Bogoliubov Transformations in Biquaternionic Form
Introduction
A Bogoliubov transformation is a linear canonical transformation of creation and annihilation operators that mixes the two: it replaces the ladder $(a_{\mathrm{fop}},a_{\mathrm{fop}}^{\dagger})$ by $$ b_{\mathrm{fop}} = u\,a_{\mathrm{fop}} + v\,a_{\mathrm{fop}}^{\dagger}, \qquad b_{\mathrm{fop}}^{\dagger} = v^*a_{\mathrm{fop}} + u^*a_{\mathrm{fop}}^{\dagger} , $$ and asks when the transformed operators again obey the canonical (anti)commutation relations. The transformation is the algebraic heart of superfluidity, superconductivity, cosmological particle production, and — of direct relevance to this corpus — the Unruh and Hawking effects, where the thermal character of the radiation is the statement that the transformation relating the inertial and accelerated or black-hole mode expansions has $v\ne0$. This article asks what the biquaternion algebra $\mathbb{B}$ contributes to that structure.
The answer has a clean shape, and it is mostly a restriction.
- Established, and recomputed below. For a single fermionic mode the biquaternion algebra contains the ladder and the transformation, and the canonical conditions can be imposed exactly. Writing $\vec b = \alpha\vec\Upsilon_x + \beta\vec\Upsilon_y$ with $\alpha,\beta\in\mathbb{C}$, the two canonical anticommutators are $$ \big\{\vec b,\vec b^{*}\big\} = \big(|\alpha|^2+|\beta|^2\big)e_t , \qquad \big\{\vec b,\vec b\big\} = 2\alpha\beta\,e_t . $$ The first is satisfied by every normalized pair, $|\alpha|^2+|\beta|^2=1$; the second forces $\alpha\beta=0$. So the one-mode canonical condition is $$ |\alpha|^2+|\beta|^2 = 1 \qquad\text{and}\qquad \alpha\beta = 0 , $$ whose solutions are: a phase, $\vec b = \exp(i\vartheta_{\mathrm{par}})\vec\Upsilon_x$; or the particle–hole exchange, $\vec b = \exp(i\vartheta_{\mathrm{par}})\vec\Upsilon_y$.
- Established (algebra). The single-mode Bogoliubov group of $\mathbb{B}$ is therefore the orthogonal group $$ O(2) = \mathrm{U}(1)\rtimes\mathbb{Z}_2 , $$ with $\mathrm{U}(1)$ the phases — inner automorphisms generated by the number operator — and the reflection the particle–hole exchange. The Fock-structure-preserving subgroup is the $\mathrm{U}(1)$ of phases alone, and the squeezing coset $O(2)/\mathrm{U}(1)\cong\mathbb{Z}_2$ is discrete: it has two points (the identity and the particle–hole exchange) and zero dimension. There is therefore no continuous one-mode squeezing — no element of the one-mode algebra rotates a mode toward its conjugate through a continuous family while preserving both canonical anticommutators. A transformation with $\alpha,\beta\ne0$ preserves $\{\vec b,\vec b^{*}\}=e_t$ but gives $\{\vec b,\vec b\}=2\alpha\beta e_t\ne0$; it is not a Bogoliubov transformation.
- Gap. A genuine fermionic Bogoliubov transformation with continuously variable $v\ne0$ needs at least two modes. For $n$ modes the Bogoliubov group is the orthogonal group $O(2n)$ of the Majorana operators, the Fock-preserving subgroup is $\mathrm{U}(n)$, and the squeezing directions are the coset $O(2n)/\mathrm{U}(n)$, of dimension $n(n-1)$. For $n=1$ that dimension is zero and the coset is discrete; for $n\ge2$ it is positive. The biquaternion algebra holds exactly one mode, so the continuous squeezing transformations live in the module, not in the algebra. This is the same module gap that the Fock, S-matrix, and Wick articles record, here with a sharp group-theoretic face.
The article proceeds as follows. The next section states the standard Bogoliubov transformation and its canonical conditions. A section constructs the one-mode transformation inside $\mathbb{B}$ and computes the two anticommutators. A section solves the one-mode condition and identifies the group. A section describes what the surviving transformations do to the vacuum. A section treats the many-mode case and the coset, and a section connects the thermal transformation to the Unruh and Hawking articles. Sections on quadratic Hamiltonians and on what is established close the article.
