Bhabha Multi-Mass Wave Equations and the Bivector Space in Biquaternionic Form
Introduction
The mass formula of the companion The Spin–Mass Formula and the Spectrum of Elementary Particles in Biquaternionic Form comes from a family of wave equations, not from a single one. These are the relativistic wave equations of Bhabha, Gel'fand and Yaglom: finite-dimensional equations for a field in an arbitrary Lorentz representation $\tau^{l\bar l}$, whose mass operator is a matrix and whose determinant is a polynomial in $N$ with several roots — one field, several masses. When the representation is an interlocking chain $(l,\bar l)\oplus(\bar l,l)$, the equation describes a particle of generalised spin $s=|l-\bar l|$ and mass $m$. The chain that carries the proton is a case in point: the proton is assigned to the member $\tau^{59/2,29}$, of degree $3540$, taken together with its conjugate.
This article records the equations, the spin chains they live on, and the third ingredient the programme attaches to them: the reduction of the equations from Minkowski space to the bivector space $\mathbb{R}^6$. The reduction is not decorative. The Lorentz group is six-dimensional and so is the space of antisymmetric two-tensors — the bivectors, the field strengths. The programme maps the wave equations into that space and writes them there. The corpus knows the space under another name: the field-strength biquaternion $\vec E+i\vec B$ of The Field-Strength Biquaternion and Its Invariants has six real components and is exactly a point of the bivector space. The article notes the identification and states where it stops.
It is worth separating this article from the companion that shares the name Bhabha. Bhabha Scattering in Biquaternionic Form is about the elastic scattering of an electron and a positron; the present article is about Bhabha's wave equations, a different piece of work by the same author. The two are cross-referenced and not merged.
The algebra and the calculus, by citation. The algebra is the biquaternion algebra of Algebra and Element Conventions, with its four units, its coefficient basis and its conjugations; the gradient, its quaternion conjugate and the d'Alembertian are those of Operator and Calculus Conventions, §The Gradient; the two sides of an action are those of that article, §The Two Sides and the Renaming Theorem. No symbol of those files is defined here, and two of their statements are used and cited rather than restated: the product of the gradient with its quaternion conjugate is the scalar $\Box_\Phi$, and the exchange by which a right action on an element is a left action on its conjugate.
Displayed formulae are written with their imaginary units reduced whenever the algebra allows it; where the source's own conventions leave a reduction undecided, the display is transcribed as printed and the undecided step is flagged rather than silently repaired.
Which fields the gradient acts on. The gradient is the derivative of every field valued in $\mathbb{B}$, and the Dirac field of the fourth section is one: it is an element of the algebra, so left multiplication by the gradient acts on it. The source's fields at a general $\tau^{l\bar l}$ are not elements of $\mathbb{B}$ — they are the carriers of the representation — so their derivative, $\partial/\partial x_\mu$ in Minkowski space and $\partial/\partial a_j$ in $\mathbb{R}^6$, is quoted as printed rather than rewritten as a gradient. Every equation of this article is written from the left; a right action on an element is the same equation on its conjugate, so nothing is lost.
The order of the article. A first section fixes the spintensors and the spin chains. A second states the wave equations and the determinant argument. A third gives the bivector-space reduction and settles its metric and its index order. A fourth works the proton chain and writes the Dirac limit as two equations on one element, and states why one equation will not do. A fifth states the biquaternion interpretation and separates what is established from what is transcribed.
The articles it leans on.
- Companion article The Spin–Mass Formula and the Spectrum of Elementary Particles in Biquaternionic Form, for the mass spectrum these equations produce.
- Companion article The Spin–Charge Hilbert Space and the SU(3) Supermultiplets in Biquaternionic Form, for the space in which the multi-mass fields are placed.
- Corpus article The Field-Strength Biquaternion and Its Invariants, for the biquaternion form of the bivector space.
- Corpus article The Dirac–Hestenes Equation and Spacetime Algebra in Biquaternionic Form, for that article's form of the Dirac and Maxwell equations used in the Dirac limit.
- Corpus article Bispinor Fields and the Fundamental Solution of the Generalized Maxwell–Dirac Equation, for the Maxwell–Dirac package in the corpus's own form.
- Corpus article Bhabha Scattering in Biquaternionic Form, for the other work bearing Bhabha's name.
Spintensors and the Spin Chains
The Spintensor
A representation of $\mathrm{SL}(2,\mathbb{C})$ is carried by a spintensor with $k$ unbarred and $r$ barred indices,
$$ S = s_{\alpha_1\alpha_2\dots\alpha_k\bar\alpha_1\bar\alpha_2\dots\bar\alpha_r}, $$
a polynomial in the products of the unbarred and barred letters $s_\alpha$ and $s_{\bar\alpha}$, with $\alpha,\bar\alpha=1,2$. The representation is $\tau^{l\bar l}$ with
$$ l = \tfrac{k}{2}, \qquad \bar l = \tfrac{r}{2}. $$
The spintensor is the image of the tensor product of $k$ copies of the fundamental representation and $r$ copies of its conjugate, and its carrier has $2^{k+r}$ components. The indices within each group can be permuted, so a spintensor may be taken symmetric; the space $\mathrm{sym}^{(k,r)}$ of symmetric spintensors has dimension
$$ \dim\mathrm{sym}^{(k,r)} = (k+1)(r+1), $$
the degree of the representation $\tau^{l\bar l}$. This is the number that the mass formula uses: because $l=k/2$ and $\bar l=r/2$, the degree is $(2l+1)(2\bar l+1)=4\bigl(l+\tfrac12\bigr)\bigl(\bar l+\tfrac12\bigr)$.
