Physics 131 Week 3 Lecture Notes

2017
Spring
 @

Four -

Vectors
-

You need that vectors

should no
convincing are an
important
and useful mathematical concept in Newtonian
physics .
The laws of

Newtonian mechanics best written in terms of vectors F- =

he
are
, eg , .

In fat when written in this the laws take the same form in #
, way ,

frames to of
inertial reference ,
a
good way guarantee the
principle
relativity ,

Well now that we have unified space and time into a

single
,

four - dimensional entity ( spacetime ! ) you might guess ,

that Four -
dimensional
vectors in
spacetime would also be useful ,

vectors
In flat spacetime it is easy to define such
,
4- in the same

way
that you
learned in
Kindergarten objects with -

as
a
magnitude and
direction ,
which be
represented geometrically
can as arrows . 4- vectors have

the the 3 vectors that to

algebra as
you
are used
-

same e.
g
,
,

÷
y
- -
-
-

g-
g +
I =
I +
an

gµ⇒se*µ
L
( We follow Hartle in a four - vector as
unity
A- ,
and a 3 -
vector as £ .
)
What's a little weird
, though ,
is that their
squared magnitude can be

direction
positive , negative ,
an zero
depending on whether the of the
vector is like time like or null
respectively ;
space , , ,

tied

|←*##
+ ^ null
r r
a null
g

" k
"
"

y y

6 ~
I
null time like null

- ×
 @

If we define Cartesian basis vectors { It ,I× ,

Ey , G } ,
we can

write vector terms of its

on in
components :

at a×€ Nea
g
=
g
+ +
ate , +

NOTE that we are

denoting the components with superfast indices
,

at but
e-
g ,
,
the basis vectors with
subscript indices . This notation
is
critically important ) especially as we more on to general relativity .

It is useful to denote the

subscript and superscript indices with

t 2 3 Then the of 4- vector

numbers 0
components
→ I →
; x→ z →
> a
.

, , ,
written form
can be in more compact as

E a.
'

g
=
e- a

x=o

or
in even more
compact form
' ,

g=a×e@
the Einstein summation convention repeated
where we are
employing -
i

indices
upstairs and downstairs are to be summed over the values 0,1 2,3
, ,

This avoids us having to write summation symbols ( £ ) all over The place ,

The components of a 4 -
vector in different inertial reference frames
are related to each other through Lorentz transformations i

'

"
)

III.
Y
O (
'
a = a -

va

signed
q3
'

of =

vector
"

The length of
"
4 defined the
squared
is in same as
a
way
-

element
"

for the line which the infinitesimal

"

,
gives squared length of an

vector
displacement :
 @
like vector

a3a3|=O
20 ⇒ time

a°a° tacit
'
rector
'
⇒ null
g. a- = t a a
-

> 0 ⇒ like vector

space

Examples
events A ( 0,1 1,0 ) B 10.5 2,0
Are the and 0 ) sqaalike
= =

,
, , ,

time like
,
or null
separated ?
Just calculate The components of the displacement vector from
A to Bt

0.5
I One
+
Is a e-
= t
go
-
-

. ,

+105
'

15 th
'

5 ( (
Is ) )
( = -
0 's ) + (
Is
.
+
z na
-
-

I > 0

Hence A and B are

space
like separated ⇒ B is outside A 's future

p 430,
light cone .
( This is the same
example that we did on
,

vectors the
now done
using 4-
,
but
physical content of the calculation

The B far from A

is same .
is
spatially too A for
light from to reach

Binthe time to -

to .
)
Note that such Finite displacement 4- vectors can
only be defined
in
Flat spacetime ,
As we will see later , 4. vectors in curved spacetime
actually have to be defined in a local
tangents to the curved

spacetime .
-

The for define

expression g. a-
can be
generalized to a

Scalar between two

product any
four -
vectors ?

a3b3
'

G. I = -

a°b° + a 'b + oib 't =

§ .

an

or

ftp.sypa# ← this

independent
is an

of
invariant
inertial
quantity
reference
,

frame .
 620

Here
"

n.mn/ojIg.giIy.IoasIII
The
ya
's are in fact the components of the
meter in the H ,× , > a)
coordinates of an inertial rest frame in
special relativity ,
We often
call
yg ,
the Mankowski .

