Symmetry. March 31, 2013

Σχετικά έγγραφα
Parallel transport and geodesics

Phys460.nb Solution for the t-dependent Schrodinger s equation How did we find the solution? (not required)

6.1. Dirac Equation. Hamiltonian. Dirac Eq.

Second Order Partial Differential Equations

Homework 8 Model Solution Section

2 Composition. Invertible Mappings

Lecture 2: Dirac notation and a review of linear algebra Read Sakurai chapter 1, Baym chatper 3

Tutorial problem set 6,

SCHOOL OF MATHEMATICAL SCIENCES G11LMA Linear Mathematics Examination Solutions

Section 8.3 Trigonometric Equations

EE512: Error Control Coding

CHAPTER 25 SOLVING EQUATIONS BY ITERATIVE METHODS

Chapter 6: Systems of Linear Differential. be continuous functions on the interval

Matrices and Determinants

C.S. 430 Assignment 6, Sample Solutions

Orbital angular momentum and the spherical harmonics

Απόκριση σε Μοναδιαία Ωστική Δύναμη (Unit Impulse) Απόκριση σε Δυνάμεις Αυθαίρετα Μεταβαλλόμενες με το Χρόνο. Απόστολος Σ.

derivation of the Laplacian from rectangular to spherical coordinates

Partial Differential Equations in Biology The boundary element method. March 26, 2013

Parametrized Surfaces

HOMEWORK 4 = G. In order to plot the stress versus the stretch we define a normalized stretch:

Exercises 10. Find a fundamental matrix of the given system of equations. Also find the fundamental matrix Φ(t) satisfying Φ(0) = I. 1.

DESIGN OF MACHINERY SOLUTION MANUAL h in h 4 0.

( y) Partial Differential Equations

Example Sheet 3 Solutions

Finite Field Problems: Solutions

Areas and Lengths in Polar Coordinates

3.4 SUM AND DIFFERENCE FORMULAS. NOTE: cos(α+β) cos α + cos β cos(α-β) cos α -cos β

Areas and Lengths in Polar Coordinates

forms This gives Remark 1. How to remember the above formulas: Substituting these into the equation we obtain with

Geometry of the 2-sphere

Concrete Mathematics Exercises from 30 September 2016

Srednicki Chapter 55

Fourier Series. MATH 211, Calculus II. J. Robert Buchanan. Spring Department of Mathematics

Inverse trigonometric functions & General Solution of Trigonometric Equations

MATH423 String Theory Solutions 4. = 0 τ = f(s). (1) dτ ds = dxµ dτ f (s) (2) dτ 2 [f (s)] 2 + dxµ. dτ f (s) (3)

Reminders: linear functions

Solutions to Exercise Sheet 5

Section 9.2 Polar Equations and Graphs

Homework 3 Solutions

Space-Time Symmetries

Chapter 6: Systems of Linear Differential. be continuous functions on the interval

b. Use the parametrization from (a) to compute the area of S a as S a ds. Be sure to substitute for ds!

Riemannian Curvature

9.09. # 1. Area inside the oval limaçon r = cos θ. To graph, start with θ = 0 so r = 6. Compute dr

4.6 Autoregressive Moving Average Model ARMA(1,1)

A Short Introduction to Tensors

Appendix A. Curvilinear coordinates. A.1 Lamé coefficients. Consider set of equations. ξ i = ξ i (x 1,x 2,x 3 ), i = 1,2,3

Spherical Coordinates

D Alembert s Solution to the Wave Equation

CRASH COURSE IN PRECALCULUS

Math221: HW# 1 solutions

Numerical Analysis FMN011

CHAPTER 101 FOURIER SERIES FOR PERIODIC FUNCTIONS OF PERIOD

Pg The perimeter is P = 3x The area of a triangle is. where b is the base, h is the height. In our case b = x, then the area is

