NON-LINEAR FIELD THEORY FOR

LEPTON AND QUARK MASSES

arXiv:hep-ph/0201193v1 21 Jan 2002

André Gsponer and Jean-Pierre Hurni

Independent Scientific Research Institute
P.O. Box 30, 1211 Geneva–12, Switzerland
e-mail: isri@vtx.ch

Received November 23, 1995

Abstract

Barut’s formula for the mass of leptons is successfully extended to quarks.

A very simple non-linear scalar field model explains both the N 4 power law
dependence of the mass, and the existence of a cut-off which limits the
number of leptons to three and the number of quarks to five, suggesting that
the mass of the sixth quark is of different origin.

PACS: 12.15.Ff, 11.10 Lm

Published in HADRONIC JOURNAL 19 (1996) 367-373

Copyright 1996
c by Hadronic Press, Inc., Palm Harbor, FL 34682, USA

1
 The foundation of the very successful Standard Model of elementary particles
is the concept of partons, i.e. leptons and quarks which interact by means of
local gauge fields. The basic characteristics of these partons, such as their mass
and charge, are not fully explained by the Model. Moreover, there is still no
explanation of why the top quark weighs much more than its siblings and why a
meaningful pattern for the masses of the partons defies understanding.
In this paper we look at the problem of lepton and quark masses from a
phenomenological point of view. We therefore take the measured lepton masses,
as well as the quark masses (although they are only indirectly measured), as input
data and look at regularities. We then give an interpretation of the resulting
sequence in terms of non-linear wave mechanics.
Our starting point is the observation that the lepton mass formula discovered
by Barut [1] in 1979 can easily be extended to quarks. Assuming that a quantized
e2
self-energy of magnitude 23 α−1 Me c2 N 4 , where α = ~c and N = 0, 1, 2, ..., is a
new quantum number, be added to the rest-mass of a lepton to get the next heavy
lepton in the chain e, µ, t, ..., Barut got the following expression:

3 X n=N
M(N) = Me (1 + α−1 n4 ) (1)
2 n=0

The agreement with the data of this rather simple formula is surprisingly good,
the discrepancy being of the order 10−4 for µ and 10−3 for τ , respectively. In
order to get the masses of the quarks, we simply take as the mass of the lightest
quark Mu = Me /7.25 . Leaving the justification of this factor for later, we see in
Figure 1 and Table 1 that the agreement between our theoretical quark masses and
the "observed" masses [2,3] is quite good.
In his paper [1], Barut suggested that magnetic self-interaction between the
charge and the anomalous magnetic moment of a lepton could explain the N 4 power
law dependence. In fact, as we found out, many different type of mechanisms lead
to a quantum number dependence or to a scaling with the fourth power of some
variable [4,5]. In particular, this will be the case in non-linear field theories with
a F 4 term in the Lagrangian [6]. In this letter, we present an example of such a
theory.
We postulate that the only stable partons are the electron and the u-quark. Let
us suppose that all "excited" states of partons can be described by a relativistic
bag-like model. The bag is taken as the "proper sphere" bounded by a surface of
constant retarded distance s := ct−~x ·~v /c = s0 . The "mass increment" associated
with the nth excitation level is then postulated to be obtained by integrating the
energy density of a scalar field over the volume of the proper sphere. Hence the

2
name of the model: the scalar "barybag".
We assume a non-linear Lagrangian density with a F 4 term:
dL 1 1 1
= ∂ν F ∂ ν F − µ2 F 2 − 2 F 4 . (2)
dV 2 2 4g
As in the traditional bag-model of hadrons [8], variation of the Lagrangian gives
the field equations, and two boundary conditions are needed: The field has to
vanish on the boundary, and its gradient has to be orthogonal to the normal to
the boundary. The second condition is trivially satisfied if F = F (ωs) where ω
is a parameter.
√ In order to put our equations in non-dimensional form, we write
F = 2ωgkf (ωs) where k is a pure number. The field equation is then
d2 µ2
f + f + 2k 2 f 3 = 0 . (3)
ds2 ω2
The solutions, finite on the real axis, are the Jacobi elliptic functions cn and sd.
2
Their modulus is k 2 = 21 (1− ωµ2 ) and the energy density is found to be independent
of s :
dE
4π = g 2 ω 4 k 2 (1 − k 2 ) . (4)
dV
We now use the boundary condition. The function cn and sd have zeros at argu-
ments of value (2n + 1)K K or 2nK K , where K (k) is the real quarter-period. Hence,
the condition that F vanishes on the boundary translates into the quantization
condition s0 ω(n) = nK K (k).
As the energy density is constant within the barybag, the calculation of the
mass increment is trivial. We get

K 4 4 g2 2
∆M(n)c2 = n k (1 − k 2 ) . (5)
3 s0
√
In the limit µ4 ≪ ω 4 , which implies k 2 = 12 , we have K 4 ( 12 ) ≈ 11.81 .
Comparing with Barut’s formula (1), or the lepton masses, and postulating that g
2
equals the electric charge quantum e, we get s0 ≈ 32 αre where re = Mee c2 is the
classical electron radius. Consequently, the barybag radius is equal to the classical
electromagnetic radius of the muon, a small fraction of the electromagnetic radius
of any hadron.
Our assumption that g = e is consistent with the idea that the elementary
particle’s basic properties might be explained by a theory in which α is somehow
the fundamental interaction constant [9]. Indeed, many correlations involving α
have been found by various authors, as much for the particle’s masses [10] as for
their lifetimes [11].

3
 Postulating that s0 is a fundamental length and that g = e for both leptons and
quarks, the only adjustable parameter left is k.
The values of k which give the lepton and quark mass spectras can be explained
in terms of a theory [12] in which the single-periodic "de Broglie waves", that
quantum mechanics associates with particles, are generalized to double-periodic
"Petiau waves" [13]. Instead of being linear combinations of sin and cos functions,
these waves are superpositions of elliptic functions sn, cn, etc. A very appealing
feature of Petiau waves is that their dependence on the modulus interpolates
between pure de Broglie waves (for k = 0) and pure solitonic waves (for k = 1):
A beautiful realization of the wave/particle duality of quantum mechanics.
The introduction of Petiau waves implies a non-linear generalization of quan-
tum theory [12]. As shown by Petiau [14], in term of the first integrals, the
Hamiltonian of a free particle has the form
H = C0 ω 4 k 2 (1 − k 2 ) (6)
where C0 is a constant. This expression has the same form as (4), confirming that
the barybag can be interpreted as the envelop of a superposition of Petiau waves
confined to a limited space-time region.
There are two non-trivial exceptional cases for elliptic functions: the harmonic
case, k = sin( π4 ), and the equianharmonic case, k = sin( 12 π
). It is very plausible
to associate the former with leptons, and the latter with quarks. Indeed, in either
case, the corresponding elliptic functions exhibit several unique symmetry and
scaling properties, which come from the fact that in the complex plane their poles
form a modular aggregate with π2 or π3 symmetries. Using these two special values
of k in (5), we obtain a mass ratio of about 7.2448 for the lepton and quark mass
sequences.
An important aspect of non-linear quantum theory is that the superposition
principle does not apply. There is thus no interference between solutions F (n, k, s)
with different values of the quantum number n. This means that the total mass is
obtained by simply adding the mass increments given by (5). Looking at Table 1
and Figure 1, the main discrepancy between our mass formula and the data is
the non-existence of partons with sequential masses larger than approximately
α−2 Me = 9.6 GeV/c2 . This fact suggests the existence of a cut-off.
A simple explanation for the cut-off is provided by Heisenberg’s uncertainty
principle. In ordinary quantum theory, when a particle is restricted to a region of
radius s0 , its minimum energy, derived from its momentum uncertainty, is roughly
given by
1 ~c 3
Es0 ≈ = α−2 Me c2 ≈ 7.2 GeV (7)
2 s0 4

4
Hence, when the energy of the barybag becomes of the order of Es0 , a plausible
explanation of the ending of the mass spectrum is that the nonspreading barybag
"decays" into a de Broglie wave packet which spreads because it is a superposition
of waves of different energies. This interpretation is consistent with Petiau’s idea
that while "progressive" waves of type exp(iωs) are fundamental in linear quantum
theory, "standing" waves of type cn(ωs) are the fundamental waves in non-linear
quantum theory [13].
In the currently Standard Model of elementary particles, a sixth quark is
needed for theoretical reasons. The experimental observation of this quark [7]
could therefore be interpreted as an argument against the present explanation of
the Barut formula. The Standard Model, however, is basically a perturbative theory
of the interactions of quarks and leptons, a theory that is logically independent of
our theory of mass quantization.
The conjunction of our theory of mass with one of interactions could possibly
explain the existence of three very light neutrinos, which have also to be predicted
by a complete model, as well as a very massive sixth quark. A more complete
model is also needed to provide a clear-cut distinction between our generalized
N 4 -Barut formula and alternative mass spectrum formulae such as the one recently
proposed by Rosen [5].
A first step in this direction is the observation that the modulus k 2 = 12 implies
µ(k) = 0. Ignoring the non-linear term, the field equation of a massive lepton
corresponds therefore to a massless particle: the neutrino. In the quark case
however, µ(k) 6= 0. For n = 5, m is about 115 GeV/c2 , i.e., on the order of the
sixth quark mass.

5
References
[1] A.O. Barut, Phys. Rev. Lett., 42, 1251 (1979).

[2] Particle Data Group, Phys. Rev., D50, 1195 (1994). According to this refer-
ence, note [j], "The estimates of u and d masses are not without controversy
... the u-quark could be essentially massless".

[3] J. Gasser and H. Leutwyler, Phys. Rep., 87, 77 (1982).

[4] Apart from Ref. 1, there are only very few articles in which a N 4 power law
has been postulated in order to explain masses of elementary particle:
G. Rosen, Nuovo Cim., 8, 403 (1960) and Nuovo Cim. Lett, 33, 45 (1982).

[5] G. Rosen, Int. J. Theor. Phys., 34, 31 (1995).

[6] For an example of N 4 effects in non-linear electromagnetic theory see:

F. Bisshopp, Int. J. Theor. Phys., 6, 225 (1972).

[7] F. Abe et al., Phys. Rev. Lett., 74, 2626 (1995).

S. Abachi et al., Phys. Rev. Lett., 74, 2632 (1995).

[8] A. Chodos et al., Phys. Rev., D9, 3471 (1974).

[9] G. Rosen, Nuovo Cim., 32, 1037 (1964) .

A.O. Barut, Phys. Lett., 73B, 310 (1978).

[10] Y. Nambu, Progr. Th. Phys., 7, 595 (1952).

H. Frohlich, Nucl. Phys., 7, 148 (1958).
A. Salam and J. Tiomno, Nucl. Phys., 9, 585 (1958/59).
Y. Tomozawa, Progr. Theor. Phys., 52, 1715 (1974).
M.H. Mac Gregor, Nuovo Cim., 103A, 983 (1990).

[11] M.H. Mac Gregor, Phys. Rev., D13, 574 (1976).

D. Aker, Nuovo Cim., 105A, 935 (1992).

[12] A. Gsponer and J.-P. Hurni, ISRI-94-03, "Lanczos-Einstein-Petiau: From

Dirac’s equation to non-linear wave mechanics", in W.R. Davis et al.,
eds., "Cornelius Lanczos Collected Published Papers With Commentaries"
(North Carolina State University, Raleigh, 1998) pages 2-1248 to 2-1277.

[13] G. Petiau, Nuovo Cim., 9, 542 (1958). For a synthesis of Petiau’s work on
this subject, see: G. Petiau, Ann. Inst. Henri Poincaré, 36, 89 (1982).

[14] G. Petiau, Nuovo Cim., 40, 84 (1965).

6
1 Table 1
N electron masses quark masses
Barut’s Ref. Barut’s Ref. Ref.
formula [2] formula [2] [3]
0 e 0.511 0.511 u 0.068 0–8 8±2
1 µ 105.55 105.66 d 14.1 5 – 15 13 ± 4
2 τ 1786.1 1784.1 s 239 100 – 300 260 ± 80
3 10294. ? c 1378 1300 – 1500 1350 ± 50
4 37184. ? b 4978 4700 – 5300 5300 ± 100
5 t 13766 ? ?
6 31989 ? ?
Comparison of lepton and quark masses in MeV/c2 calculated with Barut’s
formula (1) to measured lepton masses from Ref.2 and to quark masses given in
Ref.2 and 3. The observation of a sixth quark of mass in the range of 160’000 to
190’000 MeV/c2 has been reported at the beginning of 1995 [7].

2 Figure 1
Theoretical mass of leptons and quarks as a function of sequential quantum number
N. Open circles are hypothetical partons with masses larger than 9.6 GeV/c2 .
The lines through the points are guides for the eyes. Me = 511 keV/c2 and
Mu = Me /7.25 are input data. The uncertainty on quark masses are from Ref.2.

7
 105 ◦ ◦
◦ ◦
104 Me ◦ ◦
α2 •b
•τ •c
103
M [MeV/c2 ]

•s
102 •µ
mass
101 •d
of leptons
100 •e
and quarks
10−1 •u
10−2
0 1 2 3 4 5 6 7

Fig.1