Semi-Relativistic Two-Body States of Spinless Particles with a Scalar-Type Interaction Potential
KTH-Mechanics, Royal institute of Technology, S-100 44 Stockholm, Sweden
† Corresponding author. E-mail:
kethylwe@gmail.com
1 IntroductionRelativistic two-body effects may be studied using quantum field equations,[1] two-body Dirac equations,[2–6] and/or by equations resulting from direct quantum-operator substitutions of
and
.[7–10] The latter approach explored in this study ignores spins other than orbital angular momenta. Many relevant references can be found in those cited.
Semay et al.[2] and Ferreira[3] showed explicit results from a 16-components Dirac approach for scalar potentials of the confining type and bound states with vanishing total spin. The main interest of these authors is related to quark spectra. The relevant second-order differential equations obtained are simple and provide some understanding of important two-body effects.
Duviryak (2008),[4] also applying a Dirac-type method, presented solvable two-body models in connection with light mesons and Regge trajectories. No explicit results for scalar potentials are given. However, the general results seem to be relevant in the present context.
Moshinsky and Requer (2003)[6] studied two equal fermionic masses in the context of positronium formations. It seems close to other procedures related to sub-atomic interactions. No explicit results for scalar potentials are given.
In the present study the semi-relativistic approach[8–10] is applied with scalar (mass-type) potentials. In addition a “local-momentum” approximation is suggested to find the Dirac-type equations of Refs. [2–4] for two-body spectra with vanishing total spin.
The basic equations for calculating bound state energies are presented in Sec. 2. Section 3 is devoted to a linear quark-type potential model. Two-body effects on selected bound state energies near the non-relativistic limit are illustrated. Conclusions are in Sec. 4.
2 Semi-relativistic Local-Momentum EquationsIn this section the “local-momentum” approximation used to simplify the semi-relativistic equation is outlined. This approach appears to be closely related to the one of Krolikowski.[7] The semi-relativistic quantum (SRQ) approximation of two interacting spinless particles starts from a Hamiltonian of classical special relativity. For an instantaneous scalar potential
in the center-of-mass frame of two massive particles, the stationary SRQ quantal wave function ψ satisfies the equation

Here,
E the relativistic energy,
c is the speed of light and

the two momentum operators. The mutual scalar potential

and masses are combined:

with

being the rest masses.
The momentum operator
is the same for both particles in a centre-of-mass frame (although moving in opposite directions). The momentum operator is given by the cartesian and the radial expressions as


where

is the orbital angular momentum operator and

the reduced Planck’s constant.
The equations of the local-momentum approximation can be derived by imagining two particles entering from free space towards a finite interaction region. In free space a plane wave
, with a given wave number k, is represented by the partial wave series[11]

where

are the orbital angular momentum quantum numbers. The Legendre polynomials

are expressed in terms of the angle
θ between the initial
z-direction and the relative position vector.
By expanding the square roots in the basic SRQ Eq. (1), and using the cartesian representation (3) of
, this Eq. (1) provides the exact asymptotic wave number k for freely propagating plane waves
:

where

The radial component
of the plane wave is the spherical Bessel functions behaving as

The constant
k-eigenvalue of the partial wave components

is related to the equation:

A formal expansion of the operator in terms of

in the
l-term of Eq. (
1), i.e. in

leads after some algebra to Eq. (
6).
Hence, the plane wave satisfies Eq. (1) and its partial-wave component satisfies (9) as well as (10).
A generalization of the above semi-relativistic observations for plane waves leads to the local-momentum approximation. To this end, let a general wave be expanded as

Assume

satisfies the second-order partial wave equation

where

in Eq. (
12) is an
unspecified scalar function of
r. If

is not constant, higher powers of

are now assumed to satisfy the “
approximate” relations

being accurate for sufficiently slowly varying functions

. It follows that the wave function

in Eq. (
12) solves Eq. (
1) approximately. The left hand member in Eq. (
1) is approximated, yielding

The scalar functions on both sides of this equation have to be equal. Algebraic manipulations determine the local-momentum coefficient, yielding

where

and

. This coefficient is consistent with the 16-component two-body Dirac approach;
[2–3] see also Ref. [
7].
An alternative mass notation is given in terms of the reduced mass μ and the total mass m

leading to the explicit expression

Hence, the leading-order local-momentum approximation is based on the second-order differential equation