The Standard Bogoliubov Transformation
Let $\{a_{\mathrm{fop},i},a_{\mathrm{fop},j}^{\dagger}\}=\delta_{ij}e_t$, $\{a_{\mathrm{fop},i},a_{\mathrm{fop},j}\}=0$ be $n$ fermionic modes. A Bogoliubov transformation is the replacement $$ b_{\mathrm{fop},i} = \sum_j\Big(u_{ij}a_{\mathrm{fop},j} + v_{ij}a_{\mathrm{fop},j}^{\dagger}\Big), \qquad b_{\mathrm{fop},i}^{\dagger} = \sum_j\Big(u_{ij}^*a_{\mathrm{fop},j}^{\dagger} + v_{ij}^*a_{\mathrm{fop},j}\Big). $$ The transformed operators obey the same CAR if and only if the matrices $u,v$ satisfy the two canonical conditions $$ u u^\dagger + v v^\dagger = I , \qquad u v^T + v u^T = 0 . $$ The first is the statement $\{b_{\mathrm{fop},i},b_{\mathrm{fop},j}^{\dagger}\}=\delta_{ij}e_t$; the second is the statement $\{b_{\mathrm{fop},i},b_{\mathrm{fop},j}\}=0$. Equivalently, writing the $n$ complex modes as $2n$ Hermitian Majorana operators $$ m_{\mathrm{fop},2i-1} = a_{\mathrm{fop},i} + a_{\mathrm{fop},i}^{\dagger}, \qquad m_{\mathrm{fop},2i} = -i\big(a_{\mathrm{fop},i} - a_{\mathrm{fop},i}^{\dagger}\big) , \qquad \{m_{\mathrm{fop},\eta},m_{\mathrm{fop},l}\} = 2\delta_{\eta l}e_t , $$ the Bogoliubov transformation is an orthogonal transformation $m_{\mathrm{fop},\eta}\mapsto\sum_l O_{\eta l}m_{\mathrm{fop},l}$ with $O\in O(2n)$, and the canonical conditions above are the statement that $O$ preserves the Majorana metric. This is the form in which the group structure is transparent:
| object | group | meaning |
|---|---|---|
| Bogoliubov transformations | $O(2n)$ | all real linear canonical maps of the Majoranas |
| Fock-structure-preserving | $\mathrm{U}(n)\subset O(2n)$ | maps that keep the splitting into creation and annihilation |
| squeezing | $O(2n)/\mathrm{U}(n)$ | maps with $v\ne0$; $\dim = n(n-1)$ |
The Fock-structure-preserving subgroup is the set of $O$ that commute with the complex structure $J$ defining $a_{\mathrm{fop},i}$; in the parametrization it is the set $v=0$, $uu^\dagger=I$, i.e. $u\in \mathrm{U}(n)$. The squeezing directions are the complementary coset: they mix $a_{\mathrm{fop},i}$ with $a_{\mathrm{fop},j}^{\dagger}$ and change the vacuum. A Bogoliubov transformation with $v\ne0$ therefore maps the original vacuum to a squeezed state. For a finite number of modes the map is implemented by a $*$-unitary on the Fock space; in the field case the transformation can fail to be unitarily implementable, and that failure is where the thermal character of the Unruh and Hawking states enters.
For bosons the same structure holds with the symplectic group in place of the orthogonal group, $Sp(2n,\mathbb{R})$ in place of $O(2n)$, and $\mathrm{U}(n)$ again as the Fock-preserving subgroup; the squeezing coset $Sp(2n,\mathbb{R})/\mathrm{U}(n)$ has dimension $n(n+1)$. The distinction between the orthogonal and symplectic cases is the distinction between the CAR and the CCR, and it is standard. The biquaternion algebra supplies no bosonic ladder, so the bosonic case is transcribed rather than realized; this is the same gap the Fock article records.
The One-Mode Transformation Inside the Algebra
Now specialize to $n=1$ and to the one-mode algebra of $\mathbb{B}$. The transformed ladder is no longer an abstract Fock operator but an element of the algebra; being a combination of the two pure nilpotents, it is pure and carries the arrow, $\vec b$, with no tilde. The most general complex-linear transformation of the pair $(\vec\Upsilon_x,\vec\Upsilon_y)$ is $$ \vec b = \alpha\,\vec\Upsilon_x + \beta\,\vec\Upsilon_y , \qquad \alpha,\beta\in\mathbb{C}, $$ whose Hermitian conjugate is $$ \vec b^{*} = \beta^*\vec\Upsilon_x + \alpha^*\vec\Upsilon_y . $$ This is the $n=1$ instance of the general parametrization, and both $\vec b$ and $\vec b^{*}$ are elements of $\mathbb{B}$ for every $\alpha,\beta$.
