Chapter 5

The Equivalence Principle

There was a minutes pause perhaps. The Psychologist seemed about to speak to me, but changed his mind. Then the Time Traveller put forth his nger towards the lever. No, he said suddenly. Lend me your hand. And turning to the Psychologist, he took that individuals hand in his own and told him to put out his forenger. So that it was the Psychologist himself who sent forth the model Time Machine on its interminable voyage. We all saw the lever turn. I am absolutely certain there was no trickery. There was a breath of wind, and the lamp ame jumped.

5.1

Inertial mass
Gravitational mass is equivalent to inertial mass.

Newtons second law states that the force on an object is proportional to mass times acceleration. In this section, we will call the mass which appears in this law the inertial mass: F = mI . a (5.1) for example, consider the electro-static interaction between two particles with masses m1 , m2 , and charges q1 , q2 . Particle 2 feels a force q 1 q2 F 2 = 2 r12 . r12 (5.2)

The acceleration felt by particle 2 can then be found by combining these two equations: = a2 q2 m2 I q1 r . 2 12 r12 (5.3)

Thus, 2 depends on the ratio of charge to inertial mass, q2 /m2 I . a Next, consider a particle falling to the ground. The Newtonian gravitational force resembles the electrostatic force (both are inverse square laws), with masses replacing electrical charges. We will simply call the gravitational charge m. Let M be the mass of the Earth, r the distance to the centre of the Earth, and r the upwards unit vector.
62

5.2 Free Fall Then the force on an apple falling to the ground can be written CM m F = r, r2 where C is a constant. Now, by Newtons 2nd law F = mI a m = = a mI

63

(5.4)

(5.5) C M r2 r. (5.6)

Thus, depends on the ratio of gravitational mass to inertial mass, m/mI . Galileos a experiments should be the same for all materials, once air resistance is neglected. This a implies that m/mI is the same constant for all matter. Thus, we can combine the two constants C and m/mI into a single constant m mI GM = = 2 r. a r G = C (5.7) (5.8)

We now see a fundamental dierence between the electrical force and the gravitational force: the former depends on a charge-to-inertial-mass ratio, but the latter does not. The inertial forces (also known as ctitious forces) are similar to gravity in this respect the acceleration has no dependence on mass.

5.2

Free Fall

Einstein had diculty incorporating both gravity and inertial forces into special relativity. His great insight was to treat them together, using the principle of equivalence to eliminate the dependence on mass. Now, inertial forces can be eliminated by transferring to a nonaccelerating frame. Einstein reasoned that gravitational forces can be removed in a similar way by transferring to a free-fall frame. Example: Consider an object in the Earths gravitational eld near r = R . Let z be the vertical direction. d2 z = g dt2 GM g= R 10ms1

64 initial conditions: z0 = z(t = 0) = h z0 = V0 = z(t) = gt + V0 1 z(t) = h gt2 + V0 t 2 Lets transform to new co-ordinates: 1 (z, t) = z + gt2 2 Then: = h + V0 t = V0 = constant =0

The Equivalence Principle

Since the acceleration is zero, there is no gravitational force. 5.2.1 Locally Inertial Frames

noindentDenition Locally Inertial Frame (LIF) An LIF is a reference frame with origin at space-time event P . An object at P is in free-fall (if there are no external forces). Near P , there are no gravitational forces, and special relativity holds. In an LIF a. gab b. c gab = 0.
P P

= ab ,

(5.9)

(5.10)

for all a, b, c. The second condition follows from the isotropy of space-time.

5.3

Geodesics

noindentDenition Geodesic a. A geodesic is a path on a manifold M which is an extremum of length (i.e. max or min distance between two points). b. A geodesic is a path which has zero covariant acceleration (to be dened later).

5.3 Geodesics 5.3.1 Examples

65

a. R2 . Geodesics are straight lines. b. S 2 . Geodesics are the great circles, for example the equator. c. Minkowski Space M 4 . In special relativity, distance becomes proper time, d 2 = ds2 = dt2 dx2 dy 2 dz 2
t
111 000 Q 111 000 111 000

Path 1 Path 2

11 00 11 00 11 00 11 00 P

Path 1: 1 = Path 2: 2 = < dt2 dx2 dy 2 dz 2 dt2 = t. (5.12) (5.13) d = dt = t. (5.11)

Thus, 1 > 2 . An object at rest has maximum proper time (and follows a geodesic in space-time). 5.3.2 The Geodesic Equation

In general relativity, the orbit of a satellite is a geodesic in a space-time distorted by the mass of the Earth. How can we nd the orbit? We seek an expression for the acceleration of the satellite in ordinary coordinates xed to the Earth. In the locally inertial (free-fall) frame, of course, the acceleration is easy: it is zero! Let ( 0 , 1 , 2 , 3 ) be coordinates in the satellites LIF, while X are coordinates xed to the Earth. In other words, near the spacetime event ( 0 , 1 , 2 , 3 ) = (0, 0, 0, 0), the satellite is at rest in the LIF coordinates, and experiences no forces and no acceleration:

66

The Equivalence Principle

a ULIF =

1 d a 0 = ; 0 d 0 d2 a = 0. d 2

(5.14)

a dULIF d

(5.15)

Let U be the 4-velocity in Earth coordinates X. We now transform ULIF to U: d d U = 0= d LIF = d d d b U Xb d d d d b Ub + U d Xb Xb d (5.16) (5.17)

by the product rule. To understand the rst term, we use the fact that d d dt dx + + ... d t d x = U 0 0 + U 1 1 + . . . = U c c = (5.18) (5.19) (5.20)

to obtain 0 = U b U c c d d d b U + Xb Xb d 2d d d b U . = U bU c b c + X X Xb d (5.21) (5.22)

