Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Relativity 2

Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

Relativistic dynamics

Lorentz transformations also aect the accelerated motion of objects under the inuence of forces. In
Newtonian physics a constant force F accelerates an abject at a constant rate a = dv/dt = F/m, so
that the velocity v = at of the object can become arbitrarily large if one waits long enough. This
cannot be in the relativistic physics because nothing can move faster than light. The acceleration must
become smaller and smaller as the velocity approaches c, which can be interpreted as the growth of
mass m. In this interpretation the velocity-dependent mass m(v) is sometimes called relativistic mass,
while the mass at rest m = m(0) is called rest mass.

We will derive the expression for m(v) mostly from kinematic principles. Consider two identical objects
of (rest) mass m. Let us label by B the reference frame xed to one of the objects (#1) so that the
other object (#2) moves away from it along x axis with velocity v . The trajectory of the moving object
is xB (tB ) = vtB . The center of mass of these two objects is located at:

m(0) 0 + m(v) xB m(v)


XB = = vtB
m(0) + m(v) m + m(v)
and moves with velocity
dXB m(v)
u= = v
dtB m + m(v)
Now set up a reference frame A xed to the center of mass of the two objects. This frame moves with
velocity u relative to B along x axis, producing the same set up for Lorentz transformations as before.
The object #1 which is at rest in B simply moves with velocity u in A. The other object #2 must
then move with velocity +u in the frame A. This follows from the requirement that the center of mass
appear at rest in the frame A, so that the two objects must move in A with the same velocity u in
opposite directions (and their relativistic masses m(u) will be the same).

We can relate the velocities of the object #2 in the frames A and B by using the velocity transformation
rule (vA = u, vB = v ):
vA + u 2u
vB = = v=
1 + vA u/c2 1 + u2 /c2
Let us write v = u and obtain the equation for :
2
u2 v 2 1 v2
     
v u v
1= 1+ 2 = 1+ 2 2 = 1+ 2 2
2u c 2u v c 2 c
v2 1 2
1+ 2 2
=
c
2
v
2 2 + 2 = 0
c
This quadratic equation has solutions:
r
v2
=1 1
c2
We cannot allow the center of mass to move faster than the object #2 in the frame B, so we must take
the solution with the plus sign ( > 1 v > u).
Finally, we go back to the expression which relates u, v and the masses:
r
v m + m(v) m v2
= = =1+ =1+ 1
u m(v) m(v) c2
We nd that
m
m(v) = q
v2
1 c2

7
What role exactly does m(v) play in dynamics? The only physical purpose of mass is to establish a
relationship between dynamics (forces) and kinematics (motion) of bodies, and our real goal here is to
discover how relativistic objects respond to forces.. In order to answer this question we must recognize
that dynamics implicitly enters the discussion above through the center-of-mass. For two bodies with
momenta p1 and p2 (in some reference frame) and masses m1 and m2 respectively, the center of mass
is a ctitious point which carries the total momentum P = p1 + p2 the way a particle with the total
mass M = m1 + m2 would, that is with velocity u = P/M . Unless there are external forces, the total
momentum is conserved:
dP dp1 dp2
= + = F21 + F12 = 0
dt dt dt
because the forces Fij exerted by the object i on the object j cancel out according to the third
Newton's law (the principle of action and reaction). We now want to dene mass by expressing a
particle momentum as the product of its mass and velocity, pi = mi vi . The total momentum is

P = m1 v1 + m2 v2 = M u

and its conservation implies u = const. no matter how the two particles interact with each other. This
u = const. is what allowed us to apply Lorentz transformations to the (inertial) center-of-mass frame.
Therefore, the above calculation is implicitly consistent with the second Newton's law

dp
F =
dt
and the denition of mass from p = m(v)v .
To summarize, the momentum of a relativistic object moving at velocity v is:

mv
p= q
2
1 vc2

where m is the object's rest mass (in modern terminology, m is simply called mass). If the object
is slow, we can neglect v/c so its momentum is simply p = mv . On the other hand, by applying the
constant force F we can build up arbitrarily large momentum p = F t in suciently large time interval
t. As p , the velocity v approaches the speed of light but never exceeds it.

Relativistic energy
Here we seek the relationship between energy and momentum in the special theory of relativity.

