0 ratings0% found this document useful (0 votes) 42 views36 pagesTensor Analysis
Tenssor analysis, relativity, discrete math, numerical method.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content,
claim it here.
Available Formats
Download as PDF or read online on Scribd
BY Contravariant Tensor:[2075]
Arppase N quantities A', A? AN in a co-rdinate system (x', x?, .... x’)
are related to N other quantities A!, A2, A™ in another coordintae
Then A* is called contravariant tensor vectors of the first rank or first order.
Similarly suppose N? quantities A is in a co-ordinate system (x',x, ... x“) are
related to N? other quantities A™ in another co-ordinate system x,
., x") by the transforamtion equations.
pr=1,2..N
Then A® is called contravariant components of a tensor of the second rank or
rank two,50 | Tensor Analysis
2.2 Lovatiant Tensor{2075]
ppese N quantities Aj, Ax - | Ay in a co-ordinate | systen,
__x¥) are related to N other quantities A, Ray vos An i anothe,
') by the transformatin equations.
(x,
co-ordinate system (x"x",
N
4
A,= = APE 1, 2, cesseeseelN
gar
by using summation convention
~ ox?
Ay= we Aq
Then A, is called covariant tensor of the first rank or first order.
Similarly, the N? quantities Agsare called covariant components of tnesor of
the second rant if
ax ext
” onax
_ Be Mines Tensor{2075]
ok: quantities A’ in a co-ordinate system Gl, x, _ x¥) are related to
N? other quantities A’ in another co-ordinate system x, x) by
the transforamtion equations
Ox? @x°
ant ae
ox
Then A‘ is called mixed tensor of second rank.
Note: A superscript is used to indicate contravariant components whereas
a subscript is used to indicate covariant components; an exception
occurs in the notation for co-ordinates.
Q.1 If A” and B’ are components of two contravariant tensors show that
A+B’ is a contravariant tensor of rank two. {T.U 2059]
Solution:
v
Here A” and B° are components of two contravariant tensors in x" and x ©
ordiante system with transformatoin A” and B4 in x? — .
in x 4 co-ordinate
system, and x" co-Refresher Book of Physics BSc. 3™ Year | 51
Then, APO = a Ae
and Bo = ae
Now, A"B? = wees
Let AYBY = Cand APBY=C™
a Ox? Ox4
Then, CM = ax awe”
Which shows that C*’ (i.e. AYB”) is a contravariant tensor of rank two.
Q.2 Show that the velocity of fluid at any point is a contravariant tensor
of rank one.
Solution:
ac
The velocity of a fluid at any point has components of “¢ in the co-ordinate
_ dx)
system x*. In the co-ordinate system x, the velocity is “Gr
But by the chain rule,
de dx dx*
dt dx® dt
Hence if follows that the velocity is a contravariant tensor of rank one or a
contravariant vector.
Q.3 Show that as is not a tensor even though A, is a covariant tensor
of rank one.
Solution:
We know that,52 | Tensor Analysis
x. et
Differentiating equation (1) with respect to *', We B
P
ox OAp, OX 9
awaxt axe
a
_ ax? OA Ox, Oxy
aw axt xo
ax? &x'OAy, Ox
avon ax ow
i «can sides is present hence 22 docs no
Since the second term on the right sides is pre: Ox?
transform as a tensor should. So Bris not a tensor though A, is a tensor of
rank one.
Q.4 If A" and B, are components of contravariant and covariant vectors
respectively. Show that AB, is scalar or invariant. [T.U. 2057]
Solution:
Suppose A" and B, be contravariant and covariant tensor of rank. one in co-
ordinate system (x', x’,
covariant tensor of rank one in another co-ordinate system (x', X°, ...... X*)
“) and A? and B, are the contravariant and
Then the transformation relation is given by
~~
Now, A’B, = 3 ——A"B,
OR
WB, = ANB,
Which shows that A"B,, is scalar or invariantRefresher Book of Physics [Link]. 3“ Year | 53
Kronecker detta , *
Q.5 Define Kronecker delta. Is the Kronecker delta
i_ fl fori=
{ is a tensor?
510 fori]
Determine its rank and type [T.U. 2058 |
OR
Prove that a is a mixed tensor of the second rank.
OR
Find out whether the kronecker delta
5 iiil -{ for i#j
i 2
1 fori=j in all co-ordinate system is a tensor? If so,
determine its rank and type.
OR
Show that 3 -( er i i isatensor
Whereas 5, -{ meres is not a tensor
Solution:
i_ (0 forizj
The Kronecker delta, denoted by 8 is defined as 8'= {1 fori}
In terms of partial differentiation 3 in the co-ordinate system (x', x’, ....
x") can be written as
If, be the transformation of 3, in another co-ordinate system
OO, oocccccsecee KN) then we can write
a.
