The Conformal Group in Biquaternionic Form
Introduction
The companion articles have established the homogeneous Lorentz group inside the biquaternion algebra exactly. The unit-norm biquaternions form $SL(2,\mathbb{C})$, the double cover of the restricted Lorentz group $SO^+(1,3)$, and rotor conjugation $$ \tilde{Q}\;\longmapsto\;\tilde{U}\,\tilde{Q}\,\tilde{U}^{*}, \qquad \tilde{Q}\in\mathbb{B}_{\Phi},\quad \tilde{U}\tilde{U}^{\natural}=e_t, $$ is their action on a four-dimensional real slice of the algebra; the corpus displays it on the material slice $\mathbb{B}_{\Phi}$, and nothing below depends on that choice. The Poincaré article then showed that the translations are of a different kind: a shift is affine, not linear, so no rotor generates it, and the restricted Poincaré group is carried as pairs $(\tilde{U},\tilde{R}_{\Phi,0})$ with $\tilde{R}_{\Phi,0}\in\mathbb{B}_{\Phi}$, a semidirect product in which the algebra supplies the homogeneous factor and its action on the shift but not the shift itself. That article closed with an explicit open question — whether an enlarged algebra could absorb translations multiplicatively, and whether a conformal extension could realize them "as parabolic products of generalized inversions rather than as pure rotors."
This article takes up the conformal end of that question. The conformal group of Minkowski space is the fifteen-parameter group generated by the Poincaré group together with the dilations and the special conformal transformations; its connected component is $SO^+(2,4)$, double-covered by $SU(2,2)$. The question is how much of it the biquaternion algebra $\mathbb{B}$ carries.
The answer is a boundary rather than an embedding, and the boundary falls in a definite place. The algebra carries the conformal structure: the zero set of its biquaternion norm is the light cone of each sector, and the algebra inverse, composed with parity, is the conformal inversion, from which the special conformal transformations are built by composition. The algebra does not carry the conformal group: it is eight real dimensions and its unit-norm subgroup six, while the conformal group is fifteen; six of the fifteen generators are elements of $\mathbb{B}$, the dilation is realized finitely by a non-invertible element, and the remaining eight — translations and special conformal transformations — are not elements of $\mathbb{B}$ at all, none of them linear on a four-dimensional slice. The standard algebraic homes of the conformal group are larger, and §Larger Homes in the Literature gives their sizes; the smallest of them is twice the eight real dimensions of $\mathbb{B}$. The article states these boundaries and does not smooth them over.
The compactification is the one part of the picture the article does build, and it is built on a form the earlier sections do not use. The space on which the conformal group acts globally is the $*$-unitary group $\mathrm{U}(2)=\{\tilde Q:\tilde Q\tilde Q^{*}=e_t\}$ — compact, of real dimension $4$, and not the norm-units $SL(2,\mathbb{C})$ — and the two sectors are its Cayley charts, poled each in the other. The boundary is a slice of the zero-divisor cone, and the sphere at its base turns out to be the Bloch sphere of the informational sector read as generators. That construction is The Compactification from the Star; it does not put the group inside the algebra, and it does not force a compact universe.
The conventions are those of the read-list companions: the algebra, the invertibles, the sectors, the biquaternion norm, the four forms, the gradient and its conjugate, the d'Alembertian and the trace are theirs, and none is restated here. Every derivative acting on an algebra-valued field in the article is the corpus gradient $\tilde\nabla_\Phi$ of Operator and Calculus Conventions, §The Gradient, and no other letter names it; the coordinate derivatives $\partial/\partial R_\mu$ are declared in the section that uses them. $\Box_\Phi$ is the quaternionic norm of the gradient in Operator and Calculus Conventions, and the material-sector reading of that one operator is fixed in §The Quaternionic Bilinear Product, Its Form and Its Norm. Throughout, $c$ is the speed of light in the medium, and the temporal coordinate of $\mathbb{B}_{\Phi}$ is the corpus's $Q_t=i q'_t$.
The Conformal Group and Its Generators
A conformal transformation of a pseudo-Riemannian manifold is a diffeomorphism whose pullback of the metric is a positive multiple of the metric, $$ \xi^*g=\Omega^2\,g,\qquad \Omega>0 . $$ It need not preserve lengths or the interval; it preserves the null directions. The conformal group of Minkowski space is $O(2,4)$, of dimension $15$; the connected component is $SO^+(2,4)$, and its double cover is $SU(2,2)$. The conformal group contains the Poincaré group as the subgroup generated by Lorentz transformations and translations, and it is completed by the dilations and the special conformal transformations.
The generators fall into four families, fifteen in all. Here $x$ and $a$ are the coordinate and translation four-vectors, with components $Q_\mu$ and $a_\mu$: $x^2=\sum_\mu Q_\mu^2=-{r'_{t}}^2+\vec r^2$, $a\cdot x=\sum_\mu a_\mu Q_\mu$, and $\Lambda$ is the $4\times4$ Lorentz matrix, read with the first index the row and the second the column.
| Family | Symbol | Count | Finite action on $x$ |
|---|---|---|---|
| Lorentz | $M_{\mu\nu}$ | 6 | $x\mapsto\Lambda x$ |
| Translations | $P_\mu$ | 4 | $x\mapsto x+a$ |
| Dilation | $D$ | 1 | $x\mapsto\lambda x$ |
| Special conformal | $\mathcal{K}_\mu$ | 4 | $x\mapsto\dfrac{x+a\,x^2}{1+2a\cdot x+a^2x^2}$ |
The symbol $\mathcal{K}_\mu$ is used for the special conformal generator because the literature's usual $K_\mu$ is taken in this series by the boost generator $B_{\mathrm{boost},\eta}=i\vec e_\eta$ of the Lorentz-group companion, which carries a spatial index and is an element of the algebra. The distinction is not merely typographical below.
A concrete realization of the fifteen generators is by vector fields on the coordinates. Because the generators are operators on spacetime and not elements of $\mathbb{B}$, the display is written on the coordinates: $Q_\mu$ are the coordinates and $Q_\mu=q_\mu+iq'_\mu$, the material reading being $Q_t=i q'_t$ and $Q_\eta=q_\eta$, the derivatives are written $\partial/\partial R_\mu$, a repeated index is summed, and no index is raised or lowered. The algebra element of $\mathbb{B}_{\Phi}$ is $\tilde Q=Q_te_t+Q_x\vec e_x+Q_y\vec e_y+Q_z\vec e_z$.
$$ P_\mu=\partial/\partial R_\mu,\qquad D=\sum_\mu Q_\mu\partial/\partial R_\mu,\qquad M_{\mu\nu}=Q_\mu\partial/\partial R_\nu-Q_\nu\partial/\partial R_\mu, $$ $$ \mathcal{K}_\mu=2Q_\mu\sum_\nu Q_\nu\partial/\partial R_\nu-x^2\partial/\partial R_\mu, $$ with $x^2=\sum_\mu Q_\mu^2=-{r'_{t}}^2+\vec r^2$. In this basis the structure constants are real, matching the convention of the Lorentz-group companion, whose brackets are $[\vec e_j,\vec e_\eta]=2\epsilon_{j\eta l}\vec e_l$, $[\vec e_j,B_{\mathrm{boost},\eta}]=2\epsilon_{j\eta l}B_{\mathrm{boost},l}$, $[B_{\mathrm{boost},j},B_{\mathrm{boost},\eta}]=-2\epsilon_{j\eta l}\vec e_l$ for $\vec e_\eta$, $B_{\mathrm{boost},\eta}=i\vec e_\eta$. The brackets are $$ [D,P_\mu]=-P_\mu,\qquad [D,\mathcal{K}_\mu]=\mathcal{K}_\mu, \qquad [D,M_{\mu\nu}]=0, $$ $$ [P_\mu,P_\nu]=0,\qquad [\mathcal{K}_\mu,\mathcal{K}_\nu]=0, \qquad [P_\mu,\mathcal{K}_\nu]=2(\delta_{\mu\nu}D-M_{\mu\nu}), $$ $$ [M_{\mu\nu},P_\rho]=\delta_{\nu\rho}P_\mu-\delta_{\mu\rho}P_\nu, \qquad [M_{\mu\nu},\mathcal{K}_\rho]=\delta_{\nu\rho}\mathcal{K}_\mu-\delta_{\mu\rho}\mathcal{K}_\nu, $$ together with the standard Lorentz brackets for $[M_{\mu\nu},M_{\rho\sigma}]$.