Consider a body of (rest) mass m initially at rest. Then, let us turn on at time t = 0 a constant force
F which accelerates the body. From the second Newton's law we immediately obtain:

dp
=F p = Ft
dt
since the force is constant, and p = 0 at t = 0. The kinetic energy T acquired by the time t is equal
to the work Fx done by the force F , where x(t) is the position of the body. Since the velocity is
v = dx/dt, we can obtain position by integrating out velocity:

t
T = Fx = F dt0 v(t0 )
0

Let us express the current velocity v in terms of momentum p. We can take the square of

mv
p= q
2
1 vc2

8
and obtain
v2
 
2 2 2
m v =p 1 2
c
p pc
v=q =p
p
m2 + c2
2
m c2 + p2
2

Substitute this in the expression for energy:

t
p(t0 )c
T =F dt0 p
m2 c2 + p(t0 )2
0

and note that we can pull the constant force inside the integral and combine it with time, F t0 = p(t0 )
0 0
p. At the desired nal time t the momentum p becomes p. The above integral can be therefore written
as:
p
p0
T =c dp0 p
m2 c2 + p02
0

This is easy to solve. We rst change the integration variable = p02 by noting that p0 dp0 = d/2.
Then we use the integration formula:

(a + )n+1
d(a + )n =
n+1
for n = 1/2 and a = m2 c2 . The result is:
p0 =p p
c d c p
T = = 2 m2 c2 + p02 0 = (mc2 )2 + (pc)2 mc2

p
2 2 2
m c + 2 p =0

Even though this expression was derived by considering constant acceleration, it holds regardless of
how the force F changes in time. The Newtonian principle of converting work to energy is merely a
formal device to dene energy as a conserved quantity. We could have turned o the force F for a
while and continued accelerating later; both momentum and energy would not change during the wait
time. Therefore, energy can depend only on momentum (and mass) of the body.

It is customary to dene the total energy of a body as


p
E= (pc)2 + (mc2 )2

where
E0 = mc2
is considered to be the energy at rest when momentum p = 0. This is the famous Einstein's equation
which hints that the mere mass of objects could be a source of energy. In Newtonian physics mass is
conserved, so one could not extract energy from mass and use it to do some work. However, the theory
of relativity already allowed us to view mass as something that can exchange energy with environment:
when we accelerate an object whose velocity approaches the speed of light, the kinetic energy must be
dumped into the increasing relativistic mass because the speed of light cannot be exceeded. The true
power of this conversion between mass and energy is unleashed in nuclear reactions.

How do we obtain the non-relativistic limit? At small velocities v momentum p is


small so that
pc  mc2 . 1 + x 1 + x/2 for small x we obtain:
Expanding the square root as
r
p
2 2 2 2 2
 pc 2
2 p2 c2 p2
T = (mc ) + (pc) mc = mc 1 + mc =
mc2 2mc2 2m
which is the familiar expression for kinetic energy in Newtonian physics.

9
Covariant formulation of the special theory of relativity
The Einstein's rst postulate requires that all laws of physics be the same in all inertial frames of
reference. We say that the laws of physics must be invariant under Lorentz transformations. In this
section we discuss a powerful formalism which allows one to formulate the laws of nature in a manifestly
covariant way, so that they automatically keep their form under Lorentz transformations.

The foundation of the covariant formulation is the so called Minkowski space. This is a four-dimensional
combination of space and time with appropriately dened metric for expressing distances between
points. The coordinates in Minkowski space are 4-vectors (ct, x, y, z) (ct, r) which we label by x .
The subscript indexes any of the four coordinates.

The metric tensor is dened by a matrix


1 0 0 0
0 1 0 0
g = g =
0

0 1 0
0 0 0 1

and used in the following fashion. The length of a vector


qP x in Minkowski space is dened by L(x) =

x x where x is dened as
X
x g x

In general, vectors can be specied with a lower or upper index, and g or g is used to convert
between lower and upper index vectors by the convention that a lower index can be contracted only
with an upper index in a summation over indices.

If there weren't for the 1 element in g , the vector length would be the same as in any Euclidean
vector space (like our three-dimensional space). However, the given form of g is needed in order to
make a vector length invariant under Lorentz transformations:

X
L2 = x x g = x2 + y 2 + z 2 (ct)2

Under Lorentz transformations

ct0 = (x + ct)
x0 = (x + ct)
y0 = y
z0 = z

the length becomes:

L0 = x02 + y 02 + z 02 (ct0 )2
= 2 (x2 + 2 c2 t2 + 2xct) + y 2 + z 2 2 ( 2 x2 + c2 t2 + 2xct)
= 2 (1 2 )x2 + y 2 + z 2 2 (1 2 )(ct)2
p
and since = 1/ 1 2 we nd that L0 = L.