Note that we can also write

( ) Ep )
axes
'
as ( b
gig
=

E
=

gig
.

Hence the
components
of the Mikowski metric tensor are
simply the

scalar products of the basis vectors '

Typified
Note that in Cartesian coordinates in flat
spacetime, we therefore have

0
XFP the
orthogonal
if ⇒ Cartesian vectors
Ee Ep basis
=
.
are

:|
Also I
-

l ( q time like )
yoo I is
= .
=

, ,

⇒ The Cartesian basis vectors

G I = I (
g is like ) ortho normal
2
=
space are
.

ftp.T#IsYIeYtit
, ,

we
⇐
1
( *

733=1
sea
.
.

Is s I ) Is ) eg .

1
(e→ like )
-

G = is
space
 630

Note That the line element can be written in terms of the components
of the metric tensor as

dshtyapdi.at#=(dx5t(dx'5t(dx5t(dx35=-dt2tdx2tdy2tdzh

This the metric tensor it

is
why we call
zap ,
as enables
you
to

( ) from infinitesimal
"

compute interval distance

"

the time ds
space
in
changes coordinates .

Four Velocity
-

Newtonian mechanics instantaneous

Recall That in
,
the
velocity
vector T A) of a
particle is
everywhere tangent to the
particle 's trajectory
I A)
through three dimensional Euclidean

sPafµj
.

gay
VH DI -

%#¥*s
, =

at JGD

The IH
trajectory ) in Euclidean
space
is a curve which is

parameterized by time t :
given a time
,
I have a
particular point on

the trajectory ,

For the timdike world lines of material

particles ,
i.e. for the
trajectories in 4 -

dimensional
spacetime , it is
perhaps natural to
the ( !) he
parameterise World line
by the invariant
proper
time T
,
,

" "

teeny
T.IT#IEteYIdngkmttyttitLarrigdge7theeartiae
¥1 - x
 @

four
so the -

velocity is a 4- rector that is

everywhere tangent to

the particle 's worldlier The different

components are

he
' .

i=N =ov
uos =t n'
=dEn=d←÷dta=N× , ,
n'

or

y=fr,tI@
/ ( spatial 3 -
vector part of 4-
velocity
time component

84
¥1
'
Y TV HVY -1
-

Note that
is
+ = =
-

.z=
=
-

i. e. the four -

velocity is a
unit timelike vector .

King
[ A more fundamental derivation ?

u± II uia=nadFI¥

=aqgegx÷
stand -1
]
=

The fact unit vector suggests

that if is a that it
might be useful as

a basis vector .
This is in Fact the case
,
as we shall see later .

Four -

Momentum
-

Just as the linear momentum 3 -

vector is
mV→ in non
.
relativistic
Newtonian mechanics ,
we define the four .
momentum as

10€ or

p
'= mci
 @

Hence ( md m8V→ )
p
=
,

5
p relativistic 3- momentum
?
energy

( E=m0i ) ( F=mN )

i.
fact
Minkowski translational
As we shall see
,
the fact that
space
is
}
invariant in time and leads to energy
and momentum conservation
space .

But because and time are unified so is this conservation law :

space ,

the conservation of 4- momentum ,

Note that

ftp. m2= me
f.
-

of
= =
z =
. - -

rest
→
of
is a time like 4- vector with
magnitude equal to the mass

back
energy
( me ) of the particle .
In Fact
, putting the c 's in
,

-
Eh + P2c2 = -
med

or

Wftu
,
a famous Formula
,Y⇒y
innefeati .

Four -

Force
-

to
If seems natural
generalise Newton 's second law ,

E =
month = date

in novel atheistic mechanics to

relativistic mechanics
€=mda÷=daPg@
in ,
 660

This fact connect £ another 4 vector called the

4=4
is in and is -

,
,

Also I a-
is the tenet
ad÷
.

If we define the 3- force F by

E=d←P@
with §=m0T being the relativistic momentum
,
then

=tdaP÷
' '

f =
dad -

. date =tF

f3=TF3
'
's TF
Similarly F and .
Also ,
,

f°=ddP÷= DE =
0¥
Thate of
change
of
energy
of
particle
But 0¥ should then
equal F. V→
,
the rate at which work is
being done

on
the particle . Here
fo = y F. T

(
F¥=mdqY=
Check :

'

math )P
'
-

VY =
mtzlhv .
dad =P me .

dad

F. V =

madgctv
) . J = me .

