As an important measurable quantity, the electromagnetic cross sections of charmonium decay in an e+e- collider are valuated for fifteen final states. A general scheme of lS-coupling analysis is adopted for constructing covariant helicity-coupling amplitudes involving two-body decays with arbitrary integer spins. Based on the decay rate, the collision cross section is obtained finally. A number of illustrative examples is worked out to expound and incarnate the general formulas.
Introduction
Although the Standard Model (SM) has been accepted as a universally appreciated theory basis in the high energy community, it is still hard to calculate the wanted experimental observable from the first principle of the SM for a great many of processes, especially when the strong interaction is involved. Quantum chromodynamics (QCD) as a widely appreciated theory of strong interaction, has been proved to be very successful at high energy when the calculation can be executed perturbatively. However, its validity at the nonperturbative regime, such as J/ψ and ψ′ resonance regions, needs more experimental guidance.
At present, the phenomenological analysis plays an important role for deepening our understanding of the general principle. As a matter of fact, many phenomenological models are constructed for charmonium decay [1]-[12], the parametrization of various decay modes are obtained, such as the pseudoscalar and pseudoscalar mesons (PP), vector and pseudoscalar mesons (VP), octet baryon-pair, and so on.
Recently, a systematical parametrization scheme has been proposed [10, 11, 12], by virtue of SU(3) flavor symmetry, the effective interaction Hamiltonian in tensor form is obtained according to group representation theory. In the light of flavor singlet principle, a systematical parametrization is realized for all charmonium decays, involving binary decays, ternary decays and radiative decays. The parameters appeared in the parametrization scheme, which can be figured out by virtue of data analysis, will shed the light on charmonium decay dynamics. Nevertheless, for data analysis the collision cross section is the crucial bridge between the phenomenological model and the experimental information. Moreover, the differential cross section that contains the angular distribution of particles furnishes a pragmatical tool for analyzing the data systematically and comprehensively.
This paper is devoted to the calculation of the electromagnetic cross sections of charmonium decay. In the next section, the decay rate is expressed by covariant helicity-coupling amplitudes involving two-body decays with arbitrary integer spins. A general scheme of lS-coupling analysis is developed to construct the Lorentz invariant amplitude and lots of illustrative examples are worked out to explained the general formulas. The section that follows focuses on the calculation of the electromagnetic cross sections of charmonium decay. The relation between the decay rate and the collision cross section is derived in this section. After that is a summary section. Relegated to the appendix are materials on various calculation details.
Decay Rates
The helicity formalism is more general since it is valid for any mass and spin of particles involved in the decay. Such a formalism has been established by Suh-Urk Chung in a series of papers [13, 14, 15, 16]. Here we only recapitulate the main ideas and collect the formulas needed for the following calculations.
Consider a two-body decay V→X+Y, where parent particle with spin J is rest and daughter particles with spins s and σ respectively fly out back-to-back. See Table I for all other relevant quantities.
Quantity
Parent
Daughter 1
Daughter 2
Spin
J
s
σ
Parity
ηJ
ηs
ησ
Helicity
λ
λ
ν
Momentum
p
q
k
Energy
p0
q0
k0
Mass
w
ws
wσ
Energy/mass
γs
γσ
Velocity
βs
βσ
Wave function
ϕ∗(λ−ν)
ω(λ)
ϵ(ν)
Table I. Two-body decay: V(J)→X(s)+Y(σ).
The key idea is as follows: there exist two ket states ∣SmS⟩ and ∣lml⟩ in the decay process V(J)→X(s)+Y(σ). They act like the states at rest in the rest frame of parent particle (PPRF), so there exist three at-rest states in the problem; ∣JM⟩ in the initial system and ∣lml⟩ in the final system, at the rest of PPRF. The third state ∣SmS⟩ acts as if it is again at the rest of PPRF. It can proved [13] that J, S, and l are rotational invariants, moreover, J^=S^+l^, then the operator l^ and S^ provide a fully prescription for the conventional lS coupling scheme for two-particle final states.
