BRST Symmetry in Biquaternionic Form

Introduction

A massless gauge field has more components than physical states, and the excess is removed by a symmetry. The companion article Canonical Quantization of the Biquaternion Maxwell Field removes it by the Gupta–Bleuler route: a covariant mode expansion with an indefinite metric and a subsidiary condition that selects the physical subspace. That route exhibits the physical states but leaves the gauge fixing an external input — a choice of gauge that the biquaternion algebra does not make. The present article describes the other route, the one that makes the gauge fixing itself a symmetry: the BRST symmetry of the gauge-fixed theory.

The BRST construction promotes the gauge parameter to an anticommuting field, the Faddeev–Popov ghost, and introduces a nilpotent odd derivation $s$ under which the gauge field, the ghost, the antighost and the Nakanishi–Lautrup auxiliary field transform. The gauge-fixed Lagrangian is $s$-exact, the physical states are the cohomology of $s$, and the unitarity of the physical subspace follows from the nilpotency. This article carries that construction into the biquaternion framework, and its interest is not in the transcription — the construction is standard — but in two facts that the transcription makes visible.

First, the nilpotency $s^2=0$ rests on the Jacobi identity of the gauge algebra, and in the biquaternion framework the gauge algebra of a simple group is realized by the commutator inside the algebra: the generators lie in the informational sector $\mathbb{M}_+$, their brackets are the adjoint action, and the Jacobi identity is a consequence of the associativity of quaternion multiplication. The nilpotency is therefore not an independent axiom in the framework; it is inherited from associativity. Second, the ghost fields cannot live in $\mathbb{B}$: the algebra is ungraded and every element of it is even, so it has no anticommuting elements, and the BRST doublet requires the Grassmann envelope $\mathbb{B}\otimes\Lambda$. This is the same kind of statement as the Fock companion's result that the ladder algebra is not native to $\mathbb{B}$, and it is stated here in the same spirit: the algebra supplies the module and the symmetry, and the odd coordinates must be adjoined.

The article is organized as follows. The gauge-fixing problem and the ghost fields are stated, and the Grassmann envelope is introduced. The BRST transformation is written for the abelian and the non-abelian cases, and the nilpotency is derived and verified. The biquaternion assignment of the fields to the two sectors is made, and the role of associativity in the Jacobi identity is isolated. The gauge-fixing fermion is constructed and the gauge-fixed Lagrangian is exhibited as BRST-exact. The ghost-number grading, the BRST charge, the cohomological characterization of physical states and the quartet mechanism are then given, and the article closes with the accounting of what is standard and what is the framework's.

The Faddeev–Popov procedure that produces the ghost Lagrangian from the path-integral measure is the subject of the companion subcategory on gauge fields; here the ghost terms are taken as given, and the symmetry they carry is the subject. The massive case is the control case: the Proca field of the preceding article has no gauge freedom, hence no ghost sector and no BRST symmetry, and the contrast is stated where it matters.

Companion article Canonical Quantization of the Biquaternion Maxwell Field, for the gauge fixing, the Gupta–Bleuler route, and the constraint structure whose gauge fixing the BRST construction replaces.
Companion article The Gauge Principle in Biquaternionic Form, for the gauge transformation, the covariant derivative, and the gauge algebra.
Companion article Non-Abelian Gauge Fields in Biquaternionic Form, for the non-abelian field strength and the structure constants.
Companion article The Yang–Mills Equation in Biquaternionic Form, for the classical non-abelian equations of motion.
Companion article Fock Space and Creation/Annihilation Operators in Biquaternionic Form, for the argument that the ladder algebra, like the Grassmann coordinates, is not native to $\mathbb{B}$.

