Cross Product

Cross product of two vectors

\( \vec{\mathbf{u}} = \langle u_1, u_2, u_3 \rangle \)
\( \vec{\mathbf{v}} = \langle v_1, v_2, v_3 \rangle \)

is originaly defined in three dimensions as

\( \vec{\mathbf{u}} \times \vec{\mathbf{v}} = \begin{vmatrix} \vec{\mathbf{i}} & \vec{\mathbf{j}} & \vec{\mathbf{k}} \\ u_1 & u_2 & u_3 \\ v_1 & v_2 & v_3 \end{vmatrix} = (u_2 v_3 - u_3 v_2) \vec{\mathbf{i}} + (u_3 v_1 - u_1 v_3) \vec{\mathbf{j}} + (u_1 v_2 - u_2 v_1) \vec{\mathbf{k}} \)

This is a vector.

\( \vec{\mathbf{u}} = \langle u_1, u_2 \rangle \)
\( \vec{\mathbf{v}} = \langle v_1, v_2 \rangle \)

To find a two-dimensional definition, we first consider them as 3-vectors,

\( \vec{\mathbf{u}} \times \vec{\mathbf{v}} = \begin{vmatrix} \vec{\mathbf{i}} & \vec{\mathbf{j}} & \vec{\mathbf{k}} \\ u_1 & u_2 & 0 \\ v_1 & v_2 & 0 \end{vmatrix} = (u_1 v_2 - u_2 v_1) \vec{\mathbf{k}} \)

Then we return to two dimensions

\( \vec{\mathbf{u}} \times \vec{\mathbf{v}} = u_1 v_2 - u_2 v_1 \)

This is a scalar.

\( \vec{\mathbf{u}} = \langle u_1, u_2, u_3, u_4 \rangle \)
\( \vec{\mathbf{v}} = \langle v_1, v_2, v_3, v_4 \rangle \)

\( \vec{\mathbf{u}} \times \vec{\mathbf{v}} = \begin{vmatrix} \vec{\mathbf{e_1}} & \vec{\mathbf{e_2}} & \vec{\mathbf{e_3}} & \vec{\mathbf{e_4}} \\ \vec{\mathbf{e_1}} & \vec{\mathbf{e_2}} & \vec{\mathbf{e_3}} & \vec{\mathbf{e_4}} \\ u_1 & u_2 & u_3 & u_4 \\ v_1 & v_2 & v_3 & v_4 \end{vmatrix} \)

Since the tensor product is not commutative, this pseudo-determinant is not zero.

\( \begin{matrix} +(u_3 v_4-u_4 v_3) (\vec{\mathbf{e_1}} \otimes \vec{\mathbf{e_2}}) +(u_4 v_2-u_2 v_4) (\vec{\mathbf{e_1}} \otimes \vec{\mathbf{e_3}}) +(u_2 v_3-u_3 v_2) (\vec{\mathbf{e_1}} \otimes \vec{\mathbf{e_4}}) +(u_1 v_4-u_4 v_1) (\vec{\mathbf{e_2}} \otimes \vec{\mathbf{e_3}}) +(u_3 v_1-u_1 v_3) (\vec{\mathbf{e_2}} \otimes \vec{\mathbf{e_4}}) +(u_1 v_2-u_2 v_1) (\vec{\mathbf{e_3}} \otimes \vec{\mathbf{e_4}}) \\ -(u_3 v_4-u_4 v_3) (\vec{\mathbf{e_2}} \otimes \vec{\mathbf{e_1}}) -(u_4 v_2-u_2 v_4) (\vec{\mathbf{e_3}} \otimes \vec{\mathbf{e_1}}) -(u_2 v_3-u_3 v_2) (\vec{\mathbf{e_4}} \otimes \vec{\mathbf{e_1}}) -(u_1 v_4-u_4 v_1) (\vec{\mathbf{e_3}} \otimes \vec{\mathbf{e_2}}) -(u_3 v_1-u_1 v_3) (\vec{\mathbf{e_4}} \otimes \vec{\mathbf{e_2}}) -(u_1 v_2-u_2 v_1) (\vec{\mathbf{e_4}} \otimes \vec{\mathbf{e_3}}) \end{matrix} \)

Actually, we can use the wedge product
\(\vec{\mathbf{u}} \wedge \vec{\mathbf{v}} = \vec{\mathbf{u}} \otimes \vec{\mathbf{v}}-\vec{\mathbf{v}} \otimes \vec{\mathbf{u}} \)
to write down the final result as:

\( (u_3 v_4-u_4 v_3) (\vec{\mathbf{e_1}} \wedge \vec{\mathbf{e_2}}) +(u_4 v_2-u_2 v_4) (\vec{\mathbf{e_1}} \wedge \vec{\mathbf{e_3}}) +(u_2 v_3-u_3 v_2) (\vec{\mathbf{e_1}} \wedge \vec{\mathbf{e_4}}) +(u_1 v_4-u_4 v_1) (\vec{\mathbf{e_2}} \wedge \vec{\mathbf{e_3}}) +(u_3 v_1-u_1 v_3) (\vec{\mathbf{e_2}} \wedge \vec{\mathbf{e_4}}) +(u_1 v_2-u_2 v_1) (\vec{\mathbf{e_3}} \wedge \vec{\mathbf{e_4}}) \)

This is a 2-tensor.