The Majorana pair of the mode. The two Hermitian combinations $m_{\mathrm{fop},1}=\vec\Upsilon_x+\vec\Upsilon_y=i\vec e_x$ and $m_{\mathrm{fop},2}=-i(\vec\Upsilon_x-\vec\Upsilon_y)=i\vec e_y$ obey $\{m_{\mathrm{fop},\eta},m_{\mathrm{fop},l}\}=2\delta_{\eta l}e_t$ for $k,l=1,2$, so they are the one-mode instance of the Majorana operators of the previous section. Both are left multiplications by an invertible element, $L_a\tilde{Ξ}=a\tilde{Ξ}$ with $a$ of norm one, hence the basis operators $m_{\mathrm{fop},1}=iL_x$ and $m_{\mathrm{fop},2}=iL_y$; for $n\ge2$ the Majorana operators are operators of the mode algebra, which the algebra $\mathbb{B}$ does not contain.
The first anticommutator. Using $\{\vec\Upsilon_x,\vec\Upsilon_y\}=e_t$ and $\{\vec\Upsilon_x,\vec\Upsilon_x\}=\{\vec\Upsilon_y,\vec\Upsilon_y\}=0$, $$ \big\{\vec b,\vec b^{*}\big\} = |\alpha|^2\big\{\vec\Upsilon_x,\vec\Upsilon_y\big\} + |\beta|^2\big\{\vec\Upsilon_y,\vec\Upsilon_x\big\} = \big(|\alpha|^2+|\beta|^2\big)\,e_t . $$ The two cross terms vanish because they are proportional to $\vec\Upsilon_x^2$ and $(\vec\Upsilon_y)^2$. Nothing constrains $(\alpha,\beta)$ except normalization, so every normalized pair gives a transformation preserving the anticommutator between $\vec b$ and its conjugate.
The second anticommutator. The remaining canonical condition is where the restriction lies: $$ \boxed{\;\big\{\vec b,\vec b\big\} = 2\alpha\beta\,\vec\Upsilon_x\vec\Upsilon_y + 2\alpha\beta\,\vec\Upsilon_y\vec\Upsilon_x = 2\alpha\beta\,\big\{\vec\Upsilon_x,\vec\Upsilon_y\big\} = 2\alpha\beta\,e_t\;}. $$ The cross terms here do not vanish: they combine into the full anticommutator, which is the identity. So $$ \boxed{\;\big\{\vec b,\vec b\big\} = 0 \iff \alpha\beta = 0\;}. $$
The result is exactly the algebraic statement that the one-mode canonical condition is stricter than naive normalization. The cross term of the first computation is the diagonal part, and it is automatically fine; the cross term of the second is the off-diagonal part, and it is automatically $2\alpha\beta\,e_t$. It is the vanishing of $\{\vec b,\vec b\}$ — the statement that $\vec b$ is a fermionic mode operator, not merely an operator with a normalized conjugate — that forbids the squeezing.
Numerical check. With $\mathsf{M}_2$ the $2\times2$ complex representation, sample $\alpha=\cos\vartheta_{\mathrm{par}}$, $\beta=\exp(i\theta_{\mathrm{pha}})\sin\vartheta_{\mathrm{par}}$ on $20000$ random $(\vartheta_{\mathrm{par}},\theta_{\mathrm{pha}})$. The first anticommutator equals $e_t$ for all samples, with a maximum residual of $4.4\times10^{-16}$; the second equals $2\alpha\beta e_t$, a central element (off-diagonal part identically zero), with magnitude $|2\alpha\beta|=|\sin 2\vartheta_{\mathrm{par}}|\le1$, maximal at $\vartheta_{\mathrm{par}}=\pi/4$ where $|\alpha|=|\beta|=1/\sqrt2$ and $|2\alpha\beta|=1$. The exact cases $\beta=0$ and $\alpha=0$ give $\{\vec b,\vec b\}=0$ to machine precision. The second anticommutator is thus the sole obstruction, and its maximum violation is the maximal squeezing amplitude.
The One-Mode Canonical Group
Solving $|\alpha|^2+|\beta|^2=1$ and $\alpha\beta=0$ gives exactly two branches.
