Applied Mathematics and Computation 182 (2006) 443–446

www.elsevier.com/locate/amc

Dirac equation on hyperbolic octonions

Jens Köplinger
Prisage Ltd, 406 Aberdeen Ter, Greensboro, NC 27403, USA

Abstract

When extending complex number algebra using nonreal square roots of +1, the resulting arithmetic has long exhibited
signs for potential applicability in physics. This article provides proof to a statement by Musès [C. Musès, Hypernumbers
and quantum field theory with a summary of physically applicable hypernumber arithmetics and their geometries, Appl.
Math. Comput. 6 (1980) 63–94] that the Dirac equation in physics can be found in conic sedenions (or 16-dimensional
M-algebra). Hyperbolic octonions (or counteroctonions), a subalgebra of conic sedenions, are used to describe the Dirac
equation sufficiently in a simple form. In the example of conic sedenions, a method is then outlined on how hypernumbers
could potentially further aid mathematical description of physical law, by transitioning between different geometries
through genuine hypernumber rotation.
Ó 2006 Elsevier Inc. All rights reserved.

Keywords: Counteroctonions; Hyperbolic octonions; Sedenions; Hypernumbers; Dirac equation; Countercomplex numbers; Conic
complex numbers

1. Introduction

The Dirac equation in physics is a fundamental quantum mechanical relation, serving as equation of
motion for a free spin 1/2 particle (like e.g. an electron or neutrino). It forms an elementary building block
in current description of electromagnetism, weak, and strong force. Charles Musès stated in [1] about the
Dirac equation that ‘‘. . . a simpler version of the equation using only 16-dimensional M-algebra1 is possible
. . .’’. This mapping of a fundamental physical relation onto a non-associative number system (in this case
conic sedenions) departs from the traditional approach, which uses matrix or tensor formalisms on associative
(circular) complex numbers.
Detailed analysis of the pertaining hypernumber systems has been performed [2,3], from which notation
and definitions are adapted here unless otherwise noted.

E-mail address: jens@prisage.com

1
K. Carmody recently communicated his intention to use in future writings the term conic sedenions instead of sedenions [2,3] or 16-
dimensional M-algebra [1]. The new term creates a clear distinction with respect to number systems like in [4]. In a similar fashion, the term
hyperbolic octonions will be used in this article instead of counteroctonions to reflect their geometric quality as compared to circular
octonions (built on square roots of 1 only).

0096-3003/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.amc.2006.04.005
444 J. Köplinger / Applied Mathematics and Computation 182 (2006) 443–446

The Dirac equation will be written in this paper as a hyperbolic octonion product, which is a subalgebra of
conic sedenions. This will provide proof to Musès’ claim from above. Expressing the Dirac equation in such a
non-associative arithmetic may then offer an interesting opportunity for further exploration. Genuine hyper-
number rotation may qualify as a new class of symmetry transformations on the Dirac equation, and equip
physicists with an additional mathematical toolset to further explore and describe fundamental relations and
forces in nature. The general method will be demonstrated in the example of conic sedenions, where rotation
in the (1, i0) plane allows to transition the Dirac equation from hyperbolic to circular geometry, thus ‘‘unify-
ing’’ the classical relation with a hypothetical ‘‘other force’’.

2. Dirac equation in hyperbolic octonions

In order to keep concepts from physics to a minimum in this paper, only the most common explicit form of
the Dirac equation (the so-called ‘‘Dirac representation’’) will be examined. Physical constants c and h are set
to 1 since they are non-essential for the mathematical structure.
A particle’s wave function W is expressed as a four vector containing circular complex wl :¼ wrl þ iwil , with
l 2 {0, 1, 2, 3} and upper index {r, i} denoting a component’s real and imaginary2 parts. The wl are functions
on space x1, x2, x3 and time x0. The abbreviation ol is short for partial derivative o/oxl.
The classical Dirac equation can then be written as:
0 10 1 0 1
m þ io0 0 io3 io1  o2 w0 0
B 0 m þ io0 io1 þ o2 io3 C B C B C
B CB w1 C B 0 C
B CB C ¼ B C: ð1Þ
@ io3 io1 þ o2 m  io0 0 A@ w2 A @ 0 A
io1  o2 io3 0 m  io0 w3 0
A hyperbolic octonion zhyp8 to basis elements bhyp8 2 {1, i1, i2, i3, e4, e5, e6, e7} will be expressed through real
number coefficients c[bhyp8] in the following notation:
zhyp8 ¼ ðc½1; c½i1 ; c½i2 ; c½i3 ; c½e4 ; c½e5 ; c½e6 ; c½e7 Þ: ð2Þ
The particle’s wave function W will be mapped per definition onto:
 
