The Schrödinger Equation in Biquaternionic Polar Variables
Introduction
This article writes the biquaternionic Schrödinger equation in the polar variables of the $N$-regular element. The general element's word has all four factors: the field's rapidity is a constant of the motion and the axis of the field's own boost precesses (§The Equation in Polar Variables), so the boost is not trivial in general. The central class is the separate case, in which the word collapses to $\varrho\,e^{i\theta_{\mathrm{pha}}}\,\varepsilon e_t$ and the equation separates into two exact real equations: a continuity equation for $\rho_{\mathrm{num}}=\varrho^{2}$ and a quantum Hamilton–Jacobi equation with quantum potential $Q=-\hbar^{2}\Delta_\Phi \varrho/(2m\varrho)$. That is Madelung's pair, exact and not an approximation, away from the nodal set. The general solution of the constant-element case is the propagator $\tilde Q(t')\tilde\Xi(0)$, and the complete polar decomposition is the lens on it; the central collapse is the Madelung reading of that lens.
The Schrödinger equation of this corpus is one element of $\mathbb{B}$ acting on another by left multiplication. The Schrödinger Equation in Biquaternionic Form writes it as
$$ i\hbar\,\frac{\partial}{\partial t'}\tilde\Xi=\tilde H_\Psi\,\tilde\Xi, \qquad \tilde H_\Psi=h_t\,e_t+i\vec h_\Psi\in\mathbb{B}_{\Psi}, $$
and states what the equation does not need: no column, no minimal left ideal, no projector, no idempotent, no declared axis, no matrix array in the body of the equation. The field is one element, the operator is one element, and the multiplication is the algebra's own.
Two settings, and they are not the same equation. In §The Equation in Polar Variables the Hamiltonian is a constant element, $\tilde H_\Psi=h_t\,e_t+i\vec h_\Psi$ with $h_t$ and $\vec h_\Psi$ fixed numbers, so the equation is an ordinary differential equation in $t'$ and its solution is the closed form $\tilde Q(t')=\exp(-i\theta_{\mathrm{evo}})$. In §The Central Field the Hamiltonian is the standard operator, $\tilde H_\Psi=\hat H\,e_t$ with $\hat H=-\hbar^{2}\Delta_\Phi/2m+V$, so the equation is the ordinary Schrödinger equation, a partial differential equation, and its solution is not a finite product of exponentials. The first setting is where the flow of the word is read, and it closes because a constant element supplies two numbers. The second is where Madelung's pair is read, and it closes because a central field supplies one scale and one phase. Nothing below transfers a result from one setting to the other.
The angle is $\theta_{\mathrm{evo}}=h_t t'/\hbar$, the scalar part of the Hamiltonian times the interval over $\hbar$. The exponent $-i\tilde H_\Psi t'/\hbar$ is anti-Hermitian, not the Hamiltonian, and that is what makes it unitary: a purely imaginary scalar part makes the phase a real angle, and a real vector part makes the rotor a unit rotor.
This article substitutes a parametrisation of that one element into that one equation and reads what comes out. The parametrisation is the complete polar decomposition (The Complete Polar Decomposition of an N-Regular Biquaternion), in the exponential form it takes in The Complete Polar Decomposition of an N-Regular Biquaternion:
$$ \tilde\Xi=\varrho\,e^{i\theta_{\mathrm{pha}}}\,\tilde U_{\Psi,\mathrm b}\,\tilde u_{\mathrm r}, \qquad \tilde U_{\Psi,\mathrm b}=\exp\Bigl(i\frac{\theta_{\mathrm{rpd}}}{2}\vec u_{\mathrm b}\Bigr), \qquad \tilde u_{\mathrm r}=\exp\Bigl(\frac{\theta_{\mathrm{axr}}}{2}\vec u_{\mathrm r}\Bigr). $$
The four factors are a positive real scale, a central angle, a Hermitian positive boost and a real rotor. The word has eight real parameters against the element's eight real components, and on its domain it is a bijection: it adds no object and drops no datum. The rotor's angle $\theta_{\mathrm{axr}}$ is the angle of the spatial rotation the rotor performs, so the factor $\tfrac12$ in its exponential is the half-angle of a quaternion and not a physical factor.
The domain is the whole of the qualification of this article. In the corpus's vocabulary an element is $N$-regular when $N\neq0$ and $N$-degenerate when $N=0$, and $N$ is the determinant (The Meaning of N-Regularity and Degeneracy). The word is a parametrisation of the group of invertibles, so
$$ \boxed{\;\text{the polar word of }\tilde\Xi\text{ exists}\iff\tilde\Xi\text{ is invertible in }\mathbb{B}\iff N(\tilde\Xi)\neq0\iff\tilde\Xi\text{ is }N\text{-regular},\;} $$
and it fails on every $N$-degenerate element. The three conditions are one condition, stated three ways: in this algebra an element is invertible exactly when its determinant does not vanish, and the word is a parametrisation of the group of invertibles. There is no third case and no extension of the word to $N=0$.
The condition is the whole of the qualification, and it is weak. The cone is a set of measure zero and it is not the carrier: a generic element of $\mathbb{B}$ is invertible, and the flow preserves the cone, so a field is invertible at every event or at none (§The Domain). Nothing has to be excluded in advance and no carrier has to be declared. The polar word of this equation is therefore a parametrisation of the field itself, with no ideal, no projector and no axis behind it.
Conventions. The algebra is $\mathbb{B}=\mathbb{C}\otimes_{\mathbb{R}}\mathbb{H}$, with $e_t=1$ and $\vec e_\mu^{2}=-e_t$; $i$ is the central scalar imaginary; $\mathbb{B}_{C}=\operatorname{span}_{\mathbb{R}}\{e_t,ie_t\}$ is the centre; $\mathbb{B}_{\Phi}$ and $\mathbb{B}_{\Psi}$ are the anti-Hermitian and Hermitian fixed subspaces; $\natural$ is quaternion conjugation and $*$ the Hermitian one; $\mathrm{Sc}$ is the coefficient of $e_t$. Time is written $t'$ and the parent's material length is $r'_t=c\,t'$, so $\partial/\partial r'_t=c^{-1}\partial/\partial t'$ and the parent's $i\hbar c\,\partial/\partial r'_t$ is $i\hbar\,\partial/\partial t'$; the two forms are used interchangeably and $\hbar$ is written explicitly in both.
No wave number is used in this article, and that is deliberate. A letter $k$ earns its place only where a mass, a momentum or an energy is absorbed by it, as $k_m=mc/\hbar$ absorbs the mass in the Dirac and Klein–Gordon equations. Here nothing is absorbed: the constant that multiplies the time derivative is $\hbar$, and the mass enters only through the operator $\tilde H_\Psi$ and, in the central case, through the kinetic term $|\vec\nabla_\Phi S|^{2}/2m$. Writing $k_m$ would require the composite $mc^{2}/k_m$, in which $m$ cancels and only $\hbar c$ survives, so the letter would hide the constant instead of naming a physical wave number. Every $\hbar$ below is therefore the reduced Planck constant and nothing else.
Two points of discipline come first, because both are ways of losing the alignment.
The word is not a split of the field. It is a re-coordinatisation of the element. Every factor is read from the element and the product is the element. Nothing is put into a slot.
The word brings nothing in behind it. No module, no ideal, no projector, no column, no separate carrier stands behind the four factors. Whatever the word is used for, the object it parametrises is still one element of $\mathbb{B}$, and the equation it is put into is still one element acting on another.
What the article establishes is short, and each item is one sentence:
- the propagator's word is a central phase times a rotor, with its scale and its boost frozen at $1$ and $e_t$;
- the two rates of the word are the mean of the two eigenvalues of the Hamiltonian and half their gap;
- the field's own scale and rapidity are constants of the motion, and nothing else is: the flow moves six of the eight parameters, and the two it leaves are the two that unitarity fixes;
- every direction the field carries, the axis of its own boost included, precesses about the Hamiltonian's own axis at twice the half-gap;
- the density is the Hermitian norm and reads the scale and the boost, while the quaternionic norm reads the scale and the phase;
- the word of a central field is a scale, a phase and a sign, and Madelung's pair is exactly that case, but the class is invariant only under a central Hamiltonian, so the pair is the evolution of that class and not of a general field.
The Field and Its Word
One element, eight real numbers
The field is one element $\tilde\Xi$ of $\mathbb{B}$, with four complex coefficients and therefore eight real components. The subscript sp is the corpus's caret for the spinor role, beside scl for the scalar role and fop for the field operator, as the caret table of Algebra and Element Conventions fixes it; once the field is named, the body writes $\tilde\Xi$ for the same element. Nothing else is needed: $\mathbb{B}$ is an algebra, so a product of two elements is an element, and the equation restricts no subspace.
The equation is complex-linear, and its symbol is Hermitian. Because $i$ is central, left multiplication $L_{\tilde H_\Psi}$ satisfies $L_{\tilde H_\Psi}(i\tilde X)=iL_{\tilde H_\Psi}\tilde X$, so the equation is complex-linear on $\mathbb{B}$ read as a complex vector space of dimension four. Because $\tilde H_\Psi^{*}=\tilde H_\Psi$, it also satisfies
$$ \mathrm{Sc}\bigl((\tilde H_\Psi\tilde X)\tilde Y^{*}\bigr)=\mathrm{Sc}\bigl(\tilde X(\tilde H_\Psi\tilde Y)^{*}\bigr), $$
so the symbol is Hermitian for the sesquilinear form $\langle\tilde X,\tilde Y\rangle_{*}=\mathrm{Sc}(\tilde X\tilde Y^{*})$, and the flow preserves $\mathrm{Sc}(\tilde X\tilde X^{*})$.
The display, where a display is wanted, is the axis-free array. In the coefficient basis the symbol is $L_{\mathrm{axf}}(\tilde H_\Psi)$, a $4\times4$ complex Hermitian array with $2\,\mathrm{Sc}\,L_{\mathrm{axf}}(\tilde H_\Psi)=4h_t$ and $\det L_{\mathrm{axf}}(\tilde H_\Psi)=\langle\tilde H_\Psi\rangle_{\natural}^{2}$, and it names no spatial direction. This is the ruling of The AXF Matrices: Axis-Free, Integer Entries, §Why the Pauli and the Gamma Displays Are Refused: the equation is written with elements and read with an axis-free array, and no matrix algebra stands between the two.
The word
The word of the field is the display of the introduction. Its four factors carry, in order: the square root of the quaternionic norm's modulus, the half-angle of the norm's phase, the Hermitian boost whose rapidity and axis the field's own Hermitian part fixes, and the rotor that carries the field's orientation. On the units it is a bijection with the eight real components, so every statement below is a statement about the field and not about a class of fields.
The two norms read different factors, and neither reads all four. The quaternionic norm is
$$ \langle\tilde\Xi\rangle_{\natural}=\tilde\Xi\tilde\Xi^{\natural}=\varrho^{2}e^{2i\theta_{\mathrm{pha}}}, $$
a central element that reads the scale and the phase and nothing else; and the Hermitian norm is
$$ \rho_{\mathrm{num}}=2\,\mathrm{Sc}\bigl(\tilde\Xi^{\dagger}\tilde\Xi\bigr)=2\,\mathrm{Sc}\bigl(\tilde\Xi\tilde\Xi^{*}\bigr)=2\varrho^{2}\cosh\theta_{\mathrm{rpd}}, $$
a positive real number that reads the scale and the boost and nothing else. Neither norm sees the rotor, and the reason is structural: the rotor is real and unitary, so it cancels in both.
The domain
Proposition. The word of $\tilde\Xi$ exists exactly on the complement of the cone $\{\tilde X:\langle\tilde X\rangle_{\natural}=0\}$, and the flow carries that complement to itself.
Proof. The first half is the boxed statement of the introduction. For the second, the propagator is $\tilde Q=\exp(-i\theta_{\mathrm{evo}})$ and $\langle\tilde Q\rangle_{\natural}$ is central, so
$$ \langle\tilde Q\tilde\Xi\rangle_{\natural}=\langle\tilde Q\rangle_{\natural}\langle\tilde\Xi\rangle_{\natural}, $$
and $\langle\tilde Q\rangle_{\natural}$ is invertible in the centre at every $t'$. Hence $\langle\tilde\Xi\rangle_{\natural}$ vanishes at one time exactly when it vanishes at every time, and the word's domain is an invariant set.
What that means in practice. A field that is invertible at $t'=0$ is invertible at every $t'$, and the word exists for all time. A field on the cone at $t'=0$ never leaves it, and the word exists at no time. There is no crossing, and no field drifts into or out of the domain. The invariance is exact: nothing is integrated, and only the centrality of $\langle\tilde Q\rangle_{\natural}$ is used.
The cone is not the nodal set. A central field vanishes exactly when its single complex coefficient does, and that is a point of the cone but not the whole of it: the cone is a proper algebraic subset of the algebra and not a level set of a single coefficient, and a field can be off the cone while one of its coefficients vanishes. What fails on the cone is the parametrisation and never the equation: the equation is one element times one element, and it makes sense on zero divisors as well.
The Equation in Polar Variables
The four brackets of the logarithmic derivative
The word is put into the equation by one differentiation. Every factor of $\tilde\Xi=\varrho\,e^{i\theta_{\mathrm{pha}}}\tilde U_{\Psi,\mathrm b}\tilde u_{\mathrm r}$ is differentiated once, and the result is
$$ \partial_{t'}\tilde\Xi=\tilde\Xi\left[\tfrac12\partial_{t'}\ln \varrho^{2}+i\partial_{t'}\theta_{\mathrm{pha}} +\tilde u_{\mathrm r}^{-1}\bigl(\tilde U_{\Psi,\mathrm b}^{-1}\partial_{t'}\tilde U_{\Psi,\mathrm b}\bigr)\tilde u_{\mathrm r} +\tilde u_{\mathrm r}^{-1}\partial_{t'}\tilde u_{\mathrm r}\right], $$
so that the equation $i\hbar\,\partial_{t'}\tilde\Xi=\tilde H_\Psi\tilde\Xi$ becomes
$$ i\hbar\left[\tfrac12\partial_{t'}\ln \varrho^{2}+i\partial_{t'}\theta_{\mathrm{pha}} +\tilde u_{\mathrm r}^{-1}\bigl(\tilde U_{\Psi,\mathrm b}^{-1}\partial_{t'}\tilde U_{\Psi,\mathrm b}\bigr)\tilde u_{\mathrm r} +\tilde u_{\mathrm r}^{-1}\partial_{t'}\tilde u_{\mathrm r}\right]=\tilde\Xi^{-1}\tilde H_\Psi\tilde\Xi . $$
The four brackets are a real scale term, a central imaginary phase term, a Hermitian boost connection and a real-vector rotor connection; the right-hand side is the Hamiltonian conjugated by the field. The identity is exact and no term is dropped. Because the equation is first order in time, the four brackets are not four equations to be solved: they are the four parts of one element, and they fix the rates of the eight parameters once $\tilde\Xi^{-1}\tilde H_\Psi\tilde\Xi$ is known.
The reductions. On a central field the boost is $e_t$ and the rotor is $\varepsilon e_t$, so the third and fourth brackets vanish and the first two remain; that is the pair of §The Central Field. On a general field all four are present, and the flow they describe is the subject of this section.
The propagator is a phase times a rotor
The parent's solution, for a Hamiltonian that is a constant element, is
$$ \tilde Q(t')=\exp\Bigl(-i\frac{\tilde H_\Psi t'}{\hbar}\Bigr) =\exp\Bigl(-i\frac{h_t t'}{\hbar}\Bigr)\Bigl(\cos\theta\,e_t+\sin\theta\,\vec u_\Psi\Bigr), \qquad \theta=\frac{|\vec h_\Psi|\,t'}{\hbar}, $$
where $\vec u_\Psi=\vec h_\Psi/|\vec h_\Psi|$ when $\vec h_\Psi\neq0$, and the constant is $\hbar$ alone because the variable is the time $t'$ and not the material length $r'_t=c\,t'$. Its word is
$$ \boxed{\;\tilde Q(t')=e^{i\theta_{\mathrm{pha},Q}}\tilde u_Q,\qquad r_Q=1,\qquad \tilde U_{\Psi,\mathrm b,Q}=e_t,\qquad \theta_{\mathrm{pha},Q}=-\frac{h_t t'}{\hbar},\qquad \theta_{\mathrm{axr},Q}=\frac{2|\vec h_\Psi|t'}{\hbar},\;} $$
so the propagator's word is frozen in the scale and the rapidity, at $1$ and $0$, carries no data in the boost's axis, and takes the rotor's axis from the Hamiltonian: only the phase and the rotor's angle run. The scale is one and the boost is the identity for one reason: $\tilde Q\tilde Q^{*}=e_t$, because the two factors of the parent's closed form are a central phase and a real rotor and nothing else. The Hermitian part of the propagator is therefore trivial and its word has nothing to read from it. What remains is the central phase and the rotor, which is the parent's real unit quaternion with its half-angle made explicit.
The factor of two is the word's, not the parent's. The parent writes $\cos\theta\,e_t+\sin\theta\,\vec u_\Psi$ with $\theta=|\vec h_\Psi|t'/\hbar$. That quaternion is a rotation through $2\theta$ about $\vec u_\Psi$, so its rotor angle is $\theta_{\mathrm{axr}}=2\theta=2|\vec h_\Psi|t'/\hbar$. No factor has been inserted: the same element is read once as a half-angle and once as an angle.
The field's scale and rapidity are constants
Left multiplication by $\tilde Q$ acts on the field's word in the only way it can. The central phase passes through everything, and the propagator's rotor is real and unitary, so
$$ \tilde Q\,\tilde\Xi =\Bigl(e^{i\theta_{\mathrm{pha},Q}}\tilde u_Q\Bigr)\Bigl(\varrho\,e^{i\theta_{\mathrm{pha}}}\,\tilde U_{\Psi,\mathrm b}\,\tilde u_{\mathrm r}\Bigr) =\varrho\,e^{i(\theta_{\mathrm{pha}}+\theta_{\mathrm{pha},Q})}\,\bigl(\tilde u_Q\,\tilde U_{\Psi,\mathrm b}\,\tilde u_Q^{\natural}\bigr)\bigl(\tilde u_Q\,\tilde u_{\mathrm r}\bigr). $$
The rotor does not pass through the boost. A real rotor and a Hermitian boost commute only when their axes are parallel or one of them is zero, so $\tilde u_Q\tilde U_{\Psi,\mathrm b}\neq\tilde U_{\Psi,\mathrm b}\tilde u_Q$ in general, and the word of $\tilde Q\tilde\Xi$ is not the word of $\tilde\Xi$ with $\tilde u_Q$ moved to the front. The rotor is applied on both sides of the boost, and what that does is rotate the boost's axis and change nothing else. With $\tilde U_{\Psi,\mathrm b}=\cosh\frac{\theta_{\mathrm{rpd}}}{2}+i\sinh\frac{\theta_{\mathrm{rpd}}}{2}\vec u_{\mathrm b}$,
$$ \tilde u_Q\,\tilde U_{\Psi,\mathrm b}\,\tilde u_Q^{\natural} =\cosh\frac{\theta_{\mathrm{rpd}}}{2}+i\sinh\frac{\theta_{\mathrm{rpd}}}{2}\bigl(\tilde u_Q\,\vec u_{\mathrm b}\,\tilde u_Q^{\natural}\bigr), $$
and $\tilde u_Q\vec u_{\mathrm b}\tilde u_Q^{\natural}$ is $\vec u_{\mathrm b}$ turned through $2\theta$ about $\vec u_\Psi$. Three statements follow, and with them the count of what the flow can move.
The scale is a constant of the motion. $\varrho(t')=\varrho(0)$.
The rapidity is a constant of the motion, and the boost is not. $\theta_{\mathrm{rpd}}(t')=\theta_{\mathrm{rpd}}(0)$, because conjugation preserves the $e_t$ coefficient. The axis does not stay, because it is a vector and the rotor turns vectors:
$$ \frac{d\vec u_{\mathrm b}}{dt'}=\frac{2}{\hbar}\,\vec h_\Psi\times\vec u_{\mathrm b}, $$
The boost axis runs at the gap, the rate at which every carried direction runs, so the boost of the field's word is not a constant of the motion: its rapidity is an invariant and its axis is a frame datum.
The rotor is carried by the propagator's rotor. $\tilde u_{\mathrm r}(t')=\tilde u_Q(t')\tilde u_{\mathrm r}(0)$, a left multiplication and not a similarity. Its content is geometric: every real direction the field carries, $\hat s=\tilde u_{\mathrm r}\hat s_0\tilde u_{\mathrm r}^{\natural}$, is carried to $\tilde u_Q\hat s\tilde u_Q^{\natural}$, so the whole carried frame turns rigidly.
Two of the eight parameters are frozen, and six move. The scale and the rapidity stay where they are. The phase, the rotor's angle, the rotor's axis and the boost's axis all move: the rotor's angle because a product of two rotors is a third rotor whose angle is not that of either factor, the rotor's axis because that composite rotation has an axis of its own, and the boost's axis because the boost is conjugated. The two frozen parameters are $\varrho$ and $\theta_{\mathrm{rpd}}$, and they are the two that unitarity fixes: the flow preserves $\rho_{\mathrm{num}}=2\varrho^{2}\cosh\theta_{\mathrm{rpd}}$, and with $\varrho$ fixed that is the same statement as $\theta_{\mathrm{rpd}}$ fixed. The density is therefore preserved too, but it is not a third parameter; it is a function of the two. The six that move are the frame's.
The two rates
The eigenvalues of the symbol are the two numbers $h_t-|\vec h_\Psi|$ and $h_t+|\vec h_\Psi|$, each twice, and both statements below are readings of the word against them.
The phase runs at the mean. The phase of the evolved field is $\theta_{\mathrm{pha}}(0)+\theta_{\mathrm{pha},Q}(t')$, so
$$ \frac{d\theta_{\mathrm{pha}}}{dt'}=-\frac{h_t}{\hbar}, $$
The central phase of the field advances at minus the mean of the two eigenvalues.
Every carried direction runs at the gap. Differentiating $\hat s(t')=\tilde u_Q\hat s(0)\tilde u_Q^{\natural}$ gives
$$ \frac{d\hat s}{dt'}=\frac{2}{\hbar}\,\vec h_\Psi\times\hat s, $$
The angular velocity is $2|\vec h_\Psi|/\hbar$, which is the gap between the eigenvalues and not half of it, because a direction turns at twice the half-angle. In the word's own parameter the half-angle runs at half the gap: $d\theta_{\mathrm{axr}}/dt'=2|\vec h_\Psi|/\hbar$ for the rotor angle, that is $|\vec h_\Psi|/\hbar$ for its half. The boost's axis is one more direction carried in this way, and its precession obeys the same law (§The field's scale and rapidity are constants).
The two rates are therefore the mean and the half-gap, and there are no others. The Hamiltonian supplies two numbers, the mean and the gap, and the word has exactly two frequencies to spend them on: the central phase runs at the mean, and the rotor's angle at the gap with its half-angle at half the gap. What remains in the word adds no third frequency. The scale and the rapidity are frozen, and the two axes are not frozen but carried: the rotor's axis and the boost's axis precess at the gap, which is the rate the rotor already has.
The half turn
At the time $\theta=\pi$, that is $|\vec h_\Psi|t'=\pi\hbar$, the propagator's rotor is $-e_t$ and
$$ \tilde\Xi(t')=-\exp\Bigl(-i\frac{h_t t'}{\hbar}\Bigr)\tilde\Xi(0), $$
The field returns to minus itself up to the central phase: the sign is the rotor's and the phase is the centre's. This is the one place where the two rates separate by inspection, because the phase has advanced by $h_t\pi/|\vec h_\Psi|$ while the rotor has advanced by exactly a half turn.
What the word of the propagator does not say
It does not say that the flow is periodic. The flow is periodic only when $h_t/|\vec h_\Psi|$ is rational; otherwise the two rates are incommensurable and the word never repeats. The parent's closed form says the same thing less plainly.
It does not fix the field's rotor. The propagator's word is the same for every initial field. What the field's own rotor is, is read from the field, and the flow multiplies it on the left.
The Central Field
The degeneration of the word
A field is central when it is a complex multiple of $e_t$, $\tilde\Xi=\Xi_t\,e_t$. Its word degenerates:
$$ \varrho=|\Xi_t|,\qquad \varrho\,e^{i\theta_{\mathrm{pha}}}=\pm\Xi_t,\qquad \varepsilon=\frac{\Xi_t}{\varrho\,e^{i\theta_{\mathrm{pha}}}}=\pm1,\qquad \tilde U_{\Psi,\mathrm b}=e_t,\qquad \tilde u_{\mathrm r}=\varepsilon e_t,\qquad \theta_{\mathrm{pha}}=\arg(\varrho\,e^{i\theta_{\mathrm{pha}}}), $$
where $\varrho\,e^{i\theta_{\mathrm{pha}}}$ is the principal square root of $N=\Xi_t^{2}$ and $\varepsilon$ its branch. The sign of the rotor is the sign of the modulus: the half-turn branch of the angle is carried by $\varepsilon$ and not by a free choice. A central field's word is therefore a scale, a phase and a sign, and three of the four factors have collapsed. This is exactly Madelung's pair, with $A=\varrho$ and $S=\hbar\theta_{\mathrm{pha}}$.
The central class is not invariant under a general Hamiltonian. If $\vec h_\Psi\neq0$ then $\tilde H_\Psi\tilde\Xi$ has a vector part, so a central field is carried off the centre at once; the class is invariant exactly when the Hamiltonian is central.
No gauge defect is left over. The field has eight real components, the word has eight real parameters, and the equation is first order in time with one complex equation per coefficient, that is eight real equations. The word is therefore exactly determined by the field, with no free function: there is no gauge to fix and no sector to choose. The four parameters that Madelung's form does not show are the boost's three and the rotor's angle, and they are not dropped by a choice; they are zero because the field is central.
The exact pair
For a central field the equation is the ordinary one-component equation, and the word's two real functions satisfy a pair that is exact and not an approximation. Write $\tilde\Xi=\varrho\,e^{i\theta_{\mathrm{pha}}}\varepsilon e_t$ and $\tilde H_\Psi=V\,e_t$, with $S=\hbar\theta_{\mathrm{pha}}$ the action. Substituting into the equation and separating the real and the imaginary parts gives
$$ \boxed{\;\frac{\partial S}{\partial t'}+\frac{|\vec\nabla_\Phi S|^{2}}{2m}+V+Q=0,\qquad \frac{\partial \varrho^{2}}{\partial t'}+\vec\nabla_\Phi\cdot\Bigl(\varrho^{2}\frac{\vec\nabla_\Phi S}{m}\Bigr)=0,\;} $$
with the quantum potential
$$ Q=-\frac{\hbar^{2}}{2m}\frac{\Delta_\Phi \varrho}{\varrho}. $$
The first is a Hamilton–Jacobi equation with one extra term, and the second is a continuity equation for the density $\rho_{\mathrm{num}}=\varrho^{2}$. Together they are the standard Madelung fluid: the first is the guidance equation of the de Broglie–Bohm reading, with $Q$ the term that separates it from classical Hamilton–Jacobi, and the second is the conservation of the density along the guidance flow. Takabayasi read the same pair as a hydrodynamical formulation and Bohm as a trajectory theory; the polar word is the biquaternionic reading of it.
The operator in the quantum potential is the Laplacian and not the d'Alembertian: the equation is first order in time and second order in space, so the second derivative in $Q$ is the spatial one and the temporal derivative never enters it.
The pair is equivalent to the equation, and neither half is an approximation: the real part says that the phase's gradient is the momentum, the imaginary part says that the density is conserved.
The current and the transport equation
The density and the current are the two halves of the continuity equation:
$$ \rho_{\mathrm{num}}=\varrho^{2},\qquad \vec j=\varrho^{2}\frac{\vec\nabla_\Phi S}{m}=\rho_{\mathrm{num}}\,\vec v, \qquad \frac{\partial\rho_{\mathrm{num}}}{\partial t'}+\vec\nabla_\Phi\cdot\vec j=0. $$
The velocity is the phase gradient over the mass, and not the word's boost: on a central field the boost has collapsed to $e_t$ and carries nothing, so $\vec v=\vec\nabla_\Phi S/m$ is the whole of the velocity.
The flow is irrotational. The velocity is a gradient, so its curl vanishes wherever the action is twice differentiable, $$ \vec\nabla_\Phi\times\vec v=\frac{1}{m}\vec\nabla_\Phi\times\vec\nabla_\Phi S=0, $$ and the circulation around any closed loop that does not enclose a node is zero. That is the irrotationality of Madelung's fluid, and it is a property of the phase and not of the algebra. The exception is a nodal line: the action is multivalued there and the circulation need not vanish, which is the quantised circulation of The Nodal Set and Quantized Circulation. In the general field the current reads the scale and the boost, and the rotor contributes nothing, exactly as the two norms say.
The density is a probability density and the current is its flux. $\rho_{\mathrm{num}}$ is positive by construction, because it is $2\,\mathrm{Sc}(\tilde\Xi^{\dagger}\tilde\Xi)$ and $\tilde\Xi^{\dagger}\tilde\Xi$ is positive in the Hermitian form; the equation preserves it, because it is the norm of a $*$-unitary flow. The definiteness is a consequence of the order of the equation: here the density is a norm and not the temporal slot of a current, so it has no sign to lose, while a second-order equation's density is the temporal slot of its current and its sign is the sign of the energy.
Why there is no mass shell
A second-order equation carries a mass shell: its four-momentum obeys $\langle\tilde K_\Phi\rangle_{\natural}=-k_m^{2}$ and the word's boost is a boost exactly on that set. A first-order equation carries none, and the absence is not a defect of the analysis.
The reason is the order of the equation. The Schrödinger equation is first order in time, so the symbol has no second time derivative to constrain, and the dispersion relation is a definition of the frequency rather than a constraint on the field. The propagator's word carries no boost at all, and the field's word carries one whose rapidity the flow cannot change, so there is nothing for a shell condition to be a condition on.
What replaces it is the pair. The two real equations of the previous section are an unconstrained system: one Hamilton–Jacobi equation that defines the phase and one continuity equation that defines the density, with the quantum potential carrying the only second derivative. Nothing is required of the initial data beyond the two functions.
What the Polar Word Does and Does Not Add
The equation, no; the coordinates, yes. Nothing here predicts anything the parent does not. What the polar form adds is a set of coordinates in which the flow is a statement about two numbers, the phase and the rotor's angle, and their two rates.
The domain is settled, and it is not a carrier. The word exists exactly on the invertibles, the invertibles are generic in this algebra, and the flow preserves them. The excluded set is the cone, a real set of measure zero, and it is not the nodal set: on a central field the two coincide, on a general field the cone is far larger than the node, and only the node is where the density vanishes. The polar word of this equation is therefore total on the generic field.
The frozen parameters are the statement of unitarity. Of the propagator's four factors, the scale and the boost are frozen; the two that remain are the two that a unitary element can carry when the Hamiltonian is fixed. The field's word freezes the same two quantities, its scale and its rapidity, while its phase and its two axes turn. The count is two in both cases, and it needs no proof beyond the word's definition.
The two rates are the mean and the half-gap. The central phase runs at minus the mean of the eigenvalues and the rotor's half-angle at half the gap; a direction carried by the field runs at the gap. These are the whole of the time dependence — two rates and one carried frame — and the parent's closed form contains them in one line without separating them.
Madelung's pair is one case, and the case is characterised. The central field is the field whose word has collapsed to a scale and a phase, and it is invariant exactly under a central Hamiltonian. Madelung's amplitude and phase are therefore not an approximation and not a general chart: they are the two parameters of a class of fields, and the class is the one in which nothing turns.
What was checked. The statements above were recomputed numerically on random Hermitian $\tilde H_\Psi$ and random fields: the word, the two norms, the flow, the two rates, the half turn, the non-Hermitian generator and the central class, and three exact-solution checks of the pair. The largest residual among the exact identities was $6\times10^{-16}$, the conjugated boost $\tilde u_Q\tilde U_{\Psi,\mathrm b}\tilde u_Q^{\natural}$ included, and the two differential laws were checked against finite differences to $10^{-6}$.
Open Questions
1. The meaning of a fixed rapidity on a turning axis. The rapidity of the field's word is a constant of the motion and its axis precesses at the gap; in the central case the boost is the identity and there is no axis at all. What physical reading the pair has for a non-central field — a velocity, a strain, or nothing — is not settled here, and the precession belongs to the question rather than beside it: a rapidity that holds while its axis turns is not yet a velocity, because nothing in the equation selects the axis.
2. The non-Hermitian generator. If $h_t$ acquires an imaginary part the scale leaves $1$ and the word reads decay, $\varrho(t')=\exp(\gamma t')$; if $\vec h_\Psi$ acquires a real part the boost leaves the identity. Both were checked. Whether either is the right description of a decaying system is open, and it is the natural place for the word to earn its keep.
3. The relativistic limit. Whether this equation, with its frozen rapidity and its absent mass shell, is the non-relativistic limit of the biquaternion Dirac equation is not checked here. The companions' mass shells are the natural comparison.
4. The relation to the WKB series. The pair of this article is the first term of the series of The Non-Quantum Limit of the Free Schrödinger Equation. Whether the word's exact coordinates simplify the higher orders is not examined.
Summary
The Schrödinger equation is one element of $\mathbb{B}$ acting on another by left multiplication, and the polar word of the second element is substituted into it. The word exists exactly on the invertibles of the algebra, which is the complement of the zero-divisor cone; a generic field is invertible and the flow preserves the cone, so the word exists for all time or for none.
The propagator's word is a central phase times a rotor, with the scale frozen at $1$, the rapidity at $0$ and the boost's axis carrying no data: only the phase and the rotor's angle move. The field's own scale and rapidity are constants of the motion, and they are the only two: the phase and the rotor's angle run, the rotor's axis and the boost's axis precess at the gap, and the flow moves six of the eight parameters. The phase runs at minus the mean of the Hamiltonian's two eigenvalues and the rotor's half-angle at half their gap; a direction the field carries runs at the gap, $d\hat s/dt'=2\vec h_\Psi\times\hat s/\hbar$.
The two norms read different factors: the quaternionic norm is $\varrho^{2}e^{2i\theta_{\mathrm{pha}}}$ and reads the scale and the phase, and the Hermitian norm is $2\varrho^{2}\cosh\theta_{\mathrm{rpd}}$ and reads the scale and the boost. Neither reads the rotor.
A central field's word collapses to a scale, a phase and a sign, and that case is exactly Madelung's: the Hamilton–Jacobi equation with the quantum potential $Q=-\hbar^{2}\Delta_\Phi \varrho/(2m\varrho)$ and the continuity equation for $\varrho^{2}$. The pair is exact, the operator is the Laplacian, and there is no mass shell, because the equation is first order in time.
Summary of Notation
| Symbol | Meaning |
|---|---|
| $\mathbb{B}=\mathbb{C}\otimes_{\mathbb{R}}\mathbb{H}$ | Biquaternion algebra |
| $e_t=1,\ \vec e_x,\vec e_y,\vec e_z$ | Quaternion basis, $\vec e_\mu^{2}=-e_t$ |
| $i$ | Central scalar imaginary |
| $\tilde\Xi$ | The field: one element of $\mathbb{B}$ |
| $\tilde H_\Psi=h_t\,e_t+i\vec h_\Psi$ | Hermitian Hamiltonian |
| $h_t,\ \vec h_\Psi$ | Its scalar part and its vector part |
| $\vec u_\Psi=\vec h_\Psi/\lvert\vec h_\Psi\rvert$ | Its axis, when $\vec h_\Psi\neq0$ |
| $N(\tilde\Xi)$ | Quaternionic norm; the word exists iff $N\neq0$ |
| $\langle\tilde X\rangle_{\natural}=\tilde X\tilde X^{\natural}$ | Quaternionic norm |
| $\natural,\ *$ | Quaternion conjugation, Hermitian conjugation |
| $\mathrm{Sc}$ | Scalar part |
| $\tilde\Xi=\varrho\,e^{i\theta_{\mathrm{pha}}}\tilde U_{\Psi,\mathrm b}\tilde u_{\mathrm r}$ | The polar word of the field |
| $\varrho>0$ | The scale |
| $\theta_{\mathrm{pha}}$ | The central angle |
| $\theta_{\mathrm{rpd}},\ \vec u_{\mathrm b}$ | Rapidity and axis of the boost |
| $\theta_{\mathrm{axr}},\ \vec u_{\mathrm r}$ | Angle and axis of the rotor |
| $\Xi_t$ | The scalar coefficient of the field; on a central field the whole field |
| $\varrho\,e^{i\theta_{\mathrm{pha}}}$ | Modulus of the norm; the principal square root of $\langle\tilde\Xi\rangle_{\natural}$ |
| $\varepsilon=\Xi_t/(\varrho\,e^{i\theta_{\mathrm{pha}}})=\pm1$ | Branch of the modulus; on a central field the rotor is $\varepsilon e_t$ |
| $\tilde Q(t')=\exp(-i\theta_{\mathrm{evo}})=e^{i\theta_{\mathrm{pha},Q}}\tilde u_Q$ | The propagator and its word |
| $\theta=\lvert\vec h_\Psi\rvert t'/\hbar$ | The parent's angle; the rotor's half-angle |
| $\tilde u_Q=\cos\theta\,e_t+\sin\theta\,\vec u_\Psi$ | The propagator's rotor |
| $\rho_{\mathrm{num}}=2\,\mathrm{Sc}(\tilde\Xi^{\dagger}\tilde\Xi)$ | Density (Hermitian norm) |
| $Q=-\hbar^{2}\Delta_\Phi \varrho/(2m\varrho)$ | Quantum potential |
| $L_{\mathrm{axf}}(\tilde H_\Psi)$ | Axis-free $4\times4$ array of the symbol |
Further Reading
- The Schrödinger Equation in Biquaternionic Form, for the equation, its solution and the axis-free array.
- The Complete Polar Decomposition of an N-Regular Biquaternion, for the word and its factors.
- The Complete Polar Decomposition of an N-Regular Biquaternion, for the exponential form used here.
- The Meaning of N-Regularity and Degeneracy, for $N$, the invertibles and the cone.
- The Dirac Equation in Biquaternionic Polar Variables and The Klein–Gordon Equation in Biquaternionic Polar Variables, for the same word applied to the two relativistic equations.
- The Schrödinger Equation in the Non-Quantum Limit, for the pair read as the first term of a series.
- The AXF Matrices: Axis-Free, Integer Entries, for why the body of the equation carries elements and not arrays.
- Erwin Madelung, "Quantentheorie in hydrodynamischer Form," Zeitschrift für Physik 40 (1927) 322–326, for the continuity equation and the Hamilton–Jacobi equation of a scalar wave.
- Takehiko Takabayasi, "On the formulation of quantum mechanics associated with classical pictures," Progress of Theoretical Physics 8 (1952) 143–182, for the same pair read as a hydrodynamical formulation.
- David Bohm, "A suggested interpretation of the quantum theory in terms of hidden variables," Physical Review 85 (1952) 166–193, for the quantum potential and the guidance equation it defines.
- The Nodal Set and Quantized Circulation, for the exception at a nodal line, where the action is multivalued and the circulation is quantised.