Boeyens Chemistry 2013

Boeyens Chemistry 2013
Chemistry in Four Dimensions
Jan C. A. Boeyens
Abstract Some chemical phenomena, awkward to rationalize, are argued to originate in the four-dimensional nature of matter in curved space-time. The problem
is traced back to the separation of space and time variables in the analysis of fourdimensional events. Although mathematically sound, this operation is not physically
valid. It destroys the essential non-classical entanglement of space and time, which
is recognized in relativistic theory, but not in quantum mechanics. We show that
without this approximation the state functions of quantum theory have the same
quaternion structure that describes Lorentz transformation and spin. Hypercomplex
formulation of four-dimensional motion eliminates several bothersome concepts,
such as wave-particle duality and probability density, by providing the logical basis for non-zero commutators in non-classical systems. It shows why chiral states
are undefined in quantum theory and why many solid-state transitions appear to be
sterically forbidden. A brief introduction to hypercomplex algebra is given as an
Keywords: d’Alembertian, Dirac’s equation, harmonic function, spin function,
1 Introduction
Several generations of chemists have been conditioned to accept the notorious discrepancy between the theory and practice of chemistry as the unquestionable norm.
Sterically forbidden molecular rearrangements and phase transformations are routinely reported without comment and the flow of electronic particles, postulated to
rationalize the course of chemical reactions, is never subjected to critical scrutiny.
In reality, practising chemists design their experiments in terms of the 19th century
Jan C. A. Boeyens
Unit for Advanced Scholarship, University of Pretoria, South Africa,
e-mail: [email protected]
Jan C. A. Boeyens
notions of chemical affinity, never adequately explained by 20th century theories.
The innocent belief that quantum physics explains ′′ all of chemistry′′ is, like the
rest of quantum theory, obediently respected as just another of its deep inscrutable
The reluctance to abandon dogmatic theory often results in the introduction of
secondary ad hoc explanations to cover up any cracks in the theory, as they occur.
A prime example occurs in the quantum theory of elemental periodicity. Based on
the wave-mechanical ordering of electronic energy levels in an isolated H atom, a
logical Aufbau procedure for many-electron atoms suggests itself. Elemental ordering in shells, consisting of 2n2 successive atoms, for integral n > 0, is predicted.
This pattern breaks down at n = 3. In order to rescue the theory a poorly explained
effect, ascribed to interelectron repulsion is conjectured, without modification of the
basic theory. Even if the only effect of the additional assumption were the splitting
of energy sub-levels whereby, for instance, the 4s sub-level occurs at a lower energy
than 3d, it still fails to account for the observed periodicity. Instead of the expected
4s1→2 3d 1→10 , the sequence 4s1→2 3d 1→8 (3d 10 4s1→2 ) is observed. Contrary to Aufbau philosophy the interpolated transition series is almost uniformly characterized
by a 4s2 , rather than 3d, valence shell. Arbitrary new concepts, such as the degree of
orbital penetration towards the nucleus, without any wave-mechanical basis, are invoked next, without improving the theory in any way. A more appropriate response
would have been to admit failure of the H model, applied to non-hydrogen atoms.
The unjustified confidence in wave mechanics to account for elemental periodicity encouraged the belief in quantum theory to explain all chemistry, albeit with the
aid of supercomputers – another unfulfilled dream. The cruellest reality for chemists
has perhaps been the consistent failure to identify the quantum-mechanical basis of
the three-dimensional structure of molecules, substantiated by thousands of independent observations, and despite massive computational efforts. This negative result provides a significant clue to account for the failure of quantum theory in the
description of chemical systems.
One of the more successful devices to reconcile chemical behaviour with quantum theory was the proposed definition of atomic orbitals to regulate the distribution
of electrons on both atoms and molecules. A minor irritant in this application is the
complex nature of the relevant wave functions that underlie the definition of atomic
orbitals. As these complex functions invariably occur in orthogonal pairs, real functions can be constructed by suitable linear combinations of these pairs. The same
strategy is used in all quantum-chemical computations in order to avoid complex
variables. This strategy comes at an exorbitant price.
A wave-mechanical model of the H atom describes an electron in terms of three
quantum numbers. However, in order to account for atomic spectra it is necessary to
assume that the extranuclear electrons are not all concentrated at the lowest energy
level, but distributed over several levels as stipulated by a fourth quantum number,
postulated to represent a two-level spin system that obeys an exclusion principle.
The strict consequence of this observation is that the orbitals of a three-fold degenerate level must have the third quantum number with values of ml = −1, 0, 1, which
eliminates the possibility of three real functions (All ml = 0). The simple conclusion
Chemistry in Four Dimensions
is that any computational scheme that operates exclusively on real variables cannot
be considered to be quantum mechanical, but rather as strictly classical.
The pioneers of quantum theory were unprepared for the discovery of two related, but unforeseen, new effects – the appearance of complex variables as an essential feature of the theory and pairs of conjugate variables that do not commute;
a stark departure from classical mechanics. The appearance of complex variables is
now readily appreciated, even in two-dimensional harmonic functions, defined by
the equation:
∂ 2Φ ∂ 2Φ
= 0.
∂ x2
∂ y2
One method to solve this equation is by assuming the potential function to be the
product of two variable functions of x and y respectively; i.e.:
Φ = X ·Y
which on differentiation gives
1 ∂ 2 X 1 ∂ 2Y
X ∂ x2 Y ∂ y2
The two terms are functions of independent variables and, in order to satisfy (1)
must be independent of both variables and therefore equal to a constant, i.e.
1 ∂ 2X
= kx2
X ∂ x2
1 ∂ 2Y
= ky2 .
Y ∂ y2
To satisfy the implied condition, kx2 + ky2 = 0, either kx = ky = 0, or ikx = ky , such
Φ = ce±k(x+iy) ,
which describes a rotation in complex notation, exactly as required by wave mechanics.