Where the generators live. The Lorentz generators $\vec e_\eta$ and $B_{\mathrm{boost},\eta}=i\vec e_\eta$ are elements of $\mathbb{B}$, and the companion article builds the whole finite-dimensional Lorentz representation theory from them. The other nine generators above are differential operators on the coordinates, not elements of $\mathbb{B}$. This split — six inside the algebra, nine outside — is the theme of the sections that follow; the dilation is a partial exception, expressible by a non-invertible biquaternion, and the translations and special conformal transformations are the genuinely external part.
How the Group Entered Physics
The fifteen-parameter group did not enter physics as the conformal group. In 1909–10 Harry Bateman and Ebenezer Cunningham, following Sophus Lie's work of 1871 on the relation between sphere transformations with an imaginary radius coordinate and four-dimensional conformal transformations, showed that setting the fourth coordinate to $i r'_t$ turns Lie's sphere transformations into conformal transformations of spacetime, and that the quadratic form and Maxwell's equations are covariant under them, irrespective of a scale $\lambda$; Bateman called them spherical wave transformations. The covariance is not universal: Maxwell's equations are conformally invariant in four dimensions, while the massive laws are not, and this is the physical content of the split above. The conformal group is the symmetry of the massless electrodynamic laws; the Poincaré group is the symmetry of every relativistic law in an inertial frame. In the parametrisation of those authors the difference is a scale, and fixing $\lambda = 1$ leaves the ten-parameter Poincaré group as the subgroup of the fifteen-parameter conformal group. (The historical literature sometimes writes "Lorentz group" where the ten-parameter subgroup is meant; the parameter count makes it the Poincaré group.) The global conformal group of the Riemann sphere, the Möbius group $PSL(2,\mathbb{C})$, is isomorphic to the proper orthochronous Lorentz group $SO^+(1,3)$ — the same coincidence one dimension down.
The Null Cone: Conformal Structure the Algebra Already Has
The conformal group is the group that preserves null directions, and in this framework "null" is already algebraic. A displacement $\tilde{Q}\in\mathbb{B}_{\Phi}$ is null exactly when $$ \langle\tilde{Q}\rangle_{\natural} = \langle\tilde{Q},\tilde{Q}\rangle_{\natural}=0 \qquad(\text{equivalently }\tilde{Q}\tilde{Q}^{\natural}=0), $$ and an element of $\mathbb{B}$ is a zero divisor exactly when its norm vanishes. The light cone of Minkowski space is therefore the material sector's slice of the zero-divisor cone of the algebra, whose geometry is the subject of the companion article on the null quadric and projective geometry; here only its invariance is needed. The algebra contains the object that the conformal group preserves, even though, as the later sections show, it does not contain the group.
The four families of generators act on the zero set of $\langle\cdot\rangle_{\natural}$ as follows; each statement follows from the algebra's product, inverse and addition, and each was checked on explicit elements.
Lorentz. Rotor conjugation preserves $\langle\cdot\rangle_{\natural}$ pointwise, $$ \langle\tilde{U}\tilde{Q}\tilde{U}^{*}\rangle_{\natural}=\langle\tilde{U}\rangle_{\natural}\,\langle\tilde{Q}\rangle_{\natural}\,\langle\tilde{U}^{*}\rangle_{\natural}=\langle\tilde{Q}\rangle_{\natural}, $$ since $\langle\tilde{U}\rangle_{\natural}=1$ and $\langle\tilde{U}^{*}\rangle_{\natural}=1$. These are the isometries: the conformal transformations that leave $\langle\cdot\rangle_{\natural}$ unchanged rather than scaling it. The dilations and the special conformal transformations do not.
Dilation. The dilation $\tilde{Q}\mapsto\lambda\tilde{Q}$ scales the form, $\langle\tilde{Q}\rangle_{\natural}\mapsto\lambda^2\langle\tilde{Q}\rangle_{\natural}$. It preserves the zero set and changes the scale, which is exactly the conformal condition with constant factor.
Translation. A shift does not preserve $\langle\cdot\rangle_{\natural}$ pointwise, but a shift by $\tilde Q_3$ preserves the interval between two points, $$ \langle(\tilde Q_1+\tilde Q_3)-(\tilde Q_2+\tilde Q_3)\rangle_{\natural}=\langle\tilde Q_1-\tilde Q_2\rangle_{\natural}, $$ so it maps null displacements to null displacements. This is the parent article's observation that the interval survives translation.
Inversion. The inversion defined in the next section but one satisfies $\langle I(\tilde{Q})\rangle_{\natural}=1/\langle\tilde{Q}\rangle_{\natural}$. It preserves the zero set, where it is defined, and inverts the scale.
So all four families preserve the null cone, and the invariant object of the conformal group is native to the algebra: it is the zero set of $\langle\cdot\rangle_{\natural}$. Two qualifications belong here rather than at the end. The inversion is undefined where $\langle\tilde{Q}\rangle_{\natural}=0$, that is, on the light cone of the origin; the conformal group is only globally defined on the conformal compactification, and on $\mathbb{B}_{\Phi}$ its transformations are partly defined rational maps. This article works with those rational maps, and the compactification on which they become global is built from the star in The Compactification from the Star below. And the invariance of the cone is a statement about the object: it does not by itself produce a group action inside $\mathbb{B}$, which is the substance of the strain.
Dilations and the Non-Unit Biquaternion
The dilation $$ \tilde{Q}\;\longmapsto\;\lambda\,\tilde{Q},\qquad \lambda>0, $$ is scalar multiplication on the material sector. It can be written by multiplying on both sides by a central element, $$ \tilde{Q}\;\longmapsto\;\sqrt{\lambda}\,e_t\;\tilde{Q}\;\sqrt{\lambda}\,e_t , $$ and this displays both what the algebra provides and what it withholds. Three points, each recomputed on explicit elements.
The dilation element is not invertible-norm biquaternion. The central element $\sqrt{\lambda}e_t$ has biquaternion norm $\langle\sqrt{\lambda}e_t\rangle_{\natural}=\lambda$, which differs from $1$ for $\lambda\ne1$, so it lies outside $SL(2,\mathbb{C})$. The rotor group acts by isometries and preserves $\langle\cdot\rangle_{\natural}$; the dilation scales $\langle\cdot\rangle_{\natural}$, so it cannot be a member of that group. The dilation is realizable by the algebra but is not of the group of the algebra.
The dilation is not an algebra automorphism. Multiplication by a scalar is linear but not multiplicative: $\lambda(\tilde{Q}\tilde{Q}_1)$ is not $(\lambda\tilde{Q})(\lambda\tilde{Q}_1)$ for $\lambda\ne1$, checked by direct computation on explicit elements. So the dilation, unlike the rotations, cannot be read as a relabelling of the algebra's own product.
The dilation generator is not an inner derivation. Extended to all of $\mathbb{B}$, the dilation is $\lambda$ times the identity, of trace $8\lambda$; every inner derivation $\mathrm{ad}_{\tilde Q_4}=[\tilde Q_4,\cdot\,]$ of the eight-dimensional algebra is scalar-free, as a direct computation in a real basis confirms for several random $\tilde Q_4$. So no element of $\mathbb{B}$ generates the dilation by commutator. This is the precise sense in which one of the fifteen parameters sits half inside the algebra: expressible, but not a group element and not an inner derivation.