Branch 1, phases: $\beta=0$, $|\alpha|=1$. Then $$ \vec b = \exp(i\vartheta_{\mathrm{par}})\,\vec\Upsilon_x , \qquad \vec b^{*} = \exp(-i\vartheta_{\mathrm{par}})\,\vec\Upsilon_y , \qquad \vartheta_{\mathrm{par}}\in[0,2\pi) . $$ This is an inner automorphism of $\mathbb{B}$, but the implementing $*$-unitary is not central: it is the gauge $*$-unitary generated by the number operator, $$ \tilde Q_{\mathrm{par},\vartheta} = \exp(-i\vartheta_{\mathrm{par}}\tilde\Pi_-) = e_t + \big(\exp(-i\vartheta_{\mathrm{par}})-1\big)\tilde\Pi_-, \qquad \tilde Q_{\mathrm{par},\vartheta}\,\vec\Upsilon_x\,\tilde Q_{\vartheta_{\mathrm{par}}}^{-1} = \exp(i\vartheta_{\mathrm{par}})\vec\Upsilon_x, \qquad \tilde Q_{\mathrm{par},\vartheta}\,\vec\Upsilon_y\,\tilde Q_{\vartheta_{\mathrm{par}}}^{-1} = \exp(-i\vartheta_{\mathrm{par}})\vec\Upsilon_y , $$ where the closed form uses $\tilde\Pi_-^2=\tilde\Pi_-$. The element $\tilde Q_{\mathrm{par},\vartheta}$ is $*$-unitary and lies in $\mathbb{B}$, and it is non-central: it does not commute with $\vec\Upsilon_x$ or $\vec\Upsilon_y$. This is the $\mathrm{U}(1)$ gauge symmetry of the single mode, written as an algebra element.
Branch 2, particle–hole exchange: $\alpha=0$, $|\beta|=1$. Then $$ \vec b = \exp(i\vartheta_{\mathrm{par}})\,\vec\Upsilon_y , \qquad \vec b^{*} = \exp(-i\vartheta_{\mathrm{par}})\,\vec\Upsilon_x . $$ This exchanges the two ladder operators, which is the Hermitian conjugation of the pair up to a phase; it maps the vacuum idempotent to its complement, $$ \vec\Upsilon_y \leftrightarrow \vec\Upsilon_x, \qquad \tilde\Pi_+ = e_t-\tilde\Pi_- \;\longmapsto\; \tilde\Pi_-, $$ The particle–hole branch is the reflection, the $\mathbb{Z}_2$ of the group.
Together, $$ \boxed{\;\text{one-mode Bogoliubov group of }\mathbb{B}\;=\;O(2)\;} $$ with the $\mathrm{U}(1)$ the phases and the $\mathbb{Z}_2$ the particle–hole exchange. There are no other one-mode Bogoliubov transformations. In particular there is no continuous squeezing: the Fock-preserving subgroup is $\mathrm{U}(1)$, the coset $O(2)/\mathrm{U}(1)$ is the two-point set $\{$identity, particle–hole$\}$, and the dimension formula $n(n-1)$ vanishes at $n=1$.
The group-theoretic statement is worth restating physically. Fermionic squeezing is the pairing of two modes into a BCS-like condensate; it requires a pair, one mode at momentum $\vec k_\Phi$ and one at $-\vec k_\Phi$, and it is the $\vec k_\Phi\leftrightarrow-\vec k_\Phi$ mixing that produces the Bogoliubov coefficients. A single mode has no partner, so there is nothing to pair with, and the algebra cannot squeeze it. The absence of one-mode squeezing in $\mathbb{B}$ is therefore not a deficiency of the framework; it is the group theory of $O(2)$.
What the Surviving Transformations Do
Both branches act geometrically on the state space, and the action is the action of the group of invertibles of $\mathbb{B}$.
Phases are the gauge symmetry of the mode. A phase $\vec\Upsilon_x\mapsto \exp(i\vartheta_{\mathrm{par}})\vec\Upsilon_x$ is conjugation by $\tilde Q_{\mathrm{par},\vartheta}=\exp(-i\vartheta_{\mathrm{par}}\tilde\Pi_-)$, which commutes with the number operator and therefore fixes it and both idempotents: $$ \tilde Q_{\mathrm{par},\vartheta}\,\tilde\Pi_-\,\tilde Q_{\vartheta_{\mathrm{par}}}^{-1} = \tilde\Pi_-, \qquad \tilde Q_{\mathrm{par},\vartheta}\,\tilde\Pi(\pm \vec e_z)\,\tilde Q_{\vartheta_{\mathrm{par}}}^{-1} = \tilde\Pi(\pm \vec e_z), \qquad \tilde Q_{\mathrm{par},\vartheta}\,(i\vec e_z)\,\tilde Q_{\vartheta_{\mathrm{par}}}^{-1} = i\vec e_z . $$ The phase changes the ladder operators but leaves the occupation and both one-mode states invariant; it is the $\mathrm{U}(1)$ gauge symmetry of the mode, and its generator is the number operator. It is inner and non-central, so it is not the trivial automorphism — it acts nontrivially on the operators and trivially on the states, which is what a symmetry is. In the bilinear pairing the effect cancels: $2\,\mathrm{Sc}(\tilde Q_{\mathrm{par},\vartheta}\tilde{Q}\tilde Q_{\vartheta_{\mathrm{par}}}^{-1})=2\,\mathrm{Sc}(\tilde{Q})$ by cyclicity, so no gauge-invariant observable sees the phase.