4 FS
at. Ox" On! Ont
, =
Ox’ x ay) (BY using chain rule)$4 | Tensor Analysis
& Oy!
< Ox ox’ 5
ee = ax at?
Which shows that 8, is a mixed tensor of second rank as follows tenso
transformation.
Again, 5, = 0 ifi+j
= lifisj
If 5, is a tensor it should follows the transformation as
or =
For i=
by =
+ |fork=0
+ 0 fork =
This shows that 5; does not follow the tensor transformation but bj is
not a tensor but 6; is a mixed tensor of second a
Q.6 Write down the fundamental operations with tensors. Suppose a”
and BM are tenosrs. Prove that their sum and difference are
tensors. [T.U. 2071]
Solution:
The following are the fundamental operatoin with tensors.
1, Addition: The sum of two or more tensor of the same rank and type
(ie. same number of contravariant indices and same number o!
covariant indices) number is also a tensor of the same rank and type
Thus, if A‘? and B"? are tensors then CT = A" and BY are tensors
then CP = AN? + B"
and associative.
™ San
4 18 also tensor. Addition of tensor is commutative
2. Subtraction: The difference of two tensors of the same rank and type iS
also a tensor of the same rank and type. Thus if A"? and BY’ are tensors,
then D?~ A"? B'? is also tensor.
3. Outer multiplication: The product of two tensors is a tensor whose
rank is the sum of the ranks of the given tensors. This product, whichRefresher Book of Physics [Link]. 3™ Year | 55
involves ordinary multiplication of the components of the tensor, is
called the outer product. For example AY B? = C'” is the outer product
of AY and B?. However note that not every tensor can be written asa
Product of two tensors of lower rank, For this reason, division of
tensors is not always possible.
Contraction: If one contravariant and one covariant index of a tensor
are set equal, the result indicates that a summation over the equal
indices is to be taken according to the summation convention. This
resulting sum is a tensor of rank two less than that of the original
tensor. The process is called contraction. For example in the tensor of
rank S, A’Z , set r= s to obtain A’ = B™, a tensor of rank 3, Further
by setting p = q, we obtain By = C", a tensor of rank one.
Inner Multiplication: By the process of outer multiplication of two
tensors followed by contraction, we obtain a new tensor called an inner
product of the given tensors. The process is called inner multiplication.
For example, given the tensors Ay and B,, the outer product is A"? Bs.
Letting q = r, and p = s we obtain the inner product A"’B, is obtained.
Inner and outer multiplication of tensors is commutative and
associative.
Quotient law: Suppose it is not known whether a quantity X is a tensor
or not. If an inner product of X with an arbitrary tensor is itself a tensor
then X is also a tensor. This is called the quotient law.
By hypothesisA’? and BY are tensors, so that,
Adding them, we get
ee
Abe iyi POOL Amy ym)
' 1 ox? Ox
Subtracting56 | Tensor Analysis
Aik _ git) 8 OO oa gra
,- BD oe ont age (Ar-Bd)
Then, (A" + B) and (A — BY") are tensors of the same rank and typ,
as A™ and BY.
Q7 Suppose A™ and B; are tensors. Prove that C' = A''B) is also ,
tensor.
Solution:
We must prove that C's’ is a tensor whose components are formed by
taking the products of components of tensors A and Bi.
Since A" and Bi are tensors
and
Multiplying,
alte. Rox om
= Bx? xt axl Ox
Which shows that A’'B{ is a tensor of rank 5, with contravariant indices
P-q, s and covariant indices r, t thus warranting the notation C™
Q8 Prove that the contraction of the tensor AQ is a scalar or invariant.
Solution:
Ox! axt
ax*
We have, Ai = 25% ap
Putting j kRefresher Book of Physics BSc. 3” Year | 57
. . ,
It follows that A; must be an invariant since A is a tensor of rank two
and contraction with respect to a single index lowers the rank by two,
we are led to define an invariant as a tensor of the rank zero.
25-Symmetric tensor and Skew symmetric tensor
Q.9 Define symmetric and skew symmetric tensors. Suppose a tensor
(or skew symmetric) with respect to indices p and
q in any co-ordinate system, Show that it remains symmetric (or
skew symmetric) with respect to p and qin any co-ordinate system.
: OR
Show that a symmetric tensor is a symmetric in all co-ordinate
system. (T.U. 2058 / 2061]
Solution:
Symmetric tensor: A tensor is said to be symmetric with respect to its
covariant or contravariant components if it is unaltered by changing the
Position of the covariant or contravariant indices.
ie. If A® = A® then A?* is said to be symmetric contravariant tensor if
Aga = Ag then Aq is called symmetric covariant tensor.
