0
$\begingroup$

I'm trying to calculate the eigenvalues of the following system which originates from here (Appendix B): \begin{align} i{\xi}_{0}\hat{\tilde{\gamma}}_{ij} &= 2i\tilde{\gamma}_{k(i}\xi_{j)}\hat{\beta}^{k} - 2\alpha \hat{\tilde{A}}_{ij} - \frac{2i}{3}\tilde{\gamma}_{ij}\xi_{k}\hat{\beta}^{k}\,, \\ i{\xi}_{0}\hat{\chi}&=\frac{2}{3}\chi\left(\alpha\hat{K}-i\xi_{k}\hat{\beta}^{k}\right) \, , \\ i{\xi}_{0}\hat{\phi} &= -\alpha \hat{K}_{\phi} \,, \\ i{\xi}_{0}\hat{K} &= \hat{\alpha} + \frac{3b(x)\alpha}{4(1+b(x))}\left(\frac{\xi^{l}\xi^{k}\hat{\tilde{\gamma}}_{kl}}{\chi} - \frac{\tilde{\gamma}^{jk}\hat{\tilde{\gamma}}_{jk}}{2}\right) + i\alpha\chi\xi_{i}\hat{\hat{\Gamma}}^{i} - \left(\frac{2}{\chi}\hat{\chi}-\frac{\tilde{\gamma}^{ij}\hat{\tilde{\gamma}}_{ij}}{2}\right)\frac{\alpha(4+b(x))}{4(1+b(x))}\,, \\ i{\xi}_{0}\hat{K}_{\phi} &= - \alpha\hat{\phi}\,, \\ i{\xi}_{0}\hat{\Theta} &= - \frac{\alpha}{2(1+b(x))}\left(\frac{2}{\chi}\hat{\chi}-\frac{\tilde{\gamma}^{ij}\hat{\tilde{\gamma}}_{ij}}{2}\right) + \frac{i\alpha\chi\xi_{i}\hat{\hat{\Gamma}}^{i}}{2} + \frac{b(x)\alpha}{2(1+b(x))}\left(\frac{\xi^{l}\xi^{k}\hat{\tilde{\gamma}}_{kl}}{\chi} - \frac{\tilde{\gamma}^{jk}\hat{\tilde{\gamma}}_{jk}}{2}\right) \,,\\ i{\xi}_{0}\hat{\tilde{A}}_{ij} &= \left(\xi_{i}\xi_{j}-\frac{1}{3}\frac{\tilde{\gamma}_{ij}}{\chi}\right)\left(\chi\hat{\alpha}-\frac{\alpha}{2}\hat{\chi}\right) + i\alpha\chi\left(\tilde{\gamma}_{k(i}\xi_{j)}\hat{\hat{\Gamma}}^{k} - \frac{\tilde{\gamma}_{ij}\xi_{k}\hat{\hat{\Gamma}}^{k}}{3}\right) + \frac{1}{2}\alpha\left(\hat{\tilde{\gamma}}_{ij}-\frac{\tilde{\gamma}_{ij}\tilde{\gamma}^{kl}\hat{\tilde{\gamma}}_{kl}}{3}\right) \,,\\ i{\xi}_{0}\hat{\hat{\Gamma}}^{i} &= \frac{i\alpha}{\chi(1+b(x))}\left(-\frac{4}{3}\xi^{i}\hat{K}-2\chi b(x)\xi_{j}\hat{\tilde{A}}^{ij}+2\xi^{i}\hat{\Theta}\right) - \frac{1}{\chi}\left(\hat{\beta}^{i}+\frac{1}{3}\xi^{i}\xi_{l}\hat{\beta}^{l}\right) \,,\\ i{\xi}_{0}\hat{\alpha} &= -\frac{2\alpha}{1+a(x)}\left(\hat{K}-2\hat{\Theta}\right) \,,\\ i{\xi}_{0}\hat{\beta}^{i} &= \frac{3}{4(1+a(x))}\hat{\hat{\Gamma}}^{i} - \frac{ia(x)}{1+a(x)}\alpha\xi^{i}\hat{\alpha} \,. \end{align} The right hand side gives in matrix notation: \begin{equation} \begin{pmatrix} 0&\hat{\tilde{\gamma}}_{ef} & \hat{\chi} & \hat{\tilde{A}}_{cd} & \hat{K} & \hat{\Theta} & \hat{\hat{\Gamma}}^{g} & \hat{\alpha} & \hat{\beta}^{h} & \hat{\phi} & \hat{K}_{\phi} \\ \hat{\tilde{\gamma}}_{ij} & 0 & 0 & -2\alpha\delta_{i}^{c}\delta_{j}^{d} & 0 & 0 & 0 & 0 & 2i\tilde{\gamma}_{h(i}\xi_{j)}-\frac{2i}{3}\tilde{\gamma}_{ij}\xi_{h} & 0 & 0 \\ \hat{\chi} & 0 & 0 & 0 & \frac{2}{3}\chi\alpha & 0 & 0 & 0 & -\frac{2i}{3}\chi\xi_{h} & 0 & 0 \\ \hat{\tilde{A}}_{ab} & \frac{1}{2}\alpha\left(\delta_{a}^{e}\delta_{b}^{f}-\frac{\tilde{\gamma}_{ab}\tilde{\gamma}^{ef}}{3}\right) & -\frac{\alpha}{2}(\xi_{a}\xi_{b}-\frac{1}{3}\frac{\tilde{\gamma}_{ab}}{\chi}) & 0 & 0 & 0 & i\alpha\chi\left(\tilde{\gamma}_{g(a}\xi_{b)}-\frac{\tilde{\gamma}_{ab}\xi_{g}}{3}\right) & \chi(\xi_{a}\xi_{b}-\frac{1}{3}\frac{\tilde{\gamma}_{ab}}{\chi}) & 0 & 0 & 0 \\ \hat{K} & \frac{3b(x)\alpha}{4(1+b(x))}\left(\frac{\xi^{e}\xi^{f}}{\chi}-\frac{\tilde{\gamma}^{ef}}{2}\right) + \frac{\tilde{\gamma}^{ef}}{2}\frac{\alpha(4+b(x))}{4(1+b(x))} & -\frac{2}{\chi}\frac{\alpha(4+b(x))}{4(1+b(x))} & 0 & 0 & 0 & i\alpha\chi\xi_{g} & 1 & 0 & 0 & 0 \\ \hat{\Theta} & \frac{\alpha}{2(1+b(x))}\frac{\tilde{\gamma}^{ef}}{2} + \frac{b(x)\alpha}{2(1+b(x))}\left(\frac{\xi^{e}\xi^{f}}{\chi}-\frac{\tilde{\gamma}^{ef}}{2}\right) & -\frac{\alpha}{2(1+b(x))}\frac{2}{\chi} & 0 & 0 & 0 & \frac{i\alpha\chi\xi_{g}}{2} & 0 & 0 & 0 & 0 \\ \hat{\hat{\Gamma}}^{k} & 0 & 0 & \frac{i\alpha}{\chi(1+b(x))}(-2\chi b(x)\xi_{d}\delta_{c}^{k}) & \frac{i\alpha}{\chi(1+b(x))}\frac{-4\xi^{k}}{3} & \frac{i\alpha}{\chi(1+b(x))}2\xi^{k} & 0 & 0 & -\frac{1}{\chi}\left(\delta^{k}_{h} + \frac{\xi^{k}\xi_{h}}{3}\right) & 0 & 0 \\ \hat{\alpha} & 0 & 0 & 0 & -\frac{2\alpha}{1+a(x)} & \frac{4\alpha}{1+a(x)} & 0 & 0 & 0 & 0 & 0 \\ \hat{\beta}^{l} & 0 & 0 & 0 & 0 & 0 & \frac{3}{4(1+a(x))}\delta^{l}_{g} & -\frac{ia(x)}{1+a(x)}\alpha\xi^{l} & 0 & 0 & 0 \\ \hat{\phi} & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & -\alpha \\ \hat{K}_{\phi} & 0 & 0 & 0 & 0 & 0 & 0 & 0 & 0 & -\alpha & 0 \end{pmatrix} \end{equation} I typed the variables such that it is easier to read. $\tilde{\gamma}_{ij}$ is the conformal metric with $\chi$ the conformal factor, $\xi_{i}$ is a arbitrary vector with unit vector with respect to the induced three metric $\gamma_{ij}$.