Conventions. The algebra is $\mathbb{B}=\mathbb{C}\otimes_\mathbb{R}\mathbb{H}$ with basis $e_0=1,e_1,e_2,e_3$, $e_k^2=-e_0$, $e_je_k=-\delta_{jk}e_0+\varepsilon_{jkl}e_l$, central $i$, material sector $\mathbb{M}_-$ and informational sector $\mathbb{M}_+$. The gauge potential is $\tilde{A}\in\mathbb{M}_-$, the gradient is $\tilde{\nabla}=e_0\partial_{ict}+\sum_ke_k\partial_k$, and the field strength is $\tilde{F}=\tilde{\nabla}^{\natural}\tilde{A}-\mathrm{Sc}(\tilde{\nabla}^{\natural}\tilde{A})$, the traceless material part of $\tilde{\nabla}^{\natural}\tilde{A}$. A Lie algebra of a simple gauge group is realized inside the framework by the commutator, the canonical example being $\mathrm{SU}(2)$ with the Hermitian generators $T^a=ie_a\in\mathbb{M}_+$, the brackets $[T^a,T^b]=2i\varepsilon^{abc}T^c=if^{abc}T^c$ and the real structure constants $f^{abc}=2\varepsilon^{abc}$; equivalently, in its anti-Hermitian normalization the same algebra is the compact form $\mathrm{span}_{\mathbb{R}}\{e_1,e_2,e_3\}\subset\mathbb{M}_-$ with $[e_i,e_j]=2\varepsilon_{ijk}e_k$. The gauge coupling is $g$, the structure constants are $f^{abc}$, and the covariant derivative is $D_\mu=\partial_\mu+gA_\mu$ in the appropriate representation. Ghosts are Grassmann-odd fields, taking values in the Grassmann envelope $\mathbb{B}\otimes\Lambda$ of the algebra; the graded Leibniz rule is $s(\tilde{Q}\tilde{R})=(s\tilde{Q})\tilde{R}+(-1)^{|\tilde{Q}|}\tilde{Q}(s\tilde{R})$ with $|\tilde{Q}|$ the Grassmann parity.

The Gauge-Fixing Problem and the Ghost Fields

The Gauge-Fixed Lagrangian

The Maxwell Lagrangian $\mathcal{L}_{\mathrm{M}}=-\tfrac14F_{\mu\nu}F^{\mu\nu}$ is invariant under $\tilde{A}\mapsto\tilde{A}-\tilde{\nabla}\Gamma$, and the kinetic operator is therefore not invertible: the propagator of the companion article exists only after a gauge is chosen. The gauge is fixed by adding to the Lagrangian a term that is not gauge invariant but whose effect on physical quantities cancels, and the modern way to write that term introduces two auxiliary fields. The first is the Nakanishi–Lautrup field $B$, a Lorentz scalar of ghost number zero; the second is the pair of Faddeev–Popov ghosts $c$ and $\bar{c}$, anticommuting Lorentz scalars of opposite ghost number. For the abelian theory the gauge-fixing and ghost terms are

$$ \mathcal{L}_{\mathrm{gf}} = B\,\partial_\mu A^\mu + \tfrac12\xi B^2, \qquad \mathcal{L}_{\mathrm{gh}} = -\,\bar{c}\,\Box\,c, $$

with $\xi$ the gauge parameter. The field $B$ is auxiliary — its equation of motion is $B=-\xi^{-1}\partial_\mu A^\mu$, and eliminating it returns the familiar $-\tfrac{1}{2\xi}(\partial\cdot A)^2$ gauge fixing — and the ghosts are the fields that the Faddeev–Popov determinant contributes. For the non-abelian theory the same structure holds with the gauge-fixing term $B^a\partial_\mu A^{a\mu}+\tfrac12\xi B^aB^a$ and the ghost term $-\bar{c}^a\partial_\mu(D^\mu c)^a$, with $D_\mu c^a=\partial_\mu c^a+g f^{abc}A_\mu^b c^c$.

The gauge-fixed Lagrangian $\mathcal{L}=\mathcal{L}_{\mathrm{gauge}}+\mathcal{L}_{\mathrm{gf}}+\mathcal{L}_{\mathrm{gh}}$ is no longer gauge invariant — the gauge invariance has been destroyed by the gauge-fixing term — but it carries a new invariance that replaces it.