Inversion and the Special Conformal Transformations
The special conformal transformations are the part of the conformal group that is nonlinear on spacetime, and they are generated by the inversion.
Inversion. For $\tilde{Q}\in\mathbb{B}_{\Phi}$ with $\langle\tilde{Q}\rangle_{\natural}\ne0$, define $$ I(\tilde{Q})=\frac{\tilde{Q}}{\langle\tilde{Q}\rangle_{\natural}}. $$ This is the four-vector form of $Q_\mu\mapsto Q_\mu/x^2$, and it is algebraic: it is the quaternion conjugate followed by the algebra inverse, $$ \boxed{\;I(\tilde{Q})=(\tilde{Q}^{\natural})^{-1}\;}, $$ because $\langle\tilde{Q}^{\natural}\rangle_{\natural}=\langle\tilde{Q}\rangle_{\natural}$ and $(\tilde{Q}^{\natural})^{\natural}=\tilde{Q}$; the norm is a central scalar, so $(\tilde{Q}^{\natural})^{-1}=(\tilde{Q}^{\natural})^{\natural}/\langle\tilde{Q}^{\natural}\rangle_{\natural}=\tilde{Q}/\langle\tilde{Q}\rangle_{\natural}$. Two properties follow and were verified: $I^2=\mathrm{id}$ where defined, and $$ \boxed{\;\langle I(\tilde{Q})\rangle_{\natural}=\frac{1}{\langle\tilde{Q}\rangle_{\natural}}\;}, $$ so $I$ maps the light cone to itself and is a conformal map with the reciprocal scale factor. Unlike the Lorentz, dilation and translation parts, the inversion is a rational map of the module: it divides by $\langle\tilde{Q}\rangle_{\natural}$.
There is a small but real distinction to record. The algebra's own inverse is $$ \tilde{Q}^{-1}=\frac{\tilde{Q}^{\natural}}{\langle\tilde{Q}\rangle_{\natural}}, $$ which differs from $I$ by the spatial reflection $(r_x\,\vec e_x+r_y\,\vec e_y+r_z\,\vec e_z)\mapsto-(r_x\,\vec e_x+r_y\,\vec e_y+r_z\,\vec e_z)$, that is, by quaternion conjugation on $\mathbb{B}_{\Phi}$: $$ \tilde{Q}^{-1}=P\bigl(I(\tilde{Q})\bigr),\qquad P:\,(i r'_t,r_x\,\vec e_x+r_y\,\vec e_y+r_z\,\vec e_z)\longmapsto(i r'_t,-(r_x\,\vec e_x+r_y\,\vec e_y+r_z\,\vec e_z)). $$ The map $P$ is an improper Lorentz transformation, so it is already in the conformal group; the group generated by either choice is the same. But the two maps are different, and it is $I=(\,\cdot{}^{\natural})^{-1}$, quaternion conjugation followed by inversion, that is the standard inversion $Q_\mu\mapsto Q_\mu/x^2$, not the naive algebra inverse. Writing "inversion is the biquaternion inverse" without this qualification is false; the correction is exactly parity.
Special conformal transformations. A translation placed between two inversions is a special conformal transformation: $$ \boxed{\;\tilde{Q}_2=I\bigl(I(\tilde{Q})+\tilde Q_3\bigr) =\frac{\tilde{Q}+\tilde Q_3\,\langle\tilde{Q}\rangle_{\natural}} {1+2\langle\tilde{Q},\tilde Q_3\rangle_{\natural}+\langle\tilde Q_3\rangle_{\natural}\langle\tilde{Q}\rangle_{\natural}}, \qquad \tilde Q_3\in\mathbb{B}_{\Phi}\;}, $$ where $$ \langle\tilde Q_1,\tilde Q_2\rangle_{\natural}=\tfrac12\bigl(\langle\tilde Q_1+\tilde Q_2\rangle_{\natural}-\langle\tilde Q_1\rangle_{\natural}-\langle\tilde Q_2\rangle_{\natural}\bigr)=\sum_\mu Q_{1\mu} Q_{2\mu} $$ is the polar form of $\langle\cdot\rangle_{\natural}$ — the complex bilinear dot product, which on $\mathbb{B}_{\Phi}$ is the Minkowski inner product. In components, with $x^2=\langle\tilde{Q}\rangle_{\natural}$ and $a\cdot x=\langle\tilde{Q},\tilde Q_3\rangle_{\natural}$, this is the standard $$ Q_{2\mu}=\frac{Q_\mu+a_\mu x^2}{1+2a\cdot x+a^2x^2}. $$ The identity of the two expressions was checked directly on random four-vectors $\tilde{Q},\tilde Q_3$, and the two displayed formulas agree, to $10^{-13}$ in the $Q_t=i q'_t$ reading of the components. This answers, in part, the open question of the Poincaré article: the translation does appear as a product of generalized inversions, but the product is the special conformal transformation, not the translation, and it is not linear in $\tilde{Q}$. The translation itself remains an additive shift.
Two qualifications belong to the formula. First, it is a fractional (nonlinear) expression in $\tilde{Q}$, not the two-sided linear map $\tilde{Q}\mapsto\tilde{U}\tilde{Q}\tilde{U}^{*}$ that carries the Lorentz group; the algebraic operation that produces it is inversion, and inversion is a rational map, not a linear one. Second, it is only partly defined: the inversion is singular on the null cone of the origin, and the denominator $1+2a\cdot x+a^2x^2$ vanishes on a cone as well. Globally, these are transformations of the conformal compactification; on $\mathbb{B}_{\Phi}$ they are defined off a null set.
The generator. Expanding the special conformal formula to first order in $a$, $$ \delta Q_\mu=a_\mu x^2-2Q_\mu(a\cdot x)=-a_\nu\,\mathcal{K}_\nu Q_\mu, $$ with $\mathcal{K}_\nu=2Q_\nu\sum_\rho Q_\rho\partial/\partial R_\rho-x^2\partial/\partial R_\nu$. This is the vector field of the generator table; the algebra produces the finite transformation but not an element of $\mathbb{B}$ that generates it.
The Compactification from the Star
Two passages above deferred the compactification: the inversion is singular on the light cone, and the special conformal transformations are global only on a space this article had not built. The space is not far away. It is the $*$-unitary group $\mathrm{U}(2)$, and it is in the algebra already, with the two sectors serving as its charts. The account follows the treatment of compactified Minkowski space by the Cayley transform [Jadczyk, On Conformal Infinity and Compactifications of the Minkowski Space, arXiv:1008.4703], read in the algebra's own objects; the source's arrays are named where they are quoted, and nowhere else.
The construction has a scope, stated here rather than at the end. The algebra contains the space on which the conformal group acts globally, and the two sectors are charts on it. It contains no reason to prefer that space to the non-compact one: the compactification adds a boundary, and what the boundary means is not settled by the algebra.
Two Unit Conditions, Not One
An element is $*$-unitary when $\tilde{Q}\tilde{Q}^{*}=e_t$, and of unit norm when $\langle\tilde{Q}\rangle_{\natural}=1$. These are the two unit conditions of Norm and Pairing Conventions, §The Four Algebraic Norms and the $*$-Unitary Convention, and they belong to different forms: the Hermitian form, which is positive definite on the whole algebra, and the complex symmetric bilinear form $\langle\cdot\rangle_{\natural}$. The star-unitary elements are the group $\mathrm{U}(2)$ of that section, compact and of real dimension $4$, against $SL(2,\mathbb{C})$ for the unit-norm elements, of real dimension $6$ and not compact; neither contains the other. The two must not be merged: the compactification of this section is built on the star, because the star is the involution that defines the two sectors, and no statement in it is a statement about $SL(2,\mathbb{C})$. Every chart value below is star-unitary, and $\tilde{Q}\tilde{Q}^{*}=e_t$ is the whole of what that says.
