A note on Duffin-Kemmer-Petiau equation in (1+1) space-time dimensions

Jose T. Lunardi1, 2, a)
1)
School of Physics and Astronomy, University of Glasgow
G12 8QQ, Glasgow,UK
2)
Department of Mathematics & Statistics, State University of Ponta Grossa
Avenida Carlos Cavalcanti 4748, Cep 84030-900, Ponta Grossa, PR, Brazil
In this note we show that the massive Duffin-Kemmer-Petiau equation restricted to (1+1) space-time dimen-
sions has only one irreducible representation, which corresponds to a (pseudo)scalar field, a result which is at
odds with some claims in the recent literature.

PACS numbers: 03.65.Pm, 02.20.Qs, 03.50.Kk

Keywords: Duffin-Kemmer-Petiau equation; (1+1) space-time dimensions; irreducible representations

The Duffin-Kemmer-Petiau (DKP) equation is a first and using relations (2) we obtain
order wave equation similar to the Dirac one, which
in its original formulation in (3+1) space-time dimen- P (U ) = P (6)
sions describes spin-0 and spin-1 fields or particles14 . P (U ) = P , (7)
Some recent papers addressed the DKP equation in
(1+1) space-time dimensions and considered the sup- which shows that P transforms as a scalar and P
posed spin-1 sector of the theory in some situations in- transforms as a vector. Moreover, by applying P and
volving interactions512 . In this note we show, by us- P on the DKP equation (1) we obtain the following
ing the (1+1)-dimensional analogs of the original DKP relations
spin-0 and spin-1 projection operators, that the spin-
(P ) = im (P ) (8)
1 sector of the (1+1) theory turns out to be (unitarily)
equivalent to its spin-0 sector, which actually describes i
P = (P ) , (9)
a (pseudo)scalar field. We illustrate this equivalence by m
explicitly building the lowest dimensional (irreducible) from which we conclude that the
scalar P satisfy the
representation of the theory. Klein-Gordon equation  + m2 P = 0, with the el-
We first recall some basic results concerning the free ements of P being essentially the derivatives of the
DKP equation in (3+1) space-time dimensions. The corresponding elements of P . Similarly, by introducing
equation is given by14,13 (we use natural units ~ = c = 1) the projectors13,16
(i m) = 0 = 0, 1, 2, 3, (1) 2 2 2
R = (1 ) (2 ) (3 ) 0 g 0

(10)
where m is the particles mass, is the DKP wave func-
R =R ,
(11)
tion and are matrices satisfying the DKP algebra
from relations (2) we conclude that R transforms as a
+ = g + g , (2) vector, whereas R transforms as a second-rank asym-
where g is the Minkowski metric tensor in (3+1) di- metric tensor. Applying R and R on the DKP equa-
mensions with signature (+, , , ). It is well known tion (1) we obtain
that there are only three irreducible representations (ir-
reps) of DKP algebra in (3+1) dimensions: one is triv- (R ) = imR (12)
ial, having dimension 1, and the other two are nontrivial, i
R = U , U = [ (R ) (R )] ,(13)
having dimensions 5 and 10, corresponding respectively m
to scalar (spin-0) and vector (spin-1) fields4,14,15 . Under which, combined, show that the field R satisfy the
infinitesimal Lorentz transformations x0 = x , with Procas equation
= g + , = , the DKP spinor trans-
forms as U , where13  + m2 R = 0,


(R ) = 0, (14)
1
U = 1 + S , S = [ , ] . (3) with U being merely the strength tensor.
2
From the above results we conclude that the opera-
By introducing the projectors13,16 tors P and P select the spin-0 sector of the DKP field,
2 2 2
P = (0 ) (1 ) (2 ) (3 )
2
(4) whereas the operators R and R select the spin-1 sec-
tor. We recall that the product of one operator from
P = P , (5)
the pair (P, P ) with any other operator from the pair
(R , R ) vanishes. This means that the spin-0 sector
and the spin-1 sector are unequivalent irreps of DKP
a) Electronic mail: jttlunardi@uepg.br algebra15 . As mentioned above, the nontrivial irreps
 2

of the DKP wave function for the spin-0 and spin- to (1+1) dimensions. Conversely, if we multiply (8)-(9)
1 sectors correspond to spinors having respectively 5 (with , = 0, 1) on the left by 1 we obtain again (21)-
and 10 components. Explicit 5- and 10-dimensional ir- (23). Therefore, we conclude that the DKP spin-0 and
reps for the matrices and spinor can be easily ob- spin-1 sectors are equivalent in (1+1) dimensions. In
tained by rewriting respectively the Klein-Gordon and other words, we conclude that there is no spin-1 sector
the Proca equations to a system of first-order differential in the strict (1+1)-dimensional DKP theory. This is the
equations4 . main message of this note.
To consider the DKP theory in a space-time restricted To illustrate the above, we can easily build an explicit
to (1+1) dimensions we restrict the space-time labels in irrep for the DKP matrices ( = 0, 1) and spinor
the equations (1) and (2) to , = 0, 1. The analogs of by writing the (1+1)-dimensional

second-order Klein-
the spin-0 projectors (4)-(5) become Gordon equation  + m2 = 0 for a scalar field as
2 2
a system of first-order equations in the form (1). Doing
P = (0 ) (1 ) (15) so, we obtain the following 3 3 irrep:
P = P , , = 0, 1, (16)
0 0 i 0 i 0 m
1
whereas the analogs in (1+1) dimensions of all the re- 0 = 0 0 0 , 1 = i 0 0 , = 1 .
maining equations (6)-(9) do not change in form, only i 0 0 0 0 0 m 0
the labels are restricted to the values 0, 1. In the same (24)
way we conclude that P transform as a Lorentz scalar Similarly, we can write the (1+1)-dimensional Procas
and P as a vector, with P and P satisfying (8) equations
and (9) (with , = 0, 1) and P satisfying the Klein-
 + m2 A = 0, A = 0, , = 0, 1,


Gordon equation in (1+1) dimensions. So, as expected,
the projectors P and P select the spin-0 (scalar) sector
of the (1+1) dimensional DKP equation. as a system of first order equations in the DKP form and
Now we consider the analogs of the spin-1 projectors obtain exactly the same representation for the matri-
 01 T
(10)-(11) in (1+1) dimensions: ces as in (24), with now given by Fm , A0 , A1
2
R = (1 ) 0 g 0 (T denotes the transpose), where F 01 = 0 A1 1 A0 .