However, if I feed this matrix to Mathematica using the xAct package, I get an error saying that it found inhomogeneous indices which I don't know from where they originate.

How I implemented this in Mathematica is by doing the following.

packageList["xAct"] = {"xAct`xPerm`","xAct`xTensor`","xAct`xPert`","xAct`Invar`","xAct`xCoba`","xAct`SymManipulator`","xAct`xTras`","xAct`TexAct`","xAct`xPand`"}; System`$PrePrint = xAct`xTensor` ScreenDollarIndices; xAct`xTensor`$DefInfoQ = False; xAct`xTensor`$CovDFormat = "Prefix"; 

Next, we define a three-dimensional manifold $M$ and a metric $\gamma_{ij}$:

DefManifold[M, 3, IndexRange[a, m]]; DefMetric[+1, gamma[-a, -b], cd, SymbolOfCovD -> {"|", "D"}, PrintAs -> "\[Gamma]"]; 

On top of that, we define a conformal metric $\tilde{\gamma}_{ij} =\chi\gamma_{ij}$ which I define with a bar (instead of a tilde).

DefTensor[chi[], M, PrintAs -> "\[Chi]"] DefMetric[+1, gammabar[-a, -b], cdbar, {",", "\!\(\*OverscriptBox[\(D\), \(_\)]\)"}, ConformalTo -> {gamma[-a, -b], chi[]}, PrintAs -> "\!\(\*OverscriptBox[\(\[Gamma]\), \(_\)]\)"] DefTensor[\[Alpha][], M] DefTensor[\[Beta][a], M] DefScalarFunction[p, PrintAs -> "a"]; DefScalarFunction[q, PrintAs -> "b"]; DefTensor[x[], M]; 

As an illustration, we define the unit vector $\xi_{i}$ as follows, together with a rule stating that its length is $1$.

