL/niVKnOk5JAioUI=">AAAB6HicdVBNS8NAEN3Ur1q/qh69LBbBU0jEtPZW9OKxBfsBbSib7aRdu9mE3Y1YSn+BFw+KePUnefPfuG0jqOiDgcd7M8zMCxLOlHacDyu3srq2vpHfLGxt7+zuFfcPWipOJYUmjXksOwFRwJmApmaaQyeRQKKAQzsYX8399h1IxWJxoycJ+BEZChYySrSRGvf9YsmxPcetelW8JJXzjJQ97NrOAiWUod4vvvcGMU0jEJpyolTXdRLtT4nUjHKYFXqpgoTQMRlC11BBIlD+dHHoDJ8YZYDDWJoSGi/U7xNTEik1iQLTGRE9Ur+9ufiX1011eOFPmUhSDYIuF4UpxzrG86/xgEmgmk8MIVQycyumIyIJ1Sabggnh61P8P2md2W7Z9hrnpdplFkceHaFjdIpcVEE1dI3qqIkoAvSAntCzdWs9Wi/W67I1Z2Uzh+gHrLdPXq+NVw==</latexit>
3
<latexit sha1_base64="weoL8WmA1iDu0bFMSrtSLdtU9Sg=">AAAB6HicbVDLTgJBEOzFF+IL9ehlIjHxRHZ9H4lePEIijwQ2ZHZoYGR2djMza0I2fIEXDxrj1U/y5t84wB4UrKSTSlV3uruCWHBtXPfbya2srq1v5DcLW9s7u3vF/YOGjhLFsM4iEalWQDUKLrFuuBHYihXSMBDYDEZ3U7/5hErzSD6YcYx+SAeS9zmjxkq1826x5JbdGcgy8TJSggzVbvGr04tYEqI0TFCt254bGz+lynAmcFLoJBpjykZ0gG1LJQ1R++ns0Ak5sUqP9CNlSxoyU39PpDTUehwGtjOkZqgXvan4n9dOTP/GT7mME4OSzRf1E0FMRKZfkx5XyIwYW0KZ4vZWwoZUUWZsNgUbgrf48jJpnJW9q/Jl7aJUuc3iyMMRHMMpeHANFbiHKtSBAcIzvMKb8+i8OO/Ox7w152Qzh/AHzucPgOWMwQ==</latexit>
<latexit sha1_base64="RVv1ASpg0sgfg2HPUQrE2p+RzCY=">AAAB8HicdVDLSgMxFM3UV62vqks3wSK4KplBa7srunFZwT6kHUomk2lDk8yQZIQy9CvcuFDErZ/jzr8x01ZQ0QOBwznnkntPkHCmDUIfTmFldW19o7hZ2tre2d0r7x90dJwqQtsk5rHqBVhTziRtG2Y47SWKYhFw2g0mV7nfvadKs1jemmlCfYFHkkWMYGOluwG30RAPvWG5gqrIolaDOXHryLWk0ah7XgO6cwuhCliiNSy/D8KYpIJKQzjWuu+ixPgZVoYRTmelQappgskEj2jfUokF1X42X3gGT6wSwihW9kkD5+r3iQwLracisEmBzVj/9nLxL6+fmqjuZ0wmqaGSLD6KUg5NDPPrYcgUJYZPLcFEMbsrJGOsMDG2o5It4etS+D/peFW3Vj2/Oas0L5d1FMEROAanwAUXoAmuQQu0AQECPIAn8Owo59F5cV4X0YKznDkEP+C8fQLC2JBp</latexit>
<latexit sha1_base64="alZZkdo41BRB6rXQE6a9oM+x//I=">AAAB8HicdVDLSgMxFL3js9ZX1aWbYBFclUzV2u6KblxWsA9ph5LJZNrQZGZIMkIp/Qo3LhRx6+e482/MtBVU9EDgcM655N7jJ4Jrg/GHs7S8srq2ntvIb25t7+wW9vZbOk4VZU0ai1h1fKKZ4BFrGm4E6ySKEekL1vZHV5nfvmdK8zi6NeOEeZIMIh5ySoyV7nrCRgPSP+0XiriELSoVlBG3il1LarVquVxD7szCuAgLNPqF914Q01SyyFBBtO66ODHehCjDqWDTfC/VLCF0RAasa2lEJNPeZLbwFB1bJUBhrOyLDJqp3ycmRGo9lr5NSmKG+reXiX953dSEVW/CoyQ1LKLzj8JUIBOj7HoUcMWoEWNLCFXc7orokChCje0ob0v4uhT9T1rlklspnd+cFeuXizpycAhHcAIuXEAdrqEBTaAg4QGe4NlRzqPz4rzOo0vOYuYAfsB5+wTEXJBq</latexit>
3 2
2
<latexit sha1_base64="htDuQjORXNou1hfoTFX09yID8ig=">AAAB6HicbVDLTgJBEOzFF+IL9ehlIjHxRHaJryPRi0dI5JHAhswOvTAyO7uZmTUhhC/w4kFjvPpJ3vwbB9iDgpV0UqnqTndXkAiujet+O7m19Y3Nrfx2YWd3b/+geHjU1HGqGDZYLGLVDqhGwSU2DDcC24lCGgUCW8Hobua3nlBpHssHM07Qj+hA8pAzaqxUr/SKJbfszkFWiZeREmSo9Ypf3X7M0gilYYJq3fHcxPgTqgxnAqeFbqoxoWxEB9ixVNIItT+ZHzolZ1bpkzBWtqQhc/X3xIRGWo+jwHZG1Az1sjcT//M6qQlv/AmXSWpQssWiMBXExGT2NelzhcyIsSWUKW5vJWxIFWXGZlOwIXjLL6+SZqXsXZUv6xel6m0WRx5O4BTOwYNrqMI91KABDBCe4RXenEfnxXl3PhatOSebOYY/cD5/AH9hjMA=</latexit>
Figure 1: The momentum and spin configuration in the coor-
dinate system.
biseparable states, (2) be positive for all non-biseparable
states, and (3) not increase under LOCC. Recently, a
GME measure satisfying all these criteria has been found
for three-qubit states [35]. It corresponds to the area of
the concurrence triangle, whose three sides are given by
the three one-to-other bipartite entanglements:
 12
F3 = 16