Whyp8 :¼ wr0 ; wi0 ; wr1 ; wi1 ; wr2 ; wi2 ; wr3 ; wi3 : ð3Þ
For any l, if wrl
is mapped to c[bhyp8] its imaginary counterpart wil
is mapped to a c[bhyp8 Æ i1]. Therefore, the
circular complex imaginary basis element i is identified with the hyperbolic octonion basis element i1, or i  i1.
With definition of
5hyp8 :¼ ðm; o0 ; 0; 0; 0; o3 ; o2 ; o1 Þ: ð4Þ
The Dirac equation can be written as generic hyperbolic octonion product:
5hyp8 Whyp8 ¼ 0: ð5Þ

Proof. The four circular complex relations of the Dirac equation (1) are separated into their real and
imaginary parts:
     
ðm þ io0 Þ wr0 þ iwi0  io3 wr2 þ iwi2  ðio1 þ o2 Þ wr3 þ iwi3 ¼ 0;
mwr0  o0 wi0 þ o1 wi3  o2 wr3 þ o3 wi2 ¼ 0; ð6Þ
mwi0 þ o0 wr0  o1 wr3  o2 wi3  o3 wr2 ¼ 0; ð7Þ

2
For clarity, the imaginary base element for circular complex numbers will be written as i, without index. Such indexing would make the
Dirac equation hard to read and be unneeded. When using octonion and sedenion arithmetic, the circular complex i will subsequently be
identified with i1 by definition, i.e., i  i1.
 J. Köplinger / Applied Mathematics and Computation 182 (2006) 443–446 445

     
ðm þ io0 Þ wr1 þ iwi1 þ ðio1 þ o2 Þ wr2 þ iwi2 þ io3 wr3 þ iwi3 ¼ 0;
mwr1  o0 wi1 þ o1 wi2 þ o2 wr2  o3 wi3 ¼ 0; ð8Þ
mwi1 þ o0 wr1  o1 wr2 þ o2 wi2 þ o3 wr3 ¼ 0; ð9Þ
     
io3 wr0 þ iwi0 þ ðio1 þ o2 Þ wr1 þ iwi1  ðm þ io0 Þ wr2 þ iwi2 ¼ 0;
mwr2 þ o0 wi2  o1 wi1 þ o2 wr1  o3 wi0 ¼ 0; ð10Þ
mwi2  o0 wr2 þ o1 wr1 þ o2 wi1 þ o3 wr0 ¼ 0; ð11Þ
   r   
ðio1  o2 Þ wr0 þ iwi0  io3 w1 þ iw1  ðm þ io0 Þ wr3 þ iwi3 ¼ 0;
i

mwr3 þ o0 wi3  o1 wi0  o2 wr0 þ o3 wi1 ¼ 0; ð12Þ

mwi3  o0 wr3 þ o1 wr0  o2 wi0  o3 wr1 ¼ 0: ð13Þ
The hyperbolic octonion product 5hyp8Whyp8 (5) is explicitly:
 
ðm; o0 ; 0; 0; 0; o3 ; o2 ; o1 Þ wr0 ; wi0 ; wr1 ; wi1 ; wr2 ; wi2 ; wr3 ; wi3
 
¼ mwr0 ; mwi0 ; mwr1 ; mwi1 ; mwr2 ; mwi2 ; mwr3 ; mwi3 ð14Þ
 
þ o0 wi0 ; o0 wr0 ; o0 wi1 ; o0 wr1 ; o0 wi2 ; o0 wr2 ; o0 wi3 ; o0 wr3
 
þ o3 wi2 ; o3 wr2 ; o3 wi3 ; o3 wr3 ; o3 wi0 ; o3 wr0 ; o3 wi1 ; o3 wr1
 
þ o2 wr3 ; o2 wi3 ; o2 wr2 ; o2 wi2 ; o2 wr1 ; o2 wi1 ; o2 wr0 ; o2 wi0
 
þ o1 wi3 ; o1 wr3 ; o1 wi2 ; o1 wr2 ; o1 wi1 ; o1 wr1 ; o1 wi0 ; o1 wr0

¼ mwr0  o0 wi0 þ o3 wi2  o2 wr3 þ o1 wi3 ; mwi0 þ o0 wr0  o3 wr2  o2 wi3  o1 wr3 ; ð15Þ
mwr1  o0 wi1  o3 wi3 þ o2 wr2 þ o1 wi2 ; mwi1 þ o0 wr1 þ o3 wr3 þ o2 wi2  o1 wr2 ;
mwr2  o0 wi2 þ o3 wi0  o2 wr1 þ o1 wi1 ; mwi2 þ o0 wr2  o3 wr0  o2 wi1  o1 wr1 ;

mwr3  o0 wi3  o3 wi1 þ o2 wr0 þ o1 wi0 ; mwi3 þ o0 wr3 þ o3 wr1 þ o2 wi0  o1 wr0 :
The eight components of (15) can be identified as left-hand part of equations (6) through (13), either identi-
cally or with the opposite sign. This proves that 5hyp8Whyp8 = 0 is indeed equivalent to the Dirac equation and
therefore validates Musès’ claim as stated in the introduction of this paper. h

