All Electromagnetic Field Tensors#

by Stéphane Haussler

In this section, I systematically derive all representations of the Faraday tensor \(F\) and its Hodge dual with metric signature \((+,-,-,-)\). I cannot imagine there is no sign error in the calculations below. The Hodge dual \(⋆\:F\) is also noted \(G\). The starting point is the electromagnetic field 2-form derivated in the article Maxwell’s Equations via Differential Forms:

\[\begin{split}F^{♭♭} = \left[ \begin{aligned} - & \E^x \: dt ∧ dx \\ - & \E^y \: dt ∧ dy \\ - & \E^z \: dt ∧ dz \\ & B^x \: dy ∧ dz \\ & B^y \: dz ∧ dx \\ & B^z \: dx ∧ dy \\ \end{aligned} \right]\end{split}\]

All expressions of the Faraday Tensor \(F\) are the same object viewed from a different perspective. All these objects are strictly equivalent and their naming (2–form, tensor) merely a matter of convention.

\(F^{♭♭}\)#

The Faraday 2–form is our starting point and the expression already given above is repeated for completeness:

\[\begin{split}F^{♭♭} = \left[ \begin{aligned} - & \E^x \: dt ∧ dx \\ - & \E^y \: dt ∧ dy \\ - & \E^z \: dt ∧ dz \\ & B^x \: dy ∧ dz \\ & B^y \: dz ∧ dx \\ & B^z \: dx ∧ dy \\ \end{aligned} \right] \\\end{split}\]

From the Faraday 2–form and noting that forms are tensors, we can derive the row/column representation of the doubly covariant Faraday tensor.

\[\begin{split}F^{♭♭} = \frac{1}{2} \begin{bmatrix} & + \E^x \: dx ∧ dt & + \E^y \: dy ∧ dt & + \E^z \: dz ∧ dt \\ - \E^x \: dt ∧ dx & & - B^z \: dy ∧ dx & + B^y \: dz ∧ dx \\ - \E^y \: dt ∧ dy & + B^z \: dx ∧ dy & & - B^x \: dz ∧ dy \\ - \E^z \: dt ∧ dz & - B^y \: dx ∧ dz & + B^x \: dy ∧ dz & \\ \end{bmatrix}\end{split}\]

With implicit bivector basis, we have the standard representation with abstract index notation

\[\begin{split}F_{μν} = \begin{bmatrix} & + \E^x & + \E^y & + \E^z \\ - \E^x & & - B^z & + B^y \\ - \E^y & + B^z & & - B^x \\ - \E^z & - B^y & + B^x & \\ \end{bmatrix}\end{split}\]

Where the field 2-form is related to the Faraday tensor with:

\[F^{♭♭} = \frac{1}{2} \: F_{μν} \: dx^μ ∧ dx^ν\]

For sanity, I refer to Wikipedia for a quick double check of the link between the Faraday 2–form and the Faraday tensor.

\(F^{♭♯}\)#

The starting point is the twice flat Faraday 2–form \(F^{♭♭}\). Applying the musical ♭♯ operator \(F^{♭♯}=\left(F^{♭♭}\right)^{♭♯}\) results in:

\[\begin{split}F^{♭♯} = \left[ \begin{aligned} & \E^x \: dt ∧ ∂_x \\ & \E^y \: dt ∧ ∂_y \\ & \E^z \: dt ∧ ∂_z \\ - & B^x \: dy ∧ ∂_z \\ - & B^y \: dz ∧ ∂_x \\ - & B^z \: dx ∧ ∂_y \\ \end{aligned} \right]\end{split}\]

We derive the row/column representation of the \(F^{♭♯}\) Faraday tensor:

\[\begin{split}F^{♭♯} = \frac{1}{2} \begin{bmatrix} & + \E^x \: dt ∧ ∂_x & + \E^y \: dt ∧ ∂_y & + \E^z \: dt ∧ ∂_z \\ + \E^x \: dx ∧ ∂_t & & - B^z \: dx ∧ ∂_y & + B^y \: dx ∧ ∂_z \\ + \E^y \: dy ∧ ∂_t & + B^z \: dy ∧ ∂_x & & - B^x \: dy ∧ ∂_z \\ + \E^z \: dz ∧ ∂_t & - B^y \: dz ∧ ∂_x & + B^x \: dz ∧ ∂_y & \\ \end{bmatrix}\end{split}\]