3 Q(Q − C1(23) )(Q − C2(13) )(Q − C3(12) ) , (6)
with Q = 12 [C1(23) + C2(13) + C3(12) ]. With this definition,
F3 takes values between 0 and 1.
III. ENTANGLEMENT IN 3-BODY DECAYS
We consider a 3-body decay 0 → 123 and assume
all particles are distinguishable and have spin-1/2. We
analyse the entanglement of the spin degrees of freedom
(d.o.f.) of the final state particles 1, 2 and 3 at a given
phase-space point (p1 , p2 , p3 ).2 To parametrise the
phase space of the final state we boost into the rest frame
of the initial particle 0 and take the z axis in the direc-
tion of p1 . The x and y axes are chosen such that the y
axis is perpendicular to the decay plane and the p2 has
1 → 3 are denoted by θ2 and θ3 (0 ≤ θ2 , θ3 ≤ π), respec-
tively. We represent the spin polarisation n of the initial
particle 0 by the polar and azimuthal angles, θ and ϕ,
respectively (see Fig. 1).
We choose the spin quantisation axis of each final state
particle in the momentum direction of that particle. In
this case, the eigenvalues of the spin (multiplied by 2)
are called helicity and denoted by λi = ±1 (i = 1, 2, 3).
For a given set of interactions, the quantum field theory
framework lets us calculate the transition matrix element
(helicity amplitude)
Mn
λ1 ,λ2 ,λ3 = ⟨λ1 , λ2 , λ3 |n⟩ , (7)
2 As a related topic, the entanglement in a orthopositronium decay
into three photons has been studied in Ref. [36].
 ∗
cd
where the momentum labels are suppressed. The initial with MLL = − √ 2
· eiϕ s θ2 , MLR = cd
√
2
· eiϕ s θ2 , MRL =
state |n⟩ is expanded by the final states as ∗ ∗ ∗
− c√2d · c θ2 and MRR = c√d2 · c θ2 . We see that this is a
biseparable state
X
|n⟩ = Mnλ1 ,λ2 ,λ3 |λ1 , λ2 , λ3 ⟩ + · · · . (8)
λ1 ,λ2 ,λ3
|Ψ⟩ = ceiϕ s θ2 |−⟩1 + c∗ c θ2 |+⟩1 ⊗
 

The ellipsis represents final states of other phase-space √1 d∗ | + +⟩23 − d| − −⟩23 .