The degree does not name the representation. Degree $4$ is carried by $\tau^{\frac12,\frac12}$, of generalised spin $0$, and by $\tau^{\frac32,0}$, of generalised spin $\frac32$; degree $15$ by $\tau^{2,1}$, of spin $1$, and by $\tau^{7,0}$, of spin $7$. Inside one spin line the degree strictly increases with the chain index, so there the index is recoverable from the degree. Across lines it is not: a mass formula that reads $m=\frac{\mu_0}{4}\deg\tau$ assigns one mass to particles of different spin, because the mass is a function of the degree and the degree does not fix the particle.
The Two Schemes
The representations do not stand alone. $\tau^{l\bar l}$ and $\tau^{l_2\bar l_2}$ are interlocking when $l_2=l\pm\frac12$ and $\bar l_2=\bar l\pm\frac12$, the two signs being independent. The interlocking relations group the spin lines into two schemes. The Bose scheme carries the integer-spin lines, the spin-$0$, spin-$1$, $\dots$ lines; the Fermi scheme carries the half-integer-spin lines, the spin-$\frac12$, spin-$\frac32$, $\dots$ lines. Every spin value $s$ has its own infinite line inside its scheme, and the line is the chain of representations reached from $s$ by single interlocking steps. A field of type $(l,\bar l)\oplus(\bar l,l)$ is a field transforming in a representation and its conjugate; the wave equations of the next section are written for such fields and describe a particle of generalised spin
$$ s = |l-\bar l| . $$
The usual spin is recovered only on the restriction $\tau^{l\bar l}\to\tau^{l,0}$ (or $\to\tau^{0,\bar l}$), that is, at $\mathrm{SU}(2)\subset\mathrm{SL}(2,\mathbb{C})$, where the spin values run over $-l,-l+1,\dots,l$.
The Wave Equations and the Determinant Argument
The Equation and Its Determinant
The equations are written with the generator matrices $M^{l\bar l}_\mu$ of the representation $\tau^{l\bar l}$, one for each of the four coordinates, and read, for the field $\tilde{Ξ}$ of the representation,
$$ iM^{l\bar l}_\mu\frac{\partial\tilde{Ξ}}{\partial x_\mu}+k_m\tilde{Ξ}=0, \qquad\text{that is}\qquad M(p)\tilde{Ξ}=-k_m\tilde{Ξ}, $$
with $M(p)=M_0p_0-M_1p_1-M_2p_2-M_3p_3$ for a plane wave of momentum $p$. A nonzero solution requires
$$ D(p) = \det\bigl(M(p)+mE\bigr) = 0 . $$
The determinant is a polynomial in $p_0,p_1,p_2,p_3$ and is constant on the transitivity surfaces of the Lorentz group, the hyperboloids $s^2(p)=p_0^2-p_1^2-p_2^2-p_3^2=\text{const}$, so it depends on $s^2$ alone and factors as
$$ \boxed{\;D\bigl(s^2\bigr) = \kappa\bigl(s^2-m_1^2\bigr)\bigl(s^2-m_2^2\bigr)\cdots\bigl(s^2-m_k^2\bigr)\;}. $$
The equation has a solution exactly when $s^2=m_i^2$, so the roots $m_i$ are the masses the field can carry: a multi-mass field. Their number is bounded by the dimension. $D$ has degree at most $N$ in the momentum and depends on $s^2$ alone, so its degree in $s^2$ is at most $N/2$: a representation of dimension $N$ carries at most $N/2$ masses, which is the rest-frame $\pm$ pairing of the $N$ roots read the other way round.
The Eigenvalues at Rest
Putting $\vec p=0$ makes $M(p)=p_0M_0$ and
$$ \det\bigl(p_0M_0+mE\bigr) = \prod_{i=1}^{N}\bigl(\lambda_ip_0+m\bigr), $$
with $\lambda_i$ the eigenvalues of $M_0$ and $N$ the dimension of the representation. The source compares this with the factorisation
$$ \det\bigl(p_0M_0+mE\bigr) = \kappa\,\bigl(p_0-\mu_0\lambda_1\bigr)\bigl(p_0-\mu_0\lambda_2\bigr)\cdots\bigl(p_0-\mu_0\lambda_N\bigr), $$
and reads off the relation
$$ m_1 = \mu_0\lambda_1,\quad -m_1=\mu_0\lambda_2,\quad m_2=\mu_0\lambda_3=-\mu_0\lambda_4,\ \dots $$
which it writes first in this linear form for the general $\tau^{l\bar l}$; for the representations of type $\tau^{0,\bar l}$, where only the barred label runs, the same display holds with the barred eigenvalues alone, and in the general case the eigenvalues enter as products,
$$ m_1 = \mu_0\lambda_1\bar\lambda_1,\qquad -m_1=\mu_0\lambda_2\bar\lambda_2,\qquad m_2=\mu_0\lambda_3\bar\lambda_3=-\mu_0\lambda_4\bar\lambda_4,\ \dots $$
Why the spectrum is symmetric. The source records the $\pm$ pairing as an observation: each non-null eigenvalue $\lambda$ of $M_0$ occurs with $-\lambda$ of the same multiplicity. It is not a separate assumption, and the reason is already in hand. At $\vec p=0$ the determinant is $\prod_i(\lambda_ip_0+m)$, and it is also $D(p_0^2)$, an even function of $p_0$, because $D$ depends on the momentum only through $s^2$. Expanding the product in $p_0$, the coefficient of $p_0^{2j+1}$ is $m^{N-2j-1}e_{2j+1}$, with $e_\eta$ the $k$-th elementary symmetric function of the $\lambda_i$; evenness therefore says $e_1=e_3=\dots=0$, and that holds exactly when the multiset $\{\lambda_i\}$ is closed under $\lambda\mapsto-\lambda$ with multiplicities preserved. The roots of $D$ are thus $\pm$ pairs — which is why the source's list contains $m_1$ and $-m_1$ — and a positive mass can be read from each pair only once the normalisation is fixed.