The group is generated by the two elementary groups of invertibles of the algebra, the unit circle of the centre and the unit real quaternions. Every $*$-unitary element factors as a phase times a unit real quaternion, $$ \tilde{Q}=c\,q,\qquad |c|=1,\qquad q\in\mathbb{B}_{H},\quad q q^{*}=e_t, $$ and the correspondence is two-to-one, $(c,q)\sim(-c,-q)$, $$ \mathrm{U}(2)\;\cong\;\bigl(\mathrm{U}(1)\times SU(2)\bigr)/\{\pm(e_t,e_t)\}\;\cong\;(S^{1}\times S^{3})/\mathbb{Z}_{2}, $$ the relation the corpus already records for the internal group, with the manifold the source gives for the compactified space. Verified by factoring fifty random star-unitary elements: the phase is a square root of the biquaternion norm, $\langle c\,q\rangle_{\natural}=c^{2}$, and the remaining factor $q=\tilde{Q}/c$ has unit norm, is real-quaternionic and satisfies $q q^{*}=e_t$, to $6\times10^{-16}$. The double cover of the compactified space is therefore the product of the unit circle of $\mathbb{B}_{C}$ and the unit sphere of $\mathbb{B}_{H}$.
The Cayley Unit of Each Sector
The chart that realises a sector inside its compactification is the Cayley transform, taken at a $*$-unitary pole, $$ \tilde{U}_{\mathrm{C}}=(\tilde{Q}-p)(\tilde{Q}+p)^{-1}. $$ Two poles are used, one per sector, and each pole lies in the sector that is not being charted: $$ p=ie_t \text{ for } \tilde{Q}\in\mathbb{B}_{\Psi},\qquad p=e_t \text{ for } \tilde{Q}\in\mathbb{B}_{\Phi}, $$ and the chart is singular at $\tilde{Q}=-p$, which lies outside the charted sector too. The first is the classical chart. For $\tilde{Q}\in\mathbb{B}_{\Psi}$ the element is Hermitian, and the chart value is the source's $u(X)$ term for term, with the element $\tilde Q$ in the place of the source's array $X$. It is singular exactly where $\tilde{Q}+ie_t$ is a zero divisor, which never happens on $\mathbb{B}_{\Psi}$. The chord values below are therefore the source's Eq. (6) term for term. The second chart is the same formula applied to the exchanged element, and the two chart values at that element are reciprocal, as shown below.
Both charts land in the $*$-unitary group, $\tilde{U}_{\mathrm{C}}\tilde{U}_{\mathrm{C}}^{*}=e_t$, checked on fifty elements of each sector to $7\times10^{-16}$, and both invert by $$ \tilde{Q}=p\,(e_t+\tilde{U}_{\mathrm{C}})(e_t-\tilde{U}_{\mathrm{C}})^{-1}, $$ the round trip returning the starting element to $3\times10^{-15}$. A chart never fails on the sector it charts. In the $\mathbb{B}_{\Psi}$ chart the first chord below carries the denominator $1-\langle\tilde{Q}\rangle_{\natural}-2ix^0$; in the $\mathbb{B}_{\Phi}$ chart the same computation gives $4/(1+\langle\tilde{Q}\rangle_{\natural}+2ix^0)$, with $\langle\tilde{Q}\rangle_{\natural}=r^2-(x^0)^2$ the material interval. Either denominator vanishes only if $x^0=0$ and $r^2=-1$, which is impossible. So no point of a sector is its own infinity, and the infinity of a sector is approached only from outside it.
Infinity Is a Zero-Divisor Chord
The chord from $e_t$ to the chart value is the element $e_t-\tilde{U}_{\mathrm{C}}$, and it is a biquaternion, so the object whose vanishing locates the boundary is its norm and not an array. The source prints a determinant; the determinant of the source's array of an element is that element's biquaternion norm, which is the dictionary $\mathsf{M}_2$ of Matrix Representations Conventions, and the article needs no array beyond this sentence. The chart therefore fails exactly where the chord from $e_t$ to the chart value is a zero divisor, equivalently where $\tilde{Q}$ sits on the light cone through the pole, $\langle\tilde{Q}+p\rangle_{\natural}=0$. The boundary is the corpus's zero-divisor cone seen from one point of the other sector, not a new object.
The second chord carries the light cone of the origin. The light cone $\langle\tilde{Q}\rangle_{\natural}=0$ is exactly the locus $$ \langle e_t+\tilde{U}_{\mathrm{C}}\rangle_{\natural}=0,\qquad \langle e_t-\tilde{U}_{\mathrm{C}}\rangle_{\natural}\ne0, $$ verified in both directions to $3\times10^{-15}$. In the chart the two chords have the explicit values $$ \langle e_t-\tilde{U}_{\mathrm{C}}\rangle_{\natural}=\frac{4}{1-\langle\tilde{Q}\rangle_{\natural}-2i h_t},\qquad \langle e_t+\tilde{U}_{\mathrm{C}}\rangle_{\natural}=\frac{-4\langle\tilde{Q}\rangle_{\natural}}{1-\langle\tilde{Q}\rangle_{\natural}-2i h_t}, $$ for $\tilde{Q}=h_t\,e_t+i(h'_x\,\vec e_x+h'_y\,\vec e_y+h'_z\,\vec e_z)\in\mathbb{B}_{\Psi}$, with $h_t$ the time coordinate and $\langle\tilde{Q}\rangle_{\natural}=h_t^2-(h'_x)^2-(h'_y)^2-(h'_z)^2$, both checked to $10^{-15}$. The denominator never vanishes, the first chord never vanishes, and the second vanishes exactly on the light cone. This is the source's Eq. (6), its determinants read as the norms of the two chords. The source's quadratic form is $q(x)=-(x^0)^2+r^2$, which is the material-sector interval of the corpus; on $\mathbb{B}_{\Psi}$ the same coordinates give $\langle\tilde{Q}\rangle_{\natural}=(x^0)^2-r^2=-q(x)$, so the source's $1+q$ and the denominator above are the same denominator.
Two chords, then, and three loci. The boundary is where the first vanishes, with apex at $\tilde{U}_{\mathrm{C}}=e_t$, the image of infinity, where both eigenvalues reach $+1$. The light cone of the origin is where the second vanishes, with apex at $\tilde{U}_{\mathrm{C}}=-e_t$, the image of the origin. The sphere is where both vanish.
The source's names for these sets run through one more translation, and the translation is worth stating once, because it is where a careless reading goes wrong. The chart fails on the chord locus $\mathcal{B}=\{U:\langle e_t-U\rangle_{\natural}=0\}$, which is the boundary. The source instead studies the translate $m(\mathcal{B})=\{-U:\langle e_t-U\rangle_{\natural}=0\}$, $m$ being multiplication by $-e_t$, and splits it into $\mathcal{B}_c=m(\mathcal{B})\setminus\mathcal{B}$ and $\mathcal{B}_s=m(\mathcal{B})\cap\mathcal{B}$. The first of these, which the source calls the cone at infinity, is $\tilde{Ξ}$ of the light cone through the origin by Eq. (7) of the source; it is an interior locus, not the boundary. The second is the sphere. Read in the chart's own coordinates, the three sets are the rows below.
| Locus | Condition on $\tilde{U}_{\mathrm{C}}$ | Character |
|---|---|---|
| The chart | $\langle e_t-\tilde{U}_{\mathrm{C}}\rangle_{\natural}\ne0$ | one point per element, no boundary inside |
| Light cone of the origin | $\langle e_t+\tilde{U}_{\mathrm{C}}\rangle_{\natural}=0$, first chord non-null | interior, dimension $3$, apex at the image of the origin |
| Boundary: the cone | first chord null, second non-null | boundary, dimension $3$, apex at the image of infinity |
| Boundary: the sphere | both chords null | a two-sphere, at the base of the cone, where the future and past cones at infinity meet |
This split is the point of the source: the boundary is not only a cone, and the treatments it corrects lost the sphere. The sphere is the interface at which the future and past parts of the infinity are identified, and it is the piece the algebra describes most directly.