The last term contains what we are searching for: the acceleration in the Earth frame dU b / d . However, we need to free this expression from the transformation matrix d /Xb . To rid ourselves of this unwanted matrix, we multiply by its inverse Xa / d : Xa d = ab. d Xb This gives 0 = 2d Xa d d b U bU c b c + U d X X Xb d Xa 2 d d b = U bU c + ab U . d Xb Xc d (5.24) (5.25) (5.23)

Now, a b dU b / d = dU a / d . Rearranging terms, we nally obtain dU a Xa 2 d + d U bU c = 0 . b Xc d X (5.26)

5.3 Geodesics Equation (5.26) can be written in the form dU a + a bc U b U c = 0, d where the Christoel symbols a bc are given by a bc = Xa 2 d . d Xb Xc

67

(5.27)

(5.28)

Equation (5.26) is called the geodesic equation, and governs the motion of matter in the absence of forces. A more useful formula for the Christoel symbols will be derived in the exercise below, and (in a dierent way) in the next chapter. Equation (5.26) has been introduced here because of its physical meaning. It computes the apparent acceleration dU a / d of an object in one frame (X) in terms of transformations from the LIF. We experience gravitational forces only because we insist on viewing things from a non-inertial frame! Like any other ctitious or inertial force, gravitation arises from the acceleration of one frame of reference with respect to the inertial frames. We do feel the eects of weight, particularly after a long hike uphill, but we can now view this as the eect of forces coming from the ground under our feet, accelerating us away from our natural state free-fall. Exercise 5.1 Here we derive an expression for the Christoel symbols in terms of the metric. To simplify the notation, gab will denote the metric in the non-inertial frame, gLab the metric in the LIF, and a /Xa (i.e. the shorthand for partial derivatives applies only to the noninertial X frame). a. Show that c gab = (c a f )(b g ) + (a f )(c b g ) gLf g . b. Show that abc = gae e bc = (a f )(b c g )gLf g . c. Show that abc + acb = 2c gab . d. Hence show that 1 a bc = g ad (b gcd + c gdb d gbc ) . 2 (5.31) (5.32) (5.30) (5.29)

5.3.3

Covariant Acceleration

Theorem. If the components of a tensor vanish in one co-ordinate system, then they vanish in all frames. Proof. This follows directly from the transformation laws for a tensor. For example, ab if in frame A we have MA = 0 for all a, b, then in frame B
cd MB =

Bc Bd ab (0) = 0. Aa Ab

(5.33)

68

The Equivalence Principle

Note that the acceleration term dU a /d is zero in the LIF, but non-zero in the Earth frame. Thus it is not a tensor! To understand the motion of objects in general relativity further, we must learn how to dierentiate vectors in a covariant way (i.e. so that the result is a tensor). A vector U involves not only its components U a , but also the basis vectors of the coordinate system. If space-time is warped, or even if we are simply using non-Cartesian coordinates, these basis vectors will point in dierent directions at dierent points. In the next chapter, we will show how to dierentiate vectors by including both components and basis vectors.

Chapter 6

Covariant Derivatives
One of the candles on the mantel was blown out, and the little machine suddenly swung round, became indistinct, was seen as a ghost for a second perhaps, as an eddy of faintly glittering brass and ivory; and it was gonevanished! Save for the lamp the table was bare.

How do we dierentiate vectors (& tensors)? There are many examples in physics where derivatives need to be extended. For example, in uid mechanics the Navier-Stokes force equation reads DV = p + 2V (6.1) Dt where D/Dt is the total Lagrangian derivative D/Dt = /t + V . (6.2)

In Quantum mechanics, Schrdingers equation has the form o 1 2m

2

+V

= E.

(6.3)

With an applied magnetic eld the gradient is replaced by the gauge covariant deriva ie tive m A, where A = B. In this context the vector potential A is sometimes called the electromagnetic connection.

6.1

Non-Euclidean Geometry

A Co-ordinate Line is a line parameterized by one of the co-ordinates. A basis vector is a tangent vector to a co-ordinate line. Let eO be the basis vector tangent a to the co-ordinate line following xa . Note A circled subscript appears because the subscript chooses between the vectors in a set e.g. the set {eO, eO, eO, eO} ordinary subscripts choose the component of a 0 1 2 3 single vector.
Basis Vectors 69

70

Covariant Derivatives

Thus, in component form (for co-ordinates X) 0 dX0 /X1 dX1 /X1 1 eO = 1 dX2 /X1 = 0 0 dX3 /X1 = eO0 = 0, eO1 = 1, . . . 1 1 a a or eO = 1 1 In general,
a eOa = c . c

(6.4) (6.5) (6.6) (6.7)

Note that the basis vectors need not be orthogonal or of unit size: eO eO = gad eO a eOd b c b c = = gbc
a d gad b c

(6.8) (6.9) (6.10)

i.e. the scalar product of basis vectors eO and eO is equal to element gbc of the metric. b c All vectors can be written as sums of basis vectors: V0 V1 V = = V c eO. (6.11) c V2 V3

6.2 The Covariant Derivative

71

6.2

The Covariant Derivative

bV

Derivatives must satisfy the product rule. Thus for the derivative in the Xb direction =
b

V c eO = ( c

bV

) eO + V c ( c

c b eO).

