# Tensor product notation [closed]

In the image there is a tensor product:

$$F_{\mu\nu}F^{\mu\nu}=2(B^2-\frac{E^2}{c^2})$$

It's about how this operation on the co- and contravariant field strength tensors can give one of the invariants of the electromagnetic field.

I've tried it and it's actually the double inner product, F_lower(row,column)*F^upper(column,row) summed over all rows and columns

$$F_{\mu \nu}F^{\nu \mu}$$

Is this the way I would write it with subscript summation?

How did it come?

-

## closed as too localized by Qmechanic♦Jan 30 '13 at 18:21

This question is unlikely to help any future visitors; it is only relevant to a small geographic area, a specific moment in time, or an extraordinarily narrow situation that is not generally applicable to the worldwide audience of the internet. For help making this question more broadly applicable, visit the help center.If this question can be reworded to fit the rules in the help center, please edit the question.

The difference between $F_{\mu \nu}F^{\mu \nu}$ and $F_{\mu \nu}F^{\nu \mu}$ is a sign because $F_{\mu \nu}$ is antisymmetric. – Qmechanic Dec 1 '12 at 10:27

You should be contracting the following two objects $$F_{\mu \nu}= \begin{pmatrix} 0 & E_x & E_y & E_z \\ -E_x & 0 & -B_z & B_y \\ -E_y & B_z & 0 & -B_x \\ -E_z & -B_y & B_x & 0 \end{pmatrix} \quad \text{and}\quad F^{\mu \nu} = \begin{pmatrix} 0 & -E_x & -E_y & -E_z \\ E_x & 0 & -B_z & B_y \\ E_y & B_z & 0 & -B_x \\ E_z & -B_y & B_x & 0 \end{pmatrix}$$ Some of the tensor components change signs when you move the indices around. Now you should be doing what was mentioned by Fabian $$\sum_{\mu=0}^{3}\sum_{\nu=0}^{3}F_{\mu \nu}F^{\mu \nu}=F_{00}F^{00}+F_{01}F^{01}+...+F_{33}F^{33}$$ As you can see, the electric field multiplication will come out with an overall minus, and the magnetic field will come out positive.

-

You can write it as $$\sum_{\mu=0}^3 \sum_{\nu=0}^3 F_{\mu\nu} F^{\mu\nu}.$$ It is usually not written because of lazyness and shortness of notation (this notation is called Einstein summation convention.

-
I just realized that two sources I was looking at have different definitions of the double inner product. For example, foamcfd.org/Nabla/guides/ProgrammersGuidese3.html and people.rit.edu/pnveme/EMEM851n/constitutive/tensors_rect.html Which one is it supposed to be? – Robert Dec 1 '12 at 8:09
I don't see a difference, if I compare (1.15) with `Double Dot Product'. In any case, it should have the form in my answer. – Fabian Dec 1 '12 at 8:25
I believe you, but on this picture people.rit.edu/pnveme/EMEM851n/constitutive/tensor_rect11.gif the formula flips the row and column on the second tensor. Do you know where that derivation comes from? – Robert Dec 1 '12 at 8:39
If the matrices are [ 1 -1; 1 1] and [1 -1; 1 1] then your equation gives A00B00+A01B01+A10B10+A11B11 = 4. The equation A00B00 + A01B10 + A10B01 + A11B11 = 1 -1 -1 +1 =0. The second equation is what's given in people.rit.edu/pnveme/EMEM851n/constitutive/tensor_rect11.gif – Robert Dec 1 '12 at 8:45
@Robert: there are different ways to contract tensors (i.e., make inner products). The second choice is written as $F_{\mu\nu} F^{\nu\mu}$. – Fabian Dec 1 '12 at 8:50