( PV date
. + rdatg ]

=
m[ PVT .
data + ovodaif ] =
moiiodatf XT
#
+ i.

✓
⇒
aa÷=Ev ]

filtering
•

. .
 670

Note that E. a
= HE .V , RE ) .

( a hi )

=
-

O
'
F. V + TEN = 0

This makes sense

!
£ cannot
change the magnitude of y ,
because
unit nectar direction
is
is a .
It can
only change its .

Note that E. £ =
82ft Fort + FY 7 0

so
€ is a
space -
like four - vector

example of the of the 4 force the transformation

An
deriving
use
-

is in

laws for E and Be in

electromagnetism ,
You will do this in the
homework Lorentz transformations of the 4 force and
by using
-

remembering that for a

charged particle we have

F=q C E tvx B)
other useful of 4- vectors
There are
many examples !

-
scalar and vector
potential in
electromagnetism combine to make

a
4- vector

charge density and electric current in electromagnetism combine to

make a
4- vector .
 680

Variation -1
Principle for Free Particle Motion
i

In of the twin that the

our discussion
paradox ,
we learned
free
World line of a
particle is a
path of maximal
proper
time .
A
path
extreme distance
"
of time spacetime
' '

proper
or is a
geodesic
in
spacetime -

Hence we
have discovered the means of formulating the relativistic
of Newton 's first
generalization law :

s.ee?tEEEt:fIEeewmatT:lxteItYstetTnEontinetIafFnnn

set of like
Consider a time
paths fn
between events A and B in an inertial
B

#
•

Eying
:tI
I.
:
*

|
=§-napdxadT5 A•
-
×

let 's these X× A) t

Now
, parameterise paths as
,
where is some

which
smoothly and
monotonically varying parameter takes values from
( Just parameterise
)
.
tots at A and B
, respectively . as we an
ordinary
fourth time
trajectory in three dimensional
space by variable are
-

a we
,

parameter izing a
World line in four dimensional spacetime by a fifth
variable : ) . We could use E itself as such a
parameter but it is more
,

instructive to Keep things general , )

Then
T.rs#tsdbIIfdaxf ,
i.e.
I§ .nu#axTf =

sets will extremized for

This now
up
our action
principle .
Taos be

that satisfy the

world lines Euler Lagrange equation
-
 690

aFh¥a ,
)
.

I. =o

with 4×5 and )

FEE
=

's cyclic that

tlqa the
= 0 -

all the x× are

which means
,

momentum 4 vector is conserved Wewill

energy
-

see how to formulate the

general relationship
between spacetime symmetries and conserved momenta

later

g-
¥tvsHI¥
about
the 2 comes because
P is a

¥ :#
:*::
Iranian
:&
"

,⇒= -

⇐
mdE FT

=
-

up d¥ = -

zap day of
-
at

d.
T

ya dn⇐
-

i.
Itu .da¥s=o
# 7)
' '

coordinates diagonal
'

spacetime
'

In Cartesian xsy is
yes
our , ,

and constant ,
so I can also write this as

II
da ( )=o

_-da¥I=OadadE=O
multiplying by ¥ ,
we get

x=0
, 1,33
 700

i. e .
the world line of a free
particle straight line in inertial
is a an

reference frame in the Minkowski spacetime of special relativity ,

f
^
• B

A
•
µ = constant

- x

variation word
In other words
,
This al
principle ,
and will be the

approach free particle motion in curved spacetime

way
we .

All of this must be related to the principle of external action in

Newtonian mechanics ,
Let's nederine
geodesic equation
our in

Minkowsk spacetime in a
way
that
destroys the elegance of
special relativity
separating time and

t.fr
t

#•
by space :
•
to

In
- -
.

5.
f. dtg
=

+A
...
.

A
External time now means
proper - ×

a¥( kaka )
-

¥ ,
=0
5.1.33 ;L=
f-
But this is
really the same
thing
-

I've
just replaced ) by t .