Skew symmetric tensor
If sign of @® the tensor is changed on altering its covariant or
contravariant indices then it is called antisymmetric or skew symmetric
tensor.
ue. If A’! = — A® then A is called skew symmetric contravariant
tensor.
Ag then Agy is called skew symmetric covariant tensor.
and 18" remains symmetric in x! co-ordinate system,
If BY is skew symmetric Bm” B®
Ox
x yw
@ Ox?
Ox. ont
Be
Ox? Ox
Then, B*‘58 | Tensor Analysis
2 B= _ Be
and B™ remains skew symmetric in the x co-ordinate system. Thus the
above results are, of course, vaild for other symmetric (skey
symmetric) tensors.
Q.10 Show that every tensor can be expressed as the sum of two tensors,
one of which is symmetric and the other skew - symmetric in a pair
of covariant or contravariant indices. {TU 2071)
Solution:
Let us consider a covariant tensor Aw. Then we can write as
Aye = 4 (Age + Am) + (Ape An)
oF, Ay = By + Cu
Where, By =4 (Ay + Ary)
and Cy = (Aw An)
Now, we have to prove that By, is symmetric tensor and C,, is a
skew symmetric tensor.
Le.
1 1
By = 7 (Aw+ An) =7 (Am + Ay)
By = Buy
Hence, By. is symmetric, tensors.
1
Again, C= 3 (Aw— Ay)
1
= = 7 (Am- Aw)
Cw = Cw
Hence c,, is skew symmetric, tensors.
Finally we can say that every tensor can be expressed as the sum of t¥°
tensor, one is symmetric and other is skew symmetric,Refresher Book of Phynlcn Sc. 3” Year | 59
Q.11 Prove that any inner product of Ay and BY is a tensor of rank 3,
Solution:
We know that inner product of two tensor means outer mutliplication
followed by contraction.
go A ~ S
So, AL = apy
OF ot
and By = axl ext ie
Multiplying and letting i =j, we get
ER Ox" EX' AK" Ox!
= FRA wpe
EXP oak OKT OX 55
(@=1ift=p)
This shows that A? B® is tensor of rank three.
Hence any inner product of the tensor A? and By is a tensor of rank three.
Q.12 If ds? = gy dx! dx* is an invariant show that gy is a symmetric
covariant tensor of rank two. [T.U. 2071,2076]
Solution:Here, gy, can be chosen symmetric. Also since ds’ is an invariant.
6 x? Ox
‘
Ox!
© Ox? ox!
Showing that gy is a symmetric covariant tensor of rank two, called the
metric tensor.
~60 | Tensor Analysis
Q.13 Prove that g is a symmetric contravariant tensor of rank two,
Solution:
We have gxG (j,k) = 8
Where Gj, k) is a cofactor for gj. and g is determinant.
GGW(., 4
ee ( oe 2)
Since gq is symmetric, GG, K) is symmetric and therefore g* jg
symmetric. If B? is an aribitrary contravariant tensor, By = 2pqB? is an
arbitrary covariant vector.
Now, Multiplying by g we get,
s°B, = e888?
= BBP (gue= 85)
= B (..8)=1 forj=p)
Since B, is an arbitrary tensor hence g* is a contravariant tensor of rank 2.
Therefore g* is a symmetric contravariant tensor of rank 2.
Q.14 Find (a) g (b) g* corresponding to ds? = 5 (dx')’ + 3 (dx’y +4 (ax
— 6dx'dx’ + 4dx? dx?
Solution:
We have ds’ = gy, dx! dx*
1)
Comparing (1) with given equation, we get
81 = 5, 82 = 3, B= 4
821 = 82> ~ 3, B= 2= By
B= B= 0
Bn Biz Bis 5-30
(a) Now, 8 =| Bi 82 Bs |= | -3 3 2] =4
Bo Bs Bs 0 4
(b) We know,
on GG. KB
= gf (G G, Kis « cofactor of gy)
3Refresher Book of Physics BSc. 3 Year 161
I «|
Here,G (1, 1) = =8
fere,G (1, 1) 4
5.0
G(2,2)= =20
4
=3 2
G(1,2)=- =12
0 4
[oa
G (2,3) =- =-10
(2,3) ce
I? ; |
7 = =-6
G(1, 3) 02
3 |
GG, 1)= =-6
GB,1) 3
-3 :|
GQ, v--|oa
62 | Tensor Analysis APB"
__BegAPBY
a vector. Show that (/(A?A,) (B"B,) =
Q.15 Define length of
invariant.
Solution:
Suppose A’
scalar quantity.