In three-dimensional formulation the condition kx2 + ky2 + kz2 = 0 or (A + iB)2
is satisfied, without loss of generality on setting kz = 0, ikx = ky , to describe a threefold degenerate state by the magnetic quantum number ml = 0, ±1. Equating all
constants to zero, by the mathematical separation of the physically entangled x and
y coordinates, not only avoids the use of complex functions, but also destroys the
ability to describe the angular momentum of the system. The one-dimensional projection appears as harmonic oscillation, e.g.
∂ 2X
= k2 X
∂ x2
X = ae±kx ,
which is an incomplete description of the complex rotation.
We have reached the uncomfortable conclusion that our trusted three-dimensional
wave equation, interpreted as a projection from four-dimensional space-time must
Jan C. A. Boeyens
have the same defects as an harmonic oscillator when being considered a model of
complex rotation.
2 Four-dimensional Motion
Motion in four dimensions proceeds against the gradient of a scalar function Φ ,
Φ =
∑ ∂ xµ = ∑ ∂ Φµ ≡
µ =0
1 ∂
ic ∂ t ∂ x ∂ y ∂ z
and the potential field is described by the d’Alembertian:
2 Φ =
Φ =0
∂ x20 ∂ x21 ∂ x22 ∂ x23
1 ∂2
≡ ∇ − 2 2 Φ =0
c ∂t
Equation (2) is recognized as Laplace’s equation in four dimensions and in alternative form is known as the wave equation in 3-dimensional space.
The mathematical procedure of solving (2) is by treating Φ as the product function
Φ = X0 (x0 ) · X1 (x1 ) · X2 (x2 ) · X3 (x3 ).
By substituting the four factors,
∂ 2Φ
∂ 2 X0
(X1 X2 X3 ),
∂ x20
∂ x20
into (2) and dividing by Φ ,
1 ∂ 2Φ
1 ∂ 2Φ
1 ∂ 2Φ
1 ∂ 2Φ
1 2
= 0.
X0 ∂ x20
X1 ∂ x21
X2 ∂ x22
X3 ∂ x23
Each term is a function of a single variable and (3) can only remain valid if each
term is independent of all variables and equal to a constant. Each term can hence be
written in the form of a one-dimensional Helmholtz equation:
− k2 X = 0
in which X = ae±kx . For X = T (t) the equation
1 d 2T
+ k2 T = 0 ,
c2 dt 2
Chemistry in Four Dimensions
writing ω = ck, yields T (t) = a exp(±iω t), in which ω is interpreted physically as
an angular frequency.
Each term in (3) requires an arbitrary constant to generate the overall solution
Φ = A exp(k0 x0 + k1 x1 + k2 x2 + k3 x3 ). These four constants are not independent and
must satisfy the general condition: ∑ k2µ = 0, imposed by (3) and (4). The most general solution of 2 Φ = 0 is therefore a function that depends equally on all four
coordinates, rather than the product of four linear functions. The four-dimensional
Laplacian, 2 Φ = 0 describes the state of potential balance along the curved Riemannian manifold. Minkowski space-time, tangent to the manifold, describes a local
pseudo-Euclidean approximation. By separating space and time variables the Laplacian reduces to a wave equation (2a) in Euclidean space. This equation is the basis
of wave mechanics, which further separates space and time variables into space-like
and time-like equations.
It is at this point that wave mechanics moves out of Minkowski space. Away from
the assertion that space-time is characterized by the harmonics of a four-dimensional
Laplacian, visualized as long-wavelength undulation; like a wave field in Minkowski
space. In wave-mechanical approximation the time and space variables (collectively
represented as x) are separated by definition of the product function
u(x,t) = f (x) · e−iω0t
to yield
∇2 ψ +
ω 2
ψ =0
iω0 ∂
Ψ = 0.
∇ + 2
c ∂t
By de Broglie’s postulate, the wave vector becomes
ω0 2 π
= ,
whereby (5) and (5a) reduce to the familiar Schrödinger equations, which provide a
good practical approximation within the locally perceived three-dimensional space,
but the holistic entanglement of space-time coordinates is lost.
Developed into a theory of atomic stability the equilibrium condition, dE/dr = 0,
defines the so-called stationary states of the hydrogen atom as a function of the total
electronic energy,
p 2 e2
(in esu).
Considerd as a standing de Broglie wave of wavelength nλ = 2π rn , the distance
between proton and electron follows as
Jan C. A. Boeyens
rn =
i.e. En = −
=− 2 .
This inverse square relationship, which assumes a spherically symmetrical standing electron wave, is the fundamental equation of atomic spectroscopy and nonrelativistic wave mechanics. In four-dimensional space-time, especially in a nonzero gravitational field, the assumption is not strictly valid and the proportionality
factor, Rn may vary with n.
The mass variable is a strictly empirical assumption that only acquires meaning
in non-Euclidean space-time on distortion of the Euclidean wave field defined by
equation (2). The space-like equation (5), known as Schrödinger’s time-independent
equation is not Lorentz invariant. It is satisfied by a non-local wave function which,
in curved space, generates time-like matter-wave packets, characterized in terms
of quantized energy and three-dimensional orbital angular momentum. The fourdimensional aspect of rotation, known as spin, is lost in the process and added on
by assumption. For macroscopic systems the wave-mechanical quantum condition
h̄ω = E − V is replaced by Newtonian particle mechanics, in which E = 12 mv2 + V .
This condition, in turn, breaks down as v → c.
Wave mechanics, and particle mechanics, formulated to describe motion in threedimensional space are both incomplete by their failure to account for spin and relativistic effects. The common defect in both formulations lies in the unphysical separation of space and time variables. The proper procedure requires hypercomplex
solutions of (2), which describe motion in four-dimensional space-time1 .
Although hypercomplex solution of (2) has never been achieved, a conditional
covariant form of the wave equation was proposed by Dirac [1] on assuming spin
matrices as possible solutions. The procedure consisted of modifying Schrödinger’s
= HΨ ,
which is linear in ∂ /∂ t, by inserting the relativistic Hamiltonian
H = p2 c2 + m20c4
To ensure Lorentz invariance the Hamiltonian should also be linear in space derivatives, −h̄ipk = ∂ /∂ xk (k = 1, 2, 3), such that
∑ αk cpk + β m0c2,
β ∂
Ψ = m0 cΨ .