Lecture 15 - Root System Axiomatics

Ordinal Arithmetic: Addition, Multiplication, Exponentiation and Limit

Phys624 Quantization of Scalar Fields II Homework 3. Homework 3 Solutions. 3.1: U(1) symmetry for complex scalar

PARTIAL NOTES for 6.1 Trigonometric Identities

Every set of first-order formulas is equivalent to an independent set

The Simply Typed Lambda Calculus

Statistical Inference I Locally most powerful tests

Solutions Ph 236a Week 8

1 Lorentz transformation of the Maxwell equations

Problem Set 9 Solutions. θ + 1. θ 2 + cotθ ( ) sinθ e iφ is an eigenfunction of the ˆ L 2 operator. / θ 2. φ 2. sin 2 θ φ 2. ( ) = e iφ. = e iφ cosθ.

Quadratic Expressions

( ) 2 and compare to M.

ANSWERSHEET (TOPIC = DIFFERENTIAL CALCULUS) COLLECTION #2. h 0 h h 0 h h 0 ( ) g k = g 0 + g 1 + g g 2009 =?

PHYS606: Electrodynamics Feb. 01, Homework 1. A νµ = L ν α L µ β A αβ = L ν α L µ β A βα. = L µ β L ν α A βα = A µν (3)

Approximation of distance between locations on earth given by latitude and longitude

Second Order RLC Filters

ΚΥΠΡΙΑΚΗ ΕΤΑΙΡΕΙΑ ΠΛΗΡΟΦΟΡΙΚΗΣ CYPRUS COMPUTER SOCIETY ΠΑΓΚΥΠΡΙΟΣ ΜΑΘΗΤΙΚΟΣ ΔΙΑΓΩΝΙΣΜΟΣ ΠΛΗΡΟΦΟΡΙΚΗΣ 19/5/2007

Nowhere-zero flows Let be a digraph, Abelian group. A Γ-circulation in is a mapping : such that, where, and : tail in X, head in

Section 7.6 Double and Half Angle Formulas

Practice Exam 2. Conceptual Questions. 1. State a Basic identity and then verify it. (a) Identity: Solution: One identity is csc(θ) = 1

Exercise 1.1. Verify that if we apply GS to the coordinate basis Gauss form ds 2 = E(u, v)du 2 + 2F (u, v)dudv + G(u, v)dv 2

Variational Wavefunction for the Helium Atom

Congruence Classes of Invertible Matrices of Order 3 over F 2

= {{D α, D α }, D α }. = [D α, 4iσ µ α α D α µ ] = 4iσ µ α α [Dα, D α ] µ.

General 2 2 PT -Symmetric Matrices and Jordan Blocks 1

w o = R 1 p. (1) R = p =. = 1

Similarly, we may define hyperbolic functions cosh α and sinh α from the unit hyperbola

Other Test Constructions: Likelihood Ratio & Bayes Tests

Geodesic Equations for the Wormhole Metric

Written Examination. Antennas and Propagation (AA ) April 26, 2017.

If we restrict the domain of y = sin x to [ π, π ], the restrict function. y = sin x, π 2 x π 2

On a four-dimensional hyperbolic manifold with finite volume

Tridiagonal matrices. Gérard MEURANT. October, 2008

6.3 Forecasting ARMA processes

1 String with massive end-points

Higher Derivative Gravity Theories

Econ 2110: Fall 2008 Suggested Solutions to Problem Set 8 questions or comments to Dan Fetter 1

If we restrict the domain of y = sin x to [ π 2, π 2

TMA4115 Matematikk 3

Math 6 SL Probability Distributions Practice Test Mark Scheme

Lecture 34 Bootstrap confidence intervals

Dirac Matrices and Lorentz Spinors

Fractional Colorings and Zykov Products of graphs

CHAPTER 48 APPLICATIONS OF MATRICES AND DETERMINANTS

ST5224: Advanced Statistical Theory II

Transcript:

Symmetry March 3, 203 The Lie Derivative With or without the covariant derivative, which requires a connection on all of spacetime, there is another sort of derivation called the Lie derivative, which requires only a curve. Let C : R M be a curve in M with tangent vectors, ξ = d dλ, with components ξ = ξ µ x µ = dxµ λ x µ The Lie derivative generalizes the directional derivative of a function, df f = ξµ dλ x µ to higher rank tensors. First, consider a vector field, v, defined on M. We define the Lie derivative of v at a point P along C to be v P + εξ v P L ξ v = lim ε 0 ε where v P + εξ is the Lie transport of v along the curve. For simplicity, let P = C λ = 0. Lie transport involves taking the value of the vector field at a point on C, say, v λ, and performing a coordinate transformation to bring the point C λ back to P = C 0. The coordinate transformation we require is, for infinitesimal λ = ε, y α = x α εξ α 0 The components of v α change as ṽ α 0 = v β λ yα x β = [ v β x µ 0 + εξ µ ] δ α β ε β ξ α = [ v β 0 + εξ µ µ v β 0 ] δ α β ε β ξ α = v α 0 + εξ µ µ v α 0 εv β 0 β ξ α The derivative is then L ξ v = lim ε 0 v P + εξ v P ε = lim ε 0 v α 0 + εξ µ µ v α 0 εv β 0 β ξ α v 0 ε = ξ µ µ v α 0 v β 0 β ξ α An easy proof of the covariance of this result is that it equals the commutator of the two vectors, L ξ v = [ξ, v]

which has the same form when ξ and v are expanded in components, [ξ, v] = [ ξ α α, v β β ] = ξ α α v β β v β β ξ α α = ξ β β v α v β β ξ α α The generalization to higher rank tensors is immediate because derivations must satisfy the Leibnitz rule. Thus, for an outer product of two vectors, T αβ = u α v β we have L ξ T αβ = L ξ u α v β = L ξ u α v β + u α L ξ v β = ξ µ µ u α u µ µ ξ α v β + u α ξ µ µ v β v µ µ ξ β = ξ µ µ u α v β u µ v β µ ξ α u α v µ µ ξ β = ξ µ µ T αβ T µβ µ ξ α T αµ µ ξ β and so on for higher ranks, with one correction term, T α...µ...β µ ξ ν, for each index. For forms, we use the directional derivative of a scalar, L ξ φ = ξ µ φ x µ together with φ = v α ω α, for arbitrary v α, ξ µ vα ω α x µ = L ξ v α ω α ξ µ µ v α ω α + v α ξ µ µ ω α = L ξ v α ω α + v α L ξ ω α ξ µ µ v α ω α + v α ξ µ µ ω α = ξ β β v α ω α v β β ξ α ω α + v α L ξ ω α v α ξ µ µ ω α = v β β ξ α ω α + v α L ξ ω α Since this must hold for all v α, v α L ξ ω α = v α ξ µ µ ω α + v α α ξ β ω β L ξ ω α = ξ µ µ ω α + ω β α ξ β 2 Symmetry The Lie derivative is just the right tool for finding symmetry of a metric. The Lie derivative of the metric tensor is L ξ g αβ = ξ µ µ g αβ + g µβ α ξ µ + g αµ β ξ µ Now suppose we have a congruence of curves, so that the collected tangent vectors form a vector field. Suppose further we choose coordinates so that λ is one of the coordinates, x α0, for α 0 a single fixed direction. Then the components of ξ µ are constant, ξ µ = dxµ dλ = δ µ α 0 For example, if the curve is timelike, then we choose t = λ and we have ξ µ =, 0, 0, 0. For such a choice, β ξ µ = 0 and the Lie derivative of the metric is just L ξ g αβ = dxµ dλ µg αβ = g αβ λ 2