Then length L of a vector Ap OF
L’= ATA, = Ba AvAa
or, LI= APA, = gM APAS
* We can define the angle 0 be
A'By ATR,
con® = JAPA, BB) Lats
Where L, = Length of A”
Lp = Length of BP
ie. APA,and BB, are invariant
‘Also, 85q APB" = A°B, is invarient.
and A, are arbitrary associated vectors. Then AYA, is
P
A, is given by
tween two vectors A and B, as
goA
So, Ta aii invariant
Note: We can define
cos =——Bae?BY _ ;
VGPA) BB) as the cosine of the angle between tvo
vectors A? and BY. If g,, APB? =
ntivsgunad AYB,= 0, the vectors are calle:
a eee
Q.16 Show that the angles 0,
7 ha» O23 and 03, betw a
cars ina dimensional coordinate sateen eh ke
cos O)2= + 608 0); = 8 su
guign Vang,’ °8 Ou =F
Solution: ™
Consider three dimensional co-ordinates curv Ql x? x?
S curve (x!, x?, x’)
Along x' co-ordinate curve x? and x?
ind x
form, are constant. Then from the me!"Refresher Book of Physics BSc. 3™ Year | 63
ds? = gi (dx! + go (dx? + gss (dx)?
or, ds= Ven dx! (dx? = 0 = dx? = 0 along x')
Then, cosine of the angle between Aj and A} is given by
Lp
cos8)2 = Bra AT AT= Bese ae
6080} = 82
Ven B22
Similarly, cos @,5 =—-8—
Ven 833
and = —Su
coma = Tess gn
Q.17 Prove that for an orthogonal co-ordinates system, the off diagonal
components of the metric tensor is zero.
Solution:
If 2 be the angle between two co-ordinate curve then
= — Bi
coe Te en
For orthogonal co-ordinate 0,2 = 90°
: vo
cos 90° Vai B22
> B20
Similalry gy, ~ gas = 0
From the fact that &yq ~ Bap
B21 = B= B= Oer
64 | Tensor Analysis
ol the metric tens
Therefore the off diagonal elements / components oF sor
are zero for the orghogonal co-ordinate system
Q.18 Prove that for the orthogonal co-ordinate system,
1
aie git san and ga = 3
Solution:
We have,
{lifp=4q
eo
Beads loit pea
= Baa
If p=q=! then
e By = Be =!
or, el'gn +a" eats 7
= 0 for the orthogonal co-ordinate system
De =
fal
Since go: = B31
For orthogonal co-ordinate system
|
/ Sy = Oforpeq
i
1D 85a = wlrp=a
2.6 Christoffel's Symbol
Q.19 Define christoffel’s symbol of the first and second kind and find out
(22, 1}, {12, 2] and (21, 2] in the cylindrical co-ordinates system.
Solution:
(a) The christoffel's symbols of first kind is given by
4 (Coe , Cr Baw
(pa. 1) = 3 ax * ax? ~ ax"
If p=q = rthen—e
Refresher Book of Physics BSc. 3" Year | 65
=} (ee Se
-4 -
[pad = (ppp) Ov tax? ax”
Ifp=q#rthen
1
(pa. = (pp.]= 3 laa t aye ax?
1 fo
- $(0+0-)
= —12f00
= Text
Ifp=reqthen
1 (Op, 280
{pa.r]= [pap] = 1 (2p Be Se)
= 1 (9+ 2ee
= $(0+ 8-0)
= 12%.
= 2 oxi
(b) The christoffel’s symbols of 2nd kind are given by
(ee
Pq =e" [pat]
1
We know g!= 7
7
Ifr 4s then
/ | - (pq, 1] = 0
, - y* [pq 1] =
| pal e
Ifr=s
{s| 4.8
ie g° [pas] fous
Now,if p= q=s,
fs | [Pp | fpr] 1 1 een
| pp
| pa Yon Een OX"66 | Tensor Analysis
If p=qes
{*}.. {pp.s] = 80
tal “{ypl” eee Oe
If p=seq
{ \_{P\_ fase)
(pq! pq)
| Ope _1 0
= Zaps Ox! 2 dx (IM Boo)
Note: () Ifp. q, rare distinct then (pq, r)=0
(ii) If p, q, s are distinct then
k jai’
2nd part
In cylindrical co-ordinates system
x! = p,x=$,8=2
Wehave gi= 1,gn=p*andgy=1
Now, to evaluate [22, 1]:
Here, (pq, r] = [22, 1] where, p=q=241=1
Then, [pq,r] = [pp, 1]
= 1280
= Dag
1(g 1 dp?