The Sphere at Infinity Is the Set of Half-Turns
Let both chords vanish. Expanding them by the polar form, $$ \langle e_t\pm\tilde{U}_{\mathrm{C}}\rangle_{\natural}=1\pm2\langle e_t,\tilde{U}_{\mathrm{C}}\rangle_{\natural}+\langle\tilde{U}_{\mathrm{C}}\rangle_{\natural}, $$ so their sum gives $\langle\tilde{U}_{\mathrm{C}}\rangle_{\natural}=-1$ and their difference gives $\langle e_t,\tilde{U}_{\mathrm{C}}\rangle_{\natural}=0$. The trace of the matrix image is twice the scalar coefficient (Norm and Pairing Conventions), so a vanishing trace is a vanishing scalar part; with $\langle\tilde{U}_{\mathrm{C}}\rangle_{\natural}=-1$ the chart value is a scalar-free star-unitary element of norm $-1$. Its spectrum is $\{\lambda,-\lambda\}$ with $-\lambda^{2}=-1$, so $\lambda=\pm1$: it is a half-turn. The whole sphere is $$ \tilde{U}_{\mathrm{C}}=\pm i\vec n,\quad \vec n\in S^{2},\qquad \tilde{U}_{\mathrm{C}}^{2}=e_t,\qquad \langle\tilde{U}_{\mathrm{C}}\rangle_{\natural}=-1. $$ Verified: for fifty unit vectors both chords vanish, $\tilde{U}_{\mathrm{C}}^{2}=e_t$, the trace vanishes and the norm is $-1$, to $10^{-15}$; and every scalar-free star-unitary element satisfies $\langle\tilde{U}_{\mathrm{C}}\rangle_{\natural}=-\lambda^{2}$, so $\langle\tilde{U}_{\mathrm{C}}\rangle_{\natural}=-1$ forces $\lambda=\pm1$ and the half-turn, to $10^{-15}$. The sphere is not in the image of the chart at all, since both chords vanish on it; it is reached only in the limit. Its translate by the central $i$ is the set of unit real quaternions, $i\cdot i\vec n=-\vec n$, and that is where the unit spatial sphere $x^0=0$, $r=1$ lands: the value of the translate there is $-\vec n$, the unit real quaternion at that direction, checked on fifty directions to $10^{-15}$. This is the source's $\tilde{Ξ}^{-1}(j(\mathrm{I}_s))=\{x^0=0,\,r=1\}$ in the corpus's own terms.
The same two-sphere carries the pure states, read as generators rather than as states. The pure states are the complementary idempotents $\tilde\Pi_{1,2}(\vec n)=\tfrac12(e_t\pm i\vec n)$ of the quantum article, and the half-turn is their difference, $$ i\vec n=\tilde\Pi_1(\vec n)-\tilde\Pi_2(\vec n)=2\tilde\Pi_1(\vec n)-e_t, $$ checked to $10^{-16}$. So the sphere at infinity of a sector is the Bloch sphere of Quantum Physics in Biquaternionic Form, with the same $\vec n$ and a different object at each point: the corpus's idempotent $\tilde\Pi_1(\vec n)$, and here the half-turn built from both idempotents at that direction, $i\vec n=\tilde\Pi_1(\vec n)-\tilde\Pi_2(\vec n)$. The half-turn acts too, and on the material sector it is the $\pi$-rotation about $\vec n$: conjugation by $i\vec n$ coincides with rotor conjugation by the unit real quaternion $\vec n$, since $\vec n^{-1}=-\vec n$, and the two agree on all of $\mathbb{B}_{\Phi}$ to machine zero. This is what makes the construction relevant rather than merely available: the conformal boundary is a sphere the corpus already uses, and identifying it requires no new structure.
The Two Charts Are Reciprocal
Reading the same element through the two sectors, $\tilde{Q}\in\mathbb{B}_{\Psi}$ and $i\tilde{Q}\in\mathbb{B}_{\Phi}$, gives chart values that are mutual inverses. Writing $\tilde{U}_{\mathrm{inf},\mathrm{C}}$ for the chart with pole $ie_t$ and $\tilde{U}_{\mathrm{mat},\mathrm{C}}$ for the chart with pole $e_t$, $$ \tilde{U}_{\mathrm{mat},\mathrm{C}}(i\tilde{Q})=\bigl(\tilde{U}_{\mathrm{inf},\mathrm{C}}(\tilde{Q})\bigr)^{-1}=\bigl(\tilde{U}_{\mathrm{inf},\mathrm{C}}(\tilde{Q})\bigr)^{*}, $$ the second equality holding because a $*$-unitary element's inverse is its star. Verified on fifty elements to $5\times10^{-16}$. The chart transition is thus the star itself: the involution that defines the two sectors, as the fixed sets of ${}^{*}$ and ${}^{\flat}$, is the operation relating the charts of their compactifications. Multiplication by the central $i$ exchanges the sectors on the way in, and the star exchanges the charts on the way out. This is the statement of the section that bears on the theory rather than on the geometry alone.
What the Construction Does Not Give
The boundary of the section, in the same form as the boundaries above.
- One sector at a time. The two charts are two coordinate systems on one $\mathrm{U}(2)$. That is a statement about the compactified space, not that the two sectors are one sector, and it does not unify them. The physical question — which sector, if either, is the compactified one — is not answered here.
- Not the group inside the algebra. The global action is fractional on $\mathrm{U}(2)$, and the group remains $SU(2,2)$, outside $\mathbb{B}$, as the next section shows. What the compactified space adds is that the space is itself a group: left translation by an element of $\mathrm{U}(2)$ is a conformal map of it, a four-parameter subgroup of the fifteen-parameter group, and the source's appendix computes the four fields it generates, $Z_0=\tfrac12(K_0+P_0)$ and $Z_i=\tfrac12(K_i-P_i)+L_i$. Verified on the flows of the four one-parameter subgroups: the interval between two points is multiplied by a real factor that factorises as $\Omega(\tilde{Q}_1)\Omega(\tilde{Q}_2)$, so the cross-ratios of four intervals are preserved, to $10^{-11}$.
- Not a compact universe by necessity. Both sectors compactify and neither is privileged. The algebra is equally consistent with the non-compact reading; the compactification is available, not forced, and a compact spacetime is a choice the framework permits.
- A naming collision to avoid. The source's $M_+$ and $M_-$ denote two copies of Minkowski space, joined along their common boundary, in the double cover $\mathrm{U}(1)\times SU(2)$, not the corpus's $\mathbb{B}_{\Psi}$ and $\mathbb{B}_{\Phi}$; and its "light cone at infinity" is the second chord's locus, which is the translate $U\mapsto-U$ of the chart boundary and not the boundary itself.
- One inconsistency in the source, which does not propagate. The source's Eq. (2) states $\det\xi(x)=(x^0)^2-(x^1)^2+(x^2)^2+(x^3)^2=-q(x)$, whereas the determinant of the array in its Eq. (1) is $(x^0)^2-r^2=-q(x)$; the two spatial plus signs cannot both be right. Nothing here depends on the printed line: the invariant statement is that the source's determinant is the norm of the chord, and the source's explicit chart and its Eq. (6) were recomputed in the algebra from Eq. (1) and agree to $10^{-15}$.
Where the Algebra Strains
The obstructions that keep the conformal group out of the algebra can now be collected. None is a matter of insufficient cleverness in choosing a parametrization; each is structural.
