Access full Solution Manual only here

http://www.book4me.xyz/solution-manual-continuum-mechanics-michael-lai/

CHAPTER 2, PART A

2.1 Given
⎡1 0 2 ⎤ ⎡1 ⎤
⎡⎣ Sij ⎤⎦ = 0 1 2 and [ ai ] = ⎢ 2 ⎥
⎢ ⎥
⎢ ⎥ ⎢ ⎥
⎢⎣ 3 0 3 ⎥⎦ ⎢⎣ 3 ⎥⎦
Evaluate (a) Sii , (b) Sij Sij , (c) S ji S ji , (d) S jk Skj (e) am am , (f) Smn am an , (g) Snm am an
-------------------------------------------------------------------------------
Ans. (a) Sii = S11 + S22 + S33 = 1 + 1 + 3 = 5 .
2 2 2 2 2 2 2 2 2
(b) Sij Sij = S11 + S12 + S13 + S21 + S22 + S23 + S31 + S32 + S33 =
1 + 0 + 4 + 0 + 1 + 4 + 9 + 0 + 9 = 28 .
(c) S ji S ji = Sij Sij =28.
(d) S jk S kj = S1k S k1 + S 2 k S k 2 + S3k S k 3
= S11S11 + S12 S21 + S13 S31 + S21S12 + S22 S22 + S23 S32 + S31S13 + S32 S23 + S33 S33
= (1)(1) + ( 0 )( 0 ) + ( 2 )( 3) + ( 0 )( 0 ) + (1)(1) + ( 2 )( 0 ) + ( 3)( 2 ) + ( 0 )( 2 ) + ( 3)( 3) = 23 .
(e) am am = a12 + a22 + a32 = 1 + 4 + 9 = 14 .
(f) Smn am an = S1n a1an + S2 n a2 an + S3n a3 an =
S11a1a1 + S12 a1a2 + S13 a1a3 + S21a2 a1 + S22 a2 a2 + S23a2 a3 + S31a3a1 + S32 a3a2 + S33a3a3
= (1)(1)(1) + ( 0 )(1)( 2 ) + ( 2 )(1)( 3) + ( 0 )( 2 )(1) + (1)( 2 )( 2 ) + ( 2 )( 2 )( 3) + ( 3)( 3)(1)
+ ( 0 )( 3)( 2 ) + ( 3)( 3)( 3) = 1 + 0 + 6 + 0 + 4 + 12 + 9 + 0 + 27 = 59.
(g) Snm am an = Smn am an =59.
__________________________________________________________________
2.2 Determine which of these equations have an identical meaning with ai = Qij a 'j .
(a) a p = Q pm am' , (b) a p = Qqp aq' , (c) am = an' Qm n .
-------------------------------------------------------------------------------
Ans. (a) and (c)
__________________________________________________________________
2.3 Given the following matrices
⎡1 ⎤ ⎡ 2 3 0⎤
[ ai ] = ⎢0 ⎥ , ⎡⎣ Bij ⎤⎦ = ⎢⎢0 5 1 ⎥⎥
⎢ ⎥
⎢⎣ 2 ⎥⎦ ⎢⎣ 0 2 1 ⎥⎦
Demonstrate the equivalence of the subscripted equations and corresponding matrix equations in
the following two problems.
(a) bi = Bij a j and [b ] = [ B ][ a ] , (b) s = Bij ai a j and s = [ a ] [ B ][ a ]
T

-------------------------------------------------------------------------------
Ans. (a)
bi = Bij a j → b1 = B1 j a j = B11a1 + B12 a2 + B13a3 = ( 2 )(1) + ( 3)( 0 ) + ( 0 )( 2 ) = 2
b2 = B2 j a j = B21a1 + B22 a2 + B23a3 = 2, b3 = B3 j a j = B31a1 + B32 a2 + B33 a3 = 2 .

Copyright 2010, Elsevier Inc

2-1
Lai et al, Introduction to Continuum Mechanics

⎡ 2 3 0⎤ ⎡1 ⎤ ⎡ 2⎤
[b] = [ B ][ a ] = ⎢⎢0 5 1 ⎥⎥ ⎢⎢0 ⎥⎥ = ⎢⎢ 2⎥⎥ . Thus, bi = Bij a j gives the same results as [b] = [ B ][ a ]
⎢⎣ 0 2 1 ⎥⎦ ⎢⎣ 2 ⎥⎦ ⎢⎣ 2 ⎥⎦
(b)
s = Bij ai a j = B11a1a1 + B12 a1a2 + B13 a1a3 + + B21a2 a1 + B22 a2 a2 + B23 a2 a3
+ B31a3 a1 + B32 a3a2 + B33a3a3 = ( 2 ) (1)(1) + ( 3) (1)(0) + ( 0 ) (1)(2) + ( 0 ) (0)(1)
+ ( 5 ) (0)(0) + (1) (0)(2) + ( 0 ) (2)(1) + ( 2 ) (2)(0) + (1) (2)(2) = 2 + 4 = 6.
⎡ 2 3 0⎤ ⎡1 ⎤ ⎡2⎤
and s = [ a ] [ B ][ a ] = [1 0 2] ⎢ 0 5 1 ⎥ ⎢ 0 ⎥ = [1 0 2] ⎢⎢ 2 ⎥⎥ = 2 + 4 = 6 .
⎢ ⎥ ⎢ ⎥
T

⎢⎣ 0 2 1 ⎥⎦ ⎢⎣ 2 ⎥⎦ ⎢⎣ 2 ⎥⎦
__________________________________________________________________
Write in indicial notation the matrix equation (a) [ A] = [ B ][C ] , (b) [ D ] = [ B ] [C ] and (c)
T
2.4
[ E ] = [ B ]T [C ][ F ] .
-------------------------------------------------------------------------------
Ans. (a) [ A] = [ B ][C ] → Aij = Bi m Cm j , (b) [ D ] = [ B ] [C ] → Aij = Bm i Cmj .
T

(c) [ E ] = [ B ] [C ][ F ] → Eij = Bmi Cmk Fkj .

T

__________________________________________________________________
∂ 2φ ∂ 2φ ∂ 2φ
2.5 Write in indicial notation the equation (a) s = A12 + A22 + A32 and (b) 2 + 2 + 2 = 0 .
∂x1 ∂x2 ∂x3
-------------------------------------------------------------------------------
∂ 2φ ∂ 2φ ∂ 2φ ∂ 2φ
Ans. (a) s = A12 + A22 + A32 = Ai Ai . (b) 2 + 2 + 2 = 0 → =0.
∂x1 ∂x2 ∂x3 ∂xi ∂xi
__________________________________________________________________
2.6 Given that Si j =ai a j and Sij′ =ai′a′j , where ai′ =Qmi am and a′j =Qn j an , and Qik Q jk = δ ij .
Show that Sii′ =Sii .
-------------------------------------------------------------------------------
Ans. Sij′ =Qmi am Qn j an =Qmi Qn j am an → Sii′ =Qmi Qni am an =δ mn am an =am am = S mm = Sii .
__________________________________________________________________
∂v ∂v
2.7 Write ai = i + v j i in long form.
∂t ∂x j
-------------------------------------------------------------------------------
Ans.
∂v ∂v ∂v ∂v ∂v ∂v
i = 1 → a1 = 1 + v j 1 = 1 + v1 1 + v2 1 + v3 1 .
∂t ∂x j ∂t ∂x1 ∂x2 ∂x3
∂v2 ∂v ∂v ∂v ∂v ∂v
i = 2 → a2 = + v j 2 = 2 + v1 2 + v2 2 + v3 2 .
∂t ∂x j ∂t ∂x1 ∂x2 ∂x3
∂v3 ∂v ∂v ∂v ∂v ∂v
i = 3 → a3 = + v j 3 = 3 + v1 3 + v2 3 + v3 3 .
∂t ∂x j ∂t ∂x1 ∂x2 ∂x3
__________________________________________________________________
Copyright 2010, Elsevier Inc
2-2
Lai et al, Introduction to Continuum Mechanics

2.8 Given that Tij = 2 μ Eij + λ Ekk δ ij , show that

(a) Tij Eij = 2μ Eij Eij + λ ( Ekk ) and (b) TijTij = 4μ 2 Eij Eij + ( Ekk ) (4μλ + 3λ 2 )
2 2

-------------------------------------------------------------------------------
Ans. (a)
Tij Eij = (2μ Eij + λ Ekk δ ij ) Eij = 2μ Eij Eij + λ Ekk δ ij Eij = 2 μ Eij Eij + λ Ekk Eii = 2μ Eij Eij + λ ( Ekk )2
(b)
TijTij = (2μ Eij + λ Ekk δ ij )(2μ Eij + λ Ekk δ ij ) = 4μ 2 Eij Eij + 2μλ Eij Ekk δ ij + 2 μλ Ekk δ ij Eij

+λ 2 ( Ekk ) δ ijδ ij = 4μ 2 Eij Eij + 2 μλ Eii Ekk + 2μλ Ekk Eii + λ 2 ( Ekk ) δ ii
2 2

= 4 μ 2 Eij Eij + ( Ekk ) (4μλ + 3λ 2 ).