This means that if this Lie derivative vanishes, the metric is independent of the coordinate λ. Since the metric then does not change along the congruence of curves, we have a symmetry of the spacetime. Any direction in which the metric is not changing is called an isometry. We can find a differential equation to describe such symmetry directions. Setting the Lie derivative of the metric to zero, we have 0 = ξ µ µ g αβ + g µβ α ξ µ + g αµ β ξ µ = ξ µ µ g αβ + [ α g µβ ξ µ ξ µ α g µβ ] + [ β g αµ ξ µ ξ µ β g αµ ] = ξ µ µ g αβ + α ξ β ξ µ α g µβ + β ξ α ξ µ β g αµ = α ξ β + β ξ α ξ µ α g µβ ξ µ β g αµ + ξ µ µ g αβ = α ξ β ξ µ 2 αg µβ + β g αµ µ g αβ + β ξ α ξ µ 2 αg µβ + β g αµ µ g αβ = α ξ β ξ µ Γ µβα + β ξ α ξ µ Γ µαβ = α ξ β ξ µ Γ µ βα + βξ α ξ µ Γ µ αβ = D α ξ β + D β ξ α resulting in the Killing equation, ξ α;β + ξ β;α = 0 Given the metric, we can ask for all solutions to this equation. Solutions, if they exist, represent symmetry directions of the spacetime, i.e., directions in which the metric is unchanging. 3 Example: Symmetries of Minkowski spacetime Consider flat spacetime, for which the metric is Minkowski, η µν. In Cartesian coordinates, η µν = and the Christoffel connection vanishes, Γ α µν = 0. Then we may replace the covariant derivatives by partial derivatives, and the Killing equation is simply ξ α,β + ξ β,α = 0 Taking a further derivative, we have Now, cycle the indices twice, to give ξ α,βµ + ξ β,αµ = 0 ξ α,βµ + ξ β,αµ = 0 ξ β,µα + ξ µ,βα = 0 ξ µ,αβ + ξ α,µβ = 0 Adding the first two and subtracting the third we find 0 = ξ α,βµ + ξ β,αµ + ξ β,µα + ξ µ,βα ξ µ,αβ ξ α,µβ = 2ξ β,αµ so that the second derivative of ξ β vanishes. This means that ξ β must be linear in the coordinates, ξ α = a α + b αβ x β 3

Substituting this into the Killing equation, 0 = ξ α,β + ξ β,α = b αβ + b βα so that a α is arbitrary while b αb must be antisymmetric. We now have 0 independent vector fields, each of the form ξ α = a α + b αβ x β for independent choices of the 0 constants a α and b αβ = b βα. The simplest choice of the 0 vector fields is to take only one of the constants nonzero. If we take b αβ = 0 and one of the components say, m for m =, 0,, 2, 3 of a α nonzero, we get four constant vector fields, ξ α m = δα m This represents a unit vector in each of the coordinate directions. Since they are constant, the integral curves are just the Cartesian coordinate axes, and the metric is indeed independent of each of these. Now setting a α = 0 and choosing one of the six antisymmetric matrices b αβ, we have either rotations or boosts. For example, with b 2 = b 2 =, with all the rest zero, the vector field is ξ = ξ α α = η αβ b βµ x µ α = x y y x This is the generator of a rotation around the z axis. Similarly, b 23 = b 32 and b 3 = b 3 lead to the generators of rotations around the x and y axes. If one of the nonzero indices is time, then we have a boost because of the sign change. For b 0 = b 0 =, we find ξ = ξ α α = η αβ b βµ x µ α = x t + t x This is a generator for a Lorentz transformation. To see this, exponentiate the generator with a parameter, [ Λ = exp λ x t + t ] x = n! λn x t + t n x Consider the effect on the coordinates t, x, y, z. Clearly, Λy = Λz = 0. For t we need x t + t t = x x n=0 x t + t x 2 t = = t x t + t x and so on, alternating between x and t. The even and odd parts of the series therefore sum separately, Λt = n! λn x t + t n t x n=0 4 x