Since 4-vector length is invariant under Lorentz transformations, we can interpret Lorentz transforma-
tions as four-dimensional rotations in Minkowski space. In tensor notation the Lorentz transformations
are: X
x0 = x

10
where
0 0
0 0
=


0 0 1 0
0 0 0 1
a b
P
Any scalar product of two 4-vectors is automatically invariant under Lorentz transformations.

An interval s12 between two events described by Minkowski vectors x1 and x2 is dened by

X
s212 = (x1 x2 )(x1 x2 )

If the two events occur at the same location, then the interval is s2 = c2 (t1 t2 )2 < 0. If they occur
2 2
at the same time, the interval is s = (r1 r2 ) > 0. An interval is called time-like if s2 < 0 and
2
space-like if s > 0.
Two events separated by a space-like interval cannot be causally connected. Clearly, one can nd
a reference frame in which such events occur simultaneously at dierent locations so that physical
inuence has no chance to propagate from one location to the other. Since the laws of physics must be
the same in all reference frames, they must agree in all frames that the two events cannot be a cause
or consequence of each other.

We dene proper time as the time measured in the reference frame xed to an object whose motion
we want to track. We can express the trajectory of the object in some reference frame x (t). However,
if we attempt to dene velocity as dx /dt we nd that it does not transform between reference frames
according to Lorentz transformations. This is so because x transforms properly, but t transforms as
well. On the other hand,
dx
v =
d
is a valid 4-vector in Minkowski space which transforms according to Lorentz transformations because
the innitesimal intervals d are Lorentz-invariant. This is the covariant denition of velocity. In
a similar fashion we dene covariant acceleration a , momentum p and force F ( m is the rest
Lorentz-invariant mass):

dv dp
a = , p = mv , F = = ma
d d
These equations specify the complete relativistic dynamics in a form which is automatically compatible
with Lorentz transformations.

In terms of familiar quantities from Newtonian physics, three-dimensional position r, velocity v, ac-
celeration a, and momentum p vectors, the covariant 4-vectors are:

x = (ct, r)
v = (c, v)
 
Fv
a = , a
mc
 
E
p = ,p
c
 
Fv
F = , F
c

with
1
=p
1 v 2 /c2

11
Einstein's general theory of relativity

The special theory of relativity is a complete description of dynamics in inertial reference frames.
However, already at the level of this denition there is a fundamental problem. An inertial reference
frame is dened as a frame in which all forces acting on objects originate from other objects in a manner
that depends on the objects' positions, velocities, etc. How can we tell for sure that the net force on a
body has such a proper origin? All we can do is measure the force based on the body's acceleration.
If we knew all laws of nature exactly and could measure the positions and velocities of all bodies in
the universe, perhaps we could test whether the measured net force is a combination of forces with
proper origins. Another way to appreciate the problem is to note that a frame moving uniformly with
respect to an inertial frame is itself inertial. In order to ultimately establish that our frame of interest
is inertial, we must be able to nd its velocity relative to a universal special frame which is inertial for
sure. The special theory of relativity denies the existence of such a preferred frame.

Other problems can be found in examples containing accelerated motion. For instance, consider a
planet orbiting a very heavy compact star at a relativistically large velocity. What is the radius of the
orbit perceived from the planet's surface? The distance to the star, radius r, is always measured in
the direction perpendicular to the planet's motion. However, the orbit perimeter appears shorter than
p
2r by a factor 1 v 2 /c2 to the observer on the planet, even though the orbit looks circular to him.
On the other hand, an observer hovering at one place on the surface of the star can clearly see that the
planet moves at the distance r along the perimeter of length 2r. Whose judgment of space geometry
is wrong? Surely, we cannot trust the predictions of special relativity for the observer on the planet,
because his reference frame is not inertial (rotation is accelerated motion).