(17)
From this result we can promptly identify the fields in
R =R , , = 0, 1. (18)
the two cases:
It is straightforward to verify that R transforms like a
F 01 1 1 1 0
vector under infinitesimal Lorentz transformations, but = , A0 = , A1 = .
now we have that the only operators R which are non- m m m
vanishing in (1+1) dimensions are R01 = R10 . Accord- We also note that if a representation for the matrices
ingly, under Lorentz transformations (3) (, = 0, 1) is buit as = 21 ( 1 + 1 ), where are the
2 2 Dirac matrices in (1+1) dimensions, the obtained
R01 (U ) = R01 , (19) representation is reducible into a nontrivial 3-dimensional
irrep unitarily equivalent to (24) and an 1-dimensional
since from DKP algebra we have that R01 S01 = 0.
trivial representation (in which all = 0 and = 0).
Therefore, R01 transforms like a scalar (if we include
[It is also worth to note a mistake that is recurrent in
improper Lorentz transformations, it transforms as a
some papers5,6,8,12 , in which the factor 1/2 is missing in
pseudo-scalar). This is as expected, because in (1+1)
this expression for ; with such a mistake the obtained
dimensions any second-rank anti-symmetric tensor must
matrices do not fulfill relations (2)]
transform like a pseudo-scalar. Moreover, the spin-1
Summarizing, in this note we have shown that the
projectors (17)-(18) can now be rewritten in terms of the
strict (1+1)-dimensional DKP equation has only one
spin-0 ones:
nontrivial irrep (apart from unitary transformations)
R0 = 1 P 1 , R1 = 1 P 0 , R01 = 1 P. (20) which describes a (pseudo)scalar field and which can be
explicitly obtained in the form (24). Then we concluded
By using these relationships the analogs of equations that there is no spin-1 representation for the DKP the-
(12)-(13) in (1+1) dimensions turn out to be ory genuinely restricted to (1+1) space-times, which is
at odds with the claims of some papers in the recent
1 1 (P ) = 1 (im)P 1 (21) literature512 . Finally, we remark that some of these
1 0 (P ) = 1 (im)P 0 (22) papers actually considered the (3+1) DKP equation for
i  0 0
particles/fields having dynamics in just one space dimen-
1 (P ) = 1 P 1 P 1 .


(23) sion, instead of considering a genuine theory restricted to
m
(1+1) space-time dimensions. Then, by using a 10 10
By multiplying the above equations on the left by 1 , representation for the matrices these papers obtained

2
and taking into account that 1 P = P , we ob- for the spin-1 sector the same energy spectrum and the
tain exactly the same set of equations (8)-(9) restricted same dynamic components for the DKP wave function as

those corresponding to the spin-0 sector (see also1720 ).
We argue that this is not a surprising result since the
dynamical equation for the unidimensional propagation
in the (3+1) case in these studies was formally identical
to the (1+1)-dimensional DKP equation.
1 G.
Petiau. Acad. R. Belg. Cl. Sci. Mem. Collect. 8, 16, 1936.
2 R. J. Duffin. Phys. Rev., 54:1114, 1938.
3 N. Kemmer. Proc. R. Soc. A, 166(924):127153, 1938.
4 N. Kemmer. Proc. R. Soc. A, 173(952):91116, 1939.
5 K. Sogut and A. Havare. Class. Quantum Grav., 23(23):7129,
2006.
6 A. Boumali. Phys. Scripta, 76(6):669, 2007.
7 A. Boumali. J. Math. Phys., 49(2):022302, 2008. Erratum: J.
Math. Phys., 54(9):099902, 2013.
8 K. Sogut and A. Havare. Phys. Scripta, 82(4):045013, 2010.
3
9 Y. Chargui, A. Trabelsi, and L. Chetouani. Phys. Lett. A,
374(29):2907, 2010.
10 Y. Chargui and A. Trabelsi. Phys. Scripta, 87(6):065003, 2013.
11 A. Boumali. Z. Naturforsch. A, 70(10):867, 2015.
12 M. Darroodi, H. Hassanabadi, and N. Salehi. Eur. Phys. J. A,
51(6):69, 2015.
13 J.T. Lunardi, B.M. Pimentel, R.G. Teixeira, and J.S. Valverde.
Phys. Lett. A, 268(3):165 173, 2000.
14 Z. Tokuoka. Nucl. Phys., 78(3):681 693, 1966.
15 E. Fischbach, M. Martin Nieto, and C. K. Scott. J. Math. Phys.,
14(12):1760, 1973.
16 Fujiwara I. Prog. Theor. Phys., 10:589, 1953.
17 B. Boutabia-Chraitia and T. Boudjedaa. Phys. Lett. A,
338(2):97, 2005.
18 A. S. de Castro. J. Math. Phys., 51(10):102302, 2010.
19 B. Boutabia-Chraitia and T. Boudjedaa. J. Geom. Phys.,
62(10):2038, 2012.
20 M. Falek, M. Merad, and T. Birkandan. J. Math. Phys.,
58(2):023501, 2017.