The polarization four-vectors or wave functions appropriate for the particles “J=1”, “s=1”, and “σ=1” are well known. In the particle rest frame (PRF),
and the indices {δ1⋯δj} have been split out into three distinct sets in the second summation, i.e., {αi} with (i=1,⋯,m+), {βi} with (i=1,⋯,m0), and {γi} with (i=1,⋯,m−), where m± stands for the number of ϕ(±)’s and m0 for ϕ(0)’s. Note that
j=m++m0+m−(7)
and
m=m+−m−,(8)
in addition,
2m±=j±m−m0,(9)
which indicates that the right-hand side must be always even. The first sum in Eq. (5) goes over the allowed values of m0 given j and m. It is clear that the maximum is given by j−m, so that m0 ranges from 0(1), 2(3), ⋯ to j−m=even(odd). The second sum in Eq. (5) represents a summation on the permutations
{(+)(+)⋯(0)(0)⋯(−)(−)⋯}.
It is seen readily that the number of terms in the summation is
bj(m,m0)=m+!m0!m−!j!.(10)
The wave function for a two-body decay with the synthetic intrinsic spin S is formed out of the s and σ polarization four-vectors
where ⟨smsσmσ∣SmS⟩ is the Clebsch-Gordan coefficient; rank-s tensor ϕs(ms) and rank-σ tensor ϕσ(mσ) are the spin j=s,σ wave functions constructed in Eq. (5). Consequently, ψ(s,σ;SmS) is a tensor of rank-(s+σ).
The wave function χ(lml) for the orbital-angular momentum l is the rank-l tensor, just like ϕδ1⋯δjj(m) in Eq. (5). As a matter of fact, one can simplify the treatment of orbital-angular momentum by defining χ(lml) along the helicity axis (that is z axis), then for two-body decay, merely the collinear vectors along the z axis need to be dealt with exclusively and the wave function χ(lml) comes with a zero z component. This means that in the PRF the orbital-angular momentum does not have time component, therefore ml=0,
Now the Lorentz invariant amplitude can be constructed as follows
AlSJ(s,σ;δ)=[pμ,ψ(s,σ;SmS),χ(l0),ϕ∗(Jδ)].(13)
This amplitude corresponds to the transition ∣Jδ⟩→∣Sδ⟩+∣l0⟩ in which the decay products are aligned along z axis. Here δ=λ−ν, the wave function ϕ∗(Jδ) refers to the initial spin J state or the parent particle. The value of l will be determined by the lS-coupling recipe. First, the S and l are coupled according to the usual angular momentum coupling rule; second, only allowed is the l when the coupled values of S and l contain J.
The notation [⋯] indicates Lorentz contraction of all of the relevant Lorentz indices in the PRF. The following rules are adopted for the contraction:
with the modified metric g(w) [see Eq. (4)] and only with this metric, if s+σ+l−J= even. Here pμ never appears in the amplitudes, i.e., n=0.
with the totally antisymmetric rank-4 tensor (ϵαβμν) together with the four-momentum pμ, and with g(w) whenever necessary, if s+σ+l−J= odd. The usage of ϵαβμν and pμ is restricted to be once and only once, so that n=1.
for angular momentum l, only amplitudes proportional to rl are considered (rl rule).
By virtue of these rules, the AlSJ(s,σ;δ) can be figured out to be
The effective use of these rules serves to reduce the number of independent amplitudes to roughly one-half of the nominal number. Precisely speaking, the number of independent FλνJ’s can be evaluated by the formula
NJ=(a+b+1)(a−s+σ+1)+(s+σ−a)(2J+1),(18)
where
a=min{J,s+σ},b=min{J,s−σ},(19)
and one has assumed here that s≥σ. Some numbers of NJ for various final states are evaluated and summarized in Table II.
NJ
S
P
V
A
T
MP
0+
0−
1−
1+
2+
S
1
0
2
1
2
P
1
1
2
1
V
4
3
5
A
4
4
T
7
Table II. The number of independent FλνJ’s. The symbols S, P, V, A, and T denote scalar, pseudoscalar, vector, axial vector, and tensor, respectively.