The group of invertibles acts on the Bloch sphere. The genuinely non-central, non-gauge unitaries of $\mathbb{B}$ form the complement of the stabilizer in $\mathrm{U}(2)$, and their conjugation action on the Hermitian sector, $$ \tilde H_\Psi \;\longmapsto\; \tilde U\,\tilde H_\Psi\,\tilde U^{*} , \qquad \tilde Q\in \mathrm{U}(2), $$ is the rotation of the Bloch vector. The orbit of a minimal idempotent is the vacuum manifold $S^2$ of The Biquaternion Vacuum as a Minimal Idempotent, and the stabilizer of a point is $\mathrm{U}(1)\times \mathrm{U}(1)$ (the gauge phase about the axis, plus the rotation about the axis). The state space of the one-mode truncation is thus acted on by $\mathrm{U}(2)/\big(\mathrm{U}(1)\times \mathrm{U}(1)\big)\cong S^2$, in exact parallel with the standard Bloch-sphere geometry.
The distinction between the two paragraphs is the distinction between the gauge subgroup and the state space. The Bogoliubov transformations that preserve the mode structure are the gauge phases ($\mathrm{U}(1)$, inner, acting trivially on the states); the larger group $\mathrm{U}(2)$ acts on the vacua, and it is this larger group that carries the geometric content of the one-mode state space. Neither contains a continuous squeezing direction.
The Many-Mode Case and the Coset
For more than one mode the algebra $\mathbb{B}$ no longer contains the full operator algebra of the modes: the many-mode CAR algebra is generated by the modes of the one-particle space, and the Fock space is a module over $\mathbb{B}$, as Fock Space and Creation/Annihilation Operators in Biquaternionic Form records. The Bogoliubov structure of the many-mode case is standard field theory, and its biquaternion content is only the following.
Two modes. Take $a_{\mathrm{fop},1},a_{\mathrm{fop},2}$ and their conjugates. The Majorana operators $m_{\mathrm{fop},1},\dots,m_{\mathrm{fop},4}$ transform under $O(4)$, with the Fock-preserving subgroup $\mathrm{U}(2)$ and the squeezing coset $$ O(4)/\mathrm{U}(2) \;\cong\; S^2 , \qquad \dim = 2 . $$ The two squeezing directions are the two independent pairings of the modes, and they are precisely the two ways of forming a BCS pair. The single symmetric pair $(\alpha,\beta)$ with real coefficients reduces to the elementary case $\alpha^2+\beta^2=1$, which was checked directly: for $(\alpha,\beta)=(0.6,0.8)$ and $(0.3,\sqrt{1-0.09})$ the relation $\alpha^2+\beta^2=1$ holds exactly. Here, unlike the one-mode case, both $\alpha$ and $\beta$ can be nonzero, because the two modes supply the independent partners that the one mode lacks.
The general $n$. The dimension of the squeezing coset is $n(n-1)$ for fermions and $n(n+1)$ for bosons. Both grow with the number of modes, and at $n=1$ the fermionic coset is zero-dimensional while the bosonic one has dimension two — one complex squeezing parameter, the single-mode squeeze of quantum optics. The fermionic vanishing is the absence of one-mode squeezing found above; a single bosonic mode does admit a squeeze, since a bosonic mode is its own partner under $a_{\mathrm{fop}}\to a_{\mathrm{fop}}^{\dagger}$. The general statement that continuous squeezing needs a partner is therefore a statement about the anticommuting grading, where the partner must be a different mode.
The group-theoretic statement for the algebra. The one-mode Bogoliubov group is $O(2)$ itself, and the absence of continuous squeezing is the discreteness of the coset $O(2)/\mathrm{U}(1)$ — two points, the identity and the reflection. The algebra cannot supply a continuous squeezing direction because squeezing is the pairing of a mode with a different mode, and $\mathbb{B}$ has room for exactly one. This is a clean instance of the general principle that the algebra hosts one mode and the field lives in a module.
Thermal Bogoliubov and the Unruh and Hawking Connection
The reason Bogoliubov transformations matter to this corpus is their use in the thermal articles.