 
2
(14)
points and other decay modes. Focusing on the spin d.o.f.
one can describe the final spin state as Therefore, 1 is entangled neither with 2, 3 nor (23):
1 X
|Ψ⟩ = Mn λ1 ,λ2 ,λ3 |λ1 , λ2 , λ3 ⟩ , (9) C12 = C13 = C1(23) = 0, (15)
N
λ1 ,λ2 ,λ3
while 2 and 3 are maximally entangled:
2 1/2
where N = ( λ1 ,λ2 ,λ3 |Mn
P
λ1 ,λ2 ,λ3 | ) is the normalisa-
tion constant. In general, this is an entangled pure state C23 = 1 . (16)
of three qubits.
The monogamy inequality (5) implies 2 and 3 must also
The 2-particle entanglement and one-to-other entan-
be maximally entangled with the rest of the system:
glement defined in Eqs. (3) and (4) can be readily calcu-
lated, respectively. C2(13) = C3(12) = 1, (17)
In the following, we assume, for simplicity, that the
final state particles are massless while the extension to which can be explicitly checked from the formula (4). Be-
the massive case is straightforward. In general, there are cause the state is biseparable, the GME measure vanishes
16 non-redundant Lorentz structures formed from from
bilinear combinations of Dirac spinors ψ̄Γψ with F3 = 0 . (18)

Γ = I, γ 5 , γ µ , γ µ γ 5 , σ µν ,

(10)
B. Vector and Axialvector Interaction
where γ µ is the Dirac γ matrices, γ 5 ≡ iγ 0 γ 1 γ 2 γ 3 and
σ µν ≡ 2i [γ µ , γ ν ].
We next consider the vector interaction
As a 3-body decay 0 → 123 of one fermion into
three fermions requires two bilinears, 256 Lorentz struc- [ψ̄1 γµ (cL PL + cR PR )ψ0 ][ψ̄3 γ µ (dL PL + dR PR )ψ2 ] , (19)
tures form a complete basis. Instead, we will focus
on the matrix elements and Lorentz structures induced with PR/L ≡ (1 ± γ5 )/2 and cL , cR , dL , dR ∈ R. The
by the exchange of (pseudo)scalars, (pseudo)vectors and matrix element is found as
(pseudo)tensors. p
Mn λ1 ,λ2 ,λ3 ∝ 4 2mp1 p2 p3
h
δλ−1 δλ+2 δλ−3 · cL dL s θ23 c θ2 c θ22 + eiϕ s θ2 s θ22
 
A. Scalar and Pseudoscalar Interaction
−δλ−1 δλ−2 δλ+3 · cL dR s θ22 c θ2 c θ23 − eiϕ s θ2 s θ23
 
We consider the effective interaction operator
+δλ+1 δλ+2 δλ−3 · cR dL s θ22 c θ2 s θ23 + eiϕ s θ2 c θ23
 