(22, 1]= -1(B)- 42 --p
To evaluate [12, 2}
Here [pq,r] = [12, 2] where p= 14
Then, [par] = [pq,q]
= 5 (i a) 12
2\.0x" © Ox? Ax
=3
(12,2) = [pa,q] = 3(23)Refresher Book of Physics [Link]. 3* Year | 67
To evaluate (21, 2
Here, p=1#q where p=r=2%q=1
1 eg,
Then. [pat] = [papl= 75%
Légm_ 1égn_1ep°
* (20, 21= tna.) = Se = 5 SB a9 =P
Hence in cylindrical co-ordinates
(2,1)=-p
U2,2}=p
(21, 2]=p
ox) rive transformation law for the christoffel symbols of (a) the first
kind (b) the second kind.
Solution:
eee atiors :
(a) Since g et
Cy ex? ox! ae at
or, x WA OH He PW TY Oy)
ae aR at OX ax” a OOK AX OK OR
By cyclic permutation of indices j,k, mand p, q, r
Baw _ GS Ort Soy? , Oxt _Fet_ SH)
Rat ax a ax aXE™” aK OKO
ax? ax! ax?
OR oe
in + g o
ae at ax" ax OR” OR OR” OW
Substracting (1) from the sum of (2) and (3) and multiplying by } we obtain
‘on using the defination of the christofel symbols of the first kind,
Ox? _ xt
{pa} + Bogeesseen4)
aD atom
Ox! Ox* Ox
(b) Multiplying (4) by "= S's Org to obtainSince 8g" (pq. 1] = 8" [Pa a-{? janet et f= 5p
ion (5) are the required transformation
Hence, equation (4) and equal
low of christoffels first and second kind.
s s s
Q.21. Prove (a) [pq. Fl = (ap, FE wf } -| } (©) [pa Fl = ef \
pq) lap Pq.
Solution:
(a) [pq.1] =
L$ Bp) 1p eB
ox" Ox? Ox’
(b) (| “ete tw.a-{"}
(©) ou ‘|. aug" (pq, 1] = 8 (Pa, #] = [pa, kK]
or,
(pq, k] = wl | that is, (pq, r] = w{) ‘|
Note that multiplying {pq, r] by g™ as the effect of replacing r by s.
raising this index and replacing square brackets by braces to yield | .
pa
Similarly, muluplying i by ge, OF By has the effect of replacing s PY
r, lowering this index the replacing braces by squre brackets to yield
[par]Refresher Book of Physics BSc. 3" Year | 69
Q.22 Prove that
a
() Sea = pm, al + lam, Pl, (oy Ses =-" \} - | P }
mn.
P a
«) ("| =3aln ve
Solution:
eg,
imamate) ee
(a) [pm q] + fam. p} = +E =) ant Bae Shan “a
(by 55 (gs,) = at) = 0. Then
a Oa ag"
ot St Eo 00rg, Bem ot
Multiplying by e”
ég* ie
8's a= -e
ég*
thatis 625 =—g"e*Cim, j]+ Lim, i)
(theft
+ and the result follows on replacing r, k, i, by p. 4, n respectively
2 = 2G (j, k) (sum over k only).
(©) We have, since G(j, 1 does not contain gy expiiy, B_ GG, 9. Then,
summing over j andr
tee fe Ge ote
me * apy GG.) 50
= pe ZBt = ge" (ime) + fm.)
j r ‘}
= =2
e((ed (il) tm
1 j
Thus, 2g ax" (*t
j a
or (3 ~ ome
The result follows on replacing j by pand m by 4‘70 | Tensor Anatysis
Q.23 Determine the Christoffel symbols of the second kind in
(2) rectangular, (b) cylindrical (c) spherical coordinates,
Solution:
Since for orthogonal coordinates pq = 0 if p #q.
ml
(a) In rectangular coordiantes, g,, = 1 so that { 0
(b) In cylindrical coordinates, x! = p, x’ = , x’ = z, we have, gy, = |
2 =p’. Bis = 1. The only non-zero Christoffel symbols of the
second kind can occur where p = 2.
These are
{nb = ~ag BFE 0) =-0
tat = {a}-abSP-b2e
(©) In spherical coordiantes, x! =r, x*= @, x° = 6, we have g,, = 1, Ba
=P, gss = rsin’®. The only non-zero christofel symbols of the
second kind can occur where p = 2 or 3.
1) _ 1 ap -16
a} = “et TRO
tal > {inl -ag, BPS et
fi) _ 14 la...