The Ghosts and the Grassmann Envelope

The ghost fields are anticommuting: $\{c(x),\bar{c}(y)\}=0$ at spacelike separation and, formally, $c^2=0$ as a classical field. The biquaternion algebra contains no such element. It is not a division algebra — $\mathbb{B}\cong M_2(\mathbb{C})$ contains nonzero nilpotents, among them $(e_1+ie_2)/2$, whose square vanishes — but nilpotency is not oddness: the algebra is ungraded, every element of $\mathbb{B}$ is even, and the product of two elements cannot be odd. An element that anticommutes with its fellows does not exist in $\mathbb{B}$, whatever its square. The ghosts therefore cannot be elements of $\mathbb{B}$. They must be adjoined, and the correct structure is the Grassmann envelope

$$ \mathbb{B}\otimes\Lambda, \qquad \Lambda = \Lambda^0\oplus\Lambda^1, $$

with $\Lambda^1$ the odd part generated by anticommuting $\theta_i$. A ghost is an element of the odd part of the envelope, $c=c^aT^a\otimes\theta$, and the even fields of the theory are elements of the even part. This is the same move the Fock companion makes for the ladder operators, and it is recorded here for the same reason: the framework does not contain the odd coordinates, it is tensored with the algebra that does.

The assignment of the gauge-algebra generators $T^a$ is a framework statement. A simple gauge algebra is realized by the commutator on a subspace of the algebra; for $\mathrm{SU}(2)$ the generators can be taken as the imaginary units,

$$ T^a = ie_a, \qquad [T^a,T^b]=2i\varepsilon^{abc}T^c, $$

so that the structure constants are $f^{abc}=2\varepsilon^{abc}$, and the generators lie in the informational sector $\mathbb{M}_+$. The gauge field, by contrast, lies in the material sector $\mathbb{M}_-$. The BRST construction therefore couples the two sectors through the gauge structure: the odd parameter and the generators are informational, the field that transforms is material.

The BRST Transformation

Definition

The BRST transformation is a Grassmann-odd derivation $s$ of the field algebra. On the abelian fields it acts as

$$ sA_\mu = \partial_\mu c, \qquad sc = 0, \qquad s\bar{c} = B, \qquad sB = 0, $$

and on the non-abelian fields as

$$ sA_\mu^a = (D_\mu c)^a = \partial_\mu c^a+g f^{abc}A_\mu^b c^c, \qquad sc^a = -\tfrac{g}{2}f^{abc}c^bc^c, \qquad s\bar{c}^a = B^a, \qquad sB^a = 0 . $$

The transformation on the gauge field is the gauge transformation $\delta A_\mu^a=(D_\mu\Gamma)^a$ with the parameter $\Gamma^a$ replaced by the ghost $c^a$: that is the sense in which the gauge parameter has been promoted to a field. The transformations of $c$ and $\bar{c}$ are the new content, and they are chosen so that $s$ is nilpotent.

Nilpotency in the Abelian Case

The abelian case is immediate. Since $s$ is a derivation of Grassmann parity one and $c$ is odd,

$$ s^2 A_\mu = s(\partial_\mu c) = \partial_\mu(sc) = 0, \qquad s^2 c = 0, \qquad s^2\bar{c} = sB = 0, \qquad s^2 B = 0 . $$

The only nonvanishing entry is $s\bar{c}=B$, and $B$ is annihilated by $s$. Hence $s^2=0$ on all abelian fields, with no use made of any structure identity.

Nilpotency in the Non-Abelian Case

The non-abelian case is where the identity is needed. Applying $s$ to $sA_\mu^a$ and using the graded Leibniz rule,

$$ s^2A_\mu^a = \partial_\mu(sc^a)+gf^{abc}\left[(sA_\mu^b)c^c+A_\mu^b(sc^c)\right]. $$

The two derivative pieces cancel between themselves: $\partial_\mu(sc^a)=-\tfrac{g}{2}f^{abc}\partial_\mu(c^bc^c)$ and $(sA_\mu^b)c^c$ contributes $gf^{abc}(\partial_\mu c^b)c^c$, and since $\partial_\mu(c^bc^c)=(\partial_\mu c^b)c^c+c^b\partial_\mu c^c$ while the antisymmetry of $f^{abc}$ in its last two indices converts $f^{abc}c^b\partial_\mu c^c$ into $f^{abc}(\partial_\mu c^b)c^c$, their sum is $gf^{abc}\left[-(\partial_\mu c^b)c^c+(\partial_\mu c^b)c^c\right]=0$. What survives is the $A$-dependent part,

$$ s^2A_\mu^a = g^2 f^{abc}f^{bde}A_\mu^d c^ec^c - \tfrac{g^2}{2}f^{abc}f^{cde}A_\mu^b c^dc^e , $$

and the two surviving terms cancel by the Jacobi identity

$$ f^{abd}f^{dce}+f^{bcd}f^{dae}+f^{cad}f^{dbe}=0 . $$

The same identity controls $s^2c^a$:

$$ s^2c^a = -\tfrac{g}{2}f^{abc}\left[(sc^b)c^c-c^b(sc^c)\right] = \tfrac{g^2}{4}f^{abc}\left[f^{bde}c^dc^ec^c - f^{cde}c^bc^dc^e\right] = 0 . $$