+ ∑ β αk
c ∂ t k=1
∂ xk
For the benefit of those readers who are not familiar with hypercomplex numbers and quaternions
an elementary introduction is provided as an Appendix to this paper.
Chemistry in Four Dimensions
With γ 0 = β , γ k = β αk , x0 = ct the equation
0 ∂
1 ∂
2 ∂
3 ∂
ih̄ γ
Ψ = m0 cΨ ,
∂ x0
∂ x1
∂ x2
∂ x3
is Lorentz invariant and commonly transformed, using units of h̄ = c = 1 and the
Einstein summation convention, to read
iγ µ ∂µ − m = 0.
These conditions are fulfilled by assuming coefficients of the form
0 σk
I 0
, β=
αk =
σk 0
0 −I
where σk are Pauli matrices and I is the 2 × 2 unit matrix. As before, the spin variables are therefore introduced empirically.
The most significant difference of Dirac’s results from those of the non-relativistic
Pauli equation is that the orbital angular momentum and spin of an electron in a
central field are no longer separate constants of the motion. Only the components of
J = L + S and J2 , which commute with the Hamiltonian, emerge as conserved quantities [1]. Dirac’s equation, extended to general relativity by the method of projective
relativity [2] automatically ensures invariance with respect to gauge, coordinate and
spinor transformations, but has never been solved in this form.
The surprising implication is that Dirac’s equation does not allow of a selfconsistent single-particle interpretation, although it has been used to calculate approximate relativistic corrections to the Schrödinger energy spectrum of hydrogen.
The obvious reason is that a 4D point particle is without duration and hence undefined. An alternative description of elementary units of matter becomes unavoidable.
Prompted by such observation Dirac [3] re-examined the classical point model of the
electron only to find that it has three-dimensional size, with an interior that allows
superluminal signals. It all points at a wave structure with phase velocity vφ > c.
The equation 2 Φ = 0 has unitary quaternion solutions of the form
Φ = eθ (iα + jβ +kγ ) = cos θ + sin θ (iα + jβ + kγ ) ,
which represent the SU(2) rotational Lie group. The four-dimensional eigenfunction
is the proper spin function. With θ = ω t it is the state function that describes total
angular momentum J = L + S. For iα + jβ + kγ = i, it describes spin only. Finally,
with β = γ = 0, it defines spin in polar projection (s = ± 21 ), as shown in more detail
below. Alternatively Φ is the state function of an electromagnetic photon.
With cos θ = 0 Φ could represent the space-like state function of electric charge,
known as the angular-momentum function, described by the quantum number l in
Schrödinger notation, also known as the three-dimensional spherical harmonics.
The most general rotation of a four-vector represented by q = w + ix + jy + kz is
described by the equation
Jan C. A. Boeyens
q′ = eθ (iα + jβ +kγ ) qe−ϕ (iλ + j µ +kν )
where α 2 + β 2 + γ 2 = λ 2 + µ 2 + ν 2 = 1. When both θ and ϕ are imaginary angles,
the transformation corresponds to the Lorentz transformation [4, 5]. This is mathematical proof that quantum theory and special relativity originate from the same
basis, as the most general solution of (2) must obviously be a quaternion function.
3 Spherical Rotation and Spin
The quaternion that describes rotation through θ about the x-axis, follows as
ei(θ /2) = cos(θ /2) + i sin(θ /2) .
This hypercomplex number is given in matrix form by
0 e−iθ /2
eiθ /2 0
and rotation of 2ω t in time, by
0 e−iω t
e ωt 0
This matrix operates on a two-level object, called a spinor, e.g. {φ1 , φ2 }. A spinor
system that moves in some general direction, x, say:
φ1 e−
e−i(ω t−kx)
φ2 e+
ei(ω t−kx)
(in shorthand notation) is shown to satisfy Schrödinger’s [6] and Dirac’s [7] equations, by forming the derivatives:
∂ 2Φ
−φ1 e−
φ1 e−
2 −φ1 e
= iω
φ2 e−
φ2 e+
φ2 e+
∂ x2
from which follows that
1 ∂ 2Φ
1 ∂Φ
= 2
iω ∂ t
k ∂ x2
Equating m = h̄k2 /2ω , the equation
∂Φ ω 2
h̄ 2
= 2∇ Φ =
∇ Φ
in three dimensions, is identical to Schrödinger’s equation or its complex conjugate,
and describes both matter and antimatter, each with the spin states {φ1 , φ2 }. Writing
Chemistry in Four Dimensions
the wave vector k = 2π /λ = p/h̄, with de Broglie, the quantum condition h̄ω =
p2 /2m follows directly.
The common assertion that electron spin is a strictly relativistic effect [8] needs
modification. As shown before [9] a linearized Schrödinger equation, first order in
all space-time derivatives implies the Pauli equation [10] with the correct value of
the electron gyromagnetic ratio. The 1/2-spin appears, not as a relativistic effect, but
because of the implied four-dimensional space-time. It has the significant advantage
that the spin degrees of freedom are contained in the theory from the beginning,
without being added ad hoc. However, the spin and angular-momentum variables
remain separated.
Since there is no geometrical understanding of spin in three-dimensional space
it is not reflected in standard wave mechanics. A four-dimensional quaternion function, on the other hand, contains a four-component spinor as temporal characteristic,
entangled with the space variables. Spin is generated in four-dimensional quaternion
rotation, which is intrinsically different from the axial rotation of three-dimensional
space, and is also described as spherical rotation2 [7]. Because rotation in spin space
is governed by one-half the angles of rotation in ordinary space, it has the special
property that any entanglement of the spinor with the supporting medium, which develops during rotation of 2π , spontaneously disentangles after a further 2π rotation.
This is the result of the half-angle operation of quaternion rotation. It causes a periodic fluctuation in the medium (vacuum) that surrounds the spinor, and is observed
as spin. The undulation with λ = 2π on the surface of the unit sphere describes spin
as mω = h̄/2,
a unit of angular momentum.