33
{2\ _ 1 Oy 1 3
\33) > 2p Ox? ~~ Ie |p LF sin’®) = ~ sind coso
13) . (3). dm 1 a, 1
{suf 113) ~ 2gay Ox! “22 sint ap (P sin’®) =>
{3\ 3) 1 ogy a
\32) tal Bans Ox? ~ 2 sink a st") ~ cot 0Refresher Book of Physics [Link]. 3" Year | 71
satrove 2A [0] 28% at {ml}
Prove Ron LI ow aw Pa
\ Solution:
[m| _ ax? Ox" a fs
From Q. 21 (b) {i ode lon
ax”
Multiplying by » We get
ox
[| &", a Ot on fs} det m
= " 8
Wom amage* lea) aig
- 28 BA fm} , Fae
OR ARK 'PH Bix
-— {p] 22 _ ax? ax fm}
oa WW oe aw axt Pa
Q.25 Show that the goedesics in a Riemannian space are given by
ax [| dx? dst
ae{ ds ds 7° (TU 2074]
Solution:
&
We must determine the extremum of i Vega?! at using Euler's
equations with F ="\ gpa
2 Unit) 5
we have,
oF
er
= (Bai) 2g a?T2| Tensor Analysis
pax’
Ban 950 5 oie wi
on i
kon , Bak
Soa 505 ik" this equation becomes
Writing Batts" 1 ax? ns
°
or, Bad! Hay
0X” + [pa k] 5°"
If we use arc leat as parameter, § = 1, § = 0 and the equation
7 dx
oF tpg)
. ee dx? dx?
Multiplying by g*, we obtain “G+ re
fs
Q.26 Suppose A, and A? are tensors. Show that (a) Ava Ay Is
oe ie
and (b) ard at ef? A’ are tensors.
Solution:
(a) Since A; = a
ox
So, 2Ai A OA, at
Ray Om ope
Fromg.25 2x.
anon
Substistuting in (1),
DA, _ Ox! an,
oF” aap ot Ale A
Ox! Ox* pity
— OR Ox" A, {a} Bea le
Ox! ax* (pq
ait ox ntRefresher Book of Physics B.8e. 3” Year | 73
aA, [mlz _ OxP dx" (OA, [8
oh, a= ~~ ( { \a,)
en ox! ax \« Pq
3 [s
ind oAs 5 le is a covariant tensor of second rank, called the
and ex ipq)
From question 25 interchanging x and X coordinates,
Fe (2 ma fi)
oxox!
Substituting in (2),
oR _ Geax OAT, [Maw ax! ER oxi) ‘lat
ae Op OF Lalo gga “Oe oo
ax! dA, [M|ox Ox! ox ‘(i ©
Ox! rf ax® 958 ax! Ok Wt
R ox9 OAP {P| ae ax 4 {3}
I
@
'
ax? y+ Ox* * [sql Ox” gx
an fila aa om (dae p ‘)
gh kil ax? aes LOx® "Lash 4
ant {>|
and axe lel A’ is a mixed tsnsor of second rank, called the
mr qs)
p
covariant derivative of A” with respect to x‘ and written Ay.74 | Tensor Analysis:
Q.271f A! and A? represent the same contravariant vector in unprimeg
and primed frame of refrence respectively, show that
= ox
ra
Al ox A
Solution:
‘Suppose the transformation equation from a rectangular (x, y, 2) system
to the (x), Xz, Xs) and (X;, Xz, X3) system are given by
X= Np (Ny. Nay Nas Y= Ya(Rty Noy Nas Z= Zi (Ky Kry h
X= X2(K1, Kas ads ¥ = Ya (1, Xa, Xa), Z = Zz M1, Xa, %G)
Then there exists a transformation directly from the (x,, x2, x3) system
to the (%, X2, X3) system defined by
Xr =i Rt, ay Hs)s Xp = Xo (Ray Ray X3)y X3 = Ho (Ray Ha, H) oeeee.(2)
and conversely from (1)
= He or
dr = ox, & * xy Ot Gy, b= ay dx, tarde: + a5 des
or
and dr = Fag, +t 4
oy 2 OR
= a; dk, +a dk, +2; dx
Then, aydx; + adx, + aydx; = a, dk, + a, +3 BH 0.)
From (2)
dx, = % Mag, d%
1 dx, dk,
ty = 8 6 a,
dx, a dx,
ay Oey
dx, = 8 dns Boy ™ a,
dx, dx, dx,Refresher Book of Physica BSc. 3™ Year {75
Substituting into (3) and equating coefficients of dx,, dx;, dx; on both
sides, we get,
a=, Ss, OH
a aR,
a By 9, OH, OH
mR,
yaa Bs 9 MH, OH
OA.