The famous twin paradox illustrates a similar problem. One twin brother stays on Earth while the
other one takes a space round-trip from Earth to a nearby star and back at near light-speed velocities.
The guy staying on earth watches broadcast video reports of his brother during this (say) 50 year long
journey (the star is about 50 light years far away from Earth). Time appears to run very slowly in these
reports for the most of the trip duration because of time dilatation, so when the two brothers meet
again on Earth, the traveler had aged only a few months in the 50 years that passed for his brother
on Earth. However, the traveler also receives video reports from his brother on Earth. To the traveler
Earth appears to move at the near-light-speed relative velocities, so he should observe a very slow
passage on time on Earth. When the two brothers meet again, the guy staying on Earth should have
aged negligibly in comparison to the traveler. Well, only one of these scenarios could happen. Which
one? Surely we cannot expect the predictions of special relativity to work for the traveler because his
frame of reference is not inertial (in order to turn around and come back to Earth, the traveler must
carry out accelerated motion).

The general theory of relativity overcomes all these diculties and does not require ill-dened inertial
reference frames.

Postulates
Einstein formulated the general theory in the same fashion as the special one. The fundamental
postulates are:

1. The laws of nature must be the same in all reference frames (no preferred reference frame can be
deduced from the equations of motion alone).

2. Inertial forces in an accelerated frame of reference are fundamentally indistinguishable from gravity
(equivalence principle)

3. The principles of special theory of relativity hold on length and time scales where motion appears
approximately uniform.

The rst postulate is the same as in the special relativity, except that now it applies to any reference
frame, not necessarily inertial.

12
The second postulate starts from a benign observation that gravitational force acting on an object is
always proportional to its mass. For example, the gravitational force between two objects with masses
M and m is according to the Newton's formula

Mm
F=G r
r2
where r is the distance between the two objects, r the unit vector along the direction between the
objects, and G is the gravitational constant. We could interpret this expression as F = mg where g is
the gravitational acceleration. If the body with mass M is very far away and very massive, all objects
will experience the same gravitational acceleration g while the force on each object will be proportional
to its mass. We could simulate gravity by a fast elevator. If the elevator accelerates with acceleration a
and we set up a clock and a coordinate system in it, we will nd that there is an inertial force acting on
every object inside the elevator proportional to the object's mass and a. This looks just like gravity.
Centrifugal force is another example, but this inertial force also has radial orientation, just like gravity
from a nearby massive star. Astronauts inside the International Space Station can oat around even
though Earth pulls them down; it is the centrifugal force which cancels out the force of gravity, leaving
behind the zero net force on astronauts.

The equivalence principle is taken very seriously in the general theory of relativity. It provides the
means to link the origin of dynamics, gravity, to its kinematic eect, accelerated motion. This is
mathematically accomplished by constructing a curved Minkowski space-time. Dynamically, curvature
is caused by massive objects in the universe. Kinematically, objects are regarded to always move freely
by inertia, but in the curved space-time. It is this curvature that gives rise to acceleration. The
mathematical description of curved space-time dynamics is well beyond the scope of this course.

The third postulate is the backbone for the development of the formalism. It implies that if there are
no massive sources of gravity around, light objects will move on straight lines and obey the rules of the
special theory of relativity. Otherwise, we can obey the rules of the special theory only in very short
time intervals during which an object moves along a short straight segment.

Some predictions
Moving massive objects exert forces similar to magnetic forces in electrodynamics. Such forces are
behind the intricate precession of Mercury's orbit in our solar system.

Massive galaxies and stars can bend the trajectory of light and act like gravitational lenses. Multiple
images of the same stars are observed in many places in the sky due to gravitational lensing.

A black hole is a star whose gravitational eld is so strong that light emitted from its surface must
unavoidably fall back on it. Einsteins equations which summarize the relativistic physics of gravity
predict in fact that there is a spherical boundary around the black hole, called event horizon, beyond
which no information can be sent outside. Any object which nds itself inside the event horizon will
inevitably fall onto the black hole in a nite amount of time, and no physical force can be generated to
prevent that. The same applies to any substance at the black hole surface, there is no physical process
which can sustain pressure strong enough to prevent the gravitational collapse onto the point at the
center, the singularity. Hence, the general theory of relativity predicts that the black hole is a point
of innitely small volume, nite mass and innitely large density. The radius of the event horizon is a
function of the black hole mass, but the black hole cannot have any other properties (except angular
momentum and charge).

Strong gravitational elds cause time dilatation.

Accelerating massive objects emit gravitational waves which can then propagate independently through
space. A gravitational wave is an oscillatory wave-like stretching and contracting of the fabric of
space-time, aecting local measurements of time intervals and length. There is an ongoing experiment
attempting to detect gravitational waves produced in cataclysmic cosmic events (stars falling on black
holes, etc.). However, gravitational waves are extremely weak.

You might also like