−iω t iωhalf
e t
has an inverse state
and an antispin
. The prodEach spin
iω t
(0 iω t) eiω t
(e−iω t 0) eiω t
=0 ,
show that spin pairing yields a boson of spin zero, while the annihilation of spin and
antispin produces
p a photon of spin h̄ and which satisfies the wave equation (2a). The
condition c = E/2m, (V = 0), now implies conversion of the total mass of matter
and antimatter into a photon of energy E = 2mc2 = h̄ω .
It is ironic that spin, which is the only non-classical attribute of quantum mechanics, is absent from the pioneering formulations of Heisenberg and Schrödinger.
Even in Dirac’s equation the appearance of spin is ascribed by fiat to Lorentz invariance, without further elucidation. In reality, both Lorentz invariance and spin,
representing relativity and quantum mechanics, respectively, are properties of the
quaternion field that underpins both theories.
2 The relationship between spherical rotation and quaternions is visualized pictorially by
Jan C. A. Boeyens
4 The Physical Model
Within the time-like domain of Minkowski space the gravitational field dominates
and a quantum-potential field dominates in the space-like domain. Electromagnetic bosons, which occur in the interface, exhibit both time-like and space-like
behaviour. Events close to, or in, the interface show both quantum and relativistic
behaviour. There is only one classical–non-classical limit as quantum mechanics
and relativity appear as a single theory. Confusion sets in with failure to distinguish between waves and particles as elementary entities and with the possibility of
massive objects moving faster than light in the vacuum. It is the unphysical identification of elementary point particles with quantum waves that gives rise to the
spurious concepts of wavicle, quantum uncertainty, probability density and infinite
self-energy. Elementary wave structures exist without such complications. It is instructive to note that the authoratitive exposition of Unsöld [12] does not involve
probability density.
Standing waves, like gas molecules, exert pressure on the walls of a container.
With the container at rest or in uniform motion the force exerted on any wall is
balanced by that exerted on the opposite wall. If the container is accelerated radiation reflected off the rear wall gains more momentum than that lost by the radiation
which reflects off the front wall. It has been demonstrated [13, 14, 16] that radiation
exerts a net force, which opposes an applied force on the container, such that
m+ 2 a = F .
The radiant energy adds an effective inertial mass ER /c2 to the mass of the container.
In the same way the total mass of an electron may be interpreted as deriving from
internal motion in a phase-locked cavity.
Plane waves with phase velocity vφ form wave packets with group velocity vg ,
such that vφ vg = c2 (= 1/ε0 µ0 ). The phase and group velocities are the same only
for electromagnetic waves with vφ = c. To describe the inertial properties of a wave
packet a mass variable is introduced [17]. A group velocity vg ≶ c defines matter
waves as described by the Elbaz [18] equation:
∇2 u ±
m c 2
1 ∂ 2u
u= 2 2
c ∂t
This is interpreted to show that the curving of space produces time-like (vg < c) as
well as space-like (vg > c) wave packets with real (m) and imaginary (im) mass
[19], respectively known as bradyons and tachyons.
Equation (8), second-order in time, has solutions u(t) and u(−t), defining matter
and antimatter waves respectively. Of the four possible forms only bradyonic matter
waves are directly observed in time-like tangent space. Other types of wave are
encountered in sub-atomic quantum systems.
Chemistry in Four Dimensions
The formation of wave packets depends on the interaction of a complementary
bradyon-tachyon pair. The bradyonic group velocity corresponds to the de Broglie
wavelength of the packet, λdB = h/mvg. The tachyonic component defines the internal structure of the wave packet with Compton wavelength λC = h/mc. The two
components are said [20] to be trapped in a relativistically invariant way. We note
that vb vt = c2 = 1/ε0 µ0 , where the group velocity of the tachyon vt matches the
phase velocity of the bradyon. Such a wave-packet is not dispersive. Essentially the
same model, developed from another perspective, is described by Milo Wolff [21].
An elementary unit of charge (e.g. electron) is considered here as a standing wave
packet created by the interaction between a pair of time-symmetric spherical scalar
waves: a retarded wave radiated by the electron, in balance with an advanced wave
that represents the resultant as the radiation from all other sources in the cosmos
interferes. This is Mach’s principle in wave formalism.
Variability in the internal structure of elementary wave packets is responsible
for the appearance of protons, neutrons and electrons [22], which interact by the
exchange of intermediate massless photons. Chemical interactions, all of this type
[6], are responsible for the growth of massive structures, which on the macro scale
interact gravitationally.
All debates on the interpretation of quantum mechanics must end in confusion,
unless the classical and non-classical models of the world are clearly distinguished.
The classical model is based on the assumption that persistent fragmentation of matter terminates in a set of elementary particles that resist further subdivision, but retain the innate quality to predict the behaviour of matter in the bulk. A non-classical
alternative starts at the other extreme with a featureless plenum that develops periodic wave structures in a topologically closed universe. In projective relativity [23]
there is
... no such thing as a body in space, but matter is an aspect of the space-time structure.
These elementary waves coalesce into bigger units that exhibit all the known properties of ponderable matter.
Classical mechanics analyzes the interaction between particles, and non-classical
mechanics should study the interaction between wave structures. We repeat that the
two models do not refer to classical and non-classical domains – they both model the
same world, but from different points of view. It so happens that at different levels
of aggregation one or the other provides a more convenient description. Attempts
to describe classical structures non-classically, or vice versa, inevitably end up with
illogical conclusions.
Physics has the dilemma of irrefutable evidence for a four-dimensional world, but
a genetic inability among physicists to visualize more than three dimensions. It is
therefore not surprising to find that those instances in which reality is badly distorted
in three-dimensional projection, inevitably lead to convoluted theories, bordering on
the supernatural. Quantum mechanics is a prime example of such a theory. It was
inspired by experimental results that defied explanation based on classical theory. It
was first recognized in the study of microphysical systems, which in time came to
Jan C. A. Boeyens
be seen as deviating from the classical and therefore subject to a new theory, without
relevance in macrophysics.