\( \vec{\mathbf{u}} = \langle u_1, u_2, u_3, u_4, u_5 \rangle \)
\( \vec{\mathbf{v}} = \langle v_1, v_2, v_3, v_4, v_5 \rangle \)

In five dimensions where triple wedge product is
\( \vec{\mathbf{u}} \wedge \vec{\mathbf{v}} \wedge \vec{\mathbf{w}} = \vec{\mathbf{u}} \otimes \vec{\mathbf{v}} \otimes \vec{\mathbf{w}} -\vec{\mathbf{u}} \otimes \vec{\mathbf{w}} \otimes \vec{\mathbf{v}} +\vec{\mathbf{v}} \otimes \vec{\mathbf{w}} \otimes \vec{\mathbf{u}} -\vec{\mathbf{v}} \otimes \vec{\mathbf{u}} \otimes \vec{\mathbf{w}} +\vec{\mathbf{w}} \otimes \vec{\mathbf{u}} \otimes \vec{\mathbf{v}} -\vec{\mathbf{w}} \otimes \vec{\mathbf{v}} \otimes \vec{\mathbf{u}} \)

we have

\( \vec{\mathbf{u}} \times \vec{\mathbf{v}} = \begin{vmatrix} \vec{\mathbf{e_1}} & \vec{\mathbf{e_2}} & \vec{\mathbf{e_3}} & \vec{\mathbf{e_4}} & \vec{\mathbf{e_5}} \\ \vec{\mathbf{e_1}} & \vec{\mathbf{e_2}} & \vec{\mathbf{e_3}} & \vec{\mathbf{e_4}} & \vec{\mathbf{e_5}} \\ \vec{\mathbf{e_1}} & \vec{\mathbf{e_2}} & \vec{\mathbf{e_3}} & \vec{\mathbf{e_4}} & \vec{\mathbf{e_5}} \\ u_1 & u_2 & u_3 & u_4 & u_5 \\ v_1 & v_2 & v_3 & v_4 & v_5 \end{vmatrix} = \begin{matrix} +(u_1 v_2-u_2 v_1) (\vec{\mathbf{e_3}} \wedge \vec{\mathbf{e_4}} \wedge \vec{\mathbf{e_5}}) -(u_3 v_1-u_1 v_3) (\vec{\mathbf{e_2}} \wedge \vec{\mathbf{e_4}} \wedge \vec{\mathbf{e_5}}) \\ +(u_1 v_4-u_4 v_1) (\vec{\mathbf{e_2}} \wedge \vec{\mathbf{e_3}} \wedge \vec{\mathbf{e_5}}) -(u_1 v_5-u_5 v_1) (\vec{\mathbf{e_2}} \wedge \vec{\mathbf{e_3}} \wedge \vec{\mathbf{e_4}}) \\ +(u_2 v_3-u_3 v_2) (\vec{\mathbf{e_1}} \wedge \vec{\mathbf{e_4}} \wedge \vec{\mathbf{e_5}}) -(u_2 v_4-u_4 v_2) (\vec{\mathbf{e_1}} \wedge \vec{\mathbf{e_3}} \wedge \vec{\mathbf{e_5}}) \\ +(u_2 v_5-u_5 v_2) (\vec{\mathbf{e_1}} \wedge \vec{\mathbf{e_3}} \wedge \vec{\mathbf{e_4}}) +(u_3 v_4-u_4 v_3) (\vec{\mathbf{e_1}} \wedge \vec{\mathbf{e_2}} \wedge \vec{\mathbf{e_5}}) \\ -(u_3 v_5-u_5 v_3) (\vec{\mathbf{e_1}} \wedge \vec{\mathbf{e_2}} \wedge \vec{\mathbf{e_4}}) +(u_4 v_5-u_5 v_4) (\vec{\mathbf{e_1}} \wedge \vec{\mathbf{e_2}} \wedge \vec{\mathbf{e_3}}) \end{matrix} \)

This is a 3-tensor.

1773 Lagrange   algebraic formula
1843 Hamilton   quaternion product, scalar, vector
1844 Grassmann  bi-vector, wedge product
1881 Gibbs      cross product