2

__________________________________________________________________
2.9 ′ .
Given that ai =Tij b j , and ai′ =Tij′b′j , where ai =Qim am′ and Tij =Qim Q jnTmn
′ bn′ = Qim Q jnTmn
(a) Show that QimTmn ′ b j and (b) if Qik Qim =δ km , then Tkn
′ (bn′ − Q jn b j ) = 0 .
-------------------------------------------------------------------------------
Ans. (a) Since ai =Qim am ′ and Tij =Qim Q jnTmn ′ , therefore, ai =Tij b j → .
Qim am′ = Qim Q jnTmn
′ b j (1), Now, ai′ =Tij′b′j → am′ =Tmj
′ b′j = Tmn
′ bn′ , therefore, Eq. (1) becomes
Qi mTm′ n bn′ = Qi m Q j nTm′ n b j . (2)
(b) To remove Qim from Eq. (2), we make use of Qik Qim =δ km by multiplying the above equation,
Eq.(2) with Qik . That is,
Qik QimTm′ nbn′ = Qik QimQ jnTm′ nb j → δ kmTm′ nbn′ = δ kmQ jnTm′ nb j → Tk′nbn′ = Q jnTk′nb j
→ Tk′n (bn′ − Q jnb j ) = 0 .
__________________________________________________________________

⎡1 ⎤ ⎡0⎤
2.10 Given [ ai ] = ⎢ 2 ⎥ and [bi ] = ⎢⎢ 2 ⎥⎥ Evaluate [di ] , if d k = ε ijk ai b j and show that this result is
⎢ ⎥

⎣⎢ 0 ⎦⎥ ⎣⎢ 3 ⎦⎥
the same as d k = ( a × b ) ⋅ e k .
-------------------------------------------------------------------------------

Ans. d k = ε ijk ai b j →
d1 = ε ij1ai b j = ε 231a2b3 + ε 321a3b2 = a2b3 − a3b2 = (2)(3) − (0)(2) = 6
d 2 = ε ij 2 ai b j = ε 312 a3b1 + ε132 a1b3 = a3b1 − a1b3 = (0)(0) − (1)(3) = −3
d3 = ε ij 3 ai b j = ε123 a1b2 + ε 213a2b1 = a1b2 − a2b1 = (1)(2) − (2)(0) = 2
Next, ( a × b ) = ( e1 + 2e2 ) × ( 2e2 + 3e3 ) = 6e1 − 3e2 + 2e3 .
d1 = ( a × b ) ⋅ e1 = 6, d 2 = ( a × b ) ⋅ e 2 = −3, d3 = ( a × b ) ⋅ e3 = 2 .
__________________________________________________________________
2.11 (a) If ε ijk Tij = 0 , show that Tij = T ji , and (b) show that δ ij ε ijk =0
-------------------------------------------------------------------------------
Copyright 2010, Elsevier Inc
2-3
Lai et al, Introduction to Continuum Mechanics

Ans. (a) for k = 1, ε ij1Tij = 0 → ε 231T23 + ε 321T32 = 0 → T23 − T32 → T23 = T32 .
for k = 2, ε ij 2Tij = 0 → ε 312T31 + ε132T13 = 0 → T31 − T13 → T31 = T13 .
for k = 3, ε ij 3Tij = 0 → ε123T12 + ε 213T21 = 0 → T12 − T21 → T12 = T21 .
(b) δ ij ε ijk = δ11ε11k + δ 22ε 22k + δ 33ε 33k = (1)( 0 ) + (1)( 0 ) + (1)( 0 ) = 0 .
__________________________________________________________________
2.12 Verify the following equation: ε ijmε klm = δ ik δ jl − δ il δ jk .
(Hint): there are 6 cases to be considered (i) i = j , (2) i = k , (3) i = l , (4) j = k , (5) j = l , and (6)
k =l .
-------------------------------------------------------------------------------

Ans. There are 4 free indices in the equation. Therefore, there are the following 6 cases to consider:
(i) i = j , (2) i = k , (3) i = l , (4) j = k , (5) j = l , and (6) k = l . We consider each case below
where we use LS for left side, RS for right side and repeated indices with parenthesis are not sum:
(1) For i = j , LS=ε (i )(i ) mε klm = 0, RS = δ (i ) k δ (i )l − δ (i )l δ (i ) k = 0.
(2) For i = k , LS=ε (i ) j1ε (i )l1 + ε (i ) j 2ε (i )l 2 + ε (i ) j 3ε (i )l 3 , RS = δ (i )(i )δ jl − δ (i )l δ j (i )
⎧0 if j ≠ l
⎪
LS=RS = ⎨ 0 if j = l = i .
⎪ 1 if j = l ≠ i
⎩
(3) For i = l , LS=ε (i ) jmε k (i ) m , RS = δ (i ) k δ j (i ) − δ (i )(i )δ jk
⎧0 if j ≠ k
⎪
LS=RS = ⎨ 0 if j = k = i
⎪ −1 if j = k ≠ i
⎩
(4) For j = k , LS=ε i ( j ) mε ( j )lm , RS = δ i ( j )δ ( j )l − δ il δ ( j )( j )
⎧0 if i ≠ l
⎪
LS=RS = ⎨ 0 if i = l = j
⎪ −1 if i = l ≠ j
⎩
(5) For j = l , LS=ε i ( j ) mε k ( j ) m , RS = δ ik δ ( j )( j ) − δ i ( j )δ ( j ) k
⎧0 if i ≠ k
⎪
LS=RS = ⎨0 if i = k = j
⎪ 1 if i = k ≠ j
⎩
(6) For k = l , LS=ε ijmε ( k )( k ) m =0, RS = δ i ( k )δ j ( k ) − δ i ( k )δ j ( k ) = 0
__________________________________________________________________
2.13 Use the identity ε ijmε klm = δ ik δ jl − δ il δ jk as a short cut to obtain the following results:
(a) ε ilmε jlm = 2δ i j and (b) ε ijk ε ijk = 6 .
-------------------------------------------------------------------------------
Ans. (a) ε ilmε jlm = δ i jδ ll − δ il δ lj = 3δ i j − δ i j = 2δ i j .
(b) ε ijk ε ijk = δ iiδ jj − δ i jδ ji = (3)(3) − δ ii = 9 − 3 = 6 .
__________________________________________________________________
2.14 Use the identity ε ijmε klm = δ ik δ jl − δ il δ jk to show that a × (b × c) = (a ⋅ c)b − (a ⋅ b)c .

Copyright 2010, Elsevier Inc

2-4
Lai et al, Introduction to Continuum Mechanics

-------------------------------------------------------------------------------
Ans. a × (b × c) = am e m × (ε ijk b j ck ei ) = ε ijk am b j ck (e m × ei )
= ε ijk am b j ck (ε nmi e n ) = ε ijk ε nmi amb j ck e n = ε jkiε nmi amb j ck e n
= (δ jnδ km − δ jmδ kn ) am b j ck e n = δ jnδ km am b j ck e n − δ jmδ kn am b j ck e n
= ak bn ck e n − a j b j cne n = (a ⋅ c)b − (a ⋅ b)c .
__________________________________________________________________
2.15 (a) Show that if Tij = −T ji , Tij ai a j = 0 and (b) if Tij = −T ji , and Sij = S ji , then Tij Sij = 0
-------------------------------------------------------------------------------
Ans. Since Tij ai a j = T ji a j ai (switching the original dummy index i to j and the original index
j to i ), therefore Tij ai a j = T ji a j ai = −Tij a j ai = −Tij ai a j → 2Tij ai a j = 0 → Tij ai a j = 0 .
(b) Tij Sij = T ji S ji (switching the original dummy index i to j and the original index j to i ),
therefore, Tij Sij = T ji S ji = −Tij S ji = −Tij Sij → 2Tij Sij = 0 → Tij Sij = 0 .
__________________________________________________________________
2.16 ( ) ( )
Let Tij = Sij + S ji /2 and Rij = Sij − S ji / 2 , show that Tij = T ji , Rij = − R ji ,
and Sij = Tij + R ij .
-------------------------------------------------------------------------------
( ) (
Ans. Tij = Sij + S ji / 2 → T ji = S ji + Sij / 2 = Tij . )
( ) ( ) (
Rij = Sij − S ji / 2 → R ji = S ji − Sij / 2 = − Sij − S ji / 2 = − Rij . )
( ) ( )
Tij +Rij = Sij + S ji /2+ Sij − S ji / 2 = Sij .
__________________________________________________________________
2.17 Let f ( x1 , x2 , x3 ) be a function of x1 , x2 ,and x3 and vi ( x1 , x2 , x3 ) be three functions of
x1 , x2 ,and x3 . Express the total differential df and dvi in indicial notation.
-------------------------------------------------------------------------------
∂f ∂f ∂f ∂f
Ans. df = dx1 + dx2 + dx3 = dxi .
∂x1 ∂x2 ∂x3 ∂xi
∂v ∂v ∂v ∂v
dvi = i dx1 + i dx2 + i dx3 = i dxm .
∂x1 ∂x2 ∂x3 ∂xm
__________________________________________________________________
2.18 Let Aij denote that determinant of the matrix ⎡⎣ Aij ⎤⎦ . Show that Aij = ε ijk Ai1 A j 2 Ak 3
-------------------------------------------------------------------------------
Ans. ε ijk Ai1 A j 2 Ak 3 = ε1 jk A11 A j 2 Ak 3 + ε 2 jk A21 A j 2 Ak 3 + ε 3 jk A31 A j 2 Ak 3
= ε123 A11 A22 A33 + ε132 A11 A32 A23 + ε 231 A21 A32 A13 + ε 213 A21 A12 A33 + ε 312 A31 A12 A23 + ε 321 A31 A22 A13
= A11 A22 A33 − A11 A32 A23 + A21 A32 A13 − A21 A12 A33 + A31 A12 A23 − A31 A22 A13
A11 A12 A13
= A21 A22 A23
A31 A32 A33
__________________________________________________________________