A more plausible interpretation is that the motion of ponderous objects, projected
into tangent three-dimensional space, differs imperceptably from four-dimensional
reality in the local environment where a classical description suffices. It only becomes an issue for fast-moving objects and where particle mass approaches zero.
The real meaning of both relativity and quantum theory is obscured by their formulation as alternatives to Newtonian mechanics that kick in at some classical limit.
A notable difference between three- and four-dimensional formulation of quantum mechanics occurs in their commutation properties. In R3 commutators that involve a time-dependent variable are found to be non-zero. For instance
[p, q] = (pq − qp) = ih̄ ,
[Jx , Jy ] = ih̄Jz , etc.
Uncertainty relationships such as ∆ p∆ q > h̄, ∆ E ∆ t > h̄, etc, derive directly from
these commutation rules [24].
The unexpected appearance of complex operators is also associated with nonzero commutators and reflects the essential two-dimensional representation in M2
Minkowski space-time. In four-dimensional space-time, H4 , all commutators are
non-zero, as appropriate for wave motion of both quantum and relativity theories.
An important consequence is that local observation has no validity on global extrapolation, as evidenced by the appearance of cosmical redshifts in the curved manifold
and the illusion of an expanding universe.
5 The Chemical Model
Chemical theory, if anything, is distorted even more than physics on projection
from four-dimensional space-time. In electromagnetic and other field theories gauge
particles have mathematically assigned phase factors, which in chemistry is simulated as probability density. Whereas the purely mathematical symbolism suffices
as working models in particle physics, chemistry has the more stringent demand to
deal with extended three-dimensional entities. Even at its lowest level the known
chemical function of an electron, defined as a structureless point particle, becomes
The analysis of dynamic systems in terms of point particles originated with Newton, but the context in which the concept was introduced has been ignored and forgotten. Newton was concerned with the motion of heavenly bodies and their mutual
interaction. To avoid the complication that the moon’s effect at the proximal and
distal surfaces of the earth cannot be the same, the total interaction was assumed
mathematically equivalent to the attraction between objects with all mass concentrated at their respective centres of gravity, which is a point. Earth and moon are
therefore modelled as interacting mass points. What works for planets also works
for apples and electrons. In practical application reference to the centre of gravity
Chemistry in Four Dimensions
was made less frequently and, in time, the actual nature of elementary units of matter
as mass points became generally accepted as physical reality.
When the wave nature of electrons was discovered experimentally Newton’s perception of mass points as centres of gravity had been inactive for so long that, instead
of a wave packet with a centre of gravity, an electron was considered as the physical
union of a particle and a wave. In this approach particle nature is considered to be
of primary importance and wave nature is simulated by a probability function that
specifies particle coordinates. This interpretation was pioneered by de Broglie’s postulate [25] of a particle piloted by a wave. However, the more natural wave model
does the opposite and follows the motion of a wave packet in terms of its centre-ofgravity coordinates. The photoelectric and Compton effects can then be understood
as interaction between waves, rather than particles, as pointed out by Schrödinger
[26]. As stated in the Abstract of [26]:
A definite ψ −distribution in configuration space is interpreted as a continuous distribution
of electricity (and electric current density) in actual space.
Unforunately this level-headed interpretation was never generally accepted. As remarked by David Bohm [27]:
The current formulation of quantum mechanics must be regarded merely as a statistical
algorithm, which provides no conceptual structure in terms of which the movement of individual systems can be understood.
It is only the theory of general relativity that provides any guidance towards
an understanding of the elementary nature of matter, via Einstein’s field equations,
conveniently condensed into the form
Gµν = kTµν
µ , ν = 0, 3 ,
which balances the tensor of space-time curvature against the stress tensor of the
matter-energy field. The vital assertion is that a vanishing curvature tensor, which
implies flat Euclidean space, demands the disappearance of all matter. It is unequivocally inferred that matter appears as a result of the curving of space-time. By the
minimal assumption, known as Occam’s razor, matter must be identified as a distortion of curved space-time.
As a reasonable conjecture we now propose that curved space-time, like an inflexible sheet wrapped around a curved surface, must develop persistent wrinkles
– the elementary units of matter or energy. We envisage flat space-time in featureless undulation that develops elementary wave packets when curved. We recognize
few types of wave packet with internal wave patterns perceived as the characteristic mass, charge, spin and chirality of the four-dimensional elementary units whose
behaviour is prescribed by a potential function according to equation (2).
On projection into three-dimensional tangent space the chiral forms known as
matter and antimatter are distinct and, as c → ∞ [28], obey conjugate forms of the
wave equation (2a), the three-dimensional approximation of (2). To substantiate this
reasoning it is noted that many features of the H atomic spectrum are reproduced
Jan C. A. Boeyens
to good approximation by the solution of (5) for the single electron on a hydrogen
The crunch comes when trying to analyze non-classical many-electron systems
by the same procedure. The mathematics to solve the many-body differential equation does not exist. The popular alternative is to consider each electron as an individual particle and to describe an n-electron system by a probability density in 3ndimensional configuration space. The use of complex variables is tacitly avoided.
The result is a procedure that pretends to simulate a non-classical problem by a
classical model, with an unnecessary complicated structure, designed to resemble
quantum formalism. In this case the statistical model that works for an ideal gas,
fails to explain the behaviour of a many-electron wave.
An alternative procedure is suggested by the recognition of elementary matter as
wave-like distortions of space-time. Unlike free-floating hard particles in a void, the
wave packets envisaged here remain part of the medium and their distribution therefore depends on the symmetry of space-time. The extranuclear charge cloud on an
atom may thus be viewed as the coalescence of electronic wave packets to constitute
a common spherical standing wave around the nucleus. The internal structure of the
wave must reflect the charge distribution as optimized under the nuclear attraction,
like a three-dimensional analogue of the essentially planar solar system.
Optimization by a golden spiral predicts the correct distribution of matter in
the solar sytem [29], with the inference that the spiral structure reflects space-time
topology. Fractal models of the universe, which imply cosmic self-similarity, would
then indicate the same optimization for extranuclear electron density. The resulting
wave structure inevitably carries an imprint of the golden ratio.