[ψ̄1 (cS + icA γ5 )ψ0 ][ψ̄3 (dS + idA γ5 )ψ2 ] , (11) i

+δλ+1 δλ−2 δλ+3 · cR dR s θ23 c θ2 s θ22 − eiϕ s θ2 c θ22 , (20)

where cS , cA , dS , dA ∈ R are coupling constants. We also
define c = cS + icA and d = dS + idA and take |c| = corresponding to the state
|d| = 1 as we are not interested in the overall scale of the
amplitude. For given phase-space point (θ2 , θ3 ) and the |Ψ⟩ = MLL | − +−⟩ + MLR | − −+⟩
initial spin (θ, ϕ), the matrix element of 0 → 123 can be + MRL | + +−⟩ + MRR | + −+⟩ , (21)
calculated as
with MLL = Mn n
−+− /N , MLR = M−−+ /N , MRL =
θ2 +θ3
p
Mn λ1 ,λ2 ,λ3 ∝ 2 2mp1 p2 p3 · s 2
n n
M++− /N and MRR = M+−+ /N . From explicit calcu-
lations, we find
h
− cd · δλ−1 δλ−2 δλ−3 · eiϕ s θ2 + cd∗ · δλ−1 δλ+2 δλ+3 · eiϕ s θ2
∗ ∗
i C12 = C13 = 0 , C23 = 2|MLL MLR + MRL MRR | . (22)
− c∗ d · δλ+1 δλ−2 δλ−3 · c θ2 + c∗ d∗ · δλ+1 δλ+2 δλ+3 · c θ2 , (12)
For one-to-other entanglements, we obtain
where shorthand notations cα = cos α and sα = sin α are
used. This corresponds to the spin state C2(13) = C3(12)
p
|Ψ⟩ = MLL | − −−⟩ + MLR | − ++⟩ = 2 (|MLL |2 + |MRL |2 ) (|MLR |2 + |MRR |2 ),
+ MRL | + −−⟩ + MRR | + ++⟩, (13) C1(23) = 2 MRR MLL − MLR MRL . (23)

3
 F3, Vector, couplings = 1/ 2, = 2 , = 2 F3, Vector, couplings = 1/ 2, = 4 , = 4
Since all three Ci(jk) are non-vanishing in general, the 1.0 1.0
GME measure F3 is also non-vanishing in that case.
0.8 0.8
MXY ∝ cX dY (X, Y = L, R) and we see that both 3
4
3
4

C23 , C1(23) and F3 are proportional to |cL cR dL dR |. One 0.6 0.6

the other hand, C2(13) and C3(12) vanish only if |cL cR | =

3
2 2
0.4 0.4
|dL dR | = 0.
To discuss the momogamy relation, we define the 4
0.2 4
0.2
monogamy measure as
0 0.0 0 0.0
2 2 2 0 4 2
3
4
0 4 2
3
4
Mi = Ci(jk) − [Cij + Cik ], (24) 2 2
F3, Tensor, couplings = 1/ 2, = 2 , = 2 F3, Tensor, couplings = 1/ 2, = 4 , = 4
for i ̸= j ̸= k ̸= i. The CKW monogamy inequalities are 1.0 1.0

