0% found this document useful (0 votes)
96 views9 pages

SP 5

1) The components of a Cartesian tensor Aij in a rotated coordinate system are obtained by applying the rotation matrix to the original tensor components. 2) Rotating the coordinate system of the given tensor results in new tensor components that are functions of the rotation angle θ. 3) The metric tensor for Euclidean 3-space is the Kronecker delta, with gij = δij.

Uploaded by

Arslan Ali
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
96 views9 pages

SP 5

1) The components of a Cartesian tensor Aij in a rotated coordinate system are obtained by applying the rotation matrix to the original tensor components. 2) Rotating the coordinate system of the given tensor results in new tensor components that are functions of the rotation angle θ. 3) The metric tensor for Euclidean 3-space is the Kronecker delta, with gij = δij.

Uploaded by

Arslan Ali
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Sample Problems 5

5.1. Consider a Cartesian tensor in Euclidean 2-space with components Aij


(in a particular system of coordinates) given by the matrix
 
2 5
;
1 3

i.e., A11 = 2, A12 = 5, A21 = 1, and A22 = 3. Consider a rotation of the


′ ′
coordinate system through angle θ. What are the components Ai j in this
new system?

Solution:
 
2 5
Aij :
1 3

′ ′

cos θ sin θ
 p11 = cos θ p12 = sin θ
rotation matrix: ⇒
− sin θ cos θ ′ ′
p21 = − sin θ p22 = cos θ

2 X
X 2
′ ′
i′ j ′
pjj Aij pii pjj
′ ′
ij
A =A pii =
i=1 j=1

′ ′ ′ ′ ′ ′ ′ ′ ′ ′
A1 1 = A11 p11 p11 + A12 p11 p12 + A21 p12 p11 + A22 p12 p12

= 2 cos2 θ + 5 cos θ sin θ + 1 cos θ sin θ + 3 sin2 θ

= 2 + sin2 θ + 6 sin θ cos θ

′ ′ ′ ′ ′ ′ ′ ′ ′ ′
A1 2 = A11 p11 p21 + A12 p11 p22 + A21 p12 p21 + A22 p12 p22

= −2 cos θ sin θ + 5 cos2 θ − 1 sin2 θ + 3 sin θ cos θ

= −1 + 6 cos2 θ + sin θ cos θ

1
′ ′ ′ ′ ′ ′ ′ ′ ′ ′
A2 1 = A11 p21 p11 + A12 p21 p12 + A21 p22 p11 + A22 p22 p12

= −2 sin θ cos θ − 5 sin2 θ + 1 cos2 θ + 3 cos θ sin θ

= 1 − 6 sin2 θ + sin θ cos θ

′ ′ ′ ′ ′ ′ ′ ′ ′ ′
A2 2 = A11 p21 p21 + A12 p21 p22 + A21 p22 p21 + A22 p22 p22

= 2 sin2 θ − 5 sin θ cos θ − 1 sin θ cos θ + 3 cos2 θ

= 2 + cos2 θ − 6 sin θ cos θ

Check: returns Aij when θ = 0.

5.2. Evaluate Aij Bij , where Aij is the tensor from problem 1 above and the
components of Bij are
 
4 0
.
2 7

Solution:
2 X
X 2
ij
A Bij = Aij Bij
i=1 j=1

= A11 B11 + A12 B12 + A21 B21 + A22 B22

= 8 + 0 + 2 + 21

= 31

5.3. Show that Aii = Aij δij .

2
Solution:

With the sums written out explicitly,


N
X
Aii = Aii
i=1

and
X N
N X N
X
Aij δij = Aij δij = Aii
i=1 j=1 i=1

In the last step, we made use of the fact that δij = 1 when i = j and is zero
when j 6= i. Thus, all but one of the terms in the sum over j vanish.

5.4. Suppose Aij j hj j


klm is a tensor and Bkm = Akhm . Show that Bkm is a tensor.