Finally, we access to the decay rate [17] for the process V→X+Y
dΩdΓ(V→X+Y)=32π2∣MΓ∣2⋅mV2∣qf∣,(20)
where mV is the mass of particle V and qf is the momentum of either the X or the Y. The decay amplitude reads
MΓ=λν∑DMδJ∗(φ,θ,0)⋅FλνJ,(21)
whence
∣MΓ∣2=λν∑∣dMδJ(θ)∣2⋅∣FλνJ∣2,(22)
where M is the spin projection of J, and
Dm′mj(α,β,γ)=e−im′αdm′mje−imγ,(23)
is the usual D-function (d-function) in angular momentum theory [18].
A remark is in order here. In the light of (20), the ∣MΓ∣2/mV2 is a unitless quantity, this form is usually adopted for the evaluation of the decay amplitude.
Such a choice leads to the requirement that AlSJ(s,σ;δ) should be unitless as well. To this end, one choice is to normalize the momentum p in Eq. (13) by w, then the wn is to be disappeared from Eq. (14). As far as the barrier factor rl is concerned, the variable r is measured in units of the scale factor r0 (e.g. 1 fm), so that rl should be understood as (unit-free) quantity (r/r0)l. Nevertheless, except for the evaluation of the decay amplitude, the original symbols are kept without change.
In the following subsections, a few of examples are spell out to exhibit the application of the aforementioned rules. Generally speaking, in the rest frame of a massive parent particle, the angular distribution of the two-body decay is a trivial event distribution. However, the charmonium that we confront is produced from an e+e− collider, there are only two helicity projections, which leads to abundant situations. Since such processes are really concered for our study, the polarization due to e+e− production is left implicit henceforth.
1−→0++0−
For this mode, s,σ=0 so that λ,ν=0; in addition J=1, together with ηJ=−1,ηs=+1,ησ=−1, according to Eq. (17), it is obtained immediately
F001=−F001,
that is F001=0. Therefore, this decay mode is forbidden by parity conservation. For an example, the decay like ψ→a0(1450)π0 is forbidden.
1−→0++1−
First, the restriction due to the orbital-angular momentum, ηJ=ηsησ(−)l, leads to l=even. Second, S=s+σ=1, then from 1⊕0=1; 1⊕2=1,2,3; and 1⊕4=3,4,5, so that l can only be 0 or 2, that is only S-wave and D-wave are allowed. Third, Eq. (17) yields
F0,11=F0,−11,
so that there are merely two independent invariant helicity-amplitudes, F0,11 and F0,01. Fourth, since s+σ+l−J=even, then n=0. By virtue of Eq. (14),
The typical decays for this mode is ψ→π+π−, K+K−, KS0KL0. From ηJ=ηsησ(−)l, l must equal to 1, that is only P-wave is allowed. From s+σ+l−J=even, then n=0. By virtue of Eq. (14), A101=r; By virtue of Eq. (16), F0,01=g10A101=g10r; finally,
First, the restriction due to the orbital-angular momentum, ηJ=ηsησ(−)l, leads to l=odd. Second, the lS-coupling indicates that only P-wave is allowed, i.e., l=1. Third, Eq. (17) yields
F0,11=−F0,−11,F0,01=−F0,01,
so that there is solely one independent invariant helicity-amplitudes, F0,11. Fourth, since s+σ+l−J=odd, then n=1. By virtue of Eq. (14), A011(λ)=⟨101λ∣1λ⟩wrfλ1(γs); By virtue of Eq. (16), F0,±11=±g111/2wr; finally,
One of decays for this mode is ψ→ϕf1(1420). The relation ηJ=ηsησ(−)l leads to l=even. Unlike the aforementioned examples, S can take several values, i.e., S=1⊕1=0,1,2. According to the lS-coupling scheme, some possible couplings are listed below
S=0S=1S=2
0⊕l=l; 1⊕0=1; 2⊕0=2;
(l=even)1⊕2=1,2,3; 2⊕2=0,1,2,3,4;
1⊕4=3,4,5; 2⊕4=2,3,4,5,6.