DefTensor[xi[a], M, PrintAs -> "\[Xi]"] xi /: xi[-a_] xi[a_] := 1 

We are now ready to define the matrix

mat1 = Table[Table[0, {10}], {10}]; mat1[[1, 3]] = -2 \[Alpha][] gammabar[-i, c] gammabar[-j, d](*delta[-i,c]delta[-j,d]*); mat1[[1, 8]] = 2 I Symmetrize[gammabar[-h, -i] xi[-j], {-i, -j}] - 2 I/3 gammabar[-i, -j] xi[-h]; mat1[[2, 4]] = 2/3 chi[] \[Alpha][]; mat1[[2, 8]] = -2 I/3 chi[] xi[-h]; mat1[[3, 1]] = 1/2 \[Alpha][] (delta[-a, e] delta[-b, f] - 1/3 gammabar[-a, -b] gammabar[e, f]); mat1[[3, 2]] = -\[Alpha][]/ 2 (xi[-a] xi[-b] - 1/(3 chi[]) gammabar[-a, -b]); mat1[[3, 6]] = I \[Alpha][] chi[] (Symmetrize[gammabar[-g, -a] xi[-b], {-a, -b}] - 1/3 gammabar[-a, -b] xi[-g]); mat1[[3, 7]] = chi[] (xi[-a] xi[-b] - 1/(3 chi[]) gammabar[-a, -b]); mat1[[4, 1]] = 3 q[x[]] \[Alpha][]/(4 (1 + q[x[]])) (1/chi[] xi[e] xi[f] - 1/2 gammabar[e, f]) + 1/2 gammabar[e, f] \[Alpha][] (4 + q[x[]])/(4 (1 + q[x[]])); mat1[[4, 2]] = -2/chi[] \[Alpha][] (4 + q[x[]])/(4 (1 + q[x[]])); mat1[[4, 6]] = I \[Alpha][] chi[] xi[-g]; mat1[[4, 7]] = 1; mat1[[5, 1]] = \[Alpha][]/(2 (1 + q[x[]])) gammabar[e, f]/2 + q[x[]] \[Alpha][]/(2 (1 + q[x[]])) (xi[e] xi[f]/chi[] - gammabar[e, f]/2); mat1[[5, 2]] = -\[Alpha][]/(2 (1 + q[x[]])) 2/chi[]; mat1[[5, 6]] = I \[Alpha][] chi[] xi[-g]/2; mat1[[6, 3]] = I \[Alpha][]/(chi[] (1 + q[x[]])) (-2 chi[] q[x[]] xi[-d] delta[ k, -c]); mat1[[6, 4]] = I \[Alpha][]/(chi[] (1 + q[x[]])) (-4/3 xi[k]); mat1[[6, 5]] = I \[Alpha][]/(chi[] (1 + q[x[]])) 2 xi[k]; mat1[[6, 8]] = -1/chi[] (delta[-h, k] + xi[k] xi[-h]/3); mat1[[7, 4]] = -2 \[Alpha][]/(1 + p[x[]]); mat1[[7, 5]] = 4 \[Alpha][]/(1 + p[x[]]); mat1[[8, 6]] = 3/(4 (1 + p[x[]])) delta[l, -g]; mat1[[8, 7]] = -I p[x[]]/(1 + p[x[]]) \[Alpha][] xi[l]; mat1[[9, 10]] = -\[Alpha][]; mat1[[9, 10]] = -\[Alpha][]; 

I get an error when I apply

Eigenvalues[mat1] 

Hence my question: is it possible to write down this matrix (and if its correct) and is this the correct way to do this?

$\endgroup$
17
  • $\begingroup$ The actual code that generated the error is likely easier to work with than the latex equations. Consider either giving the code for the matrix and dropping the xact part (which diminishes your chances of getting an answer since probably few people here work specifically with tensor algebra) or directing your question to be specific about xact output/notation and providing the code you used so far. $\endgroup$ Commented Dec 28, 2022 at 15:19
  • $\begingroup$ I have edited my question with some code, if it is not sufficient, let me know and I will add more. :-) $\endgroup$ Commented Dec 28, 2022 at 18:32
  • $\begingroup$ I get a "IsIndexOf::nouse: Attempting to apply IsIndexOf on q[x[]]" and a "Throw::nocatch: Uncaught Throw[ERROR[q[x[]]]] returned to top level error" when evaluating your code $\endgroup$ Commented Dec 28, 2022 at 18:49
  • $\begingroup$ Consider copy pasting the matrix $\endgroup$ Commented Dec 28, 2022 at 18:50
  • 1
    $\begingroup$ sorry in option 1 I meant will not affect the code in EigenValues $\endgroup$ Commented Dec 28, 2022 at 20:14

0

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.