Copyright 2010, Elsevier Inc

2-5
Lai et al, Introduction to Continuum Mechanics

CHAPTER 2, PART B

2.19 A transformation T operate on any vector a to give Ta = a / a , where a is the magnitude

of a . Show that T is not a linear transformation.
-------------------------------------------------------------------------------
a a+b a b
Ans. Since Ta = for any a , therefore T(a + b) = . Now Ta + Tb = +
a a+b a b
therefore T(a + b) ≠ Ta + Tb and T is not a linear transformation.
_________________________________________________________________
2.20 (a) A tensor T transforms every vector a into a vector Ta = m × a where m is a specified
vector. Show that T is a linear transformation and (b) If m = e1 + e2 , find the matrix of the
tensor T .
-------------------------------------------------------------------------------
Ans. (a) T(α a + β b) = m × (α a + β b) = m × α a + m × β b = α m × a + β m × b = α Ta + β Tb. Thus,
the given T is a linear transformation.
(b) Te1 = m × e1 = (e1 + e2 ) × e1 = −e3 , Te2 = m × e2 = (e1 + e2 ) × e2 = e3 ,
Te3 = m × e3 = (e1 + e2 ) × e3 = −e2 + e1 . Thus,
⎡0 0 1⎤
[T] = ⎢⎢ 0 0 −1⎥⎥ .
⎢⎣ −1 1 0 ⎥⎦
_________________________________________________________________
2.21 A tensor T transforms the base vectors e1 and e2 such that Te1 = e1 + e2 and Te2 = e1 − e2 .
If a = 2e1 + 3e2 and b = 3e1 + 2e2 , use the linear property of T to find (a) Ta ,(b) Tb , and (c)
T(a + b) .
------------------------------------------------------------------------------
Ans.
(a)Ta = T(2e1 + 3e2 ) = 2Te1 + 3Te2 = 2 ( e1 + e2 ) + 3 ( e1 − e2 ) = 5e1 − e2 .
(b)Tb = T(3e1 + 2e2 ) = 3Te1 + 2Te2 =3 ( e1 + e2 ) + 2 ( e1 − e2 ) =5e1 + e2 .
(c)T(a + b) = Ta + Tb = ( 5e1 − e2 ) + ( 5e1 + e2 ) = 10e1 .
_________________________________________________________________
2.22 Obtain the matrix for the tensor T which transforms the base vectors as follows:
Te1 = 2e1 + e3 , Te2 = e2 + 3e3 , Te3 = −e1 + 3e2 .
------------------------------------------------------------------------------
⎡ 2 0 −1⎤
Ans. [ T] = ⎢⎢ 0 1 3 ⎥⎥ .
⎢⎣1 3 0 ⎥⎦
_________________________________________________________________
2.23 Find the matrix of the tensor T which transforms any vector a into a vector b = m(a ⋅ n)
where m = ( )
2 / 2 ( e1 + e 2 ) and n = ( )
2 / 2 ( −e1 + e3 ) .
------------------------------------------------------------------------------
( )( )
Ans. Te1 = m ( e1 ⋅ n ) = n1m = − 2 / 2 ⎡ 2 / 2 ( e1 + e 2 ) ⎤ = − ( e1 + e2 ) / 2 .
⎣ ⎦
Copyright 2010, Elsevier Inc
2-6
Lai et al, Introduction to Continuum Mechanics

Te2 = m ( e 2 ⋅ n ) = n2m = 0m = 0 .
Te3 = m ( e3 ⋅ n ) = n3m = ( )( )
2 / 2 ⎡ 2 / 2 ( e1 + e 2 ) ⎤ = ( e1 + e2 ) / 2 .
⎣ ⎦
⎡ −1 / 2 0 1 / 2 ⎤
Thus, [ T] = ⎢⎢ −1 / 2 0 1 / 2 ⎥⎥ .
⎢⎣ 0 0 0 ⎥⎦
_________________________________________________________________
2.24 (a) A tensor T transforms every vector into its mirror image with respect to the plane whose
normal is e 2 . Find the matrix of T . (b) Do part (a) if the plane has a normal in the e3 direction.
------------------------------------------------------------------------------
⎡1 0 0 ⎤
Ans. (a) Te1 = e1 , Te 2 = −e 2 , Te3 = e3 , thus, [ T] = ⎢⎢0 −1 0 ⎥⎥ .
⎢⎣0 0 1 ⎥⎦
⎡1 0 0 ⎤
(b) Te1 = e1 , Te 2 = e 2 , Te3 = −e3 , thus, [ T] = ⎢⎢0 1 0 ⎥⎥ .
⎢⎣0 0 −1⎥⎦
_________________________________________________________________
2.25 (a) Let R correspond to a right-hand rotation of angle θ about the x1 -axis. Find the matrix
of R . (b) do part (a) if the rotation is about the x2 -axis. The coordinates are right-handed.
------------------------------------------------------------------------------
Ans.(a) Re1 = e1 , Re 2 = 0e1 + cosθ e 2 + sinθ e3 , Re3 = 0e1 − sinθ e 2 + cosθ e3 . Thus,
⎡1 0 0 ⎤
[ R ] = ⎢0 cosθ − sin θ ⎥⎥ .
⎢
⎢⎣0 sin θ cos θ ⎥⎦
(b) Re1 = −sinθ e3 + cosθ e1 , Re2 = e2 , Re3 = cosθ e3 + sinθ e1 . Thus,
⎡ cos θ 0 sin θ ⎤
[ R ] = ⎢⎢ 0 1 0 ⎥⎥ .
⎢⎣ − sin θ 0 cos θ ⎥⎦
_________________________________________________________________
2.26 Consider a plane of reflection which passes through the origin. Let n be a unit normal
vector to the plane and let r be the position vector for a point in space. (a) Show that the reflected
vector for r is given by Tr = r − 2(r ⋅ n)n , where T is the transformation that corresponds to the
reflection. (b) Let n = (e1 + e 2 + e3 ) / 3 , find the matrix of T . (c) Use this linear transformation to
find the mirror image of the vector a = e1 + 2e 2 + 3e3 .
------------------------------------------------------------------------------
Ans. (a) Let the vector r be decomposed into two vectors rn and rt , where rn is in the direction of
n and rt is in a direction perpendicular to n . That is, rn is normal to the plane of reflection and rt is
on the plane of reflection and r = rt + rn . In the reflection given by T , we have,
Trn = −rn and Trt = rt , so that Tr = Trt + Trn = rt − rn = ( r − rn ) − rn = r − 2rn = r − 2(r ⋅ n)n .
(b) n = (e1 + e 2 + e3 ) / 3 → e1 ⋅ n = e2 ⋅ n = e3 ⋅ n = 1 / 3 .

Copyright 2010, Elsevier Inc

2-7
Lai et al, Introduction to Continuum Mechanics

( )
Te1 = e1 − 2(e1 ⋅ n)n = e1 − 2 1 / 3 ⎡⎣ (e1 + e 2 + e3 ) / 3 ⎤⎦ = ( e1 − 2e2 − 2e3 ) / 3 .

( )
Te 2 = e 2 − 2(e 2 ⋅ n)n = e 2 − 2 1 / 3 ⎡⎣ (e1 + e 2 + e3 ) / 3 ⎤⎦ = ( −2e1 + e 2 − 2e3 ) / 3 .

( )
Te3 = e3 − 2(e3 ⋅ n)n = e3 − 2 1 / 3 ⎡⎣(e1 + e 2 + e3 ) / 3 ⎤⎦ = ( −2e1 − 2e 2 + e3 ) / 3 .
⎡ 1 −2 −2 ⎤
1⎢
[T] = ⎢ −2 1 −2⎥⎥ .
3
⎢⎣ −2 −2 1 ⎥⎦
⎡ 1 −2 −2 ⎤ ⎡ 1 ⎤ ⎡ −3⎤
1⎢
(c) [ T][a ] = ⎢ −2 1 −2 ⎥⎥ ⎢⎢ 2 ⎥⎥ = ⎢⎢ −2 ⎥⎥ → Ta = − ( 3e1 + 2e 2 + e3 ) .
3
⎢⎣ −2 −2 1 ⎥⎦ ⎢⎣ 3 ⎥⎦ ⎢⎣ −1⎥⎦
_________________________________________________________________
2.27 Knowing that the reflected vector for r is given by Tr = r − 2(r ⋅ n)n (see the previous
problem), where T is the transformation that corresponds to the reflection and n is the normal to the
mirror, show that in dyadic notation, the reflection tensor is given by T = I − 2nn and find the
matrix of T if the normal of the mirror is given by n = (e1 + e 2 + e3 ) / 3 ,
------------------------------------------------------------------------------
Ans. From the definition of dyadic product, we have ,
Tr = r − 2(r ⋅ n)n = r − 2(nn)r = (Ir − 2(nn)r ) = (I − 2nn)r → T = I − 2nn .
⎡1⎤ ⎡1 1 1⎤
2⎢ ⎥ 2⎢
For n = (e1 + e 2 + e3 ) / 3 → [2nn] = ⎢1⎥ [1 1 1] = ⎢1 1 1⎥⎥ .
3 3
⎢⎣1⎥⎦ ⎢⎣1 1 1⎥⎦
⎡ 1 −2 −2 ⎤
1⎢
→ [T] = [I ] − [2nn] = ⎢ −2 1 −2 ⎥⎥ .
3
⎢⎣ −2 −2 1 ⎥⎦
_________________________________________________________________
2.28 A rotation tensor R is defined by the relation Re1 = e 2 , Re 2 = e3 , Re3 = e1 (a) Find the
matrix of R and verify that R T R = I and det R = 1 and (b) find a unit vector in the direction of the
axis of rotation that could have been used to effect this particular rotation.
------------------------------------------------------------------------------
⎡0 0 1 ⎤ ⎡ 0 1 0 ⎤ ⎡0 0 1 ⎤ ⎡1 0 0 ⎤ 0 0 1
⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢ ⎥
Ans. (a) [ R ] = ⎢1 0 0 ⎥ → [ R ] [ R ] = ⎢ 0 0 1 ⎥ ⎢1 0 0 ⎥ = ⎢ 0 1 0 ⎥ , det [ R ] = 1 0 0 = 1 .
T