Solution:

Since Aij
klm is a tensor,
′ ′ ′
Aik′jl′ m′ = Aij i j k l m ′
klm pi pj pk ′ pl′ pm′

′ ′ ′ ′ ′
Bkj ′ m′ = Ahk′ hj ′ m′ = Aik′jl′ m′ δil′


= Aij i j k l m l ′ ′
klm pi pj pk ′ pl′ pm′ δi′


= Aij i j k l m ′
klm pi pj pk ′ pi′ pm′


= Aij j k m ′
l i
klm (pi′ pi ) pj pk ′ pm′


= Aij l j k m
klm δi pj pk ′ pm′


= Alj j k m
klm pj pk ′ pm′


j
= Bkm pjj pkk′ pm
m′

3
5.5. Consider Euclidean 2-space and a rotation of the axes through angle θ.
Suppose Aij is a rank 2 tensor defined for this space and transformation and
that its components are given by Aij = i + j. Also define B = Aii . Verify
explicitly that B is a scalar.

Solution:

First, recall that


′ ′
p11 = cos θ = p11′ , p12 = sin θ = p21′

′ ′
p21 = − sin θ = p12′ , p22 = cos θ = p22′

B = Aii = A11 + A22 = 2 + 4 = 6

′ ′ ′
B ′ = Aii′ = A11′ + A22′

2 X
X 2
Aij p1i pj1′
′ ′
A11′ =
i=1 j=1

′ ′ ′ ′
= A11 p11 p11′ + A12 p11 p21′ + A21 p12 p11′ + A22 p12 p21′

= 2 cos2 θ + 3 cos θ sin θ + 3 sin θ cos θ + 4 sin2 θ

′ ′ ′ ′ ′
A22′ = A11 p21 p12′ + A12 p21 p22′ + A21 p22 p12′ + A22 p22 p22′

= 2(− sin θ)2 + 3(− sin θ) cos θ + 3 cos θ(− sin θ) + 4 cos2 θ

′ ′
B ′ = A11′ + A22′ = 2 + 0 + 0 + 4 = 6 = B

4
5.6. What is the metric tensor for Euclidean 3-space?

Solution:
(
1 , i=j
d~s 2 = dx2 + dy 2 + dz 2 ⇒ gij = δij =
0 , i 6= j

5.7. Show that g i j = δ i j .

Solution:

g i j = g ik gkj = δ i j

In the first equality, g ik raises the index k on gkj (or, alternatively, gkj lowers
the index k on g ik ). The second equality follows from the definition of g ik .

5.8. Given that gµν = diag(-1, -1, -1, 1), show that g µν = gµν .

Solution:

We just need to find g µν such that g µν gνα = δ µ α .

µ=1 : g 11 g1α + g 12 g2α + g 13 g3α + g 14 g4α = δ 1 α

α = 1 ⇒ −g 11 = 1 ⇒ g 11 = −1

α = 2 ⇒ −g 12 = 0 ⇒ g 12 = 0

α = 3 ⇒ −g 13 = 0 ⇒ g 13 = 0

α = 4 ⇒ g 14 = 0

5
Similarly when µ = 2, 3.

µ=4 : g 41 g1α + g 42 g2α + g 43 g3α + g 44 g4α = δ 4 α

α = 1 ⇒ −g 41 = 0 ⇒ g 41 = 0

α = 2 ⇒ −g 42 = 0 ⇒ g 42 = 0

α = 3 ⇒ −g 43 = 0 ⇒ g 43 = 0

α = 4 ⇒ g 44 = 1

We have specified all the components of g µν and found diag(-1, -1, -1, 1).

