JOURNAL OF MATHEMATICAL PHYSICS 52, 082303 (2011)

One real function instead of the Dirac spinor function

Andrey Akhmetelia)
LTASolid Inc., 10616 Meadowglen Ln 2708, Houston, Texas 77042, USA
(Received 23 April 2011; accepted 20 July 2011; published online 16 August 2011)

Three out of four complex components of the Dirac spinor can be algebraically
eliminated from the Dirac equation (if some linear combination of electromagnetic
fields does not vanish), yielding a partial differential equation of the fourth order for
the remaining complex component. This equation is generally equivalent to the Dirac
equation. Furthermore, following Schrödinger [Nature (London), 169, 538 (1952)],
the remaining component can be made real by a gauge transform, thus extending to the
Dirac field the Schrödinger conclusion that charged fields do not necessarily require
complex representation. One of the two resulting real equations for the real function
describes current conservation and can be obtained from the Maxwell equations in
spinor electrodynamics (the Dirac-Maxwell electrodynamics). As the Dirac equation
is one of the most fundamental equations, these results both belong in textbooks and
can be used for development of new efficient methods and algorithms of quantum
chemistry. C 2011 American Institute of Physics. [doi:10.1063/1.3624336]

I. INTRODUCTION
The Dirac equation is a most fundamental equation, crucial for high energy physics and sig-
nificant even for such mundane devices as lead-acid batteries (Ref. 1). It is well known that two
components of the Dirac spinor can be algebraically eliminated from the Dirac equation (Ref. 2,
p. 445). However, to the best of the author’s knowledge, it was not noticed before that another
component can be algebraically eliminated from the resulting equations, yielding a partial differ-
ential equation of the fourth order for the remaining component. While it was noted in Ref. 3
(p. 28) that “the set of four first-order equations comprising the Dirac equation is generally equiva-
lent to, and may be reduced to, a single fourth order equation,” according to the authors of Ref. 3,
this remark relates to currently known rigorous solutions of the Dirac equation, and this result had
not been established for the Dirac equation with arbitrary given electromagnetic field.8
It is possible to make one step further and rewrite the Dirac equation in terms of just
one real function, rather than a complex function, using Schrödinger’s approach, who noted
(Ref. 4) that for each solution of the equations of scalar electrodynamics (the Klein-Gordon-Maxwell
electrodynamics) there is a physically equivalent (i.e., coinciding with it up to a gauge transform)
solution with a real matter field, despite the widespread belief about charged fields requiring complex
representation. Schrödinger concludes his work (Ref. 4) with the following: “One is interested in
what happens when [the Klein-Gordon equation] is replaced by Dirac’s wave equation of 1927 or
other first-order equations. This and the bearing on Dirac’s 1951 theory will be discussed more fully
elsewhere.” To the best of the author’s knowledge, Schrödinger did not publish any sequel to Ref. 4,
and the lack of extension to the Dirac equation may explain the fact that Schrödinger’s work did not
get the attention it deserves. Such an extension is proposed here, as the only complex component
of the fourth-order partial differential equation equivalent to the Dirac equation can be made real in
the same way. Thus, the Dirac equation can be rewritten as an equation for one real function rather
than for a Dirac spinor. This result is both important on its own and can be used for development of
new efficient methods and algorithms of quantum chemistry.

a) Electronic mail: akhmeteli@ltasolid.com. URL: http://www.akhmeteli.org.

0022-2488/2011/52(8)/082303/5/$30.00 52, 082303-1 

C 2011 American Institute of Physics
082303-2 Andrey Akhmeteli J. Math. Phys. 52, 082303 (2011)

II. ELIMINATION OF COMPONENTS FROM THE DIRAC EQUATION

To illustrate the parallels with beautiful but little-known Schrödinger’s work,4 the Dirac equation
is considered as part of spinor electrodynamics, although most derivations are valid without any
changes for the Dirac equation in electromagnetic field independently of the Maxwell equations.
Schrödinger considered the equations of scalar electrodynamics
(∂ μ + ie Aμ )(∂μ + ie Aμ )ψ + m 2 ψ = 0, (1)

Aμ − Aν,νμ = jμ , (2)

jμ = ie(ψ ∗ ψ,μ − ψ,μ

∗
ψ) − 2e2 Aμ ψ ∗ ψ. (3)
He observed that the complex charged matter field ψ can be made real by a gauge transform (at least
locally), although it is generally believed that complex functions are required to describe charged
fields.
The equations of motion in the relevant gauge (unitary gauge (Refs. 5 and 6)) for the transformed
four-potential of electromagnetic field B μ and real matter field ϕ are as follows:
ϕ − (e2 B μ Bμ − m 2 )ϕ = 0, (4)