(6.12)

V c is a number at each point (i.e. a function of position), so we can write V c ( b V c ) = b V c = Xb = = (b V c ) eO + V c c c bV b eO . We need to dene the last term in brackets. The object will name its components with the capital Greek letter : 0 bc 1 bc c b eO = 2 . bc 3 bc In terms of the basis vectors,
c b eO c b eO

(6.13) (6.14)

is itself a vector. We

(6.15)

= 0 bc eO + 1 bc eO + 2 bc eO + 3 bc eO, 0 1 2 3

(6.16)

or
c b eO

= a bc eO . a

(6.17)

The object a bc is called a metric connection, or alternatively a Christoel symbol. Let us go back to calculating the gradient of a vector:
bV

= (b V c ) eO + V c c = (b V c ) eO + V c c

c b eO a bc eO a

(6.18) . (6.19) (6.20) (6.21)

Exchange a c in the 1st term on the RHS: =

bV

= (b V a ) eO + V c a bc eO a a a c a = (b V + V bc ) eO. a
a

This is a vector, with components

bV

= (b V a + a bc V c )
b

To recap: The covariant derivative, a. Produces tensors. b. Obeys the product rule. c. For a scalar function f ,

(derivatives in Xb direction)

bf

= b f =

f . Xb

(6.22)

d. There exists a set of numbers a bc , where

bV a

= b V a + a bc V c .

72

Covariant Derivatives

6.3

Derivatives of Other Tensors

Use the properties listed above. Example: Given a 2nd rank tensor, Mcd , nd ( b M )cd . To do this, let V c , W d be arbitrary vectors. Let f = Mcd V c W d be a scalar function. By the product rule for b , b f = (b Mcd ) V c W d + Mcd (b V c ) W d + Mcd V c b W d . Also by the product rule for
bf b,

(6.23)
d

=(

b M )cd

V c W d + Mcd

bV

W d + Mcd V c

bW

(6.24)

Meanwhile, (

bf

= b f , so V c W d = (b Mcd ) V c W d + Mcd b V c +Mcd V c b W d

bW d bV c

b M )cd

Wd

Now apply rule(4): (

b M )cd

V c W d = (b Mcd ) V c W d + Mcd (c ba V a ) W d + V c d ba W a

(6.25)

Next we factor out V and W . To do this, we swap a c in the second to last term, and a d in the last term: (
b M )cd

V c W d = (b Mcd ) V c W d (a bc Mad + a bd Mca ) V c W d .

(6.26)

As the above is true for any values of V c and W d we can cancel V c W d from both sides, with the nal result (
b M )cd

= b Mcd a bc Mad a bd Mca .

(6.27)

In general, we will obtain one for each index of a tensor, with a minus sign for each subscript (form) index and a plus sign for each superscript (vector) index. Exercise 6.1 The covariant derivative b of a vector eld X a is
bX a

= b X a + a X c . bc
af

(6.28)

Also, the covariant derivative of a scalar is the same as the partial derivative: = a f.
b

(6.29) of a form Wc ,

Use these equations to nd an expression for the covariant derivative i.e. nd b Wc .

Exercise 6.2 In general Relativity, the relation between the Faraday tensor and the vector potential a is dened by Fab = (
b )a

a )b .

(6.30)

Write out the right hand side in terms of Christoel symbols. Show that the Christoel terms cancel, leaving Fab = (b )a (a )b . (6.31)

6.3 Derivatives of Other Tensors Suppose the operator (covariant derivative) satises g=0 b. a bc = a cb (zero torsion) (g = metric)

73

Theorem:

a. (6.32) (6.33)

then the Christoel symbols are determined by 1 a bc = g ad (b gcd + c gdb d gbc ) . 2

Proof:

(6.34)

gcd is a 2nd order tensor with two lower indices, so (

b g)cd

= b gcd a bc gad a bd gca . dbc gad a bc .

(6.35) (6.36)

Dene Then (
b g)cd

= b gcd

= b gcd dbc cbd = 0 = dbc + cbd .

(6.37) (6.38) (6.39)

Next apply assumption 2: cdb = cbd , so b gcd = dbc + cdb . Obtain two more equations by cycling b c, c d, and d b: = c gdb = bcd + dbc d gbc = cdb + bcd . Take the sum (equation (6.40) + equation (6.41) - equation (6.42)): b gcd + c gdb d gbc = 2dbc + 0 + 0 1 = dbc = (b gcd + c gdb d gbc ) . 2 Finally, use gdb = gbd in the second to last term and let a bc = g ad dbc (6.45) (6.43) (6.44) (6.41) (6.42) (6.40)

to prove the theorem. Exercise 6.3 Consider a sphere of radius 1 as a 2-dimensional manifold with coordinates X1 = (colatitude) and X2 = (longitude). What is the metric gab and its inverse

74

Covariant Derivatives

g bc ? Find the Christoel symbols a bc (there are 8 of these for a 2-dimensional manifold). Suppose a geodesic on the sphere is parameterized by . Use the geodesic equation d2 Xa dXb dXc =0 + a bc d2 d d to nd 6.3.1
d2 d2

(6.46)

and

d2 . d2

The gradient of the metric in General Relativity

For an object in free-fall (no external forces) the geodesic equation gives dU a + a bc U b U c = 0. d (6.47)

This is true in all coordinate frames. But in the LIF the object will appear to be at rest (or in uniform motion). Thus the components of U in the LIF remain constant, i.e. dU a / d = 0. This implies that a bc U b U c = 0 in LIF. (6.48)

Since this holds for arbitrary 4-velocities U, we must have the Christoel symbols vanishing, a bc = 0 in LIF. (6.49) (Strictly speaking, only the part of a bc symmetric in the lower indices b and c need vanish. Einsteins theory employs the simplest assumption, that the antisymmetric part of a bc called the torsion is always zero. Some modied theories of gravity include a non-zero torsion.) Now, ( c g)ab = c gab d ca gdb d cb gad . (6.50) In the LIF, however, d ca = d cb = 0. Also, by the denition of locally inertial frames, c gab = 0. Thus g = 0. (6.51) And, since g is a tensor, it must vanish in all frames. Thus in General Relativity equation (6.34) can be used to calculate the Christoel symbols.