A check of the comparison. Taken literally, the two displays are not the same polynomial. The roots of $\det(p_0M_0+mE)$ are $p_0=-m/\lambda_i$, while the second display has roots $p_0=\mu_0\lambda_i$; the two agree only if $\mu_0\lambda_i=-m/\lambda_i$ for every eigenvalue, that is if all $|\lambda_i|$ are equal, which a general representation of the Lorentz group does not satisfy. A constant can match the two lists at one eigenvalue and at no more, so the step is not a rescaling with a scale absorbed; in terms of mass the mismatch is an inversion. The determinant gives $m_i=m/|\lambda_i|$, which falls as the eigenvalue grows, while the source's product form rises with it. The invariant content is the $\pm$ pairing; the functional form of the mass in the eigenvalues is the point the source leaves unexamined, and it is recorded below as an open question rather than silently repaired.
In the infinite-dimensional limit $l,\bar l\to\infty$ the formula becomes
$$ m^{(s)} = \mu_0\bigl(l+\tfrac12\bigr)\bigl(\bar l+\tfrac12\bigr), \qquad s=|l-\bar l| , $$
which is the companion mass formula; in degree form it is $m^{(s)}=\frac{\mu_0}{4}\deg\tau^{l\bar l}$, since $(2l+1)(2\bar l+1)=4(l+\frac12)(\bar l+\frac12)$. The present article's contribution is the equations and the chains; the companion's is the spectrum and the assignments, and the two are kept separate.
The Bivector Space $\mathbb{R}^6$
The Metric Mapping
Minkowski space and the space of bivectors are related by the algebraic mapping
$$ g_{ab} \longrightarrow g_{\alpha\beta\gamma\delta} \equiv g_{\alpha\gamma}g_{\beta\delta}-g_{\alpha\delta}g_{\beta\gamma}, $$
which sends the metric of $\mathbb{R}^{1,3}$ to a metric on the six-dimensional space of antisymmetric two-tensors. The source's Minkowski metric is $\mathrm{diag}(-1,-1,-1,+1)$ in the coordinate order $(x_1,x_2,x_3,x_0)$, with the timelike coordinate $x_0$ listed last, and the printed bivector metric is
$$ g_{ab} = \mathrm{diag}(-1,-1,-1,+1,+1,+1). $$
The index order, corrected. The source prints the collective order
$$ 23\to0,\qquad 10\to1,\qquad 20\to2,\qquad 30\to3,\qquad 31\to4,\qquad 12\to5, $$
but under that order the mapping does not reproduce the printed diagonal: componentwise it gives
$$ 23\mapsto+1,\quad 10\mapsto-1,\quad 20\mapsto-1,\quad 30\mapsto-1,\quad 31\mapsto+1,\quad 12\mapsto+1, $$
that is $(+1,-1,-1,-1,+1,+1)$. The printed diagonal is reproduced exactly by the order
$$ 10\to0,\qquad 20\to1,\qquad 30\to2,\qquad 23\to3,\qquad 31\to4,\qquad 12\to5, $$
which lists the three boosts first and the three rotations second: the pairs $10,20,30$ contain the timelike index $0$ and give $-1$, while the pairs $23,31,12$ are purely spatial and give $+1$. Two repairs would reconcile the printed table with the mapping: change the order, or change three signs of the metric. The order is the likelier slip, because the printed diagonal is the one the mapping must give — three negative boost directions and three positive rotation directions — and altering it would leave the bivector metric unrelated to the Minkowski metric it is defined from. The equations below are written in the $a_j$ basis and do not use the collective order at all, so the slip is confined to the display.
The inertia $(3,3)$ is the inertia of the Lorentz Lie algebra's Killing form, in the same order: three negative directions for the non-compact boosts and three positive for the compact rotations. The bivector space is $\mathbb{R}^6$ as a real vector space; the identification with $\mathbb{C}^3$ needs a complex structure, and the metric is not the positive-definite form that a Hermitian metric on $\mathbb{C}^3$ would give — it is the split form, with three directions of each sign. The two structures are therefore not interchangeable: $\mathbb{C}^3$ names a complex structure on $\mathbb{R}^6$, and the metric $g_{\alpha\beta\gamma\delta}$ is a separate choice on the same space.
A note on the two orders of the coordinates. The corpus's frame lists the time coordinate first and writes $\varepsilon=(1,-1,-1,-1)$, the opposite of the source's order; the two metrics are the same metric written in opposite orders. The article keeps each order where it is used and never mixes them in one display.