Dimension. The biquaternion algebra has real dimension $8$, and its unit-norm subgroup $SL(2,\mathbb{C})$ — the group that acts as the Lorentz group — has real dimension $6$. The conformal group has dimension $15$. No faithful assignment of unit-norm biquaternions to conformal elements exists, for the same counting reason that blocks the translations in the Poincaré article. In Lie-algebra terms, $\mathfrak{so}(2,4)$ is fifteen-dimensional, and the commutator of $\mathbb{B}$ closes on a Lie algebra of real dimension six, which cannot contain it. No arrangement of the existing algebra closes the gap.
Linearity. The natural self-action of the algebra is the linear map $\tilde{Q}\mapsto\tilde{U}\tilde{Q}\tilde{U}^{*}$; it preserves $\langle\cdot\rangle_{\natural}$ pointwise, hence produces isometries only, never dilations, and its group is six-dimensional. For a rotation this map is conjugation by a unit real quaternion and is an inner automorphism; for a boost it is the twisted map $\tilde U_{\Psi,\mathrm b}\tilde{Q}\tilde U_{\Psi,\mathrm b}$ (since $\tilde U_{\Psi,\mathrm b}^{*}=\tilde U_{\Psi,\mathrm b}$), which is linear but not multiplicative and so is not an algebra automorphism at all. Either way it is linear in $\tilde{Q}$, and the special conformal transformations are not. Classifying the fifteen generators by how they can be realized:
- the six Lorentz generators are elements of $\mathbb{B}$;
- the dilation is realized by a non-invertible element, and its generator is a non-inner, nonzero-trace linear map;
- the four translations act affinely, hence not linearly on $\mathbb{B}_{\Phi}$;
- the four special conformal transformations are fractional, hence not linear at all.
The size of the linearly realized part. The conformal transformations that act linearly or affinely on $\mathbb{B}_{\Phi}$ are exactly the similarities $x\mapsto\lambda\Lambda x+a$, whose group — Lorentz (6) plus dilation (1) plus translations (4) — has dimension $11$. The remaining four generators, the special conformal ones, are precisely the nonlinear part. The algebra carries all eleven: the Lorentz factor by multiplication of rotors, the dilation by a non-invertible scaling, the translations by addition on the module. The last four it reaches only through the nonlinear inversion. That is the sharp form of the boundary.
The second sector. The Lorentz action extends to the whole algebra: $\tilde{Q}\mapsto\tilde{U}\tilde{Q}\tilde{U}^{*}$ is a linear map of $\mathbb{B}$ that restricts to rotor conjugation on $\mathbb{B}_{\Phi}$ and preserves $\mathbb{B}_{\Psi}$ as well (the conjugate of a Hermitian element is Hermitian). Conformal operations behave less uniformly. The inversion does preserve each sector separately: $\langle\cdot\rangle_{\natural}$ is real on $\mathbb{B}_{\Psi}$ and on $\mathbb{B}_{\Phi}$, so the scalar factor $\tilde{Q}/\langle\tilde{Q}\rangle_{\natural}$ keeps the Hermitian or anti-Hermitian character. But the translations are $\mathbb{B}_{\Phi}$-valued and the special conformal transformations are anchored there, so the full conformal group has no canonical action on the informational sector $\mathbb{B}_{\Psi}$. Whether $\mathbb{B}_{\Psi}$ is meant to be conformally invariant, and with what weight, is not determined by the algebra. This is a gap, not a result, and it is left visible. The compactification of a sector is poled in the other sector (The Compactification from the Star), which ties the two charts together without giving the conformal group an action on either sector; the gap survives that tie.
Larger Homes in the Literature
The conformal group is a standard object, and its standard algebraic homes are larger than $\mathbb{B}$. Three are worth naming, because they locate exactly what the biquaternion algebra would have to acquire.
Vahlen matrices. The Ahlfors–Vahlen construction realizes the conformal group of $\mathbb{R}^{n,m}$ by $2\times2$ matrices $$ \begin{pmatrix} a & b\ c & d\end{pmatrix} $$ whose entries lie in the Clifford algebra $\mathrm{Cl}_{n,m}$ and are products of vectors, and which satisfy the conformality conditions that make the fractional action $$ \tilde{Q}\;\longmapsto\;(aX+b)(cX+d)^{-1} $$ reduce to a conformal map of vectors. For Minkowski space the entries live in $\mathrm{Cl}_{1,3}$, the entry algebra of real dimension $16$ — twice the eight of $\mathbb{B}$ — and the ambient matrix algebra $M_2(\mathrm{Cl}_{1,3})$ has real dimension $64$. The biquaternion algebra is the even subalgebra, $\mathbb{B}=\mathrm{Cl}^+_{1,3}$, and the natural question — whether the Vahlen conditions can be imposed with entries in the even subalgebra alone, giving a reduced Vahlen group — is a technical point that this article does not settle. It is stated here as the precise form of "the algebra is too small."
The six-dimensional embedding. Alternatively, the conformal group is a rotation group of a six-dimensional space. The conformal compactification of $\mathbb{R}^{1,3}$ is the projectivized null cone of $\mathbb{R}^{2,4}$, exactly as Minkowski space is a real slice of the complexified rotation picture of the Lorentz-transformation companion; the conformal group is $O(2,4)$, realized by rotations of that six-dimensional space. In Clifford terms this is $\mathrm{Cl}_{2,4}$: its bivectors span the fifteen-dimensional conformal algebra, and its even subalgebra, the home of the conformal rotors, has real dimension $32$. The biquaternion algebra, as the even subalgebra of the four-dimensional Clifford algebra $\mathrm{Cl}_{1,3}$, is eight-dimensional. The conformal rotor algebra is four times larger. This is the quantitative content of the Poincaré article's remark that any enlargement forced by the translations must be strictly larger than $\mathbb{B}$.
The Clifford-group interpretation, and dilations without a Weyl field. The six-dimensional embedding is sometimes stated as its own result: the conformal group of four-dimensional spacetime emerges as a subgroup of the Clifford group [Castro and Pavšič, "Clifford Algebra of Spacetime and the Conformal Group," Int. J. Theor. Phys. 42 (2003) 1693, arXiv:hep-th/0203194]. The standard part is the bivector count of a six-generator set — the four vector generators $v_\mu$ of $\mathrm{Cl}_{1,3}$, the pseudoscalar and the scalar unit, of inertia $(-1,+1)$ in the last two slots. The bivectors of two vector slots give the six Lorentz generators; those mixing a vector slot with the two extra ones give the translations $\tfrac12v_\mu(e_t-\omega)$ and the special conformal generators $\tfrac12v_\mu(e_t+\omega)$, where $\omega$ is the corpus's pseudoscalar, $\omega=-\vec e_z$ with $\omega^2=-e_t$; and the bivector of the two extra slots gives the dilation, $-\tfrac{i}{2}\omega$ — fifteen in all, the same $\mathrm{Cl}_{2,4}$ statement as above. The pseudoscalar is the source's $i$: the source's own pseudoscalar, of square $e_t$, is $i\omega$, while the central $i$ stays free. One caution belongs to the dilation: the unit is central, so its commutator with the pseudoscalar vanishes and the standard bivector formula, which is a commutator, returns nothing for that pair; the dilation is read from the product $e_t\omega$ itself, and it is not the zero that the literal bracket would give. The authors' own comment on the construction is that it "might look mathematically trivial". What the reference adds is where that six-dimensional space sits: it is the subspace of the Clifford space spanned by those six generators, its coordinates $(\eta_\mu,\eta^5,\eta^6)$ — the source's $(x_\mu,\check s,s)$, with $\check s$ the pseudoscalar and $s$ the scalar coordinate — renamed $(\zeta_\mu,\zeta_5,\zeta_6)$ here, so that the conformal group acts linearly on it and, on the cone $\sum_{a,b}g_{ab}\zeta_a\zeta_b=0$ with $g=\mathrm{diag}(+1,-1,-1,-1,-1,+1)$, reproduces the conformal transformations of Minkowski space. The remaining polyvector coordinates — the bivectors $x_{\mu\nu}$ and the pseudovectors $\check x_\mu$ — lie outside that subspace and take no part.