6.4

Covariant Directional Derivatives and Acceleration

Consider a curve parameterized by whose tangent vector is V (see section 1.5.1). For a function f , the derivative in the direction along the curve is (Recall equation (1.76)): df = V f d = V b b f. (6.52) (6.53)

6.5 Newtons Law of motion For the directional derivative of a vector along the curve, the covariant gradient includes Christoel symbols. Thus for a vector W, DW =V D or, expressed in components, DW D
a

75 now

W,

(6.54)

W
bW a

a a

(6.55) (6.56) (6.57) (6.58)

= Vb

= V b (b W + a bc W c ) dW a + a bc V b W c . = d

We can apply this to nd the acceleration 4-vector. Let the world-line of an object be parameterized by its proper time , with tangent vector the 4-velocity U. Then the 4-acceleration is

a = aa

DU = U U; D a = U b b U = U b (b U a + a bc U c ) dU a = + a bc U b U c . d

(6.59) (6.60) (6.61)

Note the similarity with the geodesic equation, equation (6.47): The geodesic equation can now be written in a very simple form:

a =0 . 6.5 Newtons Law of motion

(6.62)

Newtons 2nd law becomes, for an external force f a f = ma = m

a a

DU D

(6.63)

If f a = 0, an object follows a geodesic. We can now describe the apparent acceleration of component a of the 4-velocity, U a : dU a 1 a = f a bc U b U c . d m (6.64)

The rst term on the RHS arises from external forces, while the second term arises from ctitious inertial forces, including gravity.

76

Covariant Derivatives

6.6

Twin Paradox

In the year 2000, one twin sets o for a distant planet, while the other twin stays home. Flight Plan (in ship time = proper time): a. 5 year acceleration 1g 10 ms2 , the surface acceleration of the Earth. b. 5 year deceleration 1g. c. 1 year on planet. d. 5 year acceleration 1g. e. 5 year deceleration 1g. In relativistic units g 1.03yr1 . When does the twin arrive back on Earth? We need to compare Earth time tE with the proper time . Conveniently, as measured 0 in the Earth frame the zeroth component of the 4-velocity UE is
0 UE =

dtE = . d

(6.65)

So (setting tE = = 0 at the start of the journey) tE ( ) =

0

( ) d .

(6.66)

Strategy: Compare U a and a a in both the Earth and the spaceship frame. We will ignore the y and z components, considering only the t and x components. Ship: in the spaceship rest frame,
a US ( ) =

1 . 0

(6.67)

The astronauts feel a force of one Earth gravity, which allows them to walk around the spaceship rather than oat about. This force is the normal force from the oor acting on the feet of the astronauts. For an astronaut of mass m, Newtons second law states that the normal force (in the forward S1 = x direction) is F = mg. Let us write this in covariant form: the covariant Newtons 2nd law reads
a FS

(6.68)

=m

DUS D

= mg.

(6.69)

Expanding the covariant derivative in terms of ordinary derivatives and Christoel symbols,
a FS = m a dUS b c + a US US Sbc d

a FS . m

(6.70)

6.6 Twin Paradox But in the ships rest frame

a US =

77

1 0

constant,

(6.71) (6.72)

a dUS = 0. d

Thus
a FS = ma . S00

(6.73) (6.74)

In particular, 1 = g S00 ( 1.03 year1 ).

a Thus 1 gives the ctional (inertial) force felt by the astronauts. Also, as FS USa = 0, S00 0 we must have FS = 0 and 0 = 0. S00 Earth: Assume the Earth frame is an inertial frame, a = 0, i.e. ignore the Earths own Ebc gravity. Then a a DUE dUE + 0. (6.75) = D d

Now, Ea b U ( ). (6.76) Sb S The co-ordinate transformation between the Earth frame and the spaceship frame depends on the speed and hence the position of the ship. The position of the ship is parameterized by the proper time . For a ship travelling at speed V ( ) the Lorentz boost formula gives
a UE ( ) =

Ea = Sb 6.6.1 The rapidity

( ) V ( ) . V ( ) ( )

(6.77)

Let the rapidity be dened by = tanh1 V . Thus the quantities V , = (1 V 2 ) and V are given by simple hyperbolic functions: V = tanh , = cosh , V = sinh . In terms of the rapidity, the transformation matrix has the simple form Ea = Sb cosh ( ) sinh ( ) . sinh ( ) cosh ( )

1/2

(6.78) (6.79) (6.80)

(6.81)

78

Covariant Derivatives

Because both the velocity 4-vector and the covariant acceleration are tensors, they can be readily transformed to Earth coordinates from the spaceship frame as in equation (6.76): UE ( ) = = DUE = D = = g cosh sinh sinh cosh cosh ( ) ; sinh ( ) cosh sinh sinh cosh cosh sinh sinh cosh sinh ( ) . cosh ( ) DUS D 0 g 1 0 (6.82) (6.83) (6.84) (6.85) (6.86)

a Meanwhile, the ordinary derivative of UE is a dUE = d

d d

cosh ( ) sinh ( )

(6.87) (6.88)

sinh ( ) , = cosh ( ) where = d/ d . Thus from equation (6.75), we nd = g, which integrates to ( ) = g.