expressed by Mi ≥ 0 for i = 1, 2, 3. From Eqs. (22) and 0.8 0.8

2 2 2 3 3
(23), one can show C23 = C2(13) − C1(23) . In the vector 4 4

interaction, we therefore have 0.6 0.6

3
2 2
2 0.4 0.4
M1 = M2 = M3 = C1(23) ≥ 0. (25)
4 4
0.2 0.2

C. Tensor and Pseudotensor Interaction 0 0.0 0 0.0

0 4 2
3
4
0 4 2
3
4
2 2

We consider the tensor interaction

Figure 2: F3 on the (θ2 , θ3 ) planes.
[ψ̄1 (cM + icE γ5 )σ µν ψ0 ][ψ̄3 (dM + idE γ5 )σµν ψ2 ] , (26)
with cM , cE , dM , dE ∈ R. As in the scalar case, we define
c = cM + icE and d = dM + idE and take |c| = |d| = D. Numerical Results
1. This operator is equivalent to α[ψ̄1 σ µν ψ0 ][ψ̄3 σµν ψ2 ] −
β µνρσ
2ϵ [ψ̄1 σµν ψ0 ][ψ̄3 σρσ ψ2 ] with α = cM dM − cE dE and
In Fig. 2, we show the GME measure F3 as a function
β = cM dE + cE dM . The 0 → 123 matrix element is given
of θ2 and θ3 for vector (upper panel) and tensor (lower
by
panel) interactions. The lower left plane is empty because
this region is unphysical, θ2 + θ3 < π. √All couplings,
p
Mn λ1 ,λ2 ,λ3 ∝ −8 2mp1 p2 p3
h cX , dX (X = L, R, M, E), are fixed to 1/ 2. In the left
c∗ d∗ · δλ+1 δλ+2 δλ+3 · [2eiϕ s θ2 s θ22 s θ23 − c θ2 s θ2 −θ
2 ]
3
panel, the spin direction n of the initial particle is set
i to the y direction (perpendicular to the decay plane),
+ cd · δλ−1 δλ−2 δλ−3 · [eiϕ s θ2 s θ2 −θ
2
3
+ 2c θ θ2 θ3
s
2 2 s 2 ] , (27) while in the right panel, it is tilted with θ = ϕ = π4
(see Fig. 1). We see that when n is perpendicular to the
implying the spin quantum state decay plane, F3 for the vector interaction depends only
|Ψ⟩ = MR | + ++⟩ + ML | − −−⟩ , (28) on the combination θ2 + θ3 and symmetric under 2 ↔ 3
exchange. For the tensor interaction, in this case, the
with MR = Mn
+++ /N
and ML = Mn
−−− /N .
This state system is maximally entangled, F3 = 1, regardless√of the
interpolates between the separable states, | + ++⟩ and decay angles. This results in |ML | = |MR | = 1/ 2, as
|−−−⟩, and the maximally entangled Greenberger √ Horne inferred from Eq. (27). When the initial spin is tilted,
Zeilinger state, |GHZ⟩ = (| + ++⟩ + | − −−⟩)/ 2 [37]. F3 behaves asymmetrically under 2 ↔ 3 both for vector
For the tensor interaction, there are no entanglements and tensor interactions, as shown in the two right plots
between two individual particles of Fig. 2.
C12 = C13 = C23 = 0 , (29) Fig. 3 shows various entanglement measures as a func-
tion of the initial spin direction n. For vector in-
while one-to-other entanglements are universal teraction, we show F3 (red-solid), C1(23) (blue-dashed),
C2(13) = C3(12) (green-dashed), C23 (purple-dotted) and
C1(23) = C2(13) = C3(12) = 2|MR ML | . (30) 2
Mi = C1(23) (orange-dotted), while only F3 is shown for
√
The GME measure in this case is the tensor interaction. All couplings are set to 1/ 2. In
F3 = 4|MR ML |2 . (31) the right and left panels, the decay angles are fixed to
(θ2 , θ3 ) = ( 4π 5π 2π 5π
6 , 6 ) and ( 6 , 6 ), receptively. The hori-
The monogamy inequalities are trivially satisfied because zontal axes of the plots represent the angle between the
no entanglement of one-to-other is shared by the individ- z-axis and n. In the upper and lower panels, n rotates
ual pairs (c.f. Eq. (29)). about the y and x axes in the right-handed way, re-
As in the scalar case all entanglement measures are spectively. We observe that F3 responds differently to
independent of the phases of c and d. the rotations of n between the vector and tensor cases.

4
 1.0 2=6
4 , 3=6
5 , couplings = 1/ 2 1.0 2=6
2 , 3=6
5 , couplings = 1/ 2 torical focus has been on two-particle entanglement, we
expanded its realm to three-particle systems, revealing a
0.8 0.8
richer tapestry of quantum correlations. This advance-
F3, Ci(jk), Cij, Mi

F3, Ci(jk), Cij, Mi

0.6 0.6 ment offers profound insights into the fundamental na-
0.4 0.4
ture of quantum mechanics, extending beyond the tradi-
tionally studied bipartite systems.
0.2 0.2
Building upon the foundational concepts of entangle-
0.0 0.0 ment monotone concurrence and the monogamy prop-
0 4 2
3
4
5
4
3
2
7
4
2 0 4 2
3
4
5
4
3
2
7
4
2
y y
erty, we propose a novel approach to understanding en-
1.0 2=6
4 , 3=6
5 , couplings = 1/ 2 1.0 2=6
2 , 3=6
5 , couplings = 1/ 2 tanglement in three-body decays. This exploration paves
the way for future studies in multi-particle quantum en-
0.8 0.8
tanglement and emphasises its significance in particle
F3, Ci(jk), Cij, Mi