⎢⎣0 1 0 ⎥⎦ ⎢⎣1 0 0 ⎥⎦ ⎢⎣0 1 0 ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦ 0 1 0

(b) Let the axis of rotation be n = α1e1 + α 2e2 + α 3e3 , then
⎡ −1 0 1 ⎤ ⎡ α1 ⎤ ⎡0 ⎤
Rn = n → [ R − I ][n ] = [ 0] → ⎢⎢ 1 −1 0 ⎥⎥ ⎢⎢α 2 ⎥⎥ = ⎢⎢0 ⎥⎥ → −α1 + α 3 = 0, α1 − α 2 = 0, α 2 − α 3 =0 .
⎢⎣ 0 1 −1⎥⎦ ⎢⎣α 3 ⎥⎦ ⎢⎣0 ⎥⎦
Thus, α1 = α 2 = α 3 , so that a unit vector in the direction of the axis of rotation is
n = (e1 + e 2 + e3 ) / 3 .
_________________________________________________________________

Copyright 2010, Elsevier Inc

2-8
Lai et al, Introduction to Continuum Mechanics

2.29 A rigid body undergoes a right hand rotation of angle θ about an axis which is in the
direction of the unit vector m . Let the origin of the coordinates be on the axis of rotation and r be
the position vector for a typical point in the body. (a) show that the rotated vector of r is given by:
Rr = (1 − cos θ )( m ⋅ r ) m + cos θ r + sin θ ( m × r ) , where R is the rotation tensor. (b) Let
m = (e1 + e 2 + e3 ) / 3 , find the matrix for R .
------------------------------------------------------------------------------
Ans. (a) Let the vector r be decomposed into two vectors rm and r p , where rm is in the direction
of m and r p is in a direction perpendicular to m , that is, r = r p + rm . Let p ≡ rp / rp be the unit
vector in the direction of r p , and let q ≡ m × p . Then, ( m , p, q) forms an orthonormal set of
vectors which rotates an angle of θ about the unit vector m . Thus,
Rrm = rm and Rrp = r p ( cos θ p + sin θ q ) . From r = rp + rm , we have,

{
Rr = Rrp + Rrm = r p ( cosθ p + sin θ q ) + rm = cosθ r p p + sin θ rp ( m × p ) + rm }
{ (
= cos θ r p + sin θ m × rp )} + rm = {cosθ ( r − rm ) + sin θ ( m × ( r − rm ) )} + rm
= r cos θ + rm (1 − cos θ ) + sin θ m × ( r − rm ) = r cosθ + rm (1 − cosθ ) + sin θ m × r
We note that rm = (r ⋅ m)m , so that Rr = r cos θ + (r ⋅ m)m (1 − cos θ ) + sin θ m × r .
(b) Use the result of (a), that is, Rr = r cos θ + (r ⋅ m) (1 − cos θ ) + sin θ m × r , we have,
Re1 = e1 cos θ + (e1 ⋅ m)m (1 − cos θ ) + sin θ m × e1 ,
Re2 = e 2 cos θ + (e 2 ⋅ m)m (1 − cos θ ) + sin θ m × e 2 ,
Re3 = e3 cos θ + (e3 ⋅ m )m (1 − cos θ ) + sin θ m × e3 .
Now, m = (e1 + e 2 + e3 ) / 3 , therefore, m ⋅ e1 = m ⋅ e 2 = m ⋅ e3 = 1 / 3
( ) ( ) ( )
m × e1 = 1 / 3 (−e3 + e 2 ), m × e 2 = 1 / 3 (e3 − e1 ), m × e3 = 1 / 3 ( −e2 + e1 ) . Thus,
Re1 = e1 cos θ + (e1 ⋅ m)m (1 − cosθ ) + sin θ m × e1

( )
= e1 cosθ + (1 / 3) (e1 + e2 + e3 ) (1 − cos θ ) + sin θ 1 / 3 (−e3 + e2 )

{ ( )} { (
= (1 / 3){1 + 2cos θ } e1 + e 2 (1 / 3)(1 − cosθ ) + sin θ 1 / 3 + e3 (1 / 3)(1 − cosθ ) − sin θ 1 / 3 )}
Re2 = e 2 cos θ + (e 2 ⋅ m)m (1 − cosθ ) + sin θ m × e 2

( )
= e 2 cosθ + (1 / 3) (e1 + e 2 + e3 ) (1 − cosθ ) + sin θ 1 / 3 (e3 - e1 )

{ ( ) } { (
= (1 / 3)(1 − cos θ ) − 1 / 3 sin θ e1 + (1 / 3)(1 + 2cosθ ) e 2 + (1 / 3)(1 − cosθ ) + sin θ 1 / 3 e3 )}
Re3 = e3 cos θ + (e3 ⋅ m)m (1 − cosθ ) + sin θ m × e3

( )
= e3 cosθ + (1 / 3) (e1 + e2 + e3 ) (1 − cosθ ) + sin θ 1 / 3 (-e2 + e1 )

{ ( ) } { ( )}
= (1 / 3)(1 − cos θ ) + 1 / 3 sin θ e1 + (1 / 3)(1 − cosθ ) − sin θ 1 / 3 e2 + (1 / 3)(1 + 2cosθ ) e3
Thus,

Copyright 2010, Elsevier Inc

2-9
Lai et al, Introduction to Continuum Mechanics

⎡ 1 + 2cos θ (1 − cosθ ) − 3 sin θ (1 − cosθ ) + 3 sin θ ⎤

1⎢ ⎥
[T] = ⎢(1 − cosθ ) + 3 sin θ (1 + 2cosθ ) (1 − cosθ ) − 3 sin θ ⎥ .
3⎢ ⎥
⎢⎣(1 − cos θ ) − 3 sin θ (1 − cosθ ) + 3 sin θ (1 + 2cosθ ) ⎥⎦
_________________________________________________________________
2.30 For the rotation about an arbitrary axis m by an angle θ , (a) show that the rotation tensor is
given by R = (1 − cosθ )(mm ) + cosθ I + sin θ E , where mm denotes that dyadic product of m and
E is the antisymmetric tensor whose dual vector (or axial vector) is m , (b) find the R A , the
antisymmetric part of R and (c) show that the dual vector for R A is given by (sin θ )m . Hint,
Rr = (1 − cos θ )( m ⋅ r ) m + cosθ r + sin θ ( m × r ) (see previous problem).
------------------------------------------------------------------------------
Ans. (a) We have, from the previous problem, Rr = (1 − cos θ )( m ⋅ r ) m + cosθ r + sin θ ( m × r ) .
Now, by the definition of dyadic product, we have ( m ⋅ r ) m = (mm)r , and by the definition of dual
vector we have, m × r = Er , thus Rr = (1 − cos θ ) (mm)r + cos θ r + sin θ Er
= {(1 − cos θ ) (mm) + cos θ I + sin θ E} r , from which, R = (1 − cos θ ) (mm) + cos θ I + sin θ E .
(b) R A = ( R − R T ) / 2 →
{
2R A = {(1 − cosθ ) (mm) + cosθ I + sin θ E} − (1 − cosθ ) (mm)T + cos θ I + sin θ ET . Now }
[mm ] = ⎡⎣ mi m j ⎤⎦ = ⎡⎣ m j mi ⎤⎦ = [mm ]
T
, and the tensor E , being antisymmetric, E = −ET , therefore,
2R A = 2sin θ E , that is, R A = sin θ E .
(c) dual vector of R A = (sin θ )(dual vector of E) = sin θ m .
_________________________________________________________________
2.31 (a) Given a mirror whose normal is in the direction of e 2 . Find the matrix of the tensor S
which first transforms every vector into its mirror image and then transforms them by a 45o right-
hand rotation about the e1 -axis. (b) Find the matrix of the tensor T which first transforms every
vector by a 45o right-hand rotation about the e1 -axis, and then transforms them by a reflection with
respect to the mirror (whose normal is e2 ). (c) Consider the vector a = (e1 + 2e2 + 3e3 ) , find the
transformed vector by using the transformation S .
(d) For the same vector a = (e1 + 2e2 + 3e3 ) , find the transformed vector by using the
transformation T .
------------------------------------------------------------------------------
Ans. Let T1 and T2 correspond to the reflection and the rotation respectively. We have
⎡1 0 0 ⎤
T1e1 = e1 , T1e2 = −e2 , T1e3 = e3 → [ T1 ] = ⎢⎢0 −1 0 ⎥⎥ .
⎢⎣0 0 1 ⎥⎦
⎡1 0 0 ⎤
1 1 ⎢ ⎥
T2e1 = e1 , T2e 2 = ( e2 + e3 ) , T2e3 = ( −e2 + e3 ) → [T2 ] = ⎢0 1 / 2 −1 / 2 ⎥ .
2 2 ⎢0 1 / 2
⎣ 1 / 2 ⎥⎦