So I end with
up

d. dui
0
j= 1,33
=
=

at
s
de

lost fourth dad 0 from

I
my equation =
,
but this is
implied the

above three
equations and the Fixed normalization of a- ! ddg ( iron )=O ,

But while this all be The same it looks lot more like
may
a
,

relativistic mechanics fact

non
Lagrangian .
In
,
 7

#
÷
L= =

For V<< 1 ( i.e. V<< c)

[ a 1 -

£V2 ⇒ mL = m
-

's mV

lie L is related to the Newtonian kinetic of the

energy
, minus

from which makes difference

particle .
Apart the minus
sign
-
no to
final of motion explains why the relativistic
the equations ,
this non .

Lagrangian for a free particle is

just£mv2 .
 720

Light -
Rays
Hour do we handle
photons
in
special relativity ? Their world lines

in an inertial reference fm me are

always straight lines but which

spacetime
" "

have Zero
length :

ds2= 0 = d 5

looks terms of but realize that

This very
weird in
geometry ,
a

photon world line still passes through distinct events ,

even if they have

spacetime separation
zero ,

to photon
Clearly we can't use E
parameterise a World line ,
as

there Well let's invent parameter 4 labels

is no E .

, just a new

the events on the photon world line in a smooth and continuous fashion ,
Jr coordinate vector to
i. e ,
is a
along the world line .
A tangent
the photon World line could than be written as
having components

↳
t
DI
^

na

.
=

db
Ty
This is still a null vectors
°

a. a =

yes # If = 0
.

-
×

There infinite of photon

are an number
ways of
parameter ing it a

world line in This

way ,
but a
particularly useful one
is to choose )
that hi constant the World line
such is
along ,
i.e.

datedness
which looks like the
geodesic equation for a
free
particle (
p , 690 )
,

but with replaced

2 by 1 ,

A parameter 1 which satisfies this is called an affine

parys .
There are
many Ft
possible choices of affine er .

time t inertial
For example ,
itself in a
particular reference Fame
 730

would be an affine parameter in

special relativity ;

×=t
e.
g. consider the photon world line , 7=0 ,
7=0

date type
_u= date
twa
+
. ,

=
Eo tea

1+7=0 ✓
a =
-

Li

and also dd,÷ = 0 ✓

( general
In
,
an affine parameter is one for which the
geodesic
satisfied For of who Know about covariant
equation is , those you
derivatives and parallel transport ,
a
geodesic is a curve that
vector
parallel transports
its own
tangent ,

Lions =O

where n' =
dd¥ .
In flat
spacetime in Cartesian coordinates ,

✓
agenda ⇒
dado
Straightness ,
iie
transporting
,

parallel The tangent recta

,
is now

the more fundamental concept rather ,

than external
proper
time
,
which isn't defined
along photon world lines ,

Note that if ) t two affine parameters for the

]
,
and ,
are same

must related transform

"

then
"

they be affine
geodesic
curve
, by an
,
i. e. there exist constants a and b such that 6
,
=
alot b ,

will often affine

We use
parameters in
parameter izing photon
world lines without
worrying
about what they actually are , All we need

to know in order to do calculations is That ore exists ,

 740

Note that photon 4- momentum

the is
defined as

of
=
( Ep ,

'

ftp.O E2 =P ( '=P3c2 with

f=O
and ⇒ E
f.
-

so or

dimensions restored ) .

Note that ¥ ( how could it be when m=O ? )

if my ,

[ However could view

Ling ( date ) ]
=

f
m
you
.

all makes
Quantum mechanically ,
this sense :

E- tnw P=k I
,

7 7
wave vector
( angular frequency
)

'

k ( k) =O w2=K2
if
⇒
of of
= w and
,
,
.

hi ikh relation
i. e .
=

,
the usual
dispersion for light ,

Loreto transformations on the 4- momentum of the

photon
to Dopplershifts relativistic aberration

:-#
beam
give
rise and or
3- , .