0 We can now go back to equation (6.66). As UE = = cosh(g ), we have

(6.89)

tE ( ) =
0

cosh(g )d 1 sinh(g ). g

(6.90) (6.91)

tE ( ) =

Now g = 1.03yr1 , and so at = 5 years, tE = sinh(5.15)/1.03 = 86.6yr. At this point, Earth time is 2086, Ship time is 2005. Similarly, the 5 year deceleration takes 86.6 years on Earth; 1 year on alien planet takes 1 year on Earth; the return journey takes 2 86.6 years on Earth. So the twin returns to Earth 21 years older, in the year 2347. Note the asymmetry between the stay-at-home twin and the space-faring twin. Earth people see the spaceship moving away and returning. But people on the spaceship also see the Earth moving away and returning! However, there is no true symmetry here: only the spaceship resides in a non-inertial rest frame. This results in a true dierence between the ow of time on the spaceship and on the Earth. (6.92)

Chapter 7

Orbits
Everyone was silent for a minute. Then Filby said he was damned.

7.1

Noethers Theorem

For any continuous symmetry of a physical system, there is a conserved quantity. This theorem is most often expressed in the context of Hamiltonian or Lagrangian mechanics, either quantum or classical. For example, if H is the Hamiltonian for a physical system, then: a. If dH/ dt = 0 (H symmetric to time translation), energy is conserved. b. If H/x = 0 (H symmetric to translation in the x direction) then the x component of linear momentum is conserved. c. If H/ = 0 (H symmetric to rotation), then angular momentum is conserved d. In electromagnetism, if H is independent of gauge, then charge is conserved. e. In particle theory, gauge symmetry can imply conservation of other kinds of charge. For example SU(3) symmetry implies conservation of colour (strong force) charge. To employ symmetry arguments in the analysis of orbits, we rst prove a variant of Noethers theorem applicable to geodesics.
Theorem:

dpa m = (a gbc ) U b U c . d 2

(7.1)

Thus, for example, if 0 gbc = 0, then dp0 / d = 0. In this case the energy E = p0 will be conserved. Proof 7.1 We derive the corresponding equation for the lowered form of the 4-velocity U. Here U = p/m can be interpreted as energy-momentum per unit mass.
79

80 First, we write Ua = gae U e and apply the product rule: dUa d = (gae U e ) d d dgae dU e = gae + Ue . d d

Orbits

(7.2) (7.3)

Apply the geodesic equation to the rst term, and write d/ d = U b b in the second term: dUa = gae e bc U b U c + U e U b b gae . (7.4) d Next, we can change the dummy variable e c in the last term, so that we can factor out U b U c : dUa = gae e bc U b U c + U c U b b gac (7.5) d = (b gac gae e bc ) U b U c . (7.6) Note that g ed is the inverse metric tensor, so by equation (6.34), abc gae e bc = 1 gae g ed (b gcd + c gdb d gbc ) 2 1 d a (b gcd + c gdb d gbc ) = 2 1 (b gca + c gab a gbc ) . = 2 1 (b gca + c gab a gbc ) U b U c 2 (7.7) (7.8) (7.9)

Thus dUa = d = b gac (7.10)

1 (a gbc + b gac c gab ) U b U c , (7.11) 2 using gca = gac to combine the rst two terms in equation (7.10). Finally, the last two terms in equation (7.11) involve the factor (b gac c gab ), which is anti-symmetric in b and c. But this factor double-contracts with U b U c , which is symmetric in b and c. Contraction of symmetric and anti-symmetric tensors gives 0, i.e. b gac c gab U b U c = 0 so we are left with QED dUa 1 = (a gbc ) U b U c . d 2 (7.12) (7.13)

7.2

The Schwarzschild Metric

Consider the space surrounding a planet or star or black hole of total mass M . We will assume that the central object is spherically symmetric (so we ignore rotation) and time

7.2 The Schwarzschild Metric

81

independent (so we ignore time evolution). One can show from the Einstein eld equations that the metric line element is d 2 = 1 rs r dt2 1 rs r
1

dr2 r2 d2 r2 sin2 d2 .

(7.14)

where rs 2GM is called the Schwarzschild Radius. Equivalently the metric tensor is 1 rrs gab = 0 0 0 0 1 rrs 0 0 0 0 0 0 . 2 r 0 2 2 0 r sin

(7.15)

The Schwarzschild radius is quite small: for the Sun, M = M , rs = 3km. The radius of the sun, however, is R = 7 105 km rs . Thus for anything orbiting the sun, even in a very low orbit, r > R so the ratio rs /r 1. For the Earth, M = M , rs = 0.886cm. Again, for anything orbiting the Earth, rs /r 1. Also note that if we neglect the rs /r terms in the metric we get back to the Minkowski metric. 7.2.1 Symmetries and Conserved Quantities

To nd a planetary orbit, we solve the geodesic equation for objects moving in the curved space described by the Schwarzschild metric. This is dicult to do directly. However we can take advantage of two symmetries in the problem. The two symmetries are time invariance and rotational invariance: 0 gbc = t gbc = 0 for all b, c = 0, 1, 2, 3 3 gbc = gbc = 0 for all b, c = 0, 1, 2, 3. The corresponding conserved quantities are energy E and angular momentum L: dp0 = d dp3 = d dE = 0; d dL = 0. d (7.18) (7.19) (7.16) (7.17)

For massive particles we can dene the energy per unit mass k = E/m and the angular momentum per unit mass h = L/m. For constant rest mass h and k will also be constant along a geodesic. We rst consider k: k = U0 = g0b U b = rs 1 U 0. r (7.20) (7.21)

82 As U 0 = dt/ d , we have rs dt =k 1 d r
1

Orbits

(7.22)

Thus the Noether symmetry arguments lead to an expression for how co-ordinate time t varies with proper time . Next consider the angular momentum per unit mass h: h = U3 = g3b U b = r2 sin2 U 3 . Now U 3 = d/ d , so h d = 2 2 . d r sin (7.25) (7.23) (7.24)