Copyright 2010, Elsevier Inc

2-10
Lai et al, Introduction to Continuum Mechanics

⎡1 0 ⎤ ⎡1 0 0 ⎤ ⎡1
0 0 0 ⎤
⎢ ⎥⎢ ⎥ ⎢ ⎥
(a) [S ] = [ T2 ][ T1 ] = ⎢0 1 / 2 −1 / 2 ⎥ ⎢0 −1 0 ⎥ = ⎢0 −1 / 2 −1 / 2 ⎥ .
⎢0 1 / 2 1 / 2 ⎥⎦ ⎢⎣0 0 1 ⎥⎦ ⎢⎣0 −1 / 2 1 / 2 ⎥⎦
⎣
⎡1 0 0 ⎤ ⎡1 0 0 ⎤ ⎡1 0 0 ⎤
⎢ ⎥ ⎢ ⎥ ⎢ ⎥
(b) [ T] = [ T1 ][ T2 ] = ⎢0 −1 0 ⎥ ⎢0 1 / 2 −1 / 2 ⎥ = ⎢ 0 −1 / 2 1 / 2 ⎥ .
⎢⎣0 0 1 ⎥⎦ ⎢0 1 / 2 1 / 2 ⎥ ⎢ 0 1 / 2 1 / 2 ⎥
⎣ ⎦ ⎣ ⎦
⎡1 0 0 ⎤ ⎡1 ⎤ ⎡ 1 ⎤
⎢ ⎥ ⎢ ⎥
(c) [b ] = [S ][a ] = ⎢0 −1 / 2 −1 / 2 ⎥ ⎢⎢ 2 ⎥⎥ = ⎢ −5 / 2 ⎥ .
⎢0 −1 / 2 1 / 2 ⎥ ⎢ 3 ⎥ ⎢ 1 / 2 ⎥
⎣ ⎦⎣ ⎦ ⎣ ⎦
⎡1 0 0 ⎤ ⎡1 ⎤ ⎡ 1 ⎤
⎢ ⎥ ⎢ ⎥
(d) [c ] = [ T ][a ] = ⎢0 −1 / 2 1 / 2 ⎥ ⎢⎢ 2 ⎥⎥ = ⎢1 / 2 ⎥ .
⎢ 0 1 / 2 1 / 2 ⎥ ⎢ 3 ⎥ ⎢5 / 2 ⎥
⎣ ⎦⎣ ⎦ ⎣ ⎦
_________________________________________________________________
2.32 Let R correspond to a right-hand rotation of angle θ about the x3 -axis (a) find the matrix
of R 2 . (b) Show that R 2 corresponds to a rotation of angle 2θ about the same axis (c) Find the
matrix of R n for any integer n .
-------------------------------------------------------------------------------
⎡cos θ − sin θ 0 ⎤
Ans. (a) [ R ] = ⎢⎢ sin θ cosθ 0 ⎥⎥ .
⎢⎣ 0 0 1 ⎥⎦

0 ⎤ ⎡cos θ − sin θ 0⎤
2 2
⎡cosθ − sin θ 0 ⎤ ⎡cosθ − sin θ −2sin θ cosθ
⎢ ⎥
→ ⎡ R ⎤ = ⎢⎢ sin θ
2
cos θ 0 ⎥⎥ ⎢⎢ sin θ cosθ 0⎥⎥ = ⎢ 2sin θ cosθ cos 2 θ − sin 2 θ 0⎥ .
⎣ ⎦ ⎢ ⎥
⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ 0 0 1 ⎥⎦ ⎢ 0 0 1⎥
⎣ ⎦
(b)
⎡ cos 2 θ − sin 2 θ −2sin θ cosθ 0 ⎤ ⎡ cos 2θ − sin 2θ 0⎤
⎢ ⎥
⎡ R 2 ⎤ = ⎢ 2sin θ cosθ 2
cos θ − sin θ 2
0 ⎥ = ⎢⎢ sin 2θ cos 2θ 0 ⎥⎥ .
⎣ ⎦ ⎢ ⎥
⎢⎣ 0 0 1 ⎥ ⎢⎣ 0 0 1 ⎥⎦
⎦
Thus, R 2 corresponds to a rotation of angle 2θ about the same axis
⎡cos nθ − sin nθ 0 ⎤
(c) ⎡ R ⎤ = ⎢⎢ sin nθ cos nθ 0 ⎥⎥ .
n
⎣ ⎦
⎢⎣ 0 0 1 ⎥⎦
_________________________________________________________________
2.33 Rigid body rotations that are small can be described by an orthogonal
transformation R = I + ε R* where ε → 0 as the rotation angle approaches zero. Consider two
successive small rotations R1 and R 2 , show that the final result does not depend on the order of
rotations.
------------------------------------------------------------------------------

Copyright 2010, Elsevier Inc

2-11
Lai et al, Introduction to Continuum Mechanics

( )( ) ( )
Ans. R 2 R1 = I + ε R*2 I + ε R1* = I + ε R*2 + ε R1* + ε 2 R*2 R1* = I + ε R*2 + R1* + ε 2 R*2 R1* .

(
As ε → 0 , R 2 R1 ≈ I + ε R*2 + R1* = R1R 2 . )
_________________________________________________________________
2.34 Let T and S be any two tensors. Show that (a) TT is a tensor, (b) TT + S T = (T + S)T and (c)
(TS)T = S T TT .
-------------------------------------------------------------------------------
Ans. Let a , b , c be three arbitrary vectors and α β be any two scalars, then
(a) a ⋅ TT (α b + β c) = (α b + β c) ⋅ Ta = α b ⋅ Ta + β c ⋅ Ta = α a ⋅ TT b + β a ⋅ TT c
( ) ( )
= a ⋅ α TT b + β TT c → TT (α b + β c) = α TT b + β TT c . Thus, TT is a linear transformation, i.e.,
tensor.
(b) a ⋅ (T + S)T b = b ⋅ (T + S)a = b ⋅ Ta + b ⋅ Sa = a ⋅ TT b + a ⋅ S T b
= a ⋅ (TT + S T )b → (T + S)T = TT + S T .
(c) a ⋅ (TS)T b = b ⋅ (TS)a = b ⋅ T(Sa) = (Sa) ⋅ TT b = a ⋅ S T TT b → (TS)T = S T TT .
_________________________________________________________________
2.35 For arbitrary tensors T and S , without relying on the component form, prove that (a)
(T−1 )T = (TT )−1 and (b) (TS)−1 = S −1T−1
-------------------------------------------------------------------------------
Ans. (a) TT−1 = I → (TT−1 )T = I → (T−1 )T TT = I → (T−1 )T = (TT )−1 .
(b) (TS)(S −1T−1 ) = T(SS −1 )T−1 = TT−1 = I , thus, (TS)−1 = S −1T−1 .
_________________________________________________________________
2.36 Let {ei } and {e′i } be two Rectangular Cartesian base vectors. (a) Show that if e′i = Qmi em ,
then ei = Qim e′m and (b) verify Qmi Qmj = δ ij = Qim Q jm .
-------------------------------------------------------------------------------
Ans. (a) e′i = Qmi e m → e′i ⋅ e j = Qmi e m ⋅ e j = Qmiδ mj = Q ji → e j = Q jm e′m → ei = Qim e′m .
(b) We have, e′i ⋅ e′j = δ ij = ei ⋅ e j , thus,
δ ij = e′i ⋅ e′j = Qmi e m ⋅ Qnj e n = Qmi Qnj em ⋅ en = Qmi Qnjδ mn = Qmi Qmj . And
δ ij = ei ⋅ e j = Qimem ⋅ Q jn en = QimQ jn em ⋅ en = QimQ jnδ mn = QimQ jm .
_________________________________________________________________
2.37 The basis {e′i } is obtained by a 30o counterclockwise rotation of the {ei } basis about the e3
axis. (a) Find the transformation matrix [Q ] relating the two sets of basis, (b) by using the vector
transformation law, find the components of a = 3e1 + e 2 in the primed basis, i.e., find ai′ and (c)
do part (b) geometrically.
------------------------------------------------------------------------------
Ans. (a) e1′ = cos30o e1 + sin 30o e2 , e′2 = − sin 30o e1 + cos30o e2 , e′3 = e3 . Thus,
⎡ cos30o − sin 30o 0⎤
⎢ ⎥
[Q ]ei = ⎢ sin 30o cos30o 0⎥ .
⎢ ⎥
⎢⎣ 0 0 1⎥
⎦
Copyright 2010, Elsevier Inc
2-12
Lai et al, Introduction to Continuum Mechanics

⎡ a1' ⎤ ⎡ 3 / 2 1/ 20⎤ ⎡ 3 ⎤ ⎡ 2⎤
⎢ ⎥ ⎢ ⎥⎢ ⎥
(b) [a ]e' = [Q ] [a ]e 3 / 2 0 ⎥ ⎢ 1 ⎥ = ⎢⎢ 0 ⎥⎥ → a = 2e1′
T
→ ⎢ a2' ⎥ = ⎢ −1 / 2
⎢ ⎥ ⎢
1 ⎥⎥ ⎢⎣ 0 ⎥⎦ ⎢⎣ 0 ⎥⎦
i i

⎢⎣ a3' ⎥⎦ ⎢⎣ 0 0
⎦
(c) Clearly a = 3e1 + e 2 is a vector in the same direction as e1′ and has a length of 2. See figure
below

_________________________________________________________________
2.38 Do the previous problem with the {e′i } basis obtained by a 30o clockwise rotation of the
{ei } basis about the e3 axis.
-------------------------------------------------------------------------------
Ans.
(a) e1′ = cos30o e1 − sin 30o e 2 , e′2 = sin 30o e1 + cos30o e 2 , e′3 = e3 . Thus,
⎡ cos30o sin 30o 0⎤
⎢ ⎥
[Q ]ei = ⎢ − sin 30o cos30o 0 ⎥ .
⎢ ⎥
⎢⎣ 0 0 1⎥
⎦
⎡ a1' ⎤ ⎡ 3 / 2 −1 / 2 0⎤ ⎡ 3 ⎤ ⎡ 1 ⎤
⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥
(b) [a ]e' = [Q ] [a ]e → ⎢ a2' ⎥ = ⎢ 1 / 2
T
3 / 2 0 ⎥ ⎢ 1 ⎥ = ⎢ 3 ⎥ → a = e1′ + 3e′2
⎢ ⎥ ⎢
1 ⎥⎥ ⎢⎣ 0 ⎥⎦ ⎢⎣ 0 ⎥⎦
i i

