Convert 2x2 Matrix to 1x1 Tensor

  • Context: Undergrad 
  • Thread starter Thread starter Vitani1
  • Start date Start date
  • Tags Tags
    Matrix rank Tensor
Click For Summary

Discussion Overview

The discussion revolves around the conversion of a 2x2 matrix representing a second-order tensor into a different form, specifically addressing the transformation of indices using the metric tensor. Participants explore the implications of raising and lowering indices, as well as the distinction between obtaining a rank-1 tensor versus a rank-2 tensor through these operations.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant inquires about converting a matrix from $$M^{ab}$$ to $$M_{b}^{a}$$ and expresses confusion about the process.
  • Another participant clarifies that applying the metric tensor to a rank-2 tensor results in a (1,1) tensor, which remains a rank-2 tensor, and questions the understanding of the Einstein summation convention.
  • A participant raises the possibility of either raising an index or computing the trace of the tensor, indicating that these operations yield different results.
  • There is a discussion about the correct notation for the resulting tensor after applying the metric tensor, with some participants correcting each other on the proper use of subscripts and superscripts.
  • One participant expresses uncertainty about the Einstein summation convention and acknowledges the need to look it up.
  • Another participant confirms the correct notation and explains the summation process involved in the convention.

Areas of Agreement / Disagreement

Participants exhibit some agreement on the mechanics of raising and lowering indices and the use of the metric tensor, but there is disagreement regarding the interpretation of the operations and the resulting tensor forms. The understanding of the Einstein summation convention remains a point of contention, with some participants unsure about its application.

Contextual Notes

There are limitations in the discussion regarding the assumptions made about the operations on the tensor and the definitions of the terms used, particularly concerning the distinction between rank-1 and rank-2 tensors. The mathematical steps involved in the conversion process are not fully resolved.

Who May Find This Useful

This discussion may be useful for students or individuals studying tensor calculus, general relativity, or related fields in physics and mathematics, particularly those interested in the manipulation of tensor indices and the application of metric tensors.

Vitani1
Messages
51
Reaction score
9
TL;DR
Trying to figure out how to transform tensors appropriately.
If I have a matrix representing a 2nd order tensor (2 2) and I want to convert this matrix from M$$\textsuperscript{ab}$$ to $$M\textsubscript{b}\textsuperscript{a}$$ what do I do? I'm given the matrix elements for the 2x2 tensor. When applying the metric tensor to this matrix I understand symbolically how to get this 1st tank tensor but I don't understand how to write it out.
 
Physics news on Phys.org
Sorry - these should be proper super/subscripts.
 
To write subscripts and superscripts use _ and ^. To get ##M^{ab}## use M^{ab}. To get ##M^a{}_b## do M^a{}_b (the empty {} produce a zero width blank which the subscript hangs off - missing it out gives ##M^a_b##).

Applying the metric tensor to a rank-2 tensor doesn't give you a rank-1 tensor - it gives you a (1,1) tensor, which is also a rank-2 tensor. If you are representing the components in a matrix, it's still a 4×4 matrix. If you know how to write the expression, do you understand the Einstein summation convention? If so, can you write out explicitly what the ##\mu\nu## component of the result is in terms of a sum of products of the components of ##M## and the metric?
 
@Vitani1 I'm confused about what you are trying to do. Are you trying to raise an index on the tensor ##M##? Or are you trying to compute its trace, which is a scalar?

Raising an index, which is what is implied by your saying you want to convert the matrix from ##M_{ab}## to ##M_a{}^b##, would look like this:

$$
M_{ab} \ g^{bc} = M_a{}^c
$$

Taking the trace, which is what is implied by your saying you want to convert the matrix to a 1 x 1 tensor, which is a scalar, would look like this:

$$
M_{ab} \ g^{ab}
$$
 
  • Like
Likes   Reactions: vanhees71
The first thing you stated is correct. Say I have a tensor Mab (superscripts) and I apply this metric tensor to this matrix to lower one of its indices - how would I multiply this result out?
 
Vitani1 said:
Say I have a tensor Mab (superscripts) and I apply this metric tensor to this matrix to lower one of its indices - how would I multiply this result out?

So you have ##M^{ab} \ g_{bc} = M^a{}_c## [Edit: fixed] in the standard notation using the Einstein summation convention. Do you understand how that convention works?
 
Last edited:
  • Like
Likes   Reactions: vanhees71
I take this to mean I multiply this matrix by a scalar given by the metric tensor g?
 
No.

Do you understand the Einstein summation convention?
 
  • Like
Likes   Reactions: Vitani1
I guess not. I'll look it up. Thank you.
 
  • #10
Vitani1 said:
I'll look it up.
See the link in @PeterDonis' last post.
 
  • #11
PeterDonis said:
So you have

Shouldn't it be ##M^{a}{}_{c}##?
 
Last edited:
  • Like
Likes   Reactions: Ibix
  • #12
PeterDonis said:
So you have ##M^{ab} \ g_{bc} = M_{ac}## in the standard notation using the Einstein summation convention. Do you understand how that convention works?
It's ##M^{ab} g_{bc}={M^a}_c##, and whenever two indices are the same (where one must be an upper and the other necessarily a lower index) you sum over ##a## (in GR from 0 to 3).

The metric components ##g_{ab}## are used to convert an upper index (contravariant) to a lower index (covariant).
 
  • #13
Solved. Thanks! By the way there is a nice problem about this in A First Course In General Relativity by Schutz (Chapter 3, problem 24)
 
  • #14
weirdoguy said:
Shouldn't it be ##M^{a}{}_{c}##?

Yes, you're right, I'll fix the post.
 

Similar threads

  • · Replies 6 ·
Replies
6
Views
1K
  • · Replies 7 ·
Replies
7
Views
1K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 6 ·
Replies
6
Views
1K
  • · Replies 22 ·
Replies
22
Views
4K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 12 ·
Replies
12
Views
4K
  • · Replies 0 ·
Replies
0
Views
1K