Setting the trace of a tensor equal to zero

Click For Summary

Homework Help Overview

The problem involves a tensor defined in an orthonormal basis of R3, with components dependent on certain functions of the squared distance from the origin. The task is to calculate the trace of the tensor and set it to zero, leading to constraints among the functions involved.

Discussion Character

  • Exploratory, Assumption checking

Approaches and Questions Raised

  • Participants discuss the calculation of the trace by summing specific tensor components and question the implications of setting the trace to zero. There is exploration of whether the variables can be simplified into a single variable.

Discussion Status

Some participants have provided insights into the relationship between the functions a and b, suggesting that setting the trace to zero leads to a specific constraint. Others have noted that the original poster's question about variable simplification is addressed within the context of the problem.

Contextual Notes

There is an emphasis on the use of the Einstein summation convention and the consideration of orthogonal transformations in R3. The discussion reflects on the nature of the tensor components and their dependencies on the squared distance.

EsmeeDijk
Messages
5
Reaction score
0

Homework Statement


A tensor t has the following components in a given orthonormal basis of R3

tij(x) = a(x2)xixj + b(x2) \deltaij x2 + c(x2) \epsilonijk xk (1)​
where the indices i,j,k = 1, 2, 3.
We use the Einstein summation convention. We will only consider orthogonal transformations on R3; therefore we will not make a distinction between upper and lower indices.
a,b,c are real functions of x2 := xixi, i.e. a = a(x2) depends on xi through the rotationally-symmetric combination xixi only.
The trace of the tensor t is defined as:
Tr(t) = tii (2)​

1. Calculate the trace (2) and set it to zero, i.e. find a constraint among the functions a,b,c such that Tr(t) = 0 for all x.

Homework Equations


We only need to use equation (1) and (2)

The Attempt at a Solution


Since the indices go from 1 to 3 we have to do t11 + t22 + t33 = 0
So the index i has to be equal to j. We get:
a(x2)x1x1 + b(x2)x2 + a(x2)x2x2 + b(x2)x2 + a(x2)x3x3 + b(x2)x2
The kronecker delta after the b is going to be one in every case so we can leave it out. The epsilon tensor in the part with c will be zero because we won't be able to get 3 different numbers, making the entire part with c 0.
We can now simplify to :
x2(a(x1)2 + a(x2)2 + a(x3)2 + 3b(x2)) = 0
So either x2 = 0
or (a(x1)2 + a(x2)2 + a(x3)2 + 3b(x2) = 0

So x = 0 or (a(x1)2 + a(x2)2 + a(x3)2 = -3b(x2)

Now my question is, is it the case that x1, x2 and x3 will they become just a normal x together?
such that we will have ax2 = -3bx2
a = -3b

 
Physics news on Phys.org
EsmeeDijk said:
x2 := xixi
This should already contain the answer to your question.
 
Contract your tensor equation with \delta_{ij} and set it equal to zero.
\mbox{Tr}(T) = \delta_{ij}T_{ij} = 0, \ \Rightarrow \ a = - 3b.
So
T_{ij} = -3b X_{i}X_{j} + b X^{2} \delta_{ij} + c \ \epsilon_{ijk} X_{k},
with b and c are arbitrary functions of the invariant X^{2}.
 
  • Like
Likes   Reactions: EsmeeDijk
Never mind, others already reacted while I was composing my post.
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
Replies
10
Views
3K
Replies
1
Views
3K
Replies
2
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
Replies
1
Views
2K