Thermal fermionic transformation. For a fermionic mode in a thermal state at inverse temperature $\beta$ the relevant Bogoliubov transformation has $$ |u|^2 = \frac{1}{1+\exp(-\beta\omega)} , \qquad |v|^2 = \frac{1}{\exp(\beta\omega)+1} , \qquad |u|^2+|v|^2 = 1 , $$ with $|v|^2$ the Fermi occupation of the mode. This is a two-mode (or many-mode) transformation — the partners are the modes on the two sides of the horizon, or the two members of a pair — and it is therefore in the nontrivial coset. It is not an element of the one-mode group $O(2)$; it is not an inner automorphism of $\mathbb{B}$; and, in the field limit, the two vacua it relates need not be unitarily equivalent. This is exactly what The Unruh Effect in Biquaternionic Form and Hawking Radiation in Biquaternionic Form record about the thermal state: the transformation of the mode expansion is a Bogoliubov transformation with $v\ne0$, and the thermality of the resulting occupation is the manifestation of the coset.
The algebra's role. The biquaternion content of the thermal transformation is the same as its content elsewhere: the mode operators are the one-mode truncation, the transformation's coefficients are central scalars once contracted, and the vacuum that is changed is a minimal idempotent of $\mathbb{B}_{\Psi}$. The transformation itself is a many-mode object and lies outside $\mathbb{B}$. In the finite-temperature machinery of The Matsubara Thermal Formalism in Biquaternionic Form the Bogoliubov transformation is what diagonalizes a quadratic Hamiltonian; in the horizon calculations it is what relates the inertial and accelerated vacua; in both cases it is standard field theory written in the framework's notation.
A structural remark. That the one-mode group has no squeezing means that no single biquaternion mode can be thermally excited by a Bogoliubov transformation. Thermality is intrinsically a two-mode (or many-mode) phenomenon in this framework: it requires a pair, and the pair lives in the module. This is a sharp, checkable statement, and it is consistent with the horizon articles' finding that the thermal state is defined by a global mode relation rather than by a local property of any one mode.
Quadratic Hamiltonians
The transformations above are the canonical transformations that diagonalize quadratic Hamiltonians, and this is why they appear throughout the thermal machinery.
A general quadratic fermionic Hamiltonian on one mode has the form $$ \tilde H_\Psi = A\,\vec\Upsilon_y\vec\Upsilon_x + \tfrac12\big(\mathcal{G}_{\mathrm{gap}}^*\vec\Upsilon_x\vec\Upsilon_x + \mathcal{G}_{\mathrm{gap}}\,\vec\Upsilon_y\vec\Upsilon_y\big) + B\,e_t , \qquad A\in\mathbb{R},\ \mathcal{G}_{\mathrm{gap}}\in\mathbb{C}, $$ with the pairing term $\mathcal{G}_{\mathrm{gap}}$ the BCS gap. The Hermitian single-mode quadratic terms are generated by $e_t$, $\tilde\Pi_-$, and the two pairing elements $\vec\Upsilon_x^2=(\vec\Upsilon_y)^2=0$; the last vanish identically, so on one mode the gap term is zero and the Hamiltonian is automatically diagonal: $$ \tilde H_\Psi = A\,\tilde\Pi_- + B\,e_t . $$ Hence a single biquaternion mode is always its own eigenmode; there is no one-mode Bogoliubov rotation to perform and no gap to open. The gap requires a second mode, in agreement with the coset analysis. On two modes the pairing term is nonzero and the diagonalization is the standard BCS transformation; it is diagonalized by the two-mode Bogoliubov transformation with coefficients $(u,v)$ satisfying $|u|^2+|v|^2=1$, which is exactly the many-mode coset element. The biquaternion algebra contains the diagonal one-mode case and nothing more.
This is a small but genuine structural statement. In the framework the one-mode number operator $\tilde\Pi_-=\tfrac12(e_t-i\vec e_z)$ is a projector, so its eigenvalues are $0$ and $1$ and the one-mode Hamiltonian $A\tilde\Pi_-+Be_t$ has the two levels $B$ and $A+B$. There is no room for a squeezed ground state. The ground state is $\tilde\Pi_+$ or $\tilde\Pi_-$ according to the sign of $A$, and it is a minimal idempotent in either case, chosen by the sign of the one-mode energy — the finite-dimensional instance of the statement that the vacuum is selected by the dynamics.
