Divergence of Tensor Field

38 Views Asked by At

The author of this textbook is introducing this coordinate free definition of the divergence of tensor fields. Can someone help break this down for me, I am just really confused. First of all, does "a" represent a vector field? And then the the dot product of the two vector fields given by <div(T), a> would give us a scalar field right? Also, the right hand side of this equation would represent a scalar field as well. Any help would be greatly appreciated!

The Definition

1

There are 1 best solutions below

0
On BEST ANSWER

$ \def\r#1{\color{red}{#1}} \def\LR#1{\left(#1\right)} \def\op#1{\operatorname{#1}} \def\p{\partial_i} \def\c{\cdot} \def\n{\nabla} \def\a{a_k} \def\T{T_{ik}} $You will have a hard time proving the linked formula because it is wrong.

To see this, expand $\n\c\LR{T\c a}$ using the Einstein summation convention $$\eqalign{ \p\LR{\T\a} &= \LR{\p\T}\a + \T\LR{\p\a} \\ }$$ then rearrange the terms $$\eqalign{ \LR{\p\T}\a &= \p\LR{\T\a} - \T\LR{\p\a} \\ }$$ and convert back to vector notation $$\eqalign{ \LR{\n\c T}\c a &= \n\c\LR{\r T\c a} - T:\LR{\n a} \\ }$$ or in terms of named operators $$\eqalign{ {\op{div}\LR{T}}\c a &= \op{div}\LR{\r T\c a} - \op{trace}\LR{T^T\!\c{\n a}} \\ }$$ Note that the term in red is $\r{T^T}$ in the linked formula, which is incorrect.

$\sf NB\!:\:$ I could have expanded $\n\c\LR{T^T\c a}\,$ but then the linked document would have two typos instead of one, so the above is actually the most generous interpretation.