In the light of these results, the possible combinations are (l,S)=(0,1),(2,1), and (2,2). The relation s+σ+l−J= even, requires n=0. Moreover, Eq. (17) indicates that
By virtue of Eq. (16), the expressions for Fλ,μ1’s can be obtained, the more details can refer to the appendix.
1−→1−+2+
One of decays for this mode is ψ→ϕf2(1270). The relation ηJ=ηsησ(−1)l leads to l=even, and S=1⊕2=1,2,3. According to the lS-coupling scheme, some possible couplings are listed below
S=1
S=2
S=3
1⊕0=1;
2⊕0=2;
3⊕0=3;
1⊕2=1,2,3;
2⊕2=0,⋯,4;
3⊕2=1,⋯,5;
1⊕4=3,4,5;
2⊕4=2,⋯,6;
3⊕4=1,⋯,7;
3⊕6=3,⋯,9.
In the light of these results, the possible combinations are (l,S)=(0,1),(2,1),(2,2),(2,3), and (4,3). The relation s+σ+l−J=even, requires n=0. In order to determine the independent Fλ,μ1’s, all possibility is listed as follows
The condition ∣λ−ν∣≤1 vetoes these terms, viz., F−1,11, F−2,01, F2,01, F−2,01, F2,−11, and F1,−11. For the left nine terms, according to Eq. (17),
together with F0,01, there are totally five independent invariant helicity-amplitudes. Then by virtue of Eqs. (16), (14), and (20), the corresponding results can be acquired and the more details can refer to Table III and the appendix.
Table III. Various distributions for charmonium decays, that is ψ′ or J/ψ into final states M1 and M2, where M1,2=S, P, V, A, and T. The shorthand symbols are defined for the terseness of the table, which include fλμ≡∣Fλ,μ∣2, Cθ≡1+cos2θ, Sθ≡sin2θ, Kα≡(32γα+31) (α=s,σ), Kσs≡γσγs, κ≡1−2(ws2+wσ2)/w2, and gi stands for certain gLS.
Cross Sections
Based on field theory [19, 20], the cross section for the process a+b→V→f reads
where SV, Sa, and Sb denote the spin of particles V, a, and b, respectively; mV, ma, and mb are their corresponding masses; Γ(V→f), Γ(Vˉ→aˉ+bˉ), and Γ(V→all) (also denoted as Γt) are the partial and total decay rates in the rest frame of the particle V; λ(x,y,z) is a function with the definition λ(x,y,z)≡(x−y−z)2−4yz.
Here the inverse process of a+b→V is the process Vˉ→aˉ+bˉ, where Vˉ, aˉ, and bˉ are the antiparticles associated with V, a, and b, which may again be the same particles. This is just the case for the electron and positron collision experiment.
The annihilation processes of e+e− take place through the conversion of the pair into a virtual photon of mass equal to the total center-of-mass (CM) energy. The photon converts into the final products. These reactions proceed through a state of well-defined quantum numbers, and as consequence the possible initial and final states are essentially limited. The interaction of the final particles with the virtual photon is directly measured in the experiment. Unstable particles with the same quantum numbers as the intermediate photon can be produced singly as resonant states that soon after decay. At the appropriate energy there would appear resonance peaks in the production cross section for the final decay products, which is just the case described by Eq. (30). Specially speaking, if V=ψ, Γ(ψ→e+e−) is denoted by Γe and Γ(ψ→f) by Γf, moreover, if the masses of the electron and positron are neglected, the cross section σ(e++e−→ψ→f) (denoted by σf) can be recast as
σf(s)=(s−mV2)2+mV2Γt212π⋅ΓeΓf.(31)
In order to utilize the results obtained in the previous section, we want to derive the above relation through another way. Start with the differential cross section [17]
dΩdσ(a+b→V→X+Y)=64π2s∣Mtot∣2∣qi∣∣qf∣.(32)
where s is the CM energy of particle V and qf (qi) is the momentum of either the X (a) or the Y (b). The total decay amplitude can be constructed as follows
Mtot=MD(V→X+Y)⋅D(s)⋅MP(a+b→V),(33)
with
D(s)=s−mV2+imVΓtξ.(34)
which is the propagator of resonance V. The parameter ξ will be fixed later.