What Is Established and What Is Interpretation
Established (algebra). - The one-mode canonical conditions are $|\alpha|^2+|\beta|^2=1$ and $\alpha\beta=0$, computed directly from $\{\vec\Upsilon_x,\vec\Upsilon_y\}=e_t$ and $\vec\Upsilon_x^2=0$; equivalently $\{\vec b,\vec b^{*}\}=(|\alpha|^2+|\beta|^2)e_t$ and $\{\vec b,\vec b\}=2\alpha\beta e_t$. - The one-mode Bogoliubov group is $O(2)=\mathrm{U}(1)\rtimes\mathbb{Z}_2$: the inner gauge phases and the particle–hole exchange. There is no continuous one-mode squeezing; the coset $O(2)/\mathrm{U}(1)$ is the two-point set $\{$identity, particle–hole$\}$. - The one-mode quadratic Hamiltonian is automatically diagonal, $\tilde H_\Psi=A\tilde\Pi_-+Be_t$, because the pairing elements vanish.
Standard, and transcribed. - The general Bogoliubov transformation, the canonical conditions $uu^\dagger+vv^\dagger=I$, $uv^T+vu^T=0$, the Majorana form, the groups $O(2n)$, $\mathrm{U}(n)$ and the squeezing cosets, and their dimensions $n(n-1)$ and $n(n+1)$. - The thermal fermionic coefficients $|u|^2$, $|v|^2$ and their use in the Unruh and Hawking calculations.
Interpretation. - The interpretation of the thermal-state calculations as "the coset" and of the absence of one-mode squeezing as the group theory of $O(2)$ is the interpretive link between the algebra and the horizon articles; the group theory itself is standard.
Open. - Whether the vanishing of one-mode fermionic squeezing has observable consequences in the framework's own terms — beyond the absence of a one-mode gap — is not settled. - The bosonic Bogoliubov transformation has no algebra realization, since $\mathbb{B}$ contains no bosonic ladder; the symplectic case is transcribed.
Summary
For a single fermionic mode the biquaternion algebra contains the Bogoliubov transformation and decides it completely. Writing $\vec b=\alpha\vec\Upsilon_x+\beta\vec\Upsilon_y$, $$ \boxed{\;\big\{\vec b,\vec b^{*}\big\}=\big(|\alpha|^2+|\beta|^2\big)e_t, \qquad \big\{\vec b,\vec b\big\}=2\alpha\beta\,e_t\;}, $$ so the canonical conditions are $|\alpha|^2+|\beta|^2=1$ and $\alpha\beta=0$. The solutions are the phases $\vec b=\exp(i\vartheta_{\mathrm{par}})\vec\Upsilon_x$, an inner automorphism generated by the number operator, and the particle–hole exchange $\vec b=\exp(i\vartheta_{\mathrm{par}})\vec\Upsilon_y$, which swaps the vacuum idempotent with the number idempotent. The one-mode Bogoliubov group is $O(2)=\mathrm{U}(1)\rtimes\mathbb{Z}_2$, with the Fock-preserving subgroup the $\mathrm{U}(1)$ of phases and the discrete coset $O(2)/\mathrm{U}(1)$; there is no continuous squeezing, and the one-mode quadratic Hamiltonian is automatically diagonal, $\tilde H_\Psi=A\tilde\Pi_-+Be_t$, because the one-mode pairing elements vanish.
Fermionic squeezing requires at least two modes. For $n$ modes the Bogoliubov group is $O(2n)$ (fermions) or $Sp(2n,\mathbb{R})$ (bosons), the Fock-preserving subgroup is $\mathrm{U}(n)$, and the squeezing coset has dimension $n(n-1)$ or $n(n+1)$. The thermal transformations of the Unruh and Hawking calculations have $v\ne0$ and lie in that coset; they are many-mode objects, not inner automorphisms of $\mathbb{B}$, and, in the field limit, the two vacua they relate need not be unitarily equivalent. The biquaternion algebra holds one mode; the squeezing lives in the module.