If the problem had said “verify” rather than “show”, then we could just treat
g µν gνα as a matrix multiplication and verify that we get the identity matrix
δij :
     
−1 0 0 0 −1 0 0 0 1 0 0 0
 0 −1 0 0  0 −1 0 0 0 1 0 0
 0 0 −1 0 = 0 X
     
 0 0 −1 0 0 1 0
0 0 0 1 0 0 0 1 0 0 0 1

5.9. Consider two frames S and S′ in standard configuration. Given a 4-


tensor Aij , find
(a) A13
(b) A24
′ ′
(c) A4 1

Solution:

For a standard Lorentz transformation,


′ ′ ′ ′ ′ ′
p11 = p44 = γ , p14 = p41 = −γβ , p22 = p33 = 1

p11′ = p44′ = γ , p14′ = p41′ = γβ , p22′ = p33′ = 1

6
The other transformation coefficients are all zero.

(a) A13 = g1µ g3ν Aµν


4
X
= g3ν [ g11 A1ν + g12 A2ν + g13 A3ν + g14 A4ν ]
|{z} |{z} |{z} |{z}
ν=1 −1 0 0 0

4
X
= − g3ν A1ν
ν=1

= −( g31 A11 + g32 A12 + g33 A13 + g34 A14 )


|{z} |{z} |{z} |{z}
0 0 −1 0

= −(−A13 ) = A13

(b) A24 = g2µ g4ν Aµν = (−1) (1) A24 = −A24

′ ′ ′ ′
(c) A4 1 = Aµν p4µ p1ν
4
X ′ ′ ′ ′ ′
= p1ν [A1ν p41 + A2ν p42 + A3ν p43 + A4ν p44 ]
ν=1

4
X ′
= p1ν [−γβ A1ν + γ A4ν ]
ν=1

′ ′
= p11 [−γβ A11 + γ A41 ] + p14 [−γβ A14 + γ A44 ]
|{z} |{z}
γ −γβ

= −γ 2 β A11 + γ 2 A41 + γ 2 β 2 A14 − γ 2 β A44

= γ 2 A41 + γ 2 β 2 A14 − γ 2 β (A11 + A44 )

7
5.10. If T µν is a 4-tensor, show that under a spatial rotation T 44 remains
unchanged, T 4i and T i4 transform as 3-vectors, and T ij transforms as a 3-
tensor.

Solution:

Note: It’s not correct to multiply the transformation matrix times a tensor
to get the tensor in the new coordinate system. That only works for vectors.
We need 2 transformation matrices for a rank 2 tensor and it’s not just simple
multiplication.

The transformation matrix for a spatial rotation in Minkowski space is

R = the rotation matrix of Euclidean 3-space


′ ′ ′ ′ ′
pij = Ri j , p4i = pi4 = 0 , p44 = 1 (i, j = 1 to 3)

′ ′ ′ ′
T µ ν = pµµ pνν T µν

′ ′ ′ ′
T 4 4 = p4µ p4ν T µν = (1) (1) T 44 = T 44 X

8
If it helps to see it written out in detail, here it is:
′ ′ ′ ′ ′ ′ ′ ′ ′ ′
T4 4 = p41 p41 T 11 + p41 p42 T 12 + p41 p43 T 13 + p41 p44 T 14
′ ′ ′ ′ ′ ′ ′ ′
+p42 p41 T 21 + p42 p42 T 22 + p42 p43 T 23 + p42 p44 T 24
′ ′ ′ ′ ′ ′ ′ ′
+p43 p41 T 31 + p43 p42 T 32 + p43 p43 T 33 + p43 p44 T 34
′ ′ ′ ′ ′ ′ ′ ′
+p44 p41 T 41 + p44 p42 T 42 + p44 p43 T 43 + p44 p44 T 44

The are 16 terms; all are zero except for the last one.


T 4 j = p4µ pjν T µν = pjν T 4ν = pji T 4i = Rj i T 4i
′ ′ ′ ′ ′ ′
X
A B

′ ′
In step A, we made use of p4i = 0 and p44 = 1. In step B, we made use

of pj4 = 0. The final result is the rotation matrix acting on a 3-component
object since i = 1, 2, 3.

T j 4 = pjµ p4ν T µν = pjµ T µ4 = pji T i4 = Rj i T i4
′ ′ ′ ′ ′ ′
X


T i j = piµ pjν T µν = pii pjj T ij = Ri i Rj j T ij
′ ′ ′ ′ ′ ′ ′
X

You might also like