By virtue of the principle of detailed balancing [21],
MP(a+b→V)=MDˉ(Vˉ→aˉ+bˉ).
Now turn to the electron and positron collision experiment, since e+ and e− are defined along the z axis, there is no directionally dependent variable for MDˉ(ψ→e+e−), so after the integral of directional variable, it obtains
Γe≡Γ(ψ→e+e−)=8π∣MD∣2⋅mψ2∣qi∣.(35)
If we define Γf as
Γf≡Γ(ψ→f)=∫dΩ32π2∣MD∣2⋅mψ2∣qf∣,(36)
and put all these together, then utilizing Eq. (32), at the same time neglecting the masses of electron and positron, it obtains eventually
σf(s)=(s−mψ2)2+mψ2Γt24πξ2⋅ΓeΓf.(37)
Comparing Eq. (31) with Eq. (37), notice that V=ψ herein, ξ is fixed to be 3. Here the crucial point lies in the fact that MD in Eq. (36) is just the MΓ in Eq. (20), therefore, the results for decay rates can be adopted for the calculations of cross sections.
The collision cross sections are calculated and results are tabulated in Table III. The information involving FλνJ has been collected in the appendix, and only the lowest order of rl is adopted in the calculation.
A remark is in order here. Table III displays the distributions of charmonium decays, where the charmonium, such as ψ′ or J/ψ, is produced in an e+e− collider. The spin projections in this case can only be either +1 or −1 along the direction of the e+ or e− beams [22], so that the zero component for d-function is forbidden. This requirement decreases the number of independent FλνJ's. However, if the charmonium is from the decay product, say, J/ψ comes from e+e−→ψ′→π+π−J/ψ, then in the rest frame of J/ψ particle, the angular distribution of the two-body decay is a trivial event distribution. But the distribution may implicate more than one form factor. Let us consider the SV mode concretely. For the cascade decay ψ′→π+π−→J/ψ, then J/ψ→ϕf0(980), in the rest frame of J/ψ particle, the angular distribution reads
dΩdσ∝∣F0,11∣2+∣F0,01∣2,(38)
which can be obtained from the formula (22), where M can be ± and 0; instead of the formula (26), where M can solely be ±. Obviously, there are two form factors, that is F0,11 and F0,01, for this event distribution.
If we combine this distribution with that from an e+e− collider, we can acquire more useful information. According to (26) the decay rate reads
by virtue of which αf can be determined, and in turn it can be obtained
∣F0,11∣2∣F0,01∣2=1+αf1−αf.(42)
Such an analysis can help us obtain the relative relation among form factors.
Summary
In this paper, the invariant helicity amplitudes for various charmonium decay are studied. The general formulas are summarized and a number of illustrative examples is worked out to expound and incarnate the theoretical ideas and general criteria.
The relation between the decay rate and the collision cross section is derived, so the results of the decay rate can be utilized for the cross section calculation. The electromagnetic cross sections of charmonium decay in an e+e− collider are valuated for fifteen decay modes. Combining these results with the systematical parametrization scheme, the colossal charmonium data samples taken in BESIII detector could be analyzed systematically in the future, which will shed the light on the dynamics of charmonium decay .
Acknowledgment
The author would like to thank Prof. C.Z. yuan and Prof. G. Li for helpful discussions and thanks is also due to Prof. G. Li for program calculation.
This work is supported in part by National Key Research and Development Program of China under Contracts No. 2023YFA1606000 and No. 2023YFA1606003.
Appendix A
This appendix summarizes, in toto, the calculation results of FλνJ for the fifteen modes of charmonium decay. The factor w has been normalized out.