The Equations in $\mathbb{R}^6$
Mapping the wave equations into $\mathbb{R}^6$ turns the four spacetime derivatives into six bivector derivatives. The equation is written with the tensor-product generator
$$ \Pi_j := M^{l}_j\otimes\mathbf{1}_{2\bar l+1}-\mathbf{1}_{2l+1}\otimes M^{\bar l}_j, $$
in which $M^{l}_j$ and $M^{\bar l}_j$ are the generators of the two factors.
This is the one two-sided operator in the article. The generator is a difference of two tensor products, and a tensor product with the identity acts on one side only. At the first chain member the carrier is the algebra itself, $V\otimes\bar V\cong M_2(\mathbb{C})\cong\mathbb{B}$, and on an element $X$ the two terms read
$$ \Pi_jX=X\,(M^{l}_j)^{\mathsf T}-M^{\bar l}_j\,X, $$
one multiplication of the field from the right by $(M^{l}_j)^{\mathsf T}$ and one from the left by $M^{\bar l}_j$. The two terms therefore act from opposite sides, and which factor acts from which side is fixed by the order in which the two tensor factors are written: the unbarred factor acts on the right, the barred one on the left, and the other order carries the opposite overall sign. Only the two-sidedness is used below. The operator is nonzero against both sides, so it is neither a left action nor a right one. It is the source's own generator, quoted as the source writes it, and it is not rewritten as a gradient: the gradient is a single left multiplication, and this operator stays two-sided on a field valued in $\mathbb{B}$. Reading the two sides is what shows why this equation is not a gradient equation.
With $a_j$ the three complex coordinates of $\mathbb{C}^3$, $a_j^*$ their conjugates, $\dot{\tilde{Ξ}}_{\mathrm{sp}}$ the conjugate field and $\tilde a_j$ the dual coordinates, the source's equation and its conjugate read
$$ \boxed{\; \begin{aligned} \sum_{j=1}^{3}\Pi_j\frac{\partial\tilde{Ξ}}{\partial a_j} +i\sum_{j=1}^{3}\Pi_j^{*}\frac{\partial\tilde{Ξ}}{\partial a_j^{*}}+k_m\tilde{Ξ} &= 0,\\[6pt] \sum_{j=1}^{3}\Pi_j^{*}\frac{\partial\dot{\tilde{Ξ}}_{\mathrm{sp}}}{\partial\tilde a_j^{*}} -i\sum_{j=1}^{3}\Pi_j\frac{\partial\dot{\tilde{Ξ}}_{\mathrm{sp}}}{\partial\tilde a_j}+m\dot{\tilde{Ξ}}_{\mathrm{sp}} &= 0. \end{aligned} \;} $$
Here $\Pi_j^{*}$ is $\Pi_j$ with both factors conjugated.
The two lines are written in opposite orders. The first line's generator puts the unbarred factor first, $M^{l}_j\otimes\mathbf{1}_{2\bar l+1}-\mathbf{1}_{2l+1}\otimes M^{\bar l}_j$. The source's second line puts the barred factor first, $M^{\bar l}_j\otimes\mathbf{1}_{2l+1}-\mathbf{1}_{2\bar l+1}\otimes M^{l}_j$: the same operator on the same space, with the two tensor factors interchanged. Each line is dimensionally consistent on its own, and the interchange is legitimate, because the conjugate field $\dot{\tilde{Ξ}}_{\mathrm{sp}}$ is written on $\bar V\otimes V$ rather than on $V\otimes\bar V$. In the two-sided reading of the previous paragraph the interchange also swaps the sides: the right multiplication becomes the left one and the left one becomes the right, so the two lines differ in which factor acts from which side and not only in the order of the factors. The display above keeps one ordering for both lines; the interchange is a choice of basis, not a correction.
A check of the imaginary units. The $i$ of the second sum is a scalar coefficient, standing in front of the matrix $\Pi_j^{*}$; whether it is read as the central $i$ or as the pseudoscalar $\omega$ changes nothing in the display, and the reduction below uses only $i^2=-1$. It is not decorative, and the source's own conventions do not decide whether it can be removed. Alongside the equation it writes $g_4=ia_1,\dots,g_6=ia_3$ and $a_1^*=-ig_4,\dots,a_3^*=-ig_6$; the two relations together give $a_j^*=a_j$, which cannot be a conjugation, so at least one of them is a slip. With $a_j^*=-ig_{j+3}$ read as the definition, $\partial/\partial a_j^*=i\,\partial/\partial g_{j+3}$; the $i$ of the second sum then meets the chain-rule $i$, and $i^2=-1$ replaces the pair by a minus. If instead $g_{j+3}=ia_j^*$ is taken as the coordinate, the pair of relations is consistent and the same reduction holds. Either correction removes the $i$; the printed pair, read literally, does not. The display above keeps the $i$ as printed, and the open question below records which relation is meant to be the definition.
The equations are defined in the three-dimensional complex space $\mathbb{C}^3$, which is the bivector space in its complex form; the generalised spin is $s=|l-\bar l|$ and the mass is $m$, exactly as in Minkowski space. The structure of the equations — two tensor-product generators, one from each factor, contracted with the derivatives of one field — is the structure the Dirac equation carries at the first chain member, where the two operators are the gradient and its conjugate and the one equation is the pair of the fourth section. The correspondence is stated, not derived here: the source's generator is two-sided on a field valued in the algebra, its two sides are the two factors of the tensor product, and it is not a derivative, so this equation is not a gradient equation and its Dirac limit is written separately below.