With implicit bivector basis, we have :

\[\begin{split}F_μ{}^ν = \begin{bmatrix} & + \E^x & + \E^y & + \E^z \\ + \E^x & & - B^z & + B^y \\ + \E^y & + B^z & & - B^x \\ + \E^z & - B^y & + B^x & \\ \end{bmatrix}\end{split}\]

Where the mixed electromagnetic field is related to the covariant-contravariant Faraday tensor through:

\[F^{♭♯} = \frac{1}{2} \: F_μ{}^ν \: dx^μ ∧ ∂_ν\]

\(F^{♯♯}\)#

The starting point is the twice flattened Faraday tensor \(F^{♭♭}\) to which we apply the ♭♯ operator \(F^{♭♯}=\left(F^{♭♭}\right)^{♭♯}\) and obtain:

\[\begin{split}F^{♯♯} = \left[ \begin{aligned} \E^x \; ∂_t ∧ ∂_x \\ \E^y \; ∂_t ∧ ∂_y \\ \E^z \; ∂_t ∧ ∂_z \\ B^x \; ∂_y ∧ ∂_z \\ B^y \; ∂_z ∧ ∂_x \\ B^z \; ∂_x ∧ ∂_y \\ \end{aligned} \right]\end{split}\]

We derive the row/column representation of the \(F^{♭♯}\) Faraday tensor:

\[\begin{split}F^{♯♯} = \frac{1}{2} \begin{bmatrix} & - \E^x \; ∂_x ∧ ∂_t & - \E^y \; ∂_y ∧ ∂_t & - \E^z \; ∂_z ∧ ∂_t \\ + \E^x \; ∂_t ∧ ∂_x & & - B^z \; ∂_y ∧ ∂_x & + B^y \; ∂_z ∧ ∂_x \\ + \E^y \; ∂_t ∧ ∂_y & + B^z \; ∂_x ∧ ∂_y & & - B^x \; ∂_z ∧ ∂_y \\ + \E^z \; ∂_t ∧ ∂_z & - B^y \; ∂_x ∧ ∂_z & + B^x \; ∂_y ∧ ∂_z & \\ \end{bmatrix}\end{split}\]

With implicit bivector basis, we have the standard representation with abstract index notation, which also permits to verify the calculations here:

\[\begin{split}F_{μν} = \begin{bmatrix} & - \E^x & - \E^y & - \E^z \\ + \E^x & & - B^z & + B^y \\ + \E^y & + B^z & & - B^x \\ + \E^z & - B^y & + B^x & \\ \end{bmatrix}\end{split}\]

Where the electromagnetic field is related to the doubly contravariant Faraday tensor through:

\[F^{♯♯} = \frac{1}{2} \: F^{μν} \: ∂_μ ∧ ∂_ν\]

\(F^{♯♭}\)#

\(G^{♭♭}\)#

The Hodge dual \(G^{♭♭}\) of the Faraday 2-form \(F^{♭♭}\) is:

\[G^{♭♭} = ⋆ F^{♭♭}\]

Expanded, we obtain:

\[\begin{split}G^{♭♭} = \left[ \begin{alignedat}{1} B^x \; & dt ∧ dx \\ B^y \; & dt ∧ dy \\ B^z \; & dt ∧ dz \\ \E^x \; & dy ∧ dz \\ \E^y \; & dz ∧ dx \\ \E^z \; & dx ∧ dy \\ \end{alignedat} \right]\end{split}\]

From the dual Faraday 2-form and noting that 2-forms are tensors, we can derive the row/column representation of the doubly covariant dual Faraday tensor.

\[\begin{split}G^{♭♭} = \begin{bmatrix} & - B^x \; dx ∧ dt & - B^y \; dy ∧ dt & - B^z \; dz ∧ dt \\ + B^x \; dt ∧ dx & & - \E^z \; dy ∧ dx & + \E^y \; dz ∧ dx \\ + B^y \; dt ∧ dy & + \E^z \; dx ∧ dy & & - \E^x \; dz ∧ dy \\ + B^z \; dt ∧ dz & - \E^y \; dx ∧ dz & + \E^x \; dy ∧ dz & \\ \end{bmatrix}\end{split}\]

\(G^{♯♯}\)#

\(G^{♭♯}\)#

\(G^{♯♭}\)#