F3, Ci(jk), Cij, Mi

0.6 0.6 phenomenology, particularly in the decay dynamics of

0.4 0.4 F3 = Ci(jk)
2 [T]
heavy fermions and hadrons. Having explicitly calculated
F3 [V]
C1(23) [V]
the expected entanglement for the three-body decay via
0.2 0.2 C2(13) = C3(12) [V] (pseudo)scalar, (pseudo)vector and (pseudo)tensor me-
C23 [V]
M1 = M2 = M3 = C1(23)
2 [V]
0.0 0.0 diators, this approach can potentially uncover deviations
0 4 2
3
4
5
4
3
2
7
4
2 0 4 2
3
4
5
4
3
2
7
4
2
x x
from the predictions of the Standard Model, shedding
light on uncharted territories within the quantum realm.
Figure 3: Various entanglement measures as functions of the Thus, we emphasis the pivotal role of three-particle
initial spin direction n. entanglement in particle physics, suggesting new avenues
for exploring new observables and novel search strategies
in high-energy physics.
The only non-vanishing 2-particle entanglement C23 in V. ACKNOWLEDGEMENTS
the vector case is constant with respect to of n.

The work of K.S. is partially supported by the Na-

IV. CONCLUSION tional Science Centre, Poland, under research grant
2017/26/E/ST2/00135 and has received funding from
The exploration of quantum entanglement has been a the Norwegian Financial Mechanism for years 2014-2021,
cornerstone in understanding the non-local correlations grant nr DEC-2019/34/H/ST2/00707. M.S. is supported
inherent in quantum systems. While much of the his- by the STFC under grant ST/P001246/1

[1] M. Aspelmeyer, H. Böhm, T. Gyatso, T. Jennewein, Rev. D 106, 055007 (2022), arXiv:2203.05619 [hep-ph] .
R. Kaltenbaek, M. Lindenthal, G. Molina-Terriza, [11] A. J. Barr, Phys. Lett. B 825, 136866 (2022),
A. Poppe, K. Resch, M. Taraba, R. Ursin, P. Walther, arXiv:2106.01377 [hep-ph] .
and A. Zeilinger, Science (New York, N.Y.) 301, 621 [12] M. M. Altakach, P. Lamba, F. Maltoni, K. Mawatari,
(2003). and K. Sakurai, Phys. Rev. D 107, 093002 (2023),
[2] T. Pramanik, X. Chen, Y. Xiang, X. Li, J. Mao, J. Bao, arXiv:2211.10513 [hep-ph] .
Y. deng, T. Dai, B. Tang, Y. Yang, Z. Li, Q. Gong, [13] J. A. Aguilar-Saavedra, Phys. Rev. D 107, 076016 (2023),
Q. He, and J. Wang, Scientific Reports 12 (2022), arXiv:2209.14033 [hep-ph] .
10.1038/s41598-022-17540-1. [14] M. Fabbrichesi, R. Floreanini, and E. Gabrielli, Eur.
[3] J. S. Bell, Physics Physique Fizika 1, 195 (1964). Phys. J. C 83, 162 (2023), arXiv:2208.11723 [hep-ph] .
[4] S. J. Freedman and J. F. Clauser, Phys. Rev. Lett. 28, [15] A. J. Barr, P. Caban, and J. Rembieliński, Quantum 7,
938 (1972). 1070 (2023), arXiv:2204.11063 [quant-ph] .
[5] Y. Afik and J. R. M. n. de Nova, Eur. Phys. J. Plus 136, [16] R. Ashby-Pickering, A. J. Barr, and A. Wierzchucka,
907 (2021), arXiv:2003.02280 [quant-ph] . JHEP 05, 020 (2023), arXiv:2209.13990 [quant-ph] .
[6] M. Fabbrichesi, R. Floreanini, and G. Panizzo, Phys. [17] M. Fabbrichesi, R. Floreanini, E. Gabrielli, and L. Mar-
Rev. Lett. 127, 161801 (2021), arXiv:2102.11883 [hep- zola, Eur. Phys. J. C 83, 823 (2023), arXiv:2302.00683
ph] . [hep-ph] .
[7] C. Severi, C. D. E. Boschi, F. Maltoni, and M. Sioli, Eur. [18] A. Cervera-Lierta, J. I. Latorre, J. Rojo, and L. Rottoli,
Phys. J. C 82, 285 (2022), arXiv:2110.10112 [hep-ph] . SciPost Phys. 3, 036 (2017), arXiv:1703.02989 [hep-th] .
[8] J. A. Aguilar-Saavedra and J. A. Casas, Eur. Phys. J. C [19] S. R. Beane, D. B. Kaplan, N. Klco, and M. J. Savage,
82, 666 (2022), arXiv:2205.00542 [hep-ph] . Phys. Rev. Lett. 122, 102001 (2019), arXiv:1812.03138
[9] Z. Dong, D. Gonçalves, K. Kong, and A. Navarro, [nucl-th] .
(2023), arXiv:2305.07075 [hep-ph] . [20] I. Low and T. Mehen, Phys. Rev. D 104, 074014 (2021),
[10] R. Aoude, E. Madge, F. Maltoni, and L. Mantani, Phys. arXiv:2104.10835 [hep-th] .