The Proton Chain and the Dirac Limit
The Chain
The chain that carries the nucleon is the spin-$\frac12$ line's Fermi-scheme chain
$$ \Bigl(\tfrac12,0\Bigr)\longleftrightarrow\Bigl(0,\tfrac12\Bigr) \ \longrightarrow\ \Bigl(1,\tfrac12\Bigr)\longleftrightarrow\Bigl(\tfrac12,1\Bigr) \ \longrightarrow\ \cdots\ \longrightarrow\ \Bigl(\tfrac{59}{2},29\Bigr)\longleftrightarrow\Bigl(29,\tfrac{59}{2}\Bigr)\ \longrightarrow\ \cdots $$
The first member is the fundamental doublet: the representation $\tau^{\frac12,0}\oplus\tau^{0,\frac12}$, four complex components in all — the two spin projections $\pm\frac12$ in two copies, one copy from each factor — which is the electron. The Dirac field reads the two copies as the two parts named by the subscript, which in the massless limit are the two helicities. Each single arrow is one interlocking step, adding $\frac12$ to both labels; each double arrow pairs a representation with its conjugate.
The Dirac Equation on the Doublet
On the fundamental doublet the Bhabha equation is the Dirac equation. At that member the carrier is the algebra itself, so the field is an element of $\mathbb{B}$ and the equation is written on the element; no carrier is introduced beside it.
The two values. The subscript of an element is its chirality, and it takes two values and no more: an element is written $\tilde{Ξ}_{L}$ or $\tilde{Ξ}_{R}$, the $L$ value and the $R$ value, and only the zero element carries both. The subscript belongs to the element and not to the field alone: An invertible element and a general element carry it as well.
The equation. On the element $\tilde{Ξ}=\tilde{Ξ}_{L}+\tilde{Ξ}_{R}$ the equation is the pair
$$ \boxed{\;\tilde\nabla_\Phi\tilde{Ξ}_{R}=k_m\tilde{Ξ}_{L},\qquad (\tilde\nabla_\Phi)^{\natural}\tilde{Ξ}_{L}=k_m\tilde{Ξ}_{R}\;} $$
Each member is one left multiplication, by the gradient or by its quaternion conjugate; one equation per value; the mass is the only term that couples them. At $m=0$ the pair decouples into $\tilde\nabla_\Phi\tilde{Ξ}_{R}=0$ and $(\tilde\nabla_\Phi)^{\natural}\tilde{Ξ}_{L}=0$. Applying $(\tilde\nabla_\Phi)^{\natural}$ to the first member and using the second gives $(\tilde\nabla_\Phi)^{\natural}\tilde\nabla_\Phi\tilde{Ξ}_{R}=k_m^{2}\tilde{Ξ}_{R}$, that is
$$ \Box_\Phi\tilde{Ξ}_{R}=k_m^{2}\tilde{Ξ}_{R}, $$
and the same step on the other member gives $\Box_\Phi\tilde{Ξ}_{L}=k_m^{2}\tilde{Ξ}_{L}$: each value obeys the Klein–Gordon equation, both on the same element.
Why one equation will not do. A single gradient does not square to the scalar $\Box_\Phi$. Its square is a derived form of this section, read from the multiplication of the units,
$$ \tilde\nabla_\Phi^2=\frac{\partial^2}{\partial R_t^2}-\Delta_\Phi+2\sum_\eta\vec e_\eta\frac{\partial^2}{\partial R_t\partial R_\eta}, $$
because the time unit commutes with the three spatial units and the mixed derivatives survive. So a first-order equation on one value cannot be turned into the Klein–Gordon equation by squaring it, and the first-order operator whose square is $\Box_\Phi$ needs both values. This is why the massive equation is a pair, while the massless equation stays first order, $\tilde\nabla_\Phi\tilde{Ξ}_{R}=0$ and $(\tilde\nabla_\Phi)^{\natural}\tilde{Ξ}_{L}=0$.
How the two are one. The two members are two equations on one element: each is one left multiplication, by the gradient or by its conjugate; no derivative acts from the right; the mass is the only term that relates the two parts; and no third object stands behind them. On the material sector the same operator is written $\Box_\Phi$, the Klein–Gordon operator in the real time, as Operator and Calculus Conventions states.
The Proton
The nucleon sits on the same chain, far up it. The programme takes the chain member
$$ \tau^{\frac{59}{2},29}\oplus\tau^{29,\frac{59}{2}} $$
as the proton. The first factor has degree $(2l+1)(2\bar l+1)=60\times59=3540$, and the conjugate factor has the same dimension, so the pair carries $7080$ components. The mass comes from the formula of the previous section. Normalising so that the electron's fundamental representation gives
$$ m_e=\mu_0\Bigl(\frac12+\frac12\Bigr)\Bigl(0+\frac12\Bigr)=\frac{\mu_0}{2}, $$
the ratio is
$$ \boxed{\;\frac{m_p}{m_e} = \frac{(2l+1)(2\bar l+1)}{2} =2\Bigl(l+\frac12\Bigr)\Bigl(\bar l+\frac12\Bigr)\;} $$
and $(59/2,29)$ gives $60\times59/2=1770$. The product $(l+\frac12)(\bar l+\frac12)=30\times29.5=885$ is $m_p/\mu_0$, not the ratio: since $m_e=\mu_0/2$, the ratio is twice that product. The source's own line $m_e=\mu_0(l+\frac12)=\mu_0$ drops the second factor $\bar l+\frac12=\frac12$, and that dropped factor is the whole of the factor of two. Its condition $(l+\frac12)(\bar l+\frac12)\approx1800$ becomes $\approx918$ once the mass formula is applied to the electron as well as to the proton; the member's $885$ falls short of $918$ by $3.6\%$, which is the $1770$ against $1836$.