⎢⎣ a3' ⎥⎦ ⎢⎣ 0 0
⎦
(c) See figure below

_________________________________________________________________
2.39 The matrix of a tensor T with respect to the basis {ei } is
Copyright 2010, Elsevier Inc
2-13
Lai et al, Introduction to Continuum Mechanics

⎡ 1 5 −5⎤
[T] = ⎢⎢ 5 0 0 ⎥⎥
⎢⎣ −5 0 1 ⎥⎦
′ with respect to a right-handed basis {e′i } where e1′ is in the direction of
Find T11′ , T12′ and T31
−e 2 + 2e3 and e′2 is in the direction of e1 .
------------------------------------------------------------------------------
Ans. The basis {e′i } is given by:
e1′ = (−e 2 + 2e3 ) / 5, e′2 = e1 , e′3 = e1′ × e′2 = (2e2 + e3 ) / 5 .
⎡ 1 5 −5⎤ ⎡ 0 ⎤
⎢ ⎥
T11′ = e1′ ⋅ Te1′ = ⎡⎣0 −1 / 5 2 / 5 ⎤⎦ ⎢⎢ 5 0 0 ⎥⎥ ⎢ −1 / 5 ⎥ = 4 / 5 .
⎢⎣ −5 0 1 ⎥⎦ ⎢ 2 / 5 ⎥
⎣ ⎦
⎡ 1 5 −5⎤ ⎡1 ⎤
T12′ = e1′ ⋅ Te′2 = ⎡⎣ 0 −1 / 5 2 / 5 ⎤⎦ ⎢⎢ 5 0 0 ⎥⎥ ⎢⎢0 ⎥⎥ = −15 / 5 .
⎢⎣ −5 0 1 ⎥⎦ ⎢⎣0 ⎥⎦

⎡ 1 5 −5⎤ ⎡ 0 ⎤
⎢ ⎥
′ = e′3 ⋅ Te1′ = ⎡0 2 / 5 1 / 5 ⎤ ⎢ 5 0 0 ⎥ ⎢ −1 / 5 ⎥ = 2 / 5 .
T31 ⎣ ⎦⎢ ⎥
⎢⎣ −5 0 1 ⎥⎦ ⎢ 2 / 5 ⎥
⎣ ⎦
_________________________________________________________________
2.40 (a) For the tensor of the previous problem, find ⎡⎣Tij′ ⎤⎦ , i.e., [ T]e' if {e′i } is obtained by a
i
o
90 right hand rotation about the e3 axis and (b) obtain Tii′ and the determinant Tij′ and compare
them with Tii and Tij .
------------------------------------------------------------------------------
⎡ 0 −1 0 ⎤
Ans. (a) e1′ = e 2 , e′2 = −e1 , e′3 = e3 → [Q ] = ⎢⎢1 0 0 ⎥⎥ .
⎢⎣ 0 0 1 ⎥⎦
⎡ 0 1 0 ⎤ ⎡ 1 5 −5⎤ ⎡0 −1 0 ⎤ ⎡ 0 −5 0 ⎤
⎡⎣Tij′ ⎤⎦ = [ T] = [Q ] [ T][Q ] = ⎢ −1 0 0 ⎥ ⎢ 5 0 0 ⎥ ⎢1 0 0 ⎥ = ⎢ −5 1 5 ⎥
′ T
⎢ ⎥⎢ ⎥⎢ ⎥ ⎢ ⎥
⎢⎣ 0 0 1 ⎥⎦ ⎢⎣ −5 0 1 ⎥⎦ ⎢⎣0 0 1 ⎥⎦ ⎢⎣ 0 5 1 ⎥⎦
(b) Tii′ = T11′ + T22′ + T33′ = 0 + 1 + 1 = 2, Tij′ = −25 .
Tii = T11 + T22 + T33 = 1 + 0 + 1 = 2, Tij = −25 .
_________________________________________________________________
2.41 The dot product of two vectors a = ai ei and b = bi ei is equal to ai bi . Show that the dot
product is a scalar invariant with respect to orthogonal transformations of coordinates.
-------------------------------------------------------------------------------
Ans. From ai′ = Qmi am and bi' = Qmi bm , we have,
ai′bi' = Qmi am Qni bn = Qmi Qni am bn = δ mn am bn = am bm = ai bi .
__________________________________________________________________

Copyright 2010, Elsevier Inc

2-14
Lai et al, Introduction to Continuum Mechanics

2.42 If Tij are the components of a tensor (a) show that TijTij is a scalar invariant with respect to
orthogonal transformations of coordinates, (b) evaluate TijTij with respect to the basis {ei } for
⎡1 0 0 ⎤ ⎡0 0 1 ⎤
[T] = ⎢1 2 5⎥ , (c) find [T] , if e′i = Qei , where [Q ] = ⎢⎢1 0 0⎥⎥ and
⎢ ⎥ ′
⎢⎣1 2 3⎥⎦ e ⎢⎣ 0 1 0 ⎥⎦ e
i i

(d) verify for the above [ T] and [ T]′ that Tij′Tij′ = TijTij .
------------------------------------------------------------------------------
Ans. (a) Since Tij are the components of a tensor, Tij′ = Qmi QnjTmn . Thus,
Tij′Tij′ = Qmi Qnj Tmn (Q pi QqjT pq ) = (Qmi Q pi )(Qnj Qqj )TmnT pq = δ mpδ nqTmnTpq = TmnTmn
(b) TijTij = T112 + T122 + T132 + T21
2 2
+ T22 2
+ T23 + T312 + T322 + T332 = 1 + 1 + 4 + 25 + 1 + 4 + 9 = 45 .
⎡ 0 1 0 ⎤ ⎡1 0 0 ⎤ ⎡0 0 1 ⎤ ⎡ 0 1 0⎤ ⎡ 0 0 1⎤ ⎡ 2 5 1⎤
(c) [ T]′ = [Q ] [ T][Q ] = ⎢⎢ 0 0 1 ⎥⎥ ⎢⎢1 2 5 ⎥⎥ ⎢⎢1 0 0 ⎥⎥ = ⎢⎢ 0 0 1 ⎥⎥ ⎢⎢ 2 5 1⎥⎥ = ⎢⎢ 2 3 1⎥⎥
T

⎢⎣1 0 0 ⎥⎦ ⎢⎣1 2 3⎥⎦ ⎢⎣0 1 0 ⎥⎦ ⎢⎣1 0 0⎥⎦ ⎢⎣ 2 3 1⎥⎦ ⎢⎣ 0 0 1⎥⎦

(d) Tij′Tij′ = 4 + 25 + 1 + 4 + 9 + 1 + 1 = 45 .
_________________________________________________________________
2.43 Let [ T] and [ T]′ be two matrices of the same tensor T , show that det [ T] =det [ T]′ .
------------------------------------------------------------------------------
Ans. ⎡T ⎤′ = ⎡Q ⎤ ⎡T ⎤ ⎡Q ⎤ → det ⎡T ⎤′ = det ⎡Q ⎤ det ⎡Q ⎤ det ⎡ T⎤ = (±1)(±1) det ⎡ T⎤ = det ⎡ T⎤ .
T T
⎣ ⎦ ⎣ ⎦ ⎣ ⎦⎣ ⎦ ⎣ ⎦ ⎣ ⎦ ⎣ ⎦ ⎣ ⎦ ⎣ ⎦ ⎣ ⎦
_________________________________________________________________
2.44 (a) If the components of a third order tensor are Rijk , show that Riik are components of a
vector, (b) if the components of a fourth order tensor are Rijkl , show that Riikl are components of a
second order tensor and (c) what are components of Riik ... , if Rijk ... are components of a tensor of
nth order?
-------------------------------------------------------------------------------
Ans. (a) Since Rijk are components of a third order tensor, therefore,
′ = Qmi Qnj Q pk Rmnp → Riik
Rijk ′ = Qmi Qni Q pk Rmnp = δ mn Q pk Rmnp = Q pk Rnnp , therefore, Riik are
components of a vector.
(b) Consider a 4th order tensor Rijkl , we have,
′ = Qmi Qnj Q pk Qql Rmnpq → Riikl
Rijkl ′ = Qmi Qni Q pk Qql Rmnpq = δ m n Q pk Qql Rm npq = Q pk Qql Rnnpq ,
therefore, Riikl are components of a second order tensor.
(c) Riik ... are components of a tensor of the (n − 2)th order.
_________________________________________________________________
2.45 The components of an arbitrary vector a and an arbitrary second tensor T are related by a
triply subscripted quantity Rijk in the manner ai = Rijk T jk for any rectangular Cartesian basis {ei } .
Prove that Rijk are the components of a third-order tensor.
------------------------------------------------------------------------------

Copyright 2010, Elsevier Inc

2-15
Lai et al, Introduction to Continuum Mechanics

Ans. Since ai = Rijk T jk is true for any basis, therefore, ai′ = Rijk
′ T jk′ ; Since a is a vector, therefore,
ai′ = Qmi am and since T is a second order tensor, therefore, Tij′ = Qmi QnjTmn . Thus,
ai′ = Qmi am → Rijk
′ T jk′ = Qmi ( Rmjk T jk ) . Multiply the last equation with Qsi and noting that
Qsi Qmi = δ sm , we have,
' T ' = Q Q ( R T ) → Q R' T ' = δ R T → Q R' T ' = R T
Qsi Rijk jk si mi mjk jk si ijk jk sm mjk jk si ijk jk sjk jk

→ Qsi Rijk mj nk mn = Rsjk T jk → Qsi Rijk Qmj Qnk Tmn = RsmnTmn . Thus,
' Q Q T '