The same passage adds, and this is speculative, a physical interpretation of the dilations: the authors argue that an extended object in this arena changes size "as a result of free motion, without presence of any forces or such fields as assumed in Weyl theory", moving dilationally because of inertia rather than because a compensating gauge field has been introduced, and they cite their own earlier work for the proposal. That is a proposal about why scale changes occur, not a derivation, and it is recorded here as the programme's claim, not as a result of the algebra.
Twistors. A third realization uses the same vector space as the algebra but a different structure. The double cover $SU(2,2)$ of the conformal group acts linearly on twistor space $T=S\oplus\bar{S}\cong\mathbb{C}^4$. As complex vector spaces, $T$ and $\mathbb{B}$ are both $\mathbb{C}^4$; but the form that $SU(2,2)$ preserves is a Hermitian form of inertia $(2,2)$, whereas the biquaternion norm $\langle\cdot\rangle_{\natural}$ is a complex symmetric bilinear form. The spaces agree; the forms do not. This is the precise sense in which the conformal group, when it acts on the same four-dimensional vector space, enters as the $*$-unitary group of a form and not as the group of norm-one elements — a different kind of group from the one the framework uses. The companion article on twistors develops this in full, with the incidence relation and the Klein correspondence; the point here is only that the twistor realization does not place the conformal group inside $\mathbb{B}$, it places it on $\mathbb{B}$ read as a new module with a new form.
The Field Equation and the Mass
The group is the symmetry of an equation, and the equation has not yet been named. It is the gradient equation of the series (Operator and Calculus Conventions, §The Gradient). At $m=0$ it is first order in one field, $$ \tilde\nabla_\Phi\tilde{Ξ}=0, $$ and every derivative in it is the gradient defined in the introduction. This is the equation the fifteen parameters leave invariant: a conformal map carries a solution to a solution times a power of the conformal factor, and the mass term is the only obstruction. It is also the reason a conformal group enters the physics at all: Maxwell's equations are the massless case.
The mass cannot be a term of the same kind. The square of the gradient is not scalar: the unit $e_t$ is the identity of the algebra, so it commutes with the three spatial units, and the two orders of a mixed derivative add, not cancel, $$ \tilde\nabla_\Phi^2=\frac{\partial^2}{\partial R_t^2}-\Delta_\Phi+2\sum_\eta\vec e_\eta\frac{\partial^2}{\partial R_t\,\partial R_\eta}, $$ while the product with the conjugate is scalar, $$ (\tilde\nabla_\Phi)^{\natural}\tilde\nabla_\Phi=\tilde\nabla_\Phi(\tilde\nabla_\Phi)^{\natural}=\Box_\Phi . $$ The square of the gradient is not the d'Alembertian, and the claim that it is must not be made.
The massive equation is the chiral pair, one equation per chirality, $$ \tilde\nabla_\Phi\tilde{Ξ}_{R}=k_m\tilde{Ξ}_{L},\qquad (\tilde\nabla_\Phi)^{\natural}\tilde{Ξ}_{L}=k_m\tilde{Ξ}_{R}, $$ both members acting from the left on their own field, so no glyph names a side. Applying $(\tilde\nabla_\Phi)^{\natural}$ to the first member and substituting the second returns $\Box_\Phi\tilde{Ξ}_{R}=k_m^{2}\tilde{Ξ}_{R}$, so the pair is one dynamics on two fields: the mass couples the halves, and each half obeys the d'Alembertian at mass $m$. Here $k_m=mc/\hbar$ is the mass wave number, the inverse of the mass length.
This is the operator form of the boundary drawn above. The massless law is first order in one field and the massive law is first order in two; no rewriting puts the mass into a first-order equation for one field, and the mass is exactly what breaks the conformal invariance. The algebra is large enough for both, and the passage from one to the other is the price the boundary charges.
Summary
The conformal group of Minkowski space is $O(2,4)$; its connected component is the fifteen-parameter group $SO^+(2,4)$, double-covered by $SU(2,2)$, generated by the six Lorentz generators $M_{\mu\nu}$, the four translations $P_\mu$, the single dilation $D$, and the four special conformal generators $\mathcal{K}_\mu$. Its brackets were recomputed here on explicit test functions: $[D,P_\mu]=-P_\mu$, $[D,\mathcal{K}_\mu]=\mathcal{K}_\mu$, $[P_\mu,\mathcal{K}_\nu]=2(\delta_{\mu\nu}D-M_{\mu\nu})$, with the translations and special conformal generators abelian and the Lorentz part the companion article's.
The biquaternion algebra carries the conformal structure but not the conformal group. What it carries: the null cone — the material sector's part of the zero-divisor cone $\{\tilde{Q}:\langle\tilde{Q}\rangle_{\natural}=0\}$ — is the invariant object of the conformal group, and every generator preserves it; the inversion is algebraic, $I(\tilde{Q})=(\tilde{Q}^{\natural})^{-1}=\tilde{Q}/\langle\tilde{Q}\rangle_{\natural}$, with $I^2=\mathrm{id}$ and $\langle I(\tilde{Q})\rangle_{\natural}=1/\langle\tilde{Q}\rangle_{\natural}$; the dilation is multiplication on both sides by the non-invertible central element $\sqrt{\lambda}e_t$; and the special conformal transformation is the composition $I\circ T_{\tilde Q_3}\circ I$, with the explicit form $$ \boxed{\;\tilde{Q}_2=\frac{\tilde{Q}+\tilde Q_3\,\langle\tilde{Q}\rangle_{\natural}}{1+2\langle\tilde{Q},\tilde Q_3\rangle_{\natural}+\langle\tilde Q_3\rangle_{\natural}\langle\tilde{Q}\rangle_{\natural}}\;}, $$ which reproduces $Q_{2\mu}=(Q_\mu+a_\mu x^2)/(1+2a\cdot x+a^2x^2)$.
What it does not carry: the group. The algebra is eight real dimensions and its unit-norm subgroup six, against the conformal group's fifteen; the six Lorentz generators are elements of $\mathbb{B}$, the dilation is realized by a non-invertible element, and its generator is a non-inner, nonzero-trace linear map; the four translations are affine and the four special conformal transformations are fractional. The transformations realized linearly or affinely on $\mathbb{B}_{\Phi}$ form the eleven-dimensional similitude group, and the four remaining generators are exactly the nonlinear part. The standard homes of the full group are larger — an entry algebra of twice the real dimension of $\mathbb{B}$, a conformal rotor algebra four times it, or $SU(2,2)$ acting on a $\mathbb{C}^4$ equipped with a $(2,2)$ Hermitian form that the biquaternion norm is not — and §Larger Homes in the Literature names each. The conformal group is expressed by the algebra — by its biquaternion norm's zero set, its inverse, and its four-vector module — and is not contained in it. The operator content is the gradient: the massless law is $\tilde\nabla_\Phi\tilde{Ξ}=0$, and the mass couples the two chirality halves, $\tilde\nabla_\Phi\tilde{Ξ}_{R}=k_m\tilde{Ξ}_{L}$ and $(\tilde\nabla_\Phi)^{\natural}\tilde{Ξ}_{L}=k_m\tilde{Ξ}_{R}$. The square of the gradient is not scalar, so no first-order equation for one field carries the mass.