3. Hypernumber rotation as symmetry transformation

In the current Standard Model for electromagnetism, weak, and strong interaction, physicists have success-
fully been able to expand the Dirac equation by adding terms that warrant invariance under certain symmetry
transformations on the operand W. These transformations are generally expressed in matrix form on circular
complex numbers.
In a simple example, use of hypernumber arithmetic will now be suggested as additional candidate for
expansion of this fundamental relation in physics. Without further speculating here about its actual relevance
in describing physical law, the following is intended to be a demonstration of method only. Instead of matrix
form on circular complex numbers, hypernumber arithmetic will become the genuine method of mathematical
description.
A conic sedenion zcon16 to basis elements bcon16 2 {1, i1, . . . , i7, i0, e1, . . . , e7} will be expressed through real
number coefficients c[bcon16] in the following notation:
zcon16 ¼ ðc½1; c½i1 ; . . . ; c½i7 ; c½i0 ; c½e1 ; . . . ; c½e7 Þ: ð16Þ
Conic sedenions contain both a hyperbolic and a circular octonion subalgebra. The hyperbolic octonion sub-
algebra from above (2) will be mapped to
zhyp8 7! ðc½1; c½i1 ; c½i2 ; c½i3 ; 0; 0; 0; 0; 0; 0; 0; 0; c½e4 ; c½e5 ; c½e6 ; c½e7 Þ ð17Þ
446 J. Köplinger / Applied Mathematics and Computation 182 (2006) 443–446

and the circular octonion zcir8 subalgebra to basis elements bcir8 2 {1, i1, . . . , i7} to
zcir8 7! ðc½1; c½i1 ; c½i2 ; c½i3 ; c½i4 ; c½i5 ; c½i6 ; c½i7 ; 0; 0; 0; 0; 0; 0; 0; 0Þ: ð18Þ
With definition of
5Q1 :¼ ðm; o0 ; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0Þ; ð19Þ
5Q2 :¼ ð0; 0; 0; 0; 0; o3 ; o2 ; o1 ; 0; 0; 0; 0; 0; 0; 0; 0Þ; ð20Þ
the operator 5hyp8 from (4) and a new circular octonion counterpart 5cir8 can be written as
5hyp8 7! 5Q1 þ i0 5Q2 ; ð21Þ
5cir8 7! 5Q1 þ 5Q2 : ð22Þ
Similarly, definition of
 
WQ1 :¼ wr0 ; wi0 ; wr1 ; wi1 ; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0; 0 ; ð23Þ
 
WQ2 :¼ 0; 0; 0; 0; wr2 ; wi2 ; wr3 ; wi3 ; 0; 0; 0; 0; 0; 0; 0; 0 ; ð24Þ
allows to map the operand Whyp8 from (3) and a new relating Wcir8 to
Whyp8 7! WQ1 þ i0 WQ2 ; ð25Þ
Wcir8 7! WQ1 þ WQ2 : ð26Þ
Using a real factor a and conic sedenions 5con16 and Wcon16 like
5con16 :¼ 5Q1 þ expðai0 Þ5Q2 ;
Wcon16 :¼ WQ1 þ expðai0 ÞWQ2 ;
this allows for continuous transition of the Dirac equation 5hyp8Whyp8 = 0 into a new circular octonionic
counterpart:
5cir8 Wcir8 ¼ 0: ð27Þ
The Dirac equation corresponds to a = p/2 and its new counterpart on circular geometry (27) to a = 0.
In a physicist’s perspective and wording, the conic sedenion relation
5con16 Wcon16 ¼ 0 ð28Þ
would therefore unify the classical Dirac equation on hyperbolic geometry with a new relation on circular
geometry by means of a mixing angle a. The expression exp(ai0) could be interpreted as symmetry transfor-
mation of a new class formed by genuine conic complex rotations.
Whether or not such hypernumber arithmetic will actually offer the benefit of describing physical law is sub-
ject to further investigation. Finding the Dirac equation in hyperbolic octonion arithmetic appears encourag-
ing, since it could open a window to broadening traditional circular complex number arithmetic for
description of the different symmetries and geometries of physical forces. For further information on hyper-
numbers in general see e.g. http://www.kevincarmody.com/math/hypernumbers.html.

Acknowledgement

I am indebted to Kevin Carmody for his gracious help with hypernumber arithmetics.

References

[1] C. Musès, Hypernumbers and quantum field theory with a summary of physically applicable hypernumber arithmetics and their
geometries, Appl. Math. Comput. 6 (1980) 63–94.
[2] K. Carmody, Circular and hyperbolic quaternions, octonions, and sedenions, Appl. Math. Comput. 28 (1988) 47–72.
[3] K. Carmody, Circular and hyperbolic quaternions, octonions, and sedenions—further results, Appl. Math. Comput. 84 (1997) 27–47.
[4] K. Imaeda, M. Imaeda, Sedenions: algebra and analysis, Appl. Math. Comput. 115 (2000) 77–88.