Interatomic interaction entails the interference of extranuclear electronic waves.
Constructive interference must occur at specific interatomic distances, which should
correlate with the notion of bond order, numerically related to the golden ratio.
The feasibility of modelling chemical interaction by elementary number theory is
6 Conclusion
Contrary to popular belief new ideas in science are rarely embraced with acclaim.
This was known, in a different context, to Machiavelli who stated 500 years ago
... that there is nothing more difficult to arrange, more doubtful of success, and more dangerous to carry through than initiating changes .... The innovator makes enemies of all those
who prospered under the old order, and only lukewarm support is forthcoming from those
who would prosper under the new. Their support is lukewarm partly from fear of their adversaries, who have the existing laws on their side, and partly because men are generally
incredulous, never really trusting new things unless they have tested them by experience.
In consequence, whenever those who oppose the changes can do so, they attack vigorously,
and the defence made by the others is only lukewarm. So both the innovator and his friends
are endangered together.
Chemistry in Four Dimensions
As an example in science, Schrödinger had to endure such attacks in response to his
wave-mechanical interpretation of quantum effects, as evidenced by his statement,
from the biography by Moore [31]:
Let me say at the outset, that in this discourse, I am opposing not a few special statements
of quantum mechanics held today (1950), I am opposing as it were the whole of it, I am
opposing its basic views that have been shaped 25 years ago, when Max Born put forward
his probability interpretation, which was accepted by almost everybody. . . . I don’t like it,
and I’m sorry I ever had anything to do with it.
Schrödinger tried to introduce a wave model of matter in opposition to the contemporary dogmatic belief in elementary particles. He failed and the debate was closed
for the best part of a century. By now the ranks of those who profit by the preservation of the wave-particle model have swelled by orders of magnitude. Even the few
lukewarm defenders have to rely on commercialized software based on probabilities to generate results considered suitable for publication in the mainstream media.
There is no reason to believe that the 4D analysis proposed here will be treated
any differently. Promoting a new system remains fraught with danger, despite compelling mathematical support.
Current consensus among mathematicians [32] is that normed division algebras,
which could be the basis of high–dimensional physical vector spaces, are restricted
to 1, 2, 4 and 8 dimensions. We interpret this to explain why physical theories in 3 dimensions are plagued by confusing features such as non-commutation and complex
phases that intrude themselves in quantum theory. In four dimensions these would be
natural features. The same argument explains why five-dimensional Kaluza-Klein
models fail as unified field theories. The alternative four-dimensional unification
based on projective geometry in curved space-time [33] and a wave model of matter
works without awkward compacted dimensions and develops gauge invariance in a
natural way.
Some common practices further aggravate the situation. The accepted interpretation of special relativity considers all space outside of the Minkowski time cone
as non-physical. This prejudice obscures the non-local nature of quantum theory
and distorts the common perception of space-time topology. By an equally arbitrary
assumption advanced solutions (in −t) of the three-dimensional wave equation are
rejected. This way all perceptions of space-time chirality, the existence of antimatter
and non-local correlation are lost.
In chemical theory misreading of the superposition principle underpins the
widespread use of real orbitals and basis sets, without any mathematical meaning.
Half a century’s research results in quantum chemistry may well be wasted effort.
But this represents Machiavelli’s profit under the old system. We propose that the
utility of number theory in the description of chemical systems could provide an
escape route from this dilemma.
Acknowledgements I have discussed the mathematics of four dimensions many times with my
colleague Casper Schutte and I gratefully acknowledge his valuable input.
Jan C. A. Boeyens
7 Hypercomplex Numbers
The theory of Special Relativity is conveniently summarized by a set of equations,
known as a Lorentz transformation, which describes all relative motion, including
that of electromagnetic signals, observed to propagate with constant speed c, irrespective of the observer’s state of motion. This transformation,
x − vt
x′ = p
1 − v2/c2
t − vx/c2
t′ = p
1 − v2/c2
relates two frames of reference in relative motion and has no meaning in a system
that separates space and time variables. The resulting four-dimensional space-time
is known as Minkowski space.
From the equation for the moving front of a spherical light wave:
x2 + y2 + z2 = c2t 2 ,
a fourth coordinate is often defined as x0 = ct, redefined by Minkowski as x0 = ict,
such that
x20 + x21 + x22 + x23 = 0
at the wavefront. In terms of the velocity ratio β = v/c and γ = 1/ 1 − β 2, the
Lorentz transformation for uniform relative motion along x1 then takes the simple
form, x′ = Lx i.e.
′ x0
γ −iβ γ
iβ γ γ
This transformation matrix has the same form as an orthogonal rotation matrix
cos φ − sin φ
sin φ cos φ
The Lorentz transformation thereby defines a rotation in the (x,t) plane through an
imaginary angle φ , defined by
cos φ = p
sin φ = p
φ = tan−1 iβ .
As this rotation mathematically interchanges time and space coordinates it means
that they are symmetry related and no longer seperable in the usual way. It is therefore more appropriate to deal with four–dimensional space–time, rather than the traditional three–dimensional space and absolute time. To visualize Minkowski space
it is useful first to review some properties of the complex plane.
Chemistry in Four Dimensions
7.1 Complex Numbers
There is a similarity between two–dimensional vectors and complex numbers, but
also subtle differences. One striking difference is between the product functions of
complex numbers and vectors.
The product of two vectors is either a scalar
z3 = z1 · z2 = z1 z2 cos θ = x1 x2 + y1 y2
or a vector
z3 = z1 × z2 = z1 z2 sin θ = y1 x2 − y2 x1 .
By way of comparison, the product of two complex numbers is
z1 z2 = (x1 + iy1 )(x2 + iy2 )
= (x1 x2 − y1 y2 ) + i(x1 y2 + x2 y1 )
z3 = x3 + iy3 .
The complex product contains two terms, not unlike the scalar and vector products,
from which it differs only in a sign convention.
It is well known that the complex number x + iy is given in polar form by Euler’s
r(cos θ + i sin θ ) = reiθ .