What the algebra does carry, by contrast, is the compactification, and it carries it on a different form. The space on which the conformal group acts globally is the $*$-unitary group $\mathrm{U}(2)$ of Norm and Pairing Conventions — and not the unit-norm group $SL(2,\mathbb{C})$. Each sector charts it by a Cayley transform poled in the other sector, $\tilde{U}_{\mathrm{C}}=(\tilde{Q}-p)(\tilde{Q}+p)^{-1}$ with $p=ie_t$ for $\mathbb{B}_{\Psi}$ and $p=e_t$ for $\mathbb{B}_{\Phi}$, and the chart transition between the two is the star. The boundary is the chord $e_t-\tilde{U}_{\mathrm{C}}$ becoming a zero divisor; the light cone of the origin is the corresponding locus for $e_t+\tilde{U}_{\mathrm{C}}$; and where both vanish the boundary is the sphere $i\vec n$, $\vec n\in S^{2}$ — the half-turns, which are the differences $\tilde\Pi_1(\vec n)-\tilde\Pi_2(\vec n)$ of the two complementary pure states. The sphere at infinity of a sector is the Bloch sphere of that sector read as generators. What the construction does not give is a reason to prefer the compactified space, or an action of the conformal group inside $\mathbb{B}$: the compactification is available to the algebra, not forced by it.
Summary of Notation
The algebra, its units, the scalar imaginary, the sectors, the norm and its polar form, the rotor and its action, the four forms, the gradient, its conjugate, the d'Alembertian and the trace are those of the read-list companions, and their definitions are not repeated here. The table lists the symbols the article introduces or uses in its own right.
| Symbol | Meaning |
|---|---|
| $\mathbb{B}_{\Phi}$, $\mathbb{B}_{\Psi}$ | The material and the informational sector; the conditions on the coefficients are in Algebra and Element Conventions |
| $\tilde{U}$ | The unit-norm element, the rotor |
| $SL(2,\mathbb{C})$ | The unit-norm group, the Lorentz double cover |
| $\mathrm{U}(2)$ | The star-unitary group; the compactified space on which the conformal group acts globally |
| $\tilde{U}_{\mathrm{C}}=(\tilde{Q}-p)(\tilde{Q}+p)^{-1}$, $p=ie_t$ or $p=e_t$ | Cayley chart of a sector, poled in the other sector at $-p$; inverse $\tilde{Q}=p(e_t+\tilde{U}_{\mathrm{C}})(e_t-\tilde{U}_{\mathrm{C}})^{-1}$ |
| $\langle e_t-\tilde{U}_{\mathrm{C}}\rangle_{\natural}$, $\langle e_t+\tilde{U}_{\mathrm{C}}\rangle_{\natural}$ | The two chords: the first vanishes on the boundary of the chart, the second on the image of the light cone of the origin |
| $\pm i\vec n=\tilde\Pi_1(\vec n)-\tilde\Pi_2(\vec n)$, $\vec n\in S^{2}$ | The sphere at infinity: the half-turns, which are the Bloch sphere of the informational sector read as generators |
| $M_{\mu\nu}$, $P_\mu$, $D$, $\mathcal{K}_\mu$ | Conformal generators: Lorentz (6), translations (4), dilation (1), special conformal (4) |
| $I(\tilde{Q})=(\tilde{Q}^{\natural})^{-1}=\tilde{Q}/\langle\tilde{Q}\rangle_{\natural}$ | Inversion; $I^2=\mathrm{id}$, $\langle I(\tilde{Q})\rangle_{\natural}=1/\langle\tilde{Q}\rangle_{\natural}$ |
| $\tilde{Q}^{-1}=\tilde{Q}^{\natural}/\langle\tilde{Q}\rangle_{\natural}=P(I(\tilde{Q}))$ | Algebra inverse, which is inversion followed by parity |
| $T_{\tilde Q_3}:\tilde Q\mapsto\tilde Q+\tilde Q_3$ | Translation by $\tilde Q_3\in\mathbb{B}_{\Phi}$ |
| $I\circ T_{\tilde Q_3}\circ I$ | Special conformal transformation |
| $SO^+(2,4)$, $SU(2,2)$ | Connected conformal group, of dimension $15$, and its double cover |
| $c$ | Speed of light in the medium, entering the corpus's $Q_t=i q'_t$ |
| $\tilde{Ξ}_{R}$, $\tilde{Ξ}_{L}$ | Chirality halves of the gradient pair, $\tilde\nabla_\Phi\tilde{Ξ}_{R}=k_m\tilde{Ξ}_{L}$, $(\tilde\nabla_\Phi)^{\natural}\tilde{Ξ}_{L}=k_m\tilde{Ξ}_{R}$ |
Further Reading
- The First-Order Root of the Biquaternion Klein–Gordon Operator, for the massless gradient equation, for the square of the gradient that is not the d'Alembertian, and for the chiral pair that carries the mass.
- Philippe Di Francesco, Pierre Mathieu, and David Sénéchal, Conformal Field Theory (Springer, 1997), for the conformal algebra, its generators, and the $SO(d,2)$ form of the conformal group.
- Steven Weinberg, The Quantum Theory of Fields, Vol. 1 (Cambridge, 1995), for the conformal group and the generators of dilations and special conformal transformations in the standard physics convention.
- Roger Penrose and Wolfgang Rindler, Spinors and Space-Time, Vol. 2 (Cambridge, 1986), for the conformal compactification of Minkowski space and its relation to null geometry and twistors.
- Arkadiusz Jadczyk, "On Conformal Infinity and Compactifications of the Minkowski Space," Advances in Applied Clifford Algebras 21 (2011) 721–756, arXiv:1008.4703, for the compactification by $2\times2$ matrices and the Cayley transform, the identification of the boundary as a cone with a two-sphere at its base, and the four conformal Killing fields of the compactified group acting on itself — the construction of The Compactification from the Star.
- Lars V. Ahlfors, Möbius Transformations in Several Dimensions (University of Minnesota, 1981), for Vahlen matrices and the Clifford-algebraic realization of the conformal group.
- Harry Bateman, "The Transformation of the Electrodynamical Equations," Proceedings of the London Mathematical Society s2-8 (1910) 223–264, and Ebenezer Cunningham, "The Principle of Relativity in Electrodynamics and an Extension Thereof," Proceedings of the London Mathematical Society s2-8 (1910) 77–98, for the spherical wave transformations and the conformal covariance of Maxwell's equations.
- Chris Doran and Anthony Lasenby, Geometric Algebra for Physicists (Cambridge, 2003), for conformal geometric algebra, the embedding in $\mathrm{Cl}_{2,4}$ (written $\mathrm{Cl}_{4,2}$ in the authors' inertia convention), and conformal transformations as rotors.
- Pertti Lounesto, Clifford Algebras and Spinors (Cambridge, 2001), for the Clifford algebras $\mathrm{Cl}_{1,3}$ and $\mathrm{Cl}_{2,4}$ and the even subalgebra identification $\mathbb{B}=\mathrm{Cl}^+_{1,3}$.
- Brian C. Hall, Lie Groups, Lie Algebras, and Representations (Springer, 2015), for $\mathrm{SO}(2,4)$, its real forms, and the double cover $SU(2,2)\to SO^+(2,4)$.
- Wu-Ki Tung, Group Theory in Physics (World Scientific, 1985), for the conformal group in the context of the Poincaré and Lorentz groups treated in the companion articles.
- C. Castro and M. Pavšič, "Clifford Algebra of Spacetime and the Conformal Group," International Journal of Theoretical Physics 42 (2003) 1693, arXiv:hep-th/0203194, for the Clifford-group statement of the six-dimensional embedding (the group acting on the null cone $\sum_{a,b}g_{ab}\zeta_a\zeta_b=0$, $\zeta_a=(\zeta_\mu,\zeta_5,\zeta_6)$, $g_{ab}=\mathrm{diag}(1,-1,-1,-1,-1,1)$; the reference writes $\eta_a$), and for the speculative interpretation of the dilations as inertial motion rather than the effect of a Weyl gauge field, both recorded under Larger Homes in the Literature.