Both nilpotency conditions were verified: with the $\mathrm{SU}(2)$ structure constants in both normalizations — the realized $f^{abc}=2\varepsilon^{abc}$ and, as a rescaling check, $f^{abc}=\varepsilon^{abc}$ — a Grassmann algebra on odd generators, and random gauge-field components, the coefficient of every Grassmann monomial in $s^2A_\mu^a$ and in $s^2c^a$ was zero, the two derivative pieces of $s^2A_\mu^a$ cancelled separately, and the Jacobi combination vanished identically.

Nilpotency and Associativity

In the biquaternion framework the Jacobi identity is not an independent postulate. The gauge algebra is realized by the commutator inside $\mathbb{B}$, and the commutator of a real associative algebra automatically satisfies the Jacobi identity:

$$ [[\tilde{Q},\tilde{R}],\tilde{S}]+[[\tilde{R},\tilde{S}],\tilde{Q}]+[[\tilde{S},\tilde{Q}],\tilde{R}]=0 \qquad\text{for all }\tilde{Q},\tilde{R},\tilde{S}\in\mathbb{B}, $$

which is a rearrangement of the associativity of the product. For the realization $\mathrm{SU}(2)=\mathrm{span}_{\mathbb{R}}\{ie_1,ie_2,ie_3\}$ the identity is exactly the statement that quaternion multiplication is associative, and the numerical check of the Jacobi combination for $f^{abc}=\varepsilon^{abc}$ returning zero is a check of that associativity through a chain of identifications.

The chain is worth stating in full, because it is the framework's contribution to the BRST construction.

The gauge algebra is a Lie algebra realized by the commutator on a subspace of $\mathbb{B}$.
The structure constants satisfy the Jacobi identity because the product of $\mathbb{B}$ is associative.
The Jacobi identity is what makes the non-abelian BRST transformation nilpotent.
Nilpotency is what makes the physical-state condition $Q|\mathrm{phys}\rangle=0$ consistent and the physical subspace positive.

Therefore, to the extent that the gauge algebra is realized inside the biquaternion algebra, the nilpotency of the BRST operator — and with it the consistency of the gauge-fixed quantization — is inherited from the associativity of the algebra rather than imposed. This is a genuine algebraic statement, and it is the sharpest thing the framework says about BRST symmetry. What it does not do is select which gauge algebra is realized, or fix the gauge parameter; those remain inputs.

The Gauge-Fixing Fermion and the BRST-Exact Lagrangian

The gauge-fixing and ghost terms are not merely invariant under $s$; they are $s$-exact. Define the gauge-fixing fermion

$$ \Psi = \bar{c}\left(\partial_\mu A^\mu+\tfrac{\xi}{2}B\right) \qquad\text{(abelian)}, \qquad \Psi = \bar{c}^a\left(\partial_\mu A^{a\mu}+\tfrac{\xi}{2}B^a\right) \qquad\text{(non-abelian)} . $$

Then a direct computation with the graded Leibniz rule gives, in the abelian case,

$$ s\Psi = (s\bar{c})\left(\partial_\mu A^\mu+\tfrac{\xi}{2}B\right) - \bar{c}\,s\!\left(\partial_\mu A^\mu+\tfrac{\xi}{2}B\right) = B\,\partial_\mu A^\mu+\tfrac{\xi}{2}B^2-\bar{c}\,\Box\,c, $$

where the minus sign on the second term is the sign $(-1)^{| \bar{c}|}=-1$ that the oddness of $\bar{c}$ requires. This is exactly $\mathcal{L}_{\mathrm{gf}}+\mathcal{L}_{\mathrm{gh}}$, so