‘Now, A can be expressed in the two co-ordinate system as
A= Ati + Aga) + Ayay & A= Aja; + Azay + Asay. (5)
where A,A and Aj are the contravariant components of A in the two
system. Substituting (4) into (5)
Aya) + Azar + Asay = Ajay + Apa) + Asay
~ x x 7 = om
= (™ 4, 2a (5B Ay + A
ox, Ox Ox: ox, OR, OX:
+ (AeA
RH OK
On comparing the coefficients, we get
. = x ox.
AL = A, Ss Ay S4 Ay
~ ox = omy, ~ ox,
Ay = A, 2+ A, 24K,
&
Ay = Ay 22+ A,B A,
Ox
moh A284 Ky 284A, 22; p= 1,2,3
1 2 3
3
And hence, Ay = & A, 22 where, p=1,2,3
jel ax,
Similarly, by interchanging the co-ordinate we see that16 | Tensor Analysis
_ x
Ay Sapa l23
In the form of contravariant
Ara
Q.28 Show that total derivative
Syt _ adxt a’xt | { k \dx? dxt
=e 7 Ne at equal to Ge * | pgs at at
What does a* = 0 mean physically?
Solution:
a - Sided
at dt
byt a |e dvtdxt [k] dx? dv< fk
= [Raf h(a dxt_ dv" s dx?
FS gsi” | at ~ dxt {le sf {i wae
.
i v= gy
vis Grand vi =F then
at = dx! dx®
Changing index 's' by 'p', we get
te okt {* bax!
at = dx? dxt
at * pq) dt dt
. :
If at = 0, physically the equation is
dx fk | ax? ax
a * pal dt dt =
Which represents the equation of geodesics
P
Also, a° = 0 means that the intrinsic derivative of
along the curve x* — x(t) is zero then the en eas
parallely along the curve. It also implies that notice ee
acceleration. Christoffel symbols have ppatered| ke eee
oth in covarian
differentiation and in the differentital equation for od hi
a geodesic. This
i
ce
ls,Refresher Book of Physics BSc. 3” year | 77
sts that there may be connection between the two concepts,
‘ant different and geodesics. Since
. Bet at
a 7 Yad
is vero, the covariant derivative along the geodesic
is zero. Thus, a geodesic is tangent vector with zero covariant derivative
along the curve. Geometrically, geodesics are straight line in space.
Q.29 Let A, be a covariant tensor of rank 1. Show that
Agar ~ [Link] = RygrAy using the property of covariant derivative.
State why Ry, is a tensor. [T.U. 2062]
tel
‘C1 Bh) QI
oil is
By interch: ‘q and 'r' and subtractii find “als
sof Be bed
ltl faa
Mow
7
EL nh
{ilfky fi \ | Wil,
\pr: tka) &'Ipal ~ Ipql lke}
cj by nand we get
Aya ~ A, Ri Ag
Pa
Ap nq is a tensor, Rey Ag is a tensor; and since Ay is an
arbitrary tensor, Ry, 18 a tensor by the quotient law. This tensor is called
Riemann-Christoffel tensor,7B | Tensor Analysis
' 7 * js an invariant.
Qw Prove that dv = dx'dx? en dx" Is a
[T.U. 2062, 2063,20¢4)
_..(1) [by definition]
un ON
a
f
a
&
a
_. dR" [using (1)]
&
Se
ay [ mI
= Ve] a
=
x
& ag H gy
x Oe
Sage a
- Vel xr a
& gn 5 ix
aN
& = Vedx'ax. dx"
@ = dv
This shows that dv is an invariant
: 1a
Q.31 Show that diva? = Year Vea (T.U. 2062
Solution:
The divergence of A’ is the contraction of the covariant derivatio”
AP, i.e. the contraction of A’, q or AP, p
7 oa {Ply
Then div A"= AP, p = ee [tha svvenenned(I)Refresher Book of Physics BSc. 3“ Year | 79
Now we have to find { , \
pk
We know g = ga G(j, k) (sum over k only).
ince G{j, k) doesn't contain gy explicitly,
am .
a, 7 6G)
Then summing over j andr
Se + BS-ai.n8 ee" (Lim, ¥] + [rm j))
By CX!
(Ca ta)
ton 3ee~ {)]
7
(ml 7 ae
Hence, by replacing j by p & m by k, we get
Now, equatioin (1) becomes
ea’ (a eat (1a
diva, p= B+ (Znve) a= oo
ndivAP = ae)
Q32. Show that vo- Jeo [vee 4]
Solution:
The gradient of @ is grad 6 = Vo = &; } 4 covariant tensor of rank one
defined as the covariant derivative of 4, written ,. The contravariant
tensor of rank one associated with @ , is
. wn OY
* bx
Then, V@= [Link]=div a)
As we know that
div AP ya [Ve']RO | Tensor Analysis
[Derive this relation too, for this see Q. 32]
Then equation (1) becomes
of ade ee Od
v4 Jez [vee a
This is the required soulution
Q.33 Show that g=J?
Solution:
Wehave. P= ixtjytkz
=>
Br 83
Now, g = | 81 82 83 |=
8 Bs Bu
ox by
Gu, Ou
- x ey
Cu, Cu,
xy
cus Guy
JJ [. =
7 (uy, 3]
8 £
Q.34 Show that g = J
Solution:
BaRefresher Rook of Physics [Link]. 3™ Year | 81
, Ox! ox! ,
Ax! ax* axi
Ox! Ox ax" ox’
But gn
avax axvax
Sn Bi Bis
Now.g =| 3
ow. & B1 B22 Bs
81 82 Bu
8u 8 Bs
= 821 B22 B23
831 B32 B33
ox! ox Ox
ax ax ax
= JgJ
e- eg
Q.35 Express the divergence of a vector A? in terms
components for (a) cylindrical coordinates, (b) spherical
coordinates.
Solution:
(a) For cylindrical coordinates x
X=, =Z,
zg 0 p? 0 | =p'and Vg p
oo 4
The physical components, denoted by A,, Ay. A, are given By
Ap Ven AD AL Ay Ven Ao pA A, Ven AA
a Q
(Ag) 15, PAD
1 6
Then, div A’ gone)82 | Tensor Analysis
(b) For spherical coordinates x! =r, x” = 8, =>
10 0
eg - | 0 PO | =F sin’@ and Ve= sino
0 0 Fsin’®
The physical components, denoted by A,, Ao, Ay are given by
A, = gn AT= Al Ag = Ven A? = 17, Ay = Ves: A® = sind A?
Then, div AP= “pares
[é (sin A) + (ein OAs) + a ag]
a
- 5 a 5 CAD ap =r 2 Hsin oA) 5 sind a
Q.36 Express the Laplacian of 4, V%, in (a) cylindrical coordinates, (b)
spherical coordinates.
Solution:
~ ro
(a) _ Incylindrical coordinates gy = 1, g22 “3 8333 = 11
Then, we have
vo = habed)
[alos) a5)
2 1eo eo
(9 Ot aR se
. 1
(b) In spherical coordiantes gy, = 1, 822 = 7, 833 =F sin’o. Then
1a br O®
vo = cs (Vee om)
1 f[af>
- manal(" ala eS
10/0) 1 2/(. am) 1 oo
52 (2): ain ve 2 + Fein’ 3b?Refresher Book of Physics fe. 3” Year | 83
Q.37 Calculate the intrinsic derivatives of each of the following tensor,
assumed to be differentiable functions of t (a) an invariant ©, (b)
AN (6) Abe (d) Abe
Solution:
ey a = a = see -2, the oridinary derivatives.
=
Se ert rr
© = ae al ** lash 8) at
S| jdxt [i] dx?
wa
ted dt * Igs{ Sat
dA n dx’ (ae {f 1 {| X
gy PA gh gf (PAim 15) yw 1S | ae
@ meat” Vat Ll ma}
5 i) ow {E) y \aat
{hata (3p ata {Ef ata)
nd aa ash MJ at
Atm {8} x det [8] dx? [8] dxt
dt “ial A dt mal A! dt ~ Ingl Ai ae
J] a det fk]. at
left
Q.38 Derive tensor form of gradient, divergence and laplacian
IT.U. 2073}
Solution:
a
The gradient of @ is grad 4 ~ Vo = 5%; a covariant tensor of rank one
defined as the covariant derivative of , written ,. The contravariant
tensor of rank one associated with @, is
ta OH
An gh84 | Tensor Analysis
of AP is the contraction of the covariant derivation of
The divergence
qorA'p
AY. ie. the contraction of AY.
ve anp= oe PPL Ak nll
Then div AP= APP Gx* * [pk )
Pp
Now we have to find {
ipk
We know g = 2a GQ. K) (sum over k only).
Since G(, k) doesn't contain Bj explicitly.
a.
a Gun
Then summing over j andr
6g OBy _ ge" ao
a ay, ox" GG, ee ox" Se — oe (Gm, 1] + (rm, j))
~ (al aD bal
1 é_fJ
Thus. 5550 =
MS: 2p ax i
{3}
or, =
7 lim! ae inva
Hence, by replacing j by p & m by k, we get
Now, equatioin (1) becomes
®
ro. oA @
diva’, P= “GyFT Zr inafa) a* 2 aM
La
div A” = eae eA) .
Now. 7o= V.V=div (e" 2)
‘As we know that
pool.
div A ae [Ve.a*]
Then equation 2 ) becomes
vo Fae “| Vou" |
This is the required soulution
mw