Few-Body Systems (EFB22) manuscript No.

(will be inserted by the editor)

L. A. Trevisan · C. Mirez · F. M. Andrade

The exact solution for the Dirac equation with the

Cornell potential
arXiv:1312.4212v1 [hep-ph] 16 Dec 2013

Received: date / Accepted: date

Abstract An analytical solution of the Dirac equation with a Cornell potential, with identical scalar
and vectorial parts, is presented. The solution is obtained by using the linear potential solution, related
to Airy functions, multiplied by another function to be determined. The energy levels are obtained
and we notice that they obey a band structure.
Keywords Cornell Potential · Dirac equation · Airy Functions · interquark forces · confining · energy
levels

1 Introduction

The Cornell potential [1; 2] was proposed to describe the interaction of quarks, containing a confining
linear term and a Coulomb interaction. To our best knowledge, the potential does not possess exact
solutions within the common equations of quantum mechanics, i.e., the nonrelativistic Schrödinger
equation, relativistic Dirac, Klein-Gordon, Proca, and Duffin-Kemmer-Petiau (DKP) equations. Here,
we focus on Dirac equation, with equal scalar and vectorial components.

2 The Solution for the Cornell Potential

We consider each quark with mass mα described by a confining potential V (r) in the Dirac equation,
which is given by  
1
α · p + βmα + (1 + β)V (r) Ψ (x) = Eαn Ψ (x), (1)
2
where α and β are the usual Dirac matrices, and the index α refers to the kind of particle (u, d, or
s quarks, in the present case). By decomposing the above equation in spherical coordinates, we have
the component wave function

f (r)Ωκm (θ, φ) 1 u(r)Ωκm (θ, φ)

     
χ(x)
Ψ (x) = = m = m , (2)
ϕ(x) ig(r)Ω−κ (θ, φ) r iv(r)Ω−κ (θ, φ)

The authors would like to thank L. Tomio for helpfull discussions. L.A. Trevisan and F.M. Andrade thanks
the Fundação Araucária and Setor de Ciências Exatas e Naturais for support.
L.A. Trevisan and F.M. Andrade
Departamento de Matemática e Estatı́stica, Universidade Estadual de Ponta Grossa, 84010-790, Ponta Grossa,
PR, Brazil
E-mail: [email protected]
C. Mirez
UFVJM, Campus Teofilo Otoni
2

Table 1 The energy levels (MeV), Eαn is for the linear potential only, Epert is calculated by first order
′ ′
perturbation method, Eαnn′ + δnn′ is for the case of the Cornell potential with different n .

′ ′ ′ ′ ′ ′
State (n) Eαn Epert Eαn′ + δnn′ (n = 0) Eαn′ + δnn′ (n = 1) Eαnn′ + δ (n = 2)

1 531.20 463.54 457.33 512.21 522.71

2 807.72 769.02 695.39 778.84 794.81
3 1011.90 983.25 871.18 975.72 995.74

such that, after separating the radial parts, we can get the following two radial equations
du(r) κ
= − u(r) + [Eαn + mα ]v(r), (3a)
dr r
dv(r) κ
= v(r) − [Eαn − mα − V (r)]u(r), (3b)
dr r
where κ = − (j + 1/2) = −(l + 1) if l = j − 1/2 and κ = j + 1/2 = l if l = j + 1/2 is the Dirac kappa.
The system (3) may be reduced to a Schrödinger like equation
d2 u(r) κ(κ + 1)
− u(r) + [Eαn + mα ] [Eαn − mα − V (r)] u(r) = 0. (4)
dr2 r2
The solution for the case of a linear potential, V (r) = λr, and κ = −1 is described below (see
details in [3] and [4]): The unnormalized solution of (4) is u(r) = Ai(λ1/3 r + an ) where Ai(r) is the
Airy function
p and an are its roots. The energy levels are given by: Eαn = mα − λan /Kα , where
Kα ≡ 3 λ(Eαn + mα ). The above form avoids singularities at origin.
For the Cornell potential, V (r) = λr−σ/r , we suppose a solution such as u(r) = Ai(λ1/3 r+an )F (r).
′
The energy levels are renamed to Eαn ′ + δnn′ , with the δnn′ term to be determined in a consistent way
′
with the energy levels [5]. The relation between δnn′ and Eαnn ′ is :

′2 2
(Eαnn′ − mα )
− 2/3
= an . (5)
[(Eαnn
′
′ + mα + δnn′ )λ]

By replacing u(r) in Eq. (3), we notice that it can be separated in a part that is the Airy equation,
and another that, in the limit r → 0, has the same form of that of the hydrogen atom (the part where
the potential −σ/r dominates). Therefore, the solution has a known form: F (r) = Pn′ (r)e−ηr rl and
(if l = 0)
η 2 + δnn′ (2Eαnn
′
′ + δnn′ ) = 0, (6a)
′ ′
−2η(1 + n ) + (Eαnn′ + δnn′ + mα )σ = 0, (6b)
where n′ is the quantum number for the resulting equation for F (r).

3 Results and Conclusion

Note that there are two quantum numbers, n and n′ , due to the relation among equations (5) and (6).
In this manner, the energy spectrum has a band structure, and if n′ → ∞, Eαnn ′
′ + δnn′ → Eαn . The

Table 1 shows the numerical calculated energy values for the linear potential and also compares with
the numerical energy values for the Cornell potential obtained thorough the first-order perturbation
correction taking the potential −σ/r as a perturbation. The results shown are for λ = 400M eV /f m
and σ = 100M eV.f m and mα = 0. A more detailed work [5] will be presented soon.

References

1. Eichten E., Gottfried K., Kinoshita T., Lane K.D., and Yan T.M.: Phys. Rev. D 17, 3090 (1978).
2. Eichten E., Gottfried K., Kinoshita T., Lane K.D., and Yan T.M.: Phys. Rev. D 21, 313(E) (1980).
3. Ferreira P.L., Alcaras J.A.C.: Lett. A. Nouvo Cim. 14 500 (1975).
4. Ferreira P.L., Helayel J.A, Zagury N.: Nouvo Cim. A 55 215 (1980) .
5. Trevisan, L.A., Mirez, C., Andrade, F.M.: To be submitted.