Once Eq. (
18) is derived it may be applied to both scattering and bound states.
Eq. (18) for mutual scalar interaction potentials is identical to that of Semay et al.[2] and Ferreira.[3] These authors used a two-body Dirac approach for J = 0 with l = 0 and l = 1 particle-antiparticle bound states. The present approach treats
as a “good” quantum number.
2.1 Non-relativistic limitBy letting the relevant non-relativistic energy ϵ be defined by

the coefficient

is expanded in powers of
c
−2, yielding

For equal masses

and for extreme light-heavy mass systems

, which can be considered also the single-mass limit in an external scalar potential 2
S.
In the single-mass limit the coefficient in Eq. (20) simplifies to

Table 1
Table 1
 Table 1
Local-momentum energy levels for the linear potential in Eq. (28) with selected values of and .
.
| l, n |
μ/m |
 |
 |
 |
| 0, 0 |
0 |
2.338 107 41 |
2.159 115 25 |
2.038 207 71 |
| 0, 1 |
0 |
4.087 949 44 |
3.610 634 50 |
3.339 201 16 |
| 0, 2 |
0 |
5.520 559 83 |
4.728 796 92 |
4.324 057 22 |
| 0, 0 |
1/4 |
2.338 107 41 |
2.265 472 95 |
2.204 232 08 |
| 0, 1 |
1/4 |
4.087 949 44 |
3.879 881 78 |
3.721 580 36 |
| 0, 2 |
1/4 |
5.520 559 83 |
5.158 860 36 |
4.902 267 20 |
| 1, 0 |
0 |
3.361 254 52 |
3.020 832 18 |
2.814 360 32 |
| 1, 1 |
0 |
4.884 451 84 |
4.239 003 78 |
3.894 057 81 |
| 1, 2 |
0 |
6.207 623 29 |
5.247 054 50 |
4.777 075 20 |
| 1, 0 |
1/4 |
3.361 254 52 |
3.206 989 76 |
3.086 488 46 |
| 1, 1 |
1/4 |
4.884 451 84 |
4.588 068 12 |
4.372 802 18 |
| 1, 2 |
1/4 |
6.207 623 29 |
5.754 298 07 |
5.443 479 99 |
| Table 1
Local-momentum energy levels for the linear potential in Eq. (28) with selected values of and .
. |
This agrees with an exact spin symmetry model of the light-heavy quark-mass system in Ref. [12], provided the light mass component is represented by μ. Also,
in Eq. (21) represents the sum of the equal “external” scalar and (time-component) vector potentials in Ref. [12]. As realized from Eq. (20), two-body effects (relative to non-relativistic results) relate to the total mass m being finite rather than infinite.
3 Linear Scalar PotentialEq. (18) is transformed into non-dimensional form for a linear scalar potential defined by

A unit length scale is chosen as in Ref. [
12]:

and a dimensionless length
x is introduced by

The parameter responsible for relativistic effects in general is

so that energy eigenvalues are scaled and represented by

where

and

are independent of the speed of light. The potential
S(
r) is likewise reduced to

The reduced differential equation becomes

with

In Eq. (
29)

represents the two-body parameter being magnified by the relativistic parameter
α
2 as this becomes large. The two extreme cases are

and

. A semiclassical analysis of the coefficient

indicates that turning points are not affected by the relativistic terms. Also

in the oscillating region of the effective potential, implying that energy levels are expected to appear shifted to lower values as
α increases.
Numerical computations based on Eq. (28) are performed using an amplitude-phase method.[13] figure 1 shows how energy levels are shifted as function of
with different values of the two-body parameter
. The reduced mass μ is considered fixed and the quantum numbers are l (orbital angular momentum) and n (radial nodes). Levels corresponding to the single-mass limit are the ones most sensitive to relativistic corrections. A possible explanation is that in this limit one of the masses is as small as possible for a given reduced mass μ. Note that all levels investigated are shifted to lower values as α increases.
The level spacing with respect to n is wider than that with respect to l (see Fig. 1), and only the lowest energy levels are considered in Fig. 1.
4 SummaryAn approximation of the semi-relativistic approach, the “local-momentum approximation”, is outlined. Bound-state conditions appear similar to those of two-body approaches based on the Dirac theory for fermions.
The two-body effect found is that single-mass conditions are more sensitive to relativistic corrections. In the single-mass limit a spectrum corresponding to the spin symmetry of the single-particle Dirac equation is obtained.