5
[21] M. Carena, I. Low, C. E. M. Wagner, and M.-L. Xiao,
(2023), arXiv:2307.08112 [hep-ph] .
[22] C. H. Bennett, H. J. Bernstein, S. Popescu, and B. Schu-
macher, Phys. Rev. A 53, 2046 (1996), arXiv:quant-
ph/9511030 .
[23] C. H. Bennett, D. P. DiVincenzo, J. A. Smolin,
and W. K. Wootters, Phys. Rev. A 54, 3824 (1996),
arXiv:quant-ph/9604024 .
[24] V. Vedral, M. B. Plenio, M. A. Rippin, and P. L.
Knight, Phys. Rev. Lett. 78, 2275 (1997), arXiv:quant-
ph/9702027 .
[25] S. Hill and W. K. Wootters, Phys. Rev. Lett. 78, 5022
(1997), arXiv:quant-ph/9703041 .
[26] R. Horodecki, P. Horodecki, M. Horodecki, and
K. Horodecki, Rev. Mod. Phys. 81, 865 (2009),
arXiv:quant-ph/0702225 .
[27] E. Chitambar and G. Gour, Reviews of Modern Physics
91 (2019), 10.1103/revmodphys.91.025001.
[28] W. K. Wootters, Physical Review Letters 80, 2245
(1998).
[29] E. Schmidt, Mathematische Annalen 63, 433 (1907).
[30] S. Xie and J. H. Eberly, Physical Review Letters 127
(2021), 10.1103/physrevlett.127.040403.
[31] V. Coffman, J. Kundu, and W. K. Wootters, Phys. Rev.
A 61, 052306 (2000), arXiv:quant-ph/9907047 .
[32] T. J. Osborne and F. Verstraete, Physical Review Letters
96 (2006), 10.1103/physrevlett.96.220503.
[33] W. Dur, G. Vidal, and J. I. Cirac, Phys. Rev. A 62,
062314 (2000), arXiv:quant-ph/0005115 .
[34] Z.-H. Ma, Z.-H. Chen, J.-L. Chen, C. Spengler,
A. Gabriel, and M. Huber, Physical Review A 83 (2011),
10.1103/physreva.83.062325.
[35] Z.-X. Jin, Y.-H. Tao, Y.-T. Gui, S.-M. Fei, X. Li-Jost,
and C.-F. Qiao, Results in Physics 44, 106155 (2023).
[36] B. C. Hiesmayr and P. Moskal, Sci. Rep. 7, 15349 (2017),
arXiv:1706.06505 [quant-ph] .
[37] D. M. Greenberger, M. A. Horne, and A. Zeilinger,
(2007), 10.48550/ARXIV.0712.0921.