(R smn
'
)
− Qsi Qmj Qnk Rijk Tmn = 0 . Since this last equation is to be true for all Tmn , therefore,
′ , which is the transformation law for components of a third order tensor.
Rsmn = Qsi Qmj Qnk Rijk
_________________________________________________________________
2.46 For any vector a and any tensor T , show that (a) a ⋅ TA a = 0 and (b) a ⋅ Ta = a ⋅ TSa ,
where TA and TS are antisymmetric and symmetric part of T respectively.
------------------------------------------------------------------------------
Ans. (a) TA is antisymmetric, therefore, (TA )T = −TA , thus,
a ⋅ TA a = a ⋅ (TA )T a = −a ⋅ TA a → 2a ⋅ TA a = 0 → a ⋅ T A a = 0 .
(b) Since T = TS + TA , therefore, a ⋅ Ta = a ⋅ (TS + TA )a = a ⋅ TSa + a ⋅ TA a = a ⋅ TSa .
__________________________________________________________________
2.47 Any tensor can be decomposed into a symmetric part and an antisymmetric part, that is
T = TS + TA . Prove that the decomposition is unique. (Hint, assume that it is not true and show
contradiction).
-------------------------------------------------------------------------------
Ans. Suppose that the decomposition is not unique, then ,we have,
T = TS + TA = SS + S A → (TS − SS ) + (TA − S A ) = 0 . Let a be any arbitrary vector, we have,
a ⋅ (TS − SS )a + a ⋅ (TA − S A )a = 0 → a ⋅ TSa − a ⋅ SSa + a ⋅ TA a − a ⋅ S A a = 0 .
But a ⋅ T A a = a ⋅ S A a = 0 (see the previous problem). Therefore,
a ⋅ TSa − a ⋅ SSa = 0 → a ⋅ (TS − SS )a = 0 → TS − SS = 0 → TS = SS . It also follows from
(TS − SS ) + (TA − S A ) = 0 that TA = S A . Thus , the decomposition is unique.
_________________________________________________________________
⎡1 2 3⎤
2.48 Given that a tensor T has the matrix [ T] = ⎢⎢ 4 5 6 ⎥⎥ , (a) find the symmetric part and the
⎢⎣ 7 8 9 ⎥⎦
anti-symmetric part of T and (b) find the dual vector (or axial vector) of the anti-symmetric part of
T.
-------------------------------------------------------------------------------
⎧ ⎡1 2 3⎤ ⎡1 4 7 ⎤ ⎫ ⎡ 2 6 10 ⎤ ⎡1 3 5 ⎤
⎡ S⎤
⎣ ⎦ 2
1
{ }
T 1 ⎪⎢
2⎪
⎢ ⎥ ⎪ 1⎢
Ans. (a) T = [ T] + [ T] = ⎨ ⎢ 4 5 6 ⎥ + ⎢ 2 5 8 ⎥ ⎬ = ⎢ 6 10 14 ⎥⎥ = ⎢⎢3 5 7 ⎥⎥ .
⎥
⎪ 2 ⎢10 14 18⎥ ⎢5 7 9 ⎥
⎩ ⎣⎢7 8 9 ⎦⎥ ⎣⎢ 3 6 9 ⎦⎥ ⎭ ⎣ ⎦ ⎣ ⎦

Copyright 2010, Elsevier Inc

2-16
Lai et al, Introduction to Continuum Mechanics

⎧ ⎡ 1 2 3⎤ ⎡ 1 4 7 ⎤ ⎫ ⎡ 0 −2 −4 ⎤ ⎡ 0 −1 −2⎤
⎣ ⎦ 2 { }
⎡ TA ⎤ = 1 [ T] − [ T]T = 1 ⎪⎨ ⎢ 4 5 6 ⎥ − ⎢ 2 5 8 ⎥ ⎪⎬ = 1 ⎢ 2 0 −2 ⎥ = ⎢ 1 0 −1⎥ .
2 ⎪⎢ ⎥ ⎢ ⎥ ⎢ ⎥ ⎢
⎪ 2 ⎢ 4 2 18 ⎥ ⎢ 2 1 0 ⎥
⎥
⎩⎣⎢ 7 8 9 ⎥
⎦ ⎣ ⎢ 3 6 9 ⎥
⎦⎭ ⎣ ⎦ ⎣ ⎦
(b) t A = −(T23 A
e1 + T31A e2 + T12A e3 ) = −(−1e1 + 2e 2 − 1e3 ) = e1 − 2e2 + e3 .
_________________________________________________________________
2.49 Prove that the only possible real eigenvalues of an orthogonal tensor Q are λ = ±1 . Explain
the direction of the eigenvectors corresponding to them for a proper orthogonal(rotation) tensor and
for an improper orthogonal (reflection) tensor.
------------------------------------------------------------------------------
Ans. Since Q is orthogonal, therefore, for any vector n , we have, Qn ⋅ Qn = n ⋅ n . Let n be an
eigenvector, then Qn = λ n , so that Qn ⋅ Qn = n ⋅ n →
λ 2 (n ⋅ n) = (n ⋅ n) → (λ 2 − 1)(n ⋅ n) = 0 → λ 2 − 1 = 0 → λ = ±1 .
The eigenvalue λ = 1 ( Qn = n ) corresponds to an eigenvector parallel to the axis of rotation for a
proper orthogonal tensor (rotation tensor); Or, it corresponds to an eigenvector parallel to the plane
of reflection for an improper orthogonal tensor (reflection tensor). The eigenvalue λ = −1 ,
( Qn = −n ) corresponds to an eigenvector perpendicular to the axis of rotation for an 180o rotation;
or, it corresponds to an eigenvector perpendicular to the plane of reflection.
_________________________________________________________________
⎡ 1 −2 −2 ⎤
1⎢
2.50 Given the improper orthogonal tensor [Q ] = ⎢ −2 1 −2 ⎥⎥ . (a) Verify that det [Q ] = −1 .
3
⎢⎣ −2 −2 1 ⎥⎦
(b) Verify that the eigenvalues are λ = 1 and − 1 (c) Find the normal to the plane of reflection (i.e.,
eigenvectors corresponding to λ = −1 ) and (d) find the eigenvectors corresponding λ = 1 (vectors
parallel to the plane of reflection).
-------------------------------------------------------------------------------
Ans. (a) det [Q ] = (1 / 3) (1 − 8 − 8 − 4 − 4 − 4) = (−27) / 27 = −1 .
3

(b) I1 = 3 / 3 = 1, I 2 = (1 / 3) {(1 − 4) + (1 − 4) + (1 − 4)} = −1,

2
I 3 = −1 →
λ 3 − λ 2 − λ + 1 = 0 → (λ − 1)(λ 2 − 1) = 0 → λ = 1,1, − 1
(c) For λ = −1 ,
⎛1 ⎞ 2 2 2 ⎛1 ⎞ 2 2 2 ⎛1 ⎞
⎜ + 1⎟ α1 − α 2 − α 3 = 0, − α1 + ⎜ + 1⎟ α 2 − α 3 = 0, − α1 − α 2 + ⎜ + 1⎟ α 3 = 0 . That
⎝3 ⎠ 3 3 3 ⎝3 ⎠ 3 3 3 ⎝3 ⎠
is, 2α1 − α 2 − α 3 = 0, − α1 + 2α 2 − α 3 = 0, − α1 − α 2 + 2α 3 = 0 , thus, α1 = α 2 = α 3 , therefore,
n = ± (e1 + e 2 + e3 ) / 3 , this is the normal to the plane of reflection.
(d) For λ = 1 ,
⎛1 ⎞ 2 2 2 ⎛1 ⎞ 2 2 2 ⎛1 ⎞
⎜ − 1⎟ α1 − α 2 − α 3 = 0, − α1 + ⎜ − 1⎟ α 2 − α 3 = 0, − α1 − α 2 + ⎜ − 1⎟ α 3 = 0
⎝ 3 ⎠ 3 3 3 ⎝ 3 ⎠ 3 3 3 ⎝3 ⎠
All three equations lead to α1 + α 2 + α 3 = 0 → α 3 = −α1 − α 2 . Thus,
1 1
n= [α1e1 + α 2e 2 − (α1 + α 2 )e3 ] , e.g., n = (e1 + e2 − 2e3 ) etc. these vectors are all
α12 + α 22 + α 32 6
perpendicular to n = ±(e1 + e2 + e3 ) / 3 and thus parallel to the plane of reflection.
_________________________________________________________________
Copyright 2010, Elsevier Inc
2-17
Lai et al, Introduction to Continuum Mechanics