$$ \mathcal{L} = \mathcal{L}_{\mathrm{gauge}} + s\Psi . $$

Since $\mathcal{L}_{\mathrm{gauge}}$ is gauge invariant and hence $s$-invariant (the gauge transformation is $sA_\mu^a=(D_\mu c)^a$ with $c$ the ghost), the whole gauge-fixed Lagrangian is $s$-invariant, and the invariance is automatic once $s^2=0$:

$$ s\mathcal{L} = s\mathcal{L}_{\mathrm{gauge}} + s^2\Psi = 0 . $$

The non-abelian case is the same computation with the covariant derivative, $s\Psi=B^a\partial_\mu A^{a\mu}+\tfrac{\xi}{2}B^aB^a-\bar{c}^a\partial_\mu(D^\mu c)^a$, so that the ghost Lagrangian is $-\bar{c}^a\partial_\mu(D^\mu c)^a$ and the gauge-fixing Lagrangian is $B^a\partial_\mu A^{a\mu}+\tfrac{\xi}{2}B^aB^a$.

The structure of the result is the reason BRST is the natural framework for gauge theories: the entire unphysical sector — gauge-fixing term and ghosts — is the image of $s$, and the physical content is what remains after the image is quotiented out. In the biquaternion framework the statement is that $\Psi$ is a Grassmann-odd element of the envelope, $s\Psi$ is its image, and the physical states are the cohomology of $s$, as below.

Ghost Number, the BRST Charge, and Cohomology

Grading

The ghost number is a grading of the field algebra, additive under products, defined by

$$ \mathrm{gh}(\tilde{A})=0, \qquad \mathrm{gh}(c)=+1, \qquad \mathrm{gh}(\bar{c})=-1, \qquad \mathrm{gh}(B)=0, $$

and extended to products and to composite operators by additivity. The BRST operator raises the ghost number by one,

$$ \mathrm{gh}(s\tilde{Q})=\mathrm{gh}(\tilde{Q})+1, $$

so that $s$ is a cochain map of degree $+1$ on the graded algebra, and $s^2=0$ makes the pair $(\mathbb{B}\otimes\Lambda,s)$ into a cochain complex, the BRST complex. The gauge-fixing fermion has ghost number $-1$, consistent with $s\Psi$ having ghost number $0$.

The BRST Charge

Because the transformation is a symmetry of the action, it has a conserved charge by Noether's theorem: the BRST charge $Q$, a Grassmann-odd operator with

$$ Q^2 = 0, \qquad s\tilde{Q} = [iQ,\tilde{Q}]_\pm , $$

the bracket being a commutator for even $\tilde{Q}$ and an anticommutator for odd $\tilde{Q}$. The nilpotency $Q^2=0$ is the operator statement of $s^2=0$, and the charge is the generator of the symmetry in the same sense that the ordinary gauge charge generates the gauge transformation. In the biquaternion framework the charge is a Grassmann-odd operator on the Fock space of the gauge and ghost fields; it is not an element of $\mathbb{B}$, for the same reason the ghosts are not.

Physical States

The physical subspace is defined by

$$ \mathcal{H}_{\mathrm{phys}} = \frac{\ker Q}{\mathrm{im}\,Q}\Big|_{\mathrm{gh}=0}, $$

the cohomology of the BRST charge at ghost number zero. Equivalently, physical states satisfy $Q|\mathrm{phys}\rangle=0$, with two such states identified if they differ by $Q|\chi\rangle$. This is a sharper characterization than the Gupta–Bleuler subsidiary condition: it states that the unphysical content is exactly the image of a nilpotent operator, so that it decouples from every physical amplitude, and it makes the positivity of the physical norm a theorem rather than a prescription, provided the cohomology is positive.

The two transverse polarizations of the massless vector field are the ghost-number-zero cohomology of the abelian BRST charge: the timelike and longitudinal components of $\tilde{A}$, together with the ghost and antighost, form the quartet of the Kugo–Ojima mechanism, a set of four states in two zero-norm pairs whose members are removed by the cohomology and which never appear in the physical subspace. In the biquaternion language the quartet is the non-transverse part of the material sector $\mathbb{M}_-$ together with the odd envelope directions of the ghost pair: the algebra supplies the material quartet, and the envelope supplies the odd pair. The physical subspace is two-dimensional at each momentum, in agreement with the count reached by the Gupta–Bleuler route and by the constraint analysis.