The proton's equation in $\mathbb{R}^6$ is the equation of the $\tau^{59/2,29}$ factor — $3540$ components for the factor, doubled for the pair — with $M^{59/2,29}_j$ replacing $M^{1/2,0}_j$. Mapping the Dirac equation into $\mathbb{R}^6$ gives the three generators of the first member as the three elements $\tfrac12c_{\frac12,\frac12}\,i\vec e_j$ of the algebra, the three spatial units read with the central $i$; the value $c_{\frac12,\frac12}=2$ makes them exactly $i\vec e_j$. This is the concrete way in which the chain's first member carries the Dirac equation, and the dictionary between the algebra and its images is the corpus's, in The Daviau Map and the Space Clifford Formulation of the Dirac Equation.
The programme's stated reason for moving up the chain is that the first member is algebraically too simple to carry the proton's internal structure. The move itself is a choice, not a derivation.
The Biquaternion Reading
Three points connect this material to the framework, of three different strengths: an identification of algebras, an agreement of objects, and a relation of equations.
First, the fundamental doublet's factor is the biquaternion algebra. $\mathbb{C}^2$ is the defining module of $\mathbb{B}=\mathbb{C}\otimes_{\mathbb{R}}\mathbb{H}\cong M_2(\mathbb{C})$, and the programme's chain of tensor powers begins at the centre $\mathbb{C}$ and reaches $\mathbb{B}$ at its second member, $\mathbb{C}^2\otimes\bar{\mathbb{C}}^2\cong\mathrm{End}(\mathbb{C}^2)=M_2(\mathbb{C})$. The corpus's Higher Spin from Tensor Products: The Three Carriers and the Absence of a Fourth gives the framework's account of the same ladder of tensor powers; the present chain is the programme's version. One caution: the programme writes "$\mathbb{C}^2$ and complex conjugate $\bar{\mathbb{C}}^2$ are biquaternion algebras", which conflates the algebra with its module. $\mathbb{C}^2$ is a module on which $\mathbb{B}\cong M_2(\mathbb{C})$ acts; the algebra is the matrix algebra, and the tensor products of the chain are tensor products of modules until an algebra structure is imposed. The article keeps the distinction.
Second, the bivector space is the corpus's field-strength space. The six-dimensional space of antisymmetric two-tensors is, after the electric–magnetic split, a pair of a vector and an axial vector, and in the corpus's notation it is the field-strength biquaternion $\vec E+i\vec B$ with six real components. The mapping $g_{ab}\to g_{\alpha\beta\gamma\delta}$ gives the metric of the six-dimensional space on which the corpus's two invariants live: the first, $\vec E^2-\vec B^2$, is the quadratic form of that metric, and the second, $\vec E\cdot\vec B$, is the pairing of $\vec E$ with $\vec B$ through the duality of the same space. The programme's $\mathbb{C}^3$ is then the complex form in which $\vec B$ and $\vec E$ are the two halves of one complex vector. This is a genuine agreement of objects, and the article records it as one.
Third, the equations themselves are the corpus's Dirac–Maxwell package in the programme's form. The Dirac limit's generators are the three elements $i\vec e_\eta$ of the algebra, and the massless limit of the pair is the Weyl pair, $\tilde\nabla_\Phi\tilde{Ξ}_{R}=0$ and $(\tilde\nabla_\Phi)^{\natural}\tilde{Ξ}_{L}=0$; the Maxwell field as the programme writes it is the companion identification, and the corpus's Bispinor Fields derives the Maxwell–Dirac solution in its own form. The two are related and not identical, and the relation is cited rather than re-derived here. The higher chain members are the multi-mass generalisations, and they lie outside $\mathbb{B}$ exactly as the companion articles say.
What the Framework Establishes, Transcribes, and Does Not
Established, and recomputed.
- The spintensor degree $\dim\mathrm{sym}^{(k,r)}=(k+1)(r+1)=(2l+1)(2\bar l+1)$ at $l=k/2$, $\bar l=r/2$, and its non-injectivity across spin lines; for the proton's first factor $60\times59=3540$.
- The $\pm$ pairing of the eigenvalues of $M_0$, derived from the evenness in $p_0$ of the rest-frame determinant.
- The bivector metric's $(3,3)$ inertia, and the printed diagonal as the componentwise image in the order $10,20,30,23,31,12$; recomputed for all six collective indices.
- The Dirac limit: the product of the gradient with its quaternion conjugate is $\Box_\Phi$, the non-scalar square $\tilde\nabla_\Phi^2=\partial^2/\partial R_t^2-\Delta_\Phi+2\sum_\eta\vec e_\eta\,\partial^2/\partial R_t\partial R_\eta$, and the elimination of the pair by which each value obeys $\Box_\Phi\tilde{Ξ}=k_m^{2}\tilde{Ξ}$; recomputed on the coefficient representation over $100$ random complex four-momenta, the product at residual $1.1\times10^{-14}$, the elimination following from the second member.
- The proton-to-electron mass ratio: the product $(l+\frac12)(\bar l+\frac12)$ is $30\times29.5=885$ at $(59/2,29)$ and $1\times\frac12$ at the electron's $(1/2,0)$, so the ratio is $885/\frac12=1770$.