Note how the velocity expressed as a vector U with upper indices U a has a dierent physical meaning from the form U with lower indices Ua . The vector U shows us where the object is going (as it represents the tangent to the world line). The form U, on the other hand, tells us how much energy and momentum (per unit mass) the object carries. 7.2.2 Orbits in the Equatorial Plane

Consider geodesics in the equatorial plane = /2. For the solar system this plane is called the ecliptic. (The constellations seen on the ecliptic are known as the zodiac.) For motion on this plane d = 0 and sin = 1, so the Schwarzschild metric line element simplies to rs rs 1 d 2 = 1 dt2 1 dr2 r2 d2 . (7.26) r r Let us nd orbit equations in terms of h and k: a. Divide the metric line element by d 2 and use equations (7.22) and (7.25) (with sin = 1): 1 = k = dr = d
2

rs 1 r

rs 1 r 1

dr d

h2 r2

(7.27) (7.28)

k2 1 +

h2 r2

rs . r

This gives us a dierential equation for dr/ d . Unfortunately, it is quite non-linear and dicult to solve in this form. Exercise 7.1 (a) Starting with equation (7.28), derive an expression for dr/d in the form 1 2 dr d
2

+ V (r) = C

7.2 The Schwarzschild Metric where C is a constant, and the eective potential is V (r) = 1 2 rs h2 rs h2 2 + 3 r r r .

83

What is the eective energy C? (b) Find the radii r1 and r2 , r1 < r2 where the eective potential has an extremum (maximum or minimum). Show that if C = V (r1 ) or C = V (r2 ) then the requirements for a circular orbit (dr/d = d2 r/d 2 = 0) are satised. Show that h 3 rs for these orbits. Also show that r2 3rs . (c) Let h = 2rs . What are r1 and r2 ? Show that for the outer orbit at r2 , V (r2 ) > 0 and hence that this orbit is stable. Is the inner orbit at r1 stable? b. To simplify the equation, we change the independent variable from , and nd r(): dr = d = dr d
2

dr d

d d

= 1

dr d rs . r

h r2

(7.29) (7.30)

h2 h2 = k2 1 + 2 r4 r

c. Next we change the dependent variable r u = 1/r. We will denote dierentiation by with a prime, e.g. u = du/ d. Thus r dr dr du = d du d 1 = 2u u = (7.31) (7.32) (7.33)

u 2 u

u4 h2 = k 2 1 + h2 u2 (1 rs u) = u 2 =

k 2 (1 + h2 u2 ) (1 rs u) (7.34) h2 h2 We now have an equation for u which at least has no terms in the denominator: u2 = k2 1 h2 + rs u u 2 + rs u3 . 2 h (Einstein) (7.35)

The corresponding Newtonian orbit equation leaves out the last term: u 2 = kN + rs u u2 , h2 V 2 GM V2 rs kN = = . 2 r 2 2r (Newton) (7.36)

where the Newtonian energy per unit mass is (7.37)

84 d. We can simplify further by dierentiating with respect to : rs u 2uu + 3rs u2 u 2 h rs 3rs 2 = u + u = u. + 2 2h 2 2u u = Thus General Relativity predicts that orbits satisfy u +u= rs 3rs 2 u . + 2h2 2 (Einstein)

Orbits

(7.38) (7.39)

(7.40)

In contrast, the Newtonian orbit equation is u +u= rs . 2h2 (Newton) (7.41)

Compare the relativistic correction term (the last term in the Einstein version) to the linear term u: 3rs u2 /2 3 rs GM = =3 . (7.42) u 2r r For planets orbiting the sun at a radius r > 100R , 3 GM 107 100R (7.43)

and so the relativistic correction term results in very small deviations from the Newtonian predictions. These deviations have, however, been observed! Exercise 7.2 a. Consider a sphere of radius 1 as a 2-dimensional manifold with coordinates x1 = , x2 = , and line-element ds2 = d2 + sin2 d2 . Show that geodesics have a conserved quantity (call it H). b. Using the metric line element, or otherwise, derive an equation for d/ds in terms of H. c. Let X = cos . Show that X satises dX ds
2

= 1 X 2 H 2.

d. Obtain a second-order dierential equation for X(s) and write down its general solution. Suppose a geodesic starts at co-latitude 0 heading due East. Find X(s) and hence (s) explicitly in terms of 0 . Show that the total length of the geodesic (i.e. the length needed to go all the way around the sphere once) is independent of 0 .

7.3 Precession of Mercurys Orbit

85

Figure 7.1: A visualization of the Schwarzschild Metric. More precisely, a 2-manifold imbedded in three dimensional space which has the same spatial metric as a constant time equatorial (t = constant, = /2) slice of the Schwarzschild metric (equation (7.26)).

Exercise 7.3 a. Consider a surface embedded in 3 dimensional Euclidean space (e.g. the surface of a bowl). Using cylindrical coordinates (r, , z), the surface is specied by the function z = Z(r). Let the two coordinates on the surface be x1 = r and x2 = . Show that the metric of the surface is given by 1 + Z 2 0 . (7.44) gab = 0 r2 b. Next consider geodesics on the surface. Since the surface is purely spatial, we replace by arclength s in the geodesic equations. Show that the metric has a symmetry, and hence there exists a conserved quantity (call it h) along each geodesic. In other words, nd a quantity h such that dh/ds = 0. c. Let u = 1/r and let u = du/d. Derive the equation u2 = 1 u2 2 h 1 . 1+Z 2

d. Now suppose Z(r) = 2 r 1. Show that this gives a surface whose metric is the same as the spatial part of the Schwarzschild metric, equation (7.26), (for = /2) in units where rs = 1 (see gure 7.1). What is the equation for u 2 ? How does this compare with the orbit equation derived from the full Schwarzschild metric?