See
p ,
@ -

740 if you
are
 @

( Relativistiebopelerskfsandaberateon
Consider observer and source of
light moving at speed along

vex
an a ✓

the observer 's X -

axis ,
In the source (
primed ) reference frame let
,

photon to
angle the
' '
the more at x x
-

axis :

µ
a

Source frame :

'
- ×

ObserverFame :

÷ = ×
Source moves with velocity
in this frame
The t
'

momentum
'
has
photon four -
and ×
components
'

pt
' '
' '
=
W
p× = K cos d

in the source (
primed ) reference frame ,
Lorentz transforming to the
observer frame ,

W =D ( w 't vkusx '

) = Vw '
( It vcosx
'
)

K cos a =D ( k
'
cos a
'
+ vw
'
)

.
^

.
w cos x= T ( w
'

cos a
'
+ vw
'

) ⇒ cos ×
'
=
WE -

a.

w=8w
( wyd i
)
'

¥
It
'

,
-
=
wv cos a +
.

wiggly
1

: the nelativistncdopplereffect
-

a)
'

W = w
( Itv cos for v < < 1

Note that for

×=E
relativistic
,
w
=¥ . This
dilation
is the transverse
-
Doppler effect
that arises from time ,

~%
twhd loss )
Tustin
#
'
=
cos x= + v

( Itv co 52
'
)
 740

⇒
cosxe.gg?.gt#-ve1ativistiaaberration_
Note
'
that I tend
for
Is <
Is photons
x < < ⇒
= cos =v x
,
i.e.
,
to be
¥ in the forward direction In fact
by moving source
a
,
,

for
for ultra relativistic motion 1 and
£=E
→ →
,
v ,
x 0 , ,

cos x = 1 -

Is = v

}
⇒ I = 2 ( 1 -

v ) = ( Itv ) ( 1 .

v ) =

f-
'
-

x
, .

i. e .
an
isotropic source beans half the
power
into a come of

angle
-
opening half -

YY .

IS
on

*⇐#bE
in
÷
- ,

)
ab frame
Source frame -

half the is beamed forward

power
( isotropic ) into a narrow cone .
 750

Observers and Observations

t

Many people
fail to obtain a
deep physical understanding of relativity
( both
special and
general ) by failing to
distinguish coordinates in
spacetime
from real measurements Take another look inertial
observers and ,
at a
global
reference frame in the Minkouski spacetime of
special relativity ,

On
p , 220 we described how this
yn
constructed
coordinate system could be

the
From motions of free particles ,
and it

EIEE.ee#EEisIEEfEEgm/0t
along
receive

k 's
information
her world
and

line
perform experiments
in This - ×

,
or

spacetime . Laboratories are local

=)
small within This vast
regions
universe of Ours ,

So what the heck do the

2-
( Ix , > ,
z ) coordinates of our

inertial reference frame ? find it useful to

global really mean I
imagine
zillions of observers who sit at rest at
every single spatial point
inertial
( x
,y , z ) in this reference frame .

They each carry gyroscopes

clocks talk their immediate to synchronize
and and to
neighbors in order

those clocks and

align those
gyroscopes They .
all reached agreement long
that special member of their cohort sits at the x=y=Z=O
ago
one
origin ,

distance measurements with

and
they all
, by combining their
neighbors ,

why sit relative to the ( whole bunch

Know
they origin ,
Imagine a

of homes in a
city each home
,
with its
unique
address and all
having
clocks that are synchronised ) .

When discussed Lorentz contraction of rod

moving
a
ve on
pp
.

490
-

@ we should have
emphasized that the simultaneous measurement
,

of the end
positions of the
moving
rod has to be
accomplished by two
observers who
happen to be at each end of the rod at previously
moving
a
,

# that Note
agreed on simultaneous me

seesa
.

single a observer who

moving
rod from the light that reaches his on her
eyes
will not
just
see Lorentz
a contracted object ,
because that's a

different sort of measurement .

 760

As beautiful and
elegant a
theory is relativity ,
in order to be part of
it
physics ,
or even science
, hey to be
put on the experimental chopping block
.

by making predictions white can be tested in a real observer 's

laboratory ,

So Lox do we treat the measurement

process
of an observer
relativity
in ?

arbitrary
It
Consider observer who world Kne
generic
is
an
moving on a

# will clock which

Through space
observer
carry
a and rulers

coordinates that
define within his or her local
laboratory , Let's assume

The observer wants to make their life and chooses an

orthogonal
easy
reference frame of unit 4- vectors with vectors Ii
basis
,
i. e , basis
Is , ,

which 0
em , I , satisfy e~£
Is
. = For
L¥P ,
and
Esu Is
= -1
,
such vectors all the
Is Es
Is ( have been
using basis
1
Ii
=
We
g, one ,
.
= ° =
.

time for inertial reference frames because had

Ex Ep
= The
zap
we
.