Similarly, acting on x we get = m=0 2m +! λ2m+ x + = x sinh λ + t cosh λ m=0 Λx = t sinh λ + x cosh λ We recognize λ as the rapidity, and the full transformation, Λt = x sinh λ + t cosh λ Λx = t sinh λ + x cosh λ Λy = y Λz = z 2m! λ2m t as a boost in the x-direction. We therefore find exactly 0 isometries in Minkowski space. This is the maximum number of independent solutions to the Killing equation. The static, spherically symmetric Schwarzschild solution had one timelike Killing field and three spatial rotational Killing fields for a total of three. A generic spacetime has no isometries. 4 Example: Static, Spherically Symmetric Spacetimes We may now say what we mean by a static, spherically symmetric spacetime. To be static, there must be a timelike Killing vector field; to be spherically symmetric, we require a full set of three rotational hence spacelike Killing vectors. We use the Lie derivative to say restrict the form of the metric for a static, spherically symmetric spacetime. If we want a static spacetime, it means that we want there to exist a timelike Killing vector field. Choosing the time coordinate to be the parameter t = λ, the symmetry condition becomes 0 = L ξ g αβ = ξ µ µ g αβ + α ξ µ g µβ + β ξ µ g αµ However, with x 0 = t = λ, the components of ξ are constant, so that Therefore, α ξ µ = 0 0 = ξ µ µ g αβ = t g αβ and we have a coordinates system in which the metric is independent of the time coordinate. For the spherical symmetry, we know that we have three rotational Killing vector fields which together generate SO 3. We can pick two of these for coordinates, but they will not commute with one another, so the metric will not be independent of both coordinates. Starting with the familiar form ξ = y z z y ξ 2 = z x x z ξ 3 = x y y x 5

it is natural to choose one coordinate, ϕ, such that is a Killing vector. rotations, ϕ = x y y x To describe a second direction, we want a linear combination of the remaining two and we want this to remain orthogonal to ξ 3, α ϕ ξ + β ϕ ξ 2 0 = ξ 3, αξ + βξ 2 = x y y x, α y z z + β z y x x z = x y, α y z z + β z y x x y z x, α y z z + β z y x x z = x y, αz y y x, βz x = αzx y, βzy y x, x = z αx + βy = r sin θz α cos ϕ + β sin ϕ To get zero, we can take Then we have Compare ξ 4 = ξ sin ϕ + ξ 2 cos ϕ = sin ϕ y z z y = sin ϕ cos ϕ α = sin ϕ r sin θ sin ϕ z r cos θ y β = cos ϕ z x x z cos ϕ = r sin θ sin ϕ sin ϕ z r cos θ sin ϕ y = cos θ r cos ϕ x + r sin ϕ y = cos θ x x + y r sin θ y z x = x r y r cos θ x r sin θ cos ϕ z r cos ϕ cos θ x + r sin θ cos ϕ cos ϕ z + r sin θ z r + xz x2 + y 2 r 2 θ y x 2 + y 2 ϕ = sin θ cos ϕ r + r cos θ cos ϕ θ sin ϕ r sin θ ϕ = y r r + yz x2 + y 2 r 2 θ + x x 2 + y 2 ϕ 6

so we have z ξ 4 = cos θ = sin θ sin ϕ r + r cos θ sin ϕ θ + r = z r r x2 + y 2 r 2 θ = cos θ r sin θ r x x + y y θ + r sin θ z cos ϕ sin θ = cos θ r sin θ cos 2 ϕ r + cos θ cos2 ϕ θ + r sin θ sin2 ϕ r + cos θ sin2 ϕ θ r sin θ cos θ r sin θ r θ ϕ = r sin θ cos θ r + cos2 θ θ r sin θ cos θ r + sin2 θ θ = cos 2 θ θ + sin2 θ θ = θ We may therefore take two of the Killing vectors to be ξ 4 = θ ξ 3 = ϕ giving two coordinates, θ, ϕ, corresponding to symmetry directions. Since these do not commute, the metric cannot be independent of both. 7