7.3

Precession of Mercurys Orbit

Perihelion = Closest approach to the Sun.

86

Orbits

Observations show that the perihelion of Mercury precesses about 1000 arcsec / century. Inuences of the other planets account for all but 43 of this. Einstein found the 43 could be explained by the extra term in the orbit equation. A numerical approach might be to go back to the rst order equation, equation (7.35), and integrate directly: d. (7.45) = (k 2 1) /h2 + rs u/h2 u2 + rs u3 However, for planetary orbits we can exploit the fact that the relativistic correction to Newtonian theory is very small. This will enable us to nd a simple analytic solution (approximate, but then so are numerical solutions!). 7.3.1 Method du

The angle measures the net angle through which the planet has orbited the sun. During the rst orbit 0 2. During the second orbit, 2 4 and so on. But each successive orbit does not exactly follow the previous orbit, so for example u(2) = u(0). Thus the function u() cannot only contain terms periodic in like sin . There may be a linear term as well. We will solve for the function u(), then invert to obtain (u). Now at perihelion r = rmin , so u = umax . We will compare successive values of (umax ). Because the orbits are not exactly alike, (umax ) (umax ) = 2 +
orbit 2 orbit 1

(7.46)

for some angle . We will call the precession. But rst we need to solve for u(). We will do this by writing this function as the sum of the Newtonian solution plus a small correction term (we derive the Newtonian solution below): u() = u0 (1 + sin ) + y() ;
Newtonian Orbit Correction

(7.47)

rs . u0 (7.48) 2h2 We plug this into the orbit equation to obtain a new dierential equation for y(). As y() is small, we drop the non-linear terms (those in y 2 etc) to obtain a linear equation in y, which can then be readily solved.

7.3 Precession of Mercurys Orbit 7.3.2 Newtonian Solution

87

First, look at the solution to the Newtonian equations 7.36 and 7.41. The second order equation has sines and cosines as complementary functions and a constant as a particular solution, so the general solution can be written u() = A cos + B sin + u0 . (7.49)

Comment: Here there are two unknown constants of integration, as expected for a second order dierential equation. These could be determined, for example, by setting initial conditions u(0 ) and u (0 ) at some angle 0 . However, we started with a rst order equation, equation (7.36), which should only have one constant of integration. What has changed? When we dierentiated to obtain the second order equation, we lost some information about the orbit. In particular, we lost the term (k 2 1)/h2 . This is the only term which tells us about the energy k. A solution of the rst order equation may be characterized by just one of the initial conditions as well as k and h. The second order equation loses the explicit k dependence. Of course, the energy will be derivable as a function of the second initial condition, and vice-versa. The perihelion occurs at minimum r, hence maximum u. Let us suppose this occurs at 0 = /2. Then one initial condition gives u ( ) = 0 A = 0. 2 (7.50)

Now plug into the rst order Newtonian equation (7.36) (without the relativistic correction term rs u3 ). The result gives 1/2 r2 B = kN + s4 . (7.51) 4h Finally, dene the eccentricity by = to yield u = u0 (1 + sin ) . (7.53) The eccentricity tells us the shape of the orbit. Thus = 0 gives a circle, 0 < < 1 gives an ellipse, = 1 gives a parabola, and > 1 gives a hyperbola (the latter two are open orbits where an object comes in from , is deected, and escapes to ). The eccentricity of Mercurys orbit is about = 0.21. Note that for a circular orbit rs ucirc = u0 = 2 (7.54) 2hcirc hcirc = Thus for planetary orbits hcirc rs . rcirc 2rs
1/2

B u0

(7.52)

rs .

(7.55)

88 7.3.3 Relativistic Correction

Orbits

We now substitute equation (7.47) into the full relativistic orbit equation (7.40). This yields the following dierential equation for y(): y +y = 3rs u0 ((1 + sin ) + y)2 2 3rs u0 = (1 + sin )2 + 2 (1 + sin ) y + y 2 . 2 (7.56) (7.57)

We can ignore the y 2 term, as y 1. Next, compare the terms linear in y. There are two terms: on the left hand side, with coecient 1, and on the right, with coecient 3rs u0 (1 + sin ). The latter term is of order rs /r 1. Neglecting this term gives y +y 3rs u0 1 + 2 sin + 2
2

sin2 .

(7.58)

The terms on the right act as forcing functions for the harmonic oscillator on the left. The most interesting forcing function is the 2 sin term, as this is in resonance with the oscillator (the complementary functions include sin ). This resonance drives the precession. For initial conditions y(/2) = 0, y (/2) = 0, the solution is
2 3rs u0 cos + (1 sin ) + 2 sin2 sin 2 2 3 3rs u0 = cos + periodic terms. 2 2

y() =

(7.59) (7.60)

Recall the discussion leading to equation (7.46). The rst orbit has perihelion at 1 = /2 , while the second orbit has perihelion at 2 = 5/2 + . Solving u (2 ) = 0 gives 3rs u0 . The mean radius is r u1 = 58 106 km. Thus 0 0.1 arcseconds/orbit. (7.62) (7.61)

The Mercury year is 88 earth days, making the precession 1 arcsecond every 880 days, or 43 arcseconds per century.

7.4 Deection of Starlight

89

7.4

Deection of Starlight

As light travels from a star, its path is distorted by the curvature of space-time. Thus photons passing near a massive object like the sun will curve around the object. When the light is viewed, it will appear in the wrong place on the sky. This eect has been seen not only for starlight in the suns gravitational eld, but also for light from distant galaxies or quasars passing by nearer galaxies on the way to Earth. The eect on starlight is best seen during a solar eclipse. During an eclipse we can see stars in the sky close to the sun without being swamped by daylight. Photons from these stars receive the maximum deection. Figure 7.4 shows the path of the starlight. The photon starts at the star, with u 0; =+ . 2 (7.63)

Because of the gravitational deection, we observe the photon coming in from the angle = /2. Thus the angle of deection is . First we look at the Newtonian prediction, then derive the relativistic deection. 7.4.1 Newtonian Theory

The Newtonian orbit for m = 0 is, from equation (7.47) u() = u0 (1 + sin ) ; The angular momentum per unit mass h is h = r2 = rV . (7.65) u0 = rs GM = . 2 2h h2 (7.64)

In gure 7.4, the photon travelling from the star to our telescope has its closest approach (perihelion) at = /2. At perihelion dr/dt = 0, so Vr = 0 and thus |V | = V . But for a

90 photon |V | = c = 1 in relativistic units. So h = rmin R

Orbits

(7.66)

if the starlight just grazes the surface of the sun on its way to our telescope. Thus u() = At the star, equation (7.63) gives 0 GM R2 GM R2 1 + sin + 1 2 , 2 (7.68) (7.69) GM (1 + sin ) . R2 (7.67)

as sin( + x) x for small x. Thus 2/. We now have u() GM R2 1+ 2 sin . (7.70)

Finally, at = /2 the starlight reaches r = R , so 1 R GM R2 GM R2 2GM R 1+ 2 2 , , (7.71) (7.72)

using 2/

1. Thus . (7.73)

7.4.2

Relativistic Theory

As photons are massless, we consider the orbit equations in the limit m 0. Photons do have energy and momentum (and angular momentum), so we hold energy E = mk and angular momentum L = mh constant. The rst order equation (equation (7.35)) becomes u2 = In the limit m 0, then, u2 = E2 u2 + rs u3 . 2 L (7.75) E2 m L2 rs um2 u2 + rs u3 . L2 (7.74)

Dierentiation gives the second order equation 3 u + u = rs u2 . 2 (7.76)

7.4 Deection of Starlight First, consider the rs = 0 case. Here M = 0 and space-time is at: u +u = 0 = u = A cos + B sin

91

(7.77) (7.78) (7.79)

Apply boundary conditions that the light reaches r = at = 0, and (again) that the perihelion of the light path occurs at (r, ) = (R , /2). As a result, A = 0 and B = R1 , i.e. 1 u= sin . (7.80) R

This is the equation of a straight horizontal line (y = R ) in polar coordinates. Light passing near a massive object, however, will be perturbed by the missing rs term. As in the analysis of the precession of Mercury we try a small non-linear correction to u: u() = 1 (sin + y()) R (7.81)

where y() << 1. Then equation (7.76) gives 1 3rs 1 ( sin + y ) + (sin + y) = (sin + y)2 2 R R 2R 3rs = y + y = (sin + y)2 . 2R Include only the terms of lowest order in y and rs (as y, rs are both small): y +y 3rs 3rs sin2 = (1 cos 2). 2R 4R (7.84) (7.82) (7.83)

Combining complementary functions and particular integrals gives y = A cos + B sin + C + D cos 2 + E sin 2. We can determine the constants as follows: a. Plugging equation (7.85) into equation (7.84) gives C = 3D, D = rs /4R and E = 0. b. At perihelion, y(/2) = 0, so B = 2D. c. Also, as seen in the gure, the photon path is symmetric about = /2. Hence, for example, y(0) = y(), which implies A = 0. (7.85)

92 We now have

Orbits

rs (3 2 sin + cos 2). (7.86) 4R Finally, we can apply the boundary conditions at the star, equation (7.4). These give y= 0 = R u ( + /2) rs = sin ( + /2) + (3 2 sin ( + /2) + cos (2 + )) 4R rs = sin(/2) + (3 + 2 sin(/2) + cos 2) 4R rs + (4 + Order()). 2 4R (7.87) (7.88) (7.89) (7.90)

We can ignore the terms of order (rs /R ), giving 2rs 4GM = R R 1.74 arcseconds. (7.91)

Thus relativity predicts a deection almost exactly double the Newtonian result.

7.5

Energy Conservation on Geodesics

According to Newtonian theory, an object moving in a central gravitational eld has energy 1 GM E = mV 2 + m(r); (r) = , (7.92) 2 r where (r) is the potential energy. We now show that for weak gravitational elds (r rs ) and small velocities (V 1), the relativistic energy is approximately the same as the Newtonian energy, apart from the contribution from rest mass. Recall the Schwarzschild metric line element in the equatorial plane, equation (7.26). In terms of , d 2 = (1 + 2) dt2 (1 + 2)1 dr2 r2 d2 . (7.93) Next, divide by dt2 , using equation (7.21) for d / dt: d dt
2

1 (1 + 2)2 2 k
1

(7.94) dr dt
2

= (1 + 2) (1 + 2)

d dt

(7.95) (7.96)

2 = (1 + 2) (1 + 2)1 Vr2 V .

Now V 2 :

1 and V 2

1 so we can expand the right hand side, ignoring terms like 2 or 1 2 (1 + 2)2 (1 + 2) (1 2) Vr2 V k2 2 (1 + 2) (Vr2 + V ) = 1 + 2 V
2

(7.97) (7.98) (7.99)

7.5 Energy Conservation on Geodesics Thus k

2

93

(1 + 2)2 1 + 2 V 2 (1 + 4)(1 2 + V 2 ) 1 + V 2 + 2.

(7.100) (7.101) (7.102)

Finally, we take the square root of this approximation, using (1 + x)1/2 1 + x/2. With E = mk the Newtonian correspondence becomes clear: 1 E m + mV 2 + m. 2 (7.103)