,
.

vectors
carets are there on the indices to
emphasise the fact that the basis

are orthonormal .

tn
In 's reference frame The

§
the observer Observer 's
,

observer 's own four -

velocity is of course World line

4=4
| fens
)
Eo
=
Es

Hence the time vector that

basis
e-; e ei
's
simply
I
the observer carries is The observer ~

Y velocity ! This makes -

/
is
sense as
U~ a
Y
-

like
unity time vector which is ever
>
when

tangent to the observer 's World line .

Consider particle with four momentum What does

f
now a -

our observer measure that

particle 's energy
to be
? Well
,
the observer

just measures the time

component of by projecting onto the basis
f
-

p
ve.tw !
Is

Eland =

f. Es f. an
-

=
-

T
1
sign because
-
=
minus
go
. ,
 770

What could be easier ? For example the

particle and the
,
suppose
both the of inertial reference frame
observer move
along x. axis an

with instantaneous speeds and respectively

up v. .

µ
=
f. e=
t % Vo
I ,
←
observer 4-
velocity in inertial frame basis

=
MY It MTPVPI
←
particle 4- momentum in inertial
f , ,

Hence
E) observed
= -

f. Un
=
-

( -

my t
my vpvo )

m7g% ( )
mltveudtffu
l
v.v
-
=
= .

Is this right ? Let's check it Lorentz transformation

by performing a

to an inertial reference frame which is the instantaneous rest frame

of the observer :

% ( P° vop
)
mtp Yo ( MAY )
'

G. =
v.
- = -

i. seed

= m
KR ( l -

v.v . ) =
m ( l -

vove )
✓
air
This is the same amount of work but as we
'll see
,
the dot product
,

method
proves
to be
quite powerful far observers in curved spacetime ,
would to comfortable with it
and it be
good for
you
become
using .

In general an observer
carrying an instantaneous ortho normal basis
,

components
themomentum of 's to
b
=
Is ,
Ii , Ii ,
Is measures a
particle 4-

yqgftloosuvd.fr#= 1 NO sum
onp )

in

.LI:1?eItYiIIII.e.ea..n=e.e.W-
 780

Lorentz transformations between inertial frames

Examples ; reference

oh flat spacetime
In spacetime
flat one can talk about
displacement 4- vectors
,

with some
meaning
( not true in curved spacetime as we
'll see later ) . Then
the
position vector of some
event is

z =
1 t
,x , y it
) =
tq +
XI ,
+
YI +713
"

Let the reference frame 0 ( the ) have 4 velocity

primed origin observer -

( 7,0 0,0 ) Is
'

y =
,
= ,

What are The spatial primed basis 4- vectors ? Well

,
clearly ,

e- , = ( 0,0 ,
1,0 ) Is , = ( 0
, 0,0 ,
1)

is the hard Let ( b 0,0 ) with b to be

In one .

It , = a
,
,
,
a and

determined from O ⇒ Vat Tub =O 8 low

Iii Io
=
-

as
°
,

and
Iii .

Ee ' =
1 ⇒ -
cites =)

i 155+15=1 ⇒ b
# =D =7v
=
a
-

,
-

'

Is ' = I k 8, 0,0 )
,
.
.

OK -
we
just need to project z onto this primed basis .

'

t fttttvx
'

xo I = -

z
.

Is , =
-

or t
'
= Nt .

✓ × ) ✓
XT
'

= ×
'
=
*
.

Ee
= -

th + ×V =
7 ( ×
.
ut ) ✓
Ii
'

Iy
'
=
z
.

Iii =

y
✓ xs 's t
'
=
z.gs
, =
z
✓
 790

So Loreto transformations flat spacetime

in can
simply be

expressed as
projections of event
position 4- vectors onto the
observer 's ortho normal basis ,