2.51 Given that tensors R and S have the same eigenvector n and corresponding eigenvalue
r1 and s1 respectively. Find an eigenvalue and the corresponding eigenvector for the tensor T = RS .
------------------------------------------------------------------------------
Ans. We have, Rn = r1n and Sn = s1n , thus, Tn = RSn = Rs1n = s1Rn = r1s1n . Thus, an eigenvalue
for T = RS is r1s1 with eigenvector n .
_________________________________________________________________
2.52 Show that if n is a real eigenvector of an antisymmetric tensor T , then the corresponding
eigenvalue vanishes.
------------------------------------------------------------------------------
Ans. Tn = λ n → n ⋅ Tn = λ (n ⋅ n) . Now, from the definition of transpose, we have n ⋅ Tn = n ⋅ TT n .
But, since T is antisymmetric, i.e., TT = −T , therefore, n ⋅ TT n = −n ⋅ Tn . Thus,
n ⋅ Tn = −n ⋅ Tn → 2n ⋅ Tn = 0 → n ⋅ Tn = 0 . Thus, λ (n ⋅ n) = 0 → λ = 0 .
_________________________________________________________________
2.53 (a) Show that a is an eigenvector for the dyadic product ab of vectors a and b with
eigenvalue a ⋅ b , (b) find the first principal scalar invariant of the dyadic product ab and (c) show
that the second and the third principal scalar invariants of the dyadic product ab vanish, and that
zero is a double eigenvalue of ab .
------------------------------------------------------------------------------
Ans. (a) From the definition of dyadic product, we have, (ab )a = a(b ⋅ a) , thus a is an eigenvector
for the dyadic product ab with eigenvalue a ⋅ b .
(b) Let T ≡ ab , then Tij = ai b j and the first scalar invariant of ab is Tii = ai bi = a ⋅ b .
a1b1 a1b2 a2b2 a2b3 a1b1 a1b3
(c) I 2 = + + =0+0+0=0.
a2b1 a2b2 a3b2 a3b3 a3b1 a3b3
a1b1 a1b2 a1b3 b1 b2 b3
I 3 = a2b1 a2b2 a2b3 = a1a2 a3 b1 b2 b3 = 0 .
a3b1 a3b2 a3b3 b1 b2 b3
Thus, the characteristic equation is
λ 3 − I1λ 2 = 0 → (λ − I1 )λ 2 = 0 → λ1 = I1 , λ2 = λ3 = 0 .
_________________________________________________________________
2.54 For any rotation tensor, a set of basis {e′i } may be chosen with e′3 along the axis of rotation
so that Re1′ = cos θ e1′ + sin θ e′2 , Re′2 = − sin θ e1′ + cosθ e′2 , Re′3 = e′3 , where θ is the angle of right
hand rotation. (a) Find the antisymmetric part of R with respect to the basis {e′i } , i.e., find [R A ]e′i .
(b) Show that the dual vector of R A is given by t A = sin θ e′3 and (c) show that the first scalar
invariant of R is given by 1 + 2cosθ . That is, for any given rotation tensor R , its axis of rotation
and the angle of rotation can be obtained from the dual vector of R A and the first scalar invariant of
R.
------------------------------------------------------------------------------
Ans. (a) From Re1′ = cosθ e1′ + sin θ e′2 , Re′2 = − sin θ e1′ + cosθ e′2 , Re′3 = e′3 , we have,
⎡cosθ − sin θ 0⎤ ⎡ 0 − sin θ 0⎤
[ R ]e'i = ⎢⎢ sin θ cos θ 0 ⎥ → ⎡ R ⎤ = ⎢⎢sin θ
⎥ A
0 0 ⎥⎥
⎣ ⎦ e'i
⎢⎣ 0 0 1 ⎥⎦ e' ⎢⎣ 0 0 0 ⎥⎦ e'
i i
A
(b) the dual vector (or axial vector) of R is given by
Copyright 2010, Elsevier Inc
2-18
Lai et al, Introduction to Continuum Mechanics

t A = −(T23′ e1′ + T31

′ e′2 + T12′ e′3 ) = −(0e1′ + 0e′2 − sin θ e′3 ) = sin θ e′3 .
(c) The first scalar invariant of R is I1 = cosθ + cosθ + 1 = 1 + 2cosθ .
__________________________________________________________________
2.55 The rotation of a rigid body is described by Re1 = e2 , Re2 = e3 , Re3 = e1 . Find the axis
of rotation and the angle of rotation. Use the result of the previous problem.
------------------------------------------------------------------------------
Ans From the result of the previous problem, we have, the dual vector of R A is given by
t A = sin θ e3' , where e3' is in the direction of axis of rotation and θ is the angle of rotation. Thus, we
can obtain the direction of axis of rotation and the angle of rotation θ by obtaining the dual vector
of R A . From Re1 = e2 , Re2 = e3 , Re3 = e1 , we have,
⎡0 0 1 ⎤ ⎡ 0 −1 1 ⎤
A⎤ 1⎢ 1
[ R ] = ⎢1 0 0⎥ → ⎣ R ⎦ = ⎢ 1 0 −1⎥⎥ → t A = ( e1 + e2 + e3 ) . Thus ,
⎢ ⎥ ⎡
2 2
⎣⎢0 1 0 ⎥⎦ ⎣⎢ −1 1 0 ⎦⎥
3 ( e1 + e2 + e3 ) 3 ' 1
tA = = e3 , where e3' = ( e1 + e2 + e3 ) is in the direction of the axis of
2 3 2 3
rotation and the angle of rotation is given by sin θ = 3 / 2 , which gives θ = 60o or 120o . On the
other hand, the first scalar invariant of R is 0. Thus, from the result in (c) of the previous problem,
we have, I1 = 1 + 2cos θ = 0 , so that cosθ = −1 / 2 which gives θ = 120o or 240o . We therefore
conclude that θ = 120o .
_________________________________________________________________
⎡ −1 0 0 ⎤
2.56 Given the tensor [Q ] = ⎢⎢ 0 −1 0 ⎥⎥ . (a) Show that the given tensor is a rotation tensor. (b)
⎢⎣ 0 0 1 ⎥⎦
Verify that the eigenvalues are λ = 1 and − 1 . (c) Find the direction for the axis of rotation (i.e.,
eigenvectors corresponding to λ = 1 ). (d) Find the eigenvectors corresponding λ = −1 and (e) obtain
the angle of rotation using the formula I1 = 1 + 2cos θ (see Prob. 2.54), where I1 is the first scalar
invariant of the rotation tensor.
-------------------------------------------------------------------------------
Ans. (a) det [Q ] = +1 , and [Q ][Q ] = [ I ] therefore it is a rotation tensor.
I

(b) The principal scalar invariants are: I1 = −1, I 2 = −1, I 3 = 1 → characteristic equation is
( )
λ 3 + λ 2 − λ − 1 = ( λ + 1) λ 2 − 1 = 0 → the eigenvalues are: λ = −1,1,1 .
(c) For λ = 1 , clearly, the eigenvector are: n = ±e3 , which gives the axis of rotation.
(d) For λ = −1 , with eigenvector n = α1e1 + α 2e2 + α 3e3 , we have
0α1 = 0, 0α 2 = 0, 2α 3 = 0 . Thus, α1 = arbitrary,α 2 = arbitrary, α 3 = 0 . The eigenvectors are:
n = α1e1 + α 2e2 , α12 + α 22 = 1 . That is, all vectors perpendicular to the axis of rotation are
eigenvectors.
(e) The first scalar invariant of Q is I1 = −1 . Thus, 1 + 2cosθ = −1 → cos θ = −1 → θ = π . ( We
note that for this problem, the antisymmetric part of Q = 0 , so that t A = 0 = sin θ n , of which
θ = π is a solution).
_________________________________________________________________
Copyright 2010, Elsevier Inc
2-19
Lai et al, Introduction to Continuum Mechanics

2.57 Let F be an arbitrary tensor. (a) Show that F T F and FF T are both symmetric tensors. (b) If
F = QU = VQ , where Q is orthogonal and U and V are symmetric, show that U 2 = F T F and
V 2 = FF T (c) If λ and n are eigenvalue and the corresponding eigenvector for U , find the
eigenvalue and eigenvector for V . [note corrections for text]
-------------------------------------------------------------------------------
Ans. (a) (F T F )T = F T (F T )T = F T F , thus F T F is symmetric. Also (FF T )T = (F T )T F T = FF T ,
therefore, FF T is also symmetric.
(b) F = QU → F T = U T QT → F T F = U T QT QU = U T U → F T F = U 2 .
F = VQ → F T = QT V T → FF T = VQQT V T = VV T → FF T = V 2 .
(c) Since F = QU = VQ , and Un = λn , therefore, VQn = QUn = Q(λ n) → V (Qn) = λ (Qn) ,
therefore, Qn is an eigenvector for V with the eigenvalue λ .
_________________________________________________________________
2.58 Verify that the second principal scalar invariant of a tensor T can be written:
( )
I 2 = TiiT jj − TijT ji / 2 .
------------------------------------------------------------------------------
Ans. TiiT jj = (T11 + T22 + T33 )2 = T112 + T222
+ T332 + 2T11T22 + 2T22T33 + 2T33T11 .
TijT ji = T1 j T j1 + T2 jT j 2 + T3 jT j 3 = T112 + T12T21 + T13T31 + T21T12 + T22
2
+ T23T32 + T31T13 + T32T23 + T332 .
Thus, TiiT jj − Tij T ji = (T112 + T22
2
+ T332 + 2T11T22 + 2T22T33 + 2T33T11 )
−(T112 + T222
+ T332 + 2T12T21 + 2T13T31 + 2T23T32 ) = 2(T11T22 − T12T21 + T22T33 − T23T32 + T33T11 − T13T31 ) .
Thus,
( )
TiiT jj − TijT ji / 2 = (T11T22 − T12T21 + T22T33 − T23T32 + T33T11 − T13T31 )
T11 T12 T22 T23 T11 T13
= + + = I2 .
T21 T22 T32 T33 T31 T33
_________________________________________________________________
2.59 A tensor has a matrix [ T] given below. (a) Write the characteristic equation and find the
principal values and their corresponding principal directions. (b) Find the principal scalar
invariants. (c) If n1 , n 2 , n3 are the principal directions, write [ T]n . (d) Could the following matrix
i

[S ] represent the same tensor T with respect to some basis.

⎡5 4 0⎤ ⎡7 2 0 ⎤
[T] = ⎢ 4 −1 0⎥ , [S ] = ⎢⎢ 2 1 0 ⎥⎥ .
⎢ ⎥
⎢⎣ 0 0 3⎥⎦ ⎣⎢ 0 0 −1⎦⎥
------------------------------------------------------------------------------
Ans.
(a) The characteristic equation is:
5−λ 4 0
4 −1 − λ 0 = 0 → (3 − λ ) [ (5 − λ )(−1 − λ ) − 16] = (3 − λ )(λ 2 − 4λ − 21) = (3 − λ )(λ + 3)(λ − 7) = 0
0 0 3−λ
Thus, λ1 = 3, λ2 = −3, λ3 = 7 .
For λ1 = 3, clearly, n1 = ±e3 .
Copyright 2010, Elsevier Inc
2-20