The product
z1 z2 = r1 eiθ1 · r2 eiθ2 = r1 r2 ei(θ1 +θ2 )
is recognized immediately as the rotation of z1 through the angle θ2 and increase
of its length by a factor r2 . To summarize: an operator w = u + iv = Ceiθ , with
C2 = u2 + v2 , θ = tan−1 (v/u), when it multiplies any vector increases its p
length by
a factor C. The magnitude and phase of the complex number x + iy are r = x2 + y2
and θ = tan−1 (y/x). The complex conjugate z∗ = (x + iy)∗ = x − iy has the phase
−θ and magnitude r. The magnitude of a complex quantity is obtained from zz∗ =
(reiθ )(re−iθ ) = r2 , which
is always
real and nonnegative.
α −β
Matrices in the form
, combined by matrix addition and multiplication
β α
are isomorphic with the field of complex numbers (α + iβ ). This way,
(α + iβ ) + (γ + iδ ) = (α + γ ) + i(β + δ ) ,
α −β
β α
γ −δ
δ γ
α + γ −(β + δ )
β +δ α +γ
(α + iβ )(γ + iδ ) = (αγ − β δ ) + i(β γ + αδ )
Jan C. A. Boeyens
α −β
β α
γ −δ
δ γ
αγ − β δ −(β γ + αδ )
β γ + αδ αγ − β δ
re = r
cos θ − sin θ
sin θ cos θ
Euler’s equation in the form:
confirms the geometrical meaning of complex numbers as rotation and enlargement.
0 −1
As an example the complex number i corresponds to the matrix
, which
1 0
represents a counter–clockwise rotation of π /2 about the origin.
Another alternative form of the rotation matrix is obtained by setting α = a2 + b2
and β = 2ab, with a = cos(θ /2), b = sin(θ /2). Hence
α = cos2 (θ /2) − sin2 (θ /2) = cos θ
β = 2 cos(θ /2) sin(θ /2) = sin θ ,
(a, b) =
a2 − b2 −2ab
2ab a2 − b2
cos θ − sin θ
sin θ cos θ
7.2 Quaternions
Extension of the complex formalism to more dimensions suggests the definition of
related hypercomplex numbers. On multiplication of two three–dimensional vectors,
without defining the mathematical properties of unit vectors i, j, k, the formal result
q = (ix1 + jy1 + kz1 )(ix2 + jy2 + kz2 )
= i2 x1 x2 + j2y1 y2 + k2 z1 z2
+ ijx1 y2 + jiy1 x2 + ikx1 z2 + kiz1 x2 + jky1 z2 + kjz1 y2
This expression is rearranged into the same form as a complex product by defining
i2 = j2 = k2 = −1
ij = k , jk = i , ki = j
ji = −k , kj = −i , ik = −j ,
the result first obtained by William Hamilton who defined
q = −(x1 x2 + y1 y2 + z1 z2 ) + i(y1 z2 − y2 z1 ) + j(z1 x2 − z2 x1 ) + k(x1 y2 − x2 y1 )
Chemistry in Four Dimensions
with the rule of composition: i2 = j2 = k2 = i jk = −1.
A hypercomplex number of unit norm
√ can now be defined in the form z = a0 + ai ei ,
where the ei are generalizations of −1, in matrix notation:
i 0
0 1
0 i
, e1 =
, e2 =
, e3 =
0 −i
−1 0
i 0
For i = 1, 3 it is called a quaternion and the matrices are quaternion units. The unit
number 1 represents the unit vector in the fourth dimension.
To represent a quaternion in matrix form, as was done for complex numbers, it is
written as a pair of complex numbers,
q = (u,
v), in the same way that the complex
a −b
number a + ib is written as (a, b) =
. As for complex numbers the product
b a
of (u, v) and its conjugate should be (|u|2 + |v|2 , 0). This result is obtained
by defin
u v∗
ing the conjugate q = (u , −v) and the matrix equivalent of (u, v) as
−v u∗
This way
u v∗
u −v
uu∗ + v∗ v −uv∗ + v∗ u
qq =
−v u∗
v u
−vu∗ + u∗v vv∗ + u∗ u
≡ |u|2 + |v|2 , 0
The modulus of q,
q = (u, v), gives
√ ∗
q q = u2 + v2 . Notice that the determinant of the matrix of
u v∗ 2
−v u∗ = |u| + |v| .
By expanding the quaternion in terms of the complex pair, u = a + ib and v = c + id,
it follows directly that q2 = a2 + b2 + c2 + d 2 . In terms of four quaternion components:
q · q = q20 + q21 + q22 + q23 .
For |u|2 + |v|2 = 1,the product
of the quaternion and its conjugate (or adjoint)
is the unitary matrix
. Unit quaternions (of modulus 1) are therefore seen
to define rotations of C2 , just as the unit complex numbers correspond to rotations
of R2 . The group of unitary transformations of C2 of determinant 1 is the special
unitary Lie group SU(2), which describes the spin function.
In Hamilton’s notation the conjugate of the general quaternion q = a + ib + jc +
kd is q∗ = a − ib − jc − kd, so that the square of the magnitude, |q|2 = q∗ q = a2 +
b2 + c2 + d 2 is a simple extension of the rule for complex numbers.
As with complex numbers the product of two quaternions
(p = α + iβ + jγ + kδ )q = (α a − β b − γ c − δ d) + i(α b + β a + γ d − δ c)
+ j(α c − β d + γ a + δ b) + k(α d + β c − γ b + δ a)
Jan C. A. Boeyens
is another quaternion. Quaternion multiplication is clearly not commutative, as pq 6=
By expanding the exponential and using the rules for multiplication of i, j, k, it
can be shown that, for α 2 + β 2 + γ 2 = 1, a quaternion equivalent of Euler’s equation
for the imaginary exponent is obtained as:
eθ (iα + jβ +kγ ) = cos θ + sin θ (iα + jβ + kγ ) .
This exponential is of unit length; the exponent represents a space vector of length
θ , with direction as given by the direction cosines α , β , γ .