ν
Bμ − B,νμ = jμ , (5)

jμ = −2e2 Bμ ϕ 2 . (6)
Work4 has another unique feature. While the initial Klein-Gordon equation (1) actually contains
two equations for the real and imaginary components of the complex field ψ, the relevant Eq. (4)
for the real field ϕ contains just one equation. What happened to the other equation? The missing
equation is equivalent to the current conservation equation, and the latter can be derived from the
Maxwell equations (5), as the divergence of the antisymmetric tensor vanishes.
It turns out, however, that Schrödinger’s results also hold in the case of spinor electrodynamics.
In general, four complex components of the Dirac spinor function cannot be made real by a single
gauge transform, but three complex components out of four can be eliminated from the Dirac
equation in a general case, yielding a fourth-order partial differential equation for the remaining
component, which component can be made real by a gauge transform. The resulting two equations
for one real component can be replaced by one equation plus the current conservation equation, and
the latter can be derived from the Maxwell equations.
Spinor electrodynamics is a more realistic theory than scalar electrodynamics, so it seems
important that the charged field of spinor electrodynamics can also be described by one real function.
It is not clear if similar results can be obtained for the Standard Model.
The resulting system of equations for the electromagnetic field and the real matter field is
overdetermined. This fact suggests that it may be feasible to eliminate the matter field altogether, as
this was done in the case of scalar electrodynamics (Ref. 7).
Let us start with the equations of (non-second-quantized) spinor electrodynamics
/ = ψ,
(i ∂/ − A)ψ (7)

Aμ − Aν,νμ = e2 ψ̄γμ ψ, (8)

/ = Aμ γ μ (the Feynman slash notation). For the sake of simplicity, a system of units is
where, e.g., A
used where  = c = m = 1, and the electric charge e is included in Aμ (e Aμ → Aμ ). In the chiral
representation of γ -matrices (Ref. 6),
   
0 −I 0 σi
γ =
0
, γ = i
, (9)
−I 0 −σ i 0
082303-3 One real function instead of the Dirac spinor function J. Math. Phys. 52, 082303 (2011)

where index i runs from 1 to 3, and σ i are the Pauli matrices. If ψ has components
⎛ ⎞
ψ1
⎜ ⎟
⎜ψ ⎟
ψ = ⎜ 2⎟, (10)
⎝ ψ3 ⎠
ψ4
the Dirac equation (7) can be written in components as follows:

(A0 + A3 )ψ3 + (A1 − ı A2 )ψ4

+ ı(ψ3,3 − ıψ4,2 + ψ4,1 − ψ3,0 ) = ψ1 , (11)

(A1 + ı A2 )ψ3 + (A0 − A3 )ψ4

− ı(ψ4,3 − ıψ3,2 − ψ3,1 + ψ4,0 ) = ψ2 , (12)

(A0 − A3 )ψ1 − (A1 − ı A2 )ψ2

− ı(ψ1,3 − ıψ2,2 + ψ2,1 + ψ1,0 ) = ψ3 , (13)

− (A1 + ı A2 )ψ1 + (A0 + A3 )ψ2

+ ıψ2,3 + ψ1,2 − ı(ψ1,1 + ψ2,0 ) = ψ4 . (14)

Obviously, Eqs. (13) and (14) can be used to express components ψ3 , ψ4 via ψ1 , ψ2 and eliminate
them from Eqs. (11) and (12) (cf. Ref. 2, p. 445). The resulting equations for ψ1 and ψ2 are as
follows:
,μ
− ψ1,μ + ψ2 (−ı A1,3 − A2,3 + A0,2 + A3,2 + ı(A0,1 + A3,1 + A1,0 ) + A2,0 )

+ ψ1 (−1 + Aμ Aμ − ı Aμ,μ + ı A0,3 − A1,2 + A2,1 + ı A3,0 ) − 2ı Aμ ψ1,μ = 0, (15)

,μ
− ψ2,μ + ıψ1 (A1,3 + ı A2,3 + ı A0,2 − ı A3,2 + A0,1 − A3,1 + A1,0 + ı A2,0 )

+ ψ2 (−1 + Aμ Aμ − ı(Aμ,μ + A0,3 + ı A1,2 − ı A2,1 + A3,0 )) − 2ı Aμ ψ2,μ = 0. (16)

As Eq. (15) contains ψ2 , but not its derivatives, it can be used to express ψ2 via ψ1 ,

−1
 
ψ2 = − ı F 1 + F 2  + ı F 3 ψ1 , (17)

where F i = E i + ı H i , electric field E i and magnetic field H i are defined by the standard formulas
⎛ ⎞
0 −E 1 −E 2 −E 3
⎜ 1 ⎟
⎜E 0 −H 3 H2 ⎟
F μν = Aν,μ − Aμ,ν = ⎜ ⎜ E2
⎟, (18)
⎝ H3 0 −H 1 ⎟⎠
E 3 −H 2 H1 0

and the modified d’Alembertian  is defined as follows:

 = ∂ μ ∂μ + 2ı Aμ ∂μ + ı Aμ,μ − Aμ Aμ + 1. (19)

Using the above notation, Eq. (16) can be rewritten as follows:



−  − ı F 3 ψ2 − ı F 1 − F 2 ψ1 = 0, (20)
082303-4 Andrey Akhmeteli J. Math. Phys. 52, 082303 (2011)

so Eq. (17) can be used to eliminate ψ2 from Eq. (20), yielding an equation of the fourth order
for ψ1 ,


−1
 
 − ı F 3 ı F 1 + F 2  + ı F 3 − ı F 1 + F 2 ψ1 = 0. (21)

This equation is equivalent to the Dirac equation (if ı F 1 + F 2 ≡0).

/
It should be noted that the coefficient at ψ2 in Eq. (15) is gauge-invariant (it can be expressed
via electromagnetic fields). While this elimination could not be performed for zero electromagnetic
fields, this does not look like a serious limitation, as in reality there always exist electromagnetic
fields in the presence of charged fields, although they may be very small. However, it should be noted
that free spinor field presents a special case and is not considered in this work, as it does not satisfy
the equations of spinor electrodynamics. It is not clear how free field being a special case is related
to the divergencies in quantum electrodynamics. It should also be noted that the above procedure
could be applied to any component of the spinor function, not just to ψ1 , yielding equations similar
to Eq. (21). Presenting the above results in a more symmetric form is beyond the scope of this work.
While the above elimination of the third component of the Dirac spinor is straightforward, this
author failed to find it elsewhere, but cannot be sure that this important result, which belongs in
textbooks, was not published previously.
Using a gauge transform, it is possible to make ψ1 real (at least locally). Then the real and the
imaginary parts of Eq. (16) after substitution of the expression for ψ2 will present two equations
for ψ1 . However, it is possible to construct just one equation for ψ1 in such a way that the system
containing this equation and the current conservation equation will be equivalent to Eq. (16).
Let us consider the current conservation equation
(ψ̄γ μ ψ),μ = 0, (22)
or
(ψ̄,μ γ μ ψ) + (ψ̄γ μ ψ,μ ) = 0. (23)
On the other hand,
(ψ̄,μ γ μ ψ)∗ = (ψ̄,μ γ μ ψ)†
= ψ † (γ μ )† (ψ,μ γ ) = ψ̄γ 0 γ 0 γ μ γ 0 γ 0 ψ,μ = ψ̄γ μ ψ,μ ,
† 0 †
(24)
μ † μ
as (γ ) = γ γ γ (Ref. 6), so the current conservation equation can be written as follows:
0 0

2Re(ψ̄γ μ ψ,μ ) = 2Im(ψ̄ı ∂/ψ) (25)

/ − 1)ψ) = 0,
= 2Im(ψ̄(ı ∂/ − A
/ and ψ̄ψ are real.
as it is not difficult to check that values ψ̄ Aψ
/ − 1)ψ can be nonzero, so
If Eqs. (13)–(15) hold, only the second component of spinor (ı ∂/ − A
Eq. (25) can be written as follows:
2Im(−ψ4∗ δ) = 0, (26)
where δ is the left-hand side of Eq. (16). Therefore, if ψ4 does not vanish identically, the system
containing the current conservation Eq. (26) and the following equation:
2Re(ψ4∗ δ) = 0, (27)
is equivalent to Eq. (16).

III. CONCLUSION
Thus, three complex components of the Dirac equation out of four can be algebraically elimi-
nated, and the remaining component can be made real using a gauge transform. In particular, most
results of the work4 for scalar electrodynamics are extended to a more realistic theory – spinor elec-
trodynamics: in a certain gauge, the charged matter field can be described by one real function. This
082303-5 One real function instead of the Dirac spinor function J. Math. Phys. 52, 082303 (2011)

function satisfies a real partial differential equation (of the fourth order) and the current conservation
equation, which can be obtained from the Maxwell equations. Thus, the system of equations for the
function is overdetermined. As it was shown in work7 that the matter field can be naturally eliminated
from the equations of scalar electrodynamics, the results of this work suggest that the spinor field
can be eliminated from spinor electrodynamics as well (and therefore spinor electrodynamics can
embedded into a quantum field theory in the same way as scalar electrodynamics (cf. Ref. 7), but
this has not been proven yet.

ACKNOWLEDGMENTS
The author is grateful to V. G. Bagrov, A. Yu. Kamenshchik, and A. E. Allahverdyan for their
interest in this work and valuable remarks and to J. Noldus for useful discussions.
1 R. Ahuja, A. Blomqvist, P. Larsson, P. Pyykkö, and P. Zaleski-Ejgierd, Phys. Rev. Lett. 106, 018301 (2011).
2 Fayyazuddin and Riazuddin, Quantum Mechanics (World Scientific, Singapore, 1990).
3 V. G. Bagrov and D. M. Gitman, Exact Solutions of Relativistic Wave Equations (Kluwer, Dordrecht, 1990).
4 E. Schrödinger, Nature (London) 169, 538 (1952).
5 E. Deumens, Physica 18D, 371 (1986).
6 C. Itzykson and J.-B. Zuber, Quantum Field Theory (McGraw-Hill, New York, 1980).
7 A. Akhmeteli, Int. J. Quantum Inf. 9, 17 (2011).
8 V. G. Bagrov, private communication (2010).
Journal of Mathematical Physics is copyrighted by the American Institute of Physics (AIP). Redistribution of
journal material is subject to the AIP online journal license and/or AIP copyright. For more information, see
http://ojps.aip.org/jmp/jmpcr.jsp