Summary of Notation
| Symbol | Meaning |
|---|---|
| $\mathbb{B}=\mathbb{C}\otimes_\mathbb{R}\mathbb{H}$ | Biquaternion algebra, $\cong M_2(\mathbb{C})$ |
| $e_t=1,\vec e_x,\vec e_y,\vec e_z$ | Quaternion basis, $\vec e_\eta^2=-e_t$ |
| $\vec\Upsilon_x=\tfrac12(i\vec e_x-\vec e_y)$, $\vec\Upsilon_y=\tfrac12(i\vec e_x+\vec e_y)$ | Single-mode ladder |
| $a_{\mathrm{fop}}$, $b_{\mathrm{fop}}$ | Abstract ladder and its Bogoliubov transform, $b_{\mathrm{fop}}=u\,a_{\mathrm{fop}}+v\,a_{\mathrm{fop}}^{\dagger}$ |
| $\tilde\Pi_-=\vec\Upsilon_y\vec\Upsilon_x=\tfrac12(e_t-i\vec e_z)$ | Number operator |
| $\tilde\Pi(\pm \vec e_z)=\tfrac12(e_t\pm i\vec e_z)$ | Vacuum and occupied idempotents |
| $\vec b=\alpha\vec\Upsilon_x+\beta\vec\Upsilon_y$ | One-mode Bogoliubov transformation |
| $|\alpha|^2+|\beta|^2=1$, $\alpha\beta=0$ | One-mode canonical conditions |
| $\{\vec b,\vec b^{*}\}=(|\alpha|^2+|\beta|^2)e_t$ | First anticommutator (always normalized) |
| $\{\vec b,\vec b\}=2\alpha\beta e_t$ | Second anticommutator (forces $\alpha\beta=0$) |
| $O(2)=\mathrm{U}(1)\rtimes\mathbb{Z}_2$ | One-mode Bogoliubov group (phases and particle–hole) |
| $u,v$ | General Bogoliubov matrices, $uu^\dagger+vv^\dagger=I$, $uv^T+vu^T=0$ |
| $m_{\mathrm{fop},\eta}$ | Majorana operators, $\{m_{\mathrm{fop},\eta},m_{\mathrm{fop},l}\}=2\delta_{\eta l}e_t$; for one mode the basis operators $iL_x$, $iL_y$ |
| $O(2n)$, $\mathrm{U}(n)$, $O(2n)/\mathrm{U}(n)$ | Bogoliubov group, Fock subgroup, squeezing coset (discrete at $n=1$) |
| $n(n-1)$, $n(n+1)$ | Squeezing-coset dimensions (fermion, boson) |
| $|u|^2=(1+\exp(-\beta\omega))^{-1}$, $|v|^2=(\exp(\beta\omega)+1)^{-1}$ | Thermal fermionic coefficients |
| $\tilde H_\Psi=A\tilde\Pi_-+Be_t$ | One-mode quadratic Hamiltonian (automatically diagonal) |
Further Reading
- N. N. Bogoliubov, "On the theory of superfluidity," Journal of Physics (USSR) 11 (1947) 23–32, and Lectures on Quantum Statistics, Vol. 1 (Gordon and Breach, 1967), for the original canonical transformation.
- J. G. Valatin, "Comments on the theory of superconductivity," Il Nuovo Cimento 7 (1958) 843–857, for the fermionic Bogoliubov transformation and the BCS ground state.
- P. G. de Gennes, Superconductivity of Metals and Alloys (Benjamin, 1966), for the BCS transformation and the gap equation.
- J.-P. Blaizot and G. Ripka, Quantum Theory of Finite Systems (MIT Press, 1986), for the canonical conditions, the Majorana representation, and the $O(2n)$/$\mathrm{U}(n)$ structure.
- A. L. Fetter and J. D. Walecka, Quantum Theory of Many-Particle Systems (McGraw-Hill, 1971), for the general Bogoliubov transformation and its use in pairing theories.
- J. D. Bjorken and S. D. Drell, Relativistic Quantum Fields (McGraw-Hill, 1965), for the field-theoretic Bogoliubov transformation in the presence of a background.
- S. W. Hawking, "Particle creation by black holes," Communications in Mathematical Physics 43 (1975) 199–220, for the thermal Bogoliubov coefficients $|u|^2$, $|v|^2$ of the horizon.
- W. G. Unruh, "Notes on black-hole evaporation," Physical Review D 14 (1976) 870–892, for the accelerated-observer Bogoliubov transformation and the thermal occupation.
- N. D. Birrell and P. C. W. Davies, Quantum Fields in Curved Space (Cambridge, 1982), for the Bogoliubov formalism in external backgrounds.
- T. D. Lee, Particle Physics and Introduction to Field Theory (Harwood, 1981), for the relation between Bogoliubov transformations and quadratic Hamiltonians.
- P. Lounesto, Clifford Algebras and Spinors (Cambridge, 2001), for the Majorana representation and the orthogonal and $*$-unitary groups acting on spinors.
- Companion articles: Fock Space and Creation/Annihilation Operators in Biquaternionic Form, for the one-mode ladder and the module structure; The Wick Theorem in Biquaternionic Form, for the one-mode real form and the anticommutators; The Unruh Effect in Biquaternionic Form and Hawking Radiation in Biquaternionic Form, for the thermal Bogoliubov transformations and the thermal state; The Biquaternion Vacuum as a Minimal Idempotent, for the vacuum idempotent and the Bloch sphere; The Matsubara Thermal Formalism in Biquaternionic Form, for the diagonalization of quadratic Hamiltonians at finite temperature.