The Massive Case as a Control

The Proca field of the preceding article has no gauge freedom, and the BRST construction has nothing to remove. The Legendre transform is singular but the constraints are second class rather than first class, the gauge transformations are absent, and no ghost fields appear; the phase-space count gives three physical polarizations directly. This is the control case that isolates what BRST does: it is a device for theories with a gauge redundancy, and its entire output — the ghost Lagrangian, the nilpotency, the cohomology — is empty when the redundancy is absent. The two descriptions agree at the level of physical states where both apply: the massive theory with the limit $\mu\to0$ and the massless theory with the gauge fixing both give the two transverse polarizations, and the longitudinal state that the Proca theory keeps is the state that the gauge symmetry of the massless theory removes.

What the Algebra Supplies and What It Imports

Supplied by the algebra, and recomputed here. The realization of the gauge algebra by the commutator inside $\mathbb{B}$, with the generators in the informational sector $\mathbb{M}_+$ and the gauge field in the material sector $\mathbb{M}_-$; the Jacobi identity as a consequence of associativity, verified through the $\mathrm{SU}(2)$ realization; the resulting nilpotency $s^2=0$ of the non-abelian BRST transformation, verified on the Grassmann envelope with random gauge fields; the graded structure in which the gauge-fixing Lagrangian is BRST-exact; and the sector assignment of the fields, which makes the gauge structure a coupling between the two fixed-point sectors.

Imported, and left visible. The gauge-fixing Lagrangian and the Nakanishi–Lautrup field; the Faddeev–Popov ghost fields and the determinant that produces them (the companion subcategory on gauge fields); the Grassmann envelope $\Lambda$, which the algebra does not contain; the Noether construction of the BRST charge; the Kugo–Ojima quartet mechanism; the cohomological characterization of physical states and the positivity argument; and the Slavnov–Taylor identities, which are the Ward identities of the BRST symmetry and which are treated in the standard literature.

Not supplied. The choice of gauge group; the value of the gauge parameter $\xi$; the odd coordinates themselves; and any empirical content. The framework makes the nilpotency inherited rather than postulated, and that is the whole of its contribution to BRST symmetry.

Open Questions

A biquaternionic Grassmann envelope. The odd coordinates are adjoined externally. Is there a natural $\mathbb{Z}_2$-graded extension of $\mathbb{B}$ — a superalgebra of which $\mathbb{B}$ is the even part — in which the BRST doublet is an element rather than an adjunction, and does the associativity argument survive?
The BRST charge as an algebraic object. The charge $Q$ is an operator on the Fock space and not an element of $\mathbb{B}$. Is there a finite-dimensional object — a derivation, a supertrace, a cocycle — built from $\mathbb{B}$ that computes the cohomology without passing through the mode algebra, in the way the adjoint action computes the spin without passing through a tensor?
Which gauge algebras are realized. The Jacobi identity holds for any gauge algebra realized by a commutator inside $\mathbb{B}$, but the algebras so realized are constrained by the finite dimension of the algebra. Is the constraint physical — does it select a subclass of gauge groups — or is it an artifact of using a single copy of the algebra?
The gauge-fixing fermion and the sector split. The fermion $\Psi$ is built from the antighost, which is odd, and the gauge-fixing function, which is even and material. Does the sector split $\mathbb{M}_-\oplus\mathbb{M}_+$ organize the BRST complex into a double complex, with the two differentials corresponding to the two sectors?
The massive limit and the disappearance of the ghosts. The ghosts disappear when the mass appears and the constraints become second class. Is there a deformation of the BRST complex in which the mass is a deformation parameter and the cohomology jumps from two to three dimensions at $\mu=0$, and is the jump visible algebraically?
Empirical content. As everywhere, whether any of this yields a prediction distinguishing the framework from standard BRST quantization. The transcription given here does not.

Summary

The BRST symmetry of a gauge theory promotes the gauge parameter to an anticommuting ghost field and introduces a nilpotent odd derivation $s$ under which the gauge field, the ghost, the antighost and the Nakanishi–Lautrup field transform. The gauge-fixed Lagrangian is BRST-exact, $\mathcal{L}=\mathcal{L}_{\mathrm{gauge}}+s\Psi$ with $\Psi=\bar{c}(\partial_\mu A^\mu+\tfrac{\xi}{2}B)$, and the physical states are the ghost-number-zero cohomology of the BRST charge, $\mathcal{H}_{\mathrm{phys}}=\ker Q/\mathrm{im}\,Q$. The unphysical content — the timelike and longitudinal components of the gauge field together with the ghost pair — forms the Kugo–Ojima quartet of zero-norm states.

In the biquaternion framework two things are added to the standard transcription. The first is the natural realization of the gauge algebra inside the algebra: the generators lie in the informational sector $\mathbb{M}_+$, the gauge field in the material sector $\mathbb{M}_-$, and the structure constants are those of the commutator. The second, and the substantive one, is the origin of nilpotency. The Jacobi identity that makes the non-abelian BRST transformation nilpotent is a consequence of the associativity of quaternion multiplication, so that $s^2=0$ is inherited from associativity and not postulated. The nilpotency was verified explicitly on the Grassmann envelope with $\mathrm{SU}(2)$ structure constants and random gauge fields, and the Jacobi combination was verified to vanish.

What the framework does not supply is the Grassmann envelope itself. The algebra is four-dimensional over $\mathbb{C}$ and ungraded: it has no odd elements, so the ghosts and the odd part of the BRST charge must be adjoined as $\mathbb{B}\otimes\Lambda$. The gauge-fixing fermion, the gauge-fixing function, the choice of gauge group and of the gauge parameter, the Faddeev–Popov procedure that produces the ghost Lagrangian, and the positivity argument for the cohomology are all imported. The massive Proca theory is the control case in which the gauge redundancy and hence the entire BRST apparatus are absent, and the physical content is three positive-norm polarizations.

Summary of Notation

Symbol	Meaning
$\mathbb{B}=\mathbb{C}\otimes_\mathbb{R}\mathbb{H}$	Biquaternion algebra, $\cong M_2(\mathbb{C})$; has nilpotents, but is ungraded and has no odd elements
$\mathbb{M}_+,\mathbb{M}_-$	Informational and material sectors; gauge field in $\mathbb{M}_-$, generators in $\mathbb{M}_+$
$\tilde{A}=iA_0e_0+\mathbf{A}\in\mathbb{M}_-$	Gauge potential biquaternion
$T^a=ie_a$, $[T^a,T^b]=2i\varepsilon^{abc}T^c$	$\mathrm{SU}(2)$ generators realized by the commutator
$f^{abc}$	Structure constants; Jacobi identity required for nilpotency
$g$, $D_\mu=\partial_\mu+gA_\mu$	Coupling and covariant derivative
$c,\bar{c}$	Faddeev–Popov ghost and antighost, Grassmann-odd
$B$	Nakanishi–Lautrup auxiliary field
$\xi$	Gauge parameter
$\mathbb{B}\otimes\Lambda$, $\Lambda=\Lambda^0\oplus\Lambda^1$	Grassmann envelope; adjoined odd coordinates
$s$	BRST operator: Grassmann-odd derivation, $\mathrm{gh}(s\tilde{Q})=\mathrm{gh}(\tilde{Q})+1$
$sA_\mu^a=(D_\mu c)^a$, $sc^a=-\tfrac{g}{2}f^{abc}c^bc^c$, $s\bar{c}^a=B^a$, $sB^a=0$	Non-abelian BRST transformations
$s^2=0$	Nilpotency; from Jacobi, hence from associativity of $\mathbb{B}$
$\Psi=\bar{c}^a(\partial_\mu A^{a\mu}+\tfrac{\xi}{2}B^a)$	Gauge-fixing fermion, $\mathcal{L}=\mathcal{L}_{\mathrm{gauge}}+s\Psi$
$\mathrm{gh}(\tilde{A})=0$, $\mathrm{gh}(c)=+1$, $\mathrm{gh}(\bar{c})=-1$, $\mathrm{gh}(B)=0$	Ghost number
$Q$, $Q^2=0$, $s\tilde{Q}=[iQ,\tilde{Q}]_\pm$	BRST charge
$\mathcal{H}_{\mathrm{phys}}=\ker Q/\mathrm{im}\,Q\big\|_{\mathrm{gh}=0}$	Physical states as BRST cohomology
Kugo–Ojima quartet	$(A_0,A_3,c,\bar{c})$ zero-norm pairs removed by the cohomology