Any quaternion can now be represented in the form Qeθ (iα + jβ +kγ ) , where Q is
the length of a four vector with direction as fixed by θ , α , β , γ . By analogy with
the complex exponential it is anticipated that this operator (with Q = 1) describes a
rotation. The argument on which it operates is also a quaternion. In three dimensions
the vector f = ix + jk + kz, in quaternion notation, is rotated by an angle θ about an
axis of direction cosines α , β , γ into f ′ according to:
f ′ = ix′ + jy′ + kz′ = e(θ /2)(iα + jβ +kγ ) f e−(θ /2)(iα + jβ +kγ ) = q · f · q−1 ,
where α 2 + β 2 + γ 2 = 1.
As in (6) we define the rotation matrix by a pair of complex numbers (u, v),
q0 = a cos(θ /2)
q1 = i sin(θ /2)
q2 = j sin(θ /2)
q3 = k sin(θ /2) ,
which define the three-dimensional rotation matrix:
 2
q0 + q21 − q22 − q23 2q1q2 − 2q0q3 2q1 q3 + 2q0q2
R(q) =  2q1 q2 + 2q0q3 q20 − q21 + q22 − q23 2q2 q3 − 2q0q1 
2q1 q3 − 2q0q2 2q2q3 + 2q0q1 q20 − q21 − q22 + q23
Each row (or column) has unit magnitude, e.g.:
+ (2q1 q2 − 2q0q3 )2 + (2q1 q3 + 2q0q2 )2 = (q · q)2 = 1 ,
and the total matrix R f = q f q−1 is orthogonal, as required for 3D rotation.
Quaternions describe rotation in any number of dimensions from 1 to 4. It is
straightforward to demonstrate3 that for the special case of rotation about the x–
q20 + q21 − q22 − q23
f ′ = e(θ /2)i (ix + jy + kz)e−(θ /2)i
Using: sin(θ /2) = ±
(1 − cos θ )
, cos(θ /2) = ±
(1 + cos θ )
Chemistry in Four Dimensions
= ix + j(y cos θ − z sin θ ) + k(y sin θ + z cos θ )
 
(i j k) 1 0
 0 cos θ − sin θ   y 
0 sin θ cos θ
i.e. a rotation through the imaginary angle θ about the x–axis in the (y, z)–plane.
The right–to–left order of matrix multiplication is important. Proof of the general
case only involves more algebra. Symbolically f ′ = ζ f ζ ∗ . By a second rotation
f ′′ = η f ′ η ∗ = ηζ f ζ ∗ η ∗ .
The final result depends on the order in which the operations are applied, because
of the fact that the quaternions η and ζ do not commute. The quantity Q is called the
tensor (stretcher) and the exponential is called the versor (turner) of the operator.
In four-dimensional rotation the argument of the operation is the full quaternion
four-vector, vµ = (v0 , vi ), rather than the three-vector f with v0 = 0, considered
before. On working out the full rotation matrix it turns out to decompose into a pair
of 3D rotations, such as (8), indicating that the four-dimensional rotation amounts
to double covering of the underlying space of 3D rotations [35].
Jan C. A. Boeyens
Dirac PAM (1928) Proc Roy Soc A 117:610.
Taub AH, Veblen O, von Neumann J (1934) Proc Nat Acad Sci US 20:383
Dirac PAM (1938) Proc Roy Soc A167:148.
Morse PM, Feshbach H (1956) Methods of Theoretical Physics. McGraw-Hill, NY.
Misner CW, Thorne K, Wheeler JA (1973) Gravitation. Freeman, San Francisco, p. 1143.
Boeyens JCA (2008) Chemistry from First Principles.
Batty-Pratt EP, Racey TJ (1980) Int J Theor Phys 19: 437.
Carmeli M (1985) Nuovo Cimento Lett 42:67.
Lévy-Leblond J-M (1967) Comm math Phys 6:286.
Pauli W (1927) Z für Physik 43:601.
Kauffman LH (1987) On Knots, Princeton Univ Press, NJ.
Unsöld A (1927) Ann. d. Physik 82:355.
Bohm D (1996) The Special Theory of Relativity. Routledge, London.
Einstein A. (1911) On the influence of gravitation on the propagation of light. Translation of
German original, in [15].
Perrett P, Jeffery GB (1952) (translators). The Principle of Relativity. Dover, N.Y.
Jennison RC, Drinkwater AJ (1977) J Phys A 10:167.
Elbaz C (1986) Phys Lett 114A:445.
Elbaz C (1983) CR Acad Sc Paris 297:455.
Feinberg G (1967) Phys Rev 159:1089.
Corben HC (1977) Lett Nuovo Cimento 20:645.
Wolff M (1993) Physics Essays 6:181;
— (1995) Galilean Electrodynamics Sept/Oct:83;
— (1997) Frontier Perspectives 4:67.
Jennison RC (1978) J Phys A: Math Gen 11:1525.
Mc Crea WH (1937) Supplement to Nature of June 12, 139:1002. (Quoting Veblen and Whitehead)
Bohm D (1951) Quantum Theory. Dover reprint, Prentice-Hall, NJ.
de Broglie L (1926) Nature 118:441.
Schrödinger E (1978) Collected papers on Wave Mechanics, Translated from the 1927 German edition, Chelsea, NY, p. 124.
Bohm D, Bub J (1966) Revs Mod Phys 38:453.
Oppenheimer JR (1930) Phys Rev 35:461.
Boeyens JCA (2009) Physics Essays 22:493.
Machiavelli N (1975) The Prince. Penguin Books, Hammondsworth. Translated from the
1929 Italian edition by George Bull.
Moore W (1989) Schrödinger – Life and Thought. University Press, Cambridge.
Gowers T (ed.) (2008) The Princeton Companion to Mathematics. University Press, Princeton.
Veblen O (1933) Projektive Relativitätstheorie. Springer, Berlin. English translation in [34].
Boeyens JCA (2010) Chemical Cosmology.
A.J. Hanson, Visualizing Quaternions, Elsevier, Amsterdam, (Morgan Kaufman), (2006).
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF