Main Page: Difference between revisions

From formulasearchengine
Jump to navigation Jump to search
No edit summary
No edit summary
Line 1: Line 1:
{{Thermodynamics|cTopic=[[Thermodynamic equations|Equations]]}}
{{Merge from|jet group|date=August 2011}}
{{For|electromagnetic equations|Maxwell's equations}}
'''Maxwell's relations''' are a set of equations in [[thermodynamics]] which are derivable from the definitions of the [[thermodynamic potentials]]. These relations are named for the nineteenth-century physicist [[James Clerk Maxwell]].


==Equation==
In [[differential geometry]], the '''jet bundle''' is a certain construction which makes a new [[smooth manifold|smooth]] [[fiber bundle]] out of a given smooth fiber bundle. It makes it possible to write [[differential equation]]s on [[Fiber bundle#Sections|section]]s of a fiber bundle in an invariant form. Jets may also be seen as the coordinate free versions of [[Taylor expansions]].


The Maxwell relations are statements of equality among the second derivatives of the thermodynamic potentials. They follow directly from the fact that the order of differentiation of an [[analytic function]] of two variables is irrelevant. If Φ is a thermodynamic potential and ''x<sub>i</sub>'' and ''x<sub>j</sub>'' are two different [[Thermodynamic potential#Natural variables|natural variables]] for that potential, then the Maxwell relation for that potential and those variables is:
Historically, jet bundles are attributed to [[Ehresmann]], and were an advance on the method ([[prolongation (mathematics)|prolongation]]) of [[Élie Cartan]], of dealing ''geometrically'' with [[derivative|higher derivatives]], by imposing [[differential form]] conditions on newly-introduced formal variables. Jet bundles are sometimes called '''sprays''', although [[spray (mathematics)|sprays]] usually refer more specifically to the associated vector field induced on the corresponding bundle (''e.g.'', the [[geodesic spray]] on [[Finsler manifold]]s.)


{{Equation box 1
More recently, jet bundles have appeared as a concise way to describe phenomena associated with the derivatives of maps, particularly those associated with the [[calculus of variations]]. Consequently, the jet bundle is now recognized as the correct domain for a [[covariant classical field theory|geometrical covariant field theory]] and much work is done in [[general relativity|general relativistic]] formulations of fields using this approach.
|title = '''Maxwell's relations''' ''(general)''
|indent =:
|equation = <math>\frac{\partial }{\partial x_j}\left(\frac{\partial \Phi}{\partial x_i}\right)=
\frac{\partial }{\partial x_i}\left(\frac{\partial \Phi}{\partial x_j}\right)
</math>
|border colour = #50C878
|background colour = #ECFCF4}}


where the [[partial derivatives]] are taken with all other natural variables held constant. It is seen that for every thermodynamic potential there are ''n''(''n'' − 1)/2 possible Maxwell relations where ''n'' is the number of natural variables for that potential.
==Jets==
{{main|Jet (mathematics)}}


== The four most common Maxwell relations ==
Let <math>(\mathcal{E}, \pi, \mathcal{M})</math>  be a [[fiber bundle]] in a category of [[manifold]]s and let <math>p \in \mathcal{M}</math>, with <math>\dim\mathcal{M}=m</math>.
Let <math>\Gamma(\pi)\,</math> denote the set of all local sections whose domain contains <math>p\,</math>. Let <math>I=(I(1),I(2),\ldots,I(m))</math> be a [[multi-index]] (an ordered <math>m</math>-tuple of integers), then


The four most common Maxwell relations are the equalities of the second derivatives of each of the four thermodynamic potentials, with respect to their thermal natural variable ([[temperature]] ''T''; or [[entropy]] ''S'') and their ''mechanical'' natural variable ([[pressure]] ''P''; or [[volume]] ''V''):
:<math>|I| := \sum_{i=1}^{m} I(i)</math>


{{Equation box 1
:<math>\frac{\partial^{|I|}}{\partial x^{I}} := \prod_{i=1}^{m} \left( \frac{\partial}{\partial x^{i}} \right)^{I(i)}.</math>
|title = '''Maxwell's relations''' ''(common)''
|indent =:
|equation =


<math> \begin{align}
Define the local sections <math>\sigma, \eta \in \Gamma(\pi)</math> to have the same '''<math>r\,</math>-jet''' at <math>p\,</math> if
+\left(\frac{\partial T}{\partial V}\right)_S &=& -\left(\frac{\partial P}{\partial S}\right)_V &=& \frac{\partial^2 U }{\partial S \partial V}\\


+\left(\frac{\partial T}{\partial P}\right)_S &=& +\left(\frac{\partial V}{\partial S}\right)_P &=& \frac{\partial^2 H }{\partial S \partial P}\\
:<math>\left.\frac{\partial^{|I|} \sigma^{\alpha}}{\partial x^{I}}\right|_{p} = \left.\frac{\partial^{|I|} \eta^{\alpha}}{\partial x^{I}}\right|_{p}, \quad 0 \leq |I| \leq r.  </math>
+\left(\frac{\partial S}{\partial V}\right)_T &=& +\left(\frac{\partial P}{\partial T}\right)_V &=& -\frac{\partial^2 A }{\partial T \partial V}\\


-\left(\frac{\partial S}{\partial P}\right)_T &=& +\left(\frac{\partial V}{\partial T}\right)_P &=& \frac{\partial^2 G }{\partial T \partial P}
The relation that two maps have the same <math>r</math>-jet is an [[equivalence relation]]. An ''r''-jet is an [[equivalence class]] under this relation, and the ''r''-jet with representative <math>\sigma\,</math> is denoted <math>j^{r}_{p}\sigma</math>. The integer <math>r</math> is also called the '''order''' of the jet.
\end{align}\,\!</math>


|border colour = #0073CF
<math>p\,</math> is the '''source''' of <math>j^{r}_{p}\sigma</math>.
|background colour=#F5FFFA}}


where the potentials as functions of their natural thermal and mechanical variables are the [[internal energy]] ''U''(''S, V''), [[Enthalpy]] ''H''(''S, P''), [[Helmholtz free energy]] ''A''(''T, V'') and [[Gibbs free energy]] ''G''(''T, P''). The [[thermodynamic square]] can be used as a [[mnemonic]] to recall and derive these relations.
<math>\sigma\,(p)</math> is the '''target''' of <math>j^{r}_{p}\sigma</math>.


=== Derivation ===
==Jet manifolds==


Maxwell relations are based on simple partial differentiation rules, in particular the [[Total derivative|total]] [[differential of a function]] and the symmetry of evaluating second order partial derivatives.
The '''<math>r^{th}\,</math> jet manifold of <math>\pi\,</math>''' is the set


:{| class="toccolours collapsible collapsed" width="80%" style="text-align:left"
:<math>\{j^{r}_{p}\sigma:p \in \mathcal{M}, \sigma \in \Gamma(\pi)\}</math>
!Derivation
 
and is denoted <math>J^{r}\pi\,</math>. We may define projections <math>\pi_{r}\,</math> and <math>\pi_{r,0}\,</math> called the '''source and target projections''' respectively, by
 
:{|
|-
|<math>\pi_{r}:J^{r}\pi\, </math>
|<math>\longrightarrow \mathcal{M} </math>
|-
|align=right|<math>j^{r}_{p}\sigma </math>
||<math>\longmapsto p </math>
|-
|
|
|-
|
|
|-
|<math>\pi_{r,0}:J^{r}\pi\, </math>
|<math>\longrightarrow \mathcal{E} </math>
|-
|align=right|<math>j^{r}_{p}\sigma </math>
|<math>\longmapsto \sigma(p) </math>
|-
|}
 
If <math>1 \leq k \leq r</math>, then the '''<math>k</math>-jet projection''' is the function <math>\pi_{r,k}\,</math> defined by
 
:{|
|-
|<math>\pi_{r,k}:J^{r}\pi \,</math>
|<math>\longrightarrow J^{k}\pi </math>
|-
|align=right|<math>j^{r}_{p}\sigma </math>
|<math>\longmapsto j^{k}_{p}\sigma </math>
|-
|}
 
From this definition, it is clear that <math>\pi_{r} = \pi \circ \pi_{r,0}</math> and that if <math>0 \leq m \leq k</math>, then <math>\pi_{r,m} = \pi_{k,m} \circ \pi_{r,k}</math>. It is conventional to regard <math>\pi_{r,r}=\operatorname{id}_{J^{r}\pi}\,</math>, the [[identity function|identity map]] on <math>J^{r}\pi \,</math> and to identify <math>J^{0}\pi\,</math> with <math>\mathcal{E}</math>.
 
The functions <math>\pi_{r,k}, \pi_{r,0}\,</math> and <math>\pi_{r}\,</math> are [[smooth]] [[surjective]] [[submersion (mathematics)|submersion]]s.
 
[[File:Jet_Bundle_Image_FbN.png|500px|center]]
 
A [[coordinate system]] on <math>\mathcal{E}</math> will generate a coordinate system on <math>J^{r}\pi\,</math>. Let <math>(U,u)\,</math> be an adapted [[coordinate chart]] on <math>\mathcal{E}</math>, where <math>u = (x^{i}, u^{\alpha})\,</math>. The '''induced coordinate chart <math>(U^{r}, u^{r})\,</math>''' on <math>J^{r}\pi\,</math> is defined by
 
:{|
|-
|align=right|<math>U^{r} \,</math>
|<math>= \{ j^{r}_{p}\sigma: \sigma(p) \in U \} \,</math>
|-
|align=right|<math>u^{r} \,</math>
|<math>= (x^{i}, u^{\alpha}, u^{\alpha}_{I})\,</math>
|-
|}
 
where
 
:{|
|-
|<math>x^{i}(j^{r}_{p}\sigma) \,</math>
|<math>= x^{i}(p) \,</math>
|-
|<math>u^{\alpha}(j^{r}_{p}\sigma) \,</math>
|<math>= u^{\alpha}(\sigma(p)) \, </math>
|-
|}
 
and the <math>n \left( {}^{m+r}C_{r} -1\right)\,</math> functions
 
:<math>u^{\alpha}_{I}:U^{k} \longrightarrow \mathbb{R}\,</math>
 
are specified by
 
:<math>u^{\alpha}_{I}(j^{r}_{p}\sigma) = \left.\frac{\partial^{|I|} \sigma^{\alpha}}{\partial x^{I}}\right|_{p}</math>
 
and are known as the '''derivative coordinates'''.
 
Given an atlas of adapted charts <math>(U,u)\,</math> on <math>\mathcal{E}</math>, the corresponding collection of charts <math>(U^{r},u^{r})\,</math> is a [[finite-dimensional]] <math>C^{\infty}\,</math> atlas on <math>J^{r}\pi\,</math>.
 
==Jet bundles==
Since the atlas on each <math>J^{r}\pi\,</math> defines a manifold, the triples <math>(J^{r}\pi, \pi_{r,k}, J^{k}\pi), (J^{r}\pi, \pi_{r,0}, \mathcal{E})\,</math> and <math>(J^{r}\pi, \pi_{r}, \mathcal{M})\,</math> all define fibered manifolds.
In particular, if <math>(\mathcal{E}, \pi, \mathcal{M})\,</math> is a fiber bundle, the triple <math>(J^{r}\pi, \pi_{r}, \mathcal{M})\,</math> defines the '''<math>r^{th}\,</math> jet bundle of <math>\pi\,</math>'''.
 
If <math>W \subset \mathcal{M}\,</math> is an open submanifold, then
 
:<math> J^{r}\left(\pi|_{\pi^{-1}(W)}\right) \cong \pi^{-1}_{r}(W).\,</math>
 
If <math>p \in \mathcal{M}\,</math>, then the fiber <math>\pi^{-1}_{r}(p)\,</math> is denoted <math>J^{r}_{p}\pi\,</math>.
 
Let <math>\sigma\,</math> be a local section of <math>\pi\,</math> with domain <math>W \subset \mathcal{M}\,</math>. The '''<math>r^{th}\,</math> jet prolongation of <math>\sigma\,</math>''' is the map <math>j^{r}\sigma:W \longrightarrow J^{r}\pi\,</math> defined by
 
:<math> (j^{r}\sigma)(p) = j^{r}_{p}\sigma. \,</math>
 
Note that <math>\pi_{r} \circ j^{r}\sigma = \operatorname{id}_{W} \,</math>, so <math>j^{r}\sigma\,</math> really is a section. In local coordinates, <math>j^{r}\sigma\,</math> is given by
 
:<math> \left(\sigma^{\alpha}, \frac{\partial^{|I|} \sigma^{\alpha}}{\partial x^{|I|}}\right) \qquad 1 \leq |I| \leq r. \,</math>
 
We identify <math>j^{0}\sigma\,</math> with <math>\sigma\,</math>.
 
===Example===
If <math>\pi\,</math> is the [[trivial bundle]] <math>(\mathcal{M} \times \mathbb{R}, pr_{1}, \mathcal{M})</math>, then there is a canonical [[diffeomorphism]] between the first jet bundle <math>J^{1}\pi\,</math> and <math>T^{*}\mathcal{M} \times \mathbb{R} </math>.
To construct this diffeomorphism, for each <math>\sigma \in \Gamma_{M}(\pi)\,</math> write <math>\bar{\sigma} = pr_{2} \circ \sigma \in C^{\infty}(M)\,</math>.
 
Then, whenever <math>p \in M \,</math>
 
:<math> j^{1}_{p}\sigma = \{ \psi : \psi \in \Gamma_{p}(\pi); \bar{\psi}(p) = \bar{\sigma}(p); d\bar{\psi}_{p} = d\bar{\sigma}_{p} \}. \,</math>
 
Consequently, the mapping
 
:{|
|-
|<math>J^{1}\pi \,</math>
|<math>\longrightarrow T^{*}\mathcal{M} \times \mathbb{R}</math>
|-
|align=right|<math>j^{1}_{p}\sigma \,</math>
|<math> \longmapsto (d\bar{\sigma}_{p},\bar{\sigma}(p)) \,</math>
|-
|}
 
is well-defined and is clearly [[injective]]. Writing it out in coordinates shows that it is a diffeomorphism, because if <math>(x^{i},u)\,</math> are coordinates on <math>\mathcal{M} \times \mathbb{R}</math>, where <math>u=id_{\mathbb{R}}\,</math> is the identity coordinate, then the derivative coordinates <math>u_{i}\,</math> on <math>J^{1}\pi\,</math> correspond to the coordinates <math>\partial_{i}\,</math> on <math>T^{*}\mathcal{M}\,</math>.
 
Likewise, if <math>\pi\,</math> is the trivial bundle <math>(\mathbb{R} \times \mathcal{M}, pr_{1}, \mathbb{R})</math>, then there exists a canonical diffeomorphism between <math>J^{1}\pi\,</math> and <math>\mathbb{R} \times T\mathcal{M}\,</math>
 
==Contact forms==
A [[differential 1-form]] <math>\theta\,</math> on the space <math>J^{r}\pi\,</math> is called a '''[[contact form]]''' (i.e. <math>\theta \in \Lambda_{C}^{r}\pi\,</math>) if it is [[pullback (differential geometry)|pulled back]] to the zero form on <math>\mathcal{M}\,</math> by all prolongations.
In other words, if <math>\theta \in \Lambda^{1}J^{r+1}\pi\,</math>, then <math>\theta \in \Lambda_{C}^{1}\pi_{r+1,r}\,</math> [[if and only if]], for every open submanifold <math>W \subset \mathcal{M}\,</math> and every <math>\sigma \in \Gamma_{W}(\pi),\,</math>
 
:<math>(j^{k+1}\sigma)^{*}\theta = 0.\,</math>
 
The [[distribution (differential geometry)|distribution]] on <math>J^{r}\pi\,</math> generated by the contact forms is called the '''Cartan distribution'''. It is the main geometrical structure on jet spaces and plays an important role in the geometric theory of [[partial differential equation]]s. The Cartan distributions are not [[distribution (differential geometry)|involutive]] and are of growing dimension when passing to higher order jet spaces. Surprisingly though, when passing to the space of infinite order jets <math>J^\infty</math> this distribution is involutive and finite dimensional. Its dimension coinciding with the dimension of the base manifold <math>\mathcal{M}</math>.
===Example===
Let us consider the case <math>(\mathcal{E},\pi,\mathcal{M})</math>, where <math>\mathcal{E} \simeq \mathbb{R}^{2}</math> and <math>\mathcal{M} \simeq \mathbb{R}</math>.
Then, <math>(J^{1}\pi, \pi, \mathcal{M})</math> defines the first jet bundle, and may be coordinated by <math>(x,u,u_{1})\,</math>, where
 
:{|
|-
|align=right|<math>x(j^{1}_{p}\sigma) </math>
|align=left|<math>= x(p) = x\,</math>
|-
|align=right|<math>u(j^{1}_{p}\sigma) </math>
|align=left|<math>= u(\sigma(p)) = u(\sigma(x)) = \sigma(x) \,</math>
|-
|align=right|<math>u_{1}(j^{1}_{p}\sigma) </math>
|align=left|<math>= \left.\frac{\partial \sigma}{\partial x}\right|_{p} = \sigma'(x)</math>
|-
|}
 
for all <math> p \in \mathcal{M}</math> and <math>\sigma \in \Gamma_{p}(\pi)\,</math>. A general 1-form on <math>J^{1}\pi\,</math> takes the form
 
:<math>\theta = a(x, u, u_{1})dx + b(x, u, u_{1})du + c(x, u,u_{1})du_{1}\,</math>
 
A section  <math>\sigma \in \Gamma_{p}(\pi)\,</math> has first prolongation <math> j^{1}\sigma = (u,u_{1}) = \left(\sigma(p), \left.\frac{\partial \sigma}{\partial x}\right|_{p}\right)\,</math>.
Hence, <math>(j^{1}\sigma)^{*} \theta\,</math> can be calculated as
 
:{|
|-
|<math>(j^{1}_{p}\sigma)^{*} \theta \,</math>
|<math>= \theta \circ j^{1}_{p}\sigma \, </math>
|-
|
|<math>= a(x, \sigma(x), \sigma'(x))dx + b(x, \sigma(x), \sigma'(x))d(\sigma(x)) + c(x, \sigma(x),\sigma'(x))d(\sigma'(x)) \,</math>
|-
|
|<math>= a(x, \sigma(x), \sigma'(x))dx + b(x, \sigma(x), \sigma'(x))\sigma'(x)dx + c(x, \sigma(x),\sigma'(x))\sigma''(x)dx \,</math>
|-
|
|<math>= [\, a(x, \sigma(x), \sigma'(x)) + b(x, \sigma(x), \sigma'(x))\sigma'(x) + c(x, \sigma(x),\sigma'(x))\sigma''(x)\, ]dx \, </math>
|-
|}
 
This will vanish for all sections <math>\sigma\,</math> if and only if <math>c=0\,</math> and <math>a = -b\sigma'(x)\,</math>. Hence, <math>\theta=b(x, u, u_{1})\theta_{0}\,</math> must necessarily be a multiple of the basic contact form <math>\theta_{0}=du-u_{1}dx\,</math>.
Proceeding to the second jet space <math>J^{2}\pi\,</math> with additional coordinate <math>u_{2}\,</math>, such that
 
:<math>u_{2}(j^{2}_{p}\sigma)=\left.\frac{\partial^{2} \sigma}{\partial x^{2}}\right|_{p} = \sigma''(x)\,</math>
 
a general 1-form has the construction
 
:<math> \theta = a(x, u, u_{1},u_{2})dx + b(x, u, u_{1},u_{2})du + c(x, u, u_{1},u_{2})du_{1} + e(x, u, u_{1},u_{2})du_{2}\,</math>
 
This is a contact form [[if and only if]]
 
:{|
|-
|<math> (j^{2}_{p}\sigma)^{*} \theta \,</math>
|<math>= \theta \circ j^{2}_{p}\sigma \,</math>
|-
|
|<math>= a(x, \sigma(x), \sigma'(x),\sigma''(x))dx + b(x, \sigma(x),\sigma'(x),\sigma''(x))d(\sigma(x))+ \,</math>
|-
|
|    <math>+ c(x, \sigma(x),\sigma'(x),\sigma'(x))d(\sigma'(x)) +  e(x, \sigma(x), \sigma'(x),\sigma''(x))d(\sigma''(x)) \,</math>
|-
|
|<math>= adx + b\sigma'(x)dx + c\sigma''(x)dx + e\sigma'''(x)dx\,</math>
|-
|
|<math>= [\, a + b\sigma'(x) + c\sigma''(x) + e\sigma'''(x)\,]dx\,</math>
|-
|
|<math>= 0\,</math>
|-
|}
 
which implies that <math>e=0\,</math> and <math>a=-b\sigma'(x)-c\sigma''(x)\,</math>. Therefore, <math>\theta\,</math> is a contact form if and only if
 
:<math>\theta = b(x, \sigma(x), \sigma'(x))\theta_{0} + c(x, \sigma(x), \sigma'(x))\theta_{1}\,</math>
 
where <math>\theta_{1} = du_{1} - u_{2}dx\,</math> is the next basic contact form
(Note that here we are identifying the form <math>\theta_{0}\,</math> with its pull-back <math>(\pi_{2,1})^{*}\theta_{0}\,</math> to <math>J^{2}\pi\,</math>).
 
In general, providing <math>x,u, \in \mathbb{R}\,</math>, a contact form on <math>J^{r+1}\pi\,</math> can be written as a [[linear combination]] of the basic contact forms
 
:<math>\theta_{k} = du_{k} - u_{k+1}dx \qquad k=0, \ldots, r-1\,</math>
 
where <math> u_{k}(j^{k}\sigma)= \left.\frac{\partial^{k} \sigma}{\partial x^{k}}\right|_{p}\,</math>.
 
Similar arguments lead to a complete characterization of all contact forms.
 
In local coordinates, every contact one-form on <math>J^{r+1}\pi\,</math> can be written as a linear combination
 
:<math>\theta = \sum_{|I|=0}^{r} P_{\alpha}^{I}\theta_{I}^{\alpha}\,</math>
 
with smooth coefficients <math>P^{\alpha}_{I}(x^{i},u^{\alpha})\,</math> of the basic contact forms
 
:<math>\theta_{I}^{\alpha} = du^{\alpha}_{I} - u^{\alpha}_{I,i}dx^{i}\,</math>
 
<math>|I|\,</math> is known as the '''order''' of the contact form <math>\theta_{I}^{\alpha}</math>. Note that contact forms on <math>J^{r+1}\pi\,</math> have orders at most <math>r\,</math>.
Contact forms provide a characterization of those local sections of <math>\pi_{r+1}\,</math> which are prolongations of sections of <math>\pi\,</math>.
 
Let <math>\psi \in \Gamma_{W}(\pi_{r+1})\,</math>, then <math>\psi = j^{r+1}\sigma\,</math> where <math>\sigma \in \Gamma_{W}(\pi)\,</math> if and only if <math>\psi^{*}(\theta|_{W})=0, \forall \theta \in \Lambda_{C}^{1}\pi_{r+1,r}.\,</math>
 
==Vector fields==
A general [[vector field]] on the total space <math>\mathcal{E}</math>, coordinated by <math>(x,u) \ \stackrel{\mathrm{def}}{=}\  (x^{i},u^{\alpha})\,</math>, is
 
:<math>V \ \stackrel{\mathrm{def}}{=}\  \rho^{i}(x,u)\frac{\partial}{\partial x^{i}} + \phi^{\alpha}(x,u)\frac{\partial}{\partial u^{\alpha}}.\,</math>
 
A vector field is called '''horizontal''', meaning all the vertical coefficients vanish, if <math>\phi^{\alpha}=0\,</math>.
 
A vector field is called '''vertical''', meaning all the horizontal coefficients vanish, if <math>\rho^{i}=0\,</math>.
 
For fixed <math>(x,u)\,</math>, we identify
 
:<math> V_{(xu)} \ \stackrel{\mathrm{def}}{=}\  \rho^{i}(x,u) \frac{\partial}{\partial x^{i}} + \phi^{\alpha}(x,u) \frac{\partial}{\partial u^{\alpha}}\,</math>
 
having coordinates <math>(x,u,\rho^{i},\phi^{\alpha})\,</math>, with an element in the fiber <math>T_{xu}\mathcal{E}</math> of <math>T\mathcal{E}</math> over <math>(x,u) \in \mathcal{E}</math>, called '''a [[tangent vector]] in <math>T\mathcal{E}</math>'''. A section
 
:{|
|-
|<math>\psi : \mathcal{E} \,</math>
|<math>\longrightarrow T\mathcal{E} </math>
|-
|align=right|<math>(x,u) \,</math>
|<math>\longmapsto \psi(x,u) = V\,</math>
|-
|}
 
is called '''a vector field on <math>\mathcal{E}</math>''' with <math> V = \rho^{i}(x,u) \frac{\partial}{\partial x^{i}} + \phi^{\alpha}(x,u) \frac{\partial}{\partial u^{\alpha}}\,</math> and <math>\psi \in \Gamma(T\mathcal{E})\,</math>.
 
The jet bundle <math>J^{r}\pi\,</math> is coordinated by <math>(x,u,w) \ \stackrel{\mathrm{def}}{=}\  (x^{i},u^{\alpha},w_{i}^{\alpha})\,</math>.  For fixed <math>(x,u,w)\,</math>, identify
 
:{|
|-
|<math>V_{(xuw)} \ \stackrel{\mathrm{def}}{=}\  \,</math>
|<math>V^{i}(x,u,w) \frac{\partial}{\partial x^{i}} + V^{\alpha}(x,u,w) \frac{\partial}{\partial u^{\alpha}} \ + \ V^{\alpha}_{i}(x,u,w) \frac{\partial}{\partial w^{\alpha}_{i}} +\,</math>
|-
|
|<math>\qquad + \ V^{\alpha}_{i_{1}i_{2}}(x,u,w) \frac{\partial}{\partial w^{\alpha}_{i_{1}i_{2}}} + \cdots \ + \ \cdots + V^{\alpha}_{i_{1}i_{2} \cdots i_{r}}(x,u,w) \frac{\partial}{\partial w^{\alpha}_{i_{1}i_{2} \cdots i_{r}}}\,</math>
|-
|}
 
having coordinates <math>(x,u,w,v^{\alpha}_{i}, v^{\alpha}_{i_{1} i_{2}},\ldots,v^{ \alpha}_{i_{1}i_{2} \cdots i_{r}})\,</math>, with an element in the fiber <math>T_{xuw}(J^{r}\pi)\,</math> of <math>T(J^{r}\pi)\,</math> over <math>(x,u,w) \in J^{r}\pi\,</math>, called '''a tangent vector in <math>T(J^{r}\pi)\,</math>'''.
Here, <math>v^{\alpha}_{i}, v^{\alpha}_{i_{1}i_{2}},\ldots,v^{\alpha}_{i_{1}i_{2} \cdots i_{r}}\,</math> are real-valued functions on <math>J^{r}\pi\,</math>. A section
 
:{|
|-
|<math>\Psi : J^{r}\pi \,</math>
|<math>\longrightarrow T(J^{r}\pi) \,</math>
|-
|align=right|<math>(x,u,w) \, </math>
|<math>\longmapsto \Psi(u,w) = V \, </math>
|-
|-
|Derivation of the Maxwell relations can be deduced from the differential forms of the [[thermodynamic potentials]]:
|}


:<math>\begin{align}  
is '''a vector field on <math>J^{r}\pi\,</math>''', and we say <math>\Psi \in \Gamma(T(J^{r}\pi))\,</math>.
dU &=& TdS-PdV \\
dH &=& TdS+VdP \\
dA &=& -SdT-PdV \\
dG &=& -SdT+VdP \\
\end{align}\,\!</math>


These equations resemble [[Total derivative|total differentials]] of the form
==Partial differential equations==


:<math>dz = \left(\frac{\partial z}{\partial x}\right)_y\!dx +
Let <math>(\mathcal{E},\pi,\mathcal{M})</math> be a fiber bundle. An '''<math>r^{th}\,</math> order [[partial differential equation]]''' on <math>\pi\,</math> is a [[closed]]{{dn|date=July 2012}} [[embedding|embedded]] submanifold <math>\mathcal{S}</math> of the jet manifold <math>J^{r}\pi\,</math>.
\left(\frac{\partial z}{\partial y}\right)_x\!dy</math>
A solution is a local section <math>\sigma \in \Gamma_{W}(\pi)\,</math> satisfying <math>j^{r}_{p}\sigma \in \mathcal{S}, \forall p \in \mathcal{M}</math>.


And indeed, it can be shown for any equation of the form
Let us consider an example of a first order partial differential equation.


:<math>dz = Mdx + Ndy \,</math>
===Example===
Let <math>\pi\,</math> be the trivial bundle <math>(\mathbb{R}^{2} \times \mathbb{R}, pr_{1}, \mathbb{R}^{2})\,</math> with global coordinates <math>(x^{1}, x^{2}, u^{1})\,</math>.
Then the map <math>F:J^{1}\pi \longrightarrow \mathbb{R}\,</math> defined by


that
:<math>F = u^{1}_{1}u^{1}_{2} - 2x^{2}u^{1}\,</math>


:<math>M = \left(\frac{\partial z}{\partial x}\right)_y, \quad
gives rise to the differential equation
N = \left(\frac{\partial z}{\partial y}\right)_x</math>


Consider, as an example, the equation <math>dH=TdS+VdP\,</math>. We can now immediately see that
:<math>S = \{ j^{1}_{p}\sigma \in J^{1}\pi : (u^{1}_{1}u^{1}_{2} - 2x^{2}u^{1})(j^{1}_{p}\sigma)=0 \} \,</math>


:<math>T = \left(\frac{\partial H}{\partial S}\right)_P, \quad
which can be written
      V = \left(\frac{\partial H}{\partial P}\right)_S</math>


Since we also know that for functions with continuous second derivatives, the mixed partial derivatives are identical ([[Symmetry of second derivatives]]), that is, that
:<math>\frac{\partial \sigma}{\partial x^{1}}\frac{\partial \sigma}{\partial x^{2}} - 2x^{2}\sigma = 0. \,</math>


:<math>\frac{\partial}{\partial y}\left(\frac{\partial z}{\partial x}\right)_y =
The particular section <math>\sigma:\mathbb{R}^{2} \longrightarrow \mathbb{R}^{2} \times \mathbb{R}\,</math> defined by
\frac{\partial}{\partial x}\left(\frac{\partial z}{\partial y}\right)_x =
\frac{\partial^2 z}{\partial y \partial x} = \frac{\partial^2 z}{\partial x \partial y}</math>


we therefore can see that
:<math>\sigma(p_{1},p_{2}) = (p^{1},p^{2},p^{1}(p^{2})^{2}) \,</math>


:<math> \frac{\partial}{\partial P}\left(\frac{\partial H}{\partial S}\right)_P =
has first prolongation given by
\frac{\partial}{\partial S}\left(\frac{\partial H}{\partial P}\right)_S </math>


and therefore that
:<math> j^{1}\sigma(p_{1},p_{2}) = (p^{1},p^{2},p^{1}(p^{2})^{2},(p^{2})^{2},2p^{1}p^{2}) \,</math>


:<math>\left(\frac{\partial T}{\partial P}\right)_S = \left(\frac{\partial V}{\partial S}\right)_P</math>
and is a solution of this differential equation, because


Each of the four Maxwell relationships given above follows similarly from one of the [[Gibbs equations]].
:{|
|-
|<math>(u^{1}_{1}u^{1}_{2} - 2x^{2}u^{1})(j^{1}_{p}\sigma) \,</math>
|<math>= u^{1}_{1}(j^{1}_{p}\sigma)u^{1}_{2}(j^{1}_{p}\sigma) - 2x^{2}(j^{1}_{p}\sigma)u^{1}(j^{1}_{p}\sigma) \,</math>
|-
|
|<math>= (p^{2})^{2} \cdot 2p^{1}p^{2} - 2 \cdot p^{2} \cdot p^{1}(p^{2})^{2} \,</math>
|-
|
|<math>= 2p^{1}(p^{2})^3 - 2p^{1}(p^{2})^3 \,</math>
|-
|
|<math>= 0 \,</math>
|-
|}
|}


:{| class="toccolours collapsible collapsed" width="80%" style="text-align:left"
and so <math>j^{1}_{p}\sigma \in \mathcal{S}\,</math> for ''every'' <math>p \in \mathbb{R}^{2}\,</math>.
!Extended derivation
 
==Jet Prolongation==
A local diffeomorphism <math>\psi:J^{r}\pi \longrightarrow J^{r}\pi\,</math> defines a contact transformation of order <math>r\,</math> if it preserves the contact ideal, meaning that if <math>\theta\,</math> is any contact form on <math>J^{r}\pi\,</math>, then <math>\psi^{*}\theta\,</math> is also a contact form.
 
The flow generated by a vector field <math>V^{r}\,</math> on the jet space <math>J^{r}\,</math> forms a one-parameter group of contact transformations if and only if the [[Lie derivative]] <math>\mathcal{L}_{V^{r}}(\theta)</math> of any contact form <math>\theta\,</math> preserves the contact ideal.
 
Let us begin with the first order case. Consider a general vector field <math>V^{1}\,</math> on <math>J^{1}\pi\,</math>, given by
 
:<math> V^{1} \ \stackrel{\mathrm{def}}{=}\  \rho^{i}(u^{1})\frac{\partial}{\partial x^{i}} + \phi^{\alpha}(u^{1})\frac{\partial}{\partial u^{\alpha}} + \chi^{\alpha}_{i}(u^{1})\frac{\partial}{\partial u^{\alpha}_{i}}. \,</math>
 
We now apply <math>\mathcal{L}_{V^{1}}</math> to the basic contact forms <math>\theta^{\alpha} = du^{\alpha} - u_{i}^{\alpha}dx^{i}\,</math>, and obtain
 
:{|
|-
|<math>\mathcal{L}_{V^{1}}(\theta^{\alpha}) </math>
|<math>= \mathcal{L}_{V^{1}}(du^{\alpha} - u_{i}^{\alpha}dx^{i}) </math>
|-
|
|<math>= \mathcal{L}_{V^{1}}du^{\alpha} - (\mathcal{L}_{V^{1}}u_{i}^{\alpha})dx^{i} - u_{i}^{\alpha}(\mathcal{L}_{V^{1}}dx^{i}) \,</math>
|-
|
|<math>= d(V^{1}u^{\alpha}) - V^{1}u_{i}^{\alpha}dx^{i} - u_{i}^{\alpha}d(V^{1}x^{i}) \,</math>
|-
|
|<math>= d\phi^{\alpha} - \chi^{\alpha}_{i}dx^{i} - u_{i}^{\alpha}d\rho^{i} \,</math>
|-
|
|<math>= \frac{\partial \phi^{\alpha}}{\partial x^{i}}\, dx^{i} + \frac{\partial \phi^{\alpha}}{\partial u^{k}}\, du^{k} + \frac{\partial \phi^{\alpha}}{\partial u^{k}_{i}}\, du^{k}_{i} - \chi^{\alpha}_{i}dx^{i} - u_{i}^{\alpha}\left[ \frac{\partial \rho^{i}}{\partial x^{m}}\, dx^{m} + \frac{\partial \rho^{i}}{\partial u^{k}}\, du^{k} + \frac{\partial \rho^{i}}{\partial u^{k}_{m}}\, du^{k}_{m} \right ] \,</math>
|-
|-
|Combined form first and second law of thermodynamics,
|}
:<math>TdS = dU+PdV</math> (Eq.1)


U, S, and V are state functions.
where we have expanded the [[exterior derivative]] of the functions in terms of their coordinates.
Let,
Next, we note that
:<math>U = U(x,y)</math>
:<math>S = S(x,y)</math>
:<math>V = V(x,y)</math>
:<math>dU = \left(\frac{\partial U}{\partial x}\right)_y\!dx +
\left(\frac{\partial U}{\partial y}\right)_x\!dy</math>


:<math>dS = \left(\frac{\partial S}{\partial x}\right)_y\!dx +
:<math> \theta^{k} = du^{k} - u_{i}^{k}dx^{i} \quad \Longrightarrow \quad du^{k} = \theta^{k} + u_{i}^{k}dx^{i} \,</math>
\left(\frac{\partial S}{\partial y}\right)_x\!dy</math>


:<math>dV = \left(\frac{\partial V}{\partial x}\right)_y\!dx +
and so we may write
\left(\frac{\partial V}{\partial y}\right)_x\!dy</math>


Substitute them in Eq.1 and one gets,
:{|
:<math>T\left(\frac{\partial S}{\partial x}\right)_y\!dx +
|-
T\left(\frac{\partial S}{\partial y}\right)_x\!dy = \left(\frac{\partial U}{\partial x}\right)_y\!dx +
|<math>\mathcal{L}_{V^{1}}(\theta^{\alpha}) \,</math>
\left(\frac{\partial U}{\partial y}\right)_x\!dy + P\left(\frac{\partial V}{\partial x}\right)_y\!dx +
|<math>= \frac{\partial \phi^{\alpha}}{\partial x^{i}}\, dx^{i} + \frac{\partial \phi^{\alpha}}{\partial u^{k}}\, (\theta^{k} + u_{i}^{k}dx^{i}) + \frac{\partial \phi^{\alpha}}{\partial u^{k}_{i}}\, du^{k}_{i} - \chi^{\alpha}_{i}dx^{i} - \,</math>
P\left(\frac{\partial V}{\partial y}\right)_x\!dy</math>
|-
|
|    <math>- u_{l}^{\alpha}\left[ \frac{\partial \rho^{l}}{\partial x^{i}}\, dx^{i} + \frac{\partial \rho^{l}}{\partial u^{k}}\, (\theta^{k} + u_{i}^{k}dx^{i}) + \frac{\partial \rho^{l}}{\partial u^{k}_{i}}\, du^{k}_{i} \right ] \,</math>
|-
|
|<math>= \left[ \frac{\partial \phi^{\alpha}}{\partial x^{i}} + \frac{\partial \phi^{\alpha}}{\partial u^{k}}u_{i}^{k} - u_{l}^{\alpha}\left(\frac{\partial \rho^{l}}{\partial x^{i}} + \frac{\partial \rho^{l}}{\partial u^{k}}u_{i}^{k}\right)- \chi^{\alpha}_{i}\right]\, dx^{i} + \left[ \frac{\partial \phi^{\alpha}}{\partial u^{k}_{i}} - u_{l}^{\alpha}\frac{\partial \rho^{l}}{\partial u^{k}_{i}}\right]\, du^{k}_{i} + \,</math>
|-
|
|    <math>+ \left( \frac{\partial \phi^{\alpha}}{\partial u^{k}} - u_{l}^{\alpha}\frac{\partial \rho^{l}}{\partial u^{k}} \right)\theta^{k}.\,</math>
|-
|}


And also written as,
Therefore, <math>V^{1}\,</math> determines a contact transformation if and only if the coefficients of <math>dx^{i}\,</math> and <math>du^{k}_{i}\,</math> in the formula vanish.
:<math>\left(\frac{\partial U}{\partial x}\right)_y\!dx +
The latter requirements imply the '''contact conditions'''
\left(\frac{\partial U}{\partial y}\right)_x\!dy = T\left(\frac{\partial S}{\partial x}\right)_y\!dx +
T\left(\frac{\partial S}{\partial y}\right)_x\!dy - P\left(\frac{\partial V}{\partial x}\right)_y\!dx -
P\left(\frac{\partial V}{\partial y}\right)_x\!dy</math>


comparing the coefficient of dx and dy, one gets
:<math>\frac{\partial \phi^{\alpha}}{\partial u^{k}_{i}} - u^{\alpha}_{l} \frac{\partial \rho^{l}}{\partial u^{k}_{i}} = 0\,</math>
:<math>\left(\frac{\partial U}{\partial x}\right)_y = T\left(\frac{\partial S}{\partial x}\right)_y - P\left(\frac{\partial V}{\partial x}\right)_y</math>
:<math>\left(\frac{\partial U}{\partial y}\right)_x = T\left(\frac{\partial S}{\partial y}\right)_x - P\left(\frac{\partial V}{\partial y}\right)_x</math>


Differentiating above equations by y, x respectively<br />
The former requirements provide explicit formulae for the coefficients of the first derivative terms in <math>V^{1}\,</math>:
:<math>\left(\frac{\partial^2U}{\partial y\partial x}\right) = \left(\frac{\partial T}{\partial y}\right)_x \left(\frac{\partial S}{\partial x}\right)_y + T\left(\frac{\partial^2 S}{\partial y\partial x}\right) - \left(\frac{\partial P}{\partial y}\right)_x \left(\frac{\partial V}{\partial x}\right)_y - P\left(\frac{\partial^2 V}{\partial y\partial x}\right)</math> (Eq.2)
:and
:<math>\left(\frac{\partial^2U}{\partial x\partial y}\right) = \left(\frac{\partial T}{\partial x}\right)_y \left(\frac{\partial S}{\partial y}\right)_x + T\left(\frac{\partial^2 S}{\partial x\partial y}\right) - \left(\frac{\partial P}{\partial x}\right)_y \left(\frac{\partial V}{\partial y}\right)_x - P\left(\frac{\partial^2 V}{\partial x\partial y}\right)</math> (Eq.3)


U, S, and V are exact differentials, therefore,
:<math>\chi^{\alpha}_{i} = \widehat{D}_{i} \phi^{\alpha} - u^{\alpha}_{l}(\widehat{D}_{i}\rho^{l}) </math> where <math>\widehat{D}_{i} = \frac{\partial}{\partial x^{i}} + u^{k}_{i}\frac{\partial}{\partial u^{k}} </math>
:<math>\left(\frac{\partial^2U}{\partial y\partial x}\right) = \left(\frac{\partial^2U}{\partial x\partial y}\right)</math>
:<math>\left(\frac{\partial^2S}{\partial y\partial x}\right) = \left(\frac{\partial^2S}{\partial x\partial y}\right)
:\left(\frac{\partial^2V}{\partial y\partial x}\right) = \left(\frac{\partial^2V}{\partial x\partial y}\right)</math>


Subtract eqn(2) and (3) and one gets<br />
denotes the zeroth order truncation of the total derivative <math>D_{i}\,</math>.
:<math>\left(\frac{\partial T}{\partial y}\right)_x \left(\frac{\partial S}{\partial x}\right)_y - \left(\frac{\partial P}{\partial y}\right)_x \left(\frac{\partial V}{\partial x}\right)_y = \left(\frac{\partial T}{\partial x}\right)_y \left(\frac{\partial S}{\partial y}\right)_x - \left(\frac{\partial P}{\partial x}\right)_y \left(\frac{\partial V}{\partial y}\right)_x</math>
:''Note: The above is called the general expression for Maxwell's thermodynamical relation.''


;Maxwell's first relation
Thus, the contact conditions uniquely prescribe the prolongation of any point or contact vector field. That is, if <math>\mathcal{L}_{V^{r}}\,</math> satisfies these equations, <math>V^{r}\,</math> is called the '''<math>r^{th}\,</math> prolongation of <math>V\,</math> to a vector field on <math>J^{r}\pi\,</math>'''.
:Allow x = S and y = V and one gets
:<math>\left(\frac{\partial T}{\partial V}\right)_S = -\left(\frac{\partial P}{\partial S}\right)_V</math>


;Maxwell's second relation
These results are best understood when applied to a particular example. Hence, let us examine the following.
:Allow x = T and y = V and one gets
:<math>\left(\frac{\partial S}{\partial V}\right)_T = \left(\frac{\partial P}{\partial T}\right)_V</math>


;Maxwell's third relation
===Example===
:Allow x = S and y = P and one gets
:<math>\left(\frac{\partial T}{\partial P}\right)_S = \left(\frac{\partial V}{\partial S}\right)_P</math>


;Maxwell's fourth relation
Let us consider the case <math>(\mathcal{E},\pi,\mathcal{M})</math>, where <math>\mathcal{E} \simeq \mathbb{R}^{2}</math> and <math>\mathcal{M} \simeq \mathbb{R}</math>.
:Allow x = T and y = P and one gets
Then, <math>(J^{1}\pi, \pi, \mathcal{E})</math> defines the first jet bundle, and may be coordinated by <math>(x,u,u_{1})\,</math>, where
:<math>\left(\frac{\partial S}{\partial P}\right)_T = -\left(\frac{\partial V}{\partial T}\right)_P</math>


;Maxwell's fifth relation
:{|
:Allow x = P and y = V
|-
:<math>\left(\frac{\partial T}{\partial P}\right)_V \left(\frac{\partial S}{\partial V}\right)_P</math><math>-\left(\frac{\partial T}{\partial V}\right)_P \left(\frac{\partial S}{\partial P}\right)_V</math> = 1
|align=right|<math>x(j^{1}_{p}\sigma) \,</math>
|<math>= x(p) = x \,</math>
|-
|align=right|<math>u(j^{1}_{p}\sigma) \,</math>
|<math>= u(\sigma(p)) = u(\sigma(x)) = \sigma(x) \,</math>
|-
|align=right|<math>u_{1}(j^{1}_{p}\sigma) \,</math>
|<math>= \left.\frac{\partial \sigma}{\partial x}\right|_{p} = \dot{\sigma}(x) \,</math>
|-
|}
 
for all <math>p \in \mathcal{M}</math> and <math>\sigma \in \Gamma_{p}(\pi)\,</math>. A contact form on <math>J^{1}\pi\,</math> has the form
 
:<math>\theta = du - u_{1}dx \,</math>
 
Let us consider a vector <math>V\,</math> on <math>\mathcal{E}</math>, having the form
 
:<math>V = x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} \,</math>
 
Then, the first prolongation of this vector field to <math>J^{1}\pi\,</math> is
 
:{|
|-
|<math>V^{1} \,</math>
|<math>= V + Z \,</math>
|-
|
|<math>= x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + Z \,</math>
|-
|
|<math>= x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + \rho(x,u,u_{1})\frac{\partial}{\partial u_{1}} \,</math>
|-
|}
 
If we now take the Lie derivative of the contact form with respect to this prolonged vector field, <math>\mathcal{L}_{V^{1}}(\theta)\,</math>, we obtain


;Maxwell's sixth relation
:{|
:Allow x = T and y = S and one gets
|-
:<math>\left(\frac{\partial P}{\partial T}\right)_S \left(\frac{\partial V}{\partial S}\right)_T -\left(\frac{\partial P}{\partial S}\right)_T \left(\frac{\partial V}{\partial T}\right)_S</math> = 1
|<math>\mathcal{L}_{V^{1}}(\theta) \,</math>
|<math>= \mathcal{L}_{V^{1}}(du - u_{1}dx) \,</math>
|-
|
|<math>= \mathcal{L}_{V^{1}}du - (\mathcal{L}_{V^{1}}u_{1})dx - u_{1}(\mathcal{L}_{V^{1}}dx) \,</math>
|-
|
|<math>= d(V^{1}u) - V^{1}u_{1}dx - u_{1}d(V^{1}x) \,</math>
|-
|
|<math>= dx - \rho(x,u,u_{1})dx + u_{1}du \,</math>
|-
|
|<math>= (1 - \rho(x,u,u_{1}) )dx + u_{1}du \,</math>
|-
|}
|}


== General Maxwell relationships ==
But, we may identify <math>du = \theta + u_{1}dx\,</math>. Thus, we get


The above are not the only Maxwell relationships. When other work terms involving other natural variables besides the volume work are considered or when the [[Particle number|number of particles]] is included as a natural variable, other Maxwell relations become apparent. For example, if we have a single-component gas, then the number of particles ''N''&nbsp; is also a natural variable of the above four thermodynamic potentials. The Maxwell relationship for the enthalpy with respect to pressure and particle number would then be:
:{|
|-
|<math>\mathcal{L}_{V^{1}}(\theta) \,</math>
|<math>= [\,1 - \rho(x,u,u_{1})\,]dx + u_{1}(\theta + u_{1}dx) \,</math>
|-
|
|<math>= [\,1 + u_{1}u_{1} - \rho(x,u,u_{1})\,]dx + u_{1}\theta  \,</math>
|-
|}


:<math>
Hence, for <math>\mathcal{L}_{V^{1}}(\theta)\,</math> to preserve the contact ideal, we require
\left(\frac{\partial \mu}{\partial P}\right)_{S, N} =
\left(\frac{\partial V}{\partial N}\right)_{S, P}\qquad=
\frac{\partial^2 H }{\partial P \partial N}
</math>


where μ is the [[chemical potential]]. In addition, there are other thermodynamic potentials besides the four that are commonly used, and each of these potentials will yield a set of Maxwell relations.
:{|
|-
|
|<math>1 + u_{1}u_{1} - \rho(x,u,u_{1}) = 0 \,</math>
|-
|<math>\Longrightarrow \quad \,</math>
|<math>\rho(x,u,u_{1}) = 1 + u_{1}u_{1}\,</math>
|-
|}


Each equation can be re-expressed using the relationship
And so the first prolongation of <math>V\,</math> to a vector field on <math>J^{1}\pi\,</math> is


:<math>\left(\frac{\partial y}{\partial x}\right)_z
:<math> V^{1} = x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + (1 + u_{1}u_{1})\frac{\partial}{\partial u_{1}} \,</math>
=
 
1\left/\left(\frac{\partial x}{\partial y}\right)_z\right.</math>
Let us also calculate the second prolongation of <math>V\,</math> to a vector field on <math>J^{2}\pi\,</math>.
We have <math>\{x,u,u_{1}, y_{2}\}\,</math> as coordinates on <math>J^{2}\pi\,</math>. Hence, the prolonged vector has the form
 
:<math> V^{2} = x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + \rho(x,u,u_{1},u_{2})\frac{\partial}{\partial u_{1}} + \phi(x,u,u_{1},u_{2})\frac{\partial}{\partial u_{2}} \,</math>
 
The contacts forms are
 
:{|
|-
|align=right|<math>\theta \,</math>
|<math>= du - u_{1}dx \,</math>
|-
|<math>\theta_{1}  \,</math>
|<math>= du_{1} - u_{2}dx \,</math>
|-
|}
 
To preserve the contact ideal, we require
 
:{|
|-
|align=right|<math>\mathcal{L}_{V^{2}}(\theta) \,</math>
|<math>= 0\,</math>
|-
|<math>\mathcal{L}_{V^{2}}(\theta_{1}) \,</math>
|<math>= 0 \,</math>
|-
|}
 
Now, <math>\theta\,</math> has no <math>u_{2}\,</math> dependency. Hence, from this equation we will pick up the formula for <math>\rho\,</math>, which will necessarily be the same result as we found for <math>V^{1}\,</math>. Therefore, the problem is analogous to prolonging the vector field <math>V^{1}\,</math> to <math>J^{2}\pi\,</math>.
That is to say, we may generate the <math>r^{th}\,</math>-prolongation of a vector field by recursively applying the Lie derivative of the contact forms with respect to the prolonged vector fields, <math>r\,</math> times.
So, we have
 
:<math> \rho(x,u,u_{1}) = 1 + u_{1}u_{1} \,</math>
 
and so
 
:{|
|-
|<math>V^{2} \,</math>
|<math>= V^{1} + \phi(x,u,u_{1},u_{2})\frac{\partial}{\partial u_{2}} \,</math>
|-
|
|<math>= x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + (1 + u_{1}u_{1})\frac{\partial}{\partial u_{1}} + \phi(x,u,u_{1},u_{2})\frac{\partial}{\partial u_{2}} \,</math>
|-
|}
 
Therefore, the Lie derivative of the second contact form with respect to <math>V^{2}\,</math> is
 
:{|
|-
|<math>\mathcal{L}_{V^{2}}(\theta_{1}) \,</math>
|<math>= \mathcal{L}_{V^{2}}(du_{1} - u_{2}dx) \,</math>
|-
|
|<math>= \mathcal{L}_{V^{2}}du_{1} - (\mathcal{L}_{V^{2}}u_{2})dx - u_{2}(\mathcal{L}_{V^{2}}dx) \,</math>
|-
|
|<math>= d(V^{2}u_{1}) - V^{2}u_{2}dx - u_{2}d(V^{2}x) \,</math>
|-
|
|<math>= d(1-u_{1}u_{1}) - \phi(x,u,u_{1},u_{2})dx + u_{2}du \,</math>
|-
|
|<math>= 2u_{1}du_{1} - \phi(x,u,u_{1},u_{2})dx + u_{2}du \,</math>
|-
|}
 
Again, let us identify <math>du=\theta + u_{1}dx \,</math> and <math>du_{1}=\theta_{1} + u_{2}dx \,</math>. Then we have
 
:{|
|-
|<math>\mathcal{L}_{V^{2}}(\theta_{1}) \,</math>
|<math>= 2u_{1}(\theta_{1} + u_{2}dx) - \phi(x,u,u_{1},u_{2})dx + u_{2}(\theta + u_{1}dx) \,</math>
|-
|
|<math>= [\, 3u_{1}u_{2} - \phi(x,u,u_{1},u_{2})\,]dx + u_{2}\theta + 2u_{1}\theta_{1} \,</math>
|-
|}
 
Hence, for <math>\mathcal{L}_{V^{2}}(\theta_{1})\,</math> to preserve the contact ideal, we require
 
:{|
|-
|
|<math>3u_{1}u_{2} - \phi(x,u,u_{1},u_{2}) = 0 \,</math>
|-
|<math>\Longrightarrow \quad \,</math>
|<math>\phi(x,u,u_{1},u_{2}) = 3u_{1}u_{2} \,</math>
|-
|}
 
And so the second prolongation of <math>V\,</math> to a vector field on <math>J^{2}\pi\,</math> is
 
:<math> V^{2} = x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + (1 + u_{1}u_{1})\frac{\partial}{\partial u_{1}} + 3u_{1}u_{2}\frac{\partial}{\partial u_{2}} \, </math>
 
Note that the first prolongation of <math>V\,</math> can be recovered by omitting the second derivative terms in <math>V^{2}\,</math>, or by projecting back to <math>J^{1}\pi\,</math>.
 
==Infinite Jet Spaces==
The [[inverse limit]] of the sequence of projections <math>\pi_{k+1,k}:J^{k+1}(\pi)\to J^k(\pi)</math> gives rise to the '''infinite jet space''' <math>J^\infty(\pi)</math>.  A point <math>j_p^\infty(\sigma)</math> is the equivalence class of sections of <math>\pi</math> that have  the same <math>k</math>-jet in <math>p</math> as <math>\sigma</math>  for all values of <math>k</math>. The natural projection <math>\pi_\infty</math> maps  <math>j_p^\infty(\sigma)</math> into <math>p</math>.
 
Just by thinking in terms of coordinates, <math>J^\infty(\pi)</math> appears to be an infinite-dimensional geometric object. In fact, the simplest way of introducing a differentiable structure on <math>J^\infty(\pi)</math>, not relying on differentiable charts, is given by the [[differential calculus over commutative algebras]]. Dual to the sequence of projections <math>\pi_{k+1,k}:J^{k+1}(\pi)\to J^k(\pi)</math> of manifolds is the sequence of injections
<math>\pi_{k+1,k}^*:C^\infty(J^{k}(\pi))\to C^\infty(J^{k+1}(\pi))</math>
of commutative algebras. Let's denote <math>C^\infty(J^{k}(\pi))</math> simply by <math>\mathcal{F}_k(\pi)</math>. Take now the [[direct limit]] <math>\mathcal{F}(\pi)</math> of the <math>\mathcal{F}_k(\pi)</math>'s. It will be a commutative algebra, which can be assumed to be the smooth functions algebra over the geometric object <math>J^\infty(\pi)</math>. Observe that <math>\mathcal{F}(\pi)</math>, being born as a direct limit, carries an additional structure: it is a filtered commutative algebra.
 
Roughly speaking, a concrete element <math>\varphi\in\mathcal{F}(\pi)</math> will always belong to some <math>\mathcal{F}_k(\pi)</math>, so it is a smooth function on the finite-dimensional manifold <math>J^k(\pi)</math> in the usual sense.
 
===Infinitely prolonged PDE's===
Given a <math>k</math>-th order system of PDE's <math>\mathcal{E}\subseteq J^k(\pi)</math>, the collection <math>I(\mathcal{E})</math> of vanishing on <math>\mathcal{E}</math> smooth functions on <math>J^\infty(\pi)</math> is an [[ideal]] in the algebra <math>\mathcal{F}_k(\pi)</math>, and hence in the direct limit <math>\mathcal{F}(\pi)</math> too.
 
Enhance <math>I(\mathcal{E})</math> by adding all the possible compositions of [[total derivative]]s applied to all its elements. This way we get a new ideal <math>I</math> of  <math>\mathcal{F}(\pi)</math> which is now closed under the operation of taking total derivative. The submanifold <math>\mathcal{E}_{(\infty)}</math> of <math>J^\infty(\pi)</math> cut out by <math>I</math> is called the '''infinite prolongation''' of <math>\mathcal{E}</math>.
 
Geometrically, <math>\mathcal{E}_{(\infty)}</math> is the manifold of '''formal solutions''' of <math>\mathcal{E}</math>. A point <math>j_p^\infty(\sigma)</math> of  <math>\mathcal{E}_{(\infty)}</math> can be easily seen to be represented by a section <math>\sigma</math> whose <math>k</math>-jet's graph is tangent to <math>\mathcal{E}</math> at the point <math>j_p^k(\sigma)</math> with arbitrarily high order of tangency.
 
Analytically, if <math>\mathcal{E}</math> is given by <math>\varphi=0</math>, a formal solution can be understood as the set of Taylor coefficients of a section <math>\sigma</math> in a point <math>p</math> that make vanish the [[Taylor series]] of <math>\varphi\circ j^k(\sigma)</math> at the point <math>p</math>.
 
Most importantly, the closure properties of  <math>I</math> imply that <math>\mathcal{E}_{(\infty)}</math> is tangent to the '''infinite-order contact structure''' <math>\mathcal{C}</math> on <math>J^\infty(\pi)</math>, so that by restricting <math>\mathcal{C}</math> to <math>\mathcal{E}_{(\infty)}</math> one gets the [[diffiety]] <math>(\mathcal{E}_{(\infty)},\mathcal{C}|_{\mathcal{E}_{(\infty)}})</math>, and can study the associated [[C-spectral sequence]].
 
==Remark==
 
This article has defined jets of local sections of a bundle, but it is possible to define jets of functions <math>f:\mathcal{M} \longrightarrow \mathcal{N}\,</math>, where <math>\mathcal{M}</math> and <math>\mathcal{N}</math> are manifolds; the jet of <math>f\,</math> then just corresponds to the jet of the section
 
:{|
|-
|<math>gr_{f}:\mathcal{M} \,</math>
|<math>\longrightarrow \mathcal{M} \times \mathcal{N} \,</math>
|-
|align=right|<math>p \,</math>
|<math>\longmapsto gr_{f}(p) = (p, f(p) )\,</math>
|-
|}


which are sometimes also known as Maxwell relations.
(<math>gr_{f}\,</math> is known as the '''graph of the function <math>f\,</math>''') of the trivial bundle <math>(\mathcal{M} \times \mathcal{N}, \pi_{1}, \mathcal{M})</math>. However, this restriction does not simplify the theory, as the global triviality of <math>\pi\,</math> does not imply the global triviality of <math>\pi_{1}\,</math>.


== See also ==
== See also ==
* [[Table of thermodynamic equations]]
* [[Jet group]]
* [[Thermodynamic equations]]
* [[Jet (mathematics)]]


==External links==
==References==
*http://theory.ph.man.ac.uk/~judith/stat_therm/node48.html a partial derivation of Maxwell's relations


* Ehresmann, C., "Introduction à la théorie des structures infinitésimales et des pseudo-groupes de Lie."  ''Geometrie Differentielle,'' Colloq. Inter. du Centre Nat. de la Recherche Scientifique, Strasbourg, 1953, 97-127.
* Kolář, I., Michor, P., Slovák, J., ''[http://www.emis.de/monographs/KSM/ Natural operations in differential geometry.]''  Springer-Verlag: Berlin Heidelberg, 1993.  ISBN 3-540-56235-4, ISBN 0-387-56235-4.
* Saunders, D. J., "The Geometry of Jet Bundles", Cambridge University Press, 1989, ISBN 0-521-36948-7
* Krasil'shchik, I. S., Vinogradov, A. M., [et al.], "Symmetries and conservation laws for differential equations of mathematical physics", Amer. Math. Soc., Providence, RI, 1999, ISBN 0-8218-0958-X.
* Olver, P. J., "Equivalence, Invariants and Symmetry", Cambridge University Press, 1995, ISBN 0-521-47811-1
* Giachetta, G., Mangiarotti, L., [[Gennadi Sardanashvily|Sardanashvily, G.]], "Advanced Classical Field Theory", World Scientific, 2009, ISBN 978-981-283-895-7
* [[Gennadi Sardanashvily|Sardanashvily, G.]], Fibre bundles, jet manifolds and Lagrangian theory. Lectures for theoreticians, [http://xxx.lanl.gov/abs/0908.1886 arXiv: 0908.1886]


[[Category:Thermodynamics]]
[[Category:Differential topology]]
[[Category:Concepts in physics]]
[[Category:Differential equations]]
[[Category:James Clerk Maxwell]]
[[Category:Fiber bundles]]
[[Category:Thermodynamic equations]]


[[ar:علاقات ماكسويل]]
[[zh:节丛]]
[[ca:Relacions de Maxwell]]
[[cs:Maxwellovy relace]]
[[de:Maxwell-Beziehung]]
[[fa:روابط ماکسول (ترمودینامیک)]]
[[fr:Relations de Maxwell]]
[[ko:맥스웰 관계식]]
[[it:Relazioni di Maxwell]]
[[he:קשרי מקסוול]]
[[ja:マクスウェルの関係式]]
[[pl:Relacje Maxwella]]
[[sv:Maxwells relationer]]
[[th:ความสัมพันธ์ของแมกซ์เวลล์]]
[[zh:麦克斯韦关系式]]

Revision as of 01:14, 13 August 2014

Library Technician Anton from Strathroy, has many passions that include r/c helicopters, property developers in condo new launch singapore and coin collecting. Finds the beauty in planing a trip to spots around the globe, recently only returning from Old Town of Corfu.

In differential geometry, the jet bundle is a certain construction which makes a new smooth fiber bundle out of a given smooth fiber bundle. It makes it possible to write differential equations on sections of a fiber bundle in an invariant form. Jets may also be seen as the coordinate free versions of Taylor expansions.

Historically, jet bundles are attributed to Ehresmann, and were an advance on the method (prolongation) of Élie Cartan, of dealing geometrically with higher derivatives, by imposing differential form conditions on newly-introduced formal variables. Jet bundles are sometimes called sprays, although sprays usually refer more specifically to the associated vector field induced on the corresponding bundle (e.g., the geodesic spray on Finsler manifolds.)

More recently, jet bundles have appeared as a concise way to describe phenomena associated with the derivatives of maps, particularly those associated with the calculus of variations. Consequently, the jet bundle is now recognized as the correct domain for a geometrical covariant field theory and much work is done in general relativistic formulations of fields using this approach.

Jets

Mining Engineer (Excluding Oil ) Truman from Alma, loves to spend time knotting, largest property developers in singapore developers in singapore and stamp collecting. Recently had a family visit to Urnes Stave Church.

Let (,π,) be a fiber bundle in a category of manifolds and let p, with dim=m. Let Γ(π) denote the set of all local sections whose domain contains p. Let I=(I(1),I(2),,I(m)) be a multi-index (an ordered m-tuple of integers), then

|I|:=i=1mI(i)
|I|xI:=i=1m(xi)I(i).

Define the local sections σ,ηΓ(π) to have the same r-jet at p if

|I|σαxI|p=|I|ηαxI|p,0|I|r.

The relation that two maps have the same r-jet is an equivalence relation. An r-jet is an equivalence class under this relation, and the r-jet with representative σ is denoted jprσ. The integer r is also called the order of the jet.

p is the source of jprσ.

σ(p) is the target of jprσ.

Jet manifolds

The rth jet manifold of π is the set

{jprσ:p,σΓ(π)}

and is denoted Jrπ. We may define projections πr and πr,0 called the source and target projections respectively, by

πr:Jrπ
jprσ p
πr,0:Jrπ
jprσ σ(p)

If 1kr, then the k-jet projection is the function πr,k defined by

πr,k:Jrπ Jkπ
jprσ jpkσ

From this definition, it is clear that πr=ππr,0 and that if 0mk, then πr,m=πk,mπr,k. It is conventional to regard πr,r=idJrπ, the identity map on Jrπ and to identify J0π with .

The functions πr,k,πr,0 and πr are smooth surjective submersions.

A coordinate system on will generate a coordinate system on Jrπ. Let (U,u) be an adapted coordinate chart on , where u=(xi,uα). The induced coordinate chart (Ur,ur) on Jrπ is defined by

Ur ={jprσ:σ(p)U}
ur =(xi,uα,uIα)

where

xi(jprσ) =xi(p)
uα(jprσ) =uα(σ(p))

and the n(m+rCr1) functions

uIα:Uk

are specified by

uIα(jprσ)=|I|σαxI|p

and are known as the derivative coordinates.

Given an atlas of adapted charts (U,u) on , the corresponding collection of charts (Ur,ur) is a finite-dimensional C atlas on Jrπ.

Jet bundles

Since the atlas on each Jrπ defines a manifold, the triples (Jrπ,πr,k,Jkπ),(Jrπ,πr,0,) and (Jrπ,πr,) all define fibered manifolds. In particular, if (,π,) is a fiber bundle, the triple (Jrπ,πr,) defines the rth jet bundle of π.

If W is an open submanifold, then

Jr(π|π1(W))πr1(W).

If p, then the fiber πr1(p) is denoted Jprπ.

Let σ be a local section of π with domain W. The rth jet prolongation of σ is the map jrσ:WJrπ defined by

(jrσ)(p)=jprσ.

Note that πrjrσ=idW, so jrσ really is a section. In local coordinates, jrσ is given by

(σα,|I|σαx|I|)1|I|r.

We identify j0σ with σ.

Example

If π is the trivial bundle (×,pr1,), then there is a canonical diffeomorphism between the first jet bundle J1π and T×. To construct this diffeomorphism, for each σΓM(π) write σ¯=pr2σC(M).

Then, whenever pM

jp1σ={ψ:ψΓp(π);ψ¯(p)=σ¯(p);dψ¯p=dσ¯p}.

Consequently, the mapping

J1π T×
jp1σ (dσ¯p,σ¯(p))

is well-defined and is clearly injective. Writing it out in coordinates shows that it is a diffeomorphism, because if (xi,u) are coordinates on ×, where u=id is the identity coordinate, then the derivative coordinates ui on J1π correspond to the coordinates i on T.

Likewise, if π is the trivial bundle (×,pr1,), then there exists a canonical diffeomorphism between J1π and ×T

Contact forms

A differential 1-form θ on the space Jrπ is called a contact form (i.e. θΛCrπ) if it is pulled back to the zero form on by all prolongations. In other words, if θΛ1Jr+1π, then θΛC1πr+1,r if and only if, for every open submanifold W and every σΓW(π),

(jk+1σ)θ=0.

The distribution on Jrπ generated by the contact forms is called the Cartan distribution. It is the main geometrical structure on jet spaces and plays an important role in the geometric theory of partial differential equations. The Cartan distributions are not involutive and are of growing dimension when passing to higher order jet spaces. Surprisingly though, when passing to the space of infinite order jets J this distribution is involutive and finite dimensional. Its dimension coinciding with the dimension of the base manifold .

Example

Let us consider the case (,π,), where 2 and . Then, (J1π,π,) defines the first jet bundle, and may be coordinated by (x,u,u1), where

x(jp1σ) =x(p)=x
u(jp1σ) =u(σ(p))=u(σ(x))=σ(x)
u1(jp1σ) =σx|p=σ(x)

for all p and σΓp(π). A general 1-form on J1π takes the form

θ=a(x,u,u1)dx+b(x,u,u1)du+c(x,u,u1)du1

A section σΓp(π) has first prolongation j1σ=(u,u1)=(σ(p),σx|p). Hence, (j1σ)θ can be calculated as

(jp1σ)θ =θjp1σ
=a(x,σ(x),σ(x))dx+b(x,σ(x),σ(x))d(σ(x))+c(x,σ(x),σ(x))d(σ(x))
=a(x,σ(x),σ(x))dx+b(x,σ(x),σ(x))σ(x)dx+c(x,σ(x),σ(x))σ(x)dx
=[a(x,σ(x),σ(x))+b(x,σ(x),σ(x))σ(x)+c(x,σ(x),σ(x))σ(x)]dx

This will vanish for all sections σ if and only if c=0 and a=bσ(x). Hence, θ=b(x,u,u1)θ0 must necessarily be a multiple of the basic contact form θ0=duu1dx. Proceeding to the second jet space J2π with additional coordinate u2, such that

u2(jp2σ)=2σx2|p=σ(x)

a general 1-form has the construction

θ=a(x,u,u1,u2)dx+b(x,u,u1,u2)du+c(x,u,u1,u2)du1+e(x,u,u1,u2)du2

This is a contact form if and only if

(jp2σ)θ =θjp2σ
=a(x,σ(x),σ(x),σ(x))dx+b(x,σ(x),σ(x),σ(x))d(σ(x))+
+c(x,σ(x),σ(x),σ(x))d(σ(x))+e(x,σ(x),σ(x),σ(x))d(σ(x))
=adx+bσ(x)dx+cσ(x)dx+eσ(x)dx
=[a+bσ(x)+cσ(x)+eσ(x)]dx
=0

which implies that e=0 and a=bσ(x)cσ(x). Therefore, θ is a contact form if and only if

θ=b(x,σ(x),σ(x))θ0+c(x,σ(x),σ(x))θ1

where θ1=du1u2dx is the next basic contact form (Note that here we are identifying the form θ0 with its pull-back (π2,1)θ0 to J2π).

In general, providing x,u,, a contact form on Jr+1π can be written as a linear combination of the basic contact forms

θk=dukuk+1dxk=0,,r1

where uk(jkσ)=kσxk|p.

Similar arguments lead to a complete characterization of all contact forms.

In local coordinates, every contact one-form on Jr+1π can be written as a linear combination

θ=|I|=0rPαIθIα

with smooth coefficients PIα(xi,uα) of the basic contact forms

θIα=duIαuI,iαdxi

|I| is known as the order of the contact form θIα. Note that contact forms on Jr+1π have orders at most r. Contact forms provide a characterization of those local sections of πr+1 which are prolongations of sections of π.

Let ψΓW(πr+1), then ψ=jr+1σ where σΓW(π) if and only if ψ(θ|W)=0,θΛC1πr+1,r.

Vector fields

A general vector field on the total space , coordinated by (x,u) =def (xi,uα), is

V =def ρi(x,u)xi+ϕα(x,u)uα.

A vector field is called horizontal, meaning all the vertical coefficients vanish, if ϕα=0.

A vector field is called vertical, meaning all the horizontal coefficients vanish, if ρi=0.

For fixed (x,u), we identify

V(xu) =def ρi(x,u)xi+ϕα(x,u)uα

having coordinates (x,u,ρi,ϕα), with an element in the fiber Txu of T over (x,u), called a tangent vector in T. A section

ψ: T
(x,u) ψ(x,u)=V

is called a vector field on with V=ρi(x,u)xi+ϕα(x,u)uα and ψΓ(T).

The jet bundle Jrπ is coordinated by (x,u,w) =def (xi,uα,wiα). For fixed (x,u,w), identify

V(xuw) =def  Vi(x,u,w)xi+Vα(x,u,w)uα + Viα(x,u,w)wiα+
+ Vi1i2α(x,u,w)wi1i2α+ + +Vi1i2irα(x,u,w)wi1i2irα

having coordinates (x,u,w,viα,vi1i2α,,vi1i2irα), with an element in the fiber Txuw(Jrπ) of T(Jrπ) over (x,u,w)Jrπ, called a tangent vector in T(Jrπ). Here, viα,vi1i2α,,vi1i2irα are real-valued functions on Jrπ. A section

Ψ:Jrπ T(Jrπ)
(x,u,w) Ψ(u,w)=V

is a vector field on Jrπ, and we say ΨΓ(T(Jrπ)).

Partial differential equations

Let (,π,) be a fiber bundle. An rth order partial differential equation on π is a closedTemplate:Dn embedded submanifold 𝒮 of the jet manifold Jrπ. A solution is a local section σΓW(π) satisfying jprσ𝒮,p.

Let us consider an example of a first order partial differential equation.

Example

Let π be the trivial bundle (2×,pr1,2) with global coordinates (x1,x2,u1). Then the map F:J1π defined by

F=u11u212x2u1

gives rise to the differential equation

S={jp1σJ1π:(u11u212x2u1)(jp1σ)=0}

which can be written

σx1σx22x2σ=0.

The particular section σ:22× defined by

σ(p1,p2)=(p1,p2,p1(p2)2)

has first prolongation given by

j1σ(p1,p2)=(p1,p2,p1(p2)2,(p2)2,2p1p2)

and is a solution of this differential equation, because

(u11u212x2u1)(jp1σ) =u11(jp1σ)u21(jp1σ)2x2(jp1σ)u1(jp1σ)
=(p2)22p1p22p2p1(p2)2
=2p1(p2)32p1(p2)3
=0

and so jp1σ𝒮 for every p2.

Jet Prolongation

A local diffeomorphism ψ:JrπJrπ defines a contact transformation of order r if it preserves the contact ideal, meaning that if θ is any contact form on Jrπ, then ψθ is also a contact form.

The flow generated by a vector field Vr on the jet space Jr forms a one-parameter group of contact transformations if and only if the Lie derivative Vr(θ) of any contact form θ preserves the contact ideal.

Let us begin with the first order case. Consider a general vector field V1 on J1π, given by

V1 =def ρi(u1)xi+ϕα(u1)uα+χiα(u1)uiα.

We now apply V1 to the basic contact forms θα=duαuiαdxi, and obtain

V1(θα) =V1(duαuiαdxi)
=V1duα(V1uiα)dxiuiα(V1dxi)
=d(V1uα)V1uiαdxiuiαd(V1xi)
=dϕαχiαdxiuiαdρi
=ϕαxidxi+ϕαukduk+ϕαuikduikχiαdxiuiα[ρixmdxm+ρiukduk+ρiumkdumk]

where we have expanded the exterior derivative of the functions in terms of their coordinates. Next, we note that

θk=dukuikdxiduk=θk+uikdxi

and so we may write

V1(θα) =ϕαxidxi+ϕαuk(θk+uikdxi)+ϕαuikduikχiαdxi
ulα[ρlxidxi+ρluk(θk+uikdxi)+ρluikduik]
=[ϕαxi+ϕαukuikulα(ρlxi+ρlukuik)χiα]dxi+[ϕαuikulαρluik]duik+
+(ϕαukulαρluk)θk.

Therefore, V1 determines a contact transformation if and only if the coefficients of dxi and duik in the formula vanish. The latter requirements imply the contact conditions

ϕαuikulαρluik=0

The former requirements provide explicit formulae for the coefficients of the first derivative terms in V1:

χiα=D^iϕαulα(D^iρl) where D^i=xi+uikuk

denotes the zeroth order truncation of the total derivative Di.

Thus, the contact conditions uniquely prescribe the prolongation of any point or contact vector field. That is, if Vr satisfies these equations, Vr is called the rth prolongation of V to a vector field on Jrπ.

These results are best understood when applied to a particular example. Hence, let us examine the following.

Example

Let us consider the case (,π,), where 2 and . Then, (J1π,π,) defines the first jet bundle, and may be coordinated by (x,u,u1), where

x(jp1σ) =x(p)=x
u(jp1σ) =u(σ(p))=u(σ(x))=σ(x)
u1(jp1σ) =σx|p=σ˙(x)

for all p and σΓp(π). A contact form on J1π has the form

θ=duu1dx

Let us consider a vector V on , having the form

V=xuux

Then, the first prolongation of this vector field to J1π is

V1 =V+Z
=xuux+Z
=xuux+ρ(x,u,u1)u1

If we now take the Lie derivative of the contact form with respect to this prolonged vector field, V1(θ), we obtain

V1(θ) =V1(duu1dx)
=V1du(V1u1)dxu1(V1dx)
=d(V1u)V1u1dxu1d(V1x)
=dxρ(x,u,u1)dx+u1du
=(1ρ(x,u,u1))dx+u1du

But, we may identify du=θ+u1dx. Thus, we get

V1(θ) =[1ρ(x,u,u1)]dx+u1(θ+u1dx)
=[1+u1u1ρ(x,u,u1)]dx+u1θ

Hence, for V1(θ) to preserve the contact ideal, we require

1+u1u1ρ(x,u,u1)=0
ρ(x,u,u1)=1+u1u1

And so the first prolongation of V to a vector field on J1π is

V1=xuux+(1+u1u1)u1

Let us also calculate the second prolongation of V to a vector field on J2π. We have {x,u,u1,y2} as coordinates on J2π. Hence, the prolonged vector has the form

V2=xuux+ρ(x,u,u1,u2)u1+ϕ(x,u,u1,u2)u2

The contacts forms are

θ =duu1dx
θ1 =du1u2dx

To preserve the contact ideal, we require

V2(θ) =0
V2(θ1) =0

Now, θ has no u2 dependency. Hence, from this equation we will pick up the formula for ρ, which will necessarily be the same result as we found for V1. Therefore, the problem is analogous to prolonging the vector field V1 to J2π. That is to say, we may generate the rth-prolongation of a vector field by recursively applying the Lie derivative of the contact forms with respect to the prolonged vector fields, r times. So, we have

ρ(x,u,u1)=1+u1u1

and so

V2 =V1+ϕ(x,u,u1,u2)u2
=xuux+(1+u1u1)u1+ϕ(x,u,u1,u2)u2

Therefore, the Lie derivative of the second contact form with respect to V2 is

V2(θ1) =V2(du1u2dx)
=V2du1(V2u2)dxu2(V2dx)
=d(V2u1)V2u2dxu2d(V2x)
=d(1u1u1)ϕ(x,u,u1,u2)dx+u2du
=2u1du1ϕ(x,u,u1,u2)dx+u2du

Again, let us identify du=θ+u1dx and du1=θ1+u2dx. Then we have

V2(θ1) =2u1(θ1+u2dx)ϕ(x,u,u1,u2)dx+u2(θ+u1dx)
=[3u1u2ϕ(x,u,u1,u2)]dx+u2θ+2u1θ1

Hence, for V2(θ1) to preserve the contact ideal, we require

3u1u2ϕ(x,u,u1,u2)=0
ϕ(x,u,u1,u2)=3u1u2

And so the second prolongation of V to a vector field on J2π is

V2=xuux+(1+u1u1)u1+3u1u2u2

Note that the first prolongation of V can be recovered by omitting the second derivative terms in V2, or by projecting back to J1π.

Infinite Jet Spaces

The inverse limit of the sequence of projections πk+1,k:Jk+1(π)Jk(π) gives rise to the infinite jet space J(π). A point jp(σ) is the equivalence class of sections of π that have the same k-jet in p as σ for all values of k. The natural projection π maps jp(σ) into p.

Just by thinking in terms of coordinates, J(π) appears to be an infinite-dimensional geometric object. In fact, the simplest way of introducing a differentiable structure on J(π), not relying on differentiable charts, is given by the differential calculus over commutative algebras. Dual to the sequence of projections πk+1,k:Jk+1(π)Jk(π) of manifolds is the sequence of injections πk+1,k:C(Jk(π))C(Jk+1(π)) of commutative algebras. Let's denote C(Jk(π)) simply by k(π). Take now the direct limit (π) of the k(π)'s. It will be a commutative algebra, which can be assumed to be the smooth functions algebra over the geometric object J(π). Observe that (π), being born as a direct limit, carries an additional structure: it is a filtered commutative algebra.

Roughly speaking, a concrete element φ(π) will always belong to some k(π), so it is a smooth function on the finite-dimensional manifold Jk(π) in the usual sense.

Infinitely prolonged PDE's

Given a k-th order system of PDE's Jk(π), the collection I() of vanishing on smooth functions on J(π) is an ideal in the algebra k(π), and hence in the direct limit (π) too.

Enhance I() by adding all the possible compositions of total derivatives applied to all its elements. This way we get a new ideal I of (π) which is now closed under the operation of taking total derivative. The submanifold () of J(π) cut out by I is called the infinite prolongation of .

Geometrically, () is the manifold of formal solutions of . A point jp(σ) of () can be easily seen to be represented by a section σ whose k-jet's graph is tangent to at the point jpk(σ) with arbitrarily high order of tangency.

Analytically, if is given by φ=0, a formal solution can be understood as the set of Taylor coefficients of a section σ in a point p that make vanish the Taylor series of φjk(σ) at the point p.

Most importantly, the closure properties of I imply that () is tangent to the infinite-order contact structure 𝒞 on J(π), so that by restricting 𝒞 to () one gets the diffiety ((),𝒞|()), and can study the associated C-spectral sequence.

Remark

This article has defined jets of local sections of a bundle, but it is possible to define jets of functions f:𝒩, where and 𝒩 are manifolds; the jet of f then just corresponds to the jet of the section

grf: ×𝒩
p grf(p)=(p,f(p))

(grf is known as the graph of the function f) of the trivial bundle (×𝒩,π1,). However, this restriction does not simplify the theory, as the global triviality of π does not imply the global triviality of π1.

See also

References

  • Ehresmann, C., "Introduction à la théorie des structures infinitésimales et des pseudo-groupes de Lie." Geometrie Differentielle, Colloq. Inter. du Centre Nat. de la Recherche Scientifique, Strasbourg, 1953, 97-127.
  • Kolář, I., Michor, P., Slovák, J., Natural operations in differential geometry. Springer-Verlag: Berlin Heidelberg, 1993. ISBN 3-540-56235-4, ISBN 0-387-56235-4.
  • Saunders, D. J., "The Geometry of Jet Bundles", Cambridge University Press, 1989, ISBN 0-521-36948-7
  • Krasil'shchik, I. S., Vinogradov, A. M., [et al.], "Symmetries and conservation laws for differential equations of mathematical physics", Amer. Math. Soc., Providence, RI, 1999, ISBN 0-8218-0958-X.
  • Olver, P. J., "Equivalence, Invariants and Symmetry", Cambridge University Press, 1995, ISBN 0-521-47811-1
  • Giachetta, G., Mangiarotti, L., Sardanashvily, G., "Advanced Classical Field Theory", World Scientific, 2009, ISBN 978-981-283-895-7
  • Sardanashvily, G., Fibre bundles, jet manifolds and Lagrangian theory. Lectures for theoreticians, arXiv: 0908.1886

zh:节丛