Transcribed, not derived.
- The wave equations in Minkowski space and their images in $\mathbb{R}^6\cong\mathbb{C}^3$.
- The determinant factorisation $D(p)=D(s^2)$, which is the source's Lorentz-invariance argument and is not re-proved here.
- The interlocking schemes and the assignment of the proton to the chain member $(59/2,29)$.
- The source's eigenvalue relation $m_i=\mu_0\lambda_i$ and its product form. The literal rest-frame determinant gives $m_i=m/|\lambda_i|$ instead, an inverse dependence, so the source's reading is not the determinant's reading with a scale absorbed.
- The printed collective index order, which the article corrects to the boost-first order.
- The conjugate-coordinate relations $g_{j+3}=ia_j$ and $a_j^*=-ig_{j+3}$, which cannot both hold.
Gap, left visible.
- The functional form of the mass in the eigenvalues — direct, as the source's product, or inverse, as the determinant — is not settled by the source.
- The two printed relations for the conjugate coordinate in $\mathbb{R}^6$ are mutually inconsistent, so whether the second sum's $i$ reduces by $i^2=-1$ is not decided by the source.
- The degree fixes the mass and not the particle, so an independent spin label is needed for every assignment.
- The passage from the abstract chain to a physical particle remains a choice fixed by the mass ratio.
- The infinite-dimensional limit is used to state the mass formula while the chains used are finite-dimensional; the same weakness as the companion mass article.
- The Dirac–Maxwell relation is cited as known and not re-derived here.
Open Questions
-
The eigenvalue normalisation. The rest-frame determinant has roots $p_0=-m/\lambda_i$, so the masses it carries are $m_i=m/|\lambda_i|$, which fall as the eigenvalues grow. The source writes the roots as $p_0=\mu_0\lambda_i$ and reads off $m_i=\mu_0\lambda_i\bar\lambda_i$, which rise. The two cannot both be the statement of one determinant: they agree only if all $|\lambda_i|$ are equal, and they differ by an inversion otherwise. A reconciliation would need eigenvalues falling as the degree grows, and the generators of $\tau^{l\bar l}$ carry weights of order $l+\bar l$, which grow. As it stands, the companion's rising spectrum does not come from this determinant in the form displayed. Which dependence is intended?
-
The conjugate coordinate in $\mathbb{R}^6$. The source writes both $g_{j+3}=ia_j$ and $a_j^*=-ig_{j+3}$, which together give $a_j^*=a_j$. Which is the definition? Under the reading $g_{j+3}=ia_j^*$ the $i$ of the second sum reduces by $i^2=-1$; under the printed reading it does not.
-
$\mathbb{C}^3$ versus $\mathbb{R}^6$ in the corpus's hands. The corpus's field-strength biquaternion $\vec E+i\vec B$ is an element of the vector part of $\mathbb{B}$, which is $\mathbb{C}^3$ over the central $i$; the programme's bivector space is the real $\mathbb{R}^6$. Are the two complex structures the same — the corpus's central imaginary unit and the $i$ of the programme's $\mathbb{C}^3$ — or does the identification stop at the real space?
-
Does the chain of algebras carry the mass spectrum? The companion mass article orders a spin line by degree; the present chain orders the same line by algebra. Is the chain's algebra sequence the object that fixes which chain members are physical?
-
The multi-mass sector. The equations have several roots, only one of which is used. Are the subleading masses physical — further particles on the same chain — and does the corpus have an interpretation for them?
-
The Maxwell–Dirac package. The corpus's Bispinor Fields article derives the Maxwell–Dirac solution in the corpus's form. Does the programme's massless limit coincide with it exactly, including the conventions of the axial current?
Summary
The Bhabha–Gel'fand–Yaglom wave equations place a field in an arbitrary Lorentz representation $\tau^{l\bar l}$, carried by coefficients of symmetric spintensors of degree $(k+1)(r+1)=(2l+1)(2\bar l+1)$, and give it a matrix mass whose determinant is a polynomial in $s^{2}=-N$ with roots $m_i^2$. The $\pm$ pairing of the eigenvalues is a consequence of the determinant's evenness at rest; the source's reading of the masses from those eigenvalues is not the determinant's, and is recorded as an open question. The representations are organised into interlocking chains — the Bose and Fermi schemes — and the fields of type $(l,\bar l)\oplus(\bar l,l)$ carry generalised spin $s=|l-\bar l|$.
The equations can be mapped into the six-dimensional bivector space $\mathbb{R}^6\cong\mathbb{C}^3$. The printed bivector metric is the componentwise image of the Minkowski metric under the boost-first collective order $10,20,30,23,31,12$; the order printed in the source interleaves the two classes and does not reproduce it. With the corrected order, the three boosts carry $-1$ and the three rotations $+1$, which is the $(3,3)$ split of the Lorentz parameter space. The $\mathbb{R}^6$ equation keeps the source's explicit $i$ in its second sum: the source's two relations for the conjugate coordinate are mutually inconsistent, so the reduction of that pair of imaginary units by $i^2=-1$ is stated as a consequence of one reading and left open, rather than performed silently.
The first chain member is the electron's Dirac doublet, and its Dirac equation is the pair $\tilde\nabla_\Phi\tilde{Ξ}_{R}=k_m\tilde{Ξ}_{L}$, $(\tilde\nabla_\Phi)^{\natural}\tilde{Ξ}_{L}=k_m\tilde{Ξ}_{R}$, one left multiplication per member and one equation per value read by the test. The two members are two equations on one element, and the mass is the only term that couples them. The single gradient cannot do the work alone: $\tilde\nabla_\Phi^2$ is not scalar, the mixed derivatives through $e_t$ survive, so no first-order equation on one value has a scalar square, and the massive equation is necessarily a pair. The nucleon is taken at the chain member $(59/2,29)$ of degree $3540$; the ratio $m_p/m_e=1770$ is $3.6\%$ below the measured value.
The framework's relation to this is the fundamental doublet's factor: $\mathbb{C}^2$ is the defining module of the biquaternion algebra, the bivector space is the corpus's field-strength space, and the equations' massless limit is the corpus's Weyl pair, with the Maxwell field as the programme writes it as the companion identification. The higher chain members lie outside $\mathbb{B}$, and the physical assignments are choices fixed by the mass ratio.
Summary of Notation
| Symbol | Meaning |
|---|---|
| $s_{\alpha_1\dots\alpha_k\bar\alpha_1\dots\bar\alpha_r}$ | Spintensor of the representation $\tau^{l\bar l}$ |
| $l=k/2$, $\bar l=r/2$ | Lorentz labels from the numbers of unbarred/barred indices |
| $\dim\mathrm{sym}^{(k,r)}=(k+1)(r+1)$ | Degree of the representation |
| $s=\lvert l-\bar l\rvert$ | Generalised spin of the field type $(l,\bar l)\oplus(\bar l,l)$ |
| $\tilde{Ξ}_{L}$, $\tilde{Ξ}_{R}$ | The two parts of the Dirac element named by the subscript; one equation per part |
| $M^{l\bar l}_\mu$, $M(p)$ | Generator matrices of the representation and the momentum matrix |
| $D(p)=\det(M(p)+mE)$ | Determinant whose roots are the masses |
| $\mu_0$, $\lambda_i$ | Mass scale and eigenvalues of $M_0$ |
| $g_{ab}$, order $10,20,30,23,31,12$ | Bivector-space metric and its boost-first collective order |
| $a_j$, $a_j^*$ | Three complex bivector coordinates and their conjugates |
| $\Pi_j=M^{l}_j\otimes\mathbf{1}-\mathbf{1}\otimes M^{\bar l}_j$, $\Pi_jX=X(M^{l}_j)^{\mathsf T}-M^{\bar l}_jX$ | Tensor-product generator of the $\mathbb{R}^6$ equation; two-sided, the article's one such operator |
| $\mathbb{C}^3\cong\mathbb{R}^6$ | Complex form of the bivector space, Lorentz parameter space |
| $(59/2,29)$, degree $3540$ | The proton's chain member |
| $\tilde\nabla_\Phi$, $(\tilde\nabla_\Phi)^{\natural}$, $\Box_\Phi$ | The gradient, its quaternion conjugate and the d'Alembertian in the material reading, all of Operator and Calculus Conventions |
| $\tilde\nabla_\Phi\tilde{Ξ}_{R}=k_m\tilde{Ξ}_{L}$, $(\tilde\nabla_\Phi)^{\natural}\tilde{Ξ}_{L}=k_m\tilde{Ξ}_{R}$ | The Dirac pair: each member one left multiplication, one equation per value |
Here $k_m=mc/\hbar$ is the mass wave number, the inverse of the mass length.
Further Reading
- H. J. Bhabha, "Relativistic wave equations for the elementary particles," Reviews of Modern Physics 17 (1945) 200–216, for the finite-dimensional relativistic wave equations and their multi-mass spectra.
- I. M. Gel'fand and A. M. Yaglom, "General relativistic-invariant equations with a mass spectrum," Zhurnal Eksperimentalnoi i Teoreticheskoi Fiziki 18 (1948) 703–733, for the general theory of the equations with a mass spectrum.
- E. Majorana, "Teoria relativistica di particelle con momento intrinseco arbitrario," Nuovo Cimento 9 (1932) 335–344, for the infinite-component equation.
- I. M. Gel'fand, R. A. Minlos and Z. Ya. Shapiro, Representations of the Rotation and Lorentz Groups and Their Applications (Pergamon, 1963), for the representations, the spintensors and the interlocking schemes.
- R. Penrose and W. Rindler, Spinors and Space-Time, vol. 1 (Cambridge, 1984), for the van der Waerden two-index calculus of the Lorentz group.
- A. Z. Petrov, Einstein Spaces (Pergamon, 1969), for the mapping of the curvature tensor into $\mathbb{R}^6$ and the classification of Einstein spaces.
- V. V. Varlamov, "Spinor Structure and Internal Symmetries," International Journal of Theoretical Physics 54 (2015) 3533–3576 (arXiv:1409.1400), for the wave equations, the chains and the bivector-space reduction recorded here.
- Companion articles: The Spin–Mass Formula and the Spectrum of Elementary Particles in Biquaternionic Form; The Spin–Charge Hilbert Space and the SU(3) Supermultiplets in Biquaternionic Form; Bhabha Scattering in Biquaternionic Form; The Field-Strength Biquaternion and Its Invariants; The Dirac–Hestenes Equation and Spacetime Algebra in Biquaternionic Form; Bispinor Fields and the Fundamental